Treatise on Water Science vol III (Aquatic Chemistry and Biology)

Preface – Aquatic Chemistry and Biology FH Frimmel, Karlsruhe Institute of Technology, Karlsruhe, Germany & 2011 Elsevier B.V. All rights reserved.

The World of Aquatic Chemistry and Microbiology Aquatic chemistry and microbiology do not belong to the classical subjects taught in universities. Nevertheless, they are part of many curricula in natural sciences and engineering. It is beyond doubt that the fascination of the molecular dimension of water itself and all its constituents, which goes like a red threat through all the aspects of structure, transport, and reactions of and in aquatic systems, attracts so many people. Due to the broad and fundamental importance of water for life, including the humans, the molecular water sciences (MoWaS) have to be transdisciplinary. The discipline includes not only physics, chemistry, biology, and geology, but also mathematics, engineering, and economics and even parts of social sciences. As a consequence, several subjects have developed based on fundamental ones but focusing on the special aspects of water, examples of which include limnology, oceanography, hydrogeology, hydrology, groundwater dynamics, drinking water treatment, municipal water management, industrial water usage, wastewater treatment, and hydrothermal usage. Many of them either are cross-linked or bridge the gap to the fields of quantitative water management. The big challenge when dealing with MoWaS can be deduced from the nano- and microscale of the substances involved and their low concentrations. The related bio-response can range from subtle to acute toxic effects. Methods to obtain reliable results are still scarce, especially for applications in natural environment. Here, the influences of matrices and the synergetic or antagonistic effects in multicomponent samples are often unclear. It is well accepted that water is the fundamental basis for our known life and in its unique function cannot be replaced by anything else. The physical properties of liquid water are reflected in its properties as transport medium, reaction phase, and mediator for higher molecular structures. One of the most impressive properties of the water molecules is the ability to form intermolecular hydrogen (H)-bonds. Linus Pauling once said, ‘‘y the hydrogen bond is especially suited to play a part in reactions occurring at normal temperatures, and I believe that it will be found that the significance of the hydrogen bond for physiology is greater than of any other single structural feature.’’ In other words, the formation and breaking of H-bonds in the energy band of our common environmental situation deliver the key for understanding life and its supporting element – water. It is also obvious that all major changes in water quality and temperature, for example, as a result of climate change, must have an influence on the dynamics of reactions and on the material balances involved. This again will influence the water cycle and hence the aquatic resources. Here, water management comes into the focus. Different kinds of water use with different influences on water quality in small- or large scale must be considered. Industrial development and population growth have led to one of the biggest

challenges to supply sufficient and hygienically safe water for human consumption and food production. Severe water shortage and necessary water quality are issues that have arisen regionally and are predicted to intensify drastically during the following decades. Concepts for multiple water use and water reuse need to be developed, taking advantage of the specific hydrological, climatic, and ecological situations. In addition, the special demands of social communities such as mega cities or developing countries have to be considered. Wherever possible, the ecological functions of regions must be protected for it is most reasonable to use nature as a self-sustained system also for water cleaning. The protective function of soils and their capability to degrade and eliminate aquatic pollutants make it attractive to use groundwater as a resource for drinking water supply, especially when protective zones and assisting technical measures are established. Toxicity and hygiene reflecting criteria are, besides the technical aspects such as corrosivity, most important for the use of water. A meaningful assessment of the use-oriented water quality has also to include parameters which quantify, for example, biota friendliness, potential for bacterial growth, eutrophication, and disinfection by-product formation. Occurrence of pathogenic microorganisms and waterborne epidemic episodes belong to the most serious events often with peaks in wars, natural disasters, and badly managed camps, homes, and companies. Quite often, shortcuts between the systems for drinking water supply and wastewater discharge have been identified as reason. Economic aspects are one of the master drivers for use of water and its management. On the one hand, the availability of enough water of suitable quality has been discussed as an issue of human rights. On the other hand, water has become a trade good, which is sold directly in bottles or through pipes or as virtual water in the manifold forms of industrial products. No matter how much profit might be involved in this business, the availability of reasonable resources and economically feasible treatment technologies will play a fundamental role. The application of cheap energy sources such as sun light and the use of homogeneous and heterogeneous catalysis, including biocatalysis, lead to most promising watertreatment concepts. Intelligent combination and an optimized sequence of treatment steps can further improve the economy of water plants. Hybrid systems are suited for highly efficient water treatment in fast working small reactors with the advantage of decentralized application. Keeping these aspects in mind, it becomes obvious that understanding the details of the properties of living and nonliving water constituents, their reactivities, and transport behavior will help to tailor powerful methods for waterquality assessment and to derive efficient concepts for timely water-treatment processes. The water cycle is an ideal case study not only for its different stages and hot spots, but also as

1

2

Preface – Aquatic Chemistry and Biology

a whole which can teach us the systematic approach to complex systems and to the solutions of the related man-made problems. It also shows the necessity of transdisciplinary thinking in the sense of lifelong learning. Starting in the early days of childhood, we need to lay the foundation for a responsible care for water as a basis for our life and culture. Furthermore, we need to invest in the tools for a sustainable water management by developing measures to save the water cycle in its proper ecological function. This calls for the classical components of teaching and research and beyond that for innovative concepts to serve the daily needs of water usage in an economically affordable and socially acceptable way. To serve this aim, a comprehensive treatise on water is presented. Volume 3 of this work includes the chemistry and microbiology of MoWaS. The analytical aspects cover water-specific sum parameters, methods for the determination of trace metals and metalloids, as well as radioactive substances, and the characterization of natural organic matter (NOM). Emerging contaminants, colloids, and engineered nanoparticles are presented and data handling is described. The identification of bacteria and parasites helps to characterize the hygienic status of water. Online monitoring,

screening of estrogen activities, and enzyme-linked immunotests show the way to modern concepts for continuous quality control and bioeffect-related assessment. The development and application of standardized methods supply tools to obtain reproducible and well-comparable results. For the special needs of water treatment and distribution, it is most useful to quantify biodegradability and toxic effects. Reaction mechanisms of oxidation and disinfection processes as well as bioremediation are important not only to understand the pathways of technical transformations and natural attenuation, but also to optimize treatment strategies. All these topics are addressed by leading experts in the field. They all intend to supply for the interdisciplinary water community the molecular facts for a meaningful diagnosis of the status of aquatic systems and for efficient technical processes within the water cycle. As the editor of this volume, I would like to thank all the authors for their valuable contributions. Furthermore, I am grateful to U. Bilitewski, T. Bu¨nger, G. Donnevert, G. Gauglitz, H. Geckeis, B. Hambsch, T. Hofmann, H. Horn, T. P. Knepper, D. Knopp, V. Neitzel, R. NieXner, B. Nowack, F. Petry, H.-J. Pluta, M. Spiteller, and M. Weller for their input by peerreview.

3.01 Sum Parameters: Potential and Limitations FH Frimmel and G Abbt-Braun, Karlsruhe Institute of Technology, Karlsruhe, Germany & 2011 Elsevier B.V. All rights reserved.

3.01.1 3.01.2 3.01.3 3.01.3.1 3.01.3.1.1 3.01.3.2 3.01.3.2.1 3.01.3.3 3.01.3.4 3.01.3.5 3.01.3.5.1 3.01.3.5.2 3.01.3.5.3 3.01.3.6 3.01.4 3.01.4.1 3.01.4.2 3.01.4.2.1 3.01.4.2.2 3.01.4.2.3 3.01.4.2.4 3.01.4.3 3.01.4.3.1 3.01.4.3.2 3.01.4.3.3 3.01.4.3.4 3.01.4.4 3.01.4.4.1 3.01.4.4.2 3.01.4.4.3 3.01.4.4.4 3.01.4.4.5 3.01.4.5 3.01.5 3.01.5.1 3.01.5.2 3.01.5.3 3.01.5.4 3.01.6 3.01.6.1 3.01.6.2 3.01.6.3 3.01.6.4 3.01.7 3.01.7.1 3.01.7.2 3.01.7.2.1 3.01.7.2.2 3.01.7.2.3 3.01.7.2.4 3.01.7.3 References

Introduction General Considerations and Scope DOC and TOC Background Relevance Analytical Procedure Method variations Interferences Advanced TOC (DOC) Characterization Applications Hydrosphere Surface water Water treatment Surrogate Parameters Oxygen Demand Parameters Introduction Chemical Oxygen Demand Background Analytical procedure Interferences Applications PMC and Permanganate Index (IMn) Background Analytical procedure Interferences Applications Biochemical Oxygen Demand Background Analytical procedure Interferences Applications Related parameters (AOC) Interdependences UVA and Visible Range Absorbance Background Analytical Procedure Interferences Applications Organically Bound Halogens Adsorbable on Activated Carbon (AOX) Background Analytical Procedure Applications Related Parameters Additional Sum Parameters Background Examples for Emerging Parameters Humic substances NPs and colloids Luminescence Bioeffect quantification View

3 3 3 3 4 4 5 5 5 7 7 7 7 8 8 8 9 9 9 10 12 12 12 13 13 13 13 13 13 14 14 14 15 15 15 15 16 16 18 18 18 19 19 19 19 20 20 21 22 22 23 23

3

4

Sum Parameters: Potential and Limitations

3.01.1 Introduction The general assessment of the quality of aquatic systems and the judgment of the efficiency of a water-treatment facility quite often relies on the application of sum parameters. Sum parameters are normally based on an integrative quantification of a specific group of compounds. However, the results obtained are mostly operationally defined and are often prone to misinterpretation. Therefore, it is essential to understand the power and at the same time the limitations of the parameters applied. Like signposts, they can give first information on assessment strategies and the necessity of singlecompound analysis. They are also suited for a total balance even in the presence of compounds with unknown structure (Frimmel and Abbt-Braun, 2009; Abbt-Braun and Frimmel, 2010). All these advantages have led to a prosperous development of water-specific sum parameters and their application in legislation, in technical rules, and in environmental recommendations. It is beyond doubt that research in and development of sum parameters have been significantly influenced by the practical aspects of water quality and vice versa. The applicability of the corresponding methods has also stimulated the development of specific instrumentation (see also Chapter 3.10 Online Monitoring Sensors). Some of the instruments are well suited for continuous measurements and can be used as online detectors. This opens the door for the resolution of mixtures by chromatography or by other fractionation methods. As a consequence, the sum parameter-based detector systems have an important bridging function between unresolved sum parameter quantification and single substance determination. All these aspects have led on the one hand to a tremendous increase of valuable information, but on the other hand to often uncritical interpretation of the results. The aim of this chapter is to focus on some well established sum parameters and to highlight their characteristics such as 1. 2. 3. 4. 5. 6.

background, principle of the method, interferences and limitations, advanced method, application, and related parameters.

3.01.2 General Considerations and Scope Sum parameters such as single-compound determination have to fulfill task-specific minimum requirements concerning exactness. It has to be decided whether the principle ‘as exact as possible’ or the approach ‘as exact as necessary’ meets best the requirements of the specific task. Often the desire for a specific and sensitive measurement finds its limitation in the needs of a high throughput of samples and/or a low economic investment. A reasonable compromise can normally be reached by a sound problem analysis prior to the determination itself. In general, classical spectroscopic and electrochemical methods cover the concentration range in aqueous samples from mg down to ng l1 (Skoog et al., 2003; Standard

Methods, 2005). This puts the application of sum parameters right into the center of a comprehensive assessment concept which is open for a dynamic back bonding of the results with the selection of further analytical steps. As a consequence, sum parameters have found their way into legislation and assessment of environmental protection with all the demands of data quality acceptance in court cases. In this chapter, we discuss in depth the parameters: dissolved organic carbon (DOC) and total organic carbon (TOC), chemical oxygen demand (COD), permanganate consumption (PMC), biochemical oxygen demand (BOD) and assimilable organic carbon (AOC), the color and ultraviolet (UV) absorbance (UVA), and on activated carbon adsorbable organically bound halogens (AOX). Most of these parameters refer to the dissolved state of the matter to be determined. Filtration through membranes with nominal pore size of 0.45 mm is widely used as analytical operation even though there might be pitfalls from pore blocking, fouling layer formation, or scaling. Quite often, the water samples are analyzed without pretreatment. This has to be clearly stated in the protocol and is normally assigned as total concentration value, for example, TOC. To close the gap between the dissolved state and particulate matter, a method for the determination of the particle-size distribution in the nanometer (nm) range is presented.

3.01.3 DOC and TOC 3.01.3.1 Background The basis for the parameters DOC and TOC is the chemical definition of organic compounds. They can be of biogeogenic or anthropogenic origin. Most natural organic substances in water are the left overs of biological activities and products of a huge variety of naturally occurring physical, chemical, and biochemical reactions in air, soil, and water. The endless number of possible substances involved in these processes makes the identification tedious and, from the quantitative point of view, impossible. Therefore, the terms natural organic matter (NOM) or humic substances (HSs) as the refractory part of it are often used for an integrative description, and the parameters DOC or TOC for quantification (Thurman, 1985; Frimmel and Christman, 1988; Perdue and Gjessing, 1990; Frimmel et al., 2002). Organic compounds of anthropogenic origin can find their way into the aquatic systems from effluents of wastewater treatment plants and industrial activities, from chemical wastes and landfills, by accidents during storage and transport of organic chemicals, and from combustion and by deposition from the air (Kolpin et al., 2002; Frimmel and Mu¨ller, 2006; Reemtsma and Jekel, 2006; Ku¨mmerer, 2008). In the current industrialized environment, it is quite idle to distinguish strictly between the purely natural components and the anthropogenic ones in many cases. Concerning the quantification for C, this might be irrelevant anyhow. Table 1 gives an overview of the different C species defined according to their character and/or to the pretreatment of the sample prior to elemental C determination. In practical work, the definitions often are only semi-quantitatively accurate.

Sum Parameters: Potential and Limitations Table 1

Common terms for property-related TOC fractions

POC) is retained together with the sorbed substances in/on the filter and the volatile compounds (volatile organic carbon, VOC) are normally lost:

Synonym

Meaning, definition

AOC BOC

Assimilable OC (see Section 3.01.4.4.5) Biodegradable OC (by microorganisms) (see Section 3.01.4.4) Chromatographable OC (by LC, GC, etc.) (see Section 3.01.3.4) Dissolved OC (o0.45 mm) Dissolved OM (E50% DOC) Natural OM (geogenic) Particulate OC (40.45 mm) Persistent organic pollutants Refractory OM (poorly biodegradable) Volatile OC (e.g., boiling point (substances)o80 1C)

COC DOC DOM NOM POC POP ROM VOC

OC, organic carbon; OM, organic matter.

3.01.3.1.1 Relevance The relevance of TOC can be deduced from its character as universal parameter. Other parameters reflecting specific properties of organic matter (OM) such as DOC or AOC or surrogate parameters such as UVA or COD can preferably be related to the TOC value to provide the basis for an especially meaningful comparison of water samples. However, it has to be kept in mind that TOC as sum parameter always remains limited in the information it can supply on the chemical structure of the matter it reflects. The instrumental tools suited for continuous TOC determination can be used as detection system for chromatographic TOC fractionation and by this it can help to overcome the limitation of structural information to some extent. The total carbon (TC) includes all C in inorganic and organic form (Equation (1)). The total inorganic carbon (TIC) reflects mainly the carbonate system (CO2, HCO3  , and CO3 2 ), and by definition also the traces of CO, CN, OCN, and SCN, which might be of relevance in specific wastewaters:

TC ¼ TIC þ TOC

5

ð1Þ

TOC comprises all the C atoms which are covalently bound in organic molecules and even particulate matter like carbon black. In natural aquatic systems and water technology, the carbonate system is considered to be most relevant due to its high mass concentrations. TIC can be quantified as CO2 after acidification (pHo2) and purging with an inert gas of high purity such as N2 or Ar. The purging step would also transfer other volatile substances such as HCN or small organic molecules such as methane (CH4), methanol (CH3OH), and C1or C2-halogen compounds from the aqueous phase to the gas phase. Due to the low concentrations of such volatile compounds in most waters, this is often neglected in the mass balances, but it can be important in the assessment of specific situations such as the occurrence of toxic substances. The TOC includes the organic carbon (OC) in dissolved and particulate matter. These two types of C can be distinguished by filtration through a 0.45-mm pore-size membrane, leading to the DOC in the filtrate. The particulate part (particulate organic carbon,

TOC ¼ DOC þ POCk þ VOCm

ð2Þ

This method has been widely accepted, even though there are still controversial debates on the influence of the mostly poorly defined filter cakes, on the results and whether the pore size of the filter should be chosen to be 0.1 mm or even below that to better reflect the dissolved state. A well acceptable way out of these problems can be seen in a detailed description of the experimental protocol of the method applied. The occurrence of TOC is a consequence of life on the Earth. The ubiquity of TOC in aquatic systems has been demonstrated in many investigations (Table 1). It is mostly refractory, that is, the biologically stable part of dissolved organic matter (DOC) which leads to a kind of steady-state TOC concentration in the different aqueous phases. DOC is often used synonymously with TOC, and other properties of the OM are reflected in specific parameters (Table 1). It is obvious that some terms and their definitions must remain vague. This means that the concerned parameter values can have a considerable span of uncertainty. Keeping this in mind, it seems to be acceptable to use Equation (3) as an approximation based on many elemental analyses. Unfortunately in literature, the databases are quite often unclear. Therefore, experimental data and procedures need to be described in detail and unambiguously to be useful:

rðDOMÞE 2rðDOCÞ

ð3Þ

3.01.3.2 Analytical Procedure Due to the high importance of TOC and DOC values in water assessment, there are standardized international methods for their determination (DIN EN 1484, 1997; Standard Methods 5310 B, 5310 C, 5310 D, 2005; see also Chapter 3.11 Standardized Methods for Water-Quality Assessment). They are mostly based on a quantitative oxidation of the organic molecules to CO2 which can be determined with a very low limit of determination around 10 mg l1. Oxidation is done either by high-temperature (up to 950 1C) combustion in the presence of a catalyst (e.g., platinum-group metals, cobalt oxide, or barium chromate) and oxygen or at ambient temperature in solution using UV irradiation and/or chemical oxidants such as H2O2 or persulfate. Inorganic C (IC) has to be removed within a pretreatment step, for example, by acidification with H3PO4 and purging as CO2. This separation step is most important for reliable TOC results, because TOC is quantified also as CO2 and this value is much smaller than the IC concentration in most waters. The CO2 produced from the inorganic carbonate system and from the organic water constituents can be (a) quantified in the gas phase after drying and transfer to a nondispersive infrared (IR) detector or (b) trapped in alkaline aqueous solution with coulometric titration. The principle of a system based on continuous flow injection of the sample is shown in Figure 1. Calibration can be done with defined aqueous potassium biphthalate (C8H5O4K) solutions for OC and with sodium

6

Sum Parameters: Potential and Limitations

CO2 analyzer

CO2 analyzer

Inorganic CO2

Organic CO2

Purger

UV reactor

Aqueous sample

H3PO4

Data system

Liquid waste

K2S2O8

CO2-free air

CO2-free N2

Figure 1 System for continuous-flow TIC/TOC analysis.

carbonate (Na2CO3) solutions for IC. The different methods operate in the concentration range 10 mg l1or(C)o1 g l1.

3.01.3.2.1 Method variations There is another procedure for continuous-flow injection of the aqueous sample (Figure 2). After acidification and persulfate addition, the sample is split: one sample flow passes through the UV reactor, whereas the other one passes to a delay coil. The CO2 from each branch is separated by CO2selective membranes into high-purity water. There the increase in the electrical conductivity can be directly related to the CO2 concentration. The CO2 from the non-UV-irradiated branch represents the TIC, whereas the CO2 from the irradiated branch represents the TC. TOC results from the difference. Samples with relatively high levels of TOC (r(C)4mg l1) and/or suspended OC can be well determined by the hightemperature combustion method. This method is suited for online measurement. The inorganic carbon can be converted to CO2 by acidification (pHo2) and removed by purging or it can be quantified, for example, in a nondispersive IR detector. In the purged sample, OC can be quantified after high-temperature catalytic oxidation as CO2 (Figure 3). A variation of the method determines the TC of the sample after its direct injection into the combustion chamber which is kept at temperatures above 950 1C to decompose all carbonates. TIC and TOC or other carbon fractions can be deduced from the respective differences.

3.01.3.3 Interferences Special care has to be taken with TOC determination of suspensions. Often the analytical homogeneity and hence the representative character of a sample are endangered by sedimentation and its kinetics. A way out of that dilemma is the

Aqueous sample Acid Oxidant (K2S2O8)

Delay coil 6 min

UV reactor 6 min

Membrane module

Membrane module

CO2 detector (TIC)

CO2 detector (TC)

Data treatment TC − TIC = TOC Figure 2 Membrane-based procedure for the continuous-flow TIC/TOC analysis.

separate quantification of the concentration of particulate matter and that of the dissolved matter, for example, after applying filtration through a membrane with defined pore size. However, it has to be kept in mind that the filtration can be influenced by the type of membrane and its bleeding (Khan

Sum Parameters: Potential and Limitations

7

Inorganic CO2 (TIC) H2SO4 or H3PO4

Sample

Catalytic combustion chamber

Purging unit

Gas (air) CO2 free

O2 CO2 free

Organic CO2 (OC) (+ H2O, HCl …)

Cooler CO2 analyzer (non-disp. IR)

Figure 3 Experimental setup for the catalytic combustion method for TIC/TOC determination in aqueous solutions.

and Pillai, 2007), its surface tension, and age and state of equilibration. In addition, undefined pore blocking and sorption processes have to be considered. The DOC concentration at the beginning of a filtration experiment can be quite different to the DOC concentration at the end. For samples with high turbidity, filtration through a set of filters with decreasing pore size and determination of the fractions obtained can be a reasonable though time-consuming option. Another possibility is the determination of the colloidal index (CI, also known as silt density index SI or fouling index FI) for OC characterization (ASTM Standard D4189-07, 2007). The principle of the method is to relate the specific filtrate volumes with the time needed to obtain them as the filtration process proceeds. In case short-wavelength UV lamps are used to increase the amount of OH radicals for oxidation, care has to be taken as the intensity of the UV light may be reduced by highly turbid samples or by aging of the light source resulting in incomplete oxidation. Problems can also arise by chloride concentrations above 0.05 wt.%, due to preferential oxidation of chloride. At relatively low DOC concentrations as present in marine systems, special care has to be taken to guarantee correct results. Dafner and Wangersky (2002) showed that special attention toward the cleanness of the sampling facilities and procedure is crucial. Sample storage should be short (o2 days) at low temperature (o4 1C) and in the dark. Examples for field procedures to collect and preserve freshwater samples for DOC analysis were shown by Kaplan (1994), and Zsolnay (2003) addressed some basic problems and artifacts such as flock formation and agglomeration in sampling and preserving DOM from soil seepage water (see also Chapter 3.06 Sampling and Conservation, Chapter 3.07 Measurement Quality in Water Analysis). Blank samples should be run to determine background values of equipment, used chemicals, gases, and filters (in the case of DOC determination).

3.01.3.4 Advanced TOC (DOC) Characterization The great relevance of TOC and DOC parameters for the assessment of aquatic systems together with the available

powerful instrumentation for quantification paved the way for their advanced analysis. In addition to all, the intelligently designed experiments which are controlled by a suite of TOC or DOC measurements, a liquid chromatographic (LC) system with online DOC detection has been developed (Huber and Frimmel, 1994) for advanced OC characterization. Especially the principle of size-exclusion chromatography (SEC; e.g., TSK HW-50S or -40S turned out to be useful for the assessment of DOM and its behavior in water-treatment processes (Her et al., 2002b). The principle of the method is given in Figure 4. To reach high chromatographic resolution and low detection limits, special care has to be taken for low background levels of OC. This means that the phosphate buffer as mobile phase, the N2 carrier gas, and the phosphoric acid as acidifier for the CO2 purging of the inorganic carbonates have to be free of organic contamination. The sample can be injected to either pass the column or bypass it. This leads to the possibility of determining the amount of chromatographable OC and the TOC. Between the column and the spinning thin film photoreactor, noninvasive online UV/visible (Vis) and fluorescence detectors can be installed to give multi-dimensionally detected chromatograms. In principle, the retention times (or elution volumes) obtained for SEC columns are reversely correlated with the molecular size and in good approximation with the molecular weight of the eluted substances. The column elution can be calibrated with polyethylene glycols and/or polystyrene sulfonates. The exclusion volume (V0) and the permeation volumes (VP) can be determined by dextrane blue and methanol, respectively. However, the molecular size calibration bears some problems because aquatic TOC contains many unknown substances and hence calibration with authentic molecules is impossible (Lankes et al., 2009). Most common errors come from interfering adsorption and ion-exchange effects of the eluted substances in the stationary phase of the columns. A typical chromatogram of the OM in tap water obtained by UV (l ¼ 254 nm), fluorescence (lex ¼ 254 nm, lem ¼ 450 nm), and OC detection is given in Figure 5. The OC trace of the chromatograms of the injected water with a TOC concentration of 0.5 mg l1 shows a dominance of

8

Sum Parameters: Potential and Limitations

Aqueous sample

Data logging processing Column (e.g., SEC)

Eluent (P-buffer)

Piston pump

Injection port

UV/Vis detector

Phosphoric acid

Carrier gas (N2)

Fluorescence detector

pH 2 Piston pump

Peltier condenser

UV thin film reactor

IR detector

Inorganic CO2

IR detector

Organic CO2

Liquid waste

Relative OC-, UV (254)-, fluorescence (450)-signal

Figure 4 Experimental setup for the size exclusion chromatographic characterization of aquatic OC.

Vp

V0

OC Fluorescence UV

1.0

0.5

0 20

40

60

80

Elution volume, Ve (ml) Figure 5 Multi-dimensional size exclusion chromatograms for tap water (Karlsruhe, sampling date 07.07.09; r(DOC) ¼ 0.5 mg l1; resin: TSK HW-50 S; eluent: phosphate buffer, 26.8 mmol l1; injection volume 2.5 ml).

high molecular substances between 40 and 50 ml of elution volume followed by a less large fraction. This material has a relatively strong UVA and does fluoresce. It is attractive to assign these fractions to refractory HSs of higher and lower molecular size. The relatively sharp chromatographic peak reflects small organic acids as reported by Brinkmann et al. (2003a, 2003b) and is followed by gradually eluting unidentified OC. There are some detector-specific differences in the fractions and in their relative intensities. In general, however, the main fractions look quite similar. As a consequence, the easy-tomeasure UVA is often used as surrogate parameter for OC determination (Her et al., 2002a, 2003). In the case of very low background values, fractions of a few tens of ng l1 OC can be quantified.

3.01.3.5 Applications The DOC methods and their combination with fractionation methods (e.g., SEC-UV/OC method) are well suited for the advanced characterization and semi-quantitative assessment of environmental processes such as nutrient cycling and pollutant transport as well as technical water-treatment processes. (see also Chapter 3.15 Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter).

3.01.3.5.1 Hydrosphere Typical ranges for TOC/DOC concentrations of aquatic systems are given in Table 2.

Sum Parameters: Potential and Limitations Table 2

9

DOC in different aquatic systems

Aquatic systems

DOC concentration (mg l 1) Average

References

Range

Ocean

0.5

0.3–2.0 in 0–300 m; 0.2–0.8 in 4300 m

Williams (1971); Duursma and Dawson (1981)

Freshwater Ice and snow Rivers Lakes

0.5 7 2.2

0.1–5.0 5–9 Oligotrophic 2–3 Eutrophic 9–16 10–50

Laird et al. (1988); Frimmel et al. (2002) Malcolm (1985); Sontheimer et al. (1986) Steinberg (2003); McKnight and Aiken (1998) Aitkenhead-Peterson et al. (2003); Bertilsson and Jones (2003) Thurman (1985); Frimmel and Abbt-Braun (1999)

19–31 B0.5 up to 10

Abbt-Braun (1992); Frimmel (1992) Dinar et al. (2006); Graber and Rudich (2006) Matthess et al. (1992); Wedepohl (1969)

Brown water Soil seepage water Rain Groundwater, CaCO3 aquifer

12 25 B1 0.7

3.01.3.5.2 Surface water

3.01.3.5.3 Water treatment

The SEC-UV/OC method finds a broad application in characterizing the OM of rivers. In Figure 6, typical chromatograms for (a) the river Rhine (Germany) and (b) the river Moskva (Russia) are shown. Although the DOC concentrations are significantly different, the main fractions of the OC for both rivers are quite similar, but the small-sized substances are more abundant in the case of the river Rhine. Both rivers show a small but significant OC fraction around the exclusion volume without any UVA. It could be shown that these substances are of high molecular carbohydrate type. For comparison, the chromatograms for water from a brown water lake (c) and for wastewater (d) are shown. The brown water is dominated by a single fraction and it is attractive to assign it to plant-derived matter of humic structure. In the case of the wastewater, there are obviously plenty of low-molecular-weight organic substances (acids) which get eliminated by biological treatment. As a result of biotreatment, a large organic fraction with low UVA is generated. Based on the assignment to matter with carbohydrate structures, this fraction around the exclusion volume of the SEC column can be used for a rough estimation of the allochthonous and autochthonous part of aquatic refractory OC. In large molecular size fractions, there was a predominance of polysaccharide material. N-Acetylated polysaccharides derived from microbial leftovers. Lignin and tannin derivatives were most abundant in the intermediate size fraction (Lankes et al., 2008). However, detailed interpretation has to rely on advanced spectroscopic information on molecular structure. For a critical evaluation of OC assignment, see, for example, AbbtBraun et al. (2004), Lankes et al. (2008), Reemtsma et al. (2008), and Kunenkov et al. (2009). Also, it has to be kept in mind that photochemical OC detection often does not work quantitatively, for example, up to 70% of certain OC compounds were not detected with the organic carbon detection system in systematic investigations (Lankes et al., 2009). Assuming that the majority of refractory OM components do absorb UV radiation, UVA values are a valuable supplement for OC detection.

The SEC-OC system can also be used to follow technical separation processes such as flocculation, membrane filtration, or adsorption. Figure 7 shows the example of bog lake (brown water) OC as it decreases after (a) addition of ferric chloride (flocculation with FeCl3) and (b) equilibration with increasing amounts of powdered activated carbon (PAC; adsorption). It is obvious that in flocculation most of the OM (87%) gets eliminated. Especially, the high-molecular-size substances get better eliminated than the small ones. Interesting to note is the high elimination yield of UV-absorbing matter and the relatively, poor elimination yield of AOX forming precursors. In the case of PAC adsorption, the rest OC which remains in solution is strongly dependent on the amount of PAC added as expected but it is mainly higher molecular matter which remains in solution. These findings can be explained by the limited availability of pores with larger size. All information that can be derived from advanced OC characterization does not only supply the basis for a better understanding of the mechanisms which rule the OC distribution, but it also opens the door for the development of technically relevant elimination processes and their optimization (see also Chapter 3.15 Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter, Chapter 3.16 Chemical Basis for Water Technology).

3.01.3.6 Surrogate Parameters There are a number of sum parameters for the determination of OM which have been developed independently or supplementary to the DOC/TOC methods. Most of them work simpler and therefore find a broad application as surrogate parameters for OC. They focus on a specific character of the present organic substances and can add valuable information for the assessment of water quality. Their specific information can be related to the mass unit of OC as a universal parameter and which can supply the basis for a sound assessment and comparison of different aquatic sources or for following a

10

Sum Parameters: Potential and Limitations 4 V0

OC UV

1.5 River water Rhine (Wörth) (OC) = 1.7 mg l−1

1.0

0.5

OC UV River water Moskva (Kolomna) (OC) = 8.5 mg l−1

3

2

1

0

0.0 20

40

60

2.5

20

80

Elution volume, Ve (ml)

(a)

30

40

50

60

70

80

Elution volume, Ve (ml)

(b)

Vp

V0

V0

OC UV Brown water HO23 (OC) = 27.7 mg l−1 dilution: 1:10

2.0

Vp

4

1.5

1.0

Relative OC-, UV (254)-signal

Relative OC-, UV (254)-signal

Vp

V0

Vp Relative OC-, UV (254)-signal

Relative OC-, UV (254)-signal

2.0

Wastewater (OC) (OC) = 24.9 mg l−1, dilution: 1:3

3

Wastewater effluent after biological treatment (OC) (OC) = 9.9 mg l−1

2

a: OC b: UV

1

a b

0

0.5 20 (c)

40

60

20

80

Elution volume, Ve (ml)

(d)

40

60

80

Elution volume, Ve (ml)

Figure 6 Size exclusion chromatogram detected by OC- and UV (l ¼ 254)-detection of river water ((a) river Rhine, (b) river Moskva), brown water ((c) Hohlohsee, HO23), and wastewater (dilution 1:3) and wastewater treatment plant effluent (d) (resin: TSK HW-50S; eluent: phosphate buffer, 26.8 mmol l1).

complete treatment pathway. Most common surrogate parameters and complementary parameters for OC are given in Table 3. They are discussed in the following sections in more detail.

3.01.4 Oxygen Demand Parameters 3.01.4.1 Introduction Despite the broad distribution, the stability of the nonradioactive elements leads to their quite constant total amounts on earth. However, their appearance in different compounds and phases called speciation makes them distinguishable according to the chemical bonds in which they are engaged (Pauling, 1960). The corresponding oxidation state of the atoms in their chemical appearance is a typical guide for their reactivity. Carbon is one of the elements which covers all the range of eight oxidation state levels from the lowest one of  IV in CH4 up to the highest of one of þ IV in CO2. The elemental form is represented by the graphite and diamond structure. CO2 is the common end product of all

biochemical degradation reactions of C-compounds and chemical combustions if sufficient O2 is available. According to the high importance of the load of organic substances in water, their oxidative transformation into CO2 has become the basis for the development of sum parameters for quality assessment (Wagner, 1973). Most of them are based on the quantification of the oxygen necessary for a more or less quantitative oxidation of all organic compounds. There are purely chemical methods and there are biochemical methods, using a mixed bacterial population.

3.01.4.2 Chemical Oxygen Demand 3.01.4.2.1 Background The aim of the COD is to obtain a complete oxidation of all organic compounds of an aqueous sample to CO2. This is best reached by wet oxidation with potassium dichromate (K2Cr2O7) in hot acid solution. Problems can arise from other water constituents, for example, inorganic ones, which also get oxidized under the reaction conditions. These disturbances can be tackled by elimination of the substances concerned or

Sum Parameters: Potential and Limitations

11

Elimination in % SAK254

AOX-FP

87

96

54

Relative OC-signal

Relative UV (254)-signal

DOC

20

40 60 80 Retention time, t (min)

100

After flocculation (FeCl3) Original

20

40

60

80

100

Retention time, t (min)

(a)

Brown water

Relative OC-signal

Remaining DOC + 50 mg l−1 PAC + 500 mg l−1 PAC + 1000 mg l−1 PAC

0 (b)

10

20

30

40

50

60

Retention time, t (min)

Figure 7 Size exclusion chromatogram obtained by OC detection (a, b) and UV detection ((a), inset) of diluted brown water and the remaining DOC after flocculation (a) and after adsorption on PAC (b) (TSK HW-50 S; eluent: phosphate buffer, 26.8 mmol l1; AOX-FP, AOX-formation potential; SAK, spectral absorption coefficient).

12

Sum Parameters: Potential and Limitations

by masking them such that they do not react. The oxidation reaction is given in Equation (4) and has a standard potential of E1 ¼ 1.36 V:

Cr2 O7 2 þ 6e þ 14H3 Oþ -2Cr3þ þ 21H2 O

ð4Þ

3.01.4.2.2 Analytical procedure The redox reactions with K2Cr2O7 work best under fairly concentrated H2SO4 conditions and at the elevated boiling temperature of the sample/acid mixture. The oxidative power of the defined amount of added dichromate is partly consumed by the known volume of the aqueous sample to be analyzed. The remaining gets quantified by reductive back titration with ferrous sulfate (Equation (5)). The color change from orange-yellow (Cr2 O7 2 ) to pale green (Cr3þ) is used as indicator for the equivalence point and for the final calculation of the result: Cr2 O7 2 þ 6Fe2þ þ 14Hþ -2Cr3þ þ 6Fe3þ þ 7H2 O

ð5Þ

The final result is calculated from the amount of consumed K2Cr2O7 converted into O2 equivalents according to

6:13  rðK2 Cr2 O7 Þ in mg l1  rðO2 Þ in mg l1

ð6Þ

The whole laboratory procedure is outlined in Figure 8. Table 3

Common surrogate parameters for OC in aquatic samples

Surrogate parameter

Acronym

Quantification as

Chemical oxygen demand Permanganate consumption Spectral UV and visible absorbance Biochemical oxygen demand Adsorbable organic halogens Various other sum parameters

COD PMC SUVA SVIA BOD AOX See Section 3.01.7 OC

O2 O2 A(254 nm) A(436 nm) O2 Cl

Organic carbon

CO2

HgSO4 1g Aqueous sample 50 ml

Ag2SO4 (50 mg) in H2SO4 (conc.) 5 ml

The method is broadly used in wastewater characterization. It works best in the concentration range 50 mg l1or(O2) o900 mg l1. Concentrations steps or dilution with organic free water are recommended if the COD concentrations are below 50 mg l1 or higher than 900 mg l1. In case ferrous ammonium sulfate (FAS) titrant and ferroin indicator are used, the color changes from blue-green to reddish brown. For COD determination, several standard methods have become available (DIN 38409-41, 1980; DIN 38409-43, 1981; DIN 38409-44, 1992; DIN ISO 15705, 2003; Standard Methods 5220 B, 5220 C, 5220 D, 2005); (see also Chapter 3.11 Standardized Methods for Water-Quality Assessment). Method variations. In addition to the described open reflux method, there is the possibility to use a so-called closed reflux method which uses borosilicate culture tube-like digestion vessels of 10 ml or more capacity and 10–25 mm diameter with polytetrafluoroethylene lined tightly fitting caps. Alternatively, borosilicate ampules can be used. The tubes or ampules filled with sample and chemicals are inserted in a block digestor at 150 1C for 120 min reflux. After cooling to room temperature, the digested solutions are titrated with ferroin indicator and FAS titrant. Alternatively, the change of C2 O7 2 to Cr3þ can be quantified spectrophotometrically at l ¼ 600 nm. The first method is mostly used for COD concentrations r(O2)o90 mg l1 whereas the l ¼ 600 nm absorption turned out to be better suited for higher concentrations. Experimental kits for these methods are commercially available. Calibration of all versions of the COD method is preferably done by potassium hydrogen phthalate (C8H5O4K) standard solutions with concentrations within the concentration range concerned. The whole procedure and equipment should be the same as for the determination of the samples.

3.01.4.2.3 Interferences The COD method based on the oxidative Cr2 O7 2 reaction at boiling conditions leads to parameter values with a coefficient of variation o8%. Use of especially cleaned glassware (e.g., H2SO4 rinsing), at least duplicate determinations and the subtraction of the blank COD of reagents and dilution water in the applied procedure can improve the data. The relevance

Calculated result

Distilled water

Flask with reflux condenser 120 min boiling

K2Cr2O7 (42 mmol l−1) 25 ml

Titrator

Ferroin indicator FAS (*) (0.25 mol l−1)

Figure 8 Laboratory procedure for the determination of the COD with dichromate (*FAS, ferrous ammonium sulfate).

Waste

Sum Parameters: Potential and Limitations

of the determination of background values was pointed out by Wagner (1973). Samples below the normal concentration range ask for even more care. In these cases, a higher volume of sample and diluted K2Cr2O7 standard solution (0.004 M) together with the appropriate amount of reagents are used and all are concentrated under boiling conditions to a volume of 150 ml. Titration is done with standardized 0.025 M FAS. Substances which are prone to poor or incomplete digestion in all described versions of the COD method are pyridine, its derivatives, and straight-chain aliphatic compounds. The latter ones can be more effectively oxidized in the presence of silver sulfate as catalyst. In the open reflux methods, volatile organic compounds can also get lost. The most common interferences are the halides, bromide and iodide, and especially chloride ions. They can form insoluble silver halides and by this inactivate the catalytic effect of Agþ. In addition, under the strong oxidative conditions of the K2Cr2O7 reaction, they can be transferred to the elements and beyond that to halo-oxoacids and their ions and by this false positive results are produced. Due to the complex reactions and the undefined mixture of resulting products, a correction of the results based on simple theoretical considerations is not possible. The addition of mercury sulfate (HgSO4) before boiling which leads to a close to complete complexation of the halides can eliminate the problem to a great extent. However, in the case of halide concentrations r(X)42 g l1, the method fails. Saline water samples can be pretreated by evaporation of the hydrogen-halide acids at reduced pressure. The hydrogen-halide acids are produced by addition of concentrated sulfuric acid to the sample under rigorous agitation:

2 X  þ 2 H þ þ SO4 2 2SO4 2 k þ 2HXm

ð7Þ

Nitrite (NO2  ) exerts about 1 mg O2 per mg NO2  – N. Due to the low NO2  concentrations in most waters, this can mostly be ignored.

3.01.4.2.4 Applications The COD is well suited for the characterization of fairly polluted waters. Municipal wastewater consumes O2 in the range from 300 to 1000 mg l1. After biological treatment, the COD (O2) drops to 20–1000 mg l1. Landfill leachate can reach up to r(O2) of 3000 mg l1. The COD of surface water normally ranges from 5 to 20 mg l1 .The COD as standardized method has found its way into wastewater legislation. In Germany, for example, 50 kg is the COD unit for payment of fees (1 unit B36 h) for the direct discharge of wastewater into the aquatic environment, and the threshold amount of discharge is 20 mg l1 of 250 kg yr1. The environmental hazards of Agþ, Cr(VI), and Hg2þ used in the COD determination ask for methods which work with smaller volumes or for alternative clean methods. From this point of view, TOC (DOC) is a promising parameter for replacing COD. COD for quantitative determination of oxidizable OC does not necessarily lead to the equivalent result as TOC (DOC) measurements, even though in both cases the end product of the reactions is CO2. The simple approximation that one mass unit of COD(O2) equals one mass unit of TOC or DOC is not precise enough in most cases due to the different oxidation

13

states of the averaged carbon (DOC, TOC) in the organic load which consequently leads to different consumptions of oxidant and hence COD values. (see also Chapter 3.16 Chemical Basis for Water Technology). The O/C atomic ratios for moderately polluted rivers (e.g., Rhine, Main, Danube, and Elbe) are around 2 ranging from 1.3 to 2.7. For wastewater, the ratios are similar or a bit higher (Zanke and Ho¨pner, 1982).

3.01.4.3 PMC and Permanganate Index (IMn) 3.01.4.3.1 Background Potassium permanganate (KMnO4) is a fairly strong oxidizing agent. Therefore, it has been used as analytical tool to characterize dissolved organic water constituents. The oxidation method has been established as fairly simple wet chemical procedure since the early days of water-quality assessment. The method should be used as operationally defined determination of the oxidizability of relatively clean water samples. The results mostly do not allow a clear correlation with the OC content of the samples. The PMC is defined to be the amount of permanganate that reacts with the sample under defined conditions. The oxidative function of permanganate in acid medium (sulfuric acid) is given in Equation (8) and shows a standard potential of E0 ¼1.52 V:

MnO4  þ 5e þ 8H3 Oþ -Mn2þ þ 12H2 O

ð8Þ

3.01.4.3.2 Analytical procedure The redox reaction partners are the oxidizable organic substances which are mostly the aim of quantification. However, inorganic water constituents (e.g., Fe2þ, Mn2þ, Cl, or NH4 þ ) which can be oxidized have to be considered. According to the protocol (DIN EN ISO 8467, 1995; (see also Chapter 3.11 Standardized Methods for Water-Quality Assessment)), the sample (defined volume) is mixed with sulfuric acid and the well-defined potassium permanganate solution, and the mix is heated for 10 min. Then a defined amount of sodium oxalate (Na2C2O4) is added in excess for reduction of the unreacted MnO4  , and the remaining oxalate is quantified (Figure 9). From all this, the amount of MnO4  consumed by the sample can be calculated, and from that the resulting oxygen demand is deduced:

rðKMnO4 Þ in mg l1  3:95 rðO2 Þ in mg l1

ð9Þ

The IMn is calculated according to

IMn ¼ f ðV1  V0 Þ=V2 M ¼ 16ðV1  V0 Þ=ðV2 Þ

ð10Þ

where V1 is the volume (in ml) of consumed permanganate solution of the sample; V0 the volume (in ml) of consumed permanganate standard solution of the blank solution; V2 the volume (in ml) of the consumed permanganate standard solution of the blank solution after addition of oxalic acid; and f (16 mg mmol1) the equivalence coefficient for the conversion into oxygen. The method is applicable to samples with r(O2)41 mg l1.

14

Sum Parameters: Potential and Limitations

KMnO4 (2 mmol l−1)

H2SO4 (2 mol l−1) 5 ml Aqueous sample 25 ml

Stirring heating in boiling water bath

Hot titration Waste 30 min pale rose

10 min reaction

Na2C2O4 (5 mmol l−1) 5 ml

KMnO4 (2 mmol l−1) 5 ml

Calculated result

Figure 9 Laboratory procedure for the determination of the permanganate consumption.

3.01.4.3.3 Interferences As far as PMC is used as surrogate parameter for OC, pitfalls resulting from the presence of oxidizable inorganic water constituents have to be considered. Halide concentrations, for example, r(Cl) 4300 mg l1, can cause significant errors leading to higher values due to the complex redox reactions of the halogen species. Fe2þ can also lead to positive false results which, however, can be corrected according to

rðFe 2þ Þ ¼ 1 mg l 1  rðKMnO4 Þ ¼ 0:57 mg l1

ð11Þ

In addition, there is a risk that aqueous KMnO4 solutions can decompose, especially at elevated temperatures. Recalibration and determination of blanks are, therefore, crucial. Another important aspect is the limited oxidation potential of KMnO4 solutions which results in only partial oxidation of OM, mostly of up to 40%.

3.01.4.3.4 Applications The IMn is mainly used for the assessment of drinking water, surface water, groundwater, and bottled water. Wastewater and other polluted waters need dilution before determination. The range for application is 0.5 mg l1or(O2)o10 mg l1. It is also suited for waters with r(Cl)o300 mg l1. Many pitfalls, the often poor yields in the oxidation of OM, and the often lacking reproducibility of the results have brought the parameter to a questionable reputation, and therefore it practically plays no major role in modern water assessment. Due to the large amount of available data of PMC from the old days, however, there might be some interest in comparison to longterm trends in water quality reflected in the oxidizability. The range of application reaches from around 1 mg l1 (as O2 equivalent) to several hundreds of mg l1. The EU directive for drinking water states 5 mg l1 as maximum parameter for oxidizability and recommends 2 mg l1. (see also Chapter 3.16 Chemical Basis for Water Technology).

3.01.4.4 Biochemical Oxygen Demand 3.01.4.4.1 Background Sustainable water management includes treatment of used water. Technical wastewater-treatment systems have been

developed for this purpose. From an economical and ecological point of view, it is most attractive to use microbiological methods (Wagner, 1979). Their application can be optimized with the help of parameters suited for the assessment of wastewater and for the control of the performance of the treatment units and the secondary effluents. Closely connected to the task of quantifying biodegradability, there is the aspect of the time frame, for example, the question: how long does it take to degrade a specific amount of OM? The time window may reach from several hours and a few days (poorly biodegradable  refractory) to even several years (practically nonbiodegradable). This poor precision asks for a pragmatically defined approach to reach meaningful results. Even though the ordinary oxygen from the air has only a standard potential of þ 0.82 V and hence is a relatively weak oxdidant (Equation (12)) at ambient temperature, with the help of biocatalysis the BOD is turned out to be a powerful parameter to serve the needs of a valuable assessment. Several standardized laboratory procedures have been developed on the basis of the O2 consumption of OM and inorganic compounds such as Fe2þ, sulfides, or reduced nitrogen compounds during a specified period of incubation with a mixed microbial population. Mostly, the procedure is focused on the organic load (carbonaceous BOD, OM):

O 2 + 4H 3 O + + 4e− OM + O 2

Bacteria

pH = 7

6H 2 O

CO 2 + H2 O + biomass

ð12Þ ð13Þ

3.01.4.4.2 Analytical procedure Standard methods are available for the determination of BOD (DIN EN 1899-1, 1998; ISO 5815:1989; Standard Methods 5210 B, 5210 C, 5210 D, 2005; (see also Chapter 3.11 Standardized Methods for Water-Quality Assessment)). The BOD is mostly determined for an incubation period of 5 days (BOD5), but other incubation periods (1–50 days) can also be applied. The principle of the procedure is given in Figure 10. The air-saturated sample (if necessary seeded and/or diluted) is filled to overflow in a then airtight corked bottle of specified volume. Dissolved oxygen (DO) is measured immediately and after incubation of 5 days at 2073 1C. BOD5 is

Sum Parameters: Potential and Limitations

15

Defined dilution water

Glass bottle (e.g., 300 ml)

Aqueous sample (air saturated)

At start time After n days

O2 determination

Waste

7.0 < pH < 7.2 20 ± 3 °C

Seed suspension if needed

Nitrification inhibitor if needed

Figure 10 Experimental procedure for the determination of the BOD of aqueous samples.

calculated as concentration difference of the initial DO and the end DO. In case the oxygen consumption should exclude the demand of reduced nitrogen compounds (nitrogenous demand; e.g., ammonia and organic nitrogen), a nitrification inhibitor (e.g., 2-chloro-6-(trichloromethyl)pyrodine, TCMP, or allylthiourea (ATU)) has to be added. The DO determination is done either iodometrically (azide modification) or electrochemically by a membrane O2-electrode.

3.01.4.4.3 Interferences At the end, a proper BOD determination needs a residual concentration of oxygen of at least 2 mg l1. Water samples with high loads of OM can be measured after dilution. In case the water to be determined has a poor bacterial population, seeding is necessary. For that 0.5 ml sedimented municipal wastewater, B2 ml of biodegraded wastewater, or 5–10 ml river water are suited. The BOD of these additions has to be considered as blank. In case plankton is present, elevated BOD values have to be expected. The same applies for other O2-consuming water constituents such as Fe2þ, SO3 2 , and/or H2S/HS. A major problem for the BOD determination is the presence of poisonous or inhibiting substances (CN, CrO4 2 , Cu2þ, Hg(0, I, II), etc.) which might be overcome by dilution. In wastewater, nitrification may also lead to interferences. In order to avoid this, the addition of N-ATU to concentrations of 2–5 mg l1 is recommended. However, in this case O2 determination using the Winkler method becomes questionable. Many of the possible pitfalls can be diagnosed by online determination of O2 over the whole observation period (Figure 11).

3.01.4.4.4 Applications The method is well suited for the characterization of samples from rivers, lakes, estuaries, and wastewaters, and for their treatment efficiency in plants and effluents. BOD5 values normally range from 5 mg l1or(O2)o 250 mg l1. Other variations with shorter or longer incubation times than 5 days exist to measure rates of oxygen uptake. In special cases, incubation times of up to 90 days are used to determine the socalled ultimate BOD. Continuous oxygen monitoring (e.g., by

O2 electrodes) allow the characterization of different phases of biodegradation over time. The domain of BOD determinations consists of the wastewater and samples from its biological treatment. Typical municipal wastewater BOD5 lies around 60 g per capita equivalent. An average daily water use of 150–200 l per capita results in BOD5 concentrations of o25 and 300–350 mg l1 in treated and untreated wastewater, respectively. With respect to the changing biodegradability of wastewater constituents, it is interesting to relate the chemical oxidizability to the biochemical one, that is, to use the COD/BOD ratio for assessment (Leithe, 1971) (Table 4).

3.01.4.4.5 Related parameters (AOC) Another approach to quantify the biodegradability of OM uses the growth effect of a mixed population. After sterile filtration through a 0.2-mm nucleopore membrane, 275 ml of the water sample together with 25 ml of a sterile filtered merely inorganic nutrient solution is filled into a cuvette. The mixed population of bacteria retained by the 0.2-mm membrane filter is washed by NaCl solution and added to the mixed solution in the cuvette to reach a turbidity of 0.03 ppm SiO2 equivalents. The turbidity is measured as 121 forward scattering of a visual light beam in 30 min intervals for 60 h. The function of the relative turbidity over time gives the growth curve. Based on the assumption that the turbidity reflects the growth of the microbial population which is caused by the nutritious effect of the sample’s OM, it is attractive to relate the change of turbidity to the amount of assimilable carbon. The function of the relative turbidity over time gives the socalled growth curve (Figure 12). For the evaluation of the growth curve, the growth rate (GR; Equation (14)) and the growth factor (GF; Equation (15)) can be determined (Hambsch et al., 1992):

GR ¼

d lnðturbÞ at t ¼ tw dt

ð14Þ

turbðmaxÞ turbðstartÞ

ð15Þ

GF ¼

16

Sum Parameters: Potential and Limitations

O2-consumption (mg l−1)

35 30

d

25

c b

20 15 10 5 0

a

−5 0

2

4

6

8

10

12

14

16

Time, t (days) Figure 11 Typical O2-concentration curves for BOD determination (a: no biological degradation; b: biological degradation with lag phase; c, d: biological degradation without lag phase).

Table 4 COD/BOD ratios for the assessment of the removal efficiency of organic compounds by biochemical degradation COD/BOD

Assessment

o1.7

Organic substances show high biodegradation and mineralization Chemical degradation is insufficient due to – slow adaption of the bacteria – high amount of persistent compounds – inhibition of the reaction because of toxic substances No or practically no chemical degradation due to – persistent substances – inhibition of the reaction by highly toxic substances

1.7–10

410

where GR is the slope(s) at the inflection point of the curve for the exponential growth phase (tw). The steeper s the better assimilable the organic substances, for example, the GR gives information on the quality of the assimilable carbon. The ratio of the maximal turbidity and the initial turbidity (GF) is related to the quantity of assimilable carbon. From the shape of the curves inhibition and retardation, for example, by toxic water constituents, of the assimilation, for example, in the presence of recalcitrant fractions, can also be deduced. Figure 13 shows the growth curves for a lake water sample without and after treatment with different amounts of hydrogen peroxide (H2O2). It is obvious that in the original water sample, assimilation starts at the earliest. After about 15 h there is still a significant but slowly increasing turbidity possibly due to refractory OM. Oxidation with H2O2 (initial concentration r0(H2O2) ¼ 0.2 mg l1) leads to an increased lag phase in the assimilation of about 10 h followed by a steep exponential growth phase and finally, after 25 h, to a quite constant maximum turbidity; this reflects the higher amount

of assimilable carbon after chemical oxidation compared to the matter in the original sample. Tenfold initial H2O2 concentration leads to a further increased lag phase possibly due to the toxic effect of H2O2. The exponential growth phase shows a similar GR as the original lake water, and after 30 h a gradual increase of turbidity occurs up to 50 h which was the end of turbidity monitoring. From the gradual increase, an ongoing oxidative degradation of the organic substances to better assimilable ones can be deduced.

3.01.4.5 Interdependences There is a large amount of data on the load of OM in aquatic samples (Table 5). Despite the prosperous situation of available data, there are not too many reliable correlations between the different parameters. It might be relatively simple for defined model compounds but the complex mixture of realistic aquatic systems is difficult to assess, even though there are some reliable data for municipal wastewater (Table 6), and for drinking water and surface water (Leithe, 1971).

3.01.5 UVA and Visible Range Absorbance 3.01.5.1 Background Aquatic systems with high concentrations of OM, for example, bog lakes and organically rich aquifers, show a typical yellow to brown color. The absorption spectra for NOM samples in the UV (UVA) and visible range (VIA) are poorly resolved with a characteristic strong increase of the absorbance to lower wavelengths. This is typical for complex mixtures of substances with significant amounts of unsaturated bonds, lone pair electrons, and/or aromatic structures (Langhals et al., 2000). In addition, strong intermolecular interactions can add to

Sum Parameters: Potential and Limitations

17

Sample preparation Water sample

Sterile filtration

Cuvette

Inoculum

0.2 µm Nucleopore

275 ml of the sterile filtered sample

Mixed population of bacteria, washed from the sterile filters by NaCl solution

25 ml of a sterile filtered nutrient salt solution Registration of the growth curve Cuvette

Turbidity

Addition of inoculum until turbidity is 0.03 ppm SiO2

Measurement

Additional measures

(12° forward scattering) 60 h, every 30 min

Dissolved organic carbon (DOC) Total cell number (TCN) at the start and at the end

Evaluation of the growth curve Growth rate

Rel. turb.

Growth factor

(in the exponential phase tw) tW

GR = d ln(turb) dt

Time (h)

t = tw

GF =

turb(max) turb(start)

Figure 12 Procedure for the turbidimetric quantification of the AOC.

Turbity (12° forward scattering)

1.2 1 0.8 0.6 Original sample

0.4

Addition of (H2O2) = 0.2 mg l−1 0.2

Addition of (H2O2) = 2 mg l−1

0 0

10

20

30

40

50

Time, t (h) Figure 13 Typical growth curves for the organic carbon in lake water without and after addition of hydrogenperoxide (H2O2).

UV–Vis absorbance. Figure 14 shows two examples of typical UVA and VIA spectra of aquatic samples. This is the basis for using UV–Vis range information as surrogate parameter for a rough estimation of the dissolved OC concentration. It is quite common to use l ¼ 254 nm of the UV range and l ¼ 436 nm of the visible range for quantification. Around l ¼ 254 nm often a weak shoulder in the spectra is obvious which is assigned to chromophores with conjugated CQC and CQO double bonds.

According to Lambert–Beer’s law spectral absorbance is proportional to the concentration of the analyte:

AðlÞ ¼ kðlÞcd

ð16Þ

AðlÞ ¼ SAKðlÞd

ð17Þ

where A(l) is the absorbance at wavelength l; k(l) the molar absorption coefficient, in l (mol m)1 or l (g m)1; c the

18

Sum Parameters: Potential and Limitations

Table 5

Typical ranges for the content of organic matter in aquatic systems as reflected in sum parameters

Type of water

DOC, r(C) (mg l1)

Drinking water Groundwater Surface water Mesotrophic Eutrophic Municipal Wastewater Treated Landfill leachate

o2 0.5–4

COD, r(O2) (mg l1)

KMnO4, r(O2) (mg l1)

BOD5, r(O2) (mg l1)

o5 3–8 5–20

2–5 4–10

20–35 100–150

200 o25 4500

300–1000 20–100 o3000

AOC, r(Ac-C)a (mg l1)

AOX, r(Cl) (mg l1)

9–20 o80

30 50–80

6

250 20 200–13 000

o500 4500

a

Acetate-C calibrated.

Table 6 Transfer factors (A:B) for the values of sum parameters in wastewater assessment (Koppe and Stozek, 1990) Parameter B

KMnO4 COD BOD5 TOC

Parameter (A) KMnO4

COD

BOD5

TOC

1.0 0.6 1.4 2.0

1.6 1.0 2.2 3.1

0.7 0.5 1.0 1.5

0.5 0.3 0.7 1.0

concentration in mol l1 or g l1; d the path length, for example, of cuvette in m; and SAK(l) the spectral absorption coefficient in m1.

3.01.5.2 Analytical Procedure According to standard methods color (VIA) is determined by visual comparison of the sample with known concentrations of colored solutions (Standard Methods 2120 B, 2005) and as spectrophotometric method using l ¼ 436 nm (Standard Methods 2120 C, 2005; DIN EN ISO 7887 C1, 1994; (see also Chapter 3.11 Standardized Methods for Water-Quality Assessment). The measurement of the color is either performed in Nessler tubes by looking vertically downward through the tubes (Standard Methods 2120 B, 2005) or by spectrophotometric determination at a wavelength between l ¼ 450 and 465 nm (Standard Methods 2120 C, 2005). The color unit (CU) of 500 is related to a mixture of 1.246 g potassium chloroplatinate and 1 g cobaltous chloride in 100 ml HCl conc. and diluted to 1000 ml. As a consequence, the unit of color equals 1 mg l1 platinum (in the form of chloroplatinate ions). Calibrated glass color disks are also used for comparison. The platinum–cobalt method is applicable to natural water, drinking water, and wastewater. A special advantage of the determination of samples in Nessler tubes is the relatively long optical pathway in the tubes which leads according to Lambert–Beer’s law to low limits of determination. Besides this method, the spectral absorption at l ¼ 436 nm can be used to determine the color (DIN EN ISO 7887 C1, 1994). Here, results are given as absorption coefficient in m1 (Standard Methods 5910 B, 2005; DIN 38404-3, 2005).

Based on the gradual decrease of absorbance with increasing wavelength, the value of l ¼ 254 nm is often used as fairly sensitive characteristic information on the content of UV-absorbing organic constituents. The results are given as absorption coefficients in m1. In addition to the described quite simple methods, more sophisticated methods for the determination of color have been standardized as well. There are the multi-wavelength method (Standard Methods 2120 D, 2005) and the tristimulus spectrophotometic method (Standard Methods 2120 E, 2005). Samples have to be filtered through 0.45-mm pore-size membranes to remove turbidity as the apparent color can be higher than the true color of the solution itself.

3.01.5.3 Interferences Interferences may arise from inorganic constituents, for example, ferrous iron, nitrate, nitrite, bromide, and from certain oxidants and reducing agents (e.g., ozone, chlorate, chlorite, and thiosulfate). An absorption scan between l ¼ 200 and 400 nm can be used to determine the presence of interferences. In addition, turbidity adds to the molecular spectrometric absorption in a complex way by absorption and light scattering. Reproducible results are obtained after a separation step is clear (o0.45 mm) solutions.

3.01.5.4 Applications Typical values for color determined at l ¼ 436 nm (SAK436) of different water samples are shown in Table 7. UV absorption is often used to monitor industrial wastewater effluents, and to evaluate the DOC removal during water-treatment processes. (see also Chapter 3.16 Chemical Basis for Water Technology, Chapter 3.15 Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter, Chapter 3.10 Online Monitoring Sensors). The spectral absorption exhibits a dependence on pH values with decreasing specific absorbance as solution pH decreases (Langhals et al., 2000). This reflects the different acid– base forms of the chromophores within the molecules or as suggested by Chen et al. (1977), an increase in molecular size due to macromolecular associations. The color caused by NOM also changes with the chemical characteristics of the

Sum Parameters: Potential and Limitations

19

3.5 Brown water (HO20) Wastewater effluent (Alb5)

3.0

Extinction

2.5 2.0 1.5

λ (254 nm)

1.0 λ (436 nm)

0.5 0.0 200

300

400

500

600

Wavelength,  (nm) Figure 14 UV and visible range spectra of a brown water lake sample (Hohlohsee, HO20) and a sample from the effluent of a biological wastewater treatment plant.

Table 7

SAK436 (VIA 436) values of different types of water

Type of water

SAK436 in (m1)

Tap water River water (moderately polluted) Brown water Wastewater Visual verification limit

o0.5 (recommended) 41 2–5 45 2

water. For example, NOM–metal complexes can be formed in the presence of Ca and Fe ions, and this affects the type and the intensity of the color. The relatively simple method of the determination at l ¼ 254 nm has led to its broad application as surrogate parameter for the more complicated instrumental determination of DOC. The correlation of the two parameters depends on the origin of the water sample but is quite constant for the individual aquatic systems which is demonstrated by the experimental data for the river Rhine (Figure 15). The values allow a rough characterization of the OM according to its genesis. The high sensitivity and consequently the small sample volume demand of the method is a significant advantage. This has led to a broad application of the method for the characterization of original samples from soils or water without major pretreatment and concentration procedures. Beyond that specific SAKs (SSAKs), for example, SAK values for r(DOC) ¼ 1 mg l1, can be used for a more detailed characterization and sound comparison of aquatic OM (Table 8). Many publications have become available on the UV and visible spectroscopic characterization, including luminescence of OM from natural origin (NOM) (e.g., MacCarthy and Rice, 1985; Cabaniss and Shuman, 1987; Bloom and Leenheer, 1989; Senesi et al., 1989; Hautala et al., 2000). They all prove

the comparability power and relevance of the SSAK in waterquality assessment.

3.01.6 Organically Bound Halogens Adsorbable on Activated Carbon (AOX) 3.01.6.1 Background Halogen-containing organic compounds are widely used in industrialized countries. Due to the resulting high amounts of production and the broad application as solvent and in many products, these anthropogenic halo-compounds have found their way into aquatic systems. Many of the compounds are of toxicological relevance for men and environment. Therefore, distribution and fate of the halo-compounds in nature and technical systems such as wastewater treatment plants are of major interest. A well-suited assessment parameter is a sum parameter reflecting all organically bound halogens which adsorb on activated carbon (AOX), where X ¼ Cl, Br, and I (Ku¨hn and Sontheimer, 1973). Similar parameters are dissolved organic halogens (DOX) or total organic halogens (TOX) which are often used synonymously to AOX.

3.01.6.2 Analytical Procedure The AOX procedure is based on the equilibration of PAC with the sample solution in batch mode (Ku¨hn and Sontheimer, 1973). Unwanted adsorption of common inorganic halides on PAC is reversed by competitive displacement by nitrate ions. After filtration of the loaded PAC, it is introduced into a furnace that pyrolyzes PAC and OC to CO2 and the bound halogens to hydrogen halides (HX). A carrier gas stream (mostly O2) transports the HX to a micro-coulometric titration cell. There the halides are quantified by measuring the current produced by silver-ion precipitation of the halides. In the cell, a constant silver-ion concentration is maintained from a solid silver electrode. The current for that is

20

Sum Parameters: Potential and Limitations 9.5

SAK (254 nm) (m−1)

8.5

Ludwigshafen km 421.4 l Mainz km 500.6 l

7.5

Koblenz km 588.3 l 6.5

Köln km 684.5 l

5.5

Düsseldorf km 732.1 r Wittlaer km 757.9 r

4.5

Basel-Birsfleden

3.5

Karlsruhe

2.5 1.5

2

2.5

3

Dissolved organic carbon (DOC) (g m−3) Figure 15 UVA and DOC concentrations of the river Rhine water (monthly composite samples from the river Rhine at different sampling places (stream km); l ¼ left, r ¼ right side; ARW, 2004; AWBR, 2006).

Table 8

Spectroscopic characteristics of selected original water and fulvic acid (FA) samples (Frimmel et al., 2002)

Water sample

Brown water lake Brunnsee Hohlohsee Brown water lake, fulvic acid HO14 FA (pH ¼ 2) HO14 FA (pH ¼ 7) HO14 FA (pH ¼ 11) River water (Rhine) Lake water (Lake Constance) Groundwater (high load of humic substances, Fuhrberg) Secondary effluent, Neureut Soil seepage water

proportional to the number of moles of halogens introduced by the carrier gas. Figure 16 gives an outline of the AOXdetermination procedure. The standardized version of the parameter method has also made its way into wastewater legislation in Germany (e.g., DIN EN ISO 9562, 2005; Standard Methods 5320 B, 2005). The determination of AOX in waters with high salt content can be preferably done with the help of solid-phase extraction (SPE), for example, with styrene–divinylbenzene copolymer. After rinsing the polymer phase with sodium nitrate solution, the AOX compounds are eluted with methanol, the methanol extract is diluted with water, and the ordinary AOX procedure using activated carbon is applied to the solution (DIN 3840922, 2001). Calibration can be done, for example, by 2,4,6-trichlorophenol. Forensic analysis may aim for a differentiation of the halogens. The appearance of iodinated X-ray contrast media in aquatic systems (Putschew et al., 2000) or the formation of organically bound bromine in the course of oxidative water treatment (Tercero and Frimmel, 2008) are prominent examples where halogen-specific AOX can help to determine the distribution and to investigate the fate of the

SAK254/DOC (l (mg m)1)

SAK436/DOC (l (mg m)1)

SAK254 /SAK436

4.50 5.09

0.42 0.30

10.6 13.2

4.80 4.98 5.20 2.21 2.92 2.92 1.44 3.13

0.30 0.38 0.58 0.12 0.09 0.15 0.10 0.18

15.9 13.1 8.8 17.6 31.2 19.7 13.7 16.7

specific group of compounds. The analytical approach traps the AOX combustion gases containing the different HX molecules in an alkaline solution which then is analyzed for Cl, Br, and I by ion chromatography (Oleksy-Frenzel et al., 1995, 2000) or by atomic emission spectroscopy (OES-ICP) (Abbt-Braun et al., 2006).

3.01.6.3 Applications The standard method is applicable for samples with AOX concentrations of r(Cl)45 mg l1. Special care has to be taken for Cl-free PAC and highly pure chemicals. Blank determinations are mandatory. The Cl content of virgin PAC should not exceed r(Cl)o20 mg g1. According to the relatively low detection limit, the method can be applied to a broad range of aquatic samples, including drinking water, process water, wastewater, and water from different stages of water treatment and from the entire aquatic environment. Due to its broad applicability and ecological relevance, the parameter has found its way into water legislation and into many assessment protocols for wastewater, treatment plant effluents, and rivers and lakes. In addition to the classical AOX determination

Sum Parameters: Potential and Limitations

Aqueous sample (100 ml)

21

Discard filtrate

Adsorption batch (50 mg PAC)

Equilibration

Filtration (0.4 μm polycarbonate filter)

Washing (nitrate solution)

h

ICP-AES ICP-MS Cl−, Br−, I− determination

Microcoulometric titration of Σ(Cl−, Br−, J−)

Ionchromatographic Cl−, Br−, I− determination or

PAC combustion (in O2 gas 950 °C)

HXabsorption (75% acidic acid; 20 mmol l−1 (NH4)2CO3)

Alternatively Figure 16 Experimental procedure for the determination of adsorbable organic halogen and the amounts of the different halogen species.

which includes most of the organic halogen compounds, there is the possibility to quantify specific fractions. (see also Chapter 3.16 Chemical Basis for Water Technology, Chapter 3.15 Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter). Recently on several occasions, elevated iodinated AOX concentrations were found in aquatic systems. The family of poorly biodegradable iodinated X-ray contrast media which has been broadly applied in medical diagnosis is the reason for that (Ternes and Hirsch, 2000; Putschew et al., 2000; AbbtBraun et al., 2006; Putschew and Jekel, 2006; Wolf et al., 2006). Data from sand column experiments (Table 9) run with contaminated wastewater reveal a partial elimination and/or degradation of the iodinated contrast media.

3.01.6.4 Related Parameters There is another approach for an integral determination and characterization of organic compounds containing halogens. The method is based on repeated liquid/liquid extraction of the aqueous sample with pentane, hexane, or heptane at a volume ratio of 20:1 (EOX, extractable organically bound halogens). The extract is dried and incinerated in a hydrogen oxygen flame. The mineralized products in the condensate are quantified with the help of volumetric precipitation analysis based on silver nitrate. In comparison to the AOX determination which can be found in wastewater legislation, the quantification of EOX is of lower importance possibly due to less reliable results (Sontheimer and Schnitzler, 1982; DIN 38409-8, 1984).

3.01.7 Additional Sum Parameters 3.01.7.1 Background There are several other sum parameters for water-quality assessment. They focus on either inorganic species such as pH,

Table 9

Iodine balance of column experiments (numbers in mg1)

I Species

Total iodine (ICP-MS) Iodine of X-ray compounds (HPLC) Inorganic iodide (IC-ICP-MS) AOI (iodinated organic metabolites) AOI-iodine of X-ray compounds

Influent aerobic

Effluent column 1 unsaturated

Effluent column 2 saturated

2.3470.04

2.4870.23

2.4370.14

2.3570.01

0.5570.06

1.5370.30

0.02070.003 1.6470.22

0.03770.012

2.2570.10

1.3870.09

2.2270.04

–0.1170.16

0.8270.13

0.6970.13

Glass columns (l ¼ 1.2 m, d ¼ 20 cm) filled with medium grain quartz sand. Feed: wastewater from the inflow of a municipal sewage treatment plant spiked with X-ray contrast media (Neureut/Karlsruhe, r(DOC) ¼ 30  90 mg l1). Conditions: waterunsaturated and water-saturated conditions (V E 2 l d1). Despite the complex and highly dynamic wastewater matrix, the example demonstrates clearly the valuable information on interdependences of the different organic and inorganic iodine species.

electrical conductivity, radioactivity or metals, (see also Chapter 3.03 Sources, Risks, and Mitigation of Radioactivity in Water, Chapter 3.02 Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species) and they quantify organically bound hetero-elements or represent special types of organic compounds. Some of them reflect specific structural features, others are based on defined operations used for isolation or identification. A selection is given in Table 10. (see also Chapter 3.11 Standardized Methods for Water-Quality Assessment).

22

Sum Parameters: Potential and Limitations

Table 10

Selection of additional sum parameters (see also Chapter 3.11 Standardized Methods for Water-Quality Assessment)

Targets (concentration range)

Method principle

Reference

Lignin and tannin Volatile organic acids (up to C6)

Folin phenol reagent, blue colour (a) Adsorption on silicic acid, elution with chloroform-butanol, titration with NaOH (b) Distillation, titration with NaOH (c) GC-FID 4-aminoantipyrine (a) After distillation (b) Chloroform extraction (c) Flow analysis

Box (1983) Westerhold (1963)

Phenols (1–250 mg l1)

Surfactants (0.2–2 mg l1 non-ionic) (40.1 mg l1 non-ionic) (40.01–0.2 mg l1 anionic) Aquatic humic substances Hydrocarbons, oil, and grease (o10 mg l1)

Nano- and microparticles Metal complexation capacity

Cholinesterase

Sublation (N2, ethylacetate) Methylene blue (MBAS) (in chloroform) Dragendorff reagent Cobalt thiocyanate (CTAS) (in methylene chloride) Adsorption/desorption on/from XAD (a) Gravimetric n-hexane/MTBE (extraction) (b) Extraction trichlorotrifluoroethane IR (c) Soxhlet extraction (for sediments) (d) Gravimetric (solid phase extraction) Flow-field-flow-fractionation (F4) (a) Polarographic Cu2þ titration (b) Metal selective electrode titration (c) Fluorescence Photometry

To differentiate between sum parameters and group parameters as it was suggested in the past (e.g., Sontheimer et al., 1986) seems to be idle since sum parameters in the modern sense do all allow an integrative view on well-defined aquatic water constituents without single compound (or even particle) identification. Of course, the precise definition of the individual sum parameter is most important to avoid misinterpretation of the results. As valuable tools for assessment and orientation, they are challenged by naturally occurring matter and emerging pollutants as well.

3.01.7.2 Examples for Emerging Parameters Four examples are given to briefly demonstrate the power of actual sum parameters’ development and their application in the assessment of aquatic samples. The given approaches are selected as typical examples of modern needs for information on the aquatic environment and its sustainable use and management. The first example focuses on the vital reservoir of refractory OM called humic substances (HSs), the second one addresses the colloids and the young world of nanoparticles (NPs) (see also Chapter 3.05 Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems), the third one shows a typical impact of the availability of advanced instrumentation on the gross characterization of aquatic samples, and the fourth one helps to answer fundamental questions such as: What does it all matter? Is the determined amount dangerous? Which kind of bioeffect has to be watched?

3.01.7.2.1 Humic substances Aquatic humic substances (AHSs) consist of refractory OM ubiquitous in aquatic systems and can be isolated according to

Heukelekian and Kaplovsky (1949) Pavan et al. (2000) DIN 38409 (1984) Neufeld and Paladino (1985) Standard Methods 6420 B (2005) DIN EN ISO 14402 (1999) Schwuger (1996) Kunkel et al. (1977) Osburn (1986); DIN 38409 (1980) ISO 7875-2 (1984) Tabak and Bunch (1981) Frimmel et al. (2002) US-EPA (1998) Gruenfeld (1973) Ullmann and Sanderson (1959) US-EPA (1999) Von der Kammer and Fo¨rstner (1998); Delay et al. (2010) Lund et al. (1990) Tuschall and Brezonik (1983) Ryan and Weber (1982) Herzsprung et al. (1989)

standardized procedures recommended by the International Humic Substances Society (IHSS) (Thurman and Malcolm, 1981; Malcolm and MacCarthy, 1992; Leenheer and Croue´, 2003; IHSS, 2010). The principle of the methods is given in Figure 17 and allows to differentiate between fulvic acids (FAs) and humic acids (HAs) according to their pH-dependent solubility and adsorption/desorption on polymer resins. The procedure has opened the door for a better understanding of the occurrence and structure of AHS and their contribution to the DOM in the hydrosphere (Frimmel et al., 2002; Senesi et al., 2009). They are left over from organisms and the reservoir for new life. Their genesis and fate can now be compared for different regions and climatic zones. Based on OC concentrations, reliable balances can be made even on global scale. In addition, the multimethod approach with analytical instrumentation helps to elucidate the aquatic function and fate of operationally defined fractions.

3.01.7.2.2 NPs and colloids Colloids and particles (NPs) of the micro-scale have been recognized in the aquatic environment for quite some time (e.g., Frimmel et al., 2007). Recently, the broad application of engineered NPs (ENPs) in daily life has led to the concern of their role in the water cycle (Frimmel and Delay, 2010; (see also Chapter 3.05 Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems)). For the assessment of the resulting heterogeneous systems, powerful methods are needed. Figure 18 shows an experimental setup which is suited to gain information, for example, on particle size distribution and the carrier function of particulate matter in aquatic samples (von der Kammer and Fo¨rstner, 1998;

Sum Parameters: Potential and Limitations

23

Aquatic sample (TOC/DOC 100%)

Filtration

<0.45 μm

Suspended sediment Variable 30% ±

>0.45 μm

HCl pH ≤ 2 XAD-8 Dissolved nonhumic substances (NHS) (NHS; 30−50%) and inorganic ions

Penetration Adsorption NaOH 0.1 molar

Elution

Humic substances (HS; 20−60 %) HCl pH = 2 Precipitation Filtration >0.45 μm

Humic acids (HA; 5−20%)

<0.45 μm

Fulvic acids (FA; 10−40%) Figure 17 XAD method for the isolation of FAs and HAs from aqueous solutions.

Frimmel et al., 2007; Wilkinson and Lead, 2007; Frimmel and NieXner, 2010). Depending on the applied detection systems, summary of the element-specific data and/or spectrometric information can be obtained. An example of the carrier function of NP for heavy metals is given in Figure 19. It has to be noted that the particle-size distribution is pH dependent. At lower pH, there is agglomeration. The example of copper shows the vehicle function of the particles which also works for many other metal ions and even organic substances when they are present in the water matrix. The consequence for the transport of pollutants in heterogeneous or porous systems is their quite different distribution in comparison to merely dissolved systems. Powerful detection modes (e.g., ICP-MS, UVA, and laser-induced breakdown detection) and the advantage of aquatic sample application without major pretreatment open the door for a broad and urgently needed investigation of environmental samples (Ko¨ster et al., 2007; Siepmann et al., 2007).

3.01.7.2.3 Luminescence Luminescence (short-lived fluorescence and longer-lived phosphorescence) belongs to the most powerful spectrometric methods combining relatively simple applicability with high sensitivity. The principle of luminescence is based on excitation and retarded relaxation of electronic systems, including interactions and redox reactions. With the help of powerful instrumentation, luminescence images of a variety of aquatic substances can be obtained and a sort of mapping of structural features is possible. Figure 20 shows the excitation emission matrix for the luminescence of algae-derived substances from an aqueous algae extract (Ziegmann et al., 2010). Even though fluorescence has a quite long tradition in water-quality assessment (Christman and Ghassemi, 1966), the systematic assessment of water samples with advanced fluorescence/phosphorescence methods is still not very common (Chen et al., 2003; Cumberland and Baker, 2007). However, the availability of advanced instrumentation which leads to high sensitivity and resolutions will definitely

24

Sum Parameters: Potential and Limitations Computer ICP-MS Internal standard (20 μg l−1 Rh, In, Ir)

Eluent

Injection valve

Degazer

6 5

1 4

2 3

FD

Injection pump Waste UV

Laminar flow pump Sample injection

Separation channel

Syringe pump Cross flow Waste

8 Laponite (Mg signal) Laponite (Si signal) Laponite (UV signal)

0.0006 0.0004

4 2

0.0002

0

0.0000 101

6

102

103

15 10 5

28Si/103Rh

0.0008

24Mg/103Rh

UV absorbance (AU)

Figure 18 Experimental setup (example for coupling of A4 with different detectors) for the characterization of particulate water constituents (suspensions, colloidal systems, etc., according to Delay et al. (2010)).

0

104

63Cu/103Rh

Hydrodynamic particle diameter, dh (nm) 0.30 0.25 0.20 0.15 0.10 0.05 0.00 101

Cu (with laponite)

102

103

104

Hydrodynamic particle diameter, dh (nm) Figure 19 Typical flow-field-flow-fractionation (F4) derived particle size distribution for an aquatic suspension of laponite and its carrier function for copper (according to Metreveli and Frimmel (2007)).

stimulate a broad application of the method to fingerprint aquatic samples and follow bulk reactions (Kumke et al., 1998; Kumke and Frimmel, 2002). Due to the limited understanding of relaxation mechanisms in complex aquatic systems and the manifold influences of water constituents (e.g., light scattering and quenching), the structure-related

interpretation must remain uncertain so far and further research is urgently needed.

3.01.7.2.4 Bioeffect quantification According to the most interesting question on the bioeffects of water constituents, a new generation of biochemical tests and

Sum Parameters: Potential and Limitations

25

0

700

1.0E5 Excitation wavelength (nm)

600

2.0E5 3.0E5

500 4.0E5

400

300

300

400

500

600

700

Emission wavelength (nm−1) Figure 20 Excitation emission matrix for the luminescence of algae-derived substances from an aqueous algae extract (Microcystis aeruginosa after 24 days of growth and after ultrasonic decomposition, tentative assignment left to right: proteins, HSs, and pigments).

16000 100% Effect 14000

Peak area in AU (mm)

12000 10000 8000 6000 4000 2000 0% Effect 0 1

10

100

1000

Mass (pg) Figure 21 Typical dose/response curve for integrative quantification of endocrine activity (17b-estradiol on diol).

quantification methods has been developed. In addition to the basic determination of biodegradation (see Section 3.01.4.4), there are mainly toxic effects which need quantification (Tschmelak et al., 2005; Fent et al., 2006; Hock et al., 2006). A whole world of toxicities has gained attention reaching from acute toxicity genotoxicity, mutagenicity, etc. (Kokan et al., 1985; McDaniels et al., 1990; Hansen, 1992; McKelvey-Martin et al., 1993; Metcalfe et al., 2001) to the more subtle effects such as endocrine disruption (Routledge et al., 1998; Ternes et al., 1999; Kolpin et al., 2002; (see also Chapter 3.09 Bioassays for Estrogenic and Androgenic Effects of Water Constituents, Chapter 3.14 Drinking Water Toxicology in Its Regulatory Framework, Chapter 3.04 Emerging Contaminants)). Many

biochemical methods developed for medical application have been adapted to aquatic environmental application. Many of them are based on cell tests (Yasunaga et al., 2004) and/or use biochemical methods (see also Chapter 3.08 Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization, Chapter 3.09 Bioassays for Estrogenic and Androgenic Effects of Water Constituents). Figure 21 shows a typical concentration/response curve mostly used for quantification. The point of 50% response is often used for comparative characterization and quantification. Since the individual compounds responsible for the effect are often unknown, reference substances are used for calibration. Detectable

26

Sum Parameters: Potential and Limitations

Aquatic sample

Determination of abiotic Σ-parameters

Determination of bioeffects (bio Σ-parameters)

Assessment (a) maximum allowed values (b) adverse effects

Application of treatment methods

Determination of treatment related Σ-parameters

Identification and quantification of single compounds

Figure 22 Role of sum parameters in the use related assessment of water quality.

concentrations as they can be reached with the help of, for example, fluorescence indicators lie in the range of nanomole. It is beyond doubt that these bioeffect-related parameters supply highly relevant information on the function and possible dysfunction of aquatic systems and on their usability for mankind as well.

3.01.7.3 View The application of chemical and biochemical sum parameters has been and will be a vital part of water-quality assessment. There is a unique task to lead from the first glance to the final detailed characterization and information. Especially from an economic point of view on determinations, the principle has to hold: as integrated as possible and as detailed as necessary. In addition, reliable material balances can only be achieved by applying suitable sum parameters. Since the hydrological cycle and its compartments are highly dynamic, this also reflects on the sum parameters. They face emerging tasks for a comprehensive assessment. It is not only the demand of more powerful methods with higher precision and accuracy and hence lower limits of determination for classical parameters, it also includes the parameter itself which has to supply the information for a timely investigation of new water quality challenges such as determination of emerging classes of compounds and quantification of different bioeffects. Another important task of sum parameters is to identify domains of so far not recognized compounds and to lead the way to their molecular identification. This applies to aquatic systems in nature as well as to the technical steps of their use and by this to the entire hydrosphere (Figure 22). The well-understood key role of sum parameters in waterquality assessment is no contradiction or alternative to single compound identification. Instead, both approaches are highly complementary and urgently need one another for profound results.

References Abbt-Braun G (1992) Spectroscopic characterization of humic substances in the ultraviolet and visible region. In: Matthess G, Frimmel FH, Hirsch P, Schulz HD, and Usdowski H-E (eds.) Progress in Hydrogeochemistry, pp. 24--36. Berlin: Springer. Abbt-Braun G and Frimmel FH (2010) Determination of organic sum parameters. In: Du¨ring R, Hummel H, Ga¨th S, and Ledin A (eds.) Handbook of Environmental Analysis – Standard and Emerging Strategies in Water Sciences. Berlin: Springer (submitted). Abbt-Braun G, Glauner T, Brinkmann T, Hesse S, Ho¨rsch P, and Frimmel FH (2006) Abwasser im Grundwasser? Identifizierung der Abwasserexfiltration aus Kanalleckagen mit chemischen Summenparametern und organischen Spurenstoffen. In: DWA Deutsche Vereinigung fu¨r Wasserwirtschaft (ed.) Undichte Kana¨le-(k)ein Risiko? Gemeinschaftstagung 11/12, pp. 60–72, October 2006, Frankfurt. Abwasser und Abfall e.V. Abbt-Braun G, Lankes U, and Frimmel FH (2004) Structural characterisation of aquatic humic substances – the need for a multiple method approach. Aquatic Science 66: 151--170. Aitkenhead-Peterson JA, McDowell WH, and Neff JC (2003) Sources, production, and regulation of allochthonous dissolved organic matter inputs to surface waters. In: Findlay SEG and Sinsabaugh RL (eds.) Aquatic Ecosystems – Interactivity of Dissolved Organic Matter, pp. 25--70. Amsterdam: Academic Press. ASTM Standard D4189-07 (2007) Standard test method for silt density index (SDI) of water. ASTM International. ARW (Arbeitsgemeinschaft Rhein-Wasserwerke e.V.) (2004) Jahresbericht. Ko¨ln: Gescha¨ftsstelle ARW. AWBR (Arbeitsgemeinschaft Wasserwerke Bodensee-Rhein) (2006) Jahresbericht. Freiburg: Gescha¨ftsstelle AWBR. Bertilsson S and Jones JB, Jr. (2003) Supply of dissolved organic matter to aquatic ecosystems: Autochthonous sources. In: Findlay SEG and Sinsabaugh RL (eds.) Aquatic Ecosystems – Interactivity of Dissolved Organic Matter, pp. 3--24. Amsterdam: Academic Press. Bloom P and Leenheer JA (1989) Vibrational, electronic, and high-energy spectroscopic methods for characterizing humic substances. In: Hayes MHB, MacCarthy P, Malcolm RL, and Swift RS (eds.) Humic Substances II: In Search of Structure, pp. 409–446. Chichester: Wiley-Interscience. Box JD (1983) Investigation of the Folin–Ciocalteu phenol reagent for the determination of polyphenolic substances in natural waters. Water Research 17: 511--525. Brinkmann T, Abbt-Braun G, and Frimmel FH (2003a) Alkaline degradation of dissolved organic matter. Acta Hydrochimica et Hydrobiologica 31: 1--12. Brinkmann T, Ho¨rsch P, Sartorius D, and Frimmel FH (2003b) Photoformation of lowmolecular-weight organic acids from brown water dissolved organic matter. Environmental Science and Technology 37(18): 4190--4198.

Sum Parameters: Potential and Limitations

Cabaniss SE and Shuman MS (1987) Synchronous fluorescence spectra of natural waters: Tracing sources of dissolved organic matter. Marine Chemistry 21: 37--50. Chen Y, LeBoeuf EJ, Dai S, and Gu B (2003) Fluorescence spectroscopic studies of natural organic matter fractions. Chemosphere 51: 325--346. Chen Y, Senesi N, and Schnitzer M (1977) Information provided on humic substances by E4/E6 ratios. Soil Science Society of America Journal 41: 352--358. Christman RF and Ghassemi M (1966) Chemical nature of organic color in water. Journal of the American Water Works Association 58: 723--744. Cumberland SA and Baker A (2007) The freshwater dissolved organic matter fluorescence – total organic relationship. Hydrological Processes 21: 2093--2099. Dafner EV and Wangersky PJ (2002) A brief overview of modern directions in marine DOC studies. Part I: Methodological aspects. Journal of Environmental Monitoring 4: 48--54. Delay M, Metreveli G, Tercero L, and Frimmel FH (2010) Coupling techniques to quantify nanoparticles and to characterize their interactions with water constituents. In: Frimmel FH and NieXner R (eds.) Nanoparticles in the Water Cycle, pp. 139--164. Berlin: Springer. DIN 38404-3 (2005) Bestimmung der Absorption im Bereich der UV-Strahlung, Spektraler Absorptionskoeffizient. In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren C3, 65, Lieferung 2006. Weinheim: Wiley-VCH; Berlin: Beuth. DIN 38409 (1980) Bestimmung der methylenblauaktiven und der bismutaktiven Substanzen. In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren H23 (EN 903:1993, water quality – determination of anionic surfactants by measurement of the methylene blue index MBAS; ISO 7875-1:1984 modified) 9, Lieferung 1981. Weinheim: Wiley-VCH; Berlin: Beuth. DIN 38409 (1984) Bestimmung des Phenol-Index. In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren H16 (ISO 6437:1990, water quality – determination of phenol index – 4-aminoantipyrine spectrometric methods after distillation), 13, Lieferung 1984. Weinheim: WileyVCH; Berlin: Beuth. DIN 38409-8 (1984) Bestimmung der extrahierbaren gebundenen Halogene (EOX). In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren H8, 13, Lieferung 1984, withdrawl 2207. Weinheim: WileyVCH; Berlin: Beuth. DIN 38409-22 (2001) Bestimmung gelo¨ster adsorbierbarer organisch gebundener Halogene in stark salzhaltigen Wa¨ssern nach Festphasenanreicherung (SPE-AOX). In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren H22, 51, Lieferung 2001. Weinheim: Wiley-VCH; Berlin: Beuth. DIN 38409-41 (1980) Bestimmung des Chemischen Sauerstoffbedarfs (CSB) im Bereich u¨ber 15 mg/l. In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren H41 (ISO 6060:1989, modified, water quality – determination of the chemical oxygen demand), 9, Lieferung 1981. Weinheim: Wiley-VCH; Berlin: Beuth. DIN 38409-43 (1981) Bestimmung des Chemischen Sauerstoffbedarfs (CSB) Kurzzeitverfahren. In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren H43, 11, Lieferung 1982. Weinheim: WileyVCH; Berlin: Beuth. DIN 38409-44 (1992) Bestimmung des Chemischen Sauerstoffbedarfs (CSB) im Bereich 5 bis 50 mg/l. In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren H44, 27, Lieferung 1992. Weinheim: WileyVCH; Berlin: Beuth. DIN EN 1484 (1997) Wasseranalytik – Anleitungen zur Bestimmung des gesamten organischen Kohlenstoffs (TOC) und des gelo¨sten organischen Kohlenstoffs (DOC) (German version EN 1484:1997; water analysis – guidelines for the determination of total organic carbon (TOC) and DOC). In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren H3, 40, Lieferung 1998. Weinheim: Wiley-VCH; Berlin: Beuth. DIN EN 1899-1 (1998) Wasserbeschaffenheit – Bestimmung des Biochemischen Sauerstoffbedarfs nach n Tagen (BSBn) – Teil 1: Verdu¨nnungs- und Impfverfahren nach Zugabe von Allylthioharnstoff (German version EN 1899-1:1998; water quality – determination of biochemical oxygen demand after n days (BODn) – part 1: Dilution and seeding method with allylthiourea addition (ISO 5815:1989,

27

modified), In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), 43, Lieferung 1999. Weinheim: Wiley-VCH; Berlin: Beuth. DIN EN ISO 8467 (1995) Wasserbeschaffenheit – Bestimmung des Permanganat-Index (German version EN ISO 8467:1995; water quality – determination of permanganate index (ISO 8467:1993)), In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), 34, Lieferung 1996. Weinheim: Wiley-VCH; Berlin: Beuth. DIN EN ISO 9562 (2005) Bestimmung adsorbierbarer organisch gebundener Halogene (AOX). (German version EN ISO 9562:2004; determination of adsorbable organically bound halogens (AOX) (ISO 9562:2004)). In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren H14, 63, Lieferung 2006. Weinheim: Wiley-VCH; Berlin: Beuth. DIN EN ISO 14402 (1999) Bestimmung des Phenolindex mit der FlieXanalytik (FIA und CFA) (German version EN ISO 14402:1999; water quality – determination of phenol index by flow analysis (FIA and CFA) (ISO 14402:1999)). In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren H37, 49, Lieferung 2001. Weinheim: Wiley-VCH; Berlin: Beuth. DIN ISO 15705 (2003) Ku¨vettentest (ISO 15705:2002; determination of the chemical oxygen demand index (ST-COD) – small-scale sealed tube method). In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren H45, 57, Lieferung 2003. Weinheim: Wiley-VCH; Berlin: Beuth. DIN EN ISO 7887 C1 (1994) Wasserbeschaffenheit – Untersuchung und Bestimmung der Fa¨rbung (German version EN ISO 7887:1994; water quality – examination and determination of colour (ISO 7887:1994)). In: Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (German Standard Methods for the Examination of Water, Waste Water and Sludge), Verfahren C1, 32, Lieferung 1995. Weinheim: Wiley-VCH; Berlin: Beuth. Dinar E, Taraniuk I, Graber ER, et al. (2006) Cloud condensation nuclei properties of model and atmospheric HULIS. Atmospheric Chemistry and Physics 6: 2465--2482. Duursma EK and Dawson R (1981) Marine Organic Chemistry. Amsterdam: Elsevier. Fent K, Weston AA, and Caminada D (2006) Ecotoxicology of human pharmaceuticals. Aquatic Toxicology 76: 122--159. Frimmel FH (1992) Isolation and general characterization of organic acids from pore water. In: Matthess G, Frimmel FH, Hirsch P, Schulz HD, and Usdowski HE (eds.) Progress in Hydrogeochemistry, pp. 9--13. Berlin: Springer. Frimmel FH and Abbt-Braun G (1999) Basic characterization of reference NOM from central Europe – similarities and differences. Environment International 25: 191--207. Frimmel FH and Abbt-Braun G (2009) Dissolved organic matter (DOM) in natural environments. In: Senesi N, Xing B, and Huang PM (eds.) Biophysico-Chemical Processes Involving Natural Nonliving Organic Matter in Environmental Systems, pp. 367--406. Hoboken, NJ: Wiley. Frimmel FH, Abbt-Braun G, Heumann KG, Hock B, Lu¨demann HD, and Spiteller M (eds.) (2002) Refractory Organic Substances (ROS) in the Environment. New York: Wiley. Frimmel FH and Christman RF (1988) Humic Substances and Their Role in the Environment. Chichester: Wiley. Frimmel FH, Delay M (2010) Introducing the ‘‘nano-world’’, In: Frimmel FH NieXner R Nanoparticles in the Water Cycle, pp. 1–12. Berlin: Springer. Frimmel FH and Mu¨ller MB (eds.) (2006) Heil-Lasten. Heidelberg: Springer. Frimmel FH and NieXner R (eds.) (2010) Nanoparticles in the Water Cycle. Berlin: Springer. Frimmel FH, von der Kammer F, and Flemming H-C (eds.) (2007) Colloidal Transport in Porous Media. Berlin: Springer. Graber ER and Rudich Y (2006) Atmospheric HULIS: How humic-like are they? A comprehensive and critical review. Atmospheric Chemistry A: Physics 6: 729--753. Gruenfeld M (1973) Extraction of dispersed oils from water for quantitative analysis by infrared spectrophotometry. Environmental Science and Technology 7: 636--639. Hambsch B, Werner P, and Frimmel FH (1992) Bacterial growth measurements in treated waters from different origin. Acta Hydrochimica et Hydrobiologica 20: 9--14. Hansen PD (1992) Phagocytosis in Mytilus edulis, a system for understanding the sublethal effects of anthropogenic pollutants and the use of AOX as an integrating parameter for the study of equilibria between chlorinated organics in Dreissena polymopha following long term exposures. In: Neumann D and Jenner HA (eds.) The Zebra Mussel Dreissena Polymorpha, pp. 171--184. Jena: Gustav Fischer Verlag.

28

Sum Parameters: Potential and Limitations

Hautala K, Peravuori J, and Phihlaja K (2000) Measurement of aquatic humus content by spectroscopic analyses. Water Research 34: 246--258. Her N, Amy G, Foss D, and Cho J (2002a) Variation of molecular weight estimation by HP-size exclusion chromatography with UVA versus online DOC detection. Environmental Science and Technology 36: 3393--3399. Her N, Amy G, Foss D, and Cho J (2002b) Optimiziation of method for detecting and characterising NOM by HPLC-size exclusion chromatography with UV and on-line DOC detection. Environmental Science and Technology 36: 1069--1076. Her N, Amy G, McKnight D, Sohn J, and Yoon Y (2003) Characterisation of DOM as a function of MW by fluorescence EEM and HPLC–SEC using UVA, DOC and fluorescence detection. Water Research 37: 4295--4303. Herzsprung P, Weil L, and Quentin K-E (1989) Determination of organophosphorous compounds and carbonates by their inhibition of cholinesterase. Zeitung WasserAbwasser-Forschung 22: 67–72 (in German). Heukelekian H and Kaplovsky AJ (1949) Improved method of volatile-acid recovery from sewage sludges. Sewage Works Journal 21: 974--983. Hock B, Giersch T, Dankwardt A, Kramer K, and Pullen S (2006) Toxicity assessment and on-line monitoring: Immunoassays. Environmental Toxicology and Water Quality 9: 243--262. Huber SA and Frimmel FH (1994) Direct gel chromatographic characterisation and quantification of marine dissolved organic carbon using high-sensitivity DOC detection. Environmental Science and Technology 28: 1194--1197. IHSS (2010) International Humic Substances Society. http://www.ihss.gatech.edu/ihss2 (accessed April 2010). ISO 7875-2:1984 (1984) Water quality – determination of surfactants – part 2: Determination of non-ionic surfactants using Dragendorff reagent. Kaplan LA (1994) A field and laboratory procedure to collect, process, and preserve freshwater samples for dissolved carbon analysis. Limnology and Oceanography 37: 1470--1476. Khan E and Pillai SS (2007) Interferences contributed by leaching from filters on measurements of collective organic constituents. Water Research 41: 1841--1850. Kokan RM, Sabo KM, and Landolt MI (1985) Cytotoxicity/genotoxicity: The application of cell culture techniques to the measurement of marine sediment pollution. Aquatic Toxicology 6: 165--177. Kolpin DW, Furlong ET, Meyer MT, et al. (2002) Pharmaceuticals, hormones and other organic wastewater contaminants in U.S. streams, 1999–2000: A national reconnaissance. Environmental Science and Technology 36: 1202--1211. Koppe P and Stozek A (1990) Kommunales Abwasser. Essen: Vulkan Verlag. Ko¨ster R, Wagner T, Delay M, and Frimmel FH (2007) Release of contaminants from bottom-ashes – colloid facilitated transport and colloidal trace analysis by means of laser-induced breakdown detection (LIBD). In: Frimmel FH, von der Kammer F, and Flemming H-C (eds.) Colloidal Transport in Porous Media, pp. 251--272. Berlin: Springer. Ku¨hn W and Sontheimer H (1973) Investigations of the determination of organic halogen compounds on activated carbon. Vom Wasser 41: 65–79 (in German). Ku¨mmerer K (ed.) (2008) Pharmaceuticals in the Environment. Berlin: Springer. Kumke MU, Abbt-Braun G, and Frimmel FH (1998) Time-resolved fluorescence measurements of aquatic natural organic matter (NOM). Acta Hydrochimica et Hydrobiologica 26: 73--81. Kumke MU and Frimmel FH (2002) Stationary and time-resolved fluorescence for refractory organic substances characterization In: Frimmel FH, Abbt-Braun G, Heumann KG, Hock B, Lu¨demann HD, and Spiteller M (eds.) Refractory Organic Substances (ROS) in the Environment, pp. 215–231 Weinheim: Wiley-VCH. Kunenkov EV, Kononikhin AS, Perminova IV, et al. (2009) Total mass difference statistics algorithm: A new approach to identification of high-mass building blocks in electrospray ionization Fourier transform ion cyclotron mass spectrometry data of natural organic matter. Analytical Chemistry 81(24): 10106--10115. Kunkel E, Peitscher G, and Espeter K (1977) New developments in trace and microanalysis of surfactants. Tenside Detergents 14: 199--202. Laird LB, Taylor HE, and Kennedy VC (1988) Snow chemistry of the Cascade–Sierra Nevada mountains. Environmental Science and Technology 20: 275--290. Langhals H, Abbt-Braun G, and Frimmel FH (2000) Association of humic substances: Verification of Lambert–Beer’s law. Acta Hydrochimica et Hydrobiologica 28: 329--332. Lankes U, Lu¨demann H-D, and Frimmel FH (2008) Search for basic relationships between ‘‘molecular size’’ and ‘‘chemical structure’’ of aquatic natural organic matter – answers from 13C and 15N CPMAS NMR spectroscopy. Water Research 42: 1051--1060. Lankes U, Mu¨ller MB, Weber M, and Frimmel FH (2009) Reconsidering the quantitative analysis of organic carbon concentrations in size exclusion chromatography. Water Research 43: 915--924. Leenheer JA and Croue´ J-P (2003) Characterizing dissolved aquatic organic matter. Environmental Science and Technology 37: 18A--26A.

Leithe W (1971) Comparison of COD, BOD and PMC in drinking water and surface waters. Vom Wasser 38: 119–127 (in German). Lee W and Westerhoff P (2006) Dissolved organic nitrogen removal during water treatment by aluminium sulfate and cationic polymer coagulation. Water Research 40(20): 3767--3774. Lund W, Helbak IA, and Seip HM (1990) Studies of the complexation properties of aquatic humic material by differential pulse polarography. Science of the Total Environment 92: 269--281. MacCarthy P and Rice JA (1985) Spectroscopic methods (other than NMR) for determining functionality in humic substances. In: Aiken GR, McKnight DM, Wershaw RL, and MacCarthy P (eds.) Humic Substances in Soil, Sediment and Water, pp. 527--559. New York: Wiley. Malcolm RL (1985) Geochemistry of stream fulvic and humic substances. In: Aiken GR, McKnight DM, Wershaw RL, and MacCarthy P (eds.) Humic Substances in Soil, Sediment and Water, pp. 181--209. New York: Wiley. Malcolm RL and MacCarthy P (1992) Quantitative evaluation of XAD 8 and XAD 4 resins used in tandem for removing organic solutes from water. Environment International 18: 597--607. Matthess G, Frimmel FH, Hirsch P, Schulz HD, and Usdowski HE (eds.) (1992) Progress in Hydrogeochemistry: Organics – Carbonate Systems – Silicate Systems – Microbiology – Models Springer Berlin. McDaniels AE, Reyes AL, Wymer LJ, Rankin CC, and Stelma GN, Jr. (1990) Comparison of the Salmonella (Ames) test, umu test and the SOS chromotests for detecting genotoxins. Environmental and Molecular Mutagenesis 16: 204--215. McKelvey-Martin VJ, Green MH, Schmezer P, Pool-Zobel BL, De Me´o MP, and Collins A. The single cell gel electrophoresis assay (comet assay): A European review. Mutation Research 288: 47–63. McKnight DM and Aiken GR (1998) Sources and age of aquatic humus. In: Hessen DO and Tranvik LJ (eds.) Aquatic Humic Substances, pp. 9--39. Berlin: Springer. Metcalfe CD, Metcalfe TL, Kiparissis Y, et al. (2001) Estrogenic potency of chemical detected in sewage treatment plant effluents as determined by in vivo assays with Japanese medaka (Oryzias latipes). Environmental Toxicological Chemistry 20: 297--308. Metreveli G and Frimmel FH (2007) Influence of Na-bentoite colloids on the transport of heavy metals in porous media. In: Frimmel FH, von der Kammer F, and Flemming H-C (eds.) Colloidal Transport in Porous Media, pp. 29--53. Berlin: Springer. Neufeld RD and Paladino SB (1985) Comparison of 4-aminoantipyrine and gas–liquid chromatography techniques for analysis of phenolic compounds. Journal of the Water Pollution Control Federation 57: 1040--1044. Oleksy-Frenzel J, Wischnack S, and Jekel M (1995) Bestimmung der organischen Gruppenparameter AOCl, AOBr und AOI in Kommunalabwasser. Vom Wasser 85: 59--67. Oleksy-Frenzel J, Wischnack S, and Jekel M (2000) Determination of organic group parameters AOCl, AOBr and AOI in municipal wastewater. Journal of Analytical Chemistry 366: 89--94. Osburn QW (1986) Analytical methodology for LAS in waters and wastes. Journal of the American Oil Chemists’ Society 63: 257--263. Pauling (1960) The Nature of the Chemical Bond and the Structure of Molecules and Crystals: An Introduction to Modern Structural Chemistry, 3rd edn. New York: Cornell University Press. Pavan P, Battistoni P, Cecchi F, and Mata-Alvarez J (2000) Two-phase anaerobic digestion of source sorted OFMSW (organic fraction of municipal solid waste): Performance and kinetic study. Water Science and Technology 41: 111--118. Perdue EM and Gjessing ET (1990) Organic Acids in Aquatic Ecosystems. Chichester: Wiley. Putschew A and Jekel M (2006) Iodinated X-ray contrast media. In: Reemtsma T and Jekel M (eds.) Organic Pollutants in the Water Cycle, pp. 87--98. Weinheim: WileyVCH. Putschew A, Wischnack S, and Jekel M (2000) Occurrence of tri-iodinated X-ray contrast agents in the aquatic environment. Science of the Total Environment 255: 129--134. Reemtsma T and Jekel M (2006) Organic Pollutants in the Water Cycle. Weinheim: Wiley-VCH. Reemtsma T, These A, Springer A, and Linscheid M (2008) Differences in the molecular composition of fulvic acid size fractions detected by size-exclusion chromatography–on line Fourier transform ion cyclotron resonance (FTICR)–mass spectrometry. Water Research 42: 63--72. Routledge EJ, Sheahan D, Desbrow C, Brightly GC, Waldock M, and Sumpter JP (1998) Identification of estrogenic chemicals in STW effluent. II: In vivo responses in trout and roach. Environmental Science and Technology 32: 1559–1565.

Sum Parameters: Potential and Limitations

Ryan DK and Weber JH (1982) Flourescence quenching titration for determination of complexing capacities and stability constants of fulvic acids. Analytical Chemistry 54: 986--990. Schwuger MJ (1996) Detergents in the Environment. Surfactant Science Series, vol. 65. New York: Dekker. Senesi N, Miano TM, Provenzano MR, and Brunetti G (1989) Spectroscopic and compositional comparative characterization of IHSS reference and standard fulvic and humic acids of various origin. Science of the Total Environment 81/82: 143--156. Senesi N, Xing B, and Huang PM (eds.) (2009) Biophysico-Chemical Processes Involving Natural Nonliving Organic Matter in Environmental Systems. Hoboken, NJ: Wiley. Siepmann R, Von der Kammer F, and Fo¨rstner U (2007) Transport of colloids in filter columns: Laboratory and field experiments. In: Frimmel FH, von der Kammer F, and Flemming H-C (eds.) Colloidal Transport in Porous Media, pp. 87--115. Berlin: Springer. Skoog DA, West DM, Holler FJ, and Crouch SR (2003) Fundamentals of Analytical Chemistry, 8th edn. Belmont, CA: Brooks Cole. Sontheimer H and Schnitzler M (1982) EOX oder AOX? Zur Anwendung von Anreicherungsverfahren bei der analytischen Bestimmung von chemischen Gruppenparametern. Vom Wasser 59: 169--179. Sontheimer H, Spindler P, and Rohmann U (1986) Wasserchemie fu¨r Ingenieure. Frankfurt/Main: ZfGW-Verlag GmbH. Standard Methods (2005) Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 2120 B (2005) Color – visual comparison method. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 2120 C (2005) Color – spectrophotometric – single-wavelength method (proposed). In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 2120 D (2005) Color – spectrophotometric – multi-wavelength method. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 2120 E (2005) Color – tristimulus spectrophotometric method. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 5210 B (2005) Biochemical oxygen demand – 5-day BOD test. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 5210 C (2005) Biochemical oxygen demand – ultimate BOD test. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 5210 D (2005) Biochemical oxygen demand – respirometric method. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 5220 B (2005) Chemical oxygen demand (COD) – open reflux method. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 5220 C (2005) Chemical oxygen demand (COD) – closed reflux, titrimetric method. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 5220 D (2005) Chemical oxygen demand (COD) – closed reflux, colorimetric method. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 5310 B (2005) Total organic carbon (TOC) – high-temperature combustion method. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 5310 C (2005) Total organic carbon (TOC) – persulfate–ultraviolet or heated-persulfate oxidation method. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 5310 D (2005) Total organic carbon (TOC) – wet-oxidation method. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Standard Methods 5320 B (2005) Dissolved organic halogen – adsorption–pyrolysis– titrimetric method. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association.

29

Standard Methods 5910 B (2005) UV-absorbing organic constituents –ultraviolet absorption method. Standard Methods for the Examination of Water and Wastewater. Washington, DC: American Public Health Association. Standard Methods 6420 B (2005) Phenols – liquid–liquid extraction gas chromatographic method. In: Standard Methods for the Examination of Water and Wastewater, 21st edn. Washington, DC: American Public Health Association. Steinberg CEW (2003) Ecology of Humic Substances in Freshwaters. Berlin: Springer. Tabak HH Bunch RL (1981) Measurement of non-ionic surfactants in aqueous environments. In: Proceedings of the 36th International Waste Conference, pp. 888–907. Lafayette, IN: Purdue University. Tercero ELA and Frimmel FH (2008) Formation of brominated products in irradiated titanium dioxide suspensions containing bromide and dissolved organic carbon. Water Research 42: 1778--1784. Ternes TA and Hirsch R (2000) Occurrence and behaviour of iodinated contrast media in sewage water facilities and the aquatic environment. Environmental Science and Technology 34: 2741--2748. Ternes TA, Stumpf M, Mueller J, Haberer K, Wilken RD, and Servos M (1999) Behavior and occurrence of estrogens in municipal sewage treatment plants. I. Investigations in Germany, Canada and Brazil. Science of the Total Environment 225: 81--90. Thurman EM (1985) Organic Geochemistry of Natural Waters. Dodrecht: Martinius Nijhoff/Dr. W. Junk Publisher. Thurman EM and Malcolm R (1981) Preparative isolation of aquatic humic substances. Environmental Science and Technology 15: 463--466. Tschmelak J, Proll G, and Gauglitz G (2005) Improved strategy for biosensor-based monitoring of water bodies with diverse organic carbon level. Biosensors and Bioelectronics 21: 979--983. Tuschall RJ and Brezonik PL (1983) Complexation of heavy metals by aquatic humus: A comparative study of five analytical methods. In: Christman RF and Gjessing ET (eds.) Aquatic and Terrestrial Humic Materials, pp. 275--294. Ann Arbor, MI: Ann Arbor Science. Ullmann WW and Sanderson WW (1959) A further study of methods for the determination of grease in sewage. Sewage and Industrial Wastes 31: 8--19. US EPA (Environmental Protection Agency) (1998) Method 1664, Revision A. nHexane Extractable Material (HEM; Oil and Grease) and Silica Gel Treated n-Hexane Extractable Material by Extraction and Gravimetry. Washington, DC: US Environmental Protection Agency. US EPA (Environmental Protection Agency) (1999) Method 1664, Revision A. Washington, DC: US Environmental Protection Agency. Von der Kammer F and Fo¨rstner U (1998) Natural colloid characterisation using flowfield-flow-fractionation followed by multi-detector analysis. Water Science and Technology 37: 173--180. Wagner R (1973) Modification of the potassium dichromate method for the determination of the total oxygen demand of organic substances Vom Wasser 41: 1–26 (in German). Wagner R (1979) The application of BOD in wastewater assessment Vom Wasser 52: 253–285 (in German). Wedepohl KH (1969) Handbook of Geochemistry, Vol. 1. Berlin, Heidelberg, New York: Springer. Westerhold AF (1963) Organic acids in digester liquor by chromatography. Journal of the Water Pollution Control Federation 35: 1431--1439. Wilkinson KJ and Lead JR (eds.) (2007) Environmental Colloids and Particles: Behaviour, Separation and Characterisation. Chichester: Wiley. Williams PM (1971) The distribution and cycling of organic matter in the ocean. In: Faust SD and Hunter JV (eds.) Organic Compounds in Aquatic Environments, pp. 145--164. New York: Dekker. Wolf L, Eiswirth M, and Ho¨tzl H (2006) Assessing sewer–groundwater interaction at the city scale based on individual sewer defects and marker species distribution. Environmental Geology 49: 849--857. Yasunaga K, Kiyonari A, Oikawa T, Abe N, and Yoshikawa K (2004) Evaluation of the Salmonella umu test with 83 NTP chemicals. Environmental and Molecular Mutagenesis 44(4): 329--345. Zanke GP and Ho¨pner T (1982) Relations between COD and TOC in an experimental watercourse. Vom Wasser 58: 257–267 (in German). Ziegmann M, Abert M, Mu¨ller M, and Frimmel FH (2010) Use of fluorescence fingerprints for the estimation of bloom formation and toxin production of Microcystis aeruginosa. Water Research 44: 195--204. Zsolnay A´ (2003) Dissolved organic matter: Artefacts, definitions, and functions. Geoderma 113: 187--209.

3.02

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

AV Hirner and J Hippler, University of Duisburg-Essen, Essen, Germany & 2011 Elsevier B.V. All rights reserved.

3.02.1 Introduction 3.02.2 Natural Waters and Anthropogenic Influence 3.02.2.1 Metal(loid) Concentration 3.02.2.2 Metal(loid) Species 3.02.3 Selected Elements 3.02.3.1 Arsenic and Antimony 3.02.3.1.1 Arsenic: Introduction and overview 3.02.3.1.2 Arsenic in drinking water 3.02.3.1.3 Methylated As species 3.02.3.1.4 Thiolated As species 3.02.3.1.5 Antimony 3.02.3.2 Mercury 3.02.3.2.1 Introduction and overview 3.02.3.2.2 Impact of mining, Minamata, and Florida Everglades 3.02.3.2.3 Essentials of biomethylation 3.02.3.2.4 Biomethylation within parameter gradients 3.02.3.2.5 Abiotic methylation 3.02.3.2.6 Global concern 3.02.3.3 Other Metals (Cd, Cu, Pb, and Zn) 3.02.3.4 Platinum Group Elements 3.02.4 Conclusions Acknowledgments References

3.02.1 Introduction Common instrumental analytical techniques such as spectrometry (e.g., graphite furnace atomic absorption spectrometry (GFAAS), hydride-generation atomic absorption spectrometry (HG-AAS), hydride-generation atomic fluorescence spectrometry (HG-AFS), inductively coupled plasma-atomic emission spectroscopy (ICP-AES), and instrumental neutron activation analysis (INAA)), mass spectrometry (mostly inductively coupled plasma-mass spectrometry (ICP-MS)), or electro-analytical techniques are used for reliable determination of trace elements in waters (e.g., Kellner et al., 1998); with respect to speciation, hyphenated combinations of chromatographic separation (gas chromatography (GC), highperformance liquid chromatography (HPLC)) with online elemental (ICP-MS) as well as molecular detection (MS of mass fragments) are needed (Ko¨sters 2003; Krupp et al., 2008), and have to be carefully applied to avoid artifacts leading to misinterpretations (e.g., Hirner, 2006; Francesconi and Sperling, 2005). However, only routine application of these methods will not provide adequate detection limits for the determination of ultra-trace concentrations such as Sb or Pb in uncontaminated groundwater. At extremely low concentrations, serious challenges are extreme analytical sensitivity (e.g., requiring a sectorfield ICP-MS) and associated lowest blank levels (e.g., clean bench class 100 conditions, low-density polyethylene (LDPE) bottles; Shotyk et al., 2005; Boutron and Go¨rlach, 1990).

31 32 32 33 36 36 36 37 40 41 41 42 42 43 45 46 47 48 48 49 50 50 50

Parallel to improvements in instrumental analytical techniques, the quality of analytical data received also increases, in particular enabling lower detection limits, and thereby facilitating the measurement of hitherto unavailable ultra-trace concentrations. As a consequence of these analytical efforts, new analytes may come into focus, especially if they are of interest to environmental research. Metal(loid)s resemble the most important group of inorganic contaminants in environmental chemistry, and have become an environmental-quality target worldwide: for example, in the US, they are of concern to almost all programs of the US Environmental Protection Agency (EPA); e.g., ambient water-quality criteria for the protection of human health and aquatic life against the potential toxic effects of these elements are developed (Clean Water Act). Together with the metal(loid)s mentioned in this chapter, the EPA priority list of elements of concern also includes Al, Ba, Be, Cr, Co, Mn, Mo, Ni, Se, Ag, Sr, Tl, and V. Among the emerging metal(loid) pollutants being presently discussed in pertinent water research (Richardson, 2009, 2008, 2007) there are some included in this chapter, such as arsenic and antimony (see Section 3.02.3.1), while others, such as (1) tungsten, (2) uranium, and (3) tributyltin (TBT), are only briefly introduced here: 1. Although tungsten (W) is widely used in consumer products because of its physical–chemical properties, in the

31

32

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

industry and in the military (e.g., currently in green bullets as an environment-friendly (?) alternative to conventional lead-based ammunition), its (potential) mobility in environmental compartments, biogeochemistry, and toxicity is largely unknown yet (Koutsospyros et al., 2006; Strigul et al., 2005). 2. Besides arsenic (As), uranium (U) has also been identified as drinking-water contaminant: for example, while in the groundwater of western Bangladesh, concentrations of As exceeding World Health Organization (WHO) drinkingwater guidelines were found in 33% of the sampled wells, in 48% of the wells, elevated levels of U were detected (Frisbie et al., 2009). Although in Germany, o0.6% of all households are estimated to receive drinking water exceeding the threshold level of 10 mg l1 U, up to 75 mg l1 have been measured in Bavaria (Friedmann and Lindenthal, 2009). 3. Triorganotin biocides amount to about one-quarter of organometallic tin compounds constituting 7% of total tin (Cima et al., 2003). In particular, because of its high stability as well as mobility and toxicity, TBT is considered to represent a high-priority pollutant in aquatic systems requiring very sensitive analytical detection (Cima et al., 2003; Dopp et al., 2007; Gremm and Frimmel, 1992): for example, the present European Union (EU) environmental quality standard (Water Framework Directive) for TBT in water will be lowered to 0.2 ng l1 requiring a lower limit of quantification of 0.06 ng l1 (H. Emons, pers. commun). The main goal of this chapter is to address the importance of speciation to evaluate metal(loid) transport behavior, bioconcentration, and toxicity in aquatic systems. While association of most metal(loid)s with inorganic and macromolecular organic ligands such as dissolved organic carbon DOC is roughly known (especially for economically important metals such as Cd, Cu, Pb, and Zn), specific low-molecular organic species have been found for As (and partly for Sb) as well as Hg exhibiting remarkable toxic properties. However, toxicity is still not sufficiently understood in the case of the platinum group elements (PGEs) due to ultra-trace concentration levels posing extreme problems in their lack of their speciation. Notably, with the exception of PGEs because of their environmental importance, all elements mentioned earlier are listed as metal(loid)s of primary interest by EPA commissions, in particular those concerned with drinking-water standards and Ambient Water Quality Criteria (AWQC) (Fairbrother et al., 2007). Generally, the following text focuses on the elements listed in the title as well as on the water phase; therefore, other elements, sediment/soil chemistry, and remediation of polluted sites are not discussed here. For understanding basics of water chemistry as well as geochemistry, the reader is referred to relevant standard references such as Merian et al. (2004), Merian (1984), Stumm and Morgan (1996), Morgan and Stumm (1991), Hitchon et al. (1999), Frimmel (1999), Salomons and Fo¨rstner (1984), or Fo¨rstner and Wittmann (1981).

3.02.2 Natural Waters and Anthropogenic Influence 3.02.2.1 Metal(loid) Concentration Complex physical–chemical and biological, equilibrium and kinetic processes control the distribution of trace elements in

groundwater imported by a variety of natural and anthropogenic sources. For reasons that may be unrelated to human activity, many surface and groundwaters contain natural concentrations of metals exceeding common drinking-water standards (Runnells et al., 1992). On the other hand, when a regulatory decision is to be made to restore affected waters to a presumed earlier state, it is obviously unrealistic to assign clean-up goals being below preexistent metal levels. Thus, metal amounts that were naturally present and amounts added as a result of human activities must be distinguished from each other. In other words, for most environmental studies (at the local, regional, and global scales) it is of fundamental importance to know the (local, regional, and global, respectively) ‘background concentrations’, that is, the borderline between concentrations of a species that naturally occur in groundwater to be compared with those concentrations in the case when anthropogenic activities are involved. It is already well known that the latter may contain extreme metal loads such as Cd, Pb, or Zn in fertilizers in the g kg1 range with significant solubility (Frimmel, 1999). However, due to progress in environmental measures and technology, water pollution in developed Western countries will decrease, making the differentiation against natural background values even more critical. As an appropriate example, in Germany, heavy-metal emissions in aqueous systems have significantly reduced from 1985 to 2000 (94% of As, 72% of Hg, 76% of Cd and Cu, and 80% of Zn; SRU, 2008) and are not anymore mainly derived from industry but from diffuse sources. A closer look, however, shows that the environmental background mentioned above is not identical to the geochemical background well defined by (exploration) geochemists to differentiate between the analyte concentration within a rock matrix devoid of enrichments and those rock parts showing positive anomalies such as in fissures or veins (Matschullat et al., 2000). Thus, environmentalists must explain if they are explicitly including natural positive anomalies in the term natural background. It should also be noted that – because of significant geogenic variations due to different underlying rocks (Frimmel, 1999; White et al., 1963) – while in principle it is almost impossible to quantify a single true background value, it should be possible to define upper limits for the background with a defined statistical reliability (Matschullat et al., 2000). Generally, when comparing traceelement concentrations in waters as given earlier by Bowen (1979) as reference, with those of more recent publications (e.g., Bruland, 1983; Salbu and Steinnes, 1995; Reimann and De Caritat, 1998), it is apparent that the latter concentrations are lower, maybe because of their higher analytical quality (improvement of techniques and avoidance of artifacts and contaminations). Taking extraordinary precautions needed to measure artifact-free trace elements in the ng l1 range, Shotyk and Krachler (2009) tried to find something such as baseline or background concentrations of trace elements in natural water. They investigated ancient layers of ice from Devon Island, Nunavut (Canada), considered to be the cleanest water on earth. They found As, Cd, Cu, Pb, and Sb at the low ng l1 range; other elements such as Ag, Bi, Mo, Sb, Sc, Tl, and U were present even at concentrations below 1 ng l1. Similar concentrations for Cd, Cu, and Pb could be found in

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

groundwater from two artesian flows in Simcoe County (Canada) and for Cd and Cu in Onyx River water (Wright Valley, Antarctica) (Green et al., 2005). Shotyk and Krachler (2009) came to the conclusion that in natural waters, many trace-element concentrations can be extremely low, in the range of a few ng l1 or even lower. Although some chalcophile elements may be highly enriched in rain and snow because of atmospheric contamination, some of these, such as Cd, Cu, Pb, or Sb, may be efficiently removed by soils, leading to very low concentrations being left in groundwater. Shotyk et al. (2010) used the element scandium to relate measured metal concentrations to natural elemental abundances: as there are effectively no industrial uses of Sc and therefore no anthropogenic emissions, and Sc behaves conservatively during chemical weathering in soil, this element can be used as a reference element for quantifying the extent of anthropogenic trace-element enrichment. Thus, it was found that chalcophile elements (e.g., As, Cd, Cu, Pb, Sb, and Zn), relative to their natural abundance in rocks, are highly enriched in snow with enrichment factors of 2–3 magnitudes. The removal of these elements from water is presumably due to processes such as retention by organic and mineral soil components; for Pb, these removal processes are so effective that apparently natural Pb/Sc ratios are found in groundwater. When drinking water is filled into bottles, from the latter, trace metals may be leached from the container walls, thus increasing the metal concentration in the water. Although some constituents in bottled waters may reflect their composition in the groundwater prior to packing, others exhibit contamination from the packing: polyethylene terephthalate (PET) plastic releases Sb, while glass releases Pb and Zn. While investigating 23 elements in 32 brands of bottled water from 28 countries, Krachler and Shotyk (2009) found trace-metal

33

levels of most bottled waters being below guideline levels, but some elements such as Li, Al, Be, Mn, or U exceeded the threshold limits for drinking water. In particular, coated aluminum and stainless-steel bottles are harmless with respect to leaching of trace metals into drinking water, but pocket flasks should be selected with great care to avoid contamination of beverages with harmful amounts of potentially toxic trace metals such as Sb. Representative data for trace-element composition of groundwater throughout the USA were retrieved from the EPA’s public domain Storage and Retreival (STORET) database by Newcomb and Rimstidt (2002) using robust data-analysis techniques described by Helsel (1990) and Helsel and Hirsch (1992). In Figure 1, their pertinent results with respect to elements discussed in this chapter are illustrated. The observed trace-element concentrations are approximately log-normally distributed. Median values for o90% of the censored sample populations range from 0.2 to 35 mg l1, minimum to maximum concentrations ranging as high as seven orders of magnitude for elements such as Zn. With respect to drinking-water quality, numerous guidelines have been published internationally, for example, in 1996 and 1998 by the WHO. When discussing about human health risk assessment with respect to drinking water, it has to be mentioned that distribution systems within homes (pipes, storage containers, etc.) can contribute significant amounts of metals such as Pb or Cu to the drinking water (Graziano et al., 1996).

3.02.2.2 Metal(loid) Species For evaluating behavior and toxicity of metal(loid)s in the environments, besides total-element concentrations, it is

1010 109

1)

108

1)

1) 1)

Range (ng l−1)

107

2)

106 105 104

4)

1)

1)

2,4) 5)

5)

4)

4)

4) 3)

3)

1)

4)

4)

1)

101 3)

100

2)

4) 2)

2) 4)

103 102

1)

2)

5)

3)

4)

5)

1)

5)

1)

1,4) 5)

4)

5)

2) 4)

1,5) 3)

1)

3)

3)

1 ng l−1

10−1 Antimony

Arsenic

Cadmium

Copper

Lead

Mercury

Zinc

Figure 1 Concentration ranges of elements in water, as discussed in this study (see text) (1) Newcomb and Rimstidt, 2002; 2) US EPA, 2002; 3) Shotyk and Krachler, 2009; 4) Frimmel, 1999; 5) Merian et al., 2004).

34

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

absolutely necessary to know about elemental speciation. Chemically speaking (according to International Union of Pure and Applied Chemistry (IUPAC)), as speciation is the process yielding evidence of the atomic or molecular form of an analyte (Irgolic, 1990), a metal(loid) species is a specific form of an element defined with regard to isotopic composition, electronic or oxidation state, complex or molecular structure, and phase (Templeton et al., 2000). These forms can be free metal ions or metal complexes dissolved in solution and sorbed on solid surfaces. In the light of this definition, a relevant example is also that of Cr speciation based on redox conditions, that is, the determination of the Cr(VI)/Cr(III) ratio (e.g., Dyk et al., 1990). To explain the strength of metal complexing, the concept of hard and soft acids and bases (HSAB) is useful (Pearson, 1973). In this concept, metal cations are Lewis acids and ligands are Lewis bases, with the metal cation and ligand in a complex acting as electron acceptor and donor, respectively. Hard acids (typical respective cations such as Co3þ, Cr3þ, Fe3þ, Mn2þ, or U4þ) are not discussed in this chapter, and also ((CH3)nAs(3n)þ) build up strong, chiefly ionic bonds with hard bases, whereas soft acids (such as Cd2þ, Cuþ, Hg2þ, Hgþ, and CH3Hgþ) and soft bases (e.g., alkylated arsenic) form strong, chiefly covalent bonds; Cu2þ, Fe2þ, Pb2þ, and Zn2þ or (CH3)nPb(4n)þ and (CH3)nSb(3n)þ are classified as borderline (meaning between hard and soft) acids (Langmuir, 1997). Soft and borderline metals generate bonds with soft ligands (including halogenated and organic species) of decreasing strength in the order Pb2þ 4 Cu2þ 4 Cd2þ 4 Zn2þ (Pickering, 1986). Arsenic species of the type HnAsOn3 4 and HnAsOn3 are classified as hard and borderline bases, 3 respectively. The HSAB concept can also be applied to toxicology: For example, the amino acids cysteine and methionine, present at active sites of some enzymes contain S functional groups representing soft ligands, and thus form strong covalent bonds with Hg, Cd, Cu, and Pb, resulting in breakdown of normal enzyme function (Manahan, 1994). With respect to inorganic speciation of natural waters for elements of interest in this chapter, forming typical cations such as Cd, Cu, and Zn (Cd2þ, Cuþ, Cu2þ, and Zn2þ), the most abundant binding partners are hydroxide, (hydrogen)carbonate, (hydrogen)sulfide, sulfate, and halogen (Cl, F, I, Br) ions, while for a typical oxoanion, such as As, just (mostly negatively charged) compounds with hydroxides and fluoride are important (Frimmel, 1999); note that the physical form of these species can be dissolved, colloidal, and particulate (crystalline and amorphous) covering a size range from macroscopic dimensions down to just a few nanometers (Merian, 1984). As a consequence, there exists the problem of the colloid and part of the particulate fraction being able to pass the conventionally used 0.45 mm filters; so these two fractions could really exceed the dissolved metal concentration by an order of magnitude of one or even more (Kennedy et al., 1974; Bergseth, 1983). While for Cd, Cu, and Zn, ions are the dominant inorganic species in water – followed by CdCO31, CuCO31, and ZnCO31 (Stumm and Morgan, 1996) – for Hg, Pb, and Sb, hydroxides are dominant (for Hg, HgCl21 is also dominant); and for and AsO3 are dominant. To find out the arsenic, AsO3 4 3 specific inorganic-element speciation of a certain natural water

sample, the first step is to obtain information about the master parameters Eh and pH of these waters, which differ significantly among ocean, rain/stream, bog, and groundwaters (Baas-Becking et al., 1960). Species stability under thermodynamic equilibrium can then be found in respective Eh–pHstability diagrams (Stumm and Morgan, 1996; Brookins, 1988; Drever, 1997); for example, at pH 8, the heavy metals Hg, Cu, Pb, and Zn should be chiefly present as complexes. While, for cadmium, copper, lead, and zinc, Cd2þ, Cu2þ, 2þ Pb , and Zn2þ are the most toxic forms, for inorganic arsenic, it is AsO3 3 . In inter-element comparisons, the toxicity (in decreasing sequence) for algae, flowering plants, fungi, and freshwater phytoplankton was found to be Hg 4 Cu 4 Cd 4 Zn (Sposito, 1989). As mentioned earlier, colloids need special attention because of their small size (nm-range). Separations at this scale can be performed by sedimentation field-flow fractionation (SdFFF) or flow field-flow fractionation (FlFFF) coupled to an ICP-MS (Schmitt et al., 2002; Lyven et al., 2003). For example, the latter authors used these instruments to determine the chemical composition of colloids from a freshwater sample, and found colloidal components of organic carbon (size 1–1.5 nm) and iron (size up to 5 nm) as most abundant. In another study (Schmitt et al., 2002) it was shown that the presence of natural organic matter (NOM) decreased the adsorption of metals onto clay particles (size 0.1–1 mm). The water-soluble fraction of NOM is the major form of organic matter (OM) in water, and is usually called dissolved organic matter (DOM) or dissolved organic carbon (DOC), and it exhibits a high capacity in binding heavy metals (Buffle, 1990). While complexes of trace elements with fulvic acid (FA) are water soluble (Frimmel, 1990), those with humic acid (HA) form colloids (Lund, 1990); metals may also be complexed by saccharides (Geraldes and Castro, 1990). After centrifugation, (5000 rpm) DOM in prefiltered samples (0.45 mm polycarbonate) can be characterized by sizeexclusion chromatography (SEC) with ultraviolet (UV) detection at 254 nm (Chen et al., 2006). Pokrovsky et al. (2006) filtered water samples from pristine boreal rivers (Central Siberia) in the field through progressively decreasing pore size (5 mm 4 220 nm 4 25 nm 4 10 kDa 4 1 kDa) using cascade frontal filtration technique, and found Cu and Zn associated to small organic complexes (size o1 to 10 kDa). Even though Fe-(hydr)oxide reduction is an important factor controlling metal mobility in aqueous systems, the dominant mechanism for this process seems to be organic matter (DOM respectively DOC) release (Grybos et al., 2009) because of negatively charged functional groups, organic matter have a high capacity for cation absorption (Laveuf and Cornu, 2009). While complexation with organic ligands was found to be important for Cu and Zn (Schro¨der et al., 2008; Du Laing et al., 2009; Koretsky et al., 2007), no or low correlation has also been reported for Zn and Cd (Beesley et al., 2010; Kalbitz and Wennrich, 1998). A special group of metal(loid) species are organometallic compounds characterized by a metal(loid)–carbon bond; these bonds are generally covalent and occur between soft acid metals and soft ligands. Organometallic compounds are formed by interactions of metal(loid)s with other chemicals and biota in the environment. As, Hg, Pb, and Sb can be

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

biotically/abiotically methylated/demethylated and can form stable methylated species in the environment. These processes are influenced by several environmental parameters such as salinity or the presence of sulfides/sulfates and DOC; while high temperatures and anoxic conditions often favor methylation, the opposite is the case for demethylation (i.e., low temperatures and oxic conditions). Maximum methylation rates typically occur at the redox boundary, which varies seasonally and frequently coincides with the sediment–water interface (Ullrich et al., 2001). In higher organisms, because of their volatility and amphiphilicity, methylated metal(loid)s show high mobility, possibly leading to critical effects in metabolism and toxicity; for several metal(loid)s (As, Bi, Se, and probably Te as well) in situ biomethylation in mammals had already been demonstrated (Dopp et al., 2004; Hirner and Rettenmeier, 2010). While previously methylation of As species was regarded as a detoxification mechanism, it is now clear that methylation of any metal(loid) actually increases its toxicity, the only possible detoxification process being volatilization (e.g., as hydride or peralkylated species, for instance, for mercury in the form of Hg1, CH3HgH, or (CH3)2Hg). Evidence for the possible occurrence of up to 17 metal(loid)s in the natural environment can be found in literature (Feldmann, 2003). While most of them have been identified in air, only 10 have been detected in water so far (Table 1). Environmental concentrations along with physical–chemical properties of these species such as thermodynamic and kinetic stability or stability against water together with mechanisms of biomethylation can be found in Craig (2003). While neutral, fully alkylated (i.e., peralkylated) species are relatively volatile and present in aqueous systems in extremely low concentrations, charged, partly alkylated species are found dissolved in water, and because of their amphiphilic properties they are enriched along the food chain. Higher alkylated compounds of As, Hg, Mn, Pb, and Sn are synthesized by the chemical industry, and have been mainly used as gasoline additives

Table 1 Metal(loid)

As Bi Cd Ge Hg Mn Mo Ni Pb Sb Se Sn Te Tl W

35

(alkylated Pb) and biocides (phenylated As/Hg and butyltin). Carbonyls of Mn, Mo, Ni, and W have not yet been reported in natural waters; there are also some indications that polonium may react with methylcobalamin to form Me2Po (Feldmann, 2003). Biomethylation potential could be demonstrated in laboratory experiments (also including sterile controls) for As, Bi, Hg, Sb, Se, Te, and is probably, but not yet validated for Cd, and Ge, but it still remains questionable for Pb, Sn, and Tl. With respect to the latter three elements, due to large anthropogenic emissions, it is actually impossible to trace low geogenic background levels: for example, for Pb, in natural waters, not only Me4Pb and Et4Pb (primary gasoline additives), but also all mixed forms MeEt3Pb, Me2Et2Pb, and Me3EtPb (secondary transformation/degradation products) can be detected in the pg l1 range; given enough time, stepwise dealkylation degrades these leadalkyls eventually into inorganic Pb (Yoshinaga, 2003). Of course, Co is also to be added to the list of metals exhibiting biomethylation potential, because it certainly can be methylated biologically as proven by the existence of the natural product methylcobalamin (MeCoB12). As mentioned earlier, biomethylation in the natural environment occurs at the water–sediment interface at the bottom of water bodies or within sediment pore water. With respect to potential hot spots emitting high concentrations of methylated metal(loid)s into the environment, contaminated sediments are most interesting. However, at highly polluted spots, bacterial life is impossible. Therefore, following the decreasing metal(loid) concentration with increasing distance from the contamination spot, bacterial populations increase along with the methylation rate; however, the latter decreases again when moving further because of metal(loid) concentrations becoming too low then. Altogether, the spatial distribution occurs as indicated in Figure 2, showing that maximum metal(loid) and methylmetal(loid) concentrations do not coincide (halo hypothesis).

Organometal(loid) species found in the environment Methylated species found In Air

In Water

X X X

X X X X X

X X X X X X X X X

X X X X X

X

Biomethylation (lab. exp.)

Higher alkyls

X X (X) (X) X

X

? X X ? X ?

X X

Carbonyls

X X X

X

X

X

Modified from Feldmann J (2003) Other organometallic compounds in the environment. In: Craig P (ed.) Organometallic Compounds in the Environment, 2nd edn., pp. 353–389. New York, NY: Wiley.

36

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

Metal hot spot

B A

Rel. amount

A

(a)

B C

C

Distance from hot spot

Metal(loid) concentration Organometal(loid) concentration Biological activity

Methylated species (b)

Figure 2 Spatial distribution of metal(loid) species at contaminated sites. (a) Metal-organic species distribution independent of metal concentration, and (b) halos of methylated species around metal hot-spot contaminations.

As a consequence of the described features, it should be borne in mind that due to their behavior in the environment, wherein organometallic forms have different characteristics from inorganic metal(loid) compounds, for risk assessment, the same general principles and approaches do not apply (Fairbrother et al., 2007). While in theory, speciation targets the determination of all chemical species present in the sample, in practice, only scientifically lower goals, resulting in operationally defined fractions, may be reached: this depends on the experimental (e.g., elution tests to simulate acid-rain leachable or plantavailable fractions) or measurement technique applied (e.g., methods determining size or redox state of the analyte). A relevant empirical method, which is often used, is sequential extraction where fractions according to element mobility are isolated. For example, Singh et al. (2005) received the mobile element fraction in sediments by applying the classical fractionation scheme of Tessier et al. (1979); sequential-extraction schemes have also be optimized for individual elements such as As (Oliveira et al., 2006; Hudson-Edwards et al., 2004; Keon et al., 2001; Wenzel et al., 2001). Perhaps, the simplest and most standardized sequential-extraction method was the one proposed by the European Community Bureau of Reference (BCR), which was improved upon in later studies (Rauret et al., 1999; Sahuquillo et al., 1999). As there exist severe relevant objections from analytical chemistry (Hirner, 1996, 2000; Sulkowski and Hirner, 2006), and since sediments are not the focus of this chapter, this speciation technique is not further discussed here. When addressing questions on metal(loid) bioaccessibility (compounds arriving at the organism membrane) and bioavailability (compounds having passed the organism membrane), information on the presence of the most important elemental species is mandatory. Here, the mechanistic-based approach of the biotic ligand model (BLM), which is designed to predict acute toxicity to aquatic organisms on the basis of physical–chemical factors affecting speciation (Di Toro et al., 2001), has to be mentioned; the BLM has been most extensively developed for copper (Santore et al., 2001). It was used by Balistrieri and Blank (2008) to compute the speciation of Cd, Cu, Pb, and Zn in aquatic systems, together with the

diffuse gradients in thin films (DGT) as another approach to evaluate speciation; the latter in situ method determines concentrations of labile metal by diffusion of ions through a hydrogel to a binding agent like Chelex-100 resin (Davison and Zhang, 1994).

3.02.3 Selected Elements 3.02.3.1 Arsenic and Antimony 3.02.3.1.1 Arsenic: Introduction and overview Arsenic (As) is found in virtually every part of the geosphere (see review by Matschullat (2000)). The element dissolves relatively well in water leading to As concentrations of 1–2 mg l1 in seawater and 1–100 mg l1 in freshwater systems (Francesconi, 2005), and in groundwater 0.5–0.9 mg l1 (Allard, 1995). Concerning the latter, mineral and thermal waters are enriched by up to and more than three magnitudes (Bissen and Frimmel, 2003a): for example, Landrum et al. (2009) claim to report the highest natural As (up to 44.9 mg l1) and Sb (up to 4.3 mg l1) concentrations ever found in natural surface water (El Tatio Geyser Field, Chile). The background concentration of dissolved As is mentioned in different sources to be 0.1 mg l1 for freshwaters, although this value is still under discussion (Matschullat, 2000). In the aqueous environment, the inorganic arsenic species, arsenite (As(III)) and arsenate (As(V)), are the most abundant species. The mobility of these compounds is influenced by pH, redox potential, and the presence of adsorbents such as oxides and hydroxides of Fe(III), Al(III), Mn(III/IV), humic substances, and clay minerals (Bissen and Frimmel, 2003a). In seawater, arsenate is the dominant dissolved species (average concentration 1.7 mg l1), taken up in the photic zone by phytoplankton and reduced to arsenite and (up to 10%) transformed to methyl- and dimethylarsenate (Matschullat, 2000). Based on research in Arctic ice cores, it can be seen that significant anthropogenic contamination of the environment with arsenic began nearly 3000 years ago: while Pb enrichments increased threefold above the natural background levels during Greek/Phoenician, Roman, and Medieval times, As is

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

elevated by a factor of 5 (Krachler et al., 2009). The most massive anthropogenic import of arsenic is associated with mining: for example, during the three centuries of gold mining in the Iron Quadrangle (Brazil), at least 390 kt of As was discharged into the drainage system, leading to up to 1960 and 60 mg l1 total and trivalent As, respectively, in mine effluents; as a consequence, in surface water in the vicinity of mining areas, up to 300 mg l1 As was found (Borba et al., 2003). Quite a different kind of environmental contamination by As was its use in biocides (e.g., Pb, Na, Ca, and Zn (methyl)arsenates) before the introduction of DDT in 1947 (banned in Germany since 1974). However, Na and Zn arsenates as wood preservatives are still used and sold (Bissen and Frimmel, 2003a). In the USA, about 40% of the total As consumption is in wood preservation (Peters et al., 1999). Arsenic is ranked first in the list of hazardous substances by US EPA (2002), and has been identified as a human carcinogen leading to skin, bladder, lung, and other kinds of cancer as well as having cardiovascular effects (ATSDR, 2005; Hughes, 2002). The main problem with this element is that arsenic ingestion associated with drinking water and diet cannot be avoided. While groundwater and surface waters used for drinking contain As usually in the low mg l1 range, in some regions, such as Southeast Asia, these concentrations often exceed the recommended international health standard for drinking water, of 10 mg l1, or even the local maximal permissible value of 50 mg l1 (WHO, 2001; Smedley and Kinniburgh, 2002; Mandal et al., 1996). It is to be noted, however, that when compared to other known or suspected human carcinogens generally set to risk levels of 1:106, for As, the correspondent risks are 1:104 (for 10 mg l1) to 1:400 (for 50 mg l1) (Morales et al., 2000); an earlier evaluation by US EPA proposed only 2 mg l1 as an acceptable 1:104 risk (NRC, 1999). However, it is to be mentioned that the kind of toxicological evaluation presented earlier has to be rated to be very approximate (if not to completely wrong) if As speciation has not been considered: thus, it is well known that arsenite can be up to two magnitudes more toxic for human cells when compared to arsenate (Petrick et al., 2000; Sakurai et al., 1998), making the problem of As in drinking water strongly dependent on the environmental redox conditions (see Section 3.02.3.1.2); as a consequence, arsenite oxidation is generally considered to represent a detoxification process. The known mechanisms for arsenite toxicity are, for example, its affinity to sulfhydryl groups leading to enzyme inactivation or interferences with DNA repair (Basu et al., 2001; Gebel, 2001). As(V) competes with phosphate in cell reactions, and can decouple oxidate phosphorylation (Squibb and Fowler, 1993). Moreover, methylated and thiolated As compounds exhibit particular toxicity, and are discussed later. In order to reduce As mobilization in the environment, redox potential should be high and pH not in the alkaline range (Bissen and Frimmel, 2003a). Under reducing conditions, As bound to Mn and Fe (hydr)oxides is mobilized because Fe(III) is reduced to Fe(II) and Mn(III/IV) to Mn(II); these reductions start at a redox potential of þ 200 mV under neutral and acidic conditions. As(III) and As(V) complexation may occur in natural waters in the presence of NOM, FA, and

37

HA (Thanabalasingam and Pickering, 1986; Xu et al., 1988; Redman et al., 2002). In general, besides the environment, organisms contain many organoarsenic compounds up to 4100 mg g1 wet mass (Francesconi, 2005). Among the latter are arsenosugars and lipids (e.g., in algae), arsenobetaine and -choline (e.g., in marine animals); all of these food-arsenic species seem to be of no significant toxicity. For an overview, the mentioned environmentally important arsenic species are compiled in Figure 3.

3.02.3.1.2 Arsenic in drinking water The most important natural sources of arsenic in the environment are volcanic emissions, geothermal fluids, and sulfide-rich mineralization. However, the most problematic challenge of current water research is dealing with elevated arsenic concentrations in drinking water (NRC, 1999; Smedley and Kinniburgh, 2002). Arsenic concentrations in groundwater can vary to a great extent; for example, in groundwater from active volcanic areas in Italy, As concentrations range from 0.1 to 7000 mg l1, and are highest where active hydrothermal circulation occurs at shallow depths (Aiuppa et al., 2003). Here, intermediate redox environments, where neither sulfides nor Fe hydroxides are stable, lead to maximal As mobility. As release can also be related to bacterial reduction of Fe(III) to Fe(II), for example, leading to high-As waters in the deepest aquifer in Murshidabad, West Bengal (Stueben et al., 2003). However, in West Bengal aquifers, the spatial distribution pattern of As is patchy, with areas containing groundwater that is high in As (4200 mg l1) found in close vicinity to low As (o50 g l1) groundwater; there is no relationship between high groundwater As concentration and high groundwater abstraction (Nath et al., 2008b). Fe-oxyhydroxide is considered to act as a potential sink for As, and organic matter controls microbially mediated redox reactions (Nath et al., 2008a). Arsenopyrite (FeAsS) being the most common As-bearing sulfide mineral, upon oxidation acids of As and S are released into drainage waters with high concentrations of dissolved As. Corkhill and Vaughan (2009) showed that the oxidation of arsenopyrite in acid is more rapid than in air, water, or in alkaline solutions. It could also be demonstrated that, when this oxidation is bacterially mediated, for example, by acidophilic Fe- and S-oxidizing bacteria, such as Acidithiobacillus ferrooxidans and Acidithiobacillus caldus, it is more extensive than abiotic oxidation. The resulting oxidized forms such as Fe arsenates can again be reduced by a variety of dissimilatory Fe-reducing bacteria such as Swanella sp. strain ANA-3, thus again releasing aqueous As(III) (Babechuk et al., 2009). Elevated groundwater As concentrations in a fractured silicate bedrock aquifer in central New Hampshire are related to the presence of pegmatites bordering granites and intruding metasedimentary rocks (Peters and Blum, 2003). As concentrations in the pegmatites averaged 9.6 mg kg1, which is much higher than that in the associated granites (0.24 mg kg1) and in the metasedimentary rocks (0.8 mg kg1); As concentrations in these pegmatites were due to the partial melting of calcareous metapelites and subsequent

38

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species Some main structures of environmental arsenicals Arsenic acids/methylated arsenic acids (main structures): OH

OH HO

H3C

As

O HO

As

OH

H3C

As

OH

CH3

O

O

As

CH3

CH3

As

CH3

CH3 Trimethylarsine oxide (TMAsO)

CH3

OH CH3

CH3 O Dimethylarsinoyl acetic acid

As

H3C

O

As

H 3C

CH3 Tetramethyl arsonium cation (TetraMAsC)

CH3 Trimethylarsonio propionate

CH3 As

As

Trimethylarsine (TMAs III)

OH Dimethylarsinic acid (DMAs V) CH3

O

As

CH3

As

OH Dimethylarsinous acid (DMAs III)

H3C

OH Monomethylarsonic acid (MMAs V)

O

CH3

H3C

O

OH Arsenic acid

CH3

As

OH Methylarsonous acid (MMAs III)

OH Arsenous acid

CH3

CH3

CH3 CH2

CH3 Arsenobetaine (AsB)

COO

CH3

As

CH2

CH2

OH

CH3 Arsenocholine (AsC)

Figure 3 Structures of important arsenic species in the environment. Modified from Craig PJ (2003) Organometallic Compounds in the Environment, 2nd ed., 415pp. New York, NY: Wiley; Feldmann J (2003) Other organometallic compounds in the environment. In: Craig P (ed.) Organometallic Compounds in the Environment, 2nd ed., pp. 353–389; and Raml R, Goessler W, and Francesconi KA (2006) Improved chromatographic separation of thio-arsenic compounds by reversed-phase high performance liquid chromatography-inductively coupled plasma mass spectrometry. Journal of Chromatography A 1128: 164–170.

recrystallization as granites with low As concentrations and pegmatites with high As concentrations. In some areas of Bangladesh and West Bengal, concentrations of As in groundwater out of the sedimentary aquifer of the Bengal Delta Plain exceed internationally and nationally set guideline concentrations (10–50 mg l1), eventually reaching levels in the mg l1 range (Nickson et al., 2000). The mobilized As is considered to have been derived from reductive dissolution of Fe oxyhydroxide covering sedimentary grains, releasing sorbed As. When observed more closely, As was found to have been mobilized only after orange Fe(III) oxyhydroxides were reduced to gray or black solid phases of Fe(II) or Fe(II/III). According to Horneman et al. (2004) and Van Geen et al. (2004), much of the As concentrated in relatively labile phases can be biotically mobilized perhaps

without the need for extensive Fe dissolution. However, since FeOOH is formed only in the secondary phase and sediments of the lower part of the Holocene aquifer (where most domestic wells are installed) contain As mostly fixed in biotite and organic matter, Seddique et al. (2008) argue for biotite being the primary source of As. The same authors interpret the patchy distribution seen of the As-enriched groundwater reflecting an uneven distribution of biotite-rich sediments, and the rate of chemical biotite weathering at the depths of the wells installed (20–50 m); this hypothesis is still under scientific discussion (Anawar and Mihaljevic, 2009; Seddique et al., 2009). Itai et al. (2010) suggest that the concentration of dissolved As is controlled by an adsorption–desorption equilibrium between sediment and groundwater.

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

39

Arsenosugar/riboses(main structures): Dimethylarsinoribosides O CH3

O

CH2

As

R: O

R

CH3

H3C

OH

OH

R:

R:

O

R: CH2

CH2

CH2

CH2 CH

SO3H

CH

O

OH

R:

O

CH

OH

OH

O

CH2

CH2

OSO3H

CH2

C

O

OH

CH

OH

OH R:

R:

R: OH

O

CH2

CH2 O

R:

SO3H

CH

CH2

C O

NH2

OH

OH

N

C

NH

NH2

OH

O

OH

N

OH

O

N

))

R:

N

R: O

O

OH P

O

O

OH P

O

OH

O

O

OH

O

OCO(CH2)nCH3

OH

OH

OCO(CH2)nCH3

Trimethylarsinoribosides CH3 CH3

As

O

CH2

R CH2

CH3

O OH

OH

R:

CH2 CH OH

CH2

CH2 OH

O R:

CH

OSO3H

OH

Figure 3 (Continued).

Deeper aquifers of the Holocene and the Pleistocene both exhibit significantly lower As contents (possibly because of the more available FeOOH as sorbent), and could possibly be used for domestic water supply as long as withdrawals do not exceed recharge rates (Zheng et al., 2004; Swartz et al., 2004); however, future effects on As mobility cannot be excluded (e.g., reductive FeOOH dissolution). In this respect, it should be borne in mind that 100 years of pumping deep groundwater for the public water supply of Hanoi (Vietnam) has

likely promoted hazardous As enrichment of the deep Pleistocene aquifer (Berg et al., 2008). When comparing the Bengal with the Huhhot basin (Inner Mongolia), As in both the basins appears to be mobilized by microbially mediated reductive dissolution of metal (hydr)oxides (Mukherjee et al., 2009); the sources of the As-bearing sediments are speculated to be in the Himalayas (Bengal) and the Da Qing mountain belt (Huhhot), respectively. In Holocene aquifers of the Red River flood plain

40

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

Arsenolipids:

CH 2OCOR' CH 3 O CH3

CHOCOR"

As O

P

O

CH 2

CH3 O

Phosphatidyl arsenocholine

CH 2OCOR'

CH3

As

CHOCOR"

O

O O

P

O

CH 2

O

CH3

Phosphatidyl dimethylarsenic acid

CH 2OCOR'

O

OH O

H 3C

As

CHOCOR"

O

O

O

P

CH 3

O

CH 2

O HO

OH

Phosphatidyl arsenosugar Figure 3 (Continued).

(Vietnam), the groundwater was found to contain up to 550 mg l1 As (Postma et al., 2007); as As correlates well with ammonia, the main mechanism for As mobilization is considered to be the reduction of iron oxide by decomposition of sedimentary organic matter. Quite a different role of the latter was hypothesized by Bauer and Blodau (2009), on observing As being completely sorbed and sedimented with FeOOH precipitates at neutral pH only in the absence of DOM. In the presence of the latter, however, the precipitation and sedimentation of FeOOH and associated As was impeded by the formation of aqueous Fe complexes and inhibition of Fe colloid growth. Therefore, the authors concluded that As mobility increased in the presence of DOM due to (1) competition between As and organic molecules for sorption sites on Fe particles; and (2) due to a higher amount of As bound to more abundant Fe colloids. Scanlon et al. (2009) described high groundwater As concentrations in semiarid oxidizing systems which were explained by As adsorption onto hydrous metal (Al, Fe, or Mn) oxides and subsequent mobilization with increased pH.

Similar relevant scenarios, that is, (1) the critical species being arsenate As(V) adsorbing strongly to mineral surfaces, and the more mobile and toxic being arsenite As(III) and (2) the occurrence of chemical and biochemical reduction of As(V) (Oremland et al., 2000), are often described in literature, for example, in the USA and Argentina (Peters and Burkert, 2008; Smedley et al., 2002). Generally, a more detailed discussion of As in the groundwater of sedimentary aquifers can be found in Bhattacharya et al. (2004), and with special emphasis on Cambodia and Vietnam in Polya et al. (2008). Of course, groundwater can also be contaminated with As by anthropogenic actions such as mining activity (e.g., Gemici et al., 2008). Moreover, at the abandoned As mine in Nishinomaki (Japan), discharged water from the mining and wastedump area is acidic and rich in As. However, the As concentration in the drainage has reduced to below the maximum contaminant level of 10 mg l1 without any artificial treatments, thus resembling natural attenuation by newly formed

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

schwertmannite (Fukushi et al., 2003). In another example, Fitzmaurice et al. (2009) examined the fate and transport of As at an industrial site where groundwater contamination leading to concentrations up to 1200 mg l1 was derived from the application of As2O3 as a herbicide. Currently, as many as 140 million people worldwide may have been exposed to drinking water with concentrations of As 410 mg l1, simpler and quicker analysis techniques rather than common instrumental methods in the chemical laboratory are needed. In this respect, a novel development, worth mentioning, is based on gold nanoparticles in a simple colorimetric and dynamic light-scattering assay, requiring only 10 min for analysis, and exhibiting detection limits of 3 ng l1 (Kalluri et al., 2009). Environmental technologists know that removal of As(III) is more difficult than the that of As(V). The reason for this is that oxidation in the presence of only air or even pure oxygen is a slow process, which may be accelerated by the presence of ozone, hypochlorite, chlorine (dioxide), H2O2, or oxidecoated sands (Bissen and Frimmel, 2003b). Respective removal techniques are coprecipitation with Fe(OH)3 and MnO2, fixed-bed filters with FeOOH, activated alumina, activated carbon or zeolites as adsorbents, anion exchange, and electrocoagulation as well as membrane filtration (ultra- and nanofiltration and reverse osmosis).

3.02.3.1.3 Methylated As species The basic mechanism of biomethylation can be described as follows: (1) As(V) species are reduced by glutathione (GSH); (2) the resulting As(III) can then accept a methyl group from S-adenosylmethione (SAM) to produce the methyl-arsenic(V) species in an oxidative–addition reaction (Cullen and Reimer, 1989; Challenger, 1951); and (3) the end products of repeated cycling are trimethylarsine oxide or trimethylarsine for fungi, tetramethylarsonium ion for clams, and dimethylarsinic acid for humans (Cullen and Reimer, 1989; Cullen et al., 1994). Previously, methylation of arsenic had been considered to be a detoxification process because pentavalent mono- and dimethylated As species (MMAs(V) and DMAs(V)) were found to be less toxic than inorganic As by more than one magnitude (NRC, 1999); 400–500 mg d1 can be excreted by humans via DMAs(V). However, studies have shown that trivalent organic analogs (MMAs(III) and DMAs(III)), which are also formed in As-methylation pathways (Challenger, 1945; Hayakawa et al., 2005), are as toxic or even more toxic than the inorganic species (Styblo et al., 1997; Lin et al., 1999; Petrick et al., 2000; Thomas et al., 2001; Dopp et al., 2005, 2006b, 2006c, 2008). MMAs(III) and DMAs(III) were shown to nick DNA at very low concentrations without chemical or enzymatic activation (Mass et al., 2001), and MMAs(III) was found to be a more potent inhibitor of thioredoxin reductase by two orders of magnitude when compared to arsenite (Lin et al., 1999). Furthermore, neurotoxic effects (blockade/enhancement of glutamate receptor responses and influence on synaptic transmission) can also be induced by methylated arsenic species (Kru¨ger et al., 2006, 2009). In groundwater impacted by methylated As pesticides, indications of the existence of the trivalent arsenic species ¨ ger and London (2008). (CH3)AsO2 2 were found by Wallschla

41

3.02.3.1.4 Thiolated As species With respect to their role in microbial redox transformations, arsenic behaves similar to the commonly coexisting, more abundant sulfur leading to interlinked environmental cycling of both elements. For example, experimental evidence shows that sulfur-oxidizing bacteria use free or arsenic-bound sulfur as a growth substrate, and directly or indirectly transform arsenite and thioarsenates to arsenate during growth (Fisher et al., 2008). In general, sulfides seem to facilitate complete As(V)/As(III) redox cycling by direct coupling of sulfide oxidation to arsenate reduction and by stimulating arsenite oxidation (Fisher et al., 2008). Thus, more than 50% of total As in sulfidic waters are As–S compounds (Stauder et al., 2005; Planer-Friedrich et al., 2007); the latter are found in alkaline (Mono Lake, CA) (Hollibaugh et al., 2005) as well as acidic waters (Yellowstone NP) (Langner et al., 2001). Abundant field and laboratory evidence demonstrates that in sulfidic waters, As thioanions replace arsenates and arsenites (Beak et al., 2008; Planer-Friedrich et al., 2007). Contrary to most published hypotheses that As thioanions are either As(III) or As(V) species (Beak et al., 2008; Bostick et al., 2005; Stauder et al., 2005), other authors (e.g. Suess et al., 2009) argue that both kinds of thioanions are likely, even in the same solutions. However, preservation of the latter species is very critical, and can be currently established best by flashfreezing of the samples (Planer-Friedrich et al., 2007), while simple acidification as a common procedure in water analysis leads to As loss by precipitation of As–S phases (Smieja and Wilkin, 2003; Planer-Friedrich and Wallschla¨ger, 2009; Suess et al., 2009). Moreover, numerous novel thioarsenic species have been found in the environment (Hansen et al., 2004; Schmeisser et al., 2004; Raml et al., 2005; Kahn et al., 2005; Nischwitz et al., 2006; Nischwitz and Pergantis, 2006), some of them even synthesized in the laboratory (Planer-Friedrich et al., 2007; Wallschla¨ger and London, 2008). In particular, in the geothermal waters of the Yellowstone National Park, besides arsenite/arsenate and mono-, di-, tri-, and tetrathioarsenate, methylated arsenic oxy- and thioanions were also detected (Planer-Friedrich et al., 2007). These thioarsenates occurred over a pH range of 2–9; while they dominated under alkaline conditions (up to 83% of total As), they were also found in acidic waters (up to 34%). Based on these observations, the authors suggested three separate reaction pathways: transformation of trithioarsenate to arsenite, stepwise ligand exchange from tri- via di- and monothioarsenate to arsenate, and oxidation of arsenite to arsenate (after thioarsenates disappeared). In groundwater impacted by methylated As pesticides, several thiolated methylspecies could also be identified; their lifetimes were up to 6 months before oxidation to pentavalent oxyspecies (Wallschla¨ger and London, 2008). However, not much is known yet regarding the toxicity of inorganic and organic thioarsenic species: while formation of thioarsenic species was found to reduce acute arsenic toxicity for the bacterium Vibrio fischeri (Rader et al., 2004), PlanerFriedrich et al. (2008) observed increasing toxicity with increasing number of thiol groups for this bacterium. While dimethyl species of arsenoacetate and arsenoethanol exhibited no toxicity against human hepatocarcinoma cells (Raml et al., 2005), in the same system, thio-DMAs(V) proved to be

42

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

10 times as toxic as DMAs(V) (Raml et al., 2007). Thio-DMAs were found to be more toxic than dithio-DMAs (Ochi et al., 2008), and in the human bladder EJ-1 cells, dimethylmonothioarsenate was more toxic than arsenite (Naranmandura et al., 2009).

3.02.3.1.5 Antimony Due to its low natural abundance, no special interest for the industry and the public, and because it has no known function in living organisms, antimony (Sb) has not been in the focus of environmental scientific research during the last decades. Together with arsenic, Sb first gained public attention in mid1990s when both elements were discussed with respect to being involved in sudden infant death syndrome (SIDS); however, there is no scientifically sound argument found in this regard, yet (Cullen, 2008). Average continental crust abundance of the element antimony is 0.2 ppm; nonpolluted filtered surface waters exhibit typical concentrations of less than 1 ppb (Filella et al., 2002b; Rudnick and Gao, 2003). Hydrothermal-volcanic ore deposits, however, may contain up to hundreds of ppm in the form of sulfide (stibnite Sb2S3), Cu–Ag–Pb–As sulfosalts, or in minerals such as galena, pyrite, arsenopyrite, chalcopyrite, or sphalerite; associated hydrothermal fluids are dominated by Sb–OH–Cl complexes, in particular Sb(OH)3 (Pokrovski et al., 2006). Wilson and Webster-Brown (2009) described the behavior of such geothermally derived Sb down the large lowland Waikato River system (New Zealand). Removal of Sb from the stream was observed to occur by its adsorption onto suspended particulate material enhanced at low pH (o5), and in the anoxic base of stratified lakes. In most natural systems, antimony mainly exists in the penta- and trivalent oxidation stages (in oxic systems Sb(III) mostly o10%), both species being subjected to strong hydrolysis in aqueous solutions by forming dominantly uncharged and negatively charged hydroxide compounds at low pH (Filella and May, 2003; Filella et al., 2009). Although this strong affinity of Sb to hydroxyl groups in solution significantly can limit Sb complexing with other inorganic and organic ligands, up to 85% of total Sb in lake and pore waters was found to be associated with NOM (Deng et al., 2001; Chen et al., 2003); moderate binding of trivalent Sb with humic compounds has been demonstrated experimentally (Buschmann and Sigg, 2004; Steely et al., 2007). In particular, calculations showed that up to 35% of dissolved trivalent Sb may be bound to HAs in the form of bidendate complexes in continental waters containing 5 mg l1 of DOC (Tella and Pokrovski, 2009). The latter authors could also demonstrate that stable complexes form between uncharged Sb(OH) and oxalic, citric, lactic acids, and catechol, whereas no complexing was detected with acetic, malonic, and adipic acids in the pH range of natural waters; trivalent Sb may also be bound to trace thiol-bearing moieties in HAs. More details regarding the environmental chemistry of antimony can be found in reviews by Filella et al. (2002a, 2002b, 2007, 2009). Due to the recent widespread use of antimony in industry (e.g., flame retardant, fireworks, pigments, ceramics, plastics, glass, brake linings, batteries, semiconductors, diodes, bactericides, polycondensation catalyst in PET production)

nowadays it is difficult to find anthropogenically not-influenced samples. Based on modern analytical equipment and clean laboratory conditions, on observing pristine groundwater of 34 samples from Springwater Township (Ontario, Canada), Sb concentrations of 2.271.2 ng l1 were found (maximum concentration 5 ng l1), reflecting reactions between percolating fluids and calcite/dolomite in depths of more than 100 m (Shotyk et al., 2005). The same water filled in polypropylene and PET bottles shows significantly higher concentrations (around 8 and 160 ng l1, respectively), providing evidence for Sb leaching from the containers (Shotyk et al., 2006). In another study, elevated Sb concentrations in bottled waters (even at or above the maximum allowable Sb concentration of 2 mg l1 for drinking water in Japan) are from the Sb2O3 used as catalyst in manufacturing PET (Shotyk and Krachler, 2007a). Dated ice cores (1842–1996) and a snow pit (1994–2004) in the Canadian High Arctic showed Sb concentrations ranging from 0.07 to 108 ng l1 (Krachler et al., 2005), while another core dated between 1300 and 10 590 BP, averaged 0.0870.03 ng l1 (Krachler et al., 2008), indicating that anthropogenic emissions have dominated throughout the entire period, amounting today to approximately 99.8% of the Sb deposited in the Arctic, thus clearly exceeding that of Pb (Krachler et al., 2005, 2008). Mono- to trimethylated Sb compounds along with the corresponding As species could be identified in geothermal waters from New Zealand (Hirner et al., 1998), and in pore waters of sediments (Duester et al., 2008); maximum observed methylation rates were about 1%. However, antimony speciation still remains a challenging task (Filella et al., 2009). Even fundamental questions such as the behavior of Sb(III)/Sb(V) during sample storage and preparation are not known systematically. Thus, currently, many key aspects of Sb environmental chemistry (e.g., biogeochemical cycles) and toxicology (e.g., ecotoxicology) remain largely unknown yet; first-cell experiments could demonstrate genotoxic effects caused by trimethylantimony dichloride (Dopp et al., 2006a).

3.02.3.2 Mercury 3.02.3.2.1 Introduction and overview Mercury is present in natural waters in very low concentrations: maximum contents may be estimated in oceans to be 1 ng l1, and in surface freshwaters 20 ng l1 (Merian et al., 2004). While the dominant forms of Hg in seawater are HgCl4 2 and HgCl 3 (along with some methylmercury chloride), in freshwater habitats, binding to humic substances is more common (Stumm and Morgan, 1996). Hg distribution in aquatic environments is characterized by high stability of compounds with sulfur and carbon, and a strong affinity to particles, colloids, and organic matter; only a small part of the element is in the completely dissolved form (Merian et al., 2004). Dimethyl mercury has been only reported in deep ocean waters, but it is lost by evaporation and photolytic degradation, and is not considered to be available for aquatic organisms. In environmental chemistry, mercury is considered a global-priority pollutant that is distributed around the earth via the atmosphere (Gustin et al., 2008; Jiang et al., 2006). In the atmosphere, the fairly unreactive elemental Hg is the

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

predominant form with an average residence time of about 1 year (Wiener et al., 2003). Thereafter, elemental Hg may be directly deposited onto soils and vegetation or it may be transformed into oxidized species, which are readily deposited and more bioavailable (especially for biomethylationgenerated methylmercury). Due to various reduction processes occurring in the soil and the water bodies, oxidized fractions may again be transformed back into volatile elemental Hg reemitted into the atmosphere. Mercury speciation is thus mostly focused on the fractions of elemental (Hg1) and monomethyl mercury (MeHg); when the sum of these two fractions (also named dissolved gaseous mercury (DGHg)) is subtracted from the total Hg, the resulting residual fraction should be mostly Hg2þ (also named reactive mercury (RHg)) in water (Horvat et al., 2003). Applying this fractionation scheme, for example, average concentrations for total, methyl, and elemental Hg measured in the water column of a remote lake (Big West Dam, Canada) were 5 ng l1, 96 and 20 pg l1, respectively (Ethier et al., 2008). As for other elements mentioned in Section 3.02.2.2, mercury can be biomethylated in the environment; therefore, dimethyl and monomethyl mercury are found in the atmoand hydrosphere, respectively (Dopp et al., 2004; Hirner et al., 1998). For example, hot springs (pH about 3.0) in Yellowstone National Park contain MeHg in microbial mats in concentrations ranging from 1 to 10 mg kg1 (King et al., 2006). While inside the mats, the methylation rate is 5–10%, in hot spring water it is o0.1% and thus comparable to other aquatic systems. The concentration of MeHg was 2–5 times higher in larval tissue than mat biomass, indicating that MeHg biomagnification occurred between primary producer and primary consumer trophic levels (Boyd et al., 2009). Particulate transport for Hg is more important in particlerich fresh and coastal waters than in the open sea. Particulate Hg consists of Hg bound to inorganic particles and particulate organic matter, as well as biogenic particles such as bacteria, algae, and phytoplankton (Ullrich et al., 2001). Inorganic Hg tends to bind more strongly to mineral particles and detrital organic matter, whereas MeHg is more strongly associated with biogenic particles. Therefore, oxyhydroxides and organic matter are among the main vectors controlling Hg mobility and transport in aquatic systems. In particular, due to the high stability of Hg–humic complexes, a high percentage of Hg in natural waters is present in organically complexed form, and Hg concentrations in lake or pore water are often significantly correlated to DOM. Anthropogenic enrichment to environmental Hg cycling is estimated to be a factor between 3 and 6 when compared to natural fluxes (Torky and Foth, 2007). In particular, environmental waters are significantly affected by technical applications of this element (e.g., as biocide in agriculture or antiseptic in medicine; Tchounwou et al., 2003); more critical industrial branches are, for example, chloroalkali electrolysis or paper production (Fitzgerald and Clarkson, 1991). Another globally important contribution is the use of elemental mercury in amalgamation techniques in gold mining, contaminating, for example, Amazon sediments by 130 t Hg yearly (Cleary, 1990); despite a few national bans, these techniques are increasingly used in countries such as China or Brazil (SRU, 2008). Generally, anthropogenic sediment pollution

43

has been demonstrated in many rivers, for example, up to 120 mg l1 elemental Hg and 130 mg l1 methylmercury have been reported for the Elbe River in Germany (Dopp et al., 2004). Compared to background values, Hg concentrations in forest soil were found to be enriched by factors ranging from 2 (Arctic) to over 4 (S Sweden) to 10 (Czech Republic) (Barrega˚rd, 2005). The respective critical concentration of 0.5 mg kg1 (Meili et al., 2003) is apparently exceeded by most countries in Middle Europe (SRU, 2008). Dated sediment cores from remote Californian lakes revealed that modern (1970–2004) lake sediment concentrations of Hg have increased by an average factor of 5 times more than historic (pre-1850) Hg concentrations (Sanders et al., 2008). Fitzgerald et al. (1998), however, estimated global atmospheric Hg deposition as inferred from lake sediments to have increased by a factor of only 2 relative to natural levels prior to the industrial revolution; but lakes in closer proximity to industrial point sources showed greater enrichments. A discussion of marine biogeochemical Hg cycling, biomagnification, and global flux models can be found in Fitzgerald et al. (2007). While biomonitoring of mussels and fishes in the Atlantic revealed no clear trends, relatively high Hg concentrations were found in inland lakes in Northern Europe explained by former Hg inputs via fungicides (Nixon et al., 2003). In Germany, apparently uncontaminated aquatic systems usually exhibit Hg concentrations o0.02 mg l1, and thus concentrations in drinking water above the limit of 1 mg l1 are rare (Bundesgesetzblatt, 2001). The present EU environmental quality standard (Water Framework Directive) for mercury and its compounds in water (after 0.45 mm filtration) is planned to be lowered from 50 to 15 ng l1.

3.02.3.2.2 Impact of mining, Minamata, and Florida Everglades When high concentrations of (inorganic) mercury are present in the environment (e.g., outcropping ore deposits, and mining districts), mercury speciation may be of special public interest. As the process of Hg recovery involves roasting (calcinations), the mine waste generated is referred to as calcine or mine-waste calcine (Gray et al., 2006). However, retorting of Hg-bearing ore is an inefficient and incomplete process; thus mine wastes often contain unconverted cinnabar, Hg1, ionic Hg compounds, and Hg chlorides and sulfates – and, of course, MeHg. In this respect, detailed studies centered around the Idrija Mine (Slovenia), the second largest Hg mine in the world (ceased operation 1995), are a good example (Hines et al., 2006; Covelli et al., 2008). Hg emissions from the mine (mainly in the form of particulate cinnabar) are transported by the Idrija and Isonzo River into the northern Adriatic Sea, 100 km away. At the Gulf of Trieste, sediments of the river mouth and in the bay are capable of producing MeHg (Hines et al., 2006). In addition, high MeHg concentrations (up to 22 mg kg1) were found in the adjacent Grado Lagoon (Covelli et al., 2008). It could be demonstrated that sediments of this lagoon release MeHg into overlying water until sulfide inhibition occurs and the methylation zone is restricted to the sediment surface (exactly the first few centimeters just below

44

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

the sediment–water interface). In the latter, MeHg in porewater ranged between 0.1 and 15% of the dissolved Hg, producing diffusive benthic MeHg fluxes similar to those of the Gulf of Trieste; however, the methylation rate seems to be higher in the lagoon. However, only a very small quantity of Hg is exported from abandoned Hg mines in SW Texas, primarily due to the arid climate and lack of precipitation resulting in no runoff in this region (Gray et al., 2006). The latter is not able to reach Rio Grande, the largest ecosystem in this region being conducive to Hg methylation. The minewaste material itself was found to contain up to 1.5 mg kg1 MeHg (among the highest concentrations reported); the latter correlated positively with Hg2þ, organic C, and total S. Conversion and transfer of MeHg from active and inactive Hg mines to surrounding ecosystems have also been reported for other cases, and often is a potential concern worldwide (Gray et al., 2004, 2003). Hg methylation is generally more efficient in humid climates where temperatures and precipitation are high, mine wastes are water saturated, and methylating bacteria are more active (Rytuba, 2000; Gray et al., 2003). Nonferrous metal smelters in China are another significant anthropogenic Hg source contributing to approximately 50% of Hg emissions to the atmosphere (Wu et al., 2006). Rivers, the estuary, and the bay along the NW Bohai Sea coast (NE China) have also been heavily contaminated by Hg due to long-term Zn-smelting activity. Hg concentrations in coastal sediments were 0.5–64 mg kg1 and in water, 39–2700 ng l1, up to three orders of magnitude higher when compared to background levels (Wang et al., 2009). These values are about one magnitude higher than those in the Hudson River, USA (Heyes et al., 2004), and MeHg in the water samples is also comparable to that in the Wanshan mining area (Qiu et al., 2005). Highest concentrations of MeHg in sediment and water reached 35 mg kg1 and 3 ng l1, respectively – the latter exceeding the Chinese drinking water guideline (1.0 ng l1); MeHg concentrations in both water and sediment pore water fairly good correlated with MeHg sediment concentration (r2 ¼ 0.74, po0.001). In freshwater hydrophytes, 5–100 mg kg1 Hg and 0.1–12 mg kg1 MeHg were found (Wang et al., 2009), indicating that MeHg accumulation in plants has a strong effect on food chains (Pickhardt and Fisher, 2007); also, humans and mollusks in and around Jinzhou Bay showed high Hg and MeHg accumulation (Liang et al., 2003; Fu et al., 1992). The most notorious MeHg poisoning case via the food chain occurred in the 1950s in Minamata Bay in Japan (Harada, 1995). As much as 380–455 t of Hg was used as a catalyst for acetaldehyde and vinyl compound production, and about 250 t was deposited in the Minamata Bay from 1932 to 1968; however, the main tragedy began in 1952 when Mn dioxide was replaced by Fe sulfide as a cocatalyzer. The industrial Hg waste was discharged into the bay sediments where it was biomethylated, releasing MeHg into the overlying waters. This was directly demonstrated by Tomiyasu et al. (2008) when investigating the Hg concentration in the water column: the concentration was highest near the sediment and decreased gradually. Up to 4 ng l1 MeHg was found in the turbid layer of water at the bottom, and the respective methylation rates were 24–54% (Tomiyasu et al., 2008). As a consequence, because of its lipophilicity, MeHg was enriched in

the aquatic food chain resulting in Hg loads in fish and shellfish being more than two orders of magnitude higher when compared to uncontaminated fish (Ullrich et al., 2001). Although high Hg concentrations were found in sediments from the Minamata Bay and its vicinity, the levels decreased gradually with distance from the bay arguing against significant movement of Hg out of the bay into the Northern Yatsushiro Sea; these sediments also have high natural and anthropogenic loads of Cd, Cu, Pb, and Zn, but distributed differently when compared to Hg (Nakata et al., 2008). Although as early as 1953 local inhabitants observed that cats were dying from eating cramps (dancing disease), the first cases of Minamata disease were reported not earlier than 1956. However, the evidence became overwhelming by 1960, that MeHg was indeed the cause of this disease (Grandjean and Choi, 2008), although the major case concerning compensation to Minamata disease patients was resolved not earlier than 2005! Characteristic of Minamata disease are special clinical signs and symptoms (impairment of speech and bilateral constriction of the visual fields) known in MeHg poisoning (previously known as Hunter–Russell triad), and eventually (in human autopsy cases) pathological changes in the nervous system, especially in the cerebral cortices, and peripheral sensory nerves (Eto, 2000). Following Minamata, in Japan, a second epidemic occurred in 1965 along the Agano River. By November 1999, 2953 cases of Minamata disease were identified and 1706 patients died, both catastrophes taken together (Eto, 2000). Sixty-four infant cases with cerebral palsy in villages where adult cases had occurred, were established as having congenital Minamata disease; the developing brains of the unborn had been affected by MeHg through transplacental exposure and even by breastfeeding (Kondo, 2000). Summarizing about MeHg-caused health effects as observed by Minamata tragedy, we learned that long-term exposure to this species (1) has a strong adverse impact on neurologic signs among residents in a local community (Yorifuji et al., 2008), (2) will increase the rate of hypertension in the affected population (Yorifuji et al., 2010), and (3) will generally lead to a greater variety of complaints in polluted when compared to nonpolluted areas (Futatsuka et al., 2000). Finally, and importantly, it should be mentioned that similar to Minamata, there are many coastal zones with industrial activities worldwide (examples follow). In addition, problems with MeHg poisoning among fish-eating populations are reported from elsewhere, for example, from the Amazon (Gochfeld, 2003). While Minamata is a first milestone in global mercury biomethylation research, the Florida Everglades is a second one. The latter representing a subtropical freshwater wetland ecosystem in the Everglades National Park (ENP) at the southern end. Freshwater flow is from north to south, transporting nutrients (N and P), sulfate (for soil amendment), and DOC from the northern Everglades Agricultural Area (EAA) in a decreasing gradient from the N to ENP. Via sulfate discharge from the EAA, the Everglades ecosystem faces the problem of Hg biomethylation, posing health risks to fish-eating birds, reptiles, and mammals (e.g., Florida panther), including humans; thus, at present, the Everglades still constitute the

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

largest continuous area in Florida where fish consumption is banned or limited because of mercury contamination. As total Hg concentrations in Everglades surface water (o10 ng l1) and soil (o500 mg kg1) are typically within background levels, biogeochemical controls make Hg available for bioaccumulation, rather than high Hg loading (Gilmour et al., 1998). The high MeHg production rate in Everglades soil (i.e., at the water–sediment interface) is related to elevated Hg levels in wildlife (Gilmour et al., 1998). Relatively high MeHg concentrations were also observed in floc and periphyton (Liu et al., 2008b); from the latter, MeHg can enter higher trophic level zooplankton and fish (Cleckner et al., 1999; Loftus, 2000). Due to the spatial variability in ecological conditions (e.g., organic matter in Everglades soil ranges from o1% up to 97%), in addition to MeHg production, biogeochemical controls make this species available to aquatic organisms (e.g., DOC, soil, and floc properties), and are important in Hg bioaccumulation (Liu et al., 2009). For the latter, the wet season is more favorable, and higher levels of Hg in mosquitofish, higher bioaccumulation, as well as biomagnification factors from periphyton to mosquitofish were observed compared to the dry season (Liu et al., 2008a). It could be shown that Hg level in mosquitofish is positively correlated with periphyton MeHg and DOC-normalized water; after uptake from water, Hg bioaccumulation occurs mainly through the food web (Liu et al., 2008b). Driven by concerns about deteriorating conditions affecting the Everglades ecosystem, in the early 1970s, the public, together with federal and state governments, initiated relevant projects to improve water quality, in particular, of the now hypereutrophic Lake Okeechobee (in the N) providing the headwaters for the Everglades (Perry, 2008). As a consequence, the Florida Everglades have been diminished by over 50% of their former extent (Perry, 2008), and dramatic declines in Hg levels in Everglades’ fish, birds, and alligators have been reported (Rumbold et al., 2002; Frederick et al., 2002). However, MeHg hot spots still remain: relevant risk assessment of MeHg exposure to three piscivorous wildlife species resulted in near 100% probability that these birds would experience exposures above the acceptable dose when foraging in northern ENP (Rumbold et al., 2008). Summarizing biota in ENP currently, they still contain the highest MeHg levels in southern Florida, being similar to or greater than other known MeHg hot spots in the USA. There is also no evidence of any significant Hg decline in Florida Bay fish still averaging over 1 mg kg1 Hg (Rumbold et al., 2008).

3.02.3.2.3 Essentials of biomethylation In water bodies, elemental Hg becomes water soluble by oxidation to (mono- and) divalent Hg ions which can be transformed into organic compounds with time. Here, the most relevant process is biomethylation by microorganisms just below the oxic/anoxic interface being often near the sediment surface in aquatic systems, leading to increase in Hg toxicity by bioaccumulation in crustaceans and fish (CHA, 2001; Clarkson, 1997); consequently, long-living big fishes and carnivores reach high enrichment factors (Davidson et al., 2004). Suitable conditions for effective biomethylation are the

45

presence of mobilizable elements as well as anaerobic environments such as wetlands or sewage/sludge/wastedepositing or (biological) treatment facilities; however, several (at least macroscopically) aerobic environments also show biomethylation potential (Dopp et al., 2004). It has also been suggested that inorganic Hg in fish tissues (liver) may be methylated endogeneously (Woshner et al., 2002). As a consequence of resembling the highest members of the aquatic food chain, humans are exposed to health risks (mainly neurotoxic effects) by ingestion of methylmercury. The mobile amphiphilic species methylmercury forms water-soluble complexes in blood plasma, and binds preferentially to sulfhydryl groups of peptides and proteins (Castoldi et al., 2001; CHA, 2001). When complexed to L-cysteine, because of its structural–chemical similarity to the essential neutral amino acid L-methionine (molecular mimicry), it can cross the blood/brain barrier via the amino acid transporter channel LAT1 (Kerper et al., 1992; Bridges and Zalups, 2005); in a similar manner, methylmercury can cross the placenta and thus reach the fetus. Therefore, methylmercury is a proven neurotoxic agent for humans with highest sensitivity when exposed prenatally. While in animal experiments, methylmercury chloride generates cancer, this is not known for humans; also, there is not enough evidence yet for genotoxic effects in humans (Torky and Foth, 2007). Se and Zn can retard the toxic effects of mercury and methylmercury by forming complexes with/without involvement of GSH; vitamins C, B, and E were also discussed in this respect (Torky and Foth, 2007). Eventually, intracellularly, methylmercury may be stored as a selenide (WHO, 1990). Finally, with respect to mercury species toxicity, it should be mentioned that Hg(II) salts are not known to be mutagens in bacterial systems, and are only weak mutagens in mammalian cells (Beyersmann and Hartwig, 1994; Hartwig, 1995). The environmental behavior of the toxic biomethylation product, MeHg, is also of great interest (i.e., coordination, geometry, and binding strength), and can meanwhile adequately be studied by X-ray fine structure methods (extended X-ray absorption fine structures (EXAFS)/X-ray absorption near edge structures (XANES); Yoon et al., 2005; Qian et al., 2002). Highly reduced organic S groups seem to be the most important binding partners for MeHg in soil and sediment. While at low loading, with MeHg, the Hg atom is associated with one S atom in the first coordination shell, at higher loading (contaminated soils/sediments) O (and/or N) atoms were found in the first coordination shell of Hg. Therefore, it is concluded, that after saturation of MeHg with thiol ligands, MeHg complexation by carboxyl ligands becomes significant; no sulfide/disulfide and polysulfide complexes were found yet. Humic sulfur ligands also bind Hg2þ (Xia et al., 1999; Hesterberg et al., 2001). In humic solutions, however, Amirbahman et al. (2002) observed only a small fraction of reduced sulfur complexing MeHg. Of course, because of the described dependency of the biomethylation process on environmental parameters, the resulting contamination of fish is not uniform, but it is dependent on local factors: for example, while in freshwater fish from St. Clair Lake, (Canada) 1935 mean Hg concentrations of 0.07 mg kg1 and 1970 higher values of 0.2 mg kg1 were detected, for seawater tuna 1970 mean Hg concentrations

46

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

of 0.13 and 0.25 mg kg1 were reported (Dales et al., 1971; Hammond, 1971). Generally, seawater fish from northern seas is more highly contaminated than freshwater fish from great inland lakes; fishes from the Mediterranean Sea have higher contamination than those from the Atlantic; long-living species such as shark or tuna are more highly contaminated than short-living ones such as salmon; and for big tuna fish in certain areas, Hg concentrations of up to 10 mg kg1 have been reported (Torky and Foth, 2007) – this would mean that according to the above-cited German drinking-water guidelines, focusing on 1 mg l1 corresponding to 2 mg intake daily, only 0.2 g of this fish will be allowed per day (or 180 g yr1)! The average real daily methylmercury dose for individuals living in the Northern Hemisphere is estimated roughly to range between 0.2 and 14 mg (NRC, 2000). From the conditions described, it is important to know not only the amount of mercury present and available for methylation, but also more about the master variables deciding about the effectiveness of the biomethylation process, that is, those environmental parameters that have the greatest influence on the methylation rate (concentration ratio of methylated to total Hg), also known as mercury methylation rate (MMR), which in nature may range from just below a few 0/00 (biogeochemical background) to a high percentage range (e.g., in the Florida Everglades), up to about 30% in freshwater lakes and rivers, and 37% in the anoxic bottom waters of a stratified pristine lake (Ullrich et al., 2001); up to 65% has been found by Potgeter (1998), and in pore water, the proportion of MeHg can reach up to 85% (Merian et al., 2004). While methylation rates in riverine and coastal surface waters are 1–5%, ocean waters and ocean rain show about 1% or less (Mason et al., 1998). Biomethylation rates appear to increase under anaerobic conditions, high temperatures, and low pH, and are usually lower in marine compared to freshwater environments because of salinity effects and sulfide/chloride interferences (Merian et al., 2004). Although high MMRs produced by biofilms at liquid/ solid interfaces may appear impressive (and are decisive in environmental health evaluations), from a global geochemical point of view, water column methylation may be potentially more important, simply because the volume of water is much larger than the volume of surficial sediments. Furthermore, when attempting to establish a global MeHg cycle, other potent sources, such as the root zones of floating macrophytes in tropical systems, also have to be considered (Ullrich et al., 2001). While not being the only possible methylators (Pak and Bartha, 1998; Warner et al., 2003; Kerin et al., 2006), sulfatereducing bacteria (SRB; populations such as Desulfovibrio, Desulfobacteriaceae, or Desulfobacter) are the main acting microorganisms methylating Hg (methanogens playing only a minor role); thus their growth conditions (e.g., nutrient availability) represent an important biomethylation parameter (Compeau and Bartha, 1985; King et al., 1999, 2000). The ability to methylate Hg is not confined to one phylogenetic group of SRB, but is scattered throughout the phylogenetic tree of sulfate-reducing eubacteria. Different organisms clearly have different rates of Hg methylation, even among the SRB, and not all SRB methylate Hg (Benoit et al., 2003). A small number of Fe-reducing bacteria, that are phylogenetically

similar to methylating SRB, have been shown capable of methylating Hg in pure culture. However, bacteria can also degrade MeHg via oxidation or reduction pathways; for example, reduction of MeHg to Hg1 is achieved by microorganisms using the mer A and/or mer B genes (Marvin-DiPasquale and Oremland, 1998). Therefore, note that reporting biomethylation data from natural systems is actually related to the difference between methylation and demethylation (i.e., net or gross methylation rates). Another key parameter strongly affecting MeHg concentrations is the presence of organic material. Generally, DOM interacts very strongly with Hg (binding to reduced S), affecting its speciation, solubility, mobility, and toxicity in the aquatic environment; DOM competes with sulfide for Hg binding (Ravichandran, 2004). However, complexation with DOC generally limits the amount of inorganic Hg available for uptake by methylating bacteria, because DOC molecules are too large to cross bacterial cell membranes. At low pH, DOC is less negatively charged, and therefore less likely to complex Hg, making it more available to methylating bacteria. In sulfate-limiting environments, where microbes may utilize organic matter as an energy source, DOC may have a stimulating effect on microbial growth and thus enhance methylation rates in the water column. Maximal methylation is often observed in surface sediments, where microbial activity is greatest due to the input of fresh organic matter. As a result, systems with high levels of organic-matter production, such as wetlands, recently flooded reservoirs, or periodically flooded river plains, may exhibit extremely high rates of MeHg production. In addition, estuaries define transitional environments for active biogeochemical transformations such as biomethylation. Chemical and physical gradients characterizing these continent–ocean interfaces affect the cycles of many metals (Mota et al., 2005). The creation of new reservoirs and enlargement of lakes significantly increase MeHg production, leading to elevated Hg concentrations in fish that remain high for several decades; flooding may provide large amounts of organic matter and nutrients, thereby stimulating microbial methylation activity (reviewed by Ullrich et al. (2001)). Kelly et al. (1997) found that MeHg production increased by almost 40 times following the experimental flooding of a boreal forest wetland. Elevated Hg concentrations in fish (Big Dam West, Canada) were explained by Ethier et al. (2008) by low pH and high DOC concentrations in lake water reflecting the poor buffering capacity of peat. Artificial reservoirs such as Petit-Saut reservoir (French Guiana) can be seen as large manmade reactors exhibiting extensively altered Hg speciation in favor of methylating species: MeHg constituted 8%, 40%, and 18% of the total Hg in the dissolved phase, the particulate suspended matter, and in the unfiltered samples, respectively (Muresan et al., 2008b). It is known for long that high lake DOC concentrations and/or low pH (Hg2S more bioavailable for methylation by SRBs, via diffusive uptake) promote methylation within sediments (Bjornberg et al., 1988; Benoit et al., 2001). Typically, in coastal and marine sediments, organic matter has been shown to be the main factor regulating MeHg concentrations (Hammerschmidt et al., 2004; Lambertsson and Nilsson, 2006). Therefore, in areas dominated by wetlands, organic

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

3.02.3.2.4 Biomethylation within parameter gradients Essentially, the complex interplay of the individual parameters or environmental factors affecting mercury biomethylation (potential master variables: microbial activity; concentrations of bioavailable Hg, organic matter, and sulfide; and redox potential, pH, salinity, and temperature), described in the last section decides about the intensity of biomethylation in a given scenario (Ullrich et al., 2001). Although it often may appear that enhanced rates of MeHg production are linked in particular with low pH, low salinity, and the presence of decomposable organic matter in reducing environments (Ullrich et al., 2001), it is not possible for any real scenario to follow up the kinetics of relevant biomethylation processes with all their complex synergistic and antagonistic variables in detail (e.g., Merritt and Amirbahman, 2009). Therefore, it is interesting to observe just a few or even one gradient parameter; in this chapter, we focus on the redox potential and the presence of inorganic sulfur (acid volatile sulfide (AVS)/sulfate). Muresan et al. (2008a) stated that it is not only the redox potential that is important, but also important is the existence of redox transition zones. As a striking relevant example, these authors studied Hg speciation along the redox gradient of an anthropogenically perturbed tropical estuary, the 70-km long Sinnamary Estuary in French Guiana, characterized by an anoxic freshwater (hypolimnetic discharges from the artificial reservoir of Petit-Saut) and a saline end-member (Amazon Plume). The upper part of the estuary producing 0.5–3.5 kg MeHg per year, exceeding the amount of Hg volatilized into the atmosphere by one order of magnitude (Muresan et al., 2008a). Maximal MMRs were observed for dissolved Hg in slightly acidic and sulfidic milieus, whereas for particulate Hg, this was the case for the most acidic and oxidized waters. Trapping of dissolved MeHg (e.g., by precipitation of Fe oxyhydroxides in the lower estuary) and mobilization of particulate MeHg (e.g., accompanying AVSs dissolution in the vicinity of the reservoir) constitute dynamic processes simultaneously occurring in the Sinnamary Estuary (Muresan et al., 2008a). In part, similar observations have already been described earlier (Bloom et al., 1999; Gill et al., 1999): MeHg mobility in estuary surface sediments was found to be linked to the Fe redox cycle, while Hg mobility was controlled by the formation of soluble polysulfide and organic complexes. Under anoxic conditions, oxyhydroxides dissolve and release any associated Hg, which could be one reason for the frequently observed Hg and MeHg enrichment in anoxic waters; seasonal and diurnal trends in MeHg concentrations in sediment pore waters may also be linked with redox effects (Ullrich et al., 2001). Taking into account both the redox potential and the presence of AVS, Ouddane et al. (2008) compared Hg

methylation rates in highly industrialized salt marsh/mudflat systems from two macrotidal estuaries: the Seine (France) and the Medway (UK); Hg concentrations were higher than uncontaminated sediments by factors up to 50 (background concentrations about 0.03 mg kg1). While Medway mudflat is characterized by stable anoxic redox conditions (about  200 mV), Seine mudflat is more oxidized (about þ 100 mV); consequently, MeHg concentrations of Medway samples were fourfold higher than those of Seine samples in spite of similar total Hg concentrations. MeHg variability was associated with the activity of SRBs and the presence of AVSs. A strong correlation was observed between MeHg and AVSs in sediments from these mudflats, possibly as consequence of the common origin of AVSs and MeHg, both being produced by microorganism activity. For estuarine environments, it was also reported that sulfide enrichment varies with organic-matter enrichment resulting in increased Hg methylation at higher sulfide concentration (Sunderland et al., 2006). As SRB are thought to be the key methylating agents (Jensen and Jernelov, 1969; King et al., 1999; Benoit et al., 2003), sulfur geochemistry plays a dominant role in Hg methylation, contrary to normally methanogen-induced biomethylation as already described in Section 3.02.2.2. In particular, while sulfate stimulating both sulfate reduction rate (SRR) and MMRs at low sulfate concentrations, the buildup of dissolved sulfide at high sulfate concentrations inhibits Hg methylation (Figure 4). Figure 4 is based on literature descriptions and data (Gilmour and Henry, 1991; Ullrich et al., 2001), and in particular on the hypothesis offered by Benoit et al. (2003) implying that inorganic Hg uptake by SRB occurs by passive diffusion of neutral Hg complexes (in anoxic waters such as HgS1, Hg(SH)21, polysulfides HgSn1, or Hg(SH)(OH)1) across the cell membrane. While this happens at low sulfide levels, at high sulfide concentrations, methylation is inhibited because of the formation of charged disulfide complexes which are likely to be less bioavailable. The hypothesis is based on field (especially in the Everglades) and laboratory studies demonstrating that the balance between sulfate availability (controlling SRB activity) and sulfide production and accumulation (controlling Hg bioavailability) are the main factors for Hg biomethylation (Benoit et al., 2003). Sulfide inhibition SRR MMR

MMR/SRR

soils, and humic surface waters, high concentrations of MeHg are formed and accumulate in higher biota (Hurley et al., 1995; St. Louis et al., 1996). This has led authorities to discourage people to regularly eat fish from the 40 000 lakes in Sweden (Hakansson, 1996). Coastal wetlands along the Gulf Coast are key sites for MeHg production and may be a principal source of MeHg to foodwebs in the Gulf of Mexico (Hall et al., 2008).

47

SRBmethanogens competition 0

10

20

30

40

50

60

[SO42−] mg l−1 Figure 4 Mercury methylation rate (MMR) as a function of sulfate supply.

48

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

Correlations between SRR and MMR have often been reported in literature (reviewed in Merritt and Amirbahman (2009)), for example, in sediment assays or core-depth profiles (MartinDoimeadios et al., 2004; Choi and Bartha, 1994; King et al., 1999). From Figure 4, it can be understood that anthropogenic sulfate from the EAA can be transported with the water flow to the south toward ENP, leading to methylation rates there rising up to the high-percentage range. Field studies in the Florida Everglades covering a large gradient in sulfate and sulfide showed that the highest MMRs were at sites of intermediate SRR and sulfide concentrations. Sites with highest MMRs are those that also have the highest fish MeHg concentrations, confirming the direct link between the extent of Hg methylation and fish MeHg levels. Although the results of the investigation of Hg methylation in a tidal estuary in French Guiana (Muresan et al., 2008a) also conformed to the neutral complex theory of Benoit et al. (2003), in other scenarios, the application of this model in its present form may be more critical (Merritt and Amirbahman, 2009; Goulet et al., 2007).

3.02.3.2.5 Abiotic methylation Mercury methylation by organisms may be enzymatic or nonenzymatic. While enzymatic methylation requires the presence of actively metabolizing organisms, nonenzymatic methylation requires only the methylated products of active metabolism; thus metabolically produced methylcobalamin (resembling the most likely environmental methyl donor) can spontaneously methylate ionic Hg in aqueous solution (Ullrich et al., 2001). However, it should also be mentioned that purely abiotic pathways can methylate inorganic mercury in the environment, when suitable chemicals such as watersoluble methylsilicon compounds or humic substances are present, or methyl groups can be transferred from other Hg, Pb, Sn, or As alkyl species. Relevant hot spots may be humusrich soils, possibly because humic substances are good methyl donors for Hg2þ (Weber, 1993). DOC in water, however, seems to suppress Hg methylation (Miskimmin et al., 1992). Both Hg2þ and MeHg are known to have strong affinities for organic matter in terrestrial and aquatic environments (Hintelmann et al., 1997; Wallschla¨ger et al., 1998) – according to Schlu¨ter (1997), Hg2þ even more than MeHg. Another pathway leading to abiotic methylation is photoproduction of MeHg: Siciliano et al. (2005) observed in aerobic surface water of lakes increasing MeHg concentrations during sunshine. This generation of MeHg was also dependent on DOM concentrations and type (o5 kDa or 30–300 kDa). Furthermore, especially in coastal waters, it appeared that Hg oxidation and reduction were photochemically mediated also affecting Hg1 and DGM cycling (Whalin et al., 2007; Zhang and Dill, 2008). However, Benoit et al. (2003) came to the conclusion that abiotic transformations (in particular, methylation/demethylation) are possible but not important in most ecosystems. Last but not the least, it should be mentioned in this respect, that research in biotic/abiotic methylation/demethylation of Hg can be successfully performed since the application of isotopic spiking techniques is possible (e.g., Martin-Doimeadios et al., 2004).

3.02.3.2.6 Global concern Summarizing, although mercury emissions via rivers and air into the Atlantic Ocean and the North Sea have decreased between 1987 and 1995 by 50%, the mercury already present in the environment is because of the extreme mobility of this element (especially, its methylated variety generated in ecosystems) actually posing increasing troubles for the only chance to export Hg from the cycling mercury-pool within a 10–15-year cycle time (SRU, 2008) by precipitation as insoluble mineral (HgS) and keeping it under oxygen-free conditions (Torky and Foth, 2007). In February 2007, at the 24th meeting in Nairobi, the councilors of the United Nations Environment Programme (UNEP) passed the resolution that serious efforts have to be undertaken globally to reduce the mercury load upon our environment. Moreover, the EU expressed it will to contribute substantially to reduce the global mercury load, for example, by avoidance of Hg-dependent technologies (SRU, 2008).

3.02.3.3 Other Metals (Cd, Cu, Pb, and Zn) In contrast to the already discussed As, Sb, and Hg, these four metals are of much higher interest in the metal industry for manufacturing of alloys and other special applications; for example, Cd (in plating (Cd-coated steel), batteries, or as stabilizer in polyvinyl chloride (PVC)), Cu (in electroplating, pigments, catalysts, and biocides), or Pb (batteries, paints, or alkyllead). Thus, in accordance with economic growth, mining and processing of these metals have markedly increased during the last decades. However, from a chemical point of view, compared to the elements discussed before, not much is known concerning the speciation of these elements, just information about ordinary inorganic (complexes with major anions such as hydroxide, carbonate, sulfide, and sulfate) and organic ligands (complexes by humic materials) can be found (Merian et al., 2004). In contrast to all the elements discussed in this chapter, only Zn is not regarded as toxic. Despite this fact, the WHO has set guideline levels of 5 mg l1 for this element in drinking water, but mainly because the aesthetic quality of drinking water will be impaired at higher concentrations; such cases, however, are not often found as Zn levels in drinking water are usually below 0.2 mg l1 (Merian et al., 2004). The history of anthropogenic use of heavy metals can be best illustrated by taking lead as an example: in dated ice cores from Devon Island Ice Cap (Canada), Pb concentrations were proportional to those of Sc (as a geogenic reference) until 3100 BP, consistent with the hypothesis that soil dust particles derived from physical and chemical weathering dominate the Pb inputs to the atmosphere, with the magnitude of these sources being climate dependent (Zheng et al., 2007). With respect to estimating the Pb content of pristine groundwater, Shotyk and Krachler (2007b) reported a mean concentration of 5.1 ng l1 for six artesian flows in southern Ontario (Canada), Hirao and Patterson (1974) reported a mean concentration of 15 ng l1 for remote stream waters of the Sierra Nevada watershed, and Field and Sherrell (2003) found in 0.3–8 ng l1 water samples from Lake Superior. Notably, filling such waters in glass bottles will increase the Pb concentrations by factors ranging from 26 to 57 (Shotyk and

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

Krachler, 2007b); however, the maximum allowable concentration of 10 mg l1 in drinking water established by the EU, Health Canada, and the WHO, will usually not be attained by this effect. For Cd, mean concentrations in natural waters are reported to range from 0.01 to 0.1 mg l1 for seawater and 3 mg l1 for freshwater, respectively; Antarctic ice shows 0.3 mg l1, and surface ocean waters o5 ng l1 (Merian et al., 2004). Although there exist natural low-pH environments unaffected by human activity (Eppinger and Fuge, 2009), acid waters are often anthropogenically generated, in particular, when draining mining areas are heavily affected by runoff from mining operations. These waters are typically acid mine drainage (AMD), eroded material from mine-tailing deposits, and waste from ore-processing operations (Salomons, 1995). Acidity and high metal concentrations of AMD are the result of pyrite oxidation during weathering or industrial processes (Banks et al., 1997). At pHo4.6, pyrite oxidation progresses slowly, and direct bacterial sulfide oxidation (e.g., by Thiobacillus thiooxidans) is the dominating process (Kirby and Elder Brady, 1998). A good example is the Iberian Pyrite Belt, an important metal-rich sulfide deposit (Braungardt et al., 2003). The rivers Rio Tinto and Rio Odiel draining this area are highly acidic (pH 2.2–3.6) and show milli-molar sulfate and iron concentrations as well as micro-molar Cu and Zn contents. Dissolved metal concentrations are at a maximum during autumn and early winter (e.g., Rio Tinto 29–54 mg l1 Cu, while 8–11 mg l1 Cu during the residual year). This variability is interpreted by the production of AMD during periods of enhanced microbial activity at higher temperatures in summer, and a subsequent runoff into the rivers with the first rain in autumn (Braungardt et al., 2003). Oxidation of sulfide-rich rocks in leftover debris from Cu mining in the 1920s contributed to metal contamination of local coastal environments in Prince William Sound (Alaska), leading to inverse correlation between pH and metal concentration in pore waters of sediments showing pH down to 3 and base metal concentrations up to 25 mg l1 (Koski et al., 2008). Adverse heavy metal impacts were also reported for the aquatic ecosystem of the Lean River in S China (Liu et al., 2003): while acidic drainage from Dexing Cu mine contains large amounts of Cu, high concentrations of Pb and Zn are found in the effluents released from many smelters and mining/panning activities. Astro¨m (2001) reports about the medium-sized stream Munsala (Finland) draining areas covered with acid sulfate soils developed on sulfide-bearing marine sediments; during high flows in autumn, while there was a strong downstream increase in Cd, Cu, and Zn concentrations related to extensive acid soil leaching, other elements such as As, Sb, and Pb were not leached as much from other soils/sediments within the catchment. Note that in the headwater of Munsala stream, trace-element concentrations were similar to local background concentrations. Elevated metal concentrations (average values given in mg kg1) of 8.6 (Cd), 481 (Cu), 4450 (Pb), and 753 (Zn) were found in waste tailings of Daduk mine, Korea (Lee et al., 2001), and similar ones of 9.4 (Cd), 229 (Cu), 6160 (Pb), and 1640 (Zn) in Imcheon mine, Korea (Jung, 2001). Water samples from the latter site were very acidic (pH down to 2.2) and contained up to 0.3 (Cd), 1.9 (Cu), 2.8 (Pb), and 53 (Zn)

49

(average values given in mg l1); concentrations of metals (and anions such as sulfate or fluoride) decreased exponentially with increasing distance from the mine. In favorable circumstances, the buffer capacity of the natural environment neutralizes the AMD (natural attenuation): For example, while at the Pecos mine (New Mexico), drainage near the waste rock pile is acidic (pH between 3 and 5) and carries high loads of Cu, Pb, and Zn, because of reacting with limestone-containing bedrock, drainage flow downstream toward the Pecos river shows increasing pH and decreasing metal contents (Berger et al., 2000). Lower concentrations (ranges given in mg l1) were reported for surface water of Udden pit lake (associated with open pit mining) in N Sweden (Ramstedt et al., 2003): Cd 0.02–0.08, Cu 0.03–0.17, Pb 0.006–0.06, and Zn 2–81. However, even these concentrations are significantly higher than background concentrations and guideline concentrations for lake and surface water in Sweden (Ramstedt et al., 2003), and are explained by weathering of ore sulfide minerals. Comparison with Hall Lake shows similar behavior for Cu, As, and Zn (Balistrieri et al., 1994); Hall lake is a natural lake and most metals in this lake precipitate below the anoxic border where sulfide is generated. However, compared to other pit lakes as described by Miller et al. (1996), the water quality of Udden pit lake is relatively good in terms of pH and heavy metal content; note that the latter two parameters correlate in that pit lakes with higher pH tend to have lower metal concentrations. Mine wastes may be discharged and dispersed into nearby agricultural soils, food crops, and stream sediments, eventually posing a potential health risk to residents in the vicinity of mining areas. For example, enriched concentrations of heavy metals were found in various plants grown in the vicinity of Daduk mine in Korea, and were correlated to those in soils (Lee et al., 2001); in particular, relatively high concentrations were detected in rice leaves and stalks grown under oxidizing rather than reducing conditions. In Mexico, large quantities of waste are derived from mining and abandoned mines pollute the aqueous systems (Armienta et al., 2003). For example, for the mine-impacted tropical river, Taxco, it could be shown that in case of Pb pollution, anthropogenic sources have contributed significantly, while natural sources contributed only small amounts (Arcega-Cabrera et al., 2009). Bioavailable Pb in riverbed sediments was 450% in 80% of the sampling stations indicating a high potential environmental risk. Highest Pb concentrations were found close to tailings during the rainy and post-rainy seasons; during dry and post-rainy seasons, Pb chemistry was mainly controlled by organic matter and carbonate content. Due to recent advances in mass spectrometry, isotope scientists are now able to precisely determine stable isotope variations even in metallic elements (Bullen and Eisenhauer, 2009); with respect to the elements covered in this chapter, already a few relevant application studies exist in literature: Peel et al. (2009) determined the ratio 66Zn:64Zn in settling particles in the hypolimnion of the eutropic Lake Greifen (Switzerland). Enrichment of the light isotope was observed during the productive summer period from June to September, when the Zn in the settling particles was predominantly

50

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

associated with organic material, and when Zn concentration in the epilimnion was the lowest. By investigating stream waters draining historical mining districts in USA and Europe, Borrok et al. (2008) concluded that besides Zn, Cu isotopes also may be powerful tools for probing biogeochemical processes in surface waters.

3.02.3.4 Platinum Group Elements The PGEs (Ru, Rh, Pd, Os, Ir, and Pt) are found in the earth’s crust in very low concentrations of about 0.05–0.1 ng g1 (Wedepohl, 1995). These elements occur as sulfides (e.g., RuS2, OsS2, and IrAsS) or alloys (Os–Ir–Ru and Pt–Fe) in the form of micrometer-sized mineral inclusions or millimetersized nuggets in placer deposits associated with mafic/ultramafic complexes; while the former may be formed at 700– 1100 1C, the latter are considered to recrystallize at lower temperatures (Petrou and Economou-Eliopoulos, 2009). PGEs are considered to be very valuable in society (e.g., jewellery, coins/bars), and there is an increasing demand for them in the automotive, chemical, dental, medical/biomedical, and petroleum industries (Ek et al., 2004; Johnson Matthey, 1999). At the end of their usage or lifetime, these products become part of our environment along with contributions by emissions during mineral processing and fossil-fuel burning (Crocket and Teruta, 1976; Chyi, 1982); in particular, important PGE sources in the environment are sewage systems and automobile catalytic converters (Lottermoser and Morteani, 1993; Helmers, 1997; Jarvis et al., 2001). Eventually, significant quantities of PGEs will enter fluvial, estuarine, and coastal sediments (e.g., de Vos et al., 2002; Jarvis et al., 2001; Ravizza and Bothner, 1996; Wei and Morrison, 1994; Esser and Turekian, 1993). For example, up to 6 ng g1 PGE have been found in sediments of the Kentish Stour (de Vos et al., 2002). Along streets with heavy traffic, platinum concentrations up to the lower mg g1 range have been detected, corresponding to anthropogenic enrichment factors of 105–106 (Hoppstock and Sures, 2004); ash from sewage sludge burning and street dust contain up to 0.5 mg g1 palladium (Leopold et al., 2008). Via road runoff, these metals are also introduced into aquatic habitats. For example, while the natural background concentration of Pt in rainwater is o0.2 ng l1, in surface-water drainage along highways, Pt levels up to 80 ng l1 are reported (Merian et al., 2004). The water solubility of PGE decreases in the order Pd 4 Pt 4 Rh, resembling their biological availability for plants and mussels; solubility is increased by aging of catalysts and the presence of humic substances (Merian et al., 2004). Thus, the most significant health risk is posed by soluble PGE contents, because via mobilization by waters, a considerable amount of PGE emitted from cars is able to enter different environmental compartments. Although adsorption and surface complexation onto soil and sediment particles lead to PGE immobilization (e.g., Sako et al., 2009), there are indications for transformation effects into more reactive/bioavailable and thus mobile species (Moldovan et al., 2001; Rauch and Morrison, 1999; Lustig et al., 1996; Alt et al., 1994). In this respect, palladium seems to be more critical than platinum and rhodium (Zimmermann and Sures, 2004): While Pd was found in soil depths of 12–16 cm, in the same soil profile, Pt and Rh have been found

mainly concentrated on the surface (depths below 8 cm). Palladium is significantly incorporated in cells, plants, and animals, and exhibits stronger biological effects similar to Pt and Rh or other heavy metals (Battke et al., 2008; Frank et al., 2008; Singer et al., 2005; Sures et al., 2006; Hoppstock and Sures, 2004). Pd forms very stable complexes not only with soft donors such as sulfur (e.g., in cysteine, methione, proteins, or enzymes), but also with N and O donors (e.g., in DNA, RNA, or peptides). Compared to Pt, Pd shows reaction rates higher by three to five orders of magnitude (Kozlowski and Pettit, 1991; Al-Bazi and Chow, 1984).

3.02.4 Conclusions The most important points raised in this chapter are summarized as follows:

•

•

•

•

•

Currently, the most serious problem globally is the intoxication of millions of people with drinking water containing too much arsenic. As the latter is of geogenic origin and the world population (and therefore demand for drinking water) is still growing, this problem will become even more pressing in the future. There exists an apparent lack in arsenic toxicology even now because speciation is not accounted for: only when the total arsenic content is known, it will make an enormous difference if the arsenic in question is in the form of arseno-betaine or arsenite (or even trivalent organic species) because of the different toxicity of the As-species. Maximal methylmercury production does not usually coincide with the site of maximal mercury pollution (see Figures 2 and 4), because MMR is a product of many variables to be accounted for. Relevant examples have been discussed in the chapter (e.g., Gulf of Trieste and Everglades National Park). Officials should be aware that flooding of large areas (e.g., wetlands or reservoirs) promote mercury methylation, thus creating the risk of methylmercury enrichment along the food chain, eventually reaching mercury concentrations in fish even in the mg kg1 range (i.e., too contaminated to be eaten). As PGEs are increasingly used, there will be an urgent demand in our society to develop relevant speciation methods to evaluate the environmental and toxicological impact of these elements.

Acknowledgments The authors are grateful to the Deutsche Forschungsgemeinschaft for generously sponsoring speciation research (grant FOR 415), and to Dr. Roland A. Diaz-Bone for providing them with a computer model of the halo hypothesis (Figure 2(a)).

References Aiuppa A, D’Alessandro W, Federico C, Palumbo B, and Valenza M (2003) The aquatic geochemistry of arsenic in volcanic groundwater from southern Italy. Applied Geochemistry 18: 1283--1296.

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

Al-Bazi SJ and Chow A (1984) Platinum metals – solution chemistry and separation methods (ion-exchange and solvent extraction). Talanta 31: 815--836. Allard B (1995) Groundwater. In: Salbu B and Steinnes E (eds.) Trace Elements in Natural Waters, pp. 151--176. Boca Raton, FL: CRC Press. Alt F, Messerschmidt J, Fleischer M, and Schaller KH (1994) Platinum. In: Angerer J and Schaller KH (eds.) Analysis of Hazardous Substances in Biological Materials, vol. 4, pp. 187--205. Weinheim: VCH. Amirbahman A, Reid A, Haines TA, Kahl JS, and Arnold C (2002) Association of methylmercury with dissolved humic acids. Environmental Science and Technology 36: 690--695. Anawar H and Mihaljevic M (2009) Comment on ‘‘Arsenic release from biotite into a Holocene groundwater aquifer in Bangladesh’’ by Seddique AA, Masuda H, Mitamura M, Shinoda K, Yamanaka T, Itai T, Maruoka T, Uesugi K, Ahmed KM, Biswas DK. Applied Geochemistry 24: 483--485. Arcega-Cabrera F, Armienta MA, Daessle LW, Castillo-Blum SE, Talavera O, and Dotor A (2009) Variations of Pb in a mine-impacted tropical river, Taxco, Mexico: Use of geochemical, isotopic and statistical tools. Applied Geochemistry 24: 162--171. Armienta MA, Talavera O, Morton O, and Barrera M (2003) Geochemistry of metals from mine tailings in Taxco, Mexico. Bulletin of Environmental Contamination and Toxicology 71: 387--393. Astro¨m M (2001) The effect of acid soil leaching on trace element abundance in a medium-sized stream, W. Finland. Applied Geochemistry 16: 387--396. ATSDR (Agency for Toxic Substances and Disease Registry) (2005) Case Studies in Environmental Medicine. U.S. Department of Health and Human Services. http:// www.atsdr.cdc.gov (accessed March 2010). Baas-Becking LGM, Kaplan IR, and Moore D (1960) Limits of the natural environment in terms of pH and oxidation-reduction potentials. Journal of Geology: 68, 243–284. Babechuk MG, Weisener CG, Fryer BJ, Paktunc D, and Maunders C (2009) Microbial reduction of ferrous arsenate: Biogeochemical implications for arsenic mobilization. Applied Geochemistry 24: 2332--2341. Balistrieri LS and Blank RG (2008) Dissolved and labile concentrations of Cd, Cu, Pb, and Zn in South Fork Coeur d’Alene River, Idaho: Comparisons among chemical equilibrium models and implications for biotic ligand models. Applied Geochemistry 23: 3355--3371. Balistrieri L, Murray JW, and Paul B (1994) The geochemical cycling of trace elements in a biogenic meromictic lake. Geochimica et Cosmochimica Acta 58: 3993--4008. Banks D, Younger PL, Arnesen R-T, Iversen ER, and Banks SB (1997) Mine-water chemistry: The good, the bad and the ugly. Environmental Geology 32: 157--174. Barrega˚rd L (2005) Exposure to mercury in the general population of Europe and theArctic Circle. In: Pirrone N and Mahaffey KR (eds.) Dynamics of Mercury Pollution on Regional and Global Scales – Atmospheric Processes and Human Exposures Around the World, pp. 385--404. New York, NY: Springer. Basu A, Mahata J, Gupta S, and Giri AK (2001) Genetic toxicology of a paradoxical human carcinogen, arsenic: A review. Mutation Research 488: 171--194. Battke F, Leopold K, Maier M, Schmidhalter U, and Schuster M (2008) Palladium exposure of barley: Uptake and effects. Plant Biology 10: 272--276. Bauer M and Blodau C (2009) Arsenic distribution in the dissolved, colloidal and particulate size fraction of experimental solutions rich in dissolved organic matter and ferric ion. Geochimica et Cosmochimica Acta 73: 529--542. Beak DG, Wilkin RT, Ford RG, and Kelly SD (2008) Examination of arsenic speciation in sulfidic solutions using X-ray absorption spectroscopy. Environmental Science and Technology 42: 1643--1650. Beesley L, Moreno-Jimenez E, Clemente R, Lepp N, and Dickonsin N (2010) Mobility of arsenic, cadmium and zinc in a multi-element contaminated soil profile assessed by in-situ soil pore water sampling, column leaching and sequential extraction. Environmental Pollution 158: 155--160. Benoit JM, Gilmour CC, Heyes A, Mason RP, and Miller C (2003) Geochemical and biological controls over methylmercury production and degradation in aquatic systems. In: Chai Y and Braids OC (eds.) Biochemistry of Environmentally Important Trace Elements, pp. 262--297. Washington, DC: ACS. Benoit JM, Gilmour CC, and Mason RP (2001) The influence of sulphide on solidphase mercury bioavailability for methylation by pure cultures of Desulfobulbus propionicus (1pr3). Environmental Science and Technology 35: 127--132. Berg M, Trang PTK, Stengel C, et al. (2008) Hydrogeological and sedimentary controls leading to arsenic contamination of groundwater in the Hanoi area, Vietnam: The impact of iron-arsenic ratios, peat, river bank deposits, and excessive groundwater abstraction. Chemical Geology 249: 91--112. Berger AC, Bethke CM, and Krumhansl JL (2000) A process model of natural attenuation in drainage from a historic mining district. Applied Geochemistry 15: 655--666. Bergseth H (1983) Effect of filter type and filtrate treatment on the measured content of Al and other ions in groundwater. Acta Agriculturae Scandinavica 33: 353--359.

51

Beyersmann D and Hartwig A (1994) Genotoxic effects of metal compounds. Archives of Toxicology 16: 192--198. Bhattacharya P, Welch AH, Ahmed KM, Jacks G, and Naidu R (2004) Arsenic in groundwater of sedimentary aquifers. Applied Geochemistry 19: 163--260. Bissen M and Frimmel FH (2003a) Arsenic – a review. Part I: Occurrence, toxicity, speciation, mobility. Acta Hydrochimica et Hydrobiologica 31: 9--18. Bissen M and Frimmel FH (2003b) Arsenic – a Review. Part II: Oxidation of arsenic and its removal in water treatment. Acta Hydrochimica et Hydrobiologica 31: 97--107. Bjornberg A, Hakanson L, and Lundbergh K (1988) A theory on the mechanism regulating the bioavailability of mercury in natural waters. Environmental Pollution 49: 53--61. Bloom NS, Gill GA, Cappellino S, et al. (1999) Speciation and cycling of mercury in Lavaca Bay, Texas, sediment. Environmental Science and Technology 33: 7--13. Borba RP, Figueiredo BR, and Matschullat J (2003) Geochemical distribution of arsenic in waters, sediments and weathered gold minaralized rocks from Iron Quadrangle, Brazil. Environmental Geology 44: 39--52. Borrok DM, Nimick DA, Wanty RB, and Ridley WI (2008) Isotopic variations of dissolved copper and zinc in stream waters affected by historical mining. Geochimica et Cosmochimica Acta 72: 329--344. Bostick BC, Fendorf S, and Brown GE, Jr (2005) In situ analysis of thioarsenite complexes in natural to alkaline sulphide solutions. Mineralogical Magazine 69: 781--795. Boutron CF and Go¨rlach U (1990) The occurrence of heavy metals in Antarctic and Greenland ancient ice and recent snow. In: Broekaert JAC, Gu¨cer S, and Adams F (eds.) Metal Speciation in the Environment, NATO ASI Series G 23, pp. 137–151. Berlin: Springer. Bowen HJM (1979) Environmental Chemistry of the Elements. London: Academic Press. Boyd ES, King S, Tomberlin JK, et al. (2009) Methylmercury enters an aquatic food web through acidophilic microbial mats in Yellowstone National Park, Wyoming. Environmental Microbiology 11: 950--959. Braungardt CB, Achterberg EP, Elbaz-Poulichet F, and Morley NH (2003) Metal geochemistry in a mine-polluted estuarine system in Spain. Applied Geochemistry 18: 1757--1771. Bridges CC and Zalups RK (2005) Molecular and ionic mimicry and the transport of toxic metals. Toxicology and Applied Pharmacology 204: 274--308. Brookins DG (1988) Eh-pH Diagrams for Geochemistry. 176pp. New York, NY: Springer. Bruland KW (1983) Trace elements in sea water. Chemical Oceanography 8: 157--220. Buffle J (1990) The ecological role of aquatic organic and inorganic components, deduced from their nature, circulation and interactions. In: Broekaert JAC, Gu¨cer S, and Adams F (eds.) Metal Speciation in the Environment, NATO ASI Series G 23, pp. 469–501. Berlin: Springer. Bullen TD and Eisenhauer A (2009) Metal stable isotopes in low-temperature systems: A primer. Elements 5: 349--352. Bundesgesetzblatt (2001) Verordnung zur Novellierung der Trinkwasserverordnung. Teil L, pp. 959–980. Buschmann J and Sigg L (2004) Antimony(III) binding to humic substances: Influence of pH and type of humic acid. Environmental Science and Technology 38: 4535--4541. Castoldi AF, Coccini T, Ceccatelli S, and Manzo L (2001) Neuotoxicity and molecular effects of methylmercury. Brain Research Bulletin 55: 197--203. CHA (Child Health Alert) (2001) The problem of mercury-contaminated fish. Child Health Alert 19: 3--5. Challenger F (1945) Biological methylation. Chemical Reviews 36: 315--361. Challenger F (1951) Biological methylation. Advances in Enzymology 12: 429--491. Chen Y, Gat P, Frimmel FH, and Abbt-Braun G (2006) Metal binding by humic substances and dissolved organic matter derived from compost. In: Twardowska I, Allen HE, and Ha¨ggblom MM (eds.) Viable Methods of Soil and Water Pollution Monitoring, Protection and Remediation, NATO Science Series, vol. 69, pp. 275–297. Dordrecht: Springer. Chen YW, Deng TL, Filella M, and Belzile N (2003) Distribution and early diagenesis of antimony species in sediments and pore waters of freshwater lakes. Environmental Science and Technology 37: 1163--1168. Choi S-C and Bartha R (1994) Environmental factors affecting mercury methylation in estuarine sediments. Bulletin of Environmental Contamination and Toxicology 53: 805--812. Chyi LL (1982) The distribution of gold and platinum in bituminous coal. Economic Geology 77: 1592--1597. Cima F, Craig PJ, and Harrington C (2003) Organotin compounds in the environment. In: Craig P (ed.) Organometallic Compounds in the Environment, 2nd edn., pp. 101--149. New York, NY: Wiley.

52

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

Clarkson TW (1997) The toxicity of mercury. Critical Reviews in Clinical Laboratory Sciences 34: 369--403. Cleary D (1990) Anatomy of the Amazon Gold Rush, 245pp. London: Macmillan. Cleckner L, Gilmour C, Hurley J, and Krabbenhoft D (1999) Mercury methylation in periphyton of the Florida Everglades. Limnology and Oceanography 44: 1815--1825. Compeau GC and Bartha R (1985) Sulphate-reducing bacteria: Principal methylators of mercury in anoxic estuarine sediments. Applied and Environmental Microbiology 50: 498--502. Corkhill CL and Vaughan DJ (2009) Arsenopyrite oxidation – a review. Applied Geochemistry 24: 2342--2361. Covelli S, Faganeli J, De Vittor C, Predonzani S, Acquavita A, and Horvat M (2008) Benthic fluxes of mercury species in a lagoon environment (Grado Lagoon, Northern Adriatic Sea, Italy). Applied Geochemistry 23: 529--546. Craig P.J. (2003) Organometallic Compounds in the Environment, 2nd edn., 415pp. New York, NY: Wiley. Crocket JH and Teruta Y (1976) Pt, Pa, Au and Ir content of Kelly Lake bottom sediments. Canadian Mineralogist 14: 58--61. Cullen WR (2008) Is Arsenic an Aphrodisiac? The Sociochemistry of an Element. 412pp. Cambridge: RSC. Cullen WR, Li H, Hewitt G, Reimer KJ, and Zalunardo N (1994) Identification of extracellular arsenical metabolites in the growth medium of the microorganisms Apiotrichum humicola and Scopulariopsis brevicaulis. Applied Organometallic Chemistry 8: 303--311. Cullen WR and Reimer KJ (1989) Arsenic speciation in the environment. Chemical Reviews 89: 713--764. Dales L, Kahn E, and Wei E (1971) Methylmercury poisoning. An assessment of the sportfish hazard in California. California Medicine 114: 13--15. Davidson PW, Myers GJ, and Weiss B (2004) Mercury exposure and child development outcomes. Pediatrics 113: 1023--1029. Davison W and Zhang H (1994) In situ speciation measurements of trace components in natural waters using thin-film gels. Nature 367: 546--548. Deng TL, Chen YW, and Belzile N (2001) Antimony speciation at ultra trace levels using hydride generation atomic fluorescence spectrometry and 8-hydroxyquinoline as an efficient masking agent. Analytica Chimica Acta 432: 293--302. de Vos E, Edwards SJ, McDonald I, Wray DS, and Carey PJ (2002) A baseline survey of the distribution and origin of platinum group elements in contemporary fluvial sediments of the Kentish Stour, England. Applied Geochemistry 17: 1115--1121. Di Toro DM, Allen HE, Bergmann HL, Meyer JS, Paquin PR, and Santore RC (2001) Biotic ligand model of the acute toxicity of metals. I: Technical basis. Environmental Toxicology and Chemistry 20: 2383--2396. Dopp E, Hartmann LM, Florea AM, Rettenmeier AW, and Hirner AV (2004) Environmental distribution, analysis, and toxicology of organometal(loid) compounds. Critical Reviews in Toxicology 34: 301--333. Dopp E, Hartmann LM, Florea AM, et al. (2006a) Trimethylantimony dichloride causes genotoxic effects in Chinese hamster ovary cells after forced uptake. Toxicology In Vitro 20: 1060--1065. Dopp E, Hartmann LM, von Recklinghausen U, Rabieh S, Hirner AV, and Rettenmeier AW (2006b) Methylation, oxidation state and membrane permeability determine toxicity of arsenic compounds. Naunyn-Schmiedeberg’s Archives of Pharmacology 372: 115. Dopp E, Hartmann LM, von Recklinghausen U, et al. (2006c) Uptake, subcellular distribution and toxicity of arsenic species in methylating and non-methylating human cells. Toxicology Letters 164: S260–S260 Dopp E, Hartmann LM, von Recklinghausen U, et al. (2005) Forced uptake of trivalent and pentavalent methylated and inorganic arsenic and its cyto-/genotoxicity in fibroblasts and hepatoma cells. Toxicological Sciences 87: 46--56. Dopp E, Hartmann LM, von Recklinghausen U, et al. (2007) The cyto- and genotoxicity of organotin compounds is dependent on the cellular uptake capability. Toxicology 232: 226--234. Dopp E, von Recklinghausen U, Hartmann LM, et al. (2008) Subcellular distribution of inorganic and methylated arsenic compounds in human urothelial cells and human hepatocytes. Drug Metabolism and Disposition 36: 971--979. Drever JI (1997) The Geochemistry of Natural Waters, 3rd edn. Upper Saddle River, NJ: Prentice Hall. Duester L, Vink JPM, and Hirner AV (2008) Methylantimony and -arsenic species in sediment pore water tested with the sediment or fauna incubation experiment. Environmental Science and Technology 42: 5866--5871. Du Laing G, Rinklebe J, Vandecasteele B, Meers E, and Tack FMG (2009) Trace metal behaviour in estuarine and riverine floodplain soils and sediments: A review. Science of the Total Environment 407: 3972--3985. Dyk S, Cornelis R, Griepink B, and Verbeeck P (1990) Stability study of Cr(III) and Cr(VI) in water for production of an aqueous chromium standard reference

material. In: Broekaert JAC, Gu¨cer S, and Adams F (eds.) Metal Speciation in the Environment, NATO ASI Series G 23, pp. 361–376. Berlin: Springer. Ek KH, Morrison GM, and Rauch S (2004) Environmental routes for platinum group elements to biological material – a review. Science of the Total Environment 334(335): 21--28. Eppinger RG and Fuge R (2009) Natural low-pH environments unaffected by human activity. Applied Geochemistry 24: 189--190. Esser BK and Turekian KK (1993) Anthropogenic osmium in coastal deposits. Environmental Science and Technology 27: 2719--2724. Ethier ALM, Mackay D, Toose-Reid LK, O’Driscoll NJ, Scheuhammer AM, and Lean DRS (2008) The development and application of a mass balance model for mercury (total, elemental and methyl) using data from a remote lake (Big Dam West, Nova Scotia, Canada) and the multi-species multiplier method. Applied Geochemistry 23: 467--481. Eto K (2000) Minamata disease. Neuropathology 20: S14--S19. Fairbrother A, Wenstel R, Sappington K, and Wood W (2007) Framework for metals risk assessment. Ecotoxicology and Environmental Safety 68: 145--227. Feldmann J (2003) Other organometallic compounds in the environment. In: Craig P (ed.) Organometallic Compounds in the Environment, 2nd edn., pp. 353--389. New York, NY: Wiley. Field M and Sherrell RM (2003) Direct determination of ultra-trace levels of metals in fresh water using desolvating micronebulization and HR-ICP-MS: Application to Lake Superior waters. Journal of Analytical Atomic Spectrometry 18: 254--259. Filella M, Belzile N, and Chen Y-W (2002a) Antimony in the environment: A review focused on natural waters II. Relevant solution chemistry. Earth-Science Reviews 59: 265--285. Filella M, Belzile N, and Chen Y-W (2002b) Antimony in the environment: A review focused on natural waters I. Occurrence. Earth-Science Reviews 57: 125--176. Filella M, Belzile N, and Lett M-C (2007) Antimony in the environment: A review focused on natural waters III. Microbiota relevant interactions. Earth-Science Reviews 80: 195--217. Filella M and May PM (2003) Computer simulation of the low-molecular-weight inorganic species distribution of antimony(III) and antimony(V) in natural waters. Geochimica et Cosmochimica Acta 67: 4013--4031. Filella M, Williams PA, and Belzile N (2009) Antimony in the environment: Known and unknowns. Environmental Chemistry 6: 95--105. Fisher JC, Wallschla¨ger D, Planer-Friedrich B, and Hollibaugh JT (2008) A new role for sulfur in arsenic cycling. Environmental Science and Technology 42: 81--85. Fitzgerald WF and Clarkson TW (1991) Mercury and monomethyl mercury: Present and future concerns. Environmental Health Perspectives 96: 159--166. Fitzgerald WF, Engstrom DR, Mason RP, and Nater ED (1998) Critical review: The case for atmospheric mercury contamination in remote areas. Environmental Science and Technology 32: 1--7. Fitzgerald WF, Lamborg CH, and Hammerschmidt CR (2007) Marine biogeochemical cycling of mercury. Chemical Reviews 107: 641--662. Fitzmaurice AG, Bilgin AA, O’Day PA, et al. (2009) Geochemical and hydrologic controls on the mobilization of arsenic derived from herbicide application. Applied Geochemistry 24: 2152--2162. Fo¨rstner U and Wittmann GTW (1981) Metal Pollution in the Aquatic Environment. Berlin: Springer. Francesconi KA (2005) Current perspectives in arsenic environmental and biological research. Environmental Chemistry 2: 141--145. Francesconi KA and Sperling M (2005) Speciation analysis with HPLC-mass spectrometry: Time to take stock. Analyst 130: 998--1001. Frank S, Singer C, and Sures B (2008) Metallothionein (MT) response after chronic palladium exposure in the Zebra mussel, Dreissena polymorpha. Environmental Research 108: 309--314. Frederick PC, Spalding M, and Dusek R (2002) Wading birds as bioindicators of mercury contamination in Florida: Annual and geographic variation. Environmental Toxicology and Chemistry 21: 262--264. Friedmann L and Lindenthal W (2009) Uranium in ground and drinking water in Bavaria. Wasser und Abfall 11: 21--24. Frimmel FH (1999) Wasser und Gewa¨sser. Berlin: Spektrum Akademischer Verlag. Frimmel FH (1990) Complexation of paramagnetic metal ions by aquatic fulvic acids. In: Broekaert JAC, Gu¨cer S, and Adams F (eds.) Metal Speciation in the Environment, NATO ASI Series G 23, pp. 57–69. Berlin: Springer. Frisbie SH, Mitchell EJ, Mastera LJ, et al. (2009) Public health strategies for western Bangladesh that address arsenic, manganese, uranium, and other toxic elements in drinking water. Environmental Health Perspectives 117: 410--416. Fu W, Cheng Z, Fu Y, and Xu H (1992) The investigation of Cd and Hg accumulation in human hairs in the area of Jinzhou Bay. Marine Environmental Science 11: 61--63 67 (in Chinese).

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

Fukushi K, Sasaki M, Sato T, Yanase N, Amano H, and Ikeda H (2003) A natural attenuation of arsenic in drainage from an abundant arsenic mine dump. Applied Geochemistry 18: 1267--1278. Futatsuka M, Kitano T, Shono M, et al. (2000) Health surveillance in the population living in a methyl mercury-polluted area over a long period. Environmental Research Section A 83: 83--92. Gebel TW (2001) Genotoxicity of arsenical compounds. International Journal of Hygiene and Environmental Health 203: 249--262. Gemici U, Tarcan G, Helvac C, and Somay AM (2008) High arsenic and boron concentrations in groundwaters related to mining activity in the Bigadic borate deposits (Western Turkey). Applied Geochemistry 23: 2462--2476. Geraldes CFGC and Castro MMCA (1990) Complexes of monosaccharides with metal ions: Some bioinorganic and environmental inorganic chemstry aspects. In: Broekaert JAC, Gu¨cer S, and Adams F (eds.) Metal Speciation in the Environment, NATO ASI Series G 23, pp. 105–119. Springer: Berlin. Gill GA, Bloom NS, Cappellino S, et al. (1999) Sediment–water fluxes of mercury in Lavaca Bay, Texas. Environmental Science and Technology 33: 663--669. Gilmour CC and Henry EA (1991) Mercury methylation in aquatic systems affected by acid depositions. Environmental Pollution 71: 131--169. Gilmour CC, Riedel GS, Edrington MC, et al. (1998) Methylmercury concentrations and production rates across a trophic gradient in the northern Everglades. Biogeochemistry 40: 327--345. Gochfeld M (2003) Cases of mercury exposure, bioavailability, and absorption. Ecotoxicology and Environmental Safety 56: 174--179. Goulet RR, Holmes J, Page B, et al. (2007) Mercury transformations and fluxes in sediments of a riverine wetland. Geochimica et Cosmochimica Acta 71: 3393--3406. Grandjean P and Choi A (2008) The delayed appearance of a mercurial warning. Epidemiology 19: 10--11. Gray JE, Greaves IA, Bustod DM, and Krabbenhoft DP (2003) Mercury and methylmercury contents in mine-waste calcine, water, and stream sediment collected from the Palawan Quicksilver mine, Philippines. Environmental Geology 43: 298--307. Gray JE, Hines ME, and Biester H (2006) Mercury methylation influenced by areas of past mercury mining in the Terlingua district, Southwest Texas, USA. Applied Geochemistry 21: 1940--1954. Gray JE, Hines ME, Higueras PL, Adatto I, and Lasora BK (2004) Mercury speciation and microbial transformations in mine wastes, stream sediments, and surface waters at the Almaden mining district, Spain. Environmental Science and Technology 38: 4285--4292. Graziano JH, Blum CB, LoIacono NJ, et al. (1996) A human in vivo model for the determination of lead bioavailability using stable isotope dilution. Environmental Health Perspectives 104: 176--179. Green WJ, Stage BR, Preston A, Wagers S, Shacat J, and Newell S (2005) Geochemical processes in the Onyx River, Wright Valley, Antarctica: Major ions, nutrients, trace metals. Geochimica et Cosmochimica Acta 69: 839--850. Gremm TJ and Frimmel FH (1992) Determination of organotin compounds in aqueous samples by means of HPGC-AED. Water Research 26: 1163--1169. Grybos M, Davranche M, Gruau G, Petitjean P, and Pedrot M (2009) Increasing pH drives organic matter solubilization from wetland soils under reducing conditions. Geoderma 154: 13--19. Gustin MS, Kolker A, and Gardfeldt K (2008) Transport and fate of mercury in the environment. Applied Geochemistry 23: 343--344. Hakansson L (1996) A simple model to predict the duration of the mercury problem in Sweden. Ecological Modelling 93: 251--262. Hall BD, Aiken GR, Krabbenhoft DP, Marvin-DiPasquale M, and Swarzenski CM (2008) Wetlands as principal zones of methylmercury production in southern Louisiana and the Gulf of Mexico region. Environmental Pollution 154: 124--134. Hammond A (1971) Mercury in the environment: Natural and human factors. Science 171: 788--794. Hansen HR, Pickford R, Oates JT, Jaspers M, and Feldmann J (2004) 2Dimethylarsinothioyl acetic acid identified in a biological sample: The first occurrence of a mammalian arsinothioyl metabolite. Angewandte Chemie International Edition 43: 337--340. Harada M (1995) Minamata disease: Methylmercury poisoning in Japan caused by environmental pollution. Critical Reviews in Toxicology 25: 1--24. Hartwig A (1995) Current aspects in metal genotoxicity. Biometals 8: 3--11. Hayakawa T, Kobayashi Y, and Cui X (2005) A new metabolic pathway of arsenite: Arsenic–glutathione complexes are substrates for human arsenic methyltransferase Cyt19. Archives of Toxicology 79: 183--191. Helmers E (1997) Platinum emission rate of automobiles with catalytic converters: Comparison and assessment of results from various approaches. Environmental Science and Pollution Research 4: 100--103.

53

Helsel DR (1990) Less than obvious: Statistical treatment of data below the detection limit. Environmental Science and Technology 24: 1766--1774. Helsel DR and Hirsch RM (1992) Studies in Environmental Science 49. Statistical Methods in Water Resources. New York, NY: Elsevier. Hesterberg D, Chou JW, Hutchison KJ, and Sayers DE (2001) Bonding of Hg(II) to reduced organic sulphur in humic acid as affected by S/Hg ratio. Environmental Science and Technology 35: 2741--2745. Heyes A, Miller C, and Mason RP (2004) Mercury and methylmercury in Hudson River sediment: Impact of tidal resuspension on partitioning and methylation. Marine Chemistry 90: 75--89. Hines ME, Faganeli J, Adatto I, and Horvat M (2006) Microbial mercury transformations in marine, estuarine and freshwater sediment downstream of the Idrija Mercury Mine, Slovenia. Applied Geochemistry 21: 1924--1939. Hintelmann H, Wellbourn PM, and Evans RD (1997) Measurement of complexation of methylmercury(II) compounds by freshwater humic substances using equilibrium dialysis. Environmental Science and Technology 31: 489--495. Hirao Y and Patterson CC (1974) Lead aerosol pollution in the High Sierra overrides. Natural mechanisms which exclude lead from a food chain. Science 184: 989--992. Hirner AV (1996) Testing metal mobility in soil by elution tests. In: Reuther E (ed.) Geochemical Approaches to Environmental Engineering of Metals, Environmental Science Series, pp. 151–177. Berlin: Springer. Hirner AV (2000) Extraktions- und Elutionsverfahren zur Bestimmung mobiler anorganischer und organischer Kontaminanten in Feststoffen. In: AnalytikerTaschenbuch, band. 21, pp. 151–177. Berlin: Springer. Hirner AV (2006) Speciation of alkylated metals and metalloids in the environment. Analytical and Bioanalytical Chemistry 385: 555--567. Hirner AV, Feldmann J, Krupp E, Gruemping R, Goguel R, and Cullen WR (1998) Metal(loid)organic compounds in geothermal gases and waters. Organic Geochemistry 29: 1765--1778. Hirner AV and Rettenmeier (2010) Methylated metal(loid) species in humans. Metal Ions in Life Sciences 7: 465--521. Hitchon B, Perkins EH, and Gunter WD (1999) Introduction to Ground Water Geochemistry. 310pp. Edmonton, AB: Geoscience Publications. Hollibaugh J, Carini S, Gu¨rleyu¨k H, et al. (2005) Distribution of arsenic species in alkaline, hypersaline, Mono Lake, CA and response to seasonal stratification and anoxia. Geochimica et Cosmochimica Acta 69: 1925--1937. Hoppstock K and Sures B (2004) Platinum-group metals. In: Merian E, Anke M, Ihnat M, and Stoeppler M (eds.) Elements and Their Compounds in the Environment, 2nd edn., pp. 1047--1086. Weinheim: Wiley-VCH. Horneman A, Van Geen A, Kent DV, et al. (2004) Decoupling of As and Fe release to Bangladesh groundwater under reducing conditions. Part I: Evidence from sediment profiles. Geochimica et Cosmochimica Acta 68: 3459--3473. Horvat M, Kotnik J, Logar M, Fajon V, Zvonaric T, and Pirrone N (2003) Speciation of mercury in surface and deep-sea waters in the Mediterranean sea. Atmospheric Environment 37: S93--S108. Hudson-Edwards KA, Houghton SL, and Osborn A (2004) Extraction and analysis of arsenic in soils and sediments. Trends in Analytical Chemistry 23: 745--752. Hughes MF (2002) Arsenic toxicity and potential mechanisms of action. Toxicology Letters 133: 1--16. Hurley J, Benoit J, Babiarz C, et al. (1995) Influences of watershed characteristics on mercury levels in Wisconsin rivers. Environmental Science and Technology 29: 1867--1875. Irgolic KJ (1990) Speciation. In: Broekaert JAC, Gu¨cer S, and Adams F (eds.) Metal Speciation in the Environment, NATO ASI Series G 23, pp. 641–645. Springer: Berlin. Itai T, Takahashi Y, Seddique AA, Maruoka T, and Mitamura M (2010) Variations in redox state of As and Fe measured by X-ray absorption spectroscopy in aquifers of Bangladesh and their effect on As adsorption. Applied Geochemistry 25: 34--47. Jarvis KE, Parry SJ, and Piper JM (2001) Temporal and spatial studies of autocatalystderived platinum, rhodium, and palladium and selected vehicle-derived trace elements in the environment. Environmental Science and Technology 35: 1031--1036. Jensen S and Jernelov A (1969) Biological methylation of mercury in aquatic organisms. Nature 223: 753--754. Jiang GB, Shi JB, and Feng XB (2006) Mercury pollution in China. Environmental Science and Technology 40: 3672--3678. Matthey J (1999) Platinum 1999. London: Johnson Matthey. Jung MC (2001) Heavy metal contamination of soils and waters in and around the Imcheon Au–Ag mine, Korea. Applied Geochemistry 16: 1369--1375. Kahn M, Raml R, Schmweisser E, Vallant B, Francesconi KA, and Goessler W (2005) Two novel thio-arsenosugars in scallops identified with HPLC-ICPMS and HPLCESMS. Environmental Chemistry 2: 171--176.

54

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

Kalbitz K and Wennrich R (1998) Mobilization of heavy metals and arsenic in polluted wetland soils and its dependance on dissolved organic matter. Science of the Total Environment 209: 27--39. Kalluri JR, Arbneshi T, Khan SA, et al. (2009) Use of gold nanoparticles in a simple colorimetric and ultrasensitive dynamic light scattering assay: Selective detection of arsenic in groundwater. Angewandte Chemie 121: 1--5. Kellner R, Mermet J-M, Otto M, and Widmer HM (1998) Analytical Chemistry. Weinheim: Wiley-VCH. Kelly CA, Rudd JWM, Bodaly RA, et al. (1997) Increases in fluxes of greenhouse gases and methyl mercury following flooding of an experimental reservoir. Environmental Science and Technology 31: 1334. Keon NE, Swartz CH, Brabander DJ, Harvey C, and Hemond HF (2001) Validation of an arsenic sequential extraction method for evaluating mobility in sediments. Environmental Science and Technology 35: 2778--2784. Kerin EJ, Gilmour CC, Roden E, Suzuki MT, Coates JD, and Mason RP (2006) Mercury methylation by dissimilatory iron-reducing bacteria. Applied and Environmental Microbiology 72: 7919--7921. Kerper IE, Ballatori N, and Clarkson TW (1992) Methylmercury transport across the blood–brain barrier by an amino acid carrier. American Journal of Physiology 262: R761--R765. King JK, Kostka JE, Frischer ME, and Saunders FM (2000) Sulfate-reducing bacteria methylate mercury at variable rates in pure culture and marine sediment. Applied and Environmental Microbiology 66: 2430--2437. King JK, Saunders FM, Lee RF, and Jahnke RA (1999) Coupling mercury methylation rates to sulfate reduction rates in marine sediments. Environmental Toxicology and Chemistry 18: 1362--1369. King SA, Behnke S, Slack K, et al. (2006) Mercury in water and bioass of microbial communities in hot springs of Yellowstone National Park, USA. Applied Geochemistry 21: 1868--1879. Kirby CS and Elder Brady JA (1998) Field determination of Fe2þ oxidation rates in acid mine drainage using a continuously-stirred tank reactor. Applied Geochemistry 13: 509--520. Kondo K (2000) Congenital Minamata disease: Warnings from Japan’s experience. Journal of Child Neurology 15: 458--464. Koretsky CM, Haveman M, Beuving L, Cuellar A, Shattuck T, and Wagner M (2007) Spatial variation of redox and trace metal geochemistry in a minerotrophic fen. Biogeochemistry 86: 33--62. Koski RA, Munk LA, Foster AL, Shanks WC, III, and Stillings LL (2008) Sulfide oxidation and distribution of metals near abandoned copper mines in coastal environments, Prince William Sound, Alaska, USA. Applied Geochemistry 23: 227--254. Koutsospyros A, Braida W, Christodoulatos C, Dermatas D, and Strigul N (2006) A review of tungsten: From environmental obscurity to scrutiny. Journal of Hazardous Materials 136: 1--19. Kozlowski H and Pettit LD (1991) Amino acid and peptide complexes of the platinum group metals. In: Hartley FR (ed.) Chemistry of the Platinum Group Metals: Recent Developments, pp. 530--545. New York, NY: Elsevier. Ko¨sters J, Diaz-Bone RA, Planer-Friedrich B, Rothweiler B, and Hirner AV (2003) Identification of organic arsenic, tin, antimony and tellurium compounds in environmental samples by GC-MS. Journal of Molecular Structure 661–662: 347--356. Krachler M and Shotyk W (2009) Trace and ultratrace metals in bottled waters: Survey of sources worldwide and comparison with refillable metal bottles. Science of the Total Environment 407: 1089--1096. Krachler M, Zheng M, Fisher D, and Shotyk W (2008) Atmospheric Sb in the Arctic during the past 16,000 years: Responses to climate change and human impacts. Global Biogeochemical Cycles 22: GB1015, 9pp Krachler M, Zheng M, Fisher D, and Shotyk W (2009) Global atmospheric As and Bi contamination preserved in 3000 year old Arctic ice. Global Biogeochemical Cycles 23: GB003471, 7pp Krachler M, Zheng M, Koerner R, Zdanowicz C, Fisher D, and Shotyk W (2005) Increasing atmospheric antimony contamination in the Northern Hemisphere: Snow and ice evidence from Devon Island, Arctic Canada. Journal of Environmental Monitoring 7: 1169--1176. Krupp EM, Milne BF, Mestrot A, Meharg AA, and Feldmann J (2008) Investigation into mercury bound to biothiols: Structural identification using ESI–ion-trap MS and introduction of a method for their HPLC separation with simultaneous detection by ICP-MS and ESI-MS. Analytical and Bioanalytical Chemistry 390: 1753--1764. Kru¨ger K, Gruner J, Madeja M, et al. (2006) Blockade and enhancement of glutamate receptor responses in Xenopus oocytes by methylated arsenicals. Archives of Toxicology 80: 492--501. Kru¨ger K, Straub H, Hirner AV, Hippler J, Binding N, and MuXhoff U (2009) Effects of monomethylarsonic and monomethylarsonous acid on evoked synaptic potentials

in hippocampal slices of adult and young rats. Toxicology and Applied Pharmacology 236: 115--123. Lambertsson L and Nilsson M (2006) Organic material: The primary control on mercury methylation and ambient methyl mercury concentrations in estuarine sediments. Environmental Science and Technology 40: 1822--1829. Landrum JT, Bennett PC, Engel AS, Alsina MA, Pasten PA, and Milliken K (2009) Partitioning geochemistry of arsenic and antimony, El Tatio Geyser Field, Chile. Applied Geochemistry 24: 664--676. Langmuir D (1997) Aqueous Environmental Chemistry. Upper Saddle River, NY: Prentice Hall. Langner HW, Jackson CR, McDermott TR, and Inskeep WP (2001) Rapid oxidation of arsenite in a hot spring ecosystem, Yellowstone National Park. Environmental Science and Technology 35: 3302--3309. Laveuf C and Cornu S (2009) A review on potentiality of rare earth elements to trace pedogenetic processes. Geoderma 154: 1--12. Lee CG, Chon H-T, and Jung MC (2001) Heavy metal contamination in the vicinity of the Daduk Au–Ag–Pb–Zn mine in Korea. Applied Geochemistry 16: 1377--1386. Leopold K, Maier M, Weber S, and Schuster M (2008) Long-term study of palladium in road tunnel dust and sewage sludge ash. Environmental Pollution 156: 341--347. Liang LN, Shi JB, He B, Jiang GB, and Yuan CG (2003) Investigation of methylmercury and total mercury contamination in mollusc samples collected from coastal sites along the Chinese Bohai Sea. Journal of Agricultural and Food Chemistry 51: 7373--7378. Lin S, Cullen WR, and Thomas DJ (1999) Methylarsenicals and arsinothiols are potent inhibitors of mouse liver thioredoxin reductase. Chemical Research in Toxicology 12: 924--930. Liu G, Cai Y, Kalla P, et al. (2008a) Mercury mass budget estimates and cycling seaonality in the Florida Everglades. Environmental Science and Technology 42: 1954--1960. Liu G, Cai Y, Mao Y, et al. (2009) Spatial variability in mercury cycling and relevant biogeochemical controls in the Florida Everglades. Environmental Science and Technology 43: 4361--4366. Liu G, Cai Y, Philippi T, et al. (2008b) Distribution of total and methylmercury in different ecosystem compartments in the Everglades: Implications for mercury bioaccumulation. Environmental Pollution 153: 257--265. Liu WX, Coveney RM, and Chen JL (2003) Environmental quality assessment on a river system polluted by mining activities. Applied Geochemistry 18: 749--764. Loftus WF (2000) Accumulation and Fate of Mercury in an Everglade Aquatic Food Web. PhD Dissertation, Florida International University, Miami, FL. Lottermoser BG and Morteani G (1993) Sewage sludges: Toxic substances, fertilizers, or secondary metal resources? Episodes 16: 329--333. Lund W (1990) The complexation of metal ions by humic substances in natural waters. In: Broekaert JAC, Gu¨cer S, and Adams F (eds.) Metal Speciation in the Environment, NATO ASI Series G 23, pp. 43–56. Berlin: Springer. Lustig S, Zang S, Michalke B, Schramel P, and Beck W (1996) Transformation behaviour of different platinum compounds in a claylike humic soil: Speciation investigations. Science of the Total Environment 188: 195--204. Lyven B, Hassello¨v M, Turner DR, Haraldsson C, and Andersson K (2003) Competition between iron- and carbon-based colloidal carriers for trace metals in a freshwater assessed using field-flow fractionation coupled to ICPMS. Geochimica et Cosmochimica Acta 67: 3791--3802. Manahan SE (1994) Environmental Chemistry, 4th edn. New York, NY: Oxford University Press. Mandal BK, Chowdhury TR, Samanta G, et al. (1996) Arsenic in groundwater in seven districts of West Bengal, India – the biggest arsenic calamity in the world. Current Science 70: 976--986. Martin-Doimeadios RCC, Tessier E, Amouroux D, et al. (2004) Mercury methylation/ demethylation and volatilization pathways in estuarine sediment slurries using species-specific enriched stable isotopes. Marine Chemistry 90: 107--123. Marvin-DiPasquale MC and Oremland RS (1998) Bacterial methylmercury degradation in Florida Everglades peat sediment. Environmental Science and Technology 32: 2556--2563. Mason RP, Rolfhus KR, and Fitzgerald WF (1998) Mercury in the North Atlantic. Marine Chemistry 61: 37--53. Mass MJ, Tennant A, Roop BC, et al. (2001) Methylated trivalent arsenic species are genotoxic. Chemical Research in Toxicology 14: 355--361. Matschullat J (2000) Arsenic in the geosphere – a review. Science of the Total Environment 249: 297--312. Matschullat J, Ottenstein R, and Reimann C (2000) Geochemical background – can we calculate it? Environmental Geology 39: 990--1000.

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

Meili M, Bishop K, Bringmark L, et al. (2003) Critical levels of atmospheric pollution: Criteria and concepts for operational modelling of mercury in forest and lake ecosystems. Science of the Total Environment 304: 83--106. Merian E (1984) Metalle in der Umwelt. Weinheim: Verlag Chemie. Merian E, Anke M, Ihnat M, and Stoeppler M (2004) Elements and Their Compounds in the Environment. Wenheim: Wiley-VCH. Merritt KA and Amirbahman A (2009) Mercury methylation dynamics in estuarine and coastal marine environments – a critical review. Earth-Science Review 96: 54--66. Miller GC, Lyons WB, and Davis A (1996) Understanding the water quality of pit lakes. Environmental Science and Technology 30: 118--123. Miskimmin BM, Rudd JWM, and Kelly CA (1992) Influence of dissolved organic carbon, pH and microbial respiration rates on mercury methylation and demethylation in lake water. Canadian Journal of Fisheries and Aquatic Sciences 49: 17--22. Moldovan M, Rauch S, Gomez M, Palacios MA, and Morrison GM (2001) Bioaccumulation of palladium, platinum and rhodium from urban particulates and sediments by the freshwater isopod Asellus aquaticus. Water Research 35: 4175--4183. Morales KH, Ryan L, Kuo TL, Wu MM, and Chen CJ (2000) Risk of internal cancers from arsenic in drinking water. Environmental Health Perspectives 108: 655--661. Morgan JJ and Stumm W (1991) Chemical processes in the environment, relevance of chemical speciation. In: Merian E (ed.) Metals and Their Compounds in the Environment, pp. 67--103. Weinheim: VCH. Mota AM, Cruz P, Vilhena C, and Concalves MLS (2005) Influence of the sediment on lead speciation in the Tagus estuary. Water Research 39: 1451--1460. Mukherjee A, Bhattacharya P, Shi F, et al. (2009) Chemical evolution in the high arsenic groundwater of the Huhhot basin (Inner Mongolia, PR China) and its difference from the western Bengal basin (India). Applied Geochemistry 24: 1835--1851. Muresan B, Cossa D, Coquery M, and Richard S (2008a) Mercury sources and transformations in a man-perturbed tidal estuary: The Sinnamary Estuary, French Guiana. Geochimica et Cosmochimica Acta 72: 5416--5430. Muresan B, Cossa D, Richard S, and Dominique Y (2008b) Monomethylmercury sources in a tropical artificial reservoir. Applied Geochemistry 23: 1101--1126. Nakata H, Shimada H, Yoshimoto M, et al. (2008) Concentrations and distribution of mercury and other heavy metals in surface sediments of the Yatsushiro Sea including Minamata Bay, Japan. Bulletin of Environmental Contamination and Toxicology 80: 78--84. Naranmandura H, Suzuki N, and Suzuki KT (2009) Reaction mechanism underlying the in vitro transformation of thioarsenicals. Toxicology and Applied Pharmacology 231: 328--335. Nath B, Berner Z, Chatterjee D, Mallik SB, and Stu¨ben D (2008a) Mobility of arsenic in West Bengal aquifers conducting low and high groundwater atsenic. Part II: Comparative geochemical profile and leaching study. Applied Geochemistry 23: 996--1011. Nath B, Stu¨ben D, Mallik SB, Chatterjee D, and Charlet L (2008b) Mobility of arsenic in West Bengal aquifers conducting low and high groundwater atsenic. Part I: Comparative hydrochemical and hydrogeological characteristics. Applied Geochemistry 23: 977--995. Newcomb WD and Rimstidt JD (2002) Trace element distribution in US groundwaters: A probabilistic assessment using public domain data. Applied Geochemistry 17: 49--57. Nickson RT, McArthur JM, Ravenscroft P, Burgess WG, and Ahmed KM (2000) Mechanism of arsenic release to groundwater, Bangladesh and West Bengal. Applied Geochemistry 15: 403--413. Nischwitz V, Kanaki K, and Pergantis SA (2006) Mass spectrometric identification of novel arsinotioyl-sugars in marine bivalves and algae. Journal of Analytical Atomic Spectrometry 21: 33--40. Nischwitz V and Pergantis SA (2006) Identification of the novel thio-arsenosugars DMThioAsSugarCarboxyl, DMTioAsSugarCarbamate and DMThioAsSugarAdenine in extracts of giant clam tissues by high-performance liquid chromatography online with electrospray tandem mass spectrometry. Rapid Communications in Mass Spectrometry 20: 3579--3585. Nixon S, Trent Z, Marcuello C, and Lallana C (2003) Europe’s Water: An Indicator Based Assessment. Copenhagen: European Environment Agency. NRC (National Research Council) (1999) Arsenic in Drinking Waters. Washington, DC: National Academic Press. NRC (National Research Council) (2000) Toxicological Effects of Methylmercury. Washington, DC: National Academy Press. Ochi T, Kita K, Suzuki T, Rumpler A, Goessler W, and Francesconi KA (2008) Cytotoxic, genotoxic and cell-cycle disruptive effects of thio-dimethylarsinate in cultured human cells and the role of glutathione. Toxicology and Applied Pharmacology 228: 59--67.

55

Oliveira V, Sarmiento AM, Gomez-Ariza JL, Nieto JM, and Sanchez-Rodas D (2006) New preservation method for inorganic arsenic speciation in acid mine drainage samples. Talanta 69: 1182--1189. Oremland RS, Dowdle PR, Hoeft S, et al. (2000) Bacterial dissimilatory reduction of arsenate and sulphate in meromictic Mono Lake, California. Geochimica et Cosmochimica Acta 64: 3073--3084. Ouddane B, Mikac N, Cundy AB, Quillet L, and Fischer JC (2008) A comparative study of mercury distribution and methylation in mudflats from two macrotidal estuaries: The Seine (France) and the Medway (United Kingdom). Applied Geochemistry 23: 618--631. Pak KR and Bartha R (1998) Mercury methylation by interspecies hydrogen and acetate transfer between sulfidogens and methanogens. Applied and Environmental Microbiology 64: 1987--1990. Pearson RG (1973) Hard and Soft Acids and Bases. Stroudsburh, PA: Dowden, Hutchinson and Ross. Peel K, Weiss D, and Sigg L (2009) Zinc isotope composition of settling particles as a proxy for biogeochemical processes in lakes: Insights from the eutrophic Lake Greifen, Switzerland. Limnology and Oceanography 54: 1699--1708. Perry WB (2008) Everglades restoration and water quality challenges in south Florida. Ecotoxicology 17: 569--578. Peters SC and Blum JD (2003) The source and transport of arsenic in a bedrock aquifer, New Hampshire, USA. Applied Geochemistry 18: 1773--1787. Peters SC, Blum JD, Klaue B, and Karagas MR (1999) Arsenic occurence in New Hampshire drinking water. Environmental Science and Technology 33: 1328--1333. Peters SC and Burkert L (2008) The occurrence and geochemistry of arsenic in groundwaters of the Newark basin of Pennsylvania. Applied Geochemistry 23: 85--98. Petrick JS, Ayala-Fierro F, Cullen WR, Carter DE, and Aposhian HV (2000) Monomethylarsonous acid (MMA(III)) is more toxic than arsenite in Chang human hepatocytes. Toxicology and Applied Pharmacology 163: 203--207. Petrou AL and Economou-Eliopoulos M (2009) The activation energy values estimated by the Arrhenius equation as a controlling factor of platinum-group mineral formation. Geochimica et Cosmochimica Acta 73: 1625--1636. Pickering WF (1986) Metal ion speciation in soils and sediments (a review). Ore Geology Reviews 1: 83--146. Pickhardt PC and Fisher NS (2007) Accumulation of inorganic and methylmercury by freshwater phytoplankton in two contrasting water bodies. Environmental Science and Technology 41: 125--131. Planer-Friedrich B, Franke D, Merkel B, and Wallschla¨ger D (2008) Acute toxicity of thioarsenates to Vibrio fischeri. Environmental Toxicology and Chemistry 27: 2027--2035. Planer-Friedrich B, London J, McCleskey RB, Nordstrom DK, and Wallschla¨ger D (2007) Thioarsenates in geothermal waters of Yellowstone National Park: Determination, preservation and geochemical importance. Environmental Science and Technology 41: 5245--5251. Planer-Friedrich B and Wallschla¨ger D (2009) A critical investigation of hydride generation-based arsenic speciation in sulfidic waters. Environmental Science and Technology 43: 5007--5013. Pokrovski GS, Borisova AY, Roux J, et al. (2006) Antimony speciation in saline hydrothermal fluids: A combined X-ray absorption fine structure spectroscopy and solubility study. Geochimica et Cosmochimica Acta 70: 4196--4214. Pokrovsky OS, Schott J, and Dupre B (2006) Trace element fractionation and transport in boreal rivers and soil porewaters of permafrost-dominated basaltic terrain in Central Siberia. Geochimica et Cosmochimica Acta 70: 3239--3260. Polya DA, Berg M, Gault AG, and Takahashi Y (2008) Arsenic in groundwaters of South-East Asia: With emphasis on Campodia and Vietnam. Applied Geochemistry 23: 2968--3251. Postma D, Larsen F, Hue NTM, et al. (2007) Arsenic in groundwater of the Red River floodplain, Vietnam: Controlling geochemical processes and reactive transport modeling. Geochimica et Cosmochimica Acta 71: 5054--5071. Potgeter H (1998) Entwicklung und Anwendung eines neuartigen Analysenverfahrens zur Bestimmung quecksilberorganischer Verbindungen in Sedimenten mit Hilfe eines gekoppelten SFC-AFS-Systems. PhD Study, University of Hamburg, 210pp. Qian J, Skyllberg U, Frech W, Bleam WF, Bloom PR, and Petit PE (2002) Bonding of methylmercury to reduced sulfur groups in soil and stream organic matter as determined by X-ray absorption spectroscopy and binding affinity studies. Geochimica et Cosmochimica Acta 66: 3873--3885. Qiu GL, Feng XB, Wang SF, and Shang LH (2005) Mercury and methylmercury in riparian soil, sediments, mine-waste calcines, and moss from abandoned Hg mines in east Guithou province, southwestern China. Applied Geochemistry 20: 627--638. Rader KJ, Dombrowski PM, Farley KJ, Mahony JD, and Di Toro TM (2004) Effect of thioarsenite formation on arsenic(III) toxicity. Environmental Toxicology and Chemistry 23: 1649--1654.

56

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

Raml R, Goessler W, and Francesconi KA (2006) Improved chromatographic separation of thio-arsenic compounds by reversed-phase high performance liquid chromatography-inductively coupled plasma mass spectrometry. Journal of Chromatography A 1128: 164--170. Raml R, Goessler W, Traar P, Ochi T, and Francesconi KA (2005) Novel thioarsenic metabolites in human urine after ingestion o fan arsenosugar, 20 ,30 dihydroxypropyl 5-deoxy-5-dimethylarsinoyl-beta-D-riboside. Chemical Research in Toxicology 18: 1444--1450. Raml R, Rumpler A, Goessler W, et al. (2007) Thiodimethylarsinate is a common metabolita in urine samples from arsenic-exposed women in Bangladesh. Toxicology and Applied Pharmacology 222: 374--380. Ramstedt M, Carlsson E, and Lo¨vgren L (2003) Aqueous geochemistry in the Udden pit lake, northern Sweden. Applied Geochemistry 18: 97--108. Rauch S and Morrison GM (1999) Platinum uptake by the freshwater isopod Asellus aquaticus in urban rivers. Science of the Total Environment 235: 261--268. Rauret G, Lopez-Sanchez JF, Sahuquillo A, et al. (1999) Improvement of the BCR three step sequential extraction procedure prior to the certification of new sediment and soil reference materials. Journal of Environmental Monitoring 1: 57--61. Ravichandran M (2004) Interactions between mercury and dissolved organic matter – a review. Chemosphere 55: 319--331. Ravizza G and Bothner MH (1996) Osmium isotopes and silver as tracers of anthropogenic metals in sediments from Massachusetts and Cape Cod bays. Geochimica et Cosmochimica Acta 60: 2753--2763. Redman AD, Macalady DL, and Ahmann D (2002) Natural organic matter affects arsenic speciation and sorption onto hematite. Environmental Science and Technology 36: 2889--2896. Reimann C and De Caritat P (1998) Chemical Elements in the Environment. Berlin: Springer. Richardson SD (2007) Water analysis: Emerging contaminants and current issues. Analytical Chemistry 79: 4295--4324. Richardson SD (2008) Environmental mass spectrometry: Emerging contaminants and current issues. Analytical Chemistry 80: 4373--4402. Richardson SD (2009) Water analysis: Emerging contaminants and current issues. Analytical Chemistry 81: 4645--4677. Rudnick RL and Gao S (2003) Composition of the continental crust. In: Holland HD and Turekian KK (eds.) Treatise on Geochemistry. Amsterdam: Elsevier. Rumbold DG, Fink LE, Laine KA, et al. (2002) Levels of mercury in alligators (Alligator mississippiensis) collected along a transect through the Florida Everglades. Science of the Total Environment 297: 239--252. Rumbold DG, Lange TR, Axelrad DM, and Atkeson TD (2008) Ecological risk of methylmercury in Everglades National Park, Florida, USA. Ecotoxicology 17: 632--641. Runnells DD, Shepherd TA, and Angino EE (1992) Metals in water: Determining natural background concentrations in mineralized areas. Environmental Science and Technology 26: 2316--2322. Rytuba JJ (2000) Mercury mine drainage and processes that control its environmental impact. Science of the Total Environment 260: 57--71. Sahuquillo A, Lopez-Sanchez JF, Rubio R, et al. (1999) Use of a certified reference material for extractable trace metals to assess sources of uncertainty in BCR threestage sequential extraction procedure. Analytica Chimica Acta 382: 317--327. Sako A, Lopes L, and Roychoudhury AN (2009) Adsorption and surface complexation modelling of palladium, rhodium and platinum in surficial semi-arid soils and sediment. Applied Geochemistry 24: 86--95. Sakurai T, Kaise T, and Matsubara C (1998) Inorganic and methylated arsenic compounds induce cell death in murine macrophages via different mechanisms. Chemical Research in Toxicology 11: 273--283. Salbu B and Steinnes E (1995) Trace Elements in Natural Waters. Boca Raton, FL: CRC Press. Salomons W (1995) Environmental impact of metals derived from mining activities: Processes, predictions, prevention. Journal of Geochemical Exploration 52: 5--23. Salomons W and Fo¨rstner U (1984) Metals in the Hydrocycle. Berlin: Springer. Sanders RD, Coale KH, Gill GA, Andrews AH, and Stephenson M (2008) Recent increase in atmospheric deposition of mercury to California aquatic systems inferred from a 300-year geochronological assessment of lake sediments. Applied Geochemistry 23: 399--407. Santore RC, Di Toro DM, and Paquin PR (2001) Biotic ligand model of the acute toxicity of metals, II: Application to acute copper toxicity in freshwater fish and Daphnia. Environmental Toxicology and Chemistry 20: 2397--2402. Scanlon BR, Nicot JP, Reedy RC, Kurtzman D, Mukherjee A, and Nordstrom DK (2009) Elevated naturally occurring arsenic in a semiarid oxidizing system, Southern High Plains aquifer, Texas, USA. Applied Geochemistry 24: 2061--2071.

Schlu¨ter K (1997) Sorption of inorganic mercury and monomethyl mercury in an ironhumus podzol soil of southern Norway studied by batch experiments. Environmental Geology 30: 266--279. Schmeisser E, Raml R, Francesconi KA, et al. (2004) Thio arsenosugars identified as natural constituents of mussels by liquid chromatography mass spectrometry. Chemical Communications 2004: 1824--1825. Schmitt D, Taylor HE, Aiken GR, Roth DA, and Frimmel FH (2002) Influence of natural organic matter on the adsorption of metal ions onto clay minerals. Environmental Science and Technology 36: 2932--2938. Schro¨der TJ, van Riemsdijk WH, van der Zee SEATM, and Vink JPM (2008) Monitoring and modelling of the solid–solution partitioning of metals and As in a river floodplain redox sequence. Applied Geochemistry 23: 2350--2363. Seddique AA, Masuda H, Mitamura M, et al. (2008) Arsenic release from biotite into a Holocene groundwater aquifer in Bangladesh. Applied Geochemistry 23: 2236--2248. Seddique AA, Masuda H, Mitamura M, et al. (2009) Reply to the comment on ‘‘Arsenic release from biotite into a Holocene groundwater aquifer in Bangladesh’’ by Hossain M Anawar and Martin Mihaljevic. Applied Geochemistry 24: 468--490. Shotyk W and Krachler M (2007a) Contamination of bottled waters with antimony leaching from polyethylene terephthalate (PET) increases upon storage. Environmental Science and Technology 41: 1560--1563. Shotyk W and Krachler M (2007b) Lead in bottled waters: Contamination from glass and comparison with pristine groundwater. Environmental Science and Technology 41: 3508--3513. Shotyk W and Krachler M (2009) Determination of trace element concentrations in natural freshwaters: How low is ‘‘low’’, and how low do we need to go? Journal of Environmental Monitoring 11: 1747--1753. Shotyk W, Krachler M, Aeschbach-Hertig W, Hillier S, and Zheng J (2010) Trace elements in recent groundwater of an artesian flow system and comparison with snow: Enrichments, depletions, and chemical evolution of the water. Journal of Environmental Monitoring 12: 208--217. Shotyk W, Krachler M, and Chen B (2006) Contamination of Canadian and European bottled waters with antimony from PET containers. Journal of Environmental Monitoring 8: 288--292. Shotyk W, Krachler M, Chen B, and Zheng J (2005) Natural abundance of Sb and Sc in pristine groundwaters, Springwater Township, Ontario, Canada, and implications for tracing contamination from landfill leachates. Journal of Environmental Monitoring 7: 1238--1244. Siciliano SD, O’Driscoll NJ, Tordon R, Hill J, Beauchamp S, and Lean DRS (2005) Abiotic production of methylmercury by solar radiation. Environmental Science and Technology 39: 1071--1077. Singer C, Zimmermann S, and Sures B (2005) Induction of heat shock proteins (hsp 70) in the Zebra mussel (Dreissena polymorpha) following exposure to platinum group metals (platinum, palladium and rhodium): Comparison with lead and cadmium exposures. Aquatic Toxicology 75: 65--75. Singh KP, Mohan D, Singh VK, and Malik K (2005) Studies on distribution and fractionation of heavy metals in Gomti river sediments – a tributary of the Ganges, India. Journal of Hydrology 312: 14--27. Smedley PL and Kinniburgh DG (2002) A review of the source, behaviour and distribution of arsenic in natural waters. Applied Geochemistry 17: 517--568. Smedley PL, Nicolli HB, Macdonald DMJ, Barros AJ, and Tullio JO (2002) Hydrogeochemistry of arsenic and other inorganic constituents in groundwaters from La Pampa, Argentina. Applied Geochemistry 17: 259--284. Smieja JA and Wilkin RT (2003) Preserveation of sulfidic waters containing dissolved As(III). Journal of Environmental Monitoring 5: 913--916. Sposito G (1989) The Chemistry of Soils. New York, NY: Oxford University Press. Squibb KS and Fowler BA (1993) The toxicity of arsenic and its compounds. In: Fowler BA (ed.) Biological and Environmental Effects of Arsenic, pp. 233--269. Amsterdam: Elsevier. SRU (Sachversta¨ndigenrat fu¨r Umweltfragen) (2008) Umweltgutachten 2008. Berlin: Umweltbundesamt. St. Louis VL, Rudd JWM, Kelly CA, Beaty KG, Flett RJ, and Roulet NT (1996) Production and loss of methylmercury and loss of total mercury from Boreal forest catchments containing different types of wetlands. Environmental Science and Technology 30: 2719--2729. Stauder S, Raue B, and Sacher R (2005) Thioarsenates in sulfidic waters. Environmental Science and Technology 39: 5933--5939. Steely S, Amarasiriwardena D, and Xing B (2007) An investigation of inorganic antimony species and antimony associated with soil humic acid molar mass fractions in contaminated soils. Environmental Pollution 148: 590--598. Strigul N, Koutsospyros A, Arienti P, Christodoulatos C, Dermatas D, and Braida W (2005) Effects of tungsten on environmental systems. Chemosphere 61: 248--258.

Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species

Stueben D, Berner Z, Chandrasekharam D, and Karmakar J (2003) Arsenic enrichment in groundwater of West Bengal, India: Geochemical evidence for mobilization of As under reducing conditions. Applied Geochemistry 18: 1417--1434. Stumm W and Morgan JJ (1996) Aquatic Chemistry, 3rd edn. New York, NY: Wiley. Styblo M, Serves SV, Cullen WR, and Thomas DJ (1997) Comparative inhibition of yeast glutathione reductase by arsenicals and arsenothiols. Chemical Research in Toxicology 10: 27--33. Suess E, Scheinost A, Bostick B, Merkel B, Wallschla¨ger D, and Planer-Friedrich B (2009) Discrimination of thioarsenites and thioarsenates by X-ray absorption spectroscopy. Analytical Chemistry 81: 8318--8326. Sulkowski M and Hirner AV (2006) Element fractionation by sequential extraction in a soil with high carbonate content. Applied Geochemistry 21: 16--28. Sunderland EM, Gobas FAPC, Branfireun BA, and Heyes A (2006) Environmental controls on the speciation and distribution of mercury in coastal sediments. Marine Chemistry 102: 111--123. Sures B, Singer C, and Zimmermann S (2006) Biological effects of palladium. In: Zereini F and Alt F (eds.) Palladium Emissions in the Environment: Analytical Methods, Environmental Assessment and Health Effects, pp. 489--499. Heidelberg: Springer. Swartz CH, Blute NK, Badruzzman B, et al. (2004) Mobility of arsenic in a Bangladesh aquifer: Inferences from geochemical profiles, leaching data, and mineralogical characterization. Geochimica et Cosmochimica Acta 68: 4539--4557. Tchounwou PB, Ayensu WK, Ninashvili N, and Sutton D (2003) Environmental exposure to mercury and its toxicopathologic implications for public health. Environmental Toxicology 18: 149--175. Tella M and Pokrovski GS (2009) Antimony(III) complexing with O-bearing organic ligands in aqueous solution: An X-ray absorption fine structure spectroscopy and solubility study. Geochimica et Cosmochimica Acta 73: 268--290. Templeton DM, Ariese F, Cornelis R, et al. (2000) Guidelines for terms related to chemical speciation and fractionation of elements: Definitions, structural aspects, and methodological approaches. Pure and Applied Chemistry 72: 1453--1470. Tessier A, Campbell P, and Bisson M (1979) Sequential extraction procedure for the speciation of particulate trace metals. Analytical Chemistry 51: 844--851. Thanabalasingam P and Pickering WF (1986) Arsenic adsorption by humic acid. Environmental Pollution 12: 233--246. Thomas DJ, Styblo M, and Lin S (2001) The cellular metabolism and systemic toxicity of arsenic. Toxicology and Applied Pharmacology 176: 127--144. Tomiyasu T, Matsuyama A, Eguchi T, Marumoto K, Oki K, and Akagi H (2008) Speciation of mercury in water at the bottom of Minamata Bay, Japan. Marine Chemistry 112: 102--106. Torky A-RW and Foth H (2007) Quecksilber. In: Dunkelberg H, Hartwig A, and Gebel T (eds.) Handbuch der Lebensmitteltoxikologie, pp. 803--850. Weinheim: WileyVCH. Ullrich SM, Tanton TW, and Abdrashitova SA (2001) Mercury in the aquatic environment: A review of factors affecting methylation. Critical Reviews in Environmental Science and Technology 31: 241--293. US EPA (2002) Current Drinking Water Standards. EPA 816-F-013. Van Geen A, Rose J, Thoral S, Garnier JM, Zheng Y, and Bottero JY (2004) Decoupling of As and Fe release to Bangladesh groundwater under reducing conditions. Part II: Evidence from sediment incubations. Geochimica et Cosmochimica Acta 68: 3475--3486. Wallschla¨ger D, Desai MVM, Spengler M, and Windmo¨ller CC (1998) How humic substances dominate mercury geochemistry in contaminated floodplain soils and sediments. Journal of Environmental Quality 27: 1044--1054. Wallschla¨ger D and London CJ (2008) Determination of methylated arsenic-sulfur compounds in groundwater. Environmental Science and Technology 42: 228--234. Wang S, Jia Y, Wang S, et al. (2009) Total mercury and methylmercury in water, sediments, and hydrophytes from the rivers, estuary, and bay along the Bohai Sea coast, northeastern China. Applied Geochemistry 24: 1702--1711. Warner KA, Roden EE, and Bonzongo J-C (2003) Microbial mercury transformation in anoxic freshwater sediments under iron-reducing and other electron-accepting conditions. Environmental Science and Technology 37: 2159--2165. Weber JH (1993) Review of possible paths for abiotic methylation of mercury(II) in the aquatic environment. Chemosphere 26: 2063--2077.

57

Wedepohl KH (1995) The composition of the continental crust. Geochimica et Cosmochimica Acta 59: 1217--1232. Wei C and Morrison GM (1994) Platinum in road dusts and urban river sediments. Science of the Total Environment 146/147: 169--174. Wenzel WW, Kirchbaumer N, Prohaska T, Stingeder G, Lombi E, and Adriano DC (2001) Arsenic fractionation in soils using an improved sequential extraction procedure. Analytica Chimica Acta 436: 309--323. Whalin L, Kim E-H, and Mason R (2007) Factors influencing the oxidation, reduction, methylation and demethylation of mercury species in coastal waters. Marine Chemistry 107: 278--294. White DE, Hem JD, and Waring GA (1963) Chemical composition of subsurface waters. In: Fleischer M (ed.) Data of Geochemistry, U.S. Geological Survey Professional Paper 440-F, 6th edn. Washington, DC: US Government Printing Office. WHO (World Health Organization) (1990) Methylmercury. International Program on Chemical Safety, Geneva, Switzerland. Environmental Health Criteria 101: 1--144. WHO (World Health Organization) (2001) Arsenic compounds. Environmental Health Criteria 224. 2nd edn. Geneva: World Health Organisation. Wiener JG, Krabbenhoft DP, Heinz GH, and Scheuhamer AM (2003) Ecotoxicology of mercury. In: Hoffman DJ, Rattner BA, Burton GA, and Cairns J (eds.) Handbook of Ecotoxicology, 2nd edn., pp. 409--463. Boca Raton, FL: Lewis Publishers. Wilson N and Webster-Brown J (2009) The fate of antimony in a major lower river system, the Waikato River, New Zealand. Applied Geochemistry 24: 2283--2292. Woshner VM, O’Hara TM, Eurell JA, et al. (2002) Distribution of inorganic mercury in liver and kidney of beluga and bowhead whales through autometallographic development of light microscope tissue sections. Toxicologic Pathology 30: 209--215. Wu Y, Wang SX, Streets DG, Hao JM, Chan M, and Jiang JK (2006) Trends in anthropogenic mercury emissions in China from 1995 to 2003. Environmental Science and Technology 40: 5312--5318. Xia K, Skyllberg UL, Bleam WF, Bloom PR, Nater EA, and Helmke PA (1999) X-ray absorption spectroscopic evidence for the complexation of Hg(II) by reduced sulphur in soil humic substances. Environmental Science and Technology 33: 257--261. Xu H, Allard B, and Grimvall A (1988) Influence of pH and organic substance on the adsorption of As(V) on geological material. Water, Air, and Soil Pollution 40: 293--305. Yoon S-J, Diener LM, Bloom PR, Nater EA, and Bleam WF (2005) X-ray absorption studies of CH3Hgþ-binding sites in humic substances. Geochimica et Cosmochimica Acta 69: 1111--1121. Yorifuji T, Tsuda T, Kashima S, Takao S, and Harada M (2010) Long-term exposure to methylmercury and its effects on hypertension in Minamata. Environmental Research 110: 40--46. Yorifuji T, Tsuda T, Takao S, and Harada M (2008) Long-term exposure to methylmercury and neurologic signs in Minamata and neighboring communities. Epidemiology 19: 3--9. Yoshinaga J (2003) Organolead compounds in the environment. In: Craig P (ed.) Organometallic Compounds in the Environment, 2nd edn., pp. 151--194. New York, NY: Wiley. Zhang H and Dill C (2008) Apparent rates of production and loss of dissolved gaseous mercury (DGM) in a southern reservoir lake (Tennessee, USA). Science of the Total Environment 392: 233--241. Zheng J, Shotyk W, Krachler M, and Fisher DA (2007) A 15,800-year record of atmospheric lead deposition on the Devon Island Ice Cap, Nunavut, Canada: Natural and anthropogenic enrichments, isotopic composition, and predominant sources. Global Biogeochemical Cycles 21: GB2027. 11pp. Zheng Y, Van Geen A, Stute M, et al. (2004) Geochemical and hydrogeological controls between shallow and deeper aquifers in two villages of Araihazar, Bangladesh: Implications for deeper aquifers as drinking water sources. Geochimica et Cosmochimica Acta 69: 5203--5218. Zimmermann S and Sures B (2004) Significance of platinum group metals emitted from automobile exhaust gas converters for the biosphere. Environmental Science and Pollution Research 11: 194--199.

3.03 Sources, Risks, and Mitigation of Radioactivity in Water D Crawford-Brown, University of Cambridge, Cambridge, UK & 2011 Elsevier B.V. All rights reserved.

3.03.1 3.03.2 3.03.3 3.03.4 3.03.5 3.03.6 3.03.7 References

Introduction Establishing Limits on the Risk from Radionuclides Specific Radionuclides of Interest Mitigation Methods Geographic Areas of Special Concern Measuring Radioactivity in Water Conclusions

3.03.1 Introduction Radionuclides occur in all water sources, whether surface water or aquifers. This ubiquity stems from their multiple origins, which include the natural composition of the Earth, with long-lived radionuclides such as elements of uranium left over from the initial creation of the Earth, and production by cosmic rays that continuously bombard the atmosphere. Add the emissions from nuclear power plants, the manufacture of nuclear fuel and weapons, and the treatment and disposal of radionuclides from uses such as medical equipment, and one finds radionuclides in essentially all materials with which water might come into contact (although naturally occurring radionuclides remain by far the largest contributor). A distinction is useful here before exploring the various aspects of radionuclides in water to be discussed in this chapter. There is a distinction between radioactivity, radionuclides, and radiation. Radioactivity refers to the ability of an atom to undergo radioactive decay, generally to a more energetically stable atom. The result of this decay is some form of radiation emerging from the atom, either as an electron (beta radiation), photon (gamma or X-ray, with gammas emerging from the nucleus of an atom and X-rays from the shells of electrons), or two neutrons and two protons bound together (an alpha particle). A radionuclide is one of the nuclides of a particular element (such as uranium) that is radioactive and, hence, emits radiation. All radioactive atoms transform eventually into a stable isotope of either the original or a different element. The unit of measure for radionuclides refers to the rate at which radioactive decays occur in a sample. This does not necessarily equal the rate at which radiations are being emitted, since more than one radiation can emerge from an atom undergoing decay. The historical unit of radioactivity is the curie or Ci, with 1 Ci being equal to 3.7 1010 disintegrations per second (the rate of radioactive decay in approximately 1 g of the radium with which Marie and Pierre Curie worked). More recently, the Ci was replaced with the Becquerel or Bq, equal to 1 disintegration per second. Note that 1 Ci is therefore equal to 3.7 1010 Bq. Neither the Ci nor the Bq depends on the type of radiation emitted; these depend only on the rate of radioactive decay underlying the emission. For radionuclides in water, the relevant measure of contamination is Becquerels per liter or Bq l1, although for historical reasons

59 59 62 63 64 65 67 67

some radionuclides such as radon continue to be reported in units of pCi l1(picocuries per liter, with 1 pCi being 1 1012 Ci or 0.037 Bq) (Table 1). This chapter focuses on radionuclides as a concern in water supplies, although that concern is for the most part related to the radiation emitted rather than to the radionuclide itself. While a radionuclide is also an element, and can therefore produce toxicity quite apart from its radioactive properties, regulatory controls are almost all based on the risk from radiation emitted by the radionuclide (uranium is toxic to organs such as the kidneys quite apart from its radiotoxicity, although it is the only regulated element where chemical toxicity, rather than radiation risk, dominates). In the case of either chemical or radiological toxicity, the dose is still the relevant metric for degree of toxicity. There is no difference in methodology for estimating risks from radiation received from radionuclides in water, food, soil, or air, or even from radiation received by procedures such as X-rays. As a result, it is necessary to consider terms and methods that may be unfamiliar to readers who have dealt primarily with chemical or biological contaminants in water.

3.03.2 Establishing Limits on the Risk from Radionuclides As mentioned, the risk from a radionuclide may be either from its action as an element, in which case all nuclides of that element when in purely elemental form produce the same effect because their shells of electrons are the same, or from Table 1 world

Radionuclides generally of greatest concern around the

Beta, gamma, and X-ray emitters

Alpha emitters

K-40 H-3 C-14 Ru-87

Ra-226 Ra-228 Po-210 Isotopes of U Isotopes of Th Rn-220 Rn-222

From De Zuane J (1997) Handbook of Drinking Water Quality. New York, NY: Wiley.

59

60

Sources, Risks, and Mitigation of Radioactivity in Water

exposure to the radiation emitted during radioactive decay. The exception is when the different nuclides of an element are found in different chemical forms. The first category of effects can be set aside for this discussion because they tend to occur at concentrations much higher than any found in water supplies. The exception is uranium in water, which produces toxicity in the kidneys due to its chemical properties. The second category contains two effects from the radiation, both of which result from damage to tissue as the radiation passes through the tissue. Deterministic effects are ones for which the severity of the tissue damage caused is proportional to the amount of radiation received (this amount is called the dose, which has a specialized meaning in radiation dosimetry as discussed later) and for which a threshold dose exists below which the effect does not occur. The severity of deterministic effects such as killing of the cells that line the gastrointestinal tract increases with the dose, and also depends on other factors, including the type, energy and dose rate of the radiation, and the sensitivity of the irradiated organ. Radiation protection for deterministic effects is built around the identification of a threshold dose, and maintaining water concentrations at levels low enough to remain below this threshold dose. As mentioned previously, uranium is the only radionuclide for which these deterministic effects are significant in regulatory decisions at environmental levels of exposure. The more common effect of radionuclides in water, and the one that drives regulatory decisions, is related to the ability of radiation to damage and alter the DNA and other structures such as membranes of cells. The result is primarily leukemia and other forms of cancer. These effects are called stochastic because radiation increases the probability of the effect (cancer), but not necessarily the severity of that effect. As a result, essentially all regulatory limits on exposure to radionuclides are intended to reduce the probability of cancer down to some level judged acceptable (the concept of acceptable risk in regulatory law). The stochastic effects of radiation are related to the ability of this radiation to cause ionization in cells, including in the DNA. This ionization breaks chemical bonds, such as the bonds holding base pairs together, which in turn can lead to mutations or re-arrangements of the DNA. The bonds are broken either directly by the radiation or through the production of chemical entities such as free radicals in the water of cells, which then act chemically on the bonds. Since the number of ionizations is related to the energy deposited in a cell by radiation, the dose of radiation is defined as the density of energy deposited in tissue as the radiation passes into or through the cells. The modern or SI unit of dose is the gray or Gy, and is equal to radiation that deposits an average of 1 J of energy per kilogram of tissue, or 1 J kg1. The historical unit was the rad, equal to 0.01 J kg1 or 0.01 Gy. While energy density is important in determining the probability of effect, studies in radiobiology show that this probability is also related to the spatial pattern of the ionizations produced. More densely packed ionizations (thus with shorter distances between each individual ionization) are more effective in damaging DNA and killing cells. Since gammas, X-rays, betas, and alphas have different spatial densities, measured by the quantity linear energy transfer, they also have different probabilities of producing an effect even if

they deliver the same dose in Gy. Each radiation is therefore assigned a quality factor Q, normalized to gamma radiations with a Q of 1. The relevant values of Q at present are 1 for X-rays, gamma rays, positrons, and electrons; 3 for neutrons below 10 keV and 10 for energies above; 10 for protons and singly charged particles of unspecified energy; and 20 for alpha and other multiply charged particles. The product of the dose (in Gy) and this QF (unitless) is the dose equivalent or effective dose, characterized as the rem in historical units (the product of dose in rads and the QF) or sievert (Sv) in modern units (the product of the dose in Gy and the QF). Choosing a limit for the allowed concentration of a radionuclide in water requires first establishing an allowed probability of effect – here of cancer. This allowed probability varies from country to country according to their internal laws governing environmental risks and the ways in which the precautionary principle is applied (Wiener, 2002), but as an example, consider the US where the limiting probability generally is 1 104 excess lifetime risk or probability of cancer. The International Commission on Radiological Protection (ICRP) (ICRP, 2007) recommends a limit on effective dose of 1 mSv yr1 from all combinations of radiation apart from natural background radiation and medical or therapeutic exposures (bearing in mind that many of these radionuclides are also part of the natural background). Using the conservative assumption of linearity of risk with dose equivalent, they estimate that the excess lifetime risk of cancer from a single exposure to 1 mSv is 7.3 105 (ICRP, 2007). A rate of dose equivalent equal to 1 mSv yr1 over a lifetime of 75 years would therefore produce a lifetime excess risk of cancer equal to about 6 103 (60 times the allowed risk in the US). Again, it has to be borne in mind that this risk estimate is produced through adoption of the conservative assumption of a linear no-threshold model of radiation induced cancer; the actual risk may be significantly lower. Restricting the lifetime excess risk of cancer to 1 104 as mentioned previously indicates that the maximum allowed rate of dose equivalent from radionuclides in water alone would be 1/60 mSv or 0.017 mSv yr1 (17 mSv yr1). Regulations on radionuclides in water are not, however, specified as limits on dose equivalent (mSv yr1). They are instead specified as limits on the concentration of the radionuclide in water (Bq l1), since that is what is mitigated in treatment of the water. It is necessary, therefore, to determine for each radionuclide the concentration that will produce the limit on dose equivalent. It is necessary here to distinguish between two broad categories of ways in which a radionuclide can produce a dose equivalent in the body when humans are exposed to water bearing that radionuclide.

•

External exposures take place when the radionuclide is present in the water and a person is exposed to the radiation either when standing near the water or when immersed in the water. In these cases, the radionuclide does not enter the body but the radiation it emits passes into the body and irradiates the tissue. The dose is then related to the concentration of the radionuclide in the water, the distance a person stands from the water, and the presence and nature of intervening materials. External exposures are generally not significant for regulatory decisions except in

Sources, Risks, and Mitigation of Radioactivity in Water

•

rare cases such as exposure to the core or holding pool of a nuclear reactor. Internal exposures take place when the radionuclide enters the body and irradiates the tissue by radiations emitted while the radionuclide is still inside. Primary exposure routes are then ingestion of water (almost always dominant in regulatory decisions on waterborne contaminants), ingestion of water used in cooking, or dermal absorption. A prominent exception to these general routes of exposure is radon, an inert gas whose risks from water are caused largely by the radon emanating into the air of a building during a shower, cooking, heating water, etc., followed by inhalation of the radon and its radioactive decay products. Internal exposures represent by far the most significant risk pathway for radionuclides in water at environmental levels.

Since internal exposures are more important in setting regulatory limits, methods are needed to estimate the dose equivalent produced by taking the radionuclide into the body (via ingestion in most cases, or inhalation in the case of radon). Both the ICRP and the National Radiological Protection Board (NRPB) have calculated dose conversion factors (DCFs) for radionuclides (see, e.g., NRPB, 1991; ICRP, 1996). These factors take into account the processes that move a radionuclide into and through the body (metabolic or pharmacokinetic models), and the processes that cause radioactive decay and hence irradiation of the tissues (dosimetric models). Each DCF is an estimate of the lifetime dose equivalent resulting from a single intake of 1 Bq of a given radionuclide, from which one can calculate the dose equivalent delivered over a lifetime from continuous intakes of 1 Bq yr1. In some cases, there are data on the movement of a particular radionuclide through the body, and these metabolic and dosimetric models can be based on these radionuclidespecific data. In many cases, however, such data do not exist and it is necessary to rely on data for the movement of other elements. Radionuclides that are in soluble form and chemically analogous to essential nutrients will tend to follow pathways through the body in a fashion similar to these analogs. For example, radionuclides of strontium (Sr), barium (Ba), radium (Ra), and calcium (Ca) all are bone-seekers similar to ionic calcium and so tend to exert their effect through irradiation of bone tissue such as the endosteal cells or marrow. Radioisotopes of cesium (Cs) and potassium (K) follow the general movement of ionic potassium and distribute to tissues throughout the body, irradiating the tissues uniformly. Radioisotopes of iodine (I) accumulate in the thyroid, leading to concerns for thyroid cancer. The lifetime excess probability of cancer, P, from water containing a radionuclide at a concentration of C (Bq l1) may be calculated as

P ¼ C  IR  DCF  SF  365  75

ð1Þ

where IR is the intake rate of water (l d1); DCF is the dose conversion factor mentioned previously (mSv Bq1); SF is the slope factor (probability of cancer per mSv, this is 7.5 105 mSv1 as mentioned previously based on the ICRP recommendations); 365 is the number of days per year and 75

61

is the mean number of years of life. The allowed concentration of the radionuclide in water may then be found by rearranging Equation (1) to yield

C ¼ P=ðIR  DCF  SF  365  75Þ

ð2Þ

As an example of a regulatory calculation of allowed concentration in water (generally called the maximum allowed concentration (MAC)), assume the allowed lifetime excess probability of cancer is 1 104; the intake rate of water is 2 l d1 (regulatory calculations use this value because it is conservative or protective of health, being in the upper few percent of the probability density function for rates of water intake); the DCF is 1 105 mSv Bq1; and the value of the slope factor SF is 7.5 105 mSv1. Using Equation (2),

MAC ¼ 10 4 =ð2  1  10 5  7:5  10 5  365  75Þ ¼ 2:4 Bq l 1

ð3Þ

Note also that this radionuclide produces an annual dose equivalent of

DE ¼ 2:5  10 5  2  1  365 ¼ 1:8  10 2 mSv yr 1

ð4Þ

or an excess lifetime probability of cancer of 1 104. By way of comparison, the sum of all natural sources of radiation worldwide is approximately 2.4 mSv yr1 (UNSCEAR, 2000). About one-third of this total dose equivalent is due to external radiation (terrestrial plus cosmic); the other two-thirds are due to the inhalation and ingestion of radionuclides in air, water, and food. Clearly, the regulatory limits are designed to produce dose equivalents that are a small fraction of that from natural background radiation. There are a few assumptions underlying the calculations above, the one of most interest here being an assumption that the probability of cancer is directly proportional to the dose equivalent down to the lowest values of dose equivalent. The main epidemiological studies on which radiation risks have been based are in populations with high levels of dose, the most important being survivors of the Hiroshima and Nagasaki bombs, patients receiving high dose equivalents for medical or therapeutic reasons and occupationally exposed workers such as uranium miners, radium-dial painters, and radiologists (National Research Council, 2005). The risk at low levels of dose equivalent is extrapolated from these epidemiological studies using a linear, no-threshold model as a conservative assumption to provide a margin of safety in regulations. These same epidemiological populations received their radiation over a short period of time, and so their dose rates were high. It is known that higher dose rates are generally more effective than low dose rates in producing cancer, the possible exception being for alpha particles emitted by radionuclides such as radium, and this dose–rate effect is not reflected in regulatory calculations of risk. This is again a conservative and health-protective approach to deal with the uncertainty in extrapolating to the conditions found in environmental exposures to radionuclides.

62

Sources, Risks, and Mitigation of Radioactivity in Water

Finally, regulators must also determine how to treat the fact that water may not be the only route by which a given radionuclide irradiates the body. The same radionuclide may be present in air, food, soil, etc. To account for this, many regulatory bodies estimate the fraction of total dose equivalent contributed by water, and reduce the allowed concentration accordingly. In the previous example, the allowed concentration was 2.5 105 Bq l1. If studies show that exposure to water accounts only for half of the total annual dose equivalent from this radionuclide, the regulator might reduce the allowed concentration by half to 1.2 105 Bq l1. The result is a total risk (water plus other routes combined) that is below the allowed or acceptable risk level. This relative source contribution calculation is not, however, uniformly applied across regulatory bodies around the world, or even across offices of the same regulatory body (it is not, e.g., applied in decisions on radionuclides in air in the US Environmental Protection Agency).

3.03.3 Specific Radionuclides of Interest There are hundreds of radioisotopes of elements. Only a handful, however, are found routinely in water supplies at a concentration sufficient to produce significant risks, and hence are of regulatory interest. The major ones are described here. To broaden the discussion beyond the US regulatory system (used in Section 3.03.2), consider the limits on specific waterborne radionuclides for drinking water, published by the World Health Organization in chapter 9 of their Guidelines for Drinking Water Quality (volume 1, 2008; see table 9.3) (Table 2). In addition, many countries have established limits on exposure to mixtures of radionuclides that act on the same tissue by the same mechanism, or because measurement methods do not distinguish easily between the sources of a given kind of radiation (such as alpha particles). In the US, some of the more important mixtures are shown below. In these cases, the regulatory limit is a maximum contaminant level (MCL), essentially the same as an MAC.

•

•

Combined radium-226 and radium-228 has been assigned a total MCL of 5 pCi l1 (or 0.19 Bq l1). This combination of radionuclides is found in most water supplies that come into contact with rocks containing radium from the uranium and thorium decay chains. The limit of 5 pCi l1 (or 0.19 Bq l1) was selected because the maximally allowed level of risk implied by regulatory limits would be attained if the mixture actually contained pure Ra-226 or Ra-228. Gross alpha has been assigned a total MCL of 15 pCi l1 (or 0.6 Bq l1) after subtracting out radon and isotopes of uranium. While this MCL does not exclude radionuclidespecific MCLs, measurement methods are simple for total alpha emissions but more complex for specific alpha emitters, generally requiring chemical separation. Note that even if a sample passes this gross alpha test, it is still necessary for it to pass the combined Ra-226 and Ra-228 test because water containing either of these radionuclides at a

Table 2 The MAC values for the most significant radionuclides as determined by the World Health Organization (2008) Americium-241 Barium-140 Bismuth-210 Bromine-82 Calcium-45 Calcium-47 Carbon-14 Cesium-131 Cesium-134 Cesium-136 Cesium-137 Chromium-51 Cobalt-57 Cobalt-58 Cobalt-60 Iodine-125 Iodine-129 Iodine-131 Iron-55 Iron-59 Lead-210 Manganese-54 Mercury-197 Mercury-203 Neptunium-239 Phosphorus-32 Plutonium-238 Plutonium-239 Plutonium-240 Plutonium-241 Polonium-210 Radium-224 Radium-226 Radium-228 Ruthenium-103 Ruthenium-106 Sodium-22 Strontium-85 Strontium-89 Strontium-90 Technecium-99 Thorium-228 Thorium-230 Thorium-232 Thorium-234 Tritium Uranium-234 Uranium-235 Uranium-238 Zinc-65 Zirconium-95

•

1 100 100 100 100 100 100 1000 10 100 10 10 000 1000 100 100 10 1 10 1000 100 0.1 100 1000 100 100 100 1 1 1 10 0.1 1 1 0.1 100 10 100 100 100 10 100 1 1 1 100 10 000 1 1 10 100 100

Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1 Bq l1

concentration of 15 pCi l1 (or 0.6 Bq l1) would produce an unacceptable level of risk. Combined uranium isotopes have been assigned an MCL of 30 mg l1 based on chemical toxicity (primarily to the kidneys). Note that the MCL is unit of mass rather than radioactivity (with 1 mg l1 being equal to approximately 12.3 mBq l1), reflecting the chemical toxicity rather than radiological risk as the primary cause of concern. It

Sources, Risks, and Mitigation of Radioactivity in Water

does not matter which mixture of isotopes of uranium is present in the sample because all have the same chemical toxicity. Radon (specifically Rn-222) presents a special challenge throughout the world, because the risks are generally much higher than for other radionuclides in water and also because the route of exposure is dominated by emanation into the air of a building rather than through direct ingestion. Empirical studies and the results of modeling of indoor air exchange indicate that the ratio of the concentration of radon in air (Bq l1) over that in water (Bq l1) due to the waterborne radon alone is approximately 1 104 (i.e., 10 000 Bq l1 in water produces 1 Bq l1 in air). The risk from the radon emanated out of the water and into the air during activities in the home (showers, washing clothes, etc.) is almost a factor of 50 higher than for direct ingestion of the radon in the water (National Research Council, 1999). As a result, regulatory limits on radon in water are based on the risks imposed by this emanation and the subsequent inhalation of radon and its radioactive decay products. Mitigation is accomplished either by removing the radon from the water before it enters the home air, or by removing the radon from the air once it has been emanated. Regulatory control is complicated further by the fact that the relative source contribution for radon in water is quite small. Radon enters the air of a building by many routes, only one of which emanates from water. The risk from radon emanated from water is only a few percent of the risk from radon that enters the air from rock underlying a building or from the building materials themselves. This leaves regulators in the position of finding that radon in water poses a level of risk much higher than for most of the other substances, including radionuclides and chemicals, currently considered in regulations. Not controlling radon in water would appear to weaken the argument for controlling these other contaminants. Removing radon from water, however, will reduce the risk to health from radon in the home by only a percent or two, and hence is not a cost-effective way to protect the public health against the total risk of environmental radon. This difficulty has been at the heart of problems in developing a regulation for radon in water in the United States, where the regulatory process has been stalled for decades as this comparative risk issue is debated. How can this challenge be resolved? The approach of the US Environmental Protection Agency, still under development, has been to propose two limits (USEPA, 2009). A regulatory limit (MCL) of 300 pCi l1 (or 11 Bq l1) is likely to become the primary standard. Even this concentration will produce a lifetime excess cancer risk above the value of 1 104 mentioned previously, being closer to 2–3 104, although through rounding one could argue that the risk is right at 1 104. If a state chooses, however, it can tackle the problem of radon primarily through a program of control of radon in air, such as increased ventilation in buildings or sealing of the ground, so radon cannot enter the building from nonwater sources. If such a program were put in place, the regulatory limit on radon in water becomes 4000 pCi l1 (150 Bq l1) in recognition of the low value of the relative source contribution for radon in water compared to other routes. Therefore,

63

state level regulators are being encouraged to adopt the most cost-effective mitigation strategy across multiple exposure pathways, the first such regulatory approach in the US. As a comparison, consider the standard being developed by the European Commission EC Recommendation 2001/928/ Euratom: For public water supplies, if the radon concentration exceeds 1000 Bq l1, remediation is considered justified. Where the radon level is above 100 Bq l1 but below 1000 Bq l1, the local authority must consider whether this poses a risk to human health. If it is concluded that such a risk exists, then remedial action should be considered. For private water supplies, where water is found to have levels of radon in excess of 1000 Bq l1 remediation of the supply should be considered.

Note that the proposed US standard is far below – or more stringent than – that being considered in the EC, in large measure because the US standard is based primarily on the calculation of excess lifetime risk where the EC standard is related more directly to the feasibility of mitigation. The risks from radionuclides in drinking water are dominated by the alpha emitters, especially Ra-226, Ra-228, natural U (a mixture of uranium isotopes with primordial origins), and Rn-222. The relative risks from these four radionuclides can be seen from data produced in the US (Milvy and Cothern, 1990) based on a national survey of ground- and surface-water supplies; similar results would be obtained in most other countries. Data are summarized in Table 3. From these data, it can be seen that radon contributes almost 98% of the total risk from these four radionuclides in the case of groundwater supplies; these supplies are examined here because the concentration in groundwater is generally quite a bit higher than in surface water. Note also that the cost of mitigation, measured as millions of dollars spent per cancer averted through mitigation varies considerably, from a low of $5 M for radon to a high in excess of $10 000 M for uranium (this is largely because U is mitigated chiefly due to its toxic effect on kidneys rather for its carcinogenicity). Also, bear in mind that the predicted savings in life – through cancer cases averted – are purely the results of the theoretical application of the linear nonthreshold model. While the costs of treatment are well established, the benefits through reduced risk remain speculative at present.

Table 3 Comparison of mean concentration in groundwater, lifetime risk, and cost per cancer case averted for isotopes of radium, uranium, and radon Radionuclide

Ra-226 þ Ra228 Natural uranium Rn-222

Mean concentration (pCi l1)

Lifetime risk

$M per cancer averted

1.1

1  105

20

1.2

2  106

10 000

4  104

5

600

64

Sources, Risks, and Mitigation of Radioactivity in Water

3.03.4 Mitigation Methods Many of the methods to treat water for other elements are effective at reducing radionuclides as well – and hence reducing the risks posed by these radionuclides. In fact, there is no difference in methods used to treat water for the radioactive or stable form of the same element. As a result, the treatment methods listed below will be familiar to anyone who has been engaged in removing chemicals and turbidity from water. To ensure confidence in removal of radionuclides, methods should conform to the recommendations of the NSF International American National Standards Institute (ANSI).

•

•

•

•

•

•

Filtration involves passing the water through a filter with suitable pore size. It has been used routinely to treat for radium, with more than 50% of the radium removed if it is in colloidal form. A special case is the use of activated carbon as the filter medium, which can remove more than 90% of radon in water, assuming the filter medium is periodically cleaned. Nanofiltration membranes have been reported to achieve significantly higher removal efficiencies in laboratory testing (Annanma¨ki, 2000), although their expense has kept them from wide application to date. Ion exchange, where the water flows through resin granules that act as a softener. With maintenance of the system, including regeneration of the resins, 90% of radionuclides can be removed. This is the approach often found in small water systems as the system is cost effective even at this scale. Lime softening, in which lime is added to water and isotopes such as radium settle out. This approach can remove up to 90% of the radium. This method has also been effective in removing isotopes of iodine, strontium, and uranium. An advantage is that such method is often employed anyway to treat hard water. Preformed hydrous manganese oxide filtration is slightly less effective than some of the methods above. The addition of potassium permanganate and manganese sulfate to water before it is filtered increases the removal of radium from the value of 50% noted in the first bullet to up to 80%. This can, however, be an inexpensive approach if filters are already in place. Reverse osmosis, in which water is passed under pressure through a semi-permeable membrane. Up to 98% of many radionuclides (including radium and uranium) can be removed by this approach. Isotopes of cesium, strontium, and iodine are also removed, although with slightly lesser efficiency. This approach, however, removes 20–30% of water from the system, which can be unacceptable in areas of a country already facing a shortage of water. Aeration is used primarily for radon. Since radon is an inert gas, it is stripped easily from the water during aeration, with a well-maintained system capable of removing more than 90% of the radon. Note that the effectiveness of aeration is related to two other aspects of radon risk: (1) radon concentrations in surface water are generally low due to the ability of the radon to diffuse to the surface of these supplies and escape to the air, so only groundwater supplies are of interest and (2) radon is readily emanated from water into the air of homes, with more than 90% emanated

during cooking and more than 50% during a shower or pouring water from a tap. Since isotopes of radium have received particular attention (second only to radon), a large number of specialized methods to remove it from water have been assessed (Clifford, 1990). The more effective methods have proven to be

• • • • • • • •

combining radium with iron removal, which has efficiencies of slightly less than 40%; lime softening, with removal efficiencies of up to 95%; sodium ion exchange softening, with removal efficiencies of up to 98%; weak acid cation exchange, with efficiencies of up to 95%; reverse osmosis hyperfiltration, with removal efficiencies of up to 96%; manganese dioxide filters, with removal efficiencies of up to 97%; absorption onto a radium selective complexer, with removal efficiencies of up to 99%; and absorption onto barium sulfate impregnated alumina, with removal efficiencies of up to 95%.

As can be seen, some of the methods above are generic and provide radium removal as a side benefit to treatment for other contaminants, whereas others are tailored to removal of radium. An issue to consider with all of the methods aside from aeration is the disposal of the removal medium. The reduction in radioactivity in the water is accompanied by a similar increase in the radioactivity in the medium. This medium may, therefore, be subject in some cases to regulations on transport and disposal of radioactive materials. It should not be assumed that these products can be disposed of at landfills intended for municipal waste. Smaller water supplies in particular face issues of the cost of treatment of the system itself, which can add significant costs to the water supplied to consumers. In these areas, it has become common to consider point-of-use technologies as an alternative to system treatment. The most common method by far is activated carbon filtration placed on the tap or at the point where the pipe enters a building. These systems can in fact be more cost effective than treatment at the source, again for smaller water systems. It is, however, at times less effective as a public health measure because the owners of buildings forget to replace the filters on a regular schedule, allowing the buildup of materials, including the radionuclide, the eventual breakthrough of the radionuclide, and the reduction of filter efficiency, as well as the possibility of microbial growth in the filters. Efficiencies for point-of-use and point-of-entry treatment technologies are similar to those found in municipal-scale treatment. One such system applied routinely – second only to granular activated carbon – is water softening, involving ion-exchange technology with a rejuvenating solution (the particular solution depends on the contaminant, which can be problematic if the home owner does not know which rejuvenator to use with which contaminant). For example, uranium removal requires a strongly basic solution, while radium requires an acid-rejuvenating solution. In addition, softeners rejuvenated with common salt (sodium chloride)

Sources, Risks, and Mitigation of Radioactivity in Water

could add a significant amount of sodium into the water. As in many cases of water treatment, therefore, there is a risk-risk trade-off to be considered

3.03.5 Geographic Areas of Special Concern As mentioned in the case of radon, highest concentrations of naturally occurring radionuclides are found in groundwater, especially in deep wells drilled into aquifers with elevated mineralization of radionuclides. Although these minerals are dissolved slowly, groundwater can be in contact with the surrounding rock for hundreds or thousands of years. Concentrations will, however, be highly variable in space, albeit less so in time, although seasonal changes in groundwater flow can affect the radionuclide concentration because it affects the contact time with rock, and radon can be affected by seasonal variations in pressure (in these cases, concentration can vary by a factor of 2–3 throughout the seasons). Concentrations are not necessarily correlated with surface geology, although radon in water does correlate well with the presence of aquifers within granitic rocks. Concentrations of radionuclides are affected by the composition of the underlying bedrock, and by the particular physical and chemical conditions in the aquifer. The result can be radionuclide concentrations that vary significantly in two wells located just a few meters apart, requiring careful consideration of the location and number of samples drawn from an aquifer in getting an accurate estimate of average concentration to which a population might be exposed. An additional complication with respect to radionuclides in the natural radioactive decay chains such as that of uranium (U-238) is the degree of secular equilibrium reached. In a decay chain, all radionuclides in a sample eventually reach the same level of activity (Bq) so long as there is no process removing any of the radionuclides from the sample. However, this equilibrium can take thousands or millions of years to achieve depending on the half-lives of the radionuclides (the half-life is the time required for half of a given radionuclide to decay). Since a long-lived radionuclide such as isotopes of radium or uranium leached from the rock and into the water may be withdrawn for use before equilibrium is reached, the residence time of water in an aquifer can be a strong determinant of total radioactivity present from the decay products of the original, leached radionuclide. This issue is not significant for radon, however, because it has a very short halflife of less than 4 days, well below the residence time in essentially any aquifer. Still, radon cannot be assumed to be in equilibrium with the radium in the water, since radon is an inert gas and can enter water from the surrounding rock above and beyond the radon produced by the decay of radium in the water. As a result, there can be wide variations in the relative concentrations of radionuclides in a given water sample (Figure 1). Even natural radionuclides can enter water through human activities that perturb the environment. This is particularly true of uranium mining and milling operations, which move these radionuclides from the subsurface to the surface, where they can contaminate surface waters at levels much higher than would be expected from natural processes. Areas that contain

65

such mining operations will, therefore, generally show waterborne radionuclides at elevated levels, especially isotopes of radium and uranium. The presence of nuclear facilities may increase the level of radionuclides in water supplies. This is due to the operation of the facility, which potentially releases fission products of uranium, tritium, and activated corrosion products (often metals irradiated by neutrons of the facility) through the cooling water, although at low concentrations; due to the generation of low-level waste discharged to landfills and subsequently leached to surface- and groundwater; and due to the radioactive products at the backend of the fuel cycle, including high-level radioactive waste stored in repositories in geological structures. For a more complete review, the reader is referred to UNSCEAR (2000). Fallout from nuclear weapons testing contributes to waterborne radionuclide concentrations today, despite a ban on atmospheric testing for several decades. The radionuclides of most interest in monitoring programs have been H-3 (tritium), C-14, Sr-90, and Cs-137. Since these radionuclides were released initially to the air, they tend to be found almost entirely in surface waters, and especially waters where the radionuclides settled in to the sediment and can be re-entrained during storms. Fallout radionuclides percolating into soils or sediments will tend to bind to grains in the soil and not reach groundwater supplies. These radionuclides are now ubiquitous, but the concentrations in water are highest near the points of testing. A potentially challenging feature of the geographic distribution of radionuclides in water is illustrated by radon. Drawing on the most recent large-scale assessment of the occurrence of radionuclides in US water supplies, Longtin (1990) divided the supplies into those serving 1000 or more people and those serving fewer than 1000 people. The national average concentration for the larger supplies was 8.9 Bq l1, while that for the smaller supplies was 28.9 pCi l1. The difference is largely due to the rural nature of populations served by small supplies, where groundwater is a more common source. This difference means that the percentage of supplies requiring mitigation, and the degree of mitigation required, are significantly higher for smaller supplies than larger ones. Since there is economy of scale in mitigation, and since smaller supplies tend to serve poorer populations, these data demonstrate that the costs associated with mitigation of radionuclides, as measured by increased water bills, are higher for the poorer populations served by small supplies.

3.03.6 Measuring Radioactivity in Water Methods for measuring radioactivity in water samples generally rely on detection of the radiations (for a good review of methods, see Knoll (2000)). For photon emitters, the sample may be placed on a scintillation counter (NaI is common) or GeLi system and the spectrum is obtained. Each radionuclide has a unique signature gamma or X-ray, and the spectrum from the detector can be used to identify the specific radionuclides by their signature peaks in the spectrum. The amount of the specific radionuclide in the sample is then proportional to the area under the peak. Such an approach requires

206Pb

(stable)

(22.6 y)

(5.01 d)

210Pb

(19.7 m) (26.9 m)

214Pb

210Bi

212Po

215Po

207TI

(4.77 m)

(3.05 m)

(36.1 m)

211Pb 208TI

200Pb

(stable)

212Pb

(10.6 h)

(stable)

207Pb

211Bi

(2.14 m)

(1.78 ms) 212Bi

(3E-7 s)

Beta decay

Alpha decay

219Rn

(3.96 s)

(1.01 h)

(0.15 s)

(138.4 d)

(1.6E-4 s)

214Bi

(3.04 m)

216Po

210Po

214Po

(55.6 s)

(3.823 d)

Radionuclide (half-life)

223Ra

(11.4 d)

224Ra

220Rn

218Po

227Th

(18.7 d)

(3.66 d)

222Rn

228Ra

(5.75 a)

226Ra

(1.6E3 a)

227Ac

(21.8 a)

220Ac

(24.1 d)

231Th

(1.06 d)

(6.15 h)

220Th

(1.91 a)

232Th

(1.4E10 a)

230Th

(7.5E4 a)

234Th

231Pa

(3.3E4 a)

234P

(7E8 a)

235U

235-Uranium decay series

(6.69 h)

234U

(2.45E5 a)

238U

(4.47E9 a)

232-Thorium decay series

Figure 1 The three primary natural radioactive decay chains found in water supplies. Reproduced from the public domain USGS website at http://gulfsci.usgs.gov/tampabay/data/2_biogeochemical_cycles/ radionuclides.html.

Tl

Pb

Bi

Po

Rn

Ra

Ac

Th

Pa

U

238-Uranium decay series

Sources, Risks, and Mitigation of Radioactivity in Water

a detector, a multichannel analyzer, and calibration samples of known activity to establish the efficiency of detection. The calibration samples are placed on the detection system both to ensure the peaks appear in anticipated channels of the multichannel analyzer, and to develop the ratio of peak area to activity. For beta emitters, the most common method is scintillation counting. The water is dissolved into, or at least mixed into, a scintillation cocktail in a vial. Radiations emerging from the radionuclides excite the molecules of the scintillation cocktail, causing cascades of photons. These photons are detected by a scintillation counter, with the number of pulses being related to the activity in the sample. The systems require use of quench curves to account for absorption of scintillations by materials present in the vials, such as turbidity. The method may also be applied to radon in water; in fact, because radon is an inert gas, scintillation counting is usually the preferred method of analysis. Detection of most alpha emitters is more complex due to sample preparation. Most of the methods involve concentration of the radionuclides through precipitation, often with a carrier agent such as barium or Fe(III) salts followed by solvent extraction. The resulting material can then be plated and the alpha emissions measured in a solid-state detector. In most cases, the extracted and plated sample is allowed to stand for a period of time to allow re-growth of the daughter products back into equilibrium, increasing the detection efficiency and accuracy while significantly improving the detection limit. Uranium may be measured through alpha spectrometry described above, but it is more common to measure it through fluorometry, and more recently by inductively coupled plasma (ICP) mass spectrometry. The former is either traditional fluorometry or a laser kinetic phosphorescence method. Neither of the methods can determine the specific isotopes of uranium, but do provide measurements of the total uranium present in the sample. This is not as significant a limitation as it would be for the isotopes of radium, since the health effects and regulatory limits for uranium are down primarily to the chemical toxicity for which separation into the contributions from different isotopes is not important. If for any reason it is important to distinguish between these isotopes, alpha spectroscopy or ICP mass spectrometry is required.

3.03.7 Conclusions Radioactivity is found essentially in any water sample, whether of groundwater or surface water. This is because the sources are numerous, not least of which is the primordial composition of the earth underlying or surrounding the body of water. Wellestablished procedures have been developed, however, to assess the concentrations of most radionuclides in water samples, to calculate the health risks, and to reduce this risk to acceptable levels through mitigation. In many cases, these mitigation methods may be adjuncts to measures that must be taken to reduce other water problems such as turbidity. From the perspective of public health, radon is by far the most significant contaminant of water supplies, especially those drawn from groundwater aquifers. The health risk of radon in water alone is larger than the sum of the risks from

67

the large majority of other waterborne contaminants. Therefore, there has been significant regulatory attention directed toward radon in the world of water science and regulation. While radioactivity is ubiquitous, there remain geographic areas known to be associated with higher levels of radioactivity in water. In regards to naturally occurring radionuclides, attention is directed primarily to groundwater withdrawn from aquifers surrounded by granitic rock, with slow water movement and hence large contact times. In regards to artificial radionuclides, or to artificial enhancement of natural radioactivity, attention is directed to uranium mines and mills; facilities involved in the nuclear fuel cycle; and to poorly managed landfills with low-level radioactive waste – especially where there is disposal of radionuclides from medicine or experiments. Despite the potential health risks posed by radionuclides in water, this is the area of water science in which more is known about the quantification and control of risks than almost any other contaminant. Through careful application of the principles discussed here, risks from waterborne radionuclides can be kept below acceptable levels.

References Annanma¨ki M (ed.) (2000) Treatment techniques for removing natural radionuclides from drinking water. Final Report of the TENAWA Project, STU K-A16 9. Helsinki: STUK. Clifford DA (1990) Removal of radium from drinking water. In: Cothern CR and Rebers PA (eds.) Radon, Radium and Uranium in Drinking Water, pp. 225--247. Chelsea, MI: Lewis. De Zuane J (1997) Handbook of Drinking Water Quality. New York, NY: Wiley. Health Canada (1995) Radiological characteristics. http://www.hc-sc.gc.ca/ewh-semt/ pubs/water-eau/radiological_characteristics/index-eng.php (accessed April 2010). ICRP (International Commission on Radiological Protection) (1996) Age-Dependent Doses to Members of the Public from Intake of Radionuclides: Part 5. Compilation of Ingestion and Inhalation Dose Coefficients, Annals of the ICRP, vol. 26(1–3), ICRP Publication 72. Oxford: Pergamon. ICRP (International Commission on Radiological Protection) (2007) Recommendations of the International Commission on Radiological Protection. ICRP Publication 103. Oxford: Pergamon. Knoll G (2000) Radiation Detection and Measurement, 3rd edn. New York, NY: Wiley. Longtin J (1990) Occurrence of radionuclides in drinking water, a national study. In: Cothern CR and Rebers PA (eds.) Radon, Radium and Uranium in Drinking Water, pp. 97--140. Chelsea, MI: Lewis. Milvy P and Cothern CR (1990) Scientific background for the development of regulations for radionuclides in drinking water. In: Cothern CR and Rebers PA (eds.) Radon, Radium and Uranium in Drinking Water, pp. 1--16. Chelsea, MI: Lewis. National Research Council (1999) Biological Effects of Ionizing Radiation VI: The Health Effects of Exposure to Indoor Radon. Washington, DC: National Academies Press. National Research Council (2005) Biological Effects of Ionizing Radiation VII. Washington, DC: National Academies Press. NRPB (National Radiological Protection Board) (1991) Committed Equivalent Organ Doses and Committed Effective Doses from Intakes of Radionuclides. Chilton, NRPB R-245. London: HMSO. UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation) (2000) Sources, Effects and Risks of Ionizing Radiation. New York, NY: United Nations. USEPA (US Environmental Protection Agency) (2009) Fact Sheet on Radon and Drinking Water. http://www.epa.gov/safewater/radon/qa1.html (accessed April 2010). Wiener J (2002) Precaution in a multi-risk world. In: Paustenbach (ed.) Human and Ecological Risk Assessment, pp. 1509–1532. New York, NY: Wiley. World Health Organization (2008) Guidelines for Drinking Water Quality, vol. 1, ch. 9. Geneva: WHO.

3.04 Emerging Contaminants K Ku¨mmerer, Leuphana University, Lu¨neburg, Germany & 2011 Elsevier B.V. All rights reserved.

3.04.1 3.04.2 3.04.3 3.04.4 3.04.5 3.04.6 3.04.6.1 3.04.6.2 3.04.6.2.1 3.04.6.2.2 3.04.6.3 3.04.7 3.04.8 3.04.9 3.04.10 3.04.11 3.04.12 3.04.13 3.04.14 3.04.15 3.04.15.1 3.04.15.2 3.04.15.3 3.04.16 3.04.17 3.04.18 References

Introduction General Aspects: What Are the Emerging Contaminants and Micro-Pollutants? Parent Compounds, Metabolites, and Transformation Products A High Diversity of Chemicals Is Present in the Aquatic Environment Sources and Fate Examples of Individual Groups Aryl Sulfonates Flame Retardants Organobromine compounds Organophosphorus compounds Pesticides Endocrine Disrupting Chemicals Anticorrosive Additives – BT and TT Gasoline Additives – Methyl tert-Butyl Ether Perfluorinated Surfactants – PFOS and PFOA Personal-Care Products Fragrances and Odorants Disinfectants UV Filters Pharmaceuticals Active Pharmaceutical Ingredients Illicit Drugs Metabolites Engineered Nanoparticles Artificial Sweeteners Cyanotoxins

3.04.1 Introduction The history of chemistry and the pharmaceutical sciences is an impressive success story. The products of the chemical and pharmaceutical industries are ubiquitous in everyday life. They help us to define the modern way of living. They contribute to our health and high living standards. The production of chemicals and pharmaceuticals, their usage, and application was associated over a long period with heavy pollution of the environment and serious health effects. During the second half of the last century, tremendous progress was made to prevent the pollution of environment and to reduce the impact of such pollution on health. Nowadays, proper and effective treatment and the prevention of emissions into air, water, and soil is in place in developed countries and will spread woldwide. However, it has also been learned since the end of the last century that products of the chemical and pharmaceutical industries themselves such as medicines, disinfectants, contrast media, personal care products, laundry detergents, surfactants, pesticides, dyes, paints, preservatives food additives, and personal care products, to name a few, also constitute a new type of environmental pollution and a possible health risk for the consumer.

69 69 70 71 72 73 73 74 74 74 75 75 75 76 77 77 77 78 78 78 78 80 81 81 82 82 83

Population growth and climate change will place great pressure on water resources in the future. Even now, several regions of this planet suffer surface water shortage as an everyday challenge, because water is necessary for agricultural, industrial, recreational, laundry, personal care, and drinking purposes. Therefore, the quality and quantity of water have to be carefully surveyed and managed. Artificial recharge can contribute to groundwater resources. However, quality control of the waters to be used must ensure that the groundwater is not contaminated by the water used for charging, so the recharge water must itself meet high standards. The presence of organic pollutants such as medicines, disinfectants, contrast media, personal care products, laundry detergents, surfactants, pesticides, dyes, paints, preservatives and food additives, and their metabolites and transformation products is of growing interest in this context. Chemicals, like many of the xenobiotic organic compounds, are of increasing concern in urban water management, because water supply, urban drainage, and wastewater-treatment systems were expressly designed originally to solve other problems (supply of potable water, flooding prevention, and sanitation). Thus, there is a need to understand, in an integrated manner, the sources, flow paths, fates, and effects of hazardous chemicals on both humans and

69

70

Emerging Contaminants

ecosystems. In this chapter, a very brief overview is given of our present knowledge about the presence of emerging pollutants in the aquatic environment. With respect to the vast amount of literature available, only a rough overview for some groups of emerging contaminants will be given here for the sake of demonstration rather than claiming an exhaustive elaboration and full overview of the topic. Instead, some typical examples and results are presented to demonstrate general principles and important issues and the diversity of chemicals involved. Therefore, after some general considerations referring to the term ‘emerging contaminants’, some selected examples will be presented in the following. The purpose of this chapter is not, and cannot be, to review all the available results as the literature is vast. Instead, typical and illustrative examples and facts will be presented to demonstrate the underlying issues. The reader interested in more detailed data and findings will find help in the numerous books and reviews that have already been published. As for analysis, for example, only recently two extensive reviews were published (Richardson, 2009; Giger, 2009). Therefore, analytical issues are not addressed here. Instead, the interested reader is advised to seek help in these articles and the ones cited in this chapter. Phthalates, polychlorinated dioxins, polychlorinated biphenyls (PCBs), polycyclic aromatic hydrocarbons, the older pesticides such as the chlordienes, dichlorodiphenyltrichloroethane, and others, surfactants such as linear alkylsulfonates and nonylphenol and nonylphenol ethoxylates, their transformation products, and other compounds have long been known to be present in the aquatic environment. Hence, they will not be included here.

3.04.2 General Aspects: What Are the Emerging Contaminants and Micro-Pollutants? The presence of certain chemicals at the lower mg l1 level in the aquatic environment has become evident with the improvement of analytical techniques (Reemtsma and Jekel, 2006; Barcelo´ and Petrovic, 2008; Ku¨mmerer, 2008a). The broader availability of liquid chromatography–mass spectrometry (LC–MS) and LC–MS/MS, in particular, permits the detection of polar compounds such as most pharmaceuticals, metabolites, and transformation products that have not previously been amenable to analysis. Some will never be detected. This is one reason why these chemicals are often called emerging contaminants. However, there is neither a general definition nor a complete list of compounds available that are in general included by the term ‘emerging contaminant’. The environment is contaminated by myriads of ‘merging contaminants’, that is, chemicals present in the aquatic environment in the mg l1 range and below. Because of this concentration range, they are often called organic micropollutants too. They are released from urban, industrial, agricultural, and other anthropogenic activities. Many of these have been and are currently undetected. As most of the chemicals applied by consumers will end up in sewage, ‘emerging contaminants’ is a term that describes nowadays often the pollution of the aquatic environment. Within the groups of emerging contaminants and/or

micro-pollutants, chemicals with similar structures can be found. However, often groups of chemicals with very different structures and properties belong to the same category in terms of application and usage. Pharmaceuticals are often classified according to their purpose and biological activity. The same hold for other groups such as insecticides, biocides, dyers, plasticizers, antibiotics, analgesics, and anti-neoplastics. Classification according to chemical structure is often used within subgroups of chemicals, for example, within the group of antibiotics or the subgroups within the antibiotics such as b-lactams, cephalosporins, penicillins, or quinolones. Other classifications refer to the mode of action (MOA), for example, anti-metabolites or alkylating agents within the group of cytotoxics/anti-neoplastics. In the case of classification according to MOA, the structures of molecules within the same group can be very different and, hence, so can their environmental fates. Sometimes authors summarize a certain group of chemicals according to their chemical structure (e.g., brominated diphenyl ethers) and sometimes according to their use (e.g., pharmaceuticals and flame retardants) or both at the same time. As some chemicals with a certain chemical structure such as the nyphtylsulfonates are applied for different purposes, such chemicals are often referred to in terms of their chemical structure only. In general, there is no clear differentiation, and overlaps of different classification schemes are common. Depending on the class of compounds, the emerging contaminant is the chemical itself or it may only be the transformation products or metabolites (e.g., pesticides or some surfactants). It is the transformation products, for others such as pharmaceuticals it is the parent compounds and, since only recently, the transformation products too. Often authors used the term ‘emerging contaminant’ to emphasize the novelty of the detection of a certain chemical in the (aquatic) environment. However, how ‘first’ is first enough is judged differently by different authors and readers. Therefore, the expression ‘emerging contaminant’ is not only loosely defined and used. The chemicals included may change from one author to another and within time. Furthermore, generally, it does not mean that these compounds have only recently appeared in the environment; sometimes, it does not even mean that there have been no earlier reports on their presence, because in some cases older literature has not been searched for or cited. In some other cases, these substances may have been introduced into the environment long ago but had not been detected because they had not been searched for or their concentration is so low (mg l1 and below) that until recently they were undetectable. In cases where findings had been reported earlier but not noticed by the researches and the compounds are of interst now (again), the term ‘emerging contaminant’ is applied by some authors despite this earlier findings. Emerging contaminant is a more or less loosely defined subgroup of the micro-pollutants. The list given here reveals that the emerging contaminants are not a homogeneous group of chemicals. On the contrary, they are very different and they can be grouped according to chemical structure, properties, purpose of application, or effects, respectively. The composition of the types of compounds constituting the whole bunch of emerging contaminants undergoes subtle change with time, as compounds that have been named emerging contaminant years ago may not be in

Emerging Contaminants

the focus of the researches anymore after some time for several reasons. Instead, other groups of chemicals may be included. The presence of such pollutants in the aquatic environment, however, is one of the big challenges for a sustainable water future in any case (DFG, 2003; Schwarzenbach et al., 2006; Ledin and Patureau, 2008; Fatta-Kassinos et al., 2010). Summarizing, there is no clear definition of emerging pollutants. In contrast, the term ‘micro-pollutants’ refers clearly to a nonambiguous criterion, that is, to such compounds that are detected in the environment in the mg l1 range and below, independently of their chemical structure, usage, or MOA, whereas the term emerging contaminant seems to be most helpful in the press room, not in science. Therefore, the author recommends to abandon the term ‘emerging contaminant’ and to stick with ‘micro-pollutant’.

3.04.3 Parent Compounds, Metabolites, and Transformation Products Many pharmaceuticals undergo a structural change in the bodies of humans and animals. The results of such processes are metabolites. After their excretion and release into the environment, both parent compounds and metabolites can undergo structural changes by a variety of biotic and nonbiotic processes, including photolysis, hydrolysis, and biotransformation. Pharmaceuticals and other chemicals are often incompletely transformed, that is, they are not fully mineralized by organisms such as bacteria and fungi in the environment (Haiß and Ku¨mmerer, 2006; Gro¨ning et al., 2007; Trautwein et al., 2008) as well as by light and other nonbiotic chemical processes. Structural transformations of chemicals may also be a result of technical processes such as effluent treatment by oxidation and photolysis (Ravina et al.,

71

2002; Zu¨hlke et al., 2004; Li et al., 2008a; Me´ndez-Arriaga et al., 2008). The resulting substances are often referred to as metabolites by authors. However, this is confusing as metabolism is linked to living organisms. Chemicals resulting from nonbiotic transformation in the environment should therefore be referred to as transformation products. When considering pharmaceuticals and other chemicals in the environment that are transformed by living organisms, it is advisable to be even more restrictive. One should only refer to those substances as metabolites that have been altered in their chemical structure within a target organism (La¨ngin et al., 2008; Figure 1). All other compounds should be referred to as transformation products. Their origin can be indicated in designation by adding the process of formation, for example, phototransformation product. Generally, such a structural change results in new chemical entities with new properties. Normally, it is assumed that metabolism and transformation of pharmaceuticals and other chemicals lead to decreased toxicity. In some cases, however, metabolism (e.g., in the case of pro-drugs) and transformation lead to more active compounds. The same has been found for phototransformation and other oxidizing processes. Nowadays, several European regulations require the inclusion of transformation products in environmental risk assessment and monitoring (e.g., Drinking Water Directive, 1998; European Commission, 2003). The exposure to transformation products can be relevant as has been indicated, for example, for pesticides in groundwater (Boxall et al., 2004; Kolpin et al., 1997, 2004; Hanke et al., 2007). In these studies, several pesticide metabolites (e.g., well known from the extensively applied pesticide metolachlor) were found in higher concentrations in groundwater than the parent compounds (see also the case of tolylfluanid described in this chapter).

Active pharmaceutical (parent compound)

Metabolites

Nonbiological metabolism (hydrolysis in the stomach) Human metabolism Microbial metabolism (liver, mucosa, etc.) (skin and gut)

Transformation products Biological transformation (organisms)

Nonbiological transformation (light, oxidation, hydrolysis, etc.)

Technical transformation (ozonolysis, photolysis, chlorination, etc.)

Humans

Sewage Water Soil Manure (air)

(Water) treatment

Figure 1 Metabolites and transformation products of pharmaceuticals. For other compounds only transformation products are of interest. From Ku¨mmerer K (ed.) (2008a) Pharmaceuticals in the Environment: Sources, Fate, Effects and Risks, 3rd edn. Berlin: Springer.

72

Emerging Contaminants

3.04.4 A High Diversity of Chemicals Is Present in the Aquatic Environment In a study, more than 100 individual water samples from over 100 European rivers from 27 European Countries were analyzed for 35 selected compounds, comprising pharmaceuticals, pesticides, fluorinated surfactants, benzotriazoles (BTs), hormones, and endocrine disrupters. Around 40 laboratories participated in this sampling exercise. The compounds most frequently detected and at the highest concentration levels were BT, caffeine, carbamazepine, tolyltriazole (TT), and nonylphenoxyacetic acid (Loos et al., 2009a, 2009b). Only about 10% of the river water samples analyzed could be classified as very clean in terms of chemical pollution. Rodil et al. (2009), for example, described a method for the simultaneous determination of 53 multiclass compounds, which they call emerging organic pollutants (e.g., acidic herbicides, ultraviolet (UV) filters, insect repellents, organophosphorus flame retardants, bactericides, pharmaceuticals, and metabolites). The authors found 31 pollutants in wastewater with concentrations up to 10 mg l1 in the case of the active pharmaceutical ingredient (API) ibuprofen. In total, 13 compounds were detected in tap water with concentrations up to 0.13 mg l1 for tri(chloropropyl)phosphate. The five most frequently detected chemicals in surface water in a study performed in the USA were cholesterol (59%, natural sterol), metolachlor (53%, herbicide), cotinine (51%, nicotine metabolite), beta-sitosterol (37%, natural plant sterol), and 1,7-dimethylxanthine (27%, caffeine metabolite), whereas, in groundwater, the most frequently detected were tetrachloroethylene (24%, solvent), carbamazepine (20%, pharmaceutical), bisphenol A (BPA; 20%, plasticizer),

1,7-dimethylxanthine (16%, caffeine metabolite), and tri-(2chloroethyl)phosphate (12%, fire retardant). A median of four compounds was detected per site, indicating that chemicals generally occur in mixtures in the aquatic environment and are likely to originate from a variety of natural origin, animal and human uses, and waste sources. In another study, four wells down-gradient from a landfill were investigated for the presence of waste-indicator and pharmaceutical compounds in groundwater affected by landfill leachate. The compounds identified included detergent degradation products, plasticizers (ethanol-2-butoxy-phosphate and diethyl phthalate), BPA, triclosan, an antioxidant (5-methyl-1H-benzotriazole), fire-retardant compounds, and several pharmaceuticals and their metabolites (Buszka et al., 2009).

3.04.5 Sources and Fate Numerous studies have shown that a variety of organic compounds such as pharmaceuticals, steroids, surfactants, flame retardants, fragrances, plasticizers, and other chemicals often associated with wastewaters have been detected in the vicinity of municipal wastewater discharges and agricultural livestock facilities, as these are an important source for the introduction of chemicals into the aquatic environment. There they may undergo different distribution and transformation processes (Figure 2). More detailed information on the (incomplete) removal of some groups of emerging contaminants by different effluent treatment technologies was addressed by Barcelo´ and Petrovic (2008). The presence of the micro-pollutants in the aquatic environment demonstrates that technical emission treatment is

Municipal or industrial sewage discharge Volatilization to atmosphere

Photolysis Hydrolysis

Dilution and diffusion

Deposit io

n

Pa r ti cl

e

p ns tra

por t rans ed t v l o s Biodegradation Dis and Sorption transformation onto sediments or t

Biocencentration Deposition and resuspension Deposition and accumulation

Figure 2 The fate of pollutants in the aquatic environment. From US Geological Survey, http://toxics.usgs.gov/regional/emc/transport_fate.html.

Emerging Contaminants

not sufficiently effective. The presence of micro-pollutants in water is currently leading to much research and development effort being directed toward advances in municipal wastewater treatment. The advanced treatment of effluents has been investigated using (photochemical) oxidation processes (e.g., Qiting and Xiheng, 1988; Zwiener and Frimmel, 2004; Ravina et al., 2002; Kiffmeyer, 2003; Ternes and Joss, 2006; Watkinson et al., 2007; Isidori et al., 2007; Putschew et al., 2007; Lee et al., 2007), filtration (Schro¨der, 2002; Drewes et al., 2002; Heberer and Feldmann, 2008), application of powdered activated charcoal (Metzger et al., 2005; Nowotny et al., 2007), and constructed wetlands (Matamoros and Bayona, 2006). Reviews are available describing the advantages and disadvantages of the different technologies (Schulte-Oehlmann et al., 2007; Jones et al., 2007; Wenzel et al., 2008; Ternes and Joss, 2006). However, the approach of effluent treatment has some limitations in principle and may not in the end be a sustainable solution. Some of these limitations may also apply for the treatment of solid waste and the chemical treatment of exhaust air, some not in the case of the application of biofilters for the treatment of exhausts:

• • • •

•

• • • • • •

Efficiency may depend strongly on the type of compound to be removed. None of the technologies can remove all of the compounds (Ravina et al., 2002; Schro¨der, 2002; Wenzel et al., 2008). Will the now so-called advanced treatment technology work for new compounds in the future? Reaction products of (photo)oxidation processes have been found themselves to possess mutagenic and toxic properties (Isidori et al., 2005, 2007; Lee et al., 2007; Wei-Hsiang and Young, 2008). Prolongation of the hydraulic retention time in sewage treatment plants (STPs) results in only a little improvement of the elimination rates. It may, however, cause high costs because of the necessity to enlarge the STPs. Resistance in bio-membrane reactors: Is the enrichment of antibiotics and resistant bacteria causing increasing resistance? (No information is available on this topic.) Resistant material will not fully be retained by membranes. Combined sewer overflow: no treatment of storm water. Sewage from leaking from drains is not treated since it soaks into the ground before it reaches the STP. The advanced treatment processes depend on a high energy input and a minimum water flow. Therefore, they are often not possible/affordable in less developed countries. Costs are not clear and whether they are affordable is not known. As for the costs, different authors present different data depending on the assumptions made. It is questionable whether the additional costs are acceptable (Jones et al., 2007).

In principle, the advanced treatment processes are not compatible with a sustainable development as they are end-of-thepipe technologies, not affordable and/or applicable in all countries, and costly for manufacturers and the public. Energy demand causes high emissions of CO2 (Jones et al., 2007) and other green house gases cause additional costs for consumers and companies, as well as for the general public in the future. (See, for example, the Kyoto-Protocol and the European

73

Union Emission Trading System (EU ETS; EU, 2003). In January 2008, the European Commission proposed a number of changes to the scheme, including the inclusion of other greenhouse gases, such as nitrous oxide and perfluorocarbons. It is also under consideration whether to extend the EU ETS to other industries. Clark (2006) has described other drivers for green chemistry.) Ozonation, for example, can consume an equivalent amount of energy to that needed to run a municipal STP. Wenzel et al. (2008) investigated the advantages and disadvantages of advanced wastewater treatment for micropollutants using environmental life-cycle assessment (LCA) and a literature review of advanced treatment performance. The LCA evaluation involved sand filtration, ozonation, and membrane bioreactors, and assessed the effect of extending existing tertiary treatment with these technologies on a variety of micro-pollutants (heavy metals, endocrine disruptors, polycyclic aromatic hydrocarbons (PAH), phthalates, and detergents, flame retardants, and others). The authors assessed the environmental break-even point where the removal of micro-pollutants and reduction in (eco-)toxicity will outweigh the increased resource and energy consumption. It was found, in some of the scenarios considered, that more environmental impact may be induced than removed by the advanced treatment. Furthermore, advanced treatment of effluent and exhausts is not available and affordable worldwide. Therefore, other approaches are necessary. (Green and sustainable are sometimes used synonymously by authors; however, they are not synonymous. Green chemistry is focused on the product/ chemical/pharmaceutical itself, whereas sustainable includes all aspects of a product related to sustainability, e.g., the shareholders, the stakeholders, and the people applying and using the compounds when seeking for solutions that will work. (First International Conference on Sustainable Pharmacy, Osnabru¨ck, 24/25 April 2008).) It has been learned that the source of micro-pollutants is often not a focus source. These molecules end up in the environment not because of their improper use, but rather as a result of their proper use. Furthermore, one has to be aware that especially in the case of consumer products it is not only a single chemical compound that is involved. Most often, it is a complex mixture of compounds that constitute a product such as a shampoo, a medicine, a disinfectant, a pesticide, a cleaning agent, or facade paint. Often, it is not feasible to assess the substance flow of compounds that are ingredients of such products as not all of the ingredients have to be declared in terms of quantity. Some are not even declared at all. If the chemicals and pharmaceuticals, their metabolites, and transformation products are not eliminated during sewage treatment, they enter the aquatic environment and eventually reach drinking water. Going back to the beginning of the pipe, the products of chemical industries – the molecules themselves – come into focus. Therefore, it is reasonable to think about emission management directly related to the properties of the molecules themselves and to focus on the whole life cycle of products and substances. This is addressed by the concepts of green and sustainable chemistry (Anastas and Warner, 1998; Clark and Smith, 2005; Clark, 2006) and, more recently, by the concept of green and sustainable pharmacy (Ku¨mmerer, 2009a). Green and sustainable are sometimes used synonymously by

74

Emerging Contaminants

authors. However, they are not synonymous. Green chemistry is focused on the product/chemical/pharmaceutical itself, whereas sustainable includes all aspects of a product related to sustainability, e.g., the shareholders, the stakeholders, and the people applying and using the compounds when looking for solutions that will work (Ku¨mmerer and Hempel, 2010). A key point is the benign-by-design concept (Daughton, 2003; Boethling et al., 2007; Ku¨mmerer, 2007, 2008b).

flame retardants, for example, for textiles. Synthetic materials, usually halocarbons, include organochlorine compounds such as PCBs, chlorendic acid derivates (most often dibutyl chlorendate and dimethyl chlorendate), and chlorinated paraffins. Nowadays, compounds such as organobromine compounds and organophosphates are also of interest.

3.04.6.2.1 Organobromine compounds

3.04.6 Examples of Individual Groups 3.04.6.1 Aryl Sulfonates Aryl sulfonates are used as precursors for sulfonated azo dyes, wetting agents, leveling agents, dispersants, optical brighteners, pesticides, ion-exchange resins, pharmaceuticals, and concrete plasticizers (Knepper et al., 1999; Reemtsma, 1999; Arslan-Alaton et al., 2010). Sulfonated azo dye formulations used in the tannery and textile industry are one major source of sulfonated aromatic amines in water. Dye production today takes place mainly in Asian countries such as China and India. It is estimated that during the dyeing process, up to 10–30% of sulfonated dyes end up in the exhausted dye bath (ArslanAlaton et al., 2010). Hence, aromatic sulfonates can be present at high concentrations in industrial wastewater in these countries. Recently, the discharge into receiving water bodies has led to serious environmental pollution problems and the production of aryl sulfonates, such as H-acid, has been restricted in some countries. Sulfonates are strong organic acids; hence, they are anions over a wide range of pH and cannot be effectively trapped by conventional adsorbents such as RP 18 materials. In the aquatic environment, they do not significantly sorb on biosludge or sediments (Zerbinati et al., 1997). Sulfonated aromatic amines can be formed during the reduction of sulfonated azo dyes under anaerobic/anoxic conditions (Knepper et al., 1999). These are potentially toxic and/ or carcinogenic (Oh et al., 1997). The fate of aryl sulfonates and their degradation products in the aquatic ecosystem and in biological treatment facilities is still not very clear, since until now only limited attention has been paid to their occurrence and degradability in the natural environment and in engineered systems (Jandera et al., 2001). Laboratory testing of organic sulfonates has revealed them to be nonbiodegradable (Su¨tterlin et al., 2008). Depending on their molecular structure or other physical–chemical properties, aryl sulfonates may be biodegradable in engineered biological treatment systems at very slow rates and after acclimation (O’Neill et al., 1999; Rieger et al., 2002). Tan et al. (2005) reported concentrations in the ng l1–mg l1 levels in European rivers and they are detectable at ng l1–mg l1 levels in surface waters. However, their concentrations in industrial wastewater-treatment plants (WWTPs) can be in the mg l1–mg l1 range (Tan et al., 2005).

3.04.6.2 Flame Retardants Minerals such as asbestos, compounds such as aluminum hydroxide, magnesium hydroxide, antimony trioxide, various hydrates, red phosphorus, and boron compounds, mostly borates are applied as flame retardants. In addition to these inorganic compounds, organic compounds are also used as

Organobromine compounds such as polybrominated diphenyl ethers (PBDEs) are one of the important groups of organic chemicals used as flame retardants. PBDEs are used as flame retardants in polymeric materials such as furnishing foam, rigid plastics, and textiles. PBDEs are technical, that is, loosely defined mixtures. Many of these chemicals are considered harmful, having been linked to liver, thyroid, reproductive/ developmental, and neurological effects. The EU has included the PBDEs on a list of chemicals to be phased out of use in electrical and electronic equipment (e.g., personal computers and mobile phones). The main source of the organobromine compounds in the aquatic environment is textiles. The compounds are washed out during the washing cycles. The manufacture of textiles may be another important source for their introduction into the environment. PBDEs are included in Annex X of the Water Framework Directive (EU, 2000) and pentabromodiphenyl ether (pentaBDE) is a priority hazardous substance. The EU risk assessment of tetraBDE, sometimes also called tetrabromobisphenol A (TBBPA), suggests the classification ‘‘Very toxic to aquatic organisms, may cause long-term adverse effects in the aquatic environment.’’ In the USA, decaBDE manufacturers have successfully petitioned for an exemption for decaBDE from this ban in the USA (Illinois EPA, 2007). A different chemical structural class of flame retradand is hexabromocyclododecane (HBCD). The EU risk assessment of HBCD (ECB, 2007) recommends that HBCD be considered a persistent, bio-accumulative, and toxic (PBT) substance, although there is no official classification yet. Amounts used may be different in different countries. Important subgroups of PBDEs are pentaBDE, octaBDE, and decaBDE. tetraBDE, sometimes also called TBBPA, has the biggest share (Thuresson, 2006) in most countries. Over 40% of the use of decaBDE occurs in North America. The US has historically led the world production of these chemicals (50% of the total global demand in 2001). High pentaBDE, also called tetrabromobisphenol (A), PBDE (A), levels in the US marine environment reflect that over 90% of the pentaBDE global production has been utilized in the USA (Yogui and Sericano, 2009). PBDEs are ubiquitous in all compartments, including water, sediment, and biota. Contamination is higher in urbanized regions. The organobromine compounds are lipophilic, bioaccumulative (log Kow often six and above), and adsorptive. Therefore, only a minor proportion will pass STPs and reach the surface water. It was found that the mean concentrations of decaBDE, pentaBDE, HBCD, and TBBPA in STP sludge were 0.12, 0.11, 0.045, and 0.040 mg kg1 dry weight (d.w.), respectively (Nylund et al., 2002). This confirms the significance of elimination by sorption. Another emerging brominated flame retardant, marketed as a replacement for decaBDE, is decabromo-diphenyl ethane

Emerging Contaminants

(deBDethane). DeBDethane has a chemical structure similar to decaBDE and would accordingly have similar properties concerning bioaccumulation and persistence. Recently, a survey was conducted on decaBDE and deBDethane in sludge from 42 WWTPs in 12 different countries around the world (Ricklund et al., 2008a, 2008b). The authors found decaBDE in concentrations from 0.003 mg kg1 dry matter (d.m.) to a maximum of 19 mg kg1d.m. in all samples. DeBDethane was present in all samples but two, in levels from 0.001 to 0.22 mg kg1 d.m. The highest deBDethane/decaBDE ratios were found in Germany and neighboring countries, whereas the lowest ratios were found in the USA and the UK. The data reflect the use patterns of the substances, that is, the known high imports of deBDethane into Germany and the largest market demands for decaBDE in the USA and the UK. Chemically applied nonhalogen decaBDE substitutes are available for natural cellulose fibers such as cotton, wool, rayon, and linen. They include:

• • •

dimethylphosphono (N-methylol)propionamide; phosphonic acids such as (3-{[hydroxymethyl]amino}3oxopropyl)-dimethyl ester; and tetrakis(hydroxymethyl)phosphonium urea ammonium salt.

The future will show whether these will be part of the nextgeneration emerging contaminants.

3.04.6.2.2 Organophosphorus compounds Another important group of flame retardants already heavily applied is organophosphates in the form of halogenated phosphorus compounds such as phosphates and phosphonium salts, for example, tri-o-cresyl phosphate, tris(2,3dibromopropyl)phosphate (TRIS), bis(2,3-dibromopropyl)phosphate, tris(1-aziridinyl)-phosphine oxide, tris(2-chloro-1methylethyl)phosphate (TCPP), tris(2-chloroethyl)phosphate (TCEP), and tris(2-chloro-1-chloromethyl-ethyl)phosphate (TDCP). The main source of these organophosphates is probably construction materials (Bester et al., 2009). The phosphates are slightly less polar (log Kow 4 and below) than the organobromine compounds. Organophosphates have been found in the effluents of WWTPs (Marklund et al., 2005). Potential risks to groundwater have been reported for tetrakis(hydroxymethyl) phosphonium chloride (Illinois EPA, 2007). The concentrations of TCPP and related compounds in five WWTPs in the Rhine/Ruhr region ranged from a few hundred to 410 000 ng l1, depending on the respective activities within the catchment area (Bester et al., 2009). Because these compounds are not usually eliminated in wastewater treatment, the effluent concentrations are identical to the inflow concentrations (Marklund et al., 2005; Bester, 2007). Several wastewater-related organic micro-pollutants such as chlorinated and nonchlorinated organophosphates have been detected in a spring. TCPP was found in concentrations up to 0.13 mg l1 in tap water in the United States (Rodil et al., 2009).

3.04.6.3 Pesticides Some pesticides have been detectable in the aquatic environment for decades. Hence, they cannot be called emerging

75

pollutants. However, transformation products formed from the applied parent compounds in the environment (e.g., soil) came into focus in a broader range only recently. Nowadays, data on the fate of the pesticide itself and the possible products of biotransformation (often called ‘metabolites’) in soil and sediments have to be presented and assessed by the applicant before pesticides are authorized. Hence, these compounds are the subject of a great deal of study. The following is a typical example. The transformation product chloridazon-methyldesphenyl is formed from chloridazon (5-amino-4-chloro-2phenylpyridazin-3(2H)-one) in soil by biotransformation (Roberts and Hutson, 2002), which is further transformed to chloridazon-methyl-desphenyl. The transformation product chloridazon-methyl-desphenyl was detected at concentrations up to several hundreds of nanograms per liter in many of the groundwater samples investigated. Weber et al. (2007) detected this compound in surface-ground-, and drinking water in Germany. The transformation product was often found in higher concentrations than the parent compound. Knowledge on the fate of the transformations product, for example, in the aquifer or drinking-water treatment is little. Transformation products may give rise to further concerns. This can be illustrated by the case of the pesticide tolylfluanid. Application and microbial degradation of the fungicide tolylfluanid in soil result in a transformation product, N,N-dimethylsulfamide (DMS). DMS was found in groundwater and surface water with typical concentrations in the range of 100–1000 and 50–90 ng l1, respectively. DMS itself is not of toxicological concern. However, it exhibits high mobility in soils and water. Hence, it can enter the drinking-water-treatment process. Laboratory-scale and field investigations concerning its fate during drinking-water treatment revealed that DMS cannot be removed via riverbank filtration, activated carbon filtration, flocculation, and oxidation or disinfection procedures based on hydrogen peroxide, potassium permanganate, chlorine dioxide, or UV irradiation (Schmidt and Brauch, 2008). Even nanofiltration does not provide adequate removal efficiency. Disinfection with hypochlorous acid converts DMS to so far unknown degradation products but not to N-nitrosodimethylamine (NDMA) or 1,1-dimethylhydrazine. However, most important, during ozonation about 30–50% of DMS is converted into the carcinogenic NDMA (Figure 3). Wei-Hsiang and Young (2008) have described the NDMA formation during chlorination and chloramination of aqueous diuron solutions. NDMA is biodegradable and can be at least partially removed by subsequent biologically active drinking-water-treatment steps, including sand or activated carbon filtration.

3.04.7 Endocrine Disrupting Chemicals Endocrine-disrupting chemicals (EDCs) are chemicals of natural or synthetic origin that may interfere with the endocrine system. They may have estrogenic, anti-estrogenic, androgenic, anti-androgenic, and other hormone-like effects. For example, the egg yolk protein vitellogenin, only produced under normal circumstances in mature female fish (Sumpter and Jobling, 1995) was found in male fish that have been exposed to such substances. The same phenomenon has been observed downstream in rivers receiving STP effluent (Purdom et al., 1994;

76

Emerging Contaminants

O

O S

N

N

S

Cl F

O

O O

S H2N

N

N

N

Cl

Tolylfluanid

N,N-Dimethylsuklfamid (DMS)

NDMA

Figure 3 Formation of NDMA by ozonation of the transformation product (DMS) of the pesticide tolyfluanid by ozonation in drinking-water treatment.

Vethaak et al., 2006; Sumpter et al., 2006). Recently, an aquatic predicted no-effect concentration (PNEC) for ethinylestradiol (EE2) has been derived based on a large number of studies on the effects of EE2 on aquatic organisms. Caldwell et al. (2008) recommended the use of a PNEC of 0.35 ng EE2 l1 in order to adequately protect organisms in surface waters. Evidence that environmentally relevant concentrations of estrogens in surface waters can impact the sustainability of wild fish populations has been presented in the study by Kidd et al. (2007). In the last 15 years, the discovery of the estrogenic nature of STP effluents (Purdom et al., 1994) has resulted in a vast public concern and a huge amount of scientific literature in order to better understand and identify EDCs, their occurrence, fate, and their biological effects on wildlife. The naturally occurring estrogenic hormones (i.e., 17b-estradiol (E2) and estrone (E1)) have been studied extensively (Desbrow et al., 1998; Routledge et al., 1998) in the 1990s. Together with EE2, the synthetic active ingredient in many contraceptive pills, they were identified as the main contributors to estrogenicity of effluents of STPs. As fish feminization was one of the most striking of their effects, they were also referred to as gender benders. In the years that followed, an increasing number of chemicals were classified as EDCs. Compounds or compound classes that had been detected in the environment in the last two decades such as BPA, phthalates, alkylphenols and their ethoxylates, some pesticides, dioxins and PCBs, tributyltin compounds, and others were shown to be endocrine active (e.g., Jobling et al., 2006; Blair et al., 2000; Silva et al., 2002). Brominated organic flame retardants are also on the list now. In the last decade, a great deal of research has been dedicated to the technological improvement of the various stages of sewage treatment in terms of removal efficiency of micropollutants, and especially estrogenic chemicals, in municipal STPs. Removal of EE2 in particular is often nowhere near complete, as demonstrated by the findings of Kanda and Churchley (2008), Esperanza et al. (2007), Vethaak et al. (2006), and Lamoree et al. (2009).

TT is a mixture of 4- and 5-methyl isomers. The BT and the TT possess good water solubility, low vapor pressure, and low octanol water distribution coefficients (log Kow 1.23 and 1.89, respectively). BT is toxic and not biodegradable (Hem et al., 2003), and it can be degraded by UV irradiation at pH values below 7 in laboratory testing. Approximately 65% reduction in the BT concentration was achieved at a dose of 320 mW s cm2, and almost 90% reduction was achieved at 1070 mW s cm2 (Hem et al., 2003). BT is transformed into several compounds instead of full mineralization. Aniline and phenazine were identified as main compounds. The authors report that the transformation products show toxic effects, but that they are not as toxic as BT itself; hence, UV irradiation brings about a general decrease in toxicity (Hem et al., 2003). However, aniline itself is a confirmed animal carcinogen with unknown relevance to humans (IARC, 1987). The overall summary evaluation of carcinogenic risk to humans refers to group 3: the agent is not classifiable as to its carcinogenicity to humans (HSDB, 2009). Weiss et al. (2006) noted 2.1 mg l1 of BT and 13 mg l1 of TTs in untreated municipal wastewater. BT and TT were found in the lower mg l1 range in the samples of primary and secondary effluents STPs (Voutsa et al., 2006; Weiss et al., 2006). The elimination rate of BT in WWTP is only 30–40% (Weiss et al., 2006) and lower for TTs. Accordingly, these compounds are found in receiving waters. Concentrations of TT in receiving rivers were lower than 80 mg l1. In an EU-wide reconnaissance of the occurrence of polar organic persistent pollutants in European river waters, more than 100 individual water samples from over 100 European rivers from 27 European countries were analyzed for 35 selected compounds, including BTs among other substances. BT and TT were among the most frequently detected compounds and also had the highest concentration levels (Loos et al., 2009b). Because of extensive de-icing activities, BT was found in the groundwater below de-icing platforms at airports (Cancilla et al., 2003a, 2003b) and in the subsurface waters at airports at concentrations up to 126 mg l1 for BT and 198 mg l1 for total TT.

3.04.8 Anticorrosive Additives – BT and TT 3.04.9 Gasoline Additives – Methyl tert-Butyl Ether BT and TT are widely used as anticorrosive additives. They are also components of cooling and hydraulic fluids, antifreezing products, aircraft de-icer, and anti-icing fluid. The main input of BT into the aquatic environment stems from their use as dishwasher detergent additives where they are used for silver protection. From there, they are discharged in municipal wastewaters (Ort et al., 2005).

Methyl tert-butyl ether (MTBE) is a common gasoline additive often found in drinking water. On account of its physicochemical properties, it can contaminate large water volumes. It is to be phased out. The MTBE situation in the USA differs significantly from the one in Europe where the concentrations measured in river water and drinking water are approximately

Emerging Contaminants

two to three orders of magnitude lower than the USA, where MTBE was detected first. This example demonstrates the impact of use patterns on the presence of contaminants. The detection limits of analytical instruments and the volumes of samples used and the location of sampling determine when and where a specific compound or a group of compounds ‘emerges’ as a contaminant. MTBE concentrations in German river water, for example, show a tendency toward increasing concentrations since 1999 (Achten et al., 2002). It is not clear when the phasing out of these compounds will result in decreasing concentrations in drinking water because the residence time of contaminants in groundwater is often not known. Kolb and Pu¨ttmann analyzed 83 finished drinkingwater samples from 50 cities in Germany for MTBE. The detection frequency was 46% at a detection limit of 10 ng l1. The concentrations ranged from 17 to 712 ng l1 (100– 200 ng l1 range in rivers for example). MTBE was detected at concentrations lower than 100 ng l1 in bank-filtered drinking water from the rivers Rhine and Main and at 43–110 ng l1 in drinking water (Achten et al., 2002). All the MTBE concentrations measured were below the proposed limit values for drinking water (Achten et al., 2002). The occurrence of MTBE and gasoline hydrocarbons was examined in three types of studies of groundwater in the USA conducted by Moran et al. (2005). The detection frequency of MTBE was highest on monitoring wells located in urban areas and in public supply wells. Only 13 groundwater samples from all study types, or 0.3%, had concentrations of MTBE that exceeded the lower limit of the US EPA’s Drinking-Water Advisory Board. The probability of detecting MTBE in groundwater was strongly associated with population density, use of MTBE in gasoline, and recharge. Groundwater in areas with high population density, in areas where MTBE is used as a gasoline oxygenate, and in areas with high recharge rates had a greater probability of the presence of MTBE. In addition, groundwater from public supply wells and shallow groundwater underlying urban land-use areas had a greater probability of MTBE occurrence than groundwater from domestic wells and groundwater underlying rural land-use areas. In a review on the MTBE production, use, properties, and its behavior in the environment and occurrence in groundwater, surface water, drinking water, and wastewater, it was concluded that the conventional methods to remove volatile organic compounds from drinking water – air stripping and adsorption on granular active carbon – are not effective in the case of MTBE with concentrations of 100 mg l1 or more. Peerreviewed laboratory data on the application of nine different advanced oxidation processes (AOPs) to the degradation and mineralization of MTBE in water revealed that the most promising seem to be UV/H2O2 and O3/H2O2 processes, from both the technical and economic point of view (Siminiceanu, 2007). However, data on possibly formed transformation products were not given.

3.04.10 Perfluorinated Surfactants – PFOS and PFOA Perfluorooctanesulfonate (PFOS) and perfluorooctanoic acid (PFOA) are perfluorinated surfactants used to produce polymers and telomers whose carbon chains can be of various

77

lengths. They are also applied in the electroplating industry to prevent volatilization of chemicals from the electroplating bath. PFOA and PFOS are discussed as end products of many fluorochemical compounds in the natural environment (Nakayama et al., 2005). In a study, more than 100 individual water samples from over 100 European rivers from 27 European countries were analyzed for 35 selected compounds, comprising pharmaceuticals, pesticides, PFOS, PFOA, BTs, hormones, and endocrine disrupters. The rivers responsible for the major aqueous emissions of PFOS and PFOA from the European continent could be identified (Loos et al., 2008). Nakayama et al. (2005) gave a review on the distribution of PFOA and PFOS in Japan and their toxicities. PFOA and PFOS had a low order of toxicity in an acute toxicity study in rodents; however, they exhibited versatile toxicities in primates. Both chemicals are carcinogenic in rodents, causing reproductive toxicity, neurotoxicity, and hepatotoxicity (Nakayama et al., 2005). Nakayama et al. (2005) reported an epidemiological study conducted by a manufacturer that revealed an increase in prostate cancer mortality among workers exposed to PFOA. Another study conducted by the same manufacturer showed an increase in bladder cancer mortality among workers exposed to PFOS. In addition, peroxisome proliferation and calcium channel modulation are demonstrated effects. There are large interspecies differences in toxicokinetics. Perfluorinated surfactants (e.g., PFOS and PFOA) have shown different potentials for reproduction toxicity and carcinogenity in animal experiments as well as partly long half-lives in humans (Guruge et al., 2006; Zhang et al., 2010). They possess compound-dependent persistence in the environment. As they are nonpolar compounds, they tend to bioaccumulate in animals and humans (Houde et al., 2006). Accordingly, Nakayama et al. reported that the concentrations of PFOA and PFOS in the sera of Japanese people may reach 57.7 mg l1. Many studies in recent years have reported the ubiquitous distribution of this group of perfluorinated substances, especially PFOS and PFOA in surface and drinking waters (Skutlarek et al., 2006; Hoelzer et al., 2008). When measurements were made on samples from the Rhine, Ruhr, Moehne, and some of their tributaries, the Rhine-Herne Canal and the Wesel-Datteln Canal, the concentrations (sum of seven components usually detected) measured in the river Rhine and its main tributaries (at the confluences) were below 100 ng l1. The highest concentration (94 ng l1) was detected in the river Ruhr (tributary of the Rhine). The pattern here was different, PFOA was the major component in contrast to the other tributaries and the river Rhine. This may indicate a specific source. Remarkably, high concentrations of PS were found in the upper catchment of the river Ruhr (up to 446 ng l1) and the river Moehne (tributary of the Ruhr, up to 4385 ng l1). Illegal waste dumping was discussed as a possible source. Dilution effects were held responsible for decreasing downstream concentrations of PS in surface waters of the rivers Moehne and Ruhr. C-7–C-11 perfluorinated carboxylates and PFOS have been analyzed in selected stretches of the river Po and its major tributaries. Concentrations of about 1.3 mg l1 of PFOA were detected in the Tanaro River close to the city Alessandria. Below this tributary, levels between 60 and 337 ng l1 were measured in the Po River on several occasions. A mass load of

78

Emerging Contaminants

about 2.6 tons per year is discharged into the Adriatic Sea according to these data. PFOS concentration levels in the Po River at Ferrara were around 10 ng l1 (Loos et al., 2008). Concentrations of 0.4–123 ng l1 for PFOS and 4.2– 2600 ng l1 for PFOA were measured in the Yodo River basin (Japan). The highest concentrations were found in effluents of some STPs and tributary streams. The PFOA concentration was often higher than that of PFOS, but their profiles in the basin were somewhat similar. Load estimates revealed that the main sources of the contaminants in Katsura River were three STP effluents, whereas the sources in Uji River could be from Lake Biwa. It is suggested that wastewaters other than domestic wastewater contributed significant loads of PFOS and PFOA in highly contaminated STP effluents (Lien et al., 2008). In another study from Japan, PFOA and PFOS concentrations in surface water were in the ranges 0.1–67 000 and 0.1– 526 ng l1, respectively. Although the origin of PFOS in surface water in Japan remains unknown, PFOA present in surface water is very likely to have been released from a few industries according to the authors. In the study performed for the river Ruhr area, the major component PFOA was determined in many drinking-water samples as well as in the surface water of the same area. There the water supplies are mainly based on bank filtration and artificial recharge. Maximum concentrations in drinking-water samples were 598 ng l1 (major component PFOA: 519 ng l1).

3.04.11 Personal-Care Products Personal care involves products as diverse as deodorant, eye liner, facial tissue, lipstick, lotion, makeup, mouthwash, pomade, perfumes, personal lubricant, shampoo, shaving cream, skin cream, and toothpaste, to give a few examples. Some personal-care products such as shampoos and washing lotions can include up to 10–20 different ingredients such as surfactants, preservatives, dyes, fragrances and odorants, and others. During or after their use, a more or less big share of it is washed off from the skin or hair into sewage.

being aware of it, humans are also affected by human infochemicals (Bhutta, 2007). Every substance which can be smelled by the human nose can be used as a scent. It is assumed that humans can distinguish up to 10 000 different scents. Of these 10 000 compounds, more than 2100 substances representing 22 chemical groups are listed as scents in the Research Institute for Fragrance Materials (RIFM) database (Salvito et al., 2002; Salvito et al., 2004). As different as they are with regard to their chemical structure, they also differ in their physical and chemical properties such as vapor pressure, water solubility, and partition coefficient between octanol and water (log Pow). Fragrances such as the musk fragrances HHCB (galaxolide), AHTN (tonalide), OTNE (tso-E-Super), and others are used in washing processes, especially in softeners, in cosmetics, and in perfumery (Reiner and Kannan, 2006). According to their use, most of these compounds end up in wastewater. The musk fragrances are often lipophilic, as they are tailored to sorb onto fabric. Hence, they can bioaccumulate (Schmid et al., 2007). Some of the polycyclic musks are under discussion, because they are assumed to act as endocrine disruptors (Seinen et al., 1999; Bitsch et al., 2001). Influent concentrations in STPs are in the lower mg l1 range (Simonich et al., 2000, 2002; Bester, 2005). Effluent concentrations are only little lower indicating little elimination on sewage treatment. The elimination of HHCB and AHTN is mainly due to sorption. Because of their persistence in the aquatic environment, HHCB and AHTN have been used as markers for wastewater discharge into surface waters. Other mechanisms might also be relevant in the case of OTNE (Ternes and Joss, 2006; Andresen et al., 2007). Other fragrances such as musk xylene and metabolites (amines) as well as the musk ketone are found in the range 10–100 ng l1 in the inflow of the WWTPs in European countries (Bester et al., 2009). Newer fragrances such as the macrocyclic musks (e.g., habanolide, cyclopentadecanolide, and ethylenebrassylate) have not been detected in wastewater yet – this may be because of low usage, low dosage, or ready degradability (Gautschi et al., 2001). Terpenoid chemicals such as linalool are also used as fragrances and scents. Knowledge on their significance is still scarce, especially as there are natural source of their origin and as they may act as infochemicals (Bolek and Ku¨mmerer, 2010).

3.04.12 Fragrances and Odorants Humans depend very much on visual and acoustic stimuli to get along in their everyday life, whereas the role of scents is considered to be of secondary importance. For fish, insects, and other organisms, however, they are much more important. As they transport information for living organisms, such chemicals are also called infochemicals (Klaschka, 2008). Today, we distinguish between several classes of different infochemicals. Pheromones transfer information in which the sender and the receiver are both from the same species, whereas allelochemicals contain interspecific information. Allelochemicals are further divided into allomones (only the sender takes advantage of the information), kairomones (only the receiver takes advantage of the information), and synomones (sender and receiver take advantage of the information) (Dicke and Sabelis, 1988). The relationship of the different infochemicals can be seen in Figure 3. Although not

3.04.13 Disinfectants Compounds that kill micro-organisms (bacteriocides) or at least prevent their growth (bactriostatics) are employed to meet hygienic standards in medicine and food processing as well as for the preservation of certain chemical products such as paints or glues (Russell et al., 1992). Some of these compounds such as alcohols (e.g., ethanol and isopropyl alcohol) are readily biodegradable preservatives. Others such as trichlosan form stable transformation products (methyltrichlosan). Others are less biodegradable and they are not fully removed in sewage treatment, including some quaternary ammonium compounds (QACs, e.g., benzalkonium chloride). QACs are cationic microbiocidal substances. They are used as surfactants, anti-electrostatics, and phase-transfer catalysts. They are also important ingredients in disinfectants. They are

Emerging Contaminants

surface active compounds consisting of a hydrophobic alkyl chain and a hydrophilic group carrying a positively charged quaternary nitrogen atom. QACs are emitted via effluents from hospitals, households, and industries and finally end up in municipal sewage where they can reach surface water (Ku¨mmerer et al., 1997; Martı´nez-Carballo et al., 2007a; Su¨tterlin et al., 2008). Because of their positive charge, QACs adsorb strongly to the negatively charged surfaces of sludge, soil, and sediments (Ferrer and Furlong, 2002; Sun et al., 2003). Their widespread use and sorption behavior imply that they are expected to be present in many environmental compartments (Martı´nez-Carballo et al., 2007b). Only recently, it has been found that QACs are weakly mutagenic (Ferk et al., 2007).

3.04.14 UV Filters Typical compounds used in sun screens are benzophenone (BP), benzhydrol, 4-hydroxybenzophenone, 2-hydroxy-4methoxybenzophenone (HMB), 2,4-dihydroxybenzophenone (DHB), 2,20 -dihydroxy-4-methoxybenzophenone, and 2,3,4trihydroxylbenzophenone. Some of these compounds have endocrine effects (Schlumpf et al., 2008). UV filters applied in sun screens are directly emitted into surface water after application when people go bathing and swimming. Accordingly, they have been detected in water samples (Giokas et al., 2004, 2005). Analysis of wastewater in Spain revealed the systematic presence of HMB (BP 3) and DHB (BP 1) in raw samples with maximum concentrations close to 500 and 250 ng l1 (Negreira et al., 2009). Analysis of seven BP-type UV filters in water in Korea (Jeon et al., 2006) revealed concentrations of 500–18 380 ng kg1 in soil samples of and of 27– 204 ng l1 in water samples.

3.04.15 Pharmaceuticals 3.04.15.1 Active Pharmaceutical Ingredients The presence of pharmaceuticals for human use in the environment has been a topic for several years now. Though the study of pharmaceuticals in the environment is still a fairly recent, a vast amount of literature has already been published, making it impossible to cover all topics and issues in this chapter.

79

Pharmaceuticals and disinfectants can be classified according to their purpose and biological activity (e.g., antibiotics, analgesics, anti-neoplastics, anti-inflammatory substances, antibiotics, antihistamines, X-ray contrast media, and surface disinfectants). The classification of small molecule APIs by their chemical structure is used mainly for the active substances within subgroups of medicines, for example, within the group of antibiotics, or subgroups within the antibiotics such as b-lactams, cephalosporins, penicillins, or quinolones. In this case, one might expect that the compounds could be treated as groups with respect to chemical behavior. However, even small changes in the chemical structure may have a significant impact on solubility and polarity as well as other properties that govern their environmental fate to some extent (Cunningham 2008, Figure 4). Other classifications refer to the MOA, for example, antimetabolites or alkylating agents within the group of cytotoxics/anti-neoplastics. In the case of classification according to MOA, chemical structures of molecules within the same group can be very different and, hence, their environmental fate can also differ. Therefore, these compounds cannot be treated as a group with respect to environmental issues. Compared with most bulk chemicals, pharmaceutically active compounds are often complex molecules with specific properties, for example, dependence of the octanol–water partition coefficient (Kow) on pH (Cunningham, 2008; Ku¨mmerer, 2009a). Some medicines contain molecules based on protein (biopharmaceuticals). Biopharmaceuticals may be defined as medical drugs produced using biotechnology by means other than direct extraction from a native (i.e., nonengineered) biological source. Examples include proteins (including antibodies) and nucleic acids. The first and best-known example was recombinant human insulin. Biopharmaceuticals are not typically regarded as biopharmaceuticals by the industry. Probably, not all naturally occurring compounds that are used as drugs are biopharmaceuticals. For example, estrogen is not regarded as a biopharmaceutical. The environmental relevance of biopharmaceuticals is not yet clear and they are not the focus of environmental research and risk management. One view is that they are not relevant because they are closely related to natural products and are therefore expected to be quickly biodegraded or are denatured, that is, inactivated in the environment. The other view is that naturally occurring

HN N

N OH

F O

O

Figure 4 Zwitterionic character of ciprofloxacin. Depending on pH additional different chemical species can be formed by internal protonation, i.e., the shift of protons between the basic amino functions and acidic carboxyl groups. Calculator Plugins were used for structure property prediction and calculation, Marvin 4.1.1, 2006, ChemAxon (http://www.chemaxon.com).

80

Emerging Contaminants

compounds are not always easily biodegraded, and modified natural compounds even less so. Structurally related compounds such as plasmids have been found in the environment (Schlu¨ter et al., 2007; Ku¨mmerer, 2009a). Furthermore, it is known that the protein structures known as prions are very stable. A prion (proteinaceous infectious particle, -on by analogy to virion) is an infectious agent composed only of protein.) They cause a number of diseases in a variety of animals and Creutzfeldt–Jakob disease in humans. Prions are believed to infect and propagate by refolding abnormally into a structure which is able to convert normal molecules of the protein into the abnormally structured form. This altered structure renders them quite resistant to denaturation by chemical treatments and physical agents (proteases, heat, radiation, and formalin) making disposal and containment of these particles difficult. Prions can be denatured by subjecting them to a temperature of 134 1C in a pressurized steam autoclave. Besides the active substances, formulations may also incorporate excipients and, in some instances, pigments and dyes. They are often of minor importance for the environment. Some medicines contain EDCs as excipients. For data and findings that are more detailed, the reader is advised to seek help from the numerous books and reviews that have already been published (e.g. Ku¨mmerer, 2001a, 2001b, 2004, 2008b, 2009a, 2009b, 2009c; Heberer, 2002; Williams, 2005; Ternes and Joss, 2006). Pharmaceutically active compounds (sometimes called active pharmaceutical ingredients or APIs) are complex molecules with various functionalities and physico-chemical and biological properties. They are developed and used because of their more or less specific biological activity on human and/or organisms that may cause illness such as bacteria and other microorganisms. Therefore, they are of special scientific and public interest. Most pharmaceutically active compounds are polar compounds in order to make them orally available. Their molecular weights range typically from 200 to 500/1000 Da. Such APIs are called small molecules. These are the ones currently being researched and detected in the environment. They are among the compounds called micro-pollutants, because they are often found in the mg l1 or ng l1 range in the aquatic environment. They are often complex molecules with different functionalities and physico-chemical and biological properties. There is no data available about the total worldwide use of pharmaceuticals. The consumption and application of pharmaceuticals may vary considerably from country to country (Verbrugh and de Neeling, 2003; Goossens et al., 2005, 2007; Schuster et al., 2008). If there are legislative changes imposed on the health system, it may happen that some compounds are not used any more or others gain more importance, for example, for economical reasons. According to United Nations’ figures, 2.3% of Japanese women of reproductive age take a contraceptive pill containing EE2 as the main active compound, compared to 16% in North America and up to 59% in Europe (United Nations, 2004). Some pharmaceuticals are sold over the counter without prescription in some countries, whereas in others they are only available by prescription. Some antibiotics such as streptomycins are used in the growing of fruit (pomology) whereas others are used in bee-keeping. Again, the situation may

vary from country to country. Antimicrobials are among the most widely used pharmaceutical compounds in animals (Boxall et al., 2003a, 2003b; Sarmah et al., 2006). These drugs are used in animal husbandry for veterinary purposes, or as growth promoters (particularly in large-scale animal farming and intensive livestock treatment). The active compounds as well as excipients may enter the environment by different routes via several different nonpoint sources such as effluents of STPs, waste, and landfill effluent or treatment of animals. Because of good manufacturing practice (GMP) regulations (required for the manufacturing of pharmaceuticals) and the frequent high economic value of the active substances, the amount of emissions occurring during manufacturing had been thought to be negligible. Indeed, such emissions are assumed to be low in Europe and the North Americas. However, manufacturers have not yet published data. It has only recently been found that in Asian countries, concentrations for single compounds up to several mg l1 can be found in effluents (Larsson et al., 2007; Li et al., 2008a, 2008b). However, even in Norway the input from a local manufacturer was high (Thomas, 2008). As expected, pharmaceuticals are present in hospital wastewater (Brown et al., 2006; Steger-Hartmann et al., 1996; Ku¨mmerer and Helmers, 1997; Hartmann et al., 1999; Ku¨mmerer, 2001a, 2001b; Ha¨drich, 2006; Go´mez et al., 2006; Seifrtova´ et al., 2008; Ku¨mmerer, 2008a). The concentrations of pharmaceuticals in hospital wastewater are higher (up to 100 mg l1 in some cases) than in municipal sewage (often in the range of a few up to 20 mg l1). However, the total substance flow is much lower because of the much lower share of effluent from hospitals in municipal effluent in developed countries. The dilution of hospital wastewater by municipal wastewater is by much more than a factor of 100 (Ku¨mmerer and Helmers, 1997, 2000). In terms of volume, the public (households) is the main source for pharmaceuticals in households (Ku¨mmerer, 2009a). Concentrations in municipal wastewater are typically in the lower mg l1 range. Another, but minor, source is effluents from landfills as date-expired medicaments are often disposed of with household waste (Eckel et al., 1993; Holm et al., 1995; Ahel and Jelicˇic, 2001; Metzger, 2004). Systematic studies of the occurrence of pharmaceuticals in the environment are now available for several countries. Meanwhile, there is evidence of the occurrence of some 180 different drugs in STP effluent, surface water, and groundwater. Probably, there are many more, however, validated methods not yet available for the analysis of most of the APIs in the environment. The concentrations of pharmaceuticals in surface waters have been shown to lie most often in the ng l1 range, in rare cases in the low mg l1 range. The findings of recent years have been confirmed for various countries and different environmental compartments (Ku¨mmerer, 2009a). Compared to the free water phase, the analysis of APIs is difficult in biosolids and sewage sludge, despite the fact that information about pharmaceuticals in sewage sludge and biosolids is necessary for a proper understanding of fate and for risk assessment (Jones-Lepp and Stevens, 2007). APIs have also been detected in the arctic environment (Kallenborn et al., 2008). Some pharmaceuticals are used by hydrologists as tracers for anthropogenic impact on waters (Mo¨ller et al.,

Emerging Contaminants

2000, 2002; Elbaz-Poulichet et al., 2002; Verplanck et al., 2005; Buerge et al., 2006). Some APIs have even been detected in drinking water (Heberer, 2002; Ku¨mmerer, 2008a).

3.04.15.2 Illicit Drugs Around 200 million people in the world are estimated to have used illicit drugs at least once during the last year (United Nations Office of Drug and Crime, 2010). They are of special concern because of their psychoactive properties. Cannabis is the one most consumed, involving far more than 4% of the global population aged between 15 and 64 years old. Next come opiates (especially heroin) and cocaine, which are the two second most consumed illicit drugs on the global level (United Nations Office of Drug and Crime, 2010). Recently, psycho-active and illicit drugs such as amphetamine, cocaine and its metabolite benzoylecgonine, morphine, 6-acetylmorphine, 11-nor-9-carboxy-delta-9-tetrahydrocannabinol, methadone and its main metabolite 2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine, and amphetamines have been detected in surface water and wastewater. The high consumption rates reported for these compounds explain their relatively high concentration levels in the aquifer. In some studies, the measured concentrations were employed to estimate the use of these drugs. Seasonal variations have been found in several studies (Kasprzyk-Hordern et al., 2008). Maximum levels of cocaine, ecstasy, and methamphetamine values were found in winter but high loads were also detected in summer. There was a significant increase in the concentrations of these compounds during the last days of December and the first days of January, corresponding to the Christmas and New Year holidays. Up to now, drugs of abuse have been detected in wastewaters and surface waters in the USA, Italy, Germany, UK, and Spain. One goal of several studies was to estimate the levels abused drugs discharged. Most of the main illicit drugs consumed are excreted unaltered or as slightly transformed metabolites, which reach the sewage system. The detection of these drugs shows that the parent compound or conjugates can pass STPs and reach receiving waters that may be used for drinking-water production. Daily and seasonal variability was examined and revealed fluctuations in the concentrations of nicotine, paraxanthine, amphetamine, cocaine, and ecstasy during the week. Estimations of consumption were made using the total concentrations found in wastewater (Zuccato et al., 2008a, 2008b; Hummel et al., 2006; Boleda et al., 2009; Huerta-Fontela et al., 2007, 2008, 2009; Jones-Lepp et al., 2004). Studies are available for Italy (Zuccato et al., 2005; Castiglioni et al., 2006), Germany (Hummel et al. 2006), USA (Jones-Lepp et al., 2004), Ireland (Bones et al., 2007), Belgium (Gheorghe et al., 2008; van Nuijs et al., 2008), UK (Kasprzyk-Hordern et al., 2007; Zuccato et al., 2008a, 2008b), Poland (Kasprzyk-Hordern et al., 2008), and in Spain (Boleda et al., 2009). Concentrations of methamphetamine were 0.8 and 0.5 ng l1 for MDMA in effluent samples from three WWTPs in the USA. The presence of cocaine (42–120 ng l1) and its major metabolite, benzoylecgonine (390–750 ng l1), in wastewaters and in the Po River (2 and 25 ng l1, respectively) was demonstrated. Cocaine and its metabolite were detected in Spanish

81

wastewaters at concentrations ranging from 4 to 4700 ng l1 and from 9 to 7500 ng l1, respectively (Huerta-Fontela et al., 2008). Compounds of the amphetamine type, morphine, or methadone were detected in the influents and effluents of two Italian WWTPs and in several Spanish ones (2–688 ng l1). Removal rates for cocaine and benzoylecgonine were higher than 88%, whereas those for amphetamine type compounds varied, ranging from 40% to more than 99%. Cannabis (THC) has been detected as its main carboxylic metabolite (THCCOOH) at concentrations from 0.5 to 7 ng l1 in Italian rivers and at 1 ng l1 in one UK River. Huerta-Fontela et al. (2009) described the removal of illicit drugs through drinking-water treatment. Drinking-water treatments achieved the complete removal of all the illicit drugs or metabolites detected in the raw water except for benzoylecgonine, a cocaine metabolite, methadone, and a methadone metabolite. Amphetamine-type stimulants (except MDMA) were completely removed during prechlorination, flocculation, and sand filtration steps. Granulated activated carbon filtration removed cocaine (100%), MDMA (88%), and benzoylecgonine (72%). Post-chlorination resulted in the complete elimination of MDMA. However, no information was reported for reaction by-products.

3.04.15.3 Metabolites In general, little is known about the occurrence, fate, and activity of metabolites of pharmaceuticals. An important question to be addressed is whether the glucoronides, methylates, glycinates, acetylates, and sulfates are still active. It has been found that some compounds (e.g., conjugates of sulfamethoxazole and ethinlyestradiol) can be cleaved back by bacteria during sewage treatment (D’Ascenzo et al., 2003; Go¨bel et al., 2005). This results in the active compound being set free again. Bendz et al. (2005) detected human ibuprofen metabolites not only in the WWTP but also in the receiving river, and carbamazepine metabolites were found in WWTP effluent and even in drinking water (Hummel et al., 2006; Miao et al., 2005). Other types of metabolites are excreted too and can be detected in wastewater (Miao et al., 2005). Their effects on environmental organisms may be less than that of the parent. However, in the case of pro-drugs the situation is probably different as it may also be for the metabolites of several other pharmaceuticals as has been shown, for example, for norfluoxetin.

3.04.16 Engineered Nanoparticles Nanomaterials are of increasing technological and economical importance, and, in turn, contribute to higher standards of living. Important properties include size, shape, surface properties, aggregation state, solubility, structure, and chemical composition. According to chemical composition and shape, for example, there are several classes of engineered nanoparticles (ENPs) such as inorganic particles (e.g., TiO2, SiO2, and ZnO), organic ones such as fullerens, and multiand single-walled carbon nanotubes. Some of them are chemically modified, that is, they carry organic functionalities on their surface. Insoluble material such as fullerens may

82

Emerging Contaminants

become soluble by such a modification. As a result of their widespread use, their exposure to the human body is increasing, for example, by ingestion or application onto skin (e.g., use of sun screens), or through the use or handling of materials or products containing them. Some of them can also reach the environment via production processes (emissions into air and wastewater) and even more likely as a result of the routine use of products containing nanoparticles (e.g., sun screens). Others are the result of wash off, for example, from facades of buildings (Kaegi et al., 2008). These authors found evidence that synthetic nanoparticles are released from urban applications into the aquatic environment. They investigated TiO2 particles as these particles are used in large quantities in exterior paints as whitening pigments and are, to some extent, also present in the nano-size range. Analytical electron microscopy revealed that TiO2 particles are detached from new and aged facade paints by natural weather conditions and are then transported by facade runoff and are discharged into natural, receiving waters. Some of the ENPs are used for the remediation of environmental pollution too such as the treatment of effluents and wastewater (e.g., photocatalysis by TiO2). In such cases, the introduction of nanomaterials is intended; however, the focus of an LCA of such materials is indispensable for the sustainable use of these nanoparticles. There is an increasing concern over the safety of ENPs to humans and the environment and it is likely that the environmental risks of these particles will have to be tested under regulatory schemes such as REACH. However, to date, little is known about the occurrence, fate, and toxicity of ENPs. Knowledge on toxicological issues of nanomaterials is almost nonexistent (Albrecht et al., 2006; Landsiedel et al., 2009) and still less as far as the environment is concerned. Frequently, genotoxicity seems to be associated with nanomaterials (Landsiedel et al., 2009). Predicted environmental concentrations (PECs) of ENPs in water arising form their use in consumer products range from o0.1 ng l1 (CeO2) to 310 ng l1 for fullerenes up to 76 mg l1 for ZnO (Tiede et al., 2009). Gottschalk et al. (2009) calculated PECs based on a probabilistic material flow analysis from a life-cycle perspective of ENP-containing products. The simulated modes (most frequent values) range from 0.003 ng l1 (fullerenes) to 21 ng l1 (TiO2) for surface waters and from 4 ng l1 (fullerenes) to 4 mg l1 (TiO2) for sewage treatment effluents. These data demonstrate that there is still only knowledge on usage and introduction of ENPs into the environment. Gottschalk et al. (2009) concluded that risks to aquatic organisms may currently emanate from nano-Ag, nano-TiO2, and nano-ZnO in sewage treatment effluents for all considered regions and for nano-Ag in surface waters and that for the other environmental compartments for which ecotoxicological data were available, no risks to organisms are presently expected. However, the aggregation of ENPs plays an important role in the environmental fate and effects because the size and shape of nanoparticles will determine the magnitude of any potentially toxic effect. Aggregation is affected by pH, ionic strength, and ionic identity (inorganic and organic) of aqueous suspensions (Sharma, 2009). In Europe, the Water Framework Directive (WFD) is responsible for maintaining a good chemical and ecological

status of surface waters. In this context, ‘priority substances’ are set up. Braun et al. (2009) concluded that it is impossible to set limit values for ENPs in surface waters now and in the foreseeable future. This is not only due to the extensive lack of knowledge in relation to unknown toxic effects, degradability, and bioaccumulation of ENPs in the aquatic environment, but also due to the questionable validity of test systems and methods to establish environmental quality standards (EQS). The limitations in our knowledge are partly due to the lack of methodology for the detection and characterization of engineered nanoparticles in complex matrices, that is, water, soil, or food (Tiede et al., 2008; Perez et al., 2009; Frimmel and Nießner, 2010). Methods have been developed for the detection of natural or engineered nanomaterials in simple matrices, which could be optimized to provide the necessary information, including microscopy, chromatography, spectroscopy, centrifugation, as well as filtration and related techniques. A combination of these is often required. A number of challenges will arise when analyzing environmental materials, including extraction challenges, the presence of analytical artifacts caused by sample preparation, problems of distinction between natural and engineered nanoparticles, and lack of reference materials.

3.04.17 Artificial Sweeteners Artificial low-calorie sweeteners are consumed in considerable quantities with food and beverages. After ingestion, some sweeteners pass through the human metabolism largely unaffected, are quantitatively excreted via urine and feces, and thus reach the environment associated with domestic wastewater. Typical sweeteners are sucralose, acesulfame, cyclamate (banned in the US), and saccarin. Their presence in the aquatic environment came into focus only recently and knowledge is still little. Concentrations in two influents of German STPs were up to 190 mg l1 for cyclamate, about 40 mg l1 for acesulfame and saccharin, and less than 1 mg l1 for sucralose (Scheurer et al., 2009). Buerge et al. (2006) detected acesulfame consistently in untreated and treated wastewater (12–46 mg l1). Removal in the STPs was limited for acesulfame and sucralose and 494% for saccharin and cyclamate (Scheurer et al., 2009). In German surface waters, acesulfame was the predominant artificial sweetener with concentrations exceeding 2 mg l1. Other sweeteners were detected up to several hundred ng l1 in the order saccharin approximate to cyclamate higher than sucralose (Scheurer et al., 2009). The analysis of 120 river surface water samples from 27 European countries showed that sucralose, which is in use in Europe since the beginning of 2005, can be found in the aquatic environment, at concentrations up to 1 mg l1. Sucralose was predominately found in samples from the UK, Belgium, the Netherlands, France, Switzerland, Spain, Italy, Norway, and Sweden, suggesting an increased use of the substance in Western Europe (Loos et al., 2009a). In North American coastal and open ocean waters, the concentration of sucralose varied over several orders of magnitude in these environmental samples with the greatest abundance in a WWTP effluent (120 mg l1).

Emerging Contaminants

The concentration decreased in receiving waters where surface water concentrations at the mouth of the estuary were 374 ng l1. Sucralose was also detected in the oligotrophic waters of the Gulf Stream (33 28.61 N–76 48.21 W) where it ranged in concentration from below detection limit to 68 ng l1 (Mead et al., 2009). In the Northern and Middle Florida Keys, values were 147 and 394 ng l1, respectively. Acesulfame was measured by Buerge et al. (2006) in groundwater samples, and even in several tap water samples (up to 2.6 mg l1) from Switzerland. Up to 4.7 mg l1 were detected in groundwater that received considerable infiltration of river water, where the infiltrating water received considerable discharges from WWTPs. Concentrations increased with population in the catchment area and decreased. with water throughflow. The persistence of some artificial sweeteners during soil aquifer treatment was demonstrated by Scheurer et al. (2009). These data show that these artificial sweeteners were not eliminated in WWTPs and were quite persistent in surface waters. Like some pharmaceuticals and dadolium contract media, the artificial sweeteners are regarded as ideal marker for the detection of domestic wastewater in natural waters, particularly groundwater (Buerge et al., 2006; Scheurer et al., 2009).

3.04.18 Cyanotoxins Cyanobacteria are ubiquitous organisms found in all types of aquatic environments. During favorable environmental conditions, cyanobacteria form dense growth referred to as algal bloom or scum. Cyanobacterial blooms occur globally as a result of eutrophication – may it be for natural reasons or as a result of human activities. Eutrophication favors the growth of toxin-producing cyanobacteria, which are algae and not bacteria in the strict sense. There seems to be a worldwide increase in the occurrence of cyanobacterial harmful algal blooms in natural and man-made water reservoirs. Cyanotoxins consist of several classes such as microcystins anatoxins, saxitoxins, and cylindrospermopsins (Pelaez et al., 2009). Cyanotoxins may exert effects on humans and wildlife. In addition to the presence of toxins, water-supply problems associated with cyanobacteria include an unpleasant taste and odor imparted to the water (Watson et al., 2008). Brittain et al. (2000) reported microcystin levels as high as 3.4 mg l1 in Lake Erie. The same concentration level was found later again for microcystin and other cyanotoxins in other Great lakes (Boyer, 2008; Makarewicz et al., 2006). Concentrations up to 97 mg l1 have been measured in Florida and even higher ones in other places in the USA (Pelaeza et al., 2009). Cyanobacteria blooms have been reported for water bodies in Australia, Europe (Bla´hova´ et al., 2008; Boaru et al., 2006; Carrasco et al., 2006; Chorus, 2002; Gkelis et al., 2005; Karlsson et al., 2005), as well as in South America (Ame´ et al., 2003; Deblois et al., 2008), Asia (Li et al., 2008a, 2008b) and in Africa (Nasri et al., 2008; Oudra et al., 2008) – all more or less in the same concentration range (Pelaez et al., 2009). A median of 0.2 mg l1 has been recorded in samples from drinking-water supplies (in both 2004 and 2005) with a maximum of 17 mg l1.

83

References Achten C, Kolb A, and Pu¨ttmann W (2002) Occurrence of methyl tert-butyl ether (MTBE) in riverbank filtered water and drinking water produced by riverbank filtration. Environmental Science and Technology 36: 3662--3670. Ahel M and Jelicˇic I (2001) Phenazone analgesics in soil and groundwater below a municipal solid waste landfill. In: Daughton CG and Jones-Lepp TL (eds.) Pharmaceuticals and Personal Care Products in the Environment: Scientific and Regulatory Issues, ACS Symposium Series 791, pp. 100–115. Washington, DC: American Chemical Society. Albrecht MA, Evans CW, and Raston CL (2006) Green chemistry and the health implications of nanoparticles. Green Chemistry 8: 417--432. Ame´ MV, Diaz MP, and Wunderlin DA (2003) Ocurrence of toxic cyanobacterial blooms in San Roque Reservoir (Co´rdoba, Argentina): A field and chemometric study. Environmental Toxicology 18: 192--201. Anastas PT and Warner JC (1998) Green Chemistry: Theory and Practice. New York: Oxford University Press. Andresen JA, Muir D, Ueno D, Darling C, Theobald N, and Bester K (2007) Emerging pollutants in the North Sea in comparison to Lake Ontario, Canada, Data. Environmental Toxicology and Chemistry 26: 1081--1089. Arslan-Alaton I, Olmez-Hanci T, Gursoy BH, and Tureli G (2010) Commercially important aryl sulphonates in the environment: fate, treatability and identification of degradation products. In: Kassinos D, Bester K, and Ku¨mmerer K (eds.) Xenobiotics in the Urban Water Cycle. New York, NY: Springer. Barcelo´ D and Petrovic M (eds.) (2008) Emerging Contaminants from Industrial and Municipal Waste. Berlin: Springer. Bendz D, Paxee´us NA, Ginn TR, and Loge FJ (2005) Occurrence and fate of pharmaceutically active compounds in the environment, a case study: Ho¨je River in Sweden. Journal of Hazardous Materials 122: 195--204. Bester K (2005) Polycyclic musks in the Ruhr catchment area – transports, discharges of wastewater, and transformations of HHCB, AHTN, and HHCB-lactone. Journal of Environmental Monitoring 7: 43--51. Bester K (2007) Personal Care Compounds in the Environment. Weinheim: VCH-Wiley. Bester K, McArdell CS, Wahlberg C, and Bucheli TD (2009) Quantitative mass flows of selected xenobiotics in urban waters and waste water treatment plants. In: Kassinos D, Bester K, and Ku¨mmerer K (eds.) Xenobiotics in the Urban Water Cycle, pp. 3--26. New York, NY: Springer. Bhutta MF (2007) Sex and the nose: Human pheromonal responses. Journal of the Royal Society of Medicine 100: 268--274. Bitsch N, Dudas C, Ko¨rner W, et al. (2001) Estrogenic activity of musk fragrances detected by the E-screen assay using human mcf-7 cells. Archives of Environmental Contamination and Toxicology 43: 257--264. Bla´hova´ L, Babica P, Adamovsky´y O, Kohoutek J, Marsa´lek B, and Blaha L (2008) Analyses of cyanobacterial toxins (microcystins, cylindrospermopsin) in the reservoirs of the Czech Republic and evaluation of health risks. Environmental Chemistry Letters 6: 223--227. Blair RM, Fang H, Branham WS, et al. (2000) The estrogen receptor relative binding affinities of 188 natural and xenochemicals: Structural diversity of ligands. Toxicological Sciences 54: 138--153. Boaru DA, Dragos N, Welker M, Bauer A, Nicoara A, and Schirmer K (2006) Toxic potential of microcystin-containing cyanobacterial extracts from three Romanian freshwaters. Toxicon 47: 925--932. Boethling RS, Sommer E, and DiFiore D (2007) Designing small molecules for biodegradability. Chemical Reviews 107: 2207--2227. Boleda MR, Galceran MT, and Ventura F (2009) Monitoring of opiates, cannabinoids and their metabolites in wastewater, surface water and finished water in Catalonia, Spain. Water Research 43: 1126--1136. Bolek R and Ku¨mmerer K (2010) Fate and effects of little investigated scents in the aquatic environment. In: Fatta-Kassinos D, Bester K, and Ku¨mmerer K (eds.) Xenobiotics in the Urban Water Cycle, pp. 87--100. New York, NY: Springer. Bones J, Thomas KV, and Brett P (2007) Using environmental analytical data to estimate levels of community consumption of illicit drugs and abused pharmaceuticals. Journal of Environmental Monitoring 9: 701--707. Boxall ABA, Fogg LA, Kay P, Blackwell PA, Pemberton EJ, and Croxford A (2003a) Veterinary medicines in the environment. Reviews of Environmental Contamination and Toxicology 180: 1--91. Boxall ABA, Kolpin D, Halling-Sørensen B, and Tolls J (2003b) Are veterinary medicines causing environmental risks? Environmental Science and Technology 36: 286A--294A. Boxall ABA, Sinclair CJ, Fenner K, Kolpin D, and Maud SJ (2004) When synthetic chemicals degrade in the environment. Environmental Science and Technology 38: 368A--375A.

84

Emerging Contaminants

Boyer GL (2008) Cyanobacterial toxins in New York and the lower Great Lakes ecosystems. Advances in Experimental Medicine and Biology 619: 153--165. Braun A, Man A, Hartmann NB, and Grieger KD (2009) Setting the limits for engineered nanoparticles in European surface waters – are current approaches appropriate? Journal of Environmental Monitoring 11: 1774--1781. Brittain SM, Wang J, Babcock-Jackson L, Carmichael WW, Rinehart KL, and Culver DA (2000) Isolation and characterization of microcystins, cyclic heptapeptide hepatotoxins from a Lake Erie strain of Microcystis aeruginosa. Journal of Great Lakes Research 26: 241--249. Brown KD, Kulis J, Thomson B, Chapman TH, and Mawhinney DB (2006) Occurrence of antibiotics in hospital, residential, and dairy effluent, municipal wastewater, and the Rio Grande in New Mexico. Science of the Total Environment 366: 772--783. Buerge IJ, Poiger T, Mu¨ller MD, and Buser HR (2006) Combined sewer overflows to surface waters detected by the anthropogenic marker caffeine. Environmental Science and Technology 40: 4096--4102. Buszka PM, Yeskis DJ, Kolpin DW, Furlong ET, Zaugg SD, and Meyer MT (2009) Waste-indicator and pharmaceutical compounds in landfill-leachate-affected ground water near Elkhart, Indiana, 2000–2002. Bulletin of Environmental Contamination and Toxicology 82: 653--659. Caldwell DJ, Mastrocco R, Hutchinson TH, et al. (2008) Derivation of an aquatic predicted no-effect concentration for the synthetic hormone, 17a-ethinyl estradiol. Environmental Science and Technology 42: 7046--7054. Cancilla DA, Baird JC, Geis S, and Corsi SR (2003a) Studies of the environmental fate and effect of aircraft deicing fluids: Detection of 5-methyl-1H-benzotriazole in the fathead minnow (Pimephales promelas). Environmental Toxicology and Chemistry 22: 134--138. Cancilla DA, Baird JC, and Rosa R (2003b) Detection of aircraft deicing additives in groundwater and soil samples from Fairchild Air Force Base, a small to moderate user of deicing fluids. Bulletin of Environmental Contamination and Toxicology 70: 868--875. Carrasco D, Moreno E, Sanchis D, et al. (2006) Cyanobacterial abundance and microcystin occurrence in Mediterranean water reservoirs in Central Spain: Microcystins in the Madrid area. European Journal of Phycology 41: 281--291. Castiglioni S, Zuccato E, Crisci E, Chiabrando C, Fanelli R, and Bagnati R (2006) Identification and measurement of illicit drugs and their metabolites in urban wastewater by liquid chromatography–tandem mass spectrometry. Analytical Chemistry 178: 8421--8429. Chorus I (2002) Cyanobacterial toxin research and its application in Germany: A review of the current status. Environmental Toxicology 17: 358--360. Clark JH (2006) Green chemistry: Today (and tomorrow). Green Chemistry 8: 17--21. Clark JH and Smith P (2005) Basic principles of green chemistry. Innovations in Pharmaceutical Technology 16: 94--97. Cunningham V (2008) Special characteristics of pharmaceuticals related to environmental fate. In: Ku¨mmerer K (ed.) Pharmaceuticals in the Environment: Sources, Fate, Effects and Risks, 3rd edn., pp. 23--34. Berlin: Springer. D’Ascenzo G, Di Corcia A, Gentili A, et al. (2003) Fate of natural estrogen conjugates in municipal sewage transport and treatment facilities. Science of the Total Environment 302: 199--209. Daughton CG (2003) Cradle-to-cradle stewardship of drugs for minimizing their enviromental disposition while promoting human health. I. Rationale for and avenues toward a green pharmacy. Environmental Health Perspectives 111: 757--774. Deblois CP, Aranda-Rodriguez R, Giani A, and Bird DF (2008) Microcystin accumulation in liver and muscle of tilapia in two large Brazilian hydroelectric reservoirs. Toxicon 51: 435--448. Desbrow C, Routledge EJ, Brighty GC, Sumpter JP, and Waldock M (1998) Identification of estrogenic chemicals in STW effluent. 1. Chemical fractionation and in vitro biological screening. Environmental Science and Technology 32: 1549--1558. DFG (German Research Council, Commission for Water Reserach) (2003) Wasserforschung im Spannungsfeld zwischen Gegenwartsbewa¨ltigung und Zukunftssicherung (Denkschrift). Weinheim: Wiley-VCH. Dicke M and Sabelis MW (1988) Infochemical terminology: Based on cost-benefit analysis rather than origin of compounds? Functional Ecology 2: 131--139. Drewes JE, Heberer T, and Reddersen K (2002) Fate of pharmaceuticals during indirect potable use. Water Science and Technology 46: 73--80. Drinking Water Directive (DWD) (1998) Council Directive 98/83/EC of 3 November 1998 on the quality of water intended for human consumption. Official Journal of the European Communities L 330/32 (05 December 1998). ECB (European Chemicals Bureau) (2007) Draft risk assessment report on 2,20 ,6,60 tetrabromo-4,40 -isopropylidenediphenol (Tetrabromobisphenol A), Ispra, Italy. (Human Health, CAS 79-94-7, EINECS 201-236-9).

Eckel WP, Ross B, and Isensee R (1993) Pentobarbital found in ground water. Ground Water 31: 801--804. Elbaz-Poulichet F, Seidel JL, and Othoniel C (2002) Occurrence of an anthropogenic gadolinium anomaly in river and coastal waters of southern France. Water Research 36: 1102--1105. Esperanza M, Suidan MT, Marfil-Vega R, et al. (2007) Fate of sex hormones in two pilot-scale municipal wastewater treatment plants: Conventional treatment. Chemosphere 66: 1535--1544. EU (2000) Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy. Official Journal of the European Union L327: 1–72 (22 December 2000). EU (2003) Directive 2003/87/EC of the European Parliament and of the Council of 13 October 2003 establishing a scheme for greenhouse gas emission allowance trading within the Community and amending Council Directive 96/61/EC. Official Journal of the European Union L 275/32 (25 October 2003). European Commission (2003) Guidance document on the assessment of the relevance of metabolites in groundwater of substances regulated under council directive 91/ 414/EEC. Sanco/221/2000-rev.10, pp. 1–14. European Commission – Health and Consumer Protection Directorate-General, 25 February 2003. Fatta-Kassinos D, Bester K, and Ku¨mmerer K (eds.) (2010) Xenobiotics in the Urban Water Cycle. New York, NY: Springer. Ferk F, Misı´k M, Hoelzl C, et al. (2007) Benzalkonium chloride (BAC) and dimethyldioctadecyl-ammonium bromide (DDAB), two common quaternary ammonium compounds, cause genotoxic effects in mammalian and plant cells at environmentally relevant concentrations. Mutagenesis 22: 363--370. Ferrer I and Furlong ET (2002) Accelerated solvent extraction followed by on-line solid-phase extraction coupled to ion trap LC/MS/MS for analysis of benzalkonium chlorides in sediment samples. Analytical Chemistry 74: 1275--1280. Frimmel FH and Nießner R (eds.) (2010) Nanoparticles in the Water Cycle. Properties, Analysis and Environmental Relevance. Berlin: Springer. Gautschi M, Bajgrowicz JA, and Kraft P (2001) Fragrance chemistry – milestones and perspectives. Chimia 55: 379--387. Giger W (2009) Hydrophilic and amphiphilic water pollutants: using advanced analytical methods for classic and emerging contaminants. Analytical and Bioanalytical Chemistry 393: 37--44. Gheorghe A, van Nuijs A, Pecceu B, Bervoets L, Jorens PG, and Blust R (2008) Analysis of cocaine and its principal metabolites in waste and surface water using solid-phase extraction and liquid chromatography-ion trap tandem mass spectrometry. Analytical and Bioanalytical Chemistry 391: 1309--1319. Giokas DL, Sakkas VA, and Albanis TA (2004) Determination of residues of UV filters in natural waters by solid-phase extraction coupled to liquid chromatography– photodiode array detection and gas chromatography–mass spectrometry. Journal of Chromatography A 1026: 289--293. Giokas DL, Sakkas VA, and Albanis TA (2005) Erratum. Journal of Chromatography A 1068: 339. Gkelis S, Harjunpa V, Lanaras T, and Sivonen K (2005) Diversity of hepatotoxic microcystins and bioactive anabaenopeptins in cyanobacterial blooms from Greek freshwaters. Environmental Toxicology 20: 249--256. Go¨bel A, Thomsen A, McArdell CS, Joss A, and Giger W (2005) Occurrence and sorption behavior of sulfonamides, macrolides, and trimethoprim in activated sludge treatment. Environmental Science and Technology 39: 3981--3989. Go´mez MJ, Petrovic´ M, Ferna´ndez-Alba AR, and Barcelo´ D (2006) Determination of pharmaceuticals of various therapeutic classes by solid-phase extraction and liquid chromatography-tandem mass spectrometry analysis in hospital effluent wastewaters. Journal of Chromatography A 1114: 224--233. Goossens H, Ferech M, Coenen S, and Stephens P (2007) European surveillance of antimicrobial consumption project group. Comparison of outpatient systemic antibacterial use in 2004 in the United States and 27 European countries. Clinical Infectious Diseases 44: 1091--1095. Goossens H, Ferech M, Vander Stichele R, Elseviers M, and ESAC Project Group (2005) Outpatient antibiotic use in Europe and association with resistance: A cross-national database study. Lancet 365: 579--587. Gottschalk F, Sonderer T, Scholz RW, and Nowak B (2009) Modeled environmental concentrations of engineered nanomaterials (TiO2, ZnO, Ag, CNT, fullerenes) for different regions. Environmental Science and Technology 43: 9216--9222. Gro¨ning J, Held C, Garten C, Claussnitzer U, Kaschabek SR, and Schlo¨mann M (2007) Transformation of diclofenac by the indigenous microflora of river sediments and identification of a major intermediate. Chemosphere 69: 509--516. Guruge KS, Taniyasu S, Yamashita N, et al. (2005) Perfluorinated organic compounds in human blood serum and seminal plasma: A study of urban and rural tea worker populations in Sri Lanka. Journal of Environmental Monitoring 7: 371--377. Ha¨drich C (2006) Quellen des Eintrags von Arzneimitteln in die aquatische Umwelt – Eine regionale Bilanzierung. PhD Thesis, University of Freiburg, Germany.

Emerging Contaminants

Haiß A and Ku¨mmerer K (2006) Biodegradability of the X-ray contrast compound diatrizoic acid, identification of aerobic degradation products and effects against sewage sludge micro-organisms. Chemosphere 62: 294--302. Hanke I, Singer H, McArdell CS, et al. (2007) Arzneimittel und Pestizide im Grundwasser. Gas Wasser Abwasser 3: 187--196. Hartmann A, Golet EM, Gartiser S, Alder AC, Koller T, and Widmer RM (1999) Primary DNA damage but not mutagenicity correlates with ciprofloxacin concentrations in German hospital wastewaters. Archives of Environmental Contamination and Toxicology 36: 115--119. Heberer T (2002) Occurrence, fate, and removal of pharmaceutical residues in the aquatic environment: A review of recent research data. Toxicology Letters 131: 5--17. Heberer T and Feldmann D (2008) Removal of pharmaceutical residues from contaminated raw water sources by membrane filtration. In: Ku¨mmerer K (ed.) Pharmaceuticals in the Environment: Sources, Fate, Effects and Risks, 3rd edn., pp. 427--453. Berlin: Springer. Hem LJ, Hartnik T, Roseth R, and Breedveld GD (2003) Photochemical degradation of benzotriazole. Journal of Environmental Science and Health Part A – Environmental Science and Engineering and Toxic/Hazardous Substances 38: 471--481. Hoelzer J, Midasch O, Midasch O, et al. (2008) Exposure to perfluorooctanoate (PFOA)-contaminated drinking water resulted in increased PFOA plasma concentrations – a biomonitoring study from Arnsberg, Germany. Epidemiology 19: 112. Holm JV, Ruegge K, Bjerg PL, and Christensen TH (1995) Occurrence and distribution of pharmaceutical organic compounds in the ground water down gradient of a landfill (Grindsted, Denmark). Environmental Science and Technology 29: 1415--1420. Houde M, Martin JW, Letcher RJ, Solomon KR, and Muir DC (2006) Biological monitoring of polyfluoroalkyl substances: A review. Environmental Science and Technology 40: 3463--3473. HSDB (2009) http://toxnet.nlm.nih.gov/cgi-bin/sis/search/f?./temp/BJEnD2t:1. Huerta-Fontela M, Galceran MT, Martin-Alonso J, and Ventura F (2008) Occurrence of psychoactive stimulatory drugs in wastewaters in north-eastern Spain. Science of the Total Environment 397: 31--40. Huerta-Fontela M, Galceran MT, and Ventura F (2007) Ultraperformance liquid chromatography–tandem mass spectrometry analysis of stimulatory drugs of abuse in wastewater and surface waters. Analytical Chemistry 79: 3821--3829. Huerta-Fontela M, Galceran MT, and Ventura F (2009) Illicit drugs in the urban water cycle. In: Fatta-Kassinos D, Bester K, and Ku¨mmerer K (eds.) Xenobiotics in the Urban Water Cycle, pp. 51--72. New York, NY: Springer. Hummel D, Loeffler D, Fink G, and Ternes TA (2006) Simultaneous determination of psychoactive drugs and their metabolites in aqueous matrices by liquid chromatography mass spectrometry. Environmental Science and Technology 40: 7321--7328. IARC (1987) Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Man. 1972–PRESENT, pp. S7–S57. Geneva: World Health Organization, International Agency for Research on Cancer. Illinois EPA (2007) Report on alternatives to the flame retardant DecaBDE: Evaluation of toxicity, availability, affordability, and fire safety issues. A Report to the Governor and the General Assembly. http://www.epa.state.il.us/reports/decabde-study/ index.html (accesed March 2010). Isidori M, Lavorqna M, Nardelli A, Parella A, Previtera L, and Rubino M (2005) Ecotoxicity of naproxen and its phototransformation products. Science of the Total Environment 348: 93--101. Isidori M, Nardelli A, Pascarella M, Rubino M, and Parrella A (2007) Toxic and genotoxic impact of fibrates and their photoproducts on non-target organisms. Envrionment International 33: 635--641. Jandera P, Fischer J, and Prokes B (2001) HPLC determination of chlorobenzenes, benzenesulphonyl chlorides and benzenesulphonic acids in industrial wastewater. Chromatographia 54: 581--587. Jeon HK, Chung Y, and Ryu JC (2006) Simultaneous determination of benzophenonetype UV filters in water and soil by gas chromatography–mass spectrometry. Journal of Chromatography A 1131: 192--202. Jobling S, Williams R, Johnson A, et al. (2006) Predicted exposures to steroid estrogens in U.K. rivers correlate with widespread sexual disruption in wild fish populations. Environmental Health Perspectives 114(supplement 1): 32--39. Jones OHA, Green PG, Voulvoulis N, and Lester JN (2007) Questioning the excessive use of advanced treatment to remove organic micro-pollutants from waste water. Environmental Science and Technology 41: 5085--5089. Jones-Lepp TL, Alvarez DA, Petty JD, and Huckins JN (2004) Polar organic chemical integrative sampling and liquid chromatography electrospray/ion-trap mass spectrometry for assessing selected prescription and illicit drugs in treated sewage effluents. Archives of Environmental Contamination and Toxicology 47: 427--439.

85

Jones-Lepp TL and Stevens R (2007) Pharmaceuticals and personal care products in biosolids/sewage sludge: The interface between analytical chemistry and regulation. Analytical and Bioanalytical Chemistry 387: 1173--1183. Kaegi R, Ulrich A, Sinnet B, et al. (2008) Synthetic TiO2 nanoparticle emission from exterior facades into the aquatic environment. Environmental Pollution 156: 233--239. Kallenborn R, Fick R, Lindberg R, et al. (2008) Pharmaceutical residues in Northern European environments: Consequences and perspectives. In: Ku¨mmerer K (ed.) Pharmaceuticals in the Environment: Sources, Fate, Effects and Risks, 3rd edn., pp. 61--74. Berlin: Springer. Kanda R and Churchley J (2008) Removal of endocrine disrupting compounds during conventional wastewater treatment. Environmental Technology 29: 315--323. Karlsson KM, Kankaanpa¨a¨ H, Huttunen M, and Meriluoto J (2005) First observation of microcystin-LR in pelagic cyanobacterial blooms in the northern Baltic Sea. Harmful Algae 4: 163--166. Kasprzyk-Hordern B, Dinsdale RM, and Guwy AJ (2007) Multi-residue method for the determination of basic/neutral pharmaceuticals and illicit drugs in surface water by solid-phase extraction and ultra performance liquid chromatography–positive electrospray ionisation tandem mass spectrometry. Journal of Chromatography A 1161: 132--145. Kasprzyk-Hordern B, Dinsdale RM, and Guwy AJ (2008) The occurrence of pharmaceuticals, personal care products, endocrine disruptors and illicit drugs in surface water in South Wales, UK. Water Research 42: 3498--3518. Kidd KA, Blanchfield PJ, Mills KH, et al. (2007) Collapse of a fish population after exposure to a synthetic estrogen. Proceedings of the National Academy of Sciences of the United States of America 104: 8897--8901. Kiffmeyer T (2003) Minimisation of human drug input by oxidative treatment of toilet effluents from hospital wards. European Conference on Human and Veterinary Pharmaceuticals in the Environment. Lyon, 14–16 April 2003. Klaschka U (2008) Odorants – potent substances at minor concentrations: The ecological role of infochemicals. In: Ku¨mmerer K (ed.) Pharmaceuticals in the Environment: Sources, Fate, Effects and Risks, 3rd edn., pp. 305--320. Berlin: Springer. Knepper TP, Sacher F, Lange FT, et al. (1999) Detection of polar organic substances relevant for drinking water. Waste Management 19: 77--99. Kolpin DW, Kalkhoff SJ, Goolsby DA, Sneck-Fahrer DA, and Thurman EM (1997) Occurrence of selected herbicides and herbicide degradation products in Iowa’s ground water, 1995. Ground Water 35: 679--688. Kolpin DW, Schnoebelen DJ, and Thurman EM (2004) Degradates provide insight to spatial and temporal trends of herbicides in ground water. Ground Water 42: 601--608. Ku¨mmerer K (2001a) Drugs in the environment: emission of drugs into wastewater by hospitals in relation to other sources – a review. Chemosphere 45: 957--969. Ku¨mmerer K (ed.) (2001b) Pharmaceuticals in the Environment: Sources, Fate, Effects and Risks, 1st edn. Berlin: Springer. Ku¨mmerer K (ed.) (2004) Pharmaceuticals in the Environment: Sources, Fate, Effects and Risks, 2nd edn. Berlin: Springer. Ku¨mmerer K (2007) Sustainable from the very beginning: Rational design of molecules by life cycle engineering as an important approach for green pharmacy and green chemistry. Green Chemistry 9: 899--907. Ku¨mmerer K (ed.) (2008a) Pharmaceuticals in the Environment: Sources, Fate, Effects and Risks, 3rd edn. Berlin: Springer. Ku¨mmerer K (2008b) Design of inherently safe molecules. In: Nelson W (ed.) Oxford Handbook of Green Chemistry: From Molecular Design to Process Scale Up. Oxford University Press (submitted). Ku¨mmerer K (2009a) The presence of pharmaceuticals in the environment due to human use – present knowledge and future challenges. Journal of Environmental Management 90(8): 2354--2366. Ku¨mmerer K (2009b) Antibiotics in the aquatic environment – a review – part I. Chemosphere 75: 417--434. Ku¨mmerer K (2009c) Antibiotics in the aquatic environment – a review – part II. Chemosphere 75: 435--441. Ku¨mmerer K, Eitel A, Braun U, et al. (1997) Analysis of benzalkonium chloride in the effluent from European hospitals by solidphase extraction and high-performance liquid chromatography with post-column ion-pairing for fluorescence detection. Journal of Chromatography A 774: 281--286. Ku¨mmerer K and Helmers E (1997) Hospital effluents as a source for platinum in the environment. Science of the Total Environment 193: 179--184. Ku¨mmerer K and Helmers E (2000) Hospital effluents as a source for gadolinium in the aquatic environment. Environmental Science and Technology 34: 573--577.

86

Emerging Contaminants

Ku¨mmerer K and Hempel M (eds.) (2010) Sustainable Pharmacy. Heidelberg: Springer. Lamoree MH, Derksen JGM, van der Linden SC, Uijterlinde CA, and de Voogt P (2009) Efficiency of removal of compounds with estrogenic activity during wastewater treatment – effects of various removal techniques. In: Kassinos D, Bester K, and Ku¨mmerer K (eds.) Xenobiotics in the Urban Water Cycle, pp. 261--282. New York, NY: Springer. La¨ngin A, Schuster A, and Ku¨mmerer K (2008) Chemicals in the environment – the need for a clear nomenclature: Parent compounds, metabolites, transformation products and their elimination. CLEAN – Soil, Air, Water 36: 349--350. Landsiedel R, Kapp MD, Schulz M, Wiench K, and Oesch F (2009) Genotoxicity investigations on nanomaterials: Methods, preparation and characterization of test material, potential artifacts and limitations – many questions, some answers. Mutation Research – Reviews 681: 241--258. Larsson DG, de Pedro C, and Paxeus N (2007) Effluent from drug manufactures contains extremely high levels of pharmaceuticals. Journal of Hazardous Materials 148: 751--755. Ledin A and Patureau D (2008) COST Action 636 xenobiotics in the urban water cycyle – a network for collaboration within Europe. In: Ku¨mmerer K (ed.) Pharmaceuticals in the Environment: Sources, Fate, Effects and Risks, 3rd edn., pp. 419--426. Berlin: Springer. Lee C, Lee Y, Schmidt C, Yoon J, and von Gunten U (2007) Oxidation of Nnitrosodimethylamine (NDMA) with ozone and chlorine dioxide: Kinetics and effect on NDMA formation potential. Environmental Science and Technology 41: 2056--2063. Li D, Yang M, Hu J, Zhang Y, Chang H, and Jin F (2008a) Determination of penicillin G and its degradation products in a penicillin production wastewater treatment plant and the receiving river. Water Research 42: 307--317. Li R, Wilhelm SW, Carmichael WW, and Watanabe MM (2008b) Polyphasic characterization of water bloom forming Raphidiopsis species (cyanobacteria) from central China. Harmful Algae 7: 146--153. Lien NPH, Fujii S, Tanaka S, Nozoe N, and Tanaka H (2008) Contamination of perfluorooctane sulfonate (PFOS) and perfluorooctanoate (PFOA) in surface water of the Yodo River basin (Japan). Desalination 226: 338--347. Loos R, Gawlik BM, Boettcherher K, Locoro G, Contini S, and Bidoglio G (2009a) Sucralose screening in European surface waters using a solid-phase extraction– liquid chromatography–triple quadrupole mass spectrometry method. Journal of Chromatography A 1216: 1126--1131. Loos R, Gawlik BM, Locoro G, Rimaviciute E, Contini S, and Bidoglio G (2009b) EUwide survey of polar organic persistent pollutants in European river waters. Environmental Pollution 157: 561--568. Loos R, Locoro G, Huber T, et al. (2008) Analysis of perfluorooctanoate (PFOA) and other perfluorinated compounds (PFCs) in the river Po watershed in N-Italy. Chemosphere 71: 306--313. Makarewicz JC, Boyer GL, Guenther W, Arnold M, and Lewis TW (2006) The occurrence of cyanotoxins in the nearshore and coastal embayments of Lake Ontario. Great Lakes Research Review 7: 25--29. Marklund A, Andersson B, and Haglund P (2005) Organophosphorus flame retardants and plasticizers in Swedish sewage treatment plants. Environmental Science and Technology 39: 7423--7429. Martı´nez-Carballo E, Sitka A, Gonza´lez-Barreiro C, et al. (2007a) Determination of selected quaternary ammonium compounds by liquid chromatography with mass spectrometry. Part I. Application to surface, waste and indirect discharge water samples in Austria. Environmental Pollution 145: 489--496. Martı´nez-Carballo E, Gonza´lez-Barreiro C, Sitka A, Kreuzinger N, Scharf S, and Gans O (2007b) Determination of selected quaternary ammonium compounds by liquid chromatography with mass spectrometry. Part II. Application to sediment and sludge samples in Austria. Environmental Pollution 146: 543--547. Matamoros V and Bayona JM (2006) Elimination of pharmaceuticals and personal care products in subsurface flow constructed wetlands. Environmental Science and Technology 40: 5811--5816. Mead RN, Morgan JB, Avery GB, et al. (2009) Occurrence of the artificial sweetener sucralose in coastal and marine waters of the United States. Marine Chemistry 116: 13--17. Me´ndez-Arriaga F, Esplugas S, and Gime´nez J (2008) Photocatalytic degradation of non-steroidal anti-inflammatory drugs with TiO2 and simulated solar irradiation. Water Research 42: 585--594. Metzger JW (2004) Drugs in municipal landfills and landfill leachates. In: Ku¨mmerer K (ed.) Pharmaceuticals in the Environment. Sources Fate Effects and Risks, 2nd edn., pp. 133--138. Berlin: Springer. Metzger S, Kapp H, Seitz W, Weber WH, Hiller G, and Su¨ßmuth W (2005) Entfernung von iodierten Ro¨ntgenkontrastmitteln bei der kommunalen Abwasserbehandlung durch den Einsatz von Pulveraktivkohle. GWF Wasser Abwasser 9: 638--645.

Miao XS, Yang JJ, and Metcalfe CD (2005) Carbamazepine and its metabolites in wastewater and in biosolids in a municipal wastewater treatment plant. Environmental Science and Technology 39: 7469--7475. Mo¨ller P, Dulski P, Bau M, Knappe A, Pekdeger A, and Sommer-von Jarmersted C (2000) Anthropogenic gadolinium as a conservative tracer in hydrology. Journal of Geochemical Exploration 69/70: 409--414. Mo¨ller P, Paces T, Dulski P, and Morteania G (2002) Anthropogenic Gd in surface water, drainage system, and the water supply of the city of Prague, Czech Republic. Environmental Science and Technology 36: 2387--2394. Moran MJ, Zogorski JS, and Squillace PJ (2005) MTBE and gasoline hydrocarbons in ground water of the United States. Ground Water 43: 615--627. Nakayama S, Harada K, Inoue K, et al. (2005) Distributions of perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) in Japan and their toxicities. Environmental Sciences 12: 293--313. Nasri H, El Herry S, and Bouaıcha N (2008) First reported case of turtle deaths during a toxic Microcystis spp. bloom in Lake Oubeira, Algeria. Ecotoxicology and Environmental Safety 71: 535--544. Negreira N, Rodriguez I, Ramil M, Rubi E, and Cela R (2009) Sensitive determination of salicylate and benzophenone type UV filters in water samples using solid-phase microextraction, derivatization and gas chromatography tandem mass spectrometry. Analytica Chimica Acta 638: 36--44. Nowotny N, Epp B, von Sonntag C, and Fahlenkamp H (2007) Quantification and modeling of the elimination behavior of ecologically problematic wastewater micropollutants by adsorption on powdered and granulated activated carbon. Environmental Science and Technology 41: 2050--2055. Nylund K, Haglund M, Berggren D, et al. (2002) Brominated flame retardants in sludge (in Swedish). Swedish EPA Report No. 5188. Oh SW, Kang MN, Cho CW, and Lee MW (1997) Detection of carcinogenic amines from dyestuffs or dyed substrates. Dyes and Pigments 33: 119--135. O’Neill C, Hawkes FR, Hawkes DL, Esteves S, and Wilcox SJ (1999) Anaerobic and aerobic biotreatment of simulated textile effluent containing varied ratios of starch and azo dye. Water Research 34: 2355--2361. Ort C, Schaffner C, Gujer W, and Giger W (2005) Modeling stochastic load variations in sewer systems. Water Science and Technology 52: 113--122. Oudra B, Andaloussi Dadi-El M, and Vasconcelos VM (2008) Identification and quantification of microcystins from a Nostoc muscorum bloom occurring in Oukaı¨meden River (High-Atlas mountains of Marrakech, Morocco). Environmental Monitoring and Assessment 149: 437--444. Pelaez M, Antoniou MG, Dionysiou DD, et al. (2009) Source and occurrence of cyanotoxins worldwide. In: Kassinos D, Bester K, and Ku¨mmerer K (eds.) Xenobiotics in the Urban Water Cycle, pp. 101--127. New York, NY: Springer. Perez S, Farre M, and Barcelo D (2009) Analysis, behavior and ecotoxicity of carbonbased nanomaterials in the aquatic environment. TrAC: Trends in Analytical Chemistry 28: 820--832. Purdom CE, Hardiman PA, Bye VJ, Eno NC, Tyler CR, and Sumpter JP (1994) Estrogenic effects of effluents from sewage treatment works. Chemistry and Ecology 8: 275--285. Putschew A, Miehe U, Tellez AS, and Jekel M (2007) Ozonation and reductive deiodination of iopromide to reduce the environmental burden of iodinated X-ray contrast media. Water Science and Technology 56: 159--165. Qiting J and Xiheng Z (1988) Combination process of anaerobic digestion and ozonization technology for treating wastewater from antibiotics production. Water Treatment 3: 285--291. Ravina M, Campanella L, and Kiwi J (2002) Accelerated mineralization of the drug diclofenac via Fenton reactions in a concentric photo-reactor. Water Research 36: 3553--3560. Reemtsma T (1999) Methods of analysis of polar aromatic sulfonates from aquatic environments. Journal of Chromatography 733: 473--489. Reemtsma T and Jekel M (eds.) (2006) Organic Pollutants in the Water Cycle. Weinheim: Wiley-VCH. Reiner JL and Kannan K (2006) A survey of polycyclic musks in selected household commodities from the United States. Chemosphere 62: 867--873. Richardson SD (2009) Water analysis: Emerging contaminants and current issues. Analytical Chemistry 81: 4645--4677. Ricklund N, Kierkegaard A, and McLachlan M (2008a) An international survey of decabromodiphenyl ethane (deBDethane) and decabromodiphenyl ether (decaBDE) in sewage sludge samples. Chemosphere 73: 1799--1804. Ricklund N, Kierkegaard A, McLachlan M, and Wahlberg C (2008b) Mass balance of decabromodiphenyl ethane and decabromodiphenyl ether in a WWTP. Chemosphere 74: 389--394. Rieger PG, Meier HM, Gerle U, Groth T, and Knackmuss HJ (2002) Xenobiotics in the environment: Present and future strategies to obviate the problem of biological persistence. Biotechnology 94: 101--123.

Emerging Contaminants

Roberts T and Hutson D (2002) Metabolic Pathways of Agrochemicals on CD-ROM. Cambridge: The Royal Society of Chemistry. Rodil R, Quintana JB, Lopez-Mahia P, Muniategui-Lorenzo S, and Prada-Rodriguez D (2009) Multi-residue analytical method for the determination of emerging pollutants in water by solid-phase extraction and liquid chromatography–tandem mass spectrometry. Journal of Chromatography A 1216: 2958--2969. Routledge EJ, Sheahan D, Desbrow C, Brighty GC, Waldock M, and Sumpter PJ (1998) Identification of estrogenic chemicals in STW effluent. 2. In vivo responses in trout and roach. Environmental Science and Technology 32: 1559--1565. Russell AD, Hugo WB, and Ayliffe GAJ (1992) Principles and Practice of Disinfection, Preservation and Sterilization, 2nd edn. Oxford: Blackwell Scientific. Salvito DT, Senna RJ, and Federle TW (2002) A framework for prioritizing fragrance materials for aquatic risk assessment. Environmental Toxicology and Chemistry 6: 1301--1308. Salvito DT, Vey MGH, and Senna RJ (2004) Fragrance materials and their environmental impact. Flavour and Fragrance Journal 2: 105--108. Sarmah AK, Meyer MT, and Boxall ABA (2006) A global perspective on the use, sales, exposure pathways, occurrence, fate and effects of veterinary antibiotics (VAs). Chemosphere 65: 725--759. Scheurer M, Brauch HJ, and Lange FT (2009) Analysis and occurrence of seven artificial sweeteners in German waste water and surface water and in soil aquifer treatment (SAT). Analytical and Bioanalytical Chemistry 394: 1585--1594. Schlumpf M, Durrer S, Faass, Ehnes C, et al. (2008) Developmental toxicity of UV filters and environmental exposure: A review. International Journal of Andrology 31: 144–151. Schlu¨ter A, Szczepanowski R, Pu¨hler A, and Top EM (2007) Genomics of IncP-1 antibiotic resistance plasmids isolated from wastewater treatment plants provides evidence for a widely accessible drug resistance gene pool. Microbiology Reviews 31: 449–477. Schmid P, Kohler M, Gujer E, Zennegg M, and Lanfranchi M (2007) Persistent organic pollutants, brominated flame retardants and synthetic musks in fish from remote alpine lakes in Switzerland. Chemosphere 67: 16--21. Schmidt CK and Brauch HJ (2008) N,N-dimethosulfamide as precursor for N-nitrosodimethylamine (NDMA) formation upon ozonation and its fate during drinking water treatment. Environmental Science and Technology 42: 6340--6346. Schro¨der HF (2002) Mass spectrometric monitoring of the degradation and elimination efficiency for hardly eliminable and hardly biodegradable polar compounds by membrane bioreactors. Water Science and Technology 46: 57--64. Schulte-Oehlmann U, Oehlmann J, and Pu¨ttman W (2007) Arzneimittelwirkstoffe in der Umwelt – Eintra¨ge, Vorkommen und der Versuch einer Bestandsaufnahme. Umweltwissenschaften und Schadstoff-Forschung 19: 168--179. Schuster A, Ha¨drich C, and Ku¨mmerer K (2008) Flows of active pharmaceutical ingredients originating from health care practices on a local, regional, and nationwide level in Germany – is hospital effluent treatment an effective approach for risk reduction? Water, Air, and Soil Pollution: Focus 8: 457--471. Schwarzenbach R, Escher BI, Fenner K, et al. (2006) The challenge of micropollutants in aquatic systems. Science 313: 1072--1077. Seifrtova´ M, Pena A, Lino CM, and Solich P (2008) Determination of fluoroquinolone antibiotics in hospital and municipal wastewaters in Coimbra by liquid chromatography with a monolithic column and fluorescence detection. Analytical and Bioanalytical Chemistry 391: 799--805. Seinen W, Lemmen JG, Pieters RHH, Verbruggen EMJ, and van der Burg B (1999) AHTN and HHCB show weak estrogenic – but no uterotrophic activity. Toxicology Letters 111: 161--168. Sharma VK (2009) Aggregation and toxicity of titanium dioxide nanoparticles in aquatic environment – a review. Journal of Environmental Science and Health A 44: 1485--1495. Silva E, Rajapakse N, and Kortenkamp A (2002) Something from ‘‘nothing’’-eight weak estrogenic chemicals combined at concentrations below NOECs produce significant mixture effects. Environmental Science and Technology 36: 1751--1756. Siminiceanu I (2007) Methyl tert-butyl ether occurrence and degradation in water. Environmental Engineering and Management Journal 6: 143--151. Simonich SL, Begley WM, Debaere G, and Eckhoff WS (2000) Trace analysis of fragrance materials in wastewater and treated wastewater. Environmental Science and Technology 34: 959--965. Simonich SL, Federle TW, Eckhoff WS, et al. (2002) Removal of fragrance materials during US and European wastewater treatment. Environmental Science and Technology 36: 2839--2847. Skutlarek D, Exner M, and Fa¨rber H (2006) Perfluorinated surfactants in surface and drinking waters. Environmental Science and Pollution Research International 13: 299--307. Steger-Hartmann T, Ku¨mmerer K, and Schecker J (1996) Trace analysis of the antineoplastics ifosfamide and cyclophosphamide in sewage water by two step solid phase extraction and GC/MS. Journal of Chromatography A 726: 179--184.

87

Sumpter JP and Jobling S (1995) Vitellogenesis as a biomarker for estrogenic contamination of the aquatic environment. Environmental Health Perspectives 103(supplement 7): 173--178. Sumpter JP, Johnson AC, Williams RJ, Kortenkamp A, and Scholze M (2006) Modeling effects of mixtures of endocrine disrupting chemicals at the river catchment scale. Environmental Science and Technology 4: 5478--5489. Sun HF, Takata A, Hata N, Kasahara I, and Taguchi S (2003) Transportation and fate of cationic surfactant in river water. Journal of Environmental Monitoring 5: 891--895. Su¨tterlin H, Alexy R, Coker A, and Ku¨mmerer K (2008) Mixtures of quaternary ammonium compounds and anionic organic compounds in the aquatic environment: Elimination and biodegradability in the closed bottle test monitored by LC-MS/MS. Chemosphere 72: 479--484. Tan NCG, van Leeuwen A, van Voorthuizen EM, et al. (2005) Fate and biodegradability of sulfonated aromatic amines. Biodegradation 16: 527--537. Ternes TA and Joss A (eds.) (2006) Human Pharmaceuticals, Hormones and Fragrances. The Challenge of Micro-Pollutants in Urban Water Management. London: IWA Publishing. Tiede K, Boxall ABA, Tear SP, Lewis J, David H, and Hassellov M (2008) Detection and characterization of engineered nanoparticles in food and the environment. Food Additives and Contaminants 25: 795–821. Tiede K, Hassellov M, Breitbarth E, Chaudhry Q, and Boxall ABA (2009) Considerations for environmental fate and ecotoxicity testing to support environmental risk assessments for engineered nanoparticles. Journal of Chromatography A 1216: 503--509. Thomas KV (2008) The relevance of different point sources. Lecture given at the ‘‘First International Conference on Sustainable Pharmacy’’. Osnabru¨ck, 24–25 April 2008. http://www.dbu.de/550artikel27307_135.html (accessed March 2010). Thuresson K (2006) Substance flow analysis of polybrominated diphenyl ethers in the City of Stockholm 2005. http://www.stockholm.se/nyagifter (accessed March 2010). Trautwein C, Metzger J, and Ku¨mmerer K (2008) Aerobic biodegradability of the calcium channel antagonist verapamil and identification of a microbial dead-end transformation product studied by LC-MS/MS. Chemosphere 72: 442--450. United Nations (2004) World contraceptive use 2003. Sales No. E. 04.XIII.2. Geneva: United Nations Publication. United Nations Office of Drug and Crime (2010) World Drug Report 2010. http:// www.unodc.org/unodc/en/data-and-analysis/WDR.html (accessed October 2010). van Nuijs ALN, Theunis L, Dubois N, Charlier C, Jorens PG, and Bervoets L (2008) Cocaine and metabolites in waste and surface water across Belgium. Environmental Pollution 157: 123--129. Verbrugh HA and de Neeling AJ (eds.) (2003) Consumption of antimicrobial agents and antimicrobial resistance among medically important bacteria in the Netherlands. SWAB NETHMAP, 2003. Verplanck PL, Taylor HE, Nordstrom DK, and Barber LB (2005) Aqueous stability of gadolinium in surface waters receiving sewage treatment plant effluent, Boulder Creek, Colorado. Environmental Science and Technology 39: 6923--6929. Vethaak D, Schrap M, and De Voogt P (eds.) (2006) Estrogens and Xenoestrogens in the Aquatic Environment: An Integrated Approach for Field Monitoring and Effect Assessment. Pensacola: Society of Environmental Toxicology and Chemistry (SETAC). Voutsa D, Hartmann P, Schaffner C, and Giger W (2006) Benzotriazoles, alkylphenols and bisphenol A in municipal wastewaters and in the Glatt River, Switzerland. Environmental Science and Pollution Research 13: 333--341. Watkinson AJ, Murby EJ, and Costanzo SD (2007) Removal of antibiotics in conventional and advanced wastewater treatment: Implications for environmental discharge and wastewater recycling. Water Research 41: 4164--4176. Watson SB, Ridal J, and Boyer GL (2008) Taste and odour and cyanobacterial toxins: Impairment, prediction, and management in the Great Lakes. Canadian Journal of Fisheries and Aquatic Sciences 65: 1779--1796. Weber WH, Seitz W, Schulz W, and Wagener HA (2007) Detection of the metabolites desphenyl-chloridazon and methyldesphenyl-chloridazon in surface water, groundwater and drinking water. Vom Wasser 105: 7--14. Wei-Hsiang C and Young TM (2008) NDMA formation during chlorination and chloramination of aqueous diuron solutions. Environmental Science and Technology 42: 1072--1077. Weiss S, Jakobs J, and Reemtsma T (2006) Discharge of three benzotriazole corrosion inhibitors with municipal wastewater and improvements by membrane bioreactor treatment and ozonation. Environmental Science and Technology 40: 7193--7199. Wenzel H, Larsen HF, Clauson-Kaas J, Høibye L, and Jacobsen BN (2008) Weighing environmental advantages and disadvantages of advanced wastewater treatment of micro-pollutants using environmental life cycle assessment. Water Science and Technology 57: 27--32.

3.05 Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems M Baalousha and JR Lead, University of Birmingham, Birmingham, UK Y Ju-Nam, University of Sheffield, Sheffield, UK & 2011 Elsevier B.V. All rights reserved.

3.05.1 3.05.2 3.05.2.1 3.05.2.2 3.05.3 3.05.3.1 3.05.3.2 3.05.4 3.05.4.1 3.05.4.1.1 3.05.4.1.2 3.05.4.2 3.05.4.2.1 3.05.4.2.2 3.05.4.2.3 3.05.4.2.4 3.05.4.3 3.05.4.3.1 3.05.4.3.2 3.05.4.4 3.05.5 3.05.5.1 3.05.5.2 3.05.5.3 3.05.5.4 3.05.5.5 3.05.5.6 3.05.6 3.05.6.1 3.05.6.2 3.05.6.3 3.05.7 3.05.7.1 3.05.7.2 3.05.7.3 3.05.7.4 3.05.8 3.05.8.1 3.05.8.2 3.05.8.3 3.05.8.4 3.05.9 3.05.9.1 3.05.9.2 3.05.9.3 3.05.9.4 3.05.9.5 3.05.9.6 References

Introduction Definitions Colloids Nanoparticles Major Types of Natural Colloids Inorganic Colloids Organic Macromolecules Major Types of Manufactured NPs Carbon-Based NPs Fullerenes Carbon nanotubes Metal Oxide NPs Iron oxide NPs Zinc oxide NPs Titania NPs Ceria NPs Metal NPs Gold and silver NPs Zero-valent iron NPs Quantum Dots Important Physico-Chemical Properties of Natural Colloid Size Shape and Morphology Surface Coating Surface Charge Pollutant Binding and Behavior Interaction Forces Intrinsic Properties of Manufactured NPs Size Shape and Morphology Surface Properties Environmental Fate and Behavior of Natural Colloids Aggregation/Disaggregation Aggregate Structure and Fractal Dimension Transport and Sedimentation in Aquatic Media Transport in Porous Media Environmental Fate and Behavior of Nanomaterials Exposure/Release of NPs Fate in Water Fate in Wastewater Fate in Soil Conclusions and Recommendations Environmental Fate and Behavior Need for New Metrology and Analysis Tools Understanding Complexity on the Nanoscale Knowledge of Uptake and Toxicity of NPs Knowledge of Structure–Activity Relationships Next Generation NPs

89 90 90 91 91 93 93 94 94 94 94 97 97 97 98 98 99 99 100 100 102 102 103 104 104 106 107 109 109 109 110 111 111 112 113 114 115 115 117 117 118 118 118 119 119 119 120 120 121

89

90

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

3.05.1 Introduction All aquatic and terrestrial environmental systems contain small particles, covering the size range from 1 nm to several micrometers. Colloids (1–1000 nm) have different properties and behavior compared to dissolved species (o1 nm) and particles (41000 nm), (see discussion in Section 3.05.2.1). The small size of colloids has two important consequences: (1) colloids have enormous surface areas and surface energies, and are able to interact strongly with contaminant species (Wilkinson and Lead, 2007) and (2) the properties of the smallest fraction of colloids (nanoparticles (NPs), 1–100 nm) may be different to those of the larger particles (4100 nm) or those of the corresponding atoms or molecules forming the colloids (Wigginton et al., 2007), based on size alone. The concentration, composition, size distribution, and other properties of environmental particles, including colloids, depend on their origin, the catchment geochemistry, rainfall, industrial waste, and other factors (Table 1). They are characterized by properties such as size, shape, surface charge, conformational properties, and interaction forces as they approach each other (see detailed discussion in Sections 3.05.5.1–3.05.5.6). Some of these particles are small enough to be transported with the water current, and others are large enough to exhibit sedimentation. The behavior of colloids and particles is dominated by processes such as aggregation/disaggregation and sedimentation (see Sections 3.05.7.1 and 3.05.7.3) in aquatic systems (Buffle et al., 1998) and attachment (see Section 3.05.7.4) to surfaces in soils and aquifers (McDowellBoyer et al., 1986). Aggregation of colloids results in the formation of large structures (see Section 3.05.7.2), often porous and fractal (Chen and Eisma, 1995), which tend to sediment in the water body or attach to surfaces in soils which results in their losses together with any adsorbed chemicals from a water body in a process called colloidal pumping (Honeyman and Santschi, 1992). Aggregation/disaggregation and the structure of the formed aggregates are highly influenced by solution physico-chemistry (e.g., pH, cation types, and concentrations) and sorption of natural organic molecules (Wilkinson et al., 1997a, 1997b), and may change during transport as physico-chemical conditions change. Such variation in structure and conformation of colloids and their aggregates may result in sorption–desorption of chemical compounds (e.g., pollutants and nutrients), and possibly a permanent retention within the structure of colloids/aggregates (Kan et al., 1994). Further, the fractal nature of colloidal

aggregates influences their sedimentation behavior (see Section 3.05.7.3). Environmental colloids have important environmental functions in aquatic and terrestrial systems. They can control pollutant (trace elements and organic contaminants) and nutrient chemistry and behavior (see Section 3.05.5.5), that is, pollutant speciation, transport, and bioavailability (Lead et al., 1999; Doucet et al., 2006). Despite decades of research, the precise role of colloids and NPs in these processes is still poorly understood in a quantitative manner and much work is required to fully elucidate their role. Importantly, the analytical and modeling capabilities are increasingly available to study such complex systems. The ongoing use of manufactured NPs (deliberately manmade materials in the size range 1–100 nm, see Section 3.05.2.2) and their consequent environmental release (either accidentally or deliberately) (Nowack and Bucheli, 2007; Luoma, 2008; Blaser et al., 2008) are likely to increase in the short term. The ability of man to manipulate materials in the nanoscale level has become the core of an exciting new research area named nanotechnology. Researchers have been strongly attracted to the idea of using NPs due to their unique and tunable optical, magnetic, and electrical properties controlled by their size, morphology, and chemistry, and therefore their potential applications in a wide range of fields such as medicine, engineering, the environment and pharmaceuticals (Klaine et al., 2008; Poole and Owens, 2003; Rotello, 2004; Schmid, 2004). Although the developments in the area of nanotechnology are recent, NPs have been fabricated in the past. Glasses with colloidal gold such as the famous Lycurgus cup and church stained glass windows are examples which have been used since ancient Greek times (Daniel and Astruc, 2004). However, the design and control of novel NPs have been the basis of many advances in this technology over the last decade. Studies of manufactured NP fate and impact in the environment are becoming important due to the discharges already occurring to the environment, the likely increase in discharges as the industry grows dramatically, the known toxicity of NPs and the immense gaps in our knowledge leading to difficulties in risk assessment and management (Handy et al., 2008a). Despite some recently acquired knowledge on the effects of NPs on human toxicology and to a lesser extent in ecotoxicology, very little is known about mechanisms of biological uptake and toxicity modes of action, about transport in and between environmental and biological compartments and their chemical behavior in the

Table 1 Major types of environmental particles, their concentrations and size distribution in different environmental compartments (Buffle and Leeuwen, 1993) Groundwater

River

Marine

Atmosphere

Major types

Alumino-silicates, metal oxyhydroxides, natural organic matter

Alumino-silicates, natural organic matter, biogenic colloids

Carbonates, aluminosilicates, biogenic colloids

Salt, silicates, ash, pollen, soot

Concentration

0.001–1 mg l1

1–1000 mg l1

0.01–0.05 mg l1

0.1–1000 ng m3

Size

o10 mm

o300 mm

o100 mm

o30 mm

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

environment. Exposure studies are rare (Mueller and Nowack, 2008; Valsami-Jones et al., 2008), but increasing rapidly (Handy et al., 2008a; Klaine et al., 2008). Recent concern has been focused on the asbestos-like behavior of carbon nanotubes (CNTs) (Poland et al., 2008) and the environmental toxicity and wide exposure of nanosilver. Next generation and active NPs such as drug delivery devices used in nanomedicine will be of concern in the next few years. An important consideration to environmental studies is that NPs are not one class of potential pollutant. As there are many trace metals and many trace organic pollutants, NPs contain a wide range of different materials with different physical, chemical, and toxicological properties. NPs should not be considered as a single homogeneous group. If we take into account different surface chemistries (including from capping agents, surfactants, or co-solvents), sizes, and other properties, then it is easy to see that the range of possible nanomaterials is vast and they have many different properties which will impact on their environmental behavior substantially (Klaine, 2009). NPs can be classified as inorganic and carbon based (see Section 3.05.3). Inorganic NPs can be further divided into metal oxides, metals, and quantum dots (QDs), while carbon-based NPs can be divided into fullerenes and CNTs. Clearly, other factors such as size and surface chemistry will change the properties of materials within this class substantially (see Section 3.05.6.3). For instance, 2 nm polyvinylpyrrolidone (PVP) stabilized gold NPs will have different surface reactivity and aggregation behavior than 50 nm citrate stabilized gold. Clearly, the role of natural colloids and the environmental behavior of NPs are interlinked (see Sections 3.05.7 and 3.05.8). First, the technological developments, which have underpinned the development of nanoscience and nanotechnology, including electron and force microscopy, are fundamental to understanding the environmental structure and interactions of colloids and NPs, and NP fate. Second, there is a long history of research into the role aquatic and terrestrial colloids play in environmental processes, transport, and biouptake (Lead and Wilkinson, 2006). With the caveat that this literature is not directly transferable, it is a potentially valuable tool to help understand the environmental behavior and impacts of manufactured NPs. Third, there is a direct interaction in the environment between NPs and colloids (Hyung et al., 2007) with consequent modification of NP fate and behavior.

3.05.2 Definitions Both colloids and NPs are solid materials in the nanoscale range and are defined more specifically below. The nanometer is a metric unit of length, and represents one-billionth of a meter or 109 m. Many existing materials (natural or manufactured) are structures on the nanometer scale (nanomaterials). Figure 1 shows the nanoscale dimension in comparison to the known dimensional scale of the universe (Hochella, 2002). At the smallest end of the scale (Figure 1(a)), fundamental particles, such as electrons and quarks, are smaller than 1018 m. At the larger end of the scale are the size of the Earth (107 m in diameter) and the Sun (109 m in diameter).

91

The nanoscale with other related objects is described in Figure 1(b) and is in the range 1–100 nm. A nanometer is a billionth of a meter (i.e., 109 m). The size of a single atom is of the order of several angstroms (0.1 nm). The size of a bacterium is about 1 mm (1000 nm), approximately the limit of visibility in light microscopes. The size of a virus is approximately equal to 100 nm (the upper size range of NPs), which cannot be detected through standard light microscopes, because they are smaller than wavelengths of light (approximately 400–700 nm). They can be observed only with higher-resolution microscopes such as scanning electron microscope (SEM, resolution of the order of 10 nm) and transmission electron microscope (TEM and atomic force microscope (AFM), both have a resolution which can be down to 0.1 nm).

3.05.2.1 Colloids The formal definitions of colloids and NPs are given by international organizations such as the International Union of Pure and Applied Chemistry (IUPAC), the International Organization for Standardization (ISO), and the Organization for Economic and Co-operation and Development (OECD). Colloid dispersions can be regarded as a particular state of matter between true solutions and suspensions. In such a colloidal phase, one substance is dispersed in another phase; in principle, the dispersed phase and the phase in which it is dispersed can be solid, liquid, or gas. However, for this chapter and with some simplifications, colloidal phases can be thought of as solids dispersed in aqueous phases, generally shortened to colloids. Colloids are generally in the submicrometer size range. Suspensions may then be defined as a heterogeneous mixture containing particles large enough (usually 41 mm) to sediment. According to the IUPAC definition, natural aquatic colloids can be defined as materials with at least one dimension between 1 nm and 1 mm (see Figure 2(a)), while particles are larger than 1 mm (Hofmann et al., 2003; Lead and Wilkinson, 2007). Alternatively, colloids can be defined as organic or inorganic entities small enough to be dominated by aggregation and to remain in the water column due to Brownian motion (diffusion) over reasonable timescales (4hours to days), but large enough to have supramolecular structure and properties, for example, electrical surface charge and possibility of conformational changes (Lead and Wilkinson, 2007). Particles are large enough (41 mm) to be dominated by sedimentation, rather than aggregation (Buffle and Leppard, 1995a) and will be removed from the water column rapidly. This definition was developed and somewhat extended by Gustafsson and Gschwend (1997) where aquatic colloids can be defined as any constituent that: (1) provides a molecular milieu into and onto which chemicals can partition which has different properties to the aqueous phase (e.g., dielectric constant) and (2) its vertical movement (in water) is not significantly affected by gravitational settling over a reasonable timescale. In practice and in much of the literature, colloids are defined as materials that permeate a filter (pore size between 0.1 and 1.0 mm) while being retained by an ultrafilter (1–100 kDa, nominal pore size). The upper size limit is often chosen to reduce sample complexity and

92

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

Electrons and quarks Scale of the Earth sciences

10–20 10–16 10–12 10–8

10–4

100

104

108

1012

1016

1020 1024 1028 m

(a)

Nanoscale m 10–10

10–8

10–6

10–4

10–2

100

102

104

106

108

1010

(b)

Figure 1 (a) The known dimensional scale of the universe. On the small end, fundamental particles such as electrons and quarks are smaller than 1018 m, and may approach 1030 m in size or smaller, but such dimensions are not physically measurable at least at this time. Other stops depicted along this dimensional journey include: the scale of the solid Earth sciences, from atoms to the Earth (1010–107 m); the Sun (109 m in diameter) as seen from the Extreme UV Imaging Telescope on the SOHO satellite; expanding gas rings (1016 m in diameter) from supernova SN1987a as observed by the Hubble Space Telescope; infrared image of the inner portion of our own galaxy (the Milky Way is nearly 1021 m in diameter); and distant galaxies (the most distant are 1026 m away). (b) The dimensional scale of the Earth sciences. Stops depicted along this dimensional journey include: scanning tunneling microscope image of lead and sulfur atoms on a galena surface (atomic size 1010 m); crystallization nucleus of calcite (109–108 m); bacterial cells (106 m in length); a single crystal of quartz (102 m); a typical open pit mine (the Carlin Mine in Nevada, USA, 102–103 m); Mt. Fuji, Japan, a composite volcano (104 m); the Red Sea from space (105 m wide and 106 m long); Earth (107 m); the Earth–Moon system as seen from Apollo 11 (4  108 m). From Hochella MF (2002) There’s plenty of room at the bottom: Nanoscience in geochemistry. Geochimica et Cosmochimica Acta 66(5): 735–743.

partially sterilize the colloidal fraction. In addition, other filter cut-offs have been used to identify different colloidal fractions (e.g., ultrafine, fine, and coarse) (Guo and Santschi, 2007). It is clear that the formal, mechanistic, and practical definitions do not entirely mesh and some work on standardization is required. Within this colloidal fraction, it is becoming useful to define a nanoscale fraction of natural particles (Lead and

Wilkinson, 2006; Wigginton et al., 2007), which may be thought of as between 1 and 100 nm, as with manufactured NPs (see Section 3.05.2.2). However, the Bo10–25 nm may be the size range in which environmental properties such as metal binding, zeta potential, and redox properties change radically compared with the bulk or larger-sized phases (Lyven et al., 2003; Madden and Hochella, 2005; Madden et al., 2006).

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

Colloids or macromolecules

Organic compounds

Solutes –10

–9

–8

1A

1 nm

93

Particles Log (size / m)

–7

–6

–5

1 μm

0.45 μm

Cellular debris

Amino acids

Polysaccharides Peptides

Peptidoglycans Proteins

Fulvics

Humic aggregates Viruses

Humics

Bacteria

Algae

(a)

Inorganic compounds

Organic compounds absorbed ‘Simple’ hydrated ions (e.g., OH–, Cl, SO42–, HS– Na+, Ca2+, etc.)

on inorganic particles Clays (aluminosilicates) Fe oxyhydroxides Mn oxides Metal sulfides Carbonates, phosphates Amorphous SiO2 Filtration

Analytical techniques

Ultrafiltration

(b)

Electron microscopy Atomic force microscopy Flow-FFF

Confocal microscopy Optical microscopy

Sedimentation-FFF

FCS Light scattering X-Ray, neutron scattering

X-Ray, absorption

LIBD

(c)

Figure 2 (a) Typical example of natural colloids and aggregates (Rhine River), scale bar corresponds to 1 mm and (b) natural heteroaggregate of colloids and particles from Lake Bret, Switzerland, as shown by transmission electron microscopy, scale bar corresponds to 250 nm. From Buffle J, Wilkinson KJ, Stoll S, Filella M, and Zhang J (1998) A generalized description of aquatic colloidal interactions: The three-colloidal component approach. Environmental Science and Technology 32(19): 2887–2899. (c) Schematic representation, by size distributions, of the major environmental colloidal and particulate components. From Lead JR and Wilkinson KJ (2006) Aquatic colloids and nanoparticles: Current knowledge and future trends. Environmental Chemistry 3: 159–171.

3.05.2.2 Nanoparticles In the literature and websites related to the nanotechnology, numerous definitions of NPs can be found and many standard bodies such as ISO, SCENIHR, OECD, the US National Nanotechnology Initiative (NNI), British Standard Institution (BSI), and the American Society for Testing Materials (ASTM) are investigating definitions for nanoscience, nanotechnology, and NPs (Klaine et al., 2008). Nanoscience is generally defined as the scientific study of materials on the nanoscale (Borm et al., 2006). Nanotechnology, as defined by US-NNI, is ‘‘the research and technology development at the atomic, molecular or macromolecular levels, in the length scale approximately 1–100 nm; the creation, and use of structures, devices and systems that have novel properties and functions because

of their small size; and ability to be controlled or manipulated on the atomic scale’’ (NNI, 2004). The Royal Society and the Royal Academy of Engineering defines nanotechnology as ‘‘the design, characterization, production and application of structures, devices and systems by controlling shape and size at the nanometre scale’’ (Royal Society and Royal Academy, 2004). Nanomaterials are the major component of nanotechnology and can be defined as materials that have one or more dimensions in the range of 1–100 nm (Lead and Wilkinson, 2006). A recent attempt to develop a more structured approach has been published by the BSI (BSI, 2007). In their terminology for nanomaterials, they define nanoscale as ‘‘size range from approximately 1 nm to 100 nm’’, a nanoobject as ‘‘discrete piece of material with one or more external dimensions in the nanoscale,’’ and an NP as a ‘‘nano-object

94

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

with all three external dimensions in the nanoscale.’’ A nanomaterial is a ‘‘material having one or more external dimensions in the nanoscale or which is nanostructured,’’ with nanostructured being defined as ‘‘possessing a structure comprising continuous elements with one or more dimension in the nanoscale.’’ These definitions are based on particle size and imply that there is a size range (1–100 nm) between that of molecules and bulk materials, where particles have unique properties different than those of molecules or bulk material (Tratnyek and Johnson, 2006). This overlaps with a definition based on novel properties, that is, an NP has novel properties based on size alone. Some of these properties arise only for particles smaller than approximately 10 nm or so, where particle size approaches the length scale of certain molecular properties (Klabunde et al., 1996). For instance, below 10 nm, particlespecific surface area increases exponentially and similar trends apply to related properties such as the ratio of surface/bulk atoms. Another reason for changes is the issue of spatial confinement such as quantum confinement, which arises because the band gap of semiconducting materials increases as particle size decreases (Klabunde et al., 1996). For instance, the decrease in hematite particle size (from 37 to 7.3 nm) greatly promotes the oxidation of aqueous Mn(II) in the presence of molecular oxygen (Madden and Hochella, 2005), quite separate from the effect of specific surface area. Small magnetite NPs (9 nm) exhibit greater reactivity toward carbon tetrachloride (CCl4) relative to larger NPs (80 nm), both on mass and surface area normalized bases (Vikesland et al., 2007). The decrease in size of ceria NP alters the oxidation state of the NPs with an increase in the fraction of Ce3þ at sizes Bo15 nm with complete reduction of ceria particles to Ce3þ at sizes Bo3 nm (Wu et al., 2004). Size-dependent inhibition of nitrifying bacteria has been observed and the inhibition was correlated to the fraction Bo5 nm in the suspension (Choi and Hu, 2008).

3.05.3 Major Types of Natural Colloids Aquatic and terrestrial colloids cover a wide range of materials, including organic, inorganic, or biota (Table 1) with proportions dependent on the nature of the inputs, outputs, and within-media processes (Bertsch and Seaman, 1999; Zimmermann-Timm, 2002). They are highly heterogeneous in size, shape, structure, chemical composition, and other properties (Figures 2(a) and 2(b)). Figures 2(a) and 2(b) show typical examples of surface freshwater colloidal material of various sizes, as observed by transmission electron microscopy. Figure 2(c) summarizes the different types of environmental colloids together with the size range they cover and the analytical tools that can be used to characterize them. Environmental colloids have been simplified and modeled in terms of three major colloidal components, namely (1) inorganic colloids, (2) humic substances (HSs), and (3) biopolymers (Buffle et al., 1998). HSs and biopolymers are both organic colloids and therefore are presented together. Their formation, composition, and properties have been described in detail in Filella (2006) and Baalousha et al. (2009), and are summarized in Tables 2 and 3.

3.05.3.1 Inorganic Colloids There are two main types of inorganic colloidal particles in oxygenated terrestrial and aquatic environments, which are aluminum phyllosilicates (e.g., clay, mica, and chlorite) and oxides and hydrous oxides of iron (e.g., hematite and magnetite), manganese (e.g., pyrolusite), and silicon (e.g., silicates) (see Table 2 for a summary of their characteristics). Other inorganic colloids can also be found, but they are usually minor components (e.g., other groups of silicates), or are primarily present in anoxic waters (e.g., FeS, FeS2, and MnS). Calcium carbonate can be found in significant amount in freshwaters but is more likely in a particulate form (Sigg, 1994).

3.05.3.2 Organic Macromolecules Natural organic matter (NOM) in the aquatic or terrestrial environment can be divided into two classes of compounds: (1) HSs including humic and fulvic acids and (2) nonhumic materials (e.g., proteins, polysaccharides, nucleic acids, and small molecules such as sugars and amino acids). HSs and extracellular polymeric substances (EPSs, e.g., polysaccharides and proteins, usually present as fibrillar material) represent the major constituents of NOM and play a significant role in determining the fate and behavior of colloids and particles (Buffle et al., 1998). Other compounds are limited by their rapid turnover or low production rates and hence low concentrations in different environmental systems (Fabiano and Pusceddu, 1998; Mannino and Harvey, 2000). The formation, properties, and characteristics of HSs and EPS are summarized in Table 3. Although this separation of compounds into relatively simple classes of colloids is widespread and of significant value, inferring behavior based on extracted phases or laboratory synthesized surrogates should be performed with caution for several reasons. First, the surface nature of natural colloids is different when compared with such pure phases or homologous series (Schulthess and Sparks, 1989; Schulthess and Huang, 1990) containing complex mixtures of organic and other material. Second, the use of particles larger than colloids in many of these studies (Aldahan et al., 1999; Arnold et al., 2001) means they may be unrepresentative. Finally, extensive particle pretreatment, such as drying, grinding, and saturation with sodium ions, often leads to selective removal of or change to phases and coatings (Mukhopadhyay and Walther, 2001).

3.05.4 Major Types of Manufactured NPs NPs be classified according to their core composition, morphology, surface chemistry, or aggregation state (Buzea et al., 2007; Ju-Nam and Lead, 2008; Rotello, 2004). NP classification based on core composition is the one used in this chapter. Among these categories, metal, metal oxide, QDs, and carbon-based NPs are the most relevant. They can be composed by a single constituent material or can be a composite of two or more materials.

Permanent and variable, that is, pH-dependent

Covers colloids and particulate range

Irregular

Mobilization of particles in soil and subsurface water, or sediment resuspension at river bed

Are the most abundant inorganic colloids in aquatic and terrestrial systems Crystalline with different crystalline structures

Surface charge

Size range

Shape

Formation

Abundance

Structure

Examples

Mainly Al and Si and less concentrations of Na, K,Ca, and Mg Contains trace concentrations of Fe and Ti Clays, mica, chlorite

Aluminum phyllosilicates

Inorganic colloids

Composition

Table 2

B0.05–0.5 mm

B0.05–0.5 mm

Exist under different crystalline and/or amorphous forms such as hematite (a-Fe2O3), goethite (a-FeOOH), lepidocrocite (g-FeOOH), maghemite (g-Fe2O3), magnetite (Fe3O4), and ferrihydrite (amorphous Fe(III) phase)

Abundant in aquatic and terrestrial systems

Particulate Fe(III) oxyhydroxides are formed by oxidation of Fe(II) and hydrolysis by subsurface aeration at oxic/ anoxic boundaries in groundwater, freshwater lakes, and coastal marine water. Biogenic processes

Spherical or rod like

Variable, largely neagtive

MnS in anoxic water Variable, largely negative

Amorphous or crystalline

Abundant in aquatic and terrestrial systems

Silica colloids can be released during the diagenesis of amorphous silica

Bacteria-mediated oxidation of Mn(II) to Mn(III, IV)

Biogenic silica formed after death of some plankton (diatoms) In all natural waters, in particular at diatom blooms Crystalline

Irregular

Irregular

Covers colloids and particulate range

Quartz or its polymorphs

Si and O

Silica

MnO2 in oxic water

Mn and O

Oxides and hydrous oxides of manganese

Hematite (a-Fe2O3), goethite (a-FeOOH), lepidocrocite (g-FeOOH), maghemite (g-Fe2O3), magnetite (Fe3O4), and ferrihydrite (amorphous Fe(III) phase) FeS, FeS2 in anoxic water Variable, pzc at near neutral pH

Fe(III) stable in the presence of oxygen at neutral pH then hydrolysis to form oxides

Oxides and hydrous oxides of iron

Baalousha et al. (2009), Davison and De Vitre (1992)

Baalousha et al. (2009), Buffle et al. (1998), Filella (2007)

Chapnick et al. (1982), Helgeson et al. (1984), Murphy and Helgeson (1987), Nelson et al. (1995), Sunda and Huntsman (1987), Wolthoorn et al. (2004a, 2004b)

Buffle et al. (1998)

Lienemann et al. (1999), Tipping et al. (1981)

Buffle et al. (1998), Buffle (2006), Buffle and Leppard (1995a), Buffle and Leppard (1995b)

Berner and Holdren (1977), Davison and De Vitre (1992), Filella (2007), Helgeson et al. (1984), Murphy and Helgeson (1987)

Berner and Holdren (1977), Sigg (1994), Wolthoorn et al. (2004a, 2004b)

References

96

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

Table 3

Organic colloids

Composition

Types/examples

Surface charge

Humic substances

Extracellular polysaccharides

References

C, H, O, N, and minor component S Fresh water humic substances have a high C/N ratio (40–50) Marine humic substances have a low C/N ratio (15–20) C/N ratio of fresh microbial material is between 5 and 10 to 1 Due to different sources Fulvic acids: soluble in water under all pHs Humic acids: soluble at pH 4 2 Humins: insoluble at all pHs pH dependent charge 10 meq g1 over pH range

C, O, N, and occasionally S

Hedges et al. (1997), Thurman (1985)

Organism-specific EPS examples areyalginic acid etc.

Thurman and Malcolm (1981)

pH-dependent charge 0 to  0.8 mequiv g1, alginates represent an exception and might have a maximum surface charge of  6 mequiv g1 Few nanometers thick with a length which can be greater than 1 mm

Buffle (1988), Filella (2007)

Size range

Primarly particles 1–2 nm, globules B10–20 nm and aggregates of several micrometers, surface coatings

Formation

Degradation of higher plants in terrestrial and freshwater environments. Degradation of plankton in marine systems

Abundance

Humic acids are generally terrestrial, while aquatic HS are dominated by FA. Make up most of organic fraction

Structure

Different models depending on the technique used, including macromolecules, supramolecular association of small molecules, micelles, and semipermeable spheres

Release from phytoplankton cells and bacteria during all stages of growth, in particular during phytoplankton blooms and may comprise 80–90% of the total extracellular release EPS represent the most abundant organic compounds in the biosphere and constitute the largest fraction of cells (concentrate on noncellular organics, minor fraction B35% at most in eutrophic waters) They may represent a significant proportion of NOM in freshwater, varying seasonally B5–30% in surface waters of lakes and likely accounts for higher proportion (up to 80%) of NOM in marine systems Polysaccharides can be rigid due to the large quantity of strongly bound hydration water (up to 80%), their association into double or triple helices that may be stabilized by hydrogen or calcium bridges or helices aggregation. Flexible conformations also possible adopt variable conformation as a function of pH and ionic strength

3.05.4.1 Carbon-Based NPs 3.05.4.1.1 Fullerenes Fullerenes are molecules with 60 atoms of carbon, commonly denoted as C60 or, less commonly, with a larger number of carbon atoms, for example, C70, C76, C78, and C80 (Kikuchi

Baalousha et al. (2005), Buffle et al. (1998), Leppard et al. (1990), Santschi (1998), Thurman et al. (1982) Leenheer and Croue´ (2003), Myklestad (1995), Strycek et al. (1992), Thurman and Malcolm (1981)

Aluwihare et al. (1997), Leenheer and Croue´ (2003), Myklestad (1995), Santschi (1998), Thurman and Malcolm (1981), Wilkinson et al. (1997)

Duval et al. (2005), Leppard et al. (1990), Leppard and Arsenault (2003), Leppard (1997), Morris et al. (1980), Piccolo (2001), Rees (1981), Swift (1989), Wershaw (1999)

et al., 1992). However, the most widely studied is the C60 molecule. C60 possesses a spherical molecular structure where the carbon atoms are positioned at the vertices of a regular truncated icosahedron structure (Kroto et al., 1985). There are also higher mass fullerenes with different geometric structures.

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

A large number of applications of fullerenes in optics (Singh and Roy, 2004), electronics (Diederich, 2005; Otsubo et al., 2005), and biomedicine (Yang et al., 2007) make this molecule important and promising in the nanotechnology field; the most popular technique for nC60 production reported in the literature is based on a physical method that involves the use of an arc discharge between graphite electrodes in 200 torr of He gas (Kratschmer et al., 1990). However, the direct use of carbon-based fullerenes in high-impact applications, for instance, in biological ones, is restricted by their poor solubility in aqueous media (Kadish and Ruoff, 2000) where they aggregate and are often termed nC60. C60 is soluble in a small range of organic solvents but the insolubility in water limits utilization in biological applications. In order to overcome this problem, two different strategies to increase their solubility have been reported in the literature: (1) noncovalent encapsulation of fullerene molecules into soluble polymeric or host molecules (Atwood et al., 1994; Yamakoshi et al., 1994) and (2) covalent functionalization of fullerenes by introduction of hydrophilic groups by chemical modification (Chiang et al., 1996). The second strategy has attracted more interest as these novel building blocks can be used for further molecular constructions. Yang and co-workers have reported the syntheses and characterization of a series of fullerene-derivatized amino acids, [60]fullerene-substituted phenylalanine, and lysine derivatives, suggesting that the incorporation of fullerene-based amino acids into proteins, peptides, or antibodies could lead to new applications in medicinal chemistry (Yang et al., 2007). In water, unmodified fullerenes are largely insoluble forming large aggregates of tens to hundreds of nanometers (or larger). Dispersion in water as NPs has been achieved in two ways: solvent exchange (Deguchi et al., 2001) and ultrasonication and stirring in water (Brant et al., 2005; Deguchi et al., 2006) in order to promote the formation of smaller clusters. The mechanism of this process is not currently clear. However, it has been assumed that the self-assembly of the fullerene molecules occurs during the formation of the NPs, and aggregation of C60 has been observed in different solvents. In addition, measured zeta potentials of these fullerenes show a negative charge, although differing mechanisms have been suggested about how a negative potential is acquired (Brant et al., 2005). Interesting differences in nC60 between the cosolvent and water-stirred varieties have been observed, generally including greater toxicities in bacteria, fish, and invertebrates (Kashiwada, 2006; Lyon et al., 2006, 2005) when dissolved with co-solvent. Again, the mechanism needs to be clarified, although the direct impact of the co-solvent or its contaminants on toxicity has been suggested (Henry et al., 2007; Oberdo¨rster, 2004). In addition, in most studies physico-chemical characteristics have not been measured fully and until this is performed alongside toxicity studies, our lack of understanding of C60 toxicity will persist.

3.05.4.1.2 Carbon nanotubes CNTs have been used in multiple applications in the development of novel materials, due to their structural robustness and synthetic versatility. Ijima and co-workers, pioneers in the synthesis of CNTs in 1991 (Iijima, 1991), reported the

97

formation of these nanotubes by using a carbon cathode and the arc discharge technique. However, the discovery of the structure of these nanomaterials took place a few years after their first synthesis (Bethune et al., 1993; Iijima and Ichihashi, 1993). There are two main forms of manufactured CNTs, the single-walled CNT (SWCNT) and multi-walled CNT (MWCNT). In terms of structure, the SWCNT is a single-layer graphene sheet rolled up as cylindrical shapes, with a diameter of approximately 1 nm and a length of several micrometers, whereas the MWCNT contains two or more concentric layers with various lengths and diameters (Gao, 2004) and larger diameters. CNTs are generated by arc evaporation (Ebbesen, 1994), laser ablation (Paradise and Goswami, 2007), and pyrolysis (Endo et al., 1993). Ebbensen and co-workers demonstrated that CNTs can be generated in bulk amounts by varying arcevaporation conditions (Ebbesen and Ajayan, 1992). Inherent properties of CNTs have been widely investigated and SWCNTs possess important mechanical, thermal, photochemical, and electrical properties (Arepalli et al., 2001) which are industrially useful. These nanomaterials are robust and stiff but flexible, and they have been reported as the strongest of all the synthetic fibers (Arepalli et al., 2001), although practical experience suggests that batch to batch (and even within batch) variability is significant. Materials containing CNTs have been suggested as being strong enough to build spacecrafts, space elevators, artificial muscles, and land and sea vehicles (Kumar, 2006), although such claims are as yet unvalidated. SWCNTs can conduct twice the electricity of copper, making these materials excellent electrical conductors, and may also be used to improve rechargeable batteries and fuel cell production, for instance (Kumar, 2006). They also have a distinctive electron-transport property, and commonly in a manufactured material bulk sample B30% of the SWCNTs are conductors and 70% are semiconductors (Watts et al., 2002). In general, CNT products contain substantial amounts of metal impurities. Clearly, the presence of such metal impurities might lead to a variety of adverse biological endpoints. The similarity in structure with asbestos suggests that CNTs may behave as these fibers do when considering their toxicity. Recent evidence (Poland et al., 2008), where a variety of CNT morphologies were examined in animal studies, has suggested long, high aspect ratio, and biopersistent CNTs cause pathologies similar to asbestosis. A second-order effect appeared to be due to metal contamination, although insufficient metal data were provided to draw firm conclusions. CNTs appeared to be more hazardous than asbestos, but in these studies producing CNT aerosols appeared to be difficult, so their risk to human health may be low. Their hydrophobicity suggests that lipids and organic materials may be an ultimate sink, for example, sediments. Functionalized CNTs which are likely to be more hydrophilic may behave differently. In aquatic systems, Smith and co-workers obtained evidence of oxidative injury in rainbow trout (Oncorhynchus mykiss) exposed to SWCNTs (Smith et al., 2007). They concluded that CNTs acted as respiratory toxicants in rainbow trout and also caused brain injury, reminiscent of a stroke. A significant concern of investigations of CNT toxicity is inadequate assessment of purity, for example, the presence of metal catalysts used in

98

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

production, and their physico-chemical form (Handy et al., 2008a, 2008b; Kagan et al., 2006; Pulskamp et al., 2007). For instance, toxicity in zebrafish embryos exposed to SWCNTs has been attributed to the presence of Co and Ni catalysts used in the preparation of the SWCNTs (Cheng et al., 2007), although this remains to be confirmed.

3.05.4.2 Metal Oxide NPs This class of NPs is extensively used in a considerable number of applications in food, material, chemical, and biological sciences (Aitken et al., 2006). It is well known that bulk materials based on TiO2, SiO2, and aluminum and iron oxides have been massively produced for many years. More recently, nanoparticulate versions of these metal oxides and others have been manufactured and introduced in commercial products such as cosmetics and sunscreens (TiO2, Fe2O3, and ZnO) (Nowack and Bucheli, 2007), fillers in dental fillings (SiO2) (Balamurugan et al., 2006), in catalysis (TiO2) (Aitken et al., 2006), and as diesel additives (CeO2) (Laosiripojana et al., 2005). Several of the most commercially important NPs are discussed below.

3.05.4.2.1 Iron oxide NPs Iron oxide NPs have been extensively used for biological and medical applications such as magnetic resonance imaging (MRI) and manufacturing pigments (Cornell and Schwertmann, 1996). NPs in the form of Fe3O4 and Fe2O3 have been synthesized with a number of methods involving different compositions and phases (Neveu et al., 2002). During the last few years, many publications describing efficient synthetic methods to obtain shape-controlled, highly stable, and monodisperse magnetic NPs have been produced. Co-precipitation (Willis et al., 2005), thermal decomposition (Park et al., 2004), and hydrothermal synthesis (Wang et al., 2005) techniques are among the most used methods and are also easily scalable with high synthetic yields. Steric and electrostatic repulsion are the interactions involved in the stability of the colloidal iron oxide NPs manufactured by the previously mentioned techniques. Their stability depends on the stabilizers, such as fatty acids or amines, and the polarity of the solvent used. In the case of the synthesis of the iron oxide NPs in an aqueous medium through the coprecipitation method, NPs are stabilized by repulsive electrostatic forces due to positive charge on particles (Lu et al., 2007), dependent on pH and ionic strength (Baalousha et al., 2008), whereas NPs obtained by the thermal decomposition technique are often sterically stabilized in an organic solvent by fatty acids or surfactant (Sun et al., 2004). Despite their high usage, it is unlikely that these materials present significant environmental problems. In fact, iron oxide NPs could be beneficial to some extent. The addition of NPs to oceanic waters, where primary productivity is limited by low Fe concentrations, may increase oceanic productivity and drawdown sufficient CO2 to some extent mitigating climate change (Raiswell et al., 2006).

3.05.4.2.2 Zinc oxide NPs Zinc oxide (ZnO) is a direct band-gap semiconductor with band-gap energy of 3.36 eV at room temperature, high exciton

binding energy of 60 meV, and high dielectric constant (Singh et al., 2007). Therefore, the luminescent properties of ZnO have attracted considerable attention due to its potential application in ultraviolet (UV) light-emitting devices. This band-gap semiconductor has numerous potential applications, particularly in the form of thin films, nanowires, nanorods, or NPs (Starowicz and Stypua, 2008), and can be introduced to optoelectronic and electronic devices. They can also be used in the production of chemical sensors and solar cells (Singh et al., 2007). One of the most widely used commercial applications is their use in the production of sunscreens and cosmetics, due to their property of blocking broad UV-A and UV-B rays (Huang et al., 2008) but being transparent. This is thus a potentially important diffuse source of NP contamination, due to wash-off from individuals into the environment. ZnO NPs are believed to be a nontoxic, biosafe, and biocompatible nanomaterial (Zhou et al., 2006), although a few reports have shown some toxicological activity of ZnO NPs, for example, in algae (Adams et al., 2006) and in vitro studies of human lung cells. It has been shown that they cause membrane damage in the Escherichia coli, possibly due to oxidative stress mechanisms (Zhang et al., 2007). More recently, Huang et al. (2008) have investigated the possible interactions that govern the bactericidal activity of 60–100 nm polyvinyl alcohol (PVA)–ZnO NPs against Streptococcus agalactiae and Staphylococcus aureus. Results showed that low concentrations of ZnO NPs did not induce any cellular damage, as previously demonstrated by other research groups (Feldmann, 2003). However, they observed cellular damage when the PVA-coated ZnO NP concentrations are higher than 0.016 M in the ethylene glycol (EG) medium containing the cells. After contact with the cells, there was a significant change in the ZnO NP crystal structure (Huang et al., 2008). Heinlaan and co-workers investigated the toxicity of ZnO NPs to bacteria Vibrio fischeri, and crustaceans Daphnia magna and Thamnocephalus platyurus, and discerned the toxic effects of metal oxides and soluble metal ions (Heinlaan et al., 2008). They showed that the NPs did not necessarily have to enter the cells to cause damage in the cell membrane. In fact, the contact between the particle and the bacterial cell wall (or other cell wall/membrane) may cause changes in the vicinity of organism–particle contact area and may also increase the solubilization of metals (Heinlaan et al., 2008). It has also been demonstrated that the release of metal ions from the ZnO NP, that is, from the NP dissolution, was responsible for toxicity in lung cell lines (Brunner et al., 2006), while under realistic environmental conditions, a similar mechanism was reported (Franklin et al., 2007). As a general point, toxicity due to NP dissolution needs to be considered for all inorganic NPs and this appears to be particularly so for the highly soluble ZnO.

3.05.4.2.3 Titania NPs Titanium oxide (TiO2) exists in three main crystallographic structures, for example, anatase, rutile, and brookite (Arami et al., 2007) of which the first two are usually considered the most important in the environment. Each of these forms presents different properties and therefore different applications and environmental impacts. The phase-specific

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

environmental effects have not been much studied due to the difficulties of producing phase-pure NPs (Valsami-Jones et al., 2008). It is also well known that it is easier to obtain TiO2 NPs with good crystallinity and high specific surface area from the anatase crystalline form. This evidence can be attributed to the thermodynamic stability which is size dependent. Indeed, anatase is more stable than rutile at particle diameters below 14 nm (Zhang and Banfield, 2000). Apart from the applications of TiO2 as a catalyst support (Djenadic et al., 2007), semiconductor photocatalyst (Gra¨tzel, 2001), and sensors (Ruiz et al., 2003), there is also a great interest in the development of synthetic methods to obtain TiO2 NPs with high morphological specificity such as nanofibers, nanowires, nanorods, and nanotubes (Wu et al., 2006) due to their potential applications in solar energy conversion, photocatalysis, photovoltaic devices, and carrier for metallic NPs (Zhu et al., 2005). TiO2 NPs are, like zinc oxide NPs, high-band-gap semiconductors due to thier large energy gap (Eg ¼ 3.2 eV). This nanostructured material requires the use of near UV light in order to be photoactivated (Bellardita et al., 2007; Reijnders, 2008). Consequently, photocatalysis using TiO2 NPs has recently become very important. The use of TiO2 NPs has improved the photodegradation process and the complete mineralization of toxic organic pollutants. In fact, TiO2 NPs have been successfully used in environmental technology for the treatment of wastewater and groundwater, the removal of benzothiophene from diesel fuel, and the degradation of air pollutants, specifically nitrogen oxide, sulfur oxides, and volatile organic compounds (Toma et al., 2006; Yu et al., 2006). Although TiO2 NPs possess large specific surface area, commercial applications have not been developed rapidly due to their tendency to aggregate and coalesce very easily forming larger particles. This size increase has an undesirable effect on the catalyst efficiency and also the difficulties in the separation and recovery of TiO2 particles from the reactant mixture (Yu et al., 2002). Nevertheless, TiO2 is used widely, as with ZnO, in sunscreens, because of its ability to absorb UV radiation. However, its photoactivity, via reactive oxygen species (ROS) production, might result in harm to skin tissue, although certain crystal phases (rutile) are less photoactive than others (anatase) (Sayes et al., 2006), while doping of titania with Mn has resulted in reduced photoactivity (Kim et al., 2007). Microorganisms in the presence of light (Oberdo¨rster et al., 2007) are adversely affected by TiO2 NPs due to the production of ROS (Hirano et al., 2005). This experimental evidence suggests that these NPs can produce oxidative stress in aquatic organisms. This has been confirmed in rainbow trout, where inflammatory injury and respiratory distress were observed after the exposure to TiO2 NPs (Federici et al., 2007; Reijnders, 2008). TiO2 particles of different sizes (10, 30, and 300 nm, as described by the manufacturer) were shown to inhibit algal growth, but the physiological mode of action is not yet understood (Hartmann et al., 2009). Authors suggested that mechanism for this toxic effect is other than the generation of ROS, and possible mechanisms include adhesion of TiO2 to algal cells and physical disruption of the cell membranes; however, it is not possible to make a general conclusion on the factors determining the algal toxicity of TiO2 at the current state of knowledge. TiO2 NPs were also shown to

99

inhibit bacteria growth in the presence and absence of light. The presence of light was shown to be an important factor presumably due to its role in promoting the generation of ROS; however, the inhibition of bacterial growth in the absence of light suggests that undetermined mechanisms in addition to production of ROS were responsible for toxicity (Adams et al., 2006).

3.05.4.2.4 Ceria NPs Cerium oxide NPs (CeO2–x, where x is between 0 and 0.5) of range 1–10 nm have been the focus of most of the advances involving the use of these materials, and also it is well known that the distinct properties are strongly size dependent and would show significant quantum size effects (Xu et al., 2008). Therefore, the development of methodologies for the synthesis of monodispersed CeO2 NPs of different and well-controlled sizes is the ultimate aim in this field. These NPs have been produced using preparation routes similar to the ones used for TiO2 and ZnO NPs. Cerium dioxide, also named ceria (CeO2–x), is a cubic fluorite-type oxide which has high thermodynamic affinity for oxygen and sulfur (Bumajdad et al., 2009). Cerium is especially interesting among rare earths because of its ability to easily change its valance state from Ce(III) to Ce(IV) (Hailstone et al., 2009), and the ratio of the two oxidation states appears to be size dependent, with increasing Ce(III) at the lower size. CeO2 has been used as an oxygen sensor (Robinson et al., 2002) and as a diesel additive, increasing fuel efficiency. The latter use, with its potential to distribute NPs widely in the environment from diesel exhausts, is of particular concern. CeO2 NPs are one of the most rapidly growing nanomaterials due to its broad use in this and other areas such as polishing and computer chip manufacturing (Rothen-Rutishauser et al., 2009). More recently, CeO2–x NPs have been used to reduce oxidative stress in biological systems as a free radical scavenger (Niu et al., 2007), although they might also cause oxidative stress under different conditions. The evidence suggests that toxicity is low in humans (Park et al., 2008) in very short term exposure studies and that positive effects occur, such as reduction in the particle number concentration of ultrafine particles form diesel combustion, but further work is required given the likely high exposures from this use. Negative effects of CeO2 on E. coli have been reported (Thill et al., 2006), where the NPs can be adsorbed on the outer membrane of E. coli as shown in the TEM images of Figure 3. Ceria also has been demonstrated to show Fenton-like chemistry (Heckert et al., 2008). It has also been shown that CeO2 NPs can be reduced in biological media. This reduction has been observed during the contact between CeO2 NPs and human dermal fibroblasts in vitro (Auffan et al., 2009) with oxidative stress and strong genotoxic effects for very low doses (40.06 mg l1). The suggested mechanism of action via oxidative stress is not fully proved and has not been elucidated in algae and invertebrates (Van Hoecke et al., 2009), although close spatial association of NPs and cells was required for toxicity. Ceria NPs have been shown to inhibit algal growth more effectively than the bulk counterparts with 72-h IC50 vlaues of 10.371.7 and 66722 mg l1 respectively, and the toxicity was related to cell-membrane damage (Rogers et al., 2010).

100

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems 30

Adsorbed ceria NP (mg m–2)

25

20

15

10

5

0 0

200

400

600

800 1000 1200 1400 1600

Free ceria NP (mg l–1) Figure 3 Isotherm of adsorption of CeO2 NPs on E. coli bacteria. Insets show TEM observations of E. coli ultramicrotomic thin sections before and after contact with 12 mg m2 of adsorbed ceria. The scale bar is 0.1 mm. A zoom shows the multilayer of NPs at the cell outer membrane. From Thill A, Zeyons O, Spalla O, Chauvat F, Rose J, and Auffan M (2006) Cytotoxicity of CeO2 nanoparticles for Escherichia coli. Physico-chemical insight of the cytotoxicity mechanism. Environmental Science and Technology 40: 6151–6156.

3.05.4.3 Metal NPs Metal NPs have attracted special attention in the fields of biological and medical sciences, catalysis, and sensing (Murphy et al., 2008; Tao et al., 2008) due to their optical properties. The applications based on these materials can benefit from their sensitive and tunable electromagnetic responses. These NPs have been introduced in many commercially available products such as clothing, cosmetics, footwear, and plastic containers.

3.05.4.3.1 Gold and silver NPs Metal systems are expected to form the basis of new diagnostic biosensor technologies and novel therapeutic agents, and the most widely studied systems are based on gold and silver (Langer and Tirrell, 2004; Liu and Lin, 2007). Silver NPs (AgNPs) have been widely used in commercial applications as bactericides in fabrics, cosmetics, and other consumer products (Jeon et al., 2003; Kim et al., 2007). Risk from these NPs is potentially high for these reasons, that is, high inherent toxicity and large use in consumer products leading to high environmental exposure (Eckelman and Graedel, 2007). These NPs, especially Au, have interesting properties such as their stability, inertness (45 nm), and size-related and tunable electronic, magnetic, and optical properties. Furthermore, metal NPs have the advantages of easy preparation and the possibility of chemical modification of the surface (Haick, 2007) by a variety of capping agents. Metal NPs exhibit the so-called surface plasmon resonance (SPR), which is caused by the interaction with the incident light and the free electrons in the materials (Noguez, 2007).

The strong extinctions of conductive metal NPs arise from an electrodynamic phenomenon known as surface plasmons. These are generated by the collective excitation of free electrons in response to a characteristic electromagnetic frequency (Daniel and Astruc, 2004). The plasmonic coupling of metal NPs with light enhances a broad range of useful optical phenomena, such as resonant light scattering (RLS), SPR, and surface-enhanced Raman scattering (SERS), all of which have tremendous potential for ultrasensitive chemical and biomolecular detection and analysis (Rotello, 2004). The resonance effect depends on a number of properties such as shape, surface chemistry, aggregation state, and size. In general, the aqueous solutions of gold NPs stabilized with citrate are red colloidal solutions which present a surface plasmon band (SPB) at approximately 550 nm (Grabar et al., 1997), whereas the corresponding silver NPs have an SPB at approximately 420 nm (Doty et al., 2005); this varies though. SPR is an important analytical tool and is potentially important in environmental applications and environmental detection, although only at the most sensitive methodologies available. The surface plasmon has been used to quantify the changes in gold NP properties upon interaction with natural organic macromolecules, such as HSs, which absorb maximally at B254 nm (Diegoli et al., 2008). Diegoli and co-workers monitored aging and aggregation behavior of acrylate- and citrate-stabilized gold NPs in the presence and absence of HS by UV–visible absorption spectroscopy and TEM (Figure 4), at different pHs. AgNPs (and also dissolved silver) are known to have significant antibacterial properties, and are used in fabrics and cosmetic and have medical uses. Ag is used in dental resin composites (Herrera et al., 2001), in synthetic zeolites (Faghihian and Kamali, 2003), and in coatings of medical equipments such as catheters, infusion systems, and medical textiles (Markarian, 2006). The antibacterial activity of silver ion (Agþ) and related silver species and, to a lesser extent, AgNPs has been studied (Sondi and Salopek-Sondi, 2004; Morones et al., 2005; Pal et al., 2007; Fabrega et al., 2009a), although the exact mode of action is not fully known. The catalytic oxidation by metallic silver and reaction with dissolved monovalent silver ion likely contribute to its bactericidal effect. The antibacterial mechanism of silver-containing products possibly ends in a long-term release of silver ions (Agþ) (see Figure 5) by oxidation of zero-valent metallic silver Ag0 in contact with water (Kumar and Mu¨nstedt, 2005) and it has been shown, for instance, that the Agþ ion inhibits the enzymes for the P, S, and N cycles of nitrifying bacteria (Kumar et al., 2005). In addition, Agþ can block DNA transcription, interrupt bacterial respiration and adenosine triphosphate (ATP) production, and react with proteins by combining the –SH groups of enzymes which leads to the inactivation of the proteins (Jeon et al., 2003). When this metal is in a nanostructured form, higher antimicrobial activity is expected due to its larger specific surface area (Yoon et al., 2007) at least in comparison to the bulk. A number of detailed studies have investigated Ag toxicity to bacteria (Elechiguerra et al., 2005; Pal et al., 2007; Shahverdi et al., 2007; Sondi and Salopek-Sondi, 2004). AgNPs have been shown to increase antibacterial activity of antibiotics such as vancomycin and amoxicillin when used on S. aureus and

Absorbance

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems 0.60 0.55 0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10

AN pH 2.0

A

C

t = 290 min 400

500 600 700 Wavelength (nm)

800

AN+SRHA pH 1.5

0.35

Absorbance

B

t=0

(a)

0.30

101

t=0

0.25 0.20

A

0.15 0.10 0.05

t = 24 h

0.00 400

(b)

0.7

500

600

700

800

AN pH 12.5

0.6 Absorbance

0.5

t=0

0.4 0.3 A

0.2

B

C

0.1 0.0

t = 24 h 600 500 700 Wavelength (nm)

400 (c)

AN+SRHA pH 12.5

0.35 0.30

800

t=0

Absorbance

0.25 0.20 A

0.15 0.10 0.05

t = 24 h

0.00 400 (d)

600 700 500 Wavelength (nm)

800

Figure 4 Comparison between UV–visible absorption spectra recorded as a function of time and TEM micrographs of acrylate-stabilized gold nanoparticle dispersions. (a) AN pH 2.0, (b) AN þ SRHA pH 1.5, (c) AN pH 12.5, and (d) AN þ SRHA pH 12.5 dispersions. Scale bar: 200 nm. From Diegoli S, Manciulea AL, Begum S, Jones IP, Lead JR, and Preece JA (2008) Interactions of charge stabilised gold nanoparticles with organic macromolecules. Science of the Total Environment 402(1): 51–61.

102

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

2 wt.% 4 wt.% 8 wt.%

Ag+ concentration (gl–1 g–1)

3.0 × 10–4 2.5 × 10–4 2.0 × 10–4 1.5 × 10–4 1.0 × 10–4 5.0 × 10–5

1 × 10–7

0.0 2

4

7

14

Immersion time (days) Figure 5 Agþ release as a function of immersion time for different silver concentrations (0.78 m2 g1). The dashed black line shows the concentration of Agþ capable of rendering antimicrobial efficacy (Agþ ions released by 1 g of the composite to 100 ml aqueous environment). From: Kumar R and Mu¨nstedt H (2005) Silver ion release from antimicrobial polyamide/silver composites. Biomaterials 26: 2081–2088.

E. coli (Shahverdi et al., 2007). Both size and shape (Morones et al., 2005; Pal et al., 2007) have been shown to affect antibacterial properties, with smaller-sized NPs having more effect. In addition, triangular plates had improved antibacterial efficiency compared with spherical and rod-shaped NPs. Along with Agþ (i.e., AgNO3), triangular nanoplates show the strongest biocidal action. A number of these studies also observed aggregation of silver NPs and larger effects on bacteria on agar plates compared with bacteria grown in a culture medium. This effect was most likely due to the increased dose and certain experimental conditions such as temperature, pH, and mixing speed (Pal et al., 2007). The work of Fabrega et al. (2009a, 2009b) indicates that dissolved silver has higher antibacterial effects compared to AgNPs (20–50 nm, citrate stabilized), but that the NPs have a different toxicity mechanism compared to dissolved ions, a finding backed up by recent work on CuNPs in fish (Griffitt et al., 2007). Navarro and co-workers performed a study comparing the toxicity of dissolved silver ion (Agþ) and AgNPs in freshwater algae (Navarro et al., 2008a). They examined the short-term toxicity of Agþ and AgNPs to photosynthesis in Chlamydomonas reinhardtii using fluorometry. Their results indicate that the interaction of these particles with algae influences the toxicity of AgNPs, which is mediated by Agþ. NPs contributed to the toxicity as a source of Agþ which was formed in the presence of algae. They also showed that abiotic factors influence or affect the dissolution rates, such as size and surface area, or the chemical conditions of the environment. They also suggested that biotic interactions should be considered to assess the risks caused by NPs in natural aquatic systems (Navarro et al., 2008a). The exact role of size, surface properties, and dissolution has yet to be fully elucidated in relevant organisms.

3.05.4.3.2 Zero-valent iron NPs The zero-valent metal NPs have received great attention due to their potential applications in the remediation of

contaminated groundwater (Elliott and Zhang, 2001; Quinn et al., 2005). Several studies have shown that these iron NPs possess the capacity of transforming (nZVI, nano-zero valent iron) or sorbing (surface oxide layer) a wide range of common environmental contaminants, including chlorinated organic solvents (Nutt et al., 2005), organic dyes (Liu et al., 2005), various inorganic compounds (Alowitz and Scherer, 2002; Cao et al., 2005), and metals (Kanel et al., 2005; Xu et al., 2005). Despite the apparent lack of risk from iron itself (see argument above related to iron oxide NPs), zero-valent iron provides one of the few examples of an adverse environmental or human impact of NPs. Zhang (2003) showed that these NPs reduced high concentrations of solvents to nearly zero within days, but at the same time oxygen levels were reduced making the groundwater anoxic and pH levels changed significantly (Zhang, 2003). No further analyses were performed in this study, although presumably there were at least substantial short-term effects on groundwater ecology. In the absence of further applications, it is also likely that conditions returned to their previous state within a relatively short period of time, but continued application may have had a prolonged effect.

3.05.4.4 Quantum Dots The semiconductor NPs have attracted a special interest for their promising applications in molecular biology, medicine, and information technology (Chen, 2008; Gao, 2004), although given their composition (e.g., Cd), they are likely to be inherently highly toxic and large-scale use needs to be viewed with caution. Capping agents are likely to reduce dissolution and passivate, at least in the short term. Semiconductors are key components of devices used in computers, light-emitting diodes, sensors, etc. (Wang et al., 2007). Semiconductors are a unique class of materials, in that they can assume characteristic properties of both metals and isolators, depending on conditions that determine the electronic nature of valence and

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

103

QD

Plasma membrane lipid peroxidation

Fas NAC

ROS

FADD Caspase 8

in iplipin dridoilo r a a cC

Cardiolipin peroxidation Cytochrome c Mitochondrial membrane lipid peroxidation

Caspase cascade

Cell death

NAC

ΔM Δ Metabolic activity

Figure 6 Proposed mechanism of QD-induced cell death involving Fas, lipid peroxidation and mitochondrial impairment. Cells exposed to cadmium telluride quantum dots (unmodified and NAC-modified) induce ROS which causes Fas upregulation and plasma membrane lipid peroxidation. Apoptotic cell death is induced by activation of Fas and its downstream effectors. Lipid peroxidation also occurs at the mitochondrial membranes, degrading cardiolipin, changing the mitochondrial membrane potential, eventually leading to the release of cytochrome c, and promoting apoptotic cascades. NAC bound to the QD surface, modifies the extent of QD internalization, which is correlated with cell death, upregulation of Fas, and ROS-induced lipid peroxidation. NAC treatment (2–5 mM) abolishes oxidative stress, induces antioxidant enzymes and attenuates mitochondrial impairment. From Choi AO, Cho SJ, Desbarats J, Lovric J, and Maysinger D (2007) Quantum dot-induced cell death involves Fas upregulation and lipid peroxidation in human neuroblastoma cells. Journal of Nanotechnology 5(1): doi:10.1186/1477-3155-5-1.

conduction bands. In the ground state, the valance band is completely filled and separated from the conduction band by a narrow band gap (Eg) (Rotello, 2004; Schmid, 2004). Semiconductor nanocrystals, or QDs, have been reported from a variety of compositions, including CdSe, CdS, Si, GaAs, and PbSe (Rotello, 2004). The ecotoxicity of QDs has only recently gained interest (Moore, 2006). For instance, the toxicity of CdTe may be linked to the leaching of toxic heavy metals from the colloidal form, and derived from the intrinsic properties of the size and surface chemistry of the CdTe QDs (Clapp et al., 2004). In theory, they could transfer energy to nearby oxygen molecules and lead to the formation of ROS, which may lead to cell inflammation, damage, and death. In fact, Choi and co-workers showed that QDs can induce cell death in human neuroblastoma cells, and lipid peroxidation and mitochondrial impairment were proposed as possible mechanisms (see Figure 6) (Choi et al., 2007). More recently, the ecotoxicological effects of CdTe QDs to freshwater mussel (Elliptio complanata) have been reported showing that these NPs are immunotoxic to freshwater mussels and can cause oxidative stress in gills and DNA damage (Gagne´ et al., 2008).

3.05.5 Important Physico-Chemical Properties of Natural Colloid The behavior, fate, and environmental functions of natural colloids and manufactured NPs are strongly influenced by their physico-chemical properties such as size, shape, surface

charge, surface coating, and others. These properties are rarely, if ever, uniform. This implies the need to understand these properties, factors influencing them, the extent of their variations, and some of the experimental methods that are available to measure these properties.

3.05.5.1 Size As discussed above, size is the primary means of defining colloids (see Section 3.05.2.1) in natural systems and manufactured NPs (see Section 3.05.2.2). Size is a useful parameter that can potentially help understanding the behavior of colloid in the environment and its role in the biogeochemical cycling of trace contaminants and nutrients, since other important physical and chemical parameters relevant to colloidal behavior, for example, specific surface area, surface reactivity, diffusion coefficient (Lead et al., 2000a, 2000b), pollutant binding and speciation (Lyven et al., 2003; Stolpe and Hassellov, 2007), bioavailability and biouptake of pollutants (Guo et al., 2002; Pan and Wang, 2002), sedimentation, and transport, are influenced greatly by colloid size. Smaller particles, in general, have a larger surface area, a higher adsorption capacity, diffuse more rapidly, and travel for longer distances in the environment. In addition, small particles reduce the bioavailability of chemicals by reducing the free fraction (Pan and Wang, 2002), though they might be themselves bioavailable. Natural colloids are usually characterized by a wide range of size distribution. Monodisperse colloids have never been reported in natural systems, although laboratory studies usually use monodisperse colloids. The size distribution of

104

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

natural colloids depends on the source and nature of particles, physical, chemical, and biological processes such as erosion, degradation, aggregation, disaggregation, and aging and the physico-chemical parameters of the system such as pH, ionic strength, and redox potential. Figures 2(a) and 2(b), correspondingly, show typical TEM images of freshwater natural colloidal particles (Buffle et al., 1998). Figure 2(a) shows that the materials with different sizes from (1) 1–2 nm, presumably correspond to HSs; (2) several hundreds of nanometers with clear defined angular edges, presumably correspond to inorganic colloids (possibly clays); (3) several hundred nanometers of globular shape, presumably correspond to iron oxides; and (4) fibrillar materials of several nanometer thickness and several hundreds of nanometer length, presumably correspond to biopolymers. Figure 2(b) shows mainly roughly spherical silica particle (gray at center) aggregated with smaller iron hydroxide particles (black spheroids). A simplified classification of environmental particles together with their approximate size range is given in Figure 2(c). Several observations can be made. For instance, the size of each class spreads over several orders of magnitude and the boundary between different colloid types is somewhat artificial and rarely found in natural systems. More realistically, colloids are often found as components of heteroaggregates and sorbed to larger particles (Figure 2(b)). Although polydisperse, natural colloids can be fractionated either by physical (e.g., size exclusion chromatography (SEC), field flow fractionation (FFF), and crossflow ultrafiltration (CFUF)) or chemical (e.g., XAD resin for fractionation of HSs) methods into simpler fractions that can be studied individually, and this has improved our understanding of natural colloids and their environmental functions. In addition, the development of sample preparation methods with minimum perturbation and new analytical tools (e.g., FFF, CFUF, TEM, and AFM) for colloid fractionation and size determination along with combining these methods has increased our knowledge of colloid size and functions dramatically (Baalousha et al., 2005 b; Baalousha and Lead, 2007; Gibson et al., 2007). In particular, a multimethod approach allows the determination of particle-size and nonsize parameters that are useful to describe colloidal structure such as shape, morphology, sphericity, and permeability (Baalousha et al., 2005a; Baalousha and Lead, 2007). Therefore, a multimethod approach is preferred (Buffle and Leppard, 1995b). Size distribution obtained from the different methods, even on the same sample and prepared with minimum sample perturbation methods, is often different. This can be explained by the differences in physical principles of the different techniques, measured parameters, detection limit, and analytical window of each method (summarized in Table 4). The measured size parameters include: physical size (microscopy techniques), hydrodynamic diameter (including a hydration layer, FFF), and radius of gyration (mass distribution within a particle, laser light scattering (LLS)). In addition, the size distribution may be related to particle number (microscopy techniques), mass (FFF), or surface area (Brunauer, Emmett, and Teller (BET)). Furthermore, the average diameter can be described by the number-, weight-, or z-average (Table 4).

3.05.5.2 Shape and Morphology Shape and morphology is one of the most important parameters to describe and classify natural colloids. Natural colloids frequently have anisotropic and irregular shape and are rarely, if not ever, isotropic (Balnois et al., 2003). This can be explained by the different chemical (e.g., pH, ionic strength, and acidity), mechanical (e.g., erosion, flow, and transport), and biological conditions to which natural colloids are exposed. Particle shape reflects material composition, release from the matrix, and transportation, and may be different from one site to another. Therefore, colloids have what may be termed a particle shape distribution. Microscopy observations show natural colloids from Rhine River (Figure 2(a)) with different shapes, including spherical, globular, irregular particles with defined angular edges, and fibrillar materials. Figure 2(b) shows natural colloids from Lake Bret, again with different shapes, mainly plate-like particles with some spheroids (Buffle et al., 1998). In addition, when colloids aggregate, many different shapes may form (see discussion in Section 3.05.7.2) that do not necessarily correspond to the primary particles. This illustrates the additional complexity in describing and defining colloids and makes quantification and incorporation of this parameter into usable models of behavior very difficult. For crystalline colloids such as clays, shape is partially determined, in addition to other parameters, by crystallographic structure (e.g., hexagonal shape of kaolinite vs. tubular shape of halloysite) and they are usually characterized by sharp edges. This might be the case in laboratory studies (and it is the case for synthesized NPs, described later) but is rarely the case in the natural environment due to erosion and degradation processes. Amorphous (noncrystalline) colloids such as organic macromolecules are characterized by globular or fibrillar structures with no sharp edges. Several methods have been used to quantify the shape of particles based on image analysis such as sphericity or circularity:

rffiffiffiffiffiffiffiffiffi 4pA C¼ P2

ð1Þ

where A and P are the area and perimeter of the particle, respectively. Others, such as aspect ratio (length/width), fractal dimension (see Section 3.05.7.2), etc. (Hentschel and Page, 2003), are frequently used. A major problem with these calculated parameters from image analysis is that they give a two-dimensional (2D) description, while shape and morphology are 3D. It is possible to overcome such problems by electron tomography measurements, that is, 3D reconstructions of images obtained at different electron microscopy (EM) stage tilt angles (Gontard et al., 2006), though this is very time consuming. Furthermore, particle morphology and aggregate structure can be determined by other methods such as static laser light scattering and small angle neutron scattering. As with other colloidal parameters, combination of analytical techniques has been used to produce other parameters that can describe particle shape and aggregate structure. For instance, combination of flow FFF (FlFFF) and static light scattering (SLS) gives a shape

Number 41 nm

Scanning a probe on a surface

0.1 nm height

Number

Principle

Spatial resolution

Size distribution

Time consuming, require large number of particles for representative PSD TEM require special sample preparation Require UHV

Direct observation of the particles, gives semi 3D information

High resolution, visual observation of the particles

Difficulties

Advantages

430 nm UHV

UHV

40.5 nm

Ambient air Liquid High resolution analysis can be performed under ambient pressure and in aqueous media

Analytical window

Number

5–20 nm

Interaction of electrons with matter

PSD, morphology, and topography

SEM

Sample environment

0.1–0.2 nm

Interaction of electrons with matter

PSD, morphology crystallography structure, defects

PSD, shape, tip– specimen interaction forces

Measured parameter

HR-TEM

AFM

Particle size characterization techniques

Method

Table 4

Separation occurs in liquid media Possibility of hyphenation with other techniques such as light scattering and mass spectroscopy Time consuming, material losses, data interpretation, and representation

Nondestructive, fast and averaging of properties

Possibility of sample damage Low spatial resolution UHV

Easy, fast

Representation of the calculated x potential

nm range UHV

0.001–3 mm

Volume

10 mm

Structural analysis of crystals, amounts of different crystalline phases, crystal size Elastic interaction of X-rays with matter

XRD

Liquid

o0.45 mm Liquid

Intensity based

Measuring hysteresis in scattered light intensity due to Brownian motion -

Diffusion coefficient, hydrodynamic diameter

DLS

Mass

o1 nm

Interaction of colloids with an applied field

Diffusion coefficient, hydrodynamic diameter

FlFFF

Representation of the measured surface area, UHV

Easy, fast, direct measurement of surface area regardless particle morphology

UHV

nm–mm range

Surface area

-

Adsorption of a gas into the surface of particles

Surface area

BET

106

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

factor, defined as the ratio of the radius of gyration to the hydrodynamic radius of the colloids (Baalousha et al., 2006b). Combination of FlFFF with TEM allows 3D characterization of colloidal particles (Baalousha et al., 2005a), while combination of FlFFF with AFM allows particle sphericity and permeability to be quantified (Baalousha and Lead, 2007).

3.05.5.3 Surface Coating There is a clear evidence from the 1970s onward that NOM adsorbs to macroscopic surfaces from electrophoretic measurements (Neihof and Loeb, 1975; Tipping and Higgins, 1982; Hunter and Liss, 1982), and more recently from surface force measurement by AFM (Mosley et al., 2003; Assemi et al., 2004; Sander et al., 2004). Other analytical tools such as AFM, FlFFF, and others have been used to determine the physical dimensions and structure of these films on macroscopic and nanoscopic surfaces, that is, NP–colloid interactions (Mayer and Xing, 2001; Lead et al., 2005; Baalousha et al., 2008; Gaebel et al., 2009), with thicknesses of HSs in the range 1–5 nm (Assemi et al., 2004; Gibson et al., 2007), while natural colloids are sometimes much larger. The formation of NOM surface coatings on colloids can dominate their environmental functions as well as their fate and behavior. For instance, surface coating may alter colloidal surface charge, hence the similarity in surface charges of colloidal particles in aquatic environment and their mutual repulsion, slowing down aggregation. Almost all environmental particles, regardless of chemical composition, are negatively charged due to the dissociation of surface functional groups on sorbed NOM (Hunter and Liss, 1982; Loder and Liss, 1985). Therefore, a useful approximation in terms of surface charge and aggregation may be to treat colloids as a single class of colloidal materials, irrespective of their nature (O’Melia, 1980; Filella and Buffle, 1993). However, these surface coatings may be patchy (Gibson et al., 2007), depending on the nature of the underlying substrate, the NOM type, and the solution conditions, meaning that this assumption does not always hold. Surface charge modification will have direct role on the stability, aggregation, and disaggregation of colloids (see Section 3.05.7.1). Surface coating is also likely to alter trace contaminant interaction with colloids. For instance, their formation will change the solid–solution partitioning of trace pollutants. NOM may also increase the sorption of organic pollutants such as carbazole, dibenzothiophene, and anthracene. In general, models describing reactions at inorganic surfaces assume that the surfaces are clean and free from organic matter, and models describing both inorganic and NOM binding to metals assume nonadditivity (Tipping, 2002).

of ionic solids (e.g., AgI); or (3) specific sorption of charged species (e.g., simple ions such as Ca2þ, surfactant ions, and polyelectrolyte chains such as HS or synthetic surfactants). The total colloidal surface charge is the sum of permanent and variable charges. The colloid surface charge must be balanced by equal and opposite charge in solution so that the colloidal system is electrically neutral. This balancing charge is created by an excess number of oppositely charged ions (counterions), and a deficit of similarly charged ions (co-ions) in the vicinity of the particle surface. This distribution of ions around a charged particle surfaces may be described by the electric double layer theory. Several models have been presented, though we only describe one model (i.e., Stern–Grahame–Gouy–Chapman model) as it is one of the most elaborate descriptions of the double layer theory (Figure 7). In this model, ions are distributed across two layers, a compact inner layer (Stern layer), where the counterions are immobile and a diffuse outer layer, which extends over a certain distance from the particle surface and decays exponentially with increasing distance into the bulk liquid phase. The distribution of ions in the diffuse layer depends on the concentration of the electrolyte, the charge of the ions, and the potential at the boundary between the compact inner layer and the diffuse outer layer. The potential at this interface is called the Stern potential. The potential at

Stern layer

Diffuse layer

– –

Colloid surface charge can be either permanent or variable. The permanent charge arises from the isomorphous substitution of cations within the colloid, for example, substitution of Si(IV) by Al(III) in kaolinite. The variable charge originates from chemical reactions at the colloidal surface: (1) ionization or dissociation of the surface functional groups (e.g., the dissociation of protons from carboxylic groups); (2) dissolution

+

+

+ –

+

– +

–

+

+

–

– +

–

– +  s ζ s/e xs

3.05.5.4 Surface Charge

Bulk solution

1/

x

Figure 7 Schematic diagram of the diffuse double layer (DDL) forming from the surface of a colloidal particle into the bulk solution. Abbreviations: zeta potential (z), electrostatic potential (C), electrostatic potential at the stern layer (CS), Euler’s number (e), Boltzmann constant (k). X is a distance from the surface, Xs is the shear plane, the distance from where ions and molecules are mobile and can be sheared off. Adapted from figure 2 in Handy RD, Von der Kammer F, Lead JR, Hassello¨v M, Owen R, and Crane M (2008) The ecotoxicology and chemistry of manufactured nanoparticles. Ecotoxicology 174: 287–314.

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

the shear plane, that is, the transition plane from fixed ions and water molecules to those which can be sheared off by fluid motion, is called the zeta potential (z), which can be calculated from the electrophoretic mobility measured by electrokinetic methods (e.g., electrophoresis). Under conditions of very low ionic strength, the decay of the potential between the Stern layer and the shear plane is negligible and the zeta potential can be seen as an approximation of the Stern potential. For more details about the different models describing the double layer, the reader is referred to the literature (Elimelech et al., 1995). Colloid surface charge can be determined indirectly by determining the zeta potential (z) of the particles from the measured electrophoretic mobility (m) which is the velocity of particles (v) per electric field unit (E), m ¼ n/E. Although measuring the electrophoretic mobility is straightforward, calculation of zeta potential is more complicated, for both colloids/NPs and when performed in environmental media. For instance, the conversion of the electrophoretic mobility to zeta potential is based on the assumption that the particles are approximately hard (nonpermeable) spheres (Delgado et al., 2007). Further models have been developed for soft (permeable) particles and applied to environmental colloids, where the concept of zeta potential is not physically meaningful (Duval et al., 2005). Permeable or semi-permeable colloidal models have been applied for environmental colloids such as HSs (Duval et al., 2005; Duval, 2007), and bacterial cells (Hayashi et al., 2001; Tsuneda et al., 2003; de Kerchove and Elimelech, 2005). In addition, for nonspherical particles, such as fibrils or fractal or porous aggregates, zeta potential values are likely to be inaccurate or misleading. Colloid surface charge plays an important role in determining their fate and behavior in the environment. Surface charge is one of the primary characteristics that determines colloidal stability, aggregation, and disaggregation. Altering surface charge through changes in solution chemistry (e.g., ionic strength, pH, and NOM) is the practical means of manipulating colloid stability, aggregation/disaggregation, and

sedimentation/deposition in engineered systems, while natural changes such as increased ionic strength in estuarine conditions will have the same effect.

3.05.5.5 Pollutant Binding and Behavior Surface reactions play an important role in environmental processes such as colloid stability (Johnson et al., 2005), dissolution rate (Johnson et al., 2004), trace contaminant speciation and transport (Joo et al., 2008) and can be important in catalyzing certain degradation reactions (e.g., chlorinated volatile organics). Colloids play an important role in regulating chemicals (e.g., contaminants and nutrients) in the environment due to their small size and consequently high surface area/volume (Figure 8). For instance, they often dominate the physicochemical speciation of trace elements and organic pollutants (Buffle, 1988; Doucet et al., 2006). A large proportion of these trace compounds (typically 40–90% or more) are adsorbed to colloids (Stumm, 1992; Stumm and Morgan, 1996). The binding of trace pollutants by colloids can be interpreted as a function of colloid size (Lead et al., 1999), chemistry of the colloidal phases (Lienemann et al., 1997), or both (Lyven et al., 2003; Baalousha et al., 2006a). It has been suggested that small colloids o50 nm (Lead et al., 1999) or o25 nm (Lyven et al., 2003) are capable of binding the largest fraction of total trace metals. In addition to size colloid chemical composition plays an important role in metal binding. Lyven et al. (2003) identified iron oxides and organic carbon as the main colloidal binding phases to which trace elements were associated with elements such as Cu and Zn associated to organic carbon and others such as Pb associated to colloidal iron oxides. In another study using the same coupled FlFFF–ICP–MS technique, Hassellov and co-workers found iron oxide as a major vector for metal transport, in particular Pb (Hassellov and van der Kammer, 2008). Other phases such as manganese oxides may also be important in trace element binding (Baalousha et al., 2006a). Dong and co-workers have investigated the

Nanoparticles

Macroscopic particles

Change in reactivity (a.u.)

Atomic clusters

107

10–10

10–9

10–8

10–7 Particle size (m)

10–6

10–5

Figure 8 General tendency for size-dependent reactivity change of a material as the particle transitions from macroscopic (bulk-like) to atomic. From Wigginton NS, Haus KL, and Hochella MF (2007) Aquatic environmental nanoparticles. Journal of Environmental Monitoring 9: 1306–1316.

108

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

sorption of Cd and Pb to surface coating (Fe and Mn oxides, and NOM) and found that they sorb significantly to metal oxide (Fe and Mn) colloids, but not to NOM (Dong et al., 2000, 2001). In particular, Fe and Mn oxides play an important role in the biogeochemical cycling of trace metals at the redox interface in lakes (Balistrieri et al., 1992a, 1992b), resulting in the removal of dissolved metals to the surface of Fe and Mn oxides as they form at the redox interface. In addition, the aggregation of colloids in natural waters results in the removal of metals from water column through a process known as colloidal pumping (Honeyman and Santschi, 1991; Stordal et al., 1996; Wen et al., 1997). As with inorganic contaminants, colloids may significantly influence the distribution, and fate and behavior of organic contaminants. For instance, majority of polycyclic hydrocarbons (PAHs) were found to be present in large (420 mm) flocs (Leppard et al., 1998), which were essentially aggregates of small colloids. Marvin et al. (2004) showed that PAHs were primarily associated with particles less than 2 mm in diameter. Majority of these particles were found to be fractal aggregates of HS. In marine systems, majority of polychlorinated biphenyls (PCBs) were found to be associated with particulate matter (41.2 mm), although in the fraction o1.2 mm, colloidal binding (40–80%) was dominant (Burgess et al., 1996). Up to 93% of PCBs were found to be associated with colloids in a coastal sea area (Totten et al., 2001). The interaction of selected pharmaceuticals (Maskaoui et al., 2007) and endocrine disrupting chemicals (EDCs) (Liu et al., 2005; Zhou et al., 2007) with natural colloids has also been more recently investigated. While the more hydrophobic pharmaceuticals showed a linear dependency of the Kcoc (colloidal organic carbon sorption coefficient) and the Kow (octanol–water partition coefficient), the Kcoc of the more hydrophilic EDCs was independent of the Kow, highlighting the importance of different binding mechanisms. Polychlorinated dibenzop-dioxins and dibenzofurans (PCDD/Fs) were found to be relocated from soil to groundwater associated with colloids (Hofmann and Wendelborn, 2007). Understanding the role of colloids in regulating chemicals in the environment has significantly improved our understanding of their bioavailability, toxicity, and transport. In the metal area, models based on equilibrium chemistry such as biotic ligand model (BLM) and free ion activity model (FIAM) have helped improving our understanding of metal toxicity (Campbell, 1995; Paquin et al., 2002; Slaveykova and Wilkinson, 2005). Models based on dynamic processes, although newer, are being used fairly extensively (van Leeuwen and Koster, 2004; van Leeuwen et al., 2005). The reader is referred to these references for a detailed discussion. In addition, it is now well known that contaminant and nutrient transport processes in marine and freshwater systems, and in surface and subsurface waters are dominated by the transport of colloids and particles (Dai et al., 1995; Santschi et al., 1997; Benedetti et al., 2002). For decades, processes of contaminant relocation in soil and groundwater were believed to occur predominantly in a two-phase system (the mobile liquid phase and the immobile solid phase) and a potentially mobile solid phase was neglected (McCarthy and Zachara, 1989). Colloid-facilitated transport is now a well-recognized process in porous media such as soils and aquifers. Small

colloids compete with the solid, immobile phase for trace contaminants sorption (e.g., metals; Chen et al., 2005), organic pollutants (White et al., 2005), and nutrients (Heathwaite et al., 2005) and increase the distances traveled by pollutants with respect to those predicted from noncolloidally bound components (Kaplan et al., 1995; McCarthy, 1998; Laegdsmand et al., 1999).

3.05.5.6 Interaction Forces As colloidal particles approach each other, different interaction forces take place, including electrical double layer, van der Waals, hydration, hydrophobic, and steric forces, which act over a relatively short distance and depend on surface properties and surface coating of particles. The electric double layer (long-range, repulsive) force results from the overlap of the double layer of two colloidal particles as they approach each other, resulting in a repulsive force, which opposes further approach:

V A ðh Þ ¼ 

" A 2R 2 2R 2 þ 2 6 h þ 4Rh ðh þ 2RÞ2 !# 4R 2 þln 1  ðh þ 2RÞ2

ð2Þ

This equation applies for two spherical particles of equal radius, R, at a surface to surface separation distance, h, apart along the center to center axis (Liang et al., 2007), where A is a constant, named ‘Hamaker constant’, which depends on the material properties such as density and polarizability. The effective Hamaker constant (Equation (3)) depends also on the dispersion medium, and is generally of order of magnitude 10–20–1021 J (Elimelech et al., 1995):

Aeff E

pffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi2 Aparticle  Amedium

ð3Þ

The van der Waals forces, short-range attractive, arise from spontaneous electrical and magnetic polarizations as particles get close to each other, giving a fluctuating electromagnetic field within the media and in the gap between particles. For identical particles, sphere–sphere double layer interaction energy can be given by Equation (4). There are many expressions available based on various assumptions for sphere– sphere double layer interaction energy and readers are referred to the literature for more details (Bell et al., 1970; Carnie et al., 1994; Sader et al., 1995; McCormack et al., 1995; Stankovich and Carnie, 1996; Genxiang et al., 2001):

 2 kT VR ðhÞ ¼ 32peR g2 expðkhÞ ze

ð4Þ

For small values of surface or zeta potential (z), Equation (4) simplifies to

VR ðhÞ ¼ 2peRz2 expðkhÞ

ð5Þ

where e is the permittivity of the medium; R the particle radius; g the dimensionless functions of the surface potentials; k

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

the Boltzmann constant; T the absolute temperature (kelvin); h the surface–surface separation between particles (m); e the electron charge; and k the inverse of Debye–Huckel screening length (m–1). Equation (5) is applicable only if kR 4 5 and h oo R. For the general case of electrolyte solutions containing a number of dissolved salts, k is defined by

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  2 P 2 ni zi e k¼ ekT

ð6Þ

where n is the number concentration of ion i in the solution. Inserting numerical values appropriate to aqueous solutions at 25 1C and converting the ion concentration into molar terms gives

k ¼ 2:32  10 9

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi X  ci z2i

ð7Þ

where c is the concentration of ions expressed in mol l1 and z the valency of the ions. The length 1/k is known as the thickness of the diffuse layer. Equation (7) shows that the increase in ionic strength results in a decrease in the thickness of the diffuse layer and a consequent decrease in the repulsive interactions among particles. Typical values of the diffuse layer thickness, 1/k, are in the range of 1–100 nm. The summation of the two forces gives the DLVO theory (named after Derjaguin and Landau, Verwey and Overbeek) (Equation (8)):

VT ¼ VA þ VR

ð8Þ

where VT is the total interaction energy; VA the attractive van der Waals energy; and VR the repulsive double layer energy. Although DLVO theory was found to be able to describe colloidal stability of simple colloidal systems, it has been found unable to fully describe colloidal behavior in aquatic and terrestrial environments (Grasso et al., 2002; Sander et al., 2004). This has been related to other, non-DLVO, forces such as hydration, hydrophobic, steric, and bridging forces. Hydration force results from the hydration layer that surrounds colloidal particles which might be different from the bulk water. For true contact to occur between particles, surfaces need to become dehydrated, which usually gives an extra repulsion to that induced by the double layer (Grasso et al., 2002). The hydrophobic forces result from the migration of the water from the distances between two hydrophobic colloidal particles, resulting in an attractive force (Elimelech et al., 1995). Steric force results from the interaction between particle surface coating (usually NOM in natural waters and stabilizing agents in case of NPs, see Section 3.05.5.3). As particles approach each other, the adsorbed layers come into contact, resulting in the interaction between these molecules. As these molecules are hydrated, any interaction will induce hydration repulsive forces as described in the previous section. The steric stabilization effect increases with increased surface coating load or the thickness of the adsorbed layer. Larger particles will need a thicker coating due to the increase of van der Waals attraction energy with particle size. Bridging results from the interaction of colloidal particles with long chain molecules such as polysaccharides in the natural environment, with long

109

chain molecules attaching to two or more particles. In such a case, particles can form aggregates even though they may be charged and repel each other, forming open flocs. Quantitative physico-chemical aggregation theory (DLVO theory) exists only for identical, compact, and spherical particles (homoaggregation). However, there is no such general theory for aggregation of a mixture of different particles (heterocoaggregation), in particular for aggregation involving polymers. Non-DLVO forces are complicated and difficult to describe, and no simple, comprehensive theory is yet developed. According to DLVO theory, parameters that affect colloidal stability are: ion type and concentration, particle size and particle surface charge, and z potential. Increased ionic strength results in a decreased double layer repulsive force (Equation (2)) because of a decreased diffuse layer thickness (Equation (7)). Polyvalent electrolytes induce larger decreases in the diffuse layer thickness than monovalent electrolytes and consequently induce greater aggregation. Both attractive and repulsive forces are proportional to particle size based on Equations (2) and (5), and generally electrostatic stability increases with increasing particle size. Increased zeta potential results in a higher colloidal stability as the electrostatic repulsive force is proportional to the square of z potential (Equation (2)), and so it is a key parameter in determining the stability of colloids. Nonetheless, the stability of colloidal particles in aquatic environment is often higher than expected on the basis of particle size, zeta potential, and ionic strength governing DLVO theory. This is likely to be related to the steric stabilization effect induced by NOM surface coating (Jekel, 1986), and possibly to the hydration effect.

3.05.6 Intrinsic Properties of Manufactured NPs NPs are not merely small crystals, and the nanoscale can be considered as an intermediate state of matter placed between bulk and molecular material. Alongside the particle size, morphology and surface charge play dominant roles in the chemical reactivity of the particle. These parameters will determine their ability, for instance, to enter the cell membranes and interact with different living organisms present in natural terrestrial and aquatic systems.

3.05.6.1 Size As particle sizes are decreased within the nanoscale range (and therefore take on different properties vs. the larger fine-sized particle types), alterations in their physical and chemical properties are found. It is not unreasonable to assume that the biological effects associated with exposures to NPs may also differ from their bulk counterparts (Warheit et al., 2008). Therefore, assessment of the potential health risks due to exposure of NPs is an emerging area in toxicology, exposure assessment, and health risk evaluations. Some effects that are only found in the NP size range are in the field of physics, for example, optical properties of NPs (gold NPs in different colors depending on their size, fluorescent QDs), superparamagnetism of small magnetic NPs (Aime et al., 2002), or supercooling of fluids in confined

110

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

geometries. Some of these effects are exploited in chemistry such as the high surface area for catalysis (Alvarez-Roman et al., 2004a) and adsorption (Alvarez-Roman et al., 2004a). The general interest in the monodisperse NPs has also arisen due to the possible applications in a wide range of domains (Poole and Owens, 2003). One of the important characteristics of all NPs is their unusually high specific surface area. This feature often has a considerable effect on the particle physical properties as well as being on its fate and behavior in natural environments (Christian et al., 2008). The physical and chemical properties of a material are also determined by the type of motion allowed for its electrons to execute, as in the case of QDs. The latter is determined by the space in which the electrons are confined (Braun et al., 2002). Unbound (unconfined) electrons have motion that is not quantized and can thus absorb any amount of energy given to them and use it simply to move more rapidly (Brus, 1986). Once bound in an atom, a molecule or a material, its motion becomes highly confined and quantization occurs. The allowed types of motion in atomic or molecular orbitals are found to have well-defined energies that are separated from one another (Braun et al., 2002). The smaller the space in which motion occurs (i.e., the stronger the confinement), the larger the energy separation between the allowed energies of the different types of motion becomes. Theoretically, if the physical size of the NP is reduced, it becomes comparable to or smaller than the Bohr radius. This would decrease the space in which the charge carriers (the excitons) move and thus confine their motion (Braun et al., 2002). For instance, in the case where the size of the semiconductor NPs becomes smaller than their Bohr radius, their band-gap energy increases (Alivisatos et al., 1988). Equally important, the energy of the band-gap absorption (and thus the NP color) and that of the emission increase, and become sensitive to the size of the particles (Heath, 1995; Wang and Herron, 1991). Thus, the optical and other physical and chemical (e.g., oxidation– reduction) properties of semiconductor NPs are size and shape dependent. As particle size becomes smaller, a greater fraction of atoms are at the surface and quantum effects tend to increase surface reactivity and energy, in general (Klaine, 2009; Wiesner et al., 2006, 2009). In some cases this does not occur, for example, due to nanoscale pit formation (Arugete et al., 2009). At the same time, NPs have a tendency to agglomerate and form larger structures. Thus, agglomeration can lead to a reduction in the number of atoms at the surface with a reduction in surface energy, although aggregation state may formally have no relation to SSA, dependent on size distribution and fractal dimension of aggregate (Buffle and Leppard, 1995a; Buffle and Leppard, 1995b; Zhang et al., 2007). Since coagulation half-lives of NPs are of the order of tens of microseconds to a few milliseconds (Preining, 1998), NP concentrations can decrease rapidly by agglomeration. Manufactured NPs, however, are specially coated to reduce agglomeration in order to exploit high surface reactivity for various useful ends.

3.05.6.2 Shape and Morphology Besides size and chemical composition, other NP properties such as shape and morphology may also affect NP transport,

bioavailability, and their toxic effects (Nel et al., 2006). The morphology of NPs is a key feature for exploiting their properties in several emerging technologies and diverse applications. For instance, selective optical filters (Ahmadi et al., 1996) and biosensors (Antognozzi et al., 1997) are among the many applications that use optical properties of metal NPs related to SPRs which depend strongly on the anisotropy of the particle shape; different shapes produce greater plasmon losses (Barth and Henry, 2004). Despite the great importance of the morphology of NPs, it is generally not well characterized and in practice almost never controlled. This situation is due to the intrinsic difficulty to accurately characterize the morphology of NPs and due to the limited number of ways known for controlling shape. The morphology of NPs depends on both kinetic (i.e., growth) and thermodynamic parameters (Adair and Suvaci, 2000). If the growth takes place far from equilibrium conditions (i.e., large supersaturation), the growth shape is not uniquely defined and depends on many parameters, such as the flux of growing material, structure of the support (if present), presence of defects and impurities, and confinement (i.e., template effect) (Adair and Suvaci, 2000). Generally, these parameters are not well controlled or not controlled at all. However, in the case of well-defined systems, it is possible to reduce the number of growth parameters and attempt to control the shape of growing particles. In the case of 2D growth, use of single crystal substrates with known surface diffusion anisotropy has enabled the preparation of 2D islands with shapes tuneable by the growth conditions (Roder et al., 1993). In the case of 3D growth, it is much more complicated to control all growth parameters. In terms of the toxicity related to the shape of the NPs, little is known, and majority of the studies reported in the literature are based on human inhalation exposures. Prior experience with asbestos and other fibrous aerosols indicates that the shape of the particles (i.e., their length and diameter) has a profound effect on toxicity. Smaller diameter fibers penetrate deeper into the respiratory tract, while longer fibers are cleared more slowly (Mossman et al., 1990; Oberdo¨rster et al., 2005). Engineered NPs come in various shapes such as spheres (e.g., dendrimers), tubes (e.g., SWCNT and MWCNT), plates (e.g., nanoclay flakes), fullerenes, and needles. While it seems likely that particle shape will affect the deposition, fate, and toxicity of the particles in the human body (Jia et al., 2005), few data about these effects are available. On the other hand, it has been suggested in the literature that there are two reasons that might account for the DNA damage caused by NPs. First, ROS generation and oxidative stress in the cell may cause oxidative damage to DNA through free radical attack. Previous work demonstrated that sunscreen TiO2 and ZnO can catalyze oxidative DNA damage in cultured human fibroblasts measured by comet assay (Dunford et al., 1997). In addition, this has been proved by the determination of 8-hydroxy-deoxyguanosine, a good marker of oxidative DNA lesion (Papageorgiou et al., 2007). However, it has been reported that CNTs exhibited greater genotoxicity than ZnO NPs which elicited more oxidative stress. Therefore, it is educible that the DNA damage caused by CNTs may come from mechanical injury and not oxidative effect. It is likely that CNTs might penetrate into cell nucleus through nucleopores,

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

and then destruct the DNA double helix (Pantarotto et al., 2004). Second, although several studies have shown that some spherical NPs such as titanium dioxide or silica NPs can also enter the nucleus (Geiser et al., 2005) and it has been demonstrated that C60 NPs can bind to and deform nucleotides (Zhao et al., 2005), CNTs induced significantly more DNA damage than other NPs with the sphere shape or crystal structure in our research. To combine the above two points, the genotoxicity of different NPs may primarily be due to particle shape rather than chemical composition.

3.05.6.3 Surface Properties At the nanoscale range, the properties of materials differ significantly from those corresponding to bulk materials of the same chemical composition due to the increased specific surface area and reactivity, which may lead to increased bioavailability and toxicity. The surface properties of NPs are one of the most important factors that govern their stability and mobility as colloidal suspensions or their aggregation into larger particles and deposition in aquatic systems (Navarro et al., 2008a). The aggregation of NPs released in the environment may be caused by several environmental parameters. For instance, in the case of TiO2, it has been reported that the NP aggregation behavior strongly depends on the pH and ionic strength of the liquid medium. Cationic and anionic species, and also the presence of HSs may affect the stability of TiO2 colloidal suspensions (Ottofuelling et al., 2007). It has also been reported in the literature that NP aggregation may have repercussions on their toxicity. However, the evaluation of the toxic effect of aggregates could be a difficult task if the specific surface area is not assessed, which is also involved in solubilization, adsorption, and catalytic properties (Kahru et al., 2008). In addition, in the case of metals, it is well covered in the literature that reactive chemical metal species depend not only on solubility, but also on all associated ions and on slight changes of pH. It is well known that surfaces and interfaces of particles are key components of nanoscale materials. As the particle size decreases, the amount of atoms found at the surface increases in comparison to the proportion of atoms found in the NP core. This implies that particles in the nanoscale are likely to be more reactive compared to their larger counterparts. However, when the potential health implications are considered, reactive groups on the surface of particle are also likely to influence the potentially toxicological effects when compared to nonreactive surfaces or coatings which tend to passivate (Warheit et al., 2008). Therefore, the shell of the NPs generated by the chemical modifications on their surfaces may also be important and relevant for their toxicity. In addition, surface coatings can be utilized to alter surface properties of NPs to prevent aggregation or agglomeration with different particle types facilitating particle dispersion and, at the same time, serve to passivate the NPs to moderate the effects of UV radiation-induced reactive oxidants (Warheit et al., 2008). Besides the importance of the NP core-shell dynamics for biological effects, different atomic planes on NP surfaces can lead to different results in toxicological studies. For instance, in the case of TiO2 NPs, two different crystal structures of titanium dioxide can be found, anastase and rutile, and

111

despite their similar core, the hazard potential may be different (Hartmann et al., 2009). It has been reported in the literature that these differences in crystallinity may result in comparative differences in the potencies of pulmonary inflammatory and cytotoxic endpoints, ranging from benign to more moderate health impacts (Warheit et al., 2007). On the other hand, in the case of CNTs, it has been suggested that the small variations and defects in CNT surface, morphology, and physico-chemical features might modulate their toxicity. Mu¨ller and co-workers synthesized defect-free ground multiwall CNTs by heating at 2400 1C the material and structural defect-induced CNTs by grinding the material that had been heated at 2400 1C. They confirmed the presence of imperfections in the carbon framework by Raman spectroscopy. The role of the abundance of defects was also confirmed by microcalorimetry as the heat of adsorption of water vapor showed that heating CNTs at high temperature increased hydrophobicity and fully eliminated hydrophilic sites. Grinding of the material led to the creation of sites capable of interaction with water molecules, suggesting the formation of defects at the surface. These surface defects are subject to oxidation or the opening of the internal pores where water may condense. The CNTs were administered intratracheally (2 mg/rat) to Wistar rats to evaluate the long-term (60 days) lung response. The results show that the acute pulmonary toxicity and the genotoxicity of CNT were reduced in the case on defect-free CNTs, indicating that the intrinsic toxicity of CNT is mainly mediated by the presence of defective sites in their carbon framework (Mu¨ller et al., 2008). In general, the toxicological effects of different NPs may be attributed to their surface properties that originate from the specific nanosize but are also determined by chemical compositions. Shrinkage in particle size may create discontinuous crystal planes that increase the number of structural defects besides disrupting the well-structured electronic configuration of the material, so as to give rise to altered electronic properties on the particle surface (Oberdo¨rster et al., 2005; Donaldson and Tran, 2002). This could establish specific surface groups that could function as reactive sites. Surface groups can make NPs hydrophilic or hydrophobic, lipophilic or lipophobic, or catalytically active or passive. The extent of these changes and their importance strongly depends on the chemical composition of the material (Nel et al., 2006).

3.05.7 Environmental Fate and Behavior of Natural Colloids Although the fate and behavior of natural colloids and manufactured NPs are interlinked, this section reviews mainly the fate and behavior of natural colloids in the environment. Nonetheless, some of the references used are on manufactured NPs or model colloidal particles within the NP size range.

3.05.7.1 Aggregation/Disaggregation Colloidal aggregation and disaggregation is one of the most important processes occurring in the natural environment. It controls the fate and behavior and of natural colloids, including their transport and sedimentation. In addition, it

112

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

controls the behavior of pollutants including their fate and behavior, transport, bioavailability, and toxicity. In the absence of surface coating (NOM or surfactants), aggregation/ disaggregation is mainly governed by particle size, x potential, and solution ionic strength as described by DLVO theory. However, this is only the case in well-controlled laboratory experiments and rarely occurs in natural environmental systems. In the natural environment, the situation is more complicated due to the presence of different types of cations and NOM. In these systems, aggregation of colloids is governed by several mechanisms, including electrostatic (i.e., charge) stabilization (see Section 3.05.5.6), steric (i.e., nanoscale surface film formation) stabilization (see Section 3.05.5.3), charge enhancement by NOM, charge neutralization by ionic strength or specifically by binding cations such as Ca, bridging (see Section 3.05.5.6) by fibrils and aggregated NOM. NOM molecules such as fulvic and humic acids (HAs) can enhance colloidal stability (Wilkinson et al., 1997a) by a mechanism known as steric stabilization in addition to enhancing colloid surface charge (Sander et al., 2004). However, in the presence of high concentration of divalent cations such as Ca2þ, HS can enhance aggregation via bridging mechanisms (Chen and Elimelech, 2007). The net effect will depend on surface coverage and the degree of charge alteration. For model compounds, it has been shown that the adsorption of negatively charged HSs to positively charged iron oxide will result in destabilization only for low surface coverage (Stumm, 1992; Ferretti et al., 1997; Baalousha et al., 2008). Other NOM molecules such as polysaccharides can induce aggregation by a bridging mechanism (Filella et al., 1993; Wilkinson et al., 1997a). The adsorption of small quantities of the polymer leads to colloidal aggregation by charge neutralization or colloid bridging, whereas the adsorption of larger quantities is thought to stabilize the colloidal suspension via steric stabilization mechanism. Several mechanisms may take place together resulting in enhanced aggregation. For instance, alginate-coated hematite NPs aggregate through electrostatic destabilization in the

500 nm

(a)

presence of monovalent cations (Naþ) according to DLVO theory. However, in the presence of CaCl2, aggregation increased more than that can be explained by the DLVO theory, and was explained by the formation of an alginate-coated hematite gel network and the cross-linking between unadsorbed alginate as shown in Figure 9, via Ca2þ bridging, that might form bridges between hematite–alginate gel structures (Chen et al., 2006). Microscopy analysis of freshwater colloids gives insight into these processes. Microscopy analysis often shows small inorganic colloids embedded in networks of fibrillar materials (see Figure 2(a)). Interaction of inorganic colloids with biopolymers is likely due to the minimal electrostatic repulsion because of low surface charge density of biopolymers (see Table 2). In such a situation, highly stable colloidal suspension might produce large aggregates in the presence of biopolymers. As biopolymers are very long in comparison with the colloid diameter, they can serve as long bridges between colloids. The attached colloid may interact with another polymer, leading to the formation of loose aggregate networks extending to large dimension. Further, HSs may aggregate as small spheroids along the fibril of biopolymers (Buffle and Leppard, 1995a), suggesting that HSs might interact with fibrils similarly to inorganic colloids. A full picture of the aggregation behavior of colloids in natural systems can be understood from microscopy analysis of natural colloids (Figures 2(a) and 2(b)), and is depicted in Figure 10. Disaggregation (i.e., breakage of colloidal aggregates) is as important as aggregation in determining colloidal size distribution, fate and behavior, and interaction with trace contaminants, but few studies are available on the disaggregation of model colloids or natural colloidal particles. Most disaggregation studies have concentrated on the effect of shear force (Newman and Stolzenbach, 1996; Bergendahl and Grasso, 1998). However, effect of solution conditions on disaggregation has rarely been considered. Altering solution conditions by dilution, changing pH, ionic strength, altering surface charge and chemical composition, may induce particle disaggregation or alter their aggregate structure (fractal

100 nm

(b)

Figure 9 Combined hematite-alginate gel aggregate in the presence of 6.1 mM CaCl2 at pH 5.2. (a) and (b) are TEM images of the same aggregate, but at different magnifications. From Chen KL, Mylon SE, and Elimelech M (2006) Aggregation kinetics of alginate-coated hematite nanoparticles in monovalent and divalent electrolytes. Environmental Science and Technology 40(5): 1516–1523.

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

113

Small aggregates (stable suspension)

Gels

Large aggregate (unstable suspension)

Figure 10 Major types of aggregates formed in the three-colloidal component system: fulvic acid, small points; inorganic colloids, circles; rigid biopolymers, lines. Both fulvic acids and polysaccharides can also form gels, which are represented here as gray areas into which inorganic colloids can be embedded. From Buffle J, Wilkinson KJ, Stoll S, Filella M, and Zhang J (1998) A generalized description of aquatic colloidal interactions: The three-colloidal component approach. Environmental Science and Technology 32(19): 2887–2899.

dimension, see Section 3.05.7.2). For instance, synthetic polymers have been shown to be able to separate two aggregated colloids, even when the separation distance was on the order of few nanometers (primary minimum) (Ouali and Pefferkorn, 1994). Disaggregation of NOM (peat HA) was observed after dilution of a peat concentrate, and disaggregation rate increased with pH (Avena and Wilkinson, 2002). Recently, NOM (Suwannee River HA) has been shown to induce the disaggregation of iron oxide aggregates of NPs, likely due to formation of surface coating of NOM on the surface and pore surfaces of aggregates and thus the enhancement of surface charge as confirmed by electrophoretic mobility measurements (Baalousha, 2009). This induces an increase of the degree of repulsion within the aggregate matrix and results in aggregate rupture. There are two possible mechanisms of aggregate breakup based on aggregate structure: slow surface erosion and fast large-scale fragmentation. In surface erosion, small particles are separated from the surface of the aggregate, whereas in large-scale fragmentation, the aggregates split into pieces of comparable sizes (Jarvis et al., 2005). Disaggregation rate depends on aggregation mechanism, aggregate structure, presence of NOM, and solution composition. For instance, diffusion-limited aggregation mechanism results in the formation of highly branched aggregates with small fractal dimension that break up by fragmentation mechanism. Reaction-limited aggregation mechanism results in the formation of compact aggregates with large fractal dimension that favor breakup by surface erosion mechanism (Yeung and Pelton, 1996). Disaggregation depends also on the way aggregates interact with NOM, which may develop in two steps. The first fast step corresponds to coverage of the aggregate surface by NOM, and a second slower step corresponds to the diffusion of NOM through the already adsorbed layer and the reptation of HA into zones near to neighboring interfaces, that

is, it aggregates pores (Baalousha, 2009). It has been shown that the adsorption of polymer (polyvinylpyridine) to colloidal particle surfaces induced aggregate fragmentation after an initial lag time, which is the time required for polymer reptation within the porous matrix of aggregates and sorption to the surface of the particles (Ouali and Pefferkorn, 1994; Pefferkorn, 1995). Solution chemical composition including ionic strength and type of cations present in solution are also expected to influence disaggregation of colloidal aggregates, with faster disaggregation rate in the presence of monovalent cations and slower disaggregation rate in the presence of divalent cations. Clearly, more research is needed to investigate the possible disaggregation of natural environmental aggregates and the role played by water composition (ionic strength, cations, pH, and natural organic molecules).

3.05.7.2 Aggregate Structure and Fractal Dimension Aggregation may occur in two different modes: diffusion limited aggregation (DLA) and reaction limited aggregation (RLA). In DLA, particles are assumed to have no surface repulsion and aggregation occurs due to Brownian motion of particles. In RLA, repulsive forces due to electrostatic or other interactions may prevent the particles from aggregating and only a fraction of collisions are successful. These modes of aggregation result in formation of aggregates with different structures with open porous aggregates in the case of DLA and more compact aggregates in RLA mode (Figure 11). In low ionic strength such as freshwater systems, compact aggregates with RLA-type aggregate structures (Figure 11(b)) are likely to dominate due to the low collision efficiency. However, in high ionic strength such as marine systems, loose aggregates with a DLA-type structure (Figure 11(a)) are more likely to dominate due to the higher collision efficiency

114

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

100 nm

100 nm (a)

(b) 1

Figure 11 TEM micrographs of iron oxide NPs (100 mg l Fe) at pH 6 (a) without HA and (b) with HA (5 mg l1) showing two different aggregation modes namely reaction limited aggregation in (a) and diffusion limited aggregation in (b). Adapted from figure 4 in Baalousha M, Manciulea A, Cumberland S, Kendall K, and Lead JR (2008) Aggregation and surface properties of iron oxide nanoparticles; influence of pH and natural organic matter. Environmental Toxicology and Chemistry 27: 1875–1882.

500 nm

500 nm

500 nm

(a)

(b)

(c)

500 nm

(d)

500 nm

(e)

Figure 12 Variation of the texture of DOM along the Adour estuary with salinity (a) 0, (b) 0.1, (c) 5.2, (d) 21.7, and (e) 23. From figure 4 in Baalousha M, Motelica-Heino M, and Coustumer P (2006) Conformation and size of humic substances: Effects of major cation concentration and type, pH, salinity and residence time. Colloids and Surfaces A: Physicochemical and Engineering Aspects 272: 48–55.

(Leppard et al., 1986, 1997; Wilkinson et al., 1999). In the presence of high ionic strength, HSs aggregate to small spheroids of about 10 nm or large, porous aggregates of several micrometers (Baalousha et al., 2005c, 2006b). However, the structure of the aggregates also depends on the interaction (residence time). The highly porous aggregates formed at high ionic strength in DLA mode (Figures 12(a)–12(c)) may become more compact (Figure 12(e)) over time due to neutralization of the remaining internal (within the aggregate structure) surface charge (Baalousha et al., 2006c). The structure of natural colloidal aggregates can be described by a parameter known as fractal dimension (Senesi and Wilkinson, 2008). A fractal object has a self-similar structure at all levels of magnification, that is, it can be subdivided into parts, each of which is a reduced-size copy to the whole structure. The fractal dimension can be described by a geometric power law scaling each dimensional geometry (volume (v) or mass (m) for three dimensions D3, projected area (A) for two dimensions D2, or perimeter (P) for one dimension (D1), and characteristic length scales (L) of the aggregate (Lee and Kramer, 2004)). D1 provides information about the morphology of the aggregate related to the irregularity of the aggregate boundary or perimeter, D2 provides information about the projected area of an aggregate, and D3

provides information about the mass distribution within the aggregate:

m or vp L D3

Ap L D2

Pp L D1

ð9Þ

A summary of studies that applied the concept of fractal dimension to environmental colloidal particles and the technique used is given elsewhere (Filella, 2006). Although scattered, the fractal dimension values reflect the aggregation mechanisms; values of D3 of 1.6–1.9 indicate a DLA (Figure 11(a), D2 ¼1.7870.06, D3 ¼1.8770.06), while values B2.1–2.3 indicate an RLA (Figure 11(b), D2 ¼1.9570.01, D3 ¼ 2.0670.02). Majority of studies in the literature have applied the concept of fractal dimension to synthetic particles such as iron oxide, goethite (Hackley and Anderson, 1989), hematite (Amal et al., 1992; Zhang and Buffle, 1996), mon¨ sterberg tmorilonite, or fractionated organic compounds (O and Mortensen, 1992; Rice and Lin, 1993; Senesi et al., 1996; Senesi et al., 1997; Rice et al., 1999; Chakraborti et al., 2003). Few studies have applied the concept of fractal dimension to nonfractionated environmental samples such as fluvial particulate matter (Lartiges et al., 2001), colloids from natural aquatic colloid, for example, river and river bed sediment and agricultural field drainage (Jarvie and King, 2007), marine

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

snow, aggregates formed in mesocosm diatom bloom, estuarine and marine suspended particles, or biological aggregates in wastewater treatment plants. Besides experimental studies, computer simulation has been proved useful to understand colloidal aggregation, including the effect of physico-chemical properties, the effect of NOM, the aggregation mode, and the structural properties (fractal dimension) of the formed aggregates (Stoll and Pefferkorn, 1996; Stoll and Buffle, 1998). Fractal dimension is an important factor in understanding aggregate fate and behavior (see Section 3.05.7.3), and their interaction with other environmental components, for example, contaminants and nutrients. Adsorption/desorption hysteresis of contaminants to fractal aggregates can be explained by the blockage of the pores within the aggregates once sorption takes place, that is, variation in their fractal dimension (Cheng et al., 2004). Aggregate structure also influences disaggregation rate (see Section 3.05.7.1).

3.05.7.3 Transport and Sedimentation in Aquatic Media The settling behavior of a hard, nonpermeable sphere can be described in a relatively straightforward manner by Stokes’ law. However, aggregation of colloidal particles in natural waters results in the formation of large, fractal, and permeable aggregates (see Section 3.05.7.2) (Johnson et al., 1996; Lartiges et al., 2001). Thus, Stokes’ law is not suitable to describe the settling behavior of natural colloidal aggregate. The settling behavior of such aggregates depends on a drag force and permeability of solvent through the porous aggregates. Pores formed within the fractal aggregate will permit greater interior flow through the aggregate, resulting in a faster settling velocity. It has been demonstrated that fractal aggregates (with heterogeneous pore sizes) settle faster than predicted by Stokes’ law for impermeable spheres or permeable sphere models that specified aggregate permeability for a homogenous distribution of particles with aggregates (Logan and Hunt, 1987; Johnson et al., 1996), indicating that intra-aggregate flow reduces the drag for aggregates compared to that for the equivalent impermeable particles. As the fractal dimension increases, the permeability decreases and the fluid mechanics resembles more closely that of an impermeable sphere (Chellam and Wiesner, 1993). Nevertheless, the settling behavior of fractal aggregates is not well understood. The settling behavior of fractal aggregates depends on many properties, including porosity, size, permeability, and buoyant density, which need to be determined to predict fractal aggregate sedimentation. Several models have been developed to predict the sedimentation behavior of fractal aggregates (Tang et al., 2002; Tang and Raper, 2002). However, it is difficult to describe mathematically the nonhomogeneous distribution of pores within fractal aggregates, although this nonhomogeneous porosity has been expressed by assuming that the porosity of fractal aggregates varies radially from the center of gyration (Veerapaneni and Wiesner, 1996). Models for determining porosity, drag coefficient, and settling velocities of fractal aggregates are reviewed elsewhere (Tang and Raper, 2002), where it was concluded that, although several models have been presented to calculate the previous parameters, no single model can best describe the

115

settling behavior of fractal aggregates. Thus, the use of any of these models generally for all types of fractal aggregates still needs to be confirmed.

3.05.7.4 Transport in Porous Media Colloid-facilitated transport is a very well-known process in porous media; mobile colloidal particles may act as carriers of strongly sorbing contaminants in subsurface materials (Kretzschmar et al., 1999; Grolimund and Borkovec, 2005). Understanding the processes (depicted in Figure 13) that control colloid transport in porous media, besides understanding their reactivity and contaminant binding (see Section 3.05.5.5), is essential to efficiently manage and remediate many environmental contaminants. It is also essential to understand the transport of microbial particles (e.g., bacteria and viruses) in the natural environment. Colloids are affected by many of the physical and chemical processes that influence solute transport such as advection, diffusion, dispersion, and sorption and desorption (known as attachment and detachment in colloid literature). Thus, colloid transport in porous media is, among other processes, governed by their physico-chemical properties (e.g., size, shape, and surface properties), the physico-chemical properties of the porous medium (e.g., grain size, surface properties), and the fluid properties (e.g., velocity, ionic composition, presence of NOM, density, and viscosity) (Kretzschmar et al., 1995; Bradford et al., 2002; Grolimund and Borkovec, 2005; Ahfir et al., 2007). Colloid transport/deposition in porous media can be thought of as occurring in two steps (McDowell-Boyer et al., 1986): 1. transport to the vicinity of the soil or sediment grains themselves (collector) by surface filtration, straining, diffusion, or physical–chemical mechanisms leading to collision and 2. attachment to the collector via electrostatic interactions between the colloid and the soil. There are several restrictions to movement of colloids through soil, including: (1) straining (some times called physical restriction or mechanical filtration) and (2) true filtration (McDowell-Boyer et al., 1986; McGechan, 2002). Straining is the trapping of colloid particles in the down-gradient pore throats that are too small to allow particle passage. Colloid retention by straining depends on both colloid and porous media properties. Complete straining (mechanical filtration) occurs when colloids are large enough to be physically excluded from entering all the soil pores, resulting in the formation of a filter cake or surface mat of colloids above the media. Incomplete straining occurs when size distribution of colloids is smaller than some of the medium pores; colloid transport may occur in the larger pore sizes, and colloid retention occurs in the smaller pore sizes. Straining may occur even for particles much smaller than the average grain size of narrowly distributed grains. Under strongly repulsive conditions with no physico-chemical filtration, straining filtration to become negligible, the ratio of grain size to colloid size must be larger than 125 (Xu et al., 2006).

116

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

Translocation

Dissolution

leachates from vadoze zone seapage

Deposition

Mobilization ionic strength pH ionic composition hydraulic effects Is↓

pH?

2+

Ca

+

→ Na

Stabilization transport

Generation Secondary mineral formation Precipitation from over-saturation Dissolution of cements Filtration

Bacterial growth Eh↓ Eh ↓

pH?

pCO2 ↑2↑

Figure 13 Schematic plot of important processes influencing colloid behavior in the subsurface environment. Mobilization usually takes place when double layers expand or by changes of the surface charge (polarity þ /  to þ / þ or more often  /  ), hydrodynamic forces usually play a less important role for colloids. Generation occurs when new colloids are produced by precipitation from supersaturation or by dissolution of cements which contain colloidal particles (as carbonates or oxides) through changes in surrounding conditions as decrease in pH or redox potential or increase in the dissolved CO2. Removal of colloids is associated with dissolution of particles, their deposition onto the immobile matrix straining filtration in the pores. From v. d. Kammer F (2005) Characterization of Environmental Colloids applying Field-Flow Fractionation – Multi Detection Analysis with Emphasis on Light Scattering Techniques. Hamburg, Germany: Hamburg University of Technology.

True filtration covers a range of mechanisms with a common feature that particle dimensions are much smaller than the pores. Filtration mechanisms include diffusion, interception, and sedimentation. Diffusion is caused by the bombardment by water molecules undergoing Brownian motion; hence, it strongly depends on colloid size. Small colloids (o100 nm) in particular are deposited by collision to the porous media due to diffusion. Interception occurs when the particle passes closer than one particle radius from the collector surface and electrostatic forces come into play. This process is especially important for colloids/aggregates larger than 1 mm. Sedimentation of colloids is due to a difference in colloid to fluid density (mainly for larger (4200 nm) or dense colloids) (McDowell-Boyer et al., 1986). These three processes relate to the three main aggregation processes: perikinetic (diffusion), orthokinetic (shear), and differential settling. Small particles are often removed more efficiently by diffusive transport, whereas larger particles are often removed more efficiently by sedimentation and interception (Bradford et al., 2002).

Attachment is the removal of colloids from solution via collision and fixation to the porous media, and is typically assumed to be the primary process controlling colloids transport in porous media. Attachment depends on particle–particle, particle–solvent, and particle–porous media interactions including double layer, London–van der Waals, hydrodynamic, hydration, hydrophobic, and steric interactions (see discussion in Section 3.05.5.6). Colloid-attachment kinetics is controlled by the rate of transport to the solid surfaces and subsequent fixation to these surfaces. While the maximum size of mobile colloids will be limited by straining and pore velocity, concentration and size distribution of smaller colloids are controlled by physico-chemical filtration. Altogether, subsurface transport of small (1– 100 nm) colloids is limited mainly by the diffusion-driven collision rate, whereas the transport of larger colloids or aggregates (41 mm) is limited by straining (if colloid density equals fluid density) or sedimentation (for colloid density 4 fluid density).

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

Besides colloid and porous media properties, solution physico-chemistry (e.g., pH, ionic strength, NOM) also affects colloidal transport via aggregation/disaggregation and effective pore size distribution. For unsaturated porous media, colloid retention at the stationary, mobile, or transitional gas–water interface has to be taken into account when considering colloidal transport (Ouyang et al., 1996; Wan and Tokunaga, 1997; Lenhart and Saiers, 2002; Crist et al., 2004; Bridge et al., 2009).

3.05.8 Environmental Fate and Behavior of Nanomaterials The potential fate and behavior of nanomaterials in the environment is not yet well understood, and available studies are scarce. Determining the fate and behavior of nanomaterials in the environment requires understanding potential sources of nanomaterials, their fate in air, soil, and water, their transformation, degradation, and persistency. In addition, the fate of nanomaterials in the environment is likely to vary with the physical and chemical characteristics of the nanomaterials and the containing medium and with the interaction of nanomaterials and other environmental contaminants. Understanding the processes that control the fate and behavior of natural colloids (NPs) in water and soil (Section 3.05.7) and ultrafine particles in air (not presented here) will improve our understanding and ability to predict the fate and behavior of manufactured nanomaterials in these systems.

3.05.8.1 Exposure/Release of NPs According to The Nanotechnology Consumer Products Inventory, the most common material mentioned in the product descriptions was silver (259 products). Carbon was the second most referenced (82 products) which included fullerenes and nanotubes, followed by titanium dioxide (31), zinc oxide (24), silica (15), and cerium oxide (1). Among potential environmental applications of NPs, remediation of contaminated groundwater with nanoscale iron is one of the most wellknown examples (Tratnyek and Johnson, 2006). Regarding personal-care products, NPs of titanium dioxide and zinc oxide are included in toothpaste, beauty products, sunscreens (Serpone et al., 2007), and textiles (Yuranova et al., 2007). Metal oxide-based NPs are also increasingly used in fillers, opacifiers, ceramics, coatings, catalysts, semiconductors, microelectronics, prosthetic implants, and drug carriers (Reijnders, 2006). Photocatalytic properties of TiO2 may also be used for solar-driven self-cleaning coatings (Cai et al., 2006). Knowing the heavy manufacturing and use of NPs in commercially available products, it is therefore correct to assume that the household waste in the case of the metal oxide NPs is likely to end up in natural water bodies. For perspective on potential nanopollution, one may consider that 2 g of 100-nm-size NPs contain enough material to provide every human worldwide with 300 000 particles each (Hardman, 2006). As it was previously mentioned, decrease in particle size changes the physicochemical and structural properties of particles and in the case of NPs that is responsible for increased bioavailability and toxic effects (Nel et al., 2006). Due to the current commercial

117

development of nanotechnology, the occupational and public exposure to NPs via inhalation, dermal absorption, and gastrointestinal tract absorption is supposed to increase dramatically in the coming years as well as their potential release in the environment. Because of their unique properties, including small size and corresponding large specific surface area, NPs are supposed to impose different degrees of biological effects from their corresponding analog micro-scaled materials (Oberdo¨rster et al., 2005; Nel et al., 2006). The extent of exposure and resulting adverse effects on human health and the environment remain mostly unknown (Jeng and Swanson, 2006; Lin et al., 2009). Thus, the studies on safety and (eco)toxicity of NPs are of extreme importance in order to support the sustainable development of nanotechnology. Gottschalk et al. (2009b) recently reported predictions of environmental concentrations of nano-TiO2, nano-ZnO, nano-Ag, CNT, and fullerenes for all environmental compartments (including sediments) of US, Europe, and Switzerland. The environmental concentrations were calculated by a probabilistic material flow modeling and compared to already published data from ecotoxicological studies (Gottschalk et al., 2009a). The most frequent values predicted by the model reported range from 0.003 (fullerenes) to 21 ng l1 (nano-TiO2) for surface waters and from 4 (fullerenes) to 4 mg l1 (nano-TiO2) for sewage treatment effluents (up to 16 mg l1 for TiO2 in surface waters, but may be larger in peak flows, where waters not always treated). For Europe and the US, the annual increase of manufactured NPs on sludge-treated soil ranges from 1 ng kg1 for fullerenes to 89 mg kg1 for nano-TiO2 (Gottschalk et al., 2009b). Blaser et al. (2008) calculated total Ag concentrations in surface waters which were by a factor 10–100 higher than the simulation results for nano-Ag reported by Gottschalk et al. and they also concluded that nano-Ag contributes only 1–15% to the total Ag into the environment. Kiser et al. (2009) have carried out measurements of Ti in sewage treatment plant sludge and reported concentrations ranging from 1 to 6 g Ti/ kg. These first measurements of manufactured NPs in the environment show concentrations in the same order of magnitude to Gottschalk et al. modeling estimates. The latter allows a first validation of the model to be reported in the literature. The results of Gottschalk et al. simulations indicated that risks to aquatic organisms may currently emanate from nano-Ag, nano-TiO2, and nano-ZnO in sewage treatment effluents for all considered regions and for nano-Ag in surface waters (Gottschalk et al., 2009b).

3.05.8.2 Fate in Water The potential fate and behavior of engineered nanomaterials, once they are released into the aquatic environment, can be understood in the light of the existing knowledge on the fate and behavior of natural colloidal particles (see Section 3.05.7). This knowledge suggests that the fate of nanomaterials in the aquatic environment can be influenced by variety of processes such as dispersion/diffusion, aggregation and disaggregation, interaction between NPs and natural water components, sedimentation, biotic and abiotic degradation, transformation and photoreaction/light. These reactions may alter the physical and chemical properties of nanomaterials and so alter their behavior in the aquatic environment.

118

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

There have been few studies on the aqueous stability and aggregation of nanomaterials under environmentally relevant conditions. Brant et al. (2005) studied the aggregation and deposition of fullerene NPs in aqueous media at variable ionic strength. They found that while in the absence of electrolytes nC60 stayed stable over time, 0.001 M solution ionic strength (NaCl) was enough to destabilize the nC60 by screening their electrostatic charge and produce large aggregates that settle rapidly (Brant et al., 2005). The addition of HA has been shown to enhance the stability of fullerene suspension in the presence of NaCl and MgCl2 and low concentrations of CaCl2 (Chen and Elimelech, 2006). However, at high concentrations (above 10 mM) of CaCl2, enhanced aggregation of fullerene NPs was observed due to bridging mechanism by HA aggregates (Chen and Elimelech, 2007). Other research has found similar complex interactions between natural and manufactured nanomaterials (Baalousha et al., 2008; Giasuddin et al., 2007; Hyung et al., 2007). Extracted Suwannee River HA (SRHA) and natural surface water (actual Suwannee River water with unaltered NOM background) have been shown to stabilize MWCNT (Hyung et al., 2007). However, extensive flocculation of SWCNT (i.e., formation of floating aggregates and partial sedimentation of other aggregates) was observed when mixed with natural waters from a lake, presumably due to the high ionic strength and the presence of divalent cations such as Ca. Apparently, sorption of HSs enhances the stability and inhibits the aggregation of CNTs to a certain extent (Hyung et al., 2007). However, cations, particularly divalent cations such as Ca and Mg, reduce the stability of CNTs in the absence or presence of NOM surface coating. Disaggregation is as important as aggregation processes in determining the fate and behavior of nanomaterials, though few studies are available (Ouali and Pefferkorn, 1994; Baalousha, 2009). NOM has been shown to induce the disaggregation of iron oxide NP aggregates (5–10 mm) formed at pH 7, likely due to formation of surface coating of NOM on the surface and pore surface of the aggregates and thus the enhancement of surface charge (Baalousha, 2009). In addition, it has been shown that certain polymers are able to disaggregate latex particle (885 nm in diameter) aggregates (Ouali and Pefferkorn, 1994). However, polysaccharide or HA did not result in the disaggregation of polystyrene latex particle aggregates which was explained by the existence of strong interparticle forces within flocs which prohibited aggregate breakup upon adsorption of NOM (Walker and Bob, 2001). Fabrega et al. (2009b) found that SRHAs can cause partial disaggregation of AgNP aggregates by nanoscale film formation, although such disaggregation and film formation decreased short-term bacterial toxicity.

3.05.8.3 Fate in Wastewater NPs used in different applications can be released by material degradation or erosion and these NPs may find their way into wastewater treatment facilities and may end up either in the wastewater sludge or in the water effluent. Nonetheless, the fate and behavior of NPs, the impact they might have on wastewater treatment, and the impact that wastewater has on NPs are largely unknown and need further investigation. For

an overview of the potential behavior of NPs in different compartments of a wastewater treatment plant, the reader is referred to the review by Brar et al. (2010). Generally, NPs fate and behavior in wastewater treatment facilities are likely to be governed by processes such as aggregation/disaggregation, adsorption to colloidal or micron particles in the wastewater, and sedimentation. In a study on the behavior of model NP (cerium oxide) in a model wastewater treatment plant, Limach et al. (2008) found that the majority of the NPs could be captured through adhesion to cleaning sludge. Nonetheless, they found that a significant fraction of the NPs escaped the wastewater plant clearing system and up to 6 wt.% of cerium oxide was found in the effluent stream (Limbach et al., 2008). Another study on the removal of TiO2 in wastewater treatment plant suggests the removal of majority of the particles which was accumulated in the settled solids. Nonetheless, a small fraction in the effluent which was mainly in the size range o0.7 mm (Kiser et al., 2009). The sedimentation of NPs in sludge in wastewater treatment plant may result in the release of NPs into soil and subsequently into groundwater (Benn and Westerhoff, 2008). Surface coating and surface charge will potentially play an important role in their behavior in wastewater systems (Limbach et al., 2008). Uncoated NPs are likely to sediment and form part of the waste sludge as they are borne to aggregation. However, coated or functionalized NPs might be partitioned between the water effluent and waste sludge due to their inherent stability induced by the surface coating. In both cases, NPs need to be removed from both compartments to prevent further pollution. On the other hand, interaction of NPs with microorganisms might potentially inhibit activated sludge process, a major process in wastewater treatment, which may result in jeopardizing water treatment plant (Brar et al., 2010). Several studies have suggested different types of NPs, including silver, iron, ZnO, CuO, La2O3, SnO2, TiO2, CNTs, nC60, and other NPs are toxic to bacteria (Kang et al., 2007; Pal et al., 2007; Auffan et al., 2008; Hu et al., 2009; Fabrega et al., 2009a), and to biofilm (Fabrega et al., 2009b).

3.05.8.4 Fate in Soil The potential fate and behavior of engineered nanomaterials in soil can be understood in the light of the existent knowledge on the fate and behavior of natural colloidal particles (see Section 3.05.7.4). Nanomaterials are small enough to travel through soil pores. However, they can be sorbed to soil particles due to their high surface area, and therefore become immobilized. In addition, the formation of large aggregates of nanomaterials can immobilize them by filtration, sedimentation, or straining in smaller pores. At the moment, little information is available on the transport and fate of nanomaterials in the natural porous environment. However, some data are available from laboratory column studies using porous media (Lecoanet and Wiesner, 2004; Lecoanet et al., 2004; Schrick et al., 2004; Li et al., 2006; Yang et al., 2007), which suggests that transport is often relatively rapid and depends on the type of nanomaterials. Laboratory soil column experiment on iron oxide and zerovalent iron NPs shows that their mobility is more limited due

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

to the efficient filtration mechanisms of aquifer material. Field studies on iron oxide NP indicate that they may migrate only few centimeters to few meters from the point of injection and that their mobility depends on many factors such as particle size, solution pH, ionic strength, soil composition, and groundwater flow velocity (Schrick et al., 2004; Li et al., 2006). The zero-valent NPs are somewhat more mobile as they have been synthesized on supports acting as a delivery vehicle (Schrick et al., 2004; Yang et al., 2007). These delivery vehicles, including anionic hydrophilic carbon and poly(acrylic acid) (PAA), bind strongly to the iron, create highly negative surfaces, thus effectively reducing the aggregation among zerovalent iron particles and reducing the filtration removal by aquifer materials. Laboratory soil column experiments with Fe/hydrophilic carbon, Fe/PAA, and unsupported iron NPs suggest that the anionic surface charges can enhance the transport of iron NP through soil and sand packed columns in comparison with unsupported iron NPs (Schrick et al., 2004; Yang et al., 2007). In addition, the transport of iron NPs (2–10 nm) through porous media column (glass beads, unbaked and baked sand) can be highly enhanced by surface modification via surfactant sorption. Unmodified iron NPs were immobile and aggregated on porous media surfaces in the column inlet area (Kanel et al., 2007). Although surfactants and polymers enhance the transport of NPs, the role of NOM in NP-facilitated transport has not yet been investigated, but likely to be important. Further, the characteristics of the soil matrix may influence the diffusion and transport of NPs. PAA-modified nanoiron slurry has been found to travel easily through silica sand columns, but not loamy sand soil columns (Yang et al., 2007). The transport of water-stable nC60 aggregates underivatized C60 crystalline NPs, stable in water for months through a soil column, was investigated at different flow rates, while other column operating parameters remained fixed through all the experiments. The nC60 particles were observed to be more mobile at higher flow velocity due to less interaction time between the nC60 particles and the porous media (Cheng et al., 2005). Lecoanet and Wiesner (2004) studied the transport and removal of silica, anatase, and fullerene-based NPs in porous media. They found that the removal of anatase is less significant at higher flow rates. However, no dependence on velocity of particle passage through the porous medium was observed for silica particles and the fullerene-based NPs. This was explained by the very small value of the collision efficiency factor in the case of silica and the deposition of fullerene-based NPs on the porous media at higher flow rates after 1 void volume, which limits the interaction of these NPs with the porous media and reduces particles removal afterwards (Lecoanet and Wiesner, 2004). The discussion above, in addition to previous knowledge on colloidal transport in porous media, suggests that the mobility of NPs in soils depends on (1) NP physical–chemical characteristics, that is, size, shape, surface coatings, and stability; (2) the properties of the soil and environment, that is, clay, sand, colloids, NOM, water chemistry, and flow rates; and (3) the interaction of NPs with natural colloidal material, that is, surface coating, aggregation/disaggregation, and sorption to larger particles. Chapter Chapter 1.03 Managing Aquatic Ecosystems discusses the more theoretical aspects of particle movement in porous media.

119

3.05.9 Conclusions and Recommendations This chapter has reviewed both colloids in natural aquatic and terrestrial systems and manufactured NPs in the same environmental systems. It is clear that similar physical and chemical properties dominate their behavior in the environment and that natural materials are a useful analog for understanding the highly topical area of NP environmental risk. However, some caution is required in making this link as NPs show many differences to environmental colloids such as tuneable properties, less complexity, and low but increasing concentrations. Clearly, there is also a benefit from two-way information flow and understanding of relatively well-defined manufactured NPs will inform our understanding of the natural colloidal and nanoparticulate chemistry and behavior. In particular, the following issues need to be addressed.

3.05.9.1 Environmental Fate and Behavior The existing literature on natural colloids suggests that their environmental fate and behavior are controlled by properties and processes such as surface properties, surface film formation, aggregation/disaggregation, and transport and sedimentation. This literature is essential in order to understand the likely fate and behavior of manufactured NPs, once released to the environment, as similar processes are expected to control their fate and behavior. Although advances have been made in understanding the fate and behavior of colloids and their role in environmental systems in the last few decades, much is still unknown due to the intrinsic complexity of natural colloids, the lack of appropriate experimental techniques, and the significant gap between theories and models, which were developed to describe well-controlled, simple laboratory systems and the reality of natural systems that contain heterogeneous mixture of particles. The development of nanoscience and the controllable synthesis of NP with different properties (e.g., size, shape, morphology, surface coating, etc.) will allow a better understanding of the fate and behavior of natural colloids through a better understanding and a systematic investigation of the properties, processes, and mechanisms controlling their fate and behavior. The synthesis of simple systems of NPs with tuneable properties will allow validating the existing theories on colloidal science and possibly developing new theories to minimize the gap between the simple systems of monodisperse spherical colloidal system and the real environmental situation.

3.05.9.2 Need for New Metrology and Analysis Tools Determining the concentration and physical and chemical properties of natural colloids and manufactured NPs in environmental and biological systems is essential to predict the environmental consequences of natural colloids and manufactured NPs. However, the diversity of nanomaterials and their properties make their identification and characterization a difficult task. In addition, the interaction of nanomaterials with the natural environmental or biological components provides an additional complexity to the system and so a significant metrological and analytical challenge. Therefore, it

120

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

is important to adopt/develop new metrological and analysis tools, in particular measuring units, characterization methods, and measurement standards. This is still at an early stage and no published data are yet available. Properties that are important for the characterization of natural colloids and manufactured NPs include, but are not limited to, concentration, size and size distribution, molar mass, surface area, state of dispersion/agglomeration, composition, structure, surface charge, oxidation state, solubility, reactivity, and stability (Buffle and Leppard, 1995b; Powers et al., 2006, 2007). The characterization of natural colloids and manufactured NPs is an extensive laborious process, demanding the use of several techniques in parallel in order to achieve a high degree of accuracy and reliability. To date, few quantitative analytical tools for measuring natural colloids and NPs in natural systems are available (see Table 4 for some of these tools), which results in a serious lack of information about their occurrence and fate and behavior in the environment, and new tools are needed to investigate both natural and manufactured NPs. So far, there is an important debate on the best metrics (size, number, and surface area) to report concentration of NPs (Oberdo¨rster et al., 2005; Warheit et al., 2006; Wittmaack, 2007), though no clear conclusion is yet achieved and further research is needed. The development of structure–activity relationships (see Section 3.05.9.5) could be the best way to reach a conclusion. There is a need for certified reference materials and standards for characterization and the terminology to enable interlaboratory comparison and benchmarking in areas of natural colloids and manufactured NP characterization, behavior, toxicity, and others. A list of the available standards for manufactured NPs, which can also be applied to natural colloids, is given in Table 5. Recently, the National Institute of Standards and Technology (NIST) has issued a new reference

Table 5

standards for manufactured NPs (citrate stabilized gold particles nominally 10, 30, and 60 nm in diameter) targeted for laboratories studying the biological effects of NPs.

3.05.9.3 Understanding Complexity on the Nanoscale Although the nanoscale is a very small scale, the definitions of natural colloids (1–1000 nm) and manufactured NPs (1–100 nm) cover a relatively wide range, under which a large complexity may exist, including impurities, surface protruding, surface coating, and adsorbed materials. Such complexity inevitably alters the behavior of natural colloids and the toxicity of nanomaterials. Understanding the behavior and biological response of nanomaterials requires a good understanding of these materials at the nano, the atomic, or even, subatomic scale. This can only be achieved by combining the results of several powerful techniques (e.g., electron microscopy, AFM, and others) for studying material structure at these tiny scales. This will allow understanding how subtle nanoscale features of a material can give rise to changes in its physico-chemical and biological properties. In addition, the development of new analytical tools for nanotechnology will help understanding the variability at such a small scale, both for natural colloids and manufactured NPs.

3.05.9.4 Knowledge of Uptake and Toxicity of NPs Although the beneficial aspects of nanomaterials are well versioned and products containing nanomaterials are already in use, the potential impact of nanomaterials is still largely unclear. Several reports have suggested the negative impact of nanomaterials on the natural environment and living organisms, although more research is required to demonstrate such a negative impact and to determine the nano-properties responsible for it. So far, there is a consensus that surface

Available standards on nanotechnology and nanoparticles

Type

Standard

Description

Terminology

GB/T 19619-2004 IUPAC recommendations

Terminology for nanomaterials Nomenclature for the C60-Ih and C70-D5 h(6) fullerenes

Sizing

GB/T 13221-2004

GB/T 20307-2006 GB/T 20099-2006

Nanometer powder – determination of particle size distribution – small X-ray scattering method (ISO/TS13762) Determination of the specific surface area of solids by gas adsorption using BET methods (ISO 9277:1999) Particle sizing analysis – photon correlation spectroscopy (ISO 13321:1996) Representation of results of particle size analysis – part 2: calculation of average particle size/diameters and mements from particle size distribution (ISO 9276-2:2001, IDT) Representation of results of particle size analysis – part 2: characterization of a classification process (ISO 9276-4:2001, IDT) General rules of nanometer-scale length measurement by SEM Sample preparation dispersing procedures for powders in liquids

Nanomaterial specifications

GB/T GB/T GB/T GB/T

Nano-nickel powder Nano-zinc oxide Nano-calcium carbonate Nano-titanium dioxide

Handling/disposal

PD 6699-2:2007

GB/T 19587-2004 GB/T 19627-2005 GB/T 15445.2-2006

GB/T 15445.4-2006

19588-2004 19589-2004 19590-2004 19591-2004

Guide to safe handling and disposal of manufactured nanomaterials

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

properties (e.g., surface area, surface reactivity, atomic surfaces, redox state, surface charge, and surface coating) achieved by the reduction in particle size are responsible for their potential uptake and toxicity (Karakoti et al., 2006), though the exact nano-properties determining their toxicity are still unknown. This is largely related to the complexity of nanomaterials, the wide range of nano-properties that might be responsible for such a negative effect, the complexity of the toxicity test system, the lack of the required characterization, as well as, the lack of the analytical tools and protocols required for this purpose, and all of these aspects need further development.

3.05.9.5 Knowledge of Structure–Activity Relationships One of the key objectives of nanoscientisits is to develop benign nanomaterials, in order to allow a full benefit of the development of nanotechnology. This can only be achieved by determining the properties of nanomaterials responsible for the harmful effects and alter them when designing synthesis methods. A useful tool to achieve this goal is the structure– activity relationship. Structure–activity relationship is the process by which chemical structure is quantitatively correlated with a welldefined process such as biological (Sayes et al., 2006) or chemical (Liang et al., 2006) reactivity. For example, the biological reactivity (e.g., toxicity) can be expressed as a function of the concentration of the substance required to give a certain biological response. In addition, the physico-chemical properties or structure of the substance can be integrated into a mathematical relationship (Equation (10)) if they can be quantified by numbers. Such a relationship is called quantitative structure–activity relationship (QSAR) which can be used to predict the biological response of other materials based on their properties.

Activity ¼ f ðphysicochemical properties and= or structural propertiesÞ

ð10Þ

The development of such models is essential to understand the properties of nanomaterials controlling their behavior and toxicity and to reduce the amount of studies required to achieve these goals.

3.05.9.6 Next Generation NPs Currently nanotechnology uses primarily passive nanomaterials. However, future developments predict the increased use of the more active nanomaterials or nanostructures, for instance, in drug delivery. Four overlapping generations of nanotechnology products have been predicted to be developed in the period 2000–20: passive nanostructures, active nanostructures, systems of nanosystems, and molecular nanosystems (Roco, 2005). The first generation (after 2000) involved the basic discovery and production of passive nanostructures such as the simple components of NPs, nanotubes, nanolayers, and nanocoatings. They have steadystate structures and functions such as chemical reactivity or mechanical behavior during their usage (Renn and Roco, 2006). The second generation (B2005) involves active nanostructures that change their properties (morphology,

121

shape, mechanical, electronic, magnetic, biological, etc.) during operation such as nanobiodevices, transistors, polymerbased targeted drugs (Yih and Al-Fandi, 2006; Park, 2007), etc. The third generation (B2010 onwards) includes systems of nanosystems which might self-assemble or self-organize, networking at the nanoscale to form larger architectures (Renn and Roco, 2006) such as artificial organs and electronic devices based on state variables (electron-spin, nuclear-spin or photonic state). The fourth generation (B2015/2020) includes molecular nanosystems, where each molecule in the nanosystem has a specific structure and plays a different role. Molecular machines might be designed by atomic manipulation and may be used as devices which will approach the way biological systems work. Whatever happens in the near future, it is certainly clear that massive and rapid changes are about to be brought about and it is incumbent upon us to be aware of these changes and as a society to use them in a beneficial manner, while minimizing any attendant risks. Current environmental studies and risk assessment programs are mainly concerned with the first generation of nanomaterials or passive nanomaterials, and even here there are fundamental uncertainty arising from our lack of knowledge. Further development of nanomaterials will involve larger and more complex phenomena and problems, and much work remains to be done.

References Adair JH and Suvaci E (2000) Morphological control of particles. Current Opinion in Colloid and Interface Science 5(1–2): 160--167. Adams LDY and Alvarez PJJ (2006) Comparative eco-toxicity of nanoscale TiO2, SiO2, and ZnO water suspensions. Water Research 40(19): 3527--3532. Adams LK, Lyon DY, and Alvarez PJJ (2006) Comparative eco-toxicity of nanoscale TiO2, SiO2, and ZnO water suspensions. Water Research 40: 3527--3532. Ahfir N-D, Wang HQ, Benamar A, Alem A, Massei N, and Dupont J-P (2007) Transport and deposition of suspended particles in saturated porous media: Hydrodynamic effect. Hydrogeology Journal 15(4): 659--668. Ahmadi TS, Wang ZL, Green TC, Henglein A, and ElSayed MA (1996) Shapecontrolled synthesis of colloidal platinum nanoparticles. Science 272(5270): 1924--1926. Aime S, Cabella C, Colombatto S, Crich SG, Gianolio E, and Maggioni F (2002) Insights into the use of paramagnetic Gd(III) complexes in MR-molecular imaging investigations. Journal of Magnetic Resonance Imaging 16(4): 394--406. Aitken RJ, Chaudhry MQ, Boxall ABA, and Hull M (2006) Manufacture and use of nanomaterials: Current status in the UK and global trends. Occupational Medicine 56: 300--306. Aldahan A, Haiping Y, and Possnert G (1999) Distribution of beryllium between solution and minerals (biotite and albite) under atmospheric conditions and variable pH. Chemical Geology 156(1–4): 209--229. Alivisatos AP, Harris AL, Levinos NJ, Steigerwald ML, and Brus LE (1988) Electronic states of semiconductor clusters – homogeneous and inhomogeneous broadening of the optical-spectrum. Journal of Physical Chemistry 89(7): 4001--4011. Alowitz MJ and Scherer MM (2002) Kinetics of nitrate, nitrite, and Cr(VI) reduction by iron metal. Environmental Science and Technology 36: 299--306. Aluwihare LI, Repeta DJ, and Chen RF (1997) A major biopolymeric component to dissolved organic carbon in surface sea water. Nature 387: 166--169. Alvarez-Roman R, Naik A, Kalia YN, Guy RH, and Fessi H (2004a) Enhancement of topical delivery from biodegradable nanoparticles. Pharmaceutical Research 21(10): 1818--1825. Alvarez-Roman R, Naik A, Kalia YN, Guy RH, and Fessi H (2004b) Skin penetration and distribution of polymeric nanoparticles. Journal of Controlled Release 99(1): 53--62. Amal R, Raper JA, and Waite TD (1992) Effect of fulvic acid adsorption on the aggregation kinetics and structure of hematite particles. Journal of Colloid and Interface Science 151(1): 244--257.

122

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

Antognozzi M, Sentimenti A, and Valdre U (1997) Fabrication of nano-tips by carbon contamination in a scanning electron microscope for use in scanning probe microscopy and field emission. Microscopy Microanalysis Microstructures 8(6): 355--368. Arami H, Mazloumi M, Khalifehzadeh R, and Sadrnezhaad SK (2007) Sonochemical preparation of TiO2 nanoparticles. Materials Letters 61: 4559--4561. Arepalli S, Nikolaev P, Holmes W, and Files BS (2001) Production and measurements of individual single-wall nanotubes and small ropes of carbon. Applied Physics Letters 78: 1610--1612. Arnold T, Zorn T, Zanker H, Bernhard G, and Nitsche H (2001) Sorption behavior of U(VI) on phyllite: Experiments and modelling. Journal of Contaminant Hydrology 47(2–4): 219--231. Arugete D, Liu J, and Hochella MF (2009) Size/shape property relationships of noncarbonaceous inorganic nanoparticles and their environmental implications. In: Lead JR and Smith E (eds.) Environmental and Human Health Effects of Manufactured Nanoparticles, pp. 79--108. Chichester: Wiley. Assemi S, Hartley PG, Scales PJ, and Beckett R (2004) Investigation of adsorbed humic substances using atomic force microscopy. Colloids and Surfaces A: Physicochemical and Engineering Aspects 248(1–3): 17--23. Atwood JL, Koutsantonis GA, and Raston CL (1994) High purity fullerene-60 via molecular recognition. Nature 368: 229--231. Auffan W, Achouak M, Rose J, et al. (2008) Relation between the redox state of ironbased nanoparticles and their cytotoxicity toward Escherichia coli. Environmental Science and Technology 42(17): 6730--6735. Auffan M, Rose J, Orsiere T, et al. (2009) CeO2 nanoparticles induce DNA damage towards human dermal fibroblasts in vitro. Nanotoxicology 3(2): 161--171. Avena MJ and Wilkinson KJ (2002) Disaggregation kinetics of a peat humic acid: Mechanism and pH effects. Environmental Science and Technology 36(23): 5100--5105. Baalousha M (2009) Aggregation and disaggregation of iron oxide nanoparticles: Influence of particle concentration, pH and natural organic matter. Science of the Total Environment 407: 2093--2101. Baalousha M, Kammer F, Motelica-Heino M, Baborowski M, Hofmeister C, and Lecoustumer P (2006a) Size-based speciation of natural colloidal particles by flowfield flow fractionation, inductively coupled plasma-mass spectroscopy, and transmission electron microscopy/X-ray energy dispersive spectroscopy: Colloidstrace element interaction. Environmental Science and Technology 7: 2156--2162. Baalousha M, Kammer FV, Motelica-Heino M, and Coustumer P (2005a) 3D characterization of natural colloids by FlFFF-MALLS-TEM. Analytical and Bioanalytical Chemistry 308: 549--560. Baalousha M, Kammer FV, Motelica-Heino M, and Coustumer P (2005b) Natural sample fractionation by FlFFF-MALLS-TEM: Sample stabilization, preparation, preconcentration and fractionation. Journal of Chromatography 1093: 156--166. Baalousha M, Kammer FV, Motelica-Heino M, Hilal H, and Coustumer P (2006b) Size fractionation characterization of natural colloids by FlFFF-MALLS. Journal of Chromatography 1104: 272--281. Baalousha M and Lead JR (2007) Characterization of natural aquatic colloids (o5 nm) by flow-field flow fractionation and atomic force microscopy. Environmental Science and Technology 41: 1111--1117. Baalousha M, Lead JR, Von der Kammer F, and Hofmann TH (2009) Natural colloids and nanoparticles in aquatic and terrestrial environs. In: Lead JR and Smith E (eds.) Environmental and Human Health Effects of Nanoparticles, pp. 109--162. Chichester: Wiley. Baalousha M, Manciulea A, Cumberland S, Kendall K, and Lead JR (2008) Aggregation and surface properties of iron oxide nanoparticles: Influence of pH and natural organic matter. Environmental Toxicology and Chemistry 27: 1875--1882. Baalousha M, Motelica-Heino M, and Coustumer P (2006c) Conformation and size of humic substances: Effects of major cation concentration and type, pH, salinity and residence time. Colloids and Surfaces A: Physicochemical and Engineering Aspects 272: 48--55. Baalousha M, Motelica-Heino M, Galaup S, and Coustumer P (2005c) Supramolecular structure of humic acids by TEM with improved sample preparation and staining. Microscopy Research and Technique 66: 299--306. Balamurugan A, Sockalingum G, Michel J, Faure J, Banchet V, and Wortham L (2006) Synthesis and characterisation of sol gel derived bioactive glass for biomedical applications. Materials Letters 60(29–30): 3752--3757. Balistrieri LS, Murray JW, and Paul B (1992b) The cycling of iron and manganese in the water column of Lake Sammamish, Washington. Limnology and Oceanography 37: 510--528. Balistrieri LS, Murray JW, and Paul B (1992a) The biogeochemical cycling of trace metals in the water column of Lake Sammamish, Washington: Response to seasonally anoxic conditions. Limnology and Oceanography 37(3): 529--548.

Balnois E, Durand-Vidal S, and Levitz P (2003) Probing the morphology of laponite clay colloids by atomic force microscopy. Langmuir 19(17): 6633--6637. Barth C and Henry CR (2004) High-resolution imaging of gold clusters on KBr(001) surfaces investigated by dynamic scanning force microscopy. Nanotechnology 15(9): 1264--1272. Bell GM, Levine S, and McCartney LN (1970) Approximate methods of determining the double-layer free energy of interaction between two charged colloidal spheres. Journal of Colloid and Interface Science 33(3): 335--359. Bellardita M, Addamo M, Di Paola A, and Palmisano L (2007) Photocatalytic behaviour of metal-loaded TiO2 aqueous dispersions and films. Chemical Physics 339: 94--103. Benedetti M, Ranville JF, Ponthieu M, and Pinheiro JP (2002) Field-flow fractionation characterization and binding properties of particulate and colloidal organic matter from the Rio Amazon and Rio Negro. Organic Geochemistry 33(3): 269--279. Benn TM and Westerhoff P (2008) Nanoparticle silver released into water from commercially available sock fabrics. Environmental Science and Technology 42(11): 4133--4139. Bergendahl J and Grasso D (1998) Colloid generation during batch leaching tests: Mechanics of disaggregation. Colloids and Surfaces A: Physicochemical and Engineering Aspects 135(1–3): 193--205. Berner RA and Holdren GR (1977) Mechanism of feldspar weathering: Some observational evidence. Geology 5: 369--372. Bertsch PM and Seaman JC (1999) Characterization of complex mineral assemblages: Implications for contaminant transport and environmental remediation. Proceedings of the National Academy of Sciences 7: 3350--3357. Bethune DS, Kiang CH, de Vries MS, Gorman G, Savoy R, and Vazquez J (1993) Cobalt-catalysed growth of carbon nanotubes with single-atomic-layer walls. Nature 363: 605--607. Blaser SA, Scheringer M, MacLeod M, and Hungerbnhler K (2008) Estimation of cumulative aquatic exposure and risk due to silver: Contribution of nanofunctionalized plastics and textiles. Science of the Total Environment 390(2–3): 396--409. Borm P, Klaessig FC, Landry TD, et al. (2006) Research strategies for safety evaluation of nanomaterials, part V: Role of dissolution in biological fate and effects of nanoscale particles. Toxicological Sciences 90(1): 23--32. Bradford SA, Yates SR, Bettahar M, and Simunek J (2002) Physical factors affecting the transport and fate of colloids in saturated porous media. Water Resources Research 38(12): 1327. Brant J, Lecaotnet H, and Wiessner MR (2005) Aggregation and deposition characteristics of fullerene nanoparticles in aqueous systems. Journal of Nanoparticle Research 7: 533--545. Brar SK, Verma M, Tyagi RD, and Surampalli RY (2010) Engineered nanoparticles in wastewater and wastewater sludge – evidence and impacts. Waste Management 30(3): 504--520. Braun M, Link S, Burda C, and El-Sayed M (2002) Transfer times of electrons and holes across the interface in CdS/HgS/CdS quantum dot quantum well nanoparticles. Chemical Physics Letters 361(5–6): 446--452. Bridge JW, Heathwaite AL, and Banwart SA (2009) Measurement of colloid mobilization and redeposition during drainage in Quartz sand. Environmental Science and Technology 43(15): 5769--5775. Brunner TJ, Manser P, Spohn P, Grass RN, Limbach LK, and Bruinink A (2006) In vitro cytotoxicity of oxide nanoparticles: Comparison to asbestos, silica, and the effect of particle solubility. Environmental Science and Technology 40: 4374--4381. Brus L (1986) Electronic wave-functions in semiconductor clusters – experiment and theory. Journal of Physical Chemistry 90(12): 2555--2560. BSI (2007) Terminology for nanomaterials. PAS 136:2007, BSI British Standards. Buffle J (1988) Complexation Reactions in Aquatic Systems: An Analystical Approach. Chichester: Ellis-Horwood. Buffle J (2006) The key role of environmental colloids/nanoparticles for the sustainability of life. Environmental Chemistry 3: 155--158. Buffle J and Leppard GG (1995b) Characterization of aquatic colloids and macromolecules. 2. Key role of physical structures on analytical results. Environmental Science and Technology 29: 2176--2184. Buffle J and Leppard GG (1995a) Characterization of aquatic colloids and macromolecules. 1. Structure and behavior of colloidal material. Environmental Science and Technology 29(9): 2169--2175. Buffle J and Van Leeuwen H (eds.) (1992) Environmental Particles. IUPAC Series on Analytical and Physical Chemistry of Environmental Systems. Vol 1. Lewis, Boca Raton, FL. Buffle J, Wilkinson KJ, Stoll S, Filella M, and Zhang J (1998) A generalized description of aquatic colloidal interactions: The three-colloidal component approach. Environmental Science and Technology 32(19): 2887--2899.

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

Bumajdad A, Eastoe J, and Mathewa A (2009) Cerium oxide nanoparticles prepared in self-assembled systems. Advances in Colloid and Interface Science 147–148: 56--66. Burgess RM, McKinney RA, and Brown WA (1996) Enrichment of marine sediment colloids with polychlorinated biphenyls: Trends resulting from PCB solubility and chlorination. Environmental Science and Technology 30(8): 2556--2566. Buzea C, Pacheco II, and Robbie K (2008) Nanomaterials and nanoparticles: Sources and toxicity. Biointerphases 2(4): MR17--MR71. Cai R, Van GM, Aw PK, and Itoh K (2006) Solar-driven self-cleaning coating for a painted surface. Comptes Rendus Chimie 9: 829--835. Campbell PG (1995) Interactions between trace metals and aquatic organisms: A critique of the free-ion activity model. In: Tessier A and Turner DR (eds.) Metal Speciation and Bioavailability in Aquatic Systems, pp. 45--102. Chichester: Wiley. Cao J, Elliott D, and Zhang W (2005) Perchlorate reduction by nanoscale iron particles. Journal of Nanoparticle Research 7: 499--506. Carnie SL, Chan DYC, and Gunning JS (1994) Electrical double layer interaction between dissimilar spherical colloidal particles and between a sphere and a plate: The linearized Poisson–Boltzmann theory. Langmuir 10(9): 2993--3009. Chakraborti RK, Gardner KH, Atkinson JF, and Van Benschoten JE (2003) Changes in fractal dimension during aggregation. Water Research 37(4): 873--883. Chapnick SD, Moore WS, and Nealson KH (1982) Microbially mediated manganese oxidation in a freshwater lake. Limnology and Oceanography 27: 1004--1014. Chellam S and Wiesner MR (1993) Fluid mechanics and fractal aggregates. Water Research 27(9): 1493--1496. Chen G, Flury M, Harsh JB, and Lichtner PC (2005) Colloid-facilitated transport of cesium in variably saturated Hanford sediments. Environmental Science and Technology 39(10): 3435--3442. Chen KL and Elimelech M (2006) Aggregation and deposition kinetics of fullerene (C60) nanoparticles. Langmuir 22(26): 10994--11001. Chen KL, Mylon SE, and Elimelech M (2006) Aggregation kinetics of alginate-coated hematite nanoparticles in monovalent and divalent electrolytes. Environmental Science and Technology 40(5): 1516--1523. Chen KL and Elimelech M (2007) Influence of humic acid on the aggregation kinetics of fullerene (C60) nanoparticles in monovalent and divalent electrolyte solutions. Journal of Colloid and Interface Science 309(1): 126--134. Chen S and Eisma D (1995) Fractal geometry of in situ flocs in the estuarine and coastal environments. Netherlands J. Sea Res. 33(2): 173--182. Cheng JP, Flahaut E, and Cheng SH (2007) Effect of carbon nanotubes on developing zebrafish (Danio reio) embryos. Environmental Toxicology and Chemistry 26: 708--716. Cheng X, Kan AT, and Tomson MB (2004) Naphthalene adsorption and desorption from aqueous (C60) fullerene. Journal of Chemical and Engineering Data 49(3): 675--683. Cheng X, Kan AT, and Tomson MB (2005) Study of C60 transport in porous media and the effect of sorbed C60 on naphthalene transport. Materials Research Society 20: 3244--3254. Chiang LY, Bhonsle JB, Wang L, Shu SF, Chang TM, and Hwu JR (1996) Efficient oneflask synthesis of water-soluble (60)fullerenols. Tetrahedron 52: 4963--4972. Choi AO, Cho SJ, Desbarats J, Lovric J, and Maysinger D (2007) Quantum dot-induced cell death involves Fas upregulation and lipid peroxidation in human neuroblastoma cells. Journal of Nanotechnology 5(1): doi:10.1186/1477-3155-5-1. Choi O and Hu Z (2008) Size dependent and reactive oxygen species related nanosilver toxicity to nitrifying bacteria. Environmental Science and Technology 42(12): 4583--4588. Christian P, Von der Kammer F, Baalousha M, and Hofmann Th (2008) Nanoparticles: Structure, properties, preparation and behaviour in environmental media. Environmental Toxicology and Chemistry 17: 326--343. Clapp AR, Medintz IL, Mauro JM, Fischer BR, Bawendi MG, and Mattoussi H (2004) Fluorescence resonance energy transfer between quantum dot donors and dyelabeled protein acceptors. Journal of American Chemical Society 126: 301--310. Cornell RM and Schwertmann U (1996) The Iron Oxides. Weinheim: VCH Publishers. Crist JT, McCarthy JF, Zevi Y, Baveye P, Throop JA, and Steenhuis TS (2004) Porescale visualization of colloid transport and retention in partly saturated porous media. Vadose Zone Journal 3(2): 444--450. Dai M, Martin JM, and Cauwet G (1995) The significant role of colloids in the transport and transformation of organic carbon and associated trace metals (Cd, Cu and Ni) in the Rhone delta (France). Marine Chemistry 51: 159--175. Daniel MC and Astruc D (2004) Gold nanoparticles: Assembly, supramolecular chemistry, quantum-sizerelated properties, and applications toward biology, catalysis, and nanotechnology. Chemical Reviews 104: 293--346. Davison W and De Vitre R (1992) Iron particles in freshwater. In: Buffle J and Leeuwen HP (eds.) Environmental Particles, pp. 315--355. Boca Raton, FL: Lewis.

123

Deguchi S, Alargova RG, and Tsujji K (2001) Stable dispersions of fullerenes, C60 and C70, in water. Preparation and characterisation. Langmuir 17(19): 6013--6017. Deguchi S, Mukai S, Tsudome M, and Horikoshi K (2006) Facile generation of fullerene nanoparticles by hand-grinding. Advanced Materials 18(6): 729--732. de Kerchove AJ and Elimelech M (2005) Relevance of electrokinetic theory for ‘‘soft’’ particles to bacterial cells: Implications for bacterial adhesion. Langmuir 21: 6462--6472. Delgado AV, Gonzblez-Caballero F, Hunter RJ, Koopal LK, and Lyklema J (2007) Measurement and interpretation of electrokinetic phenomena. Journal of Colloid and Interface Science 309(2): 194--224. Diederich F (2005) Advanced opto-electronics materials by fullerene and acetylene scaffolding. Pure and Applied Chemistry 77(11): 1851--1863. Diegoli S, Manciulea AL, Begum S, Jones IP, Lead JR, and Preece JA (2008) Interactions of charge stabilised gold nanoparticles with organic macromolecules. Science of the Total Environment 402(1): 51--61. Djenadic RR, Nikolic LM, Giannakopoulos KP, Stojanovic B, and Srdic VV (2007) Onedimensional titanate nanostructures: Synthesis and characterization of more options. Journal of the European Ceramic Society 27: 4339--4343. Donaldson K and Tran CL (2002) Inflammation caused by particles and fibers. Inhalation Toxicology 14(1): 5--27. Dong D, Li Y, and Hua X (2001) Investigation of Fe, Mn oxides and organic material in surface coatings and Pb, Cd adsorption to surface coatings developed in different natural waters. Microchemical Journal 70(1): 25--33. Dong D, Nelson YM, Lion LW, Shuler ML, and Ghiorse WC (2000) Adsorption of Pb and Cd onto metal oxides and organic material in natural surface coatings as determined by selective extractions: New evidence for the importance of Mn and Fe oxides. Water Research 34(2): 427--436. Doty RC, Tshikhudo TR, Brust M, and Fernig DG (2005) Extremely stable water-soluble Ag nanoparticles. Chemistry of Materials 17: 4630--4635. Doucet F, Lead JR, and Santschi PH (2006) Role of colloids in mediating trace element fate and behaviour in aquatic systems. In: Wilkinson KJ and Lead JR (eds.) Environmental Colloids: Behaviour, Structure and Characterisation, pp. 95--158. New York: Wiley. Dunford R, Salinaro A, Cai L, et al. (1997) Chemical oxidation and DNA damage catalysed by inorganic sunscreen ingredients. FEBS Letters 418(1–2): 87--90. Duval JF (2007) Electrophoresis of soft colloids: Basic principles and applications. In: Wilkinson KJ and Lead JR (eds.) Environmental Colloids and Particles: Behaviour, Separation and Characterisation, pp. 315--344. New York: Wiley. Duval JFL, Wilkinson KJ, van Leeuwen HP, and Buffle J (2005) Humic substances are soft and permeable: Evidence from their electrophoretic mobilities. Environmental Science and Technology 39(17): 6435--6445. Ebbesen TW (1994) Carbon nanotubes. Annual Review of Materials Research 24: 235. Ebbesen TW and Ajayan PM (1992) Large-scale synthesis of carbon nanotubes. Nature 358: 220--222. Eckelman MJ and Graedel TE (2007) Silver emissions and their environmental impacts: A multilevel assessment. Environmental Science and Technology 41: 6283--6289. Elechigerra JL, Burt JL, Morones JR, Camacho-Bragado A, Gao X, and Lara HH (2005) Interaction of silver nanoparticles with HIV-1. Journal of Nanobiotechnology 3: 6. Elimelech M, Gregory J, Jia G, and Williams RA (1995) Surface interaction potentials. In: Elimelech M, Gregory J, Jia G, and Williams RA (eds.) Particle Deposition and Aggregation: Measurement, Modelling and Simulation, pp. 3--33. Woburn, MA: Butterworth-Heinemann. Elliott D and Zhang W (2001) Field assessment of nanoparticles for groundwater treatment. Environmental Science and Technology 35: 4922--4926. Endo M, Takeuchi K, Igarashi S, Kobori K, Shiraishi M, and Kroto HW (1993) The production and structure of pyrolytic carbon nanotubes (PCNTs). Journal of Physics and Chemistry of Solids 54: 1841--1848. Fabiano M and Pusceddu A (1998) Total and hydrolizable particulate organic matter (carbohydrates, proteins and lipids) at a coastal station in Terra Nova Bay (Ross Sea, Antarctica). Polar Biology 19: 125--132. Fabrega J, Fawcett SR, Renshaw JC, and Lead JR (2009a) Silver nanoparticle impact on bacterial growth: Effect of pH, concentration, and organic matter. Environmental Science and Technology 43(19): 7285--7290. Fabrega J, Renshaw JC, and Lead JR (2009b) Interactions of Silver nanoparticles with pseudomonas putida biofilms. Environmental Science and Technology 43(23): 9004--9009. Faghihian H and Kamali M (2003) Synthesis of Na–P-c zeolite from perlite and study of its ability to remove cyanide from liquid wastes. International Journal of Environment and Pollution 19: 557--566. Federici G, Shaw BJ, and Handy RD (2007) Toxicity of titanium dioxide nanoparticles to rainbow trout (Oncorhynchus mykiss): Gill injury, oxidative stress, and other physiological effects. Aquatic Toxicology 84: 415--430.

124

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

Feldmann C (2003) Polyol-mediated synthesis of nanoscale functional materials. Advanced Functional Materials 13: 101--107. Ferretti R, Zhang J, and Buffle J (1997) Kinetics of hematite aggregation by polyacrylic acid: Effect of polymer molecular weights. Colloids and Surfaces A: Physicochemical and Engineering Aspects 121: 203--215. Filella M. (2006) Colloidal properties of submicron particles in natural waters. In: Wilkinson K and Lead JR, (eds). Environmental Colloids and Particles: Behaviour, Structure and Characterisation, IUPAS Series of Analystical and Physical Chemistry of Environmental Systems, pp. 18–93. New York: Wiley. Filella M and Buffle J (1993) Factors controlling the stability of submicron colloids in natural waters. Colloids and Surfaces A: Physicochemical and Engineering Aspects 73: 255--273. Filella M, Buffle J, and Leppard GG (1993) Characterization of submicrometre colloids in freshwaters: Evidence for their bridging by organic structures. Water Science and Technology 27: 91--102. Franklin NM, Rogers NJ, Apte SC, Batley GE, Gadd GE, and Casey PS (2007) Comparative toxicity of nanoparticulate ZnO, bulk ZnO, and ZnCl2 to a freshwater microalga (Pseudokirchneriella subcapitata): The importance of particle solubility. Environmental Science and Technology 41: 8484--8490. Gaebel C, Lead JR, Renshaw JC, and Tellam JH (2009) Preliminary indications from atomic force microscopy of the presence of rapidly-formed nanoscale films on aquifer material surfaces. Journal of Contaminant Hydrology 108(1–2): 46--53. Gagne´ F, Auclair J, Turcotte P, Fournier M, Gagnon C, and Sauve´ S (2008) Ecotoxicity of CdTe quantum dots to freshwater mussels: Impacts on immune system, oxidative stress and genotoxicity. Aquatic Toxicology 86: 333--340. Gao G (2004) Nanostructures and Nanomaterials. Synthesis, Properties and Applications. London: Imperial College Press. Geiser M, et al. (2005) Ultrafine particles cross cellular membranes by nonphagocytic mechanisms in lungs and in cultured cells. Environmental Health Perspectives 113: 1555--1560. Genxiang L, Ruijiang F, Jun J, and Ping WH (2001) Strong and weak interactions of colloidal particles with high surface potentials. Journal of Colloid and Interface Science 241(1): 81--88. Giasuddin ABM, Kanel SR, and Choi H (2007) Adsorption of humic acid onto nanoscale zerovalent iron and its effect on arsenic removal. Environmental Science and Technology 41(6): 2022--2027. Gibson CT, Turner IJ, Roberts CJ, and Lead JR (2007) Quantifying the dimensions of nanoscale organic surface layers in natural waters. Environmental Science and Technology 41(4): 1339--1344. Gontard LC, Dunin-Borkowski RE, Ozkaya D, Hyde T, Midgley PA, and Ash P (2006) Crystal size and shape analysis of Pt nanoparticles in two and three dimensions. Journal of Physics: Conference Series 26: 367--370. Gottschalk F, Scholz RW, and Nowack B (2009a) Probabilistic material flow modeling for assessing the environmental exposure to compounds: Methodology and an application to engineered nano-TiO2 particles. Environmental Modelling and Software 25(3): 320--332. Gottschalk F, Sonderer T, Scholz RW, and Nowack B (2009b) Modeled environmental concentrations of engineered nanomaterials (TiO2, ZnO, Ag, CNT, fullerenes) for different regions. Environmental Science and Technology 43(24): 9216--9222. Grabar KD, Brown KR, Keating CD, Stranick SJ, Tang SL, and Natan MJ (1997) Nanoscale characterization of gold colloid monolayers: A comparison of four techniques. Analytical Chemistry 69: 471--477. Grasso D, Subramaniam K, Butkus M, Strevett K, and Bergendahl J (2002) A review of non-DLVO interactions in environmental colloidal systems. Reviews in Environmental Science and Biotechnology 1: 17--38. Gra¨tzel M (2001) Sol–gel processed TiO2 films for photovoltaic applications. Journal of Sol-Gel Science and Technology 22: 7--13. Griffitt RJ, Weil R, Hyndman KA, Denslow ND, Powers K, and Taylor D (2007) Exposure to copper nanoparticles causes gill injury and acute lethality in zebrafish (Danio rerio). Environmental Science and Technology 41: 8178--8186. Grolimund D and Borkovec M (2005) Colloid-facilitated transport of strongly sorbing contaminants in natural porous media: Mathematical modeling and laboratory column experiments. Environmental Science and Technology 39(17): 6378--6386. Guo L and Santschi PH (2007) Ultrafiltration and its applications to sampling and characterization of aquatic colloids. In: Wilkinson KJ and Lead JR (eds.) Environmental Colloids and Particles: Behaviour, Separation and Characterisation, IUPAC Series on Analytical and Physical Chemistry of Environmental Systems, pp. 159–222. Chichester: Wiley. Guo L, Santschi PH, and Ray SM (2002) Metal partitioning between colloidal and dissolved phases and its relation with bioavailability to American oysters. Marine Environmental Research 54(1): 49--64. Gustafsson O and Gschwend PM (1997) Aquatic colloids: Concepts, definitions, and current challenges. Limnology and Oceanography 42(3): 519--528.

Hackley VA and Anderson MA (1989) Effects of short-range forces on the long-range structure of hydrous iron oxide aggregates. Langmuir 5(1): 191--198. Haick H (2007) Chemical sensors based on molecularly modified metallic nanoparticles. Journal of Physics D – Applied Physics 40: 7173--7286. Hailstone RK, Di Francesco AG, Leong JG, Allston TD, and Reed KJ (2009) A study of lattice expansion in CeO2 nanoparticles by transmission electron microscopy. Journal of Physical Chemistry C 113: 15155--15159. Handy RD, Owen R, and Valsami-Jones E (2008a) The ecotoxicology of nanoparticles and nanomaterials: Current status, knowledge gaps, challenges, and future needs. Ecotoxicology 17(5): 315--325. Handy RD, Von der Kammer F, Lead JR, Hassello¨v M, Owen R, and Crane M (2008b) The ecotoxicology and chemistry of manufactured nanoparticles. Ecotoxicology 17(4): 287--314. Hardman R (2006) A toxicologic review of quantum dots: Toxicity depends on physicochemical and environmental factors. Environmental Health Perspectives 114: 165--172. Hartmann NB, Von der Kammer F, Baalousha M, Ottofuelling S, and Baun A (2009) Algal testing of titanium dioxide nanoparticles – testing considerations, inhibitory effects and modification of cadmium bioavailability. Toxicology 269(2–3): 190--197. Hassellov M and van der Kammer F (2008) Iron oxides as geochemical nanovectors for metal transport in soil–river systems. Elements 4(6): 401--406. Hayashi H, Tsuneda S, Hirata A, and Sasaki H (2001) Soft particle analysis of bacterial cells and its interpretation of cell adhesion behaviors in terms of DLVO theory. Colloids and Surfaces B: Biointerfaces 22(2): 149--157. Heath MR (1995) Size spectrum dynamics and the planktonic ecosystem of LochLinnhe. ICES Journal of Marine Science 52(3–4): 627--642. Heathwaite L, Haygarth P, Matthews R, Preedy N, and Butler P (2005) Evaluating colloidal phosphorus delivery to surface waters from diffuse agricultural sources. Journal of Environmental Quality 34(1): 287--298. Heckert EG, Seal S, and Self WT (2008) Fenton-like reaction catalyzed by the rare earth inner transition metal cerium. Environmental Science and Technology 42(13): 5014--5019. Hedges JI, Keil RG, and Benner R (1997) What happens to terrestrial organic matter in the ocean? Organic Geochemistry 27: 195--212. Heinlaan M, Ivask A, Blinova I, Dubourguier HC, and Kahru A (2008) Toxicity of nanosized and bulk ZnO, CuO and TiO2 to bacteria Vibrio fischeri and crustaceans Daphnia magna and Thamnocephalus platyurus. Chemosphere 71: 1308--1316. Helgeson HC, Murphy WM, and Aagaard P (1984) Thermodynamic and kinetic constraints on reaction rates among minerals and aqueous solutions. II. Rate constants, effective surface area, and the hydrolysis of feldspar. Geochimica et Cosmochimica Acta 48: 2405--2432. Henry TB, Menn F, Fleming JT, Wilgus J, Compton RN, and Sayler GS (2007) Attributing effects of aqueous C60 nano-aggregates to tetrahydrofuran decomposition products in larval zebrafish by assessment of gene expression. Environmental Health Perspectives 115(7): 1059--1065. Hentschel ML and Page NW (2003) Selection of descriptors for particle shape characterization. Particle and Particle Systems Characterisation 20(1): 25--38. Herrera M, Carrion P, Baca P, Liebana J, and Castillo A (2001) In vitro antibacterial activity of glass-ionomer cements. Microbios 104: 141--148. Hirano K, Nitta H, and Sawada K (2005) Effect of sonication on the photo-catalytic mineralization of some chlorinated organic compounds. Ultrasonics Sonochemistry 12: 271--276. Hochella MF (2002) There’s plenty of room at the bottom: Nanoscience in geochemistry. Geochimica et Cosmochimica Acta 66(5): 735--743. Hofmann T, et al. (2003) Aquatic colloids: Definition and relevance – a review. Grundwasser 8: 203--210. Hofmann Th and Wendelborn A (2007) Colloid facilitated transport of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) to the groundwater at Ma Da Area, Vietnam. Environmental Science and Pollution Research 14: 223--224. Honeyman BD and Santschi PH (1991) A Brownian-pumping model for oceanic trace metal scavenging: Evidence from Th isotopes. Journal of Marine Research 47(4): 951--992. Honeyman BD and Santschi PH (1992) The role of particles and colloids in the transport of radionuclides and trace metals in the oceans. In: Buffle J and van Leeuwen T (eds.) Environmental Particles. IUPAC series on Environmental Analytical and Physical Chemistry, pp. 379--423. Boca Raton, FL: Lewis Publishers. Hu X, Cook S, Wang P, and Hwang HM (2009) In vitro evaluation of cytotoxicity of engineered metal oxide nanoparticles. Science of the Total Environment 407(8): 3070--3072. Hunter KA and Liss PS (1982) Organic matter and surface charge of suspended particles in estuarine waters. Limnology and Oceanography 27: 322--335.

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

Hyung H, Fortner JD, Hughes JB, and Kim JH (2007) Natural organic matter stabilizes carbon nanotubes in the aqueous phase. Environmental Science and Technology 41(1): 179--184. Ijima S (1991) Helical microtubules of graphitic carbon. Nature 354: 56--58. Ijima S and Ichihashi T (1993) Single-shell carbon nanotubes of 1-nm diameter. Nature 363: 603--605. Jarvie HP and King SM (2007) Small-angle neutron scattering study of natural aquatic nanocolloids. Environmental Science and Technology 41(8): 2868--2873. Jarvis P, Jefferson B, Gregory J, and Parsons SA (2005) A review of floc strength and breakage. Water Research 39(14): 3121--3137. Jekel MR (1986) The stabilization of dispersed mineral particles by adsorption of humic substances. Water Research 20(12): 1543--1554. Jeng HA and Swanson J (2006) Toxicity of metal oxide nanoparticles in mammalian cells. Journal of Environmental Science and Health. Part A, Toxic/Hazardous Substances and Environmental Engineering 41: 2699--2711. Jeon HJ, Yi SC, and Oh SG (2003) Preparation and antibacterial effects of Ag–SiO2 thin films by sol–gel method. Biomaterials 24: 4921--4928. Jia Y, Morrison DJ, and Moosbrugger JC (2005) The influence of magnetostriction on the shape of the hysteresis loop of cyclically deformed single crystal nickel. Scripta Materialia 53(9): 1025--1029. Johnson CP, Li X, and Logan BE (1996) Settling velocities of fractal aggregates. Environmental Science and Technology 30: 1911--1918. Johnson SB, Gordon E, Healy TW, and Scales PJ (2005) Adsorption of organic matter at mineral/water interfaces. 6. Effect of inner-sphere versus outer-sphere adsorption on colloidal stability. Langmuir 21(14): 6356--6365. Johnson SB, Yoon TH, Slowey AJ, and Brown GE (2004) Adsorption of organic matter at mineral/water interfaces: 3. Implications of surface dissolution for adsorption of oxalate. Langmuir 20(26): 11480--11492. Joo JC, Shackelford CD, and Reardon KF (2008) Sorption of nonpolar neutral organic compounds to humic acid-coated sands: Contributions of organic and mineral components. Chemosphere 70(7): 1290--1297. Ju-Nam Y and Lead JR (2008) Manufactured nanoparticles: An overview of their chemistry, interactions and potential environmental implications. Science of the Total Environment 400: 396--414. Kadish KM and Ruoff RS (2000) Fullerenes: Chemistry, physics, and technology, p. 53. New York: Wiley. Kagan VE, Tyurina YY, Tyurin VA, Konduru NV, Potapovich AI, and Osipov AN (2006) Direct and indirect effects of single walled carbon nanotubes on RAW 264.7 macrophages: Role of iron. Toxicology Letters 165: 88--100. Kahru A, Dubourguier HC, Blinova I, Ivask A, and Kasemets K (2008) Biotests and biosensors for ecotoxicology of metal oxide nanoparticles: A minireview. Sensors 8: 5153--5170. v. d. Kammer F (2005) Characterization of Environmental Colloids applying Field-Flow Fractionation–Multi Detection Analysis with Emphasis on Light Scattering Techniques. Hamburg, Germany: Hamburg University of Technology. Kan AT, Fu G, and Tomson MB (1994) Adsorption/desorption hysteresis in organic pollutant and soil/sediment interaction. Environmental Science and Technology 28(5): 859--867. Kanel SR, Dhriti D, Bruce M, and Heechul C (2007) Transport of surface-modified iron nanoparticle in porous media and application to arsenic(III) remediation. Journal of Nanoparticle Research 9: 725--735. Kanel SR, Manning B, Charlet L, and Choi H (2005) Removal of arsenic(III) from groundwater by nanoscale zero-valent iron. Environmental Science and Technology 39: 1291--1298. Kang S, Pinault M, Pfefferle LD, and Elimelech M (2007) Single-walled carbon nanotubes exhibit strong antimicrobial activity. Langmuir 23(17): 8670--8673. Kaplan DI, Bertsch PM, and Adriano DC (1995) Facilitated transport of contaminant metals through an acidified aquifer. Ground Water 33: 708--717. Karakoti AS, Hench LL, and Seal LL (2006) The potential toxicity of nanomaterials – the role of surfaces. Journal of the Minerals Metals and Materials Society 58: 77--82. Kashiwada S (2006) Distribution of nanoparticles in the see-through medaka (Oryzias latipes). Environmental Health Perspectives 114: 1697--1702. Kikuchi K, Nakahara N, Wakabayashi T, Suzuki S, Shiromaru H, and Miyake Y (1992) NMR characterization of isomers of C78, C82 and C84 fullerenes. Nature 357: 142--145. Kim KJ, Park YR, Lee JH, et al. (2007) Room-temperature ferromagnetic properties in Mn-doped rutile TiO2-delta thin films. Journal of Magnetism and Magnetic Materials 316(2): E215--E218. Kiser MA, Westerhof P, Benn T, Wang Y, Perez-Rivera J, and Hristovski K (2009) Titanium nanomaterial removal and release from wastewater treatment plants. Environmental Science and Technology 43: 6757--6763.

125

Klabunde KJ, Stark J, Koper O, et al. (1996) Nanocrystals as stoichiometric reagents with unique surface chemistry. Journal of Physical Chemistry 100(30): 12142--12153. Klaine SJ, Alvarez PJJ, Batley GE, et al. (2008) Nanomaterials in the environment: Behavior, fate, bioavailability, and effects. Environmental Toxicology and Chemistry 27(9): 1825--1851. Kra¨tschmer W, Lamb LD, Fostiropoulos K, and Huffman DR (1990) Solid C60: A new form of carbon. Nature 347: 354--358. Kretzschmar R, Borkovec M, Grolimund D, and Elimelech M (1999) Mobile subsurface colloids and their role in contaminant transport. Advances in Agronomy 66: 121--193. Kretzschmar R, Robarge WP, and Amoozegar A (1995) Influence of natural organic matter on colloid transport through saprolite. Water Resources Research 31: 435--446. Kroto HW, Heath JR, O’Brien SC, Curl RF, and Smalley RE (1985) C60: Buckminsterfullerene. Nature 318(6042): 162--163. Kumar C (2006) Nanomaterials – Toxicity, Health and Environmental Issues. Weinheim, Germany: Wiley-VCH. Kumar R and Mu¨nstedt H (2005) Silver ion release from antimicrobial polyamide/silver composites. Biomaterials 26: 2081--2088. Laegdsmand M, Villholth KG, Ullum M, and Jensen KH (1999) Processes of colloid mobilization and transport in macroporous soil monoliths. Geoderma 93(1–2): 33--59. Langer R and Tirrell DA (2004) Designing materials for biology and medicine. Nature 428: 487--492. Laosiripojana N, Sutthisripok W, and Assabumrungrat S (2005) Synthesis gas production from dry reforming of methane over CeO2 doped Ni/Al2O3: Influence of the doping ceria on the resistance toward carbon formation. Chemical Engineering Journal 112(1–3): 13--22. Lartiges BS, Deneux-Mustin S, Villemin G, et al. (2001) Composition, structure and size distribution of suspended particulates from the Rhine River. Water Research 35(3): 808--816. Lead JR, Hamilton-Taylor J, Davison W, and Harper M (1999) Trace metal sorption by natural particles and coarse colloids. Geochimica et Cosmochimica Acta 63(11–12): 1661--1670. Lead JR, Muirhead D, and Gibson CT (2005) Characterisation of freshwater natural aquatic colloids by atomic force microscopy (AFM). Environmental Science and Technology 39: 6930--6936. Lead JR and Wilkinson KJ (2006) Aquatic colloids and nanoparticles: Current knowledge and future trends. Environmental Chemistry 3: 159--171. Lead JR and Wilkinson KJ (2007) Environmental colloids: Current knowledge and future developments. In: Lead JR and Wilkinson K (eds.) Environmental Colloids: Behaviour, Structure and Characterisation, pp. 1--16. New York: Wiley. Lead JR, Wilkinson KJ, Balnois E, et al. (2000a) Diffusion coefficients and polydispersities of the Suwannee River fulvic acid: Comparison of fluorescence correlation spectroscopy, pulsed-field gradient nuclear magnetic resonance, and flow field-flow fractionation. Environmental Science and Technology 34(16): 3508--3513. Lead JR, Wilkinson KJ, Starchev K, Canonica S, and Buffle J (2000b) Determination of diffusion coefficients of humic substances by fluorescence correlation spectroscopy: Role of solution conditions. Environmental Science and Technology 34(7): 1365--1369. Lecoanet HF, Bottero JY, and Wiesner MR (2004) Laboratory assessment of the mobility of nanomaterials in porous media. Environmental Science and Technology 38(19): 5164--5169. Lecoanet HF and Wiesner MR (2004) Velocity effects on fullerene and oxide nanoparticle deposition in porous media. Environmental Science and Technology 38: 4377--4382. Lee C and Kramer TA (2004) Prediction of three-dimensional fractal dimensions using the two-dimensional properties of fractal aggregates. Advances in Colloid and Interface Science 112(1–3): 49--57. Leenheer JA and Croue´ JP (2003) Characterizing aquatic dissolved organic matter. Environmental Science and Technology 37: 18A--26A. Lenhart JJ and Saiers JE (2002) Transport of silica colloids through unsaturated porous media: Experimental results and model comparisons. Environmental Science and Technology 36(4): 769--777. Leppard GG (1997) Colloidal organic fibrils of acid polysaccharides in surface waters: Electron-optical characteristics, activities and chemical estimates of abundance. Colloids and Surfaces A: Physicochemical and Engineering Aspects 120: 1--15. Leppard GG and Arsenault AL (2003) Note – quantification of individual native biocolloids in natural waters: Assessing indicators of aquatic events, by using transmission electron microscopy with a modified approach to image analysis. Archiv fu¨r Hydrobiologie 156: 565--573.

126

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

Leppard GG, Buffle J, and Baudat R (1986) A description of the aggregation properties of aquatic pedogenic fulvic-acids – combining physicochemical data and microscopic observations. Water Research 20(2): 185--196. Leppard GG, Burnison BK, and Buffle J (1990) Transmission electron microscopy of the natural organic matter of surface waters. Analytica Chimica Acta 232: 107--121. Leppard GG, Flannigan DT, Mavrocordatos D, Marvin CH, Bryant DW, and McCarry BE (1998) Binding of polycyclic aromatic hydrocarbons by size classes of particulate in Hamilton Harbor water. Environmental Science and Technology 32(22): 3633--3639. Leppard GG, West MM, Flannigan DT, Carson J, and Lott JNA (1997) A classification scheme for marine organic colloids in the Adriatic Sea: Colloid speciation by transmission electron microscopy. Canadian Journal of Fisheries and Aquatic Sciences 54(10): 2334--2349. Li X-Q, Elliott DW, and Zhang W-X (2006) Zero-valent iron nanoparticles for abatement of environmental pollutants: Materials and engineering aspects. Critical Reviews in Solid State and Materials Sciences 31(4): 111--122. Liang Y, Zhang H, Tian Z, Zhu X, Wang X, and Yi B (2006) Synthesis and structure– activity relationship exploration of carbon-supported PtRuNi nanocomposite as a CO-tolerant electrocatalyst for proton exchange membrane fuel cells. Journal of Physical Chemistry B 110(15): 7828--7834. Liang Y, Hilal N, Langston P, and Starov V (2007) Interaction forces between colloidal particles in liquid: Theory and experiment. Advances in Colloid and Interface Science 134–135: 151--166. Lienemann C-P, Monnerat M, Dominik J, and Perret D (1999) Identification of stoichiometric iron–phosphorus colloids produced in a eutrophic lake. Aquatic Science 61: 133--149. Lienemann CP, Taillefert M, Perret D, and Gaillard JF (1997) Association of cobalt and manganese in aquatic systems: Chemical and microscopic evidence. Geochimica et Cosmochimica Acta 61(7): 1437--1446. Limbach LK, Bereiter R, Iler EM, Krebs R, Ili R, and Stark WJ (2008) Removal of oxide nanoparticles in a model wastewater treatment plant: Influence of agglomeration and surfactants on clearing efficiency. Environmental Science and Technology 42(15): 5828--5833. Lin W, Xu Y, Huang C, et al. (2009) Toxicity of nano- and micro-sized ZnO particles in human lung epithelial cells. Journal of Nanoparticle Research 11: 25--39. Liu GD and Lin YH (2007) Nanomaterial labels in electrochemical immunosensors and immunoassays. Talanta 74: 308--317. Liu R, Wilding A, Hibberd A, and Zhou JL (2005) Partition of endocrine-disrupting chemicals between colloids and dissolved phase as determined by cross-flow ultrafiltration. Environmental Science and Technology 39(8): 2753--2761. Loder TC and Liss PS (1985) Control by organic coatings of the surface charge of estuarine suspended particles. Limnology and Oceanography 30(2): 418--421. Logan BE and Hunt JR (1987) Advantages to microbes of growth in permeable aggregates in marine systems. Limnology and Oceanography 32: 1034--1048. Lu AH, Sabalas EL, and Schu¨th F (2007) Magnetic nanoparticles: Synthesis, protection, functionalization, and application. Angewandte Chemie International Edition 46: 1222--1244. Luoma SN (2008) Silver nanotechnologies and the environment: Old problems or new challenges? PEN 15, pp. 1–66. Woodrow Wilson International Center for Scholars. Project on Emerging nanotechnologies. The PEW Charitable Trusts. Lyon DY, Adams LK, Falkner JC, and Alvarez PJJ (2006) Antibacterial activity of fullerene water suspensions: Effects of preparation method and particle size. Environmental Science and Technology 40: 4360--4366. Lyon DY, Fortner JD, Sayes CM, Colvin VL, and Hughes JB (2005) Bacterial cell association and antimicrobial activity of a C60 water suspension. Environmental Toxicology and Chemistry 24: 2757--2762. Lyven B, Hassellov M, Turner DR, Haraldsson C, and Andersson K (2003) Competition between iron- and carbon-based colloidal carriers for trace metals in a freshwater assessed using flow field-flow fractionation coupled to ICPMS. Geochimica et Cosmochimica Acta 67(20): 3791--3802. Madden AS and Hochella J (2005) A test of geochemical reactivity as a function of mineral size: Manganese oxidation promoted by hematite nanoparticles. Geochimica et Cosmochimica Acta 69(2): 389--398. Madden AS, Hochella J, and Luxton TP (2006) Insights for size-dependent reactivity of hematite nanomineral surfaces through Cu2þ sorption. Geochimica et Cosmochimica Acta 70(16): 4095--4104. Mannino A and Harvey HR (2000) Biochemical composition of particles and dissolved organic matter along an estuarine gradient: Sources and implications for DOM reactivity. Limnology and Oceanography 45: 775--788. Markarian J (2006) Steady growth predicted for biocides. Plastics, Additives and Compounding 8: 30--33.

Marvin CH, Leppard GG, West MM, Stern GA, Boden AR, and McCarry BE (2004) Refined tunable methodology for characterization of contaminant–particle relationships in surface water. Journal of Environmental Quality 33(6): 2132--2140. Maskaoui K, Hibberd A, and Zhou JL (2007) Assessment of the interaction between aquatic colloids and pharmaceuticals facilitated by cross-flow ultrafiltration. Environmental Science and Technology 41(23): 8038--8043. Mayer LM and Xing B (2001) Organic matter–surface area relationships in acid soils. Soil Science Society of America Journal 65(1): 250--258. McCarthy JF (1998) Colloid-facilitated transport of contaminants in groundwater: Mobilization of transuranic radionuclides from disposal trenches by natural organic matter. Physics and Chemistry of the Earth 23(2): 171--178. McCarthy JF and Zachara JM (1989) Subsurface transport of contaminants. Environmental Science and Technology 23(5): 496--502. McCormack D, Carnie SL, and Chan DYC (1995) Calculations of electric double-layer force and interaction free energy between dissimilar surfaces. Journal of Colloid and Interface Science 169(1): 177--196. McDowell-Boyer LM, Hunt JR, and Sitar N (1986) Particle transport through porous media. Water Resources Research 22: 1901--1921. McGechan MB (2002) SW – soil and water: Transport of particulate and colloid-sorbed contaminants through soil, part 2: Trapping processes and soil pore geometry. Biosystems Engineering 83(4): 387--395. Moore MN (2006) Do nanoparticles present ecotoxicological risks for the health of the aquatic environment? Environment International 32: 967--976. Morones JR, Elechiguerra JL, Camacho A, et al. (2005) The bactericidal effect of silver nanoparticles. Nanotechnology 16(10): 2346--2353. Morris ER, Rees DA, Robinson G, and Young GA (1980) Competitive inhibition of interchain interactions in polysaccharide systems. Journal of Molecular Biology 138: 363--374. Mosley LM, Hunter KA, and Ducker WA (2003) Forces between colloid particles in natural waters. Environmental Science and Technology 37: 3303--3308. Mossman BT, Marsh JP, Sesko A, et al. (1990) Inhibition of lung injury, inflammation, and interstitial pulmonary fibrosis by polyethylene glycol-conjugated catalase in a rapid inhalation model of asbestosis. American Review of Respiratory Disease 141(5): 1266--1271. Mueller NC and Nowack B (2008) Exposure modeling of engineered nanoparticles in the environment. Environmental Science and Technology 42(12): 4447--4453. Mukhopadhyay B and Walther JV (2001) Acid–base chemistry of albite surfaces in aqueous solutions at standard temperature and pressure. Chemical Geology 174(4): 415--443. Mu¨ller J, Huaux F, Fonseca A, et al. (2008) Structural defects play a major role in the acute lung toxicity of multiwall carbon nanotubes: Toxicological aspects. Chemical Research in Toxicology 21: 1698--1705. Murphy CJ, Gole AM, Hunyadi SE, Stone JW, Sisco PN, and Alkilany A (2008) Chemical sensing and imaging with metallic nanorods. Chemical Communications 5: 544--557. Murphy WM and Helgeson HC (1987) Thermodynamic and kinetic constraints on reaction rates among minerals and aqueous solutions. III. Activated complexes and the pH-dependence of the rates of feldspar, pyroxene, wollastonite, and olivine hydrolysis. Geochimica et Cosmochimica Acta 51: 3137--3153. Myklestad SM (1995) Release of extracellular products by phytoplankton with special emphasis on polysaccharides. Science of the Total Environment 165: 155--164. Navarro E, Baun A, Behra R, et al. (2008a) Environmental behavior and ecotoxicity of engineered nanoparticles to algae, plants, and fungi. Ecotoxicology 17: 372--386. Neihof RA and Loeb GI (1975) The surface charge of particulate matter in seawater. Limnology and Oceanography 17: 7--16. Nel A, Xia T, Madler L, and Li N (2006) Toxic potential of materials at the nanolevel. Science 311: 622--627. Nelson DM, Tre´guer P, Brzezinski MA, Leynaert A, and Que´guiner B (1995) Production and dissolution of biogenic silica in the ocean: Revised global estimates, comparison with regional data and relationship to biogenic sedimentation. Global Biogeochemical Cycles 9: 359--372. Neveu S, Bee A, Robineau M, and Talbot D (2002) Size-selective chemical synthesis of tartrate stabilized cobalt ferrite ionic magnetic fluid. Journal of Colloid and Interface Science 255: 293--298. Newman KA and Stolzenbach KD (1996) Kinetics of aggregation and disaggregation of titanium dioxide particles and glass beads in a sheared fluid suspension. Colloids and Surfaces A: Physicochemical and Engineering Aspects 107: 189--203. Niu J, Azfer A, Rogers LM, Wang X, and Kolattukudy PE (2007) Cardioprotective effects of cerium oxide nanoparticles in a transgenic murine model of cardiomyopathy. Cardiovascular Research 73(3): 549--559.

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

NNI (2004) National Nanotechnology Initiative Strategic Plan. Developed by the Nanoscience, Engineering, and Technology Subcommittee, Committtee on National Science and Technology Council, December. Noguez C (2007) Surface plasmons on metal nanoparticles: The influence of shape and physical environment. Journal of Physical Chemistry C 111: 3806--3819. Nowack B and Bucheli TD (2007) Occurrence, behavior and effects of nanoparticles in the environment. Environmental Pollution 150: 5--22. Nutt M, Hughes J, and Wong M (2005) Designing Pd on Au bimetallic nanoparticles catalysts for trichloroethene hydrodechlorination. Environmental Science and Technology 39: 1346--1353. O’Melia C (1980) ES&T features: Aquasols: The behavior of small particles in aquatic systems. Environmental Science and Technology 14(9): 1052--1060. Oberdo¨rster E (2004) Manufactured nanomaterials (fullerenes, C60) induce oxidative stress in the brain of juvenile largemouth bass. Environmental Health Perspectives 112: 1058--1062. Oberdo¨rster G, Oberdo¨rster E, and Oberdo¨rster J (2005) Nanotoxicology: An emerging discipline evolving from studies of ultrafine particles supplemental web sections. Environmental Health Perspectives 113: 823--839. O¨sterberg R and Mortensen K (1992) Fractal dimension of humic acids. European Biophysics Journal 21: 163--167. Ottofuelling S, Kammer FV, and Hofmann T (2007) Nanoparticles in the aquatic environment – aggregation behavior of TiO2 nanoparticles studied in a simplified aqueous test matrix (SAM). Geophysical Research Abstract 9: 08876. Otsubo T, Aso Y, and Takimiya K (2005) Fullerene-tethered oligothiophenes as advanced molecular electronics materials. Pure and Applied Chemistry 77(12): 2003--2010. Ouali L and Pefferkorn E (1994) Fragmentation of colloidal aggregates induced by polymer adsorption. Journal of Colloid and Interface Science 168(2): 315--322. Ouyang Y, Shinde D, Mansell RS, and Harris W (1996) Colloid-enhanced transport of chemicals in subsurface environments: A review. Critical Reviews in Environmental Science and Technology 26: 189--204. Pal S, Tak YK, and Song JM (2007) Does the antibacterial activity of silver nanoparticles depend on the shape of the nanoparticle? A study of the Gramnegative bacterium Escherichia coli. Applied and Environmental Microbiology 73(6): 1712--1720. Pan JF and Wang WX (2002) Comparison of the bioavailability of Cr and Fe bound with natural colloids of different origins and sizes to two marine bivalves. Marine Biology 414(5): 915--924. Pantarotto D, Briand JP, Prato M, and Bianco A (2004) Translocation of bioactive peptides across cell membranes by carbon nanotubes. Chemical Communications (Cambridge) 1: 16--17. Papageorgiou I, Brown C, Schins R, et al. (2007) The effect of nano- and micron-sized particles of cobalt–chromium alloy on human fibroblasts in vitro. Biomaterials 28: 2946--2958. Paquin PR, et al. (2002) The biotic ligand model: A historical overview. Comparative Biochemistry and Physiology, Part C: Toxicology and Pharmacology 133(1–2): 3--35. Paradise M and Goswami T (2007) Carbon nanotubes – production and industrial applications. Materials and Design 28(5): 1477--1489. Park J, An K, Hwang Y, Park JG, Noh HJ, and Kim JY (2004) Ultra-large-scale syntheses of monodisperse nanocrystals. Nature Materials 3: 891--895. Park K (2007) Nanotechnology: What it can do for drug delivery. Journal of Controlled Release 120(1–2): 1--3. Pefferkorn E (1995) The role of polyelectrolytes in the stabilisation and destabilisation of colloids. Advances in Colloid and Interface Science 56: 33--104. Piccolo A (2001) The supramolecular structure of humic substances. Soil Science 166: 810--832. Poland CA, Duffin R, Kinloch I, et al. (2008) Carbon nanotubes introduced into the abdominal cavity of mice show asbestos-like pathogenicity in a pilot study. Nature Nanotechnology 3(7): 423--428. Poole CP and Owens FJ (2003) Introduction to Nanotechnology. Hoboken: WileyInterscience. Powers KW, Brown SC, Krishna VB, Wasdo SC, Moudgil BM, and Roberts SM (2006) Research strategies for safety evaluation of nanomaterials. Part VI. Characterization of nanoscale particles for toxicological evaluation. Toxicological Sciences 90(2): 296--303. Powers KW, Palazuelos M, Moudgil BM, and Roberts SM (2007) Characterization of the size, shape, and state of dispersion of nanoparticles for toxicological studies. Nanotoxicology 1(1): 42--51. Preining O (1998) The physical nature of very, very small particles and its impact on their behaviour. Journal of Aerosol Science 29(5–6): 481--495.

127

Pulskamp K, Diabate S, and Krug HF (2007) Carbon nanotubes show no sign of acute toxicity but induce intracellular reactive oxygen species in dependence on contaminants. Toxicology Letters 168: 58--74. Quinn J, Geiger C, Clausen C, Brooks K, Coon C, and O’Hara S (2005) Field demonstration of DNAPL dehalogenation using emulsified zero valent iron. Environmental Science and Technology 39: 1309--1318. Raiswell R, Tranter M, Benning LG, et al. (2006) Contributions from glacially derived sediment to the global iron (oxyhydr)oxide cycle: Implications for iron delivery to the oceans. Geochimica et Cosmochimica Acta 70: 2765--2780. Rees DA (1981) Polysaccharide shapes and their interactions – some recent advances. Pure and Applied Chemistry 43: 1--14. Reijnders L (2006) Cleaner nanotechnology and hazard reduction of manufactured nanoparticles. Journal of Cleaner Production 14(2): 124--133. Reijnders L (2008) Hazard reduction for the application to titania nanoparticles in environmental technology. Journal of Hazardous Materials 152: 440--445. Renn O and Roco MC (2006) Nanotechnology and the need for risk governance. Journal of Nanoparticle Research 8: 153--191. Rice JA and Lin JS (1993) Fractal nature of humic materials. Environmental Science and Technology 27(2): 413--414. Rice JA, Tombacz E, and Malekani K (1999) Applications of light and X-ray scattering to characterize the fractal properties of soil organic matter. Geoderma 88(3–4): 251--264. Robinson RD, Spanier JE, Zhang F, Chan SW, and Herman IP (2002) Visible thermal emission from sub-band gap laser excited cerium oxide particles. Journal of Applied Physics 92: 1936--1941. Roco MC (2005) International perspective on government nanotechnology funding in 2005. Journal of Nanoparticle Research 7: 707--712. Roder H, Hahn E, Brune H, Bucher JP, and Kern K (1993) Building one-dimensional and 2-dimensional nanostructures by diffusion-controlled aggregation at surfaces. Nature 366(6451): 141--143. Rogers NJ, Franklin NM, Apte SC, Batley GE, Lead JR, and Baalousha M (2010) Physico-chemical behaviour and algal toxicity of nanoparticulate CeO2 in freshwater. Environmental Chemistry 7(1): 50--60. Rotello VM (2004) Nanoparticles: Building Blocks for Nanotechnology, 1st edn. New York: Springer. Rothen-Rutishauser B, Grass RN, Blank F, et al. (2009) Direct combination of nanoparticle fabrication and exposure to lung cell cultures in a closed setup as a method to simulate accidental nanoparticle exposure of humans. Environmental Science and Technology 43: 2634--2640. Royal Society and Royal Academy (2004) Nanoscience and nanotechnologies: Opportunities and uncertainties. Ruiz AM, Sakai G, Cornet A, Shimanoe K, Morante JR, and Yamazoe N (2003) Microstructure control of thermally stable TiO2 obtained by hydrothermal process for gas sensors. Sensors and Actuators B 103: 312--317. Sader JE, Carnie SL, and Chan DYC (1995) Accurate analytic formulas for the doublelayer interaction between spheres. Journal of Colloid and Interface Science 171(1): 46--54. Sander S, Mosley LM, and Hunter KA (2004) Investigation of interparticle forces in natural waters: Effects of adsorbed humic acids on iron oxide and alumina surface properties. Environmental Science and Technology 38(18): 4791--4796. Santschi PH, Balnois E, Wilkinson KJ, Zhang J, and Buffle J (1998) Fibrillar polysaccharides in marine macromolecular organic matter as imaged by atomic force microscopy and transmission electron microscopy. Limnology and Oceanography 43: 896--908. Santschi PH, Lenhart JJ, and Honeyman BD (1997) Heterogeneous processes affecting trace contaminant distribution in estuaries: The role of natural organic matter. Marine Chemistry 58(1–2): 99--125. Sayes CM, Wahi R, Kurian PA, et al. (2006) Correlating nanoscale titania structure with toxicity: A cytotoxicity and inflammatory response study with human dermal fibroblasts and human lung epithelial cells. Toxicological Sciences 92(1): 174--185. Schmid G (2004) Nanoparticles: From Theory to Application, 1st edn. Weinheim: Wiley-VCH. Schrick B, Hydutsky BW, Blough JL, and Mallouk TE (2004) Delivery vehicles for zerovalent metal nanoparticles in soil and groundwater. Chemistry of Materials 16(11): 2187--2193. Schulthess CP and Huang CP (1990) Adsorption of heavy metals by silicon and aluminum oxide surfaces on clay minerals. Soil Science Society of America Journal 54(3): 679--688. Schulthess CP and Sparks DL (1989) Competitive ion exchange behavior on oxides. Soil Science Society of America Journal 53(2): 366--373.

128

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

Senesi N, Rizzi FR, Dellino P, and Acquafredda P (1996) Fractal dimension of humic acids in aqueous suspension as a function of pH and time. Soil Science Society of America Journal 60: 1613--1678. Senesi N, Rizzi FR, Dellino P, and Acquafredda P (1997) Fractal humic acids in aqueous suspensions at various concentrations, ionic strengths, and pH values. Colloids and Surfaces A: Physicochemical and Engineering Aspects 127(1–3): 57--68. Senesi N and Wilkinson K (2008) Biophysical Chemistry of Fractal Structures and Processes in Environmental Systems. Chichester: Wiley. Serpone N, Dondi D, and Albini A (2007) Inorganic and organic UV filters: Their role and efficacy in sunscreens and suncare products. Inorganica Chimica Acta 360: 794--802. Shahverdi AR, Kakhimi A, Shahverdi HD, and Minaian S (2007) Synthesis and effect of silver nanoparticles on the antibacterial activity of different antibiotics against Staphylococcus aureus and Escherichia coli. Nanomedicine: Nanotechnology, Biology and Medicine 3: 168--171. Sigg L (1994) Regulation of trace elements in lakes: Role of sedimentation. In: Buffle J and DeVitre RR (eds.) Chemical and Biological Regulation of Aquatic Systems, pp. 175--196. Boca Raton, FL: Lewis. Singh CP and Roy S (2004) Dynamics of all-optical switching in C-60 and its application to optical logic gates. Optical Engineering 43(2): 426--431. Singh S, Thiyagarajan P, Kant KM, Anita D, Thirupathiah S, and Rama N (2007) Structure, microstructure and physical properties of ZnO based materials in various forms: Bulk, thin film and nano. Journal of Physics D – Applied Physics 40: 6312--6327. Slaveykova VI and Wilkinson KJ (2005) Predicting the bioavailability of metals and metal complexes: Critical review of the biotic ligand model. Environmental Chemistry 2: 9--24. Smith CJ, Shaw BJ, and Handy RD (2007) Toxicity of single walled carbon nanotubes to rainbow trout, (Oncorhynchus mykiss): Respiratory toxicity, organ pathologies, and other physiological effects. Aquatic Toxicology 82: 94--109. Sondi I and Salopek-Sondi B (2004) Silver nanoparticles as antimicrobial agent: A case study on E. coli as a model for Gram-negative bacteria. Journal of Colloid and Interface Science 275(1): 177--182. Stankovich J and Carnie SL (1996) Electrical double layer interaction between dissimilar spherical colloidal particles and between a sphere and a plate: Nonlinear Poisson–Boltzmann theory. Langmuir 12(6): 1453--1461. Starowicz M and Stypula B (2008) Eectrochemical synthesis of ZnO nanoparticles. European Journal of Inorganic Chemistry 6: 869--872. Stoll S and Buffle J (1998) Computer simulation of flocculation processes: The roles of chain conformation and chain/colloid concentration ratio in the aggregate structures. Journal of Colloid and Interface Science 205(2): 290--304. Stoll S and Pefferkorn E (1996) Monte Carlo simulation of controlled colloid growth by homo- and hetero-coagulation in two dimensions. Journal of Colloid and Interface Science 177(1): 192--197. Stolpe B and Hassellov M (2007) Changes in size distribution of fresh water nanoscale colloidal matter and associated elements on mixing with seawater. Geochimica et Cosmochimica Acta 71(13): 3292--3301. Stordal MC, Santschi PH, and Gill GA (1996) Colloidal pumping: Evidence for the coagulation process using natural colloids tagged with 203Hg. Environmental Science and Technology 30(11): 3335--3340. Strycek T, Acreman J, Kerry A, Leppard GG, Nermut MV, and Kushner DJ (1992) Extracellular fibril production by freshwater algae and cyanobacteria. Microbial Ecology 23: 53--74. Stumm W (1992) Chemistry of the Solid-Water Interface: Processes at the MineralWater and Particle-Water Interface in Natural Systems. New York, NY: Wiley Interscience. Stumm W and Morgan JJ (1996) Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters. New York, NY: Wiley Interscience. Sun S, Zeng H, Robinson DB, Raoux S, Rice PM, and Wang SX (2004) Monodisperse MFe2O4 (M = Fe, Co, Mn) nanoparticles. Journal of American Chemical Society 126: 273--279. Sunda WG and Huntsman SA (1987) Microbial oxidation of manganese in a North Carolina estuary. Limnology and Oceanography 32: 552--564. Swift RS (1989) Molecular weight, size, shape, and charge characteristics of humic substances: Some basic considerations. In: Hayes MHB, McCarthy P, Malcolm RL, and Swift RS (eds.) Humic Substances. II. In Search of Structure, pp. 449--466. Chichester: Wiley. Tang P, Greenwood J, and Raper JA (2002) A model to describe the settling behavior of fractal aggregates. Journal of Colloid and Interface Science 247(1): 210--219. Tang P and Raper JA (2002) Modelling the settling behaviour of fractal aggregates – a review. Powder Technology 123(2–3): 114--125.

Tao AR, Habas S, and Yang PD (2008) Shape control of colloidal metal nanocrystals. Small 4: 310--325. The Royal Society and Royal Academy of Engineering (2004) Nanoscience and nanotechnologies: Opportunities and uncertainties, published 29 July 2004, http:// www.nanotec.org.uk/finalReport.htm Thill A, Zeyons O, Spalla O, Chauvat F, Rose J, and Auffan M (2006) Cytotoxicity of CeO2 nanoparticles for Escherichia coli. Physico-chemical insight of the cytotoxicity mechanism. Environmental Science and Technology 40: 6151--6156. Thurman EM (1985) Organic Geochemistry of Natural Waters. Norwell, MA: Kluwer Academic. Thurman EM and Malcolm RL (1981) Preparative isolation of aquatic humic substances. Environmental Science and Technology 15: 463--466. Thurman EM, Wershaw RL, Malcolm RL, and Pinckney DJ (1982) Molecular size of aquatic humic substances. Organic Geochemistry 27--35. Tipping E (2002) Cation Binding by Humic Substances. Cambridge: Cambridge University Press. Tipping E and Higgins DC (1982) The effect of adsorbed humic substances on the colloid stability of haematite particles. Colloids and Surfaces 5(2): 85--92. Tipping E, Woof C, and Cooke D (1981) Iron oxide from a seasonally anoxic lake. Geochimica et Cosmochimica Acta 45: 1411--1419. Toma D, Bertrand G, Chwa SO, Meunier C, Klein D, and Codder C (2006) Comparative study on the photocatalytic decomposition of nitrogen oxides using TiO2 coatings prepared by conventional plasma spraying and suspension plasma spraying. Surface and Coatings Technology 200: 5855--5862. Totten LA, Brunciak PA, Gigliotti CL, et al. (2001) Dynamic air–water exchange of polychlorinated biphenyls in the New York–New Jersey Harbor Estuary. Environmental Science and Technology 35(19): 3834--3840. Tratnyek PG and Johnson RL (2006) Nanotechnologies for environmental cleanup. Nano Today 1(2): 44--48. Tsuneda S, Jung J, Hayashi H, Aikawa H, Hirata A, and Sasaki H (2003) Influence of extracellular polymers on electrokinetic properties of heterotrophic bacterial cells examined by soft particle electrophoresis theory. Colloids and Surfaces B: Biointerfaces 29(2–3): 181--188. Valsami-Jones E, Berhanu D, Dybowska A, et al. (2008) Nanomaterial synthesis and characterization for toxicological studies: TiO2 case study. Mineralogical Magazine 72(1): 515--519. Van Hoecke K, et al. (2009) Fate and effects of CeO2 nanoparticles in aquatic ecotoxicity tests. Environmental Science and Technology 43(12): 4537--4546. van Leeuwen H and Koster W (2004) Physicochemical Kinetics and Transport at Biointerfaces. Chichester: Wiley. van Leeuwen HP, Town RM, Buffle J, et al. (2005) Dynamic speciation analysis and bioavailability of metals in aquatic systems. Environmental Science and Technology 39(22): 8545--8556. Veerapaneni S and Wiesner MR (1996) Hydrodynamics of fractal aggregates with radially varying permeability. Journal of Colloid and Interface Science 177(1): 45--57. Vikesland PJ, Heathcock AM, Rebodos RL, and Makus KE (2007) Particle size and aggregation effects on magnetite reactivity toward carbon tetrachloride. Environmental Science and Technology 41(15): 5277--5283. Walker HW and Bob MM (2001) Stability of particle flocs upon addition of natural organic matter under quiescent conditions. Water Research 35(4): 875--882. Wan J and Tokunaga TK (1997) Film straining of colloids in unsaturated porous media: Conceptual model and experimental testing. Environmental Science and Technology 31(8): 2413--2420. Wang Y and Herron N (1991) Nanometer-sized semiconductor clusters – materials synthesis, quantum size effects, and photophysical properties. Journal of Physical Chemistry 95(2): 525--532. Wang X, Zhuang J, Peng Q, and Li Y (2005) A general strategy for nanocrystal synthesis. Nature 437: 121--124. Warheit DB, Sayes CM, Reed KL, and Swain KA (2008) Health effects related to nanoparticle exposures: Environmental, health and safety considerations for assessing hazards and risks. Pharmacology and Therapeutics 120: 35--42. Warheit DB, Webb TR, Reed KL, Frerichs S, and Sayes CM (2007) Pulmonary toxicity study in rats with three forms of ultrafine-TiO2 particles: Differential responses related to surface properties. Toxicology 230(1): 90--104. Warheit DB, Webb TR, Sayes CM, Colvin VL, and Reed KL (2006) Pulmonary instillation studies with nanoscale TiO2 rods and dots in rats: Toxicity is not dependent upon particle size and surface area. Toxicological Sciences 91(1): 227--236. Watts PCP, Hsu WK, Randall DP, Kroto HW, and Walton DRM (2002) Non-linear current–voltage characteristics of electrically conducting carbon nanotube– polystyrene composites. Physical Chemistry 4: 5655--5662.

Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems

Wen LS, Santschi PH, and Tang D (1997) Interactions between radioactively labeled colloids and natural particles: Evidence for colloidal pumping. Geochimica et Cosmochimica Acta 61(14): 2867--2878. Wershaw RL (1999) Molecular aggregation of humic substances. Soil Science 164: 803--813. White HK, Xu L, Lima ALC, Eglinton TI, and Reddy CM (2005) Abundance, composition, and vertical transport of PAHs in marsh sediments. Environmental Science and Technology 39(21): 8273--8280. Wiesner MR, Lowry GV, Alvarez P, Dionysiou D, and Biswas P (2006) Assessing the risks of manufactured nanomaterials. Environmental Science and Technology 40(14): 4336--4345. Wiesner MR, Lowry GV, Jones KL, Hochella MF, Di Giulio RT, Casman E, and Bernhardt ES (2009) Decreasing uncertainties in assessing environmental exposure, risk, and ecological implications of nanomaterials. Environmental Science and Technology 43(17): 6458--6462. Wigginton NS, Haus KL, and Hochella MF (2007) Aquatic environmental nanoparticles. Journal of Environmental Monitoring 9: 1306--1316. Wilkinson K and Lead JR (2007) Environmental Colloids: Behaviour, Seperation and Characterisation. New York: Wiley. Wilkinson KJ, Balnois E, Leppard GG, and Buffle J (1999) Characteristic features of the major components of freshwater colloidal organic matter revealed by transmission electron and atomic force microscopy. Colloids and Surfaces A: Physicochemical and Engineering Aspects 155(2–3): 287--310. Wilkinson KJ, Joz-Rolland A, and Buffle J (1997a) Different role of pedogenic fulvic acids and aquagenic biopolymers on colloid aggregation and stability in freshwaters. Limnology and Oceanography 42: 1714--1724. Wilkinson KJ, Ne`gre J-C, and Buffle J (1997b) Coagulation of colloidal material in surface waters: The role of natural organic matter. Journal of Contaminant Hydrology 26(1–4): 229--243. Willis AL, Turro NJ, and O’Brien S (2005) Spectroscopic characterization of the surface of iron oxide nanocrystals. Chemistry of Materials 17: 5970--5975. Wittmaack K (2007) In search of the most relevant parameter for quantifying lung inflammatory response to nanoparticle exposure: Particle number, surface area or what? Environmental Health Perspectives 115: 187--194. Wolthoorn A, Temminghoff EJM, and van Riemsdijk WH (2004a) Colloid formation in groundwater by subsurface aeration: Characterisation of the geo-colloids and their counterparts. Applied Geochemistry 19: 1391--1402. Wolthoorn A, Temminghoff EJM, Weng L, and van Riemsdijk WH (2004b) Colloid formation in groundwater: Effect of phosphate, manganese, silicate and dissolved organic matter on the dynamic heterogeneous oxidation of ferrous iron. Applied Geochemistry 19: 611--622. Wu D, Liu J, Zhao X, Li A, Chen Y, and Ming N (2006) Sequence of events for the formation of titanate nanobelts. Chemistry of Materials 18: 547--553. Wu L, Wiesmann HJ, Moodenbaugh AR, et al. (2004) Oxidation state and lattice expansion of CeO2 – x nanoparticles as a function of particle size. Physical Review B 69: 125415-1–125415-9. Xu J, Dozier A, and Bhattacharyya D (2005) Synthesis of nanoscale bimetallic particles in polyelectrolyte membrane matrix for reductive transformation of halogenated organic compounds. Journal of Nanoparticle Research 7: 449--467. Xu HX, Li GS, and Li LP (2008) CeO2 nanocrystals: Seed-mediated synthesis and size control. Materials Research Bulletin 43: 990--995.

129

Xu S, Gao B, and Saiers JE (2006) Straining of colloidal particles in saturated porous media. Water Resources Research 42: W12S16.1--W12S16.10. Yamakoshi YN, Yagami T, Fukuhara K, Sueyoshi S, and Miyata N (1994) Solubilization of fullerene into water with polyvinylpyrrolidone applicable to biological tests. Journal of the Chemical Society, Chemical Communications 21(4): 517--518. Yang GCC, Tu HC, and Hung CH (2007) Stability of nanoiron slurries and their transport in the subsurface environment. Separation and Purification Technology 58: 166--172. Yeung AKC and Pelton R (1996) Micromechanics: A new approach to studying the strength and breakup of flocs. Journal of Colloid and Interface Science 184(2): 579--585. Yih TC and Al-Fandi M (2006) Engineered nanoparticles as precise drug delivery systems. Journal of Cellular Biochemistry 97: 1184--1190. Yoon KY, Byeon JH, Park JH, and Hwang J (2007) Susceptibility constants of Escherichia coli and Bacillus subtilis to silver and copper nanoparticles. Science of the Total Environment 373: 572--575. Yu JC, Yu J, and Zhao J (2002) Enhanced photocatalytic activity of mesoporous and ordinary TiO2 thin films by sulfuric acid treatment. Applied Catalysis B: Environmental 36: 31--43. Yu K, Lee W, Huang W, Wu C, Lou C, and Yang S (2006) Effectiveness of photocatalytic filter for removing volatile organic compounds in the heating, ventilation and air conditioning system. Journal of the Air and Waste Management Association 56: 666--674. Yuranova T, Laub D, and Kiwi J (2007) Synthesis, activity and characterization of textiles showing self-cleaning activity under daylight irradiation. Catalysis Today 122: 109--117. Zhang J and Buffle J (1996) Multi-method determination of the fractal dimension of hematite aggregates. Colloids and Surfaces A: Physicochemical and Engineering Aspects 107: 175--187. Zhang Z, Buffle J, and Alemani D (2007) Metal flux and dynamic speciation at (bio)interfaces. Part II: Evaluation and compilation of physicochemical parameters for complexes with particles and aggregates. Environmental Science and Technology 41(22): 7621--7631. Zhou JL, Liu R, Wilding A, and Hibberd A (2007) Sorption of selected endocrine disrupting chemicals to different aquatic colloids. Environmental Science and Technology 41(1): 206--213. Zhou J, Xu NS, and Wang ZL (2006) Dissolving behavior and stability of ZnO wires in biofluids: A study on biodegradability and biocompatibility of ZnO nanostructures. Advanced Materials 18: 2432--2435. Zhu HY, Lan Y, Gao XP, Ringer SP, Zheng ZF, and Song DY (2005) Phase transition between nanostructures of titanate and titanium dioxide via simple wet-chemical reactions. Journal of American Chemical Society 127: 6730--6736. Zimmermann-Timm H (2002) Characteristics, dynamics and importance of aggregates in rivers – an invited review. International Review of Hydrobiology 87: 197--240.

Relevant Websites http://www.nanotechproject.org The Project on Emerging Nanotechnologies.

3.06 Sampling and Conservation T Schulze, G Streck, and A Paschke, UFZ – Helmholtz-Centre for Environmental Research, Leipzig, Germany & 2011 Elsevier B.V. All rights reserved.

3.06.1 Introduction 3.06.2 General Aspects and Requirements of Sampling Environmental Waters 3.06.2.1 Objectives and Challenges of Sampling 3.06.2.2 Design of Sampling and Monitoring Programs 3.06.2.3 Types of Water Samples and Water-Sampling Methods 3.06.2.4 Sampling Site Selection and Sampling Frequencies 3.06.2.5 QC and QA 3.06.2.6 Safety Considerations 3.06.2.7 Standards, Guidelines, and Handbooks for Sampling of Water Samples 3.06.3 Handling and Conservation of Liquid Water Samples 3.06.3.1 Introduction 3.06.3.2 Alterations of Water Samples 3.06.3.3 Handling and Conservation of Water Samples 3.06.3.4 Selection of Sample Containers and Storage 3.06.4 Water Sampling Using Traditional Methods 3.06.4.1 Sampling of Surface Water 3.06.4.2 Sampling of Groundwater 3.06.5 Water Sampling Using Passive Sampling Technology 3.06.5.1 Introduction to Passive Samplers, Their Modeling, Calibration, and Quality Control 3.06.5.2 Passive Sampling of Nonpolar Organic Compounds 3.06.5.3 Passive Sampling of Polar Compounds Acknowledgments References

3.06.1 Introduction Sampling, in general, is an important prerequisite of a successful chemical, physical, or biological analysis. In terms of an integrated approach, sampling is a part of the analytical procedure and the beginning of the analytical chain. An analytical result cannot be better than the sample on which the analysis was performed due to the vulnerability of samples to artifacts, contamination, incorrect chemical treatment, and mislabeling during sampling and processing (Madrid and Zayas, 2007; Wilde, 2005). Thus, unskilled sampling and consecutive sample processing lead to imprecise, or in worst case, useless analytical results. Therefore, it is essential to implement and to carry out sampling programs and sampling with expertise. Keeping these considerations in mind, key issues and main challenges of sampling for scientific and regulatory purposes are (1) representativeness, (2) integrity of the samples collected, and (3) accuracy of the sampling procedures and design of the sampling or monitoring programs. The water samples require to be representative regarding the variability of each chemical, physical, or biological parameter observed in time and spatial distribution of the water body under investigation. It is important to ensure the integrity of the water samples, due to changes in the samples during collection, transport, and storage, prior to sample preparation and analysis. Design and implementation of sampling programs as well as selection of sampling methods done with

131 132 132 132 132 133 134 134 134 135 135 135 136 136 137 137 137 139 139 141 143 148 148

accuracy are essential to achieve the goals of representativeness and integrity of samples and to gain valid results for correct conclusions and further decision making. Furthermore, professional training and a regular auditing of the field personnel who is performing sampling is mandatory because of their responsibility to ensure a sampling in compliance with quality assurance (QA) standards. In the year 2000, the European Community established the Water Framework Directive (WFD) (European Community, 2000) to improve and protect the quality of all water bodies (including surface, transitional, coastal, and groundwaters) at river-basin level in Europe. The WFD demands an integrative river-basin management and policy making regarding (European Community, 2000):

• • • •

protection and enhancement of the biological and chemical status of aquatic environments; a sustainable water use based on long-term protection of water bodies; a significant reduction of pollution in surface waters and groundwaters; and achievement of background levels of naturally occurring compounds

In 2006, the European Groundwater Directive (GD) was enacted to establish specific measures for the prevention and control of groundwater pollution in accordance with Article

131

132

Sampling and Conservation

17 of the WFD (European Community, 2006a):

• • • •

criteria for the assessment of good groundwater chemical status; criteria for the identification and reversal of significant and sustained upward trends and for the definition of starting points for trend reversals; completion of the provisions for preventing or limiting inputs of pollutants into groundwater included in WFD; and prevention of the deterioration of the status of groundwater bodies.

Thus, the WFD and the GD enhance the requirements for the determination of target values and environmental quality standards (EQSs) regarding the sampling techniques and the objectives for the design of sampling programs. The aim of this chapter is to discuss the general scope, purposes, and requirements of the sampling process with a focus on water types as surface water (lakes, streams, and rivers), groundwater and wastewater. The sampling approaches reflect the requirements of environmental or chemical management programs such as the WFD, the GD, and the Registration, Evaluation, Authorization and Restriction of Chemicals (European Community, 2006b) including recent passive sampling techniques. In this chapter, surface water includes all standing or flowing waters on the surface of the inland (e.g., rivers, streams, and lakes), groundwater is the water present in the subsurface at the saturation zone, and a river basin is the land area where the surface- and groundwater run-off flows through a system of consecutive surface waters into the sea at a single river mouth, estuary, or delta (European Community, 2000).

3.06.2 General Aspects and Requirements of Sampling Environmental Waters 3.06.2.1 Objectives and Challenges of Sampling The objective of sampling is to collect a portion of water for chemical, physical, or biological analysis or other kind of testing in a specified procedure. A challenge is the volume of sample that should be small enough to be transported conveniently and handled in laboratory, while representing the part of environment sampled accordingly (American Health Association 2005; Kramer, 1994; Madrid and Zayas, 2007). Sampling includes the whole process of taking, storing, conserving, and transporting samples, since these are inherent parts of every sampling procedure. Furthermore, sampling is the starting point and integral part of the analytical process; hence, it can influence the accuracy of the analytical results significantly. The main challenges and key issues of sampling are (1) representativeness, (2) integrity of the samples collected, and (3) accuracy of the sampling procedures as well as the design of the sampling or monitoring programs.

3.06.2.2 Design of Sampling and Monitoring Programs The design of sampling and monitoring programs is a holistic approach that addresses the problem(s) of concern or reason(s) for sampling unambiguously and specifies the

objective(s) of the program and the indicators to be sampled clearly (Keith, 1990; Maher et al., 1994; Strobl and Robillard, 2008; Whitfield, 1988). However, many early water-monitoring programs have been designed without a consistent or logical design strategy resulting in failed studies and useless data collections (Maher et al., 1994; Strobl and Robillard, 2008), whereas today water-monitoring programs are practiced and are more focused to specific goals with selected variables or indicators (European Commission 2003a; Strobl and Robillard, 2008). In general, a minimum of required aspects should be considered during the design to establish an appropriate, effective, and representative monitoring program to achieve the monitoring objective(s) (Dixon and Chiswell, 1996; European Community, 2000; Maher et al., 1994; Strobl and Robillard, 2008; Whitfield, 1983):

• • • • • • • •

definition of monitoring goal(s) (scientific and regulatory); identification of suitable indicator(s) or parameter(s) (physical, chemical, and biological); selection of sampling sites (representativeness and reference conditions); choice of sampling strategy (number of samples, sampling frequency, and methods); verification of cost-effectiveness (comparison of alternative strategies, information gained versus cost of sampling); data analysis using computer-based tools (spatial and statistical analysis, models, and geographical information systems (GIS)); periodic review of sampling and quality control (QC) procedures as well as refinement and optimization in case of deficiencies; and periodic review of the adequacy of the monitoring goal(s).

Besides these general requirements, a monitoring program is river-basin specific and needs to be adapted, because of the diversity in catchment pressures (e.g., diffuse and point pollution sources), water body types, and hydromorphological and physicochemical characteristics of the river basin (European Commission 2003a; Strobl and Robillard, 2008). Regarding these considerations, the WFD (European Community, 2000) is exemplary for the establishment of monitoring programs and setup of monitoring guidelines at riverbasin level. The objectives for the establishment of sampling and monitoring programs for water bodies (i.e., groundwater and surface water) within the WFD were defined as follows (European Commission, 2003a):

• • • •

assessment of the environmental status of a water body and long-term changes in natural conditions, especially resulting from anthropogenic activity; estimation of pollution concentrations and loads, for example, to calculate trends or to monitor magnitude and impacts of accidental pollution; discovery of the reasons of water bodies’ failures to achieve EQSs where these are not identified; and ascertainment of reference conditions for water bodies.

Three types of monitoring strategies for the sampling and design of sampling programs were defined to achieve these

Sampling and Conservation

goals (Allan et al., 2006; European Community, 2000; Ru¨del et al., 2009):

•

• •

Surveillance monitoring. There is a provision of information and data to complete and validate impact-assessment procedures, to plan further monitoring activities, and to assess long-term changes of (aquatic) ecosystems. Operational monitoring. This is to classify the status of water bodies according to EQS and to estimate the changes of water bodies due to action plans. Investigative monitoring. It is undertaken at sites where surveillance monitoring shows that EQSs are not reached to identify the causes of the failure as well as to assess the magnitude and impact of accidental pollution events.

3.06.2.3 Types of Water Samples and Water-Sampling Methods The sampling strategy defined in the sampling or monitoring program and the aim of a scientific investigation itemize the types of samples to be taken and the method to achieve sampling. The types of samples as well as the water-sampling method should be considered and defined in the sampling program to ensure that the goals of the program can be achieved successfully. Basically, the characteristics of water and wastewater, the variability of water and wastewater flow, and the requirements of analysis are important factors for the decision on types of samples and sampling methods (Wardencki and Namies´nik, 2002). The collected types of water samples are (1) grab or spot samples and (2) composite samples (American Health Association, 2005; Kramer, 1994; Wardencki and Namies´nik, 2002):

•

•

Grab or spot samples. Single or discrete samples collected at a given location, time, and depth of water body that are representative only for the composition of the sampled medium during sampling time (usually seconds to minutes). Composite or integrative samples. Samples collected using pooled portions of grab samples or by using continuous collecting automated sampling devices and stored in one sample container that are representative for the average conditions during sampling period or samples obtained by means of passive samplers.

The different composite sampling types are defined as:

• •

•

Time-proportional composite sampling. Equal volumes of samples collected proportional to water flow at or during constant time intervals. Flow-proportional composite sampling. Samples that are taken proportional to the water flow either collecting an equal volume of sample in an interval depending on the volume of water passed the sampling point or by alteration of the volume of sample proportional to the flow at constant time intervals. Time-weighted sampling. Collecting samples continuously during an entire sampling time using the passive sampling technique.

The overall sampling methods can be divided into manual, automatic, and passive sampling techniques:

•

Manual sampling. Collection of discrete or composite samples using a hand device (e.g., bottle, scoop, bucket, or

•

•

133

hand pump); risk of errors are due to incorrect handling of the device. Automated sampling. Gathering of discrete or continuous samples using an automated device with a predefined program without human action; it eliminates human errors, but interferences with the equipment may occur (e.g., contamination by and sorption to the materials used in tubing, pumps, and collection vessels). Passive sampling. Time-weighted sampling technique for polar and nonpolar organic pollutants and metals based on the free diffusion (according to Fick’s first law of diffusion) of the analyte molecules from the sampled environment to a receiving phase with a very high affinity for the target analytes (Greenwood et al., 2009; Namies´nik et al., 2005).

The question of how to select between the types of samples and methods for the sampling depends on the goals of the sampling or monitoring program. During the implementation of WFD for the most water bodies in Europe, spot samples were identified to be suitable to achieve the goals of the monitoring programs, but in specific situations (e.g., flow conditions and temporal variations influence pollutant concentrations or calculations of pollution load are performed) more representative flow- or time-integrated sampling should be applied (European Commission, 2009). However, in recent reviews, the disadvantages and drawbacks of spot sampling regarding the monitoring strategies of integrative environmental protection programs such as the WFD were addressed (Greenwood et al., 2009; Madrid and Zayas, 2007):

•

•

•

spot water samples represent the water quality only at the moment of sampling and may not reflect the average water quality, particularly if the concentrations of pollutants fluctuate with space and time; significant errors can occur, especially when pollutants are present in only trace levels and large volumes must be collected for analysis, even with automated sampling methods; and measurement of truly or freely dissolved contaminants by most conventional approaches due to difficulties with the separation of this fraction (see Section 3.06.3.3).

Hence, spot water samples are representatives only if the source of contamination is essentially constant (Wardencki and Namies´nik, 2002). Nevertheless, in some cases spot sampling may be appropriate (e.g., sampling of protected groundwater wells and some well-mixed surface waters as well as for parameters needed to be analyzed in situ, e.g., pH, dissolved oxygen, and temperature) (American Health Association, 2005; Madrid and Zayas, 2007). For a more representative sampling, a higher frequency of sampling (i.e., automated sequential sampling) to provide composite samples and continuous online monitoring systems is recommended. Furthermore, the usage of passive sampling approaches discussed in Section 3.06.5 could be a good opportunity to tackle the indicated problems (Greenwood et al., 2009; Madrid and Zayas, 2007). Although they are not well recognized and established in official monitoring programs (Greenwood et al., 2009; Mills et al., 2009b), the international standardization regarding ISO 5667-23, ‘‘Determination of priority pollutants in surface water using passive sampling’’

134

Sampling and Conservation

and recommendations in the guidance documents for the implementation of WFD (European Commission, 2009) as well as the activities in the Network of Reference Laboratories for Monitoring of Emerging Pollutants (Mills et al., 2009b) enhance this process. Further development of validation schemes for QC and quality assurance methods and standards for the use of passive sampling devices as well as successful demonstration of sampler performance alongside conventional sampling schemes can help to convince regulators of the fact that passive samplers are alternative and cost-efficient instruments for water monitoring. Thus, the standardization will lead to a broader acceptance of passive sampling methods for regulatory and routine monitoring purposes.

3.06.2.4 Sampling Site Selection and Sampling Frequencies The sampling site can be defined as the area, transect, or stretch of a water body where samples are collected and the sampling or monitoring point(s) as the exact location(s) where samples are taken within a sampling site. The sampling frequency is the interval in which a sample is collected. The choice of sampling sites and sampling frequencies depends on the objectives and goals of the respective sampling or monitoring program. Nevertheless, the selection of the sampling sites and points has been rather often more a decision by personal judgment based on pragmatic considerations such as accessibility and safety than on determination of the optimal sampling site for each indicator under investigation (Dixon and Chiswell, 1996; Maher et al., 1994). In consequence, the goals of monitoring programs in terms of an integrative risk assessment demanded, for example, by the WFD may fail because of nonexistent consistence of sampling areas or points for different parameters such as metals, macroinvertebrates, and organic compounds (von der Ohe et al., 2009). In the Annex V of the WFD, minimal criteria for the selection of sampling points are listed for the monitoring of surface waters (European Community, 2000, 2009):

•

•

•

Surveillance monitoring. The sampling points should be located at major river stretches and at the downstream end of relevant subcatchments, advantageously with fixed monitoring stations and automated sampling to obtain continuous and composite samples. Operational monitoring. The sampling points should be located in order to allow assessment of the pressures of point or diffusion of pollution sources such that the magnitude and impact thereof is recorded sufficiently and representatively for the entire water bodies. Investigative monitoring. The sampling points should be located such that the goals of the program are achievable (e.g., unknown exceedance of EQS or magnitude and impact of an accident).

In practice, the selection of the exact sampling points including water depths for water sampling depends on local conditions such as vertical and lateral mixing, and homogeneity of the water body (European Commission, 2009). For an integrated assessment of the chemical and biological status, the requirements of biological monitoring should be considered for

sampling site selection as well (European Community, 2000; von der Ohe et al., 2009). The distance from, for example, effluents of wastewater treatment plants and the confluence with tributaries should be considered to quarantine the vertical and horizontal mixing and homogeneity of the water body sampled (e.g., using tracer method as described in detail elsewhere) (ISO 5667-6, 2005; Kramer, 1994). The selection of groundwater sampling points depends on the setup of the groundwater monitoring network that should be representative for each groundwater body under observation to allow the assessment of possible chemical pressures and their impact on the groundwater (European Community, 2000). However, the implementation of a groundwater monitoring network is more difficult in comparison to surface water monitoring due to the three-dimensional nature of the groundwater system and the spatial and temporal variability of the groundwater body (see also Chapter 2.06 Mechanics of Groundwater Flow) (e.g., hydrogeology, hydrology, connected surface ecosystems, and pollution pressures) (European Commission, 2007a). Another critical factor of sampling is the sampling frequency, because the confidence intervals of measures are a function of the numbers of samples taken (Dixon and Chiswell, 1996; Strobl and Robillard, 2008). Frequency of monitoring predominantly depends on the characteristics of the water body and the monitoring site. The Annex V of the WFD (European Community, 2000) provides tabulated sampling frequencies for different indicators and environmental quality elements in terms of minimal required intervals to assess the quality of a water body under investigation which were refined for the practical implementation of WFD (European Commission, 2003a, 2003b, 2007, 2009). In general, computer tools and models can be applied for the selection of optimized sampling sites and sampling frequencies in order to avoid the common practice of designing monitoring networks on subjective factors using, for example, Kriging theory, analysis of variance, least-squares methods, Fuzzy logic, and GIS (European Community, 2003b; Strobl and Robillard, 2008). Further guidance for the selection of sampling site(s), sampling frequencies, and number of samples can be found for surface waters in ISO 5667-6 (2005) and Wilde (2005) and for groundwater in ISO 5667-22 (2009) and Nielsen and Nielsen (2007a) as well as in Annex V of the WFD (European Community, 2000).

3.06.2.5 QC and QA QC is essential for the appropriate application of sampling technologies in water monitoring. In general, QA measures should be implemented throughout all procedures including preparation, handling (transportation, deployment, and retrieval), and storage and processing. The level of QC applied to sampling varies depending on the project objectives and procedures involved. QC/QA procedures for sampling are outlined in a vast number of publications for traditional and passive sampling methods (Vrana et al., 2005a; Huckins et al., 2000) as well as in different ISO guidelines (ISO 5667-10, 1992; ISO 5667-14, 1998; ISO 5667-11, 2009; ISO 5667-22, 2009; ISO/DIS 5667-23, 2009; see also Chapter 3.07 Measurement Quality in Water Analysis).

Sampling and Conservation 3.06.2.6 Safety Considerations Safety considerations are issues of different national and international laws and regulations regarding personnel safety precautions. However, collection of water samples has some elements of danger due to dangerous sampling sites (e.g., river banks, boats, weirs, and sewage treatment plants) and intrinsic toxicity of the water samples collected as well as hazardous substances (e.g., acids) which are used for conservation of the samples. Personal protective clothing should be worn the whole time during sampling and handling samples (e.g., disposable gloves, eye protection, laboratory coats, respirators, and coveralls) to protect contamination and damage with toxic substances caused by skin or eye passage, respiration, or swallowing. Also, the sampling sites should be selected under safety aspects and dangerous sites should be avoided wherever possible.

3.06.2.7 Standards, Guidelines, and Handbooks for Sampling of Water Samples Table 1 lists the most important International Organization for Standardization (ISO) and European Commission guidelines and recommended handbooks related to water sampling. The standard guidelines of ISO and related national organizations are well established and validated, and hence, they are widely accepted for regulatory and legislative purposes (see also Chapter 3.11 Standardized Methods for WaterQuality Assessment). In general, these guidelines contain the principles for the design of sampling programs and sampling techniques, and QA for the sampling of surface water, groundwater, and wastewater. The National Field Manual for the Collection of Water-Quality Data of the US Geological Survey (USGS, 2008) and Standard Methods for the Examination of Water and Wastewater (American Health Association, 2005) are comprehensive manuals for the sampling and analysis of waters. The guidelines of the European Community reflect the recent developments and discussions regarding the common implementation strategy for the European WFD and are available free of charge from the website of European Communities. The Essential Handbook of Ground-Water Sampling (Nielsen and Nielsen, 2007a) and the handbook on passive sampling techniques in environmental monitoring (Greenwood et al., 2007a) contain techniques for groundwater and passive sampling in detail.

3.06.3 Handling and Conservation of Liquid Water Samples 3.06.3.1 Introduction The pretreatment of collected water samples is commonly necessary before chemical analysis due to loss of analytes, contamination, and other alterations, for example, change of pH, temperature, and dissolving of gases. This processing is the consecutive step after sampling and includes the composition, subsampling (splitting), preservation, and shipment of samples and depends on the target analytes and the intended purpose of the sampling campaign or program (Wilde et al., 2004). If the samples are not processed immediately during or after sampling (e.g., the measurement of pH and oxygen), a long-term stabilization or preservation is

135

Table 1 List of guidelines and recommended handbooks regarding sampling of water samples Guidance/handbook

Reference

Water quality – sampling – part 1: Guidance on the design of sampling programmes and sampling techniques Water quality – sampling – part 3: Guidance on the preservation and handling of water samples Water quality – sampling – part 4: Guidance on sampling from lakes, natural and manmade Water quality – sampling – part 6: Guidance on sampling of rivers and streams Water quality – sampling – part 10: Guidance on sampling of wastewaters Water quality – sampling – part 11: Guidance on sampling of groundwaters Water quality – sampling – part 14: Guidance on quality assurance of environmental water sampling and handling Water quality – sampling – part 20: Guidance on the use of sampling data for decision making – compliance with thresholds and classification systems Water quality – sampling – part 22: Guidance on the design and installation of groundwater monitoring points Water quality – sampling – part 23: Determination of priority pollutants in surface water using passive sampling National field manual for the collection of water-quality data (US Geological Survey) Common Implementation Strategy for the Water Framework Directive – Guidance document No. 7: Monitoring under the Water Framework Directive Common Implementation Strategy for the Water Framework Directive – Guidance document No. 11: Planning process Common Implementation Strategy for the Water Framework Directive – Guidance document No. 15: Guidance on groundwater monitoring Common Implementation Strategy for the Water Framework Directive – Guidance document No. 16: Guidance on groundwater monitoring in drinking water protection areas Common Implementation Strategy for the Water Framework Directive – Guidance document No. 19: Guidance on surface water chemical monitoring under the Water Framework Directive Passive Sampling Techniques in Environmental Monitoring Standard Methods for the Examination of Water and Wastewater The Essential Handbook of Ground-Water Sampling

ISO 5667-1, 2006

ISO 5667-3, 2003

ISO 5667-4, 1987

ISO 5667-6, 2005 ISO 5667-10, 1992 ISO 5667-11, 2009 ISO 5667-14, 1998

ISO 5667-20, 2008

ISO/DIS 5667-22, 2009 ISO/DIS 5667-23, 2009 USGS, 2008 European Commission, 2003a

European Commission, 2003b European Commission, 2007a

European Commission, 2007b

European Commission, 2009

Greenwood et al., 2007a American Health Association, 2005 Nielsen and Nielsen, 2007a

136

Sampling and Conservation

recommended to protect the samples against changes of physical and chemical properties as well as the disposal of each component interfering with analytes and analytical procedures from the sample. This section is related to liquid water samples. Section 3.06.5 deals with the description of handling of passive samplers.

3.06.3.2 Alterations of Water Samples A main challenge of the sampling procedure is to prevent the alteration of the water samples with respect to their natural character. The major possible alterations of the natural character of water samples are (Parr et al., 1988; USGS, 2008; Wardencki and Namies´nik, 2002) as follows:

• •

Physical–chemical processes. Oxidation, reduction, hydrolysis, (de-)polymerization, volatilization, adsorption, absorption, diffusion, photo-degradation, precipitation, and degassing. Biochemical processes. Consumption of compounds by organisms (e.g., bacteriae or algae) due to their growth including formation of metabolic compounds.

Some basic chemical and physical properties are sensitive to quick changes after collection of sample, for example, temperature, pH, and dissolved gases (e.g., oxygen and carbon dioxide). Many organic compounds as well as metals and inorganic analytes such as nitrogen and phosphate species are sensitive to changes of pH; volatiles and dissolved gases may degas; and ions and hydrophilic and organic compounds may adsorb to surfaces of glass bottles. Changes in the reduction– oxidation potential can lead to precipitation or dissolution and other alterations of substances (e.g., iron and manganese). Biological activity in the samples is a reason for a wide range of changes. It influences, for example, the oxidation state of some compounds, and some substances (e.g., nitrogen, phosphorus, and organic compounds) are potential growing substrates for microbes and get lost. If possible, analyze the following parameters immediately in situ to avoid significant changes (see also Chapter 3.01 Sum Parameters: Potential and Limitations):

• • • • • • • • •

temperature; pH; specific conductance; turbidity; alkalinity; reduction–oxidation potential; dissolved gases; color; and odor.

and bottles, maximum elapsed time between sampling and analysis, and pretreatment (e.g., filtration, addition of a preservation agent). These requirements depend on the properties of the sample and analytes to be determined. The properties of samples may be different among the sample species (groundwater, surface water, and wastewater) and among the sample locations. These differences should be considered while planning the sampling and selection of methods for sampling, handling, preservation, and storage. Conservation methods such as cooling or adding of a preservation agent may be applied to avoid alterations of the natural character thereof. The pretreatment of the samples may include filtration of the samples to avoid, for example, sorption of metals and organic compounds to suspended particulate matter (SPM) or interactions between dissolved fraction and particulate-bound fraction due to addition of a preservative. Commonly used membrane filters, for example, for metals, have a pore size of 0.4–0.45 mm, and glass-fiber filters often used for organic compounds have a nominal particle size retention of 0.7– 1.2 mm with a similar SPM retention capacity comparing with membrane filters. However, the pore-size limits are an operational and somewhat arbitrary limit to separate SPM and the dissolved fraction. Because in this concept the role of colloids – which occur in a diameter ranging from 1 nm to 1 mm (see also Chapter 3.05 Natural Colloids and Manufactured Nanoparticles in Aquatic and Terrestrial Systems) – and other dissolved organic carbon (DOC) – which is present in natural surface and groundwaters and may not be retained by the filter – is completely ignored (Danielsson, 1982; Horowitz et al., 1996; Kramer, 1994; see also Chapter 3.01 Sum Parameters: Potential and Limitations). Nevertheless, if the filtration in situ is difficult or impossible, the addition of preservatives can cause an undesirable change of equilibrium between dissolved and particulate-bound fractions, for example, of metals, and hence, the filtration and possible addition of preservatives should be done as soon as possible when reaching the laboratory (Kramer, 1994). If the sampling container is used to collect the sample, for example, by immersing it in the water, the preservative must be added after sampling to avoid the loss. The chemicals used for preservation should have the bestavailable quality and should also be controlled for their blanks to avoid any contamination. Table 2 gives an overview of suitable conservation methods for different groups of analytes. The preservation agents should be added initially to the sample containers. If there is concern of interference of some determinants with used agents, then splitting of the sample is required to avoid interactions between the preservation agents.

3.06.3.3 Handling and Conservation of Water Samples

3.06.3.4 Selection of Sample Containers and Storage

The cooling of samples should start as soon as possible, for example, using ice or commercially available cooling packs or storing in refrigeration systems. Dry ice ought to be avoided due to freezing of the sample, breaking of glass beakers, and a possible change of pH in the sample (American Health Association, 2005). It is useful to consult the analyzing laboratory to determine the requirements for handling, preservation, and storage of the samples, such as the choice of containers

Besides the choice of preservation agent, the selection of the sample containers for transport and storage is of major importance for the integrity of the sample (American Health Association, 2005; Wilde et al., 2004). The containers and the cap-liners should be made of a material, which is appropriate to keep the sample in its natural condition and for the conservation of the sample as well as for the expected range of analytes. Fluorinated polymers (e.g., polytetrafluoroethylene

Sampling and Conservation

137

Table 2 Suitable conservation methods for water samples (American Health Association 2005; Jeannot, 1994; Wardencki and Namiesnik, 2002; Wilde et al., 2004) Conservation technique

Preservation agent

Determinant

Untreated Refrigeration (2–5 1C)

NA NA

Acidification to pHo2

HCl HNO3

Anions Acidity and alkalinity, BOD, bromides and bromine compounds, chlorophyll, chromium (VI), color, COD, conductivity, cyanide, DOC, iodide, iodine, nitrate, nitrite, odor, oil and grease, organic compounds (e.g. pesticides, PAH, PCB, VOC, organochlorine and organophosphorus pesticides, base/neutrals and acids), pH, phenols, phosphorus, silicates, suspended solids, sulfate, sulfide, surfactants, TOC, turbidity Hydrocarbons, oil and grease, mercury Cations, metalloids (antimony, arsenic, and selenium), oil and grease, phosphorus (total), total hardness, trace elements Ammonia, boron, COD, free and ionized ammonia, total hardness, Kjeldahl nitrogen, oil and grease, organic chlorine, permanganate index, phosphorus (total), surfactants (anionic), TOC Phenoxylalkanoic herbicides Phenols, phenol index Cyanide Nonionic surfactants Sulfide

H2SO4

Acidification to pHo4 Alkalinize to pH412 Addition of other agents

Formic acid H3PO4 or H2SO4 NaOH Formaldehyde Zinc acetate solution

(PTFE)) and quartz or borosilicate hard glass such as Pyrexs or Durans are the best materials for most inorganic determinants and organic compounds, respectively. However, high-density polyethylene bottles are convenient for inorganic compounds (Batley and Gardner, 1977) and often used due to the high costs of fluorinated polymers. Polyethylene (PE) is not suitable for organic compounds (Barcelona et al., 1985; House, 1994), because these may absorb to the PE or polymer compounds such as softeners and antioxidants can leach from the PE walls. Soft glass containers ought to be avoided due to leaching of constituents from the glass wall, for example, silica, sodium, and boron. Use only light-adsorbing hard glass containers with PTFE cap-liners for photosensitive organic compounds. Silanization of the glassware can be suitable for the determination of hydrophilic organic compounds such as bisphenol A, alkylphenols, and some pesticides, which may adsorb the glass surfaces. Containers should be pre-leached and cleaned in laboratory using suitable agents (e.g., phosphate-free detergents, acids, organic solvents, and ultra-cleanwater) to avoid any contamination. In situ pre-rinsing with the sample is generally not recommended to avoid loss of preadded preservative and through potential biasing of the results due to compounds that adhere to the walls of the sample container (e.g., SPM) (American Health Association, 2005; House, 1994). If there is intention to preserve and store samples by freezing, the containers should consist of HDPE or PTFE to avoid breakage. Table 3 summarizes the recommendations for containers for some common groups of analytes.

3.06.4 Water Sampling Using Traditional Methods 3.06.4.1 Sampling of Surface Water The selection of a representative sampling site or sampling point is one of the main challenges of the setup of a sampling or monitoring program. The principles of sampling site

selection were discussed above. A main problem of sampling point selection is probably the heterogeneity of the water body. The distribution and fate of chemicals in the water body are controlled by diffusion and advection (Schwarzenbach et al., 2003). Hence, the concentration of a pollutant may change throughout the water body resulting in a heterogenic vertical, horizontal, and lateral distribution of the pollutant (Artiola, 2004; Whitfield, 1983). A comprehensive guideline as to how to handle the problems of sampling site selection and how to obtain representative samples even from heterogenic sampling locations can be found elsewhere (Wilde et al., 2004). The sampling equipment is selected considering the type of water (e.g., lake and river), the sample requirements (e.g., size and analytes), and the objectives of the sampling program. The used sampler must be without risks of sample contamination by the construction materials of the samplers (sorption and/or release of compounds) and from previous use for sampling in other water bodies (memory effects) to avoid analytical artifacts (Keith, 1990). Different sampling devices were developed for (1) spot sampling, (2) sampling from specific depths, and (3) automatic sampling. Spot sampling is often done manually using bottles, scoops, or buckets immersed directly to the water body in a depth of 0.5–1 m as well as using depth-water samplers for stratified water bodies (Artiola, 2004; House, 1994). Bottles can be immersed to collect a surface sample using an extendable telescope rod with a holder for the bottle. There are commercially available different types of depth-water samplers (Table 4). The depth samplers are lowered down into the water using metered tag ropes and either opened or closed at the predefined water depth according to their sampling principle. Manual sampling devices are used, for example, from a boat, a bridge, or the river bank. Automated sampling devices are available in many different configurations for stationary and mobile operation. The sampling devices may be programmed and configured to sample discrete as well as flow- and time-proportional

138

Sampling and Conservation

Table 3

Recommended containers, maximum storage times, and typical sample volumes for water samples for exemplary water analytes

Determinant

Sample container

Alkalinity–acidity

P,GB P,GB Anions P,G P P P Biological oxygen demand (BOD) P,G P,G Cations (except Cr6þ, Hg) P,G P Chemical oxygen demand (COD) P,G P,G Cr6þ P(A),G(A) P Mercury (Hg) P(A),G(A) GB G Cyanide P,G GB P P Nitrogen and phosphorus (nutrients) P,G P

Storage time

Typical sample size (ml) Reference

24 h 24 h 48 h–28 days

100 4 100–200 125–500 100–200 250 1000 1000 1000 250 100 100 1000 500 1000 250

24 h–48 h 24 h 24 h 6 months 7 days 24 h 24 h 28 days 1 month 24 h 48 h

7–28 days 48 h

Chlorinated hydrocarbons Dioxins and furans

Ga (PTFE) Ga, PTFE-LC Ga, PTFE-LC 7 days

Dissolved organic carbon Hydrocarbons Nitroaromatics

G G Ga, PTFE-LC Ga Ga, PTFE-LC G, wide-mouth

Nitrosamines Oil and grease

Organochlorine pesticides Organophosphorus pesticides Organo-nitrogen herbicides Pesticides

Phenols

Polynuclear aromatic compounds Polychlorinated Biphenyls Total organic carbon (TOC)

Volatile organic compounds

G G Ga, G Ga, Ga, Ga Ga, G

28 days 15 days 7 days

PTFE-LC 1 day PTFE-LC PTFE-LC PTFE-LC

P,G, PTDE.LC G Ga, PTFE-LC Ga Ga Ga, PTFE-LC Ga Ga, PTFE-LC GB Ga, PTFE-LC G G Ga

1000 100 250 1000 500 100 125 1000 125 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000

7 days 1000 1000 2000 1000 500 o24 h–28 days 1000 500–1000 1000 7 days 1000 1000 7 days 1000 1000 7 days 100 40 125 14 days 40 7 days 7 days

American Health Association, 2005 Wardencki and Namiesnik, 2002 American Health Association, 2005 Nielsen and Nielsen, 2007b Wardencki and Namiesnik, 2002 Wilde et al., 2004 American Health Association, 2005 Wardencki and Namiesnik, 2002 American Health Association, 2005 Wilde et al., 2004 American Health Association, 2005 Wardencki and Namiesnik, 2002 American Health Association, 2005 Nielsen and Nielsen, 2007b American Health Association, 2005 Wardencki and Namiesnik, 2002 Wilde et al., 2004 American Health Association, 2005 Wardencki and Namiesnik, 2002 Wilde et al., 2004 Nielsen and Nielsen, 2007b American Health Association, 2005 Wardencki and Namiesnik, 2002 Wilde et al., 2004 Nielsen and Nielsen, 2007b Koester and Clement, 1993 Wardencki and Namiesnik, 2002 Wilde et al., 2004 Nielsen and Nielsen, 2007b Nielsen and Nielsen, 2007b Wilde et al., 2004 Nielsen and Nielsen, 2007b American Health Association, 2005 Wardencki and Namiesnik, 2002 Wilde et al., 2004 Jeannot, 1994 Nielsen and Nielsen, 2007b Jeannot, 1994 Nielsen and Nielsen, 2007b Koester and Clement, 1993 Wilde et al., 2004 American Health Association, 2005 Wardencki and Namiesnik, 2002 Wilde et al., 2004 American Health Association, 2005 Jeannot, 1994 Nielsen and Nielsen, 2007b Wilde et al., 2004 Jeannot, 1994 Nielsen and Nielsen, 2007b Jeannot, 1994 Nielsen and Nielsen, 2007b American Health Association, 2005 Nielsen and Nielsen, 2007b Wilde et al., 2004 Jeannot, 1994 Wilde et al., 2004

P, polyethylene; G, glass; GB, borosilicate glass; PTFE-C, Teflons-lined caps; Ga, amber borosilicate glass; P(A)/G(A), acid rinsed P or G.

Sampling and Conservation Table 4

139

Depth sampling devices for surface waters

Principle

Sampler name

Sampling volume (l)

Spot/depth sampler

Water-sediment bottle Ruttner sampler MERCOS sampler LIMNOS sampler modified Van Dorn sampler

1–2 1–2 2 1.4–7

samples. The water is delivered to the systems by peristaltic, syringe, membrane, cogged submersible, and eccentric submersible pumps (Wardencki and Namies´nik, 2002). Flowproportional sampling machines are connected to flow- or discharge-gauging systems and measure the flow in the water body to take a water sample after a predefined volume of water that has passed the sampling point or to take event-controlled samples, for example, during flood events. Time-proportional sampling devices take a certain volume of sample after a predefined time interval, for example, a volume of 20 ml taken every 15 min over 24 h to obtain a 24-h composite sample of approximately 2 l. The samples are delivered to sample bottles inside the machines using internal distribution systems and frequently cooled at 4 1C. The automatic devices are normally equipped with an automatic dosing system which always obtains a constant volume of water (Wardencki and Namies´nik, 2002). For more details regarding the principles of water-sampling techniques and sampling of surface waters, the reader is referred to Sturgeon et al. (1987), Wardencki and Namies´nik (2002), Wilde et al. (2004), and the respective ISO guidelines (ISO 5667-4, 1987; ISO 5667-14, 1998; ISO 5667-6, 2005; ISO 5667-1, 2006).

3.06.4.2 Sampling of Groundwater The methods for sampling of groundwater are mainly different from surface water sampling. The main challenges are the time- and cost-intensive accessing of groundwater, that is, drilling, construction, and maintenance of observation and monitoring wells, and the three-dimensional black-box character of the groundwater bodies (i.e., lack of visibility of the subground and its inherent hydrogeological, geochemical, and biological characteristics) (Nielsen, 2007b; Roy and Fouillac, 2004). Therefore, the planning and installation of a groundwater-monitoring network and groundwater sampling demands a high grade of experience and professionalism to achieve the goals of the monitoring or sampling program successfully. Thus, this section only gives a short overview regarding different types of observation and monitoring wells as well as samplers used for groundwater sampling. In hydrogeology, three types of wells are defined: (1) water supply wells, (2) observation wells, and (3) monitoring wells (Dalton et al., 2007; Wilde, 2005). The first well type is used for extraction of water for drinking, industrial process, and irradiation purposes, and usually equipped with a dedicated high-capacity pump; the second well type is used to collect hydrogeological data for observation of the aquifer characteristics and often equipped with piezometers; the third well

Water depths (m)

Reference

o30

Artiola, 2004 Majaneva et al., 2009 Freimann et al., 1983 Majaneva et al., 2009 Finucane and May, 1961

type is used for observation of physical, chemical, and biological characteristics of the aquifer. The wells for groundwater monitoring are installed using a variety of drilling techniques and well-construction methods, for example, driven well points and nested multilevel wells in separated or single boreholes to observe shallow, unconfined saturated zones, and multiple saturated zones (Barackman et al., 2004; Dalton et al., 2007; Lerner and Teutsch, 1995). For the sampling of the groundwater-monitoring wells, there are different devices commercially available (Barackman et al., 2004; Nielsen, 2007a):

•

•

•

•

Grab samplers. Open and point-source bailers, thief samplers, and syringe samplers; lowered down in the boreholes, particularly equipped with valves and mechanical, electrical, or pneumatic triggers to open/close valves or plugs to obtain samples in a certain depth. Suction-lift pumps. Peristaltic and surface centrifugal pumps; vacuum pumps operated at the ground surface connected to the sampler tubing; problematic due to alteration of the samples due to vacuum (e.g., degassing). Submersible positive displacement samplers. Gas-displacement pumps, bladder pumps, piston pumps, helical rotor pumps, gear-drive pumps, inertial-lift pumps; pumps submersed in groundwater table and using different mechanical concepts to lift the water to surface. Passive samplers. Passive diffusion samplers are used for groundwater sampler as well; the different types of passive samplers are discussed in Section 3.06.5.

The interested reader is directed to Barackman et al. (2004), Nielsen and Nielsen (2007a) and the respective ISO guidelines (ISO 5667-1, 2006; ISO/DIS 5667-22, 2009, ISO/DIS 566723, 2009) for further information regarding groundwater well construction, well development, and sampling of groundwater.

3.06.5 Water Sampling Using Passive Sampling Technology 3.06.5.1 Introduction to Passive Samplers, Their Modeling, Calibration, and Quality Control Passive samplers have been used in environmental monitoring since the beginning of the 1970s. The early designs were used to measure concentrations of gaseous pollutants in air, but since the late 1980s, passive samplers have been developed for monitoring concentrations of pollutants in water, soils, and sediments (Greenwood et al., 2007a). The same principles of operation apply to all passive sampler devices, for use in both

140

Sampling and Conservation

Curvilinear

Equilibrium

St

Linear

In the latter case, transport of samples from field to laboratory becomes very expensive. A range of passive samplers has been developed for monitoring environmental pollutants from a range of chemical classes, including metals, polar organics, nonpolar organics, organometallics, and volatile organics. Several reviews (Go´recki and Namies´ek, 2002; Vrana et al., 2005b; Stuer-Lauridsen, 2005; Seethapathy et al., 2008; Zabiegala et al., 2010) and the worldwide first anthology (Greenwood et al., 2007a) summarize these achievements. Sections 3.06.5.2 and 3.06.5.3 describe several of these devices and their applications more detailed for nonpolar and polar organic target compounds, respectively. Passive sampling of metal ions is not covered in this chapter; the readers are referred to the relevant literature (Greenwood et al., 2007b; Warnken et al., 2007; Mills et al., 2009a). While all the samplers have the same basic components, their structural configurations, handling properties, ease of use, and performance are widely different. They all have strengths and weaknesses, and it is important to select the sampler most appropriate for each particular problem to be investigated. For all passive samplers, uptake rates depend to a greater or lesser extent on temperature and turbulence of the water (Booij et al., 2007). The rate of uptake can also be affected by the growth of microorganisms (biofouling) on the surface of the diffusion limiting membrane. Further, all depend on the availability of reliable calibration data for the pollutants of interest. For samplers used for monitoring nonpolar pollutants, performance reference compounds (PRCs) have been developed. These are compounds (typically deuterated analogs of the compounds to be measured) that are loaded onto the receiving phase of the sampler prior to deployment, and that offload at a measurable rate. If the kinetics of uptake and offloading are isotropic, that is, the offloading rates of the PRCs are affected by temperature, turbulence, and biofouling in a manner similar to the uptake rates of pollutants, then the rates of loss of PRCs from the sampler can be used to correct the uptake rates of pollutants for the effects of those environmental variables. This approach can effectively provide in situ calibration of the samplers.

R

Amount sorbed / water concentration

air and water. Uptake of a chemical from the environment is by molecular diffusion. The samplers comprise a receiving phase that accumulates contaminants, and has a very high affinity for them so that the concentration at its surface is maintained close to zero, and a diffusion-limiting layer that separates the receiving phase from the bulk water environment. Hence, the mass of a contaminant accumulated is determined by its concentration in the water (more precisely, the gradient of its chemical potential between bulk water phase and receiving phase), the exposure time, and the sampling rate (Rs) of the device (i.e., an overall compound- and samplerspecific uptake rate constant). In case of diffusion-based samplers where the solutes diffuse through capillaries (pores) to the collecting phase, Rs is a product of the compound’s diffusion coefficient in water, the exchange surface area with the surrounding (i.e., the total area of capillary/pore openings), and the inverse of the diffusion path length. For permeation-based samplers, where a nonporous membrane encloses the receiving phase, the membrane/water partition coefficient is included as additional factor in Rs (Go´recki and Namies´ek, 2002). Most passive samplers measure only concentrations of freely dissolved analytes and not the total amount of analytes present in the water column. Fractions bound to SPM or to DOC are not measured or irrelevant due to either their exclusion by the diffusion-limiting layer, slow transport through it, or poor uptake by the receiving phase. In all passive samplers, the mass accumulated is used to determine the external concentration, but depending on sampler design and mode of operation, this can reflect either the equilibrium concentration or the time-weighted average concentration over the deployment period (days to months) (Figure 1). Where environmental concentrations fluctuate with time the kinetic samplers are used, but in more constant or slowly changing conditions, the equilibrium samplers are deployed. Since the samplers accumulate substances over a prolonged period the analytes are effectively pre-concentrated, and this can bring them above the level of quantification of the analytical method. It would be necessary to collect and extract large volumes of water in order to achieve a comparable sensitivity with spot sampling.

Time Figure 1 Passive samplers mostly operate in the linear uptake phase or in equilibrium with the surrounding water phase.

Sampling and Conservation

However, PRCs have not been developed for samplers for monitoring polar organics or inorganics (Mills et al., 2007). In the following, basic mathematical equations are given for the description of the uptake of compounds into a passive sampler. A first model of this process was derived by Huckins and co-workers at the US Geological Survey in the early 1990s for the semipermeable membrane device (SPMD) in analogy to those describing the dynamic exposure of aquatic organisms. The modeling approach can easily be applied to other passive samplers such as the membrane-enclosed sorptive coating (MESCO) device or the polar organic integrative sampler (POCIS) (see Sections 3.06.5.2 and 3.06.5.3). A more detailed discussion of the theory of passive sampling can be found in the literature (Booij et al., 2007; Huckins et al., 2006). The passive solute uptake in the receiving phase of a permeation-based sampler can be described by a first-order kinetics according to

   kov A a mSðtÞ ¼ m0 þ ðcW KSW VS  m0 Þ 1  exp  t KSW VS

ð1Þ

where mS(t) is the amount of compound accumulated in the sampler after the exposure time t, m0 the amount already in the sampling phase before exposure (blank, background level), cW the concentrations in water, KSW the partition coefficient between sampling phase and water, VS the volume of the sampling phase, kov the overall mass-transfer coefficient into the sampler, A the available exchange area (membrane surface), and a the membrane porosity (a ¼ 1 for nonporous membranes or if a membrane is not present). The coefficient in the exponential function in Equation (1) can be combined to an overall exchange constant kex, whereby the product in the numerator is usually named as sampling rate RS:

kex ¼

kov A a RS ¼ KSW VS KSW V S

ð2Þ

Substance enrichment follows approximately a linear trend until half-life of accumulation t1/2 is reached:

t1=2 ¼

ln 2 kex

ð3Þ

The passive sampler accumulates integrative up to this point in time; hence, it is possible to derive (from Equation (1)) an expression for the time-weighted average (TWA) concentration in the aqueous medium monitored:

CTWA ¼ W

mðtÞ  m0 RS  t

ð4Þ

This value is the primary goal of passive sampler application in water monitoring. For longer exposure times, when the sampler reaches distribution equilibrium with surrounding water phase, the term in square brackets in Equation (1) approximately equals 1 and the aqueous concentrations Ceq W are calculated according to

Ceq W ¼

mðtÞ  m0 KSW  VS

ð5Þ

141

The result of Equation (5) cannot be considered as a TWA concentration because it is not to deduce when equilibrium was reached and thus the analyte amount accumulated does not necessarily reflect all fluctuations in concentration during the complete sampling period. Instead, the value calculated from Equation (5) provides a snapshot of the concentration representative for the equilibrium period. For more details on the principles of equilibrium samplers, the reader is referred to Mayer et al. (2003). Other more sophisticated mathematical models are possible, depending on the degree of simplifying assumptions that are taken into account. However, this simple model based on the film theory for interfacial mass transfer and the inherent assumption of the additivity of mass-transfer resistances (Cussler, 1984) is able to describe the occurring phenomena (i.e., the data from field trials) reasonably well and serves in the design and evaluation of laboratory calibration experiments. Such calibration studies under controlled conditions are necessary because the sampling rate RS is a complex parameter depending not only on the aforementioned substance-specific coefficients and sampler properties/dimensions, but is also influenced by environmental factors such as water temperature, hydrodynamic conditions, and biofilm growth on the surface of the sampler. Partly, this also holds for KSW as the only specific factor in Equation (5). Different laboratory experiments are possible for passive sampler calibration and model equations to evaluate them mathematically given by Booij et al. (2007). The static-exposure design where the passive samplers are exposed to a single volume of contaminated water is used usually for determination of sampling rates under more quiescent conditions (e.g., relevant for sampler deployment in lakes or groundwater). Another batch-wise approach is to work with a periodic renewal of the exposure water. Here, several flow regimes can be realized by adjusting stirring/shaking. Paschke et al. (2006) demonstrated the feasibility of such serial batch tests for rapid MESCO calibration. A third option for RS determination in the laboratory is the continuous flow design where depletion of the water phase in the exposure chamber is prevented by a constant supply of freshly contaminated water. Also in such experiments, different conditions (flow regime, temperature, DOC, salt concentrations, etc.) can be arranged. Vrana et al. (2001a) used long glass columns with slowly upstream flowing contaminated water for SPMD and MESCO calibration. Furthermore, Vrana et al. (2006a) designed and performed continuous flow experiments in a tank with the samplers placed on a rotating carousel (allowing higher flow rates) for calibrating the Chemcatchers device. A similar design is applied successfully in short-term exposure testing of six different passive samplers for monitoring hydrophobic contaminants in spiked river water (Allan et al., 2010). However, it will not be possible to simulate natural exposure conditions for the passive samplers in their full diversity and complexity. As already mentioned above, a promising approach is the in situ calibration of passive samplers by using PRCs. The dissipation rate constants of these preloaded compounds from the receiving phase during field deployment of the sampler can be estimated and used for deriving sampling rates under site-specific field conditions.

142

Sampling and Conservation

The PRC approach is applicable at least for receiving phases that consist of a liquid or a nonpolar polymeric film where the dominant uptake process is absorption into the bulk phase, but not adsorption to the surface of the material. A recently reported study on the field performance of seven passive sampling devices for monitoring hydrophobic substances demonstrates this in situ calibration approach (Allan et al., 2009).

3.06.5.2 Passive Sampling of Nonpolar Organic Compounds In the last two decades, various passive sampling devices were designed for monitoring nonpolar pollutants in the aquatic systems. Especially, the determination of persistent organic pollutants is of relevance due to their tendency to bioaccumulate and their high toxic potential. These pollutants are present in the aquatic environment both dissolved and particle-bound (due to their hydrophobicity). Of primary interest for risk assessment is the bioavailable fraction which corresponds to the dissolved fraction (for most of exposure routes). With conventional sampling techniques (e.g., grab sampling), only the total content of the pollutants is obtained. Furthermore, analyses of grab samples provide information about very hydrophobic organic pollutant burden only if large-volume water samples (410 l) are processed. In this section, only those passive sampling devices are considered that are commercially available or at least under commercial development. Such a stage of technical maturation implies that (reliable) calibration data are generated for several classes of compounds and made available for the enduser and that the field applicability of the device is demonstrated. More detailed information about the variety of more recently reported passive sampler types for water monitoring, including field application of solid-phase microextraction fibers, can be found in periodically appearing review papers (e.g., Go´recki and Namies´nik, 2002; Vrana et al., 2005a; StuerLauridsen 2005; Ouyang and Pawliszyn 2007; Seethapathy et al., 2008; Zabiegala et al., 2010). SMPD. Among the permeation-based samplers, the socalled SPMDs, introduced by Huckins and co-worker at US Geological Survey in the early 1990 (Huckins et al., 1990, 1993, 2006), attained the greatest importance and widespread application. A SPMD consists of lay-flat low-density polyethylene (LDPE) tubing enclosing a thin film of triolein. SPMDs are proved to be most effective in their capacity to accumulate lipophilic contaminants. Commercially available standard SPMDs are used in many field campaigns for sampling of, for example, chlorinated hydrocarbons, polycyclic aromatics, and organochlorine pesticides in different aqueous environments (e.g., freshwater and marine systems, and groundwater wells). The main disadvantage of this sampler type is, besides the relatively high price, the complex sample preparation required to recover the accumulated pollutants from the collecting phase (triolein). This is usually achieved by dialysis using considerable amounts of organic solvents, followed by pre-concentration, solvent exchanges, and cleanup of the extracts before the chromatographic analysis (Petty et al., 2000; Huckins et al., 2000; Wenzel et al., 2004). A compilation of SPMD sampling rates is given in Huckins et al.

(2006). Aspects of field application of SPMDs are also discussed there as well as by Bergqvist and Zaliauskiene (2007). In the last decade several attempts have been made to develop devices, which avoid the drawbacks with SPMDs and make the passive sampling technology more attractive also for routine monitoring programs. Single-phase polymer films such as LDPE (Booij et al., 2003) or silicone strips (Smedes, 2007; Yates et al., 2007) and rods (Paschke et al., 2006) can be used directly to avoid the triolein traces in the extracts. However, for the processing of these polymer materials, considerable amounts of organic solvents are necessary in most cases. Data evaluation is not straightforward due to the stronger influence of hydrodynamics on the substance uptake over the uncovered surface of the sampling phase, especially if PRCs are not used. PBD. For monitoring volatile (nonpolar) organic compounds in groundwater, the most widely used device is the passive diffusion bag (PDB) developed by Vroblesky at the US Geological Survey (Vroblesky, 2007). The major advantages are its simplicity (a protected PE bag filled with water) and the possibility to use the conventional headspace gas chromatography for analysis. However, one has to bear in mind that this is an equilibrium sampling device which does not provide TWA concentrations. Information on PDB handling, processing, and data interpretation can be found in Vroblesky (2001a). Many applications are described and are also reported (Vroblesky, 2001b, 2007). Other samplers (for semi-/low-volatile hydrophobic organics) contain solid materials (granular adsorbents and compact polymeric sorbents) instead of a liquid organic receiving phase. This allows simpler extraction procedures or even thermo-desorption of the accumulated pollutants without previous sample preparation. Grathwohl and co-workers at the University of Tu¨bingen, Germany, for example, designed a dosimeter for the integrative sampling of organic compounds in groundwater (Martin et al., 2001; WeiX et al., 2007). This sampler consists of a porous ceramic tube which can be filled with different grained adsorbents, for example, with ion-exchange resin Amberlite IRA-743 or Tenax, and was tested for monitoring several polynuclear aromatic hydrocarbons (PAHs) in groundwater (Martin et al., 2003; Bopp et al., 2005). Concerning the subsequent thermo-desorption of the analytes from Tenax difficulties appear due to the (unexpected) water permeability of the ceramic tube. Also, the long lag phase and thus response time of the sampler can be a problem when exposed in aqueous systems with more fluctuating concentrations of target compounds. Chemcatcher(R). The so-called Chemcatchers developed by Greenwood and co-workers at the University of Portsmouth, UK, is another promising passive sampling device (Kingston et al., 2000; Vrana et al., 2006a, 2006b; Greenwood et al., 2007b). The nonpolar Chemcatchers version consists of a PTFE body containing a C18 EmporeTM disk as receiving phase. A 40-mm-thick LDPE disk (47 mm diameter) of diffusion-limiting membrane is placed on the top of the receiving phase. The PTFE body part supports both the C18 EmporeTM disk and the LDPE membrane and sealed them in place. Currently, an optimized design of the Chemcatchers body is under investigation at the University of Portsmouth

Sampling and Conservation

which is simpler to use and more cost efficient than the original PTFE body. Conditioning and extraction of receiving phase is carried out in most steps according to protocols for the conventional solid-phase extraction of water samples with C18 EmporeTM disks (see Vrana et al. (2006a, 2006b) for specific modifications). There are sampling rates with the Chemcatchers device available for many priority organic pollutants. Both temperature- and flow-dependent (Vrana et al., 2006a, 2006b, 2007) field applications are also performed successfully with this flexible sampler configuration (Vrana et al., 2007; Greenwood 2007b; Allan et al., 2009). MESCO. A decade ago, another promising sampler for nonpolar aquatic micropollutants was developed at the Helmholtz Centre for Environmental Research – UFZ Leipzig, Germany. Vrana et al. (2001b) described the use of coarse pieces of silicone-based sorbent material as collecting phase of a passive sampler, which is enclosed in a membrane bag during field exposure and can be retrieved lostless for the following processing. Different types of the membrane-enclosed sorptive coating (MESCO) device are tested meanwhile for time-weighted average (TWA) sampling of organic compounds in water (Paschke et al., 2007). MESCOs can consist of different types of silicone-collecting phases. Commercial TwisterTM bars (PDMS-coated stir bars from GERSTEL, Mu¨lheim a.d.R., Germany), silicone tubes, and silicone rods are tested so far. In some MESCO applications, cellulosemembrane bags were applied to envelope the receiving phase (Vrana, 2001b, 2006b), whereas in others cellulose was replaced by an LDPE because it has proved to be more stable to biodegradation in the field (Wennrich et al., 2003; Paschke et al., 2006). The advantages of using silicone tubes and rods are their nonfragility, low costs, and flexibility in processing with minimal solvent consumption (Van Pinxteren et al., 2010). Current investigations deal with optimization of the used membrane thickness and material (Paschke et al., 2007). MESCO sampling rates for PAHs, polychlorinated biphenyls (PCBs), and selected organo-chlorine pesticides can be found together with discussion of optional instrumental processing variants in the original papers mentioned above. Examples of MESCO field applications are also reported (Vrana et al., 2006b; Paschke et al., 2006, 2007; Allan et al., 2009). Finally, a recently published report on a field study in the River Meuse, The Netherlands, should be mentioned (Allan et al., 2009). There, the field performance of seven passive sampling devices for monitoring dissolved concentrations of PAHs, PCBs, hexachlorobenzene, and p,p0 -DDE was evaluated through simultaneous field exposure of 7–28 days. Data generated by the Chemcatchers, LDPE membranes, two versions of the MESCO sampler, silicone rods, silicone strips, and SPMDs were evaluated through PRC dissipation data, analyte masses absorbed by various samplers, and the comparison of TWA concentration data. Despite different modes of calculation, relatively consistent TWA concentrations are obtained from different samplers. The variability observed is likely due to the uncertainty of sampler–water partition coefficients and the extrapolation of analyte uptake rates at the high n-octanol–water partition coefficient (KOW) range from a narrow PRC data range. These issues are further investigated (see, Allan et al. (2010) for first attempt in this matter).

143

3.06.5.3 Passive Sampling of Polar Compounds The following section describes passive sampling of polar organic compounds (POCs). With the development of analytical chemistry, especially the progress in mass spectrometric techniques combined with liquid chromatography, POCs gained increasing attention in the last few years (Richardson, 2008, 2009). It was realized that POCs are frequent contaminants in aquatic ecosystems (Loos et al., 2009; Richardson, 2007) since POCs that are not readily degradable can enter rivers via wastewater treatment plants (Hewitt and Marvin, 2005; Reemtsma et al., in press; Reemtsma et al., 2006). Furthermore, POCs can contribute significantly to the whole toxicity of water or sediments (Bandow et al., 2009; Biselli et al., 2005; Pomati et al., 2006; Streck, 2009a). Different approaches exist to classify a compound as polar or nonpolar, either by considering their electric-dipole moment, whether functional groups are present that could undergo particular interactions with other molecules, or by means of their KOW as a simple but somewhat arbitrary discriminator (Schwarzenbach et al., 2003). Usually, many pesticides, especially herbicides, pharmaceuticals as well as organic wastewater contaminants are counted among POCs. Two special passive samplers have been developed to sample these compounds, that is, the polar organic chemical integrative sampler (POCIS) and the Chemcatchers. POCIS. POCISs consist of a sorbent enclosed by membrane disks made of polyethersulfone (poly(oxy-1,4-phenylsulfonyl1,4-phenyl, PES)). The microporous membrane, with a typical pore size of 0.1 mm and a thickness of 132 mm, allows water and dissolved substances to pass but excludes solid material such as SPM. Commercially available material used also for solid-phase extraction is employed as sorbents. Such material has usually particle sizes between 5 and 100 mm. In order to prevent the sorbents from being washed away, they are placed between two PES disks forming a sandwich (Figure 2). PES is not amenable to heat sealing, which is the standard sealing technique for other membranes used for passive samplers (e.g., PE for SPMDs). Therefore, another way had to be found to enclose the sorbents reliably. Devices that are commercially available consist of two stainless steel rings serving as compression holders. Other inert material (e.g., aluminum or PTFE) not interfering with the sampling process is suitable (Alvarez et al., 2004). The membrane sandwich with the sorbents is placed between two metal compression holders, which are then tightened by nuts and bolts. In principle, a wide variety of sorbents is suitable for sampling POCs. However, ready-to-use POCISs are available on the market in two different configurations, either with a triphasic admixture or with Oasiss HLB (EST Inc., St. Joseph, USA) (Alvarez et al., 2007). The triphasic admixture is composed of 80% of weight of the polystyrene divinylbenzene resin Isolute ENV þ and of 20% of AmbersorbTM 572 carbon dispersed on S-X3 BiodeadsTM. First POCISs were introduced with AmbersorbTM 1500 instead of AmbersorbTM 572; however, the first one is not produced any more. AmbersorbTM 572 is found to be an equivalent substitute. POCIS with the triphasic admixture, marketed under the name AQUASENSE-P with pesticide configuration, has been applied for sampling a variety of different compound classes. This includes not only

144

Sampling and Conservation Bolt hole

Compression ring Sorbent PES membrane

Figure 2 Setup of POCIS; the sorbent is enclosed by an upper and lower membrane hold by compression rings and fixed by bolts and nuts.

pesticides, but also steroidal hormones or compounds released from municipal wastewater treatment plants (Table 5). Oasiss HLB consists of a co-polymer (poly(divinylbenzolco-N-vinylpyrrolidon)) and has been found especially suitable for sequestering pharmaceuticals. POCISs with this type of sorbent are marketed as AQUASENSE-P with pharmaceutical configuration. Pharmaceuticals often have multiple functional groups and are therefore able to bind strongly to the carbon of the triphasic admixture. This can lead to relative low recoveries of pharmaceuticals from POCIS with the triphasic admixture (Alvarez et al., 2007). Further, a commonly used sorbent with POCIS is strataTM-X, which is a surface-modified styrene divinylbenzene polymeric sorbent suitable for a wide range of acidic, basic, and neutral compounds. POCISs with strataTM-X have, for example, been used to collect pesticides, wastewater-related compounds, or steroidal hormones (Cefas, 2007; Streck, 2009b). The membrane of a standard POCIS in contact with the water phase covers 41 cm2, which thus is the effective sampling surface area. POCIS with the triphasic admixture or with Oasiss HLB exhibits a surface area to sorbent mass ratio of 180 cm2 g1 while this reported ratio for POCIS with strataTMX is approximately 150 cm2 g1. Chemcatchers . The Chemcatchers-type passive sampler consists of a solid sorbent disk as receiving phase and membranes that cover the disk and govern the uptake of the compounds. Both are usually mounted on a sampler body made of PTFE (Kingston et al., 2000). Disks with different sorbents embedded in a polymeric matrix (PTFE) are commercially available under the trademark EmporeTM. Chemcatchers can target different classes of compounds depending on the sorbent disk and the choice of membranes. Sorbent–membrane combinations have been tested, for example, for hydrophobic compounds, for metals, as well as for POCs (Greenwood et al., 2007a). Table 6 contains examples of passive samplers with EmporeTM disks and different membranes employed for sampling of POCs. Sorbent disks without membranes were used by several authors for short-term sampling. Membranes slow down the uptake of compounds and influence the transfer velocity of compounds from the water phase to the sorbent (Figure 3). Thus, they extend the period in which the receiving phase has not yet reached its sorption capacity and continues linear uptake (kinetic regime). Time-weighted average

concentrations can be derived from the amount of chemical trapped on the receiving phase as long as the passive sampling device stays in the kinetic regime. Without membranes, the linear uptake occurs for a period of less than 10 days (Stephens et al., 2005, 2009), while membranes made of polysulfone or PES enhance this time to at least 21 or 30 days (Shaw et al., 2009; Tran et al., 2007). Besides affecting the period of the kinetic regime, membranes function as a protection of the sorbent disks from biofouling (Harman et al., 2009a; Macleod et al., 2007) and decrease the sensitivity of the system to changes in flow velocity (Booij et al., 2007) as well as to changes in concentrations of compounds. Field and lab procedures. POCIS and Chemcatchers are deployed in the field often within a protective canister which is made of perforated stainless steel plates (Figure 4). The protective canister deflects debris that otherwise may damage the membranes of the passive samplers. POCISs are deployed beneath the surface water typically for a period between a week and few months. Alvarez and co-workers reported that a reliable determination of water concentrations is still possible with sampling times of 56 days (Alvarez et al., 2004; JonesLepp et al., 2004). Deployment time of Chemcatchers depends on whether the EmporeTM disks are covered with membranes. Naked disks allow sampling time for a maximum period of 7–14 days (Table 6 and Figure 3). Transport from and to the lab into the field and sample storage should be done in cooled and protected, for example, airtight containers (Alvarez et al., 2007). For Chemcatchers, the transport in water-filled plastic zip-bags has been reported (Scha¨fer et al., 2008a) in order to keep the preconditioned EmporeTM disks wet. The next step after retrieval from the water phase is to dismount the passive sampler and separate the receiving phase from the membranes (Figure 5). For POCIS, it is recommended to clean the body of the sample holder and the exterior of the membranes prior to the disassembly using clean water and a soft brush in order to prevent debris, organisms, or particulate matter which could contaminate the sorbent material. This step is not necessary for the Chemcatchers, since the receiving phase is solidly embedded in the polymer matrix of the disk. Loose sorbent material retrieved from POCIS is flushed into a glass with suitable solvents, for example, methanol. Targeted compounds can then be eluted using an appropriate solvent. For the triphasic admixture a solvent mixture of methanol, toluene, and dichloromethane

Sampling and Conservation Table 5

Application of POCIS with different receiving phase

Receiving phase

Class of compounds

Reference

Triphasic admixture

Steroids (e.g., 17a-ethynylestradiol) Pesticides (e.g., triazines, organophosphates, and chloroacetanilides) Pesticides (e.g., triazines) Musk compounds (e.g., galaxolide, tonalide, and traseolide) Other wastewater related compounds (e.g., alkyl phosphates, phthalates, and N,N-diethyltoluamide) Alkylated phenols Estrogenic activity (YES-assay) Alkylphenols Alkylphenolethoxylate Bisphenol A Steroids UV-filters (e.g., benzophenone-4, benzophenone-3, 3-(4-methylbenzylidene)camphor, 2-ethyl-hexyl-4-trimethoxycinnamate) Steroids (e.g., 17a-ethinylestradiol, estriol, 17b-estradiol, and estrone) Wastewater related compounds (e.g., caffeine) Steroids (e.g., estrogenic and androgenic steroids) Pesticides (e.g., triazines, organophosphates, and thiocarbamates) Pharmaceuticals Pesticides Wastewater related compounds Bioanalysis (YES-assay) Pharmaceuticals (e.g., azithromycin ,fluoxetine, levothyroxine, and omeprazole) Alkylated phenols Azaarenes Polycyclic aromatic compounds Alkylated phenols Cresols Pharmaceuticals (e.g., amitriyptiline, doxepine, imipramine, and carbamazepine) Alkylphenols Alkylphenolethoxylate Bisphenol A Steroids Pharmaceuticals Antibiotics Wastewater-related compounds Pharmaceuticals Pesticides Antibiotics Wastewater-related compounds Pharmaceuticals Wastewater-related compounds Pesticides Pesticides (eg., terbutylazine, diethylterbuthylazine, isoproturon) Steroids (eg., 5a-dihydrotestosterone (DHT), estrone) Pharmaceuticals (eg., flutamide, tamoxifen) Pesticides (prometryn)

Alvarez et al., 2004, 2008

Oasiss HLB

StrataTM-X

145

Alvarez et al., 2009

Harman et al., 2009a Burki et al., 2006 Arditsoglou and Voutsa, 2008

Zenker et al., 2008

Sellin et al., 2009

Sellin et al., in press

Petty et al., 2004

Alvarez et al., 2004 Harman et al., 2008a

Harman et al., 2009b Togola and Budzinski, 2007 Arditsoglou and Voutsa, 2008

Bartelt-Hunt et al., 2009

Bueno et al., 2009

Macleod et al., 2007 Mazzella et al., 2007 Mazzella et al., 2008 Cefas, 2007

Rotter et al., unpublished data

146 Table 6

Sampling and Conservation Passive sampling devices for polar compounds containing an EmporeTM disk as receiving phase

Receiving phase (EmporeTM disk)

Membranes

Targeted class of compounds

Deployment timea,b

Reference

C18 SDB-XC C18

Polysulfone PES None

o9 days o21 days o7 day

Kingston et al., 2000 Tran et al., 2007 Stephens et al., 2005

SDB-RS

PES/none

3.2–10.3 daysc

Stephens et al., 2009

SDB-RS

PES/none

o30 days/o10 days

Shaw et al., 2009

SDB-XC SDB-XC SDB-XC

PES/none None PES/none

1 day and 9 days 10–13 days 6/30 days

Scha¨fer et al., 2008a Scha¨fer et al., 2008b Vermeirssen et al., 2009

SDB-RPS

None

6 days

Vermeirssen et al., 2009

SDB-XC

None

Nonionic herbicides Nonionic and ionic herbicides Herbicides (atrazine, diuron, hexazinone, flumetoron) Herbicides (diuron, atrazine, simazine) Herbicides (tebuthiuron, hexazinone, simazine, atrazine, diuron, ametryn, metolachlor) Pesticides (fipronil and chlorpyrifos) thiacloprid Pesticides and herbicides Pharmaceuticals, herbicides and pesticides Pharmaceuticals, herbicides and pesticides Herbicides and pesticides

14 days

Gunold et al., 2008

a

with/without membrane. time at which linear uptake was observed. c the capacity of samplers without PES-membrane was probably reached at 7.2 days. b

in the ratio of 1:1:8 has frequently been applied while for the processing of Oasiss HLB methanol has been used (Alvarez et al., 2005). Disks from Chemcatchers are processed for extraction either in an ultrasonic bath (Kingston et al., 2000; Shaw et al., 2009), by using a vacuum manifold (Tran et al., 2007) or by simply shaking the sorbent disks in a glass using solvents. Using an ultrasonic extraction method can detach particles of sorbent from the disk; therefore, a filtration step after extraction is recommended (Tran et al., 2007). The membranes are usually not extracted; however, it should be mentioned that membranes can contain a substantial fraction to the total amount of chemicals collected (Tran et al., 2007). Especially when the sampling time is short, an erroneous determination of water concentrations can occur. Extracts obtained from the passive samplers can then undergo the usual steps for chemical or biological analysis. Effect of environmental conditions on uptake. Conditions in environmental waters differ, for example, in temperature, pH, salinity, and flow velocity. These conditions can affect the uptake rates of compounds in passive samplers. Thus, the dependency of uptake rates from these conditions has to be determined usually in special laboratory or field experiments before passive samplers can be employed quantitatively. In the following, an overview is given how environmental conditions influence passive sampling. Temperature. Water temperature can have a large effect on the uptake of compounds into passive samplers. Membrane– compound partition coefficients are temperature dependent besides diffusion rates or sorption constants. This effect has been investigated by several research groups, since temperature changes occur regularly in the environment, for example, during the course of the year or when wastewater enters a river. In general, an enhanced water temperature leads to faster uptake; however, the effect is strongly compound dependent. Togola

and Budzinski (2007) reported that for ketoprofen the uptake rate into POCIS almost doubled, whereas for other pharmaceuticals such as carbamazepine uptake remained constant when the temperature was enhanced by 6 1C. For POCIS with strataTM-X as sorbent, an increase of up to a factor of 2.6 was found when the temperature was raised from 13 to 20 1C (Streck, 2009b). Mainly, musk compounds were considered in this study. However, bisphenols were unaffected by a temperature change. In a study using Chemcatchers, an Arrheniustype relationship was observed for a temperature range between 4 and 20 1C, and the uptake of atrazine and diuron increased with rising temperatures (Kingston et al., 2000). There is no model today to calculate compound-specific changes in uptake rates caused by variation of the water temperature. Therefore, calibration studies at different temperatures are highly important to adjust to varying conditions in the field (So¨derstro¨m et al., 2009). pH. POCs often possess many functional groups and can be present in the environment in ionized or neutral form, based on the pH. Physicochemical properties of the compounds are changing when a neutral compound becomes ionized or vice versa. Hence, influence of pH on uptake into passive samplers has to be considered. To date, only little information is available to what extent the pH value affects the uptake of compounds into POCIS or Chemcatchers. Only one study is known to the author in which the pH value was varied: Zhang et al. (2008) determined uptake rates for estrone, 17b-estradiol, 17a-ethynylestradiol, and bisphenol A, which have pKa values of 10.8, 10.5, 10.7, and approximately 10.0, respectively (Lewis and Archer, 1979; Staples et al., 1998). Uptake rates were assessed for pH values between 4 and 10, but no differences were found. Therefore, Zhang et al. (2008) concluded that both neutral and ionized forms were accumulated equally. However, since the pH value was lower

Sampling and Conservation

147

ng sampler −1

600

400

Diu

Ame

Sim

Teb

Atr

Met

Hex

Fip

200

0 0

5

10

(a)

15 Days

20

25

30 Figure 4 POCIS and protective canister ready for deployment.

ng sampler −1

1500

1000

Diu

Ame

Sim

Teb

Atr Hex

Met

between 0 and 35 g l1, salinity will probably play a minor part in most studies. Flow velocity. Uptake of compounds into a passive sampler is governed by several resistances to mass transfer which occurs through the water boundary layer, through the membrane, and within the sorbent itself. Work by Alvarez et al. (2004) noted that the aqueous boundary layer controls the uptake into POCIS, and the sampling rate RS can be written as

Fip

500

0 0 (b)

5

10

15 Days

20

25

30

Figure 3 Uptake of herbicides in Chemcatcher with (a) and without (b) a PES-membrane; with membranes the uptake remains linear for at least 30 days. Without membranes a quasi-linear uptake occurs for several days. Lines show (a) linear regression and (b) second order regression fit. Diu, diuron; Sim, simazine; Atr, atrazine; Hex, hexazinone; Ame, Ametryn; Teb, tebuthiuron; Met, metolachlor; Fip ¼ fipronil. Reproduced from Shaw M, Eaglesham G, and Mueller JF (2009) Uptake and release of polar compounds in SDB-RPS EmporeTM disks; implications for their use as passive samplers. Chemosphere 75: 1–7, with permission from Elsevier.

or just equal (for bisphenol A) to the pKa values, ionized forms of the investigated substances were only partly addressed with the study design. Salinity. Salinity can affect the sampling efficiency of passive samplers by enhancing the water ionic strength which decreases the hydrophilicity of compounds (Togola and Budzinski, 2007; Harman et al., 2008a). Another reason for reduced uptake rates with higher salinity can be due to the salting-out effect: With an increase of ionic strength, the solubility of compounds is exceeded and they are no longer available for passive sampling (Huckins et al., 1999). In an experiment varying salinity of water between 0 and 35 g l1 of salt, Togola and Budzinski (2007) found a decreased uptake into POCIS especially for basic pharmaceuticals. Neutral and acidic pharmaceuticals were not affected. Thus, variation in uptake rates due to salinity changes is compound dependent, and in fact not fully investigated yet. However, in the abovementioned study that lasted 7 days, the decrease in sampling rates amounted to maximal 64% for imipramine. Since only few applications of POCIS see differences in the salt content

RS ¼ A

  DW dW

ð6Þ

where A is the sampler surface, DW the diffusion coefficient of the compound in water, and dW the thickness of the aqueous stagnant film layer. With increasing flow velocity, the thickness of the aqueous layer diminishes which should lead to a faster uptake till the mass transfer is limited by the diffusion within the membrane or the sorbent. Such behavior has already been demonstrated for other passive samplers, for example, SPMDs (Huckins et al., 2006; Vrana and Schu¨u¨rmann, 2002). MacLeod et al. (2007) calibrated pharmaceutical type POCIS for three different flow velocities between 3 and 12 cm s1 and for quiescent conditions. Uptake rates showed no significant differences in the calibration experiments under flow conditions. This could indicate that already at 3 cm s1 aqueous boundary layer control no longer limits the mass transfer. For a part of the compounds, significant differences were found between flowing and quiescent conditions, which is in agreement with studies from other researchers (Alvarez et al., 2004). The increase in uptake rates for these compounds ranged from a factor of 3.1 for the pharmaceutical carbamazepine to 10.4 for biocide triclosan. However, it should be mentioned that the calibration experiment with quiescent water was operated at a temperature 6 1C lower, which partly could be responsible for lower uptake rates. The independence of uptake rates for nominal flow velocities between 13 to 51 cm s1 was also demonstrated for POCIS with strataTM-X (Streck, 2009b). A similar picture can be drawn for Chemcatchers. Both Gunold et al. (2008) and Vermeirssen et al. (2009) investigated the flow dependency for samplers with a naked EmporeTM SDB-XC disk. While Gunold observed no differences in uptake rates at two velocities (13.5 and 40 cm s1) in spiked tap water,

148

Sampling and Conservation

Exterior cleaning Disassembling and retrieval of sorbent

Transport to lab in airtight can/glass

Extraction of sorbent Retrieval of POCIS

Chemical analysis

Bioanalysis

Figure 5 Processing POCIS from retrieval in the field till analysis.

Vermeirssen reported increasing rates up to 37 cm s1 when calibrating with treated sewage effluent. However, also the last study indicated an asymptotic approach of uptake rates at higher flow velocities. Clear differences were observed for membrane covered and naked EmporeTM-type passive samplers. With membranes, the velocity at which uptake rates remained constant was lower (Vermeirssen et al., 2009). In summary, it appears that uptake rates of POCIS and Chemcatchers increase from quiescent to flowing conditions until at a certain flow velocity the aqueous stagnant film layer becomes so small that it does not control the mass transfer any more. However, more calibration data are needed to back this hypothesis. Biofouling. Another problem that can affect the uptake of compounds into POCIS and thus lead to larger errors in quantifying water concentrations is biofouling. Scha¨fer et al. (2008a) pre-fouled Chemcatchers with and without PES membrane for up to 9 days before exposing them to the pesticide thiacloprid within an artificial stream. Biofouling of the naked receiving phase led to a fourfold decrease of uptake. No biofouling was observed for samplers equipped with PES membrane demonstrating the protective function of PES for the period of deployment. With an experiment, comparing uptake kinetics in fouled and nonfouled POCIS, Harman et al. (2009a) found in general an increase of the uptake rates of alkylated phenols after biofouling. The reason for this behavior remained unexplained, and other groups of compounds may behave differently. More research is necessary to clarify the role of biofouling. Burst uptake and lag phase. Several researchers reported that in the beginning of a calibration experiment uptake rates for hydrophilic compounds are higher than expected leading to positive intercepts, whereas hydrophobic compounds show negative intercepts (Harman et al., 2008b). The transition from this burst uptake to the behavior of a lag phase can be approximately related to a log KOW of 3. The burst effect represents initial wetting of the membrane, which leads to a capillary effect. For more hydrophobic compounds, sorption to the membrane dominates probably in the early stages, suggesting a biphasic uptake. However, these effects are negligible for exposure times lasting several weeks (Alvarez et al., 2007).

In recent years, scientists and authorities realized that POCs could play an important role as contaminants in the aquatic environment. Passive samplers such as POCIS and Chemcatchers provide a powerful, easy-to-handle, and cheap tool to sample POCs from the water phase. They allow for an integrative sampling over days or weeks and are thus able to provide TWA concentrations. An integrative monitoring for such long sampling times is almost impossible and financially not feasible with traditional sampling methods. Due to their integrative character, POCIS and Chemcatchers allow further to sequester POCs at concentration levels that are difficult to reach with spot sampling. Furthermore, both types of passive samplers collect dissolved compounds only, which is the bioavailable fraction in the water phase. However, to date some limitations apply to these passive samplers. Especially, their ability to serve as a routine tool for quantitative analysis is limited. The knowledge about uptake rates for different compounds is still limited. In particular, progress in the prediction of dependency of uptake rates from environmental conditions like flow velocity, salinity, temperature, or biofouling is crucial. Several researchers are looking for suitable performance reference chemicals (PRCs) to adjust for uncertainties in uptake rates due to varying environmental conditions. However, to date the number of candidates for PRCs is very limited, and the applicability of the PRC approach with POCIS and Chemcatchers is still in debate. Nevertheless, several research groups are tackling the open questions, and tremendous progress has been made in the last few years. Therefore, polar passive samplers have good prospects to become a standard in water-quality monitoring in the coming future.

Acknowledgments This work was partly supported by the European Union through the Integrated Project MODELKEY (Contract-No. 511237-GOCE) and by the Academic Research Collaboration Programme of German Academic Exchange Service (DAAD) and British Council (Contract-No. D/07/09989).

Sampling and Conservation

References Allan IJ, Booij K, Paschke A, Vrana B, Mills GA, and Greenwood R (2009) Field performance of seven passive sampling devices for monitoring of hydrophobic substances. Environmental Science and Technology 43: 5383--5390. Allan IJ, Booij K, Paschke A, Vrana B, Mills GA, and Greenwood R (2010) Short-term exposure testing of six different passive samplers for monitoring hydrophobic contaminants in water. Journal of Environmental Monitoring 12: 696–703. Allan IJ, Vrana B, Greenwood R, et al. (2006) Strategic monitoring for the European water framework directive. Trends in Analytical Chemistry 25: 704--715. Alvarez DA, Cranor WL, Perkins SD, Clark RC, and Smith SB (2008) Chemical and toxicologic assessment of organic contaminants in surface water using passive samplers. Journal of Environmental Quality 37: 1024--1033. Alvarez DA, Cranor WL, Perkins SD, et al. (2009) Reproductive health of bass in the Potomac, USA, drainage: Part 2. Seasonal occurrence of persistent and emerging organic contaminants. Environmental Toxicology and Chemistry 28: 1084--1095. Alvarez DA, Huckins JN, Petty JD, et al. (2007) Tool for monitoring hydrophilic contaminants in water: Polar organic chemical integrative sampler (POCIS). In: Greenwood R, Mills G, and Vrana B (eds.) Passive Sampling Techniques in Environmental Monitoring, Comprehensive Analytical Chemistry, vol. 48, ch. 8, pp. 171–197. Amsterdam: Elsevier. Alvarez DA, Petty JD, Huckins JN, et al. (2004) Development of a passive, in situ, integrative sampler for hydrophilic organic contaminants in aquatic environments. Environmental Toxicology and Chemistry 23: 1640--1648. Alvarez DA, Stackelberg PE, Petty JD, et al. (2005) Comparison of a novel passive sampler to standard water-column sampling for organic contaminants associated with wastewater effluents entering a New Jersey stream. Chemosphere 61: 610--622. American Health Association (2005) Standard Methods for the Examination of Water & Wastewater. Washington, DC: American Public Health Association. Arditsoglou A and Voutsa D (2008) Passive sampling of selected endocrine disrupting compounds using polar organic chemical integrative samplers. Environmental Pollution 156: 316--324. Artiola JF (2004) Monitoring surface waters. In: Artiola JF, Pepper IL, and Brusseau ML (eds.) Environmental Monitoring and Characterization, pp. 141--161. Burlington: Academic Press. Bandow N, Altenburger R, Streck G, and Brack W (2009) Effect-directed analysis of contaminated sediments with partition-based dosing using green algae cell multiplication inhibition. Environmental Science and Technology 43: 7343--7349. Barackman M, Brusseau ML, Janick FA, Ian LP, and Brusseau ML (2004) Groundwater sampling. In: Artiola JF, Pepper IL, and Brusseau ML (eds.) Environmental Monitoring and Characterization, pp. 121--139. Burlington: Academic Press. Barcelona MJ, Helfrich JA, and Garske EE (1985) Sampling tubing effects on groundwater samples. Analytical Chemistry 57: 460--464. Bartelt-Hunt SL, Snow DD, Damon T, Shockley J, and Hoagland K (2009) The occurrence of illicit and therapeutic pharmaceuticals in wastewater effluent and surface waters in Nebraska. Environmental Pollution 157: 786--791. Batley GE and Gardner D (1977) Sampling and storage of natural waters for trace metal analysis. Water Research 11: 745--756. Bergqvist PA and Zaliauskiene A (2007) Field study considerations in the use of passive sampling devices in water monitoring. In: Greenwood R, Mills G, and Vrana B (eds.) Passive Sampling Techniques in Environmental Monitoring, Comprehensive Analytical Chemistry, vol. 48, ch. 14, pp. 311–327. Amsterdam: Elsevier. Biselli S, Reineke N, Heinzel N, et al. (2005) Bioassay-directed fractionation of organic extracts of marine surface sediments from the North and Baltic Sea – Part I: Determination and identification of organic pollutants. Journal of Soils and Sediments 5: 171--181. Booij K, Hofmans HE, Fischer CV, and Van Weerlee EM (2003) Temperature-dependent uptake rates of nonpolar organic compounds by semipermeable membrane devices and low-density polyethylene membranes. Environmental Science and Technology 37: 361--366. Booij K, Vrana B, Huckins JN, Greenwood R, Mills G, and Vrana B (2007) Theory, modelling and calibration of passive samplers used in water monitoring. In: Greenwood R, Mills G, and Vrana B (eds.) Passive Sampling Techniques in Environmental Monitoring, Comprehensive Analytical Chemistry, vol. 48, ch. 7, pp. 141–169. Amsterdam: Elsevier. Bopp S, WeiX H, and Schirmer K (2005) Time-integrative monitoring of polycyclic aromatic hydrocarbons in groundwater using the ceramic dosimeter passive sampling device. Journal of Chromatography A 1072: 137--147.

149

Bueno MJM, Hernando MD, Aguera A, and Fernandez-Alba AR (2009) Application of passive sampling devices for screening of micro-pollutants in marine aquaculture using LC-MS/MS. Talanta 77: 1518--1527. Burki R, Vermeirssen ELM, Ko¨rner O, Joris C, Burkhardt-Holm P, and Segner H (2006) Assessment of estrogenic exposure in brown trout (Salmo trutta) in a Swiss midland river: Integrated analysis of passive samplers, wild and caged fish, and vitellogenin mRNA and protein. Environmental Toxicology and Chemistry 25: 2077--2086. CEFAS (2007) Aquatic environment monitoring report. Lowestoft 59: 75. Cussler EL (1984) Diffusion: Mass Transfer in Fluid Systems. Cambridge: Cambridge University Press. Dalton MG, Huntsman BE, and Bradbury K (2007) Acquisation and interpretation of water-level data. In: Nielsen DM and Nielsen GL (eds.) The Essential Handbook of Ground-Water Sampling, pp. 173--201. Boca Raton, FL: CRC Press/Taylor and Francis. Danielsson LG (1982) On the use of filters for distinguishing between dissolved and particulate fractions in natural waters. Water Research 16: 179--182. Dixon W and Chiswell B (1996) Review of aquatic monitoring program design. Water Research 30: 1935--1948. European Commission (2003a) Guidance document no. 7 – monitoring under the water framework directive. Common Implementation Strategy for the Water Framework Directive (2000/60/EC). European Commission (2003b) Guidance document no. 11 – planning process. Common Implementation Strategy for the Water Framework Directive (2000/60/ EC). European Commission (2007a) Guidance document no. 15 – guidance on groundwater monitoring. Common Implementation Strategy for the Water Framework Directive (2000/60/EC). European Commission (2007b) Guidance document no. 16 – guidance on groundwater monitoring in drinking water protection areas. Common Implementation Strategy for the Water Framework Directive (2000/60/EC). European Commission (2009) Guidance document no. 19 – guidance on surface water chemical monitoring under the water framework directive. Common Implementation Strategy for the Water Framework Directive (2000/60/EC). European Community (2000) Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for community action in the field of water policy. European Community (2006a) Directive 2000/60/EC of the European Parliament and of the Council of 12 December 2006 on the protection of groundwater against pollution and deterioration. European Community (2006b) Regulation (EC) no. 1907/2006 of the European Parliament and of the Council of 18 December 2006 concerning the registration, evaluation, authorisation and restriction of chemicals (REACH), establishing a European chemicals agency, amending directive 1999/45/EC and repealing council regulation (EEC) no. 793/93 and commission regulation (ec) no. 1488/94 as well as council directive 76/769/EEC and commission directives 91/155/EEC, 93/67/ EEC, 93/105/EC and 2000/21/EC. Finucane JH and May BZ (1961) Modified van Dorn water sampler. Limnology and Oceanography 85--87. Freimann P, Schmidt D, and Schomaker K (1983) MERCOS – a simple Teflon sampler for ultratrace metal analysis in seawater. Marine Chemistry 14: 43--48. Greenwood R, Mills GA, and Vrana B (2009) Potential applications of passive sampling for monitoring non-polar industrial pollutants in the aqueous environment in support of reach. Journal of Chromatography A 1216: 631--639. Greenwood R, Mills G, and Vrana B (eds.) (2007a) Passive Sampling Techniques in Environmental Monitoring, Comprehensive Analytical Chemistry, vol. 48. Amsterdam: Elsevier. Greenwood R, Mills GA, Vrana B, Allan I, Aguilar-Martinez R, and Morrison G (2007b) Monitoring of priority pollutants in water using Chemcatcher passive sampling devices. In: Greenwood R, Mills G, and Vrana B (eds.) Passive Sampling Techniques in Environmental Monitoring, Comprehensive Analytical Chemistry, vol. 48, ch. 9, pp. 199–229. Amsterdam: Elsevier. Go´recki T and Namies´nik J (2002) Passive sampling. Trends in Analytical Chemistry 21: 276--291. Gunold R, Scha¨fer RB, Paschke A, Schu¨u¨rmann G, and Liess M (2008) Calibration of the Chemcatchers passive sampler for monitoring selected polar and semi-polar pesticides in surface water. Environmental Pollution 155: 52--60. Harman C, Boyum O, Erik Tollefsen K, Thomas K, and Grung M (2008a) Uptake of some selected aquatic pollutants in semipermeable membrane devices (SPMDs) and the polar organic chemical integrative sampler (POCIS). Journal of Environmental Monitoring 10: 239--247. Harman C, Boyum O, Thomas KV, and Grung M (2009a) Small but different effect of fouling on the uptake rates of semipermeable membrane devices and polar organic

150

Sampling and Conservation

chemical integrative samplers. Environmental Toxicology and Chemistry 28: 2324--2332. Harman C, Thomas KV, Tollefsen KE, Meier S, Boyum O, and Grung M (2009b) Monitoring the freely dissolved concentrations of polycyclic aromatic hydrocarbons (PAH) and alkylphenols (AP) around a Norwegian oil platform by holistic passive sampling. Marine Pollution Bulletin 58: 1671--1679. Harman C, Tollefsen KE, Boyum O, Thomas K, and Grung M (2008b) Uptake rates of alkylphenols, PAHs and carbazoles in semipermeable membrane devices (SPMDs) and polar organic chemical integrative samplers (POCIS). Chemosphere 72: 1510--1516. Hewitt LM and Marvin CH (2005) Analytical methods in environmental effects-directed investigations of effluents. Mutation Research 589: 208--232. Horowitz AJ, Lum KR, Garbarino JR, Hall GEM, Lemieux C, and Demas CR (1996) Problems associated with using filtration to define dissolves trace element concentrations in natural water samples. Environmental Science and Technology 30: 954--963. House WA (1994) Sampling techniques for organic substances in surface waters. International Journal of Environmental Analytical Chemistry 57: 207--214. Huckins JN, Manuweera GK, Petty JD, Mackay D, and Lebo JA (1993) Lipid-containing semipermeable membrane devices for monitoring organic contaminants in water. Environmental Science and Technology 27: 2489--2496. Huckins JN, Petty JD, and Booij K (2006) Monitors of Organic Chemicals in the Environment – semipermeable membrane devices. Heidelberg: Springer. Huckins JN, Petty JD, Orazio CE, et al. (1999) Determination of uptake kinetics (sampling rates) by lipid-containing semipermeable membrane devices (SPMDs) for polycyclic aromatic hydrocarbons (PAHs) in water. Environmental Science and Technology 33: 3918--3923. Huckins JN, Petty JD, Prest HF, et al. (2000) A Guide for the Use of Semipermeable Membrane Devices (SPMDs) as Sampler of Waterborne Hydrophobic Organic Contaminants. API Publication No. 4690. Washington, DC: American Petroleum Institute (API). Huckins JN, Tubergen MW, and Manuweera GK (1990) Semipermeable membrane devices containing model lipid: A new approach to monitoring the bioavaiiability of lipophilic contaminants and estimating their bioconcentration potential. Chemosphere 20: 533--552. ISO 5667-1 (2006) Water quality – sampling – part 1: Guidance on the design of sampling programmes and sampling techniques. ISO 5667-3 (2003) Water quality – sampling – part 3: Preservation and handling of water. ISO 5667-4 (1987) Water quality – sampling – part 4: Guidance on sampling from lakes, natural and man-made. ISO 5667-6 (2005) Water quality – sampling – part 6: Guidance on sampling of rivers and streams. ISO 5667-10 (1992) Water quality – sampling – part 10: Guidance on sampling of waste waters. ISO 5667-11 (2009) Water quality – sampling – part 11: Guidance on sampling of groundwaters. ISO 5667-14 (1998) Water quality – sampling – part 14: Guidance on quality assurance of environmental water sampling and handling. ISO 5667-20 (2008) Water quality – sampling – part 20: Guidance on the use of sampling data for decision making – compliance with thresholds and classification systems. ISO/DIS 5667-22 (2009) Water quality – sampling – part 22: Guidance on the design and installation of groundwater monitoring points. ISO/DIS 5667-23 (2009) Water quality – sampling – part 23: Determination of priority pollutants in surface water using passive sampling. Jeannot R (1994) Preservation techniques for analysis of organic compounds in water samples – a review. International Journal of Environmental Analytical Chemistry 57: 231--236. Jones-Lepp TL, Alvarez DA, Petty JD, and Huckins JN (2004) Polar organic chemical integrative sampling and liquid chromatography-electrospray/ion-trap mass spectrometry for assessing selected prescription and illicit drugs in treated sewage effluents. Archives of Environmental Contamination and Toxicology 47: 427--439. Keith LH (1990) Environmental sampling: A summary. Environmental Science and Technology 24: 610--617. Kingston JK, Greenwood R, Mills GA, Morrison GM, and Persson LB (2000) Development of a novel passive sampling system for the time-averaged measurement of a range of organic pollutants in aquatic environments. Journal of Environmental Monitoring 2: 487--495. Koester CJ and Clement RE (1993) Analysis of drinking water for trace organics. Critical Reviews in Analytical Chemistry 24: 263--316. Kramer KJM (1994) Inorganic contaminants in the water column: Sampling and sampling strategy. International Journal of Environmental Analytical Chemistry 57: 179--188.

Lerner DN and Teutsch G (1995) Recommendations for level-determined sampling in wells. Journal of Hydrology 171: 355--377. Lewis KM and Archer RD (1979) pKa values of estrone, 17-beta-estradiol and 2methoxyestrone. Steroids 34: 485--499. Loos R, Gawlik BM, Locoro G, Rimaviciute E, Contini S, and Bidoglio G (2009) EUwide survey of polar organic persistent pollutants in European river waters. Environmental Pollution 157: 561--568. Macleod SL, Mcclure EL, and Wong CS (2007) Laboratory calibration and field deployment of the polar organic chemical integrative sampler for pharmaceuticals and personal care products in wastewater and surface water. Environmental Toxicology and Chemistry 26: 2517--2529. Madrid Y and Zayas ZP (2007) Water sampling: Traditional methods and new approaches in water sampling strategy. Trends in Analytical Chemistry 26: 293--299. Maher WA, Cullen PW, and Norris RH (1994) Framework for designing sampling programs. Environmental Monitoring and Assessment 30: 139--162. Majaneva M, Autio R, Huttunen M, Kuosa H, and Kuparinen J (2009) Phytoplankton monitoring: The effect of sampling methods using different stratification and bloom conditions in the Baltic Sea. Boreal Environment Research 14: 313--322. Martin H, Patterson BM, and Davis GB (2003) Field trial of contaminant groundwater monitoring: Comparing time-integrative ceramic dosimeters and conventional water sampling. Environmental Science and Technology 37: 1360--1364. Martin H, Piepenbrink M, and Grathwohl P (2001) Ceramic dosimeters for the timeintegrated contaminant monitoring. Journal of Process Analytical Chemistry 6: 68--73. Mayer P, Tolls J, Hermens J, and Mackay D (2003) Equilibrium sampling devices. Environmental Science and Technology 37: 184A--191A. Mazzella N, Debenest T, and Delmas F (2008) Comparison between the polar organic chemical integrative sampler and the solid-phase extraction for estimating herbicide time-weighted average concentrations during a microcosm experiment. Chemosphere 73: 545--550. Mazzella N, Dubernet JF, and Delmas F (2007) Determination of kinetic and equilibrium regimes in the operation of polar organic chemical integrative samplers – application to the passive sampling of the polar herbicides in aquatic environments. Journal of Chromatography A 1154: 42--51. Mills GA, Allan IJ, Guigues N, Knutsson J, Holmberg A, and Greenwood R (2009a) Monitoring heavy metals using passive sampling devices. In: Gonzalez C, Quevauviller P, and Greenwood R (eds.) Rapid Chemical and Biological Techniques for Water Monitoring, ch. 4.1, pp. 243–261. New York, NY: Wiley. Mills GA, Greenwood R, Vrana B, and Ocelka T (2009b) Meeting report – 3rd international passive sampling workshop and related satellites. Journal of Environmental Monitoring 11: 1911--1914. Mills GA, Vrana B, Allan IJ, Alvarez DA, Huckins JN, and Greenwood R (2007) Trends in monitoring pharmaceuticals and personal-care products in the aquatic environment by use of passive sampling devices. Analytical and Bioanalytical Chemistry 387: 1153--1157. Namies´nik J, Zabiegała B, Kot-Wasik A, Partyka M, and Wasik A (2005) Passive sampling and/or extraction techniques in environmental analysis: A review. Analytical and Bioanalytical Chemistry 381: 279--301. Nielsen DM and Nielsen GL (eds.) (2007a) The Essential Handbook of Ground-Water Sampling. Boca Raton, FL: CRC Press/Taylor and Francis. Nielsen DM (2007a) Purging and sampling device selection and operation. In: Nielsen DM and Nielsen GL (eds.) The Essential Handbook of Ground-Water Sampling, pp. 55--98. Boca Raton, FL: CRC Press/Taylor and Francis. Nielsen DM (2007b) The science behind ground-water sampling. In: Nielsen DM and Nielsen GL (eds.) The Essential Handbook of Ground-Water Sampling, pp. 1--33. Boca Raton, FL: CRC Press/Taylor and Francis. Nielsen GL and Nielsen DM (2007b) Ground-water sample pretreatment: Filtration and preservation. In: Nielsen DM and Nielsen GL (eds.) The Essential Handbook of Ground-Water Sampling, pp. 131--151. Boca Raton, FL: CRC Press/Taylor and Francis. Ouyang G and Pawliszyn J (2007) Configurations and calibration methods for passive sampling techniques. Journal of Chromatography A 1168: 226--235. Parr J, Bollinger M, Callaway O, and Carlberg K (1988) Preservation techniques for organic and inorganic compounds in water samples. In: Keith LH (ed.) Principles of Environmental Sampling, pp. 221--230. Washington, DC: American Chemical Society. Paschke A, Schwab K, Bru¨mmer J, Schu¨u¨rmann G, Paschke H, and Popp P (2006) Rapid semi-continuous calibration and field test of membrane-enclosed silicone collector as passive water sampler. Journal of Chromatography A 1124: 187--195. Paschke A, Vrana A, Popp P, Wennrich L, Paschke H, and Schu¨u¨rmann G (2007) Membrane-enclosed sorptive coating for monitoring organic compounds in water.

Sampling and Conservation

In: Greenwood R, Mills G, and Vrana B (eds.) Passive Sampling Techniques in Environmental Monitoring, Comprehensive Analytical Chemistry, vol. 48, ch. 10, pp. 231–250. Amsterdam: Elsevier. Petty JD, Huckins JN, Alvarez DA, et al. (2004) A holistic passive integrative sampling approach for assessing the presence and potential impacts of waterborne environmental contaminants. Chemosphere 54: 695--705. Petty JD, Orazio CE, Huckins JN, et al. (2000) Considerations involved with the use of semipermeable membrane devices for monitoring environmental contaminants. Journal of Chromatography A 879: 8395. Pomati F, Castiglioni S, Zuccato E, et al. (2006) Effects of a complex mixture of therapeutic drugs at environmental levels on human embryonic cells. Environmental Science and Technology 40: 2442--2447. Reemtsma T, Miehe U, Du¨nnbier U, and Jekel M (2009) Polar pollutants in municipal wastewater and the water cycle: Occurrence and removal of benzotriazoles. Water Research 44: 596--604. Reemtsma T, Weiss SJ, Petrovic M, et al. (2006) Polar pollutants entry into the water cycle by municipal wastewater: A European perspective. Environmental Science and Technology 40: 5451--5458. Richardson SD (2007) Water analysis: Emerging contaminants and current issues. Analytical Chemistry 79: 4295--4323. Richardson SD (2008) Environmental mass spectrometry: Emerging contaminants and current issues. Analytical Chemistry 80: 4373--4402. Richardson SD (2009) Water analysis: Emerging contaminants and current issues. Analytical Chemistry 81: 4645--4677. Roy S and Fouillac AM (2004) Uncertainties related to sampling and their impact on the chemical analysis of groundwater. Trends in Analytical Chemistry 23: 185--193. Ru¨del H, Schro¨der W, von der Trenck K, and Wiesmu¨ller G (2009) Substance-related environmental monitoring. Environmental Science and Pollution Research 16: 486--498. Scha¨fer RB, Paschke A, and Liess M (2008a) Aquatic passive sampling of a short-term thiacloprid pulse with the Chemcatcher: Impact of biofouling and use of a diffusion-limiting membrane on the sampling rate. Journal of Chromatography A 1203: 1--6. Scha¨fer RB, Paschke A, Vrana B, Mu¨ller R, and Liess M (2008b) Performance of the Chemcatchers passive sampler when used to monitor 10 polar and semi-polar pesticides in 16 Central European streams, and comparison with two other sampling methods. Water Research 42: 2707--2717. Schwarzenbach RP, Gschwend PM, and Imboden DM (2003) Environmental Organic Chemistry, 2nd edn. New York, NY: Wiley. Seethapathy S, Go´recki T, and Li X (2008) Passive sampling in environmental analysis. Journal of Chromatography A 1184: 234--253. Sellin MK, Snow DD, Akerly DL, and Kolok AS (2009) Estrogenic compounds downstream from three small cities in Eastern Nebraska: Occurrence and biological effect. Journal of the American Water Resources Association 45: 14--21. Sellin MK, Snow DD, and Kolok AS (2010) Reductions in hepatic vitellogenin and estrogen receptor alpha expression by sediments from an agriculturally impacted waterway. Aquatic Toxicology 96: 103--108. Shaw M, Eaglesham G, and Mueller JF (2009) Uptake and release of polar compounds in SDB-RPS EmporeTM disks; implications for their use as passive samplers. Chemosphere 75: 1--7. Smedes F (2007) Monitoring of chlorinated biphenyls and polycyclic aromatic hydrocarbons by passive sampling in concert with deployed mussels. In: Greenwood R, Mills G, and Vrana B (eds.) Passive Sampling Techniques in Environmental Monitoring, Comprehensive Analytical Chemistry, vol. 48, ch. 19, pp. 407–448. Amsterdam: Elsevier. So¨derstro¨m H, Lindberg RH, and Fick J (2009) Strategies for monitoring the emerging polar organic contaminants in water with emphasis on integrative passive sampling. Journal of Chromatography A 1216: 623--630. Staples SA, Dorn PB, and Klecka GM (1998) A review of the environmental fate, effects, and exposures of bisphenol A. Chemosphere 36: 2149--2173. Stephens BS, Kapernick A, Eaglesham G, and Mueller J (2005) Aquatic passive sampling of herbicides on naked particle loaded membranes: Accelerated measurement and empirical estimation of kinetic parameters. Environmental Science and Technology 39: 8891--8897. Stephens BS, Kapernick AP, Eaglesham G, and Mueller JF (2009) Event monitoring of herbicides with naked and membrane-covered Empore disk integrative passive sampling devices. Marine Pollution Bulletin 58: 1116--1122. Streck G (2009a) Chemical and biological analysis of estrogenic, progestagenic and androgenic steroids in the environment. Trends in Analytical Chemistry 28: 635--652. Streck G (2009b) Experimental determination of uptake rates for polar organic chemical integrative samplers (POCIS). In: 3rd International Passive Sampling Workshop

151

(IPSW 2009). Proceedings. Prague, The Czech Republic, Institute of Chemical Technology, p. 59. Strobl RO and Robillard PD (2008) Network design for water quality monitoring of surface freshwaters: A review. Journal of Environmental Management 87: 639--648. Stuer-Lauridsen F (2005) Review of passive accumulation devices for monitoring organic micropollutants in the aquatic environment. Environmental Pollution 136: 503--524. Sturgeon R, Berman SS, and Kremling K (1987) Sampling and storage of natural water for trace metals. Critical Reviews in Analytical Chemistry 18: 209--244. Togola A and Budzinski H (2007) Development of polar organic integrative samplers for analysis of pharmaceuticals in aquatic systems. Analytical Chemistry 79: 6734--6741. Tran ATK, Hyne RV, and Doble P (2007) Calibration of a passive sampling device for time-integrated sampling of hydrophilic herbicides in aquatic environments. Environmental Toxicology and Chemistry 26: 435--443. USGS (ed.) (2008) National Field Manual for the Collection of Water-Quality Data: U.S. Geological Survey Techniques of Water-Resources Investigations, U.S. Geological Survey, book 9, chs. A1–A9. Washington, DC: United Stated Government Printing Office. Van Pinxteren M, Paschke A, and Popp P (2010) Silicone rod and silicone tube sorptive extraction. Journal of Chromatography A 1217: 2589–2598. Vermeirssen ELM, Bramaz N, Hollender J, Singer H, and Escher BI (2009) Passive sampling combined with ecotoxicological and chemical analysis of pharmaceuticals and biocides – evaluation of three Chemcatcher (TM) configurations. Water Research 43: 903--914. von der Ohe PC, De Deckere E, Pru¨X A, et al. (2009) Towards an integrated European risk assessment. Integrated Environmental Assessment and Management 5: 50--61. Vrana B, Mills GA, Allan IJ, et al. (2005a) Passive sampling techniques for monitoring pollutants in water. Trends in Analytical Chemistry 24: 845--868. Vrana B, Mills GA, Dominiak E, and Greenwood R (2006a) Calibration of the Chemcatcher passive sampler for the monitoring of priority pollutants in water. Environmental Pollution 142: 333--343. Vrana B, Mills GA, Kotterman M, Leonards P, Booij K, and Greenwood R (2007) Modelling and field application of the Chemcatcher passive sampler calibration data for monitoring of hydrophobic organic pollutants in water. Environmental Pollution 145: 895--904. Vrana B, Paschke H, Paschke A, Popp P, and Schu¨u¨rmann G (2005b) Performance of semipermeable membrane devices for sampling of organic contaminants in groundwater. Journal of Environmental Monitoring 7: 500--508. Vrana B, Paschke A, and Popp P (2006b) Calibration and field performance of membrane-enclosed sorptive coating for integrative passive sampling of persistent organic pollutants in water. Environmental Pollution 144: 296--307. Vrana B, Paschke A, Popp P, and Schu¨u¨rmann G (2001a) Use of semipermeable membrane devices (SPMDs): Determination of bioavailable, organic, waterborne contaminants in the industrial region of Bitterfeld, Saxony-Anhalt, Germany. Environmental Science and Pollution Research 8: 27--34. Vrana B, Popp P, Paschke A, and Schu¨u¨rmann G (2001b) Membrane enclosed sorptive coating – an integrative passive sampler for monitoring organic contaminants in water. Analytical Chemistry 73: 5191--5200. Vrana B and Schu¨u¨rmann G (2002) Calibrating the uptake kinetics of semipermeable membrane devices in water: Impact of hydrodynamics. Environmental Science and Technology 36: 290--296. Vroblesky DA (2007) Passive diffusion samplers to monitor volatile organic compounds in ground-water. In: Greenwood R, Mills G, and Vrana B (eds.) Passive Sampling Techniques in Environmental Monitoring, Comprehensive Analytical Chemistry, vol. 48, ch. 13, pp. 295–310. Amsterdam: Elsevier. Vroblesky DA (2001a) Users guide for polyethylene-based passive diffusion bag samplers to obtain volatile organic compound concentrations in wells. Part 1: Deployment, recovery, data interpretation, and quality control and assurance. Water Resources Investigations Report 01-4060. Columbia, SC: US Geological Survey. Vroblesky DA (ed.) (2001b) Users guide for polyethylene-based passive diffusion bag samplers to obtain volatile organic compound concentrations in wells. Part 2: Field tests. Water Resources Investigations Report 01-4061. Columbia, SC: US Geological Survey. Wardencki W and Namies´nik J (2002) Sampling water and aqueous solutions. In: Pawliszyn J (ed.) Sampling and Sample Preparation for Field and Laboratory: Comprehensive Analytical Chemistry, 37, pp. 33--60. Elsevier: Amsterdam. Warnken KW, Zhang H, and Davison W (2007) In situ monitoring and dynamic speciation measurements in solution using DGT. In: Greenwood R, Mills G, and Vrana B (eds.) Passive Sampling Techniques in Environmental Monitoring, Comprehensive Analytical Chemistry, vol. 48, ch. 11, pp. 251–278. Amsterdam: Elsevier.

152

Sampling and Conservation

WeiX H, Schirmer K, Bopp S, and Grathwohl P (2007) Use of ceramic dosimeters in water monitoring. In: Greenwood R, Mills G,and Vrana B (eds.) Passive Sampling Techniques in Environmental Monitoring, Comprehensive Analytical Chemistry, vol. 48, ch. 12, pp. 279–294. Amsterdam: Elsevier. Wennrich L, Vrana B, Popp P, and Lorenz W (2003) Development of an integrative passive sampler for the monitoring of organic water pollutants. Journal of Environmental Monitoring 5: 813--822. Wenzel KD, Vrana B, Hubert A, and Schu¨u¨rmann G (2004) Dialysis of persistent organic pollutants and polycyclic aromatic hydrocarbons from semipermeable membranes. A procedure using an accelerated solvent extraction device. Analytical Chemistry 76: 5503--5509. Whitfield P (1983) Evaluation of water quality sampling locations on the Yukon river. Journal of the American Water Resources Association 19: 115--121. Whitfield P (1988) Goals and data collection designs for water quality monitoring. Journal of the American Water Resources Association 24: 775--780. Wilde FD (2005) Preparations for water sampling (ver. 2.0). In: National Field Manual for the Collection of Water-Quality Data: U.S. Geological Survey Techniques of Water-Resources Investigations, book 9, ch. A1. http://pubs.water.usgs.gov/twri9A (accessed March 2010). Wilde FD, Radtke DB, Gibs J, and Iwatsubo RT (eds.) (2004) Processing of water samples (version 2.1). In: National Field Manual for the Collection of Water-Quality Data: U.S. Geological Survey Techniques of Water-Resources Investigations, U.S.

Geological Survey, book 9, ch. A5. Washington, DC: United Stated Government Printing Office. Yates K, Davies I, Webster L, Pollard P, Lawton L, and Moffat C (2007) Passive sampling: Partition coefficients for a silicone rubber reference phase. Journal of Environmental Monitoring 9: 1116--1121. Zabiegala B, Kot-Wasik A, Urbanowicz M, and Namies´nik J (2010) Passive sampling as a tool for obtaining reliable analytical informantion in environmental quality monitoring. Analytical and Bioanalytical Chemistry 396: 273--296. Zenker A, Schmutz H, and Fent K (2008) Simultaneous trace determination of nine organic UV-absorbing compounds (UV filters) in environmental samples. Journal of Chromatography A 1202: 64--74. Zhang Z, Hibberd A, and Zhou JL (2008) Analysis of emerging contaminants in sewage effluent and river water: Comparison between spot and passive sampling. Analytica Chimica Acta 607: 37--44.

Relevant Website http://europa.eu.int Europa.

3.07 Measurement Quality in Water Analysis B Magnusson, SP Technical Research Institute of Sweden, Bora˚s, Sweden M Koch, Universita¨t Stuttgart, Stuttgart, Germany & 2011 Elsevier B.V. All rights reserved.

3.07.1 Introduction 3.07.2 Terminology 3.07.2.1 Measurand 3.07.2.2 Measurement Uncertainty 3.07.2.3 Metrological Traceability 3.07.2.4 Validation 3.07.2.5 Trueness 3.07.2.6 Precision 3.07.2.7 Limit of Detection 3.07.2.8 Limit of Quantification 3.07.3 How to Set the Analytical Requirement 3.07.4 Quality of Drinking Water Analysis 3.07.5 How to Assess the Quality in a Lab 3.07.5.1 Method Validation 3.07.5.1.1 Implementation of a standard method 3.07.5.1.2 Single laboratory (in-house) method validation 3.07.5.2 Metrological Traceability 3.07.5.3 Measurement Uncertainty 3.07.5.4 Quality Control 3.07.5.4.1 Internal QC 3.07.5.4.2 External QC – PT 3.07.6 Data Treatment 3.07.7 Conclusions Acknowledgments References

Glossary Measurand Quantity intended to be measured (VIM 2.3). Measurement uncertainty Non-negative parameter characterizing the dispersion of the quantity values being attributed to a measurand, based on the information used (VIM 2.26). Metrological traceability Property of a measurement result whereby the result can be related to a reference through a documented unbroken chain of calibrations each contributing to the measurement uncertainty (VIM 2.41). Precision Closeness of agreement between indications or measured quantity values obtained by replicate

3.07.1 Introduction Many decisions in the water sector are based on the results of measurements. Therefore, it is essential that such results are reliable. The needed measurement quality can be achieved by validation that the test method is fit for the intended purpose and by establishing traceability of the results to stated references and an estimate of the uncertainty of measurement. The

153 153 154 154 154 154 154 154 155 156 156 157 159 159 159 160 161 162 164 164 167 167 168 168 168

measurements on the same or similar objects under specified conditions (VIM 2.15). Trueness Closeness of agreement between the average of an infinite number of replicate measured quantity values (VIM 2.10) and a reference quantity value (VIM 2.14). Validation Parameter where the specified requirements are adequate for an intended use (VIM 2.45) and where verification is defined as the provision of objective evidence that a given item fulfills specified requirements (VIM 2.44).

ongoing quality control (QC; internal and external) assures that the measurement results (including uncertainty) are of the same quality as at the time of validation. Measurement quality should include both sampling and analysis. The analysis is treated in this chapter and sampling in Chapter 3.06 Sampling and Conservation. We had reliable measurement data for most constituents at the mg l1 level and nutrients at the mg l1 level many years

153

154

Measurement Quality in Water Analysis

back and we can, in many cases, compare the levels of these components in the environment over time and space. For background levels of trace elements and organic compounds at the ng l1 level, we can only in some cases compare measurement results that date back to some 20 years back, and still for many components the requirements for measurement quality are not met. Therefore, the analytical results cannot be compared. There is an ongoing development in both sampling and analysis and with increasing capability of detection of low levels of important components and improved sampling, the requirements will be fulfilled for more components. Important issues in improving quality in trace element analyses were the introduction of clean room laboratory facilities in the 1980s, intense development of samplers and hydrographical wires and trace analytical techniques such as graphite furnace atomic absorption spectrometry (AAS) and inductively coupled plasma mass spectrometry (ICP-MS). The issue of quality in analysis in the water sector is dealt with in several European Union (EU) projects. One of the most recent is EAQC-WISE (2010). The project European Analytical Quality Control in support of the Water Framework Directive (WFD) via the Water Information System for Europe (EAQC-WISE) aims to establish a general QC system and the ultimate objective is to develop a sustainable pan-EU quality assurance and QC (QA/QC) system for water, biota, sediment, and related soil monitoring data. In this chapter, measurement quality in chemical analysis in general is presented with examples from the water sector: (1) the needed terminology; (2) how to set the requirements; (3) the present status of measurement quality in water analysis for some selected parameters based on the requirements in EU directives; and (4) how each laboratory can demonstrate quality. The aim is not a basic introduction for beginners, but an overview which may stimulate further reading and study of the literature.

3.07.2 Terminology To describe and assess the measurement quality the following terms are essential: the general terms (measurand, measurement uncertainty, traceability, and validation) and the specific terms related to validation (trueness, precision, limit of detection (LOD), and limit of quantification (LOQ)). The definitions are mainly from the vocabulary in metrology (VIM), the internationally agreed vocabulary for measurements (BIPM, 2008b). VIM is a normative reference in ISO/IEC 17025:2005; therefore, this terminology applies to all accredited laboratories. Some additional terms used in validation are introduced in Section 3.07.5.1.

and sediments, the base for reporting is included in the measurand, for example, mass fraction (mg kg1) of Cd in a sediment sample delivered to the laboratory reported on dry basis (105 1C, 2 h).

3.07.2.2 Measurement Uncertainty Measurement uncertainty provides information on the level of confidence that can be placed on the measurement result. The estimate of measurement uncertainty is a requirement for accreditation and should be communicated, on request, to the customer to show the quality of the measurement. Measurement uncertainty is the quantitative expression of the doubt associated with the result. The result is often presented in the format value7expanded uncertainty. For example, 50 mg l17 10 mg l1 (expanded uncertainty k ¼ 2) corresponds to the interval 40–60 mg l1. We interpret this as: the measured value7expanded uncertainty is an estimate of the true value (VIM 2.11) and the true value is (with a stated probability of normally 95%) somewhere in this interval. The uncertainty interval and the relation between the reference (or estimate of true value) and the measured value are shown in Figure 1.

3.07.2.3 Metrological Traceability Traceability can refer to the documentation, that is, sampling procedure, laboratory, analyst, test method, etc., but when we are referring to traceability of measurement results as in ISO/ IEC 17025:2005 the results have to be traceable to the metrological references used. To be specific we here use the term ‘metrological traceability’. Ideally, the references should be values of national and international standards expressed in SI units. The metrological traceability can be achieved through chains of calibrations (VIM 2.39). For temperature and many other physical quantities (e.g., mass and time), the traceability is relatively easily established by the national metrology institutes. However, in chemistry the metrological traceability is realized by working standards for calibration which are normally prepared from reference materials: substances with defined purity, solutions of pure substances, or matrix reference materials.

Uncertainty interval y − U ... y + U

Difference (error)

y − U yR

y

y+U

3.07.2.1 Measurand The specification of the measurand normally includes the kind of quantity, the unit, the analyte to be analyzed, and the test item, for example, mass concentration (ng l1) of Cd in a seawater sample delivered to the laboratory or mass concentration (ng l1) of Cd in a seawater basin at the time of sampling. In the former case, the test item is laboratory sample, and in the latter the test item is the seawater basin. For biota

Reference value best estimate of the true value Measured value Figure 1 Relation between a measured value, y, with a given uncertainty interval and a reference value, yR. From Ivo Leito, University of Tartu, Estonia.

Measurement Quality in Water Analysis 3.07.2.4 Validation In VIM, validation implies a verification or check that the measurement procedure (VIM 2.6) or test method is fit for the intended purpose. ‘Fitness for purpose’ or ‘adequate for an intended use’ implies that the performance of the procedure (as specified by performance parameters, such as trueness, precision, LOD, and LOQ, besides the measurement uncertainty) meets the specified requirements.

3.07.2.5 Trueness Trueness is related to systematic measurement error (VIM 2.17) and is an expression of how close the mean of a set of results is to the reference value of a reference material. Trueness cannot be expressed numerically. It is normally expressed in terms of measurement bias (VIM 2.18). The bias is the difference between the mean value of several measurement results and a reference quantity value (VIM 5.18) as shown in Figure 2. Measurement bias is an estimate of the systematic error as shown in the figure. A measurement bias may be due to several causes such as an erroneous calibration, contamination, losses during sample

Mean

Bias

Reference quality value

Figure 2 Bias – the difference between the mean of several measurement results and a reference value. Reproduced with permission from LGC Limited.

treatment, or lack of selectivity. Selectivity is high, if the measurement result is independent of matrix components other than the measurand, that is, there are no significant interferences. An estimate of the bias for measurement results produced by a laboratory under intermediate precision (within-laboratory reproducibility) conditions can be obtained by applying the measuring procedure to one or several reference materials several times over a longer time period (e.g., 6 months) and calculating the mean value. The bias is then the difference between the mean value obtained with this procedure under intermediate precision conditions and the reference quantity value (VIM 5.18).

3.07.2.6 Precision Measurement precision is related to random measurement error (VIM 2.19) and is a measure of how close results are to one another. The term precision is used differently in measurement science and in common language. When we talk about measurement results within the analytical community precision expresses spread, but in common language it is synonymous with accuracy (closeness of agreement between a measured quantity value and a true quantity value of a measurand (VIM 2.13)). Measurement results cannot be corrected to remove the effect of random error but the size of the random error can be reduced by making replicate measurements and calculating the mean value. Measurement precision is expressed numerically using measures of imprecision such as the standard deviation calculated from results obtained by carrying out replicate measurements on a suitable material under specified conditions (Figure 3). Examples of specified conditions are: repeatability conditions, intermediate precision conditions (also called within-laboratory reproducibility conditions), or reproducibility conditions (for details on these terms, see Section 3.07.5.1).

Between batches

Between laboratories

Increasing s

Between injections

Within batch (replicates)

155

Measurement repeatability

Intermediate measurement precision

Measurement reproducibility

Figure 3 The relationship between different estimates of precision illustrated in terms of the magnitude of the observed imprecision. As the conditions of measurement become more variable (e.g., moving from repeatability conditions to reproducibility conditions), the standard deviation of measurement results generally increases. Reproduced with permission from LGC Limited.

156

Measurement Quality in Water Analysis

3.07.2.7 Limit of Detection

Critical value



Measured value

Many analysts are familiar with calculating the LOD for a test method by multiplying a standard deviation, s (obtained from the results from the analysis of a blank sample or a sample containing a low content of the measurand) by an appropriate factor (typically between 3 and 5). The multiplying factor is based on statistical reasoning. The following text explains the background to the commonly used factor of 3.3. The aim, when determining the LOD, is to establish the lowest amount of the concentration of the analyte present in a sample that can be detected, using a given test method, with a specified level of confidence. Defining the LOD is a two-step process. First, a critical value is established. This value is set so that the probability of obtaining a measurement result that exceeds the critical value is no greater than a, if a sample actually contains none of the measurand. The critical value sets a criterion for declaring a sample to be positive. A false positive probability of a ¼ 0.05 is generally used; this leads to a critical value of approximately 1.65s (where s is the standard deviation of a large number of results for a blank sample or a sample containing a low content of the measurand, and 1.65 is the one-tailed Student t-value for infinite degrees of freedom at a significance level, a ¼ 0.05). The critical value is indicated on the vertical axis in Figure 4 to emphasize the fact that it is a measured value. The critical value is most conveniently expressed in terms of concentration, though in principle it may be any observation, such as peak area. Any result exceeding the critical value should be declared positive. However, if the true value for the concentration in a sample was exactly equal to the critical value (expressed in terms of concentration), approximately half of the measurement results would be expected to fall below the critical value, giving a false negative rate of 50%. This is illustrated by the distribution line in the middle of Figure 4. A false negative rate of 50% is obviously too high to be of practical use; the test method does not reliably give results above the critical value if the true value for concentration is equal to the critical value. The LOD is intended to represent the true amount of substance concentration for which the false negative rate is acceptable given the critical value. The acceptable false negative error, b, is usually set equal to the acceptable false positive error; this is largely for

Limit of detection



0 0 Distribution of results

True value 50% false negative rate if analyte concentration = critical value

Figure 4 Illustration of statistical basis of limit of detection calculations. Reproduced with permission from LGC Limited.

historical reasons. The International Union of Pure and Applied Chemistry (IUPAC) recommends default values of a ¼ b ¼ 0.05 (Currie, 1995). Using a ¼ b ¼ 0.05, the LOD needs to be 1.65s above the value specified for the critical value. This is illustrated by the shaded distribution on the horizontal axis in Figure 4. The factor for calculating the LOD with a ¼ b ¼ 0.05 is thus 1.65 þ1.65 ¼ 3.3.

3.07.2.8 Limit of Quantification The IUPAC recommendation is to set the LOQ to a factor times the measured standard deviation that is used to determine the detection limit. The factor is arbitrary but a factor of 10 is usually used. When the LOD is calculated as 3.3s, the ratio between LOQ and LOD is a factor of 3. With a factor of 10 for the LOQ, the repeatability expressed as CV% (coefficient of variation) is about 10% at the concentration level of LOQ.

3.07.3 How to Set the Analytical Requirement Analytical requirements shall be governed by the intended use of the results – what purpose do we have for the measurement (see Figure 5). In Chapter 3.06 Sampling and Conservation, ‘sampling and conservation’ uncertainty in general with reference to ISO 5667-20:2008 is discussed in detail. Here, we focus on the analytical requirement for the water sector. For Europe, the analytical requirements for the monitoring of ground- and surface water in the context of the WFD as well as for drinking water analysis are set in EU directives. The WFD 2000/60/EC (European Parliament and the Council of the European Union, 2000) established a framework for community action in the field of water policy, especially for groundwater, surface water, and coastal seawater. Environmental quality standards (EQSs) have been set in the daughter directive 2008/105/EC (QA/QC directive) (European Parliament and the Council of the European Union, 2008) for the priority substances mentioned in the WFD and for certain other pollutants in the form of limits for the annual average (AA-EQS) and for maximum allowable concentrations (MACEQS). Requirements on quality assurance and QC in the laboratories performing the monitoring are set in a separate directive 2009/90/EC published in 2009 (Commission of the European Communities, 2009). A maximum expanded uncertainty of measurement (k ¼ 2) of 50% of the EQS value is required for measurement results for water monitoring. The LOQ should be less than 30% of the EQS value. The first requirement of expanded uncertainty (k ¼ 2) of 50% of the EQS value for all parameters is very ambitious since the EQS values are very low for many substances (examples in Table 1) relative to the detection capability of analytical methods. Up to now there is only very limited information from interlaboratory tests to prove that these requirements can be met by the laboratories under routine conditions. The second requirement for the WFD directive is the LOQ. Compared with the requirements in the EU drinking water directive (see below) the lower limit is here set using LOQ

Measurement Quality in Water Analysis

157

Client Decision on result

Client issue

Data presentation

Define issue

Client Report on measurement

Decision on measurement

Evaluation

interface Measuring scientist

Sampling Analysis

Measurement

Measuring scientist Figure 5 The measurement cycle starting with the client issue and ending with a decision upon the result. Reproduced from SP Technical Research Institute of Sweden.

Table 1 Environmental quality standards (EQS) from 2008/105/EC for some selected parameters and extract from the minimum performance criteria for analyses according to WFD directive 2000/60/ EC laid down in the QA/QC requirement 2009/90/EC Parameter

Environmental quality standard (EQS) (mg l 1)

Maximum expanded uncertainty 50% of the EQS (mg l 1)

Limit of quantification (LOQ) 30% of the EQS (mg l 1)

Benzene Benzo(a)pyrene Brominated diphenylether Endosulfan Mercury

10 0.05 0.000 5

5 0.025 0.000 25

3 0.015 0.000 15

0.005 0.05

0.002 5 0.025

0.001 5 0.015

instead of LOD. The LOQ is commonly used as a reporting limit in analysis – a measurement result below the LOQ is generally reported as ‘less than’. The ratio of LOQ to LOD is often equal to 3 (e.g., if the LOD is defined as 3.3s and the LOQ is 10s). The reporting of data less than can cause problems when further calculations are performed (see Section 3.07.6). In the EU drinking water directive (Council of the European Union, 1998), there are three requirements: 1. trueness, 2. precision, and 3. LOD limit (see extract in Table 2). Specified requirements are given for each component related to a limit value – so-called parametric value. They are derived on the basis of relevant and extended toxicological data to ensure that water intended for human consumption can be

consumed safely on a lifetime basis, and thus represent a high level of health protection. The prescribed requirements are clear from the analytical point of view, and the additional, explanatory notes give good guidance on how to demonstrate that these requirements are met. Further guidance on this approach can be found in the literature (Thompson et al., 2002; EURACHEM, 1998) and in ISO 5725 (ISO 5725-1:1994; ISO 5725-2:1994; ISO 57253:1994; ISO 5725-4:1994; ISO 5725-5:1998; ISO 57256:1994). The precision requirements should be assessed under intermediate precision conditions (within-laboratory reproducibility conditions) and not under repeatability conditions. We also recommend assessing the bias under intermediate precision conditions. In the EU drinking water directive the requirements are based on trueness and precision, and in the WFD the requirements are based on measurement uncertainty which can be seen to encompass both trueness and precision. This reflects the change in terminology used to describe the measurement quality from accuracy (trueness and precision) to measurement uncertainty. The relation between the terms is demonstrated in Figure 6. A question that is still under discussion is how to combine trueness and precision data in order to obtain a required measurement uncertainty. We propose the following approach on the basis of the requirements in the EU drinking water directive: 1. We could use half the value of the precision requirement as a standard uncertainty component (standard deviation) since it is defined as two standard deviations (see note 2 in Table 2). 2. If we assume that the value for trueness is the maximum allowed bias, then the standard uncertainty component can be calculated from a rectangular distribution, which is the half-width of the interval divided by root of 3 (Eurachem, 2000).

158

Measurement Quality in Water Analysis

Table 2 Extract from table ‘Parameters for which performance characteristics are specified’ from annex III of the EU drinking water directive (Council of the European Union, 1998) Parameter

Parametric value

Trueness % of parametric value (note 1)

Precision % of parametric value (note 2)

Limit of detection % of parametric value (note 3)

Ammonium Arsenic Benzene Benzo(a)pyrene Mercury Pesticides

0.5 mg l1 10 mg l1 1 mg l1 0.01 mg l1 1 mg l1 0.5 mg l1

10 10 25 25 20 25

10 10 25 25 10 25

10 10 25 25 20 25

Note 1. Trueness is the systematic error and is the difference between the mean value of the large number of repeated measurements and the true value. Note 2. Precision is the random error and is usually expressed as the standard deviation (within and between batch) of the spread of results about the mean. Acceptable precision is twice the relative standard deviation. Note 3. Limit of detection is either (1) 3 times the relative within batch standard deviation of a natural sample containing a low concentration of the parameter, or 5 times the relative within batch standard deviation of a blank sample. The terms ‘trueness’ and ‘precision’ discussed in notes 1 and 2 are further defined in ISO 5725.

Type of errors

Performance characteristics

Quantitative expression of performance characteristics

Systematic error

Trueness

Bias

(Total) error

Accuracy

Measurement uncertainty

Random error

Precision

Standard deviation repeatability / within-lab reproducibility/ reproducibility

Figure 6 Relationships between type of error, qualitative performance characteristics, and their quantitative expression (trueness and precision). Reproduced from Menditto A, Patriarca M, and Magnusson B (2007) Understanding the meaning of accuracy, trueness and precision. Accreditation and Quality Assurance 12: 45–47, with permission from Springer.

The standard uncertainties are then combined to an estimated maximum standard uncertainty (Equation (1)):

umax

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi    ffi limitprecision 2 limittrueness 2 pffiffiffi ¼ þ 2 3

ð1Þ

The EU drinking water directive specifies three different requirement levels. From these requirements an estimated maximum standard uncertainty is calculated according to Equation (1) and the maximum allowed expanded uncertainty is calculated at a confidences level of about 95% (coverage factor 2) – see Table 3.

Table 3 Requirements in the EU drinking water directive and the estimated maximum standard uncertainty and expanded measurement uncertainty Trueness (%)

Precision (%)

Maximum standard uncertainty (%)

Maximum expanded uncertainty (%)

10 20 25

10 10 25

7.6 13 20

15 25 38

The expanded uncertainty is calculated with a coverage factor of 2. All requirements are given in percent of the parametric value.

Measurement Quality in Water Analysis

These values are valid only at the parametric value stated in the drinking water directive. There is no other information on requirements at other concentration levels. The maximum expanded uncertainty requirement for WFD monitoring (QA/QC Directive) is slightly higher (50%) than for EU drinking water directive (Table 3). However, the parametric values (EQS) are much lower in the WFD and therefore the WFD-related performance criteria are highly challenging. It is still not clear whether these requirements can be met by routine water laboratories. In the next section, we look at the present status of drinking water analysis for analytical laboratories in Europe using Germany as an example.

3.07.4 Quality of Drinking Water Analysis There are a large number of interlaboratory comparisons in the field of water analysis, organized for the proficiency testing (PT – see Section 3.07.5.4.2) of water laboratories. From the results of these intercomparisons, information can be gathered on whether the analytical requirements are fulfilled. The sampling part is still to be evaluated. The following text deals with the analytical quality for drinking water analysis. The uncertainty of measurement for most components depends on the concentration, and the measurement uncertainty at the specific level (EQS or parametric value) has to be evaluated. Figures 7–11 show examples of CV% versus the mass concentration of the analyte for PTs or interlaboratory trials performed in Germany during the time period 2002–09. The CV% is calculated from all the laboratory results using the Q-method, a robust method described in ISO/TS 20612:2007. In addition, a function for CV% versus mass concentration according to ISO/TS 20612:2007 is fitted to the data (blue line). The CV% function indicates the average quality of the performance of the drinking water laboratories over some years. The parametric value from the European drinking water directive together with the required CV% is shown with a red line. The

159

CV% can be regarded as an estimate of the standard uncertainty (ISO/TS 21748:2004) if the laboratories use the same method. In order to assess the measurement quality, we have to compare these PT results with the requirements. In Table 4 the required maximum standard uncertainty estimated from the EU drinking water directive (interpreted in the way described above) is compared with the average CV% at the parametric value. This is compliant to the fact that in these PT rounds results in the range 72 standard deviations are assessed as successful. This exactly matches the requirements of the directive (expanded uncertainty with a factor of 2). Analytes where the requirements of the EU drinking water directive are fulfilled on average are indicated Yes. In cases, where the requirements on average are almost fulfilled have a No, and those where the requirements are clearly not fulfilled are indicated by a No. The results from these PT trials show that the performance requirements of the EU drinking water directive are far from the analytical reality for most laboratories. In some cases (e.g., most of the major components), the quality of the analyses is much better than the requirements, but in many cases, especially for many of the trace elements and most of the organic trace compounds it is very difficult to meet the requirements for most laboratories. However, single expert laboratories may have a higher measurement quality but maybe to an increased cost of analysis.

3.07.5 How to Assess the Quality in a Lab The main pillars for measurement quality in an analytical laboratory are: 1. 2. 3. 4.

validation, metrological traceability, measurement uncertainty, and QC.

45

Variation coefficient (%)

40 35 30 25 20 15 10 5 0 0

0.1

0.05

0.15

0.2

0.25

Mass concentration (µg l−1) BW 1/02 NW 1/03

BW 2/04 NW 2/05

BW 2/09 NW 4/07

BW 4/06

HH 10.LURV

Parametric value

Figure 7 CV% vs. mass concentration for benzo(a)pyrene. Results from PT exercises during 2002–09. The various PT rounds are indicated by different symbols. From www.aqsbw.de.

160

Measurement Quality in Water Analysis

Variation coefficient (%)

25

20

15

10

5

0 0

BW 1/06

10

20

BW 2/03

30

40 50 60 Mass concentration (µg l−1)

HH 08

NW 1/07

NW 3/03

70

80

NW 3/04

90

100

Parametric value

Figure 8 CV% vs. mass concentration for lead. Results from PT exercises during 2002–09. The various PT rounds are indicated by different symbols. From www.aqsbw.de.

45 Variation coefficient (%)

40 35 30 25 20 15 10 5 0 0

2

4

6

8

10

12

16

14

18

20

Mass concentration (µg l−1) BW 2/06

BW 3/08

BW 4/03

NW 4/02

NW 4/03

NW 4/04

NW 2/07 Parametric value

Figure 9 CV% vs. mass concentration for tetrachloroethene. Results from PT exercises during 2002–09. The various PT rounds are indicated by different symbols. From www.aqsbw.de.

Validation demonstrates that the method used in this laboratory at a given time was fit for purpose and all significant effects on the measurement result were taken into account. Metrological traceability demonstrates that the measurement results are traceable to the metrological references used and measurement uncertainty provides information on the level of confidence that can be placed on the measurement result. QC assures that the measurement results (including uncertainty) are of the same quality as at the time of validation. The requirements on how to demonstrate measurement quality for

an analytical laboratory are laid down in chapter 5 in ISO/IEC 17025:2005.

3.07.5.1 Method Validation When a laboratory implements a test method for the first time, it is essential that the performance of the method is studied prior to the analysis of test samples to confirm that the method is suitable for the required application. Table 5 lists the aspects of method performance which may require study as part of a

Measurement Quality in Water Analysis

161

35

Variation coefficient (%)

25 20 15 10 5 0 0

0.1

0.2

0..3

0..4

0..5

0.6

0.7

NW 2/09

NW 4/06

0.8

0.9

Mass concentration (µg l−1) BW 1/02

BW 5.LÜRV

HH 2/05

HH 5.LÜRV

NW 2/04

Parametric value

Figure 10 CV% vs. mass concentration for atrazine. Results from PT exercises during 2002–09. The various PT rounds are indicated by different symbols. From www.aqsbw.de.

Variation coefficient (%)

12 10 8 −

6

× −+

4

−× -

−

+×

-

2

−

+

× -

−

−

-

0 0

BW 2/08 NI 1/06 − NW 1/04

20

40

BW 4.LÜRV NI 2/05 NW 3/06

60

80 100 120 Mass concentration (µg l−1)

BW 4/02 × NI 2/06 NW 1/09

140

160

BW 4/05 HH 4.LÜRV + NI 4/06 NI 4/05 Parametric value

180

200

NI 1/05 - NI 1/05

Figure 11 CV% vs. mass concentration for nitrate. Results from PT exercises during 2002–09. The various PT rounds are indicated by different symbols. From www.aqsbw.de.

validation exercise. The extent of the validation study carried out by a laboratory depends on the history of the method being implemented and the criticality of the application. Two scenarios commonly encountered in laboratories are:

• •

implementation of a standard method which has been previously validated through an interlaboratory study and development and validation of a method within a single laboratory for its own use (often referred to as ‘in-house validation’).

Guidance on validation is available in a number of texts including documents produced by Eurachem (1998) and IUPAC (Thompson et al., 2002).

3.07.5.1.1 Implementation of a standard method A standard method is a method that has been published by a national or international standards body (e.g., ISO), or by a sectoral organization. The validation of the method will have been carried out prior to publication (often by means of an

162

Measurement Quality in Water Analysis

Table 4 Comparison of the maximum standard uncertainty with CV% obtained in proficiency tests (PT) Maximum standard uncertainty (%)

Average CV% in PT

Continued

Requirements fulfilled

Major components Ammonium Chloride Conductivity Fluoride Iron Manganese Nitrate Nitrite Oxidizability Sodium Sulfate

8 8 8 8 8 8 8 8 19 8 8

7 3 1 7 8 9 4 5 9 4 3

Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes

Trace elements and ions Aluminum Antimony Arsenic Boron Bromate Cadmium Chromium Copper Cyanide Lead Mercury Nickel Selenium

8 19 8 8 19 8 8 8 8 8 13 8 8

12 19 13 6 36 10 8 5 19 15 17 9 17

No Yes No Yes No No Yes Yes No No No No No

Polycyclic aromatic hydrocarbons Benzo(a)pyrene 19 Benzo(b)fluoranthene 19 Benzo(ghi)perylen 19 Benzo(k)fluoranthene 19 Indeno(1,2,3-cd)pyrene 19

30 21 28 22 27

No No No No No

Volatile organic trace compounds Benzene 19 1,2-Dichloroethane 19 Tetrachloroethene 19 Trichloroethene 19 Bromodichloromethane 19 Bromoform 19 Chloroform 19 Dibromochloromethane 19

26 23 20 20 15 16 15 16

No No No No Yes Yes Yes Yes

Pesticides and metabolites Aldrin Atrazine Bentazone Bromoxynil Chlortoluron 2,4-D 2,4-DB p,p0 -DDD p,p0 -DDE p,p0 -DDT Desethylatrazine Desisopropylatrazine Dichlorprop

26 17 27 29 18 29 34 25 25 30 24 30 23

No Yes No No Yes No No No No No No No No

19 19 19 19 19 19 19 19 19 19 19 19 19

Table 4

Dieldrin Dimethoate Diurno a-Endosulfan b-Endosulfan Endrin Fenoprop a-HCH d-HCH g-HCH Hexachlorobenzene Heptachlor Ioxynil Isoproturon Linuron MCPA MCPB Mecoprop Metazachlor Metolachlor Propazine Simazine 2,4,5-T Terbuthylazine

Maximum standard uncertainty (%)

Average CV% in PT

Requirements fulfilled

19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19

23 42 22 22 21 28 23 21 29 26 27 33 36 23 26 27 25 26 36 26 21 20 27 20

No No No No No No No No No No No No No No No No No No No No No No No No

The maximum standard uncertainty is estimated from EU drinking water directive according to Equation (1). In the column ‘Requirements fulfilled’ italics indicate requirement fulfilled and bolds indicate requirement not fulfilled. Bold italics indicate results not fulfilled but close to the requirement.

interlaboratory study). The organization of interlaboratory studies for method validation is described in ISO 5725. A laboratory wishing to use such a method, within its stated scope, will not generally need to carry out a full validation of the method prior to use. However, the laboratory must demonstrate that it can perform the method according to specifications and can achieve any performance criteria specified in the method such as targets for measurement repeatability or bias.

3.07.5.1.2 Single laboratory (in-house) method validation A more detailed validation study is required if a laboratory has modified a standard method or developed a method in-house. The parameters that require study depend on the scope of the method. Table 6 indicates the parameters which may require study during the validation of different types of methods. The validation study should aim to cover the scope of the method (i.e., representative analyte concentrations and sample types). Ideally, a precision study should address both repeatability and intermediate conditions (within-laboratory reproducibility) of measurement. The latter aims to provide an indication of the likely variability of results obtained from different batches of analyses (possibly produced by different analysts). Bias should be studied through the analysis of suitable certified reference materials (CRMs), if available. If there are no CRMs available, then alternative strategies are required – see Table 5.

Measurement Quality in Water Analysis Table 5

163

Method validation parameters

Parameter

Comments

Selectivity

The ability of the test method to determine the analyte(s) without interferences from other components (e.g., the sample matrix).

Precision

A measure of the dispersion of measurement results.

Repeatability, sr

Precision obtained under repeatability conditions of measurement: measurements made on the same material by a single analyst, using the same procedure, under the same operating conditions over a short time period. Often used to provide an estimate of within-batch variability of results.

Intermediate precision, sI or withinlaboratory reproducibility sRw

Precision obtained under intermediate conditions of measurement: measurements made on the same material using the same procedure, but over an extended time period and possibly by different analysts who may be using different pieces of equipment. Often used to provide an estimate of between-batch variability of results.

Reproducibility, sR

Precision obtained under reproducibility conditions of measurement: measurements being made on the same material using the same procedure but by different analysts working in different locations. Obtained from an interlaboratory study.

Trueness – bias

Estimated as the difference between the mean of several measurement results and a reference value. Experimental studies for estimating bias include:  the analysis of certified reference materials,  the analysis of spiked (fortified) samples, and  comparison with results obtained from a reference method and if none of these are available or suitable  comparison with results obtained from proficiency tests.

Measuring (working) range

The range of values over which the method has been demonstrated to produce results that are fit for purpose.

Limit of detection (LOD)

The minimum concentration of the analyte that can be detected with a specified level of confidence.

Limit of quantification (LOQ)

The lowest concentration of analyte that can be determined with an acceptable level of uncertainty.

Linear range

Part of the working range in which change in instrument/method response is directly proportional to the change in analyte concentration.

Ruggedness/robustness

The extent to which a test method is influenced by variation in operating conditions. Ruggedness testing evaluates how small changes in the method conditions affect the measurement result (e.g., changes in temperature, pH, reagent concentration).

Table 6 Method performance parameters required for the validation of in-house developed test methods for different types of analysis Parameter

Type of analysis Qualitative

Selectivity Precision Bias Limit of detection Limit of quantitation Linearity/ working range Ruggedness a

|

Quantitative Major componenta

Trace analysisb

Physical property

| | |

| | | |

| | |

|

3.07.5.2 Metrological Traceability As pointed out above, traceability can refer to the documentation, that is, sampling procedure, laboratory, analyst, method, etc., but as in ISO/IEC 17025:2005 we are dealing with traceability of measurement results. In other words, measurement results have to be traceable to the metrological references used. To be specific, we here use the wording metrological traceability (see Figure 12). An example of demonstration of traceability for a test method from a Eurachem leaflet (Eurachem, 2008) is given in Box 1.

| |

|

|

|

|

|

3.07.5.3 Measurement Uncertainty The basic requirements for uncertainty estimation are:

|

Major component: analyte concentration in the range 1–100% by mass or where detection of presence or absence is not an issue. b Trace analysis: analyte concentration less than 100 mg kg1 or where detection capability is important. Reproduced with permission from LGC Limited.

1. a clear definition of the measurand, that is, the quantity intended to be measured; 2. a comprehensive specification of the test method and the test items; and 3. a comprehensive analysis of the effects impacting the measurement results.

164

Measurement Quality in Water Analysis

Figure 12 Example of traceability to the SI – temperature. The temperature of a sample can be traced back through a chain of calibrations to the reference, an SI value of temperature at 0 1C. Illustration by Douglas Hasbun, Sweden.

Box 1

Metrological traceability to the stated references – example from a Eurachem leaflet: mercury in tuna fish

A measurement result (mass fraction) of mercury in a tuna fish sample is 4.0370.11 mg kg1. The result is reported as total Hg on dry weight basis (105 1C, 2 h, determined on a separate sample portion) and the measurement uncertainty is reported with a 95% level of confidence (k ¼ 2). In this case, mercury was determined with a mercury analyzer (AAS cold vapor) after a microwave digestion. The samples are weighed on a balance with a calibration certificate relating the weight measured to the SI unit kg. The acid digest is diluted in a volumetric flask where the manufacturer supplies the traceability of the volume of the flask to a national standard. The calibration curve was made using a CRM, a mercury standard with a certificate stating a quantity value of 0.99870.005 mg l1 (k ¼ 2) and with traceability to pure mercury. The method is validated using an appropriate matrix CRM with a total mercury concentration of 1.9770.04 mg kg1 (k ¼ 2). This validation is a check on the method performance. The evidence required by the laboratory to demonstrate traceability for the mercury result is shown below: 1. 2. 3. 4. 5. 6.

mass concentration of the Hg solution – a certificate of the CRM solution, mass of sample – calibration certificate of the balance, volume of volumetric flask – calibration certificate of the manufacturer, drying temperature – calibration of oven, digestion conditions – check according to specifications, and drying time – ordinary clock or stopwatch.

An introduction to measurement uncertainty is presented in a Eurolab report (Eurolab, 2007) and for further reading we recommend the Eurachem guide (Eurachem, 2000) and the fundamental reference document Guide to the Expression of Uncertainty in Measurement (BIPM, 2008a). Examples of different approaches for uncertainty estimation that can be used are presented in Figure 13. It is important to note that for most instrumental methods, the uncertainty is proportional to concentration and therefore a relative uncertainty is appropriate at levels well above the LOQ. At the LOQ level, it is appropriate to report the uncertainty in absolute terms, that is, in concentration unit used. In new ISO standards, measurement uncertainty may be included in the standard. The work for a single laboratory will then only be to show that their uncertainty is similar or lower than the uncertainty given in the ISO standard. Today, the reproducibility standard deviation from an interlaboratory comparison according to ISO 5725 is given in many standards and this can often be used as an estimate of standard uncertainty (ISO/TS 21748:2004). However, each laboratory has to estimate its own measurement uncertainty and for laboratories that already have the method in routine use we can recommend the single laboratory validation approach using

data for within-laboratory reproducibility from internal QC and data for the uncertainty on bias from validation or external QC (PT). The basic principle including the alternative to enlarge the uncertainty due to an observed but unknown bias (Magnusson and Ellison, 2008) is presented in Box 2 extract from the Eurolab report (Eurolab, 2007). This single validation approach described above is presented in detail in a guideline from Nordtest or Nordic Innovation Centre (NICe) from 2003 – TR 537 (Magnusson et al., 2003). An example summary from this handbook is given below for ammonium in water with data from an expert laboratory. The estimated uncertainty for ammonium–nitrogen is at least a factor of 2 lower than for routine laboratories at this mass concentration level of around 200 mg l1. No reference materials are available and the bias estimate is therefore based on results from PT. Details of the calculations are given in the handbook TR 537. Example from Nordtest handbook TR 537 is given as follows. Ammonium in water by ISO 11732:2005. Background for the NHþ 4 2N example – automatic photometric method. The laboratory has participated in six PTs or interlaboratory comparisons. All results have been somewhat

Measurement Quality in Water Analysis

165

Definition of the measurand, list of uncertainty components

Interlaboratory approach

Intralaboratory approach

Yes

Mathematical model?

Evaluation of standard uncertainties

Law of uncertainty propagation GUM

Modeling approach

Method performance

No

Method accuracy ISO 5725

Proficiency testing ISO/IEC 17043 + ISO 13528

Use of values already published + Uncertainty on the bias and factors not taken into account during interlaboratory study ISO TS 21748

Variability + Uncertainty on the bias and factors not taken into account during interlaboratory study

Organization of replicate measurements, method validation

Adding other uncertainty contributions, e.g., uncertainty on the bias

Single-laboratory validation approach

PT or PT method performance study

Interlaboratory validation approach

PT approach

Empirical approaches Figure 13 The different approaches to uncertainty estimation presented in the Eurolab report Data from Eurolab (2007) Measurement uncertainty revisited: Alternative approaches to uncertainty evaluation. Technical Report 1/2007, Paris. http://www.eurolab.org (accessed April 2010).

Box 2 Single laboratory validation approach. From Eurolab (2007) Measurement uncertainty revisited: Alternative approaches to uncertainty evaluation. Technical Report 1/2007, Paris. http://www.eurolab.org (accessed April 2010). The basic principle behind this approach is the synthesis of uncertainty estimates from estimates of precision and estimates of bias: * * *

Measurement accuracy ¼ precision and trueness Measurement uncertainty ¼ within-laboratory reproducibility and uncertainty on bias Measurement uncertainty is estimated as a root sum of squares of a standard deviation s characterizing the (im)precision of the measurement and an estimate b accounting for measurement bias, which gives the standard uncertainty u according to the schematic equation:

u¼

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi s2 þ b2

Here, it is understood that measurement bias is investigated, and corrective actions are taken to remove/reduce such bias to the greatest possible extent. The bias-related uncertainty estimate accounts for the potential bias left after correction. In practice, however, it happens quite often that significant bias is found, but the data are not sufficient for deriving a sound correction. For example, it may be doubtful whether a single-level correction, based on measurements of a single standard, is applicable to the entire measuring range. Then additional measurements, for example, including another standard, should be made in order to characterize the bias to an appropriate degree. If this is not possible or not practical, a pragmatic alternative is to increase the uncertainty to account for the observed bias instead of attempting any correction.

higher than the nominal value. The laboratory therefore concludes that there may be a small positive bias. On average, the difference has been þ 2.2%. This difference can be considered as a possible bias but since we have no traceable

reference values no correction is applied. Since it is not corrected for in their analytical results, but exists, it is treated as another uncertainty component. The data from the PTs are shown in Table 7.

166

Measurement Quality in Water Analysis

In Table 8, combined uncertainty, uc is calculated from the control sample limits and bias estimation from results from PT that the laboratory has participated in. The sR from interlaboratory comparisons (ISO/TS 21748:2004) using the same method can also be used if a higher uncertainty is acceptable. Measurement uncertainty U (95% confidence interval) is estimated to 77% (Table 9). The customer demand is 710%. The calculations are based on internal QC (control chart limits) and results from PT. The use of the reproducibility standard deviation from the standard would lead to an expanded uncertainty estimate of 716%.

3.07.5.4 Quality Control 3.07.5.4.1 Internal QC For guidance in Internal QC the Nordtest handbook Trollbook (Hovind et al., 2007) gives an introduction with detailed examples. The following text is based on this Trollbook. Internal QC at the chemical analytical laboratory involves a continuous, critical evaluation of the laboratory’s own analytical methods and working routines. The control encompasses the analytical process starting with the sample entering the laboratory and ending with the analytical report. The most important tool in this QC is the use of control charts. The basis is that the laboratory runs control samples together with the test samples. The control values are plotted in a control Table 7 Evaluation of results for a laboratory from six proficiency tests of NHþ 4 –N in water Exercise

1999 1 2 2000 1 2 2001 1 2

Assigned value xref (mg l 1)

Laboratory result xi (mg l 1)

Difference (%)

sR (%)

Number of labs

81 73

83 75

2.4 2.7

10 7

31 36

264 210

269 213

1.9 1.4

8 10

32 35

110 140

112 144

1.8 2.9

7 11

36 34

þ 2.18

8.8

34

 X RMS (root mean square)

Table 8

chart. In this way it is possible to demonstrate that the test method performs within given limits. If the control value is outside the limits, no analytical results are reported and remedial actions have to be taken to identify the sources of error, and to remove such errors. Figure 14 illustrates the most common type of control chart, the X-chart. Several types of control charts are available but the main two are X-charts for control samples, blank samples, and recovery, and R-charts (R ¼ range) for test samples. X-charts are used to control both trueness and precision, and R-charts are used purely for controlling precision (repeatability). Normally, control charts are constructed using statistically calculated warning and action limits. In order to set robust control limits (warning and action limits), it is recommended (Hovind et al., 2007) that the standard deviation is calculated from a large number of results (preferably 460) during a longer time period (preferably 4 year). In order to be able to start the QC, preliminary control limits are set which are modified after, for example, 1 year. When a QC program is established, it is essential to have in mind the requirements on the analytical results and for what purposes the analytical results are produced – the concept of fit for purpose. From the requirement on the analytical results, the analyst sets up the control program: 1. 2. 3. 4.

type of QC sample, type of QC charts, control limits – warning and action limits, and control frequency.

The control limits are then set based on customer or legislative targets or on target measurement uncertainties (ISO/TS 13530:2009). Guidance on how to use target measurement uncertainty to set control limits can be found in examples 1 and 2 in the Trollbook. It is also important to have very basic and simple rules for deciding if a control value is in control – that is, if the analyst can report the results. The normal rules (ISO 8258:1991) for Table 9

Calculation of combined and expanded uncertainty

Measurand

2.25

Ammonium– nitrogen

Combined uncertainty uc

Expanded uncertainty U

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1:67 2 þ 2:71 2 ¼ 3:18%

3.18  2 ¼ 6.4E 7%

Calculation of standard uncertainties for a laboratory measuring NHþ 4 –N in water Value

Relative u(x)

Comments

Reproducibility within-laboratory, Rw  ¼ 200 mg l1 Rw Control sample X

Warning limits are set to 73.34%

1.67%

3.34% / 2 ¼ 1.67%

Method and laboratory bias Proficiency testing

RMSbias ¼ 2.25%

2.71%

Bias

u(xref) ¼ 1.5% Reproducibility between laboratories Standard method sR

8%

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi RMS2difference þ uðx ref Þ2 sR uðx ref Þ ¼ pffiffiffiffiffiffiffiffiffi nLab uðbiasÞ ¼

EN / ISO 11732 at level 200 mg l1

Measurement Quality in Water Analysis

167

70

µg l−1

65

60

55

50 1-Feb

22-Mar

10-Mar

28-Jun

16-Aug

4-Oct

22-Nov

10-Jan

28-Feb

Date of analysis Figure 14 Example of an X control chart for the direct determination of zinc in water. All control values in the green area (within the warning limits) show that the determination of zinc performs within given limits and the routine sample results is reported. Control values in the red area (outside the action limits) show clearly that there is something wrong and no routine sample results are reported. A control value in the yellow area is evaluated according to specific rules. From Hovind H, Magnusson B, Krysell M, Lund U, and Ma¨kinen I (2007) Internal quality control – handbook for chemical laboratories, 3rd edn. Nordtest Technical Report 569. http://www.nordicinnovation.net/nordtest.cfm (accessed April 2010).

deciding if a control value is in control are based on statistical process control and are, in the authors’ view, not directly applicable to chemical analysis. We recommend the following rules for in control proposed in the Trollbook. The three rules for QC from the Trollbook are as follows: 1. The method is in control if:

 the control value is within the warning limits and  the control value is between warning and action limit and the two previous control values were within warning limits

In this case the analyst can report the analytical results. 2. The method is in control but can be regarded as out of statistical control if all the control values are within the warning limits (maximum one out of the last three between warning and action limit) and if:

 seven control values in consecutive order gradually increasing or decreasing (10) and

 10 out of 11 consecutive control values are lying on the same side of the central line (10)

In this case, the analyst can report the analytical results but a problem may be developing. Important trends should be discovered as early as possible in order to avoid serious problems in the future. Any long-term trends are taken care of in the annual long-term evaluation of the control charts – see Trollbook. 3. The method is out of control if:

 the control value is outside the action limits and  the control value is between the warning and the action limit and at least one of the two previous control values is also between warning and action limit – the rule two out of three.

In this case, normally no analytical results can be reported. All results since last value in control was obtained must be reanalyzed.

3.07.5.4.2 External QC – PT PTs are useful to help the participating laboratories to assess their own performance and to identify gaps. Of course also customers and authorities may use such results to identify laboratories that fulfill their quality requirements. The usual procedure is that a PT provider prepares water samples and distributes them to the participating laboratories. After finishing the analyses the results of the measurements are sent back to the provider, where the results are compared with an assigned value determined either from external references or from the consensus of the participants. Statistical analysis is done and an assessment is made usually using a scoring procedure delivering a z-score. This z-score compares the difference of the participant’s result (x) from the assigned value (X) with a standard deviation for proficiency assessment ð^ sÞ (Equation (2)). The latter can be determined from the data set in the PT round or from external quality requirements, for example, for the water sector the standard deviation ð^ sÞ can be set equal to the standard uncertainty calculated from the WFD or the EU water directive (see above) (for more details, see ISO 13528:2005):

z¼

xX ^ s

ð2Þ

If we assume that the requirements are exactly met, we expect s, that is, |z| p 2.0. that 95% of the data will lie within X 7 2^ For an individual laboratory this means that values between 2 and 3 are questionable and |z| p 3 require corrective actions. In most cases, a consensus mean or median is used as assigned value and the comparison of the laboratory’s result with the assigned value, strictly speaking, only delivers the comparability of the result with the results of other laboratories. Only if we have a metrologically traceable reference value or we can assume that the assigned value is a good estimate for the true value we can calculate the overall bias for all the laboratories participating in the PT. Example of PT with traceable reference values is the International Measurement

168

Measurement Quality in Water Analysis

Evaluation Programme (IMEP). For the water sector there are recent developments to introduce traceable reference values into routine PTs for water analysis (Rienitz et al., 2007; Koch and Baumeister, 2008). If several different analytical techniques are used, it is recommended to calculate the bias for each technique. PT results can demonstrate the quality of measurements to customers, authorities, and accreditation bodies.

4. Beho¨rde fu¨r Soziales, Familie, Gesundheit und Verbraucherschutz, Hamburg; and 5. Niedersa¨chsisches Landesgesundheitsamt, Aurich. The authors acknowledge the following for their valuable contributions to this chapter: Vicki Barwick, Marina Patri¨ rnemark for contributing arca, Elizabeth Prichard and Ulf O to the discussion regarding terminology. Vicki Barwick for reviewing the chapter and contributing material on method validation.

3.07.6 Data Treatment The QA/QC directive for WFD directive also gives guidance on data treatment – how to calculate mean values if the data set contains values below LOQ. This is a critical issue and can often vary from laboratory to laboratory and rules are here needed so that mean values and sums are calculated in the same way. The following are proposed in the directive:

•

•

First is to calculate a mean value for one parameter over for, for example, a year: ‘‘Where the amounts of physicochemical or chemical measurands in a given sample are below the limit of quantification, the measurement results shall be set to half of the value of the limit of quantification concerned for the calculation of mean values.’’ Second is to calculate a sum of several parameters, for example, pesticides: ‘‘ y results below the limit of quantification of the individual substances shall be set to zero.’’

3.07.7 Conclusions There is a general need for improvement of analysis of trace components in the water sector. With the needs clearly defined, measurement quality can be assessed for laboratories in general using PT. The results (2002–09) show that the requirements set in the EU drinking water directive are met for main components including nutrients and conductivity. For trace elements/ions, the requirements are met for some parameters. For trace organic compounds, the requirements are met for the different trihalomethanes but not for most of the other organic compounds. For a single laboratory the quality can be assessed when a new test method is implemented using the main pillars (traceability, validation, and uncertainty). For the ongoing assessment on the measurement quality the internal and external QC are vital and this information should be open to customers, authorities, and accreditation bodies in order to assess the quality.

Acknowledgments The permission to use the data from the following PT providers for Figures 7–11 is gratefully acknowledged: 1. AQS Baden-Wu¨rttemberg, Universita¨t Stuttgart; 2. Landesinstitut fu¨r den o¨ffentlichen Gesundheitsdienst Nordrhein-Westfalen, Mu¨nster; 3. Landesamt fu¨r Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen, Recklinghausen;

References BIPM (2008a) Evaluation of measurement data – guide to the expression of uncertainty in measurement (GUM), JCGM 100:2008. http://www.bipm.org/en/publications/ guides/gum.html (accessed April 2010). BIPM (2008b) International vocabulary of metrology – basic and general concepts and associated terms (VIM), JCGM 200:2008. (Also published as ISO/IEC Guide 99:2007.) http://www.bipm.org/en/publications/guides/vim.html (accessed April 2010). Commission of the European Communities (2009) Commission Directive 2009/90/EC of 31 July 2009 laying down, pursuant to Directive 2000/60/EC of the European Parliament and of the Council, technical specifications for chemical analysis and monitoring of water status. Official Journal of the European Union L 201/36, 1.8.2009. Council of the European Union (1998) Council Directive 98/83/EC of 3 November 1998 on the quality of water intended for human consumption. Official Journal of the European Communities L 330/32, 5.12.98. Currie LA (1995) Nomenclature in evaluation of analytical methods including detection and quantification capabilities (IUPAC recommendations). Pure and Applied Chemistry 67: 1699--1723. EAQC-WISE (2010) EAQC-WISE – European Analytical Quality Control in support of the WFD via the Water Information System for Europe. http://www.eaqc-wise.net (accessed April 2010). Eurachem (1998) The fitness for purpose of analytical methods: A laboratory guide to method validation and related topics. http://www.eurachem.org (accessed April 2010). Eurachem (2000) Quantifying Uncertainty in Analytical Measurement, 2nd edn. http:// www.eurachem.org (accessed April 2010). Eurachem (2008) Metrological traceability of analytical results – a Eurachem leaflet. http://www.eurachem.org (accessed April 2010). Eurolab (2007) Measurement uncertainty revisited: Alternative approaches to uncertainty evaluation. Technical Report 1/2007, Paris. http://www.eurolab.org (accessed April 2010). European Parliament and the Council of the European Union (2000) Directive 2000/60/ EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for community action in the field of water policy (Water Framework Directive, WFD). Official Journal of the European Communities L 327/1, 22.12.2000. European Parliament and the Council of the European Union (2008) Directive 2008/ 105/EC of the European Parliament and of the Council of 16 December 2008 on environmental quality standards in the field of water policy (QA/QC directive). Amending and subsequently repealing Council Directives 82/176/EEC, 83/513/ EEC, 84/156/EEC, 84/491/EEC, 86/280/EEC and amending Directive 2000/60/EC of the European Parliament and of the Council. Official Journal of the European Union L 348/84, 24.12.2008. Hovind H, Magnusson B, Krysell M, Lund U, and Ma¨kinen I (2007) Internal quality control – handbook for chemical laboratories, 3rd edn. Nordtest Technical Report 569. http://www.nordicinnovation.net/nordtest.cfm (accessed April 2010). ISO 5667-20:2008 Water quality – sampling – part 20: Guidance on the use of sampling data for decision making – compliance with thresholds and classification systems. ISO 5725-1:1994 Accuracy (trueness and precision) of measurement methods and results – part 1: General principles and definitions. ISO 5725-2:1994 Accuracy (trueness and precision) of measurement methods and results – part 2: Basic method for the determination of repeatability and reproducibility of a standard measurement method. ISO 5725-3:1994 Accuracy (trueness and precision) of measurement methods and results – part 3: Intermediate measures of the precision of a standard measurement method.

Measurement Quality in Water Analysis

ISO 5725-4:1994 Accuracy (trueness and precision) of measurement methods and results – part 4: Basic methods for the determination of the trueness of a standard measurement method. ISO 5725-5:1998 Accuracy (trueness and precision) of measurement methods and results – part 5: Alternative methods for the determination of the precision of a standard measurement method. ISO 5725-6:1994 Accuracy (trueness and precision) of measurement methods and results – part 6: Use in practice of accuracy values. ISO 8258:1991 Shewhart control charts. ISO 11732:2005 Water quality – determination of ammonium nitrogen – method by flow analysis (CFA and FIA) and spectrometric detection. ISO 13528:2005 Statistical methods for use in proficiency testing by interlaboratory comparisons. ISO/IEC 17025:2005 General requirements for the competence of testing and calibration laboratories. ISO/TS 13530:2009 Water quality – guidance on analytical quality control for chemical and physicochemical water analysis. ISO/TS 20612:2007 Water quality – interlaboratory comparisons for proficiency testing of analytical chemistry laboratories. ISO/TS 21748:2004 Guidance for the use of repeatability, reproducibility and trueness estimates in measurement uncertainty estimation. Koch M and Baumeister F (2008) Traceable reference values for routine drinking water proficiency testing: First experiences. Accreditation and Quality Assurance 13: 77--82. Magnusson B and Ellison SLR (2008) Treatment of uncorrected measurement bias in uncertainty estimation for chemical measurements. Analytical and Bioanalytical Chemistry 390: 201--213.

169

Magnusson B, Na¨ykki T, Hovind H, and Krysell M (2003) Handbook for calculation of measurement uncertainty in environmental laboratories. Nordtest Technical Report 537. http://www.nordicinnovation.net/nordtest.cfm (accessed April 2010). Rienitz O, Schiel D, Gu¨ttler B, Koch M, and Borchers U (2007) A convenient and economic approach to achieve SI-traceable reference values to be used in drinkingwater interlaboratory comparisons. Accreditation and Quality Assurance 12: 615--622. Thompson M, Ellison SLR, and Wood R (2002) Harmonized guidelines for singlelaboratory. Validation of methods of analysis (IUPAC Technical Report) 74: 835--855.

Relevant Websites http://irmm.jrc.ec.europa.eu European Commission Joint Reasearch Centre, Institute of Reference Materials and Measurements; IMEP, Interlaboratory Comparisons. http://www.aqsbw.de Analytische Qualita¨tssicherung Baden-Wu¨rttemberg. http://www.eptis.bam.de EPTIS. http://www.iswa.uni-stuttgart.de Institut fu¨r Siedlungswasserbau,Wassergu¨te- und Abfallwirtschaft.

3.08

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

S Thiele, BM Fuchs, and RI Amann, Max Planck Institute for Marine Microbiology, Bremen, Germany & 2011 Elsevier B.V. All rights reserved.

3.08.1 3.08.2 3.08.2.1 3.08.2.2 3.08.2.2.1 3.08.2.2.2 3.08.2.3 3.08.2.4 3.08.2.5 3.08.2.6 3.08.2.6.1 3.08.2.6.2 3.08.2.6.3 3.08.2.6.4 56.2.6.5 3.08.3 3.08.3.1 3.08.3.2 3.08.3.3 3.08.3.3.1 3.08.3.3.2 3.08.3.3.3 3.08.3.4 3.08.4 3.08.5 References

Introduction The Full-Cycle rRNA Approach Ribosomal RNA Sources of 16S rRNA Sequences Pure cultures Microbial communities Clone Libraries Sequencing Sequence Analysis Probe Design Accessibility of the probe Selection of the oligonucleotide probe label Quality check of probes Adjusting probe specificity Clone-FISH Fluorescence In Situ Hybridization Fixation and Permeabilization Hybridization with Monolabeled Oligonucleotide Probes Catalyzed Reporter Deposition Fluorescence In Situ Hybridization Embedding, permeabilization, and inactivation of endogenous peroxidases Hybridization Catalyzed reporter deposition Troubleshooting Cell Counting From Cell Detection to Ecological Function

3.08.1 Introduction The microscopic discovery of bacteria in dental plagues in 1683 by van Leeuwenhoek marks the beginning of the identification, localization, and quantification of microorganisms in their environment. In the late nineteenth century, Koch’s cultivation techniques with agar plates led to the discovery of most of the pathogens in short time. However, it took many more years to realize by staining techniques that readily culturable microorganisms represent only a minor fraction of the microbes present in the environment, for example, the marine pelagial (Jannasch and Jones, 1959). With the advent of molecular biology, uncultured organisms also became accessible to directed research in the late twentieth century. Since that time a wealth of new freshwater and marine microorganisms have been discovered, identified, and abundances in different environments have been measured. The focus has broadened from mainly free-living bacteria to complex microbial communities in various environments, such as sediments, biofilms, or tissues. Nevertheless, planktonic microorganisms in limnic and marine ecosystems remain of high interest. Bacterioplankton communities in freshwater systems were found to be mainly formed by Cyanobacteria, Betaproteobacteria, Bacteroidetes, and Gram-positive Actinobacteria

171 171 171 172 172 173 173 174 174 176 176 176 177 177 179 180 180 180 181 182 182 183 183 184 186 187

(Glo¨ckner et al., 1999; Simek et al., 2001; Zwart et al., 2002). In contrast, marine bacterioplankton communities consist of mainly Cyanobacteria (e.g., Prochlorococcus), Alphaproteobacteria (e.g., SAR11), Gammaproteobacteria (e.g., SAR86), Bacteroidetes (e.g., Polaribacter), and, interestingly, of Archaea from the marine group I (see Partensky et al. (1999) and Giovannoni and Stingl (2005) for reviews). Besides almost cosmopolitan bacteria such as SAR11, which can make up to 50% of the bacterioplankton cells in some marine environments (Morris et al., 2002), other microorganisms can be found only in specific environments. These environments may be characterized by different water temperatures or different nutrient availabilities, for example, members of the clade Roseobacter dwell in the Southern Ocean or in correlation with high concentrations of particulate organic matter (Selje et al., 2004; Buchan et al., 2005). Yet, other bacteria are restricted to extreme environments such as hot springs (Marsh and Larsen, 1953), saline lakes (Oren, 1999), hydrothermal vents (Jeanthon, 2000), and man-made systems such as activated sludge (Snaidr et al., 1997). These findings indicate that microbial life is of a broad variety and thus taxonomy and determination of microorganisms are not trivial. One of the first classification systems of microorganisms was invented by Ferdinand Cohn in 1872.

171

172

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

The system was based on morphological markers and classified six genera of bacteria as a clade of plants (Cohn, 1872). During the early twentieth century, more and more physiological and biochemical tests on pure cultures were added to the list of discriminative markers, resulting in a more detailed, but, not-yet-stable classification (Garrity et al., 2001). In the second half of the twentieth century, improved chemical analytics made it possible to include DNA base composition, the analysis of lipids, isoprenoid quinones, and cytochromes as well as cell-wall composition. In the late 1970s and early 1980s, new molecular biological techniques enabled another major step forward in the classification of microorganisms. The comparative sequence analysis of the RNA of the small subunit (SSU) of the ribosome finally stabilized our view of the tree of life and led to the recognition of the domains Bacteria, Archaea, and Eukarya (Woese et al., 1975; Woese and Fox 1977; Woese, 1987). In the following, we focus on a set of methods, called the ‘full-cycle ribosomal RNA (rRNA) approach’, and the determination and quantification of bacteria in the environment, using fluorescence in situ hybridization (FISH).

3.08.2 The Full-Cycle rRNA Approach The full-cycle rRNA approach was developed as a phylogenybased toolbox for cultivation-independent studies of microbial diversity and ecology. Figure 1 shows a flow diagram

of this approach, starting with different sources of sequence information. After DNA extraction, the 16S rRNA genes are amplified by polymerase chain reaction (PCR) using conserved primers. The amplified rRNA genes are singularized by cloning in a plasmid vector and transformation in competent Escherichia coli cells. They are then sequenced and submitted to sequence databases. Comparative sequence analysis is the basis for the design of oligonucleotide probes. Finally, these probes can be applied to environmental samples using FISH techniques (Amann et al., 1995).

3.08.2.1 Ribosomal RNA The rRNA provides some characteristics, which predestines this molecule as a phylogenetic marker. First of all, the rRNA is an integrated part of the ribosome, the protein factory of each cell, and has the same function in every organism. The rRNA molecules must have developed in early stages of life and thus are evolutionarily conserved even in its two- and three-dimensional structures. Although the primary structure, that is, the nucleotide sequences of rRNAs, is highly conserved throughout all organisms, some regions in the rRNA are more conserved than others. For example, while some regions are identical across all domains, other regions are more variable and specific for particular genera or even species. Today, the rRNA is widely accepted as a global marker for phylogenetic studies, which is essentially lacking artifacts from

Environmental/ Environmental/culture Culture Sample sample

Extracted Extracted nucleic Nucleicacid Acid DNA rRNA

Nucleic Nucleic acid acid probe probe

rDNA clones Clones

rDNA sequences Sequences Comparative analysis Analysis rDNA database Data Base Hybridization

Sequencing

Figure 1 Flow scheme of the full-cycle rRNA approach. Modified from Amann RI, Ludwig W, and Schleifer KH (1995) Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiological Reviews 59: 143–169.

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

lateral gene transfer (for reviews, see Olsen et al. (1986), Woese (1987), or Pace (2009)). The conserved sequence stretches on the rRNA provide a variety of technical advantages. Primers can be designed to amplify, for example, the 16S rRNA genes from almost all bacteria in environmental samples by PCR. rRNA is highly abundant in cells, making it suitable as a target for in situ hybridization studies. For example, a cell from a logarithmically growing E. coli culture contains up to 70 000 ribosomes. Ribosomes consist of two subunits. The large subunit (LSU) of the bacterial and archaeal ribosomes contains the 5S rRNA (B120 nucleotides) and the 23S rRNA (B3000 nucleotides), while the SSU contains the 16S rRNA (B1600 nucleotides). The unit [S] refers to Svedberg, a sedimentation coefficient in ultracentrifugation. In contrast to Bacteria and Archaea, the eukaryotic ribosome contains in the LSU 5S, 5.8S, and 28S rRNA, and in the SSU 18S rRNA. Due to the smaller size, the sequencing of the 16S/18S rRNA is more convenient, compared to the longer 23S/28S rRNA. Additionally, 16S/ 18S rRNA has a much higher information content compared to 5S or 5.8S rRNA. Therefore, the rRNA of the SSU, the 16S rRNA of Bacteria and Archaea, and the 18S rRNA of the Eukarya were chosen as the most suitable markers to build up a phylogenetic database. Since this chapter centers on bacterial and archaeal identification, we focus on 16S rRNA, although most techniques can be applied to 18S, 23S, or 28S rRNA as well.

3.08.2.2 Sources of 16S rRNA Sequences 3.08.2.2.1 Pure cultures The classification of pure cultures by the full rRNA cycle is mostly straightforward, while the identification of organisms from environmental samples needs a few more steps. In the case of cultured organisms, biomass can be either picked from an agar plate or retrieved from liquid media via centrifugation and washing. In order to extract nucleic acids, cells are broken up by enzymatic lysis or mechanical cell disruption. After removal of cell fragments and proteins by standard sodium

173

dodecyl sulfate (SDS)/chloroform extraction, DNA and RNA can be precipitated (Box 1; Zhou et al., 1996). The rRNA genes are amplified from the extracted DNA using general primers (Table 1). The amplicon is purified and directly subjected to sequencing. The achieved sequences have to be checked for sequencing errors before they can be assembled and imported into a database. By comparative sequence analysis, phylogenetic trees can be calculated. Finally, the sequences are deposited in a public database such as European Bioinformatics Institute (EBI) or GenBank. Based on the retrieved sequences, specific oligonucleotide probes can then be designed for the localization and quantification of the organisms in the environment.

3.08.2.2.2 Microbial communities The analysis of the diversity of complex microbial communities by comparative sequence analysis and the identification of microorganisms without cultivation involves some additional steps to singularize and characterize the different types of rRNA sequences. Figure 1 summarizes the steps involved in the full-cycle rRNA approach for microbial ecology. In brief, first the sample is taken directly from the environment followed by a DNA extraction, as described previously. This is a critical step since Bacteria and Archaea have very different cell-wall composition, making it often very demanding to retrieve nucleic acids from all cells present in a sample. In order to amplify the rRNA genes, PCR with general primer pairs targeting the 16S rRNA genes of, for example, almost all bacteria (Table 1) is conducted. This results in a large number of 16S gene amplicons from most of the organisms present in the sample. These DNA fragments can now be analyzed by several methods to estimate the diversity of the microorganisms present, for example, by amplified ribosomal DNA (rDNA) restriction analysis, a method using restriction patterns on agarose gels (e.g., Smit et al., 1997), or a combined method of PCR amplification and denaturing high-performance liquid chromatography used for the analysis of polymicrobial

Box 1 SDS-based DNA extraction. Modified from Zhou JZ, Bruns MA, and Tiedje JM (1996) DNA recovery from soils of diverse composition. Applied and Environmental Microbiology 62: 316–322 Remarks: Modified version for DNA extraction from polycarbonate filters. This protocol is suitable for extracting DNA of a molecular weight up to 200 kb. Careful handling of DNA to avoid shearing is mandatory. Never vortex the DNA and use tips with wide opening. 1. Add 13.5 ml extraction buffer (100 mM Tris–HCl (pH 8.0), 100 mM ethylene-diamine-tetra-acetic acid (EDTA) (pH 8.0), 100 mM Na-phosphate (pH 8.0), 1.5 M NaCl, 1% CTAB) and 100 ml proteinase K (10 mg ml1 ) to a polycarbonate filter of 25-mm diameter which has been cut into small pieces; incubate for 60 min at 37 1C on a shaker. 2. Add 1.5 ml SDS (20%) and incubate for 120 min at 65 1C. 3. Centrifuge at c.50 000 g (15 min at room temperature (RT)), transfer supernatant into a fresh tube and add 4.5 ml extraction buffer and 0.5 ml SDS (20%) and incubate for 10 min at 65 1C. 4. Centrifuge using the same conditions and add chloroform/isoamylalcohol (24:1) to the supernatant. 5. Mix carefully, centrifuge at 10 000 g for 10 min and decant upper aqueous phase into a fresh tube. 6. Repeat chloroform/isoamylalcohol extraction steps, precipitate DNA from the aqueous phase by adding 0.6 volumes of isopropanol, and incubate at RT overnight. 7. Centrifuge at c.50 000 g at RT for 20 min, decant supernatant, and wash pellet with 10 ml ethanol (80%). 8. Centrifuge for 10 min, carefully decant supernatant, and dry pellet at RT. 9. Resuspend pellet in 200 ml PCR water or TE buffer (1:10) and incubate at 4 1C overnight. 10. Transfer DNA into a 1.5 ml PP tube, reduce volume to about 50 ml in the speed vac and store at 4 1C.

174 Table 1

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization Standard primers for the amplification of 16S and 23S rRNA genes from environmental samples

Primer

Sequence (5 0 - 3 0 )

Target

Reference

GM3F (8F) GM4R (1492R) 27F Univ1390R Arch20F ARC21F Arch958 1492R L189R

AGA GTT TGA TCM TGG C TAC CTT GTT ACG ACT T GTT GAT CCT GGC TCA G GAC GGG CGG TGT GTA CAA TTC CGG TTG ATC CTG CCG GA TTC CGG TTG ATC CYG CCR G YCC GGC GTT GAM TCC AAT T GGT TAC CTT GTT ACG ACT T TAC TGA GAT GYT TMA RTT C

Bacteria Bacteria Bacteria Bacteria and Archaea Archaea Archaea Archaea Archaea Bacterial 23S

Muyzer et al. (1995) Muyzer et al. (1995) Lane (1991) Zheng et al. (1996) DeLong (1992) Massana et al. (1997) DeLong (1992) DeLong (1992) Yu and Mohn (2001)

Modifications: M ¼ adenine or cytosine, R ¼ adenine or guanine; S ¼ guanine or cytosine; V ¼ guanine, adenosine, or cytosine, Y ¼ thimidine or cytosine.

infections (Domann et al., 2003). Denaturing gradient gel electrophoresis (DGGE) can be applied to differentiate between different microorganisms based on the guanine–cytosine (GC) content of the DNA (Muyzer et al., 1993; Myers et al., 1987). In a gel electrophoresis with a polyacrylamide gel containing a linear gradient of denaturants, the amplified 16S rRNA genes are separated in the gel according to their different GC content. Denaturants in the gel, for example, urea and formamide, lead to a transition of double-stranded DNA amplicons to partially melted structures at a particular point and thus to a stop of the fragments along the gradient in the gel. Sequences with a higher GC content will stop later in the gel at stronger denaturing conditions. AT-rich amplicons will denature earlier, thereby leading to a banding pattern in the gel. DGGE is well suited to have first insights into the microbial diversity of an environment. Individual bands can be cut from the gel, the rRNA gene fragment can be re-amplified, and the partial 16S rRNA can be determined. Another method used for environmental samples is the terminal restriction length polymorphism method (T-RFLP), using a fluorescently labeled primer for the PCR amplification of the genes and a restriction digest combined with a fluorescence detection of the resulting fragments by capillary electrophoresis (Abdo et al., 2006; Liu et al., 1997). Different lengths of the restriction fragments lead to different mobility in the gel or the capillary. The detection of the fluorescent dyes and the migration time allows the calculation of the fragment length. Different fragment lengths of the 16S rDNA fragments are indicative for different taxa. Different environments will result in characteristic and discernable banding patterns. However, pattern techniques such as DGGE and T-RFLP, in general, have only a limited resolution to assess the overall microbial diversity.

3.08.2.3 Clone Libraries Compared to the other techniques described above, clone libraries provide often a more detailed picture of the microbial diversity present in a given habitat (Box 2). The PCR product is purified after the amplification by the use of a commercial kit, for example, from Qiagen (Qiagen, Hilden, Germany), and checked for purity, size, and DNA amount via an agarose gel electrophoresis. The purified amplicon gene is cloned into plasmid vectors, for example, pCR 4-TOPO (Invitrogen, Carlsbad, USA), transformed into competent E. coli cells, and

singularized by plating on media, giving rise to single clones. From isolated clones, the plasmid containing the 16S rRNA insert can be isolated using a kit for plasmid extraction, such as PureLink Quick Plasmid Miniprep Kit (Invitrogen, Carlsbad, USA). To check whether inserts of the correct size and type are present, the isolated plasmids can be screened by PCR or agarose gel electrophoresis. A screening step can also be done from the clones directly via PCR; however, for sequencing purposes, a plasmid extraction is highly recommended to avoid cell fragments, proteins, etc., that might have an influence on the sequencing process.

3.08.2.4 Sequencing Subsequently, Sanger sequencing is done with primers targeting the cloning vector, for example, M13F and M13R (Invitrogen, Carlsbad, USA) (Box 3). High-quality sequencing of the 16S rRNA insert from clones is commercially available from specialized companies, for example, GATC-Biotech AG (Konstanz, Germany). The 16S rRNA sequences have now been determined for virtually all B8400 validly described species, and additionally for more than 1000 000 PCR-retrieved sequences of yet-uncultured environmental microorganisms. High-quality 16S rRNA sequences of cultured organisms (B7700) of valid taxonomic rank are currently assembled and curated in the Living Tree Project (Yarza et al., 2008). In total 41000 000 sequences of prokaryotic 16S rRNA have been determined and collected in databases. The SILVA project (Pruesse et al., 2007) and the Ribosomal Database Project (RDP) (Cole et al., 2009) are two established databases. It should be noted that new technologies, such as pyrosequencing (Ronaghi et al., 1996; Sogin, 2009) of PCR-amplified 16S rRNA fragments, enable highly parallel sequencing of potentially millions of 16S rRNA fragments, but the sequences are still shorter and of lower quality than those determined by Sanger sequencing. However, the development of new technologies is fast. This will then allow for an even faster processing and a higher throughput of environmental samples.

3.08.2.5 Sequence Analysis For a first quick phylogenetic identification of the sequences, a BLAST search (BLAST, Basic Local Alignment Search Tool) can

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

175

Box 2 Generation of a 16S rRNA gene clone library 1. Extracted DNA (see, e.g., Box 1) is checked for integrity and size by agarose gel electrophoresis. The DNA concentration is either determined spectophotometrically or estimated by using standards on the agarose gel. Do not use fragments smaller than 20 kb. 2. Amplify 16S rRNA genes via PCR reaction. 10  buffer 2 ml dNTPs (2.5 mM each dNTP) 2 ml (Optional: BSA 3 mg ml1 2 ml) Primer (50 mM) each 0.2 ml Eppendorf Master Taq 0.04 ml ( ¼ 0.2 units) Template 10 ng Add water to 20 ml final Primer sequences specific for most Bacteria (Muyzer et al., 1995): GM3F (50 -AGAGTTTGATCMTGGC-30 ) GM4R (50 -TACCTTGTTACGACTT-30 ) Thermocycling for GM3F/GM4R primers: 4 min at 94 1C, 1 min at 94 1C, 1 min at 48 1C (needs to be optimized for the template), 3 min at 72 1C, go to step 2 and repeat for 18–35 cycles (less means less chimera), 60 min at 60 1C (in order to obtain 100% A-overhang), and hold at 15 1C. 3. Purify reactions (e.g., Qiagen kit or, if necessary, gel purification); pool reactions, and run agarose gel for quantification and purity control. 4. Use TOPO TA Cloning Kit, Invitrogen or pGEM-T-Easy Vector System, Promega. In order to clone your vector, follow the instructions in the supplied manual. 5. Pick the clones with toothpicks first onto an agar plate then into liquid medium in a microtiter plate (MTP; 100 ml LBAMP) or pick the clones directly into the liquid medium in a MTP (100 ml LBAMP). Incubate MTP overnight. 6. Check clones via a screening PCR with the vector primers M13F and M13R 10  buffer 2 ml dNTPs 2.5 mM 2 ml BSA 3 mg ml1 2 ml 5  enhancer 4 ml Primer (50 mM) each 0.2 ml Eppendorf master Taq (1 U ml1) 0.2 ml Template 0.5 ml (from overnight culture) Add water to 20 ml final 7. Run aliquot of reaction on agarose gel to select the positive clones and purify selected clones with Multiscreen HV/Sephadex G50 plates.

Box 3 Sequencing of the cloned 16S rRNA genes Remark: This reaction uses M13F and M13R primers. If you use different primers adjust annealing time and temperature, if necessary. 1. Prepare sequence reaction with M13F and M13R primer and run sequencing. 5 ml sequence reaction: 1 ml BIG DYE, 1 ml 2.5  reaction buffer, 1 ml primer (5 mM), 0.5–1 ml template ( ¼ purified product of the screening PCR), and add water to 5 ml final. Thermocycling: 20 s at 96 1C, 10 s at 96 1C, 5 s at 55 1C, 4 min at 60 1C, go to 2 and repeat 60 cycles, and hold at 20 1C. 2. Purify reactions with Multiscreen HV/Sephadex G50 plates.

be done (Altschul et al., 1990). The next step in the workflow is the alignment of the retrieved sequences for comparing them with the databases, for example, RDP or SILVA. Alignment programs, such as the one implemented in the ARB

program package (Ludwig et al., 2004) or available at the SILVA webpage (Pruesse et al., 2007), can be used to automatically align several hundreds of sequences in one batch. The aligned sequences are then subjected to phylogenetic

176

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

analyses, resulting in the reconstruction of phylogenetic trees. The construction of phylogenetic trees should follow the standard operating procedure for phylogenetic inference (SOPPI; Peplies et al., 2008) using information of type strain sequences (Yarza et al., 2008). However, an in-depth description of the reconstruction of phylogenetic trees is beyond the scope of this chapter. The retrieved sequences can now be used to design probes in order to detect microorganisms in environmental samples. The reliability of such probes for the identification of cells in complex environments is strongly dependent on the quality of the 16S rRNA database. A comprehensive and well-maintained database is vital for rational probe design. Such a database can be found at the SILVA project site (Pruesse et al., 2007). Powerful software tools are required to manage such a database. We recommend the highly integrated software package ARB released by the Technical University of Munich (Ludwig et al., 2004). This program runs on different UNIX-based operating systems (including Linux); it is capable of maintaining 4500 000 aligned 16S rRNA sequences and allows the easy import of additional sequences from various sources. In an ideal world, probe design would be solely based on highquality, full-length sequences. In the real world, the public databases are filled with partial sequences and sequences that contain sequencing errors. The sequence gaps in partial sequences are severely hindering probe design by reducing the number of potential target regions. Databases with incomplete sequences also fail to provide reliable information about the current specificity of probes that have been designed in the past. However, the SILVA databases allow for the selection of high-quality sequences (e.g., Silva-Ref database) on the basis of different sequence qualifiers for optimal probe design. As of March 2010, the Silva-Ref database contained about 461000 high-quality sequences of 1200 base pair length, while the Silva-Parc database included all sequences with a minimal length of 300 base pairs and all qualities. By referring to such

Table 2

high-quality databases, oligonucleotide probes can be designed with specificities ranging from the species level up to levels such as phyla or even domain. Table 2 shows examples of probes which are frequently used in marine water samples. Additional probes for freshwater environments or man-made habitats such as wastewater treatment plants can be found at probeBase. This probe database was established by Loy et al. (2003) at the University of Vienna and it has currently more than 1500 entries.

3.08.2.6 Probe Design When none of the existing probes can be used, a new one has to be designed. For FISH, a probe length of 15–25 nucleotides – most often 18 nucleotides – is common. Sequence signatures serving as suitable target sites for nucleic acid probing can be conveniently searched with the PROBE_DESIGN tool within the ARB software package. First, a target group of organisms must be specified, for example, in a phylogenetic tree contained in ARB. The PROBE_DESIGN tool searches for possible signature sequences that are diagnostic for the selected species. The tool automatically excludes potential probe sequences which contain self-complementary regions with more than 3 nucleotides. The GC content of probe sequences influences their melting behavior. By default, this parameter is set in PROBE_DESIGN between 50% and 100% to ensure a tight binding. We recommend that the GC content of a newly designed probe should be between 50% and 70%, since a higher GC content could result in unspecific binding. Sometimes the PROBE_DESIGN tool cannot find a suitable probe target site. However, the program provides options for modifying the search parameters to look for signatures in subsets of the group originally selected, or by choosing to allow for the signature to be found in a defined number of species outside the target group (Ludwig et al., 2004). By the combinatorial use of probes with overlapping specificity, the

List of examples of frequently used probes for marine planktonic samples

Probe

Target organisms

Sequence (5 0 - 3 0 )

Arch915

Archaea

GTGCTCCCCCGCCAATTCCT

Eury806

CACAGCGTTTACACCTAG

0

GCTGCCTCCCGTAGGAGT GCAGCCACCCGTAGGTGT GCTGCCACCCGTAGGTGT ACTCCTACGGGAGGCAGC GGTAAGGTTCTGCGCGTT CAACGCTAACCCCCTCC

35 35 35 35 35 35

Amann et al. (1990a) Daims et al. (1999) Daims et al. (1999) Wallner et al. (1993) Neef (1997) Eilers et al., (2001)

Bet42a

Euryarchaeota marine group II Bacteria Supplement to EUB338 Supplement to EUB338 Control Alphaproteobacteria Clade Roseobacter and relatives Betaproteobacteria

35

Manz et al. (1992)

Gam42a

Gammaproteobacteria

GCCTTCCCACATCGTTT

35

Manz et al. (1992)

CF319a Pla46

Bacteroidetes Planctomycetes

TGGTCCGTGTCTCAGTAC GACTTGCATGCCTAATCC

35 30

Manz et al. (1996) Neef et al. (1998)

Eub338 Eub338-II Eub338-III Non338 Alf968 Ros537

a

GCCTTCCCACTTCGTTT

Formamide concentration in CARD-FISH hybridization buffer. More probes can be found at probeBase (http://www.microbial-ecology.de/probebase).

Competitor/helper sequence (5 0 - 3 0 )

cGam42a GCCTTCCCACATCGTTT cBet42a GCCTTCCCACTTCGTTT

FAa(%)

Reference

35

Stahl and Amann (1991) Teira et al. (2004)

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

177

3.08.2.6.2 Selection of the oligonucleotide probe label

wavelength B525 nm (suitable for Alexa 488, Fluorescein, Atto 488, etc.). It is important to note that the CY5 and Alexa 647 derivatives emit light in the near infrared (see Table 3) and are only detectable with a charge-coupled device camera or on a confocal laser scanning microscope with red laser excitation. Concomitant to the selection of the fluorescent label, care should be taken such that the right optical filters are chosen for the detection of the respective dyes. Well-adapted optical filters with a high transmission in the emission spectrum of the dye and a strong and sharp blocking of the excitation light are essential for a confident detection of weak signals. In nutrient-depleted environmental samples, oligonucleotides carrying only one fluorochrome may not be sensitive enough for detection of cells with low ribosome content (Pernthaler et al., 2002). Polynucleotide probes with a length of 4100 nucleotides carrying several fluorochromes may solve this sensitivity problem (Trebesius et al., 1994; DeLong et al., 1999). However, these probes do not allow for single mismatch discrimination and they are, therefore, lacking specificity for narrow target groups on the level of species and genera. An alternative labeling technique that increases fluorescence signal intensity uses horseradish-peroxidase (HRP)labeled oligonucleotides and a subsequent detection of this enzymatic label by catalyzed reporter deposition FISH (CARD FISH). The fluorescent staining in this case is the result of a secondary incubation with fluorescently labeled tyramides. The covalent and therefore permanent deposition of these labeled reporter compounds occurs only within cells hybridizing with the HRP-labeled probe. CARD-FISH signals are significantly brighter than FISH signals obtained with the same probe (Scho¨nhuber et al., 1997). Hoshino et al. (2008) determined a signal amplification of 26- to 41-fold. However, cell permeabilization protocols need to be adjusted in order to enable the larger enzyme-labeled oligonucleotides to penetrate into cells (Pernthaler et al., 2002).

A range of fluorochromes is available for the labeling of nucleic acids. However, not all fluorochromes are equally suited as labels for oligonucleotides. Newly developed dyes, in particular, should be checked for nonspecific staining. Standard labels for in situ hybridization are the green fluorescein and the red tetramethylrhodamine derivatives (Table 3). These dyes are well suited for standard applications when the ribosome content of target cells is high. In addition, these dyes can be used in conjunction to label different probes in double staining experiments. CY3 and CY5 (cyanine dyes) are members of the indocarbocyanine family. The high signal intensity of these two dyes makes them the fluorochromes of choice for detection of small cells with lower ribosome content like bacterioplankton cells. CY3 and CY5 emit strong fluorescence due to their high quantum yields and high molar extinction coefficients (Table 3). Different excitation and emission wavelength need different filter sets. These sets are available in standard configurations, for example, 65 HE Alexa 488 shift free (Zeiss, Jena, Germany) for Alexa 488 fluorescent dye, or in more variable configurations, for example, BrightLine HC 475/35 (AHF Analysetechnik AG, Tu¨bingen, Germany) for dyes with excitation wavelength B475 nm and emission

Oligonucleotide probes are custom-made by solid-phase synthesis. In the last step of synthesis, a fluorochrome is added to the 50 end of the oligonucleotide. Purified probe stocks are frequently delivered lyophilized. Upon reconstitution with 100-ml sterile water, a probe synthesis at 0.02 mmol scale yields approximately a 1500 ng ml1 stock solution. To determine the exact probe concentration, the absorbance of the 1:100 diluted stock solutions at 260 nm should be measured, assuming that 1 OD260 nm ¼ 20 ng ml1 DNA. Furthermore, the labeling of the oligonucleotide should be checked. For a pure monolabeled oligonucleotide, the ratio of absorption of the dye (Adye) versus the absorption of the nucleic acids at 260 nm (A260) should match the ratio of the extinction coefficients (e) of the dye and oligonucleotide. The extinction coefficient e at 260 nm (e260) of an oligonucleotide can be estimated from its nucleotide composition as the sum of the extinction coefficients of the individual nucleotides (deoxyadenosine triphosphate (dATP) ¼ 15.4 cm3 mmol1, deoxycytidine deoxyguanosine triphosphate (dCTP) ¼ 7.3 cm3 mmol1,

selected group of organisms may be fully targeted. Whenever possible, the identification of microbial cells in complex environmental samples should not be based on a single oligonucleotide probe, but rather on probe sets which contain, for example, two probes targeting the population of interest.

3.08.2.6.1 Accessibility of the probe A problem that should be considered during the design of FISH probes is target site accessibility. The higher-order structure of the ribosome may hinder the binding of the probe to its target site. The 16S rRNA in situ accessibility for oligonucleotide probes has recently been studied for two members of the domain Bacteria (E. coli, Pirellula sp. strain 1), one eukaryote (Saccharomyces cerevisiae), and one archaeon (Metallosphaera sedula) (Figure 2; Fuchs et al., 1998; Behrens et al., 2003). Furthermore, a complete accessibility map for the 23S rRNA of E. coli has been published (Fuchs et al., 2001). These color-coded maps clearly demonstrate dramatic differences in the binding of different fully complementary 18-mer probes to one batch of fixed target cells, which seem to be most strongly influenced by the type of secondary structure that is targeted. Although the secondary structure of the ribosome is highly conserved and a consensus map could be developed from the 16S rRNA accessibility studies, each probe should be checked on their respective target group of organisms to ensure high probe signals (Behrens et al., 2003). Inaccessible target regions can be made accessible by the use of unlabeled oligonucleotides, called ‘helpers’ (Fuchs et al., 2000). These bind adjacent to the diagnostic probe, thereby opening the target region for the probe. Helpers should be a few nucleotides longer than the diagnostic probe, that is, if the probe is an 18-mer, the helper should be a 21-mer, to ensure a tight binding beyond the melting point of the diagnostic probe.

3.08.2.6.3 Quality check of probes

178

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

1090

42 700

710

670

1160

45

1200

371020

800

1140

1150

1060

38

26 750

1190

1130

1180

39 1030

740

660

1170 10401050

780

1210

22

1010

650

640

1000

21 760

620

840

810

770

990

36

28

35

980

590

600

610

40 1070

1080 790

23

630

41

730

44

1120

43 25

680

690

1110

1100

720

24

1220 820

580 470

27

830

850

570 520

18 460

530

34 1260

970 860

20

880

480

890

960 1230

950

870

19 490

450

430

17

1

550

380

1380

1350

48 1510

40

4

1290

1300

1390

1500

390

940

930

1340

10

16

31

32

1400

3

50

1370

47

1330

1310

3

370

1410

1490

1530

1520

1360 1320

360

49

50

15

1540

5

350

14

920

20

2

30

420

330

900

910

560 540

1270

1280

1240

510

500

410

46

33

29

30

1250

1420 60

340

70

320 310

100

6

1480

3

80

300 50

1430

1470

Class I: 81−100%

13 110

290

Class II: 61−80%

7

280 240

1440

120

1460

12

Class III: 41−60%

270

130 250

230

1450

8 260

11

140

Class IV: 21−40% 150

9

220

160

210

200

180

10

Class V: 6−20%

170

Class VI: 0−5%

190

Figure 2 Accessibility map of the 16S rRNA of Escherichia coli for fluorescently labeled oligonucleotide probes. From Behrens S, Ru¨hland C, Inacio J, et al. (2003) In situ accessibility of small-subunit rRNA of members of the domains Bacteria, Archaea and Eucarya to Cy3-labeled oligonucleotide probes. Applied and Environmental Microbiology 69: 1748–1758.

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization Table 3

179

Dye labels frequently used for oligonucleotide probes and their characteristics

Carboxy-fluorescein (FAM)b Fluoresceinb Alexa 488 Atto 488 CY3a Alexa 546 Carboxytetramethyl-rhodamine (TAMRA) Alexa 594 Atto 590 (rhodamine-derivative) CY5a Alexa 647

Excitation (7 10 nm)

Emission (7 10 nm)

492 490 494 501 512/552 554 540 590 594 625–650 651

518 520 517 523 565/615 570 565 617 624 670 672

Molecular weight (Da) 376 389 643 766 1079 466 820 792 1250

e (mol1 cm1) 79 000 77 000 71 000 90 000 150 000 112 000 91 000 92 000 12 000 250 000 270 000

Data compiled from a Amersham Biosciences. b Sambrook J, Fritsch EF, and Maniatis T (1989) Molecular Cloning: A Laboratory Manual, 2nd edn., vol. I. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, and Invitrogen. e molar extinction coefficient. Note: Fluorescein and derivatives are pH sensitive and exhibit maximum fluorescence at pHZ9.

triphosphate (dGTP) ¼ 11.7 cm3 mmol1, thymidine 50 triphosphate (dTTP) ¼ 8.8 cm3 mmol1, from Sambrook et al., 1989). Taking into account the extinction coefficient of the dye (edye; see also Table 3), the quality of labeled oligonucleotide can be estimated by calculating a ratio k according to the following:

k¼

e260 =edye A260 =Adye

Values of ko1 indicate an incomplete labeling of a probe, whereas values 41 point to the presence of additional, potentially unbound dye. Considering inaccuracies in the estimation of the extinction coefficients of oligonucleotides, k-values between 0.7 and 1.3 are acceptable. Working solutions are prepared at concentrations of 50 ng ml1 and stored in the dark at –20 1C. Only small portions of probe working solutions (50–100 ml) should be prepared, since repeated freeze–thawing may damage the probe and might result in a precipitation of the probe, which in turn leads to weak hybridization signals and high background. HRP-labeled probes can be ordered from Biomers (Ulm, Germany). They are shipped in a lyophilized state and subsequently suspended in sterile H2O. It is important not to freeze HRP-labeled probe solutions, repeatedly, but to prepare small aliquots which are stored in the refrigerator at 4 1C. For calculating the quality of the probe, it has to be taken into account that the enzyme has a broad absorption maximum at 404 nm, and therefore contributes to the measured absorbance at 260 nm. This has to be considered in the determination of the probe concentration by the following correction:

OD260 ðoligoÞ ¼ OD260  OD404  0:276 A peak ratio (A260/A404) of around 3 indicates a good labeling of probe with HRP (Pernthaler et al., 2004). For checking the enzymatic activity of the HRP, the Amplex Red Hydrogen

Peroxide/Peroxidase Assay Kit can be used (Invitrogen, Carlsbad, USA). Following the manual, first a calibration curve is done with a peroxidise standard to which the activity of the HRP-labeled probe is compared. This test is ultrasensitive, and therefore performed on a probe solution diluted 10 000 from the probe stock solution (Jo¨rg Wulf, MPI Bremen, personal communication).

3.08.2.6.4 Adjusting probe specificity It is important that previously reported probes are reevaluated regularly against an updated database, such as SILVA (Pruesse et al., 2007), for specificity and target group coverage (see also Amann and Fuchs, 2008). First, an in silico check using online tools such as probeCheck (Loy et al., 2008) should be performed. Ideally, the probe of interest shows at least a mismatch to all nontarget microorganisms. For the design of new probes, it is important to keep these discriminatory positions central because a mismatch between probe and nontarget rRNA at the 30 - or 50 - end of the oligonucleotide is only weakly destabilizing. The discriminatory effect of a single mismatch can be increased by competitor oligonucleotides (Manz et al., 1992). They help to increase probe specificity. These competitor oligonucleotides have been shown to strongly suppress unspecific probe binding to a particular one-mismatch sequence. Competitors are mostly applied as unlabeled oligonucleotides, which are fully complementary to the mismatch-containing nontarget sequence. The probes and competitors specific for the Gammaproteobacteria (GAM42a) and Betaproteobacteria (BET42a) are prominent examples for this concept (Manz et al., 1992). Subsequently, the optimal hybridization conditions need to be established to guarantee high specificity and good sensitivity of the probe. For this, a series of hybridizations is performed at increasing stringency either by increasing the temperature of hybridization or by increasing the concentrations of a denaturing agent such as formamide in the

180

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

hybridization buffer (often referred to as melting curve; Figure 3). The changes in the fluorescence intensities of individual cells can be quantified by computer-assisted image analysis (Neef et al., 1996), by flow cytometry (Fuchs et al., 1998), or they can be semiquantitatively scored by a trained microscopist. The most desirable hybridization stringency is usually the one immediately before the target cell fluorescence begins to decrease. At this formamide concentration, the hybridization signal of the target cells is still maximal and that of nontarget organism should be low or absent. As a rule of thumb, an 18-mer oligonucleotide with a GC content between 50% and 60% will start to dissociate from its fully complementary rRNA target at a formamide concentration of approximately 30–40% in our standard buffer at 46 1C. Finally, new probes should be tested by FISH of isolates which have none, one, and more mismatches to the oligonucleotide at the optimized hybridization conditions (see Box 4).

This quite laborious method is based on the assumption that the temperature of dissociation from isolated rRNA is the same as from rRNA in fixed cells. On the other hand, 16S rRNA gene clones carrying the target sequence of a new probe can be used for adjusting the hybridization conditions (Clone-FISH). For this, the rRNA gene of interest must have been ligated in correct orientation into a vector with an inducible promoter upstream of the multiple cloning sites. Such clones are then grown with chloramphenicol and isopropylb-D-thiogalactopyranoside. This induction leads to an in vivo transcription of the cloned 16S rRNA gene and the accumulation of heterologous 16S rRNA of the uncultured organism inside the E. coli cell. After standard fixation, these induced E. coli cells, displaying the heterologous rRNA, can be used as analogs to cultured organisms for determining the melting point of probes (Schramm et al., 2002).

3.08.3 Fluorescence In Situ Hybridization 56.2.6.5 Clone-FISH

Probe conferred signal (sensitivity)

New probes are quite frequently designed to target yet uncultured microorganisms, which are only known from their rRNA gene sequences. In this case, it is not possible to test the probes using pure culture isolates. Two strategies are available to optimize the hybridization conditions in such a case. On the one hand, the cloned 16S rRNA gene of interest can be transcribed in vitro to RNA which is then blotted on a nylon membrane and hybridized with a labeled oligonucleotide at increasing levels of formamide (e.g., Pernthaler et al., 1998).

Target organism Nontarget organism

Temperature, formamide (stringency)

Specificity

Low

High

Low

Signal/noise = Max

Figure 3 Theoretical melting behavior of a probe to a target (straight line) and a mismatch containing organism (dotted line). The shaded area depicts the optimal working conditions for the probe.

Box 4

FISH with rRNA-targeted oligonucleotide probes can be used for identification, localization, and quantification of defined microbial populations in complex samples (Amann et al., 1995). The principle steps necessary for this phylogenetic staining (DeLong et al., 1989) are shown in Figure 4, and are described in more detail in the following sections (see Wagner et al. (2003) and Amann and Fuchs (2008) for reviews). Additionally, other FISH techniques provide powerful tools for environmental research and can be used to detect different target nucleic acids. Thus, transfer-messenger RNA (tmRNA) and messenger RNA (mRNA) can be used as targets for oligonucleotide probes. Within certain species, tmRNA can be used for discrimination of subspecies (Scho¨nhuber et al., 2001), while mRNA can be used to link expressed metabolic functions to cell identities (Pernthaler and Amann, 2004). Moreover, recognition of individual genes FISH using polynucleotide probes can be applied to detect certain genes within environmental samples and thus link the identity of a cell to its potential function (Zwirgelmaier et al., 2004). Another link of function, activity, and identity of microorganisms is provided by microautoradiography (MAR) FISH (e.g., Lee et al., 1999). However, in this chapter, we focus on FISH and the more sensitive CARD FISH.

3.08.3.1 Fixation and Permeabilization The first step in FISH analyses is the sampling and fixation of the cells. The most common fixatives are formaldehyde and ethanol. Fixation is critical for the whole FISH analysis. On the one hand, it prevents cell lysis, and thus preserves the cell

Melting curve analysis

1. Hybridize different microorganisms with none (full match) and one or more mismatches, according to the protocol (Boxes 6, 8, and 9) using different formamide concentrations (10–70% in steps of 5%) in the hybridization buffer. 2. Take pictures of the hybridized cells in an epifluorescence microscope using a fixed exposure time. 3. Analyze the brightness of the probe pictures via the program ImageJ and calculate mean brightness for the different formamide concentrations. The concentration with the highest differences of brightness values of the full match to mismatch organism should be used for hybridization (see also Figure 3). This is usually the concentration before the signal of the target cells is decreasing (see shaded area in Figure 3).

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

181

Sample (culture or environment) Fixation and permeabilization

Fixed and permeabilized cells

Flourescent dye Probe

Target (rRNA) Hybridization with fluorescent-labled oligonucleotide probes

Ribosomes

Hybridized cells

Quantification

Flow cytometry

Epifluorescence microscopy

Figure 4 Flow scheme of the analysis of an environmental sample by the FISH approach.

morphology. On the other hand, the permeabilization of the cell wall is necessary for the access of oligonucleotide probes to intracellular target sites. Bacteria, Archaea, and also fungi, algae, and protozoa have a broad variety of cell walls. Therefore, there is no single standard protocol available that fits all microorganisms (see Box 5 for basic protocols). Empirical optimizations should consider both, modification of individual steps of the standard protocols and additional treatments depending on the type of cell wall. Enzymatic digestions of the thick peptidoglycan layer of Gram-positive bacteria by lysozyme, the digestion of pertinacious cell walls using proteases, wax removal with solvents, the use of different detergents, or even short-time treatments in hydrochloric

acid are just some examples (Burggraf et al., 1994; Roller et al., 1994; Davenport et al., 2000; Pernthaler et al., 2004). All fixation protocols also disintegrate the cell membranes. Otherwise, the oligonucleotide probe could not diffuse to its ribosomal target sites. Consequently, cells fixed for FISH are always dead.

3.08.3.2 Hybridization with Monolabeled Oligonucleotide Probes Hybridization with fluorescently labeled oligonucleotide probes can be done in two formats. One format is the quick, but nonquantitative protocol for hybridization of cells on

182

Box 5

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

Fixation protocols for different sample types

1. Formaldehyde fixation of pure cultures with Gram-negative cell wall (Amann et al., 1990). A. Harvest cells during logarithmic growth from liquid medium by centrifugation of an aliquot of 2 ml by centrifugation for 10 min. at 10 000g, discharge supernatant and resuspend cells in 750 ml PBS buffer (145 mM NaCl, 1.4 mM NaH2PO4, 8 mM Na2HPO4, pH 7.4). B. Fix cells by adding 250 ml of a 4% PFA fixative (1% final concentration) and incubate for 1 h (for robust cells) to 24 h (for fragile cells) at 4 1C. C. Pellet cells by centrifugation (10 min at 10 000 g), discharge supernatant and thoroughly resuspend fixed cells in 500 ml PBS. D. Add 500 ml absolute ethanol and mix thoroughly, at this stage samples can be used for FISH or stored at  20 1C for several months. 2. Ethanol fixation of pure cultures with Gram-positive cell wall (Roller et al., 1994). A. Harvest cells during logarithmic growth by centrifugation of an aliquot (c.2 ml) for 10 min at 4000 g, discharge supernatant and wash cells in PBS buffer. B. Pellet cells by centrifugation (10 min at 10 000 g), discharge supernatant and add 500 ml PBS to resuspend cells thoroughly. C. Add 500 ml cold, absolute ethanol, at this stage samples can be stored at  20 1C for several months. 3. Fixation of planktonic samples (modified from Glo¨ckner et al. (1999)). A. Add formalin (37% formaldehyde) to a water sample to a final concentration of 1–3% and fix for 1–24 h at 4 1C; needs to be optimized for new sample types. B. Place a moistened support filter (0.45 mm pore size, cellulose nitrate, 47 mm diameter; Sartorius, Germany) and a membrane filter (0.2 mm pore size, white polycarbonate, 47 mm diameter; Millipore, Eschborn, Germany; shiny side up!) into a filtration tower; filter a known volume of the fixed sample by applying gentle vacuum; support filters may be utilized for several samples; for cell numbers of around 106 ml1, 10 ml of sample is generally sufficient. C. After complete sample filtration, wash with 10–20 ml of sterile H2O; remove H2O by filtration, put the membrane filter in a plastic petri dish, cover and allow air-drying. D. Store at  20 1C until processing; filters can be stored frozen for several months without apparent loss of hybridization signal. 4. Fixation of sediment/soil samples (Llobet-Brossa et al., 1998). A. Fix sediment samples with fresh formaldehyde solution (end concentration 1–4%) for 1–2 h at RT or max 24 h at 4 1C. B. Centrifuge at 16 000 g for 5 min, pour off supernatant and resuspend sample with 1  PBS pH 7.6, repeat washing step twice. C. Store sediment sample in a 1:1 mix of PBS/ethanol at  20 1C until further processing.

glass slides (Manz et al., 1992). For this, cells of pure cultures, enrichments or concentrated cell suspensions from environmental samples are spotted onto gelatin-coated slides, allowed to air-dry, and fixed by immersion in formalin or ethanol. These slides are then covered with hybridization buffer and probe, and incubated in a moisture chamber for several hours. After a short washing step, the cells can be embedded in antifading reagent for microscopic visualization. This hybridization technique is robust, yet not quantitative since cell loss from the slide during hybridization and washing cannot be completely ruled out. Nevertheless, this technique can be used in wastewater and food analytics because high numbers of bacteria are present and slides can be used as colonization devices. In water science, immobilization of cells on glass slides is frequently used for the analysis of wastewater treatment or activated sludge plants (Wagner et al., 1993; Daims et al., 2001). In limnic and marine systems, low cell numbers require cell concentration and the usage of polycarbonate membrane filters for analysis on microbial communities. Therefore, fixed cells are concentrated and immobilized on membrane filters of an adequate pore size (most often 0.2-mm pore-sized polycarbonate filters). This technique is more suitable for environmental samples with low cell concentrations, for example, marine or freshwater. Care must be taken such that the filtered volume is adjusted in a way that facilitates subsequent cell counting in the microscope. Filters are then cut into pieces (eight for a 25-mm-diameter filter, 16– 20 for a 47-mm-diameter filter), and these filter pieces are then used for hybridization with different probes (Table 3, Box 6). Again, a moisture chamber is used to avoid evaporation of the hybridization buffer and to ensure a high stringency of the hybridization. Hybridization is performed using a buffer/

probe mix (10:1) to cover the filter pieces. Incubation times should be in the range of 90 min to 3 h at a temperature of 46 1C. After several washing steps, filter pieces are stained with the DNA stain 40 ,6-diamidino-2-phenylindole (DAPI), and embedded in antifading reagent for microscopic analyses (Glo¨ckner et al., 1996).

3.08.3.3 Catalyzed Reporter Deposition Fluorescence In Situ Hybridization In environmental samples, single oligonucleotides carrying only one fluorochrome may not provide enough fluorescence signals to detect cells with low ribosome contents (Pernthaler et al., 2002). Polynucleotide probes with a length of more than 100 nucleotides labeled with several fluorochromes per molecule are an alternative (Trebesius et al., 1994; DeLong et al., 1999). However, these probes lack the specificity for narrow target groups such as species or genera. An alternative labeling technique that increases the fluorescence signal intensity uses HRP-labeled oligonucleotides. When using HRP-labeled probes, fluorescent staining results from a secondary incubation with fluorescently labeled tyramide. Each HRP-labeled probe catalyzes the covalent deposition of multiple labeled tyramides, resulting in a staining that is significantly brighter than those obtained with oligonucleotides labeled with a single fluorochrome (Scho¨nhuber et al., 1997; Hoshino et al., 2008). However, cell permeabilization protocols need to be adjusted in order to enable the larger enzyme-labeled oligonucleotides to diffuse into the cells (Pernthaler et al., 2002). Depending on the sample, the protocols for CARD FISH differ substantially, yet, there are generally two additional steps compared to FISH: (1) embedding in agarose and

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

(2) incubation with fluorochrome-labeled tyramide (Pernthaler et al., 2004).

3.08.3.3.1 Embedding, permeabilization, and inactivation of endogenous peroxidases Similar to FISH with monolabeled oligonucleotide probes, samples for CARD FISH are fixed with formaldehyde and filtered on polycarbonate filters. These filters are then embedded in low gelling point agarose (0.1%) and dried at a temperature between 20 and 50 1C. The embedded cells are stabilized by the agarose, and cell loss during permeabilization is prevented. In order to permeabilize the cells for the large HRP enzyme (B40 kDa), again no single standard protocol is available. However, permeabilization with lysozyme turned Table 4 NaCl concentration in the washing buffer according to % formamide of the hybridization buffer % formamide in hybridization buffer 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70

[NaCl] in M final concentration 0.900 0.636 0.450 0.318 0.225 0.159 0.112 0.080 0.056 0.040 0.028 0.020 0.014

ml 5 M NaCl in 50 ml

9000 6300 4500 3180 2150 1490 1020 700 460 300 180 100 40

183

out to be a very good method for most marine planktonic Bacteria (see Box 7 for a basic protocol). Archaea or cells with protective substances, such as exopolymers or waxes, could be permeabilized with substances described above (Section 3.08.4.1) or they require the development of new permeabilization protocols. Due to the use of HRP as the label for the oligonucleotide probes, naturally occurring peroxidases must be inactivated. Several bacteria produce, for example, peroxidases as a protection mechanism against peroxide which are formed from reactive oxygen species in aerobic environments (Farr and Kogoma, 1991). Such peroxidases could lead to false-positive signals in cell counts and thus must be inactivated before the CARD of tyramide, for example, by a short incubation in diluted hydrochloric acid. Furthermore, autofluorescence of certain cells could also lead to false-positive counts. An incubation of the cells with peroxide solution (3%) reduces autofluorescence signals to background levels.

3.08.3.3.2 Hybridization Several minor changes from the FISH protocol were made for the hybridization of cells with HRP-labeled probes (Box 8). Similar to monolabeled probes, moisture chambers are used to ensure stringency of the hybridization. However, the concentration of the hybridization buffer/probe mix can be chosen an order of magnitude lower (300:1 to 100:1, depending on the sample). Furthermore, a prolonged hybridization time of 2–8 h at 46 1C is recommended. Subsequent washing steps are crucial; filters must not run dry before the CARD procedure.

3.08.3.3.3 Catalyzed reporter deposition CARD of tyramide by HRP is known since two decades as a method of signal amplification (Bobrow et al., 1989). First used in immunoblotting and immunosorbent assays, the

Box 6 FISH using monolabeled oligonucleotide probes on membrane filters 1. Cut sections from membrane filters with a razor blade and label filter sections with a pencil, for example, by numbering them. 2. Put filter sections on glass slides (cells facing up!), several filter sections can be placed on one slide and for simultaneous hybridization with the same probe. 3. Prepare 2 ml of hybridization buffer in a microfuge tube (360 ml 5 M NaCl, 40 ml 1 M Tris/HCl, formamide % depending on probe, add water to 2 ml, add 2 ml SDS (10%)). 4. Remove an aliquot of 20 ml per filter piece into a separate cap and add 2 ml probe working solution (50 ng probe ml1) per filter piece. 5. Prepare moisture chamber by putting a piece of blotting paper into a 50-ml polyethylene tube and soaking it with the remaining hybridization buffer without probe (see above). 6. Carefully cover the filter section with the hybridization mix and place the slide with filter sections into the polyethylene tube (in a horizontal position). 7. Incubate at 46 1C for at least 90 min (maximum: 3 h). 8. Meanwhile prepare 50 ml of washing buffer in a polyethylene tube (X ml 5 M NaCl, depending on formamide concentration in the hybridization buffer (see Table 4)), 1 ml 1 M Tris/HCl, 500 ml 0.5 M EDTA (if formamide concentration of the hybridization buffer is higher than 20%), add to 50 ml with water and add 50 ml SDS (10%). 9. Quickly transfer filter sections into preheated washing buffer and incubate for 15 min at 48 1C (water bath). 10. Pour washing buffer with filter sections into a petri dish. Pick filter sections and rinse them by placing them into a petri dish with distilled H2O for several seconds, then let them air-dry on blotting paper. 11. For counterstaining, put filter sections on a glass plate, cover with c.50 ml of DAPI solution (1 mg ml1), and incubate for 3 min. Afterwards, wash filter sections subsequently for 1 min in distilled H2O and for 1 min in 80% ethanol to remove unspecific staining. Let air-dry. 12. Samples are mounted in a 4:1 mix of Citifluor and Vecta Shield. The filter sections have to be completely dry before embedding, otherwise part of the cells might detach during inspection. 13. Double-stained and air-dried preparations as well as filters mounted on slides can be stored in the dark at  20 1C for several days without substantial loss of probe fluorescence.

184

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

Box 7

Embedding, permeabilization and inactivation of endogenous peroxidases for CARD FISH

Remark: Do not embed cells if flow cytometry should be used for analyses, and handle filters very carefully during the washing steps to prevent cell loss. Embedding 1. Boil low gelling point agarose (0.1%, gel strength should be approximately 1000 g cm2), fill the agarose in a pre-warmed petri dish and let it cool down to 35–40 1C. 2. Dip filter with both sides in the agarose and place it face down (shiny side with bacteria down!) onto a parafilm covered, even surface (e.g., glass plate), let dry; temperature for drying is not crucial, anything between 20 and 50 1C is fine. 3. Remove filters from surface by soaking in 80–96% ethanol. 4. Let the filter air-dry on a piece of tissue paper. Permeabilization Remark: Permeabilization with lysozyme proved to be the optimal method for most of the marine planktonic bacteria. 1. Incubate filter in 10–20 ml of fresh lysozyme solution in a small petri dish (10 mg ml1 in 0.05 M EDTA, pH 8.0; 0.1 M Tris–HCl, pH 8.0) for 60 min at 37 1C. 2. Wash in excess with MilliQ water. Inactivation of endogenous peroxidases 1. Incubate in 0.01 M HCl or waterous H2O2 solution (3%) for 10–20 min at RT. Wash filters well in excess MilliQ water and 96% ethanol and let the filter air-dry on a paper. 2. If you have problems with autofluorescence of your cells, an additional incubation in 3% H2O2 in water for 10 min at RT may help.

Box 8

Hybridization with HRP-labeled oligonucleotide probes

1. Prepare a humidity chamber by inserting a piece of tissue paper in a 50-ml tube and soak it with a formamide-water mix according to the formamide concentration of the hybridization buffer. Mix hybridization buffer (3.6 ml 5 M NaCl, 0.4 ml 1 M Tris–HCl pH 8.0, formamide (depending on probe, see Table 2),  ml sterile dH2O (depending on probe), 2.0 ml Blocking Reagent (10%, Roche, Basel; prepare according to manufacturer’s instructions), 2.0 g of dextran sulfate) and 20 ml SDS (20%)) with probe working solution (50 ng DNA ml1) in a ratio 300:1 (i.e., 1 ml) to 100:1 (i.e., 3 ml) depending on sample. For every filter piece 100 ml of hybridization mix should be calculated. 2. Dip each filter completely into the hybridization solution and place filters face up onto a parafilm covered glass slide; spread the rest of the solution evenly onto the filters. Close tube firmly and keep the tube in a horizontal position. 3. Incubate at 46 1C for 2–3 h (coastal water) or 6–8 h for oligotrophic/open ocean water samples. 4. Wash filters in prewarmed washing buffer (0.5 ml 0.5 M EDTA pH 8.0, 1.0 ml 1 M Tris–HCl pH 8.0,  ml NaCl (depending on probe, see Table 4 in standard FISH protocol) add dH2O to a final volume of 50 ml, then add 25 ml SDS (20%)) for 10 min at 48 1C. 5. Transfer filters to 1  PBS (do not let filter run dry!) and incubate for 15 min at RT. 6. To remove excess liquid, dab filter on blotting paper, but do not let filter run dry!

combination with oligonucleotide probes made this method suitable as an approach for microbial ecology (Box 9; Scho¨nhuber et al., 1997; Pernthaler et al., 2002). Thus, HRPlabeled probes can be used to enhance FISH signals by the use of peroxide as a catalyst for tyramide oxidation. H2O2 is the activating substrate of peroxidase. The peroxidase enters a radical form and transfers the radical to the fluorescently labeled tyramide. In this reaction, a proton is set free. The activated HRP radicalizes in a second step another fluorescently labeled tyramide molecule and is then reaching a resting stage, needing H2O2 for further activation (Veitch, 2004). The fluorescently labeled tyramide radicals bind covalently to electron-rich moieties within the cell, such as tyrosine sites in proteins close to the reaction site (Figure 5; Pernthaler, et al., 2002). These reactions are again carried out in a humidity chamber. Therefore, H2O2 (0.15% in phosphate-buffered saline (PBS) buffer) is mixed with the amplification buffer and fluorescently labeled tyramide (1 mg ml1) is added to the mix (see Box 9 for protocol). The filter pieces are then incubated at 46 1C for up to 45 min and washed thoroughly afterwards.

After DAPI counterstaining and embedding into antifading agent, the filter pieces can be analyzed via epifluorescence microscopy. For flow cytometric analyses, cells have to be removed from the filters again by mechanical and chemical treatments before counterstaining with DAPI (Box 10; Sekar et al., 2004). Subsequently, the resuspended cells can be counterstained and analyzed via flow cytometry (Sekar et al., 2004).

3.08.3.4 Troubleshooting Even though FISH and CARD FISH are rather robust methods, they are prone to errors during handling. Some of them are described in the following; however, in case of failure, we also recommend to consult the extensive check list provided by Wagner et al. (2003). First of all, some of the chemicals and buffers, such as formamide, dextran sulfide, and the hybridization buffer mix, are delicate and tend to decay over time. These reagents need to be refilled from the stock or prepared freshly in short time

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

185

Box 9 Catalyzed reporter deposition 1. Prepare a moisture chamber by inserting a piece of tissue paper in a 50-ml tube and soak it with 2 ml water. 2. Prepare a fresh solution of H2O2 (0.15% in PBS), keep it cool. 3. Mix amplification buffer with H2O2 solution in a ratio of 100:1 (as a guideline, the same volume as the hybridization mix is sufficient) and add fluorescently labeled tyramide (1 mg ml1) and mix well, keep in the dark. (The volume of labeled tyramide added strongly depends on the nature of the sample, start with 1000:1; if the signal is not sufficient: in-/decrease the ratio of added tyramide.) 4. Dip filter completely in the amplification mix, place filter sections face up on a parafilm covered glass slide and spread the rest of the amplification mix over the filters. 5. Put the slides into the humidity chamber and incubate at 46 1C for up to 45 min in the dark. 6. To remove excess liquid, dab filter on blotting paper and incubate in 1  PBS for 5–10 min at RT in the dark (or: 15 min at 46 1C on a shaker). 7. Wash filters thoroughly in excess with deionized water. Therefore, use a Bu¨chner funnel and MQ water to create a gyre in the funnel. Then wash filters thoroughly twice in excess in 96% ethanol (1–2 min), let them completely air-dry in the dark before counterstaining with DAPI. 8. For counterstaining, put filter sections on a glass plate, cover with 50 ml of DAPI solution, and incubate for 3 min; afterwards wash filter sections for several seconds in 80% ethanol to remove unspecific staining followed by rinsing in distilled H2O and air-drying. 9. Samples are mounted in a 4:1 mix of Citifluor and Vecta Shield; the filter sections have to be completely dry before embedding, otherwise part of the cells might detach during inspection. 10. Double-stained and air-dried preparations as well as filters mounted on slides can be stored in the dark at  20 1C for several days without substantial loss of probe fluorescence.

HRP HRP Hybridization with horseradish-peroxidase (HRP)-labeled oligonucleotide probes

HRP HRP

+ H2O2 Catalyzed reporter deposition (CARD) of fluorescently labeled tyramide

HRP

Protein + 2 H2O Figure 5 CARD FISH with HRP-labeled probes, using fluorescently labeled tyramines.

intervals. HRP-labeled probes should not be repeatedly frozen, because freezing might break the enzyme–oligonucleotide binding and decrease the activity of the enzyme, resulting in bad or no signals. Another frequent problem could be the lack of signals. This might have several reasons. The most trivial one is the complete loss of cells from the slide or filter during handling. It might also well be that during sampling too little cells have been transferred onto the filter or glass slide to be detected in the microscopic viewing field (no cells, no signals; Amann et al., 1995). Alternatively, the staining procedure by FISH resulted in low signal intensities due to low ribosome contents in the target cells, in which case, CARD FISH might enhance

the signals. If CARD FISH fails, most likely the cells are impermeable to the large HRP probes and need to be permeabilized more thoroughly by treatment with, for example, proteinases and lysozyme. Some newly developed probes could fail in FISH, although a positive control hybridization with an established probe such as EUB338 produces nice signals. In such a case, the target cells have been permeabilized and contain sufficient ribosomes for detection, and the most likely cause of failure is then inaccessibility of the probe target sites in the ribosome. The use of helper oligonucleotides here might increase the accessibility of the probe-binding site, resulting in a significant increase of the hybridization signals. Yilmaz and co-workers could also show that the low

186

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

Box 10 Preparation of the CARD FISH stained cells for flow cytometry. From Sekar R, Fuchs B, Amann R, and Pernthaler J (2004) Flow sorting of marine bacteria after fluorescence in situ hybridization. Applied and Environmental Microbiology 70: 6210–6219 Remark: No standard protocol is available, different methods for cell removal should be combined for different samples. 1. Mechanical cell removal A. Vortexing for 15 min at 2500 rpm. B. Sonication at maximum power for 30 min in a water bath. 2. Chemical removal. Try the following chemicals for up to 2 h at 37 1C with subsequent 10 min vortexing, depending on samples. A. 1 M Tris–HCl B. PBS (1  ) with 0.05% Triton X-100 C. 150 mM NaCl with 0.05% Tween 80 D. 0.05 M pyrophosphate E. 1 M KCl F. 0.1% Pluronic F-80 G. 0.1% sodium cholate 3. Counterstain cells with DAPI (1 mg ml1). 4. Analyze cells in a flow cytometer, for example, MoFlo flow cytometer (BeckmanCoulter) or FACScalibur (BD biosciences).

A good method proved to be the incubation with NaCl–Tween 80 for 30 min at 37 1C.

efficiencies of FISH are often a kinetic problem, and may be overcome by the alteration of the probe sequence, for example, by designing a longer oligonucleotide probe (Yilmaz and Noguera, 2004; Yilmaz et al., 2006).

3.08.4 Cell Counting One of the aims of FISH-based analyses of bacterial communities is exact quantification. The common method to achieve quantitative cell counts is either flow cytometry or manual cell counting using epifluorescence microscopes. Flow cytometry has successfully been applied to sort environmental samples and to determine the composition of the microbial communities (Sekar et al., 2004). However, this approach requires a relatively large amount of sample, and quantitative measurements are biased toward large cells with high ribosome content due to a lack of sensitivity. Generally, cells with high ribosome contents are preferred for sorting. Therefore, manual counting using epifluorescence microscopes is recommended, even though this method is time consuming and can make up a major part of the experimental time. The quantification of taxa is most often based on determining the ratio of total DAPI-stained cells to probe-stained cells. To reduce counting errors, at least 1000 DAPI-stained cells should be inspected for probe signal. Due to the low sample throughput and subjectivity of manual counting, a semi-automated counting method (Cottrell et al., 2006) and an automatic counting system for epifluorescence microscopes were developed for FISH-based quantification of microbial communities (Pernthaler et al., 2003). Additionally, a semi-automated technique for structural investigations was invented by Daims et al. 2001. Based on image analysis, the technique of Pernthaler allows a high sample throughput and thus increases efficiency of the counting (see also Schattenhofer et al., 2009). Until now, these techniques can only be applied to planktonic samples, since

more complex samples, for example, sediment samples, exceed the differentiation capability of the machine. Nevertheless, several improvements in automatic counting have recently been made by Zeder and Pernthaler (2009). Nested focusing in bright field and fluorescence illumination, continuous life-image acquisition during focusing, multiple spot focus measurements to assess quality and topology of the filter section, and the usage of z-stacks to compensate for unevenness of the filter surface are used to perform a more reliable auto-focusing process (Zeder and Pernthaler, 2009). Furthermore, an artificial neural network was implemented into the focus routine as a quality check.

3.08.5 From Cell Detection to Ecological Function From the numerous applications since the introduction of FISH into the field of microbial ecology 20 years ago (DeLong et al., 1989; Amann et al., 1990b), only some examples are mentioned in the following. One recent application of fluorescence-labeled oligonucleotides was the quantification of the probably most abundant bacterial and archaeal taxa on Earth. The alphaproteobacterium SAR11 constitutes up to 50% of the bacterial community of the open-ocean surface waters, while marine group I crenarchaeota makes up to 40% of deep ocean waters (Karner et al., 2001; Morris et al., 2002; Schattenhofer et al., 2009). In addition, taxa with lower frequencies of about 1 in 1000, but of high ecological importance, were discovered using FISH, for example, bacteria catalyzing the anaerobic ammonium oxidization (anammox) in the Black Sea (Kuypers et al., 2003). Furthermore, not only natural environments were investigated via FISH, but also anthropogenic environments such as wastewater treatment plants, where biofilm cells of the genus Nitrospira could be subdivided in two sublineages (Maixner et al., 2006). Drinking water, probably one of the biggest future concerns of humankind, can be monitored with

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

FISH methods, for example, Enterobacteriaceae were detected with direct viable counts combined with FISH and laser scanning cytometry (Baudart et al., 2005), Bacteroides of human feces could be detected in drinking water experiments (Savichtcheva et al., 2005) and Helicobacter pylori was detected in drinking water biofilms (Braganca et al., 2007). Furthermore, FISH was applied to various monitoring approaches in wastewater treatment, for example, in combination with MAR (Hagman et al., 2008), as a control in the development of new wastewater reactors (Lalbahadur et al., 2005), following earlier application of FISH in wastewater treatment plants (Manz et al., 1994). These examples show that identification of individual cells by FISH can be applied not only for basic research on bacterial community, but also on water quality monitoring, wastewater treatment, and a variety of more water-related issues. By the combination of FISH with other methods, different aspects of microbial activity can be analyzed for single cells in a spatially and phylogenetically resolved manner. Active DNA synthesis could be assigned to particular bacterial cells by the combination of CARD FISH with the immunodetection of bromodeoxyuridine-labeled nucleotides (Pernthaler et al., 2002). The amount of rRNA synthesis can be estimated from FISH-based measurements of the ratio of precursor to mature rRNA (Oerther et al., 2000). The cellspecific uptake of stable isotope or radioactively labeled substrates can be determined by, for example, MAR FISH (Lee et al., 1999), Raman FISH (Huang et al., 2007), or a combination of nano-secondary ion mass spectrometry and in situ hybridization (Behrens et al., 2008; Li et al., 2008; Musat et al., 2008). Finally, FISH is now used for in situ identification of bacteria in dental plaque (Gmu¨r and Lu¨thi-Schaller, 2007) thereby coming back to the habitat in which bacteria were discovered by Leuwenhook. All these methods allow further insight into ecological processes and the important functions of bacterial or archaeal taxa in the environment. New FISH protocols could in future provide even better possibilities for researchers in different fields of microbiology to gain information about bacterial identity and activity. In a time of massive environmental metagenome sequencing and ever-increasing sequence databases, there is an urgent need for further probe development and the continuous optimization of microscopic techniques, including high-throughput automatic counting methods.

References Abdo Z, Schuette UME, Bent SJ, Williams CJ, Forney LJ, and Joyce P (2006) Statistical methods for characterizing diversity of microbial communities by analysis of terminal restriction fragment length polymorphisms of 16S rRNA genes. Environmental Microbiology 8: 929--938. Altschul SF, Gish W, Miller W, Myers EW, and Lipman DJ (1990) Basic local alignment search tool. Journal of Molecular Biology 215: 403--410. Amann RI, Binder BJ, Olson RJ, Chisholm SW, Devereux R, and Stahl DA (1990a) Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Applied and Environmental Microbiology 56: 1919--1925. Amann RI and Fuchs BM (2008) Single-cell identification in microbial communities by improved fluorescence in situ hybridization techniques. Nature Reviews Microbiology 6: 339--348.

187

Amann RI, Krumholz L, and Stahl DA (1990b) Fluorescent-oligonucleotide probing of whole cells for determinative, phylogenetic, and environmental studies in microbiology. Journal of Bacteriology 172: 762--770. Amann RI, Ludwig W, and Schleifer KH (1995) Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiological Reviews 59: 143--169. Baudart J, Olaizola A, Coallier J, Gauthier V, and Laurent P (2005) Assessment of a new technique combining a viability test, whole-cell hybridization and laserscanning cytometry for the direct counting of viable Enterobacteriaceae cells in drinking water. FEMS Microbiology Letters 243: 405--409. Behrens S, Lo¨sekann T, Pett-Ridge J, et al. (2008) Linking microbial phylogeny to metabolic activity at the single-cell level by using enhanced element labelingcatalyzed reporter deposition fluorescence in situ hybridization (EL-FISH) and NanoSIMS. Applied and Environmental Microbiology 74: 3143--3150. Behrens S, Ru¨hland C, Inacio J, et al. (2003) In situ accessibility of small-subunit rRNA of members of the domains Bacteria, Archaea and Eucarya to Cy3-labeled oligonucleotide probes. Applied and Environmental Microbiology 69: 1748--1758. Bobrow MN, Harris TD, Shaughnessy KJ, and Litt GJ (1989) Catalyzed reporter deposition, a novel method of signal amplification: Application to immunoassays. Journal of Immunological Methods 125: 279--285. Braganca SM, Azefedo NF, Simoes LC, Keevil CW, and Vieira MJ (2007) Use of fluorescent in situ hybridization for the visualization of Helicobacter pylori in real drinking water biofilms. Water Science and Technology 55: 387--393. Buchan A, Gonzalez JM, and Moran MA (2005) Overview of the marine Roseobacter lineage. Applied and Environmental Microbiology 71: 5665--5677. Burggraf S, Mayer T, Amann R, Schadhauser S, Woese CR, and Stetter KO (1994) Identifying members of the domain Archaea with ribosomal-RNA-targeted oligonucleotide probes. Applied and Environmental Microbiology 60: 3112--3119. Cohn F (1872) Grundzu¨ge einer neuen natu¨rlichen Anordnung der kryptogamischen Pflanzen. Jahresbericht der Schlesischen Gesellschaft fu¨r Vaterla¨ndische Kultur 49: 83--89. Cole JR, Wang Q, Cardenas E, et al. (2009) The Ribosomal Database Project: Improved alignments and new tools for rRNA analysis. Nucleic Acids Research 37: D141--D145. Cottrell MT, Mannino A, and Kirchman DL (2006) Aerobic anoxygenic phototrophic bacteria in the Mid-Atlantic Bight and the North Pacific Gyre. Applied and Environmental Microbiology 72: 557--564. Daims H, Bru¨hl A, Amann R, Schleifer KH, and Wagner M (1999) The domain-specific probe EUB338 is insufficient for the detection of all bacteria: Development and evaluation of a more comprehensive probe set. Systematic and Applied Microbiology 22: 434--444. Daims H, Ramsing NB, Schleifer KH, and Wagner W (2001) Cultivation-independent, semiautomatic determination of absolute bacterial cell numbers in environmental samples by fluorescence in situ hybridization. Applied and Environmental Microbiology 67: 5810--5818. Davenport RJ, Curtis TP, Goodfellow M, Stainsby FM, and Bingley M (2000) Quantitative use of fluorescent in situ hybridization to examine relationships between mycolic acid-containing actinomycetes and foaming in activated sludge plants. Applied and Environmental Microbiology 66: 1158--1166. DeLong EF (1992) Archaea in coastal marine environments. Proceedings of the National Academy of Sciences of the United States of America 89: 5685--5689. DeLong EF, Taylor LT, Marsh TL, and Preston CM (1999) Visualization and enumeration of marine planktonic archaea and bacteria by using polyribonucleotide probes and fluorescent in situ hybridization. Applied and Environmental Microbiology 65: 5554--5563. DeLong EF, Wickham GS, and Pace NR (1989) Phylogenetic stains: Ribosomal RNAbased probes for the identification of single cells. Science 243: 1360--1363. Domann E, Hong G, Imirzalioglu C, et al. (2003) Culture-independent identification of pathogenic bacteria and polymicrobial infections in the genitourinary tract of renal transplant recipients. Journal of Clinical Microbiology 41: 5500--5510. Doungudomdacha S, Rawlinson A, and Douglas CW (2000) Enumeration of Porphyromonas gingivalis, Prevotella intermedia and Actinobacillus actinomycetemcomitans in subgingival plaque samples by quantitative-competitive PCR. Journal of Medical Microbiology 49: 861--874. Eilers H, Pernthaler J, Peplies J, Gloeckner FO, Gerdts G, and Amann R (2001) Isolation of novel pelagic bacteria from the German Bight and their seasonal contributions to surface picoplankton. Applied and Environmental Microbiology 67: 5134--5142. Farr SB and Kogoma T (1991) Oxidative stress response in Escherichia coli and Salmonella typhimurium. Microbiological Reviews 55: 561--585. Fuchs BM, Glo¨ckner FO, Wulf J, and Amann RI (2000) Unlabeled helper oligonucleotides increase the in situ accessability to 16S rRNA of fluorescently

188

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

labeled oligonucleotide probes. Applied and Environmental Microbiology 66: 3603--3607. Fuchs BM, Syutsubo K, Ludwig W, and Amann RI (2001) In situ accessability of Escherichia coli 23S rRNA to fluorescently labeled oligonucleotide probes. Applied and Environmental Microbiology 67: 961--968. Fuchs BM, Wallner G, Beisker W, Schwippl I, Ludwig W, and Amann RI (1998) Flow cytometric analysis of the in situ accessibility of Escherichia coli 16S rRNA for fluorescently labeled olinucleotide probes. Applied and Environmental Microbiology 64: 4973--4982. Garrity G, Boone DR, and Castenholz RW (eds.) (2001) Bergey’s Manual of Systematic Bacteriology, 2nd edn., vol. 1. New York: Springer. Giovannoni SJ and Stingl U (2005) Molecular diversity and ecology of microbial plankton. Nature 437: 343--348. Glo¨ckner FO, Amann R, Alfreider A, et al. (1996) An in situ hybridization protocol for detection and identification of planktonic bacteria. Systematic and Applied Microbiology 19: 403--406. Glo¨ckner FO, Fuchs BM, and Amann RI (1999) Bacterioplankton composition of lakes and oceans: A first comparison based on fluorescence in situ hybridization. Applied and Environmental Microbiology 65: 3721--3726. Gmu¨r R and Lu¨thi-Schaller H (2007) A combined immunofluorescence and fluorescent in situ hybridization assay for single cell analyses of dental plaque microorganisms. Journal of Microbiological Methods 69: 402–405. Hagman M, Nielsen JL, Nielsen PH, and Jansen JL (2008) Mixed carbon sources for nitrate reduction in activated sludge-identification of bacteria and process activity studies. Water Research 42: 1539--1546. Hoshino T, Yilmaz LS, Noguera DR, Daims H, and Wagner M (2008) Quantification of target molecules needed to by fluorescence in situ hybridization (FISH) and catalyzed reporter deposition-FISH. Applied and Environmental Microbiology 74: 5068--5077. Huang WE, Stoecker K, Griffiths R, et al. (2007) Raman-FISH: Combining stable isotope Raman spectroscopy and fluorescence in situ hybridization for the single cell analysis of identity and function. Environmental Microbiology 9: 1878--1889. Jannasch HW and Jones GE (1959) Bacterial populations in sea water as determined by different methods of enumeration. Limnology and Oceanography 4: 128--139. Jeanthon C (2000) Molecular ecology of hydrothermal vent microbial communities. Antonie van Leeuwenhoek International Journal of General and Molecular Microbiology 77: 117--133. Karner MB, DeLong EF, and Karl DM (2001) Archaeal dominance in the mesopelagic zone of the Pacific Ocean. Nature 409: 507--510. Kuypers MMM, Sliekers AO, Lavik G, et al. (2003) Anaerobic ammonium oxidation by anammox bacteria in the Black Sea. Nature 422: 608--611. Lalbahadur T, Pillay S, Rodda N, et al. (2005) Microbiological studies of an anaerobic baffled reactor: Microbial community characterisation and deactivation of healthrelated indicator bacteria. Water Science and Technology 51: 155--162. Lane DJ (1991) 16S/23S rRNA sequencing. In: Stackebrandt E and Goodfellow M (eds.) Nucleic Acid Techniques in Bacterial Systematics, pp. 115--175. Chichester: Wiley. Lee N, Nielsen PH, Andreasen KH, et al. (1999) Combination of fluorescent in situ hybridization and microautoradiography – a new tool for structure–function analyses in microbial ecology. Applied and Environmental Microbiology 65: 1289--1297. Li T, Wu TD, Mazeas L, et al. (2008) Simultaneous analysis of microbial identity and function using NanoSIMS. Environmental Microbiology 10: 580--588. Liu WT, Marsh TL, Cheng H, and Forney LJ (1997) Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16S rRNA. Applied and Environmental Microbiology 63: 4516--4552. Llobet-Brossa E, Rossello´-Mora R, and Amann R (1998) Microbial community composition of Wadden Sea sediments as revealed by fluorescence in situ hybridization. Applied and Environmental Microbiology 64: 2691--2696. Loy A, Arnold R, Tischler P, Rattei T, Wagner M, and Horn M (2008) ProbeCheck – a central resource for evaluating oligonucleotide probe coverage and specificity. Environmental Microbiology 10: 2894--2896. Loy A, Horn M, and Wagner M (2003) ProbeBase: An online resource for rRNAtargeted oligonucleotide probes. Nucleic Acids Research 31: 514--516. Ludwig W, Strunk O, Westram R, et al. (2004) ARB: A software environment for sequence data. Nucleic Acids Research 32: 1363--1371. Maixner F, Noguera DR, Anneser B, et al. (2006) Nitrite concentration influences the population structure of Nitrospira-like bacteria. Environmental Microbiology 8: 1487--1495. Manz W, Amann R, Ludwig W, Vancanneyt M, and Schleifer KH (1996) Application of a suite of 16S rRNA-specific oligonucleotide probes designed to investigate bacteria

of the phylum Cytophaga–Flavobacter–Bacteroides in the natural environment. Microbiology 142: 1097--1106. Manz W, Amann R, Ludwig W, Wagner M, and Schleifer KH (1992) Phylogenetic oligodeoxynucleotide probes for the major subclasses of Proteobacteria: Problems and solutions. Systematic and Applied Microbiology 15: 593--600. Manz W, Wagner M, Amann R, and Schleifer KH (1994) In-situ characterization of the microbial consortia active in 2 waste-water treatment plants. Water Research 28: 1715--1723. Marsh CL and Larsen DH (1953) Characterization of some thermophilic bacteria from the hot springs of Yellowstone-National Park. Journal of Bacteriology 65: 193--197. Morris RM, Rappe MS, Connon SA, et al. (2002) SAR11 clade dominates ocean surface bacterioplankton communities. Nature 420: 806--810. Musat N, Halm H, Winterholler B, et al. (2008) A single-cell view on the ecophysiology of anaerobic phototrophic bacteria. Proceedings of the National Academy of Sciences of the United States of America 105: 17861--17866. Muyzer G, de Waal EC, and Uitterlinden AG (1993) Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Applied and Environmental Microbiology 59: 695--700. Muyzer G, Teske A, Wirsen CO, and Jannasch HW (1995) Phylogenetic relationships of Thiomicrospira species and their identification in deep-sea hydrothermal vent samples by denaturing gradient gel electrophoresis of 16S rDNA fragments. Archives of Microbiology 164: 165--172. Myers RM, Maniatis T, and Lerman LS (1987) Detection and localization of single base pair changes by denaturing gradient gel electrophoresis. Methods in Enzymology 155: 501--527. Neef A (1997) Application of In Situ Identification of Bacteria to Population Analysis in Complex Microbial Communities. PhD Thesis, Technical University of Munich, Munich. Neef A, Amann R, Schlesner H, and Schleifer KH (1998) Monitoring a widespread bacterial group: In situ detection of planctomycetes with 16S rRNA-targeted probes. Microbiology 144: 3257--3266. Neef A, Zaglauer A, Meier H, Amann RI, Lemmer H, and Schleifer KH (1996) Population analysis in a denitrifying sand filter: Conventional and in situ identification of Paracoccus spp. in methanol-fed biofilms. Applied and Environmental Microbiology 62: 4329--4339. Oerther DB, Pernthaler J, Schramm A, Amann RI, and Raskin L (2000) Monitoring precursor 16S rRNAs of Acinetobacter spp. in activated sludge wastewater treatment systems. Applied and Environmental Microbiology 66: 2154--2165. Olsen GJ, Lane DJ, Giovannoni SJ, Pace NR, and Stahl DA (1986) Microbial ecology and evolution: A ribosomal RNA approach. Annual Reviews of Microbiology 40: 337--365. Oren A (1999) Bioenergetic aspects of halophilism. Microbiology and Molecular Biology Reviews 63: 334--348. Pace NR (2009) Mapping the tree of life: Progress and prospects. Microbiology and Molecular Biology Reviews 73: 565--576. Partensky F, Hess WR, and Vaulot D (1999) Prochlorococcus, a marine photosynthetic prokaryote of global significance. Microbiology and Molecular Biology Reviews 63: 106--127. Peplies J, Kottmann R, Ludwig W, and Glo¨ckner FO (2008) A standard operating procedure for phylogenetic inference (SOPPI) using (rRNA) marker genes. Systematic and Applied Microbiology 31: 251--257. Pernthaler A and Amann R (2004) Simultaneous fluorescence in situ hybridization of mRNA and rRNA in environmental bacteria. Applied and Environmental Microbiology 70: 5426--5433. Pernthaler A, Pernthaler J, and Amann RI (2002) Fluorescence in situ hybridization and catalyzed reporter deposition for the identification of marine bacteria. Applied and Environmental Microbiology 68: 3094--3101. Pernthaler A, Pernthaler J, and Amann RI (2004) Sensitive multicolor fluorescence in situ hybridization for the identification of environmental microorganisms. In: Hilario E and MacKay JF (eds.) Protocols for Nucleic Acid Analysis by Nonradioactive Probes, vol. 353, pp. 153--164. Totowa, NJ: Humana Press. Pernthaler J, Glo¨ckner FO, Unterholzner S, Alfreider A, Psenner R, and Amann R (1998) Seasonal community and population dynamics of pelagic bacteria and archaea in a high mountain lake. Applied and Environmental Microbiology 64: 4299--4306. Pernthaler J, Pernthaler A, and Amann RI (2003) Automated enumeration of groups of marine picoplankton after fluorescence in situ hybridizations. Applied and Environmental Microbiology 69: 2631--2637. Pruesse E, Quast C, Knittel K, et al. (2007) SILVA: A comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Research 35: 7188--7196.

Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization

Roller C, Wagner M, Amann R, Ludwig W, and Schleifer KH (1994) In situ probing of Gram-positive bacteria with high DNA G þ C content using 23S rRNA-targeted oligonucleotides. Microbiology 140: 2849--2858. Ronaghi M, Karamohamed S, Petterson B, Uhlen M, and Nyren P (1996) Real-time DNA sequencing using detection of pyrophosphate release. Analytical Biochemistry 242: 84--89. Sambrook J, Fritsch EF, and Maniatis T (1989) Molecular Cloning: A Laboratory Manual, 2nd edn., vol. I. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Savichtcheva O, Okayama N, Ito T, and Okabe S (2005) Application of a direct fluorescence-based live/dead staining combined with fluorescence in situ hybridization for assessment of survival rate of Bacteroides spp. in drinking water. Biotechnology and Bioengineering 92: 356--363. Schattenhofer M, Fuchs BM, Amann R, Zubkov MV, Tarran GA, and Pernthaler J (2009) Latitudinal distribution of prokaryotic picoplankton populations in the Atlantic Ocean. Environmental Microbiology 11: 2078--2093. Scho¨nhuber W, Bourhis GL, Tremblay J, Amann R, and Kulakauskas S (2001) Utilization of tmRNA sequences for bacterial identification. BMC Microbiology 1(20): 20--27. Scho¨nhuber W, Fuchs BM, Juretschko S, and Amann RI (1997) Improved sensitivity of whole-cell hybridization by the combination of horseradish peroxidase-labeled oligonucleotides and tyramide signal amplification. Applied and Environmental Microbiology 63: 3268--3273. Schramm A, Fuchs BM, Nielsen JL, Tonolla M, and Stahl DA (2002) Fluorescence in situ hybridization of 16S rRNA gene clones (Clone-FISH) for probe validation and screening of clone libraries. Environmental Microbiology 4: 713--720. Sekar R, Fuchs B, Amann R, and Pernthaler J (2004) Flow sorting of marine bacteria after fluorescence in situ hybridization. Applied and Environmental Microbiology 70: 6210--6219. Selje N, Simon M, and Brinkhoff T (2004) A newly discovered Roseobacter cluster in temperate and polar oceans. Nature 427: 445--448. Simek K, Armengol J, Comerma M, et al. (2001) Changes in the epilimnetic bacterial community composition, production, and protist-induced mortality along the longitudinal axis of a highly eutrophic reservoir. Microbial Ecology 42: 359--371. Smit E, Leeflang P, and Wernas K (1997) Detection of shifts in microbial community structure and diversity in soil caused by copper contamination using amplified ribosomal DNA restriction analysis. FEMS Microbiology Ecology 23: 249--261. Snaidr J, Amann R, Huber I, Ludwig W, and Schleifer KH (1997) Phylogenetic analysis and in situ identification of bacteria in activated sludge. Applied and Environmental Microbiology 63: 2884--2896. Sogin ML (2009) Characterizing microbial population structures through massively parallel sequencing. In: Epstein SS (ed.) Uncultured Microorganisms, pp. 19--33. Berlin: Springer. Stahl DA and Amann R (1991) Development and application of nucleic acid probes. In: Stackebrandt E and Goodfellow M (eds.) Nucleic Acid Techniques in Bacterial Systematics, pp. 205--248. Chichester: Wiley. Teira E, Reinthaler T, Pernthaler A, Pernthaler J, and Herndl GJ (2004) Combining catalyzed reporter deposition-fluorescence in situ hybridization and microautoradiography to detect substrate utilization by Bacteria and Archaea in the deep ocean. Applied and Environmental Microbiology 70: 4411--4414. Trebesius K, Amann RI, Ludwig W, Mu¨hlegger K, and Schleifer KH (1994) Identification of whole fixed bacterial cells with nonradioactive 23S rRNA-targeted polynucleotide probes. Applied and Environmental Microbiology 60: 3228--3235. Veitch NC (2004) Horseradish peroxidase: A modern view of a classic enzyme. Phytochemistry 65: 249--259. Wagner M, Amann RI, Lemmer H, and Schleifer KH (1993) Probing activated sludge with oligonucleotides specific for proteobacteria: Inadequacy of culture-dependent methods for describing microbial community structure. Applied and Environmental Microbiology 59: 1520--1525.

189

Wagner M, Horn M, and Daims H (2003) Fluorescence in situ hybridisation for the identification and characterisation of prokaryotes. Current Opinion in Microbiology 6: 302--309. Wallner G, Amann R, and Beisker W (1993) Optimizing fluorescent in situ hybridization with rRNA-targeted oligonucleotide probes for flow cytometric identification of microorganisms. Cytometry 14: 136--143. Woese CR (1987) Bacterial evolution. Microbiological Reviews 51: 221--271. Woese CR and Fox GE (1977) Phylogenetic structure of prokaryotic domain – primary kingdoms. Proceedings of the National Academy of Sciences of the United States of America 74: 5088--5090. Woese CR, Fox GE, Zablen L, et al. (1975) Conservation of primary structure in 16S ribosomal-RNA. Nature 254: 83--86. Yarza P, Richter M, Peplies J, et al. (2008) The all-species living tree project: A 16S rRNA-based phylogenetic tree of all sequenced type strains. Systematic and Applied Microbiology 31: 241--250. Yilmaz LS and Noguera DR (2004) Mechanistic approach to the problem of hybridization efficiency in fluorescent in situ hybridization. Applied and Environmental Microbiology 70: 7126--7139. Yilmaz LS, Okten HE, and Noguera DR (2006) Making all parts of the 16S rRNA of Escherichia coli accessible in situ to single DNA oligonucleotides. Applied and Environmental Microbiology 72: 733--744. Yu Z and Mohn WW (2001) Bacterial diversity and community structure in an aerated lagoon revealed by ribosomal intergenic spacer analysis and 16S ribosomal DNA sequencing. Applied and Environmental Microbiology 67: 1565--1574. Zeder M and Pernthaler J (2009) Multispot live-image autofocusing for highthroughput microscopy of fluorescently stained bacteria. Cytometry 75: 781--788. Zheng D, Alm EW, Stahl DA, and Raskin L (1996) Characterization of universal small subunit rRNA hybridization probes for quantitative molecular microbial ecology studies. Applied and Environmental Microbiology 62: 4504–4513. Zhou JZ, Bruns MA, and Tiedje JM (1996) DNA recovery from soils of diverse composition. Applied and Environmental Microbiology 62: 316--322. Zwart G, Crump BC, Kamst-van Agterveld MP, Hagen F, and Han SK (2002) Typical freshwater bacteria: An analysis of available 16S rRNA gene sequences from plankton of lakes and rivers. Aquatic Microbial Ecology 28: 141--155. Zwirgelmaier K, Ludwig W, and Schleifer KH (2004) Recognition of individual genes in a single bacterial cell by fluorescence in situ hybridization – RING-FISH. Molecular Microbiology 51: 89--96.

Relevant Websites http://www.arb-home.de ARB Project. http://www.microbial-ecology.net Department of Microbial Ecology: probeCheck. http://greengenes.lbl.gov Greengenes Project. http://rsbweb.nih.gov/ij ImageJ. http://www.ncbi.nlm.nih.gov NCBI; BLAST. http://www.probebase.net probeBase: Department of Microbial Ecology. http://rdp.cme.msu.edu Ribosomal Database Project. http://www.arb-silva.de SILVA: FISH.

3.09 Bioassays for Estrogenic and Androgenic Effects of Water Constituents K Kramer, Technische Universita¨t Mu¨nchen, Freising, Germany & 2011 Elsevier B.V. All rights reserved.

3.09.1 Introduction 3.09.2 In Vivo Bioeffect Assays 3.09.2.1 Classic In Vivo Assays at the Organism Level 3.09.2.1.1 Uterine weight of rodents 3.09.2.1.2 Sexual development of female rodents 3.09.2.1.3 Oviduct weight assay 3.09.2.1.4 Sexual differentiation of birds 3.09.2.1.5 Sexual differentiation of reptiles 3.09.2.2 Molecular Biomarkers 3.09.2.2.1 Methods to determine vtg 3.09.2.2.2 Quantitative PCR 3.09.2.2.3 Alkali-labile phosphate method 3.09.2.3 Gene Expression Analysis 3.09.2.3.1 Genome-wide DNA microarrays 3.09.2.3.2 Subset microarrays 3.09.3 In Vitro Assays at the Cellular Level 3.09.3.1 Cell Proliferation Assays 3.09.3.2 Vitellogenin Assays 3.09.3.2.1 Culture systems and cell types for in vitro vtg assays 3.09.3.2.2 Analytical considerations for vtg determination in cell cultures 3.09.3.2.3 Determination of vtg protein 3.09.3.2.4 Determination of vtg mRNA 3.09.3.3 Reporter Assays 3.09.3.3.1 ED-reporter assays based on nonestrogen hormone receptors 3.09.3.3.2 ED-reporter assays beyond transactivation 3.09.3.4 Yeast-Based Assays 3.09.3.4.1 Initial yeast estrogen screens 3.09.3.4.2 Optimization of the initial YES 3.09.3.4.3 Subtype YES 3.09.3.4.4 ER mutants 3.09.3.4.5 Extension of the YES principle 3.09.4 Subcellular Assays 3.09.4.1 ER Preparation 3.09.4.2 Enzyme-Linked Receptor Assay 3.09.4.3 Fluorescence Polarization Assays 3.09.4.4 Biosensors 3.09.5 Conclusions Acknowledgment References

3.09.1 Introduction Exposure of wildlife species to contaminants can permanently modify the development of the reproductive and endocrine systems. For example, alligators at the Lake Apopka, Florida, were exposed to a number of contaminants derived from agriculture and municipals as well as a major pesticide (dicofol and dichlorodiphenyltrichloroethane (DDT)) spill (Woodward et al., 1993; Guillette et al., 1995). Alligator eggs from Lake Apopka showed elevated levels of p,p0 - dichlorodiphenyldichloroethene (DDE), p,p0 -DDD, dieldrin, and

191 192 192 193 193 193 193 193 193 194 194 194 194 195 195 196 196 197 198 198 199 199 199 201 202 202 202 203 204 205 207 208 208 208 209 209 211 213 213

various polychlorinated biphenyls (PCBs; Heinz et al., 1991). Hatchlings from these eggs exhibited abnormal gonadal anatomy (Guillette et al., 1995). Another prominent example is 17a-ethynylestradiol (EE2), which is widely used as a human contraceptive. This substance is also administered for the treatment of prostate and/or breast cancer (Kuster et al., 2004). In addition to compounds in legal use, several strictly banned estrogens (Directive 96/22/EC), for example, dienestrol and diethylstilbestrol, which are applied as growth promoters during fattening of cattle, have been reported to occur in river sediments (Kuster et al., 2004).

191

192

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

Although environmental contaminants have been suspected to play a role in reproductive dysfunctions, developmental abnormalities, and cancer hazards to wildlife and humans (e.g., Colborn and Clement, 1992; Guillette et al., 1999; Segner, 2005), the linkage among these processes is still controversially discussed (Safe et al., 2002). Particularly, aquatic wildlife appears to be at risk by bioeffective contaminants, because watercourses constitute a natural sink of industrial chemicals, pesticides, or hormones excreted by humans, for example, the natural steroid hormone, 17b-estradiol (E2) (Ternes et al., 2002; Rutishauser et al., 2004). E2 has various physiological functions. In addition to its role in the reproductive system of women, this estrogen also plays various roles in cardiovascular health, bone integrity, cognition, and behavior. Considering the functional diversity of estrogen, it is not surprising that estrogen is also implicated in the development or progression of numerous disorders, including cancer (breast, ovarian, and prostate), neurodegenerative diseases (Alzheimer’s disease and Parkinson’s disease), cardiovascular disease, endometriosis, insulin resistance, obesity, and lupus erythematosus. These diseases can be divided into two groups: those believed to be caused by excess of estrogen, such as breast cancer (Yager and Davidson, 2006) and endometrial carcinoma (Deroo and Korach, 2006), and those diseases such as Alzheimer’s disease (Pinkerton and Henderson, 2005) that may be relieved by estrogen treatment. Natural hormones such as E2 act through the endocrine system, which essentially consists of cells producing and releasing the hormone into the bloodstream, through which it is transported to the corresponding target cells. The physiological response generated in the target cell due to a hormone is a composite reaction, which is followed by degradation and excretion from the body. Thus, the effective concentration refers to that amount of hormone capable of inducing a specific response in the target cell of an organism and, in the case of E2, is modified by binding to serum proteins such as albumin and sex hormone-binding globulin (SHBG). Albumin is a nonspecific binding protein with a low affinity and specificity for E2, whereas SHBG has a high affinity and specificity for E2. In order to accurately measure the estrogenicity of a certain compound, several factors must therefore be taken into account: (1) affinity of the compound for the estrogen receptor (ER), (2) accumulation of the compound in the environment and the body, (3) degradation or metabolism of the compound in the environment and body, and (4) the availability of the compound to the target cell (Arnold et al., 1996a). The evaluation of estrogenic compound concentrations is further hampered by the occurrence of two different ERs (ERa and ERb; Walter et al., 1985; Mosselman et al., 1996). Different tissues are characterized by a variation in the expression levels of the two ER isoforms (Couse et al., 1997), as well as in the levels of specific coactivators and corepressors. The latter are critical to elicit an estrogenic response (Shang and Brown, 2002). Mice lacking one or both ERs were created to define the receptor functions. ERa knockout mice turned out to be infertile (male and female). They were characterized by diminished bone density and a disturbed breast development. ERb knockout mice developed normally but females showed very reduced fertility due to defects in both the ovary and the uterus (Couse and Korach, 1999; Krege et al., 1998).

The human ER (hER) isoforms contain a highly conserved DNA-binding domain and an assimilable distinctive conservation of the ligand-binding domain (LBD). In addition to its ligand-binding activity, the LBD has dimerization activity and a ligand-dependent activation function-2 (AF-2; Katzenellenbogen, 1996). In the absence of ligand, the ER is associated with heatshock proteins (Hsps) and is transcriptionally inactive (cf. Figure 1; Smith and Toft, 1993). In the presence of ligand, the ER dissociates from the Hsps. This facilitates the homodimerization and binding to a regulatory DNA sequence, the estrogen response element (ERE; Figure 1). This ERE-bound, ligand-occupied ER complex can either activate or suppress the transcription of downstream target genes (e.g., vitellogenin (vtg) or zona radiata in fish; see Figure 1) in a cell- and promoter-specific manner (Fujimoto and Katzenellenbogen, 1994; Tsai and O’Malley,

Follicle cell Oocyte GtH

N E2

SbG

vtg/Zr Secretion Hepatocyte vtg/Zr

E2 ER Hsp

Hsp Hsp

Gene expression

Hsp E2

ERE ER ER E2 ER

ER

vtg and Zr gene N

Figure 1 Estrogen synthesis in oocytes and subsequent primary hormone effect on gene expression in hepatocyte of rainbow trout (Oncorhynchus mykiss). Explanation in the text. GtH, Gonadotropin; E2, 17b-estradiol; SbG, steroid-binding globuline; ER, estrogen receptor; Hsp, Heat shock protein; ERE, estrogen-responsive element; vtg, vitellogenin; Zr, zona radiata; N, nucleus. Adapted from Alberti MC (2006) Erfassung und Bewertung von Genexpressionsmustern von Zebraba¨rblingen (Danio rerio) nach Belastung mito¨strogenen Substanzen. PhD Thesis, Technische Universita¨t Mu¨nchen.

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

1994). If the complex recruits coactivators and induces the transcription of a downstream target gene, this biological function is generalized as the transactivation or transcriptional activity of the ER. However, it should be noted that in the absence of ligand, the ER could also form a homodimer and bind to the ERE (Zhuang et al., 1995). The molecular interaction of estrogenic compounds with components of the endocrine system constitutes the background of bioeffect assays, which are subsequently described. Particularly, cell-based assays are highlighted thereafter, because they have the potential to deliver rather complex information at reasonable economic means. These bioassays can essentially be used to identify whether discharges to the environment cause biological effects, such as endocrine interference. If the tests are positive, chemical evaluation should be utilized to identify the cause in an analytical chemistry/toxicity-identification evaluation exercise. This strategy enables the determination of the causative agents, and the results obtained by applying bioassays in the first stage are used to direct attention to the detailed chemical analysis of fractions until reasonable correlations are attained. The entire process can be repeated applying an additional fractionation procedure until the chemical complexity of the fractions is reduced sufficiently. Thus, nontarget chemical analysis enables the detection or identification of unknown compounds responsible for the observed effect (Petrovic et al., 2004).

3.09.2 In Vivo Bioeffect Assays Disruption of estrogen signaling is a prominent effect described for xenobiotic compounds, and this chapter focuses on bioeffect assays for the detection and characterization of endocrine disruptors (EDs), which act through interference with E2-mediated signaling components. However, due to the conserved mechanism of nuclear-receptor (NR) action, analogous assays have been already or can be established in a similar manner for the detection of compounds interfering with comparable hormone receptors (Gray et al., 1997, 2002; Vinggaard et al., 2002; Hartig et al., 2002). Numerous in vitro and in vivo assays have been suggested as screens for estrogenicity and their performance as well as correlation have critically been evaluated (e.g., Shelby et al., 1996; Zacharewski, 1997; Ashby, 1998; Combes, 2000; O’Connor et al., 2002; Charles, 2004; Scrimshaw and Lester, 2003). There exist various concepts of grouping these assays for the effect-based analysis of endocrine-disrupting compounds, depending on the individual point of view. In biological terms, these test systems can be distinguished by the corresponding organization level of the involved biological components: organism, tissue, cell, or subcellular elements. Therefore, the assays described subsequently are assigned accordingly.

3.09.2.1 Classic In Vivo Assays at the Organism Level Bioeffect assays at the organism level are performed to evaluate the impact of EDs on the endocrine system. In this context, multigeneration reproduction studies are considered as the ultimate test system for identifying adverse effects (O’Connor et al., 1998). However, it is unlikely that in vivo tests will, in the

193

long term, be utilized for monitoring the environment. Since they are expensive and time consuming, they are not regarded as ideal tools for an effective screening of EDs in economical terms (O’Connor et al., 1998). In addition, using animals for these investigations will raise critical ethical questions and encounter limited acceptance in the public (ECETOC, 1999). Despite this, in vivo studies are perfectly suited to validate in vitro techniques. Considering a reasonable range of species enables us to define different endpoints in order to obtain a comprehensive assessment of ED effects (Hoss et al., 2001). Subsequently, a few examples of well-established in vivo assays are briefly mentioned. This is followed by a more exhaustive description of in vivo assays using biomarkers and gene-expression profiling.

3.09.2.1.1 Uterine weight of rodents The rodent uterotrophic assay has, to date, been the most widely used in vivo assay. This test is considered to be the gold standard of estrogenicity (Korach and McLachlan, 1995; Gray, 1998). It is based on the ability of chemicals to stimulate uterine growth (Shelby et al., 1996; Beresford et al., 2000; Odum et al., 1997). In the classical uterotrophic assay, the test compound is administered with the diet to ovariectomized or immature female rats or mice, typically during several days (Buelbring and Burn, 1935; Kuch and Ballschmiter, 1999). Hormonally active compounds induce cell proliferation in the uterine mucous membrane. After explantation, the uterine weight is measured. The hormone effect of the test compound is determined by direct comparison to control animals treated with and without natural estrogens such as E2. The lowest observed effect dose (LOED) of this assay was determined at 104 mg E2 kg1 body weight. Significant uterine growth was observed on a daily dose of 0.4 mg kg1 body weight of E2, whereas phytoestrogen coumesterol was effective at 20 mg kg1 body weight per day (Baker et al., 1999). However, the results of the uterotrophic assay are dependent on various parameters such as animal species, procedure, and intervals of administration. Various estrogenic compounds such as o,p0 DDT, methoxychlor, chlordecone, and PCBs were identified using this in vivo assay (Guelden et al., 1998). Although the proliferative effect of natural estrogens on the female genital tract (e.g., the vaginal cornification) is considered as a reliable indicator of estrogenicity, this test is essentially not amenable for large-scale screening.

3.09.2.1.2 Sexual development of female rodents The administration of hormonally active compounds during the neonatal phase results in female rodents in premature sexual development, interference of the reproduction cycle (persistent vaginal estrus syndrome), and damage of the ovaries (Guelden et al., 1998). In addition, functional changes of the hypothalamus and the hypophyse were observed for rodents treated with o,p0 -DDT and chlordecone (Gellert et al., 1974; Gellert, 1978). These physiological alterations are considered as effect-related endpoints.

3.09.2.1.3 Oviduct weight assay This test is performed by feeding juvenile chicks up to several weeks, followed by the determination of the oviduct weight

194

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

(Dorfman and Dorfman, 1953). Based on the administered dose and test duration, the weight can increase up to 100-fold as compared to untreated controls. Besides the enhanced cell proliferation, concentrations of glycogen, ovalbumin, and conalbumin are raised in the ovaries. The estrogenic effect of the insecticide DDT using this assay was investigated (Bitman et al. (1968)) as early as in the year 1968. A variant of this initial assay focuses on various anatomical effects using an in ovo exposure strategy (Biau et al., 2007). Morphological defects in the urogenital system of the developing chick embryo were observed for estrone and estriol, whereas ethynylestradiol revealed fewer effects. Estriol caused persistence of Mu¨llerian ducts in male embryos and hypertrophic oviducts of females. Estrone exerted comparable effects though at a lower rate. Kidney dysfunction was exclusively observed with estrone, in both males and females.

3.09.2.1.4 Sexual differentiation of birds The left ovary together with the accordant oviduct is usually completely developed in female birds, whereas the corresponding right organs degenerate during the embryonic phase. Male birds lack these organs since the Mu¨llerian ducts from which these are developed, become degenerate on both sides. However, on exposure to o,p0 -DDT, p,p0 -DDT, and methoxychlor, feminization of male embryos of the seagull can be observed (Fry and Toone, 1981). The feminization results in degeneration or absence of the right testicle and the development of oviducts, of which the one on the left side is well developed.

3.09.2.1.5 Sexual differentiation of reptiles The gender development of many reptiles such as tortoise, lizards, and alligators is not genetically determined, but depends on the incubation temperature of the eggs. For example, the tortoise Trachemys scripta develops at a temperature of 32 1C into female, and at 26 1C into male progeny. If estrogens are present in the egg shell during sex determination, ovaries are developed even at lower temperatures. This effect on eggs incubated at lower temperatures also applies for antiestrogens (tamoxifen), antiandrogens (triphenylethylene), polychlorinated biphenyls, and other xenobiotics (Wibbels and Crews, 1992; Bergeron et al. 1994).

3.09.2.2 Molecular Biomarkers Biomarkers indicating exposure to pollutants and their effects are increasingly applied to assess the quality of ecosystems. Biomarkers can be defined as measurements of body fluids, cells, or tissues that indicate in biochemical or cellular terms the presence of contaminants or the magnitude of the host response (Livingstone et al., 2000). A broad spectrum of potential biomarkers was applied to study endocrine disruption. In aquatic organisms, these included changes in hormone titers (steroid hormones and thyroid hormones), abnormal gonad development (cf. above), and alterations in distinct enzyme activities (i.e., aromatases) and protein levels such as vtg, zona radiata proteins, and spiggin (Matthiessen, 2003; Kleinkauf et al., 2004). In fish, one of the most

frequently used biomarkers for screening estrogenic activity of chemicals and environmental samples includes the induction of vtg synthesis (e.g., Kime et al., 1999; Tyler et al., 1999). In egg-laying vertebrates, estrogens activate the hepatic synthesis of vtg, a calcium-containing glycolipophosphoprotein (Wallace, 1985; Arukwe and Goksøyr, 2003). Synthesized in the liver, vtg is transported via the bloodstream to the ovary where it is incorporated and sequestered by the maturing oocyte (Tyler et al., 1988; Specker and Sullivan, 1994). The vtg synthesis is generally restricted to females, while males contain no measurable or very low vtg levels in their plasma (e.g., Silversand et al., 1993; Tyler et al., 1996, 1999). However, the hepatic production of vtg can be induced in male fish by exposure to exogenous estrogens (Mommsen and Walsh, 1988). Thus, the level of vtg in male fish is indicative of exposure to estrogenic compounds. For this reason, vtg induction in fish has become an accepted indicator of exposure to estrogenic substances. The induction of this biomarker in intact fish has been utilized for chemical screening and environmental monitoring (Thorpe et al., 2003; Burki et al., 2006). The induction of vtg can be monitored in shortterm in vivo tests with fish, or, alternatively, in isolated fish hepatocytes (Navas and Segner, 2006).

3.09.2.2.1 Methods to determine vtg In fish in vivo, vtg is usually measured as circulating protein in the plasma, or as vtg mRNA in the liver. For vtg protein quantification, biochemical methods such as determination of phosphoprotein phosphorus (Craik and Harvey, 1984) or alkali-labile phosphoprotein (cf. below; Pelissero et al., 1991) have been used. Further tests include immunochemical methods such as radioimmunoassay (RIA; Jobling et al., 1998), enzyme-linked immunosorbent assay (ELISA; Bon et al., 1997; Marx et al., 2001), and Western blotting (Sole´ et al., 2000). These immunochemical tests are based on the selective binding of antibodies to the biomarker. The vtg mRNA can be determined semiquantitatively by reverse transcriptase-polymerase chain reaction (RT-PCR, Ren et al., 1996), real-time PCR (cf. below; Celius et al., 2000; Alberti et al., 2005; Burki et al. 2006), microarray technology (cf. below; Hoyt et al., 2003), or RNAse protection/dot-blot assays (Islinger et al., 1999; Navas and Segner, 2006). In order to define the effect of EDs on biomarkers such as vtg synthesis, the test organism is typically exposed to the chemical, ideally at strictly controlled conditions. In our group, the zebra fish Danio rerio was used as model organism to evaluate the bioeffects of estrogenic substances. The exposition was performed in a flow-through system. The experimental setup was as shown in Figure 2. Up to 12 male zebra fish were exposed to estrogenic compounds for 11 days in each of the tanks. Each experiment included one fish tank containing 30 ng l1 EE2 as positive and four basins without any estrogenic compound as negative controls. Four female fish were kept in an additional tank lacking any estrogenic compounds for calibration purposes. Following an exposure period of 11 days, the zebra fish were analyzed according to the methods outlined next.

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

195

1 6 4 2

5

3

7

7 8 (a)

(b)

Figure 2 Experimental setup of the exposition system for zebra fish. (a) Schematic drawing 1: freshwater inlet with filter; 2: overfall; 3: supply basin (500 l); 4: peristaltic pump for water supply; 5: peristaltic pump for dosing the test compound; 6: prepared test solutions; 7: exposition basin (20 l), 8: outlet via peristaltic pump and filter. (b) Exposition lab for zebra fish experiments showing exposition basins and peristalting pumps according to (a). Adapted from Alberti MC (2006) Erfassung und Bewertung von Genexpressionsmustern von Zebraba¨rblingen (Danio rerio) nach Belastung mit o¨strogenen Substanzen. PhD Thesis, Technische Universita¨t Mu¨nchen.

3.09.2.2.2 Quantitative PCR Quantitative PCR (qPCR) can be used to determine the effective concentrations of estrogenic compounds. Following exposition in the flow-through tank system described above, liver tissues of the exposed animals were dissected. After RNA extraction and reverse transcription into complementary DNA (cDNA), qPCR was performed using specific primers, which enabled the selective amplification of target and control genes. Fluorescence curves of a typical qPCR are shown in Figure 3. The crossing point constitutes the crucial value of these fluorescence curves. It describes the cycle number, at which the fluorescence signal gains exponentially in strength. Increasing template cDNA in a sample leads to a decreasing crossing point and vice versa. Thus, the expression level of vtg can be deduced, if a calibrator (e.g., nonexposed female zebra fish) is included in the study. The relative expression level is determined by the ratio of the crossing points of the target gene vtg and a reference gene (e.g., b-actin). Positive controls (nos. 3 and 5), which were exposed to EE2, are characterized by lower crossing points as the corresponding curves at 5000 mg l1 genistein (nos. 4 and 6). Negative controls (nos. 7 and 8) did not evoke any fluorescence signal. The calibrator curves

(nos. 1 and 2) derived from exposed female fish were used for normalization of the different qPCR runs.

3.09.2.2.3 Alkali-labile phosphate method Most of the biomarker studies mentioned above have been used in vertebrates such as rodents or fish. However, there exists an increasing interest in studying ED effects on invertebrate species, such as bivalve mollusks (Porte et al., 2006). The latter are used worldwide in biomonitoring programs (Matozzo et al., 2008). There are few specific antibodies developed against bivalve vtg-like molecules for quantitative immunochemical methods (Li et al., 1998; Kang et al., 2003; Osada et al., 2003). Therefore, indirect methods are used to study ED-triggered alterations, such as increase in RNA contents, lipid deposition, glycogen depletion, increase in protein levels, calcium, magnesium, and phosphoproteins contents (Verslycke et al., 2002; Arukwe and Goksøyr, 2003; Marin and Matozzo, 2004). Among these methods, the measurement of phosphoproteins by the alkali-labile phosphate (ALP) method has been widely used in different aquatic organisms such as fish and mollusks (Kramer et al., 1998; Verslycke et al., 2002).

196

Bioassays for Estrogenic and Androgenic Effects of Water Constituents 18 16

Relative fluorescence

14 1 2 3 4 5 6

12 10 8 6 4

7

2

8

0 −2 0

5

10

15

20

25

30

35

40

No. of cycles Figure 3 Fluorescence curves of a typical qPCR applying specific primers for vitellogenin 1 and b-actin1 (reference gene). Samples are derived from exposition of zebra fish with genistein. The experiment included 32 liver samples from exposition with genistein or EE2 (as positive control). Subsequently, experimental conditions eight curves are detailed. 1: vtg1-calibrator for comparison of different qPCR reactions; 2: X -actin1-calibrator for comparison of different qPCR reactions; 3: positive control, exposed with 30 ng l1 EE2, vtg1-primer; 4: sample, exposed with 5000 mg l1 genistein, vtg1-primer; 5: positive control, exposed with 30 ng l1 EE2, b-actin1-primer; 6: sample, exposed with 5000 mg l1 genistein, b-actin1-primer; 7: negative control, no sample added, vtg1-primer; 8: negative control, no sample needed, b-actin1-primer. Adapted from Alberti MC (2006) Erfassung und Bewertung von Genexpressionsmustern von Zebraba¨rblingen (Danio rerio) nach Belastung mit o¨strogenen Substanzen. PhD Thesis. Technische Universita¨t Mu¨nchen.

In fish, ALP levels have been shown to correlate with vtg levels determined by specific immunochemical and gene-expression techniques (Versonnen et al., 2004; Robinson et al., 2004). For example, a significant positive correlation was found between vtg levels measured by specific ELISA and ALP methods during an international intercalibration study using adult male zebra fish, which were exposed to E2 for 2–9 days (Porcher, 2003). However, the ELISA turned out to be more sensitive than the ALP technique (Ortiz-Zarragoitia and Cajaraville, 2005).

3.09.2.3 Gene Expression Analysis The modification of the gene-expression pattern in particular cell types is considered as one of the typical responses of an organism as either direct or indirect response to toxicant exposure (Nuwaysir et al., 1999; Steiner and Anderson, 2000). Some chemicals elicit toxic responses by initially damaging cellular components. Target cells typically respond by trying to repair the damage or to adapt to the injury, in part, through altering expression of appropriate repair genes. Other toxicants that modulate the endocrine system or cellular replication affect toxic responses directly by triggering signal transduction systems, leading to altered gene expression. It has been hypothesized that the spectrum of altered gene expression then determines the type and outcome of the toxic response. Thus, an approach to assess the toxicity of a given compound stems from the identification of gene-expression patterns elicited in a tissue or organ exposed to particular classes of chemicals. As gene and microarrays facilitate the quantitative analysis of thousands of gene-expression changes in a single experiment, transcript profiling can be used as a tool to predict toxic outcomes of exposure to particular chemicals (Naciff and Daston, 2004).

3.09.2.3.1 Genome-wide DNA microarrays A generic strategy was developed by Naciff and Daston (2004) to define a transcript signature, which is characteristic for chemicals with estrogenic activity. For this purpose, an oligonucleotide-based microarray was applied in standardized in vivo test systems, using the developing rat reproductive system (uterus and ovaries), at two life stages: fetal and prepubertal. For the evaluation of fetal and prepupertal response to EDs, the gene-expression profile induced by graded dosages of EE2, genistein, and bisphenol A (BPA) was evaluated, in order to establish dose–response relationships (McLachlan and Newbold, 1987; Diel et al., 2000). The gene-expression profiles were compared between treatment groups and controls (vehicle-treated animals), using a rat genome chip, which allows the evaluation of approximately 7000 rat-annotated genes and over 1740 expressed sequence tags. Scanned output files resulted in a signal value, corresponding to the level of expression for each transcript represented in the microarray, which was used to calculate the average fold change (Lockhart et al., 1996; Naciff et al., 2002). The exposure of prepubertal female rats to EE2 resulted in the expression modulation of more than 500 genes (out of 8740 evaluated genes) from the uterus and ovaries, compared with fetal exposure to the same chemical (Naciff et al., 2003). However, this response was only apparent at relatively high doses (1 and 10 mg EE2 kg1 d1). Over 45% of the genes regulated by estrogenic compounds (EE2, BPA, and genistein) in the fetal uterus were regulated, in the same direction by EE2 in the pubertal uterus/ovaries. This suggests that the exposure of female rats during fetal or prepubertal development to chemicals with estrogenic activity results in the alteration of gene expression in a developmental-stage-specific manner (Naciff and Daston, 2004; Naciff et al., 2002, 2003).

Bioassays for Estrogenic and Androgenic Effects of Water Constituents 3.09.2.3.2 Subset microarrays DNA microarray chips, which represent the entire genome of an organism, are perfectly suited to identify all ED-regulated genes (cf. above). However, once revealed, the array can be reduced to this subset of ED-responsive genes. The subset microarrays can be precisely tailored to meet individual experimental requirements in terms of number of replicates on a single chip, arrangement of individual genes, economic assay design, etc. An example for such arrays is presented in Figure 4. Yellow color indicates equal gene expression in both, exposed and control animals, whereas upregulated gene expression upon exposition results in red spots. Green color reveals downregulation. As majority of spots in the subset array on the left side of the depicted example in Figure 4 are yellow, the exposition with 1 mg l1 genistein did not evidently alter the expression levels compared to the control animal. However, following exposition with 5000 mg l1 genistein increases the number of red spots, which are indicative for upregulated genes in exposed animals (Figure 4, right). Bright red spots are located at position B7 (hemopexin, hpx), C2 (mesoderm posterior b, mespb), C10 (zona pellucida glycoprotein 2.4, zp2.4), D10 (zona pellucida glycoprotein 2.2, zp2.2), D12 (zona pellucida glycoprotein 3b, zp3b), and H12 (zgc:114012, vtg5) (Alberti, 2006).

197

modulating factors. The interplay between ligand-bound ER and these factors determines the response depending on the target cell. In vitro assays are tests for the identification of potentially endocrine-active compounds, but cannot, therefore, be solely used to deduce the risk of endocrine-related adverse effects. Furthermore, as most synthetic EDs exert activity only at high doses, non-ED-related effects have to be considered carefully to assess the actual potential of EDs to affect human health. The assays subsequently described allow a ranking of a series of compounds that could be used to prioritize compounds for studies, which enable the detection of reproductive and developmental toxicities in vivo (Mueller, 2004). The most frequently used cellular assays to screen chemicals and environmental samples for (anti)estrogenic activity in ecotoxicology include cell proliferation assays, particularly the MCF-7-based E-screen (e.g., Soto and Sonnenschein, 1985; Blom et al., 1998) and transactivation or reporter gene assays with fish, mammalian, or yeast cells (e.g., Ackermann et al., 2002; Giesy et al., 2002; Hornung et al., 2003). In addition, integrative assessment approaches are based on ER-mediated vtg synthesis induced in isolated hepatocytes of rainbow trout and quantified in nonradioactive dot-blot/RNAse protection assay in parallel to comprehensive chemical analyses of estrogenic substances (Holllert et al., 2005).

3.09.3 In Vitro Assays at the Cellular Level Even stripped down to the cellular level, EDs interfere with the complex network of ER signal transduction and modulate coupled cell signaling pathways. The physiological effect is therefore not solely defined by the ligand structure, but rather by the interaction of the receptor–ligand complex with various

3.09.3.1 Cell Proliferation Assays Predominantly based on human cell lines, proliferation techniques utilize a number of endpoints to measure the cell proliferation induced by exposure to estrogenic compounds. In the presence of estrogen, particular cells are stimulated

Figure 4 Subset microarrays showing different samples of exposed male zebra fish. (Left) Hybridization with Cy3-labeled (green) sample of a male control and a Cy5-labeled (red) sample of a male zebra fish, which was exposed to 1 mg l1 genistein. (Right) Hybridization with Cy3-labeled sample of a male control and a Cy5-labeled sample of a male zebra fish, which was exposed to 5000 mg l1 genistein. Adapted from Alberti MC (2006) Erfassung und Bewertung von Genexpressionsmustern von Zebraba¨rblingen (Danio rerio) nach Belastung mito¨strogenen Substanzen. PhD Thesis, Technische Universita¨t Mu¨nchen.

198

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

besides the enhanced proliferation for the synthesis of, for example, enzymes, growth factors, and progesterone receptors (PRs). Commonly used cells are estrogen-responsive MCF-7 or T47D human breast cancer cells and ZR-75 cells (White et al., 1994). The E-screen assay developed by Soto et al. (1992) is based on the increased growth of MCF-7 cells in the presence of estrogens (Soto et al., 1995). When a range of concentrations is tested, the method can differentiate between agonists, partial agonists, and inactive compounds (Korach and McLachlan, 1995). The E-screen assay compares cell yields in both, positive and negative controls with those from samples exposed to test compounds. Later on, the endpoint of the E-screen was modified to utilize a colorimetric endpoint, which was claimed to be faster and easier to perform than cell counting (Korner et al., 1998). A range of other means for quantifying cell growth has also been reported with the alamarblue and [3H]-thymidine-incorporation assays exhibiting enhanced sensitivity than cell counting as well as DNA and methylthiazoletetrazolium (MTT) assays (Desaulniers et al. 1998). Experimental protocols for proliferation of MCF-7 cells (and other estrogen-responsive lines) require that the media used for growth are stripped of steroids with dextran charcoal. The proliferative effect on estrogen-sensitive MCF-7 cell lines is inhibited, if they are treated in this manner. This inhibition can be compensated by addition of estrogenic compounds. Therefore, the E-screen does not determine the stimulation of cellular proliferation, rather than the reduction of the inhibitory effect on cell proliferation. The estrogenic response is calculated as relative proliferative effect (RPE) units, which corresponds to the maximum induced proliferation induced by the test compound compared to E2 (Koerner et al., 1999). This enables the discrimination between a potent agonist (RPE ¼ 100%) and a partial agonist (RPEo100%). In the past, several MCF-7 sublines were developed by cloning and by exposition to different selection conditions, for example, variation of culture media. As a consequence, these individual sublines display altered sensitivity toward estrogens. This has to be taken into account, if results obtained by E-screen techniques are compared, which are derived from different laboratories. Some of the developed cell lines even proliferate in estrogen-free media (Katzenellenbogen et al., 1987; Odum et al., 1998). This aspect essentially hampers standardization and validation of the method for the assessment of estrogenic compounds. Other disadvantages associated with cell-proliferation assays are due to the fact that mammalian cells exhibit tissue-specific differences in the expression of receptor subtypes (O’Connor et al., 1998; Scrimshaw and Lester, 2003).

3.09.3.2 Vitellogenin Assays The vtg response can be measured in isolated fish hepatocytes in a similar manner as described above for the analysis at the organism level. The application of this kind of bioassay has been suggested as an in vitro screen for identifying estrogenactive substances (Pelissero et al., 1993; Segner and Braunbeck, 2003). Compared to the corresponding in vivo assays, the benefit of the in vitro assay is that it is less laborious and more

cost efficient. Compared to other in vitro screening assays for estrogic compounds, for example, the recombinant yeast ER assay (cf. below), the hepatocyte vtg assay offers the potential to comprehensively detect effects of estrogenic metabolites, because the hepatocytes are metabolically competent and enable even the detection of antiestrogenic effects (Navas and Segner, 2006).

3.09.3.2.1 Culture systems and cell types for in vitro vtg assays One of the first reports using isolated liver cells of fish to study vtg synthesis was published by the group of Yves Valotaire and Farzad Pakdel (Maitre et al., 1986). While these first studies were focused on the physiological action of E2, it was in 1993 when Pelissero et al. (1993) as well as Jobling and Sumpter (1993) applied liver cell cultures of fish for assessing the estrogenic potency of phytoestrogens and xenoestrogens. Generally, the impact of various parameters, such as serum, medium constituents, or temperature, has to be generally considered for all kinds of cellular assays. These factors may significantly affect the endpoints of a cell assay, such as ED-induced alterations of vtg levels (Islinger et al., 1999; Pawlowski et al., 2000; Navas and Segner, 2006). The majority of published studies on vtg induction in isolated fish liver cells used monolayer culture systems (e.g., Kwon et al., 1993; Peyon et al., 1998; Navas and Segner, 2000). As an alternative to monolayer cultures, three-dimensional aggregate cultures of fish hepatocytes have been developed (Flouriot et al., 1993; Latonnelle et al., 2000). In these aggregates, the hepatocytes are arranged in a three-dimensional structure, similar to the arrangement characteristic for functional tissues in vivo. Flouriot et al. (1993) showed that for identical culture conditions including identical culture medium, the mRNA levels of vtg and ER in the aggregates were significantly higher than those observed in monolayer cultures. Moreover, the longevity and estrogen responsiveness of the liver cells were extended in the aggregates. Whereas in hepatocyte monolayers the level of ER mRNA rapidly decreased after 1 week of continuous culture, three-dimensional cultures maintained uncompromised functional vtg synthesis and secretion over a period of 30 days (Flouriot et al., 1993). Alternatively, an in vitro system can be based on liver slices (Schmieder et al., 2000; Shilling and Williams, 2000). Livertissue slices essentially benefit from the naturally preserved tissue organization. Tissue slices are not restricted to hepatocytes solely (as typically used in monolayer-based assays; Navas and Segner, 2006). They complement the basic function of hepatocytes with the entire spectrum of specialized cell types, which are necessary to sustain tissue homeostasis and therefore warrant a comparably uncompromised organ function and metabolism. Finally, one factor, which has a pronounced influence on basal and hormone-inducible vtg synthesis of isolated fish hepatocytes, is the physiological status and the endocrine history of the donor animal. The production of vtg varies between male and female hepatocytes (e.g., Smeets et al., 1999). Whereas liver cells isolated from male fish usually show no measurable vtg expression in hormone-free media, hepatocytes isolated from maturing females produce vtg

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

mRNA and protein without extracellular estrogenic stimulation. For instance, Pelissero et al. (1993) found no basal production of vtg protein in hepatocytes isolated from male rainbow trout or immature females. However, liver cells obtained from maturing females released vtg protein into the culture medium (Smeets et al., 1999; Navas and Segner, 2006). In this context, it is of crucial relevance whether the genderspecific difference of basal vtg production influences the sensitivity of the hepatocytes to estrogen treatments. As vtg synthesis essentially depends on the expression of the ER, hepatocytes of female origin, which have a higher constitutive expression of ER, may respond faster or be more sensitive to estrogen treatment than male hepatocytes. In line with this consideration, Riley et al. (2004) found that 107 M E2 was sufficient to significantly induce vtg production in hepatocytes isolated from female tilapia, while a significantly higher concentration of 104 M E2 was required in male hepatocytes. Consequently, the elevated basal secretion of vtg by female hepatocytes may obscure weak induction responses. For instance, Kordes et al. (2002) described that while 109 M EE2 were able to induce vtg secretion in liver cells isolated from male medaka (Oryzias latipes), this concentration did not elevate vtg secretion in hepatocytes isolated from mature females. Furthermore, Smeets et al. (1999) observed that the dose–response relationships for E2 were similar in hepatocytes from male and female carp. Despite this, female hepatocytes reached higher absolute levels of vtg than male hepatocytes. From these results, the authors concluded that the ER level in the hepatocytes influences only the absolute magnitude of vtg production but not the relative sensitivity for estrogens. Nevertheless, they suggested to use hepatocytes of male origin when assessing estrogenic potencies of chemicals, mainly because the induction of vtg can be detected more easily and with a higher induction factor in male hepatocytes, which show no or minor basal synthesis of vtg (Navas and Segner, 2006). The role of species/donor fish differences on the vtg response has been discussed by Latonnelle et al. (2000), who showed that rainbow trout hepatocytes cultured as aggregates and exposed to E2 after 4 days of culture (the minimal time necessary for the formation of aggregates) required 2 days to synthesize measurable levels of vtg. Sturgeon hepatocytes cultured under the same conditions required 6 days of E2 treatment until detectable levels of vtg were produced. It is tempting to ascribe this variation to a species difference. However, knowing that the rainbow trout hepatocytes originated from fish at the onset of the reproductive cycle, while the sturgeon hepatocytes were isolated from immature fish, it is more likely that a difference in the physiological status of the donor fish explains the observed variation in the lag time of vtg induction (Navas and Segner, 2006).

3.09.3.2.2 Analytical considerations for vtg determination in cell cultures The majority of studies investigating vtg induction in cultured fish hepatocytes used the natural estrogen E2 as a positive reference. In all publications, increasing vtg production has been reported with concomitantly raised concentrations of E2. Nevertheless, there seems to exist a pronounced variability

199

concerning the minimal concentration of E2 eliciting a vtg production, which is significantly different from that detected in controls. Minimal effective E2 concentrations vary at a rather wide range from 1012 to 106 M. Reported EC50 values range from 1011 to 106 M E2, with a concentration centered around a concentration of approximately 109 M E2. The lag time between the time point when estrogen treatment starts and when vtg production becomes detectable is variable and may be influenced by estrogen concentration, culture conditions, analytical methods, physiological status of the donor fish, and species differences. Lower estrogen concentrations appear to require longer incubation periods when compared to higher estrogen concentrations (e.g., Peyon et al., 1993; Latonnelle et al., 2000; Okoumassoun et al., 2002). Several authors addressed the problem of establishing significant concentration–response curves of vtg induction with xenoestrogens in the hepatocyte assay (e.g., Islinger et al., 1999; Toomey et al., 1999). Xenoestrogens typically have low estrogenic potencies, and thus relatively high concentrations (107 to 104 M range, see above) have to be applied in order to induce a vtg response in cultured fish hepatocytes. Above a critical level, these high concentrations can exert cytotoxic reactions, which in turn result in decreased vtg synthesis (Navas and Segner, 2006). Furthermore, the cellular metabolism of the test compounds affects the time- and concentration-dependent response of hepatocytes to estrogens. Cultured fish hepatocytes maintain biotransformation capabilities (Segner and Cravedi, 2001) and metabolize E2 rather rapidly. It has been shown for several species (Peyon et al., 1998; Schmieder et al., 2000) that the half-life of E2 in in vitro liver preparations of fish is limited to approximately 0.5–2 h. The biotransformation of xenoestrogens by cultured fish hepatocytes has been demonstrated accordingly (Cravedi and Zalko, 2005). Thus, differences between test substances or between species in their rate of metabolic turnover can influence estrogenic potencies (e.g., Lindholst et al., 2003). Metabolism of test agents is of particular relevance for the detection of estrogen precursors, which become estrogenic by cellular metabolism. Contrary to other in vitro screens for estrogenic activity, such as ER-binding assays, cultured hepatocytes enable the detection of these estrogen precursors. A few studies have addressed the capability of cultured hepatocytes to detect estrogenicity resulting from xenobiotic metabolism. Petit et al. (1997) tested 32 substances in a recombinant yeast assay containing the rainbow trout ER as well as in cultured trout hepatocytes. They found that 30% of the substances were estrogenic exclusively in one of the two test systems, and hypothesized that this difference is due to metabolic differences between the two test systems. A case study was provided for nonylphenol (NP) derivatives. In the recombinant yeast system, 4-nonylphenol exhibited pronounced estrogenicity, while the estrogenic effect of nonylphenol ethoxylates decreased with the increasing length of the ethoxylate substituent. This decrease in estrogenic activity was associated with a strong decrease in binding affinity to the ER. As measured in a receptor-binding assay, the binding affinities of NP, NP-diethoxylate, and NP-heptaethoxylate were 270, 6700, and 47000 times, respectively, lower than that of E2 (Petit et al., 1997). In the hepatocyte assay, however, NPdiethoxylate showed an eightfold higher estrogenic potency

200

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

than NP, suggesting that NP-diethoxylate with low ER affinity was metabolized into NP with higher ER affinity (Navas and Segner, 2006).

3.09.3.2.3 Determination of vtg protein In order to measure vtg in cultured cells, in principle, the same methods are applicable as employed for the analysis of vtg in vivo. ELISA was the analytical method of choice in most of the studies published to date. By this means, vtg protein can be measured either as secreted protein in the cell-culture medium or as intracellular protein in the cells (following homogenization). The ELISA procedure is an immunochemical technique based on detecting and quantifying antigens by selective antibodies. It can generate absolute (using a vtg standard curve) or relative values (without a standard curve, measuring only relative optical density readings). Comprehensive overviews are given, for example, by Nilsen et al. (2004).

3.09.3.2.4 Determination of vtg mRNA Alternatively to the immunochemical detection of vtg protein in hepatocyte culture supernatants, the mRNA level of transcribed vtg genes can be determined. The assay principle is similar to the methods applied for the corresponding analysis at the organism level (cf. above). Since the number of fish species for which vtg sequences are available is steadily expanding, mRNA-directed techniques can be applied for a broad range of different species. Islinger et al. (1999) introduced a dot-blot/RNAse protection assay employing digoxigenin-labeled RNA transcripts. Another assay is based on RTPCR (cf. above) as developed by Pawlowski et al. (2000) in isolated rainbow trout hepatocytes. In this method, the level of vtg mRNA is determined densitometrically. This enables at least a semiquantitative analysis of vtg mRNA levels. The amount of target mRNA is then typically calibrated by the mRNA of a housekeeping gene, for example, b-actin. The choice of a housekeeping gene can be crucial since its expression level may be not as constant as assumed and may

ER ER gene

Reporter protein

vary with culture conditions and treatments. The use of RTPCR typically improves the sensitivity of the assay, and thus enables the detection of vtg gene transcription at lower concentrations of xenoestrogens as compared to ELISA. In contrast to semiquantitative RT-PCR, properly designed real-time RT-PCR experiments (cf. above) allow a quantitative analysis of vtg mRNA levels, either by means of relative quantitation of vtg mRNA using the threshold approach (Funkenstein et al., 2004) or by means of absolute quantitation calibrated by a standard curve. The problem of calibrating the target mRNA against housekeeping genes, however, remains as in other mRNA techniques (Navas and Segner, 2006).

3.09.3.3 Reporter Assays Reporter assays as ED screening methods are based on the fact that steroid receptors are transcription factors, which induce expression of target genes after binding to specific DNA sequences in their promoter region. Transient transactivation assays, in which cells are transfected with the cDNA for ER and a reporter gene containing an ERE or an estrogen-responsive promoter, are widely used to measure ligand-induced ERmediated gene activation (Shelby et al., 1996). In this assay, yeast or mammalian cell lines lacking endogenous ER are transiently transfected with an expression plasmid carrying the gene encoding ERa, ERb, or any desired receptor variant along with an ER-responsive promoter or ERE linked to a chloramphenicol acetyltransferase (CAT) or luciferase reporter gene (cf. Figure 5). ER ligands induce a dose-dependent transcription of the reporter protein and can easily be monitored. Due to the high sensitivity of the available luciferase reporter vectors, very weak to highly potent estrogens can be analyzed. Furthermore, single compounds and chemical mixtures can be analyzed depending on ERE, ER-subtype, and cellular context for their estrogenicity and antiestrogenicity. This versatility is especially important if the tissue-specific estrogenic/antiestrogenic activity of the so-called selective ER modulators (SERMs), such as tamoxifen and raloxifene, is considered

ER ERE

Reporter gene

E2 Estrogen receptor (ER)

Gene expression

Estrogen response element (ERE)

Figure 5 General principle of reporter-gene assays based on ERE-activation by the presence of E2 or ED in the sample. ER genes can be endogenously expressed or provided by transient or stable transfection. ERE reporter fusions are located on a separate plasmid. Popular reporter genes are, for example, b-galactosidase, luciferase, CAT, and gfp. See the text for detailed explanation.

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

(McDonnell, 1999). SERMs are synthetic estrogens applied for the treatment of hormone-dependent disorders. The SERMs differentially activate wild-type ER subtypes and variant forms expressing activation function 1 and 2 in human cells transfected with a pC3-luciferase construct, and these in vitro differences reflect their unique in vivo biologies (Safe et al., 2001). In addition to these transient ER-expressing systems, tissuespecific cell lines have been developed, which stably express the ERa or ERb gene (Jiang and Jordan, 1992; Mueller, 2004). However, as the production of ER is artificially enforced in cells, whose wild type is lacking this receptor, some of the cellular reactions may not necessarily reflect the physiological response of the analyzed cell type. Therefore, cell lines with endogenous ER expression were used for transactivation assays, in order to compensate these limitations. Examples are cell lines such as human breast tumor T47D or human ovarian BG-1 cells. Similar immortalized cell lines can be employed to measure transactivation on stably integrated or transiently transfected ERE reporter vectors (Legler et al., 1999; Rogers and Denison, 2000; Mueller and Korach, 2001; Wilson et al., 2004). As both ER subtypes have different primary sequences in their activation function 2-containing hormone-binding domains (Gustafsson, 1999; Ogawa et al., 1998), some ER subtype-selective ligands have been identified. These ligands differ in their binding affinities for the two receptors and exhibit variable agonistic or antagonistic attributes according to the ER subtype considered (Veeneman, 2005). Balaguer et al. (1999) established stably transfected ERa- or ERb-responsive reporter cell lines to reveal the ER subtype-selective activities of various compounds. These cell lines enable to determine the whole-cell affinity of ligands for hERa and ERb and to precisely compare their effects on transcriptional activation (Escande et al., 2006). The human cervix adenocarcinoma HeLa cell line was used by Escande et al. (2006) as a host to generate stable reporter cell lines, because it does not express endogenous ERs. The corresponding HELN–ERa and HELN–ERb reporter cell lines were generated by first transfecting HeLa cells with an estrogen-responsive reporter to obtain the HELN cell line. The reporter gene contains a luciferase gene driven by an ERE in front of the b-globin promoter and a neomycin phosphotransferase gene under the control of the SV40 promoter (Balaguer et al., 1999). In a second step, HELN cells were transfected with the corresponding ER subtype expressing plasmid to obtain the HELN–ERa and HELN–ERb cell lines, respectively. The Kd value, calculated from saturation curves, was approximately 0.1 nM for Erb, whereas an almost threefold higher affinity was determined for ERa. These Kd values were within the range generally reported for E2 binding to ERs in various systems (Kuiper et al., 1998). The reporter cells were used to compare the induction of transcriptional activity of ERa and ERb by various estrogen agonists and antagonists. For instance, raloxifene was identified as the most selective antagonist for ERa and RU486 as the most selective antagonist for ERb. Raloxifene exhibited an ERa-selective partial agonist/ antagonist function but a pure antagonistic effect through ERb (Barkhem et al., 1998). Another example for mammalian reporter assays was established employing the human breast cancer cell line T47D.

201

This cell line was transfected with the Gal4 ER and luciferase gene constructs which resulted in sensitive and responsive cells (Legler et al., 1999). Chemical-activated luciferase gene expression (CALUX) cell line assays have been established in rat or mouse hepatoma cells, which were transfected with receptor-controlled luciferase reporter gene constructs.

3.09.3.3.1 ED-reporter assays based on nonestrogen hormone receptors Due to their anabolic effects, androgens are used to promote muscle strength in athletes and meat quantity in farm animals (Meyer, 2001; Evans, 2004). It has also been found that environmental chemicals can interfere with androgen action, thereby possibly contributing to the disruption of the endocrine system in wildlife and humans (Kelce and Wilson, 1997; Andersen et al., 2002). The androgen receptor (AR) is a liganddependent transcription factor that regulates specific gene expression by binding to specific hormone-response elements within the regulatory DNA sequences of androgen-responsive genes (Claessens et al., 2001). The basic mechanism is comparable to the corresponding estrogen system. The enhancer region of the mouse mammary tumor viral long terminal repeat (MMTV-LTR) promoter is the most widely used enhancer to study AR function. Four inverted repeats of the core sequence 50 -TGTTCT-30 within the MMTVLTR enhancer are recognized by: AR, glucocorticoid receptor (GR), PR, and mineralocorticoid receptor (MR; Glass, 1994), now classified as the members of the 3C group within the NR family. The MMTV promoter also contains several enhancer regions that can be addressed by transcription factors that may respond to other hormonal and cellular stimuli, thereby modulating steroid responses (Aurrekoetxea-Hernandez and Buetti, 2004; Uchiumi et al., 1998). Several stable reporter gene assays have been described for androgens. However, these systems still have several drawbacks, since they either have a low responsiveness, employ slowly growing prostatic cell lines, or are not selective in their response because of expression of other nuclear hormone receptors of the C3 class, activating the transfected reporter gene through non-AR-mediated mechanisms (Terouanne et al., 2000; Blankvoort et al., 2001; de Gooyer et al., 2003; Wilson et al., 2002). The full-length MMTV promoter has been used to generate a number of androgen-responsive reporter cell lines. Although this promoter is quite selective to AR, PR, and GR, it also contains a number of regulatory sites that can be targeted by different agents other than steroids (Uchiumi et al., 1998; Ouatas et al., 2002). MDA-kb2 is a derivative of a human breast-cancer cell line containing such a stably integrated MMTV-luciferase reporter (Wilson et al., 2002). In addition to responding to androgens, this cell line responds very strongly to glucocorticoids acting through the GR that is present endogenously. This makes the system unsuitable as a selective screening tool. However, the androgen specificity was improved by stable transfection of human prostatic PC-3 cells with human AR (hAR) and the MMTV-luciferase reporter, named PALM cells (Terouanne et al., 2000), CHOhARMMTVluc cells (de Gooyer et al., 2003), and COShARMMTVluc cells (Paris et al., 2002). So far, the only cell line

202

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

that uses a simpler reporter construct, thereby avoiding influences by nonsteroidal regulatory pathways, is derived from the human breast-cancer cell line T47 D, stably transfected with a luciferase reporter under transcriptional control of the PB-ARE2 androgen response element (Blankvoort et al., 2001). This stable cell line shows additional hormone class specificity, as it mainly responds to progestins, due to the known overexpression of PR in T47D cells, and relatively low endogenous AR levels (Sutherland et al., 1988). Sonneveld et al. (2005) developed an androgen-reporter cell line that combines high specificity and sensitivity. They selected the human bone cell line U2-OS, in which the stably introduced hAR was highly active, whereas expression of other C3 class receptors was insignificant. Based on earlier observations (Quaedackers et al., 2001) and transient transfections using a panel of steroid receptors, steroid reporter plasmids, and different cell lines (HEK293, T47D, U2-OS, HeLa, and CHO), the osteoblastic osteosarcoma U2-OS cell line was chosen as the best candidate to serve as an appropriate platform for androgen-, estrogen-, glucocorticoid-, and progestinresponsive reporter cell lines. The selection was mainly based on the observation that the U2-OS cell line showed little or no endogenous receptor activity using reporter plasmids only, while it supported strong hormone-mediated responses when cognate receptors were transiently introduced (Quaedackers et al., 2001). Whereas no evidence for significant endogenous activity of AR, PR, or GR upon ligand stimulation was found, cotransfection of the appropriate receptors resulted in high reporter activity upon ligand treatment. Stable transfectants were selected from U2-OS cells transfected with the hAR, hPR, and hGR and the 3x HRE-TATA-Luc reporter construct, or with the hERa in combination with the 3x ERE-TATA-Luc reporter construct (Quadackers et al., 2001; Sonneveld et al., 2005). In this manner, distinct steroid reporter cell lines with the same cellular background (U2-OS) and comparable minimal promoter reporter constructs (multimerized response elements coupled to the TATA box and the luciferase reporter gene) were generated. These lines are characterized by high levels of induction (fold induction ranging between 30 and 80), high stability (usually more than 40 passages), high sensitivity (picomolar to nanomolar range), and high selectivity (Sonneveld et al., 2005). For example, the thereby derived AR CALUX cell line (Sonneveld et al., 2003) was remarkably selective for androgens, showing no substantial agonistic response to the (androgen) precursors DHEA, pregnenolone, and other natural steroid hormones. The most potent androgen was dihydrotestosterone (DHT), activating these cells with an EC50 of 0.13 nM. The AR CALUX cells showed high sensitivity toward all natural androgens tested, with the following range of potencies (EC50 values in nM): DHT (0.13), testosterone (0.66), and androstenedione (4.5). Remarkably, in contrast to the ER that is prevalently activated by environmental pollutants, the AR seems to be prone to antagonism rather than agonism (Sohoni and Sumpter, 1998; Willemsen et al., 2004). This correlation was revealed by testing pesticides with ERa agonistic activity, which were found to be also AR antagonists (e.g., o,p 0-DDT, p,p 0-DDT, methoxychlor). This phenomenon may contribute to the

feminization of male fish as described before. While only a small number of environmental compounds are currently known as AR agonists, some of them are quite potent antagonists. The AR CALUX bioassay readily classified chemicals according to their antiandrogenic properties when tested in the presence of EC50 concentrations of DHT. The well-known and widely used AR antagonists flutamide (IC50 5 1.3 mM), vinclozolin (IC50 5 1.0 mM), and cyproterone acetate (IC50 57.1 nM) clearly showed antagonistic properties (Sonneveld et al., 2005). Compounds, particularly at micromolar levels or higher, can occasionally nonspecifically repress responses in reportergene assays. This can be due to overall cytotoxicity, ultimately leading to cell death, but can also be due to more specific effects such as inhibition of protein synthesis or mRNA transcription. The latter effects precede the more general cytotoxic effects, with cell death as the least sensitive parameter. Therefore, controls should be assessing nonspecific repression of reporter gene activity rather than overall cytotoxicity and cell death. Constitutively expressed reporter genes, which are frequently used as controls, have the drawback that no bona fide constitutive promoters have been identified so far. Therefore, the use of these controls cannot be recommended. In the case of steroid receptor-mediated responses, an appropriate control for nonspecific inhibition is considered in the determination of the effect of the test compound on the reportergene activation by an excess of high-affinity agonist. For example, all inhibitory responses of the CALUX assay can be reversed by coincubation of the sample with excess DHT, demonstrating the specificity of the response (Sonneveld et al., 2005). Squelching of common cofactors by other nuclear (hormone) receptors is a well-known mechanism of interference and might therefore produce false-negative results. An example for this type of mechanism is the interference between PR and ER (Kraus et al., 1995). However, squelching does not seem to be prominent in U2-OS-derived CALUX bioassays, because they do not express high levels of steroid receptors other than the stably introduced receptor of interest. This was shown by the fact that progestins and glucocorticoids do not interfere with DHT- or E2-induced luciferase activity in the AR and ERa CALUX bioassays, respectively, while androgens do not show reduced E2-induced luciferase activity in the ERa CALUX bioassay. Another receptor shown to possess squelching effects with nuclear hormone receptors is the aryl hydrocarbon receptor (AhR). Interference of the AhR ligand dioxin on ER signaling was demonstrated in T47D cells (ER CALUX bioassay) expressing functional AhR (Legler et al., 1999), but not in U2-OS cells (ERa CALUX bioassay) not expressing AhR (Sonneveld et al., 2003).

3.09.3.3.2 ED-reporter assays beyond transactivation In addition to the analysis of ligand binding and ER transactivation, steroidogenesis of endogenous estrogens offers an endpoint to measure the physiological response to EDs. These assays include the measurement of the activity of steroidogenic enzymes such as aromatase (cf. below; Mak et al., 1999) and the quantification of the pattern of steroid biosynthesis (Gray, 1998). Other assays include the measurement of ER/coactivator association by glutathione-S-transferase

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

pull-down, fluorescence resonance energy transfer (FRET), or two-hybrid assays (below), and analysis of ER-mediated gene expression (Routledge et al., 2000; Jorgensen et al., 2000). These assays can be very helpful for the analysis of reaction mechanisms of selected xenoestrogens and elucidation of their putative role in tissue-specific estrogenic effects (An et al., 2001). A critical shortcoming of in vitro assays in general is the lack of metabolic competence of the used cellular systems. As many compounds have to be activated by metabolic biotransformation in order to obtain estrogenic potential, a reporter gene assay may not be able to identify compounds that are potent xenoestrogens in vivo due to their metabolism. In order to compensate for this frequently raised criticism, primary human and rat hepatocytes were employed that maintain metabolic competence in culture. Primary cell cultures can be used in transactivation assays and primary hepatocytes treated with the UV-filtering compound 4-methylbenzylidenecamphor were shown to be responsive in ER reporter-gene assays, because they are capable of producing estrogenic metabolites of the tested compound (Mueller, 2004; Mueller et al., 2003).

3.09.3.4 Yeast-Based Assays 3.09.3.4.1 Initial yeast estrogen screens The yeast Saccharomyces cerivisiae is one of the classic model organisms applied in biological research since decades. In contrast to multicellular organisms, it is the most basic eukaryotic model at a single-cell level, which provides functionality of many mammalian genes. Metzger et al. (1988) have shown that despite hERa being nonfunctional when expressed in bacteria, it retains its native ligand-activated transcriptional activity in yeast cells. This indicates that the underlying regulatory mechanisms have been highly conserved during evolution and constitute the fundamental finding on which all other yeast estrogen screens (YESs) are predicated. Some of the advantages of yeast-based assays compared to mammalian cell culture are the lower costs for nutrient media, shorter assay duration as a consequence of the higher growth rate, and the robustness toward toxic effects such as endotoxins or solvents (Breithofer et al., 1998). However, the latter aspect is discussed controversially in literature. For example, the nonpolar fractions of sediment extracts were reported to show poisonous effects on yeast cells, which hampered the measurement of estrogenic activity by YESs. In contrast, the nonpolar fraction did not significantly harm human endometrial adenocarcinoma cells (Hashimoto et al., 2005). Despite this controversy, the yeast system is the model of choice for testing environmentally relevant samples such as sewage effluents, because contaminants present in these usually nonsterile samples do not impair the cellular proliferation and ER activity measurement in yeast in contrast to mammalian cells (Ramamoorthy et al., 1997). Further benefits of this system are seen in the availability of a large variety of genetic tools, in order to install yeast variants, which provide a perfect match for a given application. As a fundamental limitation, screening techniques based on recombinant yeast have been criticized as such systems may not accurately

203

mirror mammalian or human systems, for example, in terms of cell walls and molecular transport mechanisms as a prerequisite for exposure to xenoestrogens. Another potential disadvantage of YES is that it is unclear whether yeast cells metabolize proestrogens to estrogens. Therefore, results obtained by YES will require confirmation in mammalian cells. In the 1990s, a number of yeast transactivation assays were developed for the determination of estrogenic activity. One of the initial systems was developed by Lyttle et al. (1992). It consisted of an expression plasmid (YEpE12) encoding the hERa, the copper-inducible CUP promoter, and a tryptophan auxotrophy marker. The marker compensated for the lack of cell lines to grow on tryptophan-deficient media by recruiting the enzymes required for the synthesis of this amino acid. Yeasts were co-transfected with the reporter plasmid YRpE2. The reporter plasmid carried two copies of the vtg ERE, the iso1-cytochrome c (CYC1) promoter in fusion with the b-galactosidase-encoding lacZ gene and a uracil auxotrophy marker (Lyttle et al., 1992). A few years later, a recombinant yeast strain was developed by Ian Purvis in the Genetics Department at Glaxo for the identification of compounds that can interact with the hER. The DNA sequence of the hER was stably integrated into the yeast genome. In this system, the hER was functionally expressed retaining its binding capability to ERE upon hormone stimulation. These ERE sequences were located within a strong promoter sequence on the b-galactosidase-encoding reporter plasmid. In the same manner as in the assay described above, upon binding an active ligand, the estrogen-occupied receptor interacted with transcription factors and other transcriptional components (Katzenellenbogen et al., 1993) to modulate gene transcription. This caused expression of the reporter gene lac-Z and the produced b-galactosidase was secreted into the medium. Following secretion, the enzyme metabolized the chromogenic substrate, chlorophenolred-b-D-galactopyranoside (CPRG) into a red product that could be measured spectroscopically at 540 nm. This yeast assay offered a measuring range for E2 from 1.5 to 3072 ng l1 and was described to be highly reproducible (Routlegde and Sumpter, 1996). Therefore, the yeast assay could detect E2 at concentrations, which were 5 times lower than that of the MCF-7 cell system initially reported by Soto et al. (1992; cf. above). The entire procedure of these initial YES assays spanned over a duration of approximately 3–4 days including cell growth and exposition. A similar recombinant yeast screen described by Klein et al. (1994) also relied on the expression of b-galactosidase and was even applicable for screening blood plasma levels of estradiol. Arnold et al. (1996b) described an assay, analogous to the one described by Routledge and Sumpter, based on the hER-containing expression plasmid pSCW231-hER and the reporter plasmid YRPE2. This latter plasmid from McDonnell et al. (1991) contained two EREs linked to the lacZ gene. The EC50 of this assay was determined at approximately 0.2 nM for E2. Similar to the majority of YES assays, the test was performed in 96-well microtiter plates, allowing screening of multiple compounds over a wide range of concentrations. Another initial system was based on ubiquitin ER fusions, which turned out to be more stable as compared

204

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

to the unfused hormone receptor (Graumann et al., 1996). Subsequent modifications at the N-terminus according to the N-end rule from Varshavsky (1992) showed effects on receptor expression level and activation function-1 (AF-1) domain on transactivation.

3.09.3.4.2 Optimization of the initial YES Initially developed YES systems revealed several shortcomings in the routine screening of EDs. For example, Vanderperren et al. (unpublished) found high blank values when applying the assay of Routledge and Sumpter (1996). A similar effect was observed when higher cell densities were used compared to the original protocol. These were the first hints that the chromogenic b-galactosidase substrate CPRG or chlorophenol red interfered with assay results. Vanderperren et al. (unpublished) found dose-dependent estrogenic effects caused by the substrates of the YES assay at a concentration of 165 mmol l1, which corresponded to the substrate quantity applied in the original protocol. At the time when the sample was added to the yeast cells, the medium already contained the CPRG substrate. Therefore, the estrogenic response in the assay seemed to be an additive effect of the test compound and CPRG/chlorophenol red. In an attempt to compensate this interference, De Boever et al. (2001) redesigned the assay. During the initial growth phase, the cells were brought into contact with the sample in the absence of CPRG. Subsequently, CPRG and cycloheximide were added simultaneously. The postponed addition of CPRG avoided synergistic interference with sample-induced ER activation. The cycloheximide inhibited ribosomal peptidyltransferase activity, which catalyzes the formation of peptide bonds during translation (Cooper and Bossinger, 1976). Thus, cycloheximide blocked any further b-galactosidase production. These modifications resulted in a significantly lower background b-galactosidase activity for the negative controls. As a consequence, the dose–response in the modified assay was increased when compared with the original assay (De Boever et al., 2001). In addition, the initial cell density was raised. Thus, the dose–response curve was obtained after an exposure phase with E2 for 24 h incubation and an additional 18 h incubation with the chromogenic substrate CPRG in the redesigned assay as compared to 3 days of incubation required for the original protocol. One of the most time-consuming steps of the original YES protocols was the release of the expressed reporter enzyme b-galactosidase from within the cell into the CPRG-containing substrate solution. A combination of an enzymatic and a chemical digestion step was introduced, in order to enable a faster conversion of CPRG to chlorophenol red (Schultis and Metzger, 2004). A higher permeability of yeast cells was achieved by lysing the cell membrane with the enzyme lyticase from Arthrobacter luteus (LYES assay). The LYES assay offered a substantially reduced incubation time of 7 h as compared to 3–5 days for the unmodified protocol. In addition, the EC50 value of the LYES assay was reduced by one order of magnitude as compared to the YES assay (Schultis and Metzger, 2004). The cell disruption in the LYES assay demanded some additional handling steps in the analysis procedure which

constituted a potential source of error. Although accelerated compared to initial YES protocols, the disruption was still considered as a time-consuming procedure. A solution to this inherent problem of classical enzyme-based reporter assays was suggested by implementing the green fluorescent protein (GFP) instead of b-galactosidase as a reporter gene (Bovee et al., 2004; Beck et al., 2005). GFP was derived from the jelly fish Aequorea victoria and emits green fluorescing light which can be measured directly without cell disintegration (Tsien, 1998). Furthermore, the chromophore of GFP is formed by intramolecular cyclization and subsequent dehydrogenation without adding any cofactors (Heim et al., 1994). Additional benefits of GFP are related ro the small molecular size, high solubility, and stability in a broad range of pH values (Zimmer, 2002). In order to create a GFP-reporter plasmid, Beck et al. (2005) inserted the coding sequence for red-shifted GFP (rsGFP) into the plasmid backbone of YRpE2 lacking the coding sequence of the lacZ gene. Fluorescence of the cell suspensions had already been measured after 4 h incubation of E2. A clone with a highly induced fluorescence level and a low basal GFP expression was selected from a panel of various yeast clones. The sensitivity for E2 of the GFP-based transactivation assay developed thereof was in the same range as for conventional YES. Furthermore, the potencies of various substances were reproducible in the yeast transactivation assay, independent of the reporter plasmid employed. However, the absolute range between baseline and plateau of the dose–response curve was higher in the yeast assay using b-galactosidase as a reporter than in the GFP assay.

3.09.3.4.3 Subtype YES Estrogens control transcriptional responses through binding to two different NRs, ERa and ERb, which share a similar architecture (Koehler et al., 2005; Harris, 2007). The carboxylterminal domain of the receptors is crucial for ligand binding, nuclear translocation, receptor dimerization, and modulation of target gene expression associated with coregulators (Tsai and O’Malley, 1994). This domain shows a mere sequence homology of 58% between ERa and ERb in contrast to the high similarity of the amino-terminal region (Mosselman et al., 1996). The sequence variability of ER subtypes is the basic cause for the respective binding affinities of estrogenic ligands and their variable agonistic transcriptional activities (Hermenegildo and Cano, 2000). In order to identify subtype-selective ER ligands, a bipartite recombinant yeast screen (BRYS) was developed by Liang et al. (2009). This subtype-selective assay involves the reporter-gene plasmid (YRp2ERE) and the ER expression plasmids (YEp– hERa or YEp–hERb). BRYa was constructed by YRp2ERE and Yep–hERa, whereas BRYb was constructed by YRp2ERE and YEp–hERb. As BRYa or BRYb each comprises exclusively one kind of ER, the ER subtype selectivity of the ligands can be elucidated. Propyl pyrazole triol (PPT) and diarylpropionitrile (DPN) are subtype-selective ER ligands. In the transcription assays, PPT and DPN displayed obvious ER subtype selectivity. PPT showed ERa preference. The b:a ratio of relative potency

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

(defined as 100  EC50(estrogen)/EC50(ligand)) was 0.67 and the b:a ratio of relative efficiency (defined as 100  (A2 – A1)(ligand)/(A2 – A1)(estrogen)) was 0.36. In contrast, DPN demonstrated ERb selectivity with a 21.57-fold ERb potency. The phytoestrogen genistein showed strong ERb potency selectivity with b:a ratio of 2710.54. PPT, on the other hand, preferentially activates ERa. The ratio of relative efficiencies and potencies between BRYa and BRYb was nearly threefold and 1.5-fold, respectively. DPN was shown to be an ERb selective agonist. The b:a ratios in relative potencies of DPN was 21.57. The ER subtype selectivity of both PPT and DPN was lower than those in 293 human embryonal kidney cells (Meyers et al., 2001) or HELN cell line stably expressing ERa or Erb (Escande et al., 2006). Engagement of coregulators present in mammalian cells may make the ER subtype selectivity more obvious. However, it is most likely that the compound and the ER subtype are the major factors to determine the agonistic or antagonistic effects. Thus, the BYRS could yield at least valuable hints on estrogenic ligand subtype selectivity (Liang et al., 2009). Coactivator YES. The ER subtype ERb differs from ERa with respect to tissue-specific expression, ligand selectivity and affinity, and selectivity in recruiting coactivators. For example, the affinity of ERa and ERb for NR boxes of pl60 coactivators is significantly influenced by the ligand that is bound to the ERs (Bramlet et al., 2001). Although the ligand-dependent activation level of reporter genes with ERE on their promoter is generally lower with ERb than ERa, the increase in their activation by ERb is larger than that by ERa (Routledge et al., 2000). By selectively recruiting coregulators to ERb, isoflavone phytoestrogens only activate ERb-mediated transcription pathways (An et al., 2001). Increasing understanding of the transcriptional regulation by estrogens suggests the necessity of detection systems for EDs providing enhanced accuracy, in which ERb and a suitable coactivator are employed. Otherwise, some chemicals that might affect estrogen systems could be missed. Lee et al. (2002) developed a yeast two-hybrid system in order to integrate the estrogen-dependent interaction between the hER ligand-binding domain (hER LBD) and various coactivator nuclear receptor-binding domains (NRBDs). For this purpose, the genes coding for GAL4–DNA binding domain (DBD) fused to hER LBD, and GAL4-transactivation domain (TAD) fused to NRBD were expressed from vector plasmids in S. cerevisiae strain YRG-2, which has a reporter construct on its chromosome. When the two fusion proteins associate with each other through the ligand-dependent interaction between hER LBD and NRBD, the GAL4TAD recruits the basal transcriptional machinery to the CYC1 promoter, resulting in the production of b-galactosidase. The b-galactosidase activity, therefore, reflects the strength of the interaction between hER LBD and NRBD, or the ability of estrogen-like chemicals that induce hER LBD association with NRBDs of the coactivators (Nishikawa et al., 1999; Peters and Khan, 1999; Sheeler et al., 2000). Six combinations of LBDs of the two hERa and b species and the three coactivators TIF2, SRC1, and ATB1 were tested by Lee et al. (2002) using the yeast two-hybrid system described above. The ligand specificity of the two most effective combinations of hER LBD and coactivator NRBD were

205

analyzed by examining the effect of a variety of natural and synthetic steroids and phytoestrogens. For the combination of hERa LBD and TIF2 NRBD, E2 was most effective among the chemicals tested at 1010 M. The order of effectiveness was E24DES4estrone4coumestrol4genisteine4testosterone. In the case of hERb LBD and SRCl NRBD, the order of effectiveness of the above chemicals was similar, although DES produced higher b-galactosidase activity than E2 at 1010 M. This latter combination gave higher b-galactosidase activity for the chemicals tested and seemed more sensitive than the former combination. The lowest concentration that gave detectable b-galactosidase activity was 1010 M for DES, l08 M for estrone, 107 M for coumestrol, 106 M for genistein, and 106 M for testosterone. These concentration levels were one to two orders lower than those for the combination of hERa LBD and TIF2. The latter two-hybrid system that employed hERb LBD and SRCl NRBD was more sensitive to estrogenic chemicals than the former. The difference in ER subtypes seemed to be primarily influential, because the selectivity and affinity of their ligand-dependent association with coactivators are different (Mosselman et al., 1996; Ogawa et al., 1998). In the presence of xenoestrogens, ERb bound to coactivators SRC-la and TIF2 at much lower concentrations and potentiated reporter-gene activity more effectively than ERa in transiently transfected HeLa cells expressing SRC-le and TIF2 (Routledge et al., 2000). ERb showed a 30-fold enhanced binding affinity to genistein compared with ERa (An et al., 2001). Isoflavone phytoestrogens repressed the expression of a tumor necrosis factor-alpha (TNF-a) promoter region through the action of ERb but not ERa, although E2 represses this promoter more potently by binding to ERa (An et al., 2001). ERb has different binding affinities than ERa for the NR boxes of coactivators, which varied depending on the estrogens employed (Bramlet et al., 2001). Ligand-bound ERb but not ERa showed an especially high affinity for NR box IX of SRC-la, a longer splice variant of SRC-1. Thus, these observations suggest that the interactions of ERb with estrogenic chemicals were not the same as those of ERa. ERb tended to bind to coactivators SRC-la and TIF2 at much lower xenoestrogen concentrations (Routledge et al., 2000). The estrogenic activities of alkylphenols reportedly depended on their structural features in both, the position (from high to low, para4 meta4 ortho) and branching (tertiary4 secondary ¼ normal) of the alkyl groups joined to a phenyl ring (Edwin and John, 1997). In agreement with these results, Lee et al. (2002) found that 4-tert-OP has higher estrogenic activity than 4-n-OP, but that a number of 4-NP hydrocarbon isomers do not give higher activity than 4-NP, probably because of insolubility. In their two-hybrid system, a-naphthol and b-naphthol showed detectable activities. Indole, a natural metabolite of tryptophan that has not been classified as positive, was positive in the coactivator system. Polycyclic aromatic hydrocarbons such as phenanthrene and naphthalene were judged to be negative in this system, and showed only weak induction of b-galactosidase. The observed differences in the response to these aromatic chemicals probably reflect the structural differences between the two types of hER LBDs. Generally, these studies indicated that a carefully optimized combination of receptor subtype and coactivator could

206

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

enable the detection of estrogenic activity in some chemicals that were not suspected of being positive (Lee et al., 2002).

3.09.3.4.4 ER mutants Subsequent to the initially established principle of the YES assay, several approaches have been pursued, in order to modulate this amazingly generic strategy for answering basic scientific questions or meeting practical requirements. Several researchers have focused on the molecular modification of the receptor itself aiming on altered functionality or ligandbinding capability. Most ligand–ERa structure–function experiments to date have applied directed mutagenesis to produce large mutant ERa libraries, which were then screened to identify ERa variants with altered ligand binding and activation profiles (Chen and Zhao, 2003; Jakacka et al., 2002; Montano et al., 1996). Similarly, directed evolution was applied to generate mutant libraries, which were screened for optimized variants using selection-based methods (Chen et al., 2004; Sitcheran et al, 2000). Affinity mutants. A panel of screening methods was established in order to isolate hERa mutants with altered transactivation potency. Initially described genetic screens of yeast have tended to isolate constitutively active C-terminally truncated proteins which lacked functional LBDs (Pierrat et al., 1994; Vegeto et al., 1992). This is due to the fact that the N-terminal AF-1 domain is the major transactivation domain in yeast. In order to eliminate the isolation of these constitutively active truncated mutant ERs, a rearranged reverse ER was developed, in which the mutagenized LBD was placed at the N-terminus of the receptor in order to restrict the possible genetic alteration of the LBD. Mutations within the LBD of the reverse ER, which generated stop codons, resulted in truncated proteins lacking a DBD. They were therefore transcriptionally inactive in the screening process. For this reason, yeast cells were cotransformed with a reporter plasmid containing two or three copies of ERE upstream of the reporter gene lacZ and an ER expression plasmid (Wrenn and Katzenellenbogen, 1993; Whelan and Miller, 1996; Eng et al., 1997). After generating a library of ER variants, the mutants were grown on selective agar plates containing the ligand of interest. The transactivation activity of ER was revealed by the b-galactosidase activity, which could be assayed using the chromogenic substrate X-gal (5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside). A second screening method entailed the fusion of the ER to the GAL4 DBD, which interacts with the GAL4 upstream activating sequence (UAS) located upstream of an integrated GAL1-lacZ gene (Bush et al., 1996). The addition of a ligand induced the expression of the lacZ gene and could be again assayed using X-gal. However, both assays were described to suffer from moderate sensitivity and low throughput because they required multiple handling steps (Chen and Zhao, 2003). The third example for an effective screening strategy was based on the fusion of the LBD of hER.a to murine dihydrofolate reductase (DHFR) providing temperature-sensitive stability (Tucker and Fields, 2001). The association of an estrogenic compound with the LBD increased the stability or activity of the murine DHFR. This in turn resulted in increased cell growth. Although this method was sensitive and amenable to high-throughput screens, it seemed to mainly screen for

increased ligand affinity because, for example, the phytoestrogen genistein was not active in this system. The fourth approach providing a sensitive genetic screen amenable to high-throughput isolation of hERa mutants was based on the linkage of the transactivation activity of ER to the cellular growth rate (Chen and Zhao, 2003). In this system, a minimum promoter of the reporter plasmid produced a very low level of constitutive expression of the HIS3 gene: the HIS3 gene product (His3p) is an essential enzyme in the histidine biosynthesis pathway. In the presence of ligand, high-level expression of the inserted HIS3 gene was activated, enabling colony growth on histidine-deficient minimum medium. A library of variants produced by error-prone PCR in the region of LBD of hERa was used to isolate hERa mutants with altered transactivation activity. Three mutants showed increased response to E2 during agar plate screening. Based on its halfmaximal concentration, the best mutant was determined at 1.0 1011 M E2, which corresponds to 100-fold increased sensitivity to E2 as compared to the wild-type hERa. Selectivity mutants. A completely different approach of mutation experiments targeted on the alteration of the ligandbinding specificity of the ER. Thus, a new type of receptor could be generated, which enabled the analysis of target groups or combinations of different hormonal reactive compounds that are not bound by the wild-type protein. Besides this, the corresponding experiments shed light on molecular details to gain an understanding of receptor structure. For example, the ER and the AR selectively bind their physiological ligands with subnanomolar affinity. Although the chemical structures of E2 and testosterone differ only in the A-ring region, the affinity of hER. for testosterone is 10 000-fold weaker than that for E2 (Chen et al., 2004). However, the affinity of hAR for E2 is 44-fold weaker than that for testosterone (Toth et al., 1995). The reason for this strict discrimination can be deduced from crystallographic structures of the hERa LBD complexed with E2 and the human AR LBD (hAR LBD) complexed with testosterone. These studies revealed that the majority of the residues interacting with the ligand (14 out of 20) are different between the ER and the AR. Despite their low sequence homology (o20%), the ER LBD and the AR LBD share a similar structural motif consisting of 12 a-helices arranged in an antiparallel sandwich motif. Chen et al. (2004) altered the ligand-binding characteristics aiming on enhancement of the hER binding affinity toward testosterone. They applied error-prone PCR to introduce random point mutations into the LBD fragment and the yeast two-hybrid system to select ER variants with altered transactivation activity. In this system, almost the entire LBD domain together with the F domain was fused to the DNA sequence encoding the DBD of yeast transcription factor GAL4 in order to create the bait plasmid (pBD-GAL4 hERa). The corresponding prey plasmid encoded the human SRC-1 fused to the gene encoding the GAL4 activation domain (pGAD424 SRC1). In the presence of agonistic ligands, the LBD undergoes a conformational change and binds to SRC-1, which brings the GAL4 DBD and the GAL4 activation domain into close proximity. This in turn activates the transcription of the HIS3 reporter (cf. above). On minimal medium lacking histidine, the cell growth rate is proportional to the strength of the ligand–receptor interaction. Two rounds of directed evolution

Bioassays for Estrogenic and Androgenic Effects of Water Constituents were performed with a cutoff concentration of 108 M testosterone in the medium. The EC50 values of the final mutants for testosterone were thus increased 234–780-fold. The binding affinities of the variants were determined by direct and competitive hormone-binding assays to establish whether the improved transactivation potencies of these evolved hERa variants toward testosterone were the result of the corresponding changes in ligand-binding affinities. The variants exhibited affinities to testosterone in the nanomolar range (up to 38 nM corresponding to 7600-fold improvement over that of the wild-type hERa). Simultaneously, the variants had increased Kd values toward E2, ranging from 0.44 to 3.53 nM (Chen et al., 2004). Unlike the naturally occurring ARs or ERs, these in vitro evolved hERa variants represent promiscuous receptors for estrogens and androgens. Such dual-ligand class receptors could constitute the basis of new approaches in the screening of hormone-reactive compounds. By applying the same principle, new receptors could be tailored for the simultaneous detection of theoretically any combination of bioeffective compounds. Expanding this principle of protein modification, the ER was used as an evolvable template structure to design a functionally orthogonal gene switch to accommodate a variety of nonsteroidal compounds in its binding pocket, which do not activate the wild-type receptor (Chockalingam et al., 2005). Modification of the binding pocket was achieved by saturation mutagenesis. This method involves the replacement of each amino acid residue that is expected to contact the ligand with the 19 possible alternative amino acids. The generated receptor mutants were selected for binding to a synthetic ligand using the yeast two-hybrid screening system, similar to that described above, by monitoring cell growth on histidine-deficient minimal medium. Subsequently, the site-saturation mutagenesis was repeated with the resulting mutant. The ligand specificity could be further enhanced by random point mutagenesis of the LBD and phenotypic screening. Using the synthetic nonsteroidal compound 4,4’-dihydroxybenzil (DHB), a mutant hER.a that showed a 50-fold enhanced binding activity to DHB and a 140-fold reduced affinity to its natural ligand E2 was selected after two rounds of mutagenesis. After three further rounds of molecular evolution, a mutant receptor was identified that showed a fivefold enhanced binding activity to DHB and an over 106-fold reduced affinity to E2 when compared with the wild-type protein. Thus, the finally achieved affinity was enhanced by a factor of 107. Applying an analogous strategy Islam et al. (2009) created a novel receptor–ligand pair, which responded to concentrations of a synthetic ligand that did not activate the native receptor. The aromatic A-ring of the synthetic ligand carried the hydroxyl group in a position analogous to the hydroxyl group in E2. The D-ring being either a five-or a six-membered ring carried a keto-group similar to the E2 homolog estrone. Iterative rounds of site-specific saturation mutagenesis of a fixed set of ligand-contacting residues were performed, followed by random mutagenesis of the entire LBD. The affinity of the ligand to the wild-type receptor was 3.7 mM as determined by its effective concentration to promote half maximal growth (EC50) of the yeast reporter strain (Chockalingam et al., 2005). After three rounds of mutagenesis, the affinity of

207

the selected mutants was enhanced up to 0.055 mM. Thus, the resulting receptor mutant showed a 67-fold increased activity to the synthetic ligand. In contrast to directed evolution methods as described above, experimental evolution systems were expected to be even more efficient for generating useful receptor variants, because the latter involved the continuous generation of new ERa variants and benefited from the power of natural selection in large microbial populations. Experimental evolution assays require the organism’s fitness to be linked to the trait of interest being investigated. When this is accomplished, small variations in that trait, generated by spontaneous mutation in a large population, confer small differences in growth rate and fitness for individual yeast. With increasing number of generation cycles, yeast inheriting fitness-improving alleles rise in frequency and ultimately dominate the yeast population over time. The advantages of carrying out such experiments in yeast include their short generation time, well-characterized genome, and consequently, the feasibility to analyze the genetics responsible for new phenotypes during the course of experimental evolution. Corresponding investigations proved that significant differences in growth rate and fitness are induced by minute (o 10 pM) differences in ligand (E2) concentration added to the media. This indicated that variants with even subtle improvements in ERa signaling should be subjected to relatively strong selection. Experimental evolution studies in yeast have shown that small differences in fitness (o 2%) can be selected for and isolated from large populations of variants within a few hundred generations (Thatcher et al., 1998; Zeyl, 2005).

3.09.3.4.5 Extension of the YES principle The conversion of androgens to estrogens is catalyzed by the enzyme aromatase. Some ED may act in an indirect fashion, for example, by inhibiting aromatase activity. This would result in a decreased level of estrogens or an increased androgen concentration in the cell. Aromatase plays an important role in the expression of secondary sexual characteristics, maintenance of pregnancy, and brain sex differentiation (MacLusky et al., 1987). Furthermore, the enzyme has received considerable attention in breast-cancer development because high expression of the enzyme activity has been associated with a significant number of breast tumors (Miller and O’Neill, 1987; Silva et al., 1989; Reed et al., 1989). Therefore, aromatase inhibitors are administered as therapeutic agents for the treatment estrogen-dependent breast cancer by inhibiting estrogen production concomitantly to antiestrogen treatment. Phytoestrogens such as flavones are competitive inhibitors of aromatase (Kao et al., 1998; Wang et al., 1994). Some flavones can inhibit aromatase with Ki values similar to that of aminoglutethimide, an approved drug for treating breast cancer. The conventional method for aromatase inhibitor screening was based on an in vitro enzyme assay with human placental microsomes as source for the enzyme. However, the aromatase assay in such preparations depends on several components including short incubation time, the stability of NADPH–cytochrome P450 reductase, and aromatase content. Furthermore, the assay procedure is tedious and it would not be easy to adapt this assay for high-throughput screening.

208

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

Although certain intact cells expressing aromatase, including human aromatase transfected mammalian cell lines (Zhou et al., 1990), provided an ideal in vivo model for aromataseinhibitor screening, the use of radioactive-labeled substrates was a disadvantageous feature of the assay. In contrast, the aromatase inhibitor screening method developed by Mak et al. (1999) relied on the coexpression of AR and aromatase in yeast cells carrying the androgen-responsive reporter plasmid (Figure 6). If the triple transformant yeast cells carrying the three plasmids were incubated with an aromatizable androgen (testosterone or androstenedione), the androgen diffused into the cell, where it either bound to the yeast-produced AR or was converted to estrogen by aromatase produced in yeast. If the conditions (concentrations and choice of substrates) favored the enzymatic reactions, most of the androgen was converted into estrogen but not bound to the AR. Therefore, it could not transactivate the yeast basal promoter linked to the androgen-responsive element (GRE/ PRE). The androgen metabolism within the yeast cells finally culminated in the inhibition or reduction of reporter-enzyme b-galactosidase induction. However, androgen-dependent transcriptional activation was apparent as reflected by the reporter-enzyme induction if an aromatase inhibitor was included in the yeast medium. Applying this system, aromatizable androgens such as androstenedione and testosterone were not able to induce the reporter enzyme in the absence of the aromatase inhibitor aminoglutethimide. However, ligand-dependent transcriptional activation was apparent in the presence of aminoglutethimide (1 mM). In contrast to aromatizable androgens, the nonaromatizable androgen, 5a-DHT, induced the reporter gene efficiently even in the presence or absence of aminoglutethimide. Thus, the system essentially enables to monitor environmental chemicals for their antiaromatase activity and for their interaction

with AR. Furthermore, it discriminates nonandrogenic aromatase inhibitors from inhibitors with androgenic activity. Single-gene studies provide important but limited information about biological systems, primarily, because most genes function as part of gene networks and molecular pathways. Fox et al. (2007) designed a generic and modular geneintegration strategy, which enabled to monitor the interaction of several components of the ERa signalling network. This evolvable ERa activity sensor (EERAS) yeast strain expressed hER.a and three ERE-driven genes that were installed as reporters of transcriptional activity induced by the network. The entire system was based on a recyclable vector with modular components to direct integration of multiple genes of interest to any target loci in the S. cerevisiae genome. The resulting EERAS strain contained a constitutively expressed ERa gene, an ERE-driven fluorescent reporter gene (ERE-yEGFP), and two ERE-driven metabolic genes (ERE-HIS3 and ERE-URA3), which were required for growth in medium deficient of histidine and uracil (Figure 7). The yEGFP activity in the EERAS strain emerged as an efficient and sensitive dose-responsive reporter of ERa signalling activity, reaching approximately sixfold activation above background at 4 h posttreatment, with an EC50 of 3.9 1010 M for E2 (Bovee et al., 2004). E2dependent growth was measured in synthetic drop-out (SD) media lacking histidine and uracil, respectively, to validate the functionality of the ERE-driven HIS3 and URA3 genes, and to determine whether growth of the EERAS strain depends on ERa signaling. The growth-related sensitivities were even higher than for the yEGFP reporter assay: EC50 values for growth in SD-His and SD-Ura were 4.7 1011 M and 7.8 1011 M, respectively. In SD-His-Ura medium, in which both ERE-driven genes are required for growth, the EC50 was 1.1 1010 M. The EERAS generally delivered sensitive doseresponsive GFP and growth curves over a broad range of

Aromatase gene Aromatase

ARE

AR AR gene

AR ARE

Androgen

Reporter gene

Reporter gene

Reporter protein

E2 Androgen receptor (AR) Gene expression

Androgen response element (ARE)

Figure 6 Principle of aromatase inhibitor screening by reporter assays based upon aromatase-catalyzed transformation of E2 to androgen. See the text for detailed explanation.

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

ER ERE

ER

Fluorescence GFP gene

ER

ER gene

ERE

URA3

Cell growth in uracil-deficient medium

HIS3

Cell growth in histidine-deficient medium

URA3 gene

ER ERE

209

HIS3 gene

Fluorescent protein

E2 Estrogen receptor (ER)

URA3 protein

URA3

HIS3 protein

HIS3

Estrogen response element (ERE) Gene expression

Figure 7 ERa gene interaction network controls EERAS yeast strain growth and yEGFP production. The EERAS yeast strain was engineered to express multiple reporter genes that control growth and yEGFP production in a ligand–ERa–ERE activity-dependent manner and to constitutively express the hERa gene. When no ligand is added, ERE-HIS3, ERE-URA3, and ERE-yEGFP reporter genes are not transactivated by ERa, yeast are unable to grow in synthetic dropout media lacking histidine or uracil, and yEGFP is not produced. When ligand is added to the media, ERa is activated, resulting in dose-responsive yeast growth and yEGFP production.

ligand concentrations, enabling the detection of, and differentiation between, ligands with strong, moderate, and weak activation of ERa in EERAS. EC50 values for ligands were comparable to, and in most cases more sensitive than, those reported by other yeast ERa reporter assays (Bovee et al., 2004; Fang et al., 2000; Soto et al., 1995).

3.09.4 Subcellular Assays In addition to tests based at the organism and cellular level, subcellular receptor-binding assays with ER preparations were used to screen chemicals and environmental samples for estrogenic activity. Measurement of ligand binding (e.g., xenoestrogen) to ER is performed by competitive assays with, for example, radioactive- or fluorescence-labeled E2 (e.g., Kuiper et al., 1997; Mueller et al., 2003). The implication is that binding triggers subsequent biological effects. The assays enabled to establish quantitative toxicity equivalents combined with a high level of reproducibility. Since these assays benefit from easy handling and comparably higher speed than cellular or organismic test systems, they are considered to match fundamental prerequisites for high-throughput screening of environmental samples.

green anole (Anolis carolinensis), and for the corresponding receptor rtER from rainbow trout (Onchorhynkiss mykiss). Alternative expression systems for ER preparations were based, for example, on baculovirus using Sf9 insect cells (Bolger et al., 1998). In our group, the yeast expression system of McDonnell et al. (1991) was used for the recombinant production of ER. The proteinase-deficient yeast strain BJ3505 was transformed with the expression vector pYEPE-10 containing the ER gene fused to an ubiquitin gene. Expression was controlled by the coppersensitive metallothionein promoter CYP. Addition of CuSO4 (1 mM) to the culture medium induced the production of the fusion protein, from which ubiquitin was enzymatically cleaved off after a short time. Figure 8 illustrates the timedependent expression of recombinant ER protein. The Western blot shows the appearance of a single band with a molecular weight of 66 kDa, the size of the hERa. A maximum yield of approximately 4–7 pmol ER mg1 total receptor protein was obtained after 16 h of expression. Longer expression periods resulted in a higher amount of total expression product but diminished percentage of functional ER. Subsequent purification of the ER was carried out by affinity chromatography on heparin-agarose, which is known to bind several DNA-binding proteins such as the ER (Hock et al., 2002).

3.09.4.1 ER Preparation Substantial amounts of the hER. are required for binding assays. A number of natural cell lines was used in order to obtain the receptor protein. However, bacterially expressed receptors for high-throughput testing have been developed for the hERa, for a reptilian receptor (aER) from the liver of the

3.09.4.2 Enzyme-Linked Receptor Assay The enzyme-linked receptor assay (ELRA) essentially employs the same principles as competitive immunoassays based on antibody–antigen interactions. The decisive reaction is the competitive binding of the estrogenic sample compound and

210

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

66 kDa

1

4

8

16

Commercial ERα

Hours after CuSO4 induction Figure 8 Western blot of human estrogen receptor expressed by recombinant yeast. From Seifert M (2000) Bestimmung von O¨strogenen und Xenoo¨strogenen mit einem Rezeptorassay. PhD Thesis, Technische Universita¨t Mu¨nchen.

1

2

ER

E2−BSA conjugate Substrate 3

Product

4

Anti-ER antibody (biotinylated) Biotinylated horseradish peroxidase

Streptavidin

Figure 9 ELRA principle (1) E2-BSA is adsorbed to the individual wells of microtiter plates. (2) E2 standards and samples are incubated together with the ER. (3) Addition of biotinylated mouse antihuman estrogen receptor antibody to label bound ER. (4) Addition of a streptavidin–biotin enzyme system. Finally, the substrate solution is incubated and the resulting signal measured spectroscopically (not shown). Adapted from Alberti MC (2006) Erfassung und Bewertung von Genexpressionsmustern von Zebraba¨rblingen (Danio rerio) nach Belastung mito¨strogenen Substanzen. PhD Thesis, Technische Universita¨t Mu¨nchen.

the immobilized E2-derivative at a limited number of ER molecules. Essential steps of the procedure are shown in Figure 9. In order to improve the sensitivity of the ELRA, the chromogen substrate tetramethylbenzidine was replaced by the luminescent substrate luminol (Seifert, 2004). This modification significantly lowered the detection limit of the assay. Consequently, the higher sensitivity of the luminescent detection system allowed higher dilutions of several assay

components as well as the dilution of the samples. This in turn lowered the potential interference with the sample matrix. The chemiluminescent ELRA reached a detection limit below 20 ng l1 compared to 100 ng l1 for the chromogen ELRA. Figure 10 shows the calibration curve of the ELRA for E2. An inverse relation between enzymatic activity and effect concentration (E2) can be observed because of the competitive assay principle. The IC50 values derived from these calibration

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

211

Therefore the polarization values are reduced (Schultis et al., 2002; Bolger et al., 1998).

1200 1000

3.09.4.4 Biosensors RLU

800 600 400 200 0

0.001 0.01

0.1

1

10

100

1000

−1)

17β-Estradiol (μg l

Figure 10 Calibration curve obtained with the luminescence ELRA for E2. RLU, Relative luminescence units. From Seifert M (2000) Bestimmung von O¨strogenen und Xenoo¨strogenen mit einem Rezeptorassay. PhD Thesis, Technische Universita¨t Mu¨nchen.

curves were used to determine the percentage of cross-reactivity, with E2 arbitrarily set at 100%. These cross-reactivities represent effect concentrations. In addition, the competitive binding assay provides relative binding affinities compared to positive controls (usually E2 or DES) and therefore enables the ranking and prioritization of several compounds for subsequent studies. Binding assays for other receptors, such as the ERb and AR, have also been developed (Kelce et al., 1994). Compared to the cell-based assays, the ELRA has one main disadvantage: although compounds may have bound to the receptor, the tests do not distinguish between subsequent agonistic or antagonistic effects (Holmes et al., 1998; Zacharewski, 1997). An agonist binds to a cellular receptor and triggers the response by mimicking the effects of the natural molecules in cells. In contrast, an antagonist inhibits receptor action by competition with the natural molecule or by interaction with other sites in the receptor. However, if the agonists and antagonists are present simultaneously in a sample, the ELRA delivers a correct sum parameter for receptor-binding substances (Seifert, 2004). As binding to the receptor does not necessarily reflect ER activation, a prioritization of endocrine-active compounds can not be solely based on binding assays. In contrast to bioassays, the interpretation of synergistic effects is limited. However, installed as a primary screening, ER-binding assays are considered as appropriate tools to decide whether or not further investigations are required employing more informative assay schemes as provided at the cellular or organism level.

3.09.4.3 Fluorescence Polarization Assays Receptor-binding assays with hER based on fluorescence polarization offer the benefit of fast and direct measurement even in the presence of unbound ligand (Checovich et al., 1995). A separation step for the removal of unbound ligand such as in the ELRA is not necessary. In these assays, estrogenic substances displace the fluorescent ligand from a slowly tumbling ER–ligand complex. With increasing concentration of a competing compound, more fluorescent ligands are displaced, which now unbound in solution tumble more rapidly.

In ER-based biosensors, the hER. was preferentially used as sensing element so far (Habauzit et al., 2007). The primary objective of sensor techniques was initially to evaluate the binding properties between the protein ER and different estrogenic compounds rather than sample measurement. In contrast to the broad range of different biosensor principles reported (optical, electrochemical, acoustic, etc.), the molecular formats applied were of intriguing similarity to the ERbased microplate assays (cf. above). The various formats can be assigned either to direct or to indirect interaction analysis between the ER and an estrogenic compound. Indirect binding assays are based on the competition between estrogenic test compounds and the E2 derivative covalently linked to the sensor surface by chemical (pentanediamine, obtained from an amide derivative of 17bestradiol-17-hemisuccinate and 1,5-pentanediamine; Usami et al., 2002) or biological spacers (BSA coupled with estradiol; Miyashita et al., 2005; Pearson et al. 2001). The covalent coupling to the sensor chip surface is mediated, for example, in the optical sensor BIAcore through carboxymethylated dextran. The ER functions a recognition element in the competition between the tested molecules and E2 bound at the sensor surface. Then a stream containing the ER is directed over the surface (Figure 11). In the absence of analyte, the ER is maximally bound to the immobilized E2. With increasing analyte concentrations, decreasing amounts of the ER are attached. The dissociation constant Kd of 2.3 1010 M deduced from those experiments (Seifert et al., 1999) corresponded very well with the values reported by other groups, where Kd values ranging from 4.2 1010 to 2.0 1010 M were determined with radiolabeled E2 (Olea et al., 1985). In direct binding assays, the ER-LBD fragment is bound on injection by an anti-ER antibody, which is covalently immobilized on the sensor chip surface. After addition of estrogenic compounds, the association to and dissociation from the ERLBD can be monitored online without any competition included (Rich et al., 2002). Another example based (as in the case of the BIAcore instrument) on surface plasmon resonance (SPR) sensors is the portable immunosensor described by Sesay and Cullen (2001) for the detection of hormone mimics. Murata et al. (2001) proposed a bioaffinity sensor based on the specific binding of estrogens to the receptor immobilized on a gold disk with cyclic voltammetry detection. This biosensor was applied for the detection of E2. There are, however, difficulties in terms of reproducibility of direct binding assays, because of general instrumental limitations of the most frequently applied SPR-based biosensors for the detection of small molecules. In addition, the affinity constant of the ER–LBD has been reported to be not identical as compared to the entire receptor protein (Haubitz et al., 2007). The complexation of E2 with spacer molecules for competition experiments alters the interaction properties of the molecules and their ER accessibility (Usami et al., 2002; Miyashita et al., 2005; Pearson et al., 2001). In addition, the mere Kd information is not sufficient to determine the

212

Bioassays for Estrogenic and Androgenic Effects of Water Constituents 3000 2500

Association phase

2000 1500 Dissociation phase RU

1000 500 0 −500 −1000 −1500 11 500

12 000

12 500

13 000

13 500

14 000

14 500

15 000 s

Time (s) Figure 11 Typical sensorgram obtained with ER and immobilized E2 derivative in a surface plasmon resonance (SPR) biosensor for the calculation of kinetic data. Symbols as explained in Figure 9. RU, response units. From Seifert M (2000) Bestimmung von O¨strogenen und Xenoa¨strogenen mit einem Rezeptorassay. PhD Thesis, Technische Universita¨t Mu¨nchen.

agonistic or antagonistic and the SERM effects (above) in organisms. Therefore, the Hill number can be used to determine agonist or antagonist activity. A Hill coefficient greater than 1 unity is indicative of cooperative action between the ligand and the receptor and a Hill coefficient less than 1 unity is indicative of negative cooperation. Despite this, properly designed SPR assays shed light on our understanding of these interactions by monitoring the speed of complex formation (association rate constant ka) and the complex stability (dissociation rate constant kd). For example, E2 rapidly (ka ¼ 1.3 106 s1) forms an unstable complex with ER (kd ¼ 1.2 103 mol1 l1 s1), whereas 4-hydroxytamoxifen slowly (ka ¼ 2.3 103 s1) forms a stable complex with ER (kd ¼ 4.1 105 mol1 l1 s1; Habauzit et al., 2007). DNA-binding assay. A few groups extended the simple biosensor-binding formats in order to functionally implement the transactivation activity of EDs. The principle of the detection is that the ER captures the estrogenic ligand, dimerizes, and then the complex binds to the ERE, which is immobilized at the transducer surface of a biosensor. The evaluation of the affinity between ER and the consensus DNA sequence ERE by SPR was introduced by Cheskis et al. (1997). Equilibrium constants and rate constant changes between ER and ERE were determined after a previous incubation with several estrogens. This method has also been applied by other groups for assessment of estrogenic action (e.g., Asano et al., 2004; Murata et al., 2004). A rapid formation of an unstable ER/ERE complex could be observed with E2. Conversely, slow association of a very stable ER/ ERE complex was characteristic for the ICI 182,780 antagonist. These results were validated by testing the binding efficiency of more than 30 endocrine compounds. Variation of both the binding properties and the stability of the ER/ERE complex was observed, depending on the ligand employed (Asano et al., 2004).

SPR-based technology was rarely used for differential characterization of the effect of estrogenic compounds, the activity of which is difficult to investigate by conventional methods. As this approach is complementary to other assays at the organism or cellular level, gene-reporter assays, and protein-based assays, SPR-based biosensors are considered to furnish additional information on the molecular activity of EDs (Habauzit et al., 2008).

3.09.5 Conclusions Endocrine bioeffect assays can be essentially assigned to in vitro and in vivo test systems. Both groups are characterized by individual and inherent pros and cons. Generally, there are a few established short-term in vivo assays that are applicable for the assessment of ED. The above-described uterotrophic and vaginal cell cornification assays represent the most frequently employed in vivo screening tools for assessing the estrogenic potential of substances such as water contaminants. A crucial difference between both the in vivo tests is the fact that vaginal epithelial cell cornification can be induced only by compounds considered to be estrogenic, whereas the uterotrophic assay responds to progesterone or testosterone. These short-term rodent assays were criticized because they may not possess sufficient sensitivity to identify xenobiotics with weak or specific endocrine-disrupting activities. It is conceivable that ED may elicit responses at the geneexpression level that may not be translated into immediate responses at the organ or tissue level but could subsequently predispose an individual or subpopulation to adverse effects at later stages of development. Assessment of ED is further complicated by the fact that many substances elicit species-, tissue-, cell-, and response-specific effects. For example,

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

tamoxifen exhibits antiestrogenic activity in the breast and agonist activities in the uterus. This underpins the necessity for measuring a number of different endpoints in order to comprehensively evaluate ED potency. In order to assess the risk posed by ED to human and wildlife health, it has been further criticized that rodents do not express SHBG following parturition, which is a major determinant of the metabolic clearance and the bioavailability of sex steroids. Finally, in vivo assays are considered to be laborious, costly, and ethically questionable (Zacharewski, 1998). Particularly, the compensation of the latter issues is postulated as a crucial benefit of properly standardized in vitro assays, which are in addition potentially amenable to highersample throughput. Table 1 summarizes some of the pros and cons of the above-described in vitro test systems. These assays are recommended by the United States Environmental Protection Agency (US EPA), Endocrine Disruptor Screening and Testing Advisory Committee (EDSTAC), and the Organization for Economic Cooperation and Development (OECD, 2001). In vitro assays are not only suitable for the screening of ED, but are also acknowledged as powerful tools for mechanistic analysis of their mode of action. The versatility of the mechanistic assays may help to pin down a potential tissue-specific effect of suspected ED (Mueller, 2004). Competitive receptor binding assays provide relative binding affinities compared to positive controls and therefore

Table 1

213

enable a ranking and prioritization of several compounds for subsequent analyses. However, binding to the receptor does not necessarily reflect activation of the hormone receptor. Therefore, a prioritization of endocrine-active compounds should not be based solely on binding assays. Reporter gene assays are suitable screening assays that provide agonistic and antagonistic potencies. Due to the high sensitivity of available luciferase reporter vectors, very weak to highly potent ED can be analyzed. Furthermore, single compounds and chemical mixtures can be analyzed depending on response element, receptor subtype, and cellular context for their agonistic and antagonistic impact. This versatility is especially important for instance in the context of tissue-specific (anti)estrogenic activity of selective ER modulators. In this context, the YES assay is the model of choice for testing environmental samples such as sewage effluents. In contrast to mammalian cell culture, these typically nonsterile samples do not impair the cellular proliferation and ER activity in yeast. Next to these transient ER-expressing systems, tissue-specific cell lines stably expressing the ER can be considered as screening tools. However, as the exogenous receptor is forced into a cell accustomed to the lack of ER, some effects measured may not reflect the physiological response of the analyzed cell type. To account for these limitations, cell lines with endogenous ER expression can be used for transactivation assays. For this purpose, cell lines with endogenous ER expression like human breast-tumor

In vitro assays for the measurement of estrogenic and antiestrogenic compounds according to Mueller (2004)

In vitro assay

Measured endpoint

Advantages

Limitations

E-Screen

Proliferation of ERa-positive cells

No defined ER expression, no mechanistic data

Ligand bindinga

Binding affinity to ERa or ERb

Measures physiological endpoint of estrogen action, measures estrogens and antiestrogens Simple, high-throughput method

ER binding to ERE

Binding affinity of Era or ERb to ERE

High-throughput method, various EREs can be used

(GST pull-down/FRET/) two-hybrid assay

Ligand-dependent association of ERa or ERb with coactivators

Transactivation assay in yeast or mammalian cellsa

ERa- or ERb- mediated activation of reporter

Analysis of gene expression

Expression of ER-regulated genes Activity of ER-regulated enzymes

Analysis of molecular interaction, defined ER subtype or ER domain, as well as coactivators can be used, measures estrogens and antiestrogens High-throughput method, measures estrogens and antiestrogens, can be done in metabolic competent cells to account for (anti)- estrogenic metabolites Analysis of physiological response, versatile, measures estrogens and antiestrogens Analysis of physiological response, measures estrogens and antiestrogens

Induction/inhibition of estrogen biosynthesis

Analysis of physiological response, measures ER-independent pathways

Analysis of enzyme activity

Analysis of steroidogenesisa

a

Does not measure ER activation, does not measure physiological response Does not measure ER activation, low sensitivity, does not measure physiological response Does not measure direct ER activation, low throughput, does not measure physiological response Does not measure physiological response

Low throughput Cell lines or primary cell cultures with active marker enzymes suitable only Cells with active steroidogenesis suitable only

Recommended for screening of xenoestrogens by the United States Environmental Protection Agency (US EPA) Endocrine Disruptor Screening and Testing Advisory Committee (EDSTAC) and the Organization for Economic Cooperation and Development (OECD). Assay principles, advantages, and limitations apply in a similar manner to analoguos sytems for the analysis of other endocrine systems.

214

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

T47 D cells (e.g., ER-CALUX, cf. above) can be employed to measure transactivation on stably integrated or transiently transfected ERE reporter vectors. Similar to these ER activity assays, test systems to measure effects on other endocrine systems, for example, (anti)androgenic activity are used for the analysis of endocrine-active compounds. However, all these in vitro assays allow a ranking of a series of compounds that should be used to prioritize compounds for studies that can detect reproductive and developmental toxicities in vivo. These in vivo studies then enable a valid risk assessment (Mueller, 2004). Therefore, in vitro and in vivo assays are typically combined for tailored monitoring strategies termed as test batteries. Due to time and cost limitations, it is technically not yet feasible to test the ED potential of all chemicals which can be found in ecosystems. However, in silico assays complementing in vitro and in vivo tests as a third component may have the potential to solve this fundamental problem of ED-assessment studies (Escher et al., 2006). There exist different modeling approaches, most prominently SAR and QSAR methods. The procedures are based on the idea that specific structural features of a compound are associated with ED activity. As chemicals including specific structural features are potential binders of a defined hormone receptor, these substructures can be selected as structural alerts for detecting chemicals as EDs in priority-setting programs in which large collections of chemicals are ranked (Shi et al., 2002). However, one of the bottlenecks of the majority of these models is that they deal with structure-binding relationships while EDs intervene on other targets and in several different ways. For example, EDs can speed up the metabolism of hormones or modulate the number of receptors. EDs can also affect natural hormone production by interfering with other signaling pathways. Therefore, advanced modeling approaches consider activity profiles instead of unique endpoints. Multiple endpoint analysis can be made from multivariate analysis (Devillers et al., 2006). Nevertheless, ED modeling is still its infancy and there is a need to develop models in various directions in order to enable the detection and correct assessment of this complex activity.

Acknowledgment The author thanks Martin Alberti and Martin Seifert for providing the figures as indicated.

References Ackermann GE, Brombacher E, and Fent K (2002) Development of a fish reporter gene system for the assessment of estrogenic compounds and sewage treatment plant effluents. Environmental Toxicology and Chemistry 21: 1864--1875. Alberti M, Kausch U, Haindl S, and Seifert M (2005) Gene expression analysis for exposure to estrogenic substances. Acta Hydrochimica et Hydrobiologica 33: 38--44. Alberti M (2006) Erfassung und Bewertung von Genexpressionsmustern von Zebraba¨rblingen (Danio rerio) nach Belastung mit o¨strogenen Substanzen. PhD Thesis, Technische Universita¨t Mu¨nchen. An J, Tzagarakis-Foster C, Scharschmidt TC, Lomri N, and Leitman DC (2001) Estrogen receptor beta-selective transcriptional activity and recruitment of

coregulators by phytoestrogens. Journal of Biological Chemistry 276: 17808--17814. Andersen HR, Vinggaard AM, Rasmussen TH, Gjermandsen IM, and BonefeldJorgensen EC (2002) Effects of currently used pesticides in assays for estrogenicity, androgenicity, and aromatase activity in vitro. Toxicology and Applied Pharmacology 179: 1--12. Arnold SF, Collins BM, Robinson MK, Guillette LJ, and McLachlan JA (1996a) Differential interaction of natural and synthetic estrogens with extracellular binding proteins in a yeast estrogen screen. Steroids 61: 642--646. Arnold SF, Robinson MK, Notides AC, Guillette LJ, Jr., and McLachlan JA (1996b) A yeast estrogen screen for examining the relative exposure of cells to natural and xenoestrogens. Environmental Health Perspectives 104: 544--548. Arukwe A and Goksøyr A (2003) Eggshell and egg yolk proteins in fish: Hepatic proteins for the next generation: oogenetic, population, and evolutionary implications of endocrine disruption. Comparative Hepatology 2: 4--24. Asano K, Ono A, Hashimoto S, Inoue T, and Kanno J (2004) Screening of endocrine disrupting chemicals using a surface plasmon resonance sensor. Analytical Sciences 20: 611--616. Ashby J (1998) The first synthetic estrogen. Environmental Health Perspectives 106: A221. Aurrekoetxea-Hernandez K and Buetti E (2004) Transforming growth factor beta enhances the glucocorticoid response of the mouse mammary tumor virus promoter through Smad and GA-binding proteins. Journal of Virology 78: 2201--2211. Baker VA, Hepburn PA, Kennedy SJ, et al. (1999) Safety evaluation of phytosterol esters. Part 1: Assessment of estrogenicity using a combination of in vivo and in vitro assays. Food and Chemical Toxicology 37: 13–22. Balaguer P, Francois F, Comunale F, et al. (1999) Reporter cell lines to study the estrogenic effects of xenoestrogens. Science of the Total Environment 233: 47--56. Barkhem T, Carlsson B, Nilsson Y, Enmark E, Gustafsson J, and Nilsson S (1998) Differential response of estrogen receptor alpha and estrogen receptor beta to partial estrogen agonists/antagonists. Molecular Pharmacology 54: 105--112. Beck V, Pfitscher A, and Jungbauer A (2005) GFP-reporter for a high throughput assay to monitor estrogenic compounds. Journal of Biochemical and Biophysical Methods 64: 19--37. Beresford N, Routledge EJ, Harris CA, and Sumpter JP (2000) Issues arising when interpreting results from an in vitro assay for estrogenic activity. Toxicology and Applied Pharmacology 162: 22--23. Bergeron JM, Crews D, and McLachlan JA (1994) PCBs as environmental estrogens: Turtle sex determination as a biomarker of environmental contamination. Environmental Health Perspectives 102: 780--781. Biau S, Bayle S, de Santa Barbara P, and Roig B (2007) The chick embryo: An animal model for detection of the effects of hormonal compounds. Analytical and Bioanalytical Chemistry 387: 1397–1403. Bitman J, Cecil HC, Harris SJ, and Fries GF (1968) Estrogenic activity of o,p0 -DDT in the mammalian uterus and avian oviduct. Science 162: 371--372. Blankvoort BM, de Groene EM, van Meeteren-Kreikamp AP, Witkamp RF, Rodenburg RJ, and Aarts JM (2001) Development of an androgen reporter gene assay (ARLUX) utilizing a human cell line with an endogenously regulated androgen receptor. Analytical Biochemistry 298: 93--102. Blom A, Ekman E, Johannisson A, Norrgren L, and Pesonen M (1998) Effects of xenoestrogenic environmental pollutants on the proliferation of a human breast cancer cell line (MCF-7). Archives of Environmental Contamination and Toxicology 34: 306--310. Bolger R, Wiese TE, Ervin K, Nestich S, and Checovich W (1998) Rapid screening of environmental chemicals for estrogen receptor binding capacity. Environmental Health Perspectives 106: 551--557. Bon E, Barbe U, Nu´nˇez Rodrgu´ez J, et al. (1997) Plasma vitellogenin levels during the annual reproductive cycle of the female rainbowtrout (Oncorhynchus mykiss): Establishment and validation of an ELISA. Comparative Biochemistry and Physiology 117B: 75--84. Bovee TFH, Helsdingen RJR, Koks PD, Kuiper HA, Hoogenboom RLAP, and Keijer J (2004) Development of a rapid yeast estrogen bioassay, based on the expression of green fluorescent protein. Gene 325: 187--200. Bramlet KS, Wu Y, and Bums TP (2001) Ligands specify coactivator nuclear receptor (NR) box affinity for estrogen receptor subtypes. Molecular Endocrinology 15: 909--922. Breithofer A, Graumann K, Scicchitano MS, Karathanasis SK, Butt TR, and Jungbauer A (1998) Regulation of human estrogen receptor by phytoestrogens in yeast and human cells. Journal of Steroid Biochemistry and Molecular Biology 67: 421--429. Buelbring E and Burn JH (1935) The estimation of oestrin and of male hormone in oily solution. Journal of Physiology 85: 320--333.

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

Burki R, Vermeirssen ELM, Koerner O, Joris C, Burkhardt-Holm P, and Segner H (2006) Assessment of estrogenic exposure in brown trout (Salmotrutta) in a Swiss midland river: Integrated analysis of passive samplers, wild and caged fish, and vitellogenin mRNA and protein. Environmental Toxicology and Chemistry 25: 2077--2086. Bush SM, Folta S, and Lannigan DA (1996) Use of the yeast one-hybrid system to screen for mutations in the ligand-binding domain of the estrogen receptor. Steroids 61: 102--109. Celius T, Matthews JB, Giesy JP, and Zacharewski TR (2000) Quantification of rainbow trout (Oncorhynchus mykiss) zona radiata and vitellogenin mRNA levels using realtime PCR after in vivo treatment with estradiol-17 beta or alpha-zearalenol. Journal of Steroid Biochemistry and Molecular Biology 75: 109--119. Charles GD (2004) In vitro models in endocrine disruptor screening. ILAR Journal 45: 494--501. Checovich WJ, Bolger RE, and Burke T (1995) Fluorescence polarization – a new tool for cell and molecular biology. Nature 375: 254--256. Chen Z, Katzenellenbogen BS, Katzenellenbogen JA, and Zhao H (2004) Directed evolution of human estrogen receptor variants with significantly enhanced androgen specificity and affinity. Journal of Biological Chemistry 279: 33855--33864. Chen Z and Zhao H (2003) A highly efficient and sensitive screening method for transactivation activity of estrogen receptors. Gene 306: 127--134. Cheskis BJ, Karathanasis S, and Lyttle CR (1997) Estrogen receptor ligands modulate its interaction with DNA. Journal of Biological Chemistry 272: 11384--11391. Chockalingam K, Chen Z, Katzenellenbogen JA, and Zhao H (2005) Directed evolution of specific receptor–ligand pairs for use in the creation of gene switches. Proceedings of the National Academy of Sciences of the United States of America 102: 5691--5696. Claessens F, Verrijdt G, Schoenmakers E, et al. (2001) Selective DNA binding by the androgen receptor as a mechanism for hormone-specific gene regulation. Journal of Steroid Biochemistry and Molecular Biology 76: 23--30. Colborn T and Clement C (eds.) (1992) Chemically Induced Alterations in Sexual and Functional Development. Princetown: Princetown Scientific Publishing. Combes RD (2000) Endocrine disruptors: A critical review of in vitro testing strategies for assessing their toxic hazard to humans. Alternatives to Laboratory Animals 28: 81--118. Cooper TG and Bossinger J (1976) Selective inhibition of protein synthesis initiation in Saccharomyces cerevisiae by low concentrations of cycloheximide. Journal of Biological Chemistry 251: 7278--7280. Couse JF and Korach KS (1999) Estrogen receptor null mice: What have we learned and where will they lead us? Endocrine Reviews 20: 358--417. Couse JF, Lindzey J, Grandien K, Gustafsson JA, and Korach KS (1997) Tissue distribution and quantitative analysis of estrogen receptor-alpha (ERalpha) and estrogen receptor-beta (ERbeta) messenger ribonucleic acid in the wild-type and ERa-knockout mouse. Endocrinology 138: 4613--4621. Craik JCA and Harvey SM (1984) The magnitudes of three phosphorus-containing fractions in the blood plasma and mature eggs of fishes. Comparative Biochemistry and Physiology 78B: 539--543. Cravedi JP and Zalko D (2005) Metabolic fate of nonylphenols and related phenolic compounds in fish. In: Mommsen TP and Moon TW (eds.) Biochemistry and Molecular Biology of Fishes, vol. 6, pp. 153--169. Amsterdam: Elsevier. De Boever P, Demare´ W, Vanderperren E, Cooreman K, Bossier P, and Verstraete W (2001) Optimization of a yeast estrogen screen and its applicability to study the release of estrogenic isoflavones from a soygerm powder. Environmental Health Perspectives 109: 691--697. De Gooyer ME, Deckers GH, Schoonen WGEJ, Verheul HA, and Kloosterboer HJ (2003) Receptor profiling and endocrine interactions of tibolone. Steroids 68: 21--30. Deroo BJ and Korach KS (2006) Estrogen receptors and human disease. Journal of Clinical Investigation 116: 561--570. Desaulniers D, Leingartner K, Zacharewski T, and Foster WG (1998) Optimization of an MCF7-E3 cell proliferation assay and effects of environmental pollutants and industrial chemicals. Toxicology In Vitro 12: 409--422. Devillers J, Marchand-Geneste N, Carpy A, and Porcher JM (2006) SAR and QSAR modeling of endocrine disruptors. SAR and QSAR in Environmental Research 17: 393--412. Diel P, Schulz T, Smolnikar K, Strunck E, Vollmer G, and Michna H (2000) Ability of xeno- and phytoestrogens to modulate expression of estrogen-sensitive genes in rat uterus: Estrogenicity profiles and uterotropic activity. Journal of Steroid Biochemistry and Molecular Biology 73: 1--10. Dorfman RI and Dorfman AS (1953) The assay of estrogens in the chick by oral administration. Endocrinology 53: 301--305.

215

ECETOC (1999) Screening and Testing Methods for Ecotoxicological Effects of Potential Endocrine Disruptors: Response to the EDSTAC. Recommendations and a Proposed Alternative Approach. Document No. 39. Brussels, Belgium: ECETOC. Edwin JR and John PS (1997) Structural features of alkylphenolic chemicals associated with estrogenic activity. Journal of Biological Chemistry 272: 3280--3288. Eng FC, Lee HS, Ferrara J, Willson TM, and White JH (1997) Probing the structure and function of the estrogen receptor ligand binding domain by analysis of mutants with altered transactivation characteristics. Molecular and Cellular Biology 17: 4644--4653. Escande A, Pillon A, Servant N, et al. (2006) Reporter: Evaluation of ligand selectivity using reporter cell lines stably expressing estrogen receptor alpha or beta. Biochemical Pharmacology 71: 1459--1469. Escher BI, Bramaz N, Richter M, and Lienert J (2006) Comparative ecotoxicological hazard assessment of beta-blockers and their human metabolites using a mode-ofaction-based test battery and a QSAR approach. Environmental Science and Technology 40: 7402--7408. Evans NA (2004) Current concepts in anabolic-androgenic steroids. American Journal of Sports Medicine 32: 534--542. Fang H, Tong W, Perkins R, et al. (2000) Quantitative comparisons of in vitro assays for estrogenic activities. Environmental Health Perspectives 108: 723--729. Flouriot G, Vaillant C, Salbert G, Pelissero C, Guiraud JM, and Valotaire Y (1993) Monolayer and aggregate cultures of rainbow trout hepatocytes: Long-term and stable liver-specific expression in aggregates. Journal of Cell Science 105: 407--416. Fox JE, Bridgham JT, Bovee TF, and Thornton JW (2007) An evolvable oestrogen receptor activity sensor: Development of a modular system for integrating multiple genes into the yeast genome. Yeast 24: 379--390. Fry DM and Toone CK (1981) DDT-induced feminization of gull embryos. Science 213: 922--924. Fujimoto N and Katzenellenbogen BS (1994) Alteration in the agonist/antagonist balance of antiestrogens by activation of protein kinase A signaling pathways in breast cancer cells: Antiestrogen selectivity and promoter dependence. Molecular Endocrinology 8: 296--304. Funkenstein B, Dyman A, Levavi-Sivan B, and Tom M (2004) Application of real-time PCR for quantitative determination of hepatic vitellogenin transcript levels in the striped sea bream, Lithognathus mormyrus. Marine Environmental Research 58: 659--663. Gellert RJ (1978) Kepone, Mirex, Dieldrin, Aldrin: Estrogenic activity and the induction of persistent vaginal estrus and anovulation in rats following neonatal exposure. Environmental Research 16: 131--138. Gellert RJ, Heinrichs WL, and Swerdloff R (1974) Effects of neonatally-administered DDT homologs on reproductive function in male and female rats. Neuroendocrinology 16: 84--94. Giesy JP, Hilscherova K, Jones PD, Kannan K, and Machala M (2002) Cell bioassays for detection of aryl hydrocarbon (AhR) and estrogen receptor (ER) mediated activity in environmental samples. Marine Pollution Bulletin 45: 3--16. Glass CK (1994) Differential recognition of target genes by nuclear receptor monomers, dimers and heterodimers. Endocrine Reviews 15: 391--407. Graumann K, Wittliff JL, Raffelsberger W, Miles L, Jungbauer A, and Butt TR (1996) Structural and functional analysis of N-terminal point mutants of the human estrogen receptor. Journal of Steroid Biochemistry and Molecular Biology 57: 293--300. Gray LE (1998) Tiered screening and testing strategy for xenoestrogens and antiandrogens. Toxicology Letters 103: 677--680. Gray LE Jr., Kelce WR, Wiese T, et al. (1997) Endocrine screening methods workshop report: Detection of estrogenic and androgenic hormonal and antihormonal activity for chemicals that act via receptor or steroidogenic enzyme mechanisms. Reproductive Toxicology 11: 719--750. Gray LE Jr., Ostby J, Wilson V, et al. (2002) Xenoendocrine disrupters-tiered screening and testing: Filling key data gaps. Toxicology 181–182: 371--382. Guelden M, Turan A, and Seibert H (1998) Endocrinally active chemicals and their occurrence in surface waters. UBA-text 66/98. Research Report 102 04 279. Berlin: Federal Environmental Agency. Guillette LJ Jr., Gross TS, Gross D, Rooney AA, and Percival HF (1995) Gonadal steroidogenesis in vitro from juvenile alligators obtained from contaminated or control lakes. Environmental Health Perspectives 103: 31--36. Guillette LJ, Gross TS, Masson GR, Matter JM, Percival HF, and Woodward AR (1999) Developmental abnormalities of the gonad and abnormal sex hormone concentrations in juvenile alligators from contaminated and control lakes in Florida. Environmental Health Perspectives 102: 680--688. Gustafsson JA (1999) Estrogen receptor beta – a new dimension in estrogen mechanism of action. Journal of Endocrinology 163: 379--383.

216

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

Habauzit D, Armengaud J, Roig B, and Chopineau J (2008) Determination of estrogen presence in water by SPR using estrogen receptor dimerization. Analytical and Bioanalytical Chemistry 390: 873--883. Habauzit D, Chopineau J, and Roig B (2007) SPR-based biosensors: A tool for biodetection of hormonal compounds. Analytical and Bioanalytical Chemistry 387: 1215--1223. Harris HA (2007) Estrogen receptor-b: Recent lessons from in vivo studies. Molecular Endocrinology 21: 1--13. Hartig PC, Bobseine KL, Britt BH, et al. (2002) Development of two androgen receptor assays using adenoviral transduction of MMTV-luc reporter and/or hAR for endocrine screening. Toxicological Sciences 66: 82--90. Hashimoto S, Akatsuka Y, Kurihara R, et al. (2005) Evaluation of the Ishikawa cell line bioassay for the detection of estrogenic substances from sediment extracts. Environmental Toxicology and Chemistry 24: 1587--1593. Heim R, Prasher D, and Tsien R (1994) Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proceedings of the National Academy of Sciences of the United States of America 91: 12501--12504. Heinz GH, Percival HF, and Jennings ML (1991) Contaminants in American alligator eggs from lakes Apopka, Griffin and Okeechobee, Florida. Environmental Monitoring and Assessment 16: 277--285. Hermenegildo C and Cano A (2000) Modulation of the oestrogen receptor: A process with distinct susceptible steps. Human Reproduction Update l6: 237--243. Hock B, Seifert M, and Kramer K (2002) Engineering antibodies and receptors for biosensors. Biosensors and Bioelectronics 17: 239--249. Holmes P, Humphrey C, and Scullion M (1998) Appraisal of test methods for sexhormone disrupting chemicals capable of affecting the reproductive process. http:// www.oecd.org/dataoecd/58/27/1898464.pdf. Hornung MW, Ankley GT, and Schmieder PK (2003) Induction of an estrogenresponsive reporter gene in rainbow trout hepatoma cells (RTH 149) at 11 or 18 1C. Environmental Toxicology and Chemistry 22: 866--871. Hoss S, Bergtold M, Haitzer M, Traunspurger W, and Steinberg CEW (2001) Refractory dissolved organic matter can influence the reproduction of Caenorhabditis elegans (Nematoda). Freshwater Biology 46: 1--10. Hoyt PR, Doktycz MJ, Beattie KL, and Greeley MS (2003) DNA microarrays detect 4nonylphenol-induced alterations in gene expression during zebrafish early development. Ecotoxicology 12: 447--469. Islam KM, Dilcher M, Thurow C, et al. (2009) Directed evolution of estrogen receptor proteins with altered ligand-binding specificities. Protein Engineering, Design and Selection 22: 45--52. Islinger M, Pawlowski S, Hollert H, Volkl A, and Braunbeck T (1999) Measurement of vitellogenin mRNA expression in primary cultures of rainbow trout hepatocytes in a non-radioactive dot blot/RNAse protection assay. Science of the Total Environment 233: 109--122. Jakacka M, Ito M, Martinson F, et al. (2002) An estrogen receptor (ER)alpha deoxyribonucleic acid-binding domain knock-in mutation provides evidence for nonclassical ER pathway signalling in vivo. Molecular Endocrinology 16: 2188--2201. Jiang SY and Jordan VC (1992) Growth regulation of estrogen receptor-negative breast cancer cells transfected with cDNA’s for estrogen receptor. Journal of the National Cancer Institute 84: 580--591. Jobling S, Nolan M, Tyler C, Brighty G, and Sumpter JP (1998) Widespread sexual disruption in wild fish. Environmental Science and Technology 32: 2498--2506. Jobling S and Sumpter JP (1993) Detergent components in sewage effluent are weakly oestrogenic to fish: An in vitro study using rainbow trout (Oncorhynchus mykiss) hepatocytes. Aquatic Toxicology 27: 361–372. Jorgensen M, Vendelbo B, Skakkebaek NE, and Leffers H (2000) Assaying estrogenicity by quantitating the expression levels of endogenous estrogenregulated genes. Environmental Health Perspectives 108: 403--412. Kang S-G, Choi K-S, Bulgakov AA, Kim Y, and Kim S-Y (2003) Enzyme-linked immunosorbent assay (ELISA) used in quantification of reproductive output in the pacific oyster, Crassostrea gigas, in Korea. Journal of Experimental Marine Biology and Ecology 282: 1--21. Kao YC, Zhou C, Sherman M, Laughton CA, and Chen S (1998) Molecular basis of the inhibition of human aromatase (estrogen synthetase) by flavone and isoflavone phytoestrogens: A site-directed mutagenesis study. Environmental Health Perspectives 106: 85--92. Katzenellenbogen BS (1996) Estrogen receptors: Bioactivities and interactions with cell signaling pathways. Biology of Reproduction 54: 287–293. Katzenellenbogen BS, Bhardwaj B, Fang H, et al. (1993) Hormone binding and transcription activation by estrogen receptors: Analyses using mammalian and yeast systems. Journal of Steroid Biochemistry and Molecular Biology 47: 39--48. Katzenellenbogen BS, Kendra KL, Norman MJ, and Berthois Y (1987) Proliferation, hormonal responsiveness, and estrogen receptor content of MCF-7 human breast

cancer cells grown in the short-term and long-term absence of estrogens. Cancer Research 47: 4355--4360. Kelce WR, Monosson E, Gamesik MP, Laws SC, and Gray JLE (1994) Environmental hormone disruptors: Evidence that vinclozolin developmental toxicity is mediated by antiandrogenic metabolites. Toxicology and Applied Pharmacology 126: 275--285. Kelce WR and Wilson EM (1997) Environmental anti-androgens: Developmental effects, molecular mechanisms and clinical implications. Journal of Molecular Medicine 75: 198–207. Kime DE, Nash JP, and Scott AP (1999) Vitellogenesis as a biomarker of reproductive disruption by xenobiotics. Aquaculture 177: 345--352. Klein KO, Baron J, Colli MJ, McDonnell DP, and Cutler GB Jr. (1994) Estrogen levels in childhood determined by an ultrasensitive recombinant cell bioassay. Journal of Clinical Investigation 94: 2475--2480. Kleinkauf A, Macfarlane C, Yeates S, Simpson MG, and Leah RT (2004) A biomarker approach to endocrine disruption in flounder-estrogen receptors, hepatocyte proliferation, and sperm motility. Ecotoxicology and Environmental Safety 58: 324--334. Koehler KF, Helguero LA, Haldosen LA, et al. (2005) Reflections on the discovery and significance of estrogen receptor b. Endocrine Reviews 26: 465--478. Koerner W, Hanf V, Schuller W, Kempter C, Metzger JW, and Hagenmaier H (1999) Development of a sensitive E-screen for quantitative analysis of estrogenic activity in municipal sewage plant effluents. Science of the Total Environment 225: 33--48. Korach KS and McLachlan JA (1995) Techniques for detection of estrogenicity. Environmental Health Perspectives 103: 5--8. Kordes C, Rieber EP, and Gutzeit HO (2002) An in vitro vitellogenin bioassay for oestrogenic substances in the medaka (Oryzias latipes). Aquatic Toxicology 58: 151--164. Korner W, Hanf V, Schuller W, Bartsch H, Zwirner M, and Hagenmaier H (1998) Validation and application of a rapid in vitro assay for assessing the estrogenic potency of halogenated phenolic chemicals. Chemosphere 37: 2395--2407. Kramer VJ, Miles-Richardson S, Pierens SL, and Giesy JP (1998) Reproductive impairment and induction of alkaline-labile phosphate, a biomarker of estrogen exposure, in fathead minnows (Pimephales promelas) exposed to waterborne 17bestradiol. Aquatic Toxicology 40: 335--360. Kraus WL, Weis KE, and Katzenellenbogen BS (1995) Inhibitory crosstalk between steroid hormone receptors: Differential targeting of estrogen receptor in the repression of transcriptional activity by agonist- and antagonist-occupied progestin receptors. Molecular and Cellular Biology 15: 1847--1857. Krege JH, Hodgin JB, Couse JF, et al. (1998) Generation and reproductive phenotypes of mice lacking estrogen receptor beta. Proceedings of the National Academy of Sciences of the United States of America 95: 15677--15682. Kuch H and Ballschmiter K (1999) Hormonell wirksame Verbindungen in der Umwelt Baden-Wuerttembergs. Arbeitsbericht Nr 151. Akademie fu¨r Technikfolgenabschaetzung in Baden-Wuerttemberg, Stuttgart. Kuiper GG, Lemmen JG, Carlsson B, et al. (1998) Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor beta. Endocrinology 139: 4252--4263. Kuiper GGJM, Carlsson B, Grandien K, et al. (1997) Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptor a and b. Endocrinology 138: 863--870. Kuster M, Lo´pez de Alda MJ, and Barcelo´ D (2004) Analysis and distribution of estrogens and progestragens in sewage sludge, soils, and sediments. Trends in Analytical Chemistry 23: 790--798. Kwon HC, Hayashi S, and Mugiya Y (1993) Vitellogenin induction by estradiol-17b in primary hepatocyte culture in the rainbowtrout Oncorhynchus mykiss. Comparative Biochemistry and Physiology 104B: 381--386. Latonnelle K, Le Menn F, and Bennetau-Pelissero C (2000) In vitro estrogenic effects of phytoestrogens in rainbow trout and Siberian sturgeon. Ecotoxicology 9: 115--125. Lee HS, Miyauchi K, Nagata Y, et al. (2002) Employment of the human estrogen receptor beta ligand-binding domain and co-activator SRC1 nuclear receptorbinding domain for the construction of a yeast two-hybrid detection system for endocrine disrupters. Journal of Biochemistry 131: 399--405. Legler J, van den Brink CE, Brouwer A, et al. (1999) Development of a stably transfected estrogen receptor-mediated luciferase reporter gene assay in the human T47D breast cancer cell line. Toxicological Science 48: 55--66. Li Q, Osada M, Suzuki R, and Mori K (1998) Changes in vitellin during oogenesis and effect of estradiol 17-b on vitellogenesis in the Pacific oyster Crassostrea gigas. Invertebrate Reproduction and Development 33: 87--89. Liang K, Yang L, Xiao Z, and Huang J (2009) A bipartite recombinant yeast system for the identification of subtype-selective estrogen receptor ligands. Molecular Biotechnology 41: 53--62.

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

Lindholst C, Wynne PM, Marriott P, Pedersen SN, and Bierregaard P (2003) Metabolism of bisphenol A in zebrafish (Danio rerio) and rainbow trout (Oncorhynchus mykiss) in relation to estrogenic response. Comparative Biochemistry and Physiology 135C: 169--177. Livingstone DR, Chipman JK, Lowe DM, et al. (2000) Development of biomarkers to detect the effects of organic pollution on aquatic invertebrates: Recent molecular, genotoxic, cellular and immunological studies on the common mussel (Mytilus edulis L.) and other mytilids. International Journal of Environment and Pollution 13: 56--91. Lockhart DJ, Dong H, Byrne MC, et al. (1996) Expression monitoring by hybrydization to high-density oligonucleotide arrays. Nature Biotechnology 14: 1675--1680. Lyttle CR, Damian-Matsumura P, Juul H, and Butt TR (1992) Human estrogen receptor regulation in a yeast model system and studies on receptor agonists and antagonists. Journal of Steroid Biochemistry and Molecular Biology 42: 677--685. MacLusky NJ, Clark AS, Naftolin F, and Goldman-Rakic PS (1987) Estrogen formation in the mammalian brain: Possible role of aromatase in sexual differentiation of the hippocampus and neocortex. Steroids 50: 459--474. Maitre JL, Valotaire Y, and Guguen-Guillouzo C (1986) Estradiol-17b stimulation of vitellogenin synthesis in primary culture of male rainbow trout hepatocytes. In Vitro Cellular and Developmental Biology – Animal 22: 337--343. Mak P, Cruz FD, and Chen S (1999) A yeast screen system for aromatase inhibitors and ligands for androgen receptor: Yeast cells transformed with aromatase and androgen receptor. Environmental Health Perspectives 107: 855--860. Marin MG and Matozzo V (2004) Vitellogenin induction as a biomarker of exposure to estrogenic compounds in aquatic environments. Marine Pollution Bulletin 48: 835--839. Marx A, Sherry J, Hansen P-D, and Hock B (2001) A new monoclonal antibody against vitellogenin from rainbow trout (Oncorhynchus mykiss). Chemosphere 44: 393--399. Matozzo V, Gagne´ F, Marin MG, Ricciardi F, and Blaise C (2008) Vitellogenin as a biomarker of exposure to estrogenic compounds in aquatic invertebrates: A review. Environment International 34: 531--545. Matthiessen P (2003) Endocrine disruption in marine fish. Pure and Applied Chemistry 75: 2249--2261. McDonnell DP (1999) The molecular pharmacology of SERMs. Trends in Endocrinology and Metabolism 10: 301--311. McDonnell DP, Nawaz Z, Densmore C, et al. (1991) High level expression of biologically active estrogen receptor in Saccharomyces cerevisiae. Journal of Steroid Biochemistry and Molecular Biology 39: 291--297. McLachlan JA and Newbold RR (1987) Estrogens and development. Environmental Health Perspectives 75: 25--27. Metzger D, White JH, and Chambon P (1988) The human oestrogen receptor functions in yeast. Nature 334: 31--36. Meyer HH (2001) Biochemistry and physiology of anabolic hormones used for improvement of meat production. Acta Pathologica, Microbiologica et Immunologica Scandinavica 109: 1--8. Meyers MJ, Sun J, Carlson KE, et al. (2001) Estrogen receptor-potency-selective ligands: Structure–activity relationship studies of diarylpropionitriles and their acetylene and polar analogues. Journal of Medicinal Chemistry 44: 4230--4251. Miyashita M, Shimada T, Miyagawa H, and Akamatsu M (2005) Surface plasmon resonance-based immunoassay for 17b-estradiol and its application to the measurement of estrogen receptor-binding activity. Analytical and Bioanalytical Chemistry 381: 667--673. Mommsen TP and Walsh PJ (1988) Vitellogenesis and oocyte assembly. In: Hoar WS and Randall DJ (eds.) Fish Physiology, vol.XIA, pp. 347--406. New York: Academic Press. Montano MM, Ekena K, Krueger KD, Keller AL, and Katzenellenbogen BS (1996) Human estrogen receptor ligand activity inversion mutants: Receptors that interpret antiestrogens as estrogens and estrogens as antiestrogens and discriminate among different antiestrogens. Molecular Endocrinology 10: 230--242. Mosselman S, Polman J, and Dijkema R (1996) ER beta: Identification and characterization of a novel human estrogen receptor. FEBS Letters 392: 49--53. Mueller SO (2004) Xenoestrogens: Mechanisms of action and detection methods. Analytical and Bioanalytical Chemistry 378: 582--587. Mueller SO, Kling M, Arifin Firzani P, et al. (2003) Activation of estrogen receptor alpha and ERbeta by 4-methylbenzylidene-camphor in human and rat cells: Comparison with phyto- and xenoestrogens. Toxicology Letters 142: 89--101. Mueller SO and Korach KS (2001) Immortalized testis cell lines from estrogen receptor (ER) alpha knock-out and wild-type mice expressing functional ERalpha or ERbeta. Journal of Andrology 22: 652--664. Murata M, Midorikawa K, Koh M, Umezawa K, and Kawanishi S (2004) Genistein and daidzein induce cell proliferation and their metabolites cause oxidative DNA

217

damage in relation to isoflavone-induced cancer of estrogen-sensitive organs. Biochemistry 43: 2569--2577. Murata M, Nakayama MH, Yakabe K, Fukuma K, Katayama Y, and Maeda M (2001) Novel biosensor for the rapid measurement of estrogen based on a ligand–receptor interaction. Analytical Sciences 17: 387--390. Naciff JM and Daston GP (2004) Toxicogenomic approach to endocrine disrupters: Identification of a transcript profile characteristic of chemicals with estrogenic activity. Toxicologic Pathology 32: 59--70. Naciff JM, Jump ML, Torontali SM, et al. (2002) Gene expression profile induced by 17a-ethynyl estradiol, bisphenol a, and genistein in the developing female reproductive system of the rat. Toxicological Sciences 68: 184--199. Naciff JM, Overmann GJ, Torontali SM, et al. (2003) Gene expression profile induced by 17 alpha-ethynyl estradiol in the prepubertal female reproductive system of the rat. Toxicological Sciences 72: 314--330. Navas JM and Segner H (2000) Antiestrogenicity of b-naphthoflavone and PAHs in cultured rainbow trout hepatocytes: Evidence for a role of the arylhydrocarbon receptor. Aquatic Toxicology 51: 79--92. Navas JM and Segner HH (2006) Vitellogenin synthesis in primary cultures of fish liver cells as endpoint for in vitro screening of the (anti)estrogenic activity of chemical substances. Aquatic Toxicology 80: 1--22. Nilsen BM, Berg K, Eidem JK, et al. (2004) Development of quantitative vitellogeninELISAs for fish test species used in endocrine disruptor screening. Analytical and Bioanalytical Chemistry 378: 621--626. Nishikawa JI, Koichi S, Goto J, Dakeyama F, Matsuo M, and Nishihara T (1999) New screening methods for chemicals with hormonal activities using interaction of nuclear hormone receptor with co-activator. Toxicology and Applied Pharmacology 154: 76--83. Nuwaysir EF, Bittner M, Trent J, Barrett JC, and Afshari CA (1999) Microarrays and toxicology: The advent of toxicogenomics. Molecular Carcinogenesis 24: 153--159. O’Connor JC, Cook JC, Marty MS, Davis LG, Kaplan AM, and Carney EW (2002) Evaluation of Tier I screening approaches for detecting endocrine-active compounds (EACs). Critical Reviews in Toxicology 32: 521--549. O’Connor JC, Frame SR, Biegel LB, Cook JC, and Davis LG (1998) Sensitivity of a Tier I screening battery compared to an in utero exposure for detecting the estrogen receptor agonist 17 beta-estradiol. Toxicological Sciences 44: 169--184. Odum J, Lefevre PA, Tittensor S, et al. (1997) The rodent uterotrophic assay: Critical protocol features, studies with nonyl phenols, and comparison with a yeast estrogenicity assay. Regulatory Toxicology and Pharmacology 25: 176--188. Odum J, Tittensor S, and Ashby J (1998) Limitations of the MCF-7 cell proliferation assay for detecting xenobiotic oestrogens. Toxicology in Vitro 12: 273--278. OECD (2001) Appraisal of test methods for sex hormone disrupting chemicals. In: OECD Environment Health and Safety Publications: Series on Testing and Assessment No. 21. Paris: Environment Directorate. Ogawa S, Inoue S, Watanabe T, et al. (1998) The complete primary structure of human estrogen receptor beta (hER beta) and its heterodimerization with ER alpha in vivo and in vitro. Biochemical and Biophysical Research Communications 243: 122--126. Okoumassoun L-E, Averill-Bates D, Gagne´ F, Marion M, and Denizeau F (2002) Assessing the estrogenic potential of organochlorine pesticides in primary cultures of male rainbow trout (Oncorhynchus mykiss) hepatocytes using vitellogenin as a biomarker. Toxicology 178: 193--207. Olea N, Devleeschower N, and Leclercq G (1985) New method for the determination of estrogen and progesterone receptors in human breast cell lines. Revista Espan˜ola De Fisiologı´a 41: 373--380. Ortiz-Zarragoitia M and Cajaraville MP (2005) Effects of selected xenoestrogens on liver peroxisomes, vitellogenin levels and spermatogenic cell proliferation in male zebrafish. Comparative Biochemistry and Physiology C 141: 133–144. Osada M, Takamura T, Sato H, and Mori K (2003) Vitellogenin synthesis in the ovary of scallop Platinopecten yessoensis: Control by estradiol-17b and the central nervous system. Journal of Experimental Zoology 299A: 172--179. Ouatas T, Clare SE, Hartsough MT, De La Rosa A, and Steeg PS (2002) MMTVassociated transcription factor binding sites increase nm23-H1 metastasis suppressor gene expression in human breast carcinoma cell lines. Clinical and Experimental Metastasis 19: 35--42. Paris F, Servant N, Terouanne B, and Sultan C (2002) Evaluation of androgenic bioactivity in human serum by recombinant cell line: Preliminary results. Molecular and Cellular Endocrinology 198: 123--129. Pawlowski S, Islinger M, Vo¨lkl A, and Braunbeck T (2000) Temperature-dependent vitellogenin-mRNA expression in primary cultures of rainbow trout (Oncorhynchus mykiss) hepatocytes at 14 and 18 1C. Toxicology in Vitro 14: 531--540.

218

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

Pearson J, Gill A, Margison GP, Vadgama P, and Povey AC (2001) A rapid receptor– ligand assay for determination of oestrogens using surface plasmon resonance. Sensors and Actuators B 76: 1--7. Pelissero C, Bennetau B, Babin P, Le Menn F, and Dunogu`es J (1991) The estrogenic activity of certain phytoestrogens in the Siberian sturgeon Acipenser baeri. Journal of Steroid Biochemistry and Molecular Biology 38: 292--299. Pelissero C, Flouriot G, Foucher JL, et al. (1993) Vitellogenin synthesis in cultured hepatocytes; an in vitro test for the estrogenic potency of chemicals. Journal of Steroid Biochemistry 44: 263--272. Peters GA and Khan SA (1999) Estrone receptor domains E and F: Role in dimerizatdon and interaction with co-activator RIP-140. Molecular Endocrinology 13: 286--296. Petit F, Le Goff P, Crave´di J-P, Valotaire Y, and Pakdal F (1997) Two complementary bioassays for screening the estrogenic potency of xenobiotics: Recombinant yeast for trout estrogen receptor and trout hepatocytes cultures. Journal of Molecular Endocrinology 19: 321--335. Petrovic M, Eljarrat E, Lopez de Alda MJ, and Barcelo D (2004) Endocrine disrupting compounds and other emerging contaminants in the environment: A survey on new monitoring strategies and occurrence data. Analytical and Bioanalytical Chemistry 378: 549--562. Peyon P, Baloche S, and Burzawa-Ge´rard E (1993) Synthesis of vitellogenin by Eel (Anguilla anguilla L.) hepatocytes in primary culture: Requirement of 17bestradiol-priming. General and Comparative Endocrinology 91: 318–329. Peyon P, Calvayrac R, Baloche S, and Burzawa-Ge´rard E (1998) Metabolic studies on eel (Anguilla anguilla L.) hepatocytes in primary culture: Effect of 17b-estradiol and growth hormone. Comparative Biochemistry and Physiology 121A: 35--44. Pierrat B, Heery DM, Chambon P, and Losson R (1994) A highly conserved region in the hormone-binding domain of the human estrogen receptor functions as an efficient transactivation domain in yeast. Gene 143: 193--200. Pinkerton JV and Henderson VW (2005) Estrogen and cognition, with a focus on Alzheimer’s disease. Seminars in Reproductive Medicine 23: 172--179. Porcher JM (2003) Intercomparison of zebrafish vitellogenin quantification methods. INERIS Draft Report, October 2003. Verneuil-en-Halatte, France:L’ Institut National de l environnement industriel et des risques. Porte C, Janer G, Lorusso LC, et al. (2006) Endocrine disruptors in marine organisms: Approaches and perspectives. Comparative Biochemistry and Physiology – Part C: Toxicology and Pharmacology 143: 303--315. Quaedackers ME, van den Brink CE, Wissink S, et al. (2001) 4-Hydroxytamoxifen trans-represses nuclear factor-kB activity in human osteoblastic U2-OS cells through estrogen receptor (ER) alpha, and not through ERbeta. Endocrinology 142: 1156--1166. Ramamoorthy K, Vyhlidal C, Wang F, et al. (1997) Additive estrogenic activities of a binary mixture of 20 ,40 ,60 -trichloro- and 20 ,30 ,40 ,50 -tetrachloro-4-biphenylol. Toxicology and Applied Pharmacology 147: 93--100. Reed MJ, Owen AM, Lai LC, et al. (1989) In situ oestrone synthesis in normal breast and breast tumor tissues: Effect of treatment with 4-hydroxyandrostenedione. International Journal of Cancer 44: 233--237. Ren L, Lewis SK, and Lech JJ (1996) Effects of estrogen and nonylphenol on the posttranscriptional regulation of vitellogenin gene expression. Chemico-Biological Interactions 100: 67--76. Rich RL, Hoth LR, Geoghegan KF, et al. (2002) Kinetic analysis of estrogen receptor/ ligand interactions. Proceedings of the National Academy of Sciences of the United States of America 99: 8562--8567. Riley LG, Hirano T, and Grau EG (2004) Estradiol-17beta and dihydrotestosterone differentially regulate vitellogenin and insulin-like growth factor-I production in primary hepatocytes of the tilapia, Oreochromis mossambicus. Comparative Biochemistry and Physiology 138C: 177--186. Robinson CD, Brown E, Craft JA, Davies IM, and Moffat CF (2004) Effects of prolonged exposure to 4-tert-octylphenol on toxicity and indices of oestrogenic exposure in the sand goby (Pomatoschistus minutus, Pallas). Marine Environmental Research 58: 19--38. Rogers JM and Denison MS (2000) Recombinant cell bioassays for endocrine disruptors: Development of a stably transfected human ovarian cell line for the detection of estrogenic and anti-estrogenic chemicals. In Vitro and Molecular Toxicology 13: 67--82. Routledge EJ and Sumpter JP (1996) Estrogenic activity of surfactants and some of their degradation products assessed using a recombinant yeast screen. Environmental Toxicology and Chemistry 15: 241--248. Routledge EJ, White R, Parker MG, and Sumpter JP (2000) Differential effects of xenoestrogens on coactivator recruitment by estrogen receptor (ER) a and ERjB. Journal of Biological Chemistry 275: 35986--35993.

Rutishauser BV, Pesonen M, Escher BI, et al. (2004) Comparative analysis of estrogenic activity in sewage treatment plant effluents involving three in vitro assays and chemical analysis of steroids. Environmental Toxicology and Chemistry 23: 857--864. Safe SH, Pallaroni L, Yoon K, Gaido K, Ross S, and McDonnell D (2002) Problems for risk assessment of endocrine-active estrogenic compounds. Environmental Health Perspectives 110: 925--929. Safe SH, Pallaroni L, Yoon K, et al. (2001) Toxicology of environmental estrogens. Reproduction, Fertility and Development 13: 307--315. Schmieder P, Tapper M, Linnum A, Denny J, Kolanczyk R, and Johnson R (2000) Optimization of a precision-cut trout liver tissue slice assay as a screen for vitellogenin induction comparison of slice incubation techniques. Aquatic Toxicology 49: 251--268. Schultis T and Metzger JW (2004) Determination of estrogenic activity by LYES-assay (yeast estrogen screen-assay assisted by enzymatic digestion with lyticase). Chemosphere 57: 1649--1655. Schultis T, Spengler P, Ko¨nig A, and Metzger JW (2002) Determination of estrogenic activity of environmental samples by means of binding to estrogen receptors ER-a and ER-b and fluorescence polarization. Vom Wasser 98: 1--12. Scrimshaw MD and Lester JN (2003) In vitro assays for determination of oestrogenic activity. Analytical and Bioanalytical Chemistry 378: 576--581. Segner H (2005) Developmental, reproductive, and demographic alterations in aquatic wildlife: Establishing causality between exposure to endocrine-active compounds (EACs) and effects. Acta Hydrochimica et Hydrobiologica 33: 17--26. Segner H and Braunbeck T (2003) End points for in vitro toxicity testing with fish cells. In: Mothersill C and Austin B (eds.) In Vitro Methods in Aquatic Toxicology, pp. 77--141. Chichester: Springer. Segner H and Cravedi JP (2001) Metabolic activity in primary cultures of fish hepatocytes. Alternatives to Laboratory animals 29: 251--257. Seifert M (2000) Bestimmung von O¨strogenen und Xenoo¨strogenen mit einem Rezeptorassay. PhD Thesis, Technische Universita¨t Mu¨nchen. Seifert M (2004) Luminescent enzyme-linked receptor assay for estrogenic compounds. Analytical and Bioanalytical Chemistry 378: 684–687. Seifert M, Haindl S, and Hock B (1999) Development of an enzyme linked receptor assay (ELRA) for estrogens and xenoestrogens. Analytica Chimica Acta 386: 191--199. Sesay AM and Cullen DC (2001) Detection of hormone mimics in water using a miniturised SPR sensor. Environmental Monitoring and Assessment 70: 83--92. Shang Y and Brown M (2002) Molecular determinants for the tissue specificity of SERMs. Science 295: 2465--2468. Sheeler CQ, Dudly MW, and Khan SA (2000) Environmental estrogens induce transcriptionally active estrogen receptor dimers in yeast: Activity potentiated by the co-activator RIP140. Environmental Health Perspectives 108: 97--103. Shelby MD, Newbold RR, Tully DB, Chae K, and Davis VL (1996) Assessing environmental chemicals for estrogenicity using a combination of in vitro and in vivo assays. Environmental Health Perspectives 104: 1296--1300. Shi L, Tong W, Fang H, et al. (2002) An integrated ‘‘4-phase’’ approach for setting endocrine disruption screening priorities – phase I and II predictions of estrogen receptor binding affinity. SAR and QSAR in Environmental Research 13: 69--88. Shilling AD and Williams DE (2000) Determining relative estrogenicity by quantifying vitellogenin induction in rainbow trout liver slices. Toxicology and Applied Pharmacology 164: 330--335. Silva MC, Rowlands MG, Dowsett M, et al. (1989) Intratumoral aromatase as a prognostic factor in human breast carcinoma. Cancer Research 49: 2588--2591. Silversand C, Hyllner SJ, and Haux C (1993) Isolation, immunochemical detection, and observations of the instability of vitellogenin from four teleosts. Journal of Experimental Zoology 267: 587--597. Sitcheran R, Emter R, Kralli A, and Yamamoto KR (2000) A genetic analysis of glucocorticoid receptor signalling. Identification and characterization of ligandeffect modulators in Saccharomyces cerevisiae. Genetics 156: 963--972. Smeets JMW, Rankouhi TR, Nichols KM, et al. (1999) In vitro vitellogenin production by carp (Cyprinus carpio) hepatocytes as a screening method for determining (anti)estrogenic activity of xenobiotics. Toxicology and Applied Pharmacology 157: 68--76. Smith DF and Toft DO (1993) Steroid receptors and their associated proteins. Molecular Endocrinology 7: 4–11. Sohoni P and Sumpter JP (1998) Several environmental oestrogens are also antiandrogens. Journal of Endocrinology 158: 327--339. Sole´ M, Porte C, and Barcelo´ D (2000) Vitellogenin induction and other biochemical responses in Carp, Cyprinus carpio, after experimental injection with 17bethynylestradiol. Archives of Environmental Contamination and Toxicology 38: 494--500.

Bioassays for Estrogenic and Androgenic Effects of Water Constituents

Sonneveld E, Jansen HJ, Riteco JAC, Brouwer A, and van der Burg B (2003) CALUX reporter cell lines as sensitive bioassays for monitoring biological activity of various classes of steroid hormones reveal the repressive effect of TCDD on the estrogen receptor. Organohalogen Compounds 60: 263–266. Sonneveld E, Jansen HJ, Riteco JA, Brouwer A, and van der Burg B (2005) Development of androgen- and estrogen-responsive bioassays, members of a panel of human cell line-based highly selective steroid-responsive bioassays. Toxicological Sciences 83: 136--148. Soto AM, Lin TM, Justicia H, Silvia RM, and Sonnenschein C (1992) An ‘in culture’ bioassay to assess the estrogenicity of xenobiotics (E-SCREEN). In: Colborn T and Clement C (eds.) Advances in Modern Environmental Toxicology: ChemicallyInduced Alterations in Sexual and Functional Development: The Wildlife/Human Connection, Vol. 21 – Advances in Modern Environmental Toxicology pp. 295--309. Princeton, NJ: Princeton Scientific. Soto AM and Sonnenschein C (1985) The role of estrogens on the proliferation of human breast tumor cells (MCF-7). Journal Of Steroid Biochemistry 23: 87--94. Soto AM, Sonnenschein C, Chung KL, Fernandez MF, Olea N, and Serrano FO (1995) The ESCREEN assay as a tool to identify estrogens: An update on estrogenic environmental pollutants. Environmental Health Perspectives 103: 113--122. Specker JL and Sullivan CV (1994) Vitellogenesis in fishes: Status and perspectives. In: Davey KG, Peter RE, and Tobe SS (eds.) Perspectives in Endocrinology, pp. 304–315. Ottawa: National Research Council. Steiner S and Anderson NL (2000) Expression profiling in toxicology: Potential and limitations. Toxicology Letters 112–113: 467--471. Sutherland RL, Hall RE, Pang GY, Musgrove EA, and Clarke CL (1988) Genetic instability and the development of steroid hormone insensitivity in cultured T 47D human breast cancer cells. Cancer Research 48: 4340--4347. Ternes TA, Andersen H, Gieberg D, and Bonerz M (2002) Determination of estrogens in sludge and sediments by liquid extraction and GC/MS/MS. Analytical Biochemistry 74: 3498--3504. Terouanne B, Tahiri B, Georget V, et al. (2000) A stable prostatic bioluminescent cell line to investigate androgen and antiandrogen effects. Molecular and Cellular Endocrinology 160: 39--49. Thatcher JW, Shaw JM, and Dickinson WJ (1998) Marginal fitness contributions of nonessential genes in yeast. Proceedings of the National Academy of Sciences of the United States of America 95: 253--257. Thorpe KL, Cummings RI, Hutchinson TH, et al. (2003) Relative potencies and combination effects of steroidal estrogens in fish. Environmental Science and Technology 37: 1142--1149. Toomey BH, Monteverdi GH, and Di Giulio RT (1999) Octylphenol induces vitellogenin production and cell death in fish hepatocytes. Environmental Toxicology and Chemistry 18: 734--739. Toth I, Faredin I, Mesko E, Wolfling J, and Schneider G (1995) In vitro binding of 16methylated C18 and C19 steroid derivatives to the androgen receptor. Pharmacological Research 32: 217--221. Tsai MJ and O’Malley BW (1994) Molecular mechanisms of action of steroid/thyroid receptor superfamily members. Annual Review of Biochemistry 63: 451--486. Tsien RY (1998) The green fluorescent protein. Annual Review of Biochemistry 67: 509--544. Tucker CL and Fields S (2001) A yeast sensor of ligand binding. Nature Biotechnology 19: 1042--1046. Tyler CR, Sumpter JP, and Bromage NR (1988) In vivo ovarian uptake and processing of vitellogenin in the rainbow trout, Salmo gairdneri. Journal of Experimental Zoology 246: 171--179. Tyler CR, van Aerle R, Hutchinson TH, Maddix S, and Trip H (1999) An in vivo testing system for endocrine disruptors in fish early life stages using induction of vitellogenin. Environmental Toxicology and Chemistry 18: 337--347. Tyler CR, van der Eerden B, Jobling S, Panter G, and Sumpter JP (1996) Measurement of vitellogenin, a biomarker for exposure to oestrogenic chemicals, in a wide variety of cyprinid fish. Journal of Comparative Physiology B 166: 418--426. Uchiumi F, Sato T, and Tanuma S (1998) Identification and characterization of a tannic acid-responsive negative regulatory element in the mouse mammary tumor virus promoter. Journal of Biological Chemistry 273: 12499--12508. Usami M, Mitsunaga K, and Ohno Y (2002) Estrogen receptor binding assay of chemicals with a surface plasmon resonance biosensor. Journal of Steroid Biochemistry and Molecular Biology 81: 47--55.

219

Varshavsky A (1992) The N-end rule. Cell 69: 725--735. Veeneman GH (2005) Non-steroidal subtype selective estrogens. Current Medical Chemistry 12: 1077--1136. Vegeto E, Allan GF, Schrader WT, Tsai MJ, McDonnell DP, and O’Malley BW (1992) The mechanism of RU486 antagonism is dependent on the conformation of the carboxy-terminal tail of the human progesterone receptor. Cell 69: 703--713. Verslycke T, De Wasch K, De Brabander HF, and Janssen CR (2002) Testosterone metabolism in the estuarine mysid Neomysis integer (Crustacea; Mysidacea): Identification of testosterone metabolites and endogenous vertebrate-type steroids. General and Comparative Endocrinology 126: 190--199. Versonnen BJ, Goemans G, Belpaire C, and Janssen CR (2004) Vitellogenin content in European eel (Anguilla anguilla) in Flanders, Belgium. Environmental Pollution 128: 363--371. Vinggaard AM, Nellemann C, Dalgaard M, Jorgensen EB, and Andersen HR (2002) Antiandrogenic effects in vitro and in vivo of the fungicide prochloraz. Toxicological Sciences 69: 344--353. Wallace RA (1985) Oocyte growth in nonmammalian vertebrates. In: Jones RE (ed.) The Vertebrate Ovary, pp. 469--502. New York: Plenum. Walter P, Green S, Greene G, et al. (1985) Cloning of the human estrogen receptor cDNA. Proceedings of the National Academy of Sciences of the United States of America 82: 7889--7893. Wang C, Makela T, Hase T, Adlercreutz H, and Kurzer MS (1994) Lignans and flavonoids inhibit aromatase enzyme in human preadipocytes. Journal of Steroid Biochemistry and Molecular Biology 50: 205--212. Whelan J and Miller N (1996) Generation of estrogen receptor mutants with altered ligand specificity for use in establishing a regulatable gene expression system. Journal of Steroid Biochemistry and Molecular Biology 58: 3--12. White R, Jobling S, Hoare SA, Sumpter JP, and Parker MG (1994) Environmentally persistent alkylphenolic compounds are estrogenic. Endocrinology 135: 175--182. Wibbels T and Crews D (1992) Specificity of steroid hormone-induced sex determination in a turtle. Journal of Endocrinology 133: 121--129. Willemsen P, Scippo M, Kausel G, et al. (2004) Use of reporter cell lines for detection of endocrine-disrupter activity. Analytical and Bioanalytical Chemistry 378: 655--663. Wilson VS, Bobseine K, and Gray LE Jr. (2004) Development and characterization of a cell line that stably expresses an estrogenresponsive luciferase reporter for the detection of estrogen receptor agonists and antagonists. Toxicological Sciences 81: 69--77. Wilson VS, Bobseine K, Lambright CR, and Gray LE Jr. (2002) A novel cell line, MDAkb2 that stably expresses an androgen- and glucocorticoid-responsive reporter for the detection of hormone receptor agonists and antagonists. Toxicological Sciences 66: 69--81. Woodward AR, Jennings ML, Percival HF, and Moore CT (1993) Low clutch viability of American alligators on Lake Apopka. Florida Science 56: 52--63. Wrenn CK and Katzenellenbogen BS (1993) Structure–function analysis of the hormone binding domain of the human estrogen receptor by regionspecific mutagenesis and phenotypic screening in yeast. Journal of Biological Chemistry 268: 24089--24098. Yager JD and Davidson NE (2006) Estrogen carcinogenesis in breast cancer. New England Journal of Medicine 354: 270--282. Zacharewski T (1997) In vitro bioassays for assessing estrogenic substances. Environmental Science and Technology 31: 613--623. Zacharewski T (1998) Identification and assessment of endocrine disruptors: Limitations of in vivo and in vitro assays. Environmental Health Perspectives 106: 577--582. Zeyl C (2005) The number of mutations selected during adaptation in a laboratory population of Saccharomyces cerevisiae. Genetics 169: 1825--1831. Zhou D, Pompon D, and Chen S (1990) Stable expression of human aromatase complementary DNA in mammalian cells: A useful system for aromatase inhibitor screening. Cancer Research 50: 6949--6954. Zhuang Y, Katzenellenbogen BS, and Shapiro DJ (1995) Estrogen receptor mutants which do not bind 17 beta-estradiol dimerize and bind to the estrogen response element in vivo. Molecular Endocrinology 9: 457–466. Zimmer M (2002) Green fluorescent protein (GFP): Applications, structure, and related photophysical behavior. Chemical Reviews 102: 759--781.

3.10 Online Monitoring Sensors G Orellana, C Cano-Raya, J Lo´pez-Gejo, and AR Santos, Complutense University of Madrid, Madrid, Spain & 2011 Elsevier B.V. All rights reserved.

3.10.1 Introduction 3.10.2 Sensors for pH Measurements 3.10.2.1 Electrochemical pH Sensors 3.10.2.1.1 pH electrodes based on redox reactions 3.10.2.1.2 Ion-selective electrodes for pH measurements 3.10.2.2 Optical pH Sensors 3.10.2.3 Optical versus Electrochemical pH Sensors 3.10.3 Sensors for Ionic Species 3.10.3.1 Ion-Selective Electrodes 3.10.3.2 Optical Ion Sensors 3.10.4 Sensors for Dissolved Carbon Dioxide 3.10.4.1 IR Spectrometry 3.10.4.2 The pCO2 Electrode 3.10.4.3 Optical pCO2 Sensors 3.10.4.4 Miscellaneous pCO2 Sensors 3.10.5 Dissolved Oxygen Sensors 3.10.5.1 Electrochemical Oxygen Sensors 3.10.5.2 Optical Oxygen Sensors 3.10.6 Sensors for Waterborne Ozone 3.10.7 Sensors for Waterborne Hydrocarbons 3.10.7.1 Oil-Spill Detection 3.10.7.2 Water-Quality Control 3.10.7.2.1 Sensors based on refractive-index changes 3.10.7.2.2 Sensors based on light scattering 3.10.7.2.3 Sensors based on absorption changes 3.10.7.2.4 Sensors based on emission changes 3.10.8 Sensors for Waterborne Organic Matter 3.10.8.1 Sensors for COD 3.10.8.2 Sensors for BOD 3.10.8.3 Sensors for TOC 3.10.9 Waterborne Chlorophyll Sensors 3.10.10 Sensors for Waterborne Pesticides 3.10.11 Sensors for Waterborne Toxins 3.10.12 Sensors for Waterborne Bacteria 3.10.13 Turbidity Sensors 3.10.14 Oxidation–Reduction Potential Sensors 3.10.14.1 Effect of pH on Oxidation–Reduction Potential Sensors 3.10.14.2 Effect of Temperature on ORP Sensors 3.10.14.3 Frequent Problems with ORP Sensors 3.10.15 Conductivity Sensors 3.10.15.1 Effect of Temperature 3.10.16 Conclusions Acknowledgments References

3.10.1 Introduction Nowadays, sensors are regarded as the electronic mimics of the human senses. Thanks to both physical and chemical sensors, we can constantly be aware of natural and human-made phenomena. In this way, it is possible to monitor the status of watercourses, oceans, underground aquifers, water-treatment

221 221 221 222 222 223 224 224 226 229 231 231 231 232 234 234 234 236 238 239 239 240 241 241 241 242 243 243 243 245 246 247 249 252 252 254 256 256 256 256 257 258 258 258

plants, distribution networks, etc., and take corrective actions as early as possible, before any damage is inflicted. This chapter provides an overview of the currently available sensors for water-quality monitoring and analysis. The definition of a sensor includes any device that is able to provide a measurement of a physical parameter (e.g., turbidity and conductivity), or the concentration of a chemical

221

222

Online Monitoring Sensors

species (e.g., molecular oxygen, ions, toxins, and hydrocarbons) in situ, in real time and continuously. In situ implies that the monitoring device can be taken to the point of measurement and introduced directly into the target water body (e.g., a pH meter), or placed where sampling and analysis can be made without human intervention. Nevertheless, the sensor signal can be sent far away from the monitoring site to central sites where the actual control is located and decisions are made. Currently, in use are radio, global system for mobile connections (GSM), and satellite links. Real-time monitoring involves rapid response by the sensor to changes in the monitored parameter, so that immediate action can be taken if needed. However, while this is the ideal behavior of any sensing device, there are processes such as analysis of several chemical species (e.g., waterborne organic matter) that require sample pretreatment. The true sensor must perform this operation in situ and online at the expense of a longer response time. Despite this, the sensor response is much shorter than the time taken to perform manual sampling, which involves transport of samples to the laboratory and subsequent analysis. Continuous measurements imply that the sensor response is reversible (ideal situation) or, at least, it can be regenerated in situ without human intervention. The devices that do not fulfill this last feature are often called ‘dosimeters’, although many scientific and commercial publications also refer to them as ‘sensors’. This chapter is intentionally restricted to true sensors, that is, devices that meet the three requirements mentioned earlier. However, where commercially available online sensors are not accessible for an important water-quality parameter, or when manual tests are firmly established in the water-monitoring field, the corresponding dosimeters are briefly described. The dynamic range of an analyzer is normally regarded as the analyte-concentration interval that the instrument is able to measure in an effective manner, that is, from the detection limit to the maximum usable indication. The detection limit refers to the smallest amount of substance or element detectable by the sensing device (typically corresponding to an analytical signal equal to 3 times the standard deviation of the background noise). From the statistical point of view, the limit of determination is always higher than the limit of detection of the sensor. Therefore, even though many manufacturers state that a particular sensor for water monitoring is applicable within an analyte concentration range from 0 to x (units), it must always be understood that the lowest indication value is actually the limit of detection. Given the wide variety of chemical sensors developed for water monitoring and the different types of sampled waters (natural, potable, underground, industrial, recreational, recycled, wastewaters, etc.), it is impossible to provide a general critical view on the practical applicability of these devices. Nevertheless, a critical assessment of the advantages/disadvantages of the different sensors described in this chapter is included under each section.

fields. In this section, we describe pH sensors divided into two main groups, namely electrochemical and optical devices.

3.10.2.1 Electrochemical pH Sensors The potentiometric measurement of pH is based on the electrochemical cell:

Reference electrodejconcentrated KCljjtest solution jelectrode reversible to H þ ðaqÞ where, typically, the electrode reversible to hydrogen ions is a glass electrode that is assumed to exhibit Nernstian behavior. The electromotive force (emf) of this cell is given by the expression 0

E ¼ E 0 þ kN logfHþ g þ Ej

ð1Þ

where kN is the Nernst constant (RTln10/F), {Hþ} is the thermodynamic activity of the hydrogen ions, EJ is the liquid junction potential that arises from the ionic strength differences between the electrolyte solution of the reference electrode and the test solution, and E00 is the conditional potential of the cell which depends on the experimental conditions (such as the filling solution) in the pH electrode and the contribution of the reference electrode potential. During routine measurements, electrode readings in one or more reference solutions are compared to that of the test solution. Assuming that the EJ values in both the test and the reference solution are the same (DEJ ¼ 0), the pH of the test solution is, therefore, operationally defined as

pHðXÞ ¼ pHðSÞ þ ðES  EX Þ=kN

ð2Þ

where X and S indicate the test and reference solution, respectively.

3.10.2.1.1 pH electrodes based on redox reactions There are two types of indicator electrodes: metal- and carbonbased. Metal electrodes develop an electric potential in response to a redox reaction at the metal surface. The most common metal indicator electrode is made of platinum, which is relatively inert. An advantage of the metal oxide electrodes is that they have very low resistance, but may be subject to severe interference by redox reactions. Various types of carbon can be used as pH electrodes because the rates of many redox reactions on the carbon surface are fast enough. Although many such materials respond to pH without preliminary activation, the derivatization of carbon surfaces has allowed development of electrodes for pH measurements that offer advantages compared to other pH meters (Kahlert, 2008).

3.10.2.1.2 Ion-selective electrodes for pH measurements

3.10.2 Sensors for pH Measurements The acidity of water is probably the most widespread chemical parameter measured in both environmental and industrial

Ion-selective electrodes are different from metal electrodes in that the former do not depend on redox processes. The essential feature of an ideal ion-selective electrode is a thin membrane across which only the target ion can migrate. The most important ion-selective electrodes for pH determination

Online Monitoring Sensors

are glass electrodes, liquid membrane electrodes, and ionsensitive field-effect transistors (ISFETs). Glass pH electrode. The glass pH electrode used to measure water acidity is the most common example of an ion-selective electrode. The overall galvanic cell of a typical (combination) glass electrode incorporating both glass and reference electrodes can be represented by Reference electrode ( ) (internal)

H+ (internal) ( )

Glass membrane

H+ (external) analyte

Reference electrode (external)

Glass electrode

The key to electrode selectivity lies in its glass membrane. The surface layers of the latter consist of fixed silicate groups associated with sodium ions ðOSiO2  Naþ Þ. When this electrode is dipped in water, the sodium ions exchange with the solvated protons in water and the surface is then described as ‘hydrated’. The glass membrane has an inner and outer hydrated layer. In these hydrated layers, the anion sites are covalently bound to the bulk of the glass and are fixed. However, the Hþ cations are mobile, being free to exchange with the external solution or with sodium ions in the body of the glass. When the electrode is placed in an aqueous solution of unknown pH, the activity of the Hþ ions in the test solution is likely to be different from the activity of the Hþ ions in the hydrated layer. This sets up a potential difference between the solution and the surface of the membrane. This boundary potential is determined by this difference in the activities. Beckman marketed the first pH glass electrode and meter in 1935. The glass pH electrode system used nowadays consists of a pH-sensitive measurement glass electrode and a separate reference electrode in a potassium chloride (KCl) gel-conducting solution (Figure 1). These electrodes are usually housed in the combination sensor, containing both electrodes, which is connected to an electronic meter with a signal amplifier and temperature compensation. The meter displays the pH reading, which may be uploaded to a computer or controller. A silver wire enclosed in the measurement electrode forwards a signal indicating the difference in acidity between the solutions inside and outside the glass membrane. The reference electrode has a stable potential, which is independent of the measuring solution and must be calibrated outside the system in a reference solution. The most commonly used reference is a silver/silver chloride electrode in a buffer. The measurement and reference electrodes complete a circuit through the water sample (via a permeable porous junction built in the glass wall, Figure 1) allowing measurements of the voltage generated by the glass electrode. Common glass pH electrodes are extraordinary sensors in that they operate within a typical temperature range of 0– 90 1C over the full pH range of 0–14 (14 orders of magnitude of the Hþ concentration!), although they require accurate temperature measurement and compensation. The pH signal generated by a glass electrode can drift, or lose accuracy, over time due to a number of factors including fouling, sensor instability, and interference from external equipment.

223

Therefore, accurate pH measurements require an external recalibration procedure using standard solutions of known pH. Other potential pitfalls of the pH glass electrodes include fragility, difficult miniaturization, leakage of the reference electrode buffer into the sample solution, poor response in low ionic strength solutions, high background noise, and moderate signal-to-noise ratio. Liquid membrane pH electrodes. Liquid membrane pH electrodes (LMEs; including polymeric membrane electrodes) offer advantages over other types of pH sensors and their aquatic applications are promising. Their key component is a pH-sensitive membrane that contains a pH-selective material composed of a neutral proton carrier dissolved in a membrane solvent. The membrane solvent is not miscible with water forming an organic phase that separates the aqueous sample solution from the aqueous internal filling solution. The neutral carriers are capable of selectively extracting ions from aqueous solution into the membrane phase and transporting them across the organic phase by carrier translocation. Similarly to the glass electrode, a membrane potential is established during the process and can be measured against a reference electrode. Compared to glass pH electrodes, LMEs have shorter response times because of their lower inner resistance. Ion-sensitive field-effect transistors. The ISFET is an integrated device containing an ion-selective electrode and an insulatedgate field-effect transistor. In pH-sensitive FETs, the ion-selective layer consists of SiO2, Si3N4, Ta2O5, or Al2O3, currently Ta2O5 being the preferred pH-sensitive layer. Technical difficulties regarding the required encapsulation of the electronic components due to their sensitivity to moisture are the main problems of ISFET fabrication. These problems have been solved by advanced packaging technologies so that an

Wires to pH meter

Filling hole

AgCl covered silver wire Ag/AgCl reference electrode Reference electrode internal solution

Permeable porous junction Glass electrode internal solution Glass membrane Figure 1 Typical glass electrode for pH measurements.

224

Online Monitoring Sensors

economical mass production of solid pH sensors is possible. The major problems that persist are in finding a compatible reference electrode and avoiding hysteresis effects. Nevertheless, ISFETs are sensitive over a wide pH range (0–14) and are able to measure faster and with less temperature dependence than glass electrodes. Unlike the latter, ISFET-based pH sensors possess a rugged structure, small size, and low impedance, being amenable to miniaturization and automation. Moreover, ISFETs are not expected to be sensitive to organic contaminants and redox species in natural environments. Representative electrodes for pH sensing are listed in Table 1.

3.10.2.2 Optical pH Sensors Optical pH sensors, also known as pH optodes, are based on pH-dependent variations in the optical properties of an indicator dye, which reacts reversibly with the protons or bases in the aqueous sample. The most popular designs use the pHdependent absorption or fluorescence of reagents that are weak electrolytes and exist in both acidic and basic forms over the pH range of interest (typically c. 4 units). The acid (hyaluronic acid (HA)) and the corresponding conjugated base (A) participating in the pH-dependent chemical equilibrium are selected to have different absorption or fluorescent properties:

HA ðcolor AÞ" H þ þ A  ðcolor BÞ

ð3Þ

Since the degree of dissociation depends on the solution pH, the acidity level can be determined by measuring the relative concentrations of both forms of the dye. It is important to point out that most reports on pH optodes ignore the effect of ionic strength on the dissociation equilibrium of the indicator. While this omission is acceptable in dilute solutions, it can lead to serious errors in some environments. Absorption- and reflectance-based pH sensors. Absorptionbased pH sensors operate under the Beer–Lambert law that relates the absorbance at the analytical wavelength (Al) of an aqueous solution of the indicator dye to the concentration (C) of its acidic and basic forms (Equation (4)):

Al ¼ log

P0;l ¼ el ðHAÞlCHA þ el ðA ÞlCA Pl

ð4Þ

where P0,l and Pl represent the incident and transmitted spectral radiant power at the analytical wavelength of the monochromatic radiation used to interrogate the system, respectively; l is the absorption path length and el is the molar absorption coefficient at the analytical wavelength of the acidic and basic species of the indicator dye. The measured absorbance is a function of the solution pH due to the effect of the latter on the acid/base equilibrium of the indicator dye. For practical fabrication of optical pH-meters, the indicator dye is usually immobilized onto a polymer support and interrogated using optical fibers to carry the light to and from the distal end where the sensing head is placed. If the supported dye is in particulate form rather than a film, it needs to be confined and separated from the sampled water by a Hþpermeable membrane. The Beer–Lambert law also holds in

transparent polymer supports but, if the indicator dye is adsorbed onto opaque materials, then diffuse reflectance rather than absorbance must be measured. In this case, the Kubelka– Munk function, that relates the absolute diffuse reflectance of the indicator material at the analytical wavelength (Rl) to the concentration of the immobilized absorbing species (C), is used:

f ðRÞ ¼

ð1  Rl Þ2 2:303el C ¼ 2R Sl

ð5Þ

where el has been defined above and Sl is the scattering coefficient of the indicator support material at the analytical wavelength. The latter is assumed to be independent of the immobilized dye concentration. As in the case of absorptionbased pH sensors, the overall dye concentration will be distributed among its acidic and basic forms (Equation (3)) depending on the pH. Therefore, the diffuse reflected color intensity at the chosen analytical wavelength will be a function of the sampled water pH. Fluorescence-based pH sensors. Fluorescence is particularly well suited for optical sensing due to the high sensitivity and selectivity of the emission phenomenon. For weakly absorbing solutions (Alo0.05), the fluorescence intensity at the analytical wavelength (IF,l) returning from the sensing head is directly proportional to the intensity of the exciting radiation (I0) and to the concentration of the fluorescent dye (C) in the sensor:

IF ¼ k0 I0 Fel lC

ð6Þ

where l is the absorption path length through the sensing layer, el is the molar absorption coefficient, F is the fluorescence quantum yield, and k0 is the fraction of the fluorophore emission that can be measured in each particular setup. Fluorescent indicator molecules are also immobilized onto a polymer support and the emission from the fluorescent material is a function of the sample pH due to the dependence of the indicator acidic and basic forms concentration on that parameter. Absorption versus fluorescence pH sensors. Absorption measurements are simple and easy to use but are not very sensitive, requiring the use of high concentrations of pH indicator dye and/or a thick sensing layer. Reflection configurations with bifurcated fiber bundles are often used to overcome this problem. In contrast, fluorescent measurements are much more sensitive and can be used for small-size sensors and/or low indicator concentration. Fluorescent sensors can be operated in the (absolute) emission intensity mode and/or the emission lifetime mode, depending on whether steady-state or pulsed excitation of the indicator dye is performed. Fluorescence lifetimes can also be determined by sinusoidally modulating the excitation light and measuring the (sinusoidally) modulated emission phase shift. The current light-emitting diodes (LEDs) have helped to make the latter method very popular. Sensors based on fluorescence lifetime measurements are preferable because of their decisive advantages over absolute intensity recording: it is intrinsically self-referenced, it has negligible signal drift due to independence of the decay time with luminophore

Online Monitoring Sensors Table 1

225

Some academic and commercial electrochemical sensors for pH determination

pH-sensitive electrode

pH range

Slope (mV/pH unit)

Response time

Interferences

Lifetime

pH electrode type

References

W/WO3 IrOx Functionalized graphite Functionalized carbon black Functionalized glassy carbon Functionalized glassy carbon Functionalized glassy carbon Functionalized carbon epoxy Functionalized carbon, organic binder (silicone, PTFE) Polypyrrole on PTFE Octyldibenzylamine

3–6 2–12 NA 2–7 4–12

 61.1  6477  59  59 59

40 s NA 2–3 s NA NA

NA Ni2þ NA NA NA

NA 2 months NA NA NA

Redox Redox Redox Redox Redox

a

1–11

55

NA

NA

NA

Redox

f

1–11

 54.7

10 min

NA

NA

Redox

g

1–12

 60

60 s

NA

NA

Redox

h

0–9

 58

o2 s

NA

NA

Redox

i

2–12 2–10

 37.8 56.5

50–100 s NA

NA Small effect of Naþ, Kþ, Ca2þ

NA NA

Redox LME

j

(2–55 1C) 2–11 2–12 4–9 0–14 (0–95 1C) 0–14 (0–100 1C) 1–12 (o80 1C) 0–14 (25 1C) 0–14

(20 1C) NA 55–58 39–42 NA

Few seconds NA NA NA

NA NA NA NA

530 days NA NA NA

ISE ISFET ISFET Redox

l

53–60

o30 s

Naþ

NA

ISE

p

NA

NA

NA

NA

ISE

q

59

NA

NA

NA

ISE

r

NA

1–5 s

NA

ISE

s

0–14 (0–60 1C) 0–14 (o135 1C)

53.23 (25 1C) NA

NA

Naþ error at pH 4 12.3 Hysteresis

6–12 months

ISFET

t

NA

NA

NA

ISFET

u

Glass Ta2O5 film Polyelectrolyte multilayers NA Glass Glass Glass Glass NA NA a

Fenster et al. (2008). El–Giar et al. (2007). c Szepesvary and Pungor (1971). d Jankowska et al. (1981). e Lawrence and Robinson (2007). f Holm et al. (2007). g Brown et al. (1976). h Li et al. (2002). i Scholz et al. (2005) and Kahlert et al. (2004). j Prissanaroon et al. (2005). k Cho et al. (1998). l Kaden et al. (2004). m Poghossian et al. (2003). n Scho¨ning et al. (2009). o Continuously self-calibrating sensor (http://www.sensorin.com). p Thermo Scientific (http://www.thermo.com). q Orbisint Memosens Glass Electrode (http://www.emc.co.nz). r pH/ORP monitor/controllers (http://www.myronl.com). s Sensorex (http://www.sensorex.com). t Argus ISFET (http://www.sentron.nl). u Tophit-H Glassless Memosens (http://www.emc.co.nz). NA, not available; PTFE, poly(tetrafluoroethylene). b

b c d e

k

m n o

226

Online Monitoring Sensors

concentration, and fluctuations in light-source intensities and photodetector do not influence the measurements.

3.10.2.3 Optical versus Electrochemical pH Sensors The most significant advantage of spectrophotometric methods is their high precision. There are no liquid-junction or high-impedance problems with optical measurements, and calibration is inherent in prior knowledge of the thermodynamic properties of the indicator dye. However, interrogation of pH electrodes is simpler, and therefore electrochemical sensors are more economical than their optical counterparts. Moreover, the enormous dynamic range of the pH electrodes can never be beaten by the optical pHmeters. The latter can be a useful option when miniature sensors operating in a narrow pH range are needed, or when the optical sensing monitors developed successfully for other widespread water-quality parameters such as dissolved oxygen (see Section 3.10.5) can be directly used for pH measurements. A summary of optical pH sensors for water appears in Table 2.

molybdenum, sulfate, zinc, nickel, tin, etc.) can also be found in water discharged from industrial areas. Depending on the water to be monitored, the ion-concentration range may be very different. For example, according to the US Environmental Protection Agency (EPA), the highest concentration of arsenic to which an aquatic community can be exposed briefly without resulting in an unacceptable effect is 340 mg l1. However, this value is limited to only 0.018 mg l1 when considering waters for human consumption. It is obvious that a sensor system becomes more expensive with higher sensitivity, a factor that must be considered when evaluating the application needs. Some waterborne ionic species can easily be transformed into neutral ones in the gas phase by just changing the pH of the sample, facilitating their determination by indirect methods. This is the case of sulfur compounds, chlorides, and ammonium. For instance, upon acidification of the water sample to be monitored, sulfite, hydrogen sulfide, and sulfide ions yield sulfur dioxide and hydrogen sulfide gases according to the following reactions: þ SO2 3 þ 2H -SO2 ðgÞ þ H2 O

HS  þ H þ -H2 SðgÞ

3.10.3 Sensors for Ionic Species S 2 þ 2H þ -H2 SðgÞ The monitoring of ionic species in water is of particular interest when considering drinking-water applications or, in general, public health. However, it is also relevant in other aspects of environmental analysis, such as aquatic life or industrial processes. The relative importance of monitoring the different waterborne ions depends on the particular nature of the aqueous medium. Aluminum compounds, for instance, can be found in swimming pools due to their use as flocculating agents, or in wastewater discharges from aluminum smelting where it becomes harmful to aquatic life. The presence of ammonium in groundwaters is an indication of potential pollution. Cadmium and lead ions are associated with the industrial production of batteries, while calcium and magnesium ions are responsible for water hardness. Chloride and bromide are ubiquitous and, in forced irrigation systems, groundwater can be monitored for these ions to avoid excess salinity. The presence of phosphates is important for plant life, as they are nutrients; yet if a wastewater rich in phosphate detergents is found due to insufficient monitoring, it produces environmentally damaging algal blooms. Another element essential to plant growth is sulfur, which can also be harmful to humans if present as sulfite ion, a potent allergenic species. Sulfite and hydrogen sulfide are often used to eliminate residuals of chlorine in wastewaters, due to the strong reducing potential of these ions. However, dissolved free sulfides strongly promote corrosion of many metals. Fertilizers are rich in nitrogen compounds such as nitrates that are harmful when they are found at high levels in water. Nitrites and nitrates can also be pollutants in aquarium waters. Both potassium and sodium cations are present in drinking, process, and wastewaters, their monitoring being relevant in the food industry as well as in horticulture as components of fertilizers. Several other ionic species (chromate, cobalt, copper, iron, lead, manganese,

Obviously, the need for a pH change in the water sample requires more than a simple point-sensitive device. Moreover, many ion sensors require sample preconditioning. Therefore, manufacturers have developed total analytical systems containing fluidic tubing, peristaltic pumps, reagent reservoirs, sensing chambers, and cleaning devices that can be deployed in situ for continuous monitoring of waterborne ions. These systems require more frequent maintenance servicing than the simple sensors, and therefore are more expensive to operate. It has to be borne in mind that personnel costs are always much higher than the cost of any monitoring sensor network, a fact that must be carefully considered when evaluating the monitoring needs of a particular application. Most of the devices for waterborne ionic-species monitoring rely on electrochemical processes for detection of the analyte, but there have emerged in the market, an increasing number of instruments based on optical-detection schemes, due to the advantages they offer. However, many commercial systems or devices described in the literature for ion sensing cannot be defined as true sensors (see Section 3.10.1), but are rather dosimeters because of their irreversible response. However, due to their widespread use, they have been included in this subsection to let the reader judge whether they can be useful for his/her particular application.

3.10.3.1 Ion-Selective Electrodes Sensors based on electrical measurements (potentiometric, amperometric, and conductometric) are widely used for field applications due to their low cost, simplicity, and robustness. When available (see, for instance, Section 3.10.2), they allow

Table 2

Some optical pH sensors found in recent literature and commercial sources

Chemical reagent

l (nm)

pH range

Precision

Response time

Temperature range

Interferences

Lifetime

Transduction principle

References

Thymol blue p–Methyl red

435 540,

Seawater pH 1.0–3.5

0.001 pH unit NA

NA 15.8 s

NA 1–3.5 1C

NA NA

4 weeks NA

Absorption Absorption

a

4-CP-BPB Bromocresol green

445, 606 610

2.0–6.0 4.0–7.0

20% (RSD)

11.9 s 5 min

NA

NA

NA

Absorption

c

PoAnis/TSA

501

4.9–10.5

0.01 pH unit

NA

Ionic strength (Naþ, Liþ, Cl)

NA

Reflectance

d

NA

5 min (acid to basic) 5–22 min (basic to acid) r60 s

NA

NA

NA

Reflectance

e

2.0–10.0 Thymol blue

NA

Congo red Bromothymol blue

6.8–9.5

b

7.9–11.2 3.3–4.8

Organometalcarbonyl complexes

2100–1750 cm1

7–13

NA

NA

NA

NA

NA

Reflectance

f

Swellable hydrogel

NA

4–5

NA

r30 s

NA

NA

NA

g

Langmuir– Blodgett film

750, 780

11–13

0.001–0.1

E20 s

NA

NA

NA

Swellingdependent reflectance Evanescent wave absorption

[Ru(bpy)2(dhphen)] (ClO4)2

Exc. 415

1–8

r0.11 pH unit

2–5 min

NA

Oxygen

2 weeks

Fluorescence

i

Em. 612 Exc. 300

3.5–9.2

r0.5 pH unit

2 min

2073 1C

Cd2þ, Cu2þr5 mg l1

2m

Fluorescence (energy transfer)

j

Em. 535 Exc. 405/450

6.5–8.5

NA

r5 s

NA

Ionic strength

50 days

Fluorescence

k

1.5–5.0

NA

r60 s

NA

Ionic strength

1 month

Fluorescence

l

7.5–9.0

0.1 pH unit

r230 s

2172 1C

Temperature

NA

Fluorescence

m

1–9

0.01–0.06 pH unit NA

E1 min

NA

Not found

8 months

DLR Fluorescence

E90 s

NA

Not found

1 month

Fluorescence

BNS, BrN/PhR, BCP, BPB

HPTS TAPP/MBTD DHFA/DHFAE/ Ru(dpp)3 Mercurochrome APN /MBTD

Em. 520 Exc. 422/481 Em. 656/528 Exc. 516/468/ 530/505 Em. 540/554/600 Exc. 506 Em. 530 Exc. 393/479 Em. 524/530

5.80–8.80

h

n

o

(Continued )

Table 2

Continued

Chemical reagent

l (nm)

pH range

Precision

Response time

Temperature range

Interferences

Lifetime

Transduction principle

References

MAHPDS

6.5–9.0

0.05 pH unit

r216 s

NA

NA

NA

Fluorescence

p

5.5–8.6

NA

2 min

NA

NA

NA

Fluorescence

q

Phenol red

Exc. 404/457 Em. 510 Exc. 406/460/ 506 Em. 515/540 525

6.2–8.4

0.1

20 s

NA

NA

Absorption

r

NA

NA

3.5–8.5

0.005 pH unit

o40 s

2–50 1C

Total alkalinity: 40–140 ppm Ionic strength

2 years

Fluorescence

s

Phenol red

NA

6.5–8.5

0.1

NA

NA

fluorescent molecules NA

NA

Absorption

t

0.1

NA

NA

NA

NA

Reflectance

t

HPTS

Cresol red m-Cresol purple Thymol blue Brilliant yellow Phenol red Phenol red nylon Cresol red m-Cresol purple Thymol blue Brilliant yellow a

8.0–10.0 8.5–10.5 9.0–12.0 NA

7.0–9.0 6.5–8.5 8.0–10.0 8.5–10.5 9.0–12.0 9.0–12.0 7.0–9.0

Bellerby et al. (2002). Wong et al. (2005). c Lau et al. (2006). d Taboada Sotomayor et al. (1997). e Wro´blewski et al. (1998). f Creaser et al. (2002). g Michie et al. (1995). h Flannery et al. (1997). i Chan et al. (1998). j Jin et al. (2001). k Hulth et al. (2002). l Niu et al. (2005). m Schroeder et al. (2005). n Sanchez-Barragan et al. (2005). o Li et al. (2006). p Vuppu et al. (2009). q Aller and Zhu (2006). r eXacts Micro 7 þ pH (http://www.sensafe.com). s Optical pH sensors (http://www.polestartech.com). t Fiber optic pH sensor (http://www.oceanoptics.com). NA, not available; CI: confidence interval; RSD: relative standard deviation; 4-CP-BPB: (4-carboxyphenyl)-bromophenol blue; TBPSP: 3,4,5,6-tetrabromophenolsulfonephthalein; PoAnis/TSA: poly(omethoxyaniline) doped by p-toluene sulfonic acid; bpy: 2,20 -bipyridine; dhphen: 4,7-dihydroxy-1,10-phenanthroline; BNS: 6-bromo-2-naphthyl sulfate; BrN: a-bromonaphthalene; PhR: Phenol red; BCP: Bromocresol purple; BPB: Bromophenol blue; HPTS: 8-hydroxypyrene 1,3,6-trisulfonic acid trisodium salt; TAPP: meso5,10,15,20-tetra-(4-allyloxyphenyl)porphyrin; MBTD: N-(2-methacryloxyethyl)benzo[k,l]thioxanthene-3,4-dicarboximide; DHFA: 20 ,70 -dihexyl-5(6)-N-octadecylcarboxamidofluorescein; DHFAE: 20 ,70 -dihexyl-5(6)-N-octadecyl-carboxamidofluorescein ethyl ester; Ru(dpp)3: tris(4,7-diphenyl-1,10-phenanthroline)ruthenium(II); DLR: dual lifetime referencing; Mercurochrome: 20 ,70 -dibromo-50 -(hydroxymercuri)fluorescein; APN: N-allyl-4-piperazinyl-1,8-naphthalimide; MAHPDS: 6-methacryloyl-8hydroxy-1,3-pyrene disulfonic acid; FITC-dextran: fluorescein isothiocyanatedextran. b

Online Monitoring Sensors

in situ, continuous monitoring of the analyte over a wide concentration range and are unaffected by water parameters such as color and turbidity. Yet, electrochemical devices for ion sensing may lack the required precision and suffer from interfering species (false-positive readings, explained later). The so-called ion–selective electrodes (ISE; the pH electrode is actually an ion-selective electrode) are commercially available for most ionic species commonly present in water samples, namely ammonium, chloride, iodide, fluoride, nitrate, potassium, sodium, and many heavy metals. Amperometric measurements of a galvanic cell based on the electrical potential created across an ion-specific solid membrane due to the presence of the target ion, versus that of a reference electrode (Figure 2), allows determination of the analyte ion activity. The analyte-sensitive membrane is placed at the bottom of a small reservoir containing the (internal) reference electrode held at constant potential, and the sensor is introduced into the sampled water. For instance, the ISE for fluoride in water contains as an analyte-selective membrane, a single LaF3 crystal doped with EuF2 to create defects that improve conductivity of the F ions within the solid. However, the calcium ISE is based on a plasticized poly(vinyl chloride) (PVC) membrane containing an organic receptor that binds and transports Ca2þ across the film. This transportation is responsible for the buildup of an electrical potential between the two sides of the membrane, so that the measured current is proportional to the concentration of the target ion in the aqueous medium where the ISE is immersed. Table 3 summarizes some of the instruments, available in the market, that use ISEs for ion monitoring in water. This chapter does not aim to provide an extensive description of every commercial instrument, but to provide a general outline on the state of the art of this technology. In situ continuous sensing of waterborne bromide, calcium, chloride, and fluoride using solid-state transducers (i.e., the

mV Meter

ISE

RE

AgCl-coated wires

Internal electrolyte

Ion-selective membrane

AgCl-saturated KCl aq.solution

Liquid junction

Figure 2 Typical combination ion-selective electrode.

229

inner electrolyte is replaced by a semi-solid paste), and ammonium and nitrate using PVC-coated electrodes, is provided by Nexsens ISE WQSensors (Table 4). All of them show measurement reproducibility better than 4% and are based on screw-cap replaceable sensing modules, simplifying maintenance operations. They have a 1.8-m USB connector cable for direct data collection by a computer and for freeing them from individual power supplies. The Nico2000 ELIT ISE sensors (Table 5) are another example of commercial electrochemical sensors. The two-head configuration (reference þ ISE) allows the user to replace only the electrode without changing the head. A multicomponent analysis is offered with a configuration of one reference with up to six ion sensors. With a response time of just 10 s, this company offers an impressive list of sensors for waterborne ions. The Yellow Spring 6000 series is an example of combined multiparametric electrochemical/optical sensing modules (see below). Unlike the ISEs mentioned above, these sensor modules have their own power supply to improve the positional freedom for in situ monitoring. Capable of working from  5 to 50 1C, its chloride, nitrate, and ammonium sensors seem ideal for nutrient environmental monitoring. The latter two contain a PVC membrane impregnated with a specific reagent yielding a working range of 0–200 mg l1 (0.001–1 mg l1 resolution), while the chloride ISE uses a solid-state membrane that allows monitoring from 0 to 1000 mg l1 (with the same resolution as the other two). Horiba Process & Environmental Sensor Technology (NJ, USA) has developed a multiparameter sensing probe capable of simultaneously analyzing up to 13 different parameters including a variety of ions in the temperature range of 0–55 1C. The W-20 series is a probe with high-pressure tolerance allowing measurement of nitrate, chloride, and fluoride at depths up to 100 m in rivers, lakes, or even in the open sea, accompanied by its built-in memory capacity of 1 month data logging. One of its attractive features is its global positioning system (GPS) module that allows acquisition of the location and time of each measurement for detailed three-dimensional (3D) records. Both Analytical Technology Inc. (ATI) and Hach market sensors that rely on the change of the sample acidity to extract the analyte into the headspace, as described earlier. The ATI’s A15/79 total residual chlorine sensor actually measures the concentration of gaseous iodine in an indirect method, measuring 0.01–2000 mg l1 analyte in 3 min. It uses potassium iodide and a pH 4 buffer as reagents for the hypochlorite determination in water:

HClO þ 2KI þ HCl-I2 ðgÞ þ 2KCl þ H2 O

3.10.3.2 Optical Ion Sensors Optical sensors based on the absorption of light by the analyte ion and rugged miniature spectrometers are becoming very popular for in situ water monitoring due to their particular advantages compared to their electrochemical counterparts: no need of electrolyte, electrode, or membrane maintenance; sturdiness of the sensitive optical probes; absence of drift due

230

Online Monitoring Sensors

Table 3

Analytical data for some commercially available electrochemical sensors for waterborne ions

Ionic species

Sensor model

Dynamic range (mg l1)

Accuracy

Response Time (s)

Observations

Agþ, Br, NO3  , Liþ, ClO4  , Ca2þ, Naþ, Kþ, NH4 þ , S, Cl, CN, F, I, SCN Br, Cd2þ, Ca2þ, Cl, Cu2þ, CN, F, I, Pb2þ, SO4 2 , NO3  , Kþ, Agþ, S2 Br, Ca2þ, Cl, F, NH4 þ , NO3  ClO/Cl2 (as total chlorine)

AB 6000 seriesa

NA

NA

NA

Single-parameter monitoring

HI98184 and HI98185b

0.01–saturation (depending on the analyte) See Table 4

NA

NA

Multiparameter monitoring

NA

NA

0–0.2; 0–2; 0–20; 0–200 0–5 0.001–2; 0.01–10 0–50 (depending on the analyte) 0–200; 0–1000 (see text)

0.02 mg l1

60

Replaceable membrane, single-parameter sensors Polarographic gas sensor

8% NA NA

180 120 NA

Multiparameter monitoring Single-parameter monitoring Multiparameter monitoring

10%

NA

10% 5% 30%

o260 90 NA

Replaceable membrane, single-parameter monitoring Single-parameter monitoring

NA NA 10% 5% o5%

NA NA 120 180 o300

Gas-sensitive electrode

10% NA

NA 600

Single-parameter monitoring Multiparameter monitoring

10%

10

Replaceable membrane, multiparameter system

10% N and 15% Cl

60

Multiparameter monitoring

NA 10% 10% 10%

60 NA NA NA

Multiparameter monitoring Single-parameter sensors Multiparameter monitoring

0.03 mg l1 0.03 mg l1

180 180

Polarographic gas sensor Polarographic gas sensor

WQSensorsc Q45H/62-63d

Cl2, ClO2, F

PCA 330 seriesb Chlori::lyserTMe Conexs DIAf

Cl, NH4 þ and NO3 

YSI 6000 seriesg

F

CA610h A15/82d IF–250i

Kþ Naþ

C–131i C–122i SODITRACEj 9245h AMTAXTM h

NH4 þ , Ba2þ, Br, Cd2þ, Ca2þ, Cl, Cu2þ, CN, F, I, Pb2þ, Hg2þ, NO3  , NO2  , ClO4  , Kþ, Agþ, Naþ, S2, SCN NH4 þ , NO3  , Cl

ELITl

0.1–10 0–1; 0–1000 0–20; 0–200; 0–2000; 0–10 000 339–3900 23–2300 1  106–10 1  105–10 0.02–5; 0.05–20; 1–100; 10–1000 0.1–1000 0.003–1000 (depending on the analyte) See Table 5

TROLL 9500m

0.14–14 000 N

NH4 þ , NO3  , Kþ NO3 

Ammo::lyserTM proe Monitor FAM Nitratej B-343i W-20 Seriesi

NH4 þ

þ









NH4 , Cl , CN , F , NO3 , NO2 

NO3  , Cl, Ca2þ, F, Kþ

SO3 2 H2S, S2 (as dissolved sulfide) a

Monitor FAM ammoniumj ES 9010k

A15/66d A15/81d

ASTI (http://www.astisensor.com). HANNA instruments (http://www.hannainst.com). c Nexsens (http://www.nexsens.com). d ATI (http://www.analyticaltechnology.com). e S::can (http://www.s-caNAt). f Grundfos Alldos (http://www.grundfosalldos.com). g YSI (http://www.ysi.com). h Hach (http://www.hach.com). i HORIBA (http://www.horiba.com). j SWAN (http://www.swan.ch). k Environnement S.A. (http://www.environnement-sa.com). l NICO 2000 (http://www.nico2000.net). m In-Situ Inc. (http://www.in-situ.com). NA, not available. b

0.35–35 500 Cl 0.1–1000 0.1–1000 14–1400 0.02–62 000 (depending on the analyte) 0–20; 0–2000 0–20; 0–2000

Online Monitoring Sensors Table 4

Technical details on Nexsens ISE WQsensors

Table 5

231

Technical data for the ELIT ion sensors

Ion

Working range (mg l1)

Temperature range (1 C)

Known interferents

Ion

Dynamic range (mg l1)

Temperature range (1 C)

Known interferents

Br

0.4–79 900

0–80

Agþ Ba2þ

0.01–107 900 0.5–13 700

0–80 0–50

Ca2þ

0.02–40 000

0–40

I, Cl, S2, CN, NH3 Pb2þ, Hg2þ, Si2þ, Fe2þ, Cu2þ, Ni2þ, NH3, Naþ, Liþ, Trisþ, Kþ, Ba2þ, Zn2þ, Mg2þ CN, Br, I, OH, S2, NH3 OH Naþ, Kþ ClO4  , I, ClO3  , F

Br

0.4–80 000

0–80

Ca2þ

0.02–40 000

0–50

Cd2þ

0.1–11 000

0–80

Cl

1–35 000

0–80

ClO 4 CN 2þ Cu

0.2–99 600 0.03–260 0.006–64 000

0–50 0–80 0–80

F Hg2þ I

0.06–2000 0.2–201 000 0.06–127 000

0–80 0–80 0–80

Kþ Naþ NH4 þ NO3 

0.4–39 000 0.05–20 000 0.03–9000 0.3–62 000

0–50 0–50 0–50 0–50

NO2 

0.5–460

0–50

Pb2þ

0.2–20 800

0–80

S2 SCN

0.003–32 000 1–5800

0–80 0–80

Hg2þ, S2þ Ca2þ, Kþ, Naþ, Mg2þ, NH4 þ , Sr2þ Ag þ , CN, I, S2, Cl 3þ Al , Ba2þ, Fe2þ, Cu2þ, Sr2þ Agþ, S2, Cu2þ, Fe2þ, Fe3þ, Hg2þ, Pb2þ Br, CN, I, S2, Agþ Cl, I, NO3  , SCN I, S2, Agþ Agþ, Br, Cd2þ, Cl, Fe2þ, Hg2þ, S OH Agþ, S2 CN, S2, Agþ, S2 O3 2 Csþ, NH4 þ Most cations Kþ BF4  , Cl, ClO4  , CN,I, NO 2, HCO3 CN, CH3COO, F, Cl, NO3  , SO4 2 Agþ, S2,Cd2þ, Cu2þ, Fe2þ, Fe3þ, Hg2þ Ag þ , Hg2þ Br, Cl, I, Ag þ , S2, S2 O3 2

Cl

0.18–35 500

0–80

F NH4 þ NO3 

0.02 to saturation 0.014–1400 (N) 0.1–14 000 (N)

0–80 0–50 0–50

to a reference electrode; lack of electrical interferences; and ease of miniaturization. However, they may be subject to interference due to turbidity or the presence of species other than the analyte absorbing in the same region (e.g., dissolved organic matter or ions other than the monitored ones). For example, S::can (Austria) provides robust sensors for nitrate and nitrite monitoring by measuring the ultraviolet (UV)–visible (VIS) absorption spectrum of these ions in the water. Determinations are possible by using chemometrics, a term coined in the 1970s to design the use of statistical methods for the analysis of (instrumental) analytical chemistry data (Brereton, 2007). Registration of thousands of spectra from known samples allows training the optical sensors for recognizing the analyte pattern and quantifying it in the actual (complex) water matrices. The multiparameter Tethys UV400 instrument (Meylan, France) also relies on spectral absorption by the sample to provide quantitative information on ammonium, nitrate, phosphate, and hydrogen sulfide. By measuring the UV light absorption at 210–220 nm, the instrument provides a working range of 0–100 mg l1 for nitrate. The ammonium-detection method is based on increasing the pH of the water by the addition of sodium hydroxide to transform all dissolved NH4 þ into gaseous NH3, which has a distinct absorption around 200 nm. The sensor is able to measure 0–100 mg l1 of ammonia free of interferents since the detection occurs in the gaseous phase. By acidifying the sample upon addition of hydrochloric acid, this device enables the extraction of gaseous H2S from the dissolved HS. Phosphate determinations are made by colorimetric measurements, yielding a working range of 0–2 mg l1. Several other instruments also rely on absorption methods for multiparameter sensing purposes. The Swan AMI Phosphate monitor (Hinwill, Switzerland) for automatic and continuous measurements of 0.01–10 mg l1 phosphate in water uses the ammonium molybdate ISO 6878 colorimetric method to work for up to 6 months, with a response time of 10 min within a temperature range from  10 to 50 1C. Hach (Loveland, CO, USA) has developed several continuous monitoring single- and multiparameter optical chemical sensors. For instance, the MO42 Molybdate Analyzer

uses colorimetric catechol chemistry detection at 420 nm to monitor molybdenum oxoanions with a limit of detection of 0.03 mg l1 and a dynamic range of 0–5 mg l1. Being capable of working for up to 1 month of unattended operation, it shows readings every 2.5 min, ensuring proper monitorization. A cuprethol colorimetric method enables the APA 6000TM instrument to analyze copper (II) at 436 nm in two separate ranges, 0.05–2.0 mg l1 and 1–10 mg l1. This device is also capable of 1 month of unattended operation, and it measures readings every 4 min, yielding results with a resolution of 0.001 mg l1. The same company also provides several solutions to monitor water hardness (magnesium and calcium), using both the APA 6000TM low-range hardness analyzer and the SP510 instruments, monitoring in the working range of 0.05–10 mg l1 (as CaCO3). As in other instruments that use optical methods for the detection of nitrate and nitrite, the Hach NitrataxTM monitor takes advantage of the molecular N–O bond UV light absorption to obtain its concentration, using a second beam of light to eliminate interference from turbidity and dissolved organic matter. This reagent-free UV-absorption technique gives a dynamic range from 0.1 to 100 mg l1 (as N) with a resolution of 0.1 mg l1.

232

Online Monitoring Sensors

PhosphaxTM is the Hach instrument for phosphate optical detection. With a self-cleaning membrane, it is capable of detecting from 0.05 to 50 mg l1 (in phosphorous) with a 5-min response time and 3 months of unattended operation. Table 6 lists some of the instruments that use optical methods for continuous detection of ionic species in water. When considering optical monitoring of waterborne ions, there are very few instruments capable of in situ continuous sensing (basically nitrates and phosphates). However, there are several portable devices capable of detecting numerous

Table 6

ions in water samples by optical methods that require human intervention for sampling, conditioning, and testing operations. For instance, a portable photometer, such as the Hach DR 2700TM, provides fast results with its multiwavelength capability. It is able to detect an impressive number of ionic species with adequate limits of detection. Test strips are also a good example of widespread simple optical dosimeters as the changes in color of an immobilized specific reagent in the presence of the target analyte yield direct quantification. Even though this type of sensing

Some of the optical instruments commercially available for ionic species monitoring

Ionic species

Sensor model

Dynamic range (mg l1)

Accuracy

Response time (s)

Remarks

Cl2, ClO2 Cu2þ

AMI CODES-II CCa CL17b APA 6000TMb

0–1; 1–3; 3–5 0.035–5 0.05–2; 1–10

0.01; 0.06; 0.2 mg l1 5% 5%

120 o150 o240

Mo6þ

MO42b

0.03–5

5%

o150

NO3 2

ISUS V3c nitro::lyserTMd

0.007–28 0–70

0.028 mg l1 3%

NA NA

DPD method DPD method Colorimetric cuprethol chemistry Colorimetric catechol chemistry UV absorption UV–VIS spectrometry over the total range, multiparameter probe

multi::lyserTMd spectro::lyserTMd

3% 3%

NA NA

TPNA–300e

0–70 From 0 to 50 (N) (depending on the analyst) 0–2 (N)

NA

3600

YSI 9600f

0.025–10

5%

NA

NITRATAXTMb PHOSPHAXTMb

0.1–100 0.05–15; 1–50

5% 2%

60 o300

Series 5000b TPNA–300e

0.2–50 0–0.5 (P)

5% NA

660 3600

Monitor AMI Phosphatea UV400g

0.01–10

NA

600

From 0 to 79 (depending on the analyse)

NA

5



NO3 , NO2

PO4



3

NH4 þ , HS, S2, NO3  ,NO2  , PO4 3

Measures total nitrogen through alkaline potassium peroxodisulfate UV oxidation–UV absorption method Cadmium reduction– diazotization colorimetric method UV absorption Colorimetric molybdovanadate chemistry Measures total phosphorus by potassium peroxodisulfate UV oxidation–molybdenum blue absorption method Ammonium molybdate colorimetric method UV absorption and colorimetric Multiparameter instrument

a

SWAN (http://www.swan.ch). HACH (http://www.hach.com). c Satlantic (http://www.satlantic.com). d s::can (http://www.s-canat). e HORIBA (http://www.horiba.com). f YSI (http://www.ysi.com). g TETHYS Instruments (http://www.tethys-instruments.com). NA, not available. b

Online Monitoring Sensors Table 7

233

Examples of the test strips for ion analysis offered by three manufacturers

Ion

Dynamic range

Manufacturer

Ion

Dynamic range

Manufacturer

Al3þ

10–250 mg l1 0.01–0.25 mg l1 10–400 mg l1 N 1–50 mg l1 N 0.1–3 mg l1 0.005–0.5 mg l1 10–100 mg l1 0.5–20 mg l1 25–500 mg l1 50–500 mg l1 0.02–2 mg l1 0.1–50 mg l1 3–100 mg l1 10–1000 mg l1 10–300 mg l1 0.5–5 mg l1 0.05–2 mg l1 0.5–10 g l1 0.05–1 mg l1 5–100 mg l1 3–500 mg l1 0.1–3 mg l1 0.005–0.3 mg l1 0.02–5 mg l1 0.3–50 mg l1 20–500 mg l1 3–600 mg l1 (with photometer detection) 0.02–1.5 mg l1 2–100 mg l1 0.05–4 mg l1 0.02–1.6 mg l1

Merck Jenway Merck Jenway Merck

Mo6þ

5–250 mg l1 0.3–18 mg l1 50–1000 mg l1 2–80 mg l1 10–500 mg l1

Merck Jenway ITS

Merck Merck

NO 3

10–500 mg l1 1–30 mg l1 N

Merck Jenway

0.5–50 mg l1 100–3000 mg l1 2–80 mg l1 0.01–0.5 mg l1 N 0.15–10 mg l1 10–500 mg l1 0.05–4 mg l1 PO4 250–1500 mg l1 0.5–12 mg l1 200–1600 mg l1 2–100 mg l1 SO4 10–400 mg l1 0.05–4 mg l1 SO3 10–200 mg l1

ITS Merck

10–250 mg l1 0.02–1 mg l1 Zn 2–100 mg l1

Merck Jenway ITS

NH4 þ As3þ/5þ Ca2þ Cl

3þ/6þ

Cr

CrO4 2 Co2þ Cuþ/2þ

þ

Ag

Fe2þ/3þ

Pb2þ F Mn2þ

Hg2þ Ni2þ

ITS Jenway ITS Merck Merck Merck Jenway; ITS ITS Merck ITS Merck Jenway ITS

Merck ITS Jenway Merck Jenway ITS

mechanism does not constitute a true sensor (see Section 3.10.1), colorimetric test strips are very much in use for in situ semi-quantitative quick detection of ionic species in water samples. Several companies offer such convenient strips: for instance, the Merckoquants strips (Merck, Germany) allow visual detection and quantitation of numerous ions in aqueous media. Some test-strip methods require the use of a portable or handheld reflectophotometer to improve accuracy of analyte determination and reproducibility. Cases in point are the Jenway environmental test kits (Belgium) or Industrial Test Systems Inc. strips with their 3 mg l1 lead-detection test using the Hach LeadTrak Pocket Colorimeter II providing limits of detection lower than the US EPA requirement (15 mg l1). Table 7 summarizes the ions covered by the test strips offered by the above-mentioned manufacturers.

3.10.4 Sensors for Dissolved Carbon Dioxide Carbon dioxide is the major end product of organic carbon degradation in almost all marine environments. Fluctuations of the CO2 level are related to the net ecosystem metabolism. Four parameters define the marine CO2 system: pH, pCO2, dissolved inorganic carbon, and total alkalinity. The solubility of CO2 in the water is about 28 times that of other

NO 2

PO4 3 Kþ SO4 2 SO3 2 Sn2þ

Zn2þ

Merck

Jenway ITS Merck Jenway Merck Jenway Merck Jenway Merck Jenway Merck

hydrophobic gases such as O2 or N2. One of the most popular methods for carbon dioxide determination in the gas phase is infrared (IR) spectrometry. For measurements in aqueous solutions, the Severinghaus pCO2 electrode is the most common potentiometric sensor.

3.10.4.1 IR Spectrometry Carbon dioxide can be analyzed by IR spectrometry because of its strong stretching bands at 2350 and 650 cm1. By measuring the intensity of these bands, CO2 is quantified. This feature may be used to determine the carbon dioxide generated after acidification of aqueous samples and to calculate in this manner the total inorganic carbon (TIC) of the water. Total inorganic carbon (also called dissolved inorganic carbon (DIC)) includes all the carbon-containing inorganic species present in solution, namely, CO2, H2CO3, HCO3  , and CO3 2 . By acidification of the water sample, the acid–base equilibria of these species are driven to CO2 production. Nevertheless, the application of IR spectrometry to TIC determination is limited due to the strong IR absorption of water as well as the long optical path lengths required for analyses in the gas phase.

234

Online Monitoring Sensors

3.10.4.2 The pCO2 Electrode The equilibrium between the gas phase and water obeys Henry’s law, which states that, for an ideally diluted solution, the gas vapor pressure of a volatile solute is proportional to its mole fraction in the aqueous solution. For surface waters (i.e., at atmospheric pressure), Henry’s law can be simplified to Equation (7):

½CO2  ¼ KH pCO2

ð7Þ

where KH represents the so-called Henry’s constant and pCO2 is the partial pressure of this gas. The pCO2 values can be determined by using an electrode such as the Severinghaus one, based on the detection of pH changes in an internal HCO3  aqueous solution caused by the incoming CO2. This solution is entrapped between a glass pH electrode and a hydrophobic gas-permeable membrane. The latter permits the flow of small, uncharged (gas) molecules such as CO2, but prevents the entrance of charged molecules (hydrogen ions, cations, anions, etc.). In natural aquatic environments, pCO2 varies widely between 0.1–104 (e.g., coastal ocean surface waters) and almost 200 atm (nearshore sediments). The Severinghaus electrode presents important drawbacks: interferences caused by basic or acidic gases, slow response time, and effects of osmotic pressure caused by variable salt conditions in the sample and in the inner electrolyte.

3.10.4.3 Optical pCO2 Sensors Like pH sensors, optical pCO2 sensors (optodes) for water are an alternative in specific applications. While electrochemical sensors employ a glass pH electrode to monitor pH changes, optical devices rely on immobilized pH-sensitive indicator dyes (see Section 3.10.2.2) to accomplish the pH transduction process within the internal electrolyte. As in electrochemical pCO2 sensors, the acidity of the internal solution at equilibrium depends on the concentration of carbonic acid produced upon hydration of the permeated CO2, which in turn is proportional to the partial pressure of the analyte in the sample. Based on the pH changes produced by CO2 diffusion through a gas-permeable membrane in an internal reservoir of hydrogen carbonate buffer, YSI Life Sciences (Yellow Springs, OH, USA) commercialized the first optical pCO2 sensor. The latter uses the pH-sensitive fluorescent dye hydroxypyrene trisulfonic acid (HPTS, pyranine) and ratiometric fluorescent measurements (measuring the green emission upon successive excitation at two wavelengths) to determine the dissolved CO2 concentration. Mills et al. introduced a new scheme to design optical pCO2 sensors. They incorporated a pH-sensitive dye into a hydrophobic polymer membrane (e.g., cellulose acetate butyrate) and replaced the hydrogen carbonate internal buffer by a lipophilic hydrated quaternary ammonium hydroxide. Depending on the pKa and concentration of the indicator dye used, such sensors allow quantification of trace levels of CO2 and show a fast response. However, the membranes tend to fog after prolonged immersion in water and, sometimes, are prone to dye leaching. A thorough description of the different

types of optical sensors for CO2 measurements can be found in the review by Mills and Eaton (2000). Orellana and co-workers (1992) patented a CO2 sensing mechanism based on luminescent Ru(II) polyazaheterocyclic complexes immobilized in hydrogels, that undergo irreversible proton transfer in their excited state from various Bro¨nsted acids. The polymer-supported indicator dye is separated from the sample by a thin silicone membrane. Permeation of the CO2 into the gel phase modifies the concentration of the internal buffer species (with different proton transfer ability, e.g., hydrogen phthalate and phthalic acid) changing the Hþtransfer quenching of the luminescent indicator dye (Figure 3). Both emission intensity and luminescence-lifetime-based interrogation can be used to fabricate the sensor using the same instrumentation than the one developed recently for dissolved oxygen sensing (see Section 3.10.5). This version is used by OptosenTM Interlab IE (Madrid, Spain) in the luminescent dissolved O2/CO2 monitors that they market. Ru(II) polypyridyl complexes can also be used in combination with colorimetric indicator dyes to manufacture luminescent sensors based on Fo¨rster resonance energy transfer (FRET) from the photoexcited metal dye (donor) to the coimmobilized colorimetric indicator (acceptor). Permeation of CO2 into the gel phase containing the two dyes lowers its pH leading to a color change. The spectral shift of the acceptor provokes a variation in the FRET efficiency with concomitant change in the emission lifetime of the donor. This principle forms the basis of the PreSens (Regensburg, Germany) dissolved CO2 monitoring system.

3.10.4.4 Miscellaneous pCO2 Sensors Martek Instruments (Raleigh, NC, USA) offers a dissolved carbon dioxide analyzer based on conductivity measurements carried out before and after degassing the water sample. The conductivity differences are due to the carbonate and bicarbonate species that carbon dioxide forms when it is in solution. Table 8 provides a summary of the prototype and commercial sensors for dissolved CO2 measurements.

* Ru

+ HB

* Ru –H+ +

B−

Ru

+ HB

Ru –H+ +

B−

Figure 3 Working principle of the luminescent CO2 sensor based on photoinduced proton transfer to excited Ru(II) polypyridyls (Orellana et al., 2000). The ground state complex is completely non-protonated (pKa ¼  1.9); however, its basicity increases more than 106-fold in its excited state due to the high-acceptor character (low-lying p* orbital) of the pyrazine ligands. Therefore, it undergoes efficient (irreversible) proton transfer from suitable Bro¨nsted acids present in the reservoir indicator phase (phosphate, hydrogen phthalate, acetic acid, H3Oþ, etc). The incoming CO2 hydrolyzes and reversibly increases the HB/B ratio leading to strong luminescence quenching of the indicator dye.

Table 8

Some prototype and commercial sensors for carbon dioxide determination in aqueous solution

Dynamic range (ppm)

Limit of detection (LOD) (ppm)

Precision

Response time

Recovery time

Temperature(s) tested (1 C)

Interferences

Lifetime

Transduction principle

References

30–180 200–1000 NA 0.044–880 0.17–880

NA NA NA 0.044 NA

5–120 s o130 min 42.6 s o126 s 16 s

NA NA 88.8 s 240 s 30 s

NA 5–23 10–30 25 25

NA Temperature HCl Not tested NA

NA 4 months NA NA NA

Electrochemical Fluorescence Fluorescence Absorption Fluorescence

a

0.18–440

NA

NA 71 ppm NA NA 5.8% (relative standard deviation, RSD) NA

o1 min

46 min

0–40

NA

Fluorescence

f

NA 0–900 Up to 50%

0.33 0.50 NA

NA o2% o2.0%

3 min 7 min 1 min

10 min 12 min NA

NA 5–35 20

H2 S, CH3COOH, temperature NA Temperature Oxygen (421%), temperature

4 weeks NA NA

Fluorescence Fluorescence Luminescence

g

(832 ppm) (at 201C and 1 bar) 0.3–4.4 0.05–7 hPa (0.07–10 ppm) 15–1500 4.4–400

0.3 0.04 hPa (0.06 ppm) 15 NA

NA NA

1–2 min o30 s

NA o40 s

NA 25–63

NA H2 S

12 months 2 months

Fluorescence Luminescence

j

710% 72%

o120 s NA

NA NA

0–60 0–50

NA NA

Electrochemical Electrochemical

l

NA

NA

o7 min

NA

20–40

NA  NO 2 , HSO3 , HOAc, HCOOH NA

NA

Fluorescence

n

NA

70.06%

o3 min

NA

15–45

Salinity, acids, SO2, HCl

6 months

Luminescence

o

NA

72.0 ppb

NA

NA

0–100

NA

Analyzer: 5 years

Conductivity

p

1–25% (16.65–416.22 ppm) (at 20 1C and 1 bar) 1–25% (16.65–416.22 ppm) (at 20 1C and 1 bar) 0–10 a

Wiegran et al. (1999). Tabacco et al. (1999) and Walt et al. (2000). c Amao and Nakamura (2005). d Oter et al. (2006). e Ertekin and Alp (2006). f Mu¨ller and Hauser (1996). g Burke et al. (2006). h Wolfbeis et al. (1998). i Orellana et al. (1992) and Interlab OptosenTM (http://www.interlab.es). j Nivens et al. (2002). k Neurauter et al. (2000). l InPros5000 Mettler Toledo (http://www.mtpro.com). m Orion carbon dioxide electrode (http://www.thermo.com). n YSI 8500 (http://www.ysilifesciences.com). o Carbon dioxide sensor (http://www.presens.de). p Martek Dissolved Carbon Dioxide Analyzer (http://www.martekinstruments.com). NA, not available. b

b c d e

h i

k

m

236

Online Monitoring Sensors

3.10.5 Dissolved Oxygen Sensors Together with pH, dissolved molecular oxygen is one of the main analytes to be monitored in water. Oxygen levels in water are critical to determine the health of stream, river, and lake ecosystems. Efficient operation of wastewater-treatment plants requires continuous monitoring of the O2 concentration for aeration, activated sludge, and nutrient-removal control. Tight corrosion control (e.g., nuclear power plants) requires sensing of waterborne O2 at ppb levels. Moreover, O2 sensors are also used as transducers for monitoring other water-quality parameters such as biological oxygen demand (BOD), chemical oxygen demand (COD), total organic carbon (TOC), etc. (see Sections 3.10.8 and 3.10.9). The first dissolved oxygen determination was performed by L. H. Winkler in 1888 using a colorimetric method based on a titration of oxygen with thiosulfate ðS2 O3 2 Þ and iodine (I2). The amount of the dissolved oxygen is proportional to the generated tetrathionate ðS4 O6 2 Þ, which is determined by reduction of I2 to iodide (I). In spite of being difficult to use for online sensing purposes, a 100 years later, this is still employed as a reference method for calibration of electrodes. Automatic measurement of dissolved oxygen based on potentiometric determination of the produced I has also been developed.

3.10.5.1 Electrochemical Oxygen Sensors In the mid-twentieth century, electrochemical methods gained importance due to their fast response, possibility of in situ operation, and analyte nondestructive character, either in an amperometric (voltage applied or intensity of current measured) or potentiometric (intensity applied and voltage measured) mode. The DOC is proportional to the intensity or voltage measured respectively. The cumbersome, dangerous, dropping mercury electrodes gave way to amperometric sensors with solid electrodes covered with gaspermeable membranes and which were capable of being miniaturized. The so-called amperometric ‘Clark electrode’ has probably been the most used O2 sensor so far, and has been the basis of the majority of commercial electrochemical sensors sold till date (see Table 9). In a Clark electrode, oxygen is reduced on a platinum cathode covered with an oxygen-permeable membrane:

O2 þ 4e þ 2H2 O-4OH Oxidation of silver metal occurs at the anode with formation of silver chloride from the chloride ions dissolved in the inner electrolyte solution:

Ag þ Cl  -AgCl þ e  The electrochemical cell has to be polarized at about 800 mV to ensure linearity between the oxygen consumed at the cathode and the measured current. The electrode destroys the oxygen molecules, thereby requiring a minimum water flow in order to maintain equilibrium at both sides of the

gas-permeable membrane. Three problems of the Clark electrode have been identified in continuous operation mode: (1) buildup of an impermeable layer of AgCl at the active anode surface that may lead to a drift in the sensor readings and eventually to failure; (2) production of OH ions at the cathode moves the potential of the inner electrolyte to negative values leading to a zero shift; and (3) Cl ions in the electrolyte eventually become depleted. These problems are normally corrected by the periodical maintenance operations of polishing the anode and replenishing the electrolyte. At the same time, the gas-permeable layer is also cleaned or changed in order to prevent algal biofouling. Moreover, there is a need to establish a polarization of the electrochemical cell, and therefore the Clark electrode needs about 10 min before the first measurement can be obtained. This warm-up time is dependent on the sensor geometry and size. The Clark electrode readings are indeed a function of the O2 partial pressure in the water. According to Henry’s law, the O2 depends on the water temperature and salinity, and on the atmospheric pressure (at 20 1C and ambient pressure of 1013 mbar, air-saturated water contains about 9 mg l1 of O2). Therefore, appropriate corrections for these parameters must always be applied in all sensors. Hydrogen sulfide, a by-product of the anaerobic metabolism of bacteria (e.g., on decaying organic matter), produces the most prominent interference on O2 measurements using the Clark electrode. Once it permeates the electrode membrane, H2S is converted into sulfide ion at the alkaline pH of the inner electrolyte. S2 reacts at the silver anode with formation of a stable precipitate of Ag2S, which passivates the electrode that eventually stops working. A particular amperometric system is the galvanic cell, where a spontaneous O2 reduction at the (platinum or other noble metal) cathode take places combined with simultaneous oxidation of a readily oxidizable sacrificial anode (e.g., lead or zinc)

O2 þ 4e þ 2H2 O-4OH Zn-Zn 2þ þ 2e  The formed Zn(OH)2 turns into zinc oxide flakes (or PbO in case of a lead anode) that, unlike the AgCl of the Clark electrode, detach from the anode surface and avoid the electrode drift in long-term continuous monitoring. The polarization needed to reduce oxygen (B800 mV) is provided by the dissimilar metals of the cathode and anode (i.e., no need of external voltage application). The galvanic O2 sensor may be regarded as a corrosion cell, where the corrosion rate is determined by the rate of oxygen consumed at the cathode. In spite of their intrinsically limited operational lifetime (typically 5 years before having to change the anode), it overcomes some of the drawbacks of the amperometric Clark electrode (neither warm-up waiting, nor electrolyte replenishment and anode servicing are required). While the galvanic cell is less sensitive to the presence of H2S, ammonia and high levels of dissolved CO2 produce stronger interference than in the case of the Clark electrode.

Online Monitoring Sensors Table 9

237

Some commercial sensors for waterborne molecular oxygen

Transduction principle

Model

Dynamic range (mg l1)

Precision

LOD

Response time (s)

Temperature range (1 C)

Electrochemical/ Clark cell Luminescence quenching Luminescence quenching

DO100a

0–20

0.2 ppm

0.1 ppm

900

0–55

Oxi::lyserTMb

0–25

1%

0.01 ppm

NA

0–50

ROXs Optical Dissolved Oxygen Sensorc Model Q45Dd

0–20

0.1 ppm

0.01 ppm

NA

NA

0–40

0.2%

0.05%

NA

 20–60

FDOs 700 IQe

0–20

0.01%

NA

o150

 5–50

TriOxmaticse

0–60

0.1%

NA

180

0–60

DC 300f

0–20

1.5%

0.01 ppm

NA

0–50

ORBISPHERE G1100g ORBISPHERE A1100g Model 9438h

0–20

72 ppb

0.6 ppb

o30

 5–50

0.05–2000

71%

0.1 ppb

30

 5–60

0–20

75%

NA

NA

 20–55

HI 9142i

0–19.9

71.5%

0.1 ppm

NA

0–50

ODOTj

0–20

71%

0.01 ppm

o60

 10–60

COS21D-Ak

0.001–20

71%

1 ppb

o60

 5–100

OPTISENS AAS 2000l

0–20

71%

NA

NA

0–50

RDOs PROm

0–20

70.2%

0.01 ppm

NA

0–50

OPTOSENTMn

0–40

70.2%

0.01 ppm

NA

0–60

LDOs Dissolved Oxygen Probeg PSt3o

0–20

70.1 ppm

0.01 ppm

60

0–50

0–45

70.4%

0.015 ppm

NA

0–50

PSt6o

0–1.8

73%

0.001 ppm

NA

0–50

Electrochemical/ galvanic cell Luminescence quenching Electrochemical/ galvanic cell Electrochemical/ galvanic cell Luminescence quenching Electrochemical/ galvanic cell Electrochemical/ galvanic cell Electrochemical/ Clark cell Luminescence quenching Electrochemical/ Clark cell Electrochemical/ Clark cell Luminescence quenching Luminescence quenching Luminescence quenching Luminescence quenching Luminescence quenching a

Stevens Water Monitoring Systems Inc. (http://www.stevenswater.com). S::can Messtechnik GmbH (http://www.s-caNAt). c YSI Environmental (http://www.ysi.com). d Analytical Technology Inc. (http://analyticaltechnology.com). e WTW GmbH (http://www.wtw.com). f OAKTON Instruments (http://www.4oakton.com). g HACH Company (http://www.hach.com) h ABB Inc. (http://www.abb.com). i HANNA Instruments Inc. (http://www.hannainst.com). j Neotek-Ponsel (http://www.neotek-ponsel.com). k Endress þ Hauser Inc. (http://www.endress.com). l KROHNE Messtechnik GmbH & Co. (http://www.krohne.com). m In-Situ Inc. (http://www.in-situ.com). n Interlab IE (http://www.interlab.es). o PreSens Precision Sensing GmbH (http://www.presens.de). NA, not available. b

238

Online Monitoring Sensors

Although concentrations of oxygen are generally measured in situ amperometrically, potentiometry offers an alternative way of O2 sensing. Using the latter method, no analyte is consumed during the measurement and, at low oxygen concentrations, the logarithmic sensitivity of potentiometric sensors becomes an advantage over the linear sensitivity of amperometric sensors. Potentiometric oxygen sensors were initially developed to measure in the gas phase at high temperature; however, novel potentiometric sensors for determination of dissolved oxygen at ambient temperatures have been manufactured using transition metals (cobalt, zinc, and platinum) or metal oxides (ruthenium oxide, tungsten oxide, and iridium oxide) electrodes.

3.10.5.2 Optical Oxygen Sensors After the original Winkler method (mentioned earlier), optical methods for measuring O2 levels in water have also been developed. For instance, the absorption at 620 nm of indigo carmine has been employed to determine 1–8 mg l1 of dissolved oxygen. This method was an improvement on the 1925 method by Efimoff that employed an indigo carmine solution, glucose as reducing agent, and potassium carbonate as pH buffer. Further modification of this procedure also led to the detection of ppb levels of O2 in water. However, colorimetric methods require sample pretreatment, which makes their online use difficult. When H. Kaustsky discovered in 1939 the luminescence quenching of organic molecules by dissolved oxygen, he was not aware that this opened an extraordinary door to the sensor community. Bimolecular dynamic deactivation of the electronic excited state does not consume the analyte and depends on the quencher concentration (O2 in this particular case). It is observed as a reduction on both the luminescence intensity and lifetime of the indicator dye according to the

Stern–Volmer equation

I0 F0 t0 ¼ ¼ ¼ 1 þ KSV ½O2  ¼ 1 þ kq t0 ½O2  I F t

ð8Þ

where I, F, and t represent the luminescence intensity, quantum yield, and lifetime of the indicator dye, respectively (the subscript 0 means in the absence of O2), and KSV is the so-called Stern–Volmer constant. The latter is equivalent to the product of the bimolecular rate constant of the quenching reaction (kq) and the lifetime of the luminophore in the absence of O2 (t0). The linearity of the Stern–Volmer law is normally lost when the luminescent indicator dye is immobilized into a polymer support for manufacturing the actual oxygen sensor by usually attaching the luminescent film at the distal end of an optical fiber (Figure 4). Therefore, oxygen optical sensors can be based on luminescence intensity or lifetime changes as a function of the analyte level. Luminescence-based O2 sensors offer advantages over electrochemical devices including ease of miniaturization, lack of analyte consumption, faster response, robustness, and insensitivity to interfering agents (e.g., H2S, CO2, or NH3). The low maintenance, extended operational lifetime, and reliability of fiber-optic oxygen sensors based on transition metal (Ru, Pd, Pt, and Ir) luminescent complexes with polyazaheterocyclic chelating ligands (bipyridines and phenanthrolines, porphyrins, etc.) are so noticeable that every major manufacturer of environmental monitors is currently offering at least one model for in situ dissolved O2 measurements in water (see Table 9), rapidly phasing out the amperometric Clark electrode. The appropriate selection of a luminescent dye with a long excited-state lifetime and a supporting polymer material with a high O2 permeability are key issues in the design of the ideal oxygen optical sensor. In this way, materials such as fluorinated polymers, polystyrene, aerogels, organically modified silicates (ORMOSILs), or polydimethylsiloxane (silicone) have

10 000

Counts

1000

100

10

1 0

5

10

15

20

Time (µS) Figure 4 Luminescence decay profiles upon pulsed excitation of the O2 indicator dye tris(4,7-diphenyl-1,10-phenanthroline)ruthenium(II) dichloride embedded in a silicone film, for different O2 concentrations in water (from top to bottom: 0, 0.4, 0.8, 1.2, 1.6, 4.0, and 8.0 mg l1).

Online Monitoring Sensors

all been used with the aim of finding the ideal matrix for each particular application. Most of the sensing layers are microheterogeneous materials formed by the base polymer plus fillers, cross-linkers, plasticizers, etc. These components may play different roles such as reinforcement, providing dye compatibility, increasing flexibility or resistance, etc. One of the most prominent advantages of fiber-optic oxygen sensors is the ease of miniaturization at affordable cost. Therefore, micro-optical sensors have made possible in situ monitoring of aquatic environments in confined spaces such as marine sediments, microbial mats, lichens, biofilms, etc. Both electrochemical and optical sensors are subject to formation of biofilm over the luminescent sensing layer, or the electrode membrane is a critical component of continuous dissolved oxygen monitoring in high biofouling environment (wastewater-treatment plants, highly eutrophized rivers or lakes, tropical ocean bays, etc.). Vendors have developed different methods to avoid such deposits and prolong the operational lifetime of the sensors. Thus, antifouling materials (YSI Inc., Yellow Spring, OH, USA), pressurized air systems (Analytical Technology Inc., Collegeville, PA, USA), ultrasounds (WTW GmbH, Weilheim, Germany), integrated spray cleaning nozzles (KROHNE Messtechnik GmbH, Duisburg, Germany), and automatic mechanical wiping (HACH Co., Loveland, CO, USA), to name a few, are useful commercially available strategies to keep sensor biofouling at bay.

3.10.6 Sensors for Waterborne Ozone Ozone (O3) is a strong oxidizing gas with various industrial applications spanning from water treatment to microelectronics to pharmaceuticals. The needs of each industry in terms of concentration is very different and ranges from the high levels for water treatment, where ozone is used for purification and disinfection purposes on highly absorbing water media, to the residual ppb levels that can be found in cooling water reservoirs of the pharmaceutical or microelectronics factories. Therefore, the monitoring needs for process control and possible interferences are broadly different. For example, the O3 concentration in wastewater treatment should be monitored carefully to keep it at the appropriate concentration that leads to effective water disinfection but minimizes the dangerous bromate ion formation from bromide. Similar to O2 (see Section 3.10.5), continuous monitoring of ozone can be performed by two main methods: electrochemical and optical sensing. The former is based on the electrochemical reduction of ozone over the sensor electrode. The redox method has a few disadvantages in that it is a nonspecific reduction process wherein chlorine or other oxidizing gases present in the sample can interfere with the O3 measurements. To resolve this, several approaches have been proposed. New electrochemical amperometric ozone sensors have a membrane through which ozone diffuses and reaches the electrode. The presence of this selective membrane avoids major interference from other species but introduces a delay in the analysis and increases the equipment cost. Ozone monitoring in waters with high concentration of particulates or salts

239

pose additional maintenance challenges to avoid membrane obstruction, with consequent increase in operational costs. A more elaborate system is the DOM-1 monitor from Eco Sensors, Inc. (Santa Fe, NM, USA). It includes a stripping chamber to extract O3 from the liquid phase for analyzing it electrochemically in the gas phase. Apparently, the stripping separation method reduces the response time of the equipment considerably. As far as optical methods are concerned, O3 has been traditionally been measured by colorimetry after reaction with indigo derivatives:

−

O

O3S N H

H N

SO3− O3 SO3−

O

ε600 ≅ 20 000 M−1 cm−1

−

O O3S O N (SO − ) H 3

ε600 ≅ 0,0 M−1 cm−1

For instance, CHEMetrics ozone dosimeter (Calverton, VA, USA) employs indigo trisulfonate. This dye reacts instantly and quantitatively with ozone, bleaching its blue color in direct proportion to the amount of ozone present. Malonic acid is included in the formulation to prevent interference for up to 3 mg l1 of waterborne chlorine. Ozone displays a strong absorption band centered at 253.7 nm in the UV region, with an absorption cross section of 1.141 1017 cm2. In fact, such a strong absorption band has been in use for quite some time for O3 measurements in the gas phase. Recently, companies such as Horiba Advanced Techno (Northampton, UK), S::can Messtechnik GmbH (Wien, Austria), and IN USA Inc. (Norwood, MA, USA) have developed new optical O3 sensors for water analysis. In some cases (e.g., the Horiba OZ-96 sensor), the system designed for ozone monitoring in clean waters is similar to that used during the manufacturing processes of the semiconductor industry where no absorbing interferent species influence the measurements. This is not the case in the treatment of wastewater with large amounts of suspended solids and organic matter that obstructs the optical measurements. Nevertheless, some recently results from S::can Messtechnik with its Spectro::lyserTM monitor show excellent agreement between the electrochemical and optical determination of ozone concentration in wastewater (Figure 5). Such an accuracy in the O3 level monitoring with an optical sensor in a complex absorbing matrix can be obtained, thanks to an elaborate data treatment using advanced chemometrics tools. A summary of representative waterborne O3 sensors can be found in Table 10.

3.10.7 Sensors for Waterborne Hydrocarbons Hydrocarbon-in-water sensing can be divided into two applications with very different requirements and detection levels. Thus, the petrochemical industry has traditionally used sensors for detecting oil spills which have high levels of hydrocarbons in water and, very often, a very large monitoring area. To this

240

Online Monitoring Sensors 7.0

2 1.5

6.0

1

Ozone ( mg L−1)

5.0

0.5 0 50

4.0

70

90

3.0

2.0

1.0

0.0 0

50

200

150

100

250

Time (h) Figure 5 Overlaid of optical (red) and electrochemical (black) ozone sensor measurements in wastewater. Data courtesy of S::can Messtechnik GmbH.

Table 10

Some commercial sensors for waterborne ozone measurements

Transduction principle Model Amperometric Amperometric UV-VIS absorption UV-VIS absorption Amperometric Amperometric Amperometric UV-VIS absorption Amperometric

DULCOTESTsa Q45H/64b spectro::lyserTMc dFFOZ-Wd DOM-1e CRIUS 4800f 9185sc Ozone Sensorg OZ-96h OZ-50i

Dynamic range (mg l1) Precision

LOD (mg l1) Response time (s) Temp. range (1 C)

0.05–2 0–200 0–30 0–150 0–2 0–10 0–2 0–100 0.1–10

NA NA NA NA 30 1 5 NA NA

0.01 ppm 0.5% 0.015 ppm 1% 10% 75% 3% or 710 ppb 70.5 ppm 73%

5 NA NA NA 60 1800 90 NA o60

0–40  20–60 5–40 NA 20–30 0–40 0–45 5–30 5–40

a

ProMinent Dosiertechnik GmbH (http://www.prominent.de). Analytical Technology Inc. (http://www.analyticaltechnology.com). c S::can Messtechnik GmbH (http://www.s-canat). d IN USA Inc. (http://www.inusacorp.com). e Eco Sensors Inc. (http://www.ecosensors.com). f Process Instruments (UK) Ltd. (http://www.processinstruments.net). g Hach Co. (http://www.hach.com). h HORIBA Advanced Techno (http://www.horiba.com). i Bionics Instrument (http://www.bionics-instrument.com). NA, not available. b

end, sensors based on the characteristic immiscibility of hydrocarbons and water have been developed. On the other hand, environmental protection agencies concentrate their efforts in detecting low levels of hydrocarbons in drinking water or recreation areas. A similar target is aimed at by the naval industry through the International Maritime Organization (IMO), whose Marine Environment Protection Committee has established an upper limit of 15 mg l1 for the oil pollution by ships and mandates control of the emission to the oceans by installation of alarm systems (IMO resolution MEPC 107[49]). Therefore, careful consideration of the sought application for the hydrocarbon-sensing device is the first step to making the

right choice of the monitoring technology. Moreover, some of the marketed sensors (see ahead) do not discriminate among the different hydrocarbon types and volatile organic compounds (e.g., chlorinated hydrocarbons). These hydrocarbonsin-water sensors are reviewed here according to their technology rather than the exact targeted analyte.

3.10.7.1 Oil-Spill Detection Since most hydrocarbons have low solubility in water and their density is lesser than 1, the presence of relatively large amounts of hydrocarbon in water is manifested as an oily

Online Monitoring Sensors

layer on the water surface. Such a situation has been used, for instance, by GE Analytical Instruments Inc. (Boulder, CO, USA) for developing the Leakwises oil-spill sensor. It is a floating device that continuously monitors the liquid surface using a high-frequency electromagnetic absorption technique. Since water absorbs more electromagnetic energy than hydrocarbons, changes in the absorption rate of water indicate the presence or buildup of a hydrocarbon layer. The sensor uses a frequency of 2.45 GHz where the effect of the difference between the water and hydrocarbon dielectric constants is maximum and the influence of salinity is greatly reduced. This sensor has a detection range of 0.3–25 mm oil layer but, being a floating device, it has operational limitations when the water-level variations or lateral currents are too high. Another floating hydrocarbon sensor is the 2114 HCF from Arjay Engineering Ltd. (Oakville, ON, Canada) that monitors the capacitance field between the probe and its concentric shield. As the volume of separated oil increases over the water surface, the probe capacitance changes. Conductivity sensors based on the large difference between dielectric constants of water and hydrocarbons have also been developed. The Expo Instruments Cobra monitor (Sunnyvale, CA, USA) has a conductive polymer that changes its resistance when hydrocarbons are adsorbed. Conductivity measurements are also the basis of the Waterra HS-1 sensor (Bellingham, WA, USA) to make inspections in narrow wells. An additional measurement with an ultrasonic sensor offers the possibility of determining the thickness of the hydrocarbon layer over the water. The sensing devices mentioned above are useful for detection of large amounts of hydrocarbon spills at a particular point. However, they may be insufficient for environmental protection agencies that have the mission of monitoring vast water areas. Two different strategies have been adopted to tackle this problem: (1) installation of several oil point detectors and (2) development of systems for very large surface surveillance. The latter has usually been realized by installing detectors in an aircraft that can scan a large area in a short period of time, searching for an oil spill. Hydrocarbon layers, which are optically thick fluids, absorb solar radiation and re-emit a portion of this radiation as

thermal energy, primarily in the IR region (8000–14 000 nm or 1250–700 cm1). Therefore, during daylight, IR-based sensing devices do not need an excitation source to operate, an advantage in terms of cost and size against other optical sensors. However, IR sensors cannot detect oil-in-water emulsions under most circumstances and several factors can interfere with the measurements including seaweeds and the shoreline. Aromatic hydrocarbons, particularly those containing multiple condensed rings, are strongly fluorescent (Figure 6). Both, the characteristic spectral (excitation/emission) and decay features of the fluorescence can be monitored and correlated to the particular hydrocarbon for quantitative and forensic measurements, very often with the aid of chemometrics. Unfortunately, the fluorescence intensity of an oil spill in seawater, excited by sunlight, is c. 5 times lower than that required for detection. Therefore, unlike IR sensors, optical fluorosensors require the use of lasers operating in the ultraviolet (300–355 nm) or visible (488 nm, e.g., Ar ion) for excitation of the pollutant. For instance, the FLSs Fluorescent LIDAR System of Laser Diagnostic Instruments AS (Tallin, Estonia) can monitor vast territories and detect hydrocarbon pollution at ppm level in water. Airborne oil-spill detection systems working on the principle of scattering of low-frequency electromagnetic radiation are also commercially available. The interaction between microwave radiation and the waves generated by the wind on the ocean surface, results in a scattered radiation known as Bragg scattering. The presence of an oil layer on the ocean surface reduces sea-surface roughness and dampens wind waves. Therefore, the back-scattered radar power decreases, creating dark structures in the radar images that signal the polluted area. This is the principle of SeaDarQ BV (Hardinxveld-Giessendam, The Netherlands) and Miros AS (Asker, Norway) large oil-spill detection systems.

3.10.7.2 Water-Quality Control When the level of waterborne hydrocarbons lies in the mg l1 (ppm) or mg l1 (ppb) ranges, no oil layer builds up and the pollutants are either dissolved or suspended as micro-droplets. In these two situations, optical sensors may be the best choice.

0.12

25 × 106

0.10

15 0.06 10

0.04

5

0.02

0

0.00 350

400 Wavelength (nm)

Figure 6 Absorption and emission (lexc ¼ 340 nm) spectra of anthracene.

450

500

Intensity (a.u.)

Absorption

20 0.08

300

241

242

Online Monitoring Sensors

The most intuitive method to develop an optical hydrocarbon sensor is to use the intrinsic photophysical properties (absorption and/or emission of light) of the analyte itself. Aromatic hydrocarbons absorb strongly in the UV spectral region and both aromatic and aliphatic hydrocarbons display characteristic (narrow) absorption bands in the IR region. Aromatic hydrocarbons also display strong fluorescence that may be employed for more sensitive direct sensing as has been shown earlier. The difference of refractive index between water and hydrocarbons may also be the basis for alternative optosensing schemes. Moreover, some sensors capitalize on the light-scattering properties of the hydrocarbon micro-droplet suspensions in water to determine the pollutant concentration. In general, interferences from the matrix (e.g., strongly absorbing media or high levels of suspended particles) are bound to affect the performance of the above-mentioned optical sensors profoundly for waterborne hydrocarbon measurements (but not of indicator-mediated ones, see Section 3.10.7.2.4).

3.10.7.2.1 Sensors based on refractive-index changes The adsorption of hydrocarbons onto the cladding of a polymer-coated silica (PCS) optical fiber provokes a change in refractive index and therefore in the intensity of the light transmitted along the fiber by attenuated total reflection. This is the basis, for instance, of the fiber-optic chemical sensor developed by Petrosense Inc. (Las Vegas, NV, USA) CMS-4000. A popular class of fiber-optic sensors based on refractiveindex changes uses the so-called fiber Bragg gratings (FBGs) and long-period gratings (LPGs) as the analyte-sensitive device. In both cases, thanks to a periodic variation in the refractive index made to the fiber core, a selection of transmitted wavelengths is obtained. The latter shift upon a change in the grating refractive index, due to adsorption of the hydrocarbons. Nevertheless, depending on the adsorbent polymer and the monitored hydrocarbon type, reversibility of the refractive-index-based fiber-optic sensors may be an issue. Therefore, instrument manufacturers offer optional sensorcleaning systems.

3.10.7.2.2 Sensors based on light scattering The scarce solubility of hydrocarbons in water usually leads to micro-droplet formation even at ppm concentrations that produce scattering of the incident light (see Section 3.10.13). For instance, Dexil Corp. (Hamden, CT) PetroFLAGs is an offline hydrocarbon analyzer that includes extraction, filtration, and turbidity measurements to determine hydrocarbons in water samples. Multiwavelength light scattering (MWLS) with detection at different angles is used by Deckma GmbH (Hamburg, Germany) in their online OMD-7 MKII hydrocarbon detector, specially designed for installation on board ships for monitoring ocean pollution. Measurements at three different wavelengths and various angles of detection allow obtaining turbidity, and oil and solid concentrations with the same equipment. MWLS including the near-IR (NIR) region is used by Rivertrace Engineering Ltd. (Redhill, Surrey, England) in the OCD Xtra oil-in-water analytical system. Both the OMD-2005 monitor of DVZ Services GmbH (Syke, Germany) and the TF16-EX sensor of Optek-Danulat GmbH (Essen,

Germany) use combined light scattering and VIS–NIR absorption data for determination of oil concentrations in water. In general, sensors based on light-scattering measurements are strongly affected by suspended particles present in the sample. In particular, iron-oxide particles have been identified as the major source of interference in sensors installed on ships, and some manufacturers have developed special systems for their elimination or discrimination to avoid false-positive alarms of hydrocarbon presence in the water.

3.10.7.2.3 Sensors based on absorption changes The UV absorption of aromatic hydrocarbons has been used for the development of rugged optical sensors. For instance, S::can Messtechnik GmbH (Vienna, Austria) manufactures the Spectro::lyser system that measures the absorption between 220 and 390 nm to determine the concentration of benzene, toluene, ethylbenzene, and xylene (BTEX). It is a double-beam instrument that compensates light scattering from suspended solids, with a variable path length from 1 to 100 mm depending on the analyte and sensitivity required. OptekDanulat GmbH (Essen, Germany) offers its AF46-EX, a dualchannel UV absorption (254, 280, 290, 300, and 313 nm) sensor with optical paths from 1 to 500 mm. The Teledyne Analytical Instruments (City of Industry, CA) 6600 model includes an in situ ultrasound homogenizer to dissolve the oil in water before UV-absorption analysis and minimize interferences from scattering. Dissolved natural organic matter or UV-absorbing salts are other potential sources of interference. The use of NIR sensors for in situ water-quality control is limited by the strong absorption of water in these regions due to the combination and overtone O–H absorption bands. These bands are often stronger than the C–H absorptions of hydrocarbons, impairing direct determination of the latter in aqueous medium. Therefore, traditional (off-line) hydrocarbon-determination methods based on NIR spectroscopy must include a previous extraction step with freon (American Society for Testing and Materials (ASTM) Method D3921) or, nowadays, more environmentally acceptable solvents such as S-316 (ASTM Method D7066-04). An example is the portable InfraCals analyzer from Wilks Enterprise, Inc. (South Norwalk, CT, USA). Nevertheless, this procedure has particular interest for analyses where online monitoring presents special difficulties due to the presence of high levels of suspended solids or for the analysis of hydrocarbons in soil. Some online sensing devices do both extraction and analysis, with a 5– 30 min operational delay. For instance, Horiba’s OCMA-25 (Northampton, UK) includes extraction of waterborne hydrocarbons with S-316 and absorbance measurements in the IR region (3400 nm). The solvent-extraction step can be eliminated using spectroscopic techniques based on evanescent field absorption (EFA) measurements using polymer-coated optical fibers as the sensing elements and the NIR or mid-infrared (MIR) spectral ranges. The latter type of hydrocarbon sensors could only be developed after the appearance of silver halide-based optical fibers which display high transmission in the IR region. The EFASs fiber-optic sensor for in-situ monitoring of organic pollutants in water from Siegrist (Karlsruhe, Germany) capitalizes on the EFA principle. Novel approaches include planar

Online Monitoring Sensors

and liquid-core waveguide technology. On these waveguideevanescent optical sensors, the coating layer is the key part of the sensor, which determines the selectivity and sensitivity of the final device. Teflon, poly(dimethylsiloxane), polysiloxane with polysiloxane–xerogel, poly(vinyl chloride) (PVC), and low-density polyethylene (LDPE) have all been used to monitor aliphatic, aromatic, and chlorinated hydrocarbons in water. Inelastic Raman-scattering sensors based on Y-shaped fiberoptic reflection probes have also been used to monitor waterborne hydrocarbons. To improve the inherent low sensitivity associated with Raman measurements and to achieve detection limits in the low ppm range for chlorinated hydrocarbons (o1 ppm for trichloroethylene, 15 ppm for perchloroethylene, 15 ppm for chloroform, and 10 ppm for carbon tetrachloride), surface-enhanced Raman spectroscopy (SERS) has been proposed as an alternative. Nevertheless, to the best of our knowledge, no commercial Raman hydrocarbon sensor is currently available.

3.10.7.2.4 Sensors based on emission changes As mentioned above, aromatic hydrocarbons display a strong fluorescence in the UV-VIS that has been exploited for the development of sensitive monitors. Lieberman et al. (1991) described laser-induced fluorescence via optical fibers to measure the level of petroleum hydrocarbons in real time and in situ in seawater. An N2 laser (337 nm, 1.4 mJ, 800 s1 pulses) was coupled to a 10-m bifurcated silica optical fiber and the fluorescence collected by six other fibers concentrically distributed around fiber undergoing excitation. A photodiode array was used as the detector. With this equipment, emission decays and fluorescence spectra of the seawater could be collected and the levels of hydrocarbons continuously monitored. Commercial equipment based on the hydrocarbon intrinsic fluorescence has been available for long in the market. For instance, the TD-4100 system from Turner Designs Hydrocarbon Instruments Inc. (Fresno, CA, USA) detects hydrocarbons in a stream of water falling through an open chamber in which fluorescence is measured. No adsorption media or measuring flow cell is used and therefore problems of measuring-time delays, regeneration, and cell cleaning are eliminated. UV fluorescence measurements are also used by DMA Sorption ApS (Vedbæk, Denmark) in their Bilge monitor for hydrocarbons in water, reaching response times below 1 s with equipment installed on board ships. The FPM 605 from J.U.M. Engineering GmbH (Karlsfeld, Germany) and the CX6000 from Awa Instruments (Duluth, GA, USA) are further examples of hydrocarbon sensors for online water monitoring based on fluorescence measurements. In the Hydrosense 2410 from Arjay Engineering Ltd. (Oakville, ON, Canada), water flows down a special UV plate to maximize signal strength and stability of the fluorescence reading. In general, fluorescence determinations (particularly those based on time-resolved measurements) have the advantage of a lower interference from particles or bubbles that usually provoke false-positive alarms in light-scattering-based equipments. However, since most fluorescence sensors use the intrinsic fluorescence of the analyte itself, hydrocarbons

243

with low fluorescence-quantum yield or no emission at all (e.g., aliphatic or chlorinated) prevent the use of such devices for waterborne hydrocarbon monitoring. To overcome such problems and manufacture more general detectors, different strategies have been proposed. Development of indirect sensors where an immobilized luminescent indicator dye displays a solvatochromic effect on its emission is one of the most promising strategies. Thus, different luminescent dyes such as Nile Red (White et al., 1996) or a ruthenium complex (Castro et al., 2005) have been successfully used to that end. The latter is the base of the fiberoptic portable hydrocarbon-in-water sensor manufactured and recently commercialized by Interlab IEC (Madrid, Spain) within its line of OptosenTM monitoring systems. A compilation of representative sensors for waterborne hydrocarbons is listed in Table 11.

3.10.8 Sensors for Waterborne Organic Matter The level of organic pollutants in the river, reservoir, or wastewater is one of the most widely analyzed parameter because excess contamination of these substances in aquatic environments provokes serious damages to the ecosystem. However, while existing laboratory methods and analyzers are plentiful, there are very few sensors for online in situ continuous (or even near real-time) monitoring of the several indices related to the contents of waterborne organic matter (COD, BOD, and TOC, see ahead).

3.10.8.1 Sensors for COD The COD index is commonly used to indirectly measure the overall amount of organic compounds in water and is expressed in milligrams per liter (mg l1), which indicates the mass of molecular oxygen consumed per liter of solution in the complete oxidation process. COD measuring systems usually imply the total oxidation of the waterborne organic matter and concomitant measurement of the oxygen consumption for such oxidation. Traditionally, oxidation has been done using a strong oxidizing agent such as potassium permanganate (KMnO4). Currently, off-line test kits with potassium dichromate (K2Cr2O7) as oxidizing agent are the most popular and practical for routine applications. COD values are determined by means of a photometric method after digestion (Environmental Protection Agency method 410.4). COD online monitoring equipment based on traditional chemical oxidants, like TOC sensors, usually display disadvantages that mainly include the measuring delay time and the experimental error due to the partial (instead of full) oxidation of some organic and inorganic matter. To overcome these limitations, more efficient methods such as thermal (1200 1C) or ozone oxidation are included in new commercial equipments (see Table 12). A different system is the PeCODTM from Aqua Diagnostic (South Melbourne, Australia) that directly measures the photocurrent charge originating from the oxidation of organic species contained in a sample. The photocatalytic oxidation of organic matter takes places in a photoelectrochemical cell with a photoactive electrode (e.g., a layer of titanium dioxide nanoparticles coated on an inert

244 Table 11

Online Monitoring Sensors Some commercial sensors for waterborne hydrocarbons

Transduction principle

Analyzer model

Dynamic range (mg l1)

Precision

LOD (mg l1)

Response time (s)

Limitations/ interferences

Temp. range (1 C)

Electrical resistance High-frequency absorption UV fluorescence

Cobraa Leakwiseb

NA 0.3–25 mm

NA NA

NA NA

5 NA

NA NA

2–74 0–70

NA

710%

0.001–1000

o10

0–49

CMS-4000 OCMA-25e

0–20 000 0–100

715% 73%

o10 NA

o60 600

Fluorescent compounds NA None

0–50 0–40

OCD Xtraf

0–200

75 ppm

1

NA

NA

0–50

DMA ppm MonitorTMg OMD-7 MKIIh

0–40

710%

o15

1

NA

0–70

0–200

710%

5

10

NA

0–70

OMD-2005i PetroFLAGTMj

0–30 10–50 000

72% 710%

2 15

NA 900

NA Off-line system

1–65 2–35

InfraCalTMk

2–5000

70.1%

0.5%

600–900

Off-line system

4–45

0–100 0–1000 NA

70.3% 71.5% NA

0.1 ppm 0.1 1.5 mm

30 o12 NA

NA NA NA

5–40 10–39 0–50

70.1 710% 71 mm o71%

0.1 0.1 ppm NA o70.05%

NA o10 NA NA

NA NA NA NA

10–50 0–50 0–50 0–70

TF16-EXq

0–500 0–1000 0–600 mm Hydrocarbon dependent 0.5–500

o70.3%

o70.05%

NA

NA

0–40

HC 9010r

0.1–10

0.01 ppm

0.1 ppm

1800

NA

5–20

Teledyne 6600s Teledyne 4080s

0–200 0–1000

72% 72%

1% 0.1 ppm

o5 15

NA For C1 to C9 þ

0–50 4–43

EnviroFlu-HCt FLS-LIDARu SeaDarQv

0–200 NA NA

NA NA NA

0.1 ppb NA NA

NA NA NA

NA NA NA

0–40 NA NA

Miros ODSw

NA

NA

NA

NA

NA

NA

Refractive index change Solvent extraction þ IR absorption Multi-wavelength light scattering UV fluorescence Three wavelength, multiangle light scattering Absorption and scattering Solvent extraction, filtration, and turbidity Solvent extraction þ IR absorption UV absorption UV-fluorescence Conductivity and ultrasounds UV-fluorescence UV-fluorescence Capacitance UV absorption VIS–NIR absorption þ light scattering Solvent extraction þ IR absorption UV absorption Gas chromatography þ FID detection UV-fluorescence UV-fluorescence Long wavelength scattering Long wavelength scattering a

TD-4100c d

Spectro::lyserTM FPM 605 m HS-1n

l

Hydrosense 2410o CX6000p 2114-HCFo AF46-EXq

Expo Instruments Inc. (http://www.expoinstruments.com). GE Analytical Instruments Inc. (http://www.geinstruments.com). c Turner Designs Hydrocarbon Instruments Inc. (http://www.oilinwatermonitors.com). d Petrosense Inc. (http://www.petrosense.com). e Horiba (http://www.horiba.co.uk). f Rivertrace Engineering Ltd. (http://www.rivertrace.com). g DMA sorption (http://www.dma-sorption.dk). h Deckmahamburg GmbH (http://www.deckma.com). i DVZ group (http://www.dvz-services.de). j Dexil Corporation (http://www.dexsil.com). k Wilks Enterprise Inc. (http://www.wilksir.com). l S::can Messtechnik GmbH (http://www.s-caNAt). m J.U.M. Engineering GmbH (http://www.jum.com). n waterra USA Inc. (http://www.waterra.com). o Arjay Engineering Ltd. (http://www.arjayeng.com). p Awa Instruments (http://www.awa-instruments.com). q Optek-Danulat GmbH (http://www.optek.com). r Environnement S.A. (http://www.environnement-sa.com). s Teledyne Analytical Instruments Inc. (http://www.teledyne-ai.com). t TriOS Mess- und Datentechnik GmbH (http://www.trios.de). u Laser Diagnostic Instruments AS (http://www.ldi.ee). v SeaDarQ B.V. (http://www.seadarq.com). w Miros AS (http://www.miros.no). NA, not available. b

Online Monitoring Sensors Table 12

245

Some commercial sensors for the chemical oxygen demand (COD) of water

Oxidation method and transduction principle

Model

Heat þ O2 determination Oxidation þ photocurrent measurements UV-VIS absorption

QuickCODsa PeCODTM P100 analyzerb

O3 oxidation þ O3 differences UV254 absorption

CarbonVISs 700/1 IQc Phoenix 1010d UV 400e

Precision

LOD (mg l1)

Response time (s)

Temprature range (1 C)

0–100

NA

NA

60

NA

0–350

3%

0.2

30–300

NA

0–2500

3%

NA

NA

NA

0–100 000

5%

10

180–900

5–40

10 ppm

NA

10

0–50

Dynamic range (mg l1)

0–20 000

a

LAR Process Analyzers AG (http://www.lar.com). Aqua Diagnostic (http://www.aquadiagnostic.com). c WTW GmbH (http://www.wtw.com). d Endress þ Hauser Instruments AG (http://www.endress.com). e Tethys Instruments SAS (http://www.tethys-instruments.com). NA, not available. b

conductive substrate) and the changes in photocurrent are related to the COD value. No consumption of an oxidizing agent is involved but an electrolyte solution is needed. Since most organic compounds absorb in the UV region, the absorption spectra of water samples can yield COD data. No oxidation reagent is consumed as no sample treatment is performed, with consequent saving in time, device size, and system autonomy. Electrochemical methods have also been successfully applied to laboratory measurements. For instance, COD levels between 20 and 9000 mg l1 have been determined amperometrically with a boron-doped diamond electrode. Oxidation is carried out by the formed hydroxyl radical at the surfaces of the electrode upon water electrolysis. The current of the working electrode changes proportionally with the concentration of the organic matter as long as the physisorbed HO radicals are not depleted. Another amperometric sensor with a surface ground copper electrode has been used to measure 10–1000 mg l1 COD thanks to the catalytic action of copper. Table 12 summarizes representative COD sensors for water analysis.

3.10.8.2 Sensors for BOD BOD, also called biochemical oxygen demand, is another very common index of the quality of water based on quantification of the overall concentration of organic substances by their effect on the respiration of a microbial biomass. The conventional parameter of quality, dating back to 1908, is the socalled BOD-5 (or BOD5) method that measures the oxygen consumption of a sample at 20 1C over 5 days in the dark, by aerobic microorganisms deliberately introduced into the water sample in a closed container. The rate of oxygen uptake is nowadays measured by an oxygen sensor placed in the headspace. The values of the BOD-5 for the different waters can be accurately measured to comply with legislation but the index

is of no use for early warning of environmental damage (spills, runoffs, illegal discharges, etc.), industrial wastewater realtime monitoring, or for maximizing the efficiency of wastewater plant operation (optimization of the biological treatment by monitoring the instantaneous organic-matter level of the influent and the effluent). To overcome the pitfalls of the BOD-5 method, an electrochemical biosensor for BOD estimation was developed, as early as 1977, based on Karube’s work in Japan. The biosensor contains whole microorganism cells immobilized on an acetylcellulose membrane in contact with the water to be measured on the one side, and with a Clark-type oxygen electrode on the other (see Section 3.10.5). While the BOD-5 method uses a mixture of microorganism species, the Karube BOD sensor was based on the respiration of a population of Trichosporon cutaneum. The yeast degrades most organic compounds with concomitant decrease of the dissolved oxygen level producing a measurable response of the oxygen sensor. The microbial sensor BOD values linearly correlated with the BOD-5 values in the 0–60 mg l1 range of a glucose–glutamic acid (GGA) standard solution, with a 20-min response time. The sensor was marketed in 1983 (Nissin Electric Co.) and successfully used, for instance, to measure wastewaters from fermentation plants. Only phosphate buffer and GGA solutions were required for the daily measurements. Several improvements have been introduced in current BOD sensors to reduce their response time (down to 30 s), extend their operational lifetime before change of the sensitive terminal (more than a year), and to raise their sensitivity (limits of detection as low as 0.2 mg l1). These improvements have been possible thanks to the introduction of flow-injection analyzers (FIAs) to perform automatic water sampling, transport, dilutions and standardization, substitution of luminescent optical oxygen sensors (see Section 3.10.5) for the electrochemical devices, and the replacement of Pseudomonas putida, Pseudomonas fluorescens biovar, Bacillus subtilis, Stenotrophomonas maltophilia (among others), or even activated

246

Online Monitoring Sensors

sludge for the T. cutaneum, which is unable to degrade lessbiodegradable organic substances. There are, currently, only a few commercial online BOD analyzers. The Japanese ruggedized BOD 3300 and the benchtop a-1000 models of Central Kagaku Co. are based on the original Karube’s electrochemical O2 sensor respiration measurements to determine BOD levels between 0–500 and 2–50 mg l1, respectively, every 30–60 min. The Spanish Optosens-DBO in situ analyzer (Interlab Ingenierı´a Electro´nica) uses state-of-the-art luminescent measurements of dissolved O2 to interrogate respiration of the immobilized microbial biomass. The sensor allows BOD determinations in the 0–2000 mg l1 range every 20–60 min. However, the South Korean company Korbi has opted for a microbial fuel cell to degrade the sample in its online HABS-2000 analyzer, and correlate the generated electrical signal with the water BOD level (0.1–200 mg l1). In spite of the potential advantages of in situ online BOD sensing, these analyzers are not yet widespread due to (1) the lack of legislation enforcement to perform such measurements, (2) the difficulties often found to relate instant BOD readings with the traditional BOD-5 measurements for water samples with high levels of suspended organic matter, and (3) the recent availability of competing technologies such as TOC online analyzers (see Section 3.10.8.3).

3.10.8.3 Sensors for TOC TOC is the amount of bound carbon in waterborne organic compounds and is yet another nonspecific indicator of water

Table 13

quality often used as an alternative to COD or BOD measurements. In other to avoid interferences from waterborne inorganic carbon (IC), mainly from carbonate and hydrogen carbonate ions, a previous acidification and purging with inert gas of the water sample is included in some of the equipment. Traditionally, TOC analysis has included a first-digestion stage where both organic and inorganic matter is oxidized to CO2. Subsequently, the generated CO2 is quantified and the TOC calculated. A combination of persulfate acid addition, UV irradiation, ozone treatment ,and high temperature combustion (1200 1C) are the most common digesting methods depending mostly on the TOC concentration of the sample (see Table 13). The combination UV/persulfate is   based on the high oxidation potential of the SO 4 and OH radicals produced upon irradiation of S2 O8 2 and H2O, respectively. The mineralization step is the limiting process in terms of analysis times. Therefore, most commercial equipments show delay times of the order of several minutes. For very low TOC concentrations, Mettler-Toledo Thornton Inc. (Bedford, MA, USA) and GE Analytical Instruments (Boulder, CO, USA) offer the model 5000 and Check Point TOC sensors, respectively (see Table 13) that perform an online vacuum-ultraviolet (VUV) (185 nm) oxidation, reducing the total analysis time to less than a minute. Another current approach for low TOC samples is to digest the organic matter by generation of OH radicals by electrolysis or photochemical dissociation of water molecules. Such a reagent-free long operational lifetime system, coupled with a gas–liquid separator and a nondispersive infrared (NDIR) analyzer is used by National Aeronautics and

Some commercial sensors for waterborne total organic carbon (TOC)

Mineralization and transduction principle

Model

Dynamic range (mg l1)

Precision

LOD

Response time (s)

UV/heated persulfate þ CO2 IR detection O3/OH þ CO2 IR detection Combustion þ CO2 IR detection IR detection UV–VIS absorption UV/cold persulfate þ potentiometric CO2 detection UV–VIS absorption

Series 6700a

0–10 000

2%

NA

420

0–40

Series 6700a Series 6700a Series 6700a ProPS–Kitb COT 9010c

0–25 000 0–10 000 0–10 0–500 0–220

3% 3% 3% NA 0.01 ppm

NA NA NA NA 0.5 ppm

600 300 420 NA 1800

0–40 0–40 0–40 NA 5–30

0–150

NA

NA

NA

0–45

0–10 000

3%

NA

360

5–40

0.05 ppb (o5 ppb) 1% (45 ppb) 3%

0.025 ppb

o60 s

0–90

0.05 ppb

15 s

O3/persulfate þ CO2 IR detection UV oxidation þ differential conductivity

UV oxidation þ conductivity a

carbo::lyserTM II/IIId BioTectors Series 4e 5000TOCef

CheckPointg

Teledyne Technologies Company (http://www.teledyne.com). TriOS Optical Sensors (http://www.trios.de). c Environnement S.A. (http://www.environnement-sa.com). d S::can Messtechnik GmbH (http://www.s-canat). e Pollution Control Systems Ltd. (http://www.biotector.com). f Mettler-Toledo Thornton Inc. (http://us.mt.com). g GE Analytical Instruments (http://www.geinstruments.com). NA, not available. b

0–1

0–1

Temperature range (1 C)

10–40

Online Monitoring Sensors

Space Administration (NASA) for TOC analysis in the International Space Station. After the full oxidation step, detection of the generated CO2 gas takes places by means of IR absorption (at 2350 cm1 after purging the aqueous CO2 into the gas phase), potentiometric methods, or differences in conductivity of the water before and after mineralization. The continuous growth of optical sensors is slowly displacing traditional (mineralization–detection) systems in the TOC sensing field as well. Direct UV-VIS absorption of the dissolved organic and inorganic matter has been used by some companies such as S::can (Vienna, Austria) and TriOS (Oldenburg, Germany) for developing alternative in situ TOC sensors. Both analyzers use the entire UV spectral region (210–330 nm) while in other optical equipment, such as the Tethys UV 400 (Meylan, France), only absorption at a single wavelength (usually 254 nm) is monitored with the consequent loss of information. Suppression of the digestion stage dramatically reduces both the analysis time and instrument size, eliminates the consumption of oxidizing reagents, and increases the equipment power autonomy. Nevertheless, these analyzers are limited thus far to low TOC measurements. Compared to BOD measurements, COD and TOC online analyzers provide faster, more reproducible readings. However, the BOD index is more closely related to natural processes than either COD or TOC values because the former uses microorganisms to determine the level of waterborne organic matter (biodegradable organic matter). Additionally, COD readings are affected by the presence of both oxidizing and reducing inorganic matter and TOC has to be corrected by the IC values (mentioned earlier). A representative collection of TOC sensors is listed in Table 13.

Table 14

3.10.9 Waterborne Chlorophyll Sensors Phytoplankton photosynthetic efficiency is one of the biological signals that rapidly reacts to changes in nutrient availability as well as to naturally occurring or anthropogenic toxins (contaminants) and, therefore, is a useful indicator of the environmental water health. As early as in 1956, P. Latimer considered the yield of in vivo chlorophyll a fluorescence as an index of the photosynthetic efficiency. The fluorescence yield can be used as an approximation to chlorophyll concentrations and, in fact, some commercial equipments use this simple principle (see Table 14). Removal of interferences from other fluorescence substances and discrimination of the different algal groups (green Chlorophyta, blue–green Cyanobacteria, brown Heterokontophyta, Haptophyta, or Dinophyta) can be performed using several excitation wavelengths and recording an excitation spectrum characteristic of each group. However, fluorescence per unit chlorophyll is not constant but varies according to the photosynthesis rate and also in response to other factors such as prior exposure to excess irradiance. Correlations between the photosynthesis rate of algal cultures and the increase of the chlorophyll red fluorescence from the photosystem II in the presence of 3-(3,4-dichlorophenyl)1,1-dimethylurea (DCMU) were first observed by G. Samuelsson and co-workers in 1977. In the presence of the pesticide, green algae will strongly fluoresce due to the inhibition of the photosynthetic electron transport. A pronounced DCMU-induced emission increase is recorded when the photosynthetic activity is high (growing algal culture), while algae in the stationary phase of growth would be expected to show only a small DCMU-induced increase in fluorescence. Based on the comparison of fluorescence readings in the presence and absence of DCMU and using a

Some commercial sensors for waterborne chlorophyll

Transduction principle

Model

Dynamic range (mg l1)

Precision

LOD (mg l1)

Temperature range (1 C)

Multiple turnover fluorescence Fluorescence (exc. 470 nm) Fluorescence Single turnover fluorescence Fluorescence (exc. 470 nm) Fluorescence (exc. 470 nm) Fluorescence (three exc. wav.) Single and multiple turnover fluorescence Single turnover fluorescence

PhytoFlasha ECO FLNTUb YSI 6025 c FIRed ECO–FLb Manta2e Algae Torchf Fasttracka IIg

0–100 0.01–50 0–400 0.05–100 0–125 0–500 0–200 0–600

NA 1% 0.1 mg l1 NA NA 0.01 mg l1 0.2 mg l1 2%

0.15 0.01 0.1 NA 0.01 0.03 NA NA

 2–50 0–30  20–60 0–40 0–30 NA 0–30  10–40

Submersible FL3500/ SMh

NA

NA

NA

a

Turner Designs Inc. (http://www.turnerdesigns.com). WET Labs, Inc. (http://www.wetlabs.com). c YSI Environmental (http://www.ysi.com). d Satlantic Inc. (http://www.satlantic.com). e Eureka Environmental Instrumentation (http://www.eurekaenvironmental.com). f BBE Moldaenke GmbH (http://www.bbe-moldaenke.de). g Chelsea Technologies Group Ltd (http://www.chelsea.co.uk). h Photon Systems Instruments (http://www.psi.cz). NA, not available. b

247

0–55

248

Online Monitoring Sensors

parallel flow-through fluorometer, Cullen and Renger developed an online method and defined a fluorescence response index (FRI). Technical advances in electronics have lead to the development of new modulated techniques where the fluorescence yield of chlorophyll a can be determined without addition of any inhibitor agent. By repetitive application of short light pulses of saturating intensity, the fluorescence yield at complete suppression of photochemical quenching is repetitively recorded, allowing continuous plots of the photochemical and non-photochemical quenching. In the dark condition, due to emission quenching by the primary electron acceptor of the photosynthetic process (quinone), a low level of fluorescence emanating from the pigment bed is measured (F0) with a low intensity (not to drive photosynthesis, and in the absence of solar irradiance), exciting beam. When a darkadapted sample is exposed to a high-energy single turnover flash (10–100 ms), a single photoreduction of all the primary electron acceptor occurs, and the fluorescence rises from F0 to a maximum fluorescence level (Fm). Thus, the maximum quantum yield of photochemistry in PSII is given by (Fm – F0)/Fm. The use of multi-turnover systems that generate a longer saturating flash (200–10 000 ms), yields a higher increase in fluorescence due to the more effective photochemical quenching process. The combination of fluorescence techniques and satellite technology has been demonstrated to be a powerful tool for monitoring evolution of ecosystems (Figure 7). In this manner, the water-quality evolution in the Baltic sea, the California current, or the Atlantic ocean have all been studied by satellite fluorescence images of the waterborne cyanobacteria provided, among others, by NASA programs SeaWiFS and MODIS-Aqua.

3.10.10 Sensors for Waterborne Pesticides Pesticides are anthropogenic chemicals commonly used in agriculture. The increasing concern about groundwater pollution due to the use of these compounds requires a strong effort in order to detect such pollutants using reliable, economical, and rapid methods. Pesticides are toxic substances and some of them (e.g., the organophosphates) are powerful inhibitors of enzymes involved in the nerve functions. They normally display low environmental persistence but have acute toxicity, and therefore, there is a demand for fastscreening methods to detect low concentrations of these pollutants. Strict regulations are being enforced in Europe and other developed areas allowing a maximum concentration of 0.1 mg l1 of individual pesticide residues in drinking water. Unfortunately, many of the sensors that have been developed do not match such a detection limit but may still be used for other water-sensing applications (rivers, lakes, reservoirs, wastewater treatment plant inlets, consent discharges, rainfall runoff monitoring, etc.). Currently applied methods for the determination of organophosphates and other pesticides in water are mainly based on gas or liquid chromatographic analyses of water samples, which generally have the advantage of high sensitivity and selectivity. However, they are intrinsically off-line methods and involve several operations such as extraction, homogenization, clean-up of the sample, and concentration and analytical determination. Due to changes in effluent discharge rates as well as dynamic environmental conditions, the aquatic environment is subject to spatially and temporally changing concentrations of pollutants. Sampling-based techniques are usually incapable of tracking these changes and are therefore not suitable for field deployment. Consequently,

Figure 7 SeaWiFS image showing the average chlorophyll a concentration from October 1997 to April 2002. Image from the SeaWiFS Image Gallery courtesy of GeoEye.

Online Monitoring Sensors

there is still a need for in situ continuously operating sensing devices able to monitor pesticides in water at trace levels. Although different sensors have been proposed for their detection, most of them are only able to operate under the controlled laboratory environment or at best with very shortterm in situ measurements due to the fragility of the immobilized enzyme, the reagent/catalyst consumption, or insufficient sensitivity for field measurements. Consequently, only a few of them have reached the market so far. The majority of proposed pesticide sensors rely on either electrochemistry or optical measurements. The most popular electrochemical devices are biosensors based on enzyme-activity inhibition using potentiometry or amperometry as transduction principles. For instance, potentiometric methods for the assay of organophosphorous and carbamate pesticides are based on the inhibitory effect of such chemicals on the acetylcholinesterase (AChE) activity and detection using an electrode of subsequent pH changes caused by acetic acid release in the enzyme-catalyzed reaction:

Acetylcholine þ H2 OAChE -Acetic acid þ Choline Other enzymes such as organophosphate hydrolase (OPH) have also been used for sensor fabrication by monitoring the enzyme activity inhibition through amperometric detection of OPH-catalyzed electroactive hydrolysis products. Fiber-optic biosensors based on immobilized acetylcholinesterase are also known. In this case, acidity changes are monitored using pH-sensitive colorimetric or fluorometric dyes instead of a pH electrode (e.g., Abraxis LLC, Warminster, PA and Severn Trent Services, Ft. Washington, PA commercial kits for detection of aldicarb and dicrotophos pesticides). Alternatively, the enzyme-catalyzed hydrolysis of acetylated luminescent dyes can be followed using high sensitivity in automated optical dosimeters. Optical fiber sensors are of particular interest due to their robustness, remote-sensing capability, and absence of interferences by electromagnetic fields or surface potentials. Other important group of optical sensors for pesticide determination is that based on measurements of the intrinsic optical properties of the analyte (MIR absorption or native luminescence). The MIR technique is limited when dealing with aqueous solutions of an analyte due to strong background absorption of water in this region. However, fiberoptic evanescent wave spectroscopy (EWS) is a technique that allows in-situ MIR absorption spectroscopy in aqueous environments. The optical waveguides provide a rugged and versatile light-delivery system while EWS can provide suitably short path lengths through a highly absorbing medium such as water. Based on the principle of attenuated total reflection (ATR), a different approach for pesticide sensing using IR spectrometry has been developed by Janotta et al. (2003). A nonpolar organically modified sol–gel material is deposited on the optical fiber. If the thickness is 1.7 mm or more, it prevents interaction of the evanescent field with water and also extracts organophosphate pesticides from the solution. With this arrangement, detection limits below 500 ppb, for parathion, fenitrothion, and paraoxon, were attained and sensor

249

measurements could be performed directly in real-life samples such as river waters. Using the native luminescence of pesticides as an optical parameter, several sensors are found in literature for water analysis (Ca´pitan-Vallvey et al., 2001; Ruedas Rama et al., 2002; Salinas-Castillo et al., 2004; Dominguez-Vidal et al., 2007). In the course of the RIver ANAlyzer (RIANA) European project, Mallat et al. have developed a pre-commercial prototype based on a fluorescent immunoassay using labeled antibodies for pesticide determination in water (Klotz et al., 1998; Mallat et al., 1999). Nowadays, it is possible to find some commercial optical immunoassays based on absorption measurements for atrazine determination in water (e.g., Abraxis, Beacon Analytical Systems of Saco, ME and Strategic Diagnostics of Newark, DE, USA). In addition to atrazine, the latter two companies offer similar test kits for determination of other pesticides such as carbofuran, 2,4-dichlorophenoxyacetic acid, etc. For rapid qualitative analysis, Silver Lake Research Corporation (Monrovia, CA, USA) markets a colorimetric immunoassay for atrazine and simazine determination in a test-strip format. Recently, an immobilized microalgae-based fiber-optic biosensor for simazine determination based on chlorophyllfluorescence monitoring has been described (Pen˜a-Va´zquez et al., 2009). Chlorophyll fluorescence increases when toxicants such as simazine inhibit the algal photosystem II (see Section 3.10.9). Alternatively, microalgal biosensors may be based on the inhibition of the photosynthetic function (O2 production) in the presence of a pesticide or other toxicant. According to this scheme, a novel fiber-optic biosensor can selectively detect simazine at sub-microgram per liter level, using a dual head containing selected toxicant-sensitive and -resistant mutants of Dictiosphaerium chlorelloides immobilized on a porous silicone film and luminescent oxygen transduction (Orellana et al., 2009). In this manner, the lack of analyte specificity due to the nonspecific photosystem II response is overcome. One of the main problems in the development of microbial biosensors is the incorporation of the microorganisms into a suitable matrix that avoids leaching without affecting stability or rendering a significant loss of activity (Gupta and Chaudhury, 2007). Currently, sol–gel films are considered one of the best options to fabricate (reversible) robust optical chemical sensors and biosensors (Jero´nimo et al., 2007). It is also possible to find some commercially available whole-cell biosensors for pesticides and other toxic species in water. Such devices are based on the inhibition of the bioluminescence of Vibrio fischeri bacteria by the overall toxicants. These nonspecific sensors are commercialized by companies such as Strategic Diagnostics and Abraxis. A representative set of examples of pesticide sensors is listed in Table 15.

3.10.11 Sensors for Waterborne Toxins Toxins are poisonous substances produced by living cells or organisms that are active at very low concentrations (XiangHong and Shuo, 2008). Depending on the organism that produces them, toxins can be classified into bacterial toxins,

Table 15

Analytical figures-of-merit of some academic and commercial sensors for pesticide determination in water

Pesticide

Dynamic range

LOD

Precision (%)

Response time

Temperature tested (1 C)

Interferences

Lifetime

Transduction principle

References

Butoxycarboxime Trichlorfon Dimethoate Neostigmin Coumaphos

0.1–10 mmol l1 5–20 mmol l1 0.5–100 mmol l1 0.1–10 mmol l1 0.015–0.90 mg l1

NA

1% (RSD)

NA

NA

NA

NA

Electrochemical

a

0.002 mg l1

NA

NA

NA

NA

NA

Electrochemical

b

Trichlorfon Aldicarb Methiocarb

0.05–4.0 mg l1 0.045–5.0 mg l1 0.2–20 mg l1

0.04 mg l1 0.03 mg l1 0.08 mg l1

Chlorpyrifos-oxon Methyl parathion

2–8 mg l1 o5  106 M

0.5 mg l1 4  107 M

4.7 3.9

NA 60 s

NA RT

Phosphate buffer Ionic strength

8 days NA

Electrochemical Electrochemical

c

Paraoxon

4.6–46  106 M

9  107 M

5.8

8

5  10 M 5  10 5  107–5  106 M

1.5  10 M 1.1  107 M

5.8 3.7

NA

NA

NA

NA

Absorbance

e

Aldicarb

0.026–260 mg l1

NA

NA

30 min

NA

NA

Absorbance

f

Dicrotophos

0.14–1400 mg l1

Humic and fulvic acids, Ca2þ, Mg2þ

Aldicarb

(tested ranges) 0.026–260 mg l1

NA

NA

3 min

NA

NA

0.14–1400 mg l1

Colorimetric (qualitative)

g

Dicrotophos

Humic and fulvic acids, Ca2þ, Mg2þ

Alachlor

(tested ranges) 5–100 mg l1

5 mg l1

NA

15 min

NA

NA

NA

Mid-IR

h

Morestan

1.0–200.0 ng ml1

0.28 ng ml1

2.9

2h

RT

Not found

NA

i

Thiabendazole

10–800 ng ml1

2.35 ng ml1

0.93

NA

2070.5 1C

a-Naftol

NA

RT phosphorescence Fluorescence

Warfarin

2–40 mg ml1

0.54 mg ml1

1.26

Naptalam

8.1–300.0 ng ml1

8.1 ng ml1

2.7

NA

20 1C

NA

NA

k

Atrazine

0.001–100 mg l1

0.18 mg l1

1–9

15 min

NA

Desethylatrazine

NA

RT phosphorescence Fluorescence

(tested range) 1

7

pH

Carbofuran Paraoxon

Simazine

8

d

0.10 mg l

1 1

Atrazine

0.1–5 mg l

0.06 mg l

Atrazine

0.1–5 mg l1

NA

Atrazine

0.1–5 mg l1

0.1 mg l1

j

o-Phenylphenol

l

Deisopropylatrazine 3.5–15.2 (RSD) 3.9–22.8 (RSD) 2.6–16.7 (RSD)

15 min

NA

Desethylatrazine

NA

Absorbance

m

20 min

NA

Desethylatrazine

NA

Absorbance

n

50 min

RT

Desethylatrazine

NA

Absorbance

o

Atrazine

Z3 mg l1

Simazine

Z4 mg l1

Simazine

19–860 mg l1

a

Wollenberger et al. (1994). Ivanov et al. (2002). c Hildebrandt et al. (2008). d Wang et al. (1999). e Andres and Narayanaswamy (1997). f Organophosphate/Carbamate Screen Kit (http://www.abraxiskits.com). g Eclox Pesticide Strips (http://www.severntrentservices.com). h Walsh et al. (1996). i Capitan-Vallvey et al. (2001). j Ruedas Rama et al. (2002). k Salinas-Castillo et al. (2004). l Mallat et al. (1999). m Atrazine ELISA Kit (http://www.abraxiskits.com). n Atrazine Tube Kit (http://www.beaconkits.com). o Rapid Assay Kit (http://www.sdix.com). p Watersafe Pesticide Test Strip (http://www.silverlakeresearch.com). q Pen˜a-Va´zquez et al. (2009). NA, not available; RT, room temperature. b

NA

NA

10 min

RT

Not found

NA

Colorimetric (qualitative)

p

3.6 mg l1

5.6

30 min

NA

Atrazine, propazine, terbuthylazine, linuron

3 weeks

Fluorescence

q

252

Online Monitoring Sensors

mycotoxins, and invertebrate and vertebrate toxins. Due to their high toxicity, effective analysis techniques are indispensable. The typical method widely used for the detection and quantification of biological toxins is high-performance liquid chromatography (HPLC) with UV, fluorescence, or mass spectrometric detection. These methods provide sensitive and specific analyses but have problems similar to that previously mentioned for pesticide determination (see Section 3.10.10): (1) they are labor intensive and not really suitable for screening large numbers of samples; (2) the extraction and cleanup processes involve numerous time-consuming steps; and (3) derivatization reagents have been used for converting the toxins into the correspondent fluorescent derivatives, which is a complex procedure and needs skilled personnel. For these reasons, rapid, sensitive and specific methods are needed for routine analysis and monitoring of water samples contaminated by these toxins. In this regard, several sensors and biosensors have emerged in the past decade for toxicity analysis. Several toxicity sensors for drinking-water protection have been evaluated by the US EPA Environmental Technology Verification Program for concentrations at and below the estimated human lethal concentration. Most of them are immunosensors based on the specific high affinity, antibody– antigen binding interactions and optical-detection techniques, such as those developed by Tetracore (Rockville, MD, USA), ADVNT Biotechnologies (Phoenix, AZ, USA), QTL Biosystems (New Kensington, PA, USA), and Response Biomedical Corp. (Vancouver, BC, Canada). Some of them are test strips that indicate the presence or absence of a certain toxin in a preestablished range (ADVNT, Tetracore). Immunosensors display some limitations such as (1) a strong dependence of the antibody-binding capacity under the assay conditions, for example, pH and temperature, and (2) the irreversible nature of the antibody–antigen interaction. A more complete description about fiber-optic immunosensors for waterbornetoxin detection, their different assay formats, and opticaldetection techniques can be found in the review by Marazuela and Moreno-Bondi (2002). Using a fluorescent-labeled amino-acid sequence, PharmaLeads (Paris, France) has developed a test kit for botulinum toxin determination. The company introduces both the fluorogenic label and a quenching substance in the aminoacid sequence and, when botulinum toxin reaches the quenching substance, segmentation occurs generating an intense fluorescence. Another biosensor has been described for aflatoxin B1 determination in river samples based on potentiometric measurements (Marrazza et al., 1999). In this case, the toxin affinity for polynucleotides is measured by its effect on the oxidation signal of the guanine peak of calf thymus DNA immobilized on the electrode surface. Other types of biosensors used for toxins determination are the cellular structure- and whole-cell-based devices. In this case, the living microorganism or a specific cellular component is used as the biorecognition element. For instance, Abraxis has developed a biosensor based on bioluminescence quenching of Vibrio fischeri bacteria, caused by the effect of toxins on their metabolism. The AbraTox Kit responds to global toxicity in water samples and can be used for pesticide determination (see Section 3.10.10) as well. These sensors

display advantages such as (1) whole cells or microorganisms are more tolerant to pH or temperature changes; (2) some microorganisms (i.e., bacteria, fungi, yeast, etc.) can be readily isolated from natural sources (river water, sediments, soil, activated sludge, etc.); (3) a single cell can contain all the enzymes and co-factors needed for detection of the analyte; (4) measurement is frequently possible without extensive preparation of the sample, and (5) biosensors can be easily regenerated by letting the cells re-grow. Limitations of this type of biosensors include longer response times and poorer selectivity compared with enzyme-based biosensors, although this feature can sometimes be turned into an advantage for certain applications (e.g., toxicity screening), as in the case of the Abraxis sensor. Representative biosensors for toxin-in-water detection can be found in Table 16. True sensor devices for waterborne toxins are still lacking but most applications (particularly bioterrorism early alert and detection) can be fulfilled with disposable dosimeters (see Section 3.10.1).

3.10.12 Sensors for Waterborne Bacteria Development of sensors for real-time detection of bacterial contamination in water supplies is a top but highly challenging priority. For this application, sensors should be sensitive enough, rapid, and robust with long operational lifetime (Ji et al., 2004). Until now, a significant number of detection methods have been developed using the optical, electrochemical, biochemical, and physical properties of the microorganisms (Hobson et al., 1996). Some of them have been commercialized, for example, those based on impedance measurements (Don Whitley Scientific, West Yorkshire, UK; Sy-Lab, Neupurkersdorf, Austria, and BioMerieux, Marcy l’Etoile, France). However, these methods are nonspecific, respond not only to bacteria but to all types of microorganisms present in the water sample, and are time consuming because they are based on microorganism growth. One rapid unspecific colorimetric method for total bacteria determination is that proposed by Palintest (Kingsway, Team Valley, England). Their test strips comprise nutrient agar for total aerobic count of bacteria and triphenyl tetrazolium chloride (TTC) dye which stains most colonies red for easy enumeration. The range of detection for bacteria is 103–107 colony-forming units (CFU) ml1 in water. Other types of biosensors have been developed recently for bacteria determination in water samples; these devices are sensitive, specific, and rapid in comparison to the previously cited methods. Some representative examples described in the literature are presented below. Optical bacteria sensors based on fluorescent nucleic acid stains, acting both as molecular-recognition elements and fluorescent reporters, have been described (Ji et al., 2004; Chuang et al., 2001; Chang et al., 2001). The working principle of these sensors is that the fluorescence quantum yield of some nucleic acid stains significantly increases upon binding to nucleic acids. This signal increase can be correlated to the amount of nucleic acid present in the sample. Since all organisms contain nucleic acids, these sensors are not specific and respond to all bacterial species.

Table 16

Some academic and commercial sensors for toxin determination

Toxin

Tested range 10

LOD 1

4

1

Precision

Response time

NA

E5 h

NA

Interferences

Transduction principle

References

Humic and fulvic acids, Ca , Mg

Absorption

a

NA

Ca2þ, Mg2þ

Luminescence

b

2þ

2þ

Anthrax Botulinum A and B

200–10 spores ml 0.004–0.3 mg l1

2  10 spores ml 0.004 mg l1

Ricin Botulinum A Ricin

0.0015–15 mg l1 5  105–0.3 mg l1 5  105–15 mg l1

0.0015 mg l1 5  105 mg l1 5  105 mg l1

Anthrax Ricin

200–5  106 spores ml1 0.05–15 mg l1

105 spores ml1 0.05 mg l1

NA

5 min

Ca2þ, Mg2þ

Fluorescence

c

Anthrax Botulinum A

200–1010 spores ml1 0.5–25 mg l1

4  105 spores ml1

NA

o15 min

Humic and fulvic acids, Ca2þ, Mg2þ

Fluorescence

d

Botulinum B Ricin Anthrax Botulinum A

0.3–1000 mg l1 1–50 mg l1 200–1010 spores ml1 0.5–25 mg l1

0.5 mg l1 1 mg l1 106 spores ml1 0.4 mg l1

NA

15 min

Humic and fulvic acids, Ca2þ, Mg2þ

Colorimetric (qualitative)

e

Botulinum B Ricin Anthrax Botulinum A and B

0.3–1000 mg l1 0.4–2000 mg l1 200–1010 spores ml1 0.01–0.5 mg l1

0.4 mg l1 0.4 mg l1 105 spores ml1 0.01 mg l1

NA

15 min

Humic and fulvic acids, Ca2þ, Mg2þ

Colorimetric (qualitative)

f

Ricin Botulinum A Botulinum B

0.035–15 mg l1 0.01–0.5 mg l1

0.035 mg l1 0.01 mg l1 0.01 mg l1

NA

30–60 min

Humic and fulvic acids, Ca2þ, Mg2þ

Fluorescence

g

Aflatoxin B1 Botulinum B

10–30 mg l1 0.0003–0.3 mg l1

10 mg l1 NA

NA r34%

2 min 60 min

NA Zn2þ, Fe2þ, Cu2þ

Electrochemical Bioluminescence

h

Ricin

0.015–15 mg l1

a

Enzyme-linked immunosorbent assay (ELISA) (http://www.tetracore.com). Bioveris (now within Roche) BioVerify Test Kits and M-Series M1 M Analyzer. c QTL Biosensor (http://www.qtlbio.com). d RAMP Immunoassay Test Cartridges (http://www.responsebio.com). e BADD Immunoassay Test Strips (http://www.advnt.org). f BioThreat Alert Immunoassay Test Strips (http://www.tetracore.com). g EzyBots A and EzyBots B Test Kits (http://www.pharmaleads.com). h Marazza et al. (1999). i AbraTox Kit (http://www.abraxiskits.com). NA, not available. b

r27%

i

254

Online Monitoring Sensors

Specific biosensors for bacteria determination have been developed using RNA and DNA probes. Combined with the polymerase chain reaction (PCR, Belgrader et al., 1999), several devices with relatively short response times have been marketed (Applied Biosystems, Foster City, CA, USA; Idaho Technology, Salt Lake City, UT, USA; Invitrogen, Carlsbad, CA, USA). If the sequence of bases of a particular part of the DNA molecule is known, then the complementary sequence (probe) can be synthesized and labeled using a fluorescent reporter. The problem with these type of sensors is that they are limited when faced with unknown or genetically modified organisms. Strategic Diagnostics has commercialized an enzyme-based qualitative biosensor to detect the presence/absence of total coliforms and E. coli in water samples. The ColitagTM method capitalizes on the detection of two enzymes, namely b-glucuronidase and b-galactosidase, which are characteristic of E. coli and the coliform groups, respectively. Colitag detects total coliforms using the chromogenic substrate o-nitrophenyl-b-Dgalactopyranoside (ONPG). Upon hydrolysis by b-galactosidase, ONPG produces a distinct yellow color, confirming the presence of coliforms in the sample. For detecting E. coli, Colitag utilizes the fluorogenic enzyme substrate 4-methylumbelliferyl-b-D-glucuronide (MUG). Upon hydrolysis by bglucuronidase, MUG releases 4-methylumbelliferone. The latter reaction product fluoresces when exposed to UV light. The b-glucuronidase enzyme is specific to E. coli and observation of fluorescence differentiates this organism from other members of the coliform group. Colitag is able to detect just one CFU of E. coli and other coliform bacteria in 100 ml of water fulfilling the legal requirement in developed countries for potable water. Nevertheless, actual levels of bacteria present in drinking waters of less-developed countries or regions may be higher by 3–4 orders of magnitude. Bacteriophages or phage organisms can be employed as recognition elements to detect deadly bacteria such as E. Coli and Salmonella. The genetically engineered phages supplied by Biophage Pharma Inc. (Montreal, Canada), are viruses that recognize specific receptors on the surface of bacteria, to which they bind with extreme selectivity and sensitivity. The system works on the basis of monitoring the change in capacitance caused when the target bacteria attach to the sensing interface (Lei et al., 2008). The phages can also be tagged with fluorophores to render them optically responsive, and are immobilized on the surface of an addressable micro-LED array. The LEDs are used to excite the phage organisms and the fluorescence intensity is dependent on the concentration of specific bacteria attached to the phages (Lei et al., 2008b). The performance of representative sensing devices for waterborne bacteria is listed in Table 17.

3.10.13 Turbidity Sensors The main areas of application of the water-turbidity measurements lie on cell-density determinations, crystallization monitoring, filtration control, detection of suspended solids, quality testing, and flocculation monitoring. The units given by turbidity instruments are the so-called nephelometric turbidity units (NTUs) but in the sensing field it is also common

to use suspended sediment concentrations (SSCs) and particle size distribution (PSD). Among all methods, two are standardized and approved for turbidity determinations of freshwater and brackish water: Nephelometric Method 2130 B and ISO 7027. The attenuation of an IR beam (e.g., 850 nm) or visible radiation in the red region (e.g., 660 nm) is used to avoid interference due to absorption of organic matter dissolved in the water. Another option is to perform a multiwavelength analysis using a white light source and detect any effects due to absorption. To eliminate interference from extraneous light sources, the analytical beam can be pulsed at a rate of several kilohertz. Light-scattering-based instruments can operate on either of the two principles: transmission and nephelometry. In the transmission mode, the SSC is calculated using the loss of light through a determined optical path. This simple principle was used by P. A. Secchi to develop the first known turbidimeter, back in 1865. In Secchi’s method, a circular disk about 30 cm in diameter is lowered from above the water into the water column, and the point at which it disappears from sight is determined. In the nephelometric mode, SSC is calculated by the amount of light scattered by the suspended particles. This scattered light can be measured at 901 (901 nephelometry) or at 3017151 to the incident beam (backscattering). Depending on the application and the amount of suspended solid, transmission, 901 nephelometry or backscattering is chosen. Air bubbles are one of the major interferences in these types of systems but can be eliminated applying high pressure (Analytical Technology, Collegeville, PA, USA), introducing a bubble-trap chamber (Hach Co., Loveland, CO,USA), or compensating by statistical treatment of the measured values (Zu¨llig AG, Rheineck, Switzerland). In order to avoid the absorption of light and interference from bubble scattering, acoustic techniques can also be used. Acoustic measurements of suspended particles in the water are based on two approaches: the first method is to measure the attenuation of an acoustic pulse passing through the water column due to the suspended particles. Particle-size distribution and concentration within the water column can be derived but estimation of distribution, as a function of depth, cannot be inferred. Commercial equipments based on ultrasonic attenuation are currently in the market (e.g., Markland Specialty Engineering Ltd., Georgetown, ON, Canada). The second approach is by interpreting the backscattering, which is the scattering by the suspended particles back to the transducer known as Acoustic Doppler Current Profiler (ADCP). The latter is a state-of-the-art equipment in oceanography and hydrometry for current velocity. The ADCP works by transmitting pings of sound at a constant frequency into the water. As the sound waves travel, they ricochet off particles suspended in the moving water, and reflect back to the instrument. Due to Doppler effect, sound waves bouncing back from a particle moving away from the profiler have a slightly lowered frequency when they return. Particles moving toward the instrument send back higher frequency waves. This shift is used to calculate how fast the particle and the water around it are moving, while the intensity of the signal echoed by the suspended particles contains information on concentration. This technique has some intrinsic limitations. First, multifrequency ADCP instruments are needed in order to resolve

Table 17

Some academic and commercial sensors for bacteria quantitation in water

Bacteria

Tested range

LOD

Response time

Temperature range

Interferences

Lifetime

Transduction Principle

References

Escherichia coli Bacillus subtilis 23095

104–108 cells ml1

104 cells ml1

5 min

4–48 1C

Not found

7 months

Fluorescence

a

Pseudomonas aeruginosa Pseudomonas aeruginosa

2.4  105–2.4  107 cells ml1 0–5.4  107 cells ml1

2.4  105 cells ml1 NA

15 min 10 min

NA NA

NA NA

50 h (operational) 2 months

Fluorescence Fluorescence

b

Erwinia herbicola

5–500 cells

NA

7–15 min

NA

NA

NA

Fluorescence

d

Escherichia coli

0.2–106 cfu ml1

10 cfu ml1

NA

NA

Not found

NA

Fluorescence

e

Francisella

2  103–5  104 cfu ml1

103 cfu ml1

NA

NA

Humic and fulvic acids

NA

Fluorescence

f

tularensis, Yersinia

4

0.28–5  10 cfu ml

Not found

pestis, Bacillus anthracis, Brucella suis, Escherichia coli Francisella tularensis, Yersinia pestis Bacillus anthracis

200–5  104 cfu ml1 40–5  104 cfu ml1 0.2–5  104 cfu ml1 2  104–5  105 cfu ml1 0.28–5  103 cfu ml1 200–5  105 cfu ml1

Not Not Not Not

NA

Fluorescence

g

a

1

104 cfu ml1 102 cfu ml1 104 cfu ml1

o10 min

NA

found found found found

Ji et al. (2004). Chuang et al. (2001). c Chang et al. (2001). d Belgrader et al. (1999). e TaqMan E. coli 0157:H7 Detection System (http://www.appliedbiosystems.com). f R.A.P.I.D. System for the detection of Francisella tularensis, Yersinia pestis, Bacillus anthracis, Brucella suis, and Escherichia coli (http://www.idahotech.com). g PathAlert detection Kits for the detection of Francisella tularensis, Yersinis pestis, and Bacillus anthracis (http://www.invitrogen.com). NA, not available. b

c

256

Online Monitoring Sensors

whether changes in echo intensity (EI) are associated with changes in sediment concentration, or changes in particle-size distribution. Luckily, most of the manufacturers already offer multi-frequency equipment (Teledyne RD Instruments, Poway, CA, USA; SonTek/YSI, San Diego, CA, USA). Second, an acoustic surrogate is the relation between particle circumference and ADCP frequency, and an error in SSC estimates has been found to increase as the ratio of particle circumference to acoustic wavelength approaches 1. A third limitation is that ADCPs are designed to detect acoustic frequency changes in current profiles and are less accurate in measuring the amplitude changes associated with EI measurements. Laser-diffraction instruments exploit the principles of small-angle forward-scattering angles. Thus, the method is mostly insensitive to change in particle color or composition. Sequoia Scientific, Inc. (Bellevue, WA, USA) and Shimadzu Corporation (Kyoto, Japan) provide this type of equipment. For instance, the LISST100X and LISST25X (LIST ¼ Laser InSitu Scattering and Transmissometry) are equipped with a 670 nm laser. Both record scattering at 32 angles, mathematically invert the readings to get the size distribution, and also scale it to obtain the volume-scattering function (VSF). Their working range is from 1 to 750 mg l1 and can be modified by increasing or reducing the optical path since it is a forward-scattering measuring technique. For very high SSC samples, the company has developed the LISST-INFINITIVE model that includes one single sample dilution step (50:1). A range from 1 to 500.000 mg l1 is achieved, but no data of a possible delay time on the reading is provided by the manufacturer. Estimation of SSC from fluid-density values computed from pressure measurements is a strong candidate for monitoring high sediment-laden flowstream. In fact, this inexpensive technology is designed to monitor SSC values over 10 g l1 since it is based on simultaneous pressure measurements with two exceptionally sensitive pressure-transducer sensors arrayed at different fixed elevations in a water column. The precision pressure sensors are available from several companies but only Waterlog (a division of design Analysis Associates, Inc., Logan, UT, USA) used them for turbidity measurements. Unfortunately, the system is no longer available. In fact, application of this technique in the field can be complicated by low signal-to-noise ratios associated with low SSC, turbulence, significantly large dissolved-solid concentrations, and watertemperature variations. Table 18 lists the most representative equipments for turbidimetry currently found in the environmental instrumentation market.

are affected by all oxidizing and reducing agents present in water so that they also are nonspecific measurements. The ORP is determined by measuring the difference of potential between an inert sensing half-cell (indicator electrode) and a reference half-cell (reference electrode) as in pH measurements. The sensing half-cell (typically platinum) acts as a platform for electron transfer to or from the sample. The standard hydrogen electrode (SHE) is the reference from which all standard redox potentials are determined, and has been assigned an arbitrary half-cell potential of 0.00 V. However, it is fragile and impractical for routine use, and therefore Ag/AgCl and saturated calomel (SCE) reference electrodes are used instead. The latter is nowadays phased out because it contains mercury. At equilibrium, the ORP is calculated as the emf of the overall galvanic cell:

EðORPÞ ¼ ½E0 þ ð2:3RT=nFÞ  ðlog½Ox=½RedÞ  Eref

ð9Þ

where E0 is the standard potential of the redox system versus SHE, R is the universal gas constant, T is the temperature in Kelvin, n is the number of electrons transferred, F is the Faraday’s constant, [Ox] and [Red] are the activities of oxidant and reductant species ,and Eref is the half-potential of the reference electrode at 25 1C. The readout of the ORP sensor is a voltage (relative to the reference electrode), with positive values indicating an oxidizing environment (ability to accept electrons) and negative values indicating a reducing environment (ability to donate electrons).

3.10.14.1 Effect of pH on Oxidation–Reduction Potential Sensors Some redox reactions are pH-dependent, for example, the reduction of the powerful disinfectant hypochlorous acid (HClO), widely used in water treatment as a by-product of chlorine solutions in water:

HClO þ H þ þ 2e  " Cl  þ H2 OðE0 ¼ 1:49 VÞ In this case, the ORP of water would be

EðORPÞ ¼ ½E0 þ ð2:3RT=nFÞ  logð½HClO½Hþ =½Cl Þ  Eref

ð10Þ

where the proton activity [Hþ] shows the ORP dependence on pH. If the redox reaction involved depends on the acidity, the solution pH must be controlled to achieve reliable ORP measurements.

3.10.14.2 Effect of Temperature on ORP Sensors

3.10.14 Oxidation–Reduction Potential Sensors The so-called oxidation–reduction potential (ORP, or simply redox) measures the capacity of an aqueous solution to either release or gain electrons by electrochemical reactions. Oxidation and reduction reactions control the behavior of many chemical species in drinking water, wastewater, and aquatic environments. Therefore, ORP values are used in a manner similar to pH values to determine water quality. ORP values

The ORP is directly dependent on the temperature of the sensing system according to Nernst equation above. The actual temperature effect depends on the ratio of activities of each redox couple present in solution. In most cases, electroactive species in solution are unknown and for this reason, temperature is not compensated in ORP sensors. Proper use of ORP sensors require that their calibration is done at the same temperature at which the measurement will be carried out. For this reason, some vendors provide tables containing the ORP

Online Monitoring Sensors Table 18

257

Some commercial sensors for water turbidimetry

Transduction principle

Model

Dynamic range

Precision

LOD

Response time (s)

Temperature range (1 C)

Infrared backscattering Infrared 901 nephelometry Infrared backscattering Red (660 nm) light 901 nephelometry White or Infrared light 901 nephelometry Infrared 901 nephelometry

AF46 CSa A15/76b OBS–3 þ c FilterTrak 660TMd MicroTOLe

0–200 g l1 0–4000 NTU 0–4000 NTU 0–5 NTU

NA NA NA 0.007 NTU

NA NA NA NA

0–55 0–50 NA 0–40

0–1000 NTU

NA 75% or 70.02 72% or 0.25 73% or70.005 75%

0.0001 NTU

NA

0–50

NTU Digital Sensorf NTU–S10f

0–4000 NTU

71%

0.01 NTU

o1

0–50

0–20 NTU

NA

NA

NA

 10–50

270WQg LISST100Xh VisoTurbs 700 IQi 6136 Turbidity Sensorj LATS–1k Model 502–ILl TU 810m Turbimax W CUS65-Cn TML–25o CT–CENSETMp ECO VSF3q

0–1000 NTU 1–750 mg l1 0–4000 NTU

71% NA 72%

NA o1 mg l1 0.05 NTU

5 NA NA

 10–50  10–45 0–60

0–1000 NTU

72% or 0.3

0.1 NTU

NA

NA

0–3.2 NTU 1–150 g1 0–4000 NTU 0–50 g1

NA 75% or 1 g/l 1 NTU o1%

0.001 NTU NA 0.05% NA

1 60 10 NA

NA 1–45 0–50 0–50

0–4000 NTU 1–200 NTU 0–25 NTU

o1% 72% NA

0.001 NTU o0.1 NTU 0.01 NTU

2 15 NA

0–50 0–50 0–30

EL400r

0–100 NTU

72%

0.01 NTU

NA

0–60

Infrared absorption and 901 nephelometry Infrared 901 nephelometry Laser diffraction Infrared 901 nephelometry Infrared 901 nephelometry Laser diffraction Ultrasonic attenuation Infrared 901 nephelometry Infrared absorption Infrared 901 nephelometry Infrared 901 nephelometry Three (red, blue, green) wavelengths, three angle (100, 125 and 1501) backscattering Infrared 901 nephelometry a

Aquasant AG (http://www.aquasant.com). Analytical Technology, Inc. (http://www.analyticaltechnology.com). c Campbell Scientific Ltd. (http://www.campbellsci.com). d HACH Company (http://www.hach.com). e HF scientific Inc. (http://www.hfscientific.com). f Neotek-Ponsel (http://www.neotek-ponsel.com). g NovaLynx Corporation (http://www.novalynx.com). h Sequoia Scientific Inc. (http://www.sequoiasci.com). i Wissenschaftlich-Technische Werksta¨tten GmbH (http://www.wtw.com). j YSI Environmental (http://www.ysi.com). k Shimadzu Corporation (http://www.shimadzu.com). l Markland Specialty Engineering Ltd. (http://www.sludgecontrols.com). m RODI Systems Corporation (http://www.rodisystems.com). n Endress þ Hauser Inc. (http://www.endress.com). o Zu¨llig AG (http://www.zuellig.ch). p CENSAR Technologies Inc. (http://www.censar.com). q Wetlab Inc. (http://www.wetlabs.com). r Tethys Instruments SAS (http://www.tethys-instruments.com). NA, not available. b

values for the (calibration) standard solution versus the reference electrode at different temperatures.

3.10.14.3 Frequent Problems with ORP Sensors ORP sensors can show a sluggish response in environmental waters if the platinum electrode of the probe has been contaminated. Common contaminants include hard-water deposits, oil/grease, or organic matter. It may take days to reach

the final ORP value in contaminated sensors and, therefore, the typical time frame of a sampling experiment (o1 h) is not sufficient to provide a correct reading. Naturally, if the electrode contaminant is redox-active, either in itself or because it contains redox-active impurities, the sensor reading will be erroneous until the contaminant is removed. In clean environmental water, there may be very few redoxactive species present and those that are present may be in very low concentration. In many cases, the concentration can be so

258

Online Monitoring Sensors

low that the redox influence of the species is actually below the detection limit of the method. Finally, the surface composition of the electrode may not be ideal for the measurements in the selected medium. While platinum ORP electrodes are primarily composed of the metal itself (in a neutral state), it is well known that the surface of the electrode where the redox action takes place is coated with a molecular layer of platinum oxide (PtO). The Pt/PtO ratio can change over time, depending on the medium in which the probe is stored, and thus the electrode surface actually possesses its own potential that can be variable. If this surface potential is similar to the ORP potential of the medium, then the electrode response can be slow. Representative ORP sensors can be found in Table 19.

3.10.15 Conductivity Sensors Electrical conductivity or specific conductance (k) is defined as the ability of a liquid to conduct electricity. It is the reciprocal of resistivity and is measured in Siemens cm1. Conductivity is extensively used to characterize water supplies for municipal, commercial, hospital, and industrial uses. While individual ions cannot easily be differentiated, conductivity gives a measurement of the total ions present in the sample, reading being proportional to the combined effect of all the ions. Conductivity is sometimes expressed as parts per million of total dissolved solids (ppm TDS) and is used to monitor mineral concentration in some applications. Table 19

A large variety of conductivity equipment is now available to measure water quality ranging from ultrapure water (very low conductivity) to concentrated chemical water streams (high conductivity). Representative examples are listed in Table 20. Conductivity sensors for water analysis use two or four electrodes with a known surface area and are directly placed in the solution or built into a tube or vessel at a specified distance. The cell constant (kcell) refers to the distance between the electrodes divided by the electrode area. If conductivity is low (very dilute solutions) the electrodes are placed close together and the cell constants are between 0.1 and 0.01 cm1. If conductivity is high, they can be further apart and the cell constants can reach up to 10 cm1. The two-electrode conductivity sensors are based on amperometric measurements. In this case, a known potential difference (DV) is applied to the pair of electrodes and the resulting electric current (I) is measured. According to Ohm’s law (I ¼ DV/R) the resistance of the system (R) can be determined and related to conductivity by the following expression:

k ¼ kcell =R Unfortunately, the resistance in this method is not constant; salt deposition on the electrodes due to electrolysis can change it. For low to medium conductivity levels (o2 mS cm1) this method may be sufficient, but for greater accuracy and for higher conductivities, a potentiometric method is required.

Representative academic and commercial sensors for water oxidation–reduction potential (ORP) measurements

Dynamic range (mV)

Precision

Response time

Temperature range (1 C)

Pressure range

Lifetime

References

90–450 86–268

E1 s o30 s

25 20–25

NA NA

NA NA

a

NA

0–60

NA

NA

c

 2000–2000  500–500

NA 70.31,70.42, 70.49 mV (different ranges) 70.25, 70.5, 72 mV (different ranges) 72 mV 2% (RSD)

NA NA

0–50 0–55

NA NA

d

 999–999  2000–2000  2000–2000

71 mV 730 mV 71 mV

NA NA NA

0–50 0–60 0–60

NA NA NA

f

 700–1100

70.1% (RSD)

10 s

0–80

5 years

i

 1500–1500

NA

NA

0–100

NA o40 psi (276 kPa) NA NA o50 psi (345 kPa) 0–30 psi (0–207 KPa) 0–232 psi (0–1.6 MPa)

NA

j

 450–1100

a

Lee et al. (2007). Jang et al. (2005). c ORP Sensor (http://www.vernier.com). d HI 504 pH/ORP Controller with Sensor Check (http://www.hannainst.com). e WQ600 ORP Sensor (http://www.globalw.com). f ORP15 ORP/Temperature Instrument (http://www.ysiecosense.com). g P & S series ORP electrode (http://www.ionode.com.au). h 720II ORP Controller (http://www.myronl.com). i CSIM11–ORP (http://www.campbellsci.com). j Orbisint CPS12/CPS12D/CPS13 (http://www.endress.com). NA, not available. b

b

e

g h

Online Monitoring Sensors Table 20

259

Representative examples of academic and commercial sensors for water conductivity measurements

Measuring range

Precision

Cell constant (cm1)

Response time

Temperature range (1 C)

Pressure range

References

1–189 mS cm1 0.01 mS–1 mS (conductance) 0–1999 mS cm1 (different ranges) 0.02–500 mS cm1

o7% (RSD) 75.6 nS (SD) NA

2.02 NA 0.01–10

NA 160 ms NA

NA NA 0–100

NA NA 0–2 MPa

a

5%

NA

NA

0–100

d

0.005–7 mS cm1 NA

10% NA

NA 0.1–10.0

NA NA

0–50 0–70

0–40 000 mS (conductance)

1%

NA

NA

0–55

0.0001 mS cm1–2 S cm1 (different ranges) 0–4999 mS cm1 (different ranges)

1.5%

o0.933

NA

0–130

0.50%

1.0–10

NA

0–95

0–10 bar (0–1 MPa) NA o7.5 bar (750 KPa) 50 psi (345 kPa) o14 bar (1.4 MPa) NA

b c

e f

g

h

i

a

Lario-Garcı´a and Pallas-Areny (2006). Li and Meijer (2008). c Conductivity analyzers (http://www.yokogawa.com). d InPro 7100 (http://www.mt.com). e CS547A (http://www.campbellsci.com). f Conductivity sensors (http://www.sensorex.com). g WQ301 (http://www.globalw.com). h WTW (http://www.wtw.com). i YSI 3100/3200 (http://www.ysi.com). NA, not available. b

The potentiometric method for conductivity measurements uses four electrodes: the two outermost electrodes apply an alternating voltage. However, rather than directly measuring the current between these two electrodes, the four-electrode sensor measures the voltage drop across the two innermost electrodes. Operating with a known current condition, the resistance of the solution can be calculated using Ohm’s law. Unlike amperometric probes, potentiometric conductivity sensors are not limited by electrolysis and therefore can be used for a wider range of conductivities.

increasingly lower analyte levels truly online. Substance-specific monitoring is still a challenge particularly for organic water pollutants due to the vast diversity of chemical structures. Therefore, online monitoring sensors and off-line laboratory techniques, such as chromatography interfaced with mass spectrometry, will continue to coexist for many years to come. Each one fulfils an important mission and they are bound to complement each other. However, only reliable, affordable, robust sensors will provide the expected benefits to the humankind and the environment.

3.10.15.1 Effect of Temperature Conductivity in aqueous solutions increases with increasing temperature because of higher ion mobility. This dependence is usually expressed as a relative change per 1C, commonly as percent per 1C; this value known as the slope of the particular aqueous solution. For this reason, conductivity readings are normally referenced to 25 1C. Fortunately, temperature sensors are readily available and can be integrated into the electronic circuitry, it being possible to correct the conductivity value and to bring it to its equivalent value at 25 1C automatically.

3.10.16 Conclusions The text above proves to the reader that there is no shortage of sensors for water. However, legislation pressure and technology advancements are continuously driving the search for novel rugged monitoring devices capable of detecting

Acknowledgments We gratefully acknowledge support from the funding institutions that have made possible our research in this area: the Madrid Community Government (IV PRICYT ref. CM-S-505/ AMB/0374), the European Regional Development Fund, the European Social Fund, the Spanish Ministry of Science and Innovation (CTQ2006-15610-C02-01-BQU and CTQ200914565-C03-01), and the UCM-Banco Santander (GR58-08910072).

References Aller RC and Zhu Q (2006) Optical pH Sensor. US Pat. 973,663, 18 May 2006. Amao Y and Nakamura N (2005) An optical sensor with the combination of colorimetric change of a-naphtholphthalein and internal reference luminescent dye for CO2 in water. Sensors and Actuators B: Chemical 107: 861--865.

260

Online Monitoring Sensors

Andres RT and Narayanaswamy R (1997) Fibre-optic pesticide biosensor based on covalently immobilized acetylcholinesterase and thymol blue. Talanta 44: 1335. Belgrader P, Benett W, Hadley D, et al. (1999) PCR detection of bacteria in seven minutes. Science 284: 449--450. Bellerby RGJ, Olsen A, Johannessen T, and Croot P (2002) A high precision spectrophotometric method for on-line shipboard seawater pH measurements: The automated marine pH sensor (AMpS). Talanta 56: 61--69. Brereton RG (2007) Applied Chemometrics for Scientists. Chichester: Wiley. Brown AP, Koval C, and Anson FC (1976) Illustrative electrochemical behavior of reactants irreversibly adsorbed on graphite electrode surfaces. Journal of Electroanalytical Chemistry 72: 379--387. Burke CS, Markey A, Nooney RI, Byrne P, and McDonagh C (2006) Development of an optical sensor probe for the detection of dissolved carbon dioxide. Sensors and Actuators B: Chemical 119: 288--294. Ca´pitan-Vallvey LF, Fernandez Ramos MD, Avidad R, and Deheidel MKA (2001) Determination of the pesticide morestan by means of a single-use phosphorimetric sensor. Analytica Chimica Acta 440: 131--141. Castro AM, Delgado J, and Orellana G (2005) Hydrocarbon in water sensing with PTFE membranes doped with a luminescent Ru(II) poly(pyridyl) complex. Journal of Material Chemistry 15: 2952--2958. Chan CM, Fung CS, Wong KY, and Lo W (1998) Evaluation of a luminescent ruthenium complex immobilized inside Nafion as optical pH sensor. Analyst 123: 1843--1847. Chang AC, Gillespie JB, and Tabacco MB (2001) Enhanced detection of live bacteria using a dendrimer thin film in an optical biosensor. Analytical Chemistry 73: 467--470. Cho DH, Chung KC, and Park MY (1998) Hydrogen ion-selective membrane electrodes based on alkyldibenzylamines as neutral carriers. Talanta 47: 815--821. Chuang H, Macuch P, and Tabacco MB (2001) Optical sensors for detection of bacteria. 1. General concepts and initial development. Analytical Chemistry 73: 462--466. Creaser CS, Hutchinson WE, and Stephenson GR (2002) An attenuated total reflectance optrode sensor incorporating infrared-active carbonylmetal probe groups. Sensors and Actuators B: Chemical 82: 150--157. Dominguez-Vidal A, Ortega-Barrales P, and Molina-Diaz A (2007) Environmental water samples analysis of pesticides by means of chemometrics combined with fluorimetric multioptosensing. Journal of Fluorescence 17: 271--277. El-Giar EE-DM and Wipf DO (2007) Microparticle-based iridium oxide ultramicroelectrodes for pH sensing and imaging. Journal of Electroanalytical Chemistry 609: 147--154. Ertekin K and Alp S (2006) Enhanced emission based optical carbon dioxide sensing in presence of perfluorochemicals (PFCs). Sensors and Actuators B: Chemical 115: 672--677. Fenster C, Smith AJ, Abts A, Milenkovic S, and Hassel AW (2008) Single tungsten nanowires as pH sensitive electrodes. Electrochemistry Communications 10: 1125--1128. Flannery D, James SW, Tatam RP, and Ashwell GJ (1997) pH sensor using Langmuir– Blodgett overlays on polished optical fibers. Optical Letters 22: 567--569. Gupta R and Chaudhury NK (2007) Entrapment of biomolecules in sol–gel matrix for applications in biosensors: Problems and future prospects. Biosensors and Bioelectronics 22: 2387--2399. Hobson NS, Tothill I, and Turner APF (1996) Microbial detection. Biosensors and Bioelectronics 11: 455--477. Holm AH, Vase KH, Winther-Jensen B, Pedersen SU, and Daasbjerg K (2007) Evaluation of various strategies to formation of pH responsive hydroquinoneterminated films on carbon electrodes. Electrochimica Acta 53: 1680--1688. Hulth S, Aller RC, Engstrom P, and Selander E (2002) A pH plate fluorosensor (optode) for early diagenetic studies of marine sediments. Limnology and Oceanography 47: 212--220. Ivanov AN, Lukachova LV, Evtugyn GA, et al. (2002) Polyaniline-modified cholinesterase sensor for pesticide determination. Bioelectrochemistry 55: 75--77. Jang A, Lee JH, Bhadri PR, et al. (2005) Miniaturized redox potential probe for in situ environmental monitoring. Environmental Science and Technology 39(16): 6191--6197. Jankowska H, Neffe S, and Swiatkowski A (1981) Investigations of the electrochemical properties of activated carbon and carbon black. Electrochimica Acta 26: 1861--1866. Janotta M, Karlowatz M, Vogt F, and Mizaikoff B (2003) Sol–gel based mid-infrared evanescent wave sensors for detection of organophosphate pesticides in aqueous solution. Analytica Chimica Acta 496: 339--348. Jero´nimo PCA, Arau´jo AN, Conceic¸a˜o BSM, and Montenegro M (2007) Optical sensors and biosensors based on sol–gel films. Talanta 72: 13--27.

Ji J, Schanzle JA, and Tabacco MB (2004) Real-time detection of bacterial contamination in dynamic aqueous environments using optical sensors. Analytical Chemistry 76: 1411--1418. Jin WJ, Costa-Fernandez JM, and Sanz-Medel A (2001) Room temperature phosphorescence pH optosensor based on energy transfer. Analytica Chimica Acta 431: 1--9. Kaden H, Jahn H, and Berthold M (2004) Study of the glass/polypyrrole interface in an all-solid-state pH sensor. Solid State Ionics 169: 129--133. Kahlert H (2008) Functionalized carbon electrodes for pH determination. Journal of Solid State Electrochemistry 12(10): 1255--1266. Klotz A, Brecht A, Barzen C, et al. (1998) Immunofluorescence sensor for water analysis. Sensors and Actuators B: Chemical B51: 181--187. Lario-Garcia J and Pallas-Areny R (2006) Constant-phase element identification in conductivity sensors using a single square wave. Sensors and Actuators A: Physical 132: 122--128. Lau KT, Shepherd R, and Diamond D (2006) Solid state pH sensor based on light emitting diodes (LED) as detector platform. Sensors 6: 848--859. Lawrence NS and Robinson KL (2007) Molecular anchoring of anthracene-based copolymers onto carbon nanotubes: Enhanced pH sensing. Talanta 74: 365--369. Lee JH, Seo Y, Lim TS, Bishop PL, and Papautsky I (2007) MEMS Needle-type sensor array for in situ measurements of dissolved oxygen and redox potential. Environmental Science and Technology 41: 7857--7863. Lei Yao H, Ghafar-Zadeh E, Shabani A, Chodavarapu V, and Zourob M (2008a) CMOS capacitive sensor system for bacteria detection using phage organisms. Proceedings of the 21st Canadian Conference on Electrical and Computer Engineering, pp. 877–880. Lei Yao H, Chodavarapu V, Shabani A, Allain B, Zourob M, and Mandeville R (2008b) CMOS imager microsystem for multi-bacteria detection. Joint 6th International Northeast Workshop on Circuits and Systems and TAISA Conference, pp. 137–140. Lieberman SH, Inman SM, and Theriault GA (1991) Laser-induced fluorescence over optical fibers for real-time in situ measurement of petroleum hydrocarbons in seawater. Oceans 1: 509--514. Li JP, Peng TZ, and Fang C (2002) Screen-printable sol-gel ceramic carbon composite pH sensor with a receptor zeolite. Analytica Chimica Acta 455: 53--60. Li X and Meijer GCM (2008) A high-performance interface for grounded conductivity sensors. Measurement Science and Technology 19: 1--7. Li ZZ, Niu CG, Zeng GM, et al. (2006) A novel fluorescence ratiometric pH sensor based on covalently immobilized piperazinyl-1,8-naphthalimide and benzothioxanthene. Sensors and Actuators B: Chemical 114: 308--315. Mallat E, Barzen C, Klotz A, Brecht A, Gauglitz G, and Barcelo D (1999) River analyzer for chlorotriazines with a direct optical immunosensor. Environmental Science and Technology 33: 965--971. Marazuela MD and Moreno-Bondi MC (2002) Fiber-optic biosensors – an overview. Analytical and Bioanalytical Chemistry 372: 664--682. Marrazza G, Chianella I, and Mascini M (1999) Disposable DNA electrochemical biosensors for environmental monitoring. Analytica Chimica Acta 387: 297--307. Michie WC, Culshaw B, McKenzie I, et al. (1995) Distributed sensor for water and pH measurements using fiber optics and swellable polymeric systems. Optical Letters 20: 103--105. Mills A and Eaton K (2000) Optical sensors for carbon dioxide: An overview of sensing strategies past and present. Quı´mica Analı´tica 19(supplement 1): 75--86. Mu¨ller B and Hauser PC (1996) Fluorescence optical sensor for low concentrations of dissolved carbon dioxide. Analyst 121: 339--343. Neurauter G, Klimant I, and Wolfbeis OS (2000) Fiber-optic microsensor for high resolution pCO2 sensing in marine environment. Fresenius Journal of Analytical Chemistry 366: 481--487. Niu CG, Gui XQ, Zeng GM, Guan AL, Gao PF, and Qin PZ (2005) Fluorescence ratiometric pH sensor prepared from covalently immobilized porphyrin and benzothioxanthene. Analytical and Bioanalytical Chemistry 383: 349--357. Nivens DA, Schiza MV, and Angel SM (2002) Multilayer sol–gel membranes for optical sensing applications: Single layer pH and dual layer CO2 and NH3 sensors. Talanta 58: 543--550. Orellana G, Lo´pez-Rodas MV, Costas E, Haigh D, and Maneiro E (2009) Biosensors Based on Microalgae for the Detection of Environmental Pollutants. PCT Pat. Appl. ES2008000465, 29 January 2009. Orellana G, Moreno-Bondi MC, Segovia E, and Marazuela MD (1992) Fiber-optic sensing of carbon dioxide based on excited-state proton transfer to a luminescent ruthenium(II) complex. Analytical Chemistry 64: 2210--2215. Oter O, Ertekin K, Topkaya D, and Alp S (2006) Room temperature ionic liquids as optical sensor matrix materials for gaseous and dissolved CO2. Sensors and Actuators B: Chemical 117: 295--301.

Online Monitoring Sensors Pen˜a-Va´zquez E, Maneiro E, Pe´rez-Conde C, Moreno-Bondi MC, and Costas E (2009) Microalgae fiber optic biosensors for herbicide monitoring using sol–gel technology. Biosensors and Bioelectronics 24: 3538--3543. Poghossian A, Berndsen L, and Schoning MJ (2003) Chemical sensor as physical sensor: ISFET-based flow-velocity, flow-direction and diffusion-coefficient sensor. Sensors and Actuators B: Chemical 95: 384--390. Prissanaroon W, Brack N, Pigram PJ, Hale P, Kappen P, and Liesegang J (2005) Fabrication of patterned polypyrrole on fluoropolymers for pH sensing applications. Synthetic Metals 154: 105--108. Ruedas Rama MJ, Ruiz Medina A, and Molina Diaz A (2002) Use of a solid sensing zone implemented with unsegmented flow analysis for simultaneous determination of thiabendazole and warfarin. Analytica Chimica Acta 459: 235--243. Salinas-Castillo A, Fernandez-Sanchez JF, Segura-Carretero A, and FernandezGutierrez A (2004) A facile flow-through phosphorimetric sensing device for simultaneous determination of naptalam and its metabolite 1-naphthylamine. Analytica Chimica Acta 522: 19--24. Sanchez-Barragan I, Costa-Fernandez JM, and Sanz-Medel A (2005) Tailoring the pH response range of fluorescent-based pH sensing phases by sol–gel surfactants coimmobilization. Sensors and Actuators B: Chemical 107: 69--76. Scholz F, Steinhardt T, Kahlert H, Poerksen JR, and Behnert J (2005) Teaching pH measurements with a student-assembled combination quinhydrone electrode. Journal of Chemical Education 82: 782--786. Scho¨ning MJ, Abouzar MH, and Poghossian A (2009) pH and ion sensitivity of a fieldeffect EIS (electrolyte–insulator–semiconductor) sensor covered with polyelectrolyte multilayers. Journal of Solid State Electrochemistry 13: 115--122. Schroeder CR, Weidgans BM, and Klimant I (2005) pH fluorosensors for use in marine systems. Analyst 130: 907--916. Szepesvary E and Pungor E (1971) Potentiometric determination of acids and bases with a silicone rubber-based graphite electrode as indicator electrode. Analytica Chimica Acta 54: 199--208. Tabacco MB, Uttamlal M, McAllister M, and Walt DR (1999) An autonomous sensor and telemetry system for low-level pCO2 measurements in seawater. Analytical Chemistry 71: 154--161. Taboada Sotomayor MP, De Paoli MA, and de Oliveira WA (1997) Fiber-optic pH sensor based on poly(o-methoxyaniline). Analytica Chimica Acta 353: 275--280. Vuppu S, Kostov Y, and Rao G (2009) Economical wireless optical ratiometric pH sensor. Measurement Science and Technology 20: 1--7. Walsh JE, MacCraith BD, Meaney M, et al. (1996) Sensing of chlorinated hydrocarbons and pesticides in water using polymer coated mid-infrared optical fibers. Analyst 121: 789--792. Walt DR, Tabacco MB, and Utamlal M (2000) Fiber Optic Sensor for Long-Term Analyte Measurements in Fluids.US Pat. 6285807, 4 September 2001. Wang J, Chen L, Mulchandani A, Mulchandani P, and Chen W (1999) Remote biosensor for in-situ monitoring of organophosphate nerve agents. Electroanalysis 11: 866--869. White J, Kauer JS, Dickinson TA, and Walt DR (1996) Rapid analyte recognition in a device based on optical sensors and the olfactory system. Analytical Chemistry 68: 2191--2202. Wiegran K, Trapp T, and Cammann K (1999) Development of a dissolved carbon dioxide sensor based on a coulometric titration. Sensors and Actuators B: Chemical B57(1–3): 120--124. Wolfbeis OS, Kovacs B, Goswami K, and Klainer SM (1998) Fiber-optic fluorescence carbon dioxide sensor for environmental monitoring. Mikrochimica Acta 129: 181--188. Wong LS, Brocklesby WS, and Bradley M (2005) Fibre optic pH sensors employing tethered non-fluorescent indicators on macroporous glass. Sensors and Actuators B: Chemical 107: 957--962. Wro´blewski W, Rozniecka E, Dybko A, and Brzo´zka Z (1998) Cellulose-based bulk pH optomembranes. Sensors and Actuators B: Chemical 48: 471--475. Xiang-Hong W and Shuo W (2008) Sensors and biosensors for the determination of small molecule biological toxins. Sensors 8: 6045--6054.

Relevant Websites http://www.abraxiskits.com Abraxiskits. http://www.advnt.org Advnt Biotechnologies. http://www.analyticaltechnology.com Analytical Technology.

http://www.appliedbiosystems.com Applied Biosystems. http://www.astisensor.com ASTi: Advanced Sensor Technologies, Inc. http://www.beaconkits.com Beacon Analytical Systems. http://www.biomerieux.es Biomerieux. http://www.biophagepharma.net BiophagePharma. http://www.campbellsci.com Campbell Scientific. http://www.dwscientific.co.uk Don Whitley Scientific. http://www.endress.com Endress þ Hauser. http://www.environnement-sa.com Environnement S.A. http://www.grundfosalldos.com Grundfos Alldos. http://www.interlab.es Grupo Interlab. http://www.hach.com Hach. http://www.hannainst.com HANNA Instruments. http://www.horiba.com HORIBA. http://www.idahotech.com Idaho Technology. http://www.sensafe.com Industrial Test Systems. http://www.in-situ.com In-Situ. http://www.invitrogen.com Invitrogen. http://www.ionode.com.au Ionode. http://www.jenway.com Jenway. http://www.martekinstruments.com Martek Instruments. http://www.merck-chemicals.com Merck. http://in.mt.com Metller Toledo. http://www.myronl.com Myron L Company. http://earthobservatory.nasa.gov NASA: Earth Observatory. http://lsda.jsc.nasa.gov NASA: Life Science Data Archive. http://modis.gsfc.nasa.gov NASA: MODIS. http://oceancolor.gsfc.nasa.gov NASA: Ocean Color Web. http://www.nexsens.com Nexsens Technology. http://www.nico2000.net NICO 2000. http://www.palintest.com Palintest. http://www.pharmaleads.com Pharmaleads. http://www.presens.de PreSens. http://www.qtlbio.com QTL Biodetection.

261

3.11 Standardized Methods for Water-Quality Assessment$ BC Gordalla, Karlsruhe Institute of Technology, Karlsruhe, Germany & 2011 Elsevier B.V. All rights reserved.

3.11.1 3.11.2 3.11.3 3.11.3.1 3.11.3.1.1 3.11.3.1.2 3.11.3.1.3 3.11.3.2 3.11.3.2.1 3.11.3.2.2 3.11.3.3 3.11.3.4 3.11.3.5 3.11.4 3.11.4.1 3.11.4.2 3.11.4.3 3.11.4.4 3.11.4.4.1 3.11.4.4.2 3.11.4.5 3.11.4.6 3.11.4.7 3.11.4.7.1 3.11.4.7.2 3.11.4.8 3.11.5 3.11.6 References

Introduction Features of Standards and Standardization Standardization Organizations Delivering Water-Testing Standards and Their TCs International Organization for Standardization General ISO/TC 147 – water quality Further ISO committees relevant for water examination and quality aspects Comite´ Europe´en de Normalisation – European Committee for Standardization CEN/TC 230 – water analysis Further TCs in CEN relevant for water-testing issues Coordination of Activities in CEN and ISO and Mutual Adoption of Documents The Role of the NSBs ASTM Items Covered by Standardization in the Field of Water Examination Terminology, Analytical Strategies, Validation, and Quality Control Sampling, Sample Pretreatment, and Basic Operations Physical–Chemical and Other Basic Parameters for Water-Quality Assessment Methods for Determination of Individual Water Constituents and Defined Groups of Substances Inorganic water constituents Methods for determination of organic compounds or jointly determinable groups of compounds Radiological Methods Microbiological Methods Biotesting Biodegradability Ecotoxicity and bioeffect testing Methods for Assessment of Water Bodies Resume and Outlook List of Standards

3.11.1 Introduction Activities to design and collect standardized methods of water examination go back to the end of the nineteenth century. A very popular example is the often-cited Standard Methods for the Examination of Water and Wastewater (APHA et al., 2005), jointly edited by the American Public Health Organization (APHA), the American Water Works Association (AWWA), and the Water Environment Federation (WEF). This collection was first published in 1905 and has now grown to a 1200-page book that is revised every 5 years by an editorial board supported by a large number of experts. Water analysis is performed worldwide using standardized protocols that have been recommended, collected, or developed by different organizations on a national or super-national level, ranging from authorities, technical, scientific, or water management associations to organizations explicitly dedicated to standardization. Water-testing standards in the narrower sense are delivered by national, European, or international standards bodies. $

Dedicated to Dr. Sibylle Schmidt, who chaired the technical committees ISO/TC 147 – water quality and CEN/TC 230 – water analysis for many years.

263 264 265 265 265 265 266 267 267 268 268 268 269 269 270 270 273 275 275 278 280 280 283 283 285 286 286 286 301

This chapter mainly focuses on the standardization activities in the field of water examination of the Comite´ Europe´en de Normalisation (CEN) – the European Committee for Standardization – which coordinates European standardization, and of the International Organization for Standardization (ISO), which deals with standardization on an international level. In these organizations, the participating countries are represented by their national standards bodies (NSBs). In addition, the standards’ portfolio on water testing of American Society for Testing and Materials (ASTM) International will be considered, which is an international standardization organization that emerged from the former ASTM. It must be stressed that these standardization organizations develop standardized methods for water examination, but that it is not their task nor are they entitled to set legal norms for water quality. The major benefit of standardized methods is the comparability of results. Comparability is especially crucial for

1. long-term observations to evidence changes or trends; 2. large-scale monitoring programs, for example, of river catchment areas; and

263

264

Standardized Methods for Water-Quality Assessment

3. reasons of equal treatment, when measured analytical parameters serve to check compliance with legal norms or to calculate emission-based taxes.

the large group of effect-, toxicity-, and ecotoxicity-testing protocols are conventional parameters.

Justifiability is a further reason, which is why water-testing standards are useful for legal and supervision purposes. The most simple case is that distinct standardized methods are directly prescribed in laws or ordinances, as is practiced, for example, in German wastewater regulations, where the calculation of sewage taxes (AbwAG, 2005) as well as compliance with the wastewater ordinance (AbwV, 2002), which gives the minimum requirements for the quality of discharged wastewater, are based on parameters measured according to official standards. As an alternative to referring directly to standards, in some regulatory stipulations specifications are made for methods to be suitable to check compliance with legal norms. As water-testing standards are set up for a specified application range, and often statistical performance data are also given therein, use of a standard may ensure that the respective specifications are met. For instance, chemical monitoring of the so-called priority pollutants based on the European Water Framework Directive (WFD) (2000/60/EC, 2000; 2008/105/EC, 2008) has to be performed by validated and documented methods which meet specified performance criteria concerning measurement uncertainty and limit of quantification (2009/90/EC, 2009). The European Drinking Water Directive (98/83/EC, 1998) gives minimum requirements for trueness, precision, and limit of detection for most chemical ingredients to be analyzed (see Chapter 3.07 Measurement Quality in Water Analysis). For determination of microbiological parameters, this directive actually refers to distinct standards. In the US, the National Technology Transfer Act of 1995 (see the ‘Relevant websites’ section for NTTA) urges US authorities to make use of already-existing private sector standards wherever possible. Thus, the Environmental Protection Agency (EPA), which has developed a large methods portfolio for water examination of its own (see the ‘Relevant websites’ section for EPA), also refers to standardized methods published by standardization organizations, for example, in order to ensure compliance with the Safe Drinking Water Act or the Clean Water Act. A detailed inventory of standardized methods approved for national environmental analysis is available on the Internet (see the ‘Relevant websites’ section for NEMI). In many cases, standardized water-testing methods dedicated to defined analytes are based on analytical methods published in scientific literature, which have been transferred into a standardized version (1) being given a single text format and further specifications concerning scope and procedural details compared to a publication in a journal, (2) based on a consensus of experts going beyond peer review, and, often, (3) after a validation step exceeding in-house validation. Besides, there are several core parameters especially designed for water analysis to which a standardized protocol is inherent. Very prominent ones are the so-called conventional – operationally defined – parameters, such as biochemical oxygen demand (BOD), chemical oxygen demand (COD), dissolved organic carbon (DOC), or adsorbable organic halogens (AOX). Chapter 3.01 Sum Parameters: Potential and Limitations is dedicated to those parameters. In addition,

3.11.2 Features of Standards and Standardization Some features that are typical for water analytical standards delivered by standards bodies are outlined below, but many of them apply to standardized methods for water examination in general. Standards have a given format with pre-defined specifications to be made. Standardized methods and especially water-testing standards delivered by standard bodies are written in a given text format that may differ for the respective standardization organizations in wording, but for chemical parameters, generally specifications are made on the following items: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

14.

definition of the analytical parameter; definition of terms; normative references; principle of the method; scope (matrices, range of concentrations reliably determinable); interferences; sampling and conservation; reagents including specifications concerning purity, standards; apparatus, equipment; all steps of analytical procedure (calibration, determination); calculation/evaluation; expression of results; precision (and bias, where applies, e.g., results from an interlaboratory trial, repeatability, reproducibility, and recovery); and test report.

Of course, for microbiological examination or biological testing, the catalog of items has to be modified accordingly. An essential feature of water analytical standards is that they are set up for a defined scope with respect to concentration range and matrices. Specifications or requirements concerning the purity of chemicals and the performance of instruments are given in a general way, but not tailored to the commercial products of a special manufacturer, as an important objective of standards bodies is to facilitate the exchange of goods and services. Thus, materials, instruments, and methods which are unique or on which a monopoly is held are not suitable for standardization. Extended or additional information (e.g., detailed results of round-robin tests, example chromatograms, technical drawings, and additional references) is given in annexes to the standards. In CEN, ISO, and ASTM, a distinction is made between normative (mandatory) and informative (nonmandatory) annexes, the latter providing information that is illustrative or serves deeper understanding. In some cases, informative annexes specify modifications which allow an international standard to be used within respective national regulations. An example is the definition of the concept lowest ineffective dilution (LID) for wastewater testing in the informative annex of EN ISO 5667-16 and the specifications

Standardized Methods for Water-Quality Assessment

how to determine this quantity given in the informative annexes of several standards for ecotoxicity testing (e.g., EN ISO 20079, EN ISO 11348-1, -2, and -3). In standards of CEN, ISO, and ASTM, information on former versions is also given. ASTM standards can be obtained in a so-called redline version, where changes compared to the former version are marked. Extended rules for the format of a standard and the items to be covered therein are given in basic standards, for example, for ISO standards in ISO/IEC (2004). For ASTM, regulations concerning procedural details of standards development, or on format and style of the documents, are given in the Red Book (ASTM, 2007), the Green Book (ASTM, 2009a), and the Blue Book (ASTM, 2009b). Standards are the result of a formalized process of consensus building. Characteristics of the standardization process are openness to the public and participation of the so-called interested parties. In the case of standardized methods for water examination, the interested parties include 1. legislator and authorities, who set legal norms in the field of water quality and need methods to control compliance; 2. water suppliers, sewerage boards, and industries which release effluents, whose compliance is checked using the standardized methods or who might be charged with emission-based taxes based on the results; and 3. universities, manufacturers of analytical instruments, and analytical laboratories, who develop and apply the standardized methods.

The development of a standard is initiated by a formal new work item proposal to the standardization organization, more precisely, to the responsible technical committee (TC). The method protocols finally delivered as standards are developed in a regulated multistage process of elaborating, commenting, and voting. In CEN and ISO, a time schedule has to be met for this process, 2-, 3-, or 4-year development tracks are possible in ISO. In European standardization, the drafts of standards (prEN) are presented to the public in the member countries by the NSBs for comments. Further public involvement is established depending on the conditions in the respective countries, for example, by giving the opportunity to submit new work item proposals to the NSBs or by publicly announcing for participants in working groups (WGs) or interlaboratory trials. Standards are subject to quality control. In many cases, the standardization process includes a validation step to be prerequisite for a method to be considered as a standard for water examination. Validation might be based on certified reference materials or on a successful interlaboratory trial. The latter is mandatory in ISO/TC 147 and in CEN/TC 230 for standardized methods on continuously measurable variables. Interlaboratory trials are not restricted to chemical parameters, but are performed also for microbiological methods, biotests, and methods for ecological assessment. Validation is an important element of standardization, and sometimes experiences and results of interlaboratory trials are additionally published in journals (e.g., Reifferscheid et al., 2008; Stottmeister et al., 2009). In ASTM, a collaborative study to determine precision

265

and bias of proposed new methods for water testing is obligatory (ASTM D2777). In ISO, CEN, and ASTM, standards once delivered are periodically checked if they are relevant and state of the art, and then either confirmed, revised, or withdrawn. In ISO or CEN, this decision is made after a routine query among the member countries every 3 or 5 years, respectively. ASTM performs a routine reballoting every 5 years.

3.11.3 Standardization Organizations Delivering Water-Testing Standards and Their TCs Within the standardization organizations, standardization work is organized in TCs dedicated to particular subject areas. The actual technical work on specific standards is done in WGs or task groups of project-related specialists. Their task is to develop the technical content of the standards (i.e., the method protocols), to elaborate the documents, and to organize validation measures (i.e., interlaboratory trials), if mandatory. WGs are usually disbanded after having finished their standardization project.

3.11.3.1 International Organization for Standardization 3.11.3.1.1 General ISO, founded in 1946, coordinates standardization activities on an international level. The member countries are represented in ISO by their NSBs. Membership is possible as a participating member (P-member) with the right and the obligation to vote on the documents, or with an observing status (O-member). In balloting processes, the vote of each P-member weighs equally (one country – one vote). The adoption of ISO standards into the respective national collections of official standards is optional; it is up to the NSBs to decide about this.

3.11.3.1.2 ISO/TC 147 – water quality This TC was created in 1971 in order to develop standardized methods for water quality control (Schmidt, 2001, 2003). According to its scope, ISO/TC 147 is in charge of ‘‘standardization in the field of water quality, including definition of terms, sampling of waters, measurement and reporting of water characteristics.’’ In addition, in situ sediments are dealt with in this committee. It was chaired by the US standards body American National Standards Institute (ANSI) until 1983, since then the secretary has been held by the German standards body Deutsches Institut fu¨r Normung (DIN) (Schmidt and Wunder, 1988; Schmidt, 1992). At present, 35 countries participate in ISO/TC 147 as P-members, and a further 52 countries are informed about the activities of the committee as O-members. About 30 active WGs are working on current standardization projects, organized in five subcommittees (SCs) according to the different subject areas to be covered for water examination (see Table 1). The WG on radiological methods is directly allocated to ISO/TC 147. Up to 1989, it was organized in a SC of its own. The majority of the documents authored by ISO/TC 147 is in the format of international standard (IS), which is the intended main deliverable of ISO. Some methods with a normative content are delivered as so-called technical

266

Standardized Methods for Water-Quality Assessment

Table 1

Subcommittees (SCs) and working groups (WGs) of the technical committee ISO/TC 147 – water quality current as of April 2010

Panel

Title

Chair

Members

TC 147 WG 4

Water quality Radiological methods

Germany (DIN) France (AFNOR)

35 P, 52 O

SC 1

Terminology

South Africa (SABS)

18 P, 26 O

SC 2 WG WG WG WG WG WG WG WG WG WG WG WG WG WG WG WG WG

Physical, chemical, and biochemical methods Polycyclic aromatic hydrocarbons (PAH) Ion chromatography methods Flow analysis methods Precision and accuracy Antimony, arsenic, and selenium GC-MS for groups of nonpolar substances Glyphosate and AMPA PFOS and PFOA SPME Chloroalkanes Color determination Mercury determination Dissolved oxygen determination Discrete analysis Chlorinated naphthalenes Volatiles Chemical oxygen demand

Germany (DIN) The Netherlands (NEN) Germany (DIN) Germany (DIN) Germany (DIN) United Kingdom (BSI) The Netherlands (NEN) France (AFNOR) Japan (JISC) DIN Germany (DIN) DIN Germany (DIN) Norway (SN) Germany (DIN) Germany (DIN) United Kingdom (BSI) Canada (SCC) Germany (DIN) United Kingdom (BSI)

26 P, 25 O

19 33 38 48 52 53 55 56 57 59 60 61 62 63 NN NN NN

SC 4 WG WG WG WG WG WG WG WG WG

Microbiological methods E. coli and other coliforms Sulfite-reducing Clostridium Salmonella Legionella Analytical quality control of microbiological media Cryptosporidium/Giardia Uncertainty of measurement Legionella by PCR E. coli/coliforms with liquid enrichment

Germany (DIN) Germany (DIN) Austria (ASI) United Kingdom (BSI) The Netherlands (NEN) France (AFNOR) United Kingdom (BSI) Finland (SFS) France (AFNOR) USA (ANSI)

28 P, 17 O

2 5 7 10 12 13 15 17 19

SC 5 WG WG WG WG WG WG

Biological methods Toxicity – bacteria and biodegradability Toxicity – invertebrates Toxicity – fish Toxicity – algae and aquatic plants Biological classification Genotoxicity

Germany (DIN) Germany (DIN) Germany (DIN) Sweden (SIS) Norway (SN) United Kingdom (BSI) Germany (DIN)

24 P, 19 O

1 2 3 5 6 9

SC 6 WG WG WG WG WG

Sampling (general methods) Design of sampling programmes Conservation methods Rivers and streams including groundwater Drinking water and water used for food and beverage processing Sampling of sludges and sediments

United Kingdom (BSI) United Kingdom (BSI) The Netherlands (NEN) United Kingdom (BSI) United Kingdom (BSI) United Kingdom (BSI)

23 P, 20 O

1 3 4 6 11

NN, to come; P, participating members; O, orbserving members.

specification (TS) based on a lower level of consensus. Nonnormative documents with a more informative or guideline character are published in the format of a technical report (TR). The portfolio of ISO/TC 147 comprises almost 250 active standards, they are included in the list of standards at the end of this chapter. On the website of ISO TC/147, extended information about the scope and structure, current standardization projects, and the standards portfolio of this committee is available (see the ‘Relevant websites’ section for ISO/TC 147).

An overview of the different steps of the standardization process in ISO/TC 147, the subsequent document stages, and queries among the member countries is given in Figure 1. The development of a standard is initiated by a formal new work item proposal submitted by one of the participating NSBs, in most cases with a first working draft added. To have the standardization project included in the working program, not only a majority is prerequisite, but also at least five countries must be willing to actively participate in the concerned WG. In two document stages, committee draft (CD) and draft of

Standardized Methods for Water-Quality Assessment

a standard (DIS), the members are given the opportunity to submit technical comments on the method. The project leader deals with these comments. An important step for all methods on parameters that are continuously measurable is the validation by a round-robin test in order to estimate repeatability and reproducibility. The philosophy of those external interlaboratory trials in ISO/TC 147 is as follows: the mean value is the reference value. An identical sample has to be measured according to the overall procedure at least in duplicate, preferably in three or four replicates. The evaluation of the test is performed based on ISO 5725-2. The data are checked for type 1 (Grubbs test), type 2 (Grubbs test), and type 3 (Cochran test) outliers; the percentage of relative outliers should not exceed 25%. The minimum performance to be achieved is that at least eight valid data sets and 24 outlier-free single data should remain. The coefficient of variation of reproducibility (interlaboratory) CVR should not exceed 30%. If a true value does exist, there will be additional requirements concerning bias, depending on the respective method. In ISO/TC 147, consensus building, decisions, and votes are made in writing by means of queries among the participating countries. Nevertheless, ISO/TC 147, its SCs and WGs meet every 18 months for a 1-week meeting. At this meeting, TC 147 and the SCs discuss principle or strategic issues of future work, for example, possible new work item proposals as well as installation or disbanding of WGs. In parallel, the experts in the WGs deal with the technical details of the methods

to be standardized, the comments received from the member countries, and the design and organization of interlaboratory trials to be performed. At present, about 50 standardization projects are going on in ISO/TC 147, 40 dealing with development of new standards, the rest with revision of already-existing standards induced by the periodical 3-year-review inquiries.

3.11.3.1.3 Further ISO committees relevant for water examination and quality aspects Hydrometric and hydrogeological aspects of aquatic systems (e.g., flow, velocity and discharge measurements, sediment transport, and measurement of groundwater levels) are dealt with in a separate TC, the ISO/TC 113 – hydrometry. Furthermore, ISO TC/147 is in liaison with ISO/TC 190 – soil quality. In the documents delivered by ISO/TC 147, basic documents are cited that have been published by other TCs or institutions. Prominent examples are the standard ISO 3696 on water for analytical purposes authored by ISO/TC 47 – chemistry, or the standard ISO 5725 on accuracy, especially part 2 dealing with repeatability and reproducibility, which was created by ISO/TC 69 – applications of statistical methods. Concerning metrological terms, quantities, and units, ISO standards refer to the International Vocabulary of Metrology (VIM) (ISO/IEC Guide 99, 2007) and to the International Union of Pure and Applied Chemistry (IUPAC, 2007, cited in ISO 80000-9).

Document stage

A committee member (e.g., DIN) submits 1 NP (new work item proposal) P-members vote

Simple majority + 5 members nominate experts

SC members vote and comment

Approval by simple majority + <3 × disagreement

2

66% Approval + <25% disagreement

P-members vote

CD (committee draft)

Interlab. trial

DIS or ISO/TS (technical specification) ISO/TR (technical report) for non-normative contents

Enquiry on DIS (Draft International Standard)

Final text for processing as FDIS (Final Draft International Standard)

Formal vote on FDIS 5 (proof check by secretariat, resp., project leader)

Final text of international standard

Publication of International Standard

ISO International Standard

4 66% Approval +

Deliverable

WD (working draft)

Building expert consensus

Consensus building 3 within TC/SC P-members vote and comment

267

TS TR

<25% disagreement

6

ISO Standard

Figure 1 Standardization process – setting up an ISO standard. CD, DIS, and FDIS (final draft of an International Standard) are three official document stages that require balloting of the members’ standards bodies. In ISO/TC 147 – water quality, an interlaboratory trial for validation is mandatory.

268

Standardized Methods for Water-Quality Assessment

3.11.3.2 Comite´ Europe´en de Normalisation – European Committee for Standardization On a European level, standardization is organized in the CEN, founded in 1961 with the goal of European harmonization (see also the ‘Relevant websites’ section for CEN). The 31 member states are represented in CEN and in its TCs by their NSBs. Initiatives for standardization projects may come from the NSB, or directly from the European Commission (EC), in case standards are developed in order to support European legislation. In European standardization, one draft stage (prEN – draft of a European standard) is released. It is published in the three language versions English, French, and German and distributed in the member countries for public comments (CEN inquiry). A European Standard (EN) has to be adopted as a national standard in the 31 member countries and supersedes national standards on the same topic. Furthermore, once a standardization project has been adopted in CEN on a topic, on a national level, no new projects on that topic are initiated, nor are revisions of existing standards undertaken (standstill obligation).

3.11.3.2.1 CEN/TC 230 – water analysis The TC CEN/TC 230 – water analysis was created in 1990 in order to establish methods for existing and future EU Directives on water quality (Schmidt, 2003). An overview of the structure of this committee is given in Table 2. At present, this TC deals mainly with the development of methods suitable to implement the monitoring programs of the WFD (2000/60/ EC). In doing so, WG 2 focuses on biological and ecological assessment methods for aquatic systems and the characterization of water bodies, which has to be performed according to annex V of the WFD. The task of WG 1 is to develop methods that are suitable to monitor the environmental quality standards (EQS) for priority pollutants and certain other pollutants according to annexes VIII and X of the WFD, listed in the daughter Directive 2008/105/EC and that meet the performance characteristics specified in daughter Directive 2009/90/EC on technical specifications for chemical analysis, that is, LOQo30% of EQS, expanded measurement uncertainty (k ¼ 2)o50% of EQS (see Chapter 3.07 Measurement Quality in Water Analysis). A strong impulse for these activities arose from the official mandate M/424 of 2008 given

Table 2

by the EC to CEN/TC 230 in order to develop or improve standardized methods in support of the WFD. At present, the main efforts are directed at the parameters organochlorine pesticides, pentabromodiphenylethers, tributyltin compounds, chloroalkanes, polycyclic aromatic hydrocarbons (PAH), furthermore to phytoplankton sampling from inland waters, determination of algal biovolume, and fish sampling. An overview of the various efforts to coordinate research, standardization, and policy in order to implement the WFD and the role of standards bodies of the different European countries therein is given in Quevauvillier et al. (2007). A list of standards and standardization projects of CEN/TC 230 is available (see the ‘Relevant websites’ section for CEN/ TC 230). The TC, its WGs, and task groups meet annually to discuss strategic topics and technical aspects. Much of the detailed technical work is also performed in expert workshops. Official votings and comments on the documents are requested by queries among the member countries of the committee and given in writing. European standards are adopted by weighted votes, as is common on many European panels.

3.11.3.2.2 Further TCs in CEN relevant for water-testing issues TCs on water testing and TCs dealing with environmental solid matrices keep each other informed about their activities. This is because water quality is influenced by adjacent solid matrices, for example, by leaching or runoff processes. Another reason is that after a preceding digestion or extraction step, the primary step of chemical analysis for testing solid matrices often is the same as for aqueous samples. The CEN/ TC 400 – horizontal standards in the fields of sludge, biowaste, and soil, emerged from the eponymous task force CEN BT/TF 151, deals with the harmonization of analytical standards for those matrices to avoid duplicate work and standards portfolios. A particular matrix tested in water technology is the intermediate phase of sludge, being formed in wastewatertreatment plants or during flocculation in drinking water treatment. CEN/TC 308 – characterization of sludges is dedicated to this subject area. The standards portfolio and working program of CEN/TC 308 and the working program of CEN/TC 400 are available on the Internet (see the ‘Relevant websites’ section for CEN/TC 308 or CEN/TC 400, respectively.).

Working groups (WGs) and task groups (TGs) of the technical committee CEN/TC 230 – water analysis, current as of April 2010

Panel

Title

Chair

TC 230 WG 1 WG 2 TG 1 TG 3 TG 4 TG 5 TG 6 TG 7

Water analysis Physical and chemical methods Biological and ecological assessment methods Invertebrates Aquatic macrophytes and algae Fish monitoring Waterbody characteristics Quality assurance Marine ecological methods

Germany (DIN) NN United Kingdom (BSI) United Kingdom (BSI) The Netherlands (NEN) Sweden (SIS) United Kingdom (BSI) Austria (ASI) Norway (SN)

Standardized Methods for Water-Quality Assessment 3.11.3.3 Coordination of Activities in CEN and ISO and Mutual Adoption of Documents There are many cases, where standardization activities on the same subject area take place on an international level and a European level as well. Both ISO/TC 147 – water quality and CEN/TC 230 – water analysis deliver standards on waterquality examination. In principle, their working programs would overlap to a great extent, with the consequence that many of the experts would meet in WGs of CEN and additionally of ISO. As both TCs are chaired by the same country, a division of labor between ISO/TC 147 and CEN/TC 230 has been established in order to avoid double or contradictory work. This is possible, because the Vienna Agreement allows mutual adoption of standards developed in CEN and ISO. At present, technical work in CEN/TC 230 is done only on standards that are especially needed for the requirements of the WFD and that are not intended to be included into the ISO portfolio. The majority of physical, chemical, and microbiological methods, as well as testing methods for ecotoxicity or degradability, are developed in ISO/TC 147 and adopted into the European standards collection. This can be done by unique acceptance procedure (UAP). This means that the final ISO standard is presented for balloting in CEN and either accepted as an EN as is or not accepted. The alternative is the so-called parallel vote (PV), where a draft stage of the standard is circulated and commented among the members of both ISO/TC 147 and CEN/TC 230 (see the ‘Relevant websites’ section for Vienna Agreement). EN ISO standards make up 40% or 70% of the active standards of ISO/TC 147 or of CEN/ TC 230, respectively. For reasons of space, ISO standards adopted as European standards are only cited once in the list of standards at the end of this chapter, as EN ISO standards, which includes reference to ISO.

3.11.3.4 The Role of the NSBs The member countries participate in standardization work done on European and international levels by their NSBs. In detail, the NSBs 1. appoint experts to participate in the TCs, SCs, working or task groups of ISO and/or CEN; 2. may submit new work item proposals to ISO and/or CEN; 3. agree/disagree to new work item proposals of ISO and/or CEN; 4. give technical comments on draft stages (e.g., ISO/CD, ISO/DIS, and prEN) of standards in development; 5. cast national vote to final documents of ISO and/or CEN; 6. may adopt ISO standards as national standards; and 7. transpose European standards into national standards (CEN members only). In the field of water analysis, these tasks are fulfilled in Germany by the TC DIN NA 119-01-03 AA (Arbeitssausschuss – Wasseruntersuchung – TC – water analysis), which was created by the German Standards Body Deutsches Institut fu¨r Normung (DIN) – German Institute for Standardization in the early 1970s in order to deliver methods for water examination as German standards. This committee, consisting of experts representing the interested parties in Germany and supported

269

by about 30 WGs, mirrors the committees CEN/TC 230 and ISO/TC 147 and additionally develops water analytical standards on a national level, where these are required for special needs. A list of active standards and current standardization projects is available on the Internet (see the ‘Relevant websites’ section for DIN NA 119-01-03 AA). Standards passed by the TC DIN NA 119-01-03 AA Wasseruntersuchung (i.e., DIN, DIN EN, DIN ISO, or DIN EN ISO standards) are available as stand-alone standards, and besides, they are included into the German loose-leaf collection of standard methods for the examination of water, wastewater, and sludge (Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung – DEV) (Wasserchemische Gesellschaft and Normenausschuss Wasserwesen im DIN, 2010; see also the ‘Relevant websites’ section for DEV), edited and updated since 1935 by the German Water Chemical Society – a division of the German Chemical Society (GDCh). This collection of methods was taken as the basis for DIN standardization in the field of water analysis when it started in the 1970s, and up to now, the Water Chemical Society has been closely involved in the standardization activities. Thus, for national German standards authored by the AA Wasseruntersuchung (standards series 38 400), the process of method development and especially the interlaboratory trial has to be recorded in extended documentation, the so-called Validierungsdokument. These documents are hosted on the website of the GDCh (see the ‘Relevant websites’ section for GDCh).

3.11.3.5 ASTM ASTM is an industry-driven international standardization organization, which emerged from the former American Society for Testing and Materials, founded in 1898. It produces consensus-based standards for goods and services. The technical experts in the WGs represent producers, users, government, and academics. In ASTM, much impetus comes from the manufacturers of analytical instruments. New work items are often started at the initiative of the WGs. The WGs are responsible for validation of the methods. The prepared documents have to pass voting in the respective SC, in the main TC, and at ASTM society level. Besides membership of organizations, for example, companies or standards bodies, in ASTM, personal membership in the committees is possible as well. Personal members have the right to vote in ballots, with the restriction of one vote each per institution involved in the respective standardization project (one company, one vote). Methods for water examination are developed in the ASTM TC D19, created in 1932, with currently around 340 members (see the ‘Relevant websites’ section for ASTM D19). D19 meets twice a year for 4 days of technical meetings and a workshop on relevant topics, with approximately 70 participants. The technical work is organized in the SCs listed in Table 3. According to its scope (see the ‘Relevant websites’ section for ASTM D19 Scope), D19 deals with 1. sampling and analysis of water, waterborne materials, and wastes, water-formed deposits and fluvial sediments; 2. surface water hydraulics and hydrologic measurements; 3. the determination of the performance of materials or products used to modify water characteristics; and

270

Standardized Methods for Water-Quality Assessment

Table 3 water

Subcommittees of the ASTM technical committee D19 on

D19.02 D19.03

Quality systems, specification, and statistics Sampling water and water-formed deposits, analysis of water for power generation and process use, on-line water analysis, and surveillance of water Methods of radiochemical analysis Inorganic constituents in water Methods for analysis of organic substances in water Sediments, geomorphology, and open-channel flow Membranes and ion-exchange materials Water microbiology

D19.04 D19.05 D19.06 D19.07 D19.08 D19.24

4. the determination of the corrosivity or deposit-forming properties of water. ASTM D19 also covers hydrological, water technological, and corrosion aspects. A special SC is dedicated to testing methods in connection with membrane technology and ion exchange. Besides the matrices freshwater and wastewater, in D19 saline waters, brines, boiler-feed waters, and process waters are considered. Several methods deal with waters from contaminated sites and oil-contaminated waters. Furthermore, scalings and deposits constitute an item. The 290 standards are under the jurisdiction of the committee D19; they are published as stand-alone methods as well as in the Annual Book of ASTM Standards, volumes 11.01 and 11.02. In the list of standards at the end of this chapter, standards authored by the ASTM SCs D19.02, D19.04, D19.05, D19.06, and D19.24 are included. ASTM water-testing standards are delivered in the format of standard test method, dedicated to special analytes, standards practice, or standard guide, the latter dealing with general methodological aspects.

3.11.4 Items Covered by Standardization in the Field of Water Examination Water-testing standards are needed for suitable parameters to cover the following purposes: 1. to check the quality of water for human consumption and use or for technical applications; 2. to check the efficiency of treatment processes (e.g., drinking water treatment and wastewater treatment); 3. to monitor and assess the quality of natural waters; and 4. to estimate emissions and loads, especially of wastewaters and effluents. For illustration purposes, Table 4 gives an overview of the parameters which have to be analyzed according to the European Drinking Water Directive (98/83/EC, 1998) and for monitoring of natural waters as a consequence of the WFD. Many of them require very sensitive methods. This is particularly important for the EQS given in the directive 2008/ 105/EC (2008) for priority substances in the field of water quality (PS) and priority hazardous substances in the field of water quality (PHS), the latter being of concern because they are toxic, persistent, and bioaccumulative. In addition, in the

table given are the guideline values set by the World Health Organization (WHO) for some water ingredients, since many of the drinking water parameters have been included into the directive because of human-toxicological concerns (for human-toxicological aspects of water ingredients, see also Chapter 3.14 Drinking Water Toxicology in Its Regulatory Framework. Corrosiveness to the main, and taste and odor are further aspects covered. Traditional parameters for surveillance of domestic or municipal wastewaters are suspended matter, BOD, COD, phosphorus, and organically and inorganically bound nitrogen. Analytes to be checked in industrial wastewaters may be manifold and depend on the respective industry branches, for example, metals such as Hg, Cd, Cr, Ni, Pb, and Cu. Prominent examples for industrial wastewater parameters are AOX and cyanide.

3.11.4.1 Terminology, Analytical Strategies, Validation, and Quality Control Definition of terms is a prerequisite to unambiguously specify technical data and procedures. For this, most TCs in standardization deal with the definition of terms to be used in their respective standards portfolio. The SC 1 of ISO/TC 147 has defined about 900 terms given in the series ISO 6107 on vocabulary, parts 1 through 9. At present, these definitions are being transposed into a concept database, which is accessible free of charge in a read-only format (see the ‘Relevant websites’ section for ISO/CDB). In CEN, terms used in wastewater treatment are collated in EN 1085. For ASTM, the standard terminology relating to water is given in the standard ASTM D1129, for fluvial sediment in ASTM D4410. Analyte-dedicated specified methods make up the majority of the standards portfolio on water examination. Besides, there are some papers on general or on partial aspects of analytical procedures, for example, guideline papers dealing with strategies for the design of monitoring programs (ISO 5667-20, ASTM D5851, ASTM D5612, ASTM D6146, and ASTM D6145), or proficiency of laboratories (ISO/IEC 17025, ISO 13528, and ISO/TS 20612). Analytical quality control is dealt with in ISO/ TS 13530, and quality assurance of biological and ecological assessments in EN 14996. A current standardization project in ISO/TC 147/SC 2 is dedicated to the determination of measurement uncertainty (intended ISO 11352). An important issue is the validation of methods, for example, design and evaluation of interlaboratory comparisons, which is specified by ISO 5725-2 or ASTM D2777. ENV ISO 13843 gives guidance on the validation of microbiological methods. Sometimes the necessity may arise to prove that a given method is equivalent to an already-existing standard, for example, if in regulatory frameworks application of equivalent methods as alternatives to a recommended official standard is permitted. ISO/TS 16489 is a guideline for establishing the equivalency of results. Criteria for establishing equivalence among microbiological methods are given in EN ISO 17994.

3.11.4.2 Sampling, Sample Pretreatment, and Basic Operations Sampling is the first step of analysis. The manifold aspects of sampling are covered in the standard ISO 5667, which now

Standardized Methods for Water-Quality Assessment Table 4

271

Parameters relevant for the assessment of drinking water and of natural waters, and their parametric values, given in European Directives

Acrylamide Aluminum Ammonium Antimony Arsenic Alachlor Anthracene Atrazine Benzene Boron Bromate Brominated diphenylether Cadmium

Carbon tetrachloride C10–13 chloroalkanes Chloride Chlorfenvinphos Chlorpyrifos (chlorpyrifos-ethyl) Chromium Copper Cyanide Cyclodiene pesticides: aldrin, dieldrin, endrin, isodrin

Inland surface waters

Drinking water

2008/105/EC (2008)

98/83/EC (1998)

WHO (2006)

Environmental quality standard, annual average (AA-EQS)

Parametric value

Guideline value

0.10 mg l1 (CP) 200 mg l1 (IP, CM) 0.50 mg l1 (IP, CM) 5.0 mg l1 (CP) 10 mg l1 (CP) As a pesticide

0.5 mg l1a

20 mg l1 10 mg l1 (P) 20 mg l1a

As a pesticide 1.0 mg l1 (CP) 1.0 mg l1 (CP) 10 mg l1 (CP)

2 mg l1 10 mg l1a 0.5 mg l1 (T) 10 mg l1a (A,T)

5.0 mg l1 (CP)

3 mg l1

0.3 mg l1 (PS) 0.1 mg l1 (PHS) 0.6 mg l1 (PS) 10 mg l1 (PS)

0.0005 mg l1 (PHS) (PHS) r0.08 mg l1 (hardness class 1) Up to 0.25 mg l1 (hardness class 5) 12 mg l1 0.4 mg l1 (PHS) 1

0.1 mg l (PS) 0.03 mg l1 (PS)

S ¼ 0.01 mg l1

DDT total 1,2-Dichloroethane Dichloromethane Di(2-ethylhexyl)-phthalate (DEHP) Diuron Endosulfan Epichlorohydrin Fluoranthene Fluoride Hexachlorobenzene Hexachlorobutadiene Hexachlorocyclohexane Iron Isoproturon Lead Manganese Mercury Naphthalene Nickel Nitrate Nitrite

0.025 mg l1 10 mg l1 (PS) 20 mg l1 (PS) 1.3 mg l1 (PS) 0.2 mg l1 (PS) 0.005 mg l1 (PHS)

Nonylphenol (4-nonylphenol) Octylphenol (4-(1,10 ,3,30 -tetramethylbutyl)phenol) Pentachlorobenzene Pentachlorophenol

0.3 mg l1 (PHS) 0.1 mg l1 (PS)

4 mg l1 250 mg l1 (IP) As a pesticide As a pesticide 50 mg l1 (CP) 2.0 mg l1 (CP) 50 mg l1 (CP) Aldrin 0.03 mg l1 Dieldrin 0.03 mg l1

3.0 mg l1 (CP)

30 mg l1 50 mg l1 (P) 2 mg l1 70 mg l1 Aldrin þ dieldrin 0.03 mg l1

1 mg l1 30 mg l1a 20 mg l1 8 mg l1

As a pesticide As a pesticide 0.10 mg l1 (CP)

0.4 mg l1 (P)

1.5 mg l1 (CP)

1.5 mg l1

0.1 mg l1 (PS) 0.01 mg l1 (PHS) 0.1 mg l1 (9) (PHS) 0.02 mg l1 (PHS) 0.3 mg l1 (PS) 7.2 mg l1 (PS) 0.05 mg l1 (9) (PHS) 2.4 mg l1 (PS) 20 mg l1 (PS)

0.007 mg l1 (PHS) 0.4 mg l1 (PS)

0.6 mg l1 200 mg l1 (IP, CM) As a pesticide 10 mg l1 (CP) 50 mg l1 (IP) 1.0 mg l1 (CP) 20 mg l1 (CP) 50 mg l1 (CP) 0.50 mg l1 (CP, CM)

9 mg l1 10 mg l1 400 mg l1 (C) 6 mg l1 70 mg l1 50 mg l1 3 mg l1b 0.2 mg l1c

9 mg l1a (P) (Continued )

272

Standardized Methods for Water-Quality Assessment

Table 4

Continued

Pesticides – individual Pesticides – total Polycyclic aromatic hydrocarbons Benzo(a)pyrene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(g,h,i)perylene Indeno(1,2,3-cd)pyrene Selenium Simazine Sodium Sulfate Tetrachloroethylene Trichloroethylene Tributyltin compounds (tributyltin-cation) Trichlorobenzenes Trichloromethane Trifluralin Trihalomethanes-total Vinyl chloride Tritium Total indicative dose Hydrogen ion concentration Oxidizability Total organic carbon (TOC) Conductivity Turbidity Color Odor Taste Escherichia coli Enterococci Clostridium perfringens (including spores) Coliform bacteria Pseudomonas aeruginosa Colony count 22 1C Colony count 37 1C a

Inland surface waters

Drinking water

2008/105/EC (2008)

98/83/EC (1998)

WHO (2006)

Environmental quality standard, annual average (AA-EQS)

Parametric value

Guideline value

0.10 mg l1 (CP) 0.50 mg l1 (CP) 0.10 mg l1 (CP) 0.010 mg l1 (CP)

0.7 mg l1a

(PHS) 0.05 mg l1 S ¼ 0.03 mg l1 S ¼ 0.002 mg l1

1 mg l1 (PS)

10 mg l1 10 mg l1 0.0002 mg l1 (PHS) 0.4 mg l1 (PS) 2.5 mg l1 (PS) 0.03 mg l1 (PS)

10 mg l1 (CP) As a pesticide 200 mg l1 (IP) 250 mg l1 (IP) 10 mg l1 (CP)f

As a pesticide 100 mg l1 (CP) 0.50 mg l1 (CP) 100 Bq l1 (IP) 0.10 mSv a1 (IP) Z6.5 and r9.5 pH units (IP, CM) 5.0 mg l1 O2 No abnormal change (IP) 2500 mS cm1 at 20 1C (IP, CM) Acceptable to consumers and no abnormal change (IP, CM)g 0/100 ml (MP, CM) 0/250 mle(MP) 0/100 ml (MP) 0/250 mle(MP) 0/100 ml (IP, CM) 0/100 ml (IP, CM) 0/250 mle (IP, CM) 0/250 mle(MP, CM) No abnormal change (IP) 100/mle (MP, CM) 20/mle (MP, CM)

10 mg l1 2 mg l1

40 mg l1 20 mg l1 (P)

300 mg l1 20 mg l1 d

0.3 mg l1a (C) 10 000 Bq l1 0.10 mSv a1

0/100 ml

0/100 ml

Considered to be carcinogenic, value calculated for an excess lifetime cancer risk of 105. Short-term exposure. c Long-term exposure. d The guideline values for dichloromethane (20 mg l1), bromodichloromethane (60 mg l1), tribromomethane (100 mg l1), trichloromethane (300 mg l1), dibromochloromethane (100 mg l1) should be complied with individually. e For water in bottles or containers. f Sum of tetrachloroethylene and trichloroethylene. g In case of surface water treatment, 1.0 NTU turbidity in the water extreatment works should be strived for. A, provisional guideline, because the calculated level is below the achievable quantification level. C, concentration at or below the health-based guideline may affect appearance, taste, or odor; CM, subject to check monitoring, performed more frequently than the audit monitoring which comprises all drinking water parameters; CP, chemical parameter; IP, indicator parameter; MP, microbiological parameter, P, provisional guideline, because available information on health effects is limited; PHS, priority hazardous substance in the field of water quality; PS, priority substance in the field of water quality. b

Standardized Methods for Water-Quality Assessment

comprises about 20 parts, some of them adopted as European standards. EN ISO 5667-1 gives a general guidance for the design of sampling programs and sampling techniques. In EN ISO 5667-3, conservation techniques for various analytes are outlined. These have to be applied if no specifications for conservation are given in the respective individual analytical standard. ISO 5667-14 gives a modus operandi of how to systematically take field blank samples, spiked samples, and replicate samples in order to control sampling errors, contamination, and variability of sampling at the successive steps from the sampling site to analysis in the laboratory. Sampling techniques are specified for sampling from various aquatic systems, that is, lakes (ISO 5667-4), rivers and streams (ISO 5667-6), groundwaters (ISO 5667-11 and ISO 5667-18), wet deposition (ISO 5667-8), and marine waters (ISO 5667-9). Specifications are also given for sampling of wastewaters (ISO 5667-10), drinking waters from waterworks and piped distribution systems (ISO 5667-5), and waters and steams in boiler plants (ISO 5667-7). In addition, sampling methods for suspended matter (ISO 5667-17), sediments (ISO 5667-12, EN ISO 5667-19), and sludges (ISO 5667-13 and ISO 5667-15) are available. Special requirements have to be met for sampling prior to microbiological analysis (EN ISO 19458) or biotesting (EN ISO 5667-16). The sampling of living beings of different trophic levels is a crucial issue in biological– ecological assessment of natural waters (see Section 3.11.4.8). At present, standardization projects are underway in ISO/ TC 147/SC 6 which deal with the sampling of drinking water distributed by tankers or means other than distribution pipes (intended ISO 5667-21), design and installation of groundwater sampling points (intended ISO 5667-22), and with the use of passive samplers in surface waters (intended ISO 566723). The issue of sampling and passive sampling in particular is treated in more detail in 00054. Further guidance documents describe standard practices such as digestion (ASTM D1971, ASTM D4309 – microwave heating, EN ISO 15587-1 – aqua regia digestion, EN ISO 15587-2 – nitric acid digestion) or spiking into aqueous samples (ASTM D5788; ASTM D5810). In ISO 3696 and ASTM D1193, specifications are given concerning the quality of reagent water for different purposes. In the ASTM portfolio, standard guides for ultra-pure water used in the electronic and semiconductor industries (ASTM D5127) and for bioapplications grade water (ASTM D5196) are available, furthermore standard practices for the preparation of substitute ocean water (ASTM D1141) or substitute wastewater (ASTM D5905). An essential step in the quantitative analysis is calibration. The statistical evaluation of the linear calibration function and a calibration strategy for nonlinear second-order calibration functions is dealt with in ISO 8466-1 and ISO 8466-2, respectively.

3.11.4.3 Physical–Chemical and Other Basic Parameters for Water-Quality Assessment The oldest parameters to assess primary acceptance of drinking water are the so-called sensoric parameters such as taste and odor. Organoleptic methods, which try to classify odor (ASTM D1292) and to roughly quantify it by threshold numbers (EN 1622), are available.

273

From very early on testing methods were developed for raw water and drinking water to ensure the efficiency of drinking water treatment. Turbidity (EN ISO 7027, ASTM D6698, ASTM D6855, and ASTM D7315) can be read as a surrogate parameter for possible presence of microorganisms. Measurement of temperature, pH value (ISO 10523, ASTM D1293, ASTM D5128, and ASTM D5464) and determination of alkalinity (EN ISO 9963-1, EN ISO 9963-2, and ASTM D1067) and CO2 (ASTM D513) serve to prove the stability of waters concerning carbonate balance. Chlorine concentration (EN ISO 7393-1, -2, -3, ASTM D1253) is checked to make sure sufficient disinfection of drinking water or pool water. Color can indicate natural organic matter, for example, humic substances, or metal compounds (Fe and Mn). The parameter is relevant for treatment aspects and in limnology as well. The standard EN ISO 7887 specifies several methods for measurement of color, comprising visual examination and comparison with color-matching solutions (hexachloroplatinate scale) and a photometric method, specified for l ¼ 436 nm, l ¼ 525 nm, and l ¼ 620 nm. The standard is being revised at present, in addition, to specify 410 nm as a suitable wavelength as well (see prEN ISO 7887:2009), since this wavelength is often used in limnological studies (Hongve and A˚kesson, 1996). Electrical conductivity (EN 27888) is a measure for electrolyte content. According to 98/83/EC (1998), it should not exceed 2500 mS cm1 (20 1C) in drinking waters for corrosiveness concerns. Oxygen is crucial wherever respiration has to be maintained ranging from aquatic wildlife to microorganisms. Therefore, oxygen concentration or saturation is a key parameter for the status of surface waters as well as an important variable in the biological treatment of wastewaters. For routine oxygen measurement, the application of electrochemical (EN 25814) and luminescent (ASTM D888) probes is specified. In Table 5, further fundamental parameters in water analysis are listed. The table reflects that saprobization and eutrification are major issues in water monitoring. Nitrogen species listed in the table are fecal indicators and document domestic wastewater influence. Furthermore, parameters to estimate some major pollutants from industrial effluents are included. Organic load is relevant in natural waters with respect to oxygen balance, and in drinking waters, because organic matter might support bacterial growth. Moreover, organic water constituents can react to form hazardous disinfection by-products during disinfection of drinking water or pool water. Therefore, there were efforts very early on to analytically capture the phenomenon of oxygen consumption as well as the quantity of overall content of organic matter in water. This has resulted in the sum parameters DOC and total organic carbon (TOC), which quantify and specify organic carbon (OC), and furthermore the COD and permanganate index, which are indicative of oxygen consumption or easily oxidizable substances (see also Chapter 3.01 Sum Parameters: Potential and Limitations). The parameter BODn (biochemical oxygen demand after n days) was designed to approach more natural conditions, as a protocol for quantifying oxygen consumption after inoculation with aerobic microorganisms after 5 or 7 days. As halogenated organic substances

274 Table 5

Standardized Methods for Water-Quality Assessment Standards that specify important conventional parameters for water examination and assessment

Total organic carbon (TOC), dissolved organic carbon (DOC) ISO 8245, EN 1484 (OC 0.3–1000 mg l1, oxidation by combustion, by addition of an oxidant, by UV radiation or other high-energy radiation; determination of CO2 formed directly or after reduction to methane by IR, titration, thermal conductivity, conductimetry, coulometry, CO2-sensitive sensors, or FID (methane)); ASTM D4129 (high-temperature oxidation and coulometric detection, 2–20 000 mg l1); ASTM D4839 (ultraviolet or persulfate oxidation and infrared detection, 0.1–4000 mg l1); ASTM D5904 (membrane conductivity detection, ultraviolet-persulfate oxidation, CO2 selective membrane, 0.5–30 mg l1); ASTM D7573 (high-temperature catalytic combustion and infrared detection) Chemical oxygen demand (COD) ISO 6060 (dichromate method, COD 30–700 mg l1) ISO 15705 (small-scale sealed tube method) ASTM D1252 (dichromate oxygen demand, macro-COD by reflux digestion and titration, micro-COD by sealed digestion and spectrometry, up to 800 mg l1) Biochemical oxygen demand after n days (BODn) ISO 5815-1 (dilution and inoculation procedure after addition of allylthiourea) ISO 5815-2 (undiluted sample) EN 1899-1 (allylthiourea addition) EN 1899-2 (undiluted samples) Permanganate index EN ISO 8467 (chloride concentration smaller than 300 mg l1) Adsorbable organic halogens (AOX) EN ISO 9562 Bismuth-active substances (cationic surfactants) ISO 7852-2 (surfactants using Dragendorff reagent) Methylene blue active substances (MBAS) index (anionic surfactants) ISO 7875-1, EN 903, ISO 16265 (CFA), ASTM D2330 (0.03–1.5 mg l1) Phenol index ISO 6439 (aminoantipyrine, spectrometric after distillation) EN ISO 14402 (FIA and CFA, 0.01–1 mg l1) ASTM D1783 (color reaction with 4-aminoantipyrine, chloroform extraction: 0–100 mg l1, direct photometric 40.1 mg l1) Oil and grease, hydrocarbon oil ISO 9377-2 (hydrocarbon oil index, solvent extraction and GC) ASTM D4281 (oil and grease, gravimetric determination, liquid–liquid extraction, soxhlet extraction) Nitrogen Nitrogen: EN ISO 11905-1 (oxidative digestion with peroxodisulfate, up to 5 mg l1) Kjeldahl-Nitrogen: EN 25663 (corresp. ISO 5663, after mineralization with selenium); ASTM D3590 (manual digestion/distillation, semiautomated colorimetric Bertholt) Chemically bound nitrogen (TNb): ISO/TR 11905-2 (chemiluminescence detection after combustion and oxidation); EN 12260; ASTM D5176 (pyrolysis and chemiluminescence detection, 0.5–1000 mg l1) Soluble silicates EN ISO 16264 (FIA and CFA) Cyanides Total cyanide: ISO 6703-1; EN ISO 14403 (CFA); ASTM D4374; ASTM D7284 (FIA); ASTM D7511(CFA, in-line UV digestion) Easily releasable cyanide: ISO 6703-2, ASTM D4374 Free cyanide: EN ISO 14403 (CFA), ASTM D4282 (microdiffusion) Cyanogen Chloride ISO 6703-3 (0.02–15 mg l1); ASTM D4165 Alkalinity Total and composite alkalinity: EN ISO 9963-1 Total alkalinity in seawater: ISO 22719 Carbonate alkalinity: EN ISO 9963-2 Sulfides Dissolved sulfide: ISO 10530 (photometry using methylene blue) Easily released sulfide: ISO 13358 (0.04–1.5 mg l1, stripping with nitrogen at pH 4, subsequent color reaction to methylene blue) Phosphorus Phosphorus (ammonium molybdate method): EN ISO 6878 Orthophosphate and total phosphorus: EN ISO 15681 Total phosphorus: EN ISO 11885 (ICP-OES, 40.1 mg l1); EN ISO 17294-2 (ICP-MS, 45 mg l1) Suspended solids EN 872; ISO 11923 (both separation with a glass-fiber filter)

Standardized Methods for Water-Quality Assessment

are of special concern, the sum parameter AOX is still as a key parameter for survey and taxation of effluents (Pluta and Rosenberg, 2005). Further sum parameters were created to estimate the content of surfactants and phenols, the latter being of concern because of odor problems. A class of substances determined in bulk in wastewaters to monitor the efficiency of grease separators includes lipophilic hydrocarbons (oil and grease), one of the very few parameters for which a gravimetric method is still an active standard (ASTM D4281). In ISO, a gravimetric method on low-volatility lipophilic substances, avoiding fluoro-chloro-hydrocarbons by using petroleum ether or n-hexane for extraction, is in preparation (intended ISO 11394). Besides sum parameters, Table 5 contains further analytical parameters that are operationally defined, as they imply that specified steps for separation, digestion, or release have to be observed. This holds for the parameters concerning cyanide, nitrogen, and sulfide. Specification of a species as dissolved is based on the convention of 0.45-mm membrane filtration, this operation accounting for the distinction between DOC and TOC or true and apparent color. The digestion procedure and its performance specify the total element content determined by ICP-OES (HNO3 in EN ISO 11885, or HNO3/HCl in ASTM D1976). Alkalinity and hardness are convention-based parameters as well. Some of the operationally defined parameters in Table 5 are based on laborious manual procedures. As they have to be analyzed in analytical laboratories for a large number of samples, miniaturization and automation were highly attractive. The BOD is the only example in the portfolio of ISO/ TC 147 that a small-scale sealed tube (ST) version has been standardized (ISO 15705). It is pointed out in this standard that the results might differ from those of the full-scale version (ISO 6060), and results of an interlaboratory trial comparing both methods for different matrices are given. Furthermore, flow analysis (flow injection analysis (FIA) or continuous flow analysis (CFA)) allows for a miniaturization and automation of these methods, since it has become technically feasible to integrate operations such as liquid–gas separation, digestion, or distillation into flow systems. Standardized flow analysis versions are available for several parameters given in Table 5 for example, for methylene blue active substances (ISO 16265), phenol index (EN ISO 14402), and total and free cyanide (EN ISO 14403). For flow-analysis determination of total nitrogen, in-line UV digestion is an element in the intended standard ISO 29441 (Kroon, 1993). If an operationally defined manual parameter is transposed into a miniaturized flow-analysis format, it is crucial that specified recovery checks have to be met. For determination of total or free cyanide, this is, for instance, the recovery of hexacyanoferrate(III) to be X90% or p5% , respectively, and the recovery of thiocyanate to be o1% in each case.

3.11.4.4 Methods for Determination of Individual Water Constituents and Defined Groups of Substances Concerning the determination of individual analytes in waters, a division can roughly be made between major components and micropollutants, and furthermore between inorganic and organic constituents.

275

3.11.4.4.1 Inorganic water constituents For a large number of individual inorganic water constituents, photometric or colorimetric manual single-parameter methods are available (see Tables 6 and 7). For chloride, calcium, and magnesium, titrimetric methods are still included in the standards collections. Manual single-parameter methods, though they are sensitive and reliable, require much expenditure of manpower, time, and chemicals. Standardized FIA and/or CFA versions are available now for determination of the prominent analytes ammonium (EN ISO 11732), nitrite and nitrate (EN ISO 13395), cyanide (EN ISO 14403), phosphate (EN ISO 15681), chloride (EN ISO 15682), silicate (EN ISO 16264), chromate (EN ISO 23913), and sulfate (ISO 22743). An alternative way to miniaturize and automatize manual single-parameter methods is the so-called discrete analyzers offered as analytical instruments. An initiative to deal with these instruments with respect to possible standardization has been started in ISO/TC 147/SC 2 (ISO new project 15923). The application of electrochemical sensors (ion-selective electrodes, see also Chapter 3.10 Online Monitoring Sensors) has been specified for selected analytes. The most prominent example is fluoride (ISO 10359-1, ASTM D1179). In the ASTM portfolio, ion-selective electrodes are considered for detection of bromide (ASTM D1246), chloride (ASTM D512), sulfide (ASTM D4658), and ammonia (ASTM D1426). As an automated multi-analyte method, ion chromatography (IC) has become widely used and is the method of choice for analysis of anions in many laboratories. Standards on IC are available for drinking waters and wastewaters (e.g., EN ISO 10304, see also Table 6). A current project in ISO deals with the ion-chromatographic determination of bromate after post-column reaction as a very sensitive method for an application range from 0.5 mg l1 upward (intended ISO 11206). Standard protocols are also available for ion-chromatographic determination of some cations (e.g., EN ISO 14911 and ASTM D6919). For a decade, capillary electrophoresis has been in discussion as a further multi-analyte flow method for routine applications (Kaniansky et al., 1999). The standard ASTM D6508 gives a specification for determination of several inorganic anions. Tailor-made classical single-analyte methods are available in the standards collections also for species of metal or metalloids, often based on reaction of metal cations with complexing agents in order to form a colored complex which could be photometrically determined (Table 7). Today, however, these substances are determined in most cases by methods of elemental analysis, unless speciation is required for special purposes (see Chapter 3.02 Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species). For instance, this applies for chromium, which is much more toxic in the oxidation state þ VI than in þ III; therefore, the species chromate has to be analyzed separately. In routine analysis, there is a trend toward elemental analytical methods based on plasma techniques (i.e., plasma optical emission spectroscopy or plasma mass spectrometry) as fast multi-element methods. In the standards EN ISO 11885 and ASTM D1976, specifications are given for multi-element determination by inductively coupled plasma atomic optical emission spectroscopy (ICP-OES),

276 Table 6

Standardized Methods for Water-Quality Assessment Standardized methods for determination of molecular ions of nonmetals

Ammonium ISO 5664 (distillation and titration,o10 mg per test portion); ISO 6778 (potentiometric,o50 mg l1); ISO 7150-1 (manual spectrometric method); EN ISO 14911 (IC, 0.1–10 mg l1) Ammonia nitrogen EN ISO 11732 (CFA and FIA, spectrometric detection, 0.1–10 mg l1); ASTM D1426 (ion-selective electrode, 0.5–1000 mg l1) Borate ISO 9390 (spectrometric, azomethine-H, 0.01–1 mg l1 B); ASTM D3082 (curcumin colorimetric-extraction method, 0.1–1.0 mg l1) Bromate EN ISO 15061 (IC, 0.5–1000 mg l1, after suitable sample pretreatment); ASTM D6581 (IC, 5–30 mg l1) Bromide EN ISO 10304-1 (IC, 0.05–20 mg l1); ASTM D1246 (ion-selective electrode, 0.5–1000 mg l1); ASTM D3869; ASTM D4327 (IC, 0.63–21.0 mg l1); ASTM D6581 (IC, 20–200 mg l1); ASTM D6508 (CIE, 1–50 mg l1) Chlorate EN ISO 10304-4 (IC, 0.03–10 mg l1); ASTM D6581 (IC, 20–500 mg l1) Chloride ISO 9297 (titrimetric with AgNO3, Mohr); EN ISO 10304-1 (IC, 0.1–50 mg l1); EN ISO 10304-4 (IC, 0.1–50 mg l1); EN ISO 15682 (flow analysis, photometric, and potentiometric detection, 1–1000 mg l1); ASTM D512 (mercurimetric titration, silver nitrate titration, and ion-selective electrode method); ASTM D4327 (IC, 0.78–26.0 mg l1); ASTM D4458; ASTM D6508 (CIE, 1–50 mg l1) Chlorite EN ISO 10304-4 (IC, 0.05–10 mg l1); ASTM D6581 (IC, 20–500 mg l1) Cyanide ISO 6703; EN ISO 14403 (CFA, 10–100 mg l1); ASTM D2036; ASTM D4282 Fluoride ISO 10359-1 (electro-chemical sensor, weakly contaminated water); ISO 10359-2 (digestion and distillation); EN ISO 10304-1 (IC, 0.01 –10 mg l1); ASTM D1179 (distillation: 0.1–2.6 mg l1, ion-selective electrode: 1–1000 mg l1); ASTM D3868; ASTM D4327 (IC, 0.26–8.49 mg l1); ASTM D6508 (CIE, 1–25 mg l1) Iodide EN ISO 10304-3 (IC, 0.1–50 mg l1); ASTM D3869 Nitrate ISO 7890-3 (photometric, sulfosalycilic acid); EN ISO 10304-1 (IC, 0.1–50 mg l1); EN ISO 13395 (flow analysis, 0.2–20 mg l1); ASTM D3867; ASTM D6508 (CIE, 1–50 mg l1) Nitrite ISO 6777 (spectrometric); EN ISO 10304-1 (IC, 0.05–20 mg l1); EN ISO 13395 (Flow analysis, 0.01–1.0 mg l1), ASTM D3867; ASTM D6508 (CIE, 1– 50 mg l1) Orthophosphate EN ISO 6878 (photometric, ammonium molybdate); EN ISO 10304-1 (IC, 0.1–20 mg l1); EN ISO 15681-1 (FIA, 0.01–1.0 mg l1); EN ISO 15681-2 (CFA, 0.01–1.00 mg l1); ASTM D4327 (IC, 0.69–23.1 mg l1); ASTM D6508 (CIE, 1–50 mg l1) Silicate EN ISO 16264 (FIA, CFA), ASTM D859 (colorimetric, 815 nm: 20–1000 mg l1, 640 nm: 0.1–5 mg l1) Sulfate ISO 22743 (CFA); EN ISO 10304-1 (IC, 0.1–100 mg l1); ASTM D516 (turbidimetric, 5–40 mg l1); ASTM D4130; ASTM D4327 (IC, 2.85–95.0 mg l1); ASTM D6508 (CIE, 1–50 mg l1) Sulfide ISO 10530 (in solution, photometrically with methylene blue); ASTM D4658 (ion-selective electrode, 0.04–4000 mg l1) Sulfite EN ISO 10304-3 (IC, 0.1–50 mg l1) Thiocyanate EN ISO 10304-3 (IC, 0.1–50 mg l1); ASTM D4193 (colorimetric, 0.1–2.0 mg l1) Thiosulfate EN ISO 10304-3 (IC, 0.1–50 mg l1)

Standardized Methods for Water-Quality Assessment

277

Table 7 Standardized methods for determination of metals/metalloids and their ions. Methods of elemental analysis based on plasma techniques are given in Table 8 Aluminum ISO 10566 (photometry, pyrocatechol, optical pathlength of 50 mm: up to 100 mg l1; optical pathlength of 10 mm: up to 500 mg l1); EN ISO 12020 (FAAS, 5–100 mg l1; GF-AAS, 10–100 mg l1); EN ISO 15586 (GF-AAS, lowest determinable concentration 1 mg l1, optimal parameters 6–60 mg l1); ASTM D857 (F-AAS, 0.5–5 mg l1) Antimony EN ISO 15586 (GF-AAS, lowest determinable concentration 1 mg l1, optimal parameters 10–100 mg l1); ASTM D3697 (HG-AAS, 1–15 mg l1) Arsenic EN ISO 15586 (GF-AAS, lowest determinable concentration 1 mg l1, optimal parameters 10–100 mg l1); EN 26595 (photometry, silver diethyldithiocarbamate, 0.001–0.1 mg l1); ASTM D2972 (photometry, silver diethyldithiocarbamate, 5–250 mg l1; GF-AAS, 5–100 mg l1; HG-AAS, 1–20 mg l1) Barium EN ISO 14911 (IC, 1–100 mg l1), ASTM D3651 (F-AAS, in saline waters, 1–5 mg l1); ASTM D4382 (GF-AAS) Beryllium ASTM D3645 (F-AAS, 10–500 mg l1; GF-AAS, 10–50 mg l1) Cadmium ISO 8288 (F-AAS, 1–50 mg l1); EN ISO 5961 (AAS, air-ethine flame: 0.05–1 mg l1; after electrothermic atomization: 0.3–3 mg l1); EN ISO 15586 (GFAAS, lowest determinable concentration 0.1 mg l1, optimal parameters 0.4–4 mg l1); ASTM D3557 (AAS, differential pulse anodic stripping voltammetry) Calcium ISO 6058 (complexometry, 2–100 mg l1); EN ISO 7980 (AAS, up to 50 mg l1 Ca and 5 mg l1 Mg); EN ISO 14911 (IC, 0.5–50 mg l1); ASTM D511 (complexometry, 1–1000 mg l1); ASTM D1126 (calcium hardness, EDTA complexometric (titrimetry), indicator hydroxynaphthol blue); ASTM D6919 (IC, 4.0–40.0 mg l1) Chromium ISO 9174 (AAS); EN 1223 (AAS); EN ISO 15586 (GF-AAS, lowest determinable concentration 0.5 mg l1, optimal parameters 2–20 mg l1) Chromium(VI) (chromate) ISO 11083 (photometry, 1,5-diphenylcarbazide); EN ISO 18412 (photometry, weakly contaminated water); EN ISO 23913 (FIA and CFA); EN ISO 10304-3 (IC); ASTM D1687; ASTM D5257 (IC, 1–1000 mg l1) Cobalt ISO 8288 (F-AAS, 3 methods); EN ISO 15586 (GF-AAS, lowest determinable concentration 1 mg l1, optimal parameters 6–60 mg l1); ASTM D3558 (F-AAS 0.1–10 mg l1; F-AAS, chelation–extraction 10–1000 mg l1; GF-AAS, 5–100 mg l1) Copper ISO 8288 (F-AAS, 3 methods); EN ISO 15586 (GF-AAS, lowest determinable concentration 0.5 mg l1, optimal parameters 3–30 mg l1); ASTM D1688 (AAS, chelation–extraction, 50–500 mg l1; F-AAS, direct 0.05–5 mg l1) Iron ISO 6332 (photometry, phenanthroline); EN ISO 15586 (GF-AAS, lowest determinable concentration 1 mg l1, optimal parameters 3–30 mg l1); ASTM D1068 (photometry bathophenanthroline, 40–1000 mg l1; F-AAS, 0.1–5.0 mg l1; GF-AAS, 5–100 mg l1) Lead ISO 8288 (F-AAS, 3 methods, 5–200 mg l1); EN ISO 15586 (GF-AAS, lowest determinable concentration 1 mg l1, optimal parameters 10–100 mg l1); ASTM D3559 (F-AAS, direct, 1–10 mg l1; F-AAS, chelation–extraction 100–1000 mg l1; differential pulse anodic stripping voltammetry, 1– 100 mg l1; GF-AAS 5–100 mg l1) Lithium EN ISO 14911 (IC, 0.01–1 mg l1); ASTM D3561 (F-AAS, in saltwater, 0.1–70 000 mg l1); ASTM D6919 (IC, 0.4–10.0 mg l1) Magnesium EN ISO 7980 (AAS, up to 50 mg l1 Ca and 5 mg l1 Mg); EN ISO 14911 (IC, 0.5–50 mg l1); ASTM D511 (AAS); ASTM D6919 (IC, 2.4–20.0 mg l1) Calcium þ magnesium ISO 6059 (EDTA, complexometry); ASTM D511 (complexometry, EDTA, Ca: 1.0–15 mg l1, Mg: 0.25–3.5 mg l1); ASTM D1126 (with hardness indicator chrome black T3) Manganese ISO 6333 (photometry, formaldoxime spectrometry, 0.01–5 mg l1); EN ISO 14911 (IC, 0.5–50 mg l1); EN ISO 15586 (GF-AAS, lowest determinable concentration 0.5 mg l1, optimal parameters 1.5–15 mg l1); ASTM D858 (F-AAS, direct 0.1–5 mg l1; F-AAS, chelation–extraction, 10–500 mg l1; GF-AAS, 5–500 mg l1) Mercury ISO 5666 (AAS cold vapor 0.1–10 mg l1; optional digestion by KMnO4/K2S2O8; reduction by tin(II) chloride or by NaBH4); ISO 16590 (AAS cold vapor 0.1–1 mg l1 with enrichment by amalgamation; reduction by tin(II) chloride or by NaBH4 after digestion by KMnO4/K2S2O8); EN ISO 17852 (atomic fluorescence, 10 ng l1 to 10 mg l1); EN 1483 (AAS cold vapor, 0.1–10 mg l1); EN 12338 (after enrichment by amalgamation 0.01–1 mg l1); ASTM D3223-02 (AAS cold vapor, 0.5–10 mg l1, stabilization with HNO3, oxidation with KMnO4/K2S2O8, reduction with SnSO4) (Continued )

278

Standardized Methods for Water-Quality Assessment

Molybdenum EN ISO 15586 (GF-AAS, lowest determinable concentration 1 mg l1, optimal parameters 6–60 mg l1); ASTM D3372 (AAS) Nickel ISO 8288 (F-AAS, 3 methods, 0.2–2 mg l1); EN ISO 15586 (GF-AAS, lowest determinable concentration 1 mg l1, optimal parameters 7–70 mg l1); ASTM D1886 (F-AAS direct 0.1–10 mg l1; F-AAS, chelation–extraction, 10–1000 mg l1, GF-AAS, 5–100 mg l1) Potassium ISO 9964-2 (F-AAS, air-ethine flame, 1–10 mg l1); ISO 9964-3 (F-OES, 0.1–10 mg l1); EN ISO 14911 (IC, 0.1–10 mg l1); ASTM D3561 (F-AAS, in saltwater); ASTM D4191 (F-AAS, 0.20–4.0 mg l1); ASTM D6919 (IC, 1.2–20.0 mg l1) Selenium ISO 9965 (HG-AAS); EN ISO 15586 (GF-AAS, lowest determinable concentration 2 mg l1, optimal parameters 15–150 mg l1); ASTM D3859 (GF-AAS, 2–100 mg l1; HG-AAS, 1–20 mg l1) Silver EN ISO 15586 (GF-AAS, lowest determinable concentration 0.2 mg l1, optimal parameters 1–10 mg l1); ASTM D3866 (AAS, chelation–extraction; GFAAS) Sodium ISO 9964-1 (F-AAS,); ISO 9964-3 (F-OES, 0.1–10 mg l1); EN ISO 14911 (IC, 0.1–10 mg l1); ASTM D3561 (F-AAS in saltwater); ASTM D4191 (F-AAS, 0.20–3.0 mg l1); ASTM D6919 (IC, 4.0–40.0 mg l1) Strontium EN ISO 14911 (IC, 0.5–50 mg l1); ASTM D3352 (AAS in saltwater, 5–2000 mg l1); ASTM D3920 (AAS, 0.1–1 mg l1) Thallium EN ISO 15586 (GF-AAS, lowest determinable concentration 2 mg l1, optimal parameters 20–200 mg l1) Vanadium EN ISO 15586 (GF-AAS, lowest determinable concentration 1 mg l1, optimal parameters 6–60 mg l1); ASTM D3373 (GF-AAS, 10–200 mg l1) Zinc ISO 8288 (F-AAS, 3 methods, 0.01–2 mg l1); EN ISO 15586 (GF-AAS, lowest determinable concentration 0.5 mg l1, optimal parameters 0.5–5 mg l1); ASTM D1691 (F-AAS direct, 0.05–2 mg l1; chelation–extraction 20–200 mg l1)

which comprise recommended emission wavelengths for observation, information on spectral interferences, examples for mixed calibration standards, and specification of the acid digestion procedure to precede determination of total element content. For the elements covered in the standards, estimated detection limits given therein for prominent wavelengths are listed in Table 8. ASTM has standardized an optical emission spectroscopy (OES) method using also direct-current plasma (DCP) (ASTM D4190). Standardized protocols for inductively coupled mass spectrometry (ICP-MS) are given for a large number of elements in EN ISO 17294 and in ASTM D5673 (see Table 8), which specify, for example, recommended analytical masses, and inform about isobaric and molecular ion interferences. An advantage of plasma OES is its large linear working range of about five decades. For many elements, ICP-MS is the most sensitive method, with a limit of quantification below 0.1 mg l1, but for refractory elements or elements with high affinity to oxygen, ICP-OES might work better (e.g., B, P, Si, and alkaline earth metals). As it is evident from Table 8, methods of plasma elemental analysis are also applicable to some nonmetals. Further standardized methods based on elemental analysis are included in Table 7. Flame OES (F-OES) is still relevant for the determination of sodium and potassium. For many elements, individual standards have been developed for determination by flame atomic absorption spectrometry (FAAS), in many cases with an optional chelation–extraction step for enrichment to increase sensitivity. In addition,

specifications for the more recent sensitive AAS version using graphite furnace atomization (GF-AAS) are given. In the standards EN ISO 15586 and ASTM D3919, procedural details concerning GF-AAS are collated for a larger group of trace metals. For the determination of mercury in low concentrations by AAS cold-vapor technique, several standardized methods are available, which differ in specifications concerning reagents for stabilization (HNO3 or K2Cr2O7), predigestion (KMnO4/ K2S2O8 or KBr/KBrO4), elimination of excessive oxidizing reagent, and reduction of mercury (SnCl2 or NaBH4) (see Table 7). In ISO 16590 and in EN 12338, an additional enrichment step by amalgamation is specified. In a current ISO standardization project of ISO/TC 147/SC 2, efforts are made to replace these standards by a single combined revised standard (intended ISO 12846) with harmonized procedures and after exclusion of some options. Besides AAS, atomic fluorescence spectroscopy (AFS) is applied for detection of mercury (EN ISO 17852). AFS is also intended for detection of arsenic, selenium, and antimony in current standardization projects as an option besides AAS (intended ISO 17378, ISO 17379, and ISO 23914). Voltammetry as a very sensitive method for determination of heavy metals is specified in ASTM standards for Cd (ASTM D3557) and Pb (ASTM D3559), furthermore in the German standard DIN 38406-16 for Zn, Cd, Pb, Cu, Tl, Ni, and Co, and in DIN 38 406-17 for uranium, which is of concern because of its toxicity to the kidneys and therefore an item in the analysis of drinking water and mineral water.

Standardized Methods for Water-Quality Assessment

279

Table 8 Standardized multielement methods based on plasma techniques, elements covered therein, and data on sensitivity or application range given in the standards Element

ICP-MS EN ISO 17294-2 a (mg l 1)

ICP-MS ASTM D5673 b (mg l 1)

ICP-OES EN ISO 11885 c (mg l 1)

ICP-OES ASTM D1976 d (mg l 1)

DCP-OES ASTM D4190 e (mg l 1)

Ag Al As Au B Ba Be Bi Ca Cd Ce Co Cr Cs Cu Dy Er Eu Fe Ga Gd Ge Hf Hg Ho In Ir K La Li Lu Mg Mn Mo Na Nd Ni P Pb Pd Pr Pt Rb Re Rh Ru S Sb Sc Se Si Sm Sn Sr Tb Te Th Ti

1 5 1 0.5 10 0.5 0.5 0.5 10 0.1 0.1 0.2 1 0.1 1 0.1 0.1 0.1

0.05 0.05 0.9

4 1 5

7 45 53

50–100

5

50–1000

0.3

50–1000

0.1

4 2 0.1 40 0.4 0.2

4

50–1000

0.03 0.07

1 1

7 7

50–1000 50–1000

0.03

2

6

50–1000

2

7

50–1000

0.5 0.1

0.3 0.1 0.3 0.1 200–1000 0.1 0.1 0.1 50 0.1 1 0.1 1 3 0.3 10 0.1 1 5.0 0.1 0.5 0.1 0.5 0.1 0.1 0.1 0.1 0.2 5 10 0.1 1 0.3 0.1 2 0.1

20 6 1 1 2 20

30 2 8

50–800

15

50–800

0.08

2 9 5

42

200–1000

0.08

13 4

32

0.1 0.1

0.2

5.0

7 3 60 0.1

75

50–600

0.03 1 (Continued )

280 Table 8

Standardized Methods for Water-Quality Assessment Continued

Element

ICP-MS EN ISO 17294-2 a (mg l 1)

ICP-MS ASTM D5673 b (mg l 1)

Tl Tm U V W Zn Zr

0.1 0.1 0.1 1 0.3

0.09 0.02 0.02 0.2

ICP-OES EN ISO 11885 c (mg l 1)

ICP-OES ASTM D1976 d (mg l 1)

DCP-OES ASTM D4190 e (mg l 1)

40

1 10 1 0.3

8

50–1000

2

50–1000

a

Lower limit of application range for the most sensitive isotope. Estimated instrument detection limit. c Limit of detection for the recommended or most sensitive wavelength given in the standard, conventional pneumatic nebulation. d Estimated detection limit. e Range covered in the study for the standard. b

3.11.4.4.2 Methods for determination of organic compounds or jointly determinable groups of compounds In the beginning of water analysis, organic water constituents were detected and quantified only as bulk or sum parameters. The progress in analytical chromatographic techniques, especially in gas chromatography (GC) and high-performance liquid chromatography (HPLC), made it possible to separate, identify, and determine organic substances that occur in water only in traces. This is performed by highly potent chromatographic separation methods combined with detectors based on various detection principles. Often, these analyses are preceded by an enrichment (concentration) step together with suitable cleanup. In parallel, in legal framework regulations have been made which call for advanced organic analytical methods. Organic compounds that have to be monitored according to the European Drinking Water Directive (98/83/EC, 1998) comprise acryl amide, benzene, 1,2-dichloroethane, epichlorohydrine, pesticides, PAH, tetrachloroethylene and trihalomethanes, and vinyl chloride (see Table 4). They are based on the guideline values for water ingredients with respect to human toxicological concerns given by the WHO (2006) or in some cases are even stricter. Organic micropollutants are also included in the list of priority substances in the field of water policy (2008/ 105/EC, 2008; see also Table 4). As organic substances considered for possible identification as priority substances, AMPA, bentazone, bisphenol-A, dicofol, EDTA, glyphosate, mecoprop, musk xylene, perfluorooctane sulfonic acid (PFOS), quinoxyphen, dioxins, and polychlorinated biphenols are listed in annex III of the Directive. Organic trace analysis makes up the main part of the biannual reviews by Richardson on current issues in water analysis, which also give an outlook on the contaminants considered for further regulations in the US, collated in the contaminant candidate list (CCL) (Richardson, 2003, 2005, 2007, 2009) (for extended information on emerging organic contaminants, see also Chapter 3.04 Emerging Contaminants). In Table 9, standardized methods for organic micropollutants or constituents are given. The majority of the

substances reflect the lifestyle of an industrialized society with civilized urban life and intensive agriculture. Some noxious substances to be determined are formed as unintended by-products of disinfection or oxidation during water treatment. The organic microconstituents to be dealt with are not exclusively xenobiotics, microcystins, for instance, are algal toxins of concern (Tillmanns et al., 2007; see also Chapter 3.14 Drinking Water Toxicology in Its Regulatory Framework). Several methods are dedicated to pesticides, which find their way into aquatic systems preferably by diffuse input and runoff. As pesticides and plant-treatment agents belong to various classes of substances based on their chemical nature, different chromatographic techniques or detection principles are applied for their determination. Besides chromatographic methods, a guideline for the determination of plant treatment and pesticide agents using selective immunoassays is given (ISO 15089). For determination of PAH, HPLC methods with fluorescence detection after liquid–liquid extraction (EN ISO 17993, ISO 7981-2) are available. A current standardization project in ISO/TC 147/SC 2 deals with the determination of PAH by GC-MS (intended ISO 28540). The majority of standardized chromatographic methods is based on column techniques. Thin-layer techniques have been specified for pesticides (ISO/TS 11370) and for PAH (ISO 7981-1). GC-MS (with negative ion chemical ionization – NCI) is also applied for the analysis of short-chain polychlorinated alkanes (SCCP) in a current standardization project (intended ISO 12010). Quantification is based on multiple linear regression calibration. Solid-phase micro-extraction (SPME) as an enrichment method has now become a subject of standardization. A document on the determination of plant-treatment agents and biocide products by GC-MS after SPME is already in the finaldraft stage in ISO (intended ISO 21708). ASTM published a standard practice on applicability of SPME for the analysis of volatile organic compounds (ASTM D6520) and specifies the use of this technique for the extraction of PAH from sediment pore waters prior to GC/MS analysis. Further recent developments considered for standardization as powerful methods are

Standardized Methods for Water-Quality Assessment Table 9

281

Standardized methods for determination of organic micropollutants. Most of them are groups of jointly determinable substances

Organic plant treatment ISO/TS 11370 EN ISO 11369 ISO 15089

and pesticide agents Automated multiple development (AMD) technique; Z50 ng l1 HPLC-UV, after solid–liquid extraction; for drinking water; LOQ 0.1 mg l1 Selective immunoassays

Organochlorine pesticides EN ISO 6468 GC, after liquid–liquid extraction; LOD 1–10 ng l1; validated for hexachlorobenzene, b-endosulfane, PCB 180, 1,2,4,5tetrachlorobenzene, a-HCH, dieldrine, p,p-DDE, p,p-DDT, PCB 28, PCB 52, PCB 101, PCB 138, PCB 153, PCB 194 ASTM D5175 GC, microextraction; LODo1 mg l1 ASTM D5812 Aldrine, chlordane, DCPA, 4,40 -DDD, 4,40 -DDE, 4,40 -DDT, dieldrin, endosulfane Phenoxyalkanoic herbicides, including bentazones and hydroxybenzonitriles EN ISO 15913 GC-MSD, after SPE and derivatization, for groundwater and drinking water, 450 ng l1 Chlorinated phenoxyacid herbicides ASTM D5317 GC-EC; bentazone, 2,4-D, 2,4-DB, 3,5-dichlorobenzoic acid, dichlorprop, pentachlorophenol (PCP), 2,4,4-T Chlorobenzenes EN ISO 6468

GC, after liquid–liquid extraction; LOD 1–10 ng l1

Biphenyls, polychlorinated (PCB) ISO 17858 HRGC/HRMS, after extraction EN ISO 6468 GC, after liquid–liquid extraction; LOD 1–10 ng l1 ASTM D5175 GC, microextraction Organotin compounds EN ISO 17353 GC, -AED, -FPD, or -MSD, after alkylation and extraction, working range 10–1000 ng l1 Chlorophenols EN 12673

GC, ECD or MSD, after acetylation, n-hexane extraction, 0.1 mg l1 to 1 mg l1

Polycyclic aromatic hydrocarbons (PAH) EN ISO 17993 HPLC, fluorescence detection, after after liquid–liquid extraction, 40.005 mg l1 in drinking water and groundwater, 40.01 mg l1 in surface water; 15 PAH ISO 7981-2 HPLC, fluorescence detection, liquid–liquid extraction; drinking and groundwater: 40.005 mg l1, surface water: 40.01 mg l1 ISO 7981-1 HPTLC (thin-layer chromatography), after liquid–liquid extraction; 40–240 ng l1 ASTM D7363 SPME, GC/MS, SIM; in sediment pore water Monocyclic aromatic hydrocarbons, naphthalene, and chlorinated compounds (volatile organic compounds) EN ISO 15680 GC, purge-and-trap, thermal desorption, 10 ng l1 to 100 mg l1 Explosives and related compounds EN ISO 22478 HPLC-UV-DAD, after SPE, 0.1–0.5 mg l1; in groundwater near ammunition waste sites or in drinking water; nitroluenes, amino nitro toluenes, nitrobenzenes, picric acid, oktogen, and hexogen Nitrophenols EN ISO 17495

GC-MSD, after solid-phase extraction, methylation, 40.5 mg l1

Glyphosate and aminomethyl phosphonic acid (AMPA) ISO 21458 HPLC, post-column derivatization and fluorescence detection; 40.05 mg l1 Parathion, parathion-methyl and organophosphorus compounds EN 12918 GC, after dichloromethane extraction, 0.01–1 mg l1 Organophosphate compounds ASTM D7597 LC/MS-MS, ESI (SRM); ethyl hydrogen dimethylamidophosphate, ethyl methylphosphonic acid, methylphosphonic acid, and pinacolyl methylphosphonic acid Organic nitrogen and phosphorus compounds EN ISO 10695 GC-NPD, after dichloromethane extraction, LOD 0.1–1.0 mg l1; after solid-phase extraction, LOD 0.012–0.060 mg l1 ASTM D5475

GC-NPD

Haloacetic acids, trichloroacetic acid, dalapon EN ISO 23631 GC-ECD or GC-MS, after liquid–liquid extraction using MTBE and derivatization using diazomethane, working range 0,5– 10 mg l1; bromochloroacetic acid, dibromoacetic acid, dichloroacetic acid, monobromoacetic acid, and monochloroacetic acid Highly volatile halogenated hydrocarbons DIN 38407-30 Headspace-GC; swimming pool waters; validated for the trihalomethanes CHCl3, CHBrCl2, CHBr2Cl, and CBr3Cl EN ISO 10301 GC, liquid–liquid extraction, LOQ 0.1–50 mg l1; static headspace, LOQ 0.1–200 mg l1 (Continued )

282

Standardized Methods for Water-Quality Assessment

ASTM D3973 GC, halogen specific detectors or MSD; 1–200 mg l1 ASTM D5316 Microextraction GC; 1,2-dibromoethane, 1,2-dibromo-3-chloropropane Purgable organic compounds (including organohalides) ASTM D3871 GC, dynamic headspace sampling, low mg l1 to low mg l1 range Phthalates EN ISO 18856 Microcystins ISO 20179

GC-MSD, after SPE, 0.02–0.150 mg l1 RP-HPLC, UV-DAD, suitable for control of WHO guideline value (1 mg l1), after SPE. For samples containing suspended algal biomass after a preceding liquid extraction step; validated for MCYST-RR, MCYST-YR, MCYST-LR

Alkylphenols EN ISO 18857–1

GC-MSD, liquid–liquid extraction (toluene), for non-filtered samples; validated for 4-(1,1,3,3-tetramethylbutyl)phenol and 4-nonylphenol (mixture of isomers); application range 0.005–0.2 mg l1 for 4-(1,1,3,3-tetramethylbutyl)phenol, 0.02– 0.2 mg l1 for 4-nonylphenol Nonylphenol, tert-octylphenol, nonylphenol monoethoxylate, nonylphenol diethoxylate ASTM D7485 LC/MS-MS, reporting range 100–2000 ng l1 ISO 24293 SPE, GC/MS; nonylphenol, individual isomers ASTM D7065 Nonylphenol monoethoxylate, nonylphenol diethoxylate Complexing agents EN ISO 16588

GC

Phenols, monovalent ISO 8165-1 ISO 8165-2 ASTM D2580

GC, extraction GC, derivatization GC, direct aqueous injection, 41 mg l1

Perfluorooctanesulfonate (PFOAS), perfluorooctanoate (PFOA) ISO 25101 LC/MS, SPE Polybrominated diphenylethers EN ISO 22032 GC/MS, EI or NCI, extraction, in sediment and sludge, LOQ (EI) 0.05–25 mg kg1 for tetra- to octabromo congeners, 0.3– 100 mg kg1 for decabromodiphenylether, lower for NCI Bisphenol A ASTM D7574

LC/MS-MS, MDL 20–600 ng l1

N-Methyl carbamates ASTM D7600 ASTM D5315

LC/MS-MS; aldicarb, carbofuran, oxamyl, and methomyl Direct aqueous injection HPLC post-column derivatization

Benzene ISO 11423-1 ISO 11423-2

Headspace-GC Extraction, GC

Tetra- to octa-chlorinated dioxins and furans ISO 18073 Isotope dilution HRGC/HRMS

liquid-chromatography tandem mass-spectrometry (LC-MS/ MS) and ultrahigh-performance liquid chromatography (UPLC).

3.11.4.5 Radiological Methods The estimation of radioactivity and the determination of radionuclides in water require special methods. In Table 10, available standards are listed. The European Drinking Water Directive (98/83/EC) considers radioactivity by giving parametric values for tritium (100 Bq l1) and a total indicative dose of 0.10 mSv a1 (excluding tritium, potassium-40, radon, and radon decay products). Gross alpha-activity and gross beta-activity measurements are suitable as a screening steps to estimate this quantity (Aurand and Ru¨hle, 2003). The standards ISO 9696 and ISO 9697, both recently revised, are wellestablished methods for this. A standard based on the liquid

scintillation counting (LSC) method, which requires reduced sample preparation and counting time (Forte et al., 2006), is in the final-draft stage (intended ISO 11704). This method is also applied in the active standard on measurement of tritium activity concentration (ISO 9698, under revision), and is intended for measurement of carbon-14 activity (ISO 13162, in preparation). Further standardization projects in ISO/TC 147 deal with the measurement of strontium (89Sr, 90Sr, intended ISO 13160) and polonium-210 activity concentration (intended ISO 13161) (for further information on radioactivity in waters, see also Chapter 3.03 Sources, Risks, and Mitigation of Radioactivity in Water).

3.11.4.6 Microbiological Methods Generally, the most common and most weighty hazards for human health that can arise from water are water-borne

Standardized Methods for Water-Quality Assessment Table 10 Parameter

283

Standardized methods for radiochemical examination and determination of radionuclides in water Standard

Alpha-activity ISO 9696 (thick source method), ASTM D1943 Beta-activity ISO 9697 (thick source method), ASTM D1890 Gross alpha- and beta- ISO 10704 (thin source deposit method), ASTM D7283 (liquid scintillation counting) activity Radionuclides ISO 10703 (activity concentration, by high-resolution gamma-ray spectrometry), ASTM 3649 Tritium ISO 9698 (liquid scintillation counting method), ASTM D4107 Radium ASTM D3454 (radium-226), ASTM D2460 (alpha-particle-emitting isotopes) Iron ASTM D4922 Strontium-90 ASTM D5811 Lead-210 ASTM D7535 Technetium-99 ASTM D7168 (solid phase extraction disk) Uranium ASTM D5174 (pulsed laser phosphorimetry), ASTM D6239 (high-resolution alpha-liquid-scintillation spectrometry), ASTM D3972 (alpha-particle spectrometry)

infections and outbreaks, induced by bacteria, viruses, or parasites. To grant the microbial safety of the distributed water is therefore the first-priority task of water suppliers. The philosophy of microbiological examination of drinking water is not only to prove the absence of definite pathogens, but also to prove that fecal influence on the water can be excluded. This is done by checking for suitable nonpathogenic indicator organisms that have to be absent, too. Microbial safety is an item for swimming pool waters and recreational waters as well. Furthermore, microbial fouling is of concern in technical systems. In Table 11, standardized methods available for the testing of water for microorganisms are listed. For some microbiological parameters to be checked routinely in drinking water, the European Drinking Water Directive (98/83/EC, 1998) directly refers to methods for examination specified in the following standards: EN ISO 9308-1 for Escherichia coli and for coliforms; EN ISO 7899-2 for enterococci; EN ISO 12780, now replaced by EN ISO 16266, for Pseudomonas aeruginosa; and EN ISO 6222 for culturable microorganisms as colony count. All these methods are based on membrane filtration. Alternative methods can be used, provided their equivalency has been shown according to EN ISO 17994. EN ISO 8199 gives a guideline for the calculation and reporting of test results. E. coli is the most important indicator organism for fecal pollution, but not all coliform bacteria are of fecal origin. In some countries, the whole group of thermotolerant coliforms is determined (ISO 9308-2) instead. The method EN ISO 7899-2 preferably captures enterococci species that originate from human or animal intestine. P. aeruginosa, which causes, for example, pneumonia, is especially of concern for immunodeficient persons. A further microbial parameter mentioned in the directive 98/83/EC (1998) is Clostridium perfringens and its spores, which are strict obligate anaerobes and conservative tracers for past and present pollution by parasites. This parameter is especially relevant for raw waters that are influenced by surface water. For its measurement, the directive gives an extended protocol based on membrane filtration. At present, a standard on the detection and enumeration of C. perfringens, suitable to match the specifications of the directive, is being developed in ISO/TC 147/SC4 (intended ISO 14189).

In warm water distribution systems with stagnant phases, Legionella bacteria can occur, which are a health hazard by respiratory uptake (e.g., in aerosols from showers). Therefore, warm water distribution systems are checked for Legionella in buildings, which are frequented by many persons. According to the German drinking water ordinance, monitoring for Legionella is mandatory in warm water systems of public buildings (TrinkwV, 2001). Up to now, active standards for microbiological analysis are all based on cultivation methods. Recently, a standardization project (intended ISO 12869) has been started in ISO/TC 147/SC 4 to consider the new technique of polymerase chain reaction (PCR) for the detection of Legionella. This method for the detection of bacteria is faster than the cultivation techniques, with the drawback that it does not discriminate between live and dead organisms (see also Chapter 3.08 Identification of Microorganisms Using the Ribosomal RNA Approach and Fluorescence In Situ Hybridization). As suspicions concerning microbial safety entail severe measures (e.g., restriction of water use), fast microbiological testing methods would be advantageous. The test repertoire given in Table 11 also comprises methods for detection of bacteriophages, viruses, and parasites. Coliphages are generally more resistant to chlorination than coliforms and may have some advantage over coliforms as an indicator of treatment efficiency in disinfected waters. The parasites Cryptosporidium and Giardia, which can get into raw waters for drinking water treatment by an unexpected shortcut to wastewaters, were the cause of serious outbreaks (see also Chapter 3.12 Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control). The yeast Candida albicans may serve as an indicator of recreational water quality. Sulfate-reducing bacteria produce hydrogen sulfide and might cause corrosion problems in pipes and pipelines. Iron bacteria utilize ferrous iron as a source of energy and cause deposits of ferric hydroxide, for example, in wells.

3.11.4.7 Biotesting Quantification of water ingredients, or proval of their absence is only one aspect of water-quality assessment. Further, an important approach is biotesting of waters or their

284

Standardized Methods for Water-Quality Assessment

Table 11

Standardized methods for microbiological examination

General Cleaning of equipment Bacterial retention of membrane filters Evaluation of membrane filters Evaluating and controlling microbiological colony count media Enumeration by culture Validation of microbiological methods Establishing equivalence of microbiological methods Sampling for microbiological analysis Special methods Adenosine triphosphate (ATP) content Aquatic bacteria Total and respiring bacteria Culturable microorganisms Microbial colony counts Total active biomass Coliform organisms, thermotolerant coliforms, presumptive Escherichia coli E. coli and coliforms

E. coli Intestinal enterococci

Enterococci Pseudomonas aeruginosa Clostridium perfringens Legionella

Campylobacter species Sulfite-reducing anaerobes (Clostridia)

Sulfate-reducing bacteria Iron bacteria Bacteriophages

ASTM D5245 ASTM F838 ISO 7704 ISO 9998 EN ISO 8199 ENV ISO 13843 EN ISO 17994 EN ISO 19458 ATP firefly (luciferin-luciferase)

ASTM D4012

Enumeration, acridine-orange epifluorescence directcounting Enumeration, microscopy, and acridine-orange INTformazan reduction Colony count, nutrient agar medium Plating methods In cooling tower waters, Kool Kount assay (KKA)

ASTM D4455

Detection and enumeration, multiple tube (most probable number) method Spores and vegetative cells, from water and extracted sediments, detection and enumeration, membrane filtration Detection and enumeration, miniaturized method, most probable number, inoculation in liquid medium Isolation and enumeration, two-step membrane filter procedure Detection and enumeration, miniaturized method most probable number, inoculation in liquid medium Detection and enumeration, membrane filtration Detection in water, using Enterolert Isolation and enumeration, membrane filtration Detection and enumeration, membrane filtration Detection and enumeration, by membrane filtration Detection and enumeration For waters with low bacterial counts, direct membrane filtration Thermotolerant, detection and enumeration Spores, detection and enumeration, enrichment in liquid medium Spores, detection and enumeration, membrane filtration

ASTM D4454 EN ISO 6222 ASTM D5465 ASTM D6530 ISO 9308-2 EN ISO 9308-1

EN ISO 9308-3 ASTM D5392 EN ISO 7899-1 EN ISO 7899-2 ASTM D6503 ASTM D5259 EN ISO 16266 ASTM D5916 ISO 11731 EN ISO 11731-2 ISO 17995 EN 26461-1 (ISO 6461) EN 26461-2 (ISO 6461-2)

In water and water-formed deposits In water and water-formed deposits R-specific RNA bacteriophages, enumeration Somatic coliphages, enumeration Validation of methods for concentration Bacteriophages infecting Bacteroides fragilis, enumeration Coliphages, low level, in waters

ASTM D4412 ASTM D932 EN ISO 10705-1 EN ISO 10705-2 ISO 10705-3 ISO 10705-4

Viruses Enteroviruses Human enteroviruses

Recovery from wastewater sludges Recovery from waters Detection by monolayer plaque assay

ASTM D4994 ASTM D5244 EN 14486

Parasites Cryptosporidium oocysts and Giardia cysts

Isolation and identification

ISO 15553

Yeast Candida albicans

Enumeration

ASTM D4249

ASTM D6734

Standardized Methods for Water-Quality Assessment

constituents. For this, the standards portfolio of CEN and ISO contains testing methods on biodegradability and for ecotoxicity testing.

3.11.4.7.1 Biodegradability An important aspect of wastewater treatment is biodegradation; therefore, biodegradability is a key parameter for evaluation of wastewater constituents. It is an important parameter to assess the behavior of water ingredients in the aquatic environment as well. This has two aspects: (1) their elimination by biological processes or possible persistence and (2) their influences on the oxygen balance of Table 12

natural water, since oxygen is needed for aerobic degradation processes. In the 1960s, the issue of biodegradability called for attention due to scum formed on rivers from poorly degradable detergents. In Germany, an early assimilation/consumption test was elaborated in 1971 (DEV, 1971). An important impetus for testing of biodegradability was given by the initiative Registration, Evaluation, Authorization and Restriction of Chemicals (REACH, 2006). For this, the Organization for Economic Cooperation and Development (OECD) developed a series of degradability tests (OECD, 2008). In Table 12, methods in the portfolio of ISO/TC 147 and CEN/TC 230 for biodegradability testing are listed. Most of the

Standardized methods for testing biodegradability in aqueous media

Standard

Test

Static tests for ultimate aerobic biodegradability (duration 28 days) EN ISO 14593 CO2 headspace test Closed bottle with headspace, 2–40 mg l1 DOC Suitable for volatile test substances EN ISO 9408 Closed respirometer test At least 100 mg l1 ThOD Suitable for volatile test substances EN ISO 9439 Carbon dioxide evolution test 2–40 mg l1 DOC EN ISO 7827 Method by analysis of dissolved organic carbon (DOC) 2-40 mg l1 DOC ISO 10708

EN ISO 10707

EN ISO 9888

285

Two-phase closed bottle test (BODIS) Determination of biochemical oxygen demand, 100 mg l1 BOD Closed bottle test Analysis of biochemical oxygen demand Smaller than 10 mg l1 ThOD (about 2–5 mg l1 test substance), low degradation potential Suitable for volatile or inhibitory test substances Static test (Zahn–Wellens test) 50–400 mg l1, high degradation potential Specified for wastewaters

Semi-continuous test for ultimate aerobic biodegradability EN ISO 9887 Semi-continuous activated sludge method (SCAS) High degradation potential (bacteria concentration 1– 4 g l1 supended matter), duration: 12–26 weeks Effective residence time of wastewater: 36 h Tests for aerobic biodegradability at low concentrations ISO 14592-1 Shake-flask batch test with surface water or water/ sediment suspensions ISO 14592-2 Continuous flow river model with attached biomass EN ISO 11733 Activated sludge simulation test 10–20 mg l1 DOC; aim: prediction of concentration in the effluent; HRT: 6 h Test for ultimate anaerobic biodegradability in digested sludge EN ISO 11734 Measurement of biogas production Test substance: 20–100 mg l1 OC Static closed bottle test, duration up to 60 days

Analytical parameter

Increase of TIC; optional: decrease of DOC (for watersoluble substances) Oxygen demand optional: additionally decrease of DOC for water-soluble substances Release of CO2 Dissolved organic carbon, at least at 3 days within the test period Dissolved oxygen (ThOD or COD)

Dissolved oxygen

Dissolved organic carbon, after 3 h check for adsorption

Dissolved organic carbon

DOC or BOD

TIC

Methodological guideline papers and additional tests ISO TR 15462 Selection of tests for biodegradability ISO 16221 Determination of biodegradability in the marine environment ISO 18749 Adsorption of substances to activated sludge EN ISO 10634 Preparation of poorly water-soluble organic compounds for biodegradability testing

286

Standardized Methods for Water-Quality Assessment

current standards were developed in the early 1990s, many of them based on OECD protocols. The philosophy of the different approaches is discussed in Pagga (1997). The technical report ISO/TR 15462 on the selection of tests for biodegradability, revised in 2006, gives an extended overview of the different methods, and their principles and scopes. Degradation processes in the environment and in treatment plants are manifold, which are taken into account by different test designs. Biodegradation tests can roughly be divided into ones which are designed for aerobic or anaerobic conditions; they can be operated in static, semicontinuous, or continuous mode. The majority of the tests in Table 12 are static tests for ultimate aerobic biodegradability with a duration of 28 days. As inoculum, mixed populations are mandatory, in order to cover the variety of degradation processes in the environment. The inoculum and the state of adaptation of the microorganisms cannot be standardized (Pagga, 1997). In most cases, activated sludge from municipal wastewatertreatment plants is applied. The tests differ in the analytical parameters by which the degree of conversion is monitored. For ultimate aerobic biodegradation (total mineralization), this can, in principle, be performed by measurement of DOC, oxygen consumption, or CO2 formation. A comparison of CO2 production or oxygen uptake with DOC removal as a measure of ultimate biodegradation is given in Reuschenbach et al. (2003) and in Strotmann et al. (2004). In order to capture the elimination that is caused by biodegradation, biodegradability tests comprise a blank control, a test with a reference substance, an inhibition control, an abiotic elimination control, and an adsorption control. Inhibitory effects to activated sludge can be tested by standardized inhibition tests (e.g., EN ISO 8192 or ISO 15522). Suitable tests for inhibition or toxicity are also given in ISO/TR 15462. Most tests require water solubility for the tested substances. A guideline for preparation of poorly water-soluble substances for testing is given in EN ISO 10634. Headspace and closedrespirometer tests are suitable settings for volatile substances to be tested. Biodegradation tests can also be designed for marine systems, which require a different composition of the test media from that used for limnic systems (Pagga, 1997).

3.11.4.7.2 Ecotoxicity and bioeffect testing The goal of ecotoxicity testing in water analysis is to assess effects of environmental samples on survival, growth, or reproduction of aquatic wildlife. The tests are used for example to estimate the environmental impact of treated wastewaters released to natural waters and are therefore important elements in emission control (Thompson et al., 2005). Standards available for ecotoxicity testing are listed in Table 13. They are designed for different trophic levels, for example, bacteria, algae, higher plants, daphnia, rotifers, and fish. EN ISO 15088 specifies a fish egg test which was developed to replace the fish test for reasons of animal welfare (Pluta and Rosenberg, 2005). The repertoire of testing methods for aquatic toxicity covers different ecotoxic effects by estimating acute toxicity, chronic toxicity, and inhibition of growth or of other physiological functions. Bioeffects are also tested on a suborganismic level. This applies to the tests on genotoxicity given in the table.

Genotoxicity is one of the effects of main concern in the assessment of substances released to the environment. Current standardization projects in ISO/TC 147/SC 5 deal with methods suitable for the testing of sediments, for example, a contact test for the inhibition of dehydrogenase activity of Arthrobacter globiformis (intended ISO 10871) and a test for toxic effects of sediment and soil samples on growth, fertility, and reproduction of the nematode Caenorhabditis elegans (intended ISO 10872). Determination of vitellogin is being specified for physiological measurements on fish (intended ISO 23893-3). For the algal test species of EN ISO 8692, Desmosdesmus subspicatus and Pseudokirchneriella subcapitata, a growth inhibition test in a format using microtiter plates is in preparation. Concerning the expression of endpoints in algal or plant growth tests, where growth rate is used for the testing endpoint, ISO decided that in future standards the endpoint should be expressed as ErCx (concentration with x% effect), not as lowest observed effect concentration (LOEC) or no observed effect concentration (NOEC).

3.11.4.8 Methods for Assessment of Water Bodies The monitoring of water bodies is a traditional task of water analysis. The use of biological indicators for water quality, for example, for saprobic levels, there has been a long tradition as well (Vilela Junqueira et al., 2010; DIN 38410-1). Increased efforts in this field have been induced by the WFD (2000/60/ EC, 2000), which stipulates that the member states perform a classification and assessment of their water bodies. Near-naturalness is the key and target variable in this approach, which first has to be metrologically captured by appropriate descriptors in order to define a good status, and based on this, has to be maintained or restored by suitable measures. Concerning morphological features, a sophisticated classification within the main categories, rivers, lakes, transitional waters, and coastal waters, has to be made. The ecological status has to be assessed for surface waters through evaluation of the levels of phytoplankton (for lentic systems), aquatic flora, benthic invertebrate fauna, and fish fauna (for freshwaters and transitional waters) concerning composition and abundance. For phytoplankton, determination of biomass is also required. In freshwaters, fish populations have to be evaluated, in addition, with respect to their age structure. Table 14 gives an overview of already-existing standards for ecological or morphological classification of water bodies. The standards concerning the sampling of macro-invertebrates (EN 27828, EN 28265, EN ISO 9391, and EN ISO 8689-1, EN ISO 8689-2) are directly referred to in the WFD. The repertoire of methods is still in development. Current standardization projects deal with in vivo absorption techniques for the estimation of chlorophyll (CEN project 00230263), quantitative and qualitative investigation of marine phytoplankton (intended EN 15972), selection of sampling methods, and devices for benthic macro-invertebrates (intended EN ISO 10870), and with estimation of fish abundance with mobile hydroacoustic methods (intended EN 15910). Furthermore, guidance standards on selection and design of taxonomic keys (CEN project 00230275) and on design of multi-metric indices (CEN project 00230261) are in preparation. Biological–ecological assessment methods are

Standardized Methods for Water-Quality Assessment Table 13

287

Standardized methods for ecotoxicity and effects testing

General procedures, ISO/TR 11044 ISO 14442 ISO/TS 20281 EN ISO 5667-16

methodology Scientific and technical aspects of batch algae growth inhibition tests Algal growth inhibition tests with poorly soluble materials, volatile compounds, metals and wastewater Statistical interpretation of ecotoxicity data Guidance on biotesting of samples

Organismic tests EN ISO 7346-1 to -3 ISO 23893-1 ISO/TS 23893-2 ISO 10229 ISO 12890 EN ISO 15088 EN ISO 6341 ISO 10706 ISO 20665 EN ISO 16712 ISO 14669 ISO 20666 EN ISO 20079 EN ISO 10253 EN ISO 8692 EN ISO 10712 EN ISO 11348-1 to 3 EN ISO 8192

Test organism Freshwater fish (Brachydanio rerio Hamilton–Buchanan (Teleostei, Cyprinidae)) Freshwater fish Freshwater fish; rainbow trout (Oncorhynchus mykiss Walbaum (Teleostei, Salmonidae)) Embryos and larvae of freshwater fish Fish egg, Danio rerio Daphnia magna Straus (Cladocera, Crustacea) Daphnia magna Straus (Cladocera, Crustacea) Ceriodaphnia dubia (Cladocera, Crustacea) Amphipods Marine copepods Rotifers, Brachionus calyciflorus Higher water plants, Lemna minor Marine algae, Skeletonema costatum and Phaeodactylum tricornutum Freshwater algae, unicellular green algae Bacteria, Pseudomonas putida Bacteria, Vibrio fischeri

EN ISO 9509 ISO 15522 ISO 13641-1, -2

Bacteria, activated sludge for carbonaceous and ammonium oxidation Bacteria, activated sludge microorganisms Bacteria, activated sludge microorganisms Bacteria, anaerobic bacteria

Suborganismic tests EN ISO 21427-2 ISO 21427-1 ISO 13829 ISO 16240

for genotoxicity Induction of micronuclei, cell line V79 Induction of micronuclei, using amphibian larvae umu test Salmonella/microsome test (Ames test)

subject to validation measures as well, a document on design and analysis of interlaboratory comparison studies is being developed (intended EN 16101). In the field of morphological classification of water bodies, a standardization project is underway for assessment of the hydromorphological features of lakes (intended EN 16039).

3.11.5 Resume and Outlook Standards for water examination are available which are designed to assess the quality of waters concerning chemical, microbiological, biological, and ecological aspects, and to control the efficiency of water-treatment processes. They have been developed and are continuously updated by the interested parties in this field, comprising science, authorities, manufacturers of analytical instruments, industries, and users. The process of development and consensus building is moderated by standardization organizations, releasing the

Effect Acute lethal toxicity Biochemical and physiological measurement Prolonged toxicity, growth rate Toxicity Acute toxicity Inhibition of mobility, acute toxicity test Long-term toxicity Chronic toxicity Acute toxicity (marine estuarine sediment) Acute lethal toxicity Chronic toxicity, 48 h Growth inhibition (Duckweed growth inhibition test) Growth inhibition Growth inhibition Growth inhibition (Pseudomonas cell multiplication inhibition test) Inhibition of light emission (luminescent bacteria test) Inhibition of oxygen consumption Inhibition of nitrification Inhibition of growth Inhibition of gas production

standards as editors. On European and international levels, the NSBs of individual countries participate in the supernational standardization organizations CEN and ISO, whose combined and complementary efforts have created partially overlapping standards portfolios of about 300 standardized methods in total. A further collection of private sector standards on water testing has been elaborated by ASTM International, the former ASTM, who opened up for participation of international experts. Recent trends tend toward miniaturization and automation of methods. Standardization of methods consistently has to find its position between convenience of commercially available ready-to-use components and the requirements of traceability and disclosure. Furthermore, the variability of results is increasingly dealt with, not only for chemical analysis, but also for microbiological methods, ecological assessment, and sampling (Strub et al., 2009). On a European level, the requirements of the WFD and its daughter directives concerning the comparability necessary for meaningful ecological and

288

Standardized Methods for Water-Quality Assessment

Table 14

Standardized methods for assessment of water bodies

Hydromorphology EN 15843 EN 14614

Degree of modification Assessment of hydrological features

Biological and ecological assesment EN 14996

Quality assurance

Chlorophyll-a ISO 10260

Determination

Phytoplankton EN 15204

Enumeration, Utermo¨hl technique

Phytobenthos EN 15708

Surveying, sampling and laboratory analysis

Shallow running water

Benthic diatoms EN 13946 EN 14407

Routine sampling Identification, enumeration, interpretation

Rivers Running waters

Macrophytes EN 15460 EN 14184

Surveying Surveying

Lakes Running waters

Zooplankton EN 15110

Sampling

Standing waters

Macroinvertebrates EN ISO 9391 EN 27828 (ISO 7828) EN 28265 (ISO 8265) EN ISO 8689-1, -2

Sampling Handnet sampling (benthic) Quantitative samplers (benthic, on stony substrata) Interpretation and presentation (benthic)

Deep waters

Fish EN 14962 EN 14011 EN 14757

Sampling, selection of method Sampling, with electricity Sampling, with multi-mesh gillnets

Chironomidae, pupal exuviae EN 15196

Sampling and processing

Soft-bottom macrofauna EN ISO 16665

Sampling and sample processing

Marine systems

Hard substrate communities EN ISO 19493

Biological surveys

Marine systems

large-scale morphological assessment and concerning the sensitivity necessary to check for compliance with the lowlevel chemical EQS are a challenge for future work.

3.11.6 List of Standards ASTM D511-09 – Standard test methods for calcium and magnesium in water. ASTM D512-04 – Standard test methods for chloride ion in water. ASTM D513-06 – Standard test methods for total and dissolved carbon dioxide in water. ASTM D516-07 – Standard test method for sulfate ion in water. ASTM D596-01(2006) – Standard guide for reporting results of analysis of water. ASTM D857-07 – Standard test method for aluminum in water.

Rivers Rivers

Shallow freshwaters Rivers

ASTM D858-07 – Standard test methods for manganese in water. ASTM D859-05 – Standard test method for silica in water. ASTM D888-09 – Standard test methods for dissolved oxygen in water. ASTM D932-85(2009) – Standard test method for iron bacteria in water and water-formed deposits. ASTM D1066-06 – Standard practice for sampling steam. ASTM D1067-06 – Standard test methods for acidity or alkalinity of water. ASTM D1068-05e1 – Standard test methods for iron in water. ASTM D1125-95(2009) – Standard test methods for electrical conductivity and resistivity of water. ASTM D1126-02(2007)e1 – Standard test method for hardness in water. ASTM D1129-06ae1 – Standard terminology relating to water. ASTM D1141-98(2008) – Standard practice for the preparation of substitute ocean water. ASTM D1179-04 – Standard test methods for fluoride ion in water.

Standardized Methods for Water-Quality Assessment

ASTM D1193-06 – Standard specification for reagent water. ASTM D1246-05 – Standard test method for bromide ion in water. ASTM D1252-06 – Standard test methods for chemical oxygen demand (dichromate oxygen demand) of water. ASTM D1253-08 – Standard test method for residual chlorine in water. ASTM D1291-06 – Standard practice for estimation of chlorine demand of water. ASTM D1292-05 – Standard test method for odor in water. ASTM D1293-99(2005) – Standard test methods for pH of water. ASTM D1385-07 – Standard test method for hydrazine in water. ASTM D1426-08 – Standard test methods for ammonia nitrogen in water. ASTM D1429-08 – Standard test methods for specific gravity of water and brine. ASTM D1498-08 – Standard test method for oxidation-reduction potential of water. ASTM D1687-02(2007)e1 – Standard test methods for chromium in water. ASTM D1688-07 – Standard test methods for copper in water. ASTM D1691-02(2007)e1 – Standard test methods for zinc in water. ASTM D1783-01(2007) – Standard test methods for phenolic compounds in water. ASTM D1886-08 – Standard test methods for nickel in water. ASTM D1890-05 – Standard test method for beta particle radioactivity of water. ASTM D1943-05 – Standard test method for alpha particle radioactivity of water. ASTM D1971-02(2006) – Standard practices for digestion of water samples for determination of metals by flame atomic absorption, graphite furnace atomic absorption, plasma emission spectroscopy, or plasma mass spectrometry. ASTM D1976-07 – Standard test method for elements in water by inductively coupled argon plasma atomic emission spectroscopy. ASTM D2035-08 – Standard practice for coagulation–flocculation jar test of water. ASTM D2036-09 – Standard test methods for cyanides in water. ASTM D2330-02 – Standard test method for methylene blue active substances. ASTM D2460-07 – Standard test method for alpha-particleemitting isotopes of radium in water. ASTM D2580-06 – Standard test method for phenols in water by gas–liquid chromatography. ASTM D2777-08e1 – Standard practice for determination of precision and bias of applicable test methods of committee D19 on water. ASTM D2791-07 – Standard test method for on-line determination of sodium in water. ASTM D2908-91(2005) – Standard practice for measuring volatile organic matter in water by aqueous-injection gas chromatography. ASTM D2972-08 – Standard test methods for arsenic in water. ASTM D3082-09 – Standard test method for boron in water.

289

ASTM D3084-05 – Standard practice for alpha-particle spectrometry of water. ASTM D3223-02(2007)e1 – Standard test method for total mercury in water. ASTM D3325-90(2006) – Standard practice for preservation of waterborne oil samples. ASTM D3326-07 – Standard practice for preparation of samples for identification of waterborne oils. ASTM D3328-06 – Standard test methods for comparison of waterborne petroleum oils by gas chromatography. ASTM D3352-08a – Standard test method for strontium ion in brackish water, seawater, and brines. ASTM D3370-08 – Standard practices for sampling water from closed conduits. ASTM D3372-02(2007)e1 – Standard test method for molybdenum in water. ASTM D3373-03(2007)e1 – Standard test method for vanadium in water. ASTM D3414-98(2004) – Standard test method for comparison of waterborne petroleum oils by infrared spectroscopy. ASTM D3415-98(2004) – Standard practice for identification of waterborne oils. ASTM D3454-05 – Standard test method for radium-226 in water. ASTM D3557-02(2007)e1 – Standard test methods for cadmium in water. ASTM D3558-08 – Standard test methods for cobalt in water. ASTM D3559-08 – Standard test methods for lead in water. ASTM D3561-02(2007)e1 – Standard test method for lithium, potassium, and sodium ions in brackish water, seawater, and brines by atomic absorption spectrophotometry. ASTM D3590-02(2006) – Standard test methods for total Kjeldahl nitrogen in water. ASTM D3645-08 – Standard test methods for beryllium in water. ASTM D3648-04 – Standard practices for the measurement of radioactivity. ASTM D3649-06 – Standard practice for high-resolution gamma-ray spectrometry of water. ASTM D3650-93(2006) – Standard test method for comparison of waterborne petroleum oils by fluorescence analysis. ASTM D3651-07 – Standard test method for barium in brackish water, seawater, and brines. ASTM D3694-96(2004) – Standard practices for preparation of sample containers and for preservation of organic constituents. ASTM D3695-95(2007) – Standard test method for volatile alcohols in water by direct aqueous-injection gas chromatography. ASTM D3697-07 – Standard test method for antimony in water. ASTM D3856-95(2006) – Standard guide for good laboratory practices in laboratories engaged in sampling and analysis of water. ASTM D3859-08 – Standard test methods for selenium in water.

290

Standardized Methods for Water-Quality Assessment

ASTM D3864-06 – Standard guide for continual online monitoring systems for water analysis. ASTM D3865-09 – Standard test method for plutonium in water. ASTM D3866-07 – Standard test methods for silver in water. ASTM D3867-09 – Standard test methods for nitrite– nitrate in water. ASTM D3868-09 – Standard test method for fluoride ions in brackish water, seawater, and brines. ASTM D3869-09 – Standard test methods for iodide and bromide ions in brackish water, seawater, and brines. ASTM D3871-84(2003) – Standard test method for purgeable organic compounds in water using headspace sampling. ASTM D3875-08 – Standard test method for alkalinity in brackish water, seawater, and brines. ASTM D3919-08 – Standard practice for measuring trace elements in water by graphite furnace atomic absorption spectrophotometry. ASTM D3920-02(2007)e1 – Standard test method for strontium in water. ASTM D3921-96(2003)e1 – Standard test method for oil and grease and petroleum hydrocarbons in water. ASTM D3972-09 – Standard test method for isotopic uranium in water by radiochemistry. ASTM D3973-85(2003) – Standard test method for low-molecular weight halogenated hydrocarbons in water. ASTM D3974-09 – Standard practices for extraction of trace elements from sediments. ASTM D3975-93(2008) – Standard practice for development and use (preparation) of samples for collaborative testing of methods for analysis of sediments. ASTM D3976-92(2005) – Standard practice for preparation of sediment samples for chemical analysis. ASTM D3977-97(2007) – Standard test methods for determining sediment concentration in water samples. ASTM D3986-07 – Standard test method for barium in brines, seawater, and brackish water by direct-current argon plasma atomic emission spectroscopy. ASTM D4012-81(2009) – Standard test method for adenosine triphosphate (ATP) content of microorganisms in water. ASTM D4025-08a – Standard practice for reporting results of examination and analysis of deposits formed from water for subsurface injection. ASTM D4107-08 – Standard test method for tritium in drinking water. ASTM D4127-06 – Standard terminology used with ion-selective electrodes. ASTM D4128-06 – Standard guide for identification and quantitation of organic compounds in water by combined gas chromatography and electron impact mass spectrometry. ASTM D4129-05 – Standard test method for total and organic carbon in water by high-temperature oxidation and by coulometric detection. ASTM D4130-08 – Standard test method for sulfate ion in brackish water, seawater, and brines. ASTM D4165-06 – Standard test method for cyanogen chloride in water. ASTM D4188-08 – Standard practice for performing pressure in-line coagulation–flocculation–filtration test.

ASTM D4189-07 – Standard test method for silt density index (SDI) of water. ASTM D4190-08 – Standard test method for elements in water by direct-current argon plasma atomic emission spectroscopy. ASTM D4191-08 – Standard test method for sodium in water by atomic absorption spectrophotometry. ASTM D4192-08 – Standard test method for potassium in water by atomic absorption spectrophotometry. ASTM D4193-08 – Standard test method for thiocyanate in water. ASTM D4196-05 – Standard test method for confirming the sterility of membrane filters. ASTM D4199-82(2003) – Standard test methods for autoclavability of membrane filters. ASTM D4249-83(2005) – Standard test method for enumeration of Candida albicans in water. ASTM D4281-95(2005)e1 – Standard test method for oil and grease (fluorocarbon extractable substances) by gravimetric determination. ASTM D4282-02 – Standard test method for determination of free cyanide in water and wastewater by microdiffusion. ASTM D4309-02(2007) – Standard practice for sample digestion using closed vessel microwave heating technique for the determination of total metals in water. ASTM D4327-03 – Standard test method for anions in water by chemically suppressed ion chromatography. ASTM D4328-08 – Standard practice for calculation of supersaturation of barium sulfate, strontium sulfate, and calcium sulfate dihydrate (gypsum) in brackish water, seawater, and brines. ASTM D4374-06 – Standard test methods for cyanides in water-automated methods for total cyanide, weak acid dissociable cyanide, and thiocyanate. ASTM D4375-96(2006) – Standard practice for basic statistics in committee D-19 on water. ASTM D4382-02(2007)e1– Standard test method for barium in water, atomic absorption spectrophotometry, graphite furnace. ASTM D4410-03 – Terminology for fluvial sediment. ASTM D4411-03(2008) – Standard guide for sampling fluvial sediment in motion. ASTM D4412-84(2009) – Standard test methods for sulfate-reducing bacteria in water and water-formed deposits. ASTM D4453-02(2006) – Standard practice for handling of ultra-pure water samples. ASTM D4454-85(2009) – Standard test method for simultaneous enumeration of total and respiring bacteria in aquatic systems by microscopy. ASTM D4455-85(2009) – Standard test method for enumeration of aquatic bacteria by epifluorescence microscopy counting procedure. ASTM D4458-09 – Standard test method for chloride ions in brackish water, seawater, and brines. ASTM D4489-95(2006) – Standard practices for sampling of waterborne oils. ASTM D4517-04 – Standard test method for low-level total silica in high-purity water by flameless atomic absorption spectroscopy.

Standardized Methods for Water-Quality Assessment

ASTM D4519-94(2005) – Standard test method for on-line determination of anions and carbon dioxide in high-purity water by cation exchange and degassed cation conductivity. ASTM D4520-03(2008) – Standard practice for determining water injectivity through the use of on-site floods. ASTM D4581-86(2005) – Standard guide for measurement of morphologic characteristics of surface water bodies. ASTM D4638-03(2007) – Standard guide for preparation of biological samples for inorganic chemical analysis. ASTM D4658-09 – Standard test method for sulfide ion in water. ASTM D4691-02(2007) – Standard practice for measuring elements in water by flame atomic absorption spectrophotometry. ASTM D4698-92(2007) – Standard practice for total digestion of sediment samples for chemical analysis of various metals. ASTM D4763-06 – Standard practice for identification of chemicals in water by fluorescence spectroscopy. ASTM D4778-05 – Standard test method for determination of corrosion and fouling tendency of cooling water under heat transfer conditions. ASTM D4785-08 – Standard test method for low-level analysis of iodine radioisotopes in water. ASTM D4822-88(2008) – Standard guide for selection of methods of particle size analysis of fluvial sediments (manual methods). ASTM D4823-95(2008) – Standard guide for core sampling submerged, unconsolidated sediments. ASTM D4839-03 – Standard test method for total carbon and organic carbon in water by ultraviolet, or persulfate oxidation, or both, and infrared detection. ASTM D4840-99(2004) – Standard guide for sampling chainof-custody procedures. ASTM D4841-88(2008) – Standard practice for estimation of holding time for water. ASTM D4922-09 – Standard test method for determination of radioactive iron in water. ASTM D4962-02(2009) – Standard practice for NaI(Tl) gamma-ray spectrometry of water samples containing organic and inorganic constituents. ASTM D4994-89(2009) – Standard practice for recovery of viruses from wastewater sludges. ASTM D5074-90(2008) – Standard practice for preparation of natural-matrix sediment reference samples for major and trace inorganic constituents analysis by partial extraction procedures. ASTM D5127-07 – Standard guide for ultra-pure water used in the electronics and semiconductor industries. ASTM D5128-09 – Standard test method for on-line pH measurement of water of low conductivity. ASTM D5172-91(2004) – Standard guide for documenting the standard operating procedures used for the analysis of water. ASTM D5173-97(2007) – Standard test method for on-line monitoring of carbon compounds in water by chemical oxidation, by UV light oxidation, by both, or by hightemperature combustion followed by gas-phase NDIR or by electrolytic conductivity.

291

ASTM D5174-07 – Standard test method for trace uranium in water by pulsed-laser phosphorimetry. ASTM D5175-91(2003) – Standard test method for organohalide pesticides and polychlorinated biphenyls in water by microextraction and gas chromatography. ASTM D5176-08 – Standard test method for total chemically bound nitrogen in water by pyrolysis and chemiluminescence detection. ASTM D5196-06 – Standard guide for bio-applications grade water. ASTM D5241-92(2004) – Standard practice for microextraction of water for analysis of volatile and semi-volatile organic compounds in water. ASTM D5244-92(2004) – Standard practice for recovery of enteroviruses from waters. ASTM D5245-92(2005) – Standard practice for cleaning laboratory glassware, plasticware, and equipment used in microbiological analyses. ASTM D5246-92(2004) – Standard test method for isolation and enumeration of Pseudomonas aeruginosa from water. ASTM D5257-03 – Standard test method for dissolved hexavalent chromium in water by ion chromatography. ASTM D5258-02(2007) – Standard practice for acidextraction of elements from sediments using closed vessel microwave heating. ASTM D5259-92(2006) – Standard test method for isolation and enumeration of enterococci from water by the membrane filter procedure. ASTM D5315-04 – Standard test method for determination of N-methyl-carbamoyloximes and N-methylcarbamates in water by direct aqueous injection HPLC with post-column derivatization. ASTM D5316-98(2004) – Standard test method for 1,2dibromoethane and 1,2-dibromo-3-chloropropane in water by microextraction and gas chromatography. ASTM D5317-98(2003)e1 – Standard test method for determination of chlorinated organic acid compounds in water by gas chromatography with an electron capture detector. ASTM D5387-93(2007) – Standard guide for elements of a complete data set for noncohesive sediments. ASTM D5391-99(2009) – Standard test method for electrical conductivity and resistivity of a flowing high-purity water sample. ASTM D5392-93(2006) – Standard test method for isolation and enumeration of Escherichia coli in water by the two-step membrane filter procedure. ASTM D5411-05 – Standard practice for calculation of average energy per disintegration (e) for a mixture of radionuclides in reactor coolant. ASTM D5412-93(2005) – Standard test method for quantification of complex polycyclic aromatic hydrocarbon mixtures or petroleum oils in water. ASTM D5462-08 – Standard test method for on-line measurement of low-level dissolved oxygen in water. ASTM D5463-08 – Standard guide for use of test kits to measure inorganic constituents in water. ASTM D5464-07 – Standard test method for pH measurement of water of low conductivity.

292

Standardized Methods for Water-Quality Assessment

ASTM D5465-93(2004) – Standard practice for determining microbial colony counts from waters analyzed by plating methods. ASTM D5475-93(2002) – Standard test method for nitrogenand phosphorus-containing pesticides in water by gas chromatography with a nitrogen–phosphorus detector. ASTM D5540-08 – Standard practice for flow control and temperature control for on-line water sampling and analysis. ASTM D5542-04(2009) – Standard test methods for trace anions in high purity water by ion chromatography. ASTM D5543-09 – Standard test methods for low-level dissolved oxygen in water. ASTM D5612-94(2008) – Standard guide for quality planning and field implementation of a water quality measurement program. ASTM D5673-05 – Standard test method for elements in water by inductively coupled plasma-mass spectrometry. ASTM D5739-06 – Standard practice for oil spill source identification by gas chromatography and positive ion electron impact low resolution mass spectrometry. ASTM D5788-95(2005) – Standard guide for spiking organics into aqueous samples. ASTM D5790-95(2006) – Standard test method for measurement of purgeable organic compounds in water by capillary column gas chromatography/mass spectrometry. ASTM D5810-96(2006) – Standard guide for spiking into aqueous samples. ASTM D5811-08 – Standard test method for strontium-90 in water. ASTM D5812-96(2002)e1 – Standard test method for determination of organochlorine pesticides in water by capillary column gas chromatography. ASTM D5847-02(2007) – Standard practice for writing quality control specifications for standard test methods for water analysis. ASTM D5851-95(2006) – Standard guide for planning and implementing a water monitoring program. ASTM D5904-02 – Standard test method for total carbon, inorganic carbon, and organic carbon in water by ultraviolet, persulfate oxidation, and membrane conductivity detection. ASTM D5905-98(2008) – Standard practice for the preparation of substitute wastewater. ASTM D5907-09 – Standard test method for filterable and nonfilterable matter in water. ASTM D5916-96(2002) – Standard test method for detection and enumeration of Clostridium perfringens from water and extracted sediments by membrane filtration (MF). ASTM D5996-05(2009) – Standard test method for measuring anionic contaminants in high-purity water by on-line ion chromatography. ASTM D5997-96(2005) – Standard test method for on-line monitoring of total carbon, inorganic carbon in water by ultraviolet, persulfate oxidation, and membrane conductivity detection. ASTM D6071-06 – Standard test method for low-level sodium in high-purity water by graphite furnace atomic absorption spectroscopy.

ASTM D6091-07 – Standard practice for 99%/95% interlaboratory detection estimate (IDE) for analytical methods with negligible calibration error. ASTM D6104-97(2003) – Standard practice for determining the performance of oil/water separators subjected to surface runoff. ASTM D6145-97(2007) – Standard guide for monitoring sediment in watersheds. ASTM D6146-97(2007) – Standard guide for monitoring aqueous nutrients in watersheds. ASTM D6157-97(2003) – Standard practice for determining the performance of oil/water separators subjected to a sudden release. ASTM D6238-98(2003) – Standard test method for total oxygen demand in water. ASTM D6239-09 – Standard test method for uranium in drinking water by high-resolution alpha-liquid-scintillation spectrometry. ASTM D6301-08 – Standard practice for collection of on-line composite samples of suspended solids and ionic solids in process water. ASTM D6317–98(2004) – Standard test method for low level determination of total carbon, inorganic carbon and organic carbon in water by ultraviolet, persulfate oxidation, and membrane conductivity detection. ASTM D6362-98(2008) – Standard practice for certificates of reference materials for water analysis. ASTM D6501-09 – Standard test method for phosphonate in brines. ASTM D6502-08 – Standard test method for continuous measurement of on-line composite samples of low level filterable matter (suspended solids) and non-filterable matter (ionic solids) in process water by X-ray fluorescence (XRF). ASTM D6503-99(2009) – Standard test method for enterococci in water using Enterolert. ASTM D6504-07 – Standard practice for on-line determination of cation conductivity in high-purity water. ASTM D6508-00(2005)e2 – Standard test method for determination of dissolved inorganic anions in aqueous matrices using capillary ion electrophoresis and chromate electrolyte. ASTM D6512-07 – Standard practice for interlaboratory quantitation estimate. ASTM D6520-06 – Standard practice for the solid-phase micro-extraction (SPME) of water and its headspace for the analysis of volatile and semivolatile organic compounds. ASTM D6530-00(2006) – Standard test method for total active biomass in cooling tower waters (kool kount assay; KKA). ASTM D6568-00(2006) – Standard guide for planning, carrying out, and reporting traceable chemical analyses of water samples. ASTM D6569-05(2009) – Standard test method for on-line measurement of pH. ASTM D6581-08 – Standard test methods for bromate, bromide, chlorate, and chlorite in drinking water by suppressed ion chromatography. ASTM D6592-01 – Standard test method for portable chemiluminescent water quality determination.

Standardized Methods for Water-Quality Assessment

ASTM D6689-01(2006) – Standard guide for optimizing, controlling, and reporting test method uncertainties from multiple workstations in the same laboratory organization. ASTM D6696-05e1 – Standard guide for understanding cyanide species. ASTM D6697-01 – Standard test method for determination for chemical oxygen demand (manganese III oxygen demand) of water. ASTM D6698-07 – Standard test method for on-line measurement of turbidity below 5 NTU in water. ASTM D6734-01(2009) – Standard test method for low levels of coliphages in water. ASTM D6764-02(2007) – Standard guide for collection of water temperature, dissolved-oxygen concentrations, specific electrical conductance, and pH data from open channels. ASTM D6800-02(2007)e1 – Standard practice for preparation of water samples using reductive precipitation preconcentration technique for ICP-MS analysis of trace metals. ASTM D6808-02(2007) – Standard practice for competency requirements of reference material producers for water analysis. ASTM D6850-03(2008) – Standard guide for QC of screening methods in water. ASTM D6855-03 – Standard test method for determination of turbidity below 5 NTU in static mode. ASTM D6888-04 – Standard test method for available cyanide with ligand displacement and flow injection analysis (FIA) utilizing gas diffusion separation and amperometric detection. ASTM D6889-03 – Standard practice for fast screening for volatile organic compounds in water using solid phase microextraction (SPME). ASTM D6919-09 – Standard test method for determination of dissolved alkali and alkaline earth cations and ammonium in water and wastewater by ion chromatography. ASTM D6994-04 – Standard test method for determination of metal cyanide complexes in wastewater, surface water, groundwater, and drinking water using anion exchange chromatography with UV detection. ASTM D7065-06 – Standard test method for determination of nonylphenol, bisphenol A, p-tert-octylphenol, nonylphenol monoethoxylate and nonylphenol diethoxylate in environmental waters by gas chromatography mass spectrometry. ASTM D7066-04e1 – Standard test method for dimer/trimer of chlorotrifluoroethylene (S-316) recoverable oil and grease and nonpolar material by infrared determination. ASTM D7126-06 – Standard test method for on-line colorimetric measurement of silica. ASTM D7168-05e1 – Standard test method for 99Tc in water by solid phase extraction disk. ASTM D7237-06 – Standard test method for aquatic free cyanide with flow injection analysis (FIA) utilizing gas diffusion separation and amperometric detection. ASTM D7282-06 – Standard practice for set-up, calibration, and quality control of instruments used for radioactivity measurements.

293

ASTM D7283-06 – Standard test method for alpha- and betaactivity in water by liquid scintillation counting. ASTM D7284-08 – Standard test method for total cyanide in water by micro distillation followed by flow injection analysis with gas diffusion separation and amperometric detection. ASTM D7315-07a – Standard test method for determination of turbidity above 1 turbidity unit (TU) in static mode. ASTM D7316-06 – Standard guide for interpretation of existing field instrumentation to influence emergency response decisions. ASTM D7362-07 – Standard guide for rapid screening of vegetation for radioactive strontium aerial deposition. ASTM D7363-07 – Standard test method for determination of parent and alkyl polycyclic aromatics in sediment pore water using solid-phase microextraction and gas chromatography/mass spectrometry in selected ion monitoring mode. ASTM D7365-09 – Standard practice for sampling, preservation and mitigating interferences in water samples for analysis of cyanide. ASTM D7366-08 – Standard practice for estimation of measurement uncertainty for data from regression-based methods. ASTM D7485-09 – Standard test method for determination of nonylphenol, p-tert-octylphenol, nonylphenol monoethoxylate and nonylphenol diethoxylate in environmental waters by liquid chromatography/tandem mass spectrometry. ASTM D7511-09 – Standard test method for total cyanide by segmented flow injection analysis, in-line ultraviolet digestion and amperometric detection. ASTM D7535-09 – Standard test method for lead-210 in water. ASTM D7572-09 – Standard guide for recovery of aqueous cyanides by extraction from mine rock and soil after remediation of process releases. ASTM D7573-09 – Standard test method for total carbon and organic carbon in water by high-temperature catalytic combustion and infrared detection. ASTM D7574-09 – Standard test method for determination of bisphenol A in environmental waters by liquid chromatography/tandem mass spectrometry. ASTM D7575-10 – Standard test method for solvent-free membrane recoverable oil and grease by infrared determination. ASTM D7597-09 – Standard test method for determination of diisopropyl methylphosphonate, ethyl hydrogen dimethylamidophosphate, ethyl methylphosphonic acid, isopropyl methylphosphonic acid, methylphosphonic acid and pinacolyl methylphosphonic acid in water by liquid chromatography. ASTM D7598-09 – Standard test method for determination of thiodiglycol in water by single reaction monitoring liquid chromatography/tandem mass spectrometry. ASTM D7599-09 – Standard test method for determination of diethanolamine, triethanolamine, N-methyldiethanolamine and N-ethyldiethanolamine in water by single reaction monitoring liquid chromatography/tandem mass spectrometry.

294

Standardized Methods for Water-Quality Assessment

ASTM D7600-09 – Standard test method for determination of aldicarb, carbofuran, oxamyl and methomyl by liquid chromatography/tandem mass spectrometry. ASTM F660-83(2007) – Standard practice for comparing particle size in the use of alternative types of particle counters. ASTM F838-05 – Standard test method for determining bacterial retention of membrane filters utilized for liquid filtration. DIN 820-1:2009 – Normungsarbeit; Grundsa¨tze. DIN 8202:2009 – Normungsarbeit – Teil 2: Gestaltung von Dokumenten. DIN 820-34:2009 – Normungsarbeit – Teil 3 – Begriffe. DIN 820-4:2009 – Normungsarbeit – Teil 4: Gescha¨ftsgang. DIN 38406-16:1990 – Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung – Kationen (Gruppe E) – Teil 16: Bestimmung von 7 Metallen (Zink, Cadmium, Blei, Kupfer, Thallium, Nickel, Cobalt) mittels Voltammetrie (E 16) (German standard methods for the examination of water, wastewater, and sludge – cations (group E) – part 16: determination of 7 metals (zinc, cadmium, lead, copper, thallium, nickel, cobalt) by voltammetry (E 16)). DIN 38406-17:2009 – Deutsche Einheitsverfahren zur Wasser, Abwasser- und Schlammuntersuchung – Kationen (Gruppe E) – Teil 17: Bestimmung von Uran – Verfahren mittels adsorptiver Stripping-Voltammetrie in Grund-, Roh- und Trinkwa¨ssern (E 17) (German standard methods for the examination of water, wastewater, and sludge – cations (group E) – part 17: determination of uranium – method using adsorptive stripping voltammetry in surface water, raw water and drinking water (E 17)). DIN 38407-30:2007 – Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung – Gemeinsam erfassbare Stoffgruppen (Gruppe F) – Teil 30: Bestimmung von Trihalogenmethanen (THM) in Schwimm- und Badebeckenwasser mit Headspace-Gaschromatographie (F 30) (German standard methods for the examination of water, wastewater and sludge – jointly determinable substances (group F) – part 30: determination of trihalogenmethanes in bathing water and pool water with headspace-gas chromatography (F 30)). DIN 38410-1:2004 – Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung – Verfahren der biologisch-o¨kologischen Untersuchung (Gruppe M) – Teil 1: Bestimmung des Saprobienindex in FlieXgewa¨ssern (M 1) (German standard methods for the examination of water, wastewater, and sludge – biological–ecological analysis of water (group M) – part 1: determination of the saprobic index in running waters (M 1)). EN 872:2005 – Water quality – determination of suspended solids – method by filtration through glass fiber filters. EN 903:1993 – Water quality – determination of anionic surfactants by measurement of the methylene blue index MBAS (ISO 7875-1:1984 modified). EN 1085:2007 – Wastewater treatment – vocabulary; trilingual version EN 1085:2007. EN 1233:1996 – Water quality – determination of chromium – atomic absorption spectrometric methods. EN 1483:2007 – Water quality – determination of mercury – method using atomic absorption spectrometry.

EN 1484:1997 – Water quality – guidelines for the determination of total organic carbon (TOC) and dissolved organic carbon (DOC). EN 1622:2006 – Water quality – determination of the threshold odor number (TON) and threshold flavor number (TFN). EN 1899-1:1998 – Water quality – determination of biochemical oxygen demand after n days (BODn) – part 1: dilution and seeding method with allylthiourea addition (ISO 5815:1989, modified). EN 1899-2:1998 – Water quality – determination of biochemical oxygen demand after n days (BODn) – part 2: method for undiluted samples (ISO 5815:1989, modified). EN 12260:2003 – Water quality – determination of nitrogen – determination of bound nitrogen (TNb), following oxidation to nitrogen oxides. EN 12338:1998 – Water quality – determination of mercury – enrichment methods by amalgamation. EN 12673:1998 – Water quality – gas chromatographic determination of some selected chlorophenols in water. EN 12918:1999 – Water quality – determination of parathion, parathion-methyl, and some other organophosphorus compounds in water by dichloromethane extraction and gas chromatographic analysis. EN 13946:2003 – Water quality – guidance standard for the routine sampling and pretreatment of benthic diatoms from rivers. EN 14011:2003 – Water quality – sampling of fish with electricity. EN 14184:2003 – Water quality – guidance standard for the surveying of aquatic macrophytes in running waters. EN 14207:2003 – Water quality – determination of epichlorohydrin. EN 14407:2004 – Water quality – guidance standard for the identification, enumeration, and interpretation of benthic diatom samples from running waters. EN 14486:2005 – Water quality – detection of human enteroviruses by monolayer plaque assay. EN 14614:2004 – Water quality – guidance standard for assessing the hydromorphological features of rivers. EN 14757:2005 – Water quality – sampling of fish with multimesh gillnets. EN 14962:2006 – Water quality – guidance on the scope and selection of fish sampling methods. EN 14996:2006 – Water quality – guidance on assuring the quality of biological and ecological assessments in the aquatic environment. EN 15110:2006 – Water quality – guidance standard for the sampling of zooplankton from standing waters. EN 15196:2006 – Water quality – guidance on sampling and processing of the pupal exuviae of Chironomidae (order Diptera) for ecological assessment. EN 15204:2006 – Water quality – guidance standard on the enumeration of phytoplankton using inverted microscopy (Utermo¨hl technique). EN 15460:2007 – Water quality – guidance standard for the surveying of macrophytes in lakes. EN 15708:2009 – Water quality – guidance standard for the surveying, sampling, and laboratory analysis of phytobenthos in shallow running water.

Standardized Methods for Water-Quality Assessment

EN 15843:2010 – Water quality – guidance standard on determining the degree of modification of river hydromorphology. EN 25663:1993 – Water quality – determination of Kjeldahl nitrogen – method after mineralization with selenium (ISO 5663:1984). EN 25813:1992 – Water quality – determination of dissolved oxygen – iodometric method (ISO 5813:1983). EN 25814:1992 – Water quality – determination of dissolved oxygen – electrochemical probe method (ISO 5814:1990). EN 26461-1:1993 – Water quality – detection and enumeration of the spores of sulfite-reducing anaerobes (clostridia) – part 1: method by enrichment in a liquid medium (ISO 6461-1:1986). EN 26461-2:1993 – Water quality – detection and enumeration of the spores of sulfite-reducing anaerobes (clostridia) – part 2: method by membrane filtration (ISO 64612:1986). EN 26595:1992 – Water quality – determination of total arsenic – silver diethyldithiocarbamate spectrophotometric method (ISO 6595:1982). EN 26595:1992/AC:1992 – Water quality – determination of total arsenic – silver diethyldithiocarbamate spectrophotometric method (ISO 6595:1982). EN 26777:1993 – Water quality – determination of nitrite – molecular absorption spectrometric method (ISO 6777:1984). EN 27828:1994 – Water quality – methods of biological sampling – guidance on handnet sampling of aquatic benthic macro-invertebrates (ISO 7828:1985). EN 27888:1993 – Water quality – determination of electrical conductivity (ISO 7888:1985). EN 28265:1994 – Water quality – design and use of quantitative samplers for benthic macro-invertebrates on stony substrata in shallow freshwaters (ISO 8265:1988). EN ISO 5667-1:2006 – Water quality – sampling – part 1: guidance on the design of sampling programmes and sampling techniques (ISO 5667-1:2006). EN ISO 5667-1:2006/AC:2007 – Water quality – sampling – part 1: guidance on the design of sampling programmes and sampling techniques (ISO 5667-1:2006). EN ISO 5667-3:2003 – Water quality – sampling – part 3: guidance on the preservation and handling of water samples (ISO 5667-3:2003). EN ISO 5667-3:2003/AC:2007 – Water quality – sampling – part 3: guidance on the preservation and handling of water samples (ISO 5667-3:2003). EN ISO 5667-16:1998 – Water quality – sampling – part 16: guidance on biotesting of samples (ISO 566716:1998). EN ISO 5667-19:2004 – Water quality – sampling – part 19: guidance on sampling in marine sediments (ISO 566719:2004). EN ISO 5961:1995 – Water quality – determination of cadmium by atomic absorption spectrometry (ISO 5961:1994). EN ISO 6222:1999 – Water quality – enumeration of culturable microorganisms – colony count by inoculation in a nutrient agar culture medium (ISO 6222:1999).

295

EN ISO 6341:1996 – Water quality – determination of the inhibition of the mobility of Daphnia magna Straus (Cladocera, Crustacea) – acute toxicity test (ISO 6341:1996). EN ISO 6341:1996/AC:1998 – Water quality – determination of the inhibition of the mobility of Daphnia magna Straus (Cladocera, Crustacea) – acute toxicity test (ISO 6341:1996). EN ISO 6468:1996 – Water quality – determination of certain organochlorine insecticides, polychlorinated biphenyls and chlorobenzenes – gas chromatographic method after liquid–liquid extraction (ISO 6468:1996). EN ISO 6878:2004 – Water quality – determination of phosphorus – ammonium molybdate spectrometric method (ISO 6878:2004). EN ISO 7027:1999 – Water quality – determination of turbidity (ISO 7027:1999). EN ISO 7346-1:1997 – Water quality – determination of the acute lethal toxicity of substances to a freshwater fish (Brachydanio rerio Hamilton–Buchanan (Teleostei, Cyprinidae)) – part 1: static method (ISO 7346-1:1996). EN ISO 7346-2:1997 – Water quality – determination of the acute lethal toxicity of substances to a freshwater fish (Brachydanio rerio Hamilton–Buchanan (Teleostei, Cyprinidae)) – part 2: semi-static method (ISO 73462:1996). EN ISO 7346-3:1997 – Water quality – determination of the acute lethal toxicity of substances to a freshwater fish (Brachydanio rerio Hamilton–Buchanan (Teleostei, Cyprinidae)) – part 3: flow-through method (ISO 7346-3:1996). EN ISO 7393-1:2000 – Water quality – determination of free chlorine and total chlorine – part 1: titrimetric method using N,N-diethyl-1,4-phenylenediamine (ISO 73931:1985). EN ISO 7393-2:2000 – Water quality – determination of free chlorine and total chlorine – part 2: colorimetric method using N,N-diethyl-1,4-phenylenediamine, for routine control purposes (ISO 7393-2:1985). EN ISO 7393-3:2000 – Water quality – determination of free chlorine and total chlorine – part 3: iodometric titration method for the determination of total chlorine (ISO 73933:1990). EN ISO 7827:1995 – Water quality – evaluation in an aqueous medium of the ‘‘ultimate’’ aerobic biodegradability of organic compounds – method by analysis of dissolved organic carbon (DOC) (ISO 7827:1994). EN ISO 7887:1994 – Water quality – examination and determination of color (ISO 7887:1994). EN ISO 7899-1:1998 – Water quality – detection and enumeration of intestinal enterococci in surface and wastewater – part 1: miniaturized method (most probable number) by inoculation in liquid medium (ISO 78991:1998). EN ISO 7899-1:1998/AC:2000 – Water quality – detection and enumeration of intestinal enterococci in surface and wastewater – part 1: miniaturized method (most probable number) by inoculation in liquid medium (ISO 78991:1998). EN ISO 7899-2:2000 – Water quality – detection and enumeration of intestinal enterococci – part 2: membrane filtration method (ISO 7899-2:2000).

296

Standardized Methods for Water-Quality Assessment

EN ISO 7980:2000 – Water quality – determination of calcium and magnesium – atomic absorption spectrometric method (ISO 7980:1986). EN ISO 8192:2007 – Water quality – test for inhibition of oxygen consumption by activated sludge for carbonaceous and ammonium oxidation (ISO 8192:2007). EN ISO 8199:2007 – Water quality – general guidance on the enumeration of microorganisms by culture (ISO 8199:2005). EN ISO 8467:1995 – Water quality – determination of permanganate index (ISO 8467:1993). EN ISO 8689-1:2000 – Water quality – biological classification of rivers – part 1: guidance on the interpretation of biological quality data from surveys of benthic macroinvertebrates (ISO 8689-1:2000). EN ISO 8689-2:2000 – Water quality – biological classification of rivers – part 2: guidance on the presentation of biological quality data from surveys of benthic macroinvertebrates (ISO 8689-2:2000). EN ISO 8692:2004 – Water quality – freshwater algal growth inhibition test with unicellular green algae (ISO 8692:2004). EN ISO 9308-1:2000 – Water quality – detection and enumeration of Escherichia coli and coliform bacteria – part 1: membrane filtration method (ISO 9308-1:2000). EN ISO 9308-1:2000/AC:2008 – Water quality – detection and enumeration of Escherichia coli and coliform bacteria – part 1: membrane filtration method (ISO 9308-1:2000/Cor 1:2007). EN ISO 9308-3:1998 – Water quality – detection and enumeration of Escherichia coli and coliform bacteria in surface and wastewater – part 3: miniaturized method (most probable number) by inoculation in liquid medium (ISO 9308-3:1998). EN ISO 9308-3:1998/AC:2000 – Water quality – detection and enumeration of Escherichia coli and coliform bacteria in surface and wastewater – part 3: miniaturized method (most probable number) by inoculation in liquid medium (ISO 9308-3:1998). EN ISO 9377-2:2000 – Water quality – determination of hydrocarbon oil index – part 2: method using solvent extraction and gas chromatography (ISO 9377-2:2000). EN ISO 9391:1995 – Water quality – sampling in deep waters for macro-invertebrates – guidance on the use of colonization, qualitative and quantitative samplers (ISO 9391:1993). EN ISO 9408:1999 – Water quality – evaluation of ultimate aerobic biodegradability of organic compounds in aqueous medium by determination of oxygen demand in a closed respirometer (ISO 9408:1999). EN ISO 9439:2000 – Water quality – evaluation of ultimate aerobic biodegradability of organic compounds in aqueous medium – carbon dioxide evolution test (ISO 9439:1999). EN ISO 9509:2006 – Water quality – toxicity test for assessing the inhibition of nitrification of activated sludge microorganisms (ISO 9509:2006). EN ISO 9562:2004 – Water quality – determination of adsorbable organically bound halogens (AOX) (ISO 9562:2004).

EN ISO 9887:1994 – Water quality – evaluation of the aerobic biodegradability of organic compounds in an aqueous medium – semicontinuous activated sludge method (SCAS) (ISO 9887:1992). EN ISO 9888:1999 – Water quality – evaluation of ultimate aerobic biodegradability of organic compounds in aqueous medium – static test (Zahn–Wellens method) (ISO 9888:1999). EN ISO 9963-1:1995 – Water quality – determination of alkalinity – part 1: determination of total and composite alkalinity (ISO 9963-1:1994). EN ISO 9963-2:1995 – Water quality – determination of alkalinity – part 2: determination of carbonate alkalinity (ISO 9963-2:1994). EN ISO 10253:2006 – Water quality – marine algal growth inhibition test with Skeletonema costatum and Phaeodactylum tricornutum (ISO 10253:2006). EN ISO 10301:1997 – Water quality – determination of highly volatile halogenated hydrocarbons – gas-chromatographic methods (ISO 10301:1997). EN ISO 10304-1:2009 – Water quality – determination of dissolved anions by liquid chromatography of ions – part 1: determination of bromide, chloride, fluoride, nitrate, nitrite, phosphate, and sulfate (ISO 10304-1:2007). EN ISO 10304-3:1997 – Water quality – determination of dissolved anions by liquid chromatography of ions – part 3: determination of chromate, iodide, sulfite, thiocyanate, and thiosulfate (ISO 10304-3:1997). EN ISO 10304-4:1999 – Water quality – determination of dissolved anions by liquid chromatography of ions – part 4: determination of chlorate, chloride, and chlorite in water with low contamination (ISO 10304-4:1997). EN ISO 10634:1995 – Water quality – guidance for the preparation and treatment of poorly water-soluble organic compounds for the subsequent evaluation of their biodegradability in an aqueous medium (ISO 10634:1995). EN ISO 10695:2000 – Water quality – determination of selected organic nitrogen and phosphorus compounds – gas chromatographic methods (ISO 10695:2000). EN ISO 10705-1:2001 – Water quality – detection and enumeration of bacteriophages – part 1: enumeration of Fspecific RNA bacteriophages (ISO 10705-1:1995). EN ISO 10705-2:2001 – Water quality – detection and enumeration of bacteriophages – part 2: enumeration of somatic coliphages (ISO 10705-2:2000). EN ISO 10707:1997 – Water quality – evaluation in an aqueous medium of the ‘‘ultimate’’ aerobic biodegradability of organic compounds – method by analysis of biochemical oxygen demand (closed bottle test) (ISO 10707:1994). EN ISO 10712:1995 – Water quality – Pseudomonas putida growth inhibition test (pseudomonas cell multiplication inhibition test) (ISO 10712:1995). EN ISO 11348-1:2008 – Water quality – determination of the inhibitory effect of water samples on the light emission of Vibrio fischeri (Luminescent bacteria test) – part 1: method using freshly prepared bacteria (ISO 11348-1:2007). EN ISO 11348-2:2008 – Water quality – determination of the inhibitory effect of water samples on the light emission of Vibrio fischeri (luminescent bacteria test) – part 2: method using liquid-dried bacteria (ISO 11348-2:2007).

Standardized Methods for Water-Quality Assessment

EN ISO 11348-3:2008 – Water quality – determination of the inhibitory effect of water samples on the light emission of Vibrio fischeri (luminescent bacteria test) – part 3: method using freeze-dried bacteria (ISO 11348-3:2007). EN ISO 11369:1997 – Water quality – determination of selected plant treatment agents – method using high-performance liquid chromatography with UV detection after solid–liquid extraction (ISO 11369:1997). EN ISO 11731-2:2008 – Water quality – detection and enumeration of Legionella – part 2: direct membrane filtration method for waters with low bacterial counts (ISO 117312:2004). EN ISO 11732:2005 – Water quality – determination of ammonium nitrogen – method by flow analysis (CFA and FIA) and spectrometric detection (ISO 11732:2005). EN ISO 11733:2004 – Water quality – determination of the elimination and biodegradability of organic compounds in an aqueous medium – activated sludge simulation test (ISO 11733:2004). EN ISO 11734:1998 – Water quality – evaluation of the ‘‘ultimate’’ anaerobic biodegradability of organic compounds in digested sludge – method by measurement of the biogas production (ISO 11734:1995). EN ISO 11885:2009 – Water quality – determination of 33 elements by inductively coupled plasma atomic emission spectroscopy (ISO 11885:2007). EN ISO 11905-1:1998 – Water quality – determination of nitrogen – part 1: method using oxidative digestion with peroxodisulfate (ISO 11905-1:1997). EN ISO 11969:1996 – Water quality – determination of arsenic – atomic absorption spectrometric method (hydride technique) (ISO 11969:1996). EN ISO 12020:2000 – Water quality – determination of aluminum – atomic absorption spectrometric methods (ISO 12020:1997). EN ISO 13395:1996 – Water quality – determination of nitrite nitrogen and nitrate nitrogen and the sum of both by flow analysis (CFA and FIA) and spectrometric detection (ISO 13395:1996). EN ISO 14402:1999 – Water quality – determination of phenol index by flow analysis (FIA and CFA) (ISO 14402:1999). EN ISO 14403:2002 – Water quality – determination of total cyanide and free cyanide by continuous flow analysis (ISO 14403:2002). EN ISO 14593:2005 – Water quality – evaluation of ultimate aerobic biodegradability of organic compounds in aqueous medium – method by analysis of inorganic carbon in sealed vessels (CO2 headspace test) (ISO 14593:1999). EN ISO 14911:1999 – Water quality – determination of disþ 2þ 2þ 2þ 2þ solved Liþ, Naþ, NHþ 4 , K , Mn , Ca , Mg , Sr , and 2þ Ba using ion chromatography – method for water and wastewater (ISO 14911:1998). EN ISO 15061:2001 – Water quality – determination of dissolved bromate – method by liquid chromatography of ions (ISO 15061:2001). EN ISO 15088:2008 – Water quality – determination of the acute toxicity of wastewater to zebrafish eggs (Danio rerio) (ISO 15088:2007).

297

CEN ISO/TR 15462:2009 – Water quality – selection of tests for biodegradability (ISO/TR 15462:2006). EN ISO 15586:2003 – Water quality – determination of trace elements using atomic absorption spectrometry with graphite furnace (ISO 15586:2003). EN ISO 15587-1:2002 – Water quality – digestion for the determination of selected elements in water – part 1: Aqua regia digestion (ISO 15587-1:2002). EN ISO 15587-2:2002 – Water quality – digestion for the determination of selected elements in water – part 2: nitric acid digestion (ISO 15587-2:2002). EN ISO 15680:2003 – Water quality – gas-chromatographic determination of a number of monocyclic aromatic hydrocarbons, naphthalene, and several chlorinated compounds using purge-and-trap and thermal desorption (ISO 15680:2003). EN ISO 15681-1:2004 – Water quality – determination of orthophosphate and total phosphorus contents by flow analysis (FIA and CFA) – part 1: method by flow injection analysis (FIA) (ISO 15681-1:2003). EN ISO 15681-2:2004 – Water quality – determination of orthophosphate and total phosphorus contents by flow analysis (FIA and CFA) – part 2: method by continuous flow analysis (CFA) (ISO 15681-2:2003). EN ISO 15682:2001 – Water quality – determination of chloride by flow analysis (CFA and FIA) and photometric or potentiometric detection (ISO 15682:2000). EN ISO 15839:2006 – Water quality – on-line sensors/analyzing equipment for water – specifications and performance tests (ISO 15839:2003). EN ISO 15913:2003 – Water quality – determination of selected phenoxyalkanoic herbicides, including bentazones and hydroxybenzonitriles by gas chromatography and mass spectrometry after solid phase extraction and derivatization (ISO 15913:2000). EN ISO 16264:2004 – Water quality – determination of soluble silicates by flow analysis (FIA and CFA) and photometric detection (ISO 16264:2002). EN ISO 16266:2008 – Water quality – detection and enumeration of Pseudomonas aeruginosa – method by membrane filtration (ISO 16266:2006). EN ISO 16588:2003 – Water quality – determination of six complexing agents – gas-chromatographic method (ISO 16588:2002). EN ISO 16588:2003/A1:2005 – Water quality – determination of six complexing agents – gas-chromatographic method (ISO 16588:2002/Amd 1:2004). EN ISO 16665:2005 – Water quality – guidelines for quantitative sampling and sample processing of marine softbottom macrofauna (ISO 16665:2005). EN ISO 16712:2006 – Water quality – determination of acute toxicity of marine or estuarine sediment to amphipods (ISO 16712:2005). EN ISO 17294-1:2006 – Water quality – application of inductively coupled plasma mass spectrometry (ICP-MS) – part 1: general guidelines (ISO 17294-1:2004). EN ISO 17294-2:2004 – Water quality – application of inductively coupled plasma mass spectrometry (ICP-MS) – part 2: determination of 62 elements (ISO 17294-2:2003).

298

Standardized Methods for Water-Quality Assessment

EN ISO 17353:2005 – Water quality – determination of selected organotin compounds – gas chromatographic method (ISO 17353:2004). EN ISO 17495:2003 – Water quality – determination of selected nitrophenols – method by solid-phase extraction and gas chromatography with mass spectrometric detection (ISO 17495:2001). EN ISO 17852:2008 – Water quality – determination of mercury – method using atomic fluorescence spectrometry (ISO 17852:2006). EN ISO 17993:2003 – Water quality – determination of 15 polycyclic aromatic hydrocarbons (PAH) in water by HPLC with fluorescence detection after liquid–liquid extraction (ISO 17993:2002). EN ISO 17994:2004 – Water quality – criteria for establishing equivalence between microbiological methods (ISO 17994:2004). EN ISO 18412:2006 – Water quality – determination of chromium(VI) – photometric method for weakly contaminated water (ISO 18412:2005). EN ISO 18856:2005 – Water quality – determination of selected phthalates using gas chromatography/mass spectrometry (ISO 18856:2004). EN ISO 18857-1:2006 – Water quality – determination of selected alkylphenols – part 1: method for non-filtered samples using liquid–liquid extraction and gas chromatography with mass selective detection (ISO 188571:2005). EN ISO 19458:2006 – Water quality – sampling for microbiological analysis (ISO 19458:2006). EN ISO 19493:2007 – Water quality – guidance on marine biological surveys of hard-substrate communities (ISO 19493:2007). EN ISO 20079:2006 – Water quality – determination of the toxic effect of water constituents and wastewater on duckweed (Lemna minor) – duckweed growth inhibition test (ISO 20079:2005). EN ISO 21427-2:2009 – Water quality – evaluation of genotoxicity by measurement of the induction of micronuclei – part 2: mixed population method using the cell line V79 (ISO 21427-2:2006). EN ISO 22032:2009 – Water quality – determination of selected polybrominated diphenylethers in sediment and sewage sludge – method using extraction and gas chromatography/mass spectrometry (ISO 22032:2006). EN ISO 22478:2006 – Water quality – determination of certain explosives and related compounds – method using high-performance liquid chromatography (HPLC) with UV detection (ISO 22478:2006). EN ISO 23631:2006 – Water quality – determination of dalapon, trichloroacetic acid, and selected haloacetic acids – method using gas chromatography (GC-ECD and/or GCMS detection) after liquid–liquid extraction and derivatization (ISO 23631:2006). EN ISO 23631:2006/AC:2007 – Water quality – determination of dalapon, trichloroacetic acid, and selected haloacetic acids – method using gas chromatography (GC-ECD and/ or GC-MS detection) after liquid–liquid extraction and derivatization (ISO 23631:2006).

EN ISO 23913:2009 – Water quality – determination of chromium(VI) – method using flow analysis (FIA and CFA) and spectrometric detection (ISO 23913:2006). ENV ISO 13530:1998 – Water quality – guide to analytical quality control for water analysis (ISO/TR 13530:1997). ENV ISO 13843:2001 – Water quality – guidance on validation of microbiological methods (ISO/TR 13843:2000). ISO 78-2:1999 – Chemistry layouts for standards – part 2: methods of chemical analysis. ISO 3696:1987 – Water for analytical laboratory use; specification and test methods. ISO 5663:1984 – Water quality – determination of Kjeldahl nitrogen – method after mineralization with selenium. ISO 5664:1984 – Water quality – determination of ammonium – distillation and titration method. ISO 5666:1999 – Water quality – determination of mercury. ISO 5667-4:1987 – Water quality – sampling – part 4: guidance on sampling from lakes, natural and man-made. ISO 5667-5:2006 – Water quality – sampling – part 5: guidance on sampling of drinking water from treatment works and piped distribution systems. ISO 5667-6:2005 – Water quality – sampling – part 6: guidance on sampling of rivers and streams. ISO 5667-7:1993 – Water quality – sampling – part 7: guidance on sampling of water and steam in boiler plants. ISO 5667-8:1993 – Water quality – sampling – part 8: guidance on the sampling of wet deposition. ISO 5667-9:1992 – Water quality – sampling – part 9: guidance on sampling from marine waters. ISO 5667-10:1992 – Water quality – sampling – part 10: guidance on sampling of wastewaters. ISO 5667-11:2009 – Water quality – sampling – part 11: guidance on sampling of groundwaters. ISO 5667-12:1995 – Water quality – sampling – part 12: guidance on sampling of bottom sediments. ISO 5667-13:2009 – Water quality – sampling – part 13: guidance on sampling of sludges from sewage and water treatment works. ISO 5667-14:1998 – Water quality – sampling – part 14: guidance on quality assurance of environmental water sampling and handling. ISO 5667-15:2009 – Water quality – sampling – part 15: guidance on preservation and handling of sludge and sediment samples. ISO 5667-17:2008 – Water quality – sampling – part 17: guidance on sampling of bulk suspended solids. ISO 5667-18:2001 – Water quality – sampling – part 18: guidance on sampling of groundwater at contaminated sites. ISO 5667-18:2001/Cor 1:2008. ISO 5667-20:2008 – Water quality – sampling – part 20: guidance on the use of sampling data for decision making – compliance with thresholds and classification systems. ISO 5725-2:1994 including Technical Corrigendum 1:2002 – accuracy (trueness and precision) of measurement methods and results – part 2: basic method for the determination of repeatability and reproducibility of a standards measurement method. ISO 5813:1983 – Water quality – determination of dissolved oxygen – iodometric method.

Standardized Methods for Water-Quality Assessment

ISO 5814:1990 – Water quality – determination of dissolved oxygen – electrochemical probe method. ISO 5815-1:2003 – Water quality – determination of biochemical oxygen demand after n days (BODn) – part 1: dilution and seeding method with allylthiourea addition. ISO 5815-2:2003 – Water quality – determination of biochemical oxygen demand after n days (BODn) – part 2: method for undiluted samples. ISO 6058:1984 – Water quality – determination of calcium content – EDTA titrimetric method. ISO 6059:1984 – Water quality – determination of the sum of calcium and magnesium – EDTA titrimetric method. ISO 6060:1989 – Water quality – determination of the chemical oxygen demand. ISO 6107-1:2004, ISO 6107-2:2006, ISO 6107-3:1993, ISO 6107-4:1993, ISO 6107-5:2004, ISO 6107-6:2004, ISO 61077:2006, ISO 6107-8:1993 – Water quality – vocabulary. ISO 6107-2:2006/DAmd 1. ISO 6107-3:1993/Amd 1:2001. ISO 6107-8:1993/Amd 1:2001. ISO 6107-9:1997 – Water quality – vocabulary – part 9: alphabetical list and subject index. ISO 6332:1988 – Water quality – determination of iron – spectrometric method using 1,10-phenanthroline. ISO 6333:1986 – Water quality – determination of manganese – formaldoxime spectrometric method. ISO 6340:1995 – Water quality – detection and enumeration of Salmonella. ISO 6439:1990 – Water quality – determination of phenol index – 4-aminoantipyrine spectrometric methods after distillation. ISO 6461-1:1986 – Water quality – detection and enumeration of the spores of sulfite-reducing anaerobes (clostridia) – part 1: method by enrichment in a liquid medium. ISO 6461-2:1986 – Water quality – detection and enumeration of the spores of sulfite-reducing anaerobes (clostridia) – part 2: method by membrane filtration. ISO 6703-1:1984 – Water quality – determination of cyanide – part 1: determination of total cyanide. ISO 6703-2:1984 – Water quality – determination of cyanide – part 2: determination of easily liberatable cyanide. ISO 6703-3:1984 – Water quality – determination of cyanide – part 3: determination of cyanogen chloride. ISO 6777:1984 – Water quality – determination of nitrite – molecular absorption spectrometric method. ISO 6778:1984 – Water quality – determination of ammonium – potentiometric method. ISO 7150-1:1984 – Water quality – determination of ammonium – part 1: manual spectrometric method. ISO 7704:1985 – Water quality – evaluation of membrane filters used for microbiological analyses. ISO 7828:1985 – Water quality – methods of biological sampling – guidance on handnet sampling of aquatic benthic macro-invertebrates. ISO 7875-1:1996 – Water quality – determination of surfactants – part 1: determination of anionic surfactants by measurement of the methylene blue index (MBAS). ISO 7875-1:1996/Cor 1:2003

299

ISO 7875-2:1984 – Water quality – determination of surfactants – part 2: determination of non-ionic surfactants using Dragendorff reagent. ISO 7888:1985 – Water quality – determination of electrical conductivity. ISO 7890-3:1988 – Water quality – determination of nitrate – part 3: spectrometric method using sulfosalicylic acid. ISO 7981-1:2005 – Water quality – determination of polycyclic aromatic hydrocarbons (PAH) – part 1: determination of six PAH by high-performance thin-layer chromatography with fluorescence detection after liquid– liquid extraction. ISO 7981-2:2005 – Water quality – determination of polycyclic aromatic hydrocarbons (PAH) – part 2: determination of six PAH by high-performance liquid chromatography with fluorescence detection after liquid–liquid extraction. ISO 8165-1:1992 – Water quality – determination of selected monovalent phenols – part 1: gas-chromatographic method after enrichment by extraction. ISO 8165-2:1999 – Water quality – determination of selected monovalent phenols – part 2: method by derivatization and gas chromatography. ISO 8245:1999 – Water quality – guidelines for the determination of total organic carbon (TOC) and dissolved organic carbon (DOC). ISO 8265:1988 – Water quality – design and use of quantitative samplers for benthic macro-invertebrates on stony substrata in shallow freshwaters. ISO 8288:1986 – Water quality – determination of cobalt, nickel, copper, zinc, cadmium, and lead – flame atomic absorption spectrometric methods. ISO 8466-1:1990 – Water quality – calibration and evaluation of analytical methods and estimation of performance characteristics – part 1: statistical evaluation of the linear calibration function. ISO 8466-2:2001 – Water quality – calibration and evaluation of analytical methods and estimation of performance characteristics – part 2: calibration strategy for non-linear second-order calibration functions. ISO 9174:1998 – Water quality – determination of chromium – atomic absorption spectrometric methods. ISO 9297:1989 – Water quality – determination of chloride – silver nitrate titration with chromate indicator (Mohr’s method). ISO 9308-2:1990 – Water quality – detection and enumeration of coliform organisms, thermotolerant coliform organisms and presumptive Escherichia coli – part 2: multiple tube (most probable number) method. ISO 9377-2:2000 – Water quality – determination of hydrocarbon oil index – part 2: method using solvent extraction and gas chromatography. ISO 9308-3:1998/Cor 1:2000 ISO 9390:1990 – Water quality – determination of borate – spectrometric method using azomethine-H. ISO 9696:2007 – Water quality – measurement of gross alpha activity in nonsaline water – thick source method. ISO 9697:2008 – Water quality – measurement of gross beta activity in nonsaline water – thick source method.

300

Standardized Methods for Water-Quality Assessment

ISO 9698:1989 – Water quality – determination of tritium activity concentration – liquid scintillation counting method. ISO 9964-1:1993 – Water quality – determination of sodium and potassium – part 1: determination of sodium by atomic absorption spectrometry. ISO 9964-2:1993 – Water quality – determination of sodium and potassium – part 2: determination of potassium by atomic absorption spectrometry. ISO 9964-3:1993 – Water quality – determination of sodium and potassium – part 3: determination of sodium and potassium by flame emission spectrometry. ISO 9965:1993 – Water quality – determination of selenium – atomic absorption spectrometric method (hydride technique). ISO 9998:1991 – Water quality – practices for evaluating and controlling microbiological colony count media used in water quality tests. ISO 10229:1994 – Water quality – determination of the prolonged toxicity of substances to freshwater fish – method for evaluating the effects of substances on the growth rate of rainbow trout (Oncorhynchus mykiss Walbaum (Teleostei, Salmonidae)). ISO 10260:1992 – Water quality – measurement of biochemical parameters – spectrometric determination of the chlorophyll-a concentration. ISO 10359-1:1992 – Water quality – determination of fluoride – part 1: electrochemical probe method for potable and lightly polluted water. ISO 10359-2:1994 – Water quality – determination of fluoride – part 2: determination of inorganically bound total fluoride after digestion and distillation. ISO 10523:2008 – Water quality – determination of pH. ISO 10530:1992 – Water quality – determination of dissolved sulfide – photometric method using methylene blue. ISO 10566:1994 – Water quality – determination of aluminum – spectrometric method using pyrocatechol violet. ISO 10703:2007 – Water quality – determination of the activity concentration of radionuclides – method by high-resolution gamma-ray spectrometry. ISO 10705-3:2003 – Water quality – detection and enumeration of bacteriophages – part 3: validation of methods for concentration of bacteriophages from water. ISO 10705-4:2001 – Water quality – detection and enumeration of bacteriophages – part 4: enumeration of bacteriophages infecting Bacteroides fragilis. ISO 10706:2000 – Water quality – determination of long-term toxicity of substances to Daphnia magna Straus (Cladocera, Crustacea). ISO 10708:1997 – Water quality – evaluation in an aqueous medium of the ultimate aerobic biodegradability of organic compounds – determination of biochemical oxygen demand in a two-phase closed bottle test. ISO 11083:1994 – Water quality – determination of chromium(VI) – spectrometric method using 1,5diphenylcarbazide. ISO/TR 11044:2008 – Water quality – scientific and technical aspects of batch algae growth inhibition tests.

ISO/TS 11370:2000 – Water quality – determination of selected organic plant-treatment agents – automated multiple development (AMD) technique. ISO 11423-1:1997 – Water quality – determination of benzene and some derivatives – part 1: head-space gas chromatographic method. ISO 11423-2:1997 – Water quality – determination of benzene and some derivatives – part 2: method using extraction and gas chromatography. ISO 11731:1998 – Water quality – detection and enumeration of Legionella. ISO 11734:1995 – Water quality – evaluation of the ‘‘ultimate’’ anaerobic biodegradability of organic compounds in digested sludge – method by measurement of the biogas production. ISO 11923:1997 – Water quality – determination of suspended solids by filtration through glass-fiber filters. ISO 12890:1999 – Water quality – determination of toxicity to embryos and larvae of freshwater fish – semi-static method. ISO 13358:1997 – Water quality – determination of easily released sulfide. ISO 13528:2005 Statistical methods for use in proficiency testing by interlaboratory comparisons. ISO/TR 11905-2:1997 – Water quality – determination of nitrogen – Part 2: determination of bound nitrogen, after combustion and oxidation to nitrogen dioxide, chemiluminescence detection. ISO/TS 13530:2009 – Water quality – guidance on analytical quality control for chemical and physicochemical water analysis. ISO 13641-1:2003 – Water quality – determination of inhibition of gas production of anaerobic bacteria – part 1: general test. ISO 13641-2:2003 – Water quality – determination of inhibition of gas production of anaerobic bacteria – part 2: test for low biomass concentrations. ISO 13829:2000 – Water quality – determination of the genotoxicity of water and wastewater using the umu-test. ISO 14442:2006 – Water quality – guidelines for algal growth inhibition tests with poorly soluble materials, volatile compounds, metals, and wastewater. ISO 14592-1:2002 – Water quality – evaluation of the aerobic biodegradability of organic compounds at low concentrations – part 1: shake-flask batch test with surface water or surface water/sediment suspensions. ISO 14592-2:2002 – Water quality – evaluation of the aerobic biodegradability of organic compounds at low concentrations – part 2: continuous flow river model with attached biomass. ISO 14669:1999 – Water quality – determination of acute lethal toxicity to marine copepods (Copepoda, Crustacea). ISO 15089:2000 – Water quality – guidelines for selective immunoassays for the determination of plant treatment and pesticide agents. ISO/TR 15462:2006 – Water quality – selection of tests for biodegradability. ISO 15522:1999 – Water quality – determination of the inhibitory effect of water constituents on the growth of activated sludge microorganisms.

Standardized Methods for Water-Quality Assessment

ISO 15553:2006 – Water quality – isolation and identification of Cryptosporidium oocysts and Giardia cysts from water. ISO 15705:2002 – Water quality – determination of the chemical oxygen demand index (ST-COD) – small-scale sealed-tube method. ISO 16221:2001 – Water quality – guidance for determination of biodegradability in the marine environment. ISO 16240:2005 – Water quality – determination of the genotoxicity of water and wastewater – Salmonella/microsome test (Ames test). ISO 16265:2009 – Water quality – determination of the methylene blue active substances (MBAS) index – method using continuous flow analysis (CFA). ISO/TS 16489:2006 – Water quality – guidance for establishing the equivalency of results. ISO/TS 16489:2006/Cor 1:2006. ISO 16590:2000 – Water quality – determination of mercury – methods involving enrichment by amalgamation. ISO 17381:2003 – Water quality – selection and application of ready-to-use test kit methods in water analysis. ISO 17858:2007 – Water quality – determination of dioxinlike polychlorinated biphenyls – method using gas chromatography/mass spectrometry. ISO 17995:2005 – Water quality – detection and enumeration of thermotolerant Campylobacter species. ISO 18073:2004 – Water quality – determination of tetra- to octa-chlorinated dioxins and furans – method using isotope dilution HRGC/HRMS. ISO 18749:2004 – Water quality – adsorption of substances on activated sludge – batch test using specific analytical methods. ISO 20179:2005 – Water quality – determination of microcystins – method using solid-phase extraction (SPE) and high-performance liquid chromatography (HPLC) with ultraviolet (UV) detection. ISO/TS 20281:2006 – Water quality – guidance on statistical interpretation of ecotoxicity data. ISO/TS 20612:2007 – Water quality – interlaboratory comparisons for proficiency testing of analytical chemistry laboratories. ISO 20665:2008 – Water quality – determination of chronic toxicity to Ceriodaphnia dubia. ISO 20666:2008 – Water quality – determination of the chronic toxicity to Brachionus calyciflorus in 48 h. ISO 21427-1:2006 – Water quality – evaluation of genotoxicity by measurement of the induction of micronuclei – part 1: evaluation of genotoxicity using amphibian larvae. ISO 21458:2008 – Water quality – determination of glyphosate and AMPA – method using high-performance liquid chromatography (HPLC) and fluorometric detection. ISO 22719:2008 – Water quality – determination of total alkalinity in seawater using high-precision potentiometric titration. ISO 22743:2006 – Water quality – determination of sulfates – method by continuous flow analysis (CFA). ISO 22743:2006/Cor 1:2007. ISO 23893-1:2007 – Water quality – biochemical and physiological measurements on fish – part 1: sampling of fish, handling and preservation of samples.

301

ISO/TS 23893-2:2007 – Water quality – biochemical and physiological measurements on fish – part 2: determination of ethoxyresorufin-O-deethylase (EROD). ISO 24293:2009 – Water quality – determination of individual isomers of nonylphenol – method using solid-phase extraction (SPE) and gas chromatography/mass spectrometry (GC/MS). ISO 25101:2009 – Water quality – determination of perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) – method for unfiltered samples using solid phase extraction and liquid chromatography/mass spectrometry. ISO 80000-9:2009 – Quantities and units – part 9: physical chemistry and molecular physics. ISO/IEC 17025:2005 – General requirements for the competence of testing and calibration laboratories. ISO/DIS 7887:2009 – prEN ISO 7887:2009 – Water quality – examination and determination of color.

References 2000/60/EC (2000) WFD – Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy. Official Journal of the European Communities L 327: 1--73. 2008/105/EC (2008) Directive 2008/105/EC of European Parliament and Council on environmental quality standards in the field of water policy, amending and subsequently repealing Council Directives 82/176/EEC, 83/513/EEC, 84/156/EEC, 84/419/EEC, 86/280/EEC and amending Directive 2000/60/EC. Official Journal of the European Communities L 348: 84--97. 2009/90/EC (2009) Commission Directive 2009/90/EC of 31 July 2009 laying down, pursuant to Directive 2000/60/EC of the European Parliament and of the Council, technical specifications for chemical analysis and monitoring of water status. Official Journal of the European Communities L 201: 36--38. 98/83/EC (1998) Council Directive 98/83/EC of 3 November 1998 on the quality of water intended for human consumption. Official Journal of the European Communities L 330: 32--54. Abw AG (2005) Gesetz u¨ber Abgaben fu¨r das Einleiten von Abwasser in Gewa¨sser (Abwasserabgabengesetz – AbwAG) in der Fassung vom 18. Januar 2005 (Act on charges levied for discharging wastewaters into waters (wastewater charges act)). Bundesgesetzblatt Teil I Nr. 5: 114–119. Abw V (2002) Verordnung u¨ber Anforderungen an das Einleiten von Abwasser in Gewa¨sser (Abwasserverordnung – AbwV) in der Fassung vom 15. Oktober 2002 (Ordinance on requirements for the discharge of wastewaters into waters (wastewater ordinance)). Bundesgesetzblatt Teil I Nr. 74: 4047–4121. APHA (American Public Health Association), AWWA (American Water Works Association), and WEF (Water Environment Federation) (ed.) (2005) Standard Methods for the Examination of Water and Wastewater, 21st edn. Baltimore, MD: Port City Press. ASTM (2007) ASTM Technical Committee Officer Handbook (‘‘Red Book’’). http:// www.astm.org/COMMIT/RedBook5.pdf (accessed April 2010). ASTM (2009a) Regulations governing ASTM Technical Committees (‘‘Green Book’’). http://www.astm.org/COMMIT/Regs.pdf (accessed April 2010). ASTM (2009b) Form and Style for ASTM Standards (‘‘Blue Book’’). http:// www.astm.org/COMMIT/Blue_Book.pdf (accessed April 2010). Aurand K and Ru¨hle H (2003) Radioaktive Stoffe und die Trinkwasserverordnung. In: Grohmann A, Ha¨sselbarth U, and Schwerdtfeger W (eds.) Die Trinkwasserverordnung. Einfu¨hrung und Erla¨uterungen fu¨r Wasserversorgungsunternehmen und U¨berwachungsbeho¨rden, 4., neu bearbeitete Auflage, pp. 377–387. Berlin: Erich Schmidt Verlag. DEV (Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung) (1971) Verfahren L 12 – Assimilationszehrungstest. Weinheim: Wiley-VCH. Berlin: Beuth. Forte M, Bertolo A, D’Alberti F, et al. (2006) Standardized methods for measuring radionuclides in drinking water. Journal of Radioanalytical and Nuclear Chemistry 269: 397--401. Hongve D and A˚kesson G (1996) Spectrophotometric determination of water colour in Hazen units. Water Research 30: 2771--2775.

302

Standardized Methods for Water-Quality Assessment

IUPAC (International Union of Pure and Applied Chemistry) (2007) Quantities, Units and Symbols in Physical Chemistry, 3rd edn. Cambridge: Royal Society of Chemistry. ISO/IEC (2004) ISO/IEC Directives, Part 2 – Rules for the Structure and Drafting of International Standards, 5th edn. Geneva: ISO/IEC. ISO/IEC (2007) Guide 99 International Vocabulary of Metrology – Basic and General Concepts and Associated Terms (VIM). Geneva: ISO/IEC. Kaniansky D, Masa´r M, Mara´k J, and Bodor R (1999) Capillary electrophoresis of inorganic anions. Journal of Chromatography A 843: 133--178. Kroon H (1993) Determination of nitrogen in water: Comparison of a continuous flow method with on-line UV-digestion with the original Kjeldahl method. Analytica Chimica Acta 276: 287--293. OECD (Organisation for Economic Cooperation and Development) (2008) OECD guidelines for the testing of chemicals, pdf edn., ISSN 1607-310X. http:// oberon.sourceoecd.org (accessed April 2010). Pagga U (1997) Testing biodegradability with standardized methods. Chemosphere 35: 2953--2972. Pluta H-J and Rosenberg M (2005) German perspective. In: Thompson KC, Wahida K, and Loibner AP (eds.) Environmental Toxicity Testing, 1st edn., pp. 290--301. Boca Raton, FL: Blackwell and CRC Press. Quevauvillier P, Borchers U, and Gawlik BM (2007) Coordinating links among research, standardisation and policy in support of water framework directive chemical monitoring requirements. Journal of Environmental Monitoring 9: 915--923. REACH (2006) Regulation (EC) No 1907/2006 of the European Parliament and of the Council of 18 December 2006 concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency, amending Directive 1999/45/EC and repealing council regulation (EEC) No 793/93 and commission regulation (EC) No 1488/94 as well as Council Directive 76/769/EEC and Commission Directives 91/155/EEC, 93/67/ EEC, 93/105/EC and 2000/21/EC. Official Journal of the European Communities 396: 1--851. Reifferscheid G, Ziemann C, Fieblinger D, et al. (2008) Measurement of genotoxicity in wastewater samples with the in vitro micronucleus test – results of a round robin study in the context of standardisation according to ISO. Mutation Research 649: 15--27. Reuschenbach P, Pagga U, and Strotmann U (2003) A critical comparison of respirometric biodegradation tests based on OECD 301 and related test methods. Water Research 37: 1571--1582. Richardson SD (2003) Water analysis: Emerging contaminants and current issues. Analytical Chemistry 75: 2831--2857. Richardson SD (2005) Water analysis: Emerging contaminants and current issues. Analytical Chemistry 77: 3807--3838. Richardson SD (2007) Water analysis: Emerging contaminants and current issues. Analytical Chemistry 79: 4295--4324. Richardson SD (2009) Water analysis: Emerging contaminants and current issues. Analytical Chemistry 81: 4645--4677. Schmidt S (1992) Wasseranalytik – Normung von Verfahren (Water analysis – standardization of methods). GIT Fachzeitschrift fu¨r das Laboratorium 1992: 621--630. Schmidt S (2001) No water, no life – water quality in ISO. ISO Bulletin January 2001: 10--14. Schmidt S (2003) International standardization of water analysis – basis for comparative assessment of water quality. Environmental Science and Pollution Research 10: 183--187. Schmidt S and Wunder H (1988) International anerkannte Verfahren in der Wasseranalytik. Zeitschrift fu¨r Wasser- und Abwasserforschung 21: 118--122. Stottmeister E, Heemken OP, Hendel P, et al. (2009) Interlaboratory trial on the analysis of alkyphenols, alkylphenol ethoxylates, and bisphenol A in water samples according to ISO/CD 18857-2. Analytical Chemistry 81: 6765--6773. Strotmann U, Reuschenbach P, Schwarz H, et al. (2004) Development and evaluation of an online CO2 evolution test and a multicomponent biodegradation test system. Applied Environmental Microbiology 70: 4621--4628. Strub MP, Lepot B, and Morin A (2009) Metrological aspects of collaborative field trials, including coping with unexpected events. Trends in Analytical Chemistry 28: 245--261.

Thompson KC, Wahida K, and Loibner AP (eds.) (2005) Environmental Toxicity Testing. Boca Raton, FL: Blackwell and CRC Press. Tillmanns AR, Pick FR, and Aranda-Rodriguez R (2007) Sampling and analysis of microcystins: Implications for the development of standardized methods. Environmental Toxicology 22: 132--143. TrinkwV (2001) Verordnung u¨ber die Qualita¨t von Wasser fu¨r den menschlichen Gebrauch vom 21. Mai 2001 (Trinkwasserverordnung – TrinkwV 2001). Bundesgesetzblatt 2001 I Nr. 24 S. 959 ff. http://bundesrecht.juris.de/ trinkwv_2001/index.html (accessed April 2010). Vilela Junqueira M, Friedrich G, and Pereira de Araujo PR (2010) A saprobic index for biological assessment of river water quality in Brazil (Minas Gerais and Rio de Janeiro states). Environmental Monitoring and Assessment 163: 545--554. Wasserchemische Gesellschaft and Normenausschuss Wasserwesen im DIN (eds.) (2010) Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung. Herausgegeben von der Wasserchemischen Gesellschaft – Fachgruppe in der Gesellschaft Deutscher Chemiker in Gemeinschaft mit dem NormenausschuX Wasserwesen (NAW) im DIN Deutsches Institut fu¨r Normung e. V. 77. Lieferung 2010 (German standard methods for the examination of water, waste water and sludge. Edited by the Water Chemical Society – a Division of the German Chemical Society and the Standards Committee ‘‘Water Practice Standards’’ of the German Institute for Standardisation). Weinheim: Wiley-VCH; Berlin: Beuth. WHO (2006) Guidelines for Drinking-Water Quality. First Addendum to Third Edition. Volume 1: Recommendations. Geneva. http://www.who.int/ water_sanitation_health/dwq/gdwq0506.pdf (accessed April 2010).

Relevant Websites http://www.astm.org ASTM International; ASTM D19 and ASTM D19 Scope. ftp://ftp.cen.eu CEN; CEN Compass: The World of European Standards. http://www.naw.din.de DIN NA 119 Normenausschuss Wasserwesen (NAW) Startseite, DIN NA 119-01-03 AA. http://www.epa.gov Environmental Protection Agency (EPA): Test Method Collections. http://www.cen.eu European Committee for Standardization (CEN); Technical Committees, Workshops and other bodies; CEN/TC 230 Water analysis, CEN/TC 308 Characterization of sludges, and CEN/TC 400 Horizontal standards in the field of sludge, biowaste and soil. http://www.gdch.de Gesellschaft Deutscher Chemiker (GDCh), Liste vorhandener Validierungsdokumente. http://www.iso.org International Organization for Standardization; Standards Development, Technical Committees, List of ISO Technical Committees, TC 147, Work Programme; Vienna Agreement. http://cdb.iso.org ISO Concept Database (ISO/CDB). https://www.nemi.gov National Environmental Methods Index (NEMI), Chemical Methods. http://standards.gov Standards.gov; NTTA (National Technology Transfer Act). http://www.wiley-vch.de Wiley-VCH, DEV (Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlamm-Untersuchung).

3.12 Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control TN Petney and H Taraschewski, Karlsruhe Institute of Technology, Karlsruhe, Germany & 2011 Elsevier B.V. All rights reserved.

3.12.1 3.12.2 3.12.2.1 3.12.2.1.1 3.12.2.1.2 3.12.2.1.3 3.12.2.2 3.12.2.2.1 3.12.2.2.2 3.12.2.2.3 3.12.2.2.4 3.12.2.2.5 3.12.2.2.6 3.12.2.2.7 3.12.2.3 3.12.2.4 3.12.2.5 3.12.2.5.1 3.12.2.5.2 3.12.2.5.3 3.12.2.5.4 3.12.2.5.5 3.12.2.5.6 3.12.2.5.7 3.12.3 3.12.3.1 3.12.3.1.1 3.12.3.1.2 3.12.3.1.3 3.12.3.1.4 3.12.3.1.5 3.12.3.1.6 3.12.3.1.7 3.12.3.2 3.12.3.2.1 3.12.3.2.2 3.12.3.2.3 3.12.3.2.4 3.12.3.2.5 3.12.3.2.6 3.12.3.2.7 3.12.3.3 3.12.3.3.1 3.12.3.3.2 3.12.3.3.3 3.12.3.3.4 3.12.3.3.5 3.12.3.3.6 3.12.3.3.7 3.12.3.4 3.12.3.4.1 3.12.3.4.2 3.12.3.4.3

Introduction Parasites Transmitted through Drinking Water Cryposporidiosis and Giardiasis Cryptosporidiosis Giardiasis Toxoplasmosis Amoebiasis Parasite characterization Developmental cycle Human involvement Disease characteristics in humans Anthropogenic alterations to the environment Prevention and cure Recommendations Free-Living Amoeba Microsporidian Infections Dracunculiasis Parasite characterization Developmental cycle Human involvement Disease characteristics in humans Prevention and cure Anthropogenic alterations to the environment Recommendations Food-Borne Parasites Transmitted through Freshwater and Marine Foods Opisthorchiasis and Clonorchiasis Parasite characterization Developmental cycles Human involvement Disease characteristics in humans Prevention and cure Anthropogenic alterations to the environment Recommendations Intestinal Flukes Parasite characterization Developmental cycle Human involvement Disease characteristics in humans Prevention and cure Anthropogenic alterations to the environment Recommendations Paragonimiasis Parasite characterization Developmental cycle Human involvement Disease characteristics in humans Prevention and cure Anthropogenic alterations to the environment Recommendations Diphyllobothriosis Parasite characterization Developmental cycle Human involvement

303 304 304 304 305 308 308 308 310 310 311 311 311 311 311 313 313 313 313 314 314 314 314 314 314 315 315 315 315 317 317 318 319 319 319 319 320 320 320 321 321 321 321 321 323 323 323 323 323 323 323 324 324

303

304

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

3.12.3.4.4 3.12.3.4.5 3.12.3.4.6 3.12.3.4.7 3.12.3.5 3.12.3.5.1 3.12.3.5.2 3.12.3.5.3 3.12.3.5.4 3.12.3.5.5 3.12.3.5.6 3.12.3.5.7 3.12.4 3.12.4.1 3.12.4.1.1 3.12.4.1.2 3.12.4.1.3 3.12.4.1.4 3.12.4.1.5 3.12.4.1.6 3.12.4.1.7 3.12.5 3.12.5.1 3.12.5.1.1 3.12.5.1.2 3.12.5.1.3 3.12.5.1.4 3.12.5.1.5 3.12.5.1.6 3.12.5.1.7 3.12.6 3.12.6.1 3.12.6.1.1 3.12.6.1.2 3.12.6.1.3 3.12.6.1.4 3.12.6.1.5 3.12.6.1.6 3.12.6.1.7 3.12.6.2 3.12.6.2.1 3.12.6.2.2 3.12.6.2.3 3.12.6.2.4 3.12.6.2.5 3.12.6.2.6 3.12.6.2.7 3.12.6.3 3.12.6.3.1 3.12.6.3.1 3.12.6.3.3 3.12.6.3.4 3.12.6.3.5 3.12.6.3.6 3.12.6.3.7 3.12.7 3.12.7.1 3.12.7.2 3.12.7.3 3.12.7.4

Disease characteristics in humans Prevention and cure Anthropogenic alterations to the environment Recommendations Anisakiasis Parasite characterization Developmental cycle Human involvement Disease characteristics in humans Prevention and cure Anthropogenic alterations to the environment Recommendations Other Parasites with a Water-Dependent Life Cycle Fascioliasis Parasite characterization Developmental cycle Human involvement Disease characteristics in humans Prevention and cure Anthropogenic alterations to the environment Recommendations Parasites Penetrating Human Skin on Contact with Freshwater Schistosomiasis (bilharziosis) Parasite characterization Developmental cycle Human involvement Disease characteristics in humans Prevention and cure Anthropogenic alterations to the environment Recommendations Water-Dependent Vector-Borne Parasites Malaria Parasite characterization Developmental cycle Human involvement Disease characteristics in humans Prevention and cure Anthropogenic alterations to the environment Recommendations Onchocerciasis Parasite characterization Developmental cycle Human involvement Disease characteristics in humans Prevention and cure Anthropogenic alterations to the environment Recommendations Lymphatic Filariasis Parasite characterization Developmental cycle Human involvement Disease characteristics in humans Prevention and cure Anthropogenic alterations to the environment Recommendations Environmental Factors Influencing the Dynamics of Water-Associated Parasites Dam Construction and Irrigation Projects Land-Use Changes Mass Animal Husbandry: Cryptosporidia and Giardia Human Conflict, Political Considerations, and Healthcare Systems

324 324 324 325 326 326 326 326 326 328 328 328 329 329 329 329 329 329 331 331 332 333 333 333 333 334 334 334 335 335 337 337 337 337 338 338 339 339 342 342 342 342 343 343 343 344 345 346 346 346 346 347 347 347 349 349 349 350 350 351

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control 3.12.7.5 El Nino 3.12.7.6 Climate Change 3.12.8 Synopsis 3.12.9 Conclusion Acknowledgments References

3.12.1 Introduction Humans are reliant on limited supplies of freshwater for drinking, hygiene, recreation, agriculture, and industry. The subsequent recycling of wastewater into usable surface water is thus necessary to maintain even reasonable water supplies (Postel et al., 1996; Pimentel et al., 1997). In addition, both freshwater and marine ecosystems provide the human population with a necessary but increasingly depleted source of highquality protein (Tidwell and Allan, 2001). Indeed, some human communities derive most of their protein from fish or waterdwelling invertebrates (Middendorp, 1992; Allan et al., 2005). This reliance makes the contact between humans and water an essential component in human life. However, water, as well as the animals harbored in it, also provides a continual source of contact with parasites and pathogens which can potentially cause major problems for humans, the subject of this chapter. Historically, there is evidence that for thousands of years, humans have either deliberately manipulated local hydrological patterns in order to ensure an adequate water supply (Shaw and Sutcliffe, 2003; Abdel Khaleq and Alhaj Ahmed, 2007), or unintentionally changed such patterns by modifying land use, such as deforestation, aimed at increasing agricultural land availability (O’Sullivan et al., 2008). Over the last 100 years, this manipulation has reached immense proportions (Vo¨ro¨smarty and Sahagian, 2000). It alters not only the hydrological patterns, but also the diversity of organisms populating the new habitat, among which are human parasites and pathogens. Thus, deforestation, with the aim of increasing agricultural land, often leads to a rise in the water table and the formation of wetlands suitable as breeding grounds for mosquitoes causing malaria (Yasuoka and Levins, 2007; O’Sullivan et al., 2008). For historical and technical reasons, diseases due to bacteria, fungi, and viruses fall under the responsibility of microbiologists and physicians. These pathogens usually do not have complex developmental cycles, and the likelihood of human infections can be substantially diminished by improving general sanitation and hygiene (Jacobsen and Koopman, 2004; Ashbolt, 2004), as well as vaccination for certain pathogens (e.g., hepatitis A virus) (Dagan et al., 2005). In contrast, disease agents belonging to the protozoa, parasitic worms, and arthropods, such as lice and ticks, fall under the scope of parasitologists and physicians. These species may be zoonotic, that is, they are predominantly animal parasites which can also infect humans, or parasites specific for humans (anthroponotic), such as Plasmodium falciparum, the cause of the most severe form of malaria, which does not have wild or domestic animal hosts. Parasites often show characteristic life cycles, for instance, being dependent on a specific snail serving as a first intermediate

305 351 351 351 353 353 353

host and a variety of fish species as the second intermediate hosts (e.g., Andrews et al., 2008). Humans can act as the major final host in terms of supporting the persistence of the parasite in a certain area, or may be as an accidental host of a parasite which usually occurs in the birds or mammals, which function as reservoir hosts (Cox, 2002). In a final host, the parasite reproduces sexually, while in intermediate hosts it merely grows or reproduces asexually. Monoxenic life cycles, that is, those not requiring an intermediate host species, also exist. Studies on the epidemiology of disease agents often involve different aspects of ecology: the first intermediate host may benefit from eutrophication of its aquatic habitat or it may become displaced by a competing invasive species which is not able to transmit the parasite (Taraschewski, 2006). In the present times of global change, it is likely that the competitor is unintentionally introduced from a different continent or it may be naturalized as part of an eradication campaign directed against the parasite (Hudson and Greenman, 1998; Torchin and Mitchell, 2004; Taraschewski, 2006). As we explain in this chapter, parasite transmission is also affected by changes in land use, including irrigation, the building of dams, wetland restoration, etc. (Walsh et al., 1993; Patz et al., 2004). In industrialized countries, serious waterborne health hazards, such as malaria, either do not exist or are controlled; however, humans cannot be efficiently protected from parasites hosted in high abundance by stock animals or water birds or other reservoir hosts (Daszak et al., 2000). In contrast to infections by viruses or bacteria, thus far, no suitably effective vaccines against parasitic protozoa or worms are available. Thus, measures for water management are of great significance in view of the control or spread of waterborne parasitic diseases (Feenstra et al., 2000; Marino, 2007; Montgomery and Elimelech, 2007; Zhou et al., 2009). All wellplanned hydrological projects, such as the construction of dams or irrigation projects, should therefore always take into consideration the project’s potential impact on the human (and animal) populations caused by changes in the epidemiology of water-associated parasites. This is especially true for developing countries where the population is particularly susceptible to such hazardous pathogens (Oomen et al., 1994; Carr et al., 2004; Moe and Rheingans, 2006). Of the 14 major human diseases caused by parasites listed by Crompton (1999) as modified by Bush et al. (2001), seven of them have obligatory freshwater periods in their life cycles. All of these have a very substantial impact on human morbidity and mortality, particularly in developing countries. These include malaria, the ninth leading cause of human mortality worldwide, and several parasites which lead to diarrheal disease, the seventh leading cause of death (Mathers et al., 2007). Indeed, one of the key risk factors for human

306

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

mortality in developing countries is ‘‘unsafe water, sanitation and hygiene’’ (Mathers et al., 2007). Here, we review human parasites associated with freshwater, brackish, or marine environments, emphasizing their transmission cycles in relation to their aquatic habitats and to hydrological engineering projects which may influence their distribution and rate of transmission and thus directly impact human health and well-being.

3.12.2 Parasites Transmitted through Drinking Water 3.12.2.1 Cryposporidiosis and Giardiasis These two diseases are commonly discussed together due to the similarities in their epidemiology (World Health Organization, 2002; Caccio` et al., 2005). The tough, long-lived cysts of both species (Figures 1–4) occur worldwide in aquatic environments with fecal contamination (World Health Organization, 2002). Their resistance to disinfectants used in water treatment, which is particularly true for Cryptosporidium spp., as well as the low infective dose make them a health danger even in developed countries such as the United States and countries of the European Union (MacKenzie et al., 1994; Ward et al., 2002; World Health Organization, 2002). Although not dealt with in detail, Cyclospora cayentanensis, which is related to Cryptosporidium, is a recently recognized species which causes a disease similar to cryptosporidiosis (Ortega et al., 1993; Shields and Olson, 2003).

3.12.2.1.1 Cryptosporidiosis Parasite characterization. Cryptosporidium spp., which are enteric coccidian protozoa lacking multicellular stages. They do not have an intermediate host (Figure 1). They belong to the phylum Apicomplexa, order Eucoccidiorida. There are at least 13 species of parasites, belonging to this genus (Xiao et al., 2004), which are capable of infecting a wide variety of vertebrate hosts, including fish, reptiles, birds, and mammals, but rarely amphibians (Fayer, 2004; Ramirez et al., 2004; Graczyk, 2007; Jirku et al., 2008). Of these, Cryptosporidium parvum is the most commonly reported species infecting mammals in general (Ramirez et al., 2004). There are a variety of vertebrate reservoir hosts, such as cattle (particularly calves), rodents, and potentially chickens (Chalmers et al., 1997; Sre´ter and Varga, 2000; Ramirez et al., 2004). In humans, two distinct types of Cryptosporidia usually occur, the oocysts of which are morphologically very similar but can be differentiated genetically (Clark, 1999). Until recently, divided into type 1 and type 2, C. parvum type 1 is now recognized as C. hominis while type 2 is C. parvum sensu stricto (Morgan-Ryan et al., 2002). Both species are pathogenic to humans (Chappell et al., 2006); however, only C. parvum appears to be frequently transferred from animals, usually cattle, to humans (Monis and Thompson, 2003). Developmental cycle. Cryptosporidium species occur in the environment as approximately 5 mm in diameter, round oocysts each of which contain four sporozoites (Figures 1 and 2). These oocysts are covered by a tough protective coat which is resistant to external factors, enabling them to survive in suitably moist, cool environments for 6 months or more

(Fayer et al., 1998; Fayer, 2004). Infection occurs through the ingestion of these oocysts (Clark, 1999). Once in their host, the oocysts move through the gut to the small intestine where they rupture and release the sporozoites. This tapering, slender stage adheres to the epithelial cells lining the gastrointestinal tract and then breaks the epithelial barrier leading to inflammation (Clark, 1999; Savioli et al., 2006). Within the cells, the parasite replicates by forming merozoites, which escape from ruptured cells to infect new cells and complete the asexual multiplication phase of the life cycle. Eventually, the merozoites differentiate into gamonts. These undergo sexual reproduction and generate oocysts containing sporozoites which are excreted in the feces and are capable of spreading the infection (Clark, 1999). In addition, as shown in Figure 1, oocysts may also rupture before being discharged with the feces leading to an autoinfection of the respective host individual. Human involvement. Cryptosporidiosis is one of the major causes of diarrheal disease in developing countries, while in industrialized countries it can also lead to serious outbreaks (Guerrant, 1997). Transmission to humans is usually via drinking water, which may or may not have been previously treated, including in developed countries such as Australia, Canada, the United Kingdom, and the United States (LeChevallier et al., 1991; Glaberman et al., 2002). The largest outbreak so far recorded was in Milwaukee, USA, with over 400 000 humans estimated to have been infected (MacKenzie et al., 1994). In the UK and the US, between 4% and 100% of surface-water samples were found to be contaminated with Cryptosporidium oocysts with contamination rates ranging from 0.1 to 10 000 oocysts per 100 l (Lisle and Rose, 1995). Infection can occur after ingestion of 10 or less oocysts (Okhuysen et al., 1999; Pereira et al., 2002). In addition, groundwater may also be contaminated (Hancock et al., 1998) and even if the water is treated before consumption, contamination may remain. In Canada, for example, 3.5% of treated water samples were contaminated compared to 6.1% of raw sewage samples and 4.5% of raw water samples (Wallis et al., 1996). Water used in recreation has also been implicated as a source of infection. In 2001, 358 cases of cryptosporidiosis were reported associated with a water park in Illinois in which contaminated water was present (Causer et al., 2006). Chlorine-resistant oocysts, frequent swimming, high density of bathers, and frequent use by very young children probably contributed to the outbreak. An unmonitored splash park, where guests have the opportunity to spray, splash, or pour water on one another, was the site of 154 cases of cryptosporidiosis in Idaho, USA (Jue et al., 2009). This outbreak was related to deficiencies in the technical control of backflow water into the system. In developing countries, children under the age of 5 years, and especially under 3 years, are most commonly affected, with an estimated 25% of all children being infected for the first time during this period (Tzipori and Ward, 2002). A number of Cryptosporidium species can be involved in these infections, and multiple species infections can occur in children as well as adults (Cama et al., 2007). In a study in Kampala, Uganda, children with persistent diarrhea (31%) were more likely to be infected by Cryptosporidium species than

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

3

307

Thick-walled oocyst ingested by host

Recreational water 2

Drinking water Contamination of water and food with oocysts

1

j Thick-walled oocyst (sporulated) exits host

a Oocyst

Thick-walled oocyst (sporulated) exits host

b Sporozoite

c Trophozoite

Auto-infection k Thin-walled oocyst (sporulated)

d

Type I meront

Asexual cycle

Merozoite Microgamont

Microgametes

f g

Undifferentiated gamont Merozoites

Macrogamont Zygote i

e

Type II meront Sexual cycle

h

Figure 1 The life cycle of Cryptosporidium spp. Sporulated oocysts, containing four sporozoites, are excreted by the infected host through feces and possibly other routes such as respiratory secretions 1 . Transmission of Cryptosporidium parvum and C. hominis occurs mainly through contact with contaminated water (e.g., drinking or recreational water). Occasionally food sources, such as chicken salad, may serve as vehicles for transmission. Many outbreaks in the United States have occurred in waterparks, community swimming pools, and day-care centers. Zoonotic and anthroponotic transmission of C. parvum and anthroponotic transmission of C. hominis occur through exposure to infected animals or exposure to water contaminated by feces of infected animals 2 . Following ingestion (and possibly inhalation) by a suitable host 3 , excystation a occurs. The sporozoites are released and they parasitize the epithelial cells ( b , c ) of the gastrointestinal tract or other tissues such as the respiratory tract. In these cells, the parasites undergo asexual multiplication (schizogony or merogony) ( d , e , f ) and then sexual multiplication (gametogony) producing microgamonts (male) g and macrogamonts (female) h . Upon fertilization of the macrogamonts by the microgametes ( i ), oocysts ( j , k ) develop that sporulate in the infected host. Two different types of oocysts are produced, the thick-walled, which is commonly excreted from the host j , and the thin-walled oocyst k , which is primarily involved in autoinfection. Oocysts are infective upon excretion, thus permitting direct and immediate fecal–oral transmission. Reproduced with permission from CDC.

308

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Figure 2 Oocysts of Cryptosporidium sp. (c. 5 mm in diameter) released with the feces of a human host. The sporozoites inside the cysts are difficult to differentiate. Reproduced with permission from CDC.

those with acute disease (22%) and were also more likely to suffer from reduced growth and malnutrition than nutritionally healthy children (Tumwine et al., 2003). Mortality rate in cases of chronic cryptosporidiosis was high in this study, with 24/191 children dying (12.6%) compared to 34/545 (6.2%) for those without a C. parvum infection (Tumwine et al., 2003). A study in the Venda region of South Africa also showed a high prevalence of 50% in the older population group, 50–59 years (Samie et al., 2006). Transmission can also occur from animals to humans (particularly on farms); however, this would appear to be rare (Ramirez et al., 2004). There is little information available on the economic impact of cryptosporidiosis; however, a study of the 1993 outbreak in Milwaukee estimated total costs to be US$93 million divided into 31.7 million medical costs and 64.6 million in lost productivity (Corso et al., 2003). Disease characteristics in humans. In humans, cryptosporidia frequently cause diarrheal diseases. This is characterized by nausea, diarrhea, abdominal pain, low-grade fever, and weight loss. Symptoms usually last for 10–14 days and the disease is self-limiting (Clark, 1999), and in some areas high prevalences are not associated with significant disease possibly due to the development of an immune response (Estaban et al., 1998; Pantenburg et al., 2008). Cryptosporidia infection, including adventitious species other than C. parvum, in patients suffering from immunodeficiency diseases, such as human immunodeficiency virus– acquired immune deficiency syndrome (HIV-AIDS), as well as those with compromised immune systems such as transplant patients, can be life threatening (Kibbler et al., 1987; Matos et al., 2004). In these patients, cryptosporidiosis is characterized by chronic, profound diarrhea and, potentially, vomiting (Blanshard et al., 1992; Cama et al., 2007). Recent research has shown that immunosuppressed patients are susceptible to a variety of different Cryptosporidium species, and that infection with different species leads to different clinical pictures (Cama

et al., 2007). In addition, prolonged infections in immunocompromised patients can lead to spread of this intestinal pathogen to the hepatobiliary and pancreatic ducts causing pathology in these organs (Tzipori and Ward, 2002). Such chronic infections lead, for example, to disruption of the epithelial surface, fibrosis, and cellular infiltration. Lack of effective treatment exacerbates the problem (Tzipori and Ward, 2002). Diagnosis relies mainly on detecting Cryptosporidium oocysts in fecal samples, although this method is unreliable due to the similarity between Cryptospoidium species and certain other protozoan organisms occurring in the feces (Jex et al., 2008). Recently, molecular diagnostic methods using immunological assays and the detection and differentiation of nucleic acids have been developed (Jex et al., 2008). Prevention, cure. The inactivation or removal of cryptosporidia from drinking water is relatively difficult, with filtration and coagulation being most effective, while chlorination is relatively ineffective (Di Giovanni et al., 1999). In order to remove cryptosporidia, multiple processes should be employed including prevention of source contamination, and performing coagulation filtration and disinfection (Ro¨delsperger et al., 1999; Betancourt and Rose, 2004). Recently, The United States Environmental Protection Agency established new drinking-water standards (Long-Term 2 Enhanced Surface Water Treatment Rule) with the aim of reducing the risk of cryptosporidiosis. Chlorination is of little value and before disinfection, an effective particle separation is necessary. In addition, ultraviolet (UV) disinfection is effective (Clancy et al., 1998; World Health Organization, 2004). Fortunately, cryptosporidiosis is usually self-limiting in healthy, well-nourished, immunocompetent humans. Even though pharmaceutical research has been ongoing for many years, ‘‘there are no consistently effective, approved products for either animals or humans’’ (Fayer, 2004). This statement is supported by a meta-analysis carried out by Abubakar et al. (2007a, 2007b). Anthropogenic alterations to the environment. Untreated wastewater is a major source of potential infection with Cryptosporidium spp. (Ramirez et al., 2004). In an Israeli study, both surface and subsurface irrigation with effluent led to the accumulation of Cryptosporidium oocysts at depths ranging from the surface to 90 cm below the soil surface (Armon et al., 2002). In England, infected cattle, along with high rainfall which washed infective stages into a water supply, are suspected to have contributed to an outbreak of 55 000 cases (Rose, 1997). Ong et al. (1996) showed that significantly more Cryptosporidium oocysts were found downstream of cattle farms than upstream and that peak concentrations occurred during calving. Recommendations. Control of cryptosporidiosis can best be affected via adequate treatment of drinking water. The resistant nature of the oocysts, however, requires relatively sophisticated, multilevel treatment (Ro¨delsperger et al., 1999; Hambsch and Lipp, 2000) which is difficult to obtain in developing countries. Here, drinking water should be boiled before consumption. At a more basic level, the correct, hygienic disposal of human and animal feces and adequate wastewater treatment prior to recycling, for example, for irrigation, can considerably

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

309

2 i

Contamination of water, food, or hands/fomites with infective cysts

Trophozoites are also passed in stool but they do not survive in the environment

1 i = infective stage d = diagnostic stage

Cyst

i d

Trophozoites

d Cyst

3

4

5

Figure 3 The life cycle of Giardia lamblia. Cysts are resistant forms and are responsible for transmission of giardiasis. Both cysts and trophozoites can be found in the feces (diagnostic stages) 1 . The cysts are hardy and can survive several months in cold water. Infection occurs by the ingestion of cysts in contaminated water, food, or by the fecal–oral route (hands or fomites) 2 . In the small intestine, excystation releases trophozoites (each cyst produces two trophozoites) 3 . Trophozoites multiply by longitudinal binary fission, remaining in the lumen of the proximal small bowel where they can be free or attached to the mucosa by a ventral sucking disk 4 . Encystation occurs as the parasites transit toward the colon. The cyst is the stage found most commonly in nondiarrheal feces 5 . As the cysts are infectious when passed in the stool or shortly afterward, person-to-person transmission is possible. While animals are infected with Giardia, their importance as a reservoir is unclear. Reproduced with permission from CDC.

reduce the likelihood of infection. Thus, the protection of surface-water reservoirs is recommended by restrictions to animal breeding and grazing in the surrounding area.

3.12.2.1.2 Giardiasis Parasite characterization. Giardia lamblia (synonyms G. duodenalis, G. intestinalis) is a flagellate protozoan or complex of morphologically similar protozoan species belonging to the phylum Sarcomastigophora, order Diplomonadida, which are characterized by duplication of their major organelles (Andrews et al., 1989; Thompson, 2000; Adam, 2001)

(Figures 3 and 4). It is estimated that there are 2.8 108 cases per year worldwide, making Giardia the major source of intestinal infections in developing countries, although it is also found in developed countries with increasing incidences of infection (Lane and Lloyd, 2002). The current taxonomic situation is complex with the genus now containing a number of new species. These appear to be relatively host specific (Monis et al., 2009). Developmental cycle. G. lamblia, like Cryptosporidium spp., has a monoxenic life cycle, that is, it does not depend on any intermediate or paratenic host (Figure 3). The 10–12-mm-long cyst (Figure 4) is the infectious stage. This form is resistant to

310

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Figure 4 Giardia lamblia cyst (c. 10–12 mm in length) excreted with human feces. Infection occurs by oral ingestion. From the slide collection of Werner Frankw, University of Stuttgart-Hohenheim, Germany.

various methods of water treatment, including chlorination and the use of ozone (Lane and Lloyd, 2002). Once in the host, excystation takes place to form an excyzoite before dividing into four trophozoites. These are 15 mm long and motile, possessing four pairs of flagella and a ventral disk for attachment to the intestinal wall, and nutrition is taken via phagocytosis from the intestinal contents. These remain in the intestine (Mu¨ller and Von Allmen, 2005) where they can either move in the intestinal lumen or attach to the microvillous brush-border of enterocytes by way of the adhesive ventral disk (Lane and Lloyd, 2002; Mu¨ller and Von Allmen, 2005) (Figure 3). G. lamblia is normally found on the surface of the lumen of the small intestine in a wide variety of mammals where the trophozoites reproduce asexually. The environmentally resistant cysts (Figure 4) are excreted with the feces and can be transmitted by the fecal–oral route either by direct contact or via contaminated food or, most commonly, through water (Thompson, 2000; Stuart et al., 2003). Contamination of freshwater occurs via fecal contamination from humans and possibly also from animal hosts such as livestock (Thompson, 2000). Human involvement. Humans are usually infected by ingesting contaminated water; this may be as drinking water or by swallowing water while swimming or during water-related recreational activity (Stuart et al., 2003). Direct transmission among children in day-care centers and between male homosexuals has also been reported (Meyers et al., 1977; Phillips et al., 1981; Polis et al., 1986; Rauch et al., 1990). Infection can also occur by eating raw vegetables, presumably irrigated with contaminated water (Osterholm et al., 1981; Stuart et al., 2003). G. lamblia is the most common cause of waterborne diarrhea in the USA in 46 states (Hlavsa et al., 2005) and outbreaks have also occurred in Australia, Canada, New Zealand, Sweden, Norway, and the United Kingdom (Thompson, 2000; Stuart et al., 2003; Nygard et al., 2006). Although zoonotic Giardia infection has been reported (Traub et al., 2004; Leonhard et al., 2007; Geurden et al., 2008), the extent to which this is the case currently requires more research based on recent molecular taxonomic work on the complex of Giardia species which has been divided into different

molecularly defined assemblages (Monis and Thompson, 2003; Thompson et al., 2008). As with Cryptosporidium, Giardia cysts cannot reliably be eliminated from drinking water by filtration, although the cysts are larger (LeChevallier et al., 1991). Indeed, in Canada, 18.2% of treated water samples were found contaminated with Giardia cysts compared to 73% of raw sewage and 21% of raw water samples (Wallis et al., 1996). In industrialized countries, Giardia transmission can also be related to inadequate hygiene at public recreational facilities such as swimming pools (Porter et al., 1988). In many developing countries, giardiasis occurs with high prevalences due to poor hygienic conditions (Sullivan et al., 1991; World Health Organization, 2002). Prevalences reach 19.5% in rural communities in Malaysia with those most at risk being less than 13 years old (Norhayati et al., 1998); a similar picture is found for the native Orang Asli children (24.9% prevalence; Al Mekhlafi et al., 2005). In Africa, prevalence rates of 9.8% (Sudan; Magambo et al., 1998), 45% (vegetable farmers in Eritrea; Srikanth and Naik, 2004), and 31% (Maasai children, Kenya; Joyce et al., 1996) have been reported. Disease characteristics in humans. The symptoms caused by G. lamblia in humans are very variable. In many individuals, the infection is asymptomatic (Adam, 1991). Should overt disease occur, G. lamblia may cause stomach and abdominal pain, and nausea followed by either severe acute or chronic diarrhea potentially leading to dehydration and weight loss due to malabsorption as the absorptive surface of the small intestine may become blocked by the high density of the flagellates (Adam, 1991; Thompson, 2004; Mu¨ller and Von Allmen, 2005). Infection may result in poor condition and growth in children due to reduced nutrient uptake (Sullivan et al., 1991; World Health Organization, 2002). The pathogenesis of giardiasis is still unclear but includes microvillus shortening, flattening, or atrophy. This subject has been reviewed in detail by Mu¨ller and Von Allmen (2005). The disease is often self-limiting with a spontaneous cure occurring after 2–6 weeks. If, however, it should become chronic, there are short or persistent periods of diarrhea (Mu¨ller and Von Allmen, 2005). Diagnosis was usually made by determining the presence of cysts in fecal samples using light microscopy. Recently, immunoassays and polymerase chain reaction (PCR)-based diagnostics have been developed (Regnath et al., 2006; Haque et al., 2007). Prevention, cure. As G. lamblia is larger than Cryptosporidium, its removal by filtration is easier and it is also substantially more susceptible to chlorination (Betancourt and Rose, 2004). Prevention or a substantial reduction in contamination can be achieved by adequate treatment of water, such as sedimentation and retention (Betancourt and Rose, 2004). Preventing contamination of water sources used for drinking and irrigation would cut the transmission route and prevent infection. In areas with insufficient hygienic standards, drinking water should be boiled. Unlike Cryptospridium spp., Giardia can be effectively treated in the human host. Tinidazole is the drug of preference, with a single dose being effective for most individuals (Petri, 2005). Metronidazole, a related drug, is also effective but

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

treatment 3 times a day for 5–7 days is recommended. Limited studies also indicate that nitrazoxanide may be as effective as trinidazole without the latter’s bitter taste. During pregnancy, paromomycin is recommended (Petri, 2005). Anthropogenic alterations to the environment. Substantial evidence is available showing that irrigation with wastewater leads to contamination of a wide variety of crop plants in various areas of the world (Thurston-Enriquez et al., 2002; Ensink et al., 2006). Experimental watering of mint, coriander, radish, and carrots with untreated wastewater, treated wastewater with sedimentation and 16 days retention, and freshwater, showed high levels of contamination for all crops when untreated wastewater was used (96, 254, 59, and 155 cysts kg1, respectively) (Amahmid et al., 1999). When these cultures were irrigated with treated or freshwater, no contamination could be detected. Cifuentes et al. (2000) found up to 300 Giardia cysts per liter of untreated wastewater in an agricultural area of Mexico. In this case, water retention in reservoirs led to a substantial reduction in the number of cysts to less than 6 l1. However, if sewage sludge was used for fertilizing fields, the likelihood of introducing Giardia (and Cryptosporidium) cysts to the environment was high (Straub et al., 1993; Gale, 2005). In the study by Ong et al. (1996) comparing Giardia and Cryptosporidium in two adjacent watersheds in which cattle production occurs, Giardia was not found in water samples collected from lakes and headwaters in either watershed, but was collected in almost 100% of samples of water which had passed through cattle pastureland, with a maximum of about 2000 cysts per 100 l at both sites. As for Cryptosporidium, the sites downstream of cattle ranches had significantly higher levels of Giardia cysts than those upstream, and the peak concentrations occurred with calving. The importance of global warming for Giardia are currently unknown, although increased temperature is likely to result in the colonization of areas previously too cold to support this species (Polley and Thompson, 2009). There is evidence, however, that the increased or extreme precipitation predicted for some areas is likely to increase the risk of contamination of otherwise safe water sources (Curriero et al., 2001). Recommendations. As for cryptosporidiosis, giardiasis control can be affected via adequate treatment of drinking water, including boiling. Again, as for cryptosporidia, the resistant nature of the cysts requires relatively sophisticated, multilevel treatment in water-purification plants, which is likely to be difficult to obtain in developing countries. In addition, the use of human and animal feces and inadequate wastewater treatment prior to recycling, for example, for irrigation, can considerably increase the likelihood of disease outbreaks. Plants deriving from such sources should be cooked before consumption and not used as salad.

3.12.2.1.3 Toxoplasmosis Toxoplasma gondii, the causative agent of toxoplasmosis, is a coccidian parasite (order Eimeriida) which has a very high prevalence in most human populations worldwide (Tenter et al., 2000). A wide variety of avian and mammal intermediate hosts, including pigs, sheep, and goats as well as poultry, game animals, and rabbits, have been recorded as intermediate

311

hosts (Tenter et al., 2000). Infection usually occurs via eating meat-containing cysts or by ingestion of oocysts which occur in environments contaminated with the infected feces of the final hosts, cats. These are unsporulated and therefore not infectious when they are excreted. Sporogony is temperature dependent and the oocysts only become infectious 2–4 days after exposure to air (Lucius and Loos-Frank, 2008). Despite the high prevalence of infection with T. gondii in humans and cats (Tenter et al., 2000), the waterborne route of infection has usually been considered to play a minor role, with most infection considered to be caused by eating undercooked, contaminated meat (Tenter et al., 2000). Recently, however, more interest in possible infection through contact with water has been generated after a major outbreak of the disease involving an estimated 2894–7718 humans, including 100 acute cases, in Victoria, British Columbia, Canada. This outbreak was associated with an unfiltered but chlormainated municipal water supply (Bowie et al., 1997). In north Rio de Janeiro State, Brazil, there was an increased risk of seropositivity associated with drinking unfiltered water from sites accessible to contamination (Bahia-Oliveira et al., 2003). Toxoplasmosis is usually considered to be a benign infection in immunocompetent individuals except when the initial infection occurs in the months prior to or during pregnancy, in which case transplacental transmission to the fetus can occur. This can cause severe disease leading to mental retardation, microcephaly or hydrocephalus, or even prenatal death (Kravetz and Federman, 2005; Rorman et al., 2006). Clinical manifestation may also only become apparent after birth with the neonate being asymptomatic (Rorman et al., 2006). In the case of retinal toxoplasmosis, which may affect about 2% of the infected population of the United States, most individuals are thought to have been infected post-natally (Smith and Cunningham, 2002; Holland, 2003). T. gondii is one of the most common protozoan parasites causing opportunistic disease in immunocompromised individuals (Ferreira and Borges, 2002). Reduced immunity can lead to reactivation of a latent infection resulting in acute disease including severe meningoencephalitis and myocarditis (Ferreira and Borges, 2002). Dubey (2004) suggests that waterborne infection with toxoplasmosis can be avoided by not drinking unfiltered or, if this is not possible, uncooked water, and that access to water sources used for human consumption should be prevented for cats.

3.12.2.2 Amoebiasis 3.12.2.2.1 Parasite characterization Entamoeba histolytica is a protozoan parasite, belonging to the phylum Sarcomastigophora, order Entamoebida, which causes amoebic dysentery, predominantly in the tropics and subtropics (Figures 5 and 6). It has been estimated that this pathogen causes about 40 000–100 000 human deaths a year while 50 million individuals show clinical disease (Petri and Singh, 1999; Petri et al., 2000, 2002; Ackers and Mirelman, 2006). Information on E. histolytica published prior to the 1990s is of doubtful value as the single species recognized prior to this time was finally classified into two species, E. histolytica, which is potentially highly pathogenic to

312

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

2 i Mature cysts ingested

d

C

A

d

d

B d

i = infective stage d = diagnostic stage

i

d

1

d

Cysts and tophozoites passed in feces A = noninvasive colonization B = intestinal disease C = extraintestinal disease 4 Trophozoites dd

Exits host

Multiplication Excystation 3

Trophozoites 4

d 5 Cysts

d

d

d

d

Figure 5 The life cycle of Entamoeba histolytica. Cysts and trophozoites are passed in feces 1 . Cysts are typically found in formed stool, whereas trophozoites are typically found in diarrheal stool. Infection by Entamoeba histolytica occurs by ingestion of mature cysts 2 in fecally contaminated food, water, or hands. Excystation 3 occurs in the small intestine and trophozoites 4 are released, which migrate to the large intestine. The trophozoites multiply by binary fission and produce cysts 5 , and both stages are passed in the feces 1 . Due to the protection conferred by their walls, the cysts can survive days to weeks in the external environment and are responsible for transmission. Trophozoites passed in the stool are rapidly destroyed once outside the body, and if ingested would not survive exposure to the gastric environment. In many cases, the trophozoites remain confined to the intestinal lumen ( A : noninvasive infection) of individuals who are asymptomatic carriers, passing cysts in their stool. In some patients the trophozoites invade the intestinal mucosa ( B : intestinal disease), or, through the bloodstream, extraintestinal sites such as the liver, brain, and lungs ( C : extraintestinal disease), with resultant pathologic manifestations. Reproduced with permission from CDC, modified by the authors of this chapter.

humans, and H. dispar, which is considered to be a commensal species in the human gut (Hamzah et al., 2006; Stauffer and Ravdin, 2003; Fotedar et al., 2007). However, these two species are morphologically indistinguishable making traditional microscopic diagnostic methods obsolete (Jackson, 1998;

Stauffer and Ravdin, 2003). Diagnosis today therefore relies on serological and molecular methods (Stauffer and Ravdin, 2003). Recently, a third morphologically identical species, E. moshkovskii, which can also be distinguished by molecular methods, has been detected in humans. Its cysts are also found

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

313

Figure 6 Mature fecal cyst of Entamoeba histolytica (10–15 mm) typically showing four nuclei. From the slide collection of Werner Frankw.

in fecally contaminated water and, although it is considered to be primarily a free-living species, evidence is accumulating that it can cause diarrhea and other intestinal problems in humans (Tanyuksel et al., 2007; Fotedar et al., 2008). In addition to these species, Tanyuksel and Petri (2003) list Entamoeba coli, E. hartmanni, E. polecki, Iodamoeba bu¨tschlii, and Endolimax nana which are also found in the human gut but are considered to be harmless. Coinfections with different Entamoeba species are possible (Fotedar et al., 2007; Tanyuksel et al., 2007).

protein, must be considered a maladaptation of the parasite to its host (Stauffer and Ravdin, 2003). Infective cysts deriving from intra-intestinal trophozoites are shed in the feces to potentially be picked up by the next host (Figure 5; Lucius and Loos-Frank, 2008). The cysts can survive in moist, cool environmental conditions for 2–4 weeks with a maximum of 3 months if temperatures are above freezing – freezing and high temperatures are lethal (Schuster and Visvesvara, 2004a; Keene, 2006).

3.12.2.2.2 Developmental cycle

E. histolytica is the dominant, globally distributed, amebic pathogen of the human gut (Schuster and Visvesvara, 2004a). It is largely an anthropogenic pathogen but a variety of nonhuman primates, possibly pigs and occasionally dogs can be infected, although in the latter, infective cysts are rarely produced (Petri et al., 2002; Verweij et al., 2003; Schuster and Visvesvara, 2004a). Transmission occurs via the oral route, usually with consumption of water or food contaminated with human fecal material (Lucius and Loos-Frank, 2008). Most cases are confined to the intestine and are self-limiting, with a study in Vietnam showing infections in asymptomatic adult carriers having an average half life of 12.9 months; however, approximately 10% of carriers develop extra-intestinal disease (Blessmann et al., 2003; Stauffer and Ravdin, 2003). Tissue invasion, in particular if the brain is involved, can lead to death within a few days and immunologically naive individuals from industrial countries who spend some months in developing countries often become severely ill (Okhuysen, 2001; Lucius and Loos-Frank, 2008). Such invasive amoebiasis is 3 or more times more common in males than in females, although the sex ratio for asymptomatic individuals is the same (Acuna-Soto et al., 2000; Blessmann et al., 2002). Amebic dysentery is a relatively uncommon colitis with severe, often bloody, stools. In this form of the disease, the trophozoites lyse the epithelial cells of the large intestine and enter the mucosa and submucosa from where they can invade the

3.12.2.2.3 Human involvement As with cryptosporidia and G. lamblia, E. histolytica has a simple, direct life cycle (Figure 5). Infection occurs when metacysts (10–15 mm in diameter with four nuclei, Figure 6) are ingested orally via fecally contaminated water or food. Excystation takes place in the small intestine with the formation of eight metacystic trophozoites (Petri et al., 2002). Once mature, the trophozoites, which have a diameter of 20–60 mm, reproduce by binary fission in the large intestine where most remain in a commensal (apathogenic) relationship. However, some may change through some currently unknown mechanism, to larger, polyploid, hemophagous, metabolically more active forms which are pathogenic (Figure 6; Lucius and Loos-Frank, 2008). These penetrate into the colon wall causing flask-shaped ulcers. Less frequently, E. histolytica spreads through the portal vein to the liver causing amebic liver abscess, and rarely to the lungs or brain where it can cause potentially lethal abscesses (Ackers and Mirelman, 2006). If a lectin on the parasite’s surface attaches to the muco-glycoproteins that line the host’s intestinal lumen, a noninvasive gut infection ensues. In contrast, if the trophozoite penetrates the mucin layer and its lectin attaches directly to the surface of the host cell, a cascade of events occurs ultimately leading to invasive disease. Under these circumstances, replication and cyst formation do not occur. Thus, invasive disease, as determined by the lectin

314

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

organs via the bloodstream (Lucius and Loos-Frank, 2008). The pathological process appears to be enhanced by the inflammatory response of the host (Ackers and Mirelman, 2006). Prevalence rates can be high, especially in children, with 28.4% of under-15-year-olds being infected in the southern Sudan (Magambo et al., 1998). A large-scale study of 289 children aged 2–5 years living in an urban slum in Bangladesh revealed that 80% were infected with E. histolytica over a 4-year period, 53% had a repeat infection, and that 4% developed clinically significant disease over a 2-year period (Haque et al., 2002, 2006). Both innate and acquired immune responses are capable of limiting the infection (Haque et al., 2002, 2006). Although developing countries in warm and temperate areas are most strongly affected, E. histolytica infections can also show high rates of infection, and potentially be a health hazard in industrialized nations, where it is often found predominantly in male homosexuals, with HIV infection possibly being an additional risk factor (Phillips et al., 1981; Ohnishi et al., 2004; Chen et al., 2007; Hung et al., 2008). The route of transmission in this case is likely to involve oral–anal sexual contact, thus differing from the usual water-based route (Allason-Jones et al., 1986). Although Campos-Rodrı´guez and Jarillo-Luna (2005) claim that such infections are usually asymptomatic and extra-intestinal amoebiasis is rare, invasive pathology has been reported from the USA, Japan, and Taiwan (Seeto and Rocky, 1999; Mitarai et al., 2001; Hung et al., 2008). In developed countries where fecal–oral transmission is rare among the general population, this disease is most commonly seen in immigrants coming from, and individuals who had visited, countries with prevalent disease (Haque et al., 2003).

3.12.2.2.4 Disease characteristics in humans Intestinal infections are often asymptomatic, although they may also cause dysentery. In the pathogenic, extra-intestinal form, symptoms of amebic colitis include abdominal cramps, watery or bloody diarrhea, and weight loss over a period of several weeks (Haque et al., 2003). These general symptoms make definitive diagnosis difficult as they are also commonly found in a variety of intestinal bacterial infections, although finding cysts with four nuclei in stool samples is diagnostic for the E. histolytica group (Haque et al., 2003; Tanyuksel and Petri, 2003). Severe but unusual disease progression includes acute necrotizing colitis, toxic megacolon, ameboma causing a bowel lesion, and peri-anal ulceration, as well as abscesses in the liver, lungs, or brain (Haque et al., 2002). These complications require early recognition and medical intervention, for example, acute necrotizing colitis, although rare, has an associated mortality of 40% (Ellyson et al., 1986). Interestingly, clinical characteristics vary geographically: in Egypt, invasive colitis is most common, while in South Africa and central Vietnam as well as in an outbreak in the Republic of Georgia in 1998 liver abscesses predominate (Barwick et al., 2002; Blessmann et al., 2002; Stauffer and Ravdin, 2003). In Vietnam, the use of river water was identified as a major risk factor (Blessmann et al., 2002). In addition to differences in food-preparation techniques between these countries, evidence of genetic variability in this species, even in samples from the same geographic location (Haque et al., 2002;

Haghighi et al., 2003; Simonishvili et al., 2005), suggests that more research should be invested in determining the population genetic structure of this parasite in relation to disease epidemiology and pathology (Bhattacharya et al., 2005). Antigen tests are used to confirm positive stool ova, and computer tomography can be used to visualize tissue abscesses for which molecular biological techniques have also been developed (Tanyuksel and Petri, 2003; Stauffer and Ravdin, 2003).

3.12.2.2.5 Anthropogenic alterations to the environment An outbreak of amebic disease involving liver abscesses in Tbilisi, Republic of Georgia, in 1998 was thought to be caused by either ‘‘inadequate municipal water treatment or contamination of municipal water in the distribution system’’ (Barwick et al., 2002). Data from El Azzouzia, Marrakesh (Morocco), where untreated, fecally contaminated wastewater is used in agriculture, indicate that amoebiasis affects 28% of the population compared to 6% in a control area without such irrigation (Melloul et al., 2002). Thus, it appears that parasite transmission is encouraged by expanding irrigation using wastewater.

3.12.2.2.6 Prevention and cure As with other diseases transmitted by the consumption of fecally contaminated water or uncooked food which had come into contact with such water, the adequate treatment of wastewater is necessary to control the disease (Smith and Perdek, 2004). Effective methods for eliminating Entamoeba spp. date back to the US army during World War II with precoat filtration (LeChevallier and Au, 2004). The cysts are resistant to chlorination and may require prolonged contact times before they are inactivated, although reducing pH and increasing temperature reduces the time needed for inactivation (LeChevallier and Au, 2004). In a Mexican study, advanced primary treatment with high sedimentation combined with sand filtration and chlorination reduced the number of protozoan cysts (Giardia and Entamoeba) by two orders of magnitude (Jiminez et al., 2001). Ozonation and UV radiation can also be effectively used (Schaefer et al., 2004). In areas with poor hygienic standards only safe, bottled, or boiled water should be consumed. Once the pathogen is present, the intra-intestinal stage can be treated with paramomycin and diloxanide furoate with the former showing a higher cure rate (85% vs. 51% in Hue, Vietnam; Blessmann et al., 2005). The visceral form responds to nitroimidazoles, especially metronidazole, to which about 90% of patients with mild-to-moderate symptoms respond (Haque et al., 2003; who provide a comprehensive treatment schedule). Recent studies also show that nitazoxanide is effective against both intestinal infection with E. histolytica as well as invasive amoebiasis (Rossignol et al., 2007).

3.12.2.2.7 Recommendations Adequate treatment of drinking water is capable of reducing this threat; however, the use of contaminated wastewater for irrigation can also lead to cysts occurring on vegetables grown for human consumption. Thus, the use of untreated wastewater for agriculture should be curtailed. Cooking vegetables

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

and drinking boiled water substantially reduce the chances of infection.

3.12.2.3 Free-Living Amoeba A number of species of free-living amoeba belonging to the genera Acanthamoeba, Naegleria, Balamuthia, and Sappinia are known to cause serious disease in humans and potentially in other animals including dogs, sheep, primates, horses, and cattle as well as birds, reptiles, and fish (Rideout et al., 1997; Schuster and Visvesvara, 2004b; Daft et al., 2005; Visvesvara, 2007b; Matin et al., 2008). Transmission is either via contact with contaminated soil or aquatic environments, potentially including seawater, where the free-living amoeba feed on bacteria (Schuster and Visvesvara, 2004b; Lorenzo-Morales et al., 2005a, 2005 b). Those capable of infecting humans must be thermotolerant to survive the normal human-body temperature of 37 1C (Schuster and Visvesvara, 2004b). Infection occurs via breaks in the skin, by cysts carried to the upper respiratory tract by air, or by amoeba carried by water (Schuster, 2002). Both immunocompetent and immunocompromised individuals may be affected (Schuster and Visvesvara, 2004b). Infection with Naegleria fowleri, which is the cause of the usually fatal primary amebic meningoencephalitis, is found in warm-water sources, such as swimming pools which lack sufficient chlorination, ponds, and flowing water (Cabanes et al., 2001; Tiewcharoen and Junnu, 2001; Schuster and Visvesvara, 2004b), and even wells (Shenoy et al., 2002; Blair et al., 2008). This species usually infects healthy children, adolescents, and young adults swimming or washing in such water, with infection occurring via the olfactory neuroepithelium (Schuster, 2002). A number of Acanthamoeba spp. as well as Balamuthia mandrillaris can also cause encephalitis and other disease syndromes, usually in immunocompromised hosts (Teknos et al., 2000; Torno et al., 2000; Marciano-Cabral and Cabral, 2003), and Acanthamoeba can also cause keratitis in healthy humans (Marciano-Cabral and Cabral, 2003; Kilvington et al., 2004; Jeong and Yu, 2005). A single case of amebic encephalitis has been reported in an immunocompetent patient caused by Sappinia pedata (originally identified as S. diploidea; Qvarnstrom et al., 2009), although the route of infection is unknown (Rocha-Azevedo et al., 2009).

3.12.2.4 Microsporidian Infections Microsporidia are single-celled, obligate intracellular, sporeforming, primitive fungal parasites which infect every major animal group from protozoans to humans (Bush et al., 2001; James et al., 2006). Although a wide variety of microsporidian species are known to infect humans, these will not be discussed in detail here as they are usually opportunistic parasites of immunocompromised patients. Occasionally, however, microsporidians have also been detected in healthy humans, often as self-limiting traveler’s diarrhea (Didier et al., 2004; Mathis et al., 2005). In Spain, Enterocytozoon bieneusi infection has been associated with diarrhea in geriatric patients (Mathis et al., 2005). In another study, infection prevalence reached 67.5% in immunocompetent individuals in Cameroon, suggesting that this situation may be much more common than

315

previously considered. (Nkinin et al., 2007). Transmission can occur via several routes including the fecal–oral, oral–oral, ingestion of contaminated food, aerosol, and potentially direct person to person. In addition, contact with or drinking contaminated water has been associated with infections (Hutin et al., 1998; Mathis et al., 2005). Irrigation water, river water, as well as groundwater can be contaminated (ThurstonEnriquez et al., 2002; Didier, 2005). Sixteen of 25 samples taken from the River Seine in France over a period of 1 year were found to be contaminated with E. bieneusi (Fournier et al., 2000). More work is needed to determine the significance of this group of pathogens.

3.12.2.5 Dracunculiasis 3.12.2.5.1 Parasite characterization Dracunculus medinensis, the Guinea worm (Figures 7 and 8), is a nematode (round worm) belonging to the suborder Spirurina, family Dracunculidae. Dracunculiasis, which it causes, once occurred throughout much of the semiarid tropics of the Old World from Central and West Africa to Yemen and parts of India and Pakistan. An estimated 3.5 million people were infected in 1986 prior to the current eradication program which began in the mid-1980s (Hopkins et al., 2005). Eradication efforts have now restricted it predominantly to Sudan and Ghana, and to a lesser extent neighboring countries, with a mere 16 000 cases reported in 2004 (Steib, 1987; Hopkins et al., 2005). Recent data indicate that this trend is continuing at least for southern Sudan, with 5815 cases reported in 2007 but only 3618 in 2008 (Rumunu et al., 2009). Eradication has been possible as D. medinensis is one of the few anthroponoses dealt with here, that is, it is not zoonotic with the only final host being humans (Lucius and Loos-Frank, 2008).

3.12.2.5.2 Developmental cycle The life cycle of D. medinensis (Figure 7) is dependent on still freshwater bodies containing small, commonly occurring, crustaceans known as copepods which act as intermediate hosts (Figure 8; Hopkins, 1983; Steib, 1987). The more these pools are frequented by humans collecting drinking water, the better the chances are of transmission. Temperatures over 19 1C are also necessary for larval development of the most important intermediate host, in West Africa, Thermocyclops inopinus (Steib, 1987). Gravid female worms, which reach a length of 70–100 cm, take up a subcutaneous position usually on the lower leg or foot of their human host where they cause a prominent, painful blister (Hopkins, 1983). On contact with water, this blister ruptures, releasing about a half a million minute (0.3–0.6 mm long), motile L1 larvae into the water (Steib, 1987; Mehlhorn and Walldorf, 1988). Usually, the females die after the first mass discharge of the larvae. The males, which only reach lengths of 3–4 cm, live until they have fertilized a female. Some of the larvae are ingested by copepod crustaceans which act as intermediate hosts. The parasites develop in the hemocoel of this host to the infective thirdstage (L3) larvae (Figure 8), which infect humans who drink unboiled or unfiltered water containing the copepod intermediate hosts. Digestion of the copepods releases the larvae which move to the small intestine before penetrating the intestinal wall and migrating through the connective tissue to

316

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Human drinks unfiltered water containing copepods with L3 larvae Larvae undergoes two molts in the copepod and becomes an L3 larvae 6

1

i Larvae are released when copepods die. Larvae penetrate the host’s stomach and intestinal wall. They mature and reproduce 2

L1 larvae consumed by a copepod 5

Female worm begins to emerge from skin 1 year after infection 3

d

Fertilized female worm migrates to surface of skin, causes a blister, and discharges larvae

4

L1 larvae released into water from the emerging female worm

i = infective stage d = diagnostic stage

Figure 7 The life cycle of Dracunculus medinensis. Humans become infected by drinking unfiltered water containing copepods (small crustaceans) which are infected with larvae of D. medinensis 1 . Following ingestion, the copepods die and release the larvae, which penetrate the host stomach and intestinal wall and enter the abdominal cavity and retroperitoneal space 2 . After maturation into adults and copulation, the male worms die and the females (length: 70–120 cm) migrate in the subcutaneous tissues toward the skin surface 3 . Approximately 1 year after infection, the female worm induces a blister on the skin, generally on the distal lower extremity, which ruptures. When this lesion comes into contact with water, a contact that the patient seeks to relieve the local discomfort, the female worm emerges and releases larvae 4 . The larvae are ingested by a copepod 5 and after two weeks (and two molts) have developed into infective larvae 6 . Ingestion of the copepods closes the cycle 1 . Reproduced with permission from CDC.

period of prepatency, the discharge of larvae into the transmission site takes about a year, ensuring that this happens under the favorable conditions of the dry season (Steib, 1987).

3.12.2.5.3 Human involvement

Figure 8 Dracunculus medinensis : infectious L3 larva (arrow) in a Thermocyclops copepod. Courtesy of Karl Steib.

the thorax (Lucius and Loos-Frank, 2008). Mating between male and female worms occurs after 60–90 days with 10–14 months being required before the eggs are mature and the female can migrate to be body surface. Due to this extended

Humans are the only final hosts for this species. The disease is distributed only in areas with a well-defined dry season with limited water availability and high densities of copepods (Steib, 1987). Transmission sites include ponds, cisterns, pools, and wells, which are frequented by many people and where humans can move into the water with their feet and legs. This is the case, for example, with the traditional step wells in which humans reach the water over a number of steps (Steib, 1987; Hopkins et al., 2005). Copepods can disperse from water body to water body in a variety of ways including transport by animals, such as waterbirds, although wind dispersal of eggs is also likely to be important (Ca´ceres and Soluk, 2002; Cohen and Shurin, 2003). In addition, it is also possible that copepod eggs lie dormant in sediment until the pond refills with rain (Bohonak and Jenkins, 2003). Thus, temporary water sources are likely to be infected (Steib, 1987).

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control 3.12.2.5.4 Disease characteristics in humans Infection is usually asymptomatic until the female worm is ready to lay her eggs and moves to the skin surface (Ruiz-Tiben and Hopkins, 2007). During this movement within the body, symptoms such as slight fever, allergic symptoms, nausea, diarrhea, dizziness, rash with pruritis, and local erythema may occur (Ruiz-Tiben and Hopkins, 2007). The blister itself is painful once it develops, which often leads the patient to seek relief in water. Although the worm usually emerges from the legs or feet (85–90% of cases), other parts of the body, such as the arms, buttocks, and genitalia, may also be involved (RuizTiben and Hopkins, 2007). Migration of the worm through the tissues may also lead to the presence of the worm in other locations, potentially leading to a space-occupying lesion and abscess formation. Chronic disease may then occur, including joint inflammation and arthritis (Ruiz-Tiben and Hopkins, 2007). The open wound is also subject to secondary bacterial infections (Steib, 1987). No diagnostic methods, except the emergence of the female worm, are available, a fact which is negated by the success of control programs not requiring determination of the disease (Molyneux, 2009).

3.12.2.5.5 Prevention and cure There is neither a drug nor a vaccine capable of curing or preventing infection with D. medinensis (Ruiz-Tiben and Hopkins, 2007). It is therefore interesting that the near eradication of the disease, through the Dracunculiasis Eradication Program (Hopkins et al., 2008), is not based on medication or vaccination, but on human education and the distribution of water-filtration equipment (including simple cloth filters) disrupting the freshwater transmission cycle of the disease by removing the copepod intermediate host before drinking (Nwaorgu, 1991; Barry, 2007). Denying contact between the gravid female worm and water, for example, by using an occlusive bandage, will also interrupt the transmission cycle (Ruiz-Tiben and Hopkins, 2007). A current infection is traditionally eliminated by extracting the gravid female manually through the ruptured blister without allowing contact to a freshwater source, a method dating back to at least 1500 BC (Cox, 2002). Metronidazole is sometimes recommended as an anti-inflamatory agent (Bogitsh et al., 2005).

3.12.2.5.6 Anthropogenic alterations to the environment It has been shown that large dams with permanent water sources are associated with lower levels of dracunculiasis than when small, impermanent dams and ponds were used as water sources of people working in the fields, suggesting that small water sources, particularly those constructed by humans (Steib, 1987), are of more epidemiological significance than large ones (see also Scott, 1960; Belcher et al., 1975; Steib and Mayer, 1988; Tayeh and Cairncross, 1998). However, AdekoluJohn (1983) showed that construction of the Kainji Reservoir in northern Nigeria was responsible for the development of surrounding ponds by raising the water table, providing suitable habitat for the intermediate hosts, and thus promoting parasite transmission. Prevalences of dracunculiasis increased in the area from approximately 0% in 1960–64 to 25.6% in 1975–79, 10 years after the construction of the dam

317

(Adekolu-John, 1983). Adekolu-John suggested that future studies take into account the role of the water level in Kainji Reservoir on the formation of ponds and surface water bodies along estuaries and streams. In times of civil unrest, large areas may not be accessible to Guinea-worm-control programs, as has occurred in southern Sudan. This kind of anthropogenic alteration could prevent the aim of complete eradication of D. medinensis in the near future (Hopkins et al., 2008; Rumunu et al., 2009).

3.12.2.5.7 Recommendations Hydrological projects in the affected areas of those countries in which D. medinensis is still found (Sudan, Ghana, Mali, Nigeria, and Niger; Hopkins et al., 2008) should aim at interrupting the transmission cycle of the disease either by providing the possibility of water filtration (even at a simple level) and/or by preventing gravid females from releasing their eggs into water bodies. The construction of sources allowing the collection of water without contact between the water and human extremities would also interrupt transmission. An accompanying program of educating the local population on the dangers of this disease should be carried through.

3.12.3 Food-Borne Parasites Transmitted through Freshwater and Marine Foods Aquaculture is rapidly overtaking natural harvesting as the main source of marine and freshwater food, with 47% of such food currently coming from this area (Food and Agriculture Organization, 2009; Subasinghe et al., 2009). This massive increase in aquaculture, particularly in Asia, has paved the way for the introduction to and propagation of a variety of parasites into the human food-production system in many parts of the world (Naylor et al., 2001). Surveys of aquaculture fish in different ponds in Vietnam showed that between 0.7% and 6.5% to as high as 44.6% contained zoonotic trematode species compared to 10.3% of wild fish (Chi et al., 2007; Hop et al., 2007; Thu et al., 2007). There are about 70 species of food-borne digenean trematodes known to be capable of infecting humans. All are hermaphroditic flatworms which have one or more aquatic snail species as intermediate hosts (World Health Organization, 2004). Most transmission takes place in freshwater but brackish and seawater are occasionally involved. These flatworms cause significant infections such as fascioliasis, clonorchiasis, opisthorchiasis, and paragonimiasis (Keiser and Utzinger, 2005). Somewhat dated estimates suggest that 18 million people are infected by fish-borne trematodes (World Health Organization, 1995a). We now have estimates of over 35 million being infected with opisthorchids trematodes alone (Lun et al., 2005; Andrews et al., 2008).

3.12.3.1 Opisthorchiasis and Clonorchiasis 3.12.3.1.1 Parasite characterization The digenetic trematode family Opisthorchiidae contains three species which are significant parasites of humans in East (Clonorchis sinensis) and Southeast Asia (C. sinensis and Opisthorchis viverrini), and in Europe, as well as all of the

318

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Russian federation except northern Siberia and the Far East (Opisthorchis felineus) (Kaewkes, 2003; Yossepowitch et al., 2004; Pozio, 2008) (Figure 9). Species of both of these genera, the adults of which inhabit the bile ducts of the human liver, are remarkably similar genetically (Park, 2006; Saijuntha et al., 2008), but differ in testicular morphology (with the testes occurring posteriorly) and the arrangement of the vitelline glands (Kaewkes, 2003). All three flatworms measure about 10 mm in length and have an anterior and a central ventral sucker for attachment to their host. Recent work suggests that O. viverrini is a species complex containing at least two species (Saijuntha et al., 2007; Andrews et al., 2008). As with most of the parasites discussed here, these helminths have long plagued human civilization. Archaeological records of C. sinensis come from Korea (several centimeters to 5 m below the level dated 668–935 AD; Han et al., 2003), China (Ming dynasty: 1368–1644 AD; Lun et al., 2005), and Japan (seventh century AD; Matsui et al., 2003). A corpse buried in 167 BC, the time of the Western Han dynasty in China (202 BC–24 AD), contained C. sinensis eggs in the gall

bladder which are morphologically identical to those found in recent times (Yang and Wei, 1984), but which differ genetically from current populations (Liu et al., 2007). The review by Sithithaworn et al. (2007) should be referred to for more detailed information on all species, while Lun et al. (2005) provide an excellent summary on C. sinensis.

3.12.3.1.2 Developmental cycles The life cycles of all three species of liver flukes are dependent on snail and cyprinid fish intermediate hosts inhabiting freshwater (Figure 9; Keiser and Utzinger, 2005; Sithithaworn et al., 2007). Adult C. sinensis worms are usually between 10–25 mm long and 3–5 mm wide (Lun et al., 2005). Final hosts are carnivorous mammals (such as dogs and cats) and humans eating raw or undercooked fish. Sexual reproduction occurs in these hosts with each mature worm producing between 1000 and 4000 eggs per day for at least 6 months (Lun et al., 2005). Once the operculated eggs (25–35 mm long and 15–17 mm wide; Figure 10) are released from the worms

Metacercariae in flesh or skin of fresh water fish are ingested by human host i

4 i = infective stage d = diagnostic stage

Free-swimming cercariae encyst in the skin or flesh of freshwater fish 3

5

Excyst in duodenum

Eggs are ingested by the snail 2

Miracidia

Sporocysts

Rediae

Cercariae

2a

2b

2c

2d

1

Embryonated eggs passed in feces d

6 Adults in biliary duct

Figure 9 The life cycle of a member of a liver fluke of the family Opisthorchiidae, modified by the authors of this chapter. The adult flukes deposit fully developed eggs that are passed in the feces 1 . After ingestion by a suitable snail (first intermediate host) 2 , the eggs release miracidia 2a, which undergo in the snail several developmental stages (sporocysts 2b, rediae 2c , cercariae 2d). The eyed and finned cercariae are released from the snail 3 and penetrate freshwater fish (second intermediate host), encysting as metacercariae in the muscles or under the scales 4 . The mammalian definitive host (cats, dogs, and various fish-eating mammals including humans) become infected by ingesting undercooked fish containing metacercariae. After ingestion, the metacercariae excyst in the duodenum 5 and ascend through the ampulla of Vater into the biliary ducts, where they attach and develop into adults, which lay eggs after 3–4 weeks 6 . The adult flukes (O. viverrini: 5–10 mm  1–2 mm; O. felineus: 7–12 mm  2– 3 mm) reside in the biliary and pancreatic ducts of the mammalian host, where they attach to the mucosa. Reproduced with permission from CDC, modified by the authors of this chapter.

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

319

not appear to be a common phenomenon with the high rates of infection in humans not correlating with infection rates in other potential hosts in Thailand, Taiwan, or Korea (Sithithaworn et al., 2007).

3.12.3.1.3 Human involvement

Figure 10 Egg (25–35 mm  15–17 mm) of Clonorchis sinensis. Note the operculum (arrows) through which the miracidium-larva hatches. Reproduced with permission from CDC.

inhabiting the bile ducts, they reach the intestinal lumen and are excreted with the feces. If they are subsequently washed into a freshwater body which contains suitable snail first intermediate hosts (e.g., Bithynia spp., Parafossarulus spp.), then they may be ingested by these species. Embryonated eggs that are eaten by these intermediate hosts contain the miracidium stage which transforms to a sporocyst which then undergoes asexual reproduction giving rise to rediae (Lun et al., 2005). The rediae produce large numbers of long-tailed cercariae which are shed by the snails into the water. Prevalences in the snail hosts are low (0.05–0.07% for O. viverrini and 0.09–0.60% for C. sinensis; Sithithaworn et al., 2007). The cercariae actively search for the second intermediate, cyprinid fish host which they penetrate, forming encapsulated metacercariae which are infectious to the final host on ingestion (Lun et al., 2005). Prevalences in fish, in contrast to snails, are very high reaching 90–95% for both O. viverrini and C. sinensis (Lun et al., 2005; Sithithaworn et al., 2007). On ingestion, the metacercariae hatch in the small intestine and the juvenile worms migrate into the intra-hepatic bile ducts, attaching to the epithelium. As many as 1500 worms have been recorded in a severe human infestation (Lun et al., 2005). After about 4 weeks, egg production starts. Potential survival of individual worms may be over 25 years in untreated patients (Attwood and Chou, 1978) although Sithithaworn et al. (2007) suggest that O. viverrini survives in humans for about 10 years. The life cycles of O. felineus and O. viverrini are similar to that of C. sinensis, differing mainly in geographic distribution and the choice of snail and fish hosts (Kaewkes, 2003; Sithithaworn and Haswell-Elkins, 2003; Sithithaworn et al., 2007). In Europe, there is a low prevalence of O. felineus in humans suggesting that they are not a major host in the life cycle of this species. In areas in Italy where recent human outbreaks have been reported, the cats examined showed infection prevalences ranging from 23.5% to 40% (Pozio, 2008). This does

Worldwide there are an estimated 35 million people infected with these species, with 15 million in China alone, although this is likely to be an underestimate as very little data are available on O. viverrini from Laos, Cambodia, and southern Vietnam (Lun et al., 2005; Andrews et al., 2008). The frequent contamination of freshwater with egg-bearing human fecal matter is a significant component in the life cycles of C. sinensis and O. viverrini (Sithithaworn and Haswell-Elkins, 2003). Indeed, in some regions of China where clonorchiasis is common, toilets are deliberately built adjacent to fish ponds leading to high rates of contamination (Lun et al., 2005). Thus, inadequate, unhygienic disposal of human fecal matter plays a major role in effectively maintaining the life cycle of both C. sinensis and O. viverrini (Sithithaworn et al., 2007). Human infection occurs through eating raw, marinated, or inadequately cooked fish (Sithithaworn and Haswell-Elkins, 2003), which is a dietary component of the cultural system in many of the areas affected (Petney, 2001; Lun et al., 2005). Men are often more frequently and potentially more heavily infected than women for both C. sinensis and O. viverrini (Lun et al., 2005; Sithithaworn et al., 2007), which is also likely to be related to traditional eating patterns (Petney, 2001). Prevalence is lowest in preschool children, increasing to early adulthood after which no consistent variation can be found for either C. sinensis or O. viverrini; intensity of infection for both species tends to increase with age (Sithithaworn et al., 2007). Distribution and prevalence of the infection with these fish-borne liver flukes vary substantially depending on the suitability of the environmental condition for the intermediate hosts and the frequency of consumption of raw fish. C. sinensis is present in all Chinese provinces except those in the far west of the country (Lun et al., 2005). Mean provincial prevalences range from 0.04% to over 4.5% in Sichuan, but local prevalences may be over 75%, such as those recorded from Guanyuan, Dongyoung, Sanshui, and Shunde, where raw fish is a common component of the diet. In Guangdong, 18% of the 862 393 people examined were positive for C. sinensis (Lun et al., 2005). In the Republic of Korea, prevalence of infection is highly variable, ranging from 2.1% in Chuncheon to 31.3% in Haman (Lim et al., 2006). A similar pattern is present for O. viverrini in Thailand, where the south (0%) and central (3.8%) parts of the country are relatively free of infection compared with the north (19.3%) and the northeast (15.7%) where infection is highest (Sithithaworn et al., 2007). There is also a great deal of local variation in prevalence within provinces, ranging in Khon Kaen province from 2% to 71% (Jongsuksuntigul, 2002; Sriamporn et al., 2004). Nevertheless, education and control programs have reduced the overall prevalence in Thailand from 34% in 1992 to 10% in 2002 (Jongsuksuntigul and Imsomboon, 2003). A number of recent studies have shown high prevalences in Laos (Sithithaworn et al., 2006; Sayasone

320

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

et al., 2007), but data for both Cambodia and southern Vietnam are largely lacking (Andrews et al., 2008). Epidemiological data for O. felineus are few and scattered, but rates of human infection usually seem to be low, probably due to the reduced likelihood of raw fish being consumed as a traditional part of the diet. In some areas of Byelorussia, the Ukraine, and western Siberia, however, prevalences range from 40% to 95% (see Yossepowitch et al., 2004). As it has become more fashionable to consume raw fish in Western countries, outbreaks on infection with O. felineus are becoming more common, with recent outbreaks in Italy being caused by the consumption of tench (Tinca tinca) (Armignacco et al., 2008; Pozio, 2008).

3.12.3.1.4 Disease characteristics in humans The symptoms and pathology associated with clonorchiasis depend on the intensity of the infection, its duration, the number of reinfections, and the susceptibility of the host individual (Min, 1984; Lun et al., 2005). Pathology is due to both local trauma and toxic irritation and can be divided into a number of phases eventually leading to hyperplasia, proliferation of connective tissue resulting in fibrosis of the bileduct wall within the liver (Min, 1984). These changes may eventually affect the liver generally. The onset of symptoms may be gradual or sudden. These include chills or fever up to 40 1C within a few weeks of infection followed by a potentially asymptomatic course. Should symptoms occur, these may be mild with gastrointestinal discomfort, diarrhea, anorexia, weight loss, fatigue, and hepatomegaly, or severe including portal cirrhosis and hypertension (Min, 1984). Complications can be severe, including pyogenic cholangitis, cholelithiasis, cholecystitis, and pancreatitiis (Min, 1984). Although infection with any of the three species leads directly to human morbidity (Sripa, 2003; Lun et al., 2005), to date O. viverrini is the only member to be officially recognized as a carcinogen, with long-term exposure potentially leading to cholangiocarcinoma (IARC, 1994; Meyer and Fried, 2007), although in Korea the prevalence of C. sinensis in different areas is positively correlated with incidence of this cancer (Choi et al., 2004; Lim et al., 2006). Worldwide cholangiocarcinoma is rare comprising 18.3% and 29.8% of all liver cancers in males and females, respectively, in the United States (Vatanasapt et al., 2002). The highest incidences of this cancer, however, occur in the distributional range of O. viverrini in Southeast Asia (Vatanasapt et al., 2002; Khuhaprema et al., 2007; Mathers et al., 2007). In the Khon Kaen area, liverfluke-related cancer comprises 89% of all liver cancers with an incidence of 97.4 per 100 000 for males and 39.0 per 100 000 for females, the highest in the world (Vatanasapt et al., 2002). Unfortunately, this cancer, which can eventually invade the liver, is usually not discovered in developing countries before it becomes inoperable and therefore fatal (Gores, 2003; Andrews et al., 2008). It has been responsible for millions of deaths in Southeast Asia (World Health Organisation, 2004). Although it is known that O. viverrini elicits a systemic immune response in its host, the mechanisms by which this occurs have been little investigated (Sithithaworn et al., 2007). In Korea, both the incidence and mortaliy rate due to cholangiocarcinoma are correlated with local prevalence of

C. sinensis (Lim et al., 2006). The most important risk factors for contracting cancer are being male, alcohol consumption, eating freshwater fish, and area of residence of which the last is the most significant (Lim et al., 2006). For O. felineus, the incubation period ranges from 2 to 4 weeks (Pozio, 2008). Infection with this species often begins with an acute infection characterized by high fever, nausea, abdominal pain, myalgia, and eosinophilia before becoming chronic (Sithithaworn et al., 2007; Armignacco et al., 2008). To date, the evidence implicating O. felineus with cholangiocarcinoma is insufficient to draw any definite conclusions (Watanapa and Watanapa, 2002); although some studies suggest that this is likely (Ruditzky, 1928; Bohl and Jakowlew, 1931; Ilyinskikh et al., 1998; Yossepowitch et al., 2004). Diagnosis of infection with opisthorchids is usually accomplished by detection of eggs in fecal samples (Khandelwal et al., 2008). This method, however, has low sensitivity and specificity, particularly in cases with light infections, and relies on the skill of the microscopist in indentifying the egg (Duenngai et al., 2008). Recently, PCR-based methods have been developed which substantially improve diagnosis (Duenngai et al., 2008). The presence of eggs in fecal samples is diagnostic for infection but care must be taken with identification due to the similarity of eggs to those of other food-borne trematodes (Sithithaworn et al., 2007).

3.12.3.1.5 Prevention and cure Control or prevention of food-borne trematode disease in human populations can be brought about by improved sanitation, health education, including dietary advice, and treatment (World Health Organization, 1995a; Jongsuksuntigul and Imsomboon, 2003). Health education, aimed at explaining the significance of the disease, how it is transmitted, the likelihood of reinfection, and how it can be controlled, particularly the role of eating raw fish, plays an important role in disease control and prevention (Jongsuksuntigul and Imsomboon, 2003). Effective prevention occurs when the consumption of raw or undercooked fish is stopped. In the past, this has proven to be difficult because of the traditional place of this food in the cultures of the people affected (Petney, 2001). Evidence from Thailand suggests that education programs were responsible for a decrease in the frequent consumption of fish from 14% to 7% between 1990 and 1994 (Jongsuksuntigul and Imsomboon, 1997). In Laos, an education program and treatment were associated with a reduction in disease prevalence from 62% to 34%. At the end of the education program, the village population understood the relationship between liver disease and the consumption of raw fish. In a control village without education, this was not the case (Strandgaard et al., 2008). Nevertheless, antihelminthic treatment needs to be provided regularly as infection rates can return to the original high levels within a year after initial treatment (Sornmani et al., 1984; Upatham et al., 1988). Treatment with a single or double dose of praziquantel is effective in elimination of adult worms in 98–100% of cases (Lun et al., 2005; Sithithaworn et al., 2007). In a recent outbreak in Italy, fish containing the metacercariae had been frozen at  10 1C for 3 days. Freezing at

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

this temperature should be prolonged and requires 5–7 days to be effective at killing the metacercariae but freezing at  28 1C is effective within 24 h (Armignacco et al., 2008).

321

time, however, conditions may become suitable for the snails, and prevalences can again rise (Morley, 2007).

3.12.3.1.7 Recommendations 3.12.3.1.6 Anthropogenic alterations to the environment The complicated three-host life cycle of this group makes predications on changes in prevalence due to anthropogenic manipulation of the environment difficult. Range extensions and increases in abundance are dependent not only on suitable environmental conditions, but also on the presence of suitable snail and fish intermediate hosts. Sithithaworn et al. (2007) speculate that the increase in prevalence and transmission of C. sinensis in China may be due to the concurrent increase in aquaculture in the affected areas. Silver carp (Hypophthalmichthys molitrix) from aquaculture ponds in Vietnam had C. sinensis prevalences of 45% (Garrett et al., 1997). The freshwater gastropod Melanoides tuberculatus, which thrives in aquaculture ponds and has been introduced from its endemic range in Africa and Asia to much of the tropical and subtropical Americas, is a potential first intermediate host for C. sinensis (Lun et al., 2005; De Kock and Wolmarans, 2009), which is also found in aquaculture ponds (Garrett et al., 1997). This would have significant implications if C. sinensis was introduced, either in fish or in humans, into areas colonized by this species. Aquaculture products from O. viverrini endemic countries have shown an immense expansion in recent years (Keiser and Utzinger, 2005), including a variety of cyprinid species (Naylor et al., 2000). As O. viverrini metacercariae are resilient and can remain viable in fish muscle even if the fish is pickled or fermented (World Health Organization, 1995a), the export and import of fish and fish products worldwide provide a potential transfer route for these parasites. In addition, migrant workers or refugees coming from endemic areas are known to have carried infection with them (Schwartz, 1986; Peng et al., 1993). In Laos, the heightening of roads above flood levels has led to the creation of large numbers of ponds in paddy fields as well as in home gardens (Haylor et al., 1997; Bush, 2004). These ponds harbor both the snail and cyprinid fish intermediate hosts of O. viverrini (Sithithaworn et al., unpublished data). The fish are eaten raw or marinated leading to human infection, while lack of adequate, hygienic toilet facilities completes the epidemiological cycle. The predicted increase in temperatures for Southeast Asia (Christensen et al., 2007) is likely to reduce the developmental time of immature stages living in ectothermic hosts and potentially reduce the time available for cercariae to find a host (Poulin, 2005; Hudson et al., 2006), while the predicted increase in rainfall could lead to an increase in habitat suitable for the intermediate hosts. It is thus not clear how the climatic changes will affect the epidemiology of these parasites. Dam construction is known to have a substantial impact on the prevalence/presence of O. felineus. The construction of dams in the former Soviet Union initially led to a substantial reduction in the infection in second intermediate host fish due to the unsuitability of the new habitat for the first intermediate host snails (Potseluev, 1991; Morley, 2007). With

Recommendations follow those of Jongsuksuntigul and Imsomboon (2003) for O. viverrini and Lun et al. (2005) for C. sinensis both of which aim at breaking the cycle of transmission. From the point of view of the freshwater component of the life cycle, this involves the improvement of sanitary condition, and preventing the contamination of ponds and streams which contain snail and cyprinid intermediate hosts with infected fecal material. Toilets built alongside or even over fish ponds should be eliminated (Lun et al., 2005). In addition, all potential food sources of infection should be cooked before consumption. Pharmaceutical treatment of infection can reduce the number of eggs being introduced into freshwater bodies, as can effective education programs aimed at reducing the consumption of raw fish. However, due to the presence of reservoir hosts, the opportunities for completely interrupting the transmission cycle are limited.

3.12.3.2 Intestinal Flukes 3.12.3.2.1 Parasite characterization About 70 species of intestinal trematodes are known to infect humans worldwide (Yu and Mott, 1994); most of these belong to the families Heterophyidae and Echinostomatidae, although zoonotic representatives are found in at least 11 other families (Yu and Mott, 1994; Fried et al., 2004). Of these, 59 species occur in Southeast Asia (Chai et al., 2009). About 35 species contracted by eating raw or insufficiently prepared fish are significant for human health, potentially causing morbidity but rarely mortality, although many infections are probably asymptomatic (Yu and Mott, 1994; Chai et al., 2005). These are usually less well studied than opisthorchid and fasciolid liver flukes discussed elsewhere in this chapter (Yu and Mott, 1994; Graczyk and Fried, 1998). Nevertheless, most parts of the world have representatives of intestinal flukes infecting humans. Infection with metacercariae can occur via a variety of snail, crustacean, and other invertebrate hosts as well as fish, amphibian, and reptile vertebrate second intermediate hosts (Figure 11; Fried et al., 2004). Most species commonly parasitizing the human gut belong to the family Heterophylidae. They are related to the Opisthorchiidae, also possessing testes in the posterior part of their bodies, but unlike the latter, the hermaphrodite genital pore is surrounded by a specific structure called the gonotyl, often forming a genital sucker ornamented with tipped rodlets (Lucius and Loos-Frank, 2008).

3.12.3.2.2 Developmental cycle Minute intestinal flukes usually do not exceed 2–3 mm in length; accordingly, their eggs are usually a little less than 20 mm long (Figure 12). Like opisthorchids, all heterophyids use freshwater, prosobranch snails as first intermediate hosts inhabiting fresh, brackish, or seawater (Figure 11; Lucius and Loos-Frank, 2008). These consume the parasite’s eggs (Figure 12), excreted with the feces of the host, which contain a fully developed miracidium (Figure 12). The further development follows the same pattern as described for

322

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

5 Cercariae penetrate the skin of fresh/brackish water fish and encyst as metacercariae in the tissue of the fish

4

Host becomes infected by ingesting undercooked fish containing metacercariae i

8 6

Metacercariae excyst in the small intestine

3 Cercariae released from snail Fish-eating mammals and birds can be infected as well 2 Snail host ingests eggs, miracidia emerge from eggs and penetrate the snail’s intestine

2a Sporocysts

2b Rediae in snail tissue

2c Cercariae

7 Adult in small intestine

d Embryonated eggs each with 1 a fully developed miracidium are passed in feces

i

= infective stage

d = diagnostic stage

Figure 11 Life cycle of the minute intestinal fluke Heterophyes spp. Adults release embryonated eggs each with a fully developed miracidium, and eggs are passed in the host’s feces 1 . After ingestion by a suitable snail (first intermediate host), the eggs hatch and release miracidia which penetrate the snail’s intestine 2 . Pirenella conica and Cerithidea cingulata are the snail hosts in the Middle East and Asia, respectively. The miracidia undergo several developmental stages in the snail, that is, sporocysts 2a , rediae 2b , and cercariae 2c . Many cercariae are produced from each redia. The cercariae are released from the snail. They have a finned tail and two eye spots resembling those of Clonorchis species 3 , and encyst as metacercariae in the tissues of a suitable euryhaline (brackish water) fish (second intermediate host) 4 . The definitive host becomes infected by ingesting undercooked or salted fish containing metacercariae 5 . After ingestion, the metacercariae excyst, attach to the mucosa of the small intestine 6 and mature into adults (measuring 1.0–1.7 mm  0.3–0.4 mm) 7 . In addition to humans, various fish-eating mammals (e.g., cats and dogs) and birds can be infected by Heterophyes species 8 . Reproduced with permission from CDC, modified by the authors of this chapter.

Figure 12 Eggs of Heterophyes heterophyes (c. 24 mm  14 mm) from the uterus of a gravid worm collected from the gut of a dog. From Taraschewski (unpublished).

opisthorchiids. The rediae shed a huge number of fin-tailed, eyed cercariae which actively swim in search of a suitable fish second intermediate host (Taraschewski, 1984). While penetrating the fish, the cercaria loses its finned tail, forming a metacercaria. On ingestion, this encapsulated, dormant larva is infectious to suitable final hosts, where it attaches to the intestinal mucosa with its ventral sucker, feeding on the outer layer of the intestinal wall using its oral sucker. The habitat where transmission occurs depends on the species involved. Heterophyids are found in water with a wide range of salinities. The miracidia and cercariae of Metagonimus yokogawai, are exposed to pure freshwater, while others, such as Stellanthchasmus falcatus, inhabit brackish water. The free-living stages of the genus Heterophyes also tolerate high salinities. H. heterophyes, which occur in mullet (a euryhaline fish of the family Mugilidae), in Egypt, tolerates salinities ranging from 20% to 60%, which corresponds with the euryhaline habitat

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

of the intermediate snail host, Pirenella conica. These mullet are traditionally eaten raw as salted fessikh (Taraschewski, 1984). For M. yokogawai and Haplorchis taichui, the consumption of uncooked, infected, cyprinid freshwater fish, in which the metacercariae are found, is the route of infection to humans (Yu and Mott, 1994). In a study in northern Thailand, Kumchoo et al. (2005) found all 15 species of cyprinid fish examined infected with Haplorchis taichui, with 540 of a total of 615 fish being infected. Heterophyid flukes, as well as members of the Echinostomatidae, parasitize birds and mammals, including dogs and cats, with species-specific host preferences. Heterophyid infections are usually zoonotic, and although in the Nile Delta, humans seem to be the major hosts, recent data are scarce (Taraschewski, 1984; Elsheikha and Elshazly, 2008). Not all intestinal flukes have three-host life cycles. Fasciolopsis buski cercariae encyst on the surface of aquatic plants (similar to Fasciola hepatica discussed later) (Fried et al., 2004). The resultant metacercariae are transmitted to humans by ingestion of raw or undercooked aquatic plants or by handling such plants resulting in the oral ingestion of metacercariae, or by drinking contaminated water. Fasciolopsis buski (family Fasciolidae) is the largest fluke parasitizing humans, reaching a length of approximately 8–10 cm and a width of 1–3 cm (Fried et al., 2004).

3.12.3.2.3 Human involvement Intestinal food-borne trematodes are acquired by eating raw or undercooked fresh, or brackish water or marine organisms, such as mollusks, crustaceans, insect larvae, squid, fish, and amphibians, which contain the infectious metacercariae (Fried et al., 2004). This includes the consumption of raw (koi pla) or fermented (pla som, pla ra) fish in the northeast of Thailand (Sithithaworn et al., 2007), and raw fish dipped in salt and vinegar (kinilaw) or snails (kuhol, kiambu-ay) in the Philippines (Belizario et al., 2007). The prevalence of intestinal fluke infection in human populations is variable (Hinz, 1996), but can be well above 50% in some areas (Bundy et al., 1991; Radomyos et al., 1998; Belizario et al., 2004), ranging to 100% in the Nghia Hung district of Vietnam where Haplorchis spp. made up 90.4% of all worms recovered (Dung et al., 2007). In such areas, the prevalence of heterophyid metacercariae in susceptible local fish species may be more than 80% with mean intensities between 100 and 250 metacercariae per fish (Kumchoo et al., 2005). Minute intestinal heterophyid flukes are widely distributed throughout the world, but human infections only occur commonly where fish is customarily consumed raw: in East Asia (M. yokogawai, Heterophyes nocens, and others), Southeast Asia (Haplorchis spp. and others), and Egypt (H. heterophyes). The highest heterophyid diversities have been recorded from Korea where 12 species of Heterophyidae and four species of Echinostoatidae and three species from other families have been found to infect humans (Chai and Lee, 2002), and from Thailand (23 indigenous species) (Nawa et al., 2005). In Tokyo, one of the most commonly found human parasite infections is due to Metagonimus yokogawai which is ingested with the local sushi and sashimi dishes (Nawa et al., 2005). Interestingly, along the Persian Gulf, the local population is

323

not affected but expatriates from Korea were found to harbor H. heterophyes and H. dispar (Chai et al., 1986). It seems likely that the increased consumption of aquatic organisms in some countries and their export on a worldwide basis, the increasing popularity of eating raw and undercooked food, as well as growing levels of international travel have led to an increase in the incidence of infection with these helminths (Fried et al., 2004).

3.12.3.2.4 Disease characteristics in humans Disease characteristics vary depending on the species of trematode involved. Infection with members of the family Heterophyidae usually causes a mild inflammatory reaction; however, high worm burdens are associated with a more severe pathology including diarrhea, abdominal pain, anorexia, nausea, and vomiting (Taraschewski, 1984). Eggs may invade the mesenteric lymphatic system (Yu and Mott, 1994; Fried et al., 2004), although many questions on the method of invasion and pathology remain open (Elsheikha, 2007). H. taichui has been shown to imbed itself deeply in the human intestinal mucosa with the severity of the damage being proportional to the worm burden, as in other helminth species (Sukontason et al., 2005). For the Echinostomatidae, the symptoms depend on the intensity of infection and can lead to different degrees of focal necrosis and inflammation of the intestinal mucosa (Yu and Mott, 1994). Heavy infections may cause eosinophilia, abdominal pain, severe diarrhea, anemia, and anorexia (Fried et al., 2004). Diagnosis involves microscopic examination of stool samples for eggs, although the morphological similarity of the eggs makes an accurate diagnosis, to species level, difficult (Fried et al., 2004).

3.12.3.2.5 Prevention and cure Abdussalam et al. (1995) have detailed the measures required to control food-borne trematode infections. The two major phases include prevention of contamination of food with infective stages (metacercariae), and inactivation of the metacercariae that have entered the food. As with our discussion on the measures for the prevention of Clonorchis sinensis and Opisthorchis spp., both aim at blocking or at least reducing the rate of transmission. Prevention can be effected by eliminating intermediate hosts, such as freshwater snails, but such methods cannot be carried out on such a widespread basis to have more than local, temporary success, and, in addition, can cause potentially unwanted changes in the local environment (El-Sayed, 2001). Reducing infection of the snails by sewage treatment and better hygienic conditions will reduce the egg input from humans into the transmission cycle but not eggs from natural-reservoir hosts such as dogs and cats. In addition, education to avoid the consumption of infected food by deep freezing or heating to destroy the infective stages can prevent human infection (Fried et al., 2004). The usual method of treatment of adult intestinal fluke infection is with praziquantel (Fried et al., 2004).

3.12.3.2.6 Anthropogenic alterations to the environment The presence of intestinal trematodes from a variety of families has been confirmed in introduced aquaculture fish species in

324

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Southeast Asia where it presents a potential risk to humans both locally and as an export item (Hop et al., 2007; Thien et al., 2007, 2009). For example, the tilapia, Oreochromis niloticus, originating from Africa which is now cultivated worldwide in subtropical countries, is known to host metacercariae from various species of intestinal trematodes (Costa-Pierce, 2003; Hop et al., 2007; Thien et al., 2009). The use of human feces to fertilize fishponds provides a direct route for first intermediate host infection (Graczyk and Fried, 1998). The presence of domestic animals such as dogs, cats, and pigs as reservoir hosts for these parasites is likely to play a role in their transmission success. In a fish-farming community in Vietnam, fecal samples from 46.6% of cats, 35% of dogs, and 14.4% of pigs contained fishborne zoonotic intestinal trematode eggs including human pathogens such as H. taichui and H. yokogawai (Anh et al., 2009). Several heterophyids naturally occurring in the temperate Old World have been introduced to the Americas (Scholz et al., 2001). This happened following the dispersal of the freshwater/brackish water snail Melanoides tuberculata, which is possibly the most invasive aquatic gastropod known, and which benefits from the rapid growth of pond aquaculture. This species spreads along international trade routes with aquaculture organisms. Furthermore, M. tuberculata was intentionally introduced in South America and certain Caribbean Islands in order to outcompete and exclude the native Biomphalaria species which transmit schistosomiasis (see below, Taraschewski, 2006). The introduction of intermediate hosts of intestinal trematodes can lead to substantial problems in both cultured and native fish species (Mitchell et al., 2002). In Asia as well as in North America and Hawaii, the metacercariae of the gill trematode Centrocestus formosanus has led to losses among both cultured and wild fish species by causing severe gill damage due to high metacercarial burdens (Mitchell et al., 2000, 2005). In the United States, the introduction of a suitable snail intermediate host, Melanoides tuberculatus, was responsible for the spread of this species (Mitchell et al., 2000, 2005). This snail is also likely to be responsible for the introduction of C. formosanus and/or Haplorchis pumilio, which are potentially pathogenic to humans as well as fish, into the United States, Mexico, Brazil, and Venezuela (Radomyos et al., 1983; Giboda et al., 1991; Scholz et al., 2001; Boge´a et al., 2005; Sommerville, 2006; Diaz et al., 2008). In a study from Vietnam, heterophyid infections in fish raised in peri-urban wastewater-fed freshwater aquaculture systems were compared to specimens from units without wastewater influx. The latter showed heavier parasite infections (Hop et al., 2007). Data on the levels of oxygen and environmental toxins in the wastewater-fed ponds, however, were not provided. Nor was the density of snail intermediate hosts determined. Thus, whether the wastewater effected the snail population or the parasites themselves remains to be determined.

3.12.3.2.7 Recommendations The wide variety of intestinal flukes, together with their different intermediate and final hosts, makes recommendations difficult. In general, however, our recommendation for

C. sinensis should also be followed for this group. In endemic areas, control programs must take into account domestic animals as potential reservoir hosts (Anh et al., 2009). The introduction of M. tuberculata, and potentially of other intermediate hosts, to new areas will assist the spread of these parasites and should, if possible, be prevented.

3.12.3.3 Paragonimiasis 3.12.3.3.1 Parasite characterization The genera Paragonimus and Euparagonimus belong to the trematode family Paragonimidae. They contain over 50 species of flukes inhabiting the lungs, of which about 10 species are potentially harmful to humans (Doanh et al., 2007). These are distributed over much of East, South, and Southeast Asia as well as in parts of South and North America including the United States (DeFrain and Hooker, 2002) and Africa, with human infections being found in all of these areas (Blair et al., 1999; Liu et al., 2008). Final hosts include a variety of rodent species and also carnivores and primates. Domestic dogs and cats were found to be highly infected with prevalences reaching 84.6% and 66.5%, respectively, in certain areas of China (Liu et al., 2008). The prevalence of the most common human parasite Paragonimus westermani (the oriental lung fluke) and of congeners in samples of the crab intermediate hosts can reach high levels: Ecuador to 76.5% (Viera et al., 1992), Columbia 50% (Ve´lez et al., 2000), Laos 59% (Odermatt et al., 2007), and between 7.9% and 85.4% in various areas of China (mean 51.2%; Liu et al., 2008). Natural foci of infection are commonly found around small streams with waterside vegetation, which are inhabited by suitable intermediate hosts, and are frequented by mammals, for drinking purposes as well as for defecation (Ve´lez et al., 2000).

3.12.3.3.2 Developmental cycle Adult worms (15  8  5 mm), which are relatively short and compact with an oval shape and have cuticular spines, occur in the lungs of their mammalian hosts. Eggs are released into the environment either by sputum or, if swallowed, in the feces (Figure 13; Soh, 1962; Blair et al., 1999; Lucius and Loos-Frank, 2008). The dark brown eggs of P. westermani are approximately 90  55 mm in length (Figure 14). They hatch to miracidia in fresh or brackish water which actively swim to find a first intermediate snail host. A number of prosobranch genera, including Melanoides, Thiara, and Semisulcospira, have been incriminated as suitable intermediate host species (Lucius and Loos-Frank, 2008). After asexual reproduction involving a progression from sporocysts to rediae to cercariae, the second intermediate crab or freshwater crayfish host can be infected either by active-seeking cercariae or by ingestion of the first intermediate host (Liu et al., 2008). The cercariae encyst in this host to metacercariae. In China, 80 species of crab (Brachiura) belonging to five families are recognized as second intermediate hosts (Liu et al., 2008). If an inadequately cooked infected crab is eaten by the human final host, infection takes place when the metacercariae excyst in the duodenum. The immature worms then penetrate the intestinal wall 3–6 h after infection, migrate through the abdominal cavity, penetrate the diaphragm, and move into the pleural

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

5 4b

4a

Sporocysts

Rediae

Cercariae invade the crustacean and encyst into metacercariae

4c

i i

= infective stage

d

= diagnostic stage

325

Cercariae Humans ingest 6 inadequately cooked or pickled crustaceans containing metacercariae

4

7

Excyst in duodenum 3 Miracidia hatch and penetrate snail

8

2 Embryonated eggs d 1 Unembryonated eggs

Adults in cystic cavities in lungs lay eggs which are excreted in sputum. Alternately, eggs are swallowed and passed with stool

Figure 13 The life cycle of Paragonimus spp. The eggs are excreted unembryonated in the sputum, or alternately, they are swallowed and passed with stool 1 . In the external environment, the eggs become embryonated 2 , and miracidia hatch and seek the first intermediate host, a suitable snail, and penetrate its soft tissues 3 . Miracidia go through several developmental stages inside the snail 4 : sporocysts 4a , rediae 4b, with the latter giving rise to many cercariae 4c , which emerge from the snail. The cercariae invade the second intermediate host, a crustacean, such as a crab or crayfish, where they encyst and become metacercariae. This is the infective stage for the mammalian host 5 . Human infection with Paragonimus species occurs by eating inadequately cooked or pickled crab or crayfish that harbor metacercariae of the parasite 6 . The metacercariae excyst in the duodenum 7 , penetrate through the intestinal wall into the peritoneal cavity, then through the abdominal wall and diaphragm into the lungs, where they become encapsulated and develop into adults 8 (7.5–12 mm  4–6 mm). The worms can also reach other organs and tissues, such as the brain and striated muscles, respectively. However, when this takes place, completion of the life cycles is not achieved because the eggs laid cannot exit these sites. Time from infection to oviposition is 65–90 days. Infections may persist for 20 years in humans. Animals such as pigs, dogs, and a variety of feline species can also harbor lung flukes. Reproduced with permission from CDC.

cavity. Immature worms can also be ingested by eating undercooked pork from wild boar which can act as paratenic hosts (Liu et al., 2008). Adult worms are found in cystic cavities in the lungs where they lay their eggs starting 65–90 days after infection (Liu et al., 2008). P. skrjabini, which is found predominantly in China, differs in behavior from P. westermani. Although it has a similar life cycle, only few individuals reach the lungs and develop into adults, with most juveniles migrating to other organs such as the muscles, liver, and brain where they cause ectopic lesions (Hu et al., 1982; Cui et al., 1998; Liu et al., 2008).

3.12.3.3.3 Human involvement A number of species can cause human infection with the most medically important species being P. westermani from Asia. Prevalences in humans vary but can be substantial, reaching up to 22.75% of the 17–22 age group of the population of the Cross River basin area of Nigeria (caused by Paragonimus uterobilateralis; Arene et al., 1998), and 20.9% in children under 15 years of age in a hyper-endemic area of Arunachal Pradesh in India, while antigen positivity against the excretory–secretory protein of adult worms was 51.7%, indicating contact with the parasite (P. hererotremus; Devi et al., 2007). Summarizing the data for China, Liu et al. (2008)

326

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control 3.12.3.3.5 Prevention and cure As with all infectious agents that are ingested with undercooked fresh- and brackish water, or marine invertebrates or vertebrates, the transmission cycle can be broken by educating the populations of infected areas to heat their food sufficiently to denature protein and thus kill the metacercariae before consumption. This is also true for immature worms from paratenic hosts. The World Health Organization (WHO) recommends the use of either praziquantel, the drug of choice, or triclabendazole for the treatment of human paragonimiasis (World Health Organization, 2004).

3.12.3.3.6 Anthropogenic alterations to the environment

Figure 14 Eggs (c. 90 mm  55 mm) of a Paragonimus species. Reproduced with permission from CDC.

found reported prevalences between 1.5% and 33.7% in various Chinese provinces. There is currently a steady increase in the number of people infected with P. westermani in Japan which may be partly related to immigration from Southeast Asia and China (Obara et al., 2004). The tradition of these immigrants of eating raw freshwater food, including crustaceans, may lead to increased rates of autochthonous infections contracted within Japan (Takagi et al., 2009). Food-preparation methods other than heating, such as pickling, marinating, or salting, are not effective at killing the metacercariae (Yokogawa, 1965).

3.12.3.3.4 Disease characteristics in humans Following the ingestion of infected intermediate hosts, the progression of the disease is slow. The first symptoms for pulmonary infections include cough, chest pain, dyspnea, blood-tinged sputum, and, occasionally, fever (Obara et al., 2004; Liu et al., 2008). In Laos, where both P. westermani and P. heterotremus are known to infect humans, 12.7% of patients presenting with a chronic cough were infected by Paragonimus sp. (Odermatt et al., 2007). Extra-pulmonary paragonimiasis can also occur, although this is less common than the pulmonary form. It most commonly involves the central nervous system with invasion of the brain potentially leading to seizures, epilepsy, motor and sensory disturbances, and other neurological syndromes (Mac et al., 2007; Liu et al., 2008). Other sites potentially affected include the eyes (Wang et al., 1984), urinary bladder, skin, liver, and the pericardial area (Liu et al., 2008). The severity of disease is dependent on the length and intensity of infection (Liu et al., 2008). Paragonimiasis can be diagnosed by finding eggs in stool or sputum samples, although these are not present until 2–3 months after infection (Liu et al., 2008). Effective immunological tests are also available (Lee et al., 2007).

Economic globalization processes, such as aquaculture with the import of potentially parasitized crustaceans (Keiser and Utzinger, 2005), are likely to have been responsible for the recent finding of the first human infection in Brazil (Lemos et al., 2007). It has been suggested that the construction of the Three Gorges Dam in China will lead to an increase in human infection rates with P. skrjabini as the habitat provided by the dam will favor the population growth of suitable crab second intermediate hosts (Morley, 2007).

3.12.3.3.7 Recommendations The problem of paragonimiasis can be addressed at two levels: (1) by breaking or inhibiting the transmission cycle which can be effected by appropriate, hygienic sewage-disposal practices and by educating the population to reduce contamination of freshwater sources with saliva and to eat only sufficiently heated water-based food products and (2) by treating the affected population to reduce the number of eggs input into the environment. For paragonimiasis, unlike infection with Opisthorcis viverrini with humans as the dominant final hosts, the presence of a variety of zoonotic hosts, especially with companion animals living integrated in human communities, there is an alternative source of introducing large numbers of eggs into water bodies, which complicates control possibilities.

3.12.3.4 Diphyllobothriosis 3.12.3.4.1 Parasite characterization The zoonotic genus Diphyllobothrium (class Cestoda, order Pseudophyllidea) contains up to 14 species of tapeworm which occasionally infect humans, of which D. latum occurs most commonly (Scholz et al., 2009). These fish tapeworms occur throughout the Northern Hemisphere (Dick et al., 2001) as well as in some countries south of the equator such as Peru (Baer et al., 1967), Chile (Torres et al., 2004a, 2004b), Brazil (Tavares et al., 2005), and Argentina (Revenga, 1993). The distribution of the various pathogenic species differs; for example, D. dendriticum occurs with a circumpolar distribution at high latitudes above the distribution of D. latum (Curtis and Bylund, 1991). The two latter species, as well as D. nihonkaiense, have recently invaded and partly colonized new continents (Torres et al., 2004a, 2004b; Arizono et al., 2009).

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control 3.12.3.4.2 Developmental cycle The natural life cycle includes fish-eating birds and mammals as definitive hosts, and freshwater or marine copepods and fish as first and second intermediate hosts, respectively (Figure 15; Scholz et al., 2009). The bodies of tapeworms are long and flat, consisting of a scolex (head) with two slit-like attachment grooves (bothria), a proliferation zone (neck), and a strobila with many proglottids (Figure 16). Diphyllobothrium spp. reach a length of up to 25 m with as many as 4000 hermaphrodite segments. The growth rate may be as high as 22 cm d1 and the adult worms can live 20 years or longer (Scholz et al., 2009). The eggs, which are operculate (35–80 mm long and 25–65 mm wide depending on species,

Infected crustacean ingested by small freshwater fish Procercoid larva released from crustacean, develops into plerocercoid larva 5

327

Figure 17) are excreted with the feces after which they mature within 2–3 weeks. In pseudophyllid cestodes, the onchosphere larva hatching from the egg is surrounded by a ciliated epithelium which enables the larva to swim about in order to attract a copepod. On ingestion by such a first intermediate host, the coracidium develops into procercoid larvae. If the copepod is in turn eaten by a small freshwater fish, the procercoid is released and migrates into fish muscle where it develops into a plerocercoid larva, the stage which is infective to humans. Diphyllobothrium species differ in their preference for the second intermediate host. Larvae of D. latum are usually found in perch and pike, whereas D. dendriticum occurs more commonly in salmonoid fish (Curtis and Bylund, 1991).

Predator fish eats 6 infected small fish

i

= infective stage

d = diagnostic stage

i

Procercoid larvae in body cavity of crustaceans

Human ingests 7 raw or undercooked, infected fish

4

Scolex

8 Coracidia hatch from eggs and are ingested by crustaceans 3

Adults in small intestine

d 2

Eggs embryonate in water

1 Unembryonated eggs passed in feces

Proglottids release 9 immature eggs

Figure 15 The life cycle of Diphyllobothrium latum. Immature eggs are passed in feces 1 . Under appropriate conditions, the eggs mature (approximately 18–20 days) 2 and yield oncospheres which develop into ciliated coracidia 3 . After ingestion by a suitable freshwater crustacean (the copepod first intermediate host), the coracidia develop into procercoid larvae 4 . Following ingestion of the copepod by a suitable second intermediate host, typically minnows and other small freshwater fish, the procercoid larvae are released from the crustacean and migrate into the fish flesh where they develop into a plerocercoid larvae (sparganum) 5 . The plerocercoid larvae are the infective stage for humans. As humans do not generally eat undercooked minnows and similar small freshwater fish, these do not represent an important source of infection. Nevertheless, these small second intermediate hosts can be eaten by larger predator species, for example, trout, perch, and pike 6 . In this case, the sparganum can migrate to the musculature of the larger predator fish and humans can acquire the disease by eating these later intermediate infected host fish raw or undercooked 7 . After ingestion of the infected fish, the plerocercoid develop into immature adults and then into mature adult tapeworms which will reside in the small intestine. The adults of D. latum attach to the intestinal mucosa by means of the two bilateral groves (bothria) of their scolex 8 . The adults can reach more than 10 m in length, with more than 3000 proglottids. Immature eggs are discharged from the proglottids (up to 1 000 000 eggs per day per worm) 9 and are passed in the feces 1 . Eggs appear in the feces 5–6 weeks after infection. In addition to humans, many other mammals can also serve as definitive hosts for D. latum. Reproduced with permission from CDC.

328

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Figure 16 Diphyllobothrium egg; size c. 90 mm  55 mm. Note that the pseudophyllid tapeworm egg has an operculum (arrows), similar to that of trematode eggs, through which the coracidium larva escapes from the shell. Courtesy of Dr. Yves Jackson, Travel and Migration Medicine Unit, Geneva University Hospitals.

Humans are seldom infected by eating small uncooked fish; rather, infection occurs by eating larger predatory fish which have themselves consumed the smaller prey fish species in which the plerocercoid larvae have migrated into muscle tissue. If a human ingests such tissue, the plerocercoid larva develops into an immature, followed by a mature, tapeworm in the small intestine. Adult worms can produce more than a million eggs per day (Scholz et al., 2009), and D. latum proglottids containing eggs are visible in the stool around 15–45 days after fish consumption (Jackson et al., 2007; Scholz et al., 2009). D. latum has been reported to survive in humans for up to 25 years (Scholz et al., 2009). Prevalences in fish may be high; Oshima and Wakai (1983) report from Japan that the yearly infection rates for cherry salmon (Oncorhynchus masou) ranged from 15.9% to 48.8%. In Argentina, introduced rainbow (Oncorhynchus mykiss) and brook trout (Salvelinus fontinalis) were heavily infested with both D. latum (28% and 9%, respectively) and D. dendriticum (58% and 27%, respectively) (Revenga, 1993).

3.12.3.4.3 Human involvement A recent estimate suggests that up to 20 million people are infected worldwide (Chai et al., 2005). Human infections are generally associated with cold waters, with Finland and Alaska being the most affected areas (Scholz et al., 2009). Nevertheless, diphyllobothriasis, caused by a member of the family Diphyllobothriidae, has also been reported from tropical areas such as southern India (Pancharatnam et al., 1998) and Malaysia (Rohela et al., 2002). There is evidence, however, that human disease rates are declining in the United States, Asia, and most of Europe, although incidence in Russia, South Korea, Japan, and South America seem to be increasing (Scholz et al., 2009). In Europe, D. latum has reappeared in lakes around the Alps (Terramocci et al., 2001; Jackson et al., 2007). The association between Diphyllobothrium tapeworms and humans is ancient, with the eggs of Diphyllobothrium having

Figure 17 Diphyllobothrium latum: plerocercoid larva in the musculature of a pike from Como Lake, Italy. From Scholz T, Garcia HH, Kuchta R, and Wicht B (2009) Update on the human broad tapeworm (genus Diphyllobothrium) including clinical relevance. Clinical Microbiological Reviews 22: 146–160 and courtesy of the American Society of Microbiologists.

been found in coprolites or other human remains from Europe, North and South America, and the Middle East (Goncalves et al., 2003). Eggs from Peruvian mummies dated between 4000 and 5000 years old were recently identified as belonging to D. pacificum (Reinhard and Urban, 2003). The route of infection involves the consumption of uncooked, inadequately cooked, or smoked freshwater or marine fish (including sushi and sashimi), the intermediate hosts of Diphyllobothrium (Terramocci et al., 2001; Nawa et al., 2005). In Switzerland, marinated but uncooked perch (Perca fluviatilis) filets were consumed by 26 guests at a wedding, seven of whom had a confirmed D. latum infection with one additional case being probable (Jackson et al., 2007).

3.12.3.4.4 Disease characteristics in humans In many cases, infection is asymptomatic; however, diarrhea and abdominal pain occur in about 20% of cases and prolonged heavy infection may lead to intestinal obstruction, cholecystitis, or cholangitis (King, 2005; Scholz et al., 2009). Other symptoms reported include constipation, fatigue,

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

headache, and occasionally allergic reactions (Scholz et al., 2009). In addition, anemia due to vitamin B12 deficiency has been reported (Stabler and Allen, 2004). Diagnosis is usually based on the recovery of eggs from fecal samples but molecular techniques are also available (Scholz et al., 2009). In general, the parasite is only identified to the genus level using egg morphology (Scholz et al., 2009).

3.12.3.4.5 Prevention and cure As nonhuman final hosts such as dogs, cats, foxes, and wild pigs are also involved in the transmission cycle, sewage treatment alone cannot eliminate the disease (Curtis and Bylund, 1991). Avoiding eating raw fish means that transmission cannot take place. Cooking fish for at least 5 min at 55 1C kills the larvae, as does freezing at  20 1C for at least 24 h (Jackson et al., 2007). If an infection occurs, praziquantel is the drug of choice with niclosamide a possible alternative (Scholz et al., 2009).

3.12.3.4.6 Anthropogenic alterations to the environment The extensive increase in aquaculture with a subsequent transfer of fish for culture purposes could lead to transfer of infected individuals to new habitats. It has been suggested, for example, that cases of diphyllobothriasis in Brazil, where the disease was previously unknown, were caused by eating imported salmon from aquaculture production (Tavares et al., 2005; Cabello, 2007). In Chile, the establishment of salmon aquaculture coincided with the appearance and increase in abundance of diphyllobothriasis. Prevalences and mean intensities of D. latum and D. dendriticum are higher in introduced rainbow trout than in native fish species (Torres et al., 2004b; Cabello, 2007). Data from Russia, including Siberia, show that the construction of reservoirs is often followed by an increase in infected fish. This becomes apparent 3–4 years after impoundment and may result in the development of stable foci of diphyllobothriasis, especially if dam construction leads to an influx of human population into the area (Morley, 2007). Accordingly, the tapeworm changes its life cycle from primarily occurring in animals (zoophilic) to humans (anthropophilic). This is facilitated by the continuous discharge of untreated domestic sewage. However, the number of infected fish is lower where sewage mixes with industrial wastewater from chemical plants, reducing the copepod density (Morley, 2007).

3.12.3.4.7 Recommendations Adequate sewage treatment can reduce contamination of water sources with eggs excreted by infected human hosts, although alternative reservoir hosts may be present. Eating of marinated or cold smoked, raw fish should only occur after deep freezing.

329

Pseudoterranova decipiens sensu lato, are parasitic nematodes, both of the family Anisakidae. Herring and cod worms usually spend their complete life cycle in the marine environment. The final hosts of these worms include marine mammals, while aquatic invertebrates, predominantly pelagic calanoid copepods and euphausiids (for Anisakis spp.) and different benthopelagic harpacticoid copepods, amphipods, and isopods (for Pseudoterranova spp.), act as intermediate, and a wide variety of cephalopod and fish species act as paratenic hosts (e.g., Abollo et al., 2001; Audicana et al., 2002; Klimpel et al., 2004). Recent molecular taxonomic examinations have shown that there are at least 9–14 species in the genus Anisakis and eight Pseudoterranova species, many of which had previously been lumped together either as A. simplex or as P. decipiens (Nadler et al., 2005; Mattiucci and Nascetti, 2006).

3.12.3.5.2 Developmental cycle Anisakis species are usually found in cetaceans (baleen and toothed whales), while Pseudoterranova species occur especially in pinnipeds such as seals (Chai et al., 2005). The heteroxenous life cycle of both nematodes principally follows the nematode life-cycle pattern, including four larval stages (L1–L4) and the adults in the final host. Unembryonated eggs of Anisakis spp. are produced by adult worms living in the intestinal tract of marine mammals (Figure 18). These are shed with the feces and become embryonated in water. There is some discrepancy in the literature on whether L2 or L3 larvae then hatch from the eggs (e.g., Smith, 1983; Koie, 2001; Klimpel et al., 2004). The third stage larvae of Anisakis spp., which are ingested by humans, are 2–3 cm long and 0.5– 1.0 mm in diameter (Bogitsh et al., 2005). The hatching larvae become free swimming and may be eaten by a pelagic copepod or euphausiid first intermediate host. If, in turn, this crustacean host is eaten by a cephalopod or fish, the L3 larvae migrate to the body cavity or viscera of this new transport host where they become encysted (Figure 19). Thus, the larvae can potentially be transferred from fish to fish. In this paratenic transfer, very large numbers of L3 larvae may accumulate in high trophic order carnivorous fish (Abollo et al., 2001). If the cephalopod or fish is eaten by a marine mammal, the larvae molt twice before reaching the adult stage (Audicana et al., 2002; Klimpel et al., 2004). In the life cycle of Pseudoterranova spp., partially embryonated eggs passed in seal feces settle onto the seabed where they complete development to the third stage larvae (L3) and hatch. Nematode larvae ingested by benthic and/or benthopelagic crustaceans hatch in their intestine and migrate to the peritoneal cavity. Various fish species serve as paratenic hosts, acquiring Pseudoterranova third-stage larvae through the food chain. Death of the fish host may lead to migration of both parasites from the visceral organs to muscle tissue, a process which may be enhanced by cold storage (Abollo et al., 2001). Humans are accidental hosts which are unsuitable for the parasites to continue their life cycle.

3.12.3.5 Anisakiasis 3.12.3.5.1 Parasite characterization

3.12.3.5.3 Human involvement

The long, thin and tapering herring worms, A. simplex sensu lato (i.e., a complex of closely related species), and a similar pathogenic species group, the cod- or seal worms

Humans are accidental hosts of both Anisakis and Pseudoterranova species both of which cause similar illnesses (Sohn and Seol, 1994; McClelland, 2002). Anisakiasis is caused by

330

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control Humans become incidental hosts through eating infected raw or undercooked seafood

7

i = infective stage d = diagnostic stage

Diagnosis of anisakiasis can be made by gastroscopic examination during which the 2-cm larvae can be removed

6

When fish or squid containing L3 larvae are ingested by marine mammals, the larvae molt twice and develop into adult worms. Adult worms produce eggs that are shed by marine mammals

d

1 Marine mammals excrete unembryonated eggs

Eggs become embryonated 2a in water and L2 larvae form in the eggs i 5 Fish and squid maintain L3 larvae that are infective to humans and marine mammals

2b After the L2 larvae hatch from eggs, they become free swimming

4 Infected crustaceans are eaten by fish and squid. Upon the host’s death, larvae migrate to the muscle tissues, and through predation, the larvae are transferred from fish to fish

3

Free-swimming larvae are ingested by crustaceans and they mature into L3 larvae

Figure 18 The life cycle of Anisakis spp. Adult stages of Anisakis simplex or Pseudoterranova decipiens reside in the stomach of marine mammals, where they are embedded in the mucosa, in clusters. Unembryonated eggs produced by adult females are passed in the feces of marine mammals 1 . The eggs become embryonated in water, and first-stage larvae are formed in the eggs. The larvae molt, become second-stage larvae 2a, and after the larvae hatch from the eggs, they become free swimming 2b. Larvae released from the eggs are ingested by crustaceans 3 . The ingested larvae develop into third-stage larvae that are infective to fish and squid 4 . The larvae migrate from the intestine to the tissues in the peritoneal cavity and grow up to 3 cm in length. Upon the host’s death, larvae migrate to the muscle tissues, and through predation, the larvae are transferred from fish to fish. Fish and squid maintain third-stage larvae that are infective to humans and marine mammals 5 . When fish or squid containing third-stage larvae are ingested by marine mammals, the larvae molt twice and develop into adult worms. The adult females produce eggs that are shed by marine mammals 6 . Humans become infected by eating raw or undercooked infected marine fish 7 . After ingestion, the anisakid larvae penetrate the gastric and intestinal mucosa, causing the symptoms of anisakiasis. Reproduced with permission from CDC.

the ingestion of nematodes of the genus Anisakis (Mattiucci et al., 1997) with raw or undercooked marine fish (Sakanari and McKerrow, 1989; Audicana et al., 2002; Nawa et al., 2005). Japan, where consumption of such food is common, accounts

for about 95% of the known anisakiasis cases worldwide, with about 2000 recorded every year (Audicana et al., 2002). In Europe, only about 500 cases had been reported up to 2002, of which almost all were from the Netherlands, Germany,

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Figure 19 Larval Anisakis in a herring (arrow). Note the typical watchspring coil of the encysted L3 larvae. Courtesy of Dr. V. Etzel, Cruxhaven.

France, and Spain (Audicana et al., 2002). The number of cases appears to be increasing (Audicana et al., 2002), perhaps as a result of increased recognition of the disease and/or because of the more frequent consumption of raw or undercooked seafood. Another hypothesis suggests that increasing eutrofication of the sea leads to higher densities of mesozooplankton, comprising mostly herbivorous crustaceans (Micheli, 1999), although variation in infestation prevalence and intensity can also be related to changes in sea temperature and the population levels of the final host (McClelland et al., 2002; Midtgaard et al., 2003). Physicians do not usually have the capability of determining precisely which nematode species is involved.

3.12.3.5.4 Disease characteristics in humans Not only have pathological effects been associated with the ingestion of live worms but allergic effects have also been noted after ingesting either live or dead worms (Del Pozo et al., 1997; Audicana et al., 2002). The clinical course of infection by Anisakis or Pseudoterranova species is variable depending on the localization of the parasites (Casta´n et al., 2002). Two forms are recognized, gastric and intestinal, with the former predominating in some areas, such as Japan, while the latter is more common in Europe and is symptomatically more severe (Casta´n et al., 2002; Akbar and Ghosh, 2005). In noninvasive anisakiasis, the worms remain within the gastrointestinal tract and the infection is usually asymptomatic. Acute gastric anisakiasis usually involves abdominal pain, nausea, and vomiting, and occurs 2–5 h after eating infected seafood. The larvae can be detected by gastroscopy and removed (Akbar and Ghosh, 2005). In Japan, there is more widespread use of endoscopic techniques due to greater awareness of the disease (Audicana et al., 2002). Intestinal anisakiasis, which is difficult to diagnose, usually appears between 10 h and 2 days of ingesting the larvae, with the major symptom being abdominal pain (Matsui et al., 1985). This is usually a self-limiting condition which resolves within 3–12 days (Matsui et al., 1985). Migration of the larvae to other organs such as the pancreas, spleen, and lungs has been reported (Akbar and Ghosh, 2005). Allergic manifestations caused by the ingestion of anisakids, either in cooked or insufficiently cooked fish, can be

331

serious, with more than 50% of patients requiring emergency treatment (Audicana et al., 2002). Symptoms include anaphylaxis, urticaria, angioedema, dermatitis, and airway hyperreactivity (Lopez-Serrano et al., 2000; Audicana et al., 2002; Nieuwenhuizen et al., 2006). Indeed, of the 625 submissions to the Sandiago Apo´stel Hospital in northern Spain for anaphylaxis between 1994 and 1999, 67 (10.7%) were caused by an allergy to A. simplex (Audicana and Kennedy, 2008). Allergens have been found in A. simplex which are resistant to both body heat and the digestive enzyme pepsin indicating that they can potentially lead to an allergic response in spite of digestive processes (Caballero and Moneo, 2004). The variability and generality of symptoms, particularly of chronic gastric anisakiasis, can lead to confusion in the diagnosis with other, potentially severe diseases such as peptic ulcer, appendicitis, Crohn’s disease, and cancer (Akbar and Ghosh, 2005). A definitive diagnosis can be made by finding worms in the stomach using gastroscopy. Abdominal ultrasonography can be used to diagnose intestinal infections but about half of the 15 cases reported by Casta´n et al. (2002) required histological examination of biopsy specimens. Immunoassays have also been developed both for diagnosing infections (Akbar and Ghosh, 2005) and for detecting allergic responses (Audicana et al., 2002).

3.12.3.5.5 Prevention and cure Humans are accidental hosts of these species and reproduction in humans does not occur. Thus, curing the disease will not lead to changes in the natural abundance of these parasites; in fact, humans act as an ecological sink. Abollo et al. (2001) suggest that better fisheries and aquaculturemanagement practices may help reduce the problem. Commercial fisheries often dispose of heavily infested viscera in the sea. Any fish feeding on these will be infected and act as paratenic hosts. In addition, the storing of whole fish (i.e., those still containing the viscera) on ice for several hours may enhance the number of L3 larvae migrating from the viscera to the muscle tissue, which is the source of infection for humans. This must be prevented. Heating fish to above 60 1C effectively kills the larvae and denatures the protein, a fact which should be propagated in education programs by the health authorities in affected regions (Audicana et al., 2002).

3.12.3.5.6 Anthropogenic alterations to the environment As the cycles of these parasites are predominantly marine, changes in land-based hydrological systems will have little influence on transmission dynamics. The effect of the substantial changes in such parameters as the temperature, salinity, and pH of seawater associated with global climate change are unclear, although eutrification can lead to an increase in intermediate host density (Micheli, 1999; McClelland et al., 2002; Midtgaard et al., 2003).

3.12.3.5.7 Recommendations As indicated by Abollo et al. (2001, discussed earlier) certain changes in fisheries practices may help reduce the likelihood of humans eating infected fish. Heating fish to temperatures

332

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

causing protein denaturation before consumption will also prevent human infection.

3.12.4 Other Parasites with a Water-Dependent Life Cycle 3.12.4.1 Fascioliasis 3.12.4.1.1 Parasite characterization Fasciola hepatica, the common liver fluke, belongs to the trematodes as do the opisthorchids, the heterophyids, and Paragonimus spp. These hermaphroditic flatworms inhabit the biliary system of the liver (Figure 20; Lucius and Loos-Frank, 2008). Naturally occurring in Europe, it is now one of the world’s most widely distributed parasites, having been recorded from at least 51 countries ranging from Europe,

4a Sporocysts

4b Rediae

4c Cercariae

North and South America, Africa, Asia, Australia, and Oceania (Mas-Coma et al., 1999; Taraschewski, 2006; Laird and Boray, 2008). A second species, F. gigantica, the distribution of which overlaps that of F. hepatica in Africa and Asia, is also known to cause similar disease in humans and animals (Mas-Coma et al., 2005). Both are zoonoses infecting a variety of domestic and wild animals including cattle, sheep, buffaloes, equids, elk, red deer, hares, and occasionally kangaroos (Presidente and Beveridge, 1978; Mas-Coma et al., 2005). Prevalences in these hosts are variable but may be very high in some areas: a mean of 56.3% (range: 41.8–61.1 for nine sites) in sheep in the Upper Awash River basin in Ethiopia (Asrat et al., 2005), 44.2% (determined by enzyme-linked immunosorbent assay (ELISA), range 38.4–62.0 for four sites) for sheep in Spain (Ferre et al., 1995), and 0–56.8% in cattle in Cambodia (F. gigantica; Tum et al., 2004).

Metacercariae on water plant ingested by human, sheep, or cattle

5 Free-swimming cercariae encyst on water plants

6 i

In snail tissue

4 Snail

7 Excyst in duodenum 7

3 Miracidia hatch, penetrate snail

8

2 Embryonated eggs in water i = infective stage d = diagnostic stage

d 1 Unembroynated eggs passed in feces

8 Adults in hepatic biliary ducts

Figure 20 The life cycle of Fasciola hepatica. Immature eggs are discharged in the biliary ducts and in the stool 1 . Eggs become embryonated in water 2 , eggs release miracidia 3 , which invade a suitable snail intermediate host 4 , including the genera Lymnaea (Galba), Fossaria and Pseudosuccinea. In the snail, the parasites undergo several developmental stages (sporocysts 4a , rediae 4b, and cercariae 4c ). The cercariae are released from the snail 5 and encyst as metacercariae on aquatic vegetation or other surfaces. Mammals acquire the infection by eating vegetation containing metacercariae. Humans can become infected by ingesting metacercariae-containing freshwater plants, especially watercress and by metacercariae floating on water 6 . After ingestion, the metacercariae excyst in the duodenum 7 and migrate through the intestinal wall, the peritoneal cavity, and the liver parenchyma into the biliary ducts, where they develop into adults 8 . In humans, maturation from metacercariae into adult flukes takes approximately 3–4 months. The adult flukes (Fasciola hepatica: up to 30 mm  13 mm; F. gigantica: up to 75 mm) reside in the large biliary ducts of the mammalian host. Fasciola hepatica infect various animal species, mostly herbivores. Reproduced with permission from CDC, modified by the authors of this chapter.

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control 3.12.4.1.2 Developmental cycle The eggs (Figure 21; c. 132  70 mm) are excreted in the feces which, when they have been washed or defecated into freshwater, embryonate over a period of 2–3 weeks, depending on the temperature, and release the miracidia larvae. These then invade the snail intermediate host. In Europe, the lesser pond snail, Lymnaea (Galba) truncatula (family Lymnaeidae) is the major first intermediate host. This is also true for the Andean Altiplano in South America and the higher altitudes in East and South Africa where this gastropod has been naturalized. Many lowland areas of the tropics and subtropics have been colonized by Lymnaea (Pseudosuccinea) columella deriving from the southern USA and the Caribbean. This species transmits the fluke in Argentina and Brazil. In other areas, such as South Africa, to which it has also been introduced, this snail shows no indication of field transmission. Currently, for Australia and New Zealand, the only known snail intermediate host is the indigenous Lymnaea tomentosa, although laboratory trials with introduced local L. columella showed that these could be infected successfully (Taraschewski, 2006). Development in the snail continues through sporocysts, redia, and cercariae, which are released into the water. These move preferentially onto the underside of partially submerged vegetation, and form a cyst which takes about 24 h to becoming infective (Figure 22). The cyst, which can also float on the surface of the water, must be ingested, with or without the vegetation, by the final hosts such as grazing ruminants, or by humans. Excystation takes place in the duodenum prior to penetration of the intestinal wall. The immature cysts enter the liver capsule through the peritoneum, and then migrate through the liver parenchyma where they feed on liver cells and blood before reaching the bile ducts. There, they develop into adults (5.0 1.3 cm length by width), attaining sexual maturity in about 12 weeks. The adults can live for up to 12–15 years (Bogitsh et al., 2005; Saba et al., 2004; Lucius and Loos-Frank, 2008).

333

indicating that transmission can occur through water used for drinking and washing dishes, etc., which is collected from irrigation and drainage canals (Estaban et al., 2002; Mas-Coma et al., 2005). This hypothesis is supported by a significant positive association between F. hepatica and other waterborne protozoan diseases with direct life cycles, such as Cryptosporidium spp. and Giardia lamblia (Estaban et al., 2002). Although originating in Europe, the highest prevalences are now found in South America, in Peru, and particularly Bolivia where up to almost 100% infection rates in the rural human population have been recorded locally (Estaban et al., 1999; Mas-Coma et al., 1999, 2005). It is estimated that 180 million people are at risk worldwide and that between 2.5 and over 17 million are infected (Mas-Coma et al., 2005). In the Nile Delta in Egypt, where F. hepatica and F. gigantica co-occur, a mean of 12.8% of the human population was found to harbor liver flukes. Here, the number of cases has risen conspicuously during the last few decades (Estaban et al., 2003). In Egypt, as in the Andean countries of South America, women are more commonly infected than men, with school children, especially girls, having the highest prevalences and intensities of infection (Estaban et al., 1999, 2002, 2003; Marcos et al., 2006). In Vietnam and Cuba too, females are more likely to be infected than males, but adults have the highest infection rates (World Health Organization, 2007; Rojas et al., 2009). This species also has a long association with humans, having been found in mummified remains from over 5000 years ago in a continual progression through to the present day (Bouchet et al., 2003; Goncalves et al., 2003).

3.12.4.1.4 Disease characteristics in humans

Eating contaminated vegetable material is undoubtedly the most common mode of infection in humans (Mas-Coma et al., 2005); however, there is some suppositional evidence

The severity of symptoms depends on the intensity of infection, with low infection levels potentially being asymptomatic (Chen and Mott, 1990). There are several phases involved with infection by Fasciola spp. (World Health Organization, 2007). After ingesting infective metacercariae, there is an incubation phase lasting from a few days to several months. The acute hepatic stage begins with the migration of parasites from the body cavity into the liver capsule where they ingest hepatic tissue causing hemorrhage and inflammation. This stage is characterized by abdominal pain and

Figure 21 Unembryonated Fasciola hepatica egg (arrows: operculum) from fresh feces. Size c. 130 mm  70 mm. From the slide collection of Werner Frankw.

Figure 22 Squash prepared Fasciola hepatica metacercariae encysted on a plant. Diameter c. 250 mm. From the slide collection of Werner Frankw.

3.12.4.1.3 Human involvement

334

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

fever accompanied by a variety of other symptoms including fatigue, weight loss, nausea, vomiting, respiratory symptoms, and chest pain (Saba et al., 2004; World Health Organization, 2007). These symptoms usually clear once the flukes reach the bile ducts. On having reached their final habitat, the chronic latent phase begins during which the parasites mature and start laying eggs. This phase can last from months to years, with symptoms being nonspecific involving abdominal pain, nausea, fatigue, and weight loss (Saba et al., 2004). The latent phase can progress to a chronic obstructive phase with worms and debris periodically blocking the bile duct. This results in swelling, potentially with acute pancreatitis, jaundice, and cholestatic hepatitis. Bacterial superinfection can lead to acute cholangitis and cholecystitis (World Health Organization, 2007). In the post-infection stage, there are a variety of clinical sequelae and complications including biliary cirrhosis and gall stones (World Health Organization, 2007). Chronic fascioliasis can be a long-term disease lasting well over 10 years (Bogitsh et al., 2005; Saba et al., 2004). It is possible that the flukes may migrate to organs other than the bile ducts such as the pancreas, spleen, and kidney (Zali et al., 2004), brain (Ying et al., 2007), eye (Dalimi and Jabarvand, 2005), and spinal cord (Vatsal et al., 2006), where they cause pathological damage. In the Altiplano of Bolivia, the synergistic associations between fascioliasis and a variety of other pathogens are believed to cause the high morbidity rates and significant mortality in Aymara children (Mas-Coma et al., 2005). For a definitive diagnosis, it is necessary to confirm the presence of the parasite or its eggs. For the latter, repeated stool examinations are usually required. Immunological methods have also been developed. These are particularly important for early diagnosis when eggs have not yet been produced (Saba et al., 2004).

3.12.4.1.5 Prevention and cure Fascioliasis is predominantly a disease of domestic and herbivorous mammals in most parts of the world. In developed countries, where large herds of stock animals are managed, control programs involving treatment of infected stock are effective (Roberts and Suhardono, 1996; Kaplan, 2001). In developing countries, in which families are often poor and possess few cattle, particularly in tropical regions where water bodies are omnipresent, effective control at this level is much more difficult and stock are often left untreated (Roberts and Suhardono, 1996). This presents a substantial problem in terms of human infection as domestic stock act as reservoir hosts continually contaminating the environment. Various attempts to control the snail intermediate hosts have met with little success (Roberts and Suhardono, 1996). Furthermore, suitable mud snails, such as Lymnaea (Pseudosuccinea) columella, are highly invasive and their accidental introduction to areas that are unaffected by F. hepatica is difficult to prevent (Taraschewski, 2006). On the other hand, trials aimed at the introduction of a nonlocal gastropod capable of outcompeting and thus reducing population levels of the local Fasciola intermediate host have progressed substantially less than in the case of schistosome flukes (see later). In Cuba, biological control of the snail host Lymnaea (Fossaria)

cubensis with the planorbid Helsioma duryi and the thiarid Thiara granifera has been successful in certain habitats. However, Lymnaea (Pseudosuccinea) columella is able to coexist with the introduced competitors (Canete et al., 2004; Rojas et al., 2009). The globally invasive North American snail Physa acuta could be a candidate for such campaigns where nontropical temperatures occur (Dreyfuss et al., 2002). Recently, it was found to have invaded Lake Titicaca (Albrecht et al., 2008). Human infection can be prevented or strongly reduced by eliminating uncooked, potentially contaminated vegetable food from the diet, and by avoiding hand–water–mouth contacts, as well as by drinking water, coming from potentially contaminated sources, only after it has been boiled (Mas-Coma et al., 2005; Ashrafi et al., 2006). In addition, to reduce the chances of infection, cattle and sheep should be prevented from feeding near water sources such as ponds and streams. The currently recommended treatment for fascioliasis is triclabendazole 10 mg kg1 body weight as a single dose. Both immature and adult parasites are killed and cure rates are high (World Health Organization, 2007).

3.12.4.1.6 Anthropogenic alterations to the environment Endemic cycles of fascioliasis are dependent on the presence of water required by the lymnaeid snail intermediate hosts and of suitable final hosts such as cattle (or indeed humans if conditions are suitable). There is evidence from various areas of the world that irrigation has led to the introduction or increased the rate of infection of humans with Fasicola species, in part through the introduction of suitable intermediate snail hosts (Mas-Coma et al., 2005). In the Punto region of Peru, for example, school children had high prevalences of infection with F. hepatica after man-made irrigation systems were established (Estaban et al., 2002). Prevalences in humans ranged from 18.8% to 31.3%. After the construction of the irrigation system, both the intermediate lymnaeid snail host as well as the liver fluke quickly adapted to this environment (Estaban et al., 2002). In Egypt, the large Nile Delta has been transformed into an agricultural plain due to extensive irrigation resulting in high abundance of fascioliasis in animals and humans (Estaban et al., 2003). In the Upper Awash River Basin of Ethiopia, irrigation was associated with significantly increased prevalence of ovine fascioliasis in mid-altitude sites during the dry season and in lowland sites during both the wet and dry seasons (Asrat et al., 2005). In addition to developing countries, fascioliasis can also be a problem enhanced by irrigation in developed countries, such as Spain (Uriarte et al., 1985; Ferre et al., 1995), where irrigation canals increase habitat in which the survival of eggs over winter is possible (Luzon-Pena et al., 1992), the USA (Malczewski et al., 1975; Kaplan, 2001), and Australia, where irrigation was the variable providing the best explanation for the observed distribution of fascioliasis (Durr et al., 2005). In Cambodia, the irrigation of rice at the edges of flooded areas as the water level falls provides suitable aquatic habitat for the snail intermediate hosts of F. gigantica. This, together with the widespread contamination of dams and canals by cattle and water buffalo feces at certain times of the year when they are used as draught animals for preparing the fields, leads

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

to increased transmission rates for this species (Tum et al., 2004; Suon et al., 2006).

3.12.4.1.7 Recommendations In areas with a high abundance of Fasciola infections, the local populations should be educated not to consume raw plants or plant juices from water sources or their vicinity, or from irrigated fields. They should also not drink uncooked or unfiltered water from the environment and avoid washing kitchen utensils in potentially contaminated water. In many or most regions of developing countries, no potable water systems inside dwellings exist. Thus, inhabitants must obtain their water from irrigation or drainage canals (Estaban et al., 2002). Any initiative for improving agricultural or semi-agricultural landscapes should be accompanied by the establishment of a clean, tapwater-supply system. If this is financially not feasible, each village should at least be provided with the opportunity of obtaining drinking water and water for washing the dishes from a safe source. An Egyptian endemic area for human infection showed a markedly reduced prevalence of the disease after the construction and utilization of so-called ‘washing units’ in which water is appropriately filtered (Mas-Coma et al., 2005). In addition, irrigation canals and wet areas should be fenced off, at least in the vicinity of villages where people might be tempted to collect water. Municipalities in irrigated areas should provide veterinary and medical control of animal and human feces for Fasciola eggs. Should these be found, treatment with triclabendazole should be initiated. Irrigation schemes should always be planned including a parasitological assessment taking into consideration the likelihood of spreading fascioiasis (Estaban et al., 2002; Asrat et al., 2005).

3.12.5 Parasites Penetrating Human Skin on Contact with Freshwater 3.12.5.1 Schistosomiasis (bilharziosis) 3.12.5.1.1 Parasite characterization Schistosomes are digenean trematodes of the family Schistosomatidae (Loker and Brant, 2006). There are a large number of species belonging to the genus Schistosoma of which at least 12 are capable of infecting humans (Lucius and Loos-Frank, 2008). They inhabit the mesenteric veins, in most cases of the posterior intestine, feeding on blood (Figure 23). In addition to using the blood vessels as their microhabitat in their final hosts, schistosomes also differ from the usual flukes in other ways: they are dioecious, that is, are either male or female; and in both sexes, the body is not flat, but rounded. Along the ventral surface of the male, there is a slit, the canalis gynaecophorus, in which the long, thin female is kept, ensuring a permanently available sexual partner. Once mated, they can survive inside the human host for decades due to specific morphological and biochemical characteristics of their outer surface. Schistosome eggs do not have an operculum (Figure 24). Moreover, schistosomes do not have a second intermediate host; thus no metacercariae exist (Lucius and Loos-Frank, 2008).

335

Molecular taxonomic characterization is likely to increase the number of known species infecting humans (Agatsuma et al., 2001, 2002; Webster et al., 2006; Zhao et al., 2009). In Schistosoma mansoni, the species with the largest distribution and the highest number of infected humans, 85 distinct haplotypes have been found belonging to five distinct lineages, with particularly high variation being present in East Africa (Morgan et al., 2005). Of the currently recognized species, the most important for humans include S. mansoni from Africa, parts of the Arabic Peninsula, Caribbean, and South America; S. intercalatum from Central African countries; S. haematobium causing urinary schistosomiasis in Africa, Madagascar, Iraq, and parts of the Arabian Peninsula; S. japonicum from East and Southeast Asia and the Western Pacific; and S. mekongi from the Mekong River Basin (Gryseels et al., 2006).

3.12.5.1.2 Developmental cycle The life cycles or the various Schistosoma species infecting humans are similar (Figure 23). Eggs released by mated females into blood capillaries passively penetrate the walls of the vessels via antigenic activity either to the intestine or urinary bladder to be excreted with the feces (e.g., S. mansoni) or urine (S. haematobium). On excretion, eggs contain a ciliated miracidium which hatches and seeks a suitable snail host when it contacts water – a process which can last for several hours before the miracidium dies if a host is not found. Intermediate hosts of the S. mansoni and S. haematobium groups are planorbid, air-breathing aquatic snails (Basommatophora) belonging to the genera Biomphalaria and Bulinus, respectively. These inhabit shallow water with a minimum temperature of 20 1C. In contrast, the S. japonicum group is transmitted by small aquatic or amphibic prosobranch snails of the genus Oncomelania. S. mekongi has the prosobranch snail Neotricula aperta as intermediate host. Once having penetrated the snail, the miracidia develop into sporocysts which finally shed cercariae (c. 500 mm long). These elongated, fork-tailed larvae (Figure 25) are released into the water, where they can survive for up to 5–10 h, during which time they actively seek a suitable final host (Lucius and Loos-Frank, 2008). On finding a host, the cercariae penetrate the skin, lose their tail, and develop into an immature worm which must enter a blood or lymph vessel to be transported to the lungs, and then the heart where they enter the arterial system and finally the portal artery. Here, the sexually mature, flat males (S. haematobium: 8–15 mm in length) find round female partners, which slip into the male’s gynaecophoral fold, mate, and then migrate to the organ of preference: S. mansoni, the mesentary of the veins usually of the large intestine or rectum; S. japonicum the mesentary of the veins of the large or small intestine; and S. haematobium the urinary bladder (Lucius and Loos-Frank, 2008). In addition to humans, all human pathogenic schistosomes have mammalian reservoir hosts. S. mansoni is predominantly a human parasite, although some primate species and rodents can harbor the parasite (Cameron, 1928; Duplantier and Se`ne, 2000). S. mekongi has been found in pigs and dogs (Urbani et al., 2002), while S. japonicum has a very wide host spectrum including at least 40 mammal species.

336

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

5 Sporocysts in snail 4 (successive generations)

i = infective stage

Cercariae released by snail into water and free swimming

d = diagnostic stage

i Cercariae lose tails during 7 penetration and become schistosomulae

Penetrate skin 6

8 Circulation

3

Miracidia penetrate snail tissue

A Migrate to portal blood in liver and mature into adults 9

B

2

In feces

d

In urine

Eggs hatch releasing miracidia

C

C

S. japonicum A S. haematobium S. mansoni C B 1

10 Paired adult worms migrate to: A B mesenteric venules of bowel/rectum (laying eggs that circulate to the liver and shed in stools) C Venous plexus of bladder

Figure 23 The life cycle of Schistosoma spp. Eggs are eliminated with feces or urine 1 . Under optimal conditions, the eggs hatch and release miracidia 2 , which swim and penetrate specific snail intermediate hosts 3 . The stages in the snail include two generations of sporocysts 4 and the production of cercariae 5 . Upon release from the snail, the infective cercariae swim, penetrate the skin of the human host 6 , and shed their forked tail, becoming schistosomulae 7 . The schistosomulae migrate through several tissues and stages to their residence in the veins ( 8 , 9 ). Adult worms in humans reside in the mesenteric venules in various locations, which at times seem to be specific for each species 10 . For instance, S. japonicum is more frequently found in the superior mesenteric veins draining the small intestine A , and S. mansoni occurs more often in the superior mesenteric veins draining the large intestine B . However, both species can occupy either location, and they are capable of moving between sites; so, it is not possible to state unequivocally that only one species occurs in one location. S. haematobium most often occurs in the venous plexus of bladder C , but it can also be found in the rectal venules. The females (size 7–20 mm; males slightly smaller) deposit eggs in the small venules of the portal and perivesical systems. The eggs are moved progressively toward the lumen of the intestine (S. mansoni and S. japonicum) and of the bladder and ureters (S. haematobium), and are eliminated with feces or urine, respectively. Reproduced with permission from CDC.

Water buffalo, cattle, and pigs are known to be important hosts in China (Hotez et al., 1997; Ross et al., 1997; Wang et al., 2006). S. haematobium is found almost exclusively in humans but primates can also harbor this parasite (Taylor et al., 1972; Lucius and Loos-Frank, 2008).

3.12.5.1.3 Human involvement

Figure 24 Embryonated egg of Schistsoma mansoni from human feces. Note the lateral spine and the lack of an operculum (compare Figures 10, 12, and 14). Reproduced with, ermission from CDC.

Schistosomiasis is one of the major human diseases with an estimated 200 million people infected, of whom 120 million have symptomatic infections and 20 million have severe disease (Chitsulo et al., 2000). About 650 million people from 76 countries are at risk of contracting the disease. Most infections are in Africa (c. 97%), which contains 85% of the endangered global population. Children under and around the age of 14 are particularly at risk (Chitsulo et al., 2000; Engels et al., 2002; Steinmann et al., 2006).

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Figure 25 Cercariae of Schistosoma mansoni showing the typical forked tail and lacking eyes, photographed with indirect fluorescent antibody stain to enable better visualization. Reproduced with permission from CDC.

Transmission of the disease to humans involves skin contact with water containing active cercariae. In certain parts of the world where schistosomiasis is endemic, most if not all water bodies are potentially infected and humans contacting water during their daily activities, such as irrigating fields, washing dishes or clothes, fishing, and collecting water for consumption, are at risk (Watts and El Katsha, 1997; Seto et al., 2007).

3.12.5.1.4 Disease characteristics in humans The disease characteristics in humans depend on which species of Schistosoma is present and the intensity of the infection and the individual immune responses (Gryseels et al., 2006). Double infections may be locally common. Initial penetration of the skin can cause a rash similar to swimmer’s itch, found in Europe and caused by the related species, Trichobilharzia ocellata, maturing in water birds, which is otherwise not pathogenic to humans (Hora´k and Kola´rˇova´, 2001). In endemic areas, chronic infections are common. Disease symptoms in these patients are usually associated with the deposition of eggs in various organs where they become the core of granulomas (Gryseels et al., 2006). Depending on the species and age of a female worm, between 20 and 3000 eggs (Figures 23 and 24) are discharged per day. Only about 50% of these succeed in passing through the wall of the intestine or urinary bladder, after which they are embryonated and excreted into the environment. The other half becomes stuck in the wall of organs, mainly the liver, or in other tissues to which they are dispersed in the blood stream. Urinary schistosomiasis is caused by S. haematobium. The most common symptom is blood in the urine due to inflammation caused by the ulceration induced by the eggs in the vesical and ureteral walls (Gryseels et al., 2006). Chronic infection can lead to fibrosis (the accumulation of collagen fibers around the eggs) and calcification of the urinary bladder and lower urinary tract. Chronic urinary schistosomiasis is associated with bladder cancer in Africa, including about 31% of all cancers in Egypt where the prevalence of S. haematobium is high (El-Rifai et al., 2000; El Mawla et al., 2001; Michaud, 2007).

337

Hepatic schistosomiasis can develop on infection with S. japonicum, S. mansoni, and S. mekongi, with S. intercalatum causing limited symptoms and pathology. Two distinct syndromes are present. Inflammatory hepatic schistosomiasis is caused by the deposition of eggs in the presinusoidal periportal spaces of the liver (Gryseels et al., 2006). This is the main cause of hepatomegaly found in up to 80% of children and adolescents with schistosomiasis, with the symptoms being difficult to differentiate from those of malaria (Gryseels et al., 2006). Fibrotic schistosomiasis is a late consequence of infection occurring predominantly in young to middle-aged adults with a history of intense, long-term infections and showing high morbidity and potential mortality (Gryseels et al., 2006). Fibrosis in the periportal spaces leads to progressive occlusion of the portal veins, portal hypertension, and splenomegaly. This development can take 5–15 years in S. mansoni infections with more rapid development in S. japonicum (Gryseels et al., 2006). Hemorrhage from gastroesophageal varices is a serious, often fatal complication (Bandeira Ferraz et al., 2001; Lacerda et al., 1999). Acute schistosomiasis is a hypersensitivity response to the migration of the immature worms which occurs a few weeks to months after infection (Gryseels et al., 2006). It is most often described in individuals, such as tourists, who become infected while visiting endemic areas (Jelinek et al., 1996; Bottieau et al., 2006). Symptoms include the sudden onset of fever, myalgia, headache, fatigue, and cough. Later, abdominal symptoms can occur as the worms migrate to their final microhabitat (Bottieau et al., 2006; Gryseels et al., 2006). Symptoms usually disappear within 2–12 weeks, but some individuals become seriously ill with abdominal pain, weight loss, and diarrhea, shortness of breath, toxemia, and hepatosplenomegaly (Gryseels et al., 2006). Movement of eggs to organs other than the liver, such as the lungs (Waldman et al., 2001; Schwartz, 2002), kidneys (Barsoum, 2004), both male and female reproductive organs (Poggensee and Feldmaier, 2001; Leutscher et al., 2000), and the central nervous system (Abreu Ferrari de, 2004), have been reported, all of which can cause significant pathology. Determination of the presence of eggs in feces or urine remains the most important diagnostic technique (Gryseels et al., 2006). Immunoassays are also available but are unable to distinguish exposure from active infection and may crossreact with other helminth species (Gryseels et al., 2006).

3.12.5.1.5 Prevention and cure Although efforts to control schistosomiasis have been partially successful in some parts of the world, the number of people infected has probably not changed significantly and it remains a major health problem, particularly in Africa (Chitsulo et al., 2000; Engels et al., 2002). The main strategy aimed at controlling schistosomiasis is based on reducing transmission from humans to local water supplies by curing infections using the drug praziquantel (Engels et al., 2002; Gryseels et al., 2006). Treatment is also aimed at preventing morbidity, including bladder cancer, associated with long-term infections (El Mawla et al., 2001; Engels et al., 2002). The opportunities to interrupt the transmission cycle of these important waterborne parasites through

338

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

chemotherapy, health education, and the prevention of sewage influx into surface waters and defecation on the banks of water bodies containing suitable intermediate hosts, each a difficult task, are limited due to the existence of reservoir hosts for most species. Some decades ago, various trials were carried out attempting to eliminate the intermediate snail hosts by releasing molluskicides into their habitats. However, no long-term, positive effects could be achieved despite conspicuous environmental damage (Perrett and Whitfield, 1996), although, recently, latex from various Euphorbia species has shown potential (Schall et al., 2001; Sermsart et al., 2005; Dos Santos et al., 2007). Biomanipulation, such as the introduction of a nonsusceptible competitor (Melanoides tuberculata, Thiara granifera, and other nonindigenous Old World snails; Pointier and Giboda, 1999), or a potentially predatory gastropod species (Lanistes carinatus, Marisa cornuarietis, Pila ovata; Pointier and McCullough, 1989; Hofkin et al., 1991; Pointier and David, 2004) into South America and the Caribbean habitats of Biomphalaria spp. transmitting S. mansoni, has been more successful. The introduction of M. tuberculata into these sites resulted in the interruption of transmission and the near-total disappearance of native planorbid (schistosome transmitting) snails, which were locally either partly or totally outcompeted and population levels sank dramatically (Taraschewski, 2006). Currently, the thiarid snails M. tuberculata and T. granifera have colonized the whole hydrographic system on the island of Martinique and maintain dense populations preventing an eventual recolonization by the planorbid intermediate hosts, thus allowing sustainable control (Pointier and Jourdane, 2000). M. tuberculata, however, can also act as an intermediate host for other helminth species of medical or veterinary importance (described earlier).

3.12.5.1.6 Anthropogenic alterations to the environment The construction of dams, lakes, and irrigation systems has substantially aided the spread of schistosomiasis (Steinmann et al., 2006). This can occur through increased habitat availability for the intermediate snail hosts, as well as a higher density of humans due to migration to agriculturally favorable irrigated sites, thus increasing fecal input into water sources. Such migration from infected to uninfected areas can lead to new disease foci (Malan et al., 1977; Idris et al., 2003). In their meta-analysis, Steinmann et al. (2006) show that, worldwide, 8.1% of the at-risk population live in proximity to irrigated areas and 5.4% in proximity to dams – a total of 106 million people. In general, studies comparing the prevalence of infection either in the presence or in the absence of a dam consistently show higher prevalences where dams are present. In Ghana, the construction of clay-core dams was associated with a change in prevalence of S. haematobium from 17% to 51% in 3 years (Hunter, 2003), while the extremely low levels of infection of about 1% prior to the formation of Lake Volta increased to levels of between 68% and 87% after the dam had filled (Paperna, 1969; Scott et al., 1982). For S. mansoni and S. haematobium, both irrigation and dam proximity can increase the risk of infection (Steinmann et al., 2006). In Burkina Faso, for example, non-irrigated areas had a prevalence of

S. haematobium in 14% of schoolchildren, but of 80% in areas where irrigation was established. For S. mansoni, which also occurs in Burkina Faso, the respective prevalences were 1.3% and 45% (Poda et al., 2003). Similar data are available from Liberia for the total population with S. haematobium relative prevalences being 11% and 42% and those for S. mansoni being 9% and 87% (Kazura et al., 1985). In Egypt, the building of the Aswan High Dam led to a shift of irrigation practices from natural, annual flooding events to perennial irrigation with permanent water in irrigation canals and drains (Lanoix, 1958; Malek, 1975, 1976). Prior to dam construction, the prevalence of infection ranged from 2% to 11%, but after the dam was built, it increased from 44% to 75% (Khalil, 1949; cited by Lanoix, 1958). In addition, a variety of studies show that the development and management of water resources can lead to the introduction of Schistosoma spp. in new areas where the human population had had no previous contact with the disease (Steinmann et al., 2006). In the Senegal River Basin, only S. haematobium was present before the construction of the Diama Dam. In less than 10 years after dam construction, not only had the prevalence of this species increased, but also S. mansoni occurred with prevalences ranging from 4% to 71% (Picquet et al., 1996). In former Zaire, the introduction of schistosomiasis was associated with mining activities which created new habitat for intermediate host snails and led to highly increased contact rates between humans (miners) and water (Polderman et al., 1985; Polderman, 1986).

3.12.5.1.7 Recommendations Oomen et al. (1994) provided effective, although somewhat dated, recommendations for regional disease-control measures in irrigation areas for schistosomiasis as well as filariasis, malaria, and onchocerciasis, diseases which are discussed next. Unfortunately, these recommendations have, in many cases, not been followed. Infected humans shed eggs in their urine (S. haematobium) or feces (other species), which begin the cycle of transmission when they reach a freshwater source with the appropriate snail host species. People working in fields who clean themselves in the water after defecation can pose a substantial problem. Supplying uncontaminated water for human consumption and sanitation to stop the eggs entering a suitable environment can break this cycle. This situation is complicated by the potential infection of freshwater sources via zoonotic hosts, particularly for S. japonicum for which water buffalo are effective reservoir hosts. In general, the development and management of water resources is an important risk factor for schistosomiasis, and hence strategies to mitigate negative effects should become integral parts in the planning, implementation, and operation of future water projects (Steinmann et al., 2006). There should, for example, be as little contact as possible between humans and infectious water sources. The Goral irrigation scheme (Oomen et al., 1994) shows that new housing for agricultural workers should be located outside of the irrigated area, canals, and river banks should be fenced off and foot bridges supplied to prevent wading through freshwater bodies, and newly erected villages should be provided with a safe

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

water supply. In addition, long, concrete watering troughs should be offered to herdsmen and the habitat made as unsuitable as possible for the snail intermediate hosts. During the construction of dams and canals, the workers should be made aware of the dangers of coming into contact with freshwater bodies and a base laboratory with medical personnel should be set up to control disease in the workers both before and during construction. If possible, dams, including microdams, should be constructed at altitudes above 2000 m where factors affecting the life cycle, such as temperature, are not as suitable as at lower altitudes (Alemayehu et al., 1998; Ghebreyesus et al., 2002).

3.12.6 Water-Dependent Vector-Borne Parasites 3.12.6.1 Malaria 3.12.6.1.1 Parasite characterization Plasmodium species are protozoa belonging to the Sporozoa (Apicomplexa, class Hematozoa) which have only single-cell stages in their development. All species in this taxon are obligatory endoparasites mostly living inside cells. Their developmental cycle includes a sporogonic, schizogonic, and a gamogonic phase of reproduction (Figures 26–28; Lucius and Loos-Frank, 2008). Within the hematozoa, the infection alternates between vertebrates, such as mammals, and bloodsucking arthropods. For the four Plasmodium species parasitizing red blood cells and causing human malaria, the vectors all belong to the mosquito genus Anopheles (Figure 29; Service, 2004). These are the final hosts of the disease, as sexual reproduction of the parasites occurs in them. Thus, humans act as intermediate hosts. The larvae of Anopheles mosquitoes live in nonpermanent aquatic environments, showing a parallel orientation to the water surface (Figure 30). As one of the major diseases infecting humans, malaria has a vast literature ranging back to over many years. The four major species of malaria infecting humans, Plasmodium falciparum, P. vivax, P. malariae, and P. ovale, are all anthroponoses, exclusively infecting humans in natural situations. Recently, however, substantial rates of infection with the zoonotic, simian malaria, P. knowelsi, have been found in humans in Malaysia (Singh et al., 2004; Cox-Singh et al., 2008). P. falciparum is pathogenically and numerically the most important malaria species worldwide, occurring predominantly in tropical and subtropical areas (Lucius and Loos-Frank, 2008). P. vivax is the most widely distributed species and accounts for 43% of human cases worldwide and more than 50% outside of Africa (Sattabongkot et al., 2004). It is present in warmer areas as well as in temperate zones with the 16 1C summer isotherm building the distributional boundary. P. malariae is substantially less common and occurs most commonly in West and East Africa, as well as the southwest Pacific, while P. ovale is uncommon and occurs predominantly in tropical Africa, New Guinea, the eastern parts of Indonesia, and the Philippines (Mueller et al., 2007; Lucius and Loos-Frank, 2008). There is a great deal of species overlap and mixed infections are not uncommon (Snounou and White, 2004; Mueller et al., 2007). Historically, malaria was much more widely distributed than it is today, occurring between latitudes 641 N and 321 S;

339

reaching as far north as the United Kingdom and Scandinavia (Reiter, 2000; Hay et al., 2004; Sallares et al., 2004). In Finland, indigenous malaria caused by P. vivax died out within the last 200 years even though no or only limited countermeasures were carried out. The last autochthonous case occurred in 1954, with a gradual decrease in household size being considered responsible (Hulden and Hulden, 2009). Major control programs, directed at the vectors, including drainage of swampy areas which act as habitat for larval stages, and against the pathogens themselves by the use of antimalarial drugs, effectively eliminated the disease elsewhere in Europe by the late 1950s (Chwatt and De Zulueta, 1980). Malaria is also one of the parasitic diseases which is known to have influenced human history, at least at a local scale (Sallares et al., 2004). Economically, it is still a major burden at the individual, family and state levels (Sachs and Malaney, 2002; Russell, 2004). Sachs and Malaney (2002), for example, show that malarial countries have higher levels of poverty and lower rates of economic growth than their non-malarial counterparts.

3.12.6.1.2 Developmental cycle The life cycle of P. falciparum is illustrated in Figure 26. However, it is sufficiently similar to that of the other human malaria parasites for it to be used as a general life cycle. The female Anopheles mosquitoes from susceptible species, of which there are about 40 significant species worldwide (Service 2004) (Figure 29), ingest gametocytes (Figures 26 and 27) when they take a blood meal from an infected human. These undergo gamogony (the sexual cycle) in the mosquito. Fertilization of the female macrogamete with male microgametes leads to the formation, first, of a zygote followed by a motile ookinete which undergoes sporogony to form an oocyst. This then breaks releasing infective sporozoites. If the female mosquito then takes another blood meal, the slender, tipped sporozoites, which occur in the salivary gland, are injected into the blood of the human host where they follow an asexual reproductive cycle (hepatic schizogony) in the liver (extra-erythrocytic cycle). The ovoid to round schizonts (meronts) and the hepatic cells enclosing them rupture and merozoites are released which then invade the erythrocytes where many asexual reproductive cycles take place (Figure 28). Eventually, some of the merozoites inside the erythrocytes differentiate to gametocytes which are infective to female mosquitoes, completing the cycle (Figure 27). The female mosquitoes lay their eggs in a wide variety of different aquatic habitats ranging from clear, unpolluted water (e.g., A. culicifacies) to polluted waters in and near human settlements (e.g., A. stephensi), the choice of which is species dependent (Service, 2004). A. merus, for example, is a coastal species which shows peak densities at salinities 30–50% that of seawater (Mosha and Mutero, 1982; Tsy et al., 2003). Between 50 and 200 eggs are laid per oviposition event. These may be laid singly or as small, floating rafts and usually hatch into larvae (Figure 30) within a few days to weeks, depending on temperature (Service, 2004). Feeding on algae, bacteria, or other surface microorganisms, they develop through four stages before developing into motile pupae. After a few days,

Human liver stages Liver cell

Infected liver cell 2

Mosquito stages 12

11 Oocyst i

Ruptured oocyst

Release of sporozoites

1 i Mosquito takes a blood meal (injects sporozoites)

A Exo-erythrocytic cycle

4

Ruptured schizont

3 Schizont

C Sporogonic cycle

Human blood stages 5

10 Ookinete

Macrogametocyte

8 Mosquito takes a blood meal (ingests gametocytes)

Immature trophozoite (ring stage) d

B Erythrocytic cycle Microgamete entering macrogamete 9

P. falciparum

Exflagellated microgametocyte

i

= infective stage

d = diagnostic stage

6 Ruptured schizont

7 Gametocytes d P. vivax P. ovens P. malariae

Mature d trophozoite

Schizont d 7 Gametocytes

Figure 26 The life cycle of Plasmodium spp. The species infesting humans differ for instance in the shape of the gametocytes which can be detected in stained blood smears (Figure 27) providing information on the specificity of the infection. In contrast, the schizonts (Figure 28), which in the erythryocytic cycle lead to the synchronized mass rupture of red blood cells and cause most of the pathogenicity, are more difficult to use for species-distinguishing diagnosis. The malaria parasite life cycle involves two hosts. During a blood meal, a malaria-infected female Anopheles mosquito inoculates sporozoites into the human host 1 . Sporozoites infect liver cells 2 and mature into schizonts 3 , which rupture and release merozoites 4 . (Of note, in P. vivax and P. ovale, a dormant stage (hypnozoites) can persist in the liver and cause relapses by invading the bloodstream weeks, or even years later.) After this initial replication in the liver (exo-erythrocytic schizogony A ), the parasites undergo asexual multiplication in the erythrocytes (erythrocytic schizogony B ). Merozoites infect red blood cells 5 . The ring-stage trophozoites mature into schizonts, which rupture releasing merozoites 6 . Some parasites differentiate into sexual erythrocytic stages (gametocytes) 7 . Blood-stage parasites are responsible for the clinical manifestations of the disease. The gametocytes, male (microgametocytes) and female (macrogametocytes), are ingested by an Anopheles mosquito during a blood meal 8 . The parasites’ multiplication in the mosquito is known as the sporogonic cycle C . While in the mosquito’s stomach, the microgametes penetrate the macrogametes generating zygotes 9 . The zygotes in turn become motile and elongated (ookinetes), 10 which invade the midgut wall of the mosquito where they develop into oocysts 11 . The oocysts grow, rupture, and release sporozoites 12 , which make their way to the mosquito’s salivary glands. Inoculation of the sporozoites into a new human host perpetuates the malaria life cycle 1 . Reproduced with permission from CDC.

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

341

Figure 27 Typical banana-shaped gametocytes of Plasmodium falciparum (between erythrocytes not taking up the stain) in a Giemsa-stained smear of human blood. From the slide collection of Werner Frankw.

Anopheles

Culex

Figure 28 Giemsa-stained smear of human blood showing a late schizont of Plasmodium vivax. The infected, inflated erythrocyte has not yet ruptured. From the slide collection of Werner Frankw.

Figure 30 A larva of Anopheles sp. lying under the water surface and of a Culex sp., both in a typical position. Some species belonging to the family Culicidae are vectors of filariasis (see further). Reproduced with permission from CDC.

the pupal case splits and the adult mosquito emerges (Service, 2004). Males and females feed on sugary plant juices such as nectar; however, the females require a blood meal prior to egg development. The average longevity of females in tropical countries varies from 10 days to more than a month (Sattabongkot et al., 2004) which is particularly important epidemiologically, as a female mosquito must survive long enough to feed on an infected host, to allow time for the intraAnopheles sexual life-cycle stage, and to bite a new host. There are very substantial differences in the vectorial efficiency of the different Anopheles species which are able to transmit Plasmodium spp. to humans (Kiszewski et al., 2004). Figure 29 A feeding female Anopheles gambiae, one of the major vectors of malaria. Note the typical anopheline posture with the head and body in a straight line forming an angle to the surface compared to the typical culicine posture with the head bent at an angle to the body, which lies horizontal to the feeding surface (Figure 34). Reproduced with permission from CDC.

3.12.6.1.3 Human involvement The most recent major summary of human malaria, including information on prevalence and incidence, and treatment and control strategies is the 2008 malaria report published by the WHO. This report indicates that in 2006, there were an

342

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

estimated 247 million cases of human malaria with some 3.3 billion people at risk. The yearly death toll is over a million people, mostly children (c. 91%) under the age of 5 years. In 2008, 109 countries were reported to have autochthonous malaria, while with growing tourism, an increasing number of travelers are bringing malaria back to non-malarial countries (Jelinek et al., 2002). The malaria situation in Africa is worse now than it was 20 years ago. Mortality has increased at least twofold and the efficiency of available low-cost antimalarial drugs is decreasing (Trape et al., 2002). Malaria caused by P. falciparum is the most pathogenic form and results in the highest mortality (Bogitsh et al., 2005). In highly endemic areas, regular reinfection after early childhood leads to partial immunity which reduces both morbidity and mortality in later infections (Kiszewski et al., 2004). In areas with low endemicity (e.g., mountainous regions or areas near the parasite’s distributional limits), however, such immunity cannot develop, and morbidity and mortality occur through all age groups. Malaria epidemics tend to occur where endemicity is relatively low (Kiszewski and Teklehaimanot, 2004). Forty to fifty years ago, there were great hopes that control programs would lead to the eradication of human malaria worldwide (Russell (1955), quoted by Hay et al., 2004). These hopes have not been fulfilled, predominantly because the mosquito vectors rapidly developed resistance to the various insecticides used for their control and because the Plasmodium pathogens developed resistance to the drugs developed for their destruction (Hemingway and Ranson, 2000; Wongsrichanalai et al., 2002). Today, malaria is present over large areas, particularly in tropical and subtropical countries, in many of which multiple drug resistance is present (Hay et al., 2004). However, the moderately pathogenic species, P. vivax, also occurs in higher latitudes (Chwatt and De Zulueta, 1980; Sattabongkot et al., 2004). Data from the USA document its potential reintroduction to states such as Virginia where suitable vectors are available (Pastor et al., 2002). Malaria is very much a disease associated with poverty (Sachs and Malaney, 2002). Even in historical times, high disease burdens of malaria occurred during times, for example, in Roman Italy, when social structures were breaking down (Carter and Mendis, 2002). In a general, worldwide analysis, McCarthy et al. (2000) calculated a significant negative correlation between malaria morbidity and the per growth rate of per-capita gross domestic product. The absolute negative growth impact of malaria exceeded 0.25% per year in a quarter of the 101 countries considered; in sub-Saharan Africa, the average annual growth reduction was 0.55%. A number of genetic mutations in humans have been favored by their ability to reduce human susceptibility to malarial parasites (Tishkoff and Verrelli, 2003). These include sickle-cell anemia, thalessemias, G6PD deficiency, and Duffy blood-antigen variations (Carter and Mendis, 2002).

3.12.6.1.4 Disease characteristics in humans There are variations between the timing and other characteristics of the human phase of the life cycles of the different malaria species (for a detailed tabular summary see Bogitsh

et al. (2005)). The duration of schizogony is critical for the development of sequential symptoms in humans. The initial symptoms of malaria are usually nonspecific and include nausea, fatigue, muscular pains, jaundice, fever, and/or diarrhea (Jelinek et al., 2002; Bogitsh et al., 2005). These can be mistaken for a less-severe disease such as influenza, hepatitis, or gastrointestinal infection (Lalloo et al., 2007). An additional complication in the diagnosis of malaria is the potential delay between contracting the disease and the outbreak of symptoms. The minimum incubation time is 6 days, but falciparum malaria may only occur one or more months after infection (Lalloo et al., 2007). Infections with P. vivax and P. ovale often only become apparent by 6 or more months, or even years, after infection (Jelinek et al., 2002). These species are known to have dormant stages in liver cells. Both the nonspecificity of the symptoms, as well as the potential delay between infection and the presentation of symptoms, mean that malaria should be suspected in any patient with a history of fever and travel to malarial countries (Lalloo et al., 2007). If the initial disease progresses to severe malaria, the clinical situation becomes more complex and potentially life threatening (World Health Organization, 2000; Trampuz et al., 2003; Greenwood et al., 2005). High rhythmic fever periods indicating the synchronized mass bursting of infected erythrocytes resulting from mature schizogonies is a typical feature of malaria infections. Infections with P. vivax and P. ovale are characterized by peaks of fever returning every second day, and for P. malariae, every third day. P. falciparum shows no highly specific pattern. Symptoms include cerebral involvement, pulmonary edema, acute renal failure, and severe anemia (World Health Organization, 2000; Trampuz et al., 2003). These may lead to the rapid deterioration in the clinical condition of the patient, including death, within hours or days (World Health Organization, 2000). Death is often caused, at least in part, by the formation of projections on the outer surface of P. falciparum-infected erythrocytes which lead to the entanglement of the red-blood cells and blocking of capillaries obstructing blood flow (Silamut et al., 1999; Mackintosh et al., 2004). Various forms of acute malaria occur depending in part on the immune status of the patient (Miller et al., 2002). In African children with a high exposure rate to falciparum-malaria hypoglycemia, severe anemia and cerebral malaria occur frequently, while acute respiratory distress is the most dangerous (World Health Organization, 2000; Schellenberg et al., 1999). Non-immune adults commonly suffer from jaundice, pulmonary edema, and renal failure (World Health Organization, 2000; Greenwood et al., 2005). Non-immune pregnant women represent a particularly susceptible group, with infection being associated with maternal anemia, intrauterine growth retardation, low birth weight, congenital infection, and neonatal mortality (Steketee et al., 2001). In the past, the diagnosis of malaria was based on finding infected blood cells in Giemsa-stained blood smears. This method is still used because it is cheap, can differentiate between the parasite species, and can quantify the parasitemia (Wongsrichanalai et al., 2007). Gametocytes of the most pathogenic species, P. falciparum, show a typical bananashaped appearance (Figure 27). Nevertheless, a variety of

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

other sensitive assays, including immunodiagnosis and detecting parasite DNA are available (Wongsrichanalai et al., 2007).

3.12.6.1.5 Prevention and cure Prevention can be effected by reducing the rate of transmission. Mosquito-control programs have been effective in the past, but the mass spraying of insecticides such as dichlorodiphenyltrichloroethane (DDT) has led to substantial ecological damage (Roberts et al., 1997; Keiser et al., 2005b) as well as the development of resistance in many species of mosquitoes (Hemingway and Ranson, 2000). Vector biocontrol by the mosquitofish, Gambusia affinis, today classified as one of the world’s 100 worst invasive alien species (Lowe et al., 2000), has caused extensive ecological damage and is no longer recommended (Bence, 1988; Howarth, 1991). Today, ecologically more acceptable methods, such as the use of Bacillus thuringiensis var. israelensis d-endotoxin which is specific against larvae of mosquitoes and blackflies and certain other freshwater insects, has been successfully used in North and South America, parts of Asia, as well as Africa and Europe (Lu¨thy and Studer, 1986; Walker and Lynch, 2007). Recent techniques to prevent biting or reduce biting rates, for example, by the use of insecticide-impregnated bed-nets have proved remarkably successful in some areas (Sexton, 1994; Shiff, 2002). Treatment regimes with various antimalarial drugs are provided in the following articles: Trampuz et al. (2003), Greenwood et al. (2005) and Lalloo et al. (2007). The drug regimen used is dependent on the potential resistance of the malarial parasites in the area in question. Multiple antimalarial drug resistance is unfortunately common and increasing in many parts of the world (Wongsrichanalai et al., 2002; White, 2004; Alfonso et al., 2006). Generally, the current treatment of preference involves artemisinin and its derivatives with the recommendation to combine this with other drugs in an attempt to reduce the development of resistance (World Health Organization, 2008). This recommendation, however, is currently not being adequately complied with (Butler, 2009). Although numerous attempts have been and are being made to develop a vaccine against the Plasmodium parasite, these have to date been unsuccessful (Girard et al., 2007; Todryk and Hill, 2007).

3.12.6.1.6 Anthropogenic alterations to the environment As pointed out by Patz et al. (2000), the various anthropogenic alterations constantly imposed on aquatic environments lead to habitat-dependent changes in mosquito communities. However, the wide variety of conditions under which at least one mosquito species which is able to transmit malaria can thrive, ensures that malaria remains endemic throughout Africa. In other regions, this rule may not apply. In areas of Thailand bordering Cambodia, the observed decrease in the abundance of Anopheles dirus, the main vector of P. falciparum in this region, was accompanied by a concurrent increase in the abundance of members of the Anopheles barbirostris/ campestris group. It has been suggested that these species might be important secondary vectors of P. vivax because of their

343

high biting intensity, anthropophily, and susceptibility to only P. vixax (Sattabongkot et al., 2004). Thus, the effects of anthropogenic changes may be complex and not always predictable. Nevertheless, some patterns are apparent: 1. Dams and irrigation. In Africa, within areas of unstable malaria, increased densities of (new) vector species following the introduction of irrigation schemes usually leads to an increase in malaria incidence (Ijumba and Lindsay, 2001), for example, in Ethiopia (Ghebreyesus et al., 1999). In contrast, in most of sub-Saharan Africa with stable high endemicity levels, the establishment of irrigation has little impact or may even have a slightly positive effect if Anopheles funestus is replaced by A. arabiensis, which has a lower vectorial capacity (Ijumba and Lindsay, 2001). Indeed, irrigation may lead to greater wealth so that local people can afford more bed-nets and have better access to healthcare (Ijumba and Lindsay, 2001). However, based on literature dealing with both Fasciola spp. and Schistosoma spp., this hypothesis should be examined in more detail, as most of the humans populations affected by these parasites do not directly benefit from the introduction of irrigation schemes. In general, the higher the dams are built in mountainous areas, the lesser is the risk of malaria transmission. For example, the rate of development and indeed the abundance of the vector and parasite are determined by the temperature (Bayoh and Lindsay, 2004). With increasing altitude, the rate of development and potential for the successful completion of the life cycle decreases; as altitude decreases, prevalence of malaria increases (Drakeley et al., 2005; Kulkarni et al., 2006; Mboera et al., 2008). 2. Deforestation. This is a well-documented factor which often intensifies the malaria hazard. Microclimatic changes, that is, higher temperatures and sun-exposed habitats resulting from deforestation exert a strong effect on mosquito larval development. In large parts of Africa and South America, deforestation can lead to increases in malaria (Guerra et al., 2006). In Africa, members of the Anopheles gambiae complex, which has a high vectorial capacity, a long life span, and a preference for human blood, became more prevalent after deforestation (Minakawa et al., 2005). In Brazil, the local removal of rain forest, combined with mining and an increase in the human population, resulted in a 70% increase in malaria prevalence. Interestingly, the number of cases of P. falciparum increased disproportionately compared to P. vivax (Patz et al., 2000). In Trinidad, following deforestation, Erythrina spp. trees, which support large numbers of bromeliads, were planted to provide shade for cocoa growing underneath. A malaria epidemic followed as the bromeliads accumulated small pools of water which became the preferred breeding sites for the vector Anopheles bellator (Patz et al., 2000). Vittor et al. (2009) carried out a large-scale survey of the breeding sites of 17 Anopheles species in Peru. They showed that deforestation and associated ecological alterations were conductive to high larval and adult densities of Anopheles darlingi, one of the most important malaria vectors in South America. In the Bolivian part of the Amazon Basin, malaria increased fourfold between 1991 and 1998, largely owing to forest clearance

344

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

(Harris et al., 2009). Deforestation does not always lead to an increase in malaria prevalence. In some areas, such as those parts of Southeast Asia where deep forest vectors, such as Anopheles dirus, have a high vectorial capacity, deforestation may also lead to a reduction in malaria incidence (Guerra et al., 2006; Obsomer et al., 2007; Petney et al., 2009). 3. Urbanization. For urban human populations, the malaria situation is similar to areas with low endemicity: low transmission, lack of immunity, high morbidity, and mortality in infected individuals in all age groups (Trape et al., 2002). Keiser et al. (2004) estimate an annual incidence of 24.8–103.2 million cases of urban malaria every year in Africa although prevalences vary greatly between urban areas. With a few exceptions, anopheline malaria vectors have not generally succeeded in adapting to urban life (Lines et al., 1994; Hay et al., 2005). This is markedly different to the vectors of the filarial worm Wuchereria bancrofti, some of which are well adapted to urban environments (see next). In the urban environment, anopheline breeding sites can usually be easily localized and the mosquitoes controlled by classical methods such as drainage and larvicide and indoor spraying (Trape et al., 2002). In rural towns in western Kenya, about 70% of all available mosquito habitats were found to be man made, half of them being cement-lined pits (Fillinger et al., 2004). Robert et al. (2003) carried out a meta-analysis of entomological inoculation rates from sub-Saharan African cities. These were 7.1 for city centers, 45.8 for peri-urban areas, and 167.7 for rural areas. The impact of urbanization in reducing transmission of malaria was most clear in areas with low seasonal rainfall. Thus, urbanization does not usually support high malaria prevalence (Hay et al., 2005).

3.12.6.1.7 Recommendations An understanding of the ecological requirements of potential malaria vectors should be obtained before carrying out any large-scale changes in land use or hydrological patterns in potential malarial areas. As indicted above, mosquitoes are sensitive to environmental change, and either increases or decreases in human malaria may result from such change. Changes in mosquito communities should be monitored and an assessment made of the vector capacity of the species currently present. In cases of deforestation/reforestation, the potential for migration of non-forest or deep forest, respectively, vectors of malaria into the area should be assessed and monitoring programs initiated. In urban environments, anopheline breeding sites can be easily localized and control using classical methods, including drainage, application of larvicide, and indoor spraying of surrounding buildings, carried out. In areas with low malaria endemicity, the introduction or intensification of irrigation should be accompanied by the establishment of healthcare stations for the local population.

3.12.6.2 Onchocerciasis 3.12.6.2.1 Parasite characterization Onchocerciasis is caused by infection with the nematode Onchocerca volvulus (family Onchocercidae) which is transmitted

via the bite of a female blackfly (Nematocera, family Simuliidae) which acts as an obligate intermediate host vector (Udall, 2007). For the species O. volvulus, only humans serve as final hosts. The round, pointed nematodes inhabit the subcuticular connective tissue and lymph system. One or more females are associated with nodules of connective tissue, seen as elevations of the skin. The microfilariae, which reach 100 million in a single host, can be found in various tissues. Long-term infections with these L2 larvae in the eye can lead to loss of sight due to which the disease is also called river blindness (Lucius and Loos-Frank, 2008). A number of other Onchocerca species parasitize stock and domestic animals as well as wildlife and can be of economic importance (Webster and Dukes, 1978; Uni et al., 2001; Marques and Scroferneker, 2004; Sre´ter and Sze´ll, 2008).

3.12.6.2.2 Developmental cycle Female O. vulvulus can live for up to 15 years, while males live for a shorter period. Once mated, females produce up to about 1000 active microfilariae (220–360 mm long) per day which then move around typically in the subcutaneous tissue, the eye, and the lymphatic system, and also in the peripheral blood, urine, and sputum, where they can live for 1–2 years (Figure 31). On biting a human, simulids ingest microfilariae with the blood meal. Some of these penetrate the blackfly’s midgut and migrate to the thoracic muscles where they develop over the next few weeks into infective L3 larvae. These in turn migrate to the head or proboscis of the blackfly and are transmitted to the human host during the fly’s next blood meal. Simulium species, which occur worldwide, are responsible for the transmission of the nematode throughout its range. As with mosquitoes, only the females require blood meals and bite the host. All Simulium species spend their larval and pupal life stages mainly as filter feeders in flowing freshwater habitats, although a few species are predatory (Service, 2004). The actual breeding habitat is species dependent, ranging from slow to rapidly flowing streams with high or low-flow volumes in mountainous areas or lowland (Service, 2004). The females of most species lay their strings of 150–500 eggs on partially immersed objects including vegetation and stones, although the Central American S. ochraceum lays individual eggs on the water surface while in flight (Service, 2004). The eggs are usually resistant to desiccation and the larvae sessile (Figure 32), although they can detach from the substrate, for example, when threatened. Pupae are protected by a silken cocoon which is attached to submerged stones or vegetation. S. neavei is unusual with immature stages being found on freshwater crabs (Lewis, 1960). Once the adults emerge, often simultaneously in very large numbers, they either float or crawl to the surface and take flight (Service, 2004). The taxonomy of the various species groups of the simulian vectors of O. volvulus is highly complex and has only become accessible with the development of molecular taxonomic techniques (Krueger and Hennings, 2006). There are currently over 50 recognized forms (species, cytoforms, and morphoforms) of the S. damnosum group ranging throughout sub-Saharan Africa to the Arabian Peninsula (Crosskey and Howard, 2004; Krueger and Hennings, 2006; Post et al., 2007).

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Blackfly stages

1

Blackfly (genus Simulium) takes a blood meal (L3 larvae enter bite wound)

345

Human stages 2

9

Subcutaneous tissues

Migrate to head and blackfly’s proboscis

i 8 L3 larvae

3 Adults in subcutaneous nodule

7

L1 larvae 5 Blackfly takes a blood meal (ingests microfilariae) 6 Microfilariae pentrate blackfly’s midgut and migrate to thoracic muscles

i

= infective stage

d

= diagnostic stage

4 Adults produce unsheathed microfilariae that typically are found in skin and in lymphatics of connective tissues, but also occasionally in peripheral blood, urine, and sputum.

d

Figure 31 The life cycle of Onchocerca volvulus. During a blood meal, an infected blackfly (genus Simulium) introduces third-stage filarial larvae onto the skin of the human host, where they penetrate into the bite wound 1 . In subcutaneous tissues, the larvae 2 develop into adult filariae, which commonly reside in nodules in subcutaneous connective tissues 3 . Adults can live in the nodules for approximately 15 years. Some nodules may contain numerous male and female worms. Females measure 33–50 cm in length and 270–400 mm in diameter, while males measure 19– 42 mm  130–210 mm. In the subcutaneous nodules, the female worms are capable of producing microfilariae for approximately 9 years. The microfilariae, measuring 220–360 mm  5–9 mm unsheathed, have a life span that may reach 2 years. They are occasionally found in peripheral blood, urine, and sputum but are typically found in the skin and in the lymphatics of connective tissues 4 . A blackfly ingests the microfilariae during a blood meal 5 . After ingestion, the microfilariae migrate from the blackfly’s midgut through the hemocoel to the thoracic muscles 6 . There, the microfilariae develop into first-stage larvae 7 and subsequently into third-stage infective larvae 8 . The third-stage infective larvae migrate to the blackfly’s proboscis 9 and can infect another human when the fly takes a blood meal 1 . Reproduced with permission from CDC.

3.12.6.2.3 Human involvement

Figure 32 Larvae of Simulium sp. on the underside of a stone in flowing water. Courtesy of Wikipedia.

Onchocerciasis is one of the major human parasitic diseases with an estimated 17.7 million people infected worldwide in 34 countries and a population at risk of 123 million (Udall, 2007). It is endemic over substantial parts of tropical Africa, the Arabian Peninsula, and parts of Central and South America (Udall, 2007). Movement of microfilariae to the eye can lead to significant pathological effects, with an estimated 500 000 people suffering from visual impairment and 270 000 from blindness (World Health Organization, 1995b). The microfilariae can survive for 1–2 years before dying. In the past, blindness caused by microfilarial infection of the eyes affected up to 50% of adults in some areas of the West African highlands while today, 99% of all humans suffering from river blindness live in Africa. Only the West African

346

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

savannah form of O. volvulus leads to river blindness. This form is transmitted by different members of the S. damnosum species group (Basa´n˜ez et al., 2006). River blindness has led to significant socioeconomic effects, with the abandonment of fertile agricultural land near rivers and streams, as well as to psychological trauma due to disfigurement of the skin (Evans, 1989; Dadzie et al., 2002; Lazdins-Helds et al., 2003; May, 2007). Prior to the initiation of control programs in West Africa, microfilarial infection was higher in males than females and increased rapidly in both sexes until about the age of 25 years when it plateaued at 90% for males and 85% for females (Kirkwood et al., 1983). Average prevalence in the villages studied was 60% (range 5–93%) (Kirkwood et al., 1983).

3.12.6.2.4 Disease characteristics in humans Human disease is caused predominantly by the inflammatory response to dying microfilariae which can occur over a period of many years. The initial symptoms usually involve an intensely itchy, diffuse papular dermatitis. Once the disease becomes chronic, the cutaneous manifestation differentiates across a spectrum ranging from pruritic lichenification to asymptomatic depigmentation (Udall, 2007). These can be severe, with skin lesions, inflammation, and swelling accompanied by severe itching. Wasting and loss of elasticity are consequences. The formation of sub-dermal nodules (onchocercomata containing 2–50 female and 1–10 male worms) are also a common symptom (Udall, 2007). Several hundred such nodules may be present, ranging from 1 to 5 cm in diameter. In Africa, these occur frequently on the torso and hips while in South America they are more common on the head and shoulders (Udall, 2007). Microfilariae frequently migrate to the eyes where inflammation gradually leads to sclerosal opacification and blindness (Udall, 2007). Not only does onchocerciasis cause major morbidity in humans, a recent report also suggests that the life expectancy of those affected is reduced and that mortality is significantly correlated with microfilarial burden (Little et al., 2004). Savannah ocular involvement is severe while that found in forests is mild (Basa´n˜ez et al., 2006). Diagnosis of onchocerciasis in the past was based on finding evidence of microfilariae in skin biopsies (skin snips). Currently, biochemical and molecular biological methods, including antigen assays and PCR to detect parasite deoxyribose nucleic acid (DNA), are available (Udall, 2007).

3.12.6.2.5 Prevention and cure A number of onchocerciasis control programs have been largely successful in reducing the burden of this disease in both Africa and South America and reducing the morbidity and socioeconomic burden in these areas (Richards et al., 2001; Molyneux, 2005). Today, prevention involves two major strategies, vector control and treatment of humans, both aimed at breaking the transmission cycle (Basa´n˜ez et al., 2006). Vector control currently involves killing larval blackfly by applying insecticides to the rivers and streams in which they live (Davies, 1994; Le´veˆque et al., 2003). The insecticide of choice depends on the development of resistance of the blackflies, with different chemicals potentially being used on a

rotational basis (Hemingway and Ranson, 2000). The concentration of the larvae in these limited habitats makes the control effort much easier than that which would be required for the widely dispersed adult flies (Le´veˆque et al., 2003). The Onchocerciasis Control Program (OCP) in West Africa, which involved 11 countries and was financed by the WHO, the Food and Agriculture Organization (FAO), the United Nations Development Program (UNDP), the World Bank, and 22 additional donors from 1974 to 2002, was the most comprehensive and successful program to combat this disease (Hodgkin et al., 2007). More recently, the African Program for Onchocerciasis Control (APOC) was initiated aimed at functioning from 1995 to 2010, with additional promised support for national task forces until 2015 (Hodgkin et al., 2007). The concept of the OPC program was to combine chemotherapy for humans with ivermectin (Mectizan generously donated by Merck & Co. Inc.) with helicopter application of insecticides (the organophosphates temphos and chlorphoxin) as well as the biological larvicidal toxin produced by Bacillus thuringiensis var. israelensis (B.t.i.) into the riverine habitats of the simuliid larvae. The latter was mainly sprayed during the dry season in areas showing resistance to temphos. B.t.i. showed the lowest degree of environmental damage and impact on nontarget invertebrates including the most important natural predators of the vectors (Richards et al., 2001; Le´veˆque et al., 2003). However, it had only a limited effect in areas with algal blooms, invasive water hyacinths, or strong water flow. The insecticides need to be applied on a weekly basis as the development of larvae to pupae takes about this time. Together with the vast stretches of water courses needing treatment (40 000 km of river treated over 106 km2 by the OCP; Hougard et al., 1997) and the requirement for aerial spraying in some areas, control efforts become both expensive and time consuming (Le´veˆque et al., 2003). In addition, although claimed to be environmentally safe, Le´veˆque et al. (2003) have shown a limited effect on fish but changes in the community structure and taxa present to the level of family. Treatment of patients with ivermectin paralyses and kills microfilariae, preventing disease progress as well stopping the females from producing microfilariae for some months after treatment, thus interrupting the transmission cycle (Diawara et al., 2009). Ivermectin treatment should be repeated every 6–12 months. Although effective in controlling the disease, the elimination of transmission has proven difficult as, after an initial drastic reduction in the number of cutaneous microfilariae, these begin to reappear at 20% or more of the initial intensity within a year (Hoerauf et al., 2003; LazdinsHelds et al., 2003). However, recent evidence has shown the feasibility of eliminating onchocerciasis in certain endemic foci via ivermectin treatment alone (Diawara et al., 2009). The possibility of eliminating adult worms by killing their Wolbachia endosymionts using doxycycline is also being examined (Hoerauf et al., 2003; Udall, 2007). Other onchocerciasis control programs are more limited. The APOC only involves treatment of the infected human populations with ivermectin, as does the Onchocerciasis Elimination Program for the Americas (OEPA; 1991 to 2010) involving six Latin American countries (Richards et al., 2001; Le´veˆque et al., 2003).

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control 3.12.6.2.6 Anthropogenic alterations to the environment Some of the alterations to the environment brought about by humans are responsible for increasing the spread of disease: 1. Dams. The Simulium species which transmit O. volvulus inhabit a wide variety of flowing water environments with specific microhabitat preferences. Any changes to these habitats are likely to lead to changes in the species or community of Simulium species present. Thus, the damming of these watercourses can reduce or eliminate water flow reducing the amount of suitable habitat available for immature blackflies (Sutherst, 2004). However, the introduction of new habitats, such as spillways on dams, is likely to provide suitable habitat for certain species and may lead to the establishment of new foci (Patz et al., 2000; Sutherst, 2004). According to Taylor et al. (2009) the simuliid density along the the Sanaga River has increased due to the initiation hydroelectric schemes, perhaps due to the construction of spillways. Adewale et al. (1999) found abundant vectors around the Owena dam in Nigeria. This study showed that 0.4% of the blackflies were infected, while 0.3% had infective larvae. An annual transmission rate of 109 larvae per person per year was calculated which is slightly lower than in a rural area without direct association to a dam (131–189 larvae per person per year, Opara et al., 2008). This suggests that construction of the dam has not substantially reduced the abundance of the blackflies and the prevalence of infection with O. volvulus. 2. Deforestation. The effect of deforestation is variable depending on the local simuliid vector. For those species, such as S. neavei and S. woodi, which require shaded streams for breeding or are attracted to dense forest patches as adults, deforestation can lead to substantial reductions in abundance or to extinction (Muro and Raybould, 1990; Taye et al., 2000; Garms et al., 2009). However, an increase in open savannah areas can lead to a change in the vector species present, as we saw above for malaria, with an increased likelihood of transmission by savannah species. In the case of onchocerciasis, this is associated with the occurrence of the more severe form of the disease (Wilson et al., 2002; Adjami et al., 2004). Where forest vectors are present, movement of infected humans into forest areas, such as occurred in Central and South America, can lead to the introduction of O. volvulus to new areas (Sutherst, 2004). 3. Eradication programs. The OCP eradication project also emphasized a concern for local biodiversity and the implementation of long-term freshwater monitoring, a very progressive position among international development projects for this time. An independent ecological committee was in charge of risk assessment for all applied larvicides and for the documentation of aquatic biodiversity. This resulted in a substantial increase in our knowledge of the fauna and ecology of West African rivers. However, when the success of the project made available 25 million hectares (ha) of fertile riverside areas for resettlement and agriculture, ecological aspects were no longer taken into consideration. These newly available areas became subject to unplanned and unsuitable colonization. This led to

347

land-use changes on an enormous scale with rampant deforestation, soil erosion, and loss of biodiversity. Pesticides far more hazardous than temephos and chlorphoxin, as well as fertilizers used in agriculture, leached into the rivers. Migrant fishermen, partly fishing with agricultural poisons, depleted local fish stocks and subsequently moved quickly to more pristine stretches of other rivers. Both fish and invertebrate species became endangered (Le´veˆque et al., 2003). Thus, although commencing with the praiseworthy aim of maintaining an intact, functioning ecosystem during the initial phase of the project, no forward ecological planning, based on the new availability of large natural areas, was made. Had this been done, the local population would have had access to a wide range of ecosystem services, such as sustainable resources and tourism, based on the conversion of part of the river and its surroundings to a national park containing a wide diversity of endangered plants and animals including attractions such as pygmy hippoppotami. West Africa, where onchocerciasis is most common, is undergoing extensive deforestation (Caspary, 1999; Schipper et al., 2008) and lags substantially behind East Africa in the area of ecotourism (Sournia, 1996). The intensive, uncontrolled use of riverine areas by humans implies that these options are either no longer possible or are severely reduced.

3.12.6.2.7 Recommendations Prior to the anthropogenic change of hydrological patterns, it is necessary to determine whether new, suitable habitats for Simulium vectors will be created allowing new foci of onchocerciasis to develop. Should such habitats be created, they should be monitored and if necessary, sprayed with an appropriate insecticide (dependent on the resistance status of the Simulium vectors) to prevent the establishment of the vector. Modern technology has provided a substantial impetus to the control of onchocerciasis. Satellite detection of water-flow rate has been used to calculate the amount of insecticide to be used in vector-control programs (Boyer et al., 1990). Geographic information system analysis can also be used to predict the areas where onchocerciasis prevalence is likely to be high and to apply appropriate control measures (Noma et al., 2002; Botto et al., 2005; Gebre-Michael et al., 2005). The d-endotoxin produced by B.t.i. is effective against blackfly larvae at very low concentrations (a few nanograms per milliliter). In West Africa, this has proved effective in large-scale programs aimed at blackfly control (Lu¨thy and Studer, 1986). As seen in the OCP project, it is not enough to control the vectors and the disease. Detailed planning on sustainable land use and the maintenance of functioning ecosystems and their services must also be made for the time after the project succeeds.

3.12.6.3 Lymphatic Filariasis 3.12.6.3.1 Parasite characterization Lymphatic filariasis is caused by any one of three species of nematode parasites (family Onchocercidae) transmitted by mosquitoes (Figure 33). Bancroftian filariasis, due to Wuchereria bancrofti, occurs throughout the tropics, while brugian filariasis, is found in the region from India across

348

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control Mosquito takes a blood meal (L3 larvae enter skin)

1 Mosquito stages

Human stages

8 Migrate to head and mosquito’s proboscis

7 L3 larvae

i 2 Adults in lymphatics

6

L1 larvae

3 4 Mosquito takes a blood meal (ingests microfilariae)

Adults produce sheathed microfilariae that migrate into lymph and blood channels

5 Microfilariae shed sheaths, penetrate mosquito’s midgut, and migrate to thoracic muscles d

i

= infective stage

d

= diagnostic stage

Figure 33 The life cycle of Wuchereria bancrofti. Different species of mosquitoes are vectors of W. bancrofti filariasis depending on geographical distribution. During a blood meal, an infected mosquito introduces third-stage filarial larvae onto the skin of the human host, where they penetrate into the bite wound 1 . They develop in adults that commonly reside in the lymphatics 2 . The female worms measure 80–100 mm in length and 0.24– 0.30 mm in diameter, while the males measure about 40  0.1 mm. Adults produce microfilariae measuring 244–296 mm  7.5–10 mm, which are sheathed and have nocturnal periodicity, except the South Pacific microfilariae which have the absence of marked periodicity. The microfilariae migrate into lymph and blood channels moving actively through lymph and blood 3 . A mosquito ingests the microfilariae during a blood meal 4 . After ingestion, the microfilariae lose their sheaths and some of them work their way through the wall of the proventriculus and cardiac portion of the mosquito’s midgut and reach the thoracic muscles 5 . There, the microfilariae develop into first-stage larvae 6 and subsequently into third-stage infective larvae 7 . The third-stage infective larvae migrate through the hemocoel to the mosquito’s proboscis 8 and can infect another human when the mosquito takes a blood meal 1 . Reproduced with permission from CDC.

Southeast Asia to China and the Philippines (due to Brugia malayi). On Timor, Flores, and other Indonesian islands, a third species, B. timori is involved (Michael and Bundy, 1997; Melrose, 2002; Service, 2004). The three species are similar in appearance with females reaching a length of 65–100 mm by 0.2–0.3 mm wide and males only about 40 mm by 0.1 mm. Microfilariae vary between 222 mm and 310 mm, depending on species, and are sheathed, that is, the L2 larva still carries the cuticle of the L1 larva as a loose sheath (Lucius and LoosFrank, 2008). There are more than a billion people at risk of contracting lymphatic filariasis (lymphedema and elephantiasis) worldwide, with an estimated 120 million affected in 80 countries and 40 million incapacitated or disfigured (Dreyer et al., 2000; Molyneux, 2003). The numbers infected are relatively equally divided between Africa, India, and Southeast Asia and the Pacific. Wuchereria bancrofti is responsible for about 90% of

cases of lymphatic filariasis. Bancroftian filariasis, caused by this species, is largely an urban disease (Michael and Bundy, 1997; Melrose, 2002). Only W. bancrofti is completely anthroponotic, lacking animal hosts (Service, 2004); the subperiodic form of B. malayi is a zoonosis known to infect primates as well as domestic cats and dogs (Chansiri et al., 2002; Melrose, 2004). All species are transmitted by anopheline (genus Anopheles) and by several genera of culicine mosquitoes (e.g., Aedes, Mansonia, and Ochlerotatus: Figure 33 and 34; Michael et al., 1994; Service, 2004), which in turn spend their egg, larval, and pupal stages in and/or on water (Service, 2004).

3.12.6.3.1 Developmental cycle Adult worms of all three species occur in the human lymphatic system (Figure 33) where they produce sheathed microfilariae

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Figure 34 Armigeres subalbatus, a culicine mosquito which occurs from Pakistan in the west, all through Southeast Asia to Indonesia, and north to Japan and Korea. Note the typical culicine posture with the body horizontal to the surface and only the head bent downward. Larva, compare Figure 29. Reproduced with permission from CDC.

(L2 larvae) that migrate into the lymph and blood vessels. Mosquitoes feeding on the blood ingest the microfilariae which then shed their sheaths, penetrate the mosquito’s midgut, and migrate through the hemocoel to the thoracic muscles where they continue their development to L3 larvae (1.2–1.6 mm long). As for Onchocerca volvulus, there is no multiplication of the parasite in the vector. The larvae migrate through the head of the mosquitoes to the labium of the proboscis. During the next blood meal on a suitable host, the L3 larvae penetrate the labium and crawl on to the host’s skin. Then a lesion, usually the bite site of the mosquito, is required before they can enter the host and move to the lymphatic system. Larval development of W. bancrofti and B. malayi can take place in a number of mosquito species, but B. timori is probably only vectored by Anopheles barbirostris (Fischer et al., 2002; Service, 2004). In rural areas, in Africa, for example, W. bancrofti is transmitted by the Anopheles funestus group and members of the Anopheles gambiae complex: in freshwater A. gambiae sensu stricto and A. arabiensis; and in salty water A. melas and A. merus (Bockarie et al., 2009).

3.12.6.3.3 Human involvement Prevalence is initially age dependent, plateauing in early adulthood, but with infection often occurring early in childhood and potentially leading to disfigurement for much of the individual’s life (Weil et al., 1999; Witt and Ottesen, 2001; Melrose, 2004). In two studies from the 1990s, 25% of 5-yearolds on Haiti were already infected (Lammie et al., 1994, 1998), while in Tanzania, prevalence for the same age group reached 28% (Simonsen et al., 1996). In the Nile Delta of Egypt, antibody prevalences reach 30% in the 10–19-year age group, peak at over 40% for 20–29-year-olds then slowly decline to about 30% again between 60 and 80 years (Weil et al., 1999). There is some indication that females of reproductive age are less commonly infected than age-matched males (Melrose, 2004), and in India 79% of chronic cases occur in men (Ramaiah et al., 2000).

349

Acute filarial disease is generally more common than the chronic disfiguring manifestations (Melrose, 2004). In some tropical and subtropical areas, the prevalence is continuing to increase, although elimination is easy and inexpensive, and various strategies have eliminated transmission in China, Japan, Korea, Thailand, and the Solomon Islands (Molyneux, 2003). It is mainly a disease of the poor, being associated with the expansion of slums in urban areas. There are two forms of both bancroftian and brugian filariasis: nocturnal periodic and nocturnal sub-periodic. In the periodic form, most microfilariae are found in the pulmonary blood vessels during the day while at night they migrate to the peripheral blood and lymph systems. These strains have adapted to mosquito vectors which feed during the night. Reduced periodicity is found in the sub-periodic form in which the microfilariae are present in the peripheral blood during the day as well as night (Service, 2004). The filarial worms live in the lymphatic system and have a fecund life span estimated at 5–15 years, producing millions of microfilariae in their lives (Melrose, 2004). Prevalences in endemic areas can be high. In their 1997 analysis, Michael and Bundy reported 17 of the 34 African countries where the disease is found to have countrywide prevalences of over 10%, with the four highest being in Guinea Bissau (37%), Comoros (27%), the Seychelles (24%), and Nigeria (22%). The highest prevalences, however, occur on Tonga (48%) and Papua New Guinea (39%) (Michael and Bundy, 1997). The socioeconomic and psychological burden of lymphatic filariasis is immense. India harbors 40% of all individuals with lymphatic filariasis. Ramaiah et al. (2000) estimated that this disease costs US$842 million per annum in treatment time and lost working time, equivalent to 0.62% of the gross national product at this time. A chronically infected male patient with acute episodes could lose up to US$50 per year, 15% of the total available income. In addition to the physical effects of the disease, the deformities which it causes lead to stigmatization of affected individuals (Melrose, 2004).

3.12.6.3.4 Disease characteristics in humans Filariasis shows various manifestations in humans. Many of those infected, including children, appear to be asymptomatic while in fact the adult worms and circulating microfilaria cause observable damage to the lymphatic system, tissue, and organs (Witt and Ottesen, 2001; Melrose, 2004). These infections may be a preliminary to chronic disease in later life (Witt and Ottesen, 2001). Acute disease can occur at any age with reports from babies as young as 3 months (Melrose, 2004). Fever (elephantoid fever) and chills accompany adeno-lymphadenitis characterized by severe inflammation of the lymph node and lymphatic vessels which lasts for about a week before resolving spontaneously (Melrose, 2004). Chronic disease leads to the most spectacular symptomatology, with elephantiasis being the final form. It usually starts with the accumulation of fluid in the extremities, scrotum, vulva, or breasts due to obstruction of the lymph vessels and can lead to severe hypertrophy of the skin and subcutaneous tissues (Dreyer et al., 2000; Melrose, 2004).

350

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Fortunately, elephantiasis only occurs in a small proportion of those suffering from the disease (Melrose, 2004). Infection has been reported to affect a wide variety of other organs including the kidneys, joints, eye, heart, and spleen. Details are provided in the comprehensive review of Melrose (2004). A variety of methods are available for diagnosing filariasis including the demonstration of microfilariae by either direct or concentration techniques, and the detection of filarial antigens, the detection of filarial antibodies, and the detection of parasite DNA (Melrose, 2004; Melrose et al., 2004).

3.12.6.3.5 Prevention and cure Reduction or elimination of the microfilariae can be achieved by administering a single dose of diethylcarbamazine (DEC) together with albendazole, which is effective over a year, in areas where onchocerciasis is not endemic (Hotez et al., 2007). The simultaneous administration of both albendazole and ivermectin removes 99% of filariae from the blood, also for a full year, and is recommended in areas where onchocerciasis is also endemic (Hotez et al., 2007). At the community level, these treatments effectively reduce transmission by preventing microfilarial uptake by mosquitoes. Both albendazole and DEC effectively kill the adult worms (Horton et al., 2000). Although this can lead to improvement in the disfiguring elephantiasis and hydrocele, particularly when the disease is in an early stage, it cannot cure them as progression of the disease is due to superinfection by fungi or bacteria. The latter can be treated by strict hygiene, minimizing the risk of infection, pharmaceutical treatment against the superinfecting agent, and measures to increase lymph flow (Dreyer et al., 2000). Lymphatic filariasis is also subject to a number of international control programs which have shown considerable success in recent years (Molyneux, 2003; Molyneux et al., 2003). In 2000, 2 million treatments were made available worldwide. This was increased to 60 million in 2002 through the generosity of GlaxoSmithKline which has donated albendazole, the drug of choice since 1998, and Merck, which now also supplies ivermectin for these programs (Molyneux, 2003). These treatment regimens have been successful; however, they must be applied continuously over many years and in many areas. Educating the local population is a necessary addition to the treatment. It is important to note, however, that individuals with densities of microfilariae as low as 3 ml1 blood can still infect mosquitoes and that, therefore, residual low-density infections after mass-treatment programs have the potential to cause a rapid re-emergence of the disease (Melrose, 2004). Vector control is also possible. Open breeding sites such as areas of flooded land, cess pits, and blocked drains can be treated with modern insecticides such as pyriproxyfen, an insect growth regulator, or the biological control agent, Bacillus sphaericus (Bockarie et al., 2009). This is related to B.t.i. but has a narrower range of potential target organisms of which the highly susceptible Culex quinquefasciatus, a major vector of W. bancrofti and a highly successful invading species, is one (Bockarie et al., 2009). In addition, residual house spraying and the use of permethrin-impregnated bed-nets have proven

to be highly effective by reducing the mosquito biting rate (Bockarie et al., 2009).

3.12.6.3.6 Anthropogenic alterations to the environment The nocturnal periodic form of W. bancrofti is transmitted over much of its range by C. quinquefasciatus which breeds preferentially in polluted water containing organic waste (including human and animal feces) such as cess pits, latrines, and drains (Subra, 1981; Raccurt et al., 1988). The unsanitary conditions in rapidly growing urban environments therefore provide ideal breeding grounds for this vector (Mian and Mulla, 1986; Calhoun et al., 2007; Chaves et al., 2009). C. quinquefasciatus is the cause for high prevalences of bancroftian filariasis in urban areas with low sanitary standards. As it is a night-biting mosquito, urban filariasis is associated with nocturnal periodicity of the microfilariae (Melrose, 2004). According to Bockarie et al. (2009), the distribution and abundance of C. quinquefasciatus is increasing in parallel to urbanization and human activity. Many rural pockets, which were relatively free of this mosquito, are becoming increasingly colonized. On the Ivory Coast, where bancroftian filariasis used to be an urban phenomenon, two villages were compared to each other: one with a traditional way of life, the other with modern conveniences (Dossou-Yovo et al., 1995). In the latter, where water pollution and household rubbish, cess pits, and septic tanks were abundant, the biting rate of the C. quinquefasciatus was significantly higher throughout the year compared to that of the traditional village. In Polynesia, a sub-periodic form of bancroftian filariasis occurs with a diurnal peak of microfilariae in the peripheral blood. An important vector of this form is Aedes polynesiensis which uses natural sites of water collection, such as leaf bracts and crab holes for breeding, as well as containers conveniently distributed by the human population such as tires, tins, jars, etc. (Service, 2004). In southern Ghana, the availability of irrigation canals late in the dry season led to a massive increase in the number of Anopheles gambiae mosquitoes. These had a filarial infection rate with W. bancrofti of 8.3%, comparable to those of the wet season (Dzodzomenyo et al., 1999). Indeed, Erlanger et al. (2005) indicate that ‘‘environmental changes due to water resource development and management consistently led to a shift in vector species composition and generally a strong proliferation of vector populations.’’ Irrigation projects in West Africa have been associated with an increase in W. bancrofti infections. Almost every house in villages bordering a man-made irrigation ditch contained a case of bancroftian filariasis. By contrast, only scattered cases were seen in villages two or more kilometers away from these ditches. Anopheles gambiae and A. funestus complex mosquito densities were up to 25 times higher in areas under water-resource development as compared to non-irrigated sites (Erlanger et al., 2005).. Little information is available on the role of deforestation in changing the prevalence of lymphatic filariasis. By analogy with malaria caused by Plasmodium falciparum, which in rural Africa is transmitted by the same vectors as W. bancrofti, we would predict that deforestation should also lead to an increased incidence of bancroftian filariasis. There is, however, a

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

negative spatial association between the two parasite species (Kelly-Hope et al., 2006).

3.12.6.3.7 Recommendations As both bancroftian and brugian filariasis are transmitted by a wide variety of different mosquito species, each with its own breeding habitat requirements, complete control of vectors is unlikely. However, in the case of Culex quinquefasciatus discussed above, increased hygiene, water treatment, and the removal of water-filled tires and other containers containing organic material, such as rotting vegetation, could reduce the breeding success of this species in urban slums. Control using insecticides has been locally successful (Bockarie et al., 2009). Aedes polynesiensis and C. quinquefasciatus populations could potentially be reduced by removing litter which acts to collect water and thus can be used as breeding sites. In Africa, the non-urbanophilic Anopheles species can be controlled as indicated above for malaria, as these are often the same species transmitting the filarial parasites (Melrose, 2004).

3.12.7 Environmental Factors Influencing the Dynamics of Water-Associated Parasites The epidemiology of the parasites discussed above, although following set patterns, is not static but dynamic and potentially unstable. This dynamism is based on the way in which the parasite and its various intermediate and final hosts react to natural and man-made global and local changes in the environment. We provide general examples of the environmental factors influencing these epidemiological cycles and summarize the information presented for the various waterborne parasites treated earlier.

3.12.7.1 Dam Construction and Irrigation Projects With the continual increase in the human population, a reliable supply of high-quality freshwater, for human consumption, agriculture, and industry, has become of critical importance (Vo¨ro¨smarty and Sahagian, 2000; Oki and Kanae, 2006). This has led to a major increase in the construction of dams and the initiation of irrigation projects. Keiser et al. (2005a) cite the construction of 40 000 large dams, 800 000 small dams, and 272 million ha of irrigated land worldwide over the last 50 years. As discussed here, such projects can lead to major deterioration in the health status of the local population due to increases in parasite and pathogen-transmission rates. This is the result of increasing the amount of habitat required for the survival and/or reproduction of the vectors (Aedes, Anopheles, Culex, etc.), or intermediate hosts (e.g., snails) traditionally transmitting the disease in the respective area, or related indigenous or introduced species which also act as vectors colonizing the habitat after its alteration by a hydrological engineering project. This shift may take years before a new equilibrium is reached. Irrigation projects are aimed at increasing the availability of reliable water supplies for agriculture in areas with limited or only seasonal water supplies. Thus, such irrigation projects frequently supply the conditions required for the maintenance

351

of parasitic diseases in areas where they would otherwise either not exist or would be of only limited significance. Schistosomiasis (Steinmann et al., 2006), malaria (Keiser et al., 2005a), and lymphatic filariasis (Erlanger et al., 2005) fit this pattern with known examples of increases in disease incidence with inadequately planned water-development projects. In all cases, consultation with specialists on the epidemiology of parasitic diseases including the ecology of their vectors or intermediate hosts could have predicted and reduced the health-risk consequences of the development projects. Zhang et al. (2002) and Li et al. (2007), for example, provided models showing a projected increase in schistosomiasis around the Three Gorge Dam and the Dongling Lake Region in China based on predicted hydrological and epidemiological data. Thus, parasitological predictions for hydrological projects are feasible. The Gu¨Neydogu Anadolu Projesi (GAP) project in southeastern Turkey includes the construction of 22 dams on the Euphrates and Tigris rivers with the aim of providing irrigation water to 1.7 million ha (10% of the surface area of the country) of arid land as well as to about 10% of the Gu¨neydogu Anadolu Projesi population (Aksoy et al., 1995). This ambitious project also appears to have more than doubled the incidence of malaria between 1990 and 1992 from 8680 cases to 18 676 cases (Aksoy et al., 1995). Other water-associated diseases from the area, including Entamoeba histolytica and Giardia lamblia, have also increased in incidence. Ak et al. (2006) have shown that dam construction and irrigation per se have not influenced the spread of these intestinal parasites, but that the elevated levels of parasitization are related to contamination of drinking water, raw green vegetables irrigated with feces-contaminated wastewater, the use of human fecal material as fertilizer, and the use of contaminated water to clean raw vegetables for consumption. Predicting the effects of hydrological change requires knowledge of all the parasite groups present or likely to move into the affected area. The creation of dams may have quite different effects on different parasite groups. For instance, it has been shown that the elimination of flowing-water habitats required by the simulian vectors reduced levels of onchocerciasis. The same hydrological changes, however, increase the area of standing-water habitat available for the breeding of mosquito species transmitting malaria and filariasis and for the intermediate snail hosts of schistosomiasis (Sutherst, 2004).

3.12.7.2 Land-Use Changes All free-living organisms have specific habitat requirements which will be impinged upon by changes in land use. These organisms include the vectors and intermediate hosts for the variety of parasite species which we have discussed above. Thus, as land use changes, the epidemiological cycles of the parasites actually or potentially present in a given area will also be modified; indeed, depending on the degree of impact, they may disappear or appear in areas where they were not previously present. Today, land-use changes are considered a major driving force in shifting epidemiological patterns (Patz et al., 2004). In terms of parasites with a waterborne transmission phase, land use is likely to have a major effect by

352

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

changing both water-flow patterns as well as the quality of the water available (Scanlon et al., 2007). The massive increases in rainfed crop and pasture land over the last several hundred years at the expense of forest and grasslands has led to decreased rates of evapotranspiration and increased rates of recharge of groundwater and streamflow, together with changes in the chemical composition of the water with mobilization of salts, salinization, and leaching of fertilizers (Scanlon et al., 2007). All of these parameters can have a significant impact on species assemblages; for example, calcium is required for the growth of most snail species, including those involved as intermediate hosts for Schistosoma spp. (Williams, 1970; Nduku and Harrison, 1976; Madson, 1987). Seasonal changes in the abundance of Biomphalaria pfeifferi, an intermediate host of Schistosoma mansoni, are related to water temperature and flow rate (Woolhouse, 1992), while high magnesium levels have a negative impact on egg production (Harrison et al., 1966). Moreover, this species is rarely found in conditions with a sodium/calcium ratio of more than 2.4 (Schutte and Frank, 1964) and not at all with conductivities below 12 mS cm1 (Polderman et al., 1985). This indicates that multiple factors are involved in determining the distribution of B. pfeifferi (Utzinger et al., 1997). Deforestation provides many examples of an increased impact on the spread and development of Anopheles spp. mosquitoes leading to an increase in malaria incidence (Yasuoka and Levins, 2007). However, exception does occur. In Thailand, however, malaria can be caused by Plasmodium falciparum and P. vivax, the two most common species, and more rarely by P. malariae, P. ovale, and the zoonotic species P. knowlesi, the importance of which may have been underestimated (Chareonviriyaphap et al., 2000; Jongwutiwes et al., 2004; Cox-Singh et al., 2008). Malaria is most common in forest and scrub areas with the major vectors in the northeast of the country, mosquitoes of the Anopheles dirus complex, occurring preferentially in shady forest and forest-fringe habitats (Walsh et al., 1993; Yasuoka and Levins, 2007). During the first half of the last century, malaria was common in the northeast of Thailand (Petney et al., 2009), but the incidence of infection decreased rapidly as deforestation progressed (Petney et al., 2009). By 1992, only two of the then 17 northeastern provinces fell within the top 15 malarial provinces in Thailand (Thimasarn et al., 1995). Both these adjoining provinces were low on the list (13 and 15, respectively) and both are close to malarial areas in Laos and Cambodia. Between 25% and 31% of cases were considered to be imported (Thimasarn et al., 1995). Today, the northeast area is considered to lave a low risk of infection (Chareonviriyaphap et al., 2000). This reduction in risk is likely to be directly related to the reduction in forest area and thus the habitat required by the vector mosquito. Areas currently requiring control are predominantly hilly and forested on the borders with Burma and Cambodia (Chareonviriyaphap et al., 2000). Today, major increases in the area covered by rubber plantations, which provide the necessary shade for the freshwater habitats needed for breeding malaria mosquitoes, could herald the return of malaria to this area (Petney et al., 2009). Patz et al. (2004) discuss policy recommendations on landuse change in relation to infectious diseases. They stress the complexity of the cascades of factors influenced by even a

single change, which can affect disease emergence. Policy decisions involve specific health-risk factors, landscape and habitat change, and economic and education considerations. The recommendations involve (1) providing a conceptual model integrating land use into public health policy; (2) more research on the relationship between deforestation and infectious diseases; (3) the development of policies to reduce pathogen contamination; and (4) the creation of centers of excellence in ecology and health research with appropriate training courses.

3.12.7.3 Mass Animal Husbandry: Cryptosporidia and Giardia Wild and domestic animals are known hosts for a variety of zoonotic diseases affecting humans (Slifko et al., 2000; Simpson, 2002). The introduction of mass animal husbandry of cattle, pigs, and poultry, with the close confinement of many individuals in a small space, provided ideal conditions for the long-term survival and the transmission of a wide variety of parasites (Gajadhar et al., 2006). Cryptosporidium parvum and Giardia lamblia, which have been dealt with briefly earlier, are two species commonly contracted by humans through drinking water contaminated with fecal material. Outbreaks of C. parvum sensu stricto involving both cattle and humans have also been reported (Monis and Thompson, 2003; Peng et al., 2003; Soba and Logar, 2008), although the mass outbreak of cryptosporidiosis in Milwaukee, initially thought to be due to water contamination with cattle feces, was in fact due to human specific C. hominis (Gajadhar and Allen, 2004). Aquaculture is undergoing a massive growth phase. In 1970, only 3.9% of total aquatic animal food, including fish, crustaceans, mollusks, and other aquatic animals, came from this source. By 2006, this had increased almost 10-fold to 36%, while if fish alone were considered, 47% came from aquaculture (Food and Agriculture Organization, 2009). In China, 90% of fish food production derived from this source. The introduction of parasites to populations with such high densities is likely to lead to increased rates of transmission, depending on the presence of suitable intermediate hosts and contamination of the water with human waste (e.g., the eggs of fish-borne trematodes). The fact that a number of important intermediate hosts, such as the snail Melanoides tuberculata, are highly successful invading species, which are associated with aquaculture, facilitates the establishment of transmission cycles in new areas (Garrett et al., 1997).

3.12.7.4 Human Conflict, Political Considerations, and Healthcare Systems Human conflict, which in the future has been predicted to increasingly include the right to access freshwater (Toset et al., 2000; Hensel et al., 2006), is also capable of limiting the effectiveness of control campaigns against parasites with a freshwater transmission cycle. In 1980, the Centers for Disease Control in Atlanta, Georgia, proposed a global eradication campaign for dracunculiasis; this was later accepted by the United Nations. In 1988, the ministers of health from African countries agreed

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

upon the eradication program aimed at being completed by 1995 (Hopkins et al., 1997, 2002). Today, 14 years after this deadline, a great deal of progress has been made toward the goal of eradication (Hopkins et al., 2002; Barry, 2007). Indeed, the number of cases has plummeted from almost 900 000 in 1989 to about 25 000, and of the 20 countries with endemic dracunculiasis at the beginning of the campaign, seven had interrupted transmission by 2002, including all countries outside of Africa (Hopkins et al., 2002). By 2007, dracunculiasis was only recorded from five countries, of which three had a total of a mere 170 cases, and only two, Sudan and Ghana, had more than a 1000 cases (Iriemenam et al., 2008). One stumbling block preventing complete eradication, however, remains – the civil war in Sudan and ethnic fighting in Ghana (Hopkins and Withers, 2002; Iriemenam et al., 2008). Data from 2007 indicate that 60.3% of all cases of dracunculiasis came from the Sudan and 37.7% from Ghana (Iriemenam et al., 2008); however, although there is commitment to and progress in fighting the disease in Ghana, and eradication has been successful in northern Sudan, there are still substantial difficulties in southern Sudan (Hopkins et al., 2005; Iriemenam et al., 2008) and it is highly unlikely that global eradication of this disease will be possible as long as these conflicts continue (Hopkins and Withers, 2002). The prevalence of malaria is also susceptible to the impact of conflict situations. Malaria transmission is often higher than usual in areas containing refugee populations due both to lack of control and medication, and to high population densities (Na´jera, 1996; Martens and Hall, 2000). In addition, the collapse of health services due to civil unrest after the break-up of the Soviet Union led to a substantial reemergence of malaria in Tajikistan (Pitt et al., 1998), while population displacement from Afghanistan to Pakistan after the Soviet invasion of the former in 1979 led to an increase in the local number of cases in refugees from 11 200 in1981 to 118 000 in 1991 (Rowland et al., 2002). Population movement within Afghanistan has led to the introduction of malaria to the Bamian Valley at an altitude of 2250–2400 m, well above its normal range (Rab et al., 2003).

3.12.7.5 El Nino The El Nino Southern Oscillation (ENSO) is a complex geophysical phenomena in the Pacific Ocean leading to major cyclic shifts in weather patterns, including changes in rainfall on a worldwide scale (Cane, 2005). These shifts in turn can have a significant influence on ecological processes in the affected areas (Jaksic, 2001; Stenseth et al., 2002). Historical evidence shows a correlation between ENSO events and malaria epidemics, which can be related to patterns of monsoon rainfall (Bouma and Van der Kaay, 1996; Zubair et al., 2008). Kovats et al. (2003), point out that malaria epidemics usually occur in areas where transmission rates are too low to lead to widespread immunity. In those areas where transmission is restricted by climate, even small climatic changes can trigger an epidemic. Increases in malaria cases have been associated with the ENSO in South America (Bouma and Dye, 1997; Poveda et al., 2001), Africa (Lindblade et al. (1999), but see Lindsay et al. (2000) who found a significant reduction in

353

Malaria in highland Tanzania) and Southern Asia (Zubair et al., 2008). The relationship between malaria incidence and El Ninodriven climatic phenomena is sufficiently great to be incorporated into predictive models of malaria outbreaks (Thomson and Connor, 2001; Anyamba et al., 2006).

3.12.7.6 Climate Change Anthropogenically caused global climate change will lead to significantly elevated temperatures and changing rainfallpatterns worldwide (IPCC, 2007). The parasites dealt with here, all of which have temperature-dependent developmental phases and rely, at least during part of their lifecycle, on aquatic environments (i.e., rainfall), are climate dependent (Sutherst, 2004; Mas-Coma et al., 2008). Moreover, their dissemination within the environment is dependent on runoff patterns (Rose et al., 2001). Nevertheless, considering the importance of these parasites for human health, remarkably little concrete information is available on how climate change is likely to influence parasite transmission rates, and the information which we do have suggests that this influence may be highly complex and the resulting patterns not readily predictable (Marcogliese, 2001; Poulin, 2005). The relationship between the incidence of El Nino and malaria shows the potential close association between vectorborne pathogens and climate. This, together with the anthropogenically driven current global warming suggests that a new and unstable dynamic has entered such epidemiological cycles (Sutherst, 2004; Patz et al., 2005).

3.12.8 Synopsis Estimates indicate that about 1.1 billion people lack access to improved water supplies and that 2.6 billion lack adequate sanitation (United Nations Children’s Fund and World Health Organization, 2004). This coupled with the fact that between 4000 and 6000 children alone die in developing countries everyday due to diseases associated with contaminated water supply and poor hygenic conditions, presents a damning picture of water-resource policies in many parts of the world (Moe and Rheingans, 2006). As we have seen above, inadequately planned human manipulation of water resources aimed at improving access and agricultural availability can be directly involved in increasing the burden of disease (Montgomery and Elimelech, 2007). Waterborne human parasites represent a major component of this medical problem and its social and economic consequences. Of the 10 neglected tropical parasitic diseases listed by May (2007) as requiring urgent political action, five have been dealt with above indicating the importance of waterborne parasites generally. Some of the diseases caused by these parasites, such as malaria, have been strongly implicated in reducing the economic strength of entire countries and regions (McCarthy et al., 2000). Superimposed on the basic epidemiological characteristics of the parasites are the dynamic, human-based environmental changes taking place on a global basis. How are these changes likely to affect the epidemiology of the parasites discussed

354

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

above? Can we predict future disease patterns? What precautions should be put in place to prevent significant local or regional increases in the burden of these parasites on the human population? How should water-development projects be planned to avoid the problem of potentially increasing the burden of human parasite infection? All of these questions are currently the subject of intense research and discussion. The freshwater ecosystems of the world are currently undergoing major modifications in their ecology brought about largely by anthropogenic means (Vo¨ro¨smarty and Sahagian, 2000; Jackson et al., 2001; Dudgeon et al., 2005). These changes have a number of major causes; directly by influencing flow patterns, and indirectly through land use, population, and climatic shifts (Bunn and Arthington, 2002; Dudgeon et al., 2005). Direct changes in flow pattern modify the physical characteristics of the stream and thus the habitat present for the aquatic plant and animal species which are present. As plants and animals often have well-defined ecological requirements and therefore preferences for certain habitats, modified flow patterns are likely to influence the entire biotic community, including the parasites and their hosts present (Bunn and Arthington, 2002). Nevertheless, it is not always easy to predict, either qualitatively or quantitatively, how such alterations in flow regimes will influence either the biotic community in general or the parasite species involved (Marcogliese, 2001; Bunn and Arthington, 2002). In many developed countries, progress in water management has usually been accompanied by adequate sewage treatment strongly reducing the contamination of surface and groundwater and ensuring the supply of safe drinking water to most segments of the population (Tulchinsky et al., 2000). In developing countries, there is still much to be done in this regard (Cave and Kolsky, 1999). Water works and sewagetreatment plants financed by foreign aid often prove to be problematic for the the developing countries involved, which do not have the funding, infrastructure, or technical knowledge to maintain them (e.g., Obi et al., 2007). Scanning through the recommendations for each species or group of parasite species, a pattern can be found which is dependent on the life cycle of the parasite. For the few species discussed, which use only humans as final hosts, such as the Guinea worm Dracunculus medinensis, breaking of the transmission cycle to humans is capable of eliminating the disease. This is much harder for both directly and indirectly transmitted diseases for which nonhuman reservoir hosts are present (zoonotic diseases), as in such cases breaking the transmission cycle to humans leaves a natural cycle in which the pathogen remains present. If control measures are then lessened, this cycle can again expand to include the human population. It is also clear from studies conducted within a parasite species or higher taxonomic groups, that each individual situation must be analyzed separately. Plasmodium falciparum malaria provides a good example: in Southeast Asia; deforestation has led to a very substantial reduction in the number of malaria cases, while in parts of South America and Africa, the opposite has been documented. In the case of diseases which are transferred directly via water, such as giaridiasis and cryptosporidiosis, control can be attained by appropriate water treatment. Even simple filtration can be effective for some parasites, such as the removal of the

intermediate copepod host of Dracunculus medinensis. This is, however, an unsatisfactory situation, particularly in developing countries, as a variety of organisms other than relatively large parasites, such as bacteria, which cannot be readily eliminated by filtration, can also act as major human pathogens (Lee and Schwab, 2005). Thus, for completely effective water treatment, additional procedures, such as chlorination, are recommended. In very poor areas where this is not feasible, boiling drinking water offers a safe alternative. The situation becomes even more difficult for those pathogens which are transmitted via contact between the human body and water, for example, Schistosoma spp., or which require a vector with an aquatic component in its life cycle, for example, the mosquitoes which transmit malaria, or the black flies which transmit Onchocerca volvulus. Filtration or other water-treatment options are unrealistic as these vectors occur throughout freshwater bodies covering large areas. Attempts at large-scale control of malaria using insecticides sprayed into such water bodies are often only partially effective, lead to insecticide resistance in the mosquitoes, and often cause very substantial environmental damage as the insecticides are nonspecific (here exceptions are the biocontrol agents B.t.i. and B. sphaericus). Pharmaceutical development and reducing contact rates between the vector and the human host, and potentially the development of vaccines are important goals for control. An often-neglected topic in discussing waterborne parasite control, which is of particular importance, is the status of state-run healthcare systems. The major eradication and control projects carried out through the WHO are run by internationally recognized experts on the parasites involved, include large teams of trained fieldworkers and have access to substantial funding. In many developing countries and some developed countries, such as Russia, however, basic medical care at the national level is inadequate, particularly for the poorer segments of society, and treatment outside of international schemes, is limited, poor, or nonexistent (Stilwell et al., 2004; Streefland, 2005; Gwatkin et al., 2007). Thus, in order to reduce the burden of waterborne parasitic diseases, not only is adequate planning of the use of national water resources required, but also the effective treatment of humans is necessary. Interestingly, a few waterborne parasites, such as Onchocerca volvulus, have stimulated the community of nations to develop ambitious eradication campaigns (discussed earlier), while others such as schistosomes, although being equally hazardous, have attracted less coordinated attention either by large international organizations or by the countries affected. Thus, there is also an irrational aspect to the way in which we evaluate the parasites and diseases discussed in this chapter.

3.12.9 Conclusion Parasitic diseases with a water-based transmission cycle present major medical problems in developing countries and potential problems in the industrialized world. Any changes in the hydrodynamic pattern of a given area are likely to affect these transmission cycles, most often to the detriment of the human population. Indeed, major increases in the incidence

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

and prevalence of highly hazardous pathogens have been recorded on numerous occasions following anthropogenic alterations. It is therefore critical when designing hydrodynamic interventions, even on a small scale, anywhere in the world to assess the likelihood of increasing the medical and socioeconomic burden of parasites on the human population involved. Accordingly, all hydrological engineering projects should be accompanied by an evaluation of the likely parasitological consequences caused by changes in the characteristics of the water bodies involved.

Acknowledgments We would like to thank the Centers for Disease Control (CDC) in Atlanta, Georgia for permission to use the life cycle figures and many of the other images presented in this manuscript. Permission to use Figure 16 was obtained from the American Society for Microbiology. We also thank the estate of Prof. Werner Frank, Dr. Karl Steib, Dr. Yves Jackson, and Dr. V. Etzel for supplying images. Dr. Sven Kimpel (University of Du¨sseldorf) was kind enough to read and improve the section on anisakiasis. Alexandra Wenz and Jasmin Skuballa provided invaluable help with organizing and arranging the figures and formatting the manuscript. Pascal Baumgartner, Ba¨rbel Dausmann, Emanuel Heitlinger, Miriam Pfa¨ffle, Alexandra Plokarz, Ula Weclawski, and Claudia Zetlmeisl helped with the preparation of the manuscript.

References Abdel Khaleq RA and Alhaj Ahmed I (2007) Rainwater harvesting in ancient civilizations in Jordan. Water Science and Technology: Water Supply 7: 85--93. Abdussalam M, Kaferstein FK, and Mott KE (1995) Food safety measures for the control of foodborne trematode infections. Food Control 6: 71--79. Abollo E, Gestal C, and Pascual S (2001) Anisakis infestation in marine fish and cephalopods from Galician waters: An updated perspective. Parasitology Research 87: 492--499. Abreu Ferrari de TC (2004) Involvement of central nervous system in the schistosomiasis. Memorias do Instituto Oswaldo Cruz 99(supplement 1): 59--62. Abubakar I, Aliyu SH, Arumugam C, Hunter PR, and Usman NK (2007a) Prevention and treatment of cryptosporidiosis in immunocompromised patients. Cochrane Database of Systematic Reviews 24: CD004932. [online] Abubakar I, Aliyu SH, Arumugam C, Usman NK, and Hunter PR (2007b) Treatment of cryptosporidiosis in immunocompromised individuals: Systematic review and meta-analysis. British Journal of Clinical Pharmacology 63: 387--393. Ackers JP and Mirelman D (2006) Progress in research on Entamoeba histolytica pathogenesis. Current Opinion in Microbiology 9: 367--373. Acuna-Soto R, Maguire JH, and Wirth DF (2000) Gender distribution in asymptomatic and invasive amebiasis. American Journal of Gastroenterology 95: 1277--1283. Adam RD (1991) The biology of Giardia spp. Microbiology and Molecular Biology Reviews 55: 706--732. Adam RD (2001) Biology of Giardia lamblia. Clinical Microbiology Reviews 14: 447--475. Adekolu-John EO (1983) The impact of lake creation on guinea-worm transmission in Nigeria on the eastern side of Kainji Lake. International Journal for Parasitology 13: 427--432. Adewale B, Mafe MA, and Oyerinde JP (1999) Infectivity and transmission dynamics of Simulium damnosum s.l. around Owena dam (Ondo State). West African Journal of Medicine 18: 257--260. Adjami AG, Toe L, Bissan Y, et al. (2004) The current status of onchocerciasis in the forest/savanna transition zone of Cote d’Ivoire. Parasitology 128: 407--414. Agatsuma T, Iwagami M, Liu CX, et al. (2001) Molecular phylogenetic position of Schistosoma sinensium in the genus Schistosoma. Journal of Helminthology 75: 215--221.

355

Agatsuma T, Iwagami M, Liu CX, et al. (2002) Affinities between Asian non-human Schistosoma species, the S. indicum group, and the African human schistosomes. Journal of Helminthology 76: 7--19. Ak M, Keles¸ E, Karacasu F, et al. (2006) The distribution of the intestinal parasitic diseases in the Southeast Anatolian (GAP ¼ SEAP) region of Turkey. Parasitology Research 99: 146--152. Akbar A and Ghosh S (2005) Anisakiasis – a neglected diagnosis in the West. Digestive and Liver Disease 37: 7--9. Aksoy S, Ariturk S, Armstrong MYK, et al. (1995) The GAP Project in southeastern Turkey: The potential for emergence of diseases. Emerging Infectious Diseases 1: 62--63. Albrecht C, Kroll O, Terrazas EM, and Wilke T (2008) Invasion of ancient Lake Titicaca by the globally invasive Physa acuta (Gastropoda: Pulmonata: Hygrophila). Biological Invasions 11: 1821--1826. Alemayehu T, Ye-ebiyo Y, Ghebreyesus TA, Witten KH, Bosman A, and Teklehaimanot A (1998) Malaria, schistosomiasis, and intestinal helminths in relation to microdams in Tigray, northern Ethiopia. Parassitologia 40: 259--267. Alfonso A, Hunt P, Cheesman A, et al. (2006) Malaria parasites can develop stable resistance to artemisinin but lack mutations in candidate genes Atp6 (encoding the sarcoplasmic and endoplasmic reticulum Ca2þ ATPase), tctp, mdr1, and cg10. Antimicrobial Agents and Chemotherapy 50: 480--489. Allan JD, Abell R, Hogan Z, et al. (2005) Overfishing of inland waters. BioScience 55: 1041--1051. Allason-Jones E, Mindel A, Sargeaunt P, and Williams P (1986) Entamoeba histolytica as a commensal intestinal parasite in homosexual men. New England Journal of Medicine 315: 353--356. Al Mekhlafi MSH, Azlin M, Aini UN, et al. (2005) Giardiasis as a predictor of childhood malnutrition in Orang Asli children in Malaysia. Transactions of the Royal Society of Tropical Medicine and Hygiene 99: 686--691. Amahmid O, Asmama S, and Bouhoum K (1999) The effect of waste water reuse in irrigation on the contamination level of food crops by Giardia cysts and Ascaris eggs. International Journal of Food Microbiology 49: 19--26. Andrews RH, Adams M, Boreham PF, Mayrhofer G, and Meloni BP (1989) Giardia intestinalis: Electrophoretic evidence for a species complex. International Journal for Parasitology 19: 183--190. Andrews RH, Sithithaworn P, and Petney TN (2008) Opisthorchis viverrini: An underestimated parasite in world health. Trends in Parasitology 24: 497--501. Anh NTL, Phuong NT, Murrell KD, et al. (2009) Animal reservoir hosts and fish-borne zoonotic trematode infections on fish farms, Vietnam. Emerging Infectious Diseases 15: 540--546. Anyamba A, Chretien JP, Small J, Tucker CJ, and Linthicum KJ (2006) Developing global climate anomalies suggest potential disease risks for 2006–2007. International Journal of Health Geographics 5: 60--67. Arene FOI, Ibanga E, and Asor JE (1998) Epidemiology of paragonimiasis in Cross River Basin, Nigeria: Prevalence and intensity of infection due to Paragonimus uterobilateralis in Yakurr local government area. Public Health 112: 119--122. Arizono N, Yamada M, Nakamura-Uchiyama F, and Ohnishi K (2009) Diphylobothriasis associated with eating raw Pacific salmon. Emerging Infectious Diseases 15: 866--870. Armignacco O, Caterini L, Marucci G, et al. (2008) Human illness caused by Opisthorchis felineus flukes, Italy. Emerging Infectious Diseases 14: 1902--1905. Armon R, Gold D, Brodsky M, and Oron G (2002) Surface and subsurface irrigation with effluents of different qualities and presence of Cryptosporidium oocysts in soil and on crop. Water Science and Technology 46: 115--122. Ashbolt NJ (2004) Microbial contamination of drinking water and disease outcomes in developing countries. Toxicology 198: 229--238. Ashrafi K, Valero MA, Massoud J, et al. (2006) Plant-borne human contamination by fascioliasis. American Journal for Tropical Medicine and Hygiene 75: 295--302. Asrat M, Petros B, Jobre Y, et al. (2005) Infection prevalence of ovive fascioliasis in small-scale irrigation schemes along the Upper Awash River Basin. Ethiopian Veterinary Journal 9: 19--27. Attwood HD and Chou ST (1978) The longevity of Clonorchis sinensis. Pathology 10: 153--156. Audicana MT, Ansotegui IJ, de Corres LF, and Kennedy MW (2002) Anisakis simplex: Dangerous – dead and alive. Trends in Parasitology 18: 20--25. Audicana MT and Kennedy MW (2008) Anisakis simplex: From obscure infectious worm to inducer of immune hypersensitivity. Clinical Microbiology Reviews 21: 360--379. Baer JG, Miranda H, Fernandez W, and Medina J (1967) Human diphyllobothriasis in Peru. Zeitschrift fu¨r Parasitenkunde 28: 277--289. Bahia-Oliveira LMG, Jones JL, Azevedo-Silva J, Alves CCF, Ore´fice F, and Addiss DG (2003) Highly endemic, water-borne toxoplasmosis in north Rio de Janeiro State, Brazil. Emerging Infectious Diseases 9: 55--62.

356

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Bandeira Ferraz AA, Souto Bacelar T, De Castro Silveira MJ, et al. (2001) Surgical treatment of schistosomal portal hypertension. International Surgery 86: 1--8. Barry M (2007) The tail end of Guinea worm – global eradication without a drug or a vaccine. New England Journal of Medicine 356: 2561--2564. Barsoum R (2004) The changing face of schistosomal glomerulopathy. Kidney International 66: 2472--2484. Barwick RS, Uzicanin A, Lareau S, et al. (2002) Outbreak of amebiasis in Tbilisi, Republic of Georgia, 1998. American Journal of Tropical Medicine and Hygiene 67: 623--631. Basa´n˜ez MG, Pion SDS, Churcher TS, Breitling LP, Little MP, and Boussinesq M (2006) River blindness: A success story under threat? PloS Medicine 3: 1454--1460. Bayoh MN and Lindsay SW (2004) Temperature-related duration of auqatic stages of the Afrotropical malaria vector mosquito Anopheles gambiae in the laboratory. Medical and Veterinary Entomology 18: 174--179. Belcher DW, Wurapa FK, Ward WB, and Lourie IM (1975) Guinea worm in southern Ghana: Its epidemiology and impact on agricultural productivity. American Journal of Tropical Medicine and Hygiene 24: 243--249. Belizario VY, Bersabe MJJ, De Los Reyes ABE, and De Leon WU (2004) School-based assessment of soil-transmitted helminthiases and food-borne parasitosis (intestinal fluke infection) in Monkayo, Compostela Valley. Southeast Asian Journal of Tropical Medicine and Public Health 35(supplement 1): 123--139. Belizario VY, Geronilla GG, Anastacio MBM, et al. (2007) Echinostoma malayanum infection, the Philippines. Emerging Infectious Diseases 13: 1130--1131. Bence JR (1988) Indirect effects and biological control of mosquitoes by mosquitofish. Journal of Applied Ecology 25: 505--521. Betancourt WQ and Rose JB (2004) Drinking water treatment processes for removal of Cryptosporidium and Giardia. Veterinary Parasitology 16: 219--234. Bhattacharya D, Haque R, and Singh U (2005) Coding and noncoding genomic regions of Entamoeba histolytica have significantly different rates of sequence polymorphisms: Implications for epidemiological studies. Journal of Clinical Microbiology 43: 4815--4819. Blair B, Sarkar P, Bright KR, Maciano-Cabral F, and Gerba CP (2008) Naegleria fowleri in well water. Emerging Infectious Diseases 14: 1499--1500. Blair D, Xu ZB, and Agatsuma T (1999) Paragonimiasis and the genus Paragonimus. Advances in Parasitology 42: 113--222. Blanshard C, Jackson AM, Shanson DC, Francis N, and Gazzard BG (1992) Cryptosporidiosis in HIV seropositive patients. Quarterly Journal of Medicine 85: 813--823. Blessmann J, Ali IKM, Nu PAT, et al. (2003) Longitudinal study of intestinal Entamoeba histolytica infections in asymptomatic adult carriers. Journal of Clinical Microbiology 41: 4745--4750. Blessmann J, Van Le A, and Tannich E (2005) Epidemiology and treatment of amebiasis in Hue´, Vietnam. Archives of Medical Research 37: 27--272. Blessmann J, Van Linh P, Nu PA, et al. (2002) Epidemiology of amebiasis in a region of high incidence of amebic liver abscess in central Vietnam. American Journal of Tropical Medicine and Hygiene 66: 578--583. Bockarie MJ, Pedersen EL, White GB, and Michael E (2009) Role of vector control in the global program to eliminate lymphatic filariasis. Annual Review of Entomology 54: 469--487. Boge´a T, Cordeiro FM, and De Gouveia JS (2005) Melanoides tuberculatus (Gastropoda: Thiaridae) as an intermediate host of heterophyidae (Trematoda: Digenea) in Rio De Janeiro metropolitan area, Brazil. Revista do Instituto de Medicina Tropical de Sa˜o Paulo 47: 87--90. Bogitsh BJ, Carter CE, and Oeltmann TN (2005) Human Parasitology, 3rd edn. London: Elsevier. Bohl BK and Jakowlew JJ (1931) Die pathogenische Anatomie der Katzenleber bei Opisthorchis felineus. Virchows Archiv 164: 724--730. Blessmann J, Ali IKM, Ton Nu PA, et al. (2003) Longitudinal study of intestinal Entamoeba histolytica infections in asymptomatic adult carriers. Journal of Clinical Microbiology 41: 4745--4750. Bohonak AJ and Jenkins DG (2003) Ecological and evolutionary significance of dispersal by freshwater invertebrates. Ecology Letters 6: 783--796. Bottieau E, Clerinx J, De Vega MR, et al. (2006) Imported Katayama fever: Clinical and biological features at presentation and during treatment. Journal of Infection 52: 339--345. Botto C, Escalona E, Vivas-Martinez S, Behm V, Delgado L, and Coronel P (2005) Geographical patterns of onchocerciasis in southern Venezuela: Relationships between environment and infection prevalence. Parassitlogia 47: 145--150. Bouchet F, Harter S, and Le Bailly (2003) The state of the art of paleoparasitological research in the Old World. Memo´rias do Instituto Oswaldo Cruz 98(supplement 1): 95--101.

Bouma MJ and Dye C (1997) Cycles of malaria associated with El Nino in Venezuela. Journal of the American Medical Association 278: 1772--1774. Bouma MJ and Van der Kaay HJ (1996) The El Nino Southern Oscillation and the historic malaria epidemics on the Indian subcontinent and Sri Lanka: An early warning system for future epidemics? Tropical Medicine and International Health 1: 86--96. Bowie WR, King AS, Werker DH, et al. (1997) Outbreak of toxoplasmosis associated with municipal drinking water. Lancet 350: 173--177. Boyer JF, Bader JC, Lapetite JM, and Servat E (1990) Satellite data transmission and hydrological forecasting in the fight against onchocerciasis in West Africa. Journal of Hydrology 117: 187--198. Bundy DA, Chandiwana SK, Homeida MM, Yoon S, and Mott KE (1991) The epidemiological implications of a multiple-infection approach to the control of human helminth infections. Transactions of the Royal Society for Tropical Medicine and Hygiene 85: 274--276. Bunn SE and Arthington AH (2002) Basic principles and ecological consequences of altered flow regimes for aquatic biodiversity. Environmental Management 30: 492--507. Bush AO, Ferna´ndez JC, Esch GW, and Seed JR (2001) Parasitism: The Diversity and Ecology of Animal Parasites. Cambridge: Cambridge University Press. Bush SR (2004) A Political Ecology of the Living Aquatic Resources in Lao PDR. PhD Thesis, University of Sydney. Butler D (2009) Malaria drug-makers ignore WHO ban. Nature 460: 310--311. Cabanes PA, Wallet F, Pringuez E, and Pernin P (2001) Assessing the risk of primary amoebic miningoencephalitis from swimming in the presence of environmental Naegleria fowleri. Applied and Environmental Microbiology 67: 2927--2931. Caballero ML and Moneo I (2004) Several allergens from Anisakis simplex are highly resistant to heat and pepsin treatments. Parasitology Research 93: 248--251. Cabello FC (2007) Aquaculture and public health. The emergence of diphyllobothriosis in Chile and the world. Revista Medica de Chile 135: 1064--1071. Caccio` SM, Thompson RCA, McLauchlin J, and Smith HW (2005) Unravelling Cryptosporidium and Giardia epidemiology. Trends in Parasitology 21: 430--437. Calhoun LM, Avery M, Jones LA, et al. (2007) Combined sewage overflows (CSO) are major urban breeding sites of Culex quinquefasciatus in Atlanta, Georgia. American Journal of Tropical Medicine and Hygiene 77: 478--484. Cama VA, Ross JM, Crawford S, et al. (2007) Differences in clinical manifestations among Cryptosporidium species and subtypes in HIV-infected persons. Journal of Infectious Diseases 196: 684--691. Cameron TWM (1928) A new defintive host for Schistosoma mansoni. Journal of Helminthology 6: 219--222. Campos-Rodrı´guez R and Jarillo-Luna A (2005) The pathogenicity of Entamoeba histolytica is related to the capacity of evading innate immunity. Parasite Immunology 27: 1--8. Cane MA (2005) The evolution of El Nino, past and future. Earth and Planetary Science Letters 230: 227--240. Canete R, Young M, Sanchez J, Wong L, and Gutierrez A (2004) Population dynamics of intermediate snail hosts of Fasciola hepatica and some environmental factors in San Juan y Martinez municipality, Cuba. Memo´rias do Instituto Oswaldo Cruz 99: 257--262. Carr RM, Blumenthal UJ, and Mara DD (2004) Guidelines for the safe use of wastewater in agriculture: Revisiting WHO guidelines. Water Science and Technology 50: 31--38. Carter R and Mendis KN (2002) Evolutionary and historical aspects of the burden of malaria. Clinical Microbiology Reviews 15: 564--594. Caspary HU (1999) Wildlife Utilization in Cote d’Ivoire and West Africa – Potentials and Constraints for Development Cooperation. Eschborn: Deutsche Gesellschaft fu¨r Technische Zusammenarbeit. Casta´n B, Borda F, In˜arrairaegui M, Pastor G, Vila J, and Zozaya JM (2002) Digestive anisakiasis: Clinical manifestations and diagnosis according to localization. Revista Espan˜ola de Enfermedades Digestivas 94: 463--472. Causer LM, Handzel T, Welch P, et al. (2006) An outbreak of Cryptosporidium hominis infection at an Illinois recreational waterpark. Epidemiology and Infection 134: 147--156. Cave B and Kolsky P (1999) Ground Water, Latrines and Health. Loughborough: Loughborough University, Resource Centre for Water, Sanitation and Environmental Health. Ca´ceres CE and Soluk DA (2002) Blowing in the wind: A field test of overland dispersal and colonization by aquatic invertebrates. Oecologia 131: 402--408. Chai JY and Lee SH (2002) Food-borne intestinal trematode infections in the Republic of Korea. Parasitology International 51: 129--154. Chai JY, Murrell KD, and Lymbery AJ (2005) Fish-borne parasitic zoonoses: Status and issues. International Journal for Parasitology 35: 1233--1254.

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Chai J-Y, Seo B-S, Lee S-H, Hong S-J, and Sohn W-M (1986) Human infections by Heterophyes heterophyes and H. dispar imported from Saudi Arabia. Korean Journal of Parasitology 24: 82--88. Chai JY, Shin EH, Lee SH, and Rim HJ (2009) Foodborne intestinal flukes in Southeast Asia. Korean Journal of Parasitology 47(supplement): S69--S102. Chalmers RM, Sturdee AP, Bull SA, Miller A, and Wright SE (1997) The prevalence of Cryptosporidium parvum and C. muris in Mus domesticus, Apodemus sylvaticus and Clethrionomys glareolus in an agricultural system. Parasitology Research 83: 478--482. Chansiri K, Tejangkura T, Kwaosak P, Sarataphan N, Phantana S, and Sukhumsirichart W (2002) PCR based method for identification of zoonostic Brugia malayi microfilariae in domestic cats. Molecular and Cellular Probes 16: 129-135. Chappell CL, Okhuysen PC, Langer-Curry R, et al. (2006) Cryptosporidium hominis: Experimental challenge of healthy adults. American Journal of Tropical Medicine and Hygiene 75: 851--857. Chareonviriyaphap T, Bangs MJ, and Ratanatham S (2000) Status of malaria in Thailand. Southeast Asian Journal of Tropical Medicine and Public Health 31: 225--237. Chaves LF, Keogh CL, Vazquez-Prokopec GM, and Kitron UD (2009) Combined sewage overflow enhances oviposition of Culex quinquefasciatus (Diptera: Culicidae) in urban areas. Journal of Medical Entomology 46: 220--226. Chen MG and Mott KE (1990) Progress in assessment of morbidity due to Fasciola hepatica infection: A review of recent literature. Tropical Diseases Bulletin 87: 1--38. Chen Y, Zhang YZ, Yang B, et al. (2007) Seroprevalence of Entamoeba histolytica infection in HIV-infected patients in China. American Journal of Tropical Medicine and Hygiene 77: 825--828. Chi TTK, Dalsgaard A, Turnbull JF, Tuan PA, and Murrell KD (2007) Prevalences of zoonotic trematodes in fish from a Vietnamese fish-farming community. Journal of Parasitology 94: 423--428. Chitsulo L, Engels D, Montresor A, and Savioli L (2000) The global status of schistosomiasis and its control. Acta Tropica 77: 41--51. Choi BI, Han JK, Hong ST, and Lee KH (2004) Clonorchiasis and cholangiocarcinoma: Etiologic relationship and imaging diagnosis. Clinical Microbiology Reviews 17: 540--552. Christensen JH, Hewitson B, Busuioc A, et al. (2007) Regional Climate Projections. Cambridge: Cambridge University Press. Chwatt B and De Zulueta J (1980) The Rise and Fall of Malaria in Europe: A HistoricoEpidemiological Study. Oxford: Oxford University Press. Cifuentes E, Gomez M, Blumenthal U, et al. (2000) Risk factors for Giardia intestinalis infection in agricultural villages practicing wastewater irrigation in Mexico. American Journal of Tropical Medicine and Hygiene 62: 388--392. Clancy JL, Hargy TM, Marshall MM, and Dyksen JE (1998) UV-light inactivation of Crytosporidium oocysts. Journal of the American Water Works Association 90: 92--102. Clark DP (1999) New insights into human cryptosporidiosis. Clinical Microbiology Reviews 12: 554--563. Cohen GM and Shurin JB (2003) Scale-dependence and mechanisms of dispersal in freshwater zooplankton. Oikos 103: 603--617. Corso PS, Kramer MH, Blair KA, Addiss DG, Davis JP, and Haddix AC (2003) Cost of illness in the 1993 waterborne Cryptosporidium outbreak, Milwaukee, Wisconsin. Emerging Infectious Diseases 9: 426--431. Costa-Pierce BA (2003) Rapid evolution of an established tilapia (Oreochromis spp.): The need to incorporate invasion science into regulatory structures. Biological Invasions 5: 71--84. Cox FEG (2002) History of human parasitology. Clinical Microbiology Reviews 15: 595--612. Cox-Singh J, Davis TME, Lee KS, et al. (2008) Plasmodium knowlesi malaria in humans is widely distributed and potentially life threatening. Clinical Infectious Diseases 46: 165--171. Crompton DW (1999) How much human helminthiasis is there in the world? Journal of Parasitology 85: 397--403. Crosskey RW and Howard TM (2004) A Revised Taxonomiy and Geographical Inventory of World Blackflies (Diptera: Simuliidae). London: Department of Entomology, The Natural History Museum. Cui J, Wang ZQ, Wu F, and Jin XX (1998) An outbreak of paragonimiosis in Zhenzhou city, China. Acta Tropica 70: 211--216. Curriero FC, Patz JA, Rose JB, and Lele S (2001) The association between extreme precipitation and waterborne disease outbreaks in the United States, 1948–1994. American Journal of Public Health 91: 1194--1199. Curtis MA and Bylund G (1991) Diphyllobothriasis: Fish tapeworms in the circumpolar north. Arctic Medical Research 50: 18--24.

357

Dadzie Y, Hopkins DR, and Neira M (2002) Final Report of the Conference on the Eradicability of Onchocerciasis. Atlanta, GA: The Carter Center. Daft BM, Visvesvara GS, Read DH, Kinde H, Uzal FA, and Manzer MD (2005) Seasonal meningoencephalitis in Holstein cattle caused by Naegleria fowleri. Journal of veterinary Diagnostic Investigation 17: 605--609. Dagan R, Leventhal A, Anis E, Slater P, Ashur Y, and Shouval D (2005) Incidence of hepatitis A in Israel following universal immunization of toddlers. Journal of the American Medical Association 294: 202--210. Dalimi A and Jabarvand M (2005) Fasciola hepatica in the human eye. Transactions of the Royal Society of Tropical Medicine and Hygiene 99: 798--800. Daszak P, Cunninghamm AA, and Hyattm AD (2000) Emerging infectious diseases of wildlife-threats to biodiversity and human health. Science 287: 443--449. Davies JB (1994) Sixty years of onchocerciasis vector control: A chronological summary with comments on eradication, reinvasion, and insecticide resistance. Annual Review of Entomology 1994: 23--45. De Kock KN and Wolmarans CT (2009) Distribution and habitats of Melanoides tuberculata (Mu¨ller, 1774) and M. victoriae (Dohrn, 1865) (Mollusca: Prosobranchia: Thiaridae) in South Africa. Water South Africa 35: 713--720. DeFrain M and Hooker R (2002) North American paragonimiasis – case report of a severe clinical infection. Chest 121: 1368--1372. Del Pozo MD, Audicana M, Diez JM, et al. (1997) Anisakis simplex, a relevant etiologic factor in acute urticaria. Allergy 52: 576--579. Devi KR, Narain K, Bhattacharya S, et al. (2007) Pleuropulminary paragonimiasis due to Paragonimus heterotremus: Molecular diagnosis, prevalence of infection and clinicoradiological features in an endemic area of northeastern India. Transactions of the Royal Society of Tropical Medicine and Hygiene 101: 786--792. Diawara L, Traore MO, Badji A, et al. (2009) Feasibility of onchocerciasis elimination with ivermectin treatment in endemic foci in Africa: First evidence from studies in Mali and Senegal. PLoS Neglected Tropical Diseases 3: e497. Diaz MT, Hernandez LE, and Bashirullah AK (2008) Studies on the life cycle of Haplorchis pumilio (Looss, 1896) (Trematoda: Heterophyidae) in Venezuela. Revista Cientı´fica (Maracaibo) 18: 35--42. Dick TA, Nelson PA, and Choudhury A (2001) Diphyllobothriasis: Update on human cases, foci, patterns and sources of human infections and future considerations. Southeast Asian Journal of Tropical Medicine and Public Health 32(supplement 2): 59--76. Didier ES (2005) Microsporidiosis: An emerging and opportunistic infection in humans and animals. Acta tropica 94: 61--76. Didier ES, Stovall ME, Green LC, Brindley PJ, Sestak K, and Didier PJ (2004) Epidemiology of microsporidiosis: Sources and modes of transmission. Veterinary Parasitology 126: 145--166. Doanh PN, Shinohara A, Horii Y, Habe S, Nawa Y, and Le NT (2007) Description of a new lung fluke species, Paragonimus vietnamensis sp. nov. (Trematoda, Paragonimidae), found in nothern Vietnam. Parasitology Research 101: 1495--1501. Dos Santos AF, De Azevedo DPL, Dos Santoa Mata RdaC, Dinis de Mendonco DIM, and Sant’Ana AEG (2007) The lethality of Euphorbia conspicua to adults of Biomphalaria glabrata, cercaria of Schistosoma mansoni and larvae of Artemia salina. Bioresourse Technology 98: 135--139. Dossou-yovo J, Doannio J, Revie`re F, and Chauvancy G (1995) Urbanization and establishment of Culex quinquefasciatus in a West African rural area. Acta Tropica 59: 251--253. Drakeley CJ, Carneiro H, Reyburn H, et al. (2005) Altitude-dependent and independent variations in Plasmodium falciparum prevalence in northeast Tanzania. Journal of Infectious Diseases 191: 1589--1598. Dreyer G, Noro˜es J, Figueredo-Silva J, and Piessens WF (2000) Pathogenesis of lymphatic disease in bancroftian filariasis: A clinical perspective. Parasitology Today 16: 544--548. Dreyfuss G, Vignoles P, Abrous M, and Rondelaud D (2002) Unusual snail species involved in the transmission of Fasciola hepatica in watercress beds in central France. Parasite 9: 113--120. Dubey JP (2004) Toxoplasmosis – a waterborne zoonosis. Veterinary Parasitology 126: 57--72. Dudgeon D, Arthington AH, Gessner MO, et al. (2005) Freshwater biodiversity: Importance, threats, status and conservation challenges. Biological Reviews 81: 163--182. Duenngai K, Sithithaworn P, Rudrappa UK, et al. (2008) Improvement of PCR for detection of Opisthorchis viverrini DNA in human stool samples. Journal of Clinical Microbiology 46: 366--368. Dung DT, De NV, Waikagul J, et al. (2007) Fishborne zoonotic intestinal trematodes, Vietnam. Emerging Infectious Diseases 13: 1828--1833.

358

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Duplantier JM and Se`ne M (2000) Rodents as reservoir hosts in the transmission of Schistosoma mansoni in Richard-Toll, Senegal, West Africa. Journal of Helminthology 74: 129--135. Durr PA, Tait N, and Lawson AB (2005) Bayesian hierarchical modelling to enhance the epidemiological values of abattoir surveys for bovine fascioliasis. Preventive Veterinary Medicine 71: 157--172. Dzodzomenyo M, Dunyo S, Ahorlu CK, et al. (1999) Bancroftian filariasis in an irrigation project community in southern Ghana. Tropical Medicine and International Health 4: 13--18. Ellyson JH, Bezmalinovic Z, Parks SN, and Lewis FR, Jr. (1986) Necrotizing amebic colitis: A frequently fatal complication. American Journal of Surgery 152: 21--26. El Mawla NG, El Bolkainy MN, and Khaled HM (2001) Bladder cancer in Africa: Update. Seminars in Oncology 28: 174--178. El-Rifai W, Kamel D, Larramendy ML, et al. (2000) DNA copy number changes in Schistosoma-associated and non-Schistosoma-associated bladder cancer. American Journal of Pathology 156: 871--878. El-Sayed KA (2001) Effect of ammonium chloride on Biomphalaria alexandrina and on its infection with Schistosoma mansoni and Echinostoma liei miracidia. Journal of the Egyptian Society of Parasitology 31: 593--602. Elsheikha HM (2007) Heterophyosis: Risk of ectopic infection. Veterinary Parasitology 147: 341--342. Elsheikha HM and Elshazly AM (2008) Preliminary observations on infection of brackish and fresh water fish by heterophyid encysted metacercariae in Egypt. Parasitology Research 103: 971--977. Engels D, Chitsulo L, Montresor A, and Savioli L (2002) The global epidemiological situation of schistosomiasis and new approaches to control and research. Acta Tropica 82: 139--146. Ensink JH, van der Hoek W, and Amerasinghe FP (2006) Giardia duodenalis infection and wastewater irrigation in Pakistan. Transactions of the Royal Society of Tropical Medicine and Hygiene 100: 538--542. Erlanger TE, Keiser J, De Castro MC, et al. (2005) Effect of water resource development and management on lymphatic filariasis, and estimates of populations at risk. American Journal of Tropical Medicine and Hygiene 73: 523--533. Estaban JG, Aguirre C, Flores A, Strauss W, Angles R, and Mas-Coma S (1998) High Cryptosporidium prevalences in healthy Aymara children from the northern Bolivian Altiplano. American Journal of Tropical Medicine and Hygiene 58: 50--55. Estaban JG, Flores A, Angles R, and Mas-Coma S (1999) High endemicity of human fascioliasis between Lake Titicaca and La Paz valley, Bolivia. Transactions of the Royal Society of Tropical Medicine and Hygiene 93: 151--156. Estaban JG, Gonzalez C, Bargues MD, et al. (2002) High fascioliasis infection in children linked to a man-made irrigation zone in Peru. Tropical Medicine and International Health 7: 339--348. Estaban JG, Gonzalez C, Curtale F, et al. (2003) Hyperendemic fascioliasis associated with schistosomiasis in villages in the Nile Delta of Egypt. American Journal of Tropical Medicine and Hygiene 69: 429--437. Evans T (1989) The impact of permanent disability on rural households: River blindness in Guinea. Institute of Developmental Studies Bulletin 20: 41--48. Fayer R (2004) Cryptosporidium: A water-borne zoonotic parasite. Veterinary Parasitology 126: 37--56. Fayer R, Trout JM, and Jenkins MC (1998) Infectivity of Cryptosporidium parvum oocysts stored in water at environmental temperatures. Journal of Parasitology 84: 1165--1169. Feenstra S, Hussain R, and Van der Hoek W (2000) Health Risks of Irrigation with Untreated Urban Wastewater in the Southern Punjab, Pakistan. Lahore: International Water Managment Institute. Ferre I, Ortega-Mora LM, and Rojo-Va´zquez FA (1995) Seroprevalence of Fasciola hepatica infection in sheep in northwestern Spain. Parasitology Research 81: 137--142. Ferreira MS and Borges AS (2002) Some aspects of protozoan infections in immunocompromised patients – a review. Memo´rias do Instituto Oswaldo Cruz 97: 443--457. Fillinger U, Sonye G, Killeen GF, Knols BGJ, and Becker N (2004) The practical importance of permanent and semipermanent habitats for controlling aquatic stages of Anopheles gambiae sensu lato mosquitoes: Operational observations from a rural town in Western Kenya. Tropical Medicine and International Health 9: 1274--1289. Fischer P, Wibowo H, Oischke S, et al. (2002) PCR-based detection and identification of the filarial parasite Brugia timori from Alor Island, Indonesia. Annals of Tropical Medicine and Parasitology 96: 809--821. Food and Agriculture Organization (2009) The State of World Fisheries and Aquaculture 2008. Rome: Food and Agriculture Organization. Fotedar R, Stark D, Beebe N, Marriott D, Ellis J, and Harkness J (2007) PCR detection of Entamoeba histolytica, Entamoeba dispar, and Entamoeba moshkovskii in stool samples from Sydney, Australia. Journal of Clinical Microbiology 45: 1035--1037.

Fotedar R, Stark D, Marriott D, Ellis J, and Harkness J (2008) Entamoeba moshkovskii infections in Sydney, Australia. European Journal of Clinical Microbiology and Infectious Diseases 27: 133--137. Fournier S, Liguory O, Santillana-Hayat M, et al. (2000) Detection of microsporidia in surface water: A one year follow-up study. Immunology and Medical Microbiology 29: 95--100. Fried B, Graczyk TK, and Tamang L (2004) Food-borne intestinal trematodiases in humans. Parasitology Research 93: 159--170. Gajadhar AA and Allen JR (2004) Factors contributing the public health and economic importance of waterborne zoonotic parasites. Veterinary Parasitology 126: 3--14. Gajadhar AA, Scandrett WB, and Forbes LB (2006) Overview of food- and waterborne zoonotic parasites at the farm level. Revue Scientifique et Technique Office International des Epizooties 25: 595--606. Gale P (2005) Land application of treated sewage sludge: Quantifying pathogen risks from consumption of crops. Journal of Applied Microbiology 98: 380--396. Garms R, Lakwo TL, Noyomugyenyi R, et al. (2009) The elimination of the vector Simulium neavei from the Itwara onchocerciasis focus in Uganda by ground larviciding. Acta Tropica 111: 203--210. Garrett ES, Dos Santos CL, and Jahncke ML (1997) Public, animal, and environmental implications of aquaculture. Emerging Infectious Diseases 3: 453--457. Gebre-Michael T, Malone JB, and McNally K (2005) Use of geographic information systems in the development of prediction models for onchocerciasis control in Ethiopia. Parassitologia 47: 135--144. Geurden T, Geldhof P, Levecke B, et al. (2008) Mixed Giardia duodenalis assemblage A and E infections in calves. International Journal for Parasitology 38: 259--264. Ghebreyesus TA, Halie M, Witten KH, et al. (1999) Incidence of malaria among children living near dams in northern Ethiopia: Community based incidence survey. British Medical Journal 319: 663--666. Ghebreyesus TA, Witten KH, Getachew A, et al. (2002) Schistosome transmission, water-resource development and altitude in northern Ethiopia. Annals of Tropical Medicine and Parasitology 96: 489--495. Giboda M, Ditrich O, Scholz T, Viengsay T, and Bouaphanh S (1991) Human Opisthorchis and Haplorchis infections in Laos. Transactions of the Royal Society of Tropical Medicine and Hygiene 85: 538--540. Giovanni GD, Hashemi FH, Shaw NJ, Abrams FA, LeChevallier MW, and Abbaszadegan M (1999) Detection of infectious Cryptosporidium parvum oocysts in surface and filter backwash water samples by immunomagnetic separation and integrated cell culture-PCR. Applied and Environmental Microbiology 65: 3427--3432. Girard MP, Reed ZH, Friede M, and Kieny MP (2007) A review of human vaccine reseach and development: Malaria. Vaccine 25: 1567--1580. Glaberman S, Moore JE, Lowery CL, et al. (2002) Three drinking-water-associated cryptosporidiosis outbreaks, Nothern Ireland. Emerging Infectious Diseases 8: 631--633. Goncalves MLC, Araujo A, and Ferreira LF (2003) Human intestinal parasites in the past: New findings and a review. Memo´rias do Instituto Oswaldo Cruz 98(supplement 1): 103--118. Gores GJ (2003) Cholangiocarcinoma: Current concepts and insights. Hepatology 37: 961--969. Graczyk TA (2007) Fish, amphibians and reptiles. In: Fayer R and Xiao LH (eds.) Cryptosporidium and Cryptosporidiosis, 2nd edn., pp. 387--394. London: CRC Press. Graczyk TK and Fried B (1998) Echinostomiasis: A common but forgotten food-borne disease. American Journal of Tropical Medicine and Hygiene 58: 501--504. Greenwood BM, Bojang K, Whitty JM, and Targett GAT (2005) Malaria. Lancet 365: 1487--1498. Gryseels B, Polman K, Clerinx J, and Kestens L (2006) Human schistosomiasis. Lancet 368: 1106--1118. Guerra CA, Snow RW, and Hay SI (2006) A global assessment of closed forests, deforestation and malaria risk. Annals of Tropical Medicine and Parasitology 100: 189--204. Guerrant RL (1997) Cryptosporidiosis: An emerging, highly infectious threat. Emerging Infectious Diseases 3: 51--57. Gwatkin DR, Rutstein S, Johnson K, Suliman E, Wagstaff A, and Amouzou A (2007) Socio-Economic Differences in Health, Nutrition, and Population within Developing Countries: An Overview. New York: The World Bank. Haghighi A, Kobayashi S, Takeuchi T, Thammapalerd N, and Nozaki T (2003) Geographic diversity among genotypes of Entamoeba histolytica field isolates. Journal of Clinical Microbiology 41: 3748--3756. Hambsch B and Lipp P (2000) Entfernung von Giardien und Cryptosporidien bei der Trinkwasseraufbereitung. Umwelt Technologie Aktuell 11: 152--153. Hamzah Z, Petmitr S, Mungthin M, Leelayoova S, and Chavalitshewinkoon-Petmitr P (2006) Differential detection of Entamoeba histolytica, Entamoeba dispar and

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Entamoeba moshkovskii by a single-round PCR assay. Journal of Clinical Microbiology 44: 3196--3200. Han ET, Guk SM, Kim JL, Joeng HJ, Kim SN, and Chai JY (2003) Detection of parasite eggs from archaeological excavations in the Republic of Korea. Memo´rias do Instituto Oswaldo Cruz 98: 123--126. Hancock CM, Rose JB, and Callahan M (1998) Crypto and Giaridia in US groundwater. Journal of the American Water Works Association 90: 58--61. Haque R, Duggal P, Ali IM, et al. (2002) Innate and acquired resistance to amebiasis in Bangladeshi children. Journal of Infectious Diseases 186: 547--552. Haque R, Huston CD, Hughes M, Houpt E, and Petri WA, Jr. (2003) Amebiasis. New England Journal of Medicine 348: 1565--1573. Haque R, Mondal D, Duggal P, et al. (2006) Entamoeba histolytica infection in children and protection from subsequent amebiasis. Infection and Immunity 74: 904--909. Haque R, Roy S, Siddique A, et al. (2007) Multiplex real-time PCR assay for detection of Entamoeba histolytica, Giardia intestinalis and Cryptosporidium spp. Americam Journal of Tropical Medicine and Hygiene 76: 713--717. Harris A, Matias-Arne´z A, and Hill N (2009) Biting times of A. darlingi in the Bolivian Amazon and its implications for control of malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 100: 45--47. Harrison AD, Nduku W, and Hooper AS (1966) The effects of high magnesium-tocalcium ratio on the egg-laying rate of an aquatic planorbid snail, Biomphalaria pfeifferi. Annals of Tropical Medicine and Parasitology 60: 212--214. Hay SI, Guerra CA, Tatem AJ, Atkinson PM, and Snow RW (2005) Urbanization, malaria transmission and disease burden in Africa. Nature Reviews of Microbiology 3: 81--90. Hay SI, Guerra CA, Tatem AJ, Noor AM, and Snow RW (2004) The global distribution and population at risk of malaria: Past, present, and future. Lancet 4: 327--336. Haylor G, Lawrence A, and Meusch E (1997) Identification of Technical, Social and Economic Constraints to Rearing of Fish in Rice Fields in Lao PDR, Project Report Vol. 3. Oregon Department of Agriculture Natural Resources Programme. London: Aquatic Research. Hemingway J and Ranson H (2000) Insecticide resistance in insect vectors of human disease. Annual Review of Entomology 45: 371--391. Hensel PR, Mitchell SM, and Sowers TEII (2006) Conflict management of riparian disputes. Political Geography 25: 383--411. Hinz E (1996) Helminthiasen des Menschen in Thailand. Frankfurt am Main: Europa¨ischer Verlag der Wissenschaft. Hlavsa MC, Watson JC, and Beach MJ (2005) Giardiasis surveillance – United States, 1998–2002. Morbidity and Mortality Weekly Reports Surveillance Summaries 54: 9--16. Hodgkin C, Molyneux DH, Abiose A, et al. (2007) The future of onchocerciasis control in Africa. PLoS Neglected Tropical Diseases 1: e74. Hoerauf A, Bu¨ttner DW, Adjei O, and Pearlman E (2003) Onchocerciasis. British Medical Journal 326: 207--210. Hofkin BV, Stryker GA, Koech DK, and Loker ES (1991) Consumption of Biomphalaria glabra egg masses and juveniles by the ampullariid snails Pila ovata, Lanistes carinatus and Marisa cornuarietis. Acta Tropica 49: 37--44. Holland GN (2003) Ocular toxoplasmosis: A global reassessment. Part I: Epidemiology and course of disease. American Journal of Ophthalmology 136: 973--988. Hop NT, Van De N, Murrell KD, and Dalsgaard A (2007) Occurrence and species distribution of fishborne zoonotic trematodes in wastewater fed aquaculture in northern Vietnam. Tropical Medicine and International Health 12(supplement 2): 66--72. Hopkins DR (1983) Dracunculiasis: An eradicable scourge. Epidemic Reviews 5: 208--219. Hopkins DR, Ruiz-Tiben E, Diallo N, Withers PC, Jr., and Maguire JH (2002) Dracunculiasis eradication: And now, Sudan. American Journal of Tropical Medicine and Hygiene 67: 415--422. Hopkins DR, Ruiz-Tiben E, Downs P, Withers PC, Jr., and Maguire JH (2005) Dracunculiasis eradication: The final inch. American Journal of Tropical Medicine and Hygiene 73: 669--675. Hopkins DR, Ruiz-Tiben E, and Roy ML (2008) Update: Progress toward global eradication of dracunculiasis, January–June 2008. Morbidity and Mortality Weekly Report 57: 1173--1176. Hopkins DR, Ruiz-Tiben E, and Ruebush TK (1997) Dracunculiasis eradication: Almost a reality. American Journal of Tropical Medicine and Hygiene 57: 252--259. Hopkins DR and Withers PC, Jr. (2002) Sudan’s war and eradication of dracunculiasis. Lancet 360(supplement): 21--22. Hora´k P and Kola´ˇrova´ L (2001) Bird schistosomes: Do they die in mammalian skin? Trends in Parasitology 17: 66--69. Horton J, Witt C, Ottesen EA, et al. (2000) An analysis of the safety of the single dose, two drug regimens used in programmes to eliminate lymphatic filariasis. Parasitology 121: 147--160.

359

Hotez PJ, Feng Z, Xu LQ, et al. (1997) Emerging and reemerging helminthiases and public health of China. Emerging Infectious Diseases 3: 303--310. Hotez PJ, Molyneux DH, Fenwick A, et al. (2007) Control of neglected tropical diseases. New England Journal of Medicine 357: 1018--1027. Hougard M, Yame´ogo L, Se´ke´te´li A, Programme Manager for the African Programme for Onchocerciasis (APOC), Boatin B, and Dadzie KY (1997) Twenty-two years of blackfly control in the onchocerciasis control programme in West Africa. Parasitology Today 13: 425–431. Howarth FG (1991) Environmental impacts of classical biological control. Annual Review of Entomology 36: 485--509. Hu XS, Feng RY, Zheng ZR, Liang JH, Wang HX, and Lu JH (1982) Hepatic damage in experimental and clinical paragonimiasis. American Journal of Tropical Medicine and Hygiene 31: 1148--1155. Hudson P and Greenman J (1998) Competition mediated by parasites: Biological and theoretical progress. Trends in Ecology and Evolution 13: 387--390. Hudson PJ, Cattadori IM, Boag B, and Dobson AP (2006) Climate disruption and parasite–host dynamics: Patterns and processes associated with warming and the frequency of extreme climatic events. Journal of Helminthology 80: 175--182. Hulden L and Hulden L (2009) The decline of malaria in Finland – the impact of the vector and social variables. Malaria Journal 7: 94. Hung CC, Ji DD, Sun HY, et al. (2008) Increased risk for Entamoeba histolytica infection and invasive amebiasis in HIV seropositive men who have sex with men in Taiwan. PLoS Neglected Tropical Diseases 2: e175. Hunter JM (2003) Inherited burden of disease: Agricultural dams and the persistence of bloody urine (schistosomiasis hematobium) in the Upper East Region of Ghana, 1959–1997. Social Science and Medicine 56: 219--234. Hutin YJF, Sombardier MN, Liguory O, et al. (1998) Risk factors for intestinal microsporidiosis in patients with human immunodeficiency virus infection: A case–control study. Journal of Infectious Diseases 178: 904--907. IARC (1994) IARC monographs on the evaluation of carcinogenic risks to humans. 61: 121–175. Idris MA, Shaban M, Richter J, Mone´ H, Mouahid G, and Ruppel A (2003) Emergence of infections with Schistosoma mansoni in the Dhofar Governorate, Oman. Acta Tropica 88: 137--144. Ijumba JN and Lindsay SW (2001) Impact of irrigation on malaria in Africa: Paddies paradox. Medical and Veterinary Entomology 15: 1--11. Ilyinskikh EN, Lepyehin AV, Logviniv SV, and Ilyinskikh NN (1998) Estimation of the mutagenic potential of the trematode Opisthorchis felineus in experimentally infected guinea pigs. Parasitology Research 84: 570--572. IPCC (2007) Climate change 2007: The physical science basis. In: Solomon S, Qin D, Chen Z, et al. (eds.) Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge: Cambridge University Press. Iriemenam NC, Oyibo WA, and Fagbenro-Beyioku AF (2008) Dracunculiasis – the saddle is virtually ended. Parasitology Research 102: 343--347. Jackson RB, Carpenter SR, Dahm CN, et al. (2001) Water in a changing world. Ecological Applications 11: 1027--1045. Jackson TF (1998) Entamoeba histolytica and Entamoeba dispar are distinct species; clinical, epidemiological and serological evidence. International Journal for Parasitology 28: 181--186. Jackson Y, Pastore R, Sudre P, Loutan L, and Chappuis F (2007) Diphyllobothrium latum outbreak from marinated raw perch, Lake Geneva, Switzerland. Emerging Infectious Diseases 13: 1957--1958. Jacobsen KH and Koopman JS (2004) Declining hepatitis A seroprevalence: A global review and analysis. Epidemiology and Infection 132: 1005--1022. Jaksic FM (2001) Ecological effects of El Nino in terrestrial ecosystems in western South America. Ecography 24: 241--250. James TY, Kauff F, Schoch CL, et al. (2006) Reconstructing the early phylogeny of fungi using a six-gene phylogeny. Nature 443: 818--822. Jelinek T, Nothdurft HD, and Lo¨scher T (1996) Schistosomiasis in travelers and expatriates. Journal of Travel Medicine 3: 160--164. Jelinek T, Schulte C, Behrens R, and Grobusch MP (2002) Imported falciparum malaria in Europe: Sentinel surveillance data from the European network on surveillance of imported infectious diseases. Clinical Infectious Diseases 34: 572--576. Jeong HJ and Yu HS (2005) The role of domestic tap water in Acanthamoeba contamination in contact lens storage cases in Korea. Korean Journal of Parasitology 43: 47--50. Jex AR, Smith HV, Monis PT, Campbell BE, and Gasser RB (2008) Cryptosporidium – biotechnological advances in the detection, diagnosis and analysis of genetic variation. Biotechnology Advances 26: 304--317. Jiminez B, Chavez A, Maya C, and Jardines L (2001) Removal of microorganisms in different stages of wastewater treatment for Mexico City. Water Science and Technology 43: 155--162.

360

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Jirku M, Valigurova´ A, Koudela B, Krizek J, Modry´ D, and Slapeta J (2008) New species of Cryptosporidium Tyzzer, 1907 (Apicomplexa) from amphibian host: Morphology, biology and phylogeny. Folia Parasitologica 55: 81--94. Jongsuksuntigul P (2002) Parasitic disease in northeast Thailand Seminar in Parasitic Diseases in Northeast Thailand, pp. 3--18. Khon Kaen, Thailand: Klungnana Vitaya. Jongsuksuntigul P and Imsomboon T (1997) The impact of a decade long opisthorchiasis control program in northeastern Thailand. Southeast Asian Journal of Tropical Medicine and Public Health 28: 551--557. Jongsuksuntigul P and Imsomboon T (2003) Opisthorchiasis control in Thailand. Acta Tropica 88: 229--232. Jongwutiwes S, Putaporntip C, Iwasaki T, Sata T, and Kanbara H (2004) Naturally acquired Plasmodium knowlesi malaria in human, Thailand. Emerging Infectious Diseases 10: 2211--2213. Joyce T, McGuigan KG, Elmore-Meegan M, and Conroy RM (1996) Prevalence of enteropathogens in stools of rural Maasai children under 5 years of age in the Maasail region of the Kenyan Rift Valley. East African Medical Journal 73: 59--62. Jue R, Schmalz T, Carter K, and Nett RJ (2009) Outbreak of cryptosporidiosis associated with a splash park – Idaho, 2007. Morbidity and Mortality Weekly Reports 58: 615--618. Kaewkes S (2003) Taxonomy and biology of liver flukes. Acta Tropica 88: 177--186. Kaplan RM (2001) Fasciola hepatica: A review of the economic impact on cattle and considerations for control. Veterinary Therapeutics 2: 40--50. Kazura JW, Neill M, Peters PAS, and Dennis E (1985) Swamp rice farming: Possible effects on endemicity of schistosomiasis mansoni and haematobuim in a population in Liberia. American Journal of Tropical Medicine and Hygiene 34: 107--111. Keene WE (2006) Entamoeba histolytica. In: American Water Works Association (ed.) Waterborne Pathogens, 310pp. Denver, CO: American Water Works Association. Keiser J, De Castro MC, Maltese MF, et al. (2005a) Effect of irrigation and large dams on the burden of malaria on a global and regional scale. American Journal of Tropical Medicine and Hygiene 72: 392--406. Keiser J, Singer BH, and Utzinger J (2005b) Reducing the burden of malaria in different eco-epidemiological settings with environmental managment: A systematic review. Lancet Infectious Diseases 5: 695--708. Keiser J and Utzinger J (2005) Emerging foodborne trematodiasis. Emerging Infectious Diseases 11: 1507--1514. Keiser J, Utzinger J, De Castro MC, Smith TA, Tanner M, and Singer BH (2004) Urbanization in sub-Saharan Africa and implication for malaria control. American Journal of Tropical Medicine and Hygiene 71(supplement 2): 118--127. Kelly-Hope LA, Diggle PJ, Rowlingson BS, et al. (2006) Negative spatial association between lymphatic filariasis and malaria in West Africa. Tropical Medicine and International Health 11: 129--135. Khalil M (1949) The national campaign for the treatment and control of schistosomiasis from the scientific and economic aspects. Journal of the Royal Egyptian Medical Association 32: 817--856. Khandelwal N, Shaw J, and Jain MK (2008) Biliary parasites: Diagnostic and therapeutic strategies. Current Treatment Options in Gastroenterology 11: 85--95. Khuhaprema T, Srivatanakul P, Sriplung H, et al. (2007) Cancer in Thailand, Vol. IV, 1998–2000. Bangkok: Bangkok Medical Publishers. Kibbler CC, Smith A, Hamilton-Dutoit SJ, Milburn H, Pattinson JK, and Prentice HG (1987) Pulmonary cryptosporidiosis occurring in a bone marrow transplant patient. Scandinavian Journal of Infectious Diseases 19: 581--584. Kilvington S, Gray T, Dart J, et al. (2004) Acanthamoeba keratitis: The role of domestic tap water contamination in the United Kingdom. Investigative Ophthalmology and Visual Science 45: 165--169. King CH (2005) Cestodes. In: Mandell GL, Bennett JE, and Dolin R (eds.) Principles and Practice of Infectious Diseases, 6th edn., pp. 3285--3293. Philadelphia, PA: Elsevier Churchill Livinstone. Kirkwood B, Smith P, Marshall T, and Prost A (1983) Variations in the prevalence and intensity of microfilarial infections by age, sex, place and time in the area of the Onchocerciasis Control Program. Transactions of the Royal Society of Tropical Medicine and Hygiene 77: 857--861. Kiszewski A, Mellinger A, Spielman A, Malaney P, Ehrlich Sachs S, and Sachs J (2004) A global index representing the stability of malaria transmission. American Journal of Tropical Medicine and Hygiene 70: 486--498. Kiszewski A and Teklehaimanot A (2004) A review of the clinical and epidemiologic burdens of epidemic malaria. American Journal of Tropicl Medicine and Hygiene 71: 128--135. Klimpel S, Palm HW, Ru¨ckert S, and Piatkowski U (2004) The life cycle of Anisakis simplex in the Norwegian Deep (northern North Sea). Parasitology Research 94: 1--9. Koie M (2001) Experimental infections of copepods and sticklebacks Gasterosteus aculeatus with small ensheathed and large third-stage larvae of Anisakis simplex (Nematoda, Ascaridoidea, Anisakidae). Parasitology Research 87: 32--36.

Kovats RS, Bouma MJ, Hajat S, Worrall E, and Haines A (2003) El Nino and health. Lancet 362: 1481--1489. Kravetz JD and Federman DG (2005) Toxoplasmosis in pregnancy. American Journal of Medicine 118: 212--216. Krueger A and Hennings IC (2006) Molecular phylogenetics of blackflies of the Simulium damnosum complex and cytophylogenetic implications. Molecular Phylogenetics and Evolution 39: 83--90. Kulkarni MA, Kweka E, Nyale E, et al. (2006) Entomological evaluation of malaria vectors at different altitudes in Hai district, northeastern Tanzania. Journal of Medical Entomology 43: 580--588. Kumchoo K, Wongsawad C, Chai JY, Vanittanakom P, and Rojanapaibul A (2005) High prevalence of Haplorchis taichui metacercariae in cyprinoid fish from Chiang Mai province, Thailand. Southeast Asian Journal of Tropical Medicine and Public Health 36: 451--455. Lacerda CM, Freire W, Vieria de Melo PS, Lacerda HR, and Carvalho G (1999) Splenectomy and ligation of the left gastric vein in schistosomiasis mansoni: The effect of esophageal variceal pressure measured by a non-invasive technique. Revista Brasileira de Cirurgia 26: 15--19. Lalloo DG, Shingadia D, Pasvol G, et al. (2007) UK malaria treatment guidelines. Journal of Infection 54: 111--121. Laird PP and Boray JC (2008) Human fascioliasis successfully treated with triclabendazole. International Medical Journal 22: 45--47. Lammie PJ, Hightower AW, and Eberhard ML (1994) Age-specific prevalence of antigenemia in a Wuchereria bancrofti-exposed population. American Journal of Tropical Medicine and Hygiene 51: 348--355. Lammie PJ, Reiss MD, Dimock KA, Streit TG, Roberts JM, and Eberhard ML (1998) Longitudinal analysis of the development of filarial infection and antifilarial immunity in a cohort of Haitian children. American Journal of Tropical Medicine and Hygiene 59: 217--221. Lane S and Lloyd D (2002) Current trends in research into the water-borne parasite Giardia. Critical Reviews of Microbiology 28: 123--147. Lanoix JN (1958) Relation between irrigation engineering and bilhaziosis. Bulletin of the World Health Organisation 18: 1011--1035. Lazdins-Helds JK, Remme JHF, and Boakye B (2003) Onchocerciasis. Nature Reviews 1: 178--179. LeChevallier MW and Au KK (2004) Water Treatment and Pathogen Control: Process Efficiency in Achieving Safe Drinking Water. Geneva: World Health Organization. LeChevallier MW, Norton WD, and Lee RG (1991) Giardia and Cryptosporidium spp. in filtered drinking water supplies. Applied and Environmental Microbiology 57: 2617--2621. Lee EJ and Schwab KJ (2005) Deficiencies in drinking water distribution systems in developing countries. Journal of Water and Health 3(2): 109--127. Lee JS, Lee J, Kim SH, and Yong TS (2007) Molecular cloning and characterization of a major egg antigen in Paragonimus westermani and its use in ELISA for the immunodiagnosis of paragonimiasis. Parasitology Research 100: 677-681. Lemos ACM, Coelho JC, Matos ED, Montal G, Aguiar F, and Badaro´ R (2007) Paragonimiasis: First case reported in Brazil. Brazilian Journal of Infectious Diseases 11: 153--156. Leonhard S, Pfister K, Beelitz P, Wielinga C, and Thompson RCA (2007) The molecular characterisation of Giardia from dogs in southern Germany. Veterinary Parasitology 150: 33--38. Leutscher P, Ramarokoto CE, Reimert C, Feldmeier H, Esterre P, and Vennervald BJ (2000) Communitiy-based study of genital schistosomiasis in men from Madagascar. Lancet 355: 117--118. Lewis DJ (1960) Observations on the Simulium neavei complex at Amani in Tanganyika. Bulletin of Entomological Research 51: 95--113. Le´veˆque C, Hougard JM, Resh V, Statzner B, and Yame´ogo L (2003) Freshwater ecology and biodiversity in the tropics: What did we learn from 30 years of onchocerciasis control and the associated biomonitoring of West African rivers? Hydrobiologia 500: 23--49. Li YS, Raso G, Zhao ZY, He YK, Ellis MK, and McManus DP (2007) Large water management projects and schistosomiasis control, Dongting Lake Region, China. Emerging Infectious Diseases 13: 973--979. Lim MK, Ju YH, Franceschi S, et al. (2006) Clonorchis sinensis infection and increasing risk of cholangiocarcinoma in the Republic of Korea. American Society of Tropical Medicine and Hygiene 75: 93--96. Lindblade KA, Walker ED, Onapa AW, Katunga J, and Wilson ML (1999) Highland malaria in Uganda: Prospective analysis of an epidemic associated with El Nino. Transactions of the Royal Society of Tropical Medicine and Hygiene 93: 480--487. Lindsay SW, Bodker R, Malima R, Msangeni HA, and Kisinza W (2000) Effect of 1997– 98 El Nino on highland malaria in Tanzania. Lancet 355: 989--990.

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Lines J, Harpham T, Leake C, and Schofield C (1994) Trends, priorities and policy directions in the control of vector-borne diseases in urban environments. Health Policy and Planning 9: 113--129. Little MP, Breitling LP, Basa´n˜ez MG, Alley ES, and Boatin BA (2004) Association between microfilarial load and excess mortality in onchocerciasis: An epidemiological study. Lancet 363: 1514--1521. Liu Q, Wei F, Liu W, Yang S, and Zhang X (2008) Paragonimiasis: An important foodborne zoonosis in China. Trends in Parasitology 24: 318--323. Liu WQ, Liu J, Zhang JH, Long XC, Lei JH, and Li YL (2007) Comparison of ancient and modern Clonorchis sinensis based on ITS1 and ITS2 sequences. Acta Tropica 101: 91--94. Lisle JT and Rose JB (1995) Cryptosporidium contamination of water in the USA and UK: A mini-review. Journal of Water Supply Research Technology 44: 103--105. Loker ES and Brant SV (2006) Diversification, dioecy and dimorphism in schistosomes. Trends in Parasitology 22: 521--528. Lopez-Serrano MC, Gomez AA, Daschner A, et al. (2000) Gastroallergic anisakiasis: Findings in 22 patients. Journal of Gastroenterology and Hepatology 15: 503--506. Lorenzo-Morales J, Lindo JF, Martinez E, et al. (2005a) Pathogenic Acanthamoeba strains from water sources in Jamaica, West Indies. Annals of Tropical Medicine and Parasitology 99: 751--758. Lorenzo-Morales J, Ortega-Rivas A, Forondo P, Martinez E, and Valladares B (2005b) Isolation and identification of pathogenic Acanthamoeba strains in Tenerife, Canary Islands, Spain from water sources. Parasitology Research 95: 273--277. Lowe S, Browne M, Boudjelas S, and De Poorter M (2000) 100 of the World’s Worst Invasive Alien Species: A Selection from the Global Invasive Species Database. 12pp. Auckland: The Invasive Species Specialist Group Commission (SSC) of the World Conservation Union (IUCN). Lucius R and Loos-Frank B (2008) Biologie von Parasiten, 2nd edn. Heidelberg: Springer. Lun ZR, Gasser RB, Lai DH, Li AX, Xing BY, and Fang YY (2005) Clonorchiasis: A key foodborne zoonosis in China. Lancet Infectious Diseases 5: 31--41. Luzon-Pena M, Rojo-Vazquez FA, and Go´mez-Bautista M (1992) The overwintering of Fasciola hepatica eggs under semi-arid and temperate Mediterranean climate (Madrid, Spain). Zentralblatt fu¨r Veterina¨rmedizin B 39: 369--375. Lu¨thy P and Studer D (1986) Control of simuliid blackflies and mosquitoes with Bacillus thuringiensis subsp. israelensis. World Journal of Microbiology and Biotechnology 2: 91--99. Mac TL, Tran DS, Quet F, Odermatt P, Preux PM, and Tan CT (2007) Epilepsy in Asia: Epidemiology, aetiology and clinical management. Lancet Neurology 6: 533--543. Mackenzie WR, Hoxie NJ, Proctor ME, et al. (1994) A massive outbreak in Milwaukee of Cryptosporidium infection transmitted through the public water supply. New England Journal of Medicine 331: 161--167. Mackintosh CL, Beeson JG, and Marsh K (2004) Clinical features and pathogenesis of severe malaria. Trends in Parasitology 20: 597--603. Madson H (1987) Effect of calcium concentration on growth and egg laying of Helisoma duryi, Biomphalaria alexandrina, B. camerunensis and Bulinus truncatus (Gastropoda: Planorbidae). Journal of Applied Ecology 24: 823--836. Magambo JK, Zeyhle E, and Wachira TM (1998) Prevalence of intestinal parasites among children in southern Sudan. East African Medical Journal 75: 288--290. Malan RM, Haratani J, Schneider C, and Spielman A (1977) The Northern Zone, Lake Chad Basin environmental health assessment (Bol Polder Project Area). In: American Public Health Association (ed.) Sahel Epidemiological and Environmental Assessment Project, vol. 3, section 1, part A, pp. 1–92. Washington, DC: American Public Health Association. Malczewski A, Wescott RB, Spratling BM, and Gorhan JR (1975) Internal parasites of Washington cattle. American Journal of Veterinary Research 36: 1671--1675. Malek EA (1975) Effect of the Aswan High Dam on prevalence of schistosomiasis in Egypt. Tropical and Geographical Medicine 27: 359--364. Malek EA (1976) Man-made lakes and human health. American Journal of Tropical Medicine and Hygiene 25: 916--917. Marciano-Cabral F and Cabral G (2003) Acanthamoeba spp. as agents of disease in humans. Clinical Microbiology Reviews 16: 273--307. Marcogliese DJ (2001) Implications of climate change for parasitism of animals in the aquatic environment. Canadian Journal of Zoology 79: 1331--1352. Marcos L, Maco V, Samalvides F, Terashima A, Espinoza JR, and Gotuzzo E (2006) Risk factors for Fasciola hepatica infection in children: A case–control study. Transactions of the Royal Society of Tropical Medicine and Hygiene 100: 158--166. Marino DD (2007) Water and food safety in the developing world: Global implications for health and nutrition of infants and young children. Journal of the American Dietetic Association 107: 1930--1934. Marino DD (2007) Water and food safety in the developing world: Global implications for health and nutrition of infants and young children. Journal of the American Dietetic Association 107: 1930--1934.

361

Martens P and Hall L (2000) Malaria on the move: Human population movement and malaria transmission. Emerging Infectious Diseases 6: 103--109. Mas-Coma S, Bargues MD, and Valero MA (2005) Fascioliasis and other plant-borne trematode zoonoses. International Journal for Parasitology 35: 1255--1278. Mas-Coma S, Estaban JG, and Bargues MD (1999) Epidemiology of human fascioliasis: A review and proposed new classification. Bulletin of the World Health Organization 77: 340--346. Mas-Coma S, Valero MA, and Bargues MD (2008) Effects of climate change on animal and zoonotic helminthisases. Revue Scientifique et Technique Office International des Epizooties 27: 443--457. Mathers CD, Ezzati M, and Lopez AD (2007) Measuring the burden of neglected tropical diseases: The global burden of disease framework. PLoS Neglected Tropical Diseases 1: e114. Mathis A, Weber R, and Deplazes P (2005) Zoonotic potential of the microsporidia. Clinical Microbiology Reviews 18: 423--445. Matin A, Siddiqui R, Jayasekera S, and Khan NA (2008) Increasing importance of Balamuthia mandrillaris. Clinical Microbiology Reviews 21: 435--448. Matos O, Alves M, Xiao L, Cama V, and Antunes F (2004) Cryptosporidium felis and C. meleagridis in persons with HIV, Portugal. Emerging Infectious Diseases 10: 2256--2257. Matsui A, Kanehara M, and Kanehara M (2003) Palaeoparasitology in Japan – discovery of toilet features. Memo´rias do Instituto Oswaldo Cruz 98: 127--136. Matsui T, Iida M, Murakami M, et al. (1985) Intestinal anisakiasis: Clinical and radiologic features. Radiology 157: 299--302. Mattiucci S and Nascetti G (2006) Molecular systematics, phylogeny and ecology of anisakid nematodes of the genus Anisakis. Parasite 13: 99--113. Mattiucci S, Nascetti G, Cianchi R, et al. (1997) Genetic and ecological data on the Anisakis simplex complex, with evidence for a new species (Nematoda, Ascaroidea, Anisakidae). Journal of Parasitology 83: 401--416. May RM (2007) Parasites, people and policy: Infectious diseases and the Millennium Development Goals. Trends in Ecology and Evolution 22: 497--503. Mboera LEG, Kamugisha MI, Rumisha SF, Kisinza WN, Senkoro KP, and Kitua AY (2008) Malaria and mosquito net utilization among school children in villages with and without healthcare facilities at different altitudes in Iringa Dsitrict, Tanzania. African Health Sciences 8: 114--119. McCarthy FD, Wolf H, and Wu Y (2000) The Growth Costs of Malaria. Working paper 7541. Cambridge: National Bureau of Economic Growth. McClelland G (2002) The trouble with sealworms (Pseudoterranova decipiens species complex, Nematoda): A review. Parasitology 124(supplement): S183--S203. Mehlhorn H and Walldorf V (1988) Life cycles. In: Mehlhorn H (ed.) Parasitology in Focus: Facts and Trends, pp. 1--148. Heidelberg: Springer. Melloul A, Amahmid O, Hassani L, and Bouhoum K (2002) Health effect of human wastes use in agriculture in El Azzouzia (the wastewater spreading area of Marrakesh city, Morocco). International Journal of Environmental Health Research 12: 17--23. Melrose WD (2002) Lymphatic filariasis: New insights into an old disease. International Journal for Parasitology 32: 947--960. Melrose WD (2004) Lymphytic Filariasis: A Review 1862–2002. Killarney: Warwick Educational Publishing. Melrose WD, Durrheim DD, and Burgess GW (2004) Update on immunological tests for lymphatic filariasis. Trends in Parasitology 20: 255--257. Meyer DA and Fried B (2007) The role of helminth infections in carcinogenesis. Advances in Parasitology 65: 239--296. Meyers JD, Kuharic HA, and Holmes KK (1977) Giardia lamblia infection in homosexual men. British Journal of Venereal Disease 53: 54--55. Mian LS and Mulla MS (1986) Survival and ovopositional response of Culex quinquefasciatus Say (Diptera: Culicidae) to sewage effluent. Bulletin of the Society of Vector Ecologists 11: 194--196. Michael E and Bundy DA (1997) Global mapping of lymphatic filariasis. Parasitology Today 13: 472--476. Michael E, Grenfell BT, and Bundy DAP (1994) The association between microfilaraemia and disease in lymphatic filariasis. Proceeding of the Royal Society B 256: 33--40. Michaud DS (2007) Chronic inflammation and bladder cancer. Urologic Oncology 25: 260--268. Micheli F (1999) Eutrophication, fisheries, and consumer-resource dynamics in marine pelagic ecosystems. Science 285: 1396--1398. Middendorp HAJ (1992) Contribution of stocked and wild fish in rice fields to fish production and farmer nutrition in Northeast Thailand. Asian Fisheries Science 5: 145--161. Midtgaard T, Andersen K, and Halvorsen O (2003) Population dynamics of sealworm, Pseudoterranova decipiens sensu lato, in sculpins, Myoxocephalus scorpius, from two areas in Norway between 1990 and 1996. Parasitology Research 89: 387--392.

362

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Miller LH, Baruch DI, Marsh K, and Doumbo OK (2002) The pathogenic basis of malaria. Nature 415: 673--679. Min HK (1984) Clonorchis sinensis: Parthenogenesis and clinical features of infection. Arzneimittelforschung 34: 1151--1153. Minakawa N, Munga S, Atieli F, et al. (2005) Spatial distribution of anopheline larval habitats in western Kenyan highlands: Effects of land cover types and topography. American Journal of Tropical Medicine and Hygiene 73: 157--165. Mitarai S, Nagai H, Satoh K, Hebisawa A, and Shishido H (2001) Amebiasis in Japanese homosexual men with human immunodeficiency virus infection. Internal Medicine 40: 671--675. Mitchell AJ, Goodwin AE, Salmon MJ, and Brandt TM (2002) Experimental infection of an exotic heterophyid trematode, Centrocestus formosanus, in four aquaculture fishes. North American Journal of Aquaculture 64: 55--59. Mitchell AJ, Overstreet RM, Gioodwin AE, and Brandt TM (2005) Spread of an exotic fish-gill trematode: A far-reaching and complex problem. Fisheries 30: 11--16. Mitchell AJ, Salmon MJ, Huffman DG, Goodwin AE, and Brandt TM (2000) Prevalence and pahogenicity of a heterophyid trematode infecting the gills of an endangered fish, the fountain darter, in two central Texas spring-fed rivers. Journal of Aquatic Animal Health 12: 283--289. Moe CL and Rheingans RD (2006) Global challenges in water, sanitation and health. Journal of Health and Water 4: 41--57. Molyneux DH (2003) Lymphatic filariasis (elephantiasis) elimination: A public health success and development opportunity. Filaria Journal 13: 1--6. Molyneux DH (2005) Onchocerciasis control and elimination: Coming of age in resource-constrained health systems. Trends in Parasitology 21: 525--529. Molyneux DH (2009) Filaria control and elimination: Diagnostic, monitoring and surveillance needs. Transactions of the Royal Society of Tropical Medicine and Hygiene 103: 338--341. Molyneux DH, Bradley M, Hoerauf A, Kyelem D, and Taylor MJ (2003) Mass drug treatment for lymphatic filariasis and onchocerciasis. Trends in Parasitology 19: 516--522. Monis PT, Caccio SM, and Thompson RCA (2009) Variation in Giardia: Towards a taxonomic revision of the genus. Trends in Parasitology 25: 93--100. Monis PT and Thompson RCA (2003) Cryptosporidium and Giardia-zoonoses: Fact or fiction? Infection, Genetics and Evolution 3: 233--244. Montgomery MA and Elimelech M (2007) Water and sanitation in developing countries: Including health in the equation. Environmental Science and Technology 41: 17--24. Morley NJ (2007) Anthropogenic effects of reservoir construction on the parasite fauna of aquatic wildlife. EcoHealth 4: 374--383. Morgan JA, Dejong RJ, Adeoye GO, et al. (2005) Origin and diversification of the human parasite Schistosoma mansoni. Molecular Ecology 14: 3889--3902. Morgan-Ryan UM, Fall A, Ward LA, et al. (2002) Cryptosporidium hominis n. sp. (Apicomplexa: Cryptosporidiidae) from Homo sapiens. Journal of Eukaryotic Microbiology 49: 433--440. Mosha FW and Mutero CM (1982) The influence of salinity on larval development and population dynamics Anopheles merus Do¨nitz (Diptera: Culicidae). Bulletin of Entomological Research 72: 119--128. Mueller I, Zimmerman PA, and Reeder JC (2007) Plasmodium malariae and Plasmodium ovale – the ‘bashful’ malaria parasites. Trends in Parasitology 23: 278--283. Muro AI and Raybould JN (1990) Population decline of Simulium woodi and reduced onchocerciasis transmission at Amani, Tanzania, in relation to deforestation. Acta Leiden 59: 153--159. Mu¨ller N and Von Allmen N (2005) Recent insights into the mucosal reactions associated with Giardia lamblia infections. International Journal for Parasitology 35: 1339--1347. Nadler SA, D’Amelio S, Dailey MD, Paggi L, Siu S, and Sakanari JA (2005) Molecular phylogenetics and diagnosis of Anisakis pseudoterranova and contracaecum from northern Pacific marine mammals. Journal of Parasitology 91: 1413--1429. Nawa Y, Hatz C, and Blum J (2005) Sushi delights and parasites: The risk of fishborne and foodborne parasitic zoonoses in Asia. Travel Medicine 41: 1297--1303. Naylor RL, Goldburg RJ, Primavera JH, et al. (2000) Effect of aquaculture on world fish supplies. Nature 405: 1017--1024. Naylor RL, Williams SL, and Strong DR (2001) Aquaculture – a gateway for exotic species. Science 294: 1655--1666. Na´jera JA (1996) Malaria Control among Refugees and Displaced Populations. Geneva: World Health Organization. Nduku WK and Harrison AD (1976) Calcium as a limiting factor in the biology of Biomphalaria pfeifferi (Krauss), (Gastropoda: Planorbidae). Hydrobiologia 49: 143--170. Nieuwenhuizen N, Lopata A, Jeebhay M, Herbert D, Robins T, and Brombacher F (2006) Exposure to the fish parasite Anisakis causes allergic airway hyper-reactivity and dermatitis. Journal of Allergy and Clinical Immunology 117: 1098--1105.

Nkinin SW, Asonganyi T, Didier ES, and Kaneshiro ES (2007) Microsporidian infection is prevalent in healthy people in Cameroon. Journal of Clinical Microbiology 45: 2841--2846. Noma M, Nwoke BE, Nutall I, Tambala PA, and Enyong P (2002) Rapid entomological mapping of onchocerciasis (REMO): Its application by the African Programme for Onchocerciasis Control (APOC). Annals of Tropical Medicine and Parasitology 96(supplement 1): 29--39. Norhayati M, Penggabean M, Oothuman P, and Fatmah MS (1998) Prevalence and some risk factors of Giardia duodenalis infection in rural community in Malaysia. Southeast Asian Journal of Tropical Medicine and Public Health 29: 735--738. Nwaorgu OC (1991) The effectiveness of combined control measures on the prevalence of guinea worm disease in Anambra State, Nigeria. Journal of Helminthology 65: 226--231. Nygard K, Schimmer B, Sobstad O, et al. (2006) A large community outbreak of waterborne giardiasis – delayed detection in a non-endemic urban area. BMC Public Health 6: 141--150. Obara A, Nakamura-Uchiyama F, Hiromatsu K, and Nawa Y (2004) Paragonimiasis cases recently found among immigrants in Japan. Internal Medicine 43: 388--392. Obi CL, Momba MNB, Samie A, Igumbor JO, Green E, and Musie E (2007) Microbiological, pysico-chemical and management parameters impinging on the efficiency of small water treatment plants in the Limpopo and Mpumalanga provinces of South Africa. Water South Africa 33: 229--238. Obsomer V, Defourny P, and Coosemans M (2007) The Anopheles dirus complex: Spatial distribution and environmental drivers. Malaria Journal 6: 26 (doi: 10.1186/1475-2875-6-26). Odermatt P, Habe S, Manichanh S, et al. (2007) Paragonimiasis and its intermediate hosts in a transmission focus in Lao People’s Democratic Republic. Acta Tropica 103: 108--115. Ohnishi K, Kato Y, Imamura A, et al. (2004) Present characteristics of symptomatic Entamoeba histolytica infection in the big cities of Japan. Epidemiology and Infection 132: 57--60. Oki T and Kanae S (2006) Global hydrological cycles and world water resources. Science 313: 68--72. Okhuysen PC (2001) Traveler’s diarrhea due to intestinal protozoa. Clinical Infectious Diseases 33: 110--114. Okhuysen PC, Chappell CL, Crabb JH, Sterling CR, and DuPont HL (1999) Virulence of three distinct Cryprosporidium parvum isolates for healthy adults. Journal of Infectious Diseases 180: 1275--1281. Ong C, Moorehead W, Ross A, and Isaac-Renton J (1996) Studies of Giardia spp. and Cryptosporidium spp. in two adjacent watersheds. Applied and Environmental Microbiology 62: 2798--2805. Oomen JMV, De Wold J, and Jobin WR (1994) Health and Irrigation: Incorporation of Disease-Control Measures in Irrigation, a Multi-Faceted Task in Design, Construction, Operation, 2nd edn. Wageningen: International Institute for Land Reclamation and Improvement. Opara KN, Usip LP, and Akpabio EE (2008) Transmission dynamics of Simulium damnosum in rural communities of Akwa Ibom State, Nigeria. Journal of Vector Borne Diseases 45: 225--230. Ortega YR, Sterling CR, Gilman RH, Cama VA, and Diaz F (1993) Cyclospora species – a new protozoan pathogen of humans. New England Journal of Medicine 328: 1308--1312. Oshima T and Wakai R (1983) Epidemiology of Diphyllobothrium latum infection in Japan, with special reference to infection of cherry salmon. Japanese Journal of Antibiotics 36: 566--572. Osterholm MT, Fortfang JC, Ristien TL, et al. (1981) An outbreak of food-borne giardiasis. New England Journal of Medicine 304: 24--28. O’Sullivan L, Jardine A, Cook A, and Weinstein P (2008) Deforestation, mosquitoes, and ancient Rome: Lessons for today. BioScience 58: 756--760. Pancharatnam S, Jacob E, and Kang G (1998) Human diphyllobaothriasis: First report from India. Transactions of the Royal Society of Tropical Medicine and Hygiene 92: 179--180. Pantenburg B, Dann SM, Wang H-C, et al. (2008) Intestinal immune response to human Cryptosporidium sp. infection. Infection and Immunity 76: 23--29. Paperna I (1969) Study of an outbreak of schistosomiasis in the newly formed Volta lake in Ghana. Zeitschrift fu¨r Tropenmedizin und Parasitologie 21: 411--424. Park GM (2006) Genetic comparison of liver flukes, Clonorchis sinensis and Opisthorchis viverrini, based on rDNA and mtDNA gene sequences. Parasitology Research 100: 351--357. Pastor A, Neely J, Goodfriend D, et al. (2002) Local transmission of Plasmodium vivax malaria – Virginia, 2002. Morbidity and Mortality Weekly Report 51: 921--922. Patz JA, Campbell-Lendrum D, Holloway T, and Foley JA (2005) Impact of regional climate change on human health. Nature 438: 310--317.

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Patz JA, Daszak P, Tabor GM, et al. (2004) Unhealthy landscapes: Policy recommendations on land use change and infectious disease emergence. Environmental Health Perspectives 112: 1092--1098. Patz JA, Graczyk TK, Geller N, and Vittor AY (2000) Effects of environmental change on emerging parasitic diseases. International Journal for Parasitology 30: 1395--1405. Peng HW, Chao HL, and Fan PC (1993) Imported Opisthorchis viverrini and parasite infections from Thai labourers in Taiwan. Journal of Helminthology 67: 102--106. Peng MM, Wilson ML, Holland RE, Meshnick SR, Lal AA, and Xiao L (2003) Genetic diversity of Cryptosporidium spp. in cattle in Michigan: Implications for understanding the transmission dynamics. Parasitology Research 90: 175--180. Pereira SJ, Ramirez NE, Xiao L, and Ward LA (2002) Pathogenesis of human and bovine Cryptosporidium parvum in gnotobiotic pigs. Journal of Infectious Diseases 186: 715--718. Perrett S and Whitfield PJ (1996) Currently available molluscicides. Parasitology Today 12: 156--159. Petney TN (2001) Environmental and social change in the tropics and its influence on parasite infection. International Journal for Parasitology 31: 919--932. Petney TN, Sithithaworn P, Satrawaha R, et al. (2009) Potential malaria re-emergence, northeastern Thailand. Emerging Infectious Diseases 15: 1330--1331. Petri WA Jr. (2002) Pathogenesis of amebiasis. Current Opinion in Microbiology 5: 443--447. Petri WA, Jr. (2005) Treatment of giardiasis. Current Treatment Options in Gastroenterology 8: 13--17. Petri WA, Jr., Haque R, Lyerly D, and Vines RR (2000) Estimating the impact of amebiasis on health. Parasitology Today 16: 320--321. Petri WA, Jr., Haque R, and Mann BJ (2002) The bittersweet interface of parasite and host: Lectin-carbohydrate interactions during human invasion by the parasite Entamoeba histolytica. Annual Review of Microbiology 56: 39--64. Petri WA, Jr. and Singh U (1999) Diagnosis and management of amoebiasis. Clinical Infectious Diseases 29: 1117--1125. Phillips SC, Mildvan D, William DC, Gelb AM, and White MC (1981) Sexual transmission of enteric protozoa and helminths in a veneral-disease-clinic population. New England Journal of Medicine 305: 603--606. Picquet M, Ernould JC, Vercruysse J, et al. (1996) The epidemiology of human schistosomiasis in the Senegal river basin. Transactions of the Royal Society of Tropical Medicine and Hygiene 90: 340--346. Pimentel D, Houser J, Preiss E, et al. (1997) Water resources: Agriculture, the environment, and society. BioScience 47: 97--106. Pitt S, Pearcy BE, Stevens RH, Sharipov A, and Banatvala N (1998) War in Tajikistan and re-emergence of Plasmodium falciparum. Lancet 352: 1279. Poda JN, Sondo B, and Parent G (2003) Influence des hydro-ame´nagements sur la distribution des bilharzioses et de leurs hoˆtes interme´diaires au Burkina Faso. Sante 13: 49--53. Poggensee G and Feldmaier H (2001) Female genital schistosomiasis: Facts and hypotheses. Acta Tropica 79: 193--210. Pointier JP and David P (2004) Biological control of Biomphalaria glabrata, the intermediate host of schistosomes, by Marisa cornuarietis in ponds of Guadeloupe: Long-term impact on the local snail fauna and aquatic flora. Biological Control 29: 81--89. Pointier JP and Giboda M (1999) The case for biological control of snail intermediate hosts of Schistosoma mansoni. Parasitology Today 15: 395--397. Pointier JP and Jourdane J (2000) Biological control of the snail hosts of schistosomiasis in areas of low transmission: The example of the Caribbean area. Acta Tropica 77: 53--60. Pointier JP and McCullough F (1989) Biological control of the snail hosts of Schistosoma mansoni in the Caribbean area using Thiara spp. Acta Tropica 46: 147--155. Polderman AM (1986) Schistosomiasis in a mining area: Intersectoral implications. Tropical Medicine and Parasitology 37: 195--199. Polderman AM, Mpamila K, Manshande JP, Gryseels B, and Van Schayk O (1985) Historical, geological and ecological aspects of transmission of intestinal schistosomiasis in Maniema, Kivu Province, Zaire. Annales de la Societe Belge de Medecine Tropicale (Antwerpen) 65: 251--261. Polderman AM, Mpamila K, Manshande JP, and Bouwhuis-Hoogerwerf ML (1985) Methodology and interpretation of parasitological surveillance of intestinal schistosomiasis in Maniema, Kivu Province, Zaire. Annales de la Societe Belge de Medecine Tropicale (Antwerpen) 65: 243--249. Polis MA, Tuazon CU, Alling DW, and Talmanis E (1986) Transmission of Giardia lamblia from a day care center to the community. American Journal of Public Health 76: 1142--1144. Polley L and Thompson RCA (2009) Parasite zoonoses and climate change: Molecular tools for tracking shifting boundaries. Trends in Parasitology 25: 285--291.

363

Porter JD, Ragazzoni HP, Buchanon JD, Waskin HA, Juranek DD, and Parkin WE (1988) Giardia transmission in a swimming pool. American Journal of Public Health 78: 659--662. Post RJ, Mustapha M, and Krueger A (2007) Taxonomy and inventory of the cytotypes of the Simulium damnosum complex (Diptera: Simuliidae) in relation to onchocerciasis. Tropical Medicine and International Health 12: 1342--1353. Postel SL, Daily GC, and Ehrlich PR (1996) Human appropriation of renewable fresh water. Science 271: 785--788. Potseluev AN (1991) The role of small hydraulic engineering installations in altering the habitat conditions for molluscs – the prime intermediate hosts of Opisthorchis. Meditsinskaia Parazitologiia 1991(5): 32--34 (in Russian). Poulin R (2005) Global warming and temperature-mediated increases in cercarial emergence in trematode parasites. Parasitology 132: 143--151. Poveda G, Rojas W, Quinones ML, et al. (2001) Coupling between annual and ENSO timescales in the malaria–climate association in Colombia. Environmental Health Perspectives 109: 489--493. Pozio E (2008) Epidemiology and control prospects of foodborne parasitic zoonoses in the European Union. Parassitologia 50: 17--24. Presidente PJA and Beveridge I (1978) Cholangitis associated with species of Progamotaenia (Cestoda: Anoplocephalidae) in the bile ducts of marsupials. Journal of Wildlife Diseases 14: 371--377. Qvarnstrom Y, Da Silva AJ, Schuster FL, Gelman BB, and Visvervara GS (2009) Molecular confirmation of Sappinia pedata as a causative agent of amoebic encephalitis. Journal of Infectious Diseases 199: 1139--1142. Rab MA, Freeman TW, Rahim S, Durrani N, Simon-Taha A, and Rowland M (2003) High altitude epidemic malaria in Bamian province, central Afghanistan. La Revue de Sante´ de la Me´diterrane´e Orientale 9: 232--239. Raccurt CP, Lowrie RC, Jr., Katz SP, and Duverseau YT (1988) Epidemiology of Wuchereria bancrofti in Leogane, Haiti. Transactions of the Royal Society of Tropical Medicine and Hygiene 82: 721--725. Radomyos B, Bunnag D, and Harinasuta T (1983) Haplorchis pumilio (Looss) infection in man in northeastern Thailand. Southeast Asian Journal of Tropical Medicine and Public Health 14: 223--227. Radomyos B, Wongsaroj T, Wilairatana P, Radomyos P, Praevanich R, Meesomboon V, and Jongsuksuntikul P (1998) Opisthorchiasis and intestinal fluke infections in Northern Thailand. Southeast Asian Journal of Tropical Medicine and Public Health 29: 123--127. Ramaiah KD, Das PK, Michael E, and Guyatt H (2000) The economic burden of lymphatic filariasis in India. Parasitology Today 16: 251--253. Ramirez NE, Ward LA, and Sreevastan S (2004) The biology and epidemiology of cryptosporidiosis in humans and animals. Microbes and Infection 6: 733--785. Rauch AM, Van R, Bartlett AV, and Pickering LK (1990) Longitudinal study of Giardia lamblia in a day care center population. Pediatric Infectious Diseases Journal 9: 186--189. Regnath T, Klemm T, and Ignatius R (2006) Rapid and accurate detection of Giardia lamblia and Cryptosporidium spp. antigens in human fecal specimens by new commercially available qualitative immunochromatagraphic assays. European Journal of Clinical Microbiology and Infectious Diseases 25: 807--809. Reinhard K and Urban O (2003) Diagnosing ancient diphyllobothriasis from Chinchorro mummies. Memo´rias do Instituto Oswaldo Cruz 98(supplement 1): 191--193. Reiter P (2000) From Shakespeare to Defoe: Malaria in England in the little Ice Age. Emerging Infectious Diseases 6: 1--11. Revenga JE (1993) Diphyllobothrium dendriticum and Diphyllobothrium latum in fishes from southern Argentina: Association, abundance, distribution, pathological effects, and risk of human infection. Parasitology Research 79: 379--383. Richards FO, Jr., Boatin B, Sauerbrey M, and Se´ke´te´li A (2001) Control of onchocerciasis today: Status and challenges. Trends in Parasitology 17: 558--563. Rideout BA, Gardiner CH, Stalis IH, Zuba JR, Hadfield T, and Visvesvara GS (1997) Fatal infections with Balamuthia mandrillaris (a free-living amoeba) in gorillas and other Old World primates. Veterinary Pathology 34: 15--22. Robert V, MacIntyre K, Keating J, et al. (2003) Malaria transmission in urban Sub-Saharan Africa. American Journal of Tropical Medicine and Hygiene 68: 169--176. Roberts DR, Laughlin LL, Hscheih P, and Legters LJ (1997) DDT, global strategies, and a malaria control crisis in South America. Emerging Infectious Diseases 3: 309--311. Roberts JA and Suhardono (1996) Approaches to the control of fasciolosis in ruminants. International Journal for Parasitology 26: 971--981. Rocha-Azevedo Bda, Tanowitz HB, and Marciano-Cabral F (2009) Diagnosis of infections caused by pathogenic free-living amoebae. Interdisciplinary Perspectives on Infectious Diseases 1: doi:10.1155/2009/251406 (14pp.).

364

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Rohela M, Jamaiah I, Chan KW, and Wan WY (2002) Diphyllobothriasis: The first case report from Malaysia. Southeast Asian Journal of Tropical Medicine and Public Health 33: 229--230. Rojas L, Vazquez A, Domenech I, and Robertson LJ (2009) Fascioliasis: Can Cuba conquer this emerging parasitosis? Trends in Parasitology 26: 26--34. Rorman E, Zamir CS, Rilkis I, and Ben-David H (2006) Congenital toxoplasmosis – prenatal aspects of Toxoplasma gondii infection. Reproductive Toxicology 21: 458--472. Rose JB (1997) Environmental ecology of Cryptosporidium and public health implications. Annual Reviews of Public Health 18: 135--161. Rose JB, Epstein PR, Lipp EK, Sherman BH, Bernard SM, and Patz JA (2001) Climate variability and change in the United States: Potential impacts on water- and foodborne diseases caused by microbiologic agents. Environmental Health Perspectives 109(supplement 2): 211--220. Ross AGP, Li YS, Sleigh AC, and McManus DP (1997) Schistosomiasis control in the People’s Republic of China. Parasitology Today 13: 152--155. Rossignol JF, Kabil SM, El-Gohary Y, and Younis AM (2007) Nitazoxanide in the treatment of amoebiasis. Transactions of the Royal Society of Tropical Medicine and Hygiene 101: 1025--1031. Rowland M, Rab MA, Freeman T, Durrani N, and Rehman N (2002) Afghan refugees and the temporal and spatial distribution of malaria in Pakistan. Social Science and Medicine 55: 2061--2072. Ro¨delsperger M, Rohmann U, Hambsch B, Baldauf G, and Wricke B (1999) Mo¨glichkeiten der Rohwasserbehandlung zur Inaktivierung von Cryptosporidien und Giardien sowie zur Elimination von Parasiten mit Filtrationsverfahren: Literaturstudie. Karlsruhe: Technologiezentrum Wasser Karlsruhe. Ruditzky MG (1928) Zur Frage u¨ber die Entstehung des Leber- und pankreaskrebses im Zusammenhang mit Distomatose. Journal of Cancer Research and Clinical Oncology 124: 402--416. Ruiz-Tiben E and Hopkins DR (2007) Dracuncuiliais (guinea worm disease) eradication. Advances in Parasitology 61: 276--310. Rumunu J, Brooker S, Hopkins A, Chane F, Emerson P, and Kolaczinski J (2009) Southern Sudan: An opportunity for NTD control and elimination? Trends in Parasitology 25: 301--307. Russell PF (1955) Man’s Mastery of Malaria. Oxford: Oxford University Press. Russell S (2004) The economic burden of illness for households in developing countries: A review of studies focusing on malaria, tuberculosis, and human immunodeficiency virus/acquired immunodeficiency syndrome. American Journal of Tropical Medicine and Hygiene 71(supplement 2): 147--155. Saba R, Korkmaz M, Inan D, et al. (2004) A human fascioliasis. Clinical Microbiology and Infection 10: 385--387. Sachs J and Malaney P (2002) The economic and social burden of malaria. Nature 415: 680--685. Saijuntha W, Sithithaworn P, Wongkham S, et al. (2007) Evidence of a species complex within the food-borne trematode Opisthorchis viverrini, and possible coevolution with their first intermediate hosts. International Journal for Parasitology 37: 695--703. Saijuntha W, Sithithaworn P, Wongkham S, et al. (2008) Mitochondrial DNA sequence variation among geographical isolates of Opisthorchis viverrini in Thailand and Lao PDR, and phylogenetic relationships with other trematodes. Parasitology 135: 1479--1486. Sakanari JA and McKerrow JH (1989) Anisakiasis. Clinical Microbiology Reviews 2: 278--284. Sallares R, Bouwman A, and Anderung C (2004) The spread of malaria to Southern Europe in antiquity: New approaches to old problems. Medical History 48: 311--328. Samie A, Bessong PO, Obi CL, et al. (2006) Cryptosporidium species: Preliminary descriptions of the prevalence and genotype distribution among school children and hospital patients in the Venda region, Limpopo Province, South Africa. Experimental Parasitology 114: 314--322. Sattabongkot J, Tsuboi T, Zollner GE, Sirichaisinthop J, and Cui L (2004) Plasmodium vivax transmission: Chances for control? Trends in Parasitology 20: 192--198. Savioli L, Smith H, and Thompson A (2006) Giardia and Cryptosporidium join the ‘‘Neglected Diseases Initiative’’. Trends in Parasitology 22: 203--208. Sayasone S, Odermatt P, Phoumindr N, et al. (2007) Epidemiology of Opisthorchis viverrini in a rural district of southern Lao PDR. Transactions of the Royal Society of Tropical Medicine and Hygiene 101: 40--47. Scanlon BR, Jolly I, Sophocleous M, and Zhang L (2007) Global impacts of conversions from natural to agricultural ecosystems on water resources: Quantity versus quality. Water Resources Research 43: W03437 (doi: 10.1029/ 2006WR005486). Schaefer FW III, Marshal MM, and Clancy JL (2004) Inactivation and removal of enteric protozoa in water. In: Stirling CR and Adam RD (eds.) The Pathogenic

Enteric Protozoa: Giardia, Entamoeba, Cryptosporidium and Cyclospora, pp. 117--127. Heidelberg: Springer. Schall VT, Vasconcellos MC, Rocha RS, Souza CP, and Mendes NM (2001) The control of the schistosomes-transmitting snail Biomphalaria glabrata by the plant mollucicide Euphorbia splendens var. hislopii (syn milli Des. Moul.): A longitudinal field study in an endemic area in Brazil. Acta Tropica 79: 165--170. Schellenberg D, Menendez C, Kahigwa E, et al. (1999) African children with malaria in an area of intense Plasmodium falciparum transmission: Features on admission to the hospital and risk factors for death. American Journal of Tropical Medicine and Hygiene 61: 431--438. Schipper J, Chanson JS, Chiozza F, et al. (2008) The status of the world’s land and marine mammals: Diversity, threat, and knowledge. Science 322: 225--230. Scholz T, Aguirre-Macedo ML, and Salgado-Maldonado G (2001) Trematodes of the family Heterophyidae (Digenea) in Mexico: A review of species and new host and geographical records. Journal of Natural History 35: 1733--1773. Scholz T, Garcia HH, Kuchta R, and Wicht B (2009) Update on the human broad tapeworm (genus Diphyllobothrium) including clinical relevance. Clinical Microbiological Reviews 22: 146--160. Schuster FL (2002) Cultivation of pathogenic and opportunistic free-living amebas. Clinical Microbiology Reviews 15: 342--354. Schuster FL and Visvesvara GS (2004a) Amebae and ciliated protozoa as causal agents of waterborne zoonotic disease. Veterinary Parasitology 126: 91--120. Schuster FL and Visvesvara GS (2004b) Free-living amoebae as opportunistic and non-opportunistic pathogens of humans and animals. International Journal for Parasitology 34: 1001--1027. Schutte CHJ and Frank GH (1964) Observations on the distribution of freshwater mollusca and chemistry of the natural waters in the south-eastern Transvaal and adjacent northern Swaziland. Bulletin of the World Health Organization 30: 389--400. Schwartz DA (1986) Cholangiocarcinoma associated with liver fluke infection: A preventable source of morbidity in Asian immigrants. American Journal of Gastroenterology 81: 76--79. Schwartz E (2002) Pulmonary schistosomiasis. Clinics in Chest Medicine 23: 433--443. Scott D (1960) An epidemiological note on guinea-worm infection in north-west Ashanti, Ghana. Annals of Tropical Medicine and Parasitology 54: 32--43. Scott D, Senher K, and England EC (1982) Epidemiology of human Schistosoma haematobium infection around Volta lake, Ghana, 1973–1975. Bulletin of the World Health Organisation 60: 89--100. Sournia G (1996) Wildlife tourism in West and Central Africa. Ecodecision 20: 52--54. Seeto RK and Rocky DC (1999) Amebic liver abscess: Epidemiology, clinical features and outcome. Western Journal of Medicine 170: 104--109. Sermsart B, Sripochang S, Suvajeejarum T, and Kiatfuengfoo R (2005) The molluscicidal activities of some Euphorbia milii hybrids against the snail Indoplanorbis exustus. Southeast Asian Journal of Tropical Medicine and Public Health 36(supplement 4): 192--195. Service M (2004) Medical Entomology for Students. Cambridge: Cambridge University Press. Seto EYW, Lee YJ, Liang S, and Zhong B (2007) Individual and village-level study of water contact patterns and Schistosoma japonicum infection in mountainous rural China. Tropical Medicine International Health 12: 1199--1209. Sexton JD (1994) Impregnated bed nets for malaria control: Biological success and social responsibility. American Journal of Tropical Medicine and Hygiene 50: 72--81. Shaw J and Sutcliffe J (2003) Ancient dams, settlement archaeology and Buddhist propagation in central India: The hydrological background. Hydrological Sciences 48: 277--291. Shenoy S, Wilson G, Prashanth HV, Vidyalakshmi K, Dhanashree B, and Bharath R (2002) Primary meningoencephalitis by Naegleria fowleri: First reported case from Mangalore, South India. Journal of Clinical Microbiology 40: 309--310. Shields JM and Olson BH (2003) Cyclospora cayetanensis: A review of an emerging parasitic coccidian. International Journal for Parasitology 33: 371--391. Shiff C (2002) Integrated approach to malaria control. Clinical Microbiology Reviews 15: 278--293. Silamut K, Phu NH, Whitty C, et al. (1999) A quantitative analysis of the microvascular sequestration of malaria parasites in the human brain. American Journal of Pathology 155: 395--410. Simonishvili S, Tsanava S, Sanadze K, et al. (2005) Entamoeba histolytica: The serinerich gene polymorphism-based genetic variability of clinical isolates from Georgia. Experimental Parasitology 110: 313--317. Simonsen PE, Lemnge MM, Msangeni HA, Jakobsen PH, and Bygbjerg (1996) Bancroftian filariasis: The patterns of filarial-specific immunoglobulin G1 (IgG1),

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

IgG4, and circulating antigens in an endemic community of northeastern Tanzania. American Journal of Tropical Medicine and Hygiene 55: 69--75. Simpson VR (2002) Wild animals as reservoirs of infectious diseases in the UK. Veterinary Journal 163: 128--146. Singh B, Kim Sung L, Matusop A, et al. (2004) A large focus of naturally acquired Plasmodium knowlesi infections in human beings. Lancet 363: 1017--1024. Sithithaworn P and Haswell-Elkins M (2003) Epidemiology of Opisthorchis viverrini. Acta Tropica 88: 187--194. Sithithaworn P, Sukavat K, Vannachone B, et al. (2006) Epidemiology of food-borne trematodes and other parasite infections in a fishing community on the Nam Ngum reservoir, Lao PDR. Southeast Asian Journal of Tropical Medicine and Public Health 37: 1083--1090. Sithithaworn P, Yongvanit P, Tesana S, and Pairojkul C (2007) Liver flukes. In: Murrell KD and Fried B (eds.) World-Class Parasites, Vol. 11. Food-Borne Parasitic Zoonoses: Fish and Plant-Borne Parasites, pp. 3--52. Heidelberg: Springer. Slifko TR, Smith HW, and Rose JB (2000) Emerging parasite zoonoses associated with water and food. International Journal for Parasitology 30: 1379--1393. Smith JE, Jr. and Perdek JM (2004) Assessment and management of watershed microbial contaminants. Critical Reviews in Environmental Science and Technology 34: 109--139. Smith JR and Cunningham ET, Jr. (2002) Atypical presentations of ocular toxoplasmosis. Current Opinion in Ophthalmology 13: 387--392. Smith JW (1983) Anisakis simplex (Rudolphi, 1809, det. Krabbe, 1878) (Nematoda: Anisakidoidea): Morphology and morphometry of larvae from euphausiids and fish, and a review of the life history and ecology. Journal of Helminthology 57: 205--224. Snounou G and White NJ (2004) The co-existence of Plasmodium: Sidelights from falciparum and vivax malaria in Thailand. Trends in Parasitology 20: 333--339. Soba B and Logar J (2008) Genetic classification of Cryptosporidium isolates from humans and calves in Slovenia. Parasitology 135: 1263--1270. Soh CT (1962) Paragonimiasis in Korea: Biological aspects. Yonsei Medical Journal 3: 79--84. Sohn WM and Seol SY (1994) A human case of gastric anisakiasis by Pseudoterranova decipiens larva. Korean Journal of Parasitology 32: 53--56. Sommerville C (2006) The pathology of Haplorchis pumilio (Looss, 1896) infections in cultured tilapias. Journal of Fish Diseases 5: 243--250. Sornmani S, Schlep FP, Vivatanasesth P, et al. (1984) A pilot project for controlling O. viverrini infection in Nong Wai, Northeast Thailand, by applying praziquantel and other measures. Arzeimittelforschung 34: 1231--1234. Sre´ter T and Sze´ll Z (2008) Onchocerciasis: A newly recognized disease in dogs. Veterinary Parasitology 151: 1--13. Sre´ter T and Varga I (2000) Cryptosporidiosis in birds – a review. Veterinary Parasitology 87: 261--279. Sriamporn S, Pisani P, Pipitgool V, Suwanrungruang K, Kamsa-ard S, and Parkin DM (2004) Prevalence of Opisthorchis viverrini infection and incidence of cholangiocarcinoma in Khon Kaen, Northeast Thailand. Tropical Medicine and International Health 9: 588--594. Srikanth R and Naik D (2004) Prevalence of giardiasis due to wastewater reuse for agriculture in the suburbs of Asmara City, Eritrea. International Journal of Environmental Health Research 14: 43--52. Sripa B (2003) Pathobiology of opisthorchiasis: An update. Acta Tropica 88: 209--220. Stabler SP and Allen RH (2004) Vitamin B12 deficiency as a worldwide problem. Annual Review of Nutrition 24: 299--326. Stauffer W and Ravdin JI (2003) Entamoeba histolytica: An update. Current Opinion in Infectious Diseases 16: 479--485. Steib K (1987) Der Einfluss Klimatologischer und O¨kologischer Faktoren auf das Auftreten der Dracunculose. In: Fricke W and Hinz E (eds.) Ra¨umliche Persistenz und Diffusion von Krankheiten, pp. 114--126. Frankfurt am Main: Peter Lang. Steib K and Mayer P (1988) Epidemiology and vectors of Dracunculus medinensis in northwest Burkina Faso, West Africa. Annals of Tropical Medicine and Parasitology 82: 189--199. Steinmann P, Keiser J, Bos R, Tanner M, and Utzinger J (2006) Schistosomiasis and water resources development: Systematic review, meta-analysis and estimates of people at risk. Lancet Infectious Diseases 6: 411--425. Steketee RW, Nahlen BL, Parise ME, and Mendez C (2001) The burden of malaria in pregnancy in malaria-endemic areas. American Journal of Tropical Medicine and Hygiene 64: 28--35. Stenseth NC, Mysterud A, Ottersen G, Hurrell JW, Chan KS, and Lima M (2002) Ecological effects of climate fluctuation. Science 297: 1292--1296. Stilwell B, Diallo K, Zurn P, Vucicic M, Adams O, and Dal Poz M (2004) Migration of health-care workers from developing countries: Strategic approaches to its management. Bulletin of the World Health Organization 82: 595--600.

365

Strandgaard H, Johansen MV, Aagaard-Hansen J, Petlueng P, and Ornbjerg N (2008) Local perceptions and practices in regard to opisthorchiasis in two villages in Lao PDR. Southeast Asian Journal of Tropical Medicine and Public Health 39: 19--26. Straub TM, Pepper IL, and Gerba CP (1993) Hazards from pathogenic microorganisms in land-disposed sewage sludge. Review of Environmental Contamination and Toxicology 132: 55--91. Streefland P (2005) Public health care under pressure in Sub-Saharan Africa. Health Policy 71: 375--382. Stuart JM, Orr HJ, Warburton FG, et al. (2003) Risk factors for sporadic giardiasis: A case study in southwest England. Emerging Infectious Diseases 9: 229--233. Subasinghe R, Soto D, and Jia J (2009) Global aquaculture and its role in sustainable development. Reviews in Aquaculture 1: 2--9. Subra R (1981) Biology and control of Culex pipiens quinquefasciatus Say, 1823 (Diptera, Culicidae) with special reference to Africa. Insect Science and Application 1: 319--338. Sukontason K, Unpunyo P, Sukontason KL, and Piangjai S (2005) Evidence of Haplorchis taichui infection as pathogenic parasite: Three case reports. Scandinavian Journal of Infectious Diseases 37: 388--390. Sullivan PB, Marsh MN, and Phillips MB (1991) Prevalence and treatment of giardiasis in chronic diarrhoea and malnutrition. Archives of Disease in Childhood 66: 304--306. Suon S, Hol D, Siek S, McLean M, and Copeman B (2006) Seasonal differences in the incidence of infection with Fasciola gigantica in Cambodian cattle. Tropical Animal Health and Production 38: 23--28. Sutherst RW (2004) Global change and human vulnerability to vector-borne diseases. Clinical Microbiology Reviews 17: 136--173. Takagi Y, Kashiyama T, Sakashita K, et al. (2009) A cluster of paragonimiasis westermani infections in Thai immigrants (in Japanese). Nihon Kokyuki Gakkai Zasshi 47: 249--253. Tanyuksel M and Petri WA, Jr. (2003) Laboratory diagnosis of amebiasis. Clinical Microbiology Reviews 16: 713--729. Tanyuksel M, Ulukanligil M, Guclu Z, Araz E, Koru O, and Petri WA, Jr. (2007) Two cases of rarely recognized infection with Entamoeba moshkovskii. American Journal of Tropical Medicine and Hygiene 76: 723--724. Taraschewski H (1984) Heterophyiasis, an intestinal fluke infection of man and vertebrates transmitted by euryhaline gastropods and fish. Helgola¨nder Meeresuntersuchungen 37: 463--478. Taraschewski H (2006) Hosts and parasites as aliens. Journal of Helminthology 80: 99--129. Tavares LER, Luque JL, and Do Bomfim TCB (2005) Human diphyllobothriasis: Reports from Rio de Janeiro, Brazil. Revista Brasileira de Parasitologia Veterinaria 14: 85--87. Taye A, Gebre-Michael T, and Taticheff S (2000) Onchocerciasis in Gilgel Ghibe River valley southwest Ethiopia. East African Medical Journal 77: 116--120. Tayeh A and Cairncross S (1998) The effect of size of surface drinking water sources on dracunculiasis prevalence in the Northern Region of Ghana. International Journal of Enviromental Health Research 8: 285--292. Taylor MG, Nelson GS, and Andrews BJ (1972) A case of natural infection of S. haematobium in a Senegalese baboon (Papio sp.). Transactions of the Royal Society of Tropical Medicine and Hygiene 66: 16--17. Taylor MJ, Awadzi K, Basa´n˜ez MG, et al. (2009) Onchocerciasis control: Vision for the future from a Ghanian perspective. Parasites and Vectors 2: 1--7. Teknos TN, Poulin MD, Laruentano AM, and Li KK (2000) Acanthamoeba rhinosinusitis: Characterization, diagnosis and treatment. American Journal of Rhinology 14: 387--391. Tenter AM, Heckeroth AR, and Weiss LM (2000) Toxoplasma gondii: From animals to humans. International Journal for Parasitology 30: 1217--1258. Terramocci R, Pagani L, Brunati P, Gatti S, Bernuzzi A, and Scaglia M (2001) Reappearance of human diphyllobothriasis in a limited area of Lake Como, Italy. Infection 29: 93--95. Thien CP, Dalsgaard A, Nhan NT, Olsen A, and Murrell KD (2009) Prevalence of zoonotic trematode parasites in fish fry and juveniles in fish farms of the Mekong Delta, Vietnam. Aquaculture 295: 1--5. Thien PC, Dalsgaard A, Thangh BN, Olsen A, and Murrell KD (2007) Prevalence of fishborne zoonotic parasites in important cultured fish species in the Mekong delta, Vietnam. Parasitology Research 101: 1277--1284. Thimasarn K, Jatapadma S, Vijaykada S, Sirichaisinthop J, and Wongsrichanalai C (1995) Epidemiology of malaria in Thailand. Travel Medicine 2: 59--65. Thompson RCA (2000) Giardiasis as a re-emerging infectious disease and its zoonotic potential. International Journal for Parasitology 30: 1259--1267. Thompson RCA (2004) The zoonotic significance and molecular epidemiology of Giardia and giardiasis. Veterinary Parasitology 126: 15--35.

366

Waterborne Parasitic Diseases: Hydrology, Regional Development, and Control

Thompson RCA, Palmer CS, and O’Handley R (2008) The public health and clinical significance of Giardia and Cryptosporidium in domestic animals. Veterinary Journal 177: 18--25. Thomson MC and Connor SJ (2001) The development of malaria early warning systems for Africa. Trends in Parasitology 17: 438--445. Thu ND, Dalsgaard A, Loan LTT, and Murrell KD (2007) Survey for zoonotic liver and intestinal trematode metacercariae in cultured and wild fish in An Giang province, Vietnam. Korean Journal of Parasitology 45: 45--54. Thurston-Enriquez JA, Watt P, Dowd SE, Enriquez R, Pepper IL, and Gerba CP (2002) Detection of protozoan parasites and microsporidia in irrigation waters used for crop production. Journal of Food Protection 65: 378--382. Tidwell JH and Allan GL (2001) Fish as food: Aquaculture’s contribution. EMBO Reports 2: 958--963. Tiewcharoen S and Junnu V (2001) Distribution of pathogenic Naegleria spp. in Thailand. Southeast Asian Journal of Tropical Medicine and Public Health 32(supplement 2): 172--178. Tishkoff SA and Verrelli BC (2003) Patterns of human genetic diversity: Implications for human evolutionary history and disease. Annual Review of Genomics and Human Genetics 4: 293--340. Todryk SM and Hill AVS (2007) Malaria vaccines: The stage we are at. Nature Reviews Microbiology 5: 487--489. Torchin ME and Mitchell C (2004) Parasites, pathogens, and invasion by plants and animals. Frontiers in Ecology and the Environment 2: 183--190. Torno MS, Babapour R, Gurevitch A, and Witt MD (2000) Cutaneous acanthamoebiasis in AIDS. Journal of the American Academy of Dermatology 42: 351--354. Torres P, Cuevas C, and Tang M (2004a) Introduced and native fishes as infection foci of Diphyllobothrium spp. in humans and dogs from two localities at Lake Panguipulli in southern Chile. Comparative Parasitology 71: 111--117. Torres P, Villalobos L, Woelfl S, and Puga S (2004b) Identification of the copepod intermediate host of the introduced broad fish tapeworm, Diphyllobothrium latum, in southern Chile. Parasitology Research 90: 1190--1193. Toset HPW, Gleditsch NP, and Hegre H (2000) Shared rivers and interstate conflict. Political Geography 9: 971--996. Trampuz A, Jereb M, Muzlovic I, and Prabhu RM (2003) Clinical review: Severe malaria. Critical Care 7: 315--323. Trape JF, Pison G, Spiegel A, Enel C, and Rogier C (2002) Combating malaria in Africa. Trends in Parasitology 18: 224--230. Traub JR, Monis PT, Robertson I, Irwin P, Menke N, and Thompson RCA (2004) Epidemiological and molecular evidence supports the zoonotic transmission of Giardia among humans and dogs living in the same community. Parasitology 128: 253--262. Tsy JMLP, Duchemin JB, Marrama L, et al. (2003) Distribution of the species of the Anopheles gambiae complex and first evidence of Anopheles merus as a malaria vector in Madagascar. Malaria Journal 2: 33. Tulchinsky TH, Burla E, Clayman M, Sadik C, Brown A, and Goldberger S (2000) Safety of community drinking-water and outbreaks of waterborne enteric disease: Israel 1976–97. Bulletin of the World Health Organization 78: 1466--1473. Tum S, Puotinen ML, and Copeman DB (2004) A geographic information systems model for mapping risk of fascioliasis in cattle and buffaloes in Cambodia. Veterinary Parasitology 122: 141--149. Tumwine JK, Kekitiinwa A, Nabukeera N, et al. (2003) Cryptosporidium parvum in children with diarrhea in Mulago Hospital, Kampala, Uganda. American Journal of Tropical Medicine and Hygiene 68: 710--715. Tzipori S and Ward H (2002) Cryptosporidiosis: Biology, pathogenesis and disease. Microbes and Infection 4: 1047--1058. Udall DN (2007) Recent updates on onchocerciasis: Diagnosis and treatment. Clinical Infectious Diseases 44: 53--60. Uni S, Bain O, Takaoka H, Miyashita M, and Suzuki Y (2001) Onchocerca dewittei japonica n. subsp., a common parasite from wild boar in Kyushu Island, Japan. Parasite 8: 215--222. United Nations Children’s Fund and World Health Organization (2004) Target. Joint Monitoring Programme for Water Supply and Sanitation: Meeting the MDG Drinking Water and Saitation Target: A Mid-Term Assessment of Progress, 2004. Geneva: World Health Organization and United Nations Children’s Fund. Upatham ES, Viyanant V, Brockelman WY, Kurathong S, Lee P, and Kraengraeng R (1988) Rate of re-infection by Opisthorchis viverrini in an endemic northeast Thai community after chemotherapy. International Journal for Parasitology 18: 643--649. Urbani C, Sinoun M, Socheat D, et al. (2002) Epidemiology and control of Mekongi schistosomiasis. Acta Tropica 82: 157--168. Uriarte J, Cabaret J, and Tanco JA (1985) The distribution and abundance of parasitic infections in sheep grazing on irrigated or on non-irrigated pastures in Northeastern Spain. Annales de Recherches Veterinaires 16: 321--325.

Utzinger J, Mayombana C, Mez K, and Tanner M (1997) Evaluation of chemical and physical–morphological factors as potential determinants of Biomphalaria pfeifferi (Krauss, 1848) distribution. Memorias do Instituto Oswaldo Cruz 92: 323--328. Vatanasapt V, Sriamporn S, and Vatanasapt P (2002) Cancer control in Thailand. Japanese Journal of Clinical Oncology 32(supplement 1): 82--91. Vatsal D, Kapoor S, Venkatesh V, Vatsal P, and Husain N (2006) Ectopic fascioliasis in the dorsal spine: Case report. Neurosurgery 59: 706--707. Verweij JJ, Vermeer J, Brienen EA, et al. (2003) Entamoeba histolytica infections in captive primates. Parasitology Research 90: 100--103. Ve´lez I, Ortega J, Hurtado M, et al. (2000) Epidemiology of paragonimiasis in Colombia. Transactions of the Royal Society of Tropical Medicine and Hygiene 94: 661--663. Viera JC, Blankespoor HD, Cooper PJ, and Guderian RH (1992) Paragonimiasis in Ecuador: Prevalence and geographical distribution of parasitisation of second intermediate hosts with Paragonimus mexicanus in Esmeraldas province. Tropical Medicine and Parasitology 43: 249--252. Visvesvara GS, Booton GC, Kelley DJ, et al. (2007a) In vitro culture, serologic and molecular analysis of Acanthamoeba isolated from the liver of a keel-billed toucan (Ramphastos sulfuratus). Veterinary Parasitology 143: 74--78. Visvesvara GS, Moura H, and Schuster FL (2007b) Pathologic and opportunistic free-living amoebae: Acanthamoeba spp. Balamuthia mandrillaris, Naegleria fowleri, and Sappinia diploidea. FEMS Immunology and Medical Microbiology 50: 1--26. Vittor AY, Pan W, Gilman RH, et al. (2009) Linking deforestation to malaria in the Amazon: Characterization of the breeding habitat of the principal malaria vector, Anopheles darlingi. American Journal of Tropical Medicine and Hygiene 81: 5--12. Vo¨ro¨smarty CJ and Sahagian D (2000) Anthropogenic disturbance of the water cycle. BioScience 50: 753--765. Waldman ADB, Day JH, Shaw P, and Bryceson ADM (2001) Subacute pulmonary granulomatous schistosomiasis: High resolution CT appearances – another cause of the halo sign. British Journal of Radiology 74: 1052--1055. Walker K and Lynch M (2007) Contributions of Anopheles larval control to malaria suppression in tropical Africa: Review of achievements and potential. Medical and Veterinary Entomology 21: 2--21. Wallis PM, Erlandsen SL, Isaac-Renton JL, Olson ME, Robertson WJ, and van Keulen H (1996) Prevalence of Giardia cysts and Cryptospiridium oocysts and characterization of Giardia spp. isolated from drinking water in Canada. Applied and Environmental Microbiology 62: 2789--2797. Walsh JF, Molyneux DH, and Birley MH (1993) Deforestation: Effects on vector-borne disease. Parasitology 106(supplement): S55--S75. Wang T, Zhang S, Wu W, et al. (2006) Treatment and reinfection of water buffaloes and cattle infected with Schistosoma japonicum in Yangtze River Valley, Anhui Province, China. Journal of Parasitology 92: 1088--1091. Wang WJ, Xin YJ, Robinson NL, Ting HW, Ni C, and Kuo PK (1984) Intraocular paragonimiasis. British Journal of Ophthalmology 68: 85--88. Ward PI, DePlazes P, Regli W, Rinder H, and Mathis A (2002) Detection of eight Cryptosporidium genotypes in surface and waste waters in Europe. Parasitology 124: 359--368. Watanapa P and Watanapa WB (2002) Liver fluke-associated cholangiocarcinoma. British Journal of Surgery 89: 962--970. Watts S and El Katsha S (1997) Irrigation, farming and schistosomiasis: A case study in the Nile Delta. International Journal of Environmental Health Research 7: 101--113. Webster BL, Southgate VR, and Littlewood DTJ (2006) A revision of the interrelationship of Schistosoma including the recently described Schistosoma guineensis. International Journal of Parasitology 36: 947--955. Webster WA and Dukes TW (1978) Bovine and equine onchocerciasis in Eastern North America with a discussion on cuticular morphology of Onchocerca spp. in cattle. Canadian Journal of Comparative Medicine 43: 330--332. Weil GJ, Ramzy RMR, El Setouhy M, Kandil AM, Ahmed ES, and Faris R (1999) A longitudinal study of bancroftian filariasis in the Nile Delta of Egypt: Baseline data and one-year follow-up. American Journal of Tropical Medicine and Hygiene 61: 53--58. White NJ (2004) Antimalarial drug resistance. Journal of Clinical Investigations 113: 1084--1092. Williams NV (1970) Studies on aquatic pulmonate snails in Central Africa. I: Field distribution in relation to water chemistry. Malacologia 10: 153--164. Wilson MD, Cheke RA, Fiasse SPJ, et al. (2002) Deforestation and the spatial– temporal distribution of savannah and forest members of the Simulium damnosum complex in southern Ghana and south-western Togo. Transactions of the Royal Society of Tropical Medicine and Hygiene 96: 632--639. Witt C and Ottesen EA (2001) Lymphatic filariasis: An infection of childhood. Tropical Medicine and International Health 6: 582--606.

3.13 Bioremediation: Plasmid-Mediated Bioaugmentation of Microbial Communities – Experience from Laboratory-Scale Bioreactors M Hausner and M Starek, Ryerson University, Toronto, ON, Canada S Bathe, Studienstiftung des deutschen Volkes, Bonn, Germany & 2011 Elsevier B.V. All rights reserved.

3.13.1 3.13.1.1 3.13.1.2 3.13.1.3 3.13.1.4

Horizontal Gene Transfer-Mediated Bioaugmentation Microbial Aggregates Xenobiotics in Aquatic Environments Role of Catabolic Genes in Bioremediation Monitoring of Gene Transfer: Detection of Donor Strains and Emerging Transconjugants Using Culture-Independent Techniques 3.13.1.5 Objectives 3.13.2 Plasmid pWW0 3.13.2.1 Characteristics and Host Range 3.13.2.2 Plasmid pWW0 Transfer to Indigenous Microorganisms in Wastewater or Activated Sludge 3.13.2.3 Introduction of Plasmid pWW0 to Groundwater-Derived Microbial Communities 3.13.2.4 Utilization of Plasmid pWW0 for Bioaugmentation Purposes 3.13.3 Plasmid pJP4 3.13.3.1 Characteristics and Host Range 3.13.3.2 Plasmid pJP4 Transfer to Indigenous Microorganisms in Wastewater or Activated Sludge 3.13.3.3 Utilization of Plasmid pJP4 for Bioaugmentation Purposes 3.13.4 Plasmid pNB2 3.13.4.1 Characteristics and Host Range 3.13.4.2 Plasmid pNB2 Transfer to Indigenous Microorganisms in Wastewater or Activated Sludge 3.13.4.3 Utilization of Plasmid pNB2 for Bioaugmentation Purposes 3.13.5 Conclusions and Recommendations Acknowledgments References

3.13.1 Horizontal Gene Transfer-Mediated Bioaugmentation 3.13.1.1 Microbial Aggregates In natural environments, as much as 99% of microorganims are found in biofilms (Dalton and March, 1998) or other bioaggregates (Adav et al., 2010). Bioaggregates, such as activated sludge flocs, aerobic and anaerobic biofilms, microbial mats or marine snow, and aerobic or anaerobic granules, can be defined as accumulations of microbes embedded in extracellular polymeric substance (EPS) (Adav et al., 2010). EPSs are biopolymers of microbial origin consisting of polysaccharides, a wide variety of proteins, glycoproteins, glycolipids, and in some cases, extracellular DNA (e-DNA) (Flemming et al., 2007). From an ecological point of view, bioaggregates can be broadly subdivided based on their location relative to the surrounding environment. Bioaggregates can be associated with a surface (biofilms), can float at a liquid–air interfaces (microbial mats), or can be free-floating in a liquid in the form of microbial flocs. There exists a special type of bioaggregate referred to as granular sludge, with a minimum size of 200 mm (de Kreuk et al., 2007). These granular bioaggregates differ from microbial flocs in that they settle faster and do not coagulate with each other (de Kreuk et al., 2007).

369 369 369 369 370 370 371 371 371 371 372 372 372 372 373 373 373 373 375 375 375 375

3.13.1.2 Xenobiotics in Aquatic Environments Bioremediation, microbially mediated degradation or detoxification of contaminants, shows potential for remediation of contaminated aquatic environments. Many xenobiotics become associated with bioaggregates (Bouwer, 1989; Wolfaardt et al., 1994) where they exert selective pressure on attached microbial communities. Given sufficient time, adaptation of microorganisms to the contaminant usually results. Microorganisms acquire the ability to survive in the presence of xenobiotics or to even utilize contaminants as carbon sources by one or a combination of the following mechanisms: (1) an increase in the numbers of specific degraders, (2) microbial community adaptation through mutations, or (3) acquisition of relevant genetic information by the community through horizontal gene transfer (HGT), leading to an eventual increase in the community’s biodegradation potential (Top et al., 2002; Top and Springael, 2003). By definition, bioaggregates are likely to possess microorganisms positioned in close proximity and are thus ideal environments for the occurrence of HGT (Wuertz et al., 2004). It is in this vein of thought that natural attenuation of environmental contaminants can be anthropologically altered. Practically speaking, it would be a difficult task to perform site-directed mutagenesis to a population of microorganisms in a natural environment with a high degree of precision, such that the

369

370

Bioremediation: Plasmid-Mediated Bioaugmentation of Microbial Communities

population would be able to degrade the contaminant in question. The addition of nutrients to an environment with the intension of stimulating the growth of specific degraders is perhaps a more practical avenue of procedure, and is commonly referred to as biostimulation (Pandey et al., 2008). The approach of biostimulation, in essence, depends on the existence of a bacterial strain with specific metabolic capabilities in the vicinity of the contaminated area. The addition of new genetic capabilities, either chromosomally or plasmid encoded, can circumvent this caveat to biostimulation. This approach is referred to as bioaugmentation and it can initiate and accelerate bioremediation (Van Limbergen et al., 1998).

3.13.1.3 Role of Catabolic Genes in Bioremediation In general, genes encoding enzymes active in degradative pathways and genes encoding antibiotic or metal resistance are often located on plasmids or other mobile genetic elements such as transposons (Top et al., 2002). Plasmid transfer via conjugation (exchange of genetic information from one cell to another mediated by cell-to-cell contact) has been frequently detected in many environments, including soils, activated sludge, sediments, the rhizosphere and phyllosphere of plants, and model, engineered and natural biofilms. Bioaugmentation can be accomplished through the addition of a bacterial strain or a mixed culture with the required metabolic properties (i.e., the capability to degrade a relevant pollutant) to the indigenous microbial community (Kasai et al., 2007; Dabert et al., 2005). Nevertheless, previous work has shown that strains cultured in the lab under optimal conditions and then introduced to existing microbial communities may not survive under the new environmental

conditions and eventually are outcompeted to levels below detection limits (Bouchez et al., 2000a, 2000b; Tchelet et al., 1999). Interestingly, bioaugmentation aimed at improving nitrogen removal from wastewater has been associated with process inhibition (Bouchez et al., 2000a, 2000b), demonstrated by nitrification breakdown upon addition of a denitrifying microorganism. The authors in that particular study attributed the rapid decline in denitrifying microorganisms to grazing by higher-order protozoan ciliates. On the other hand, bioaugmentation via in situ genetic manipulation (the introduction of catabolic genes into an existing indigenous community by means of HGT via conjugation or transformation) may result in a lasting presence of the introduced degradative genes in an existing microbial community (Bathe, 2004; Bathe et al., 2004b, 2005, 2009). In this manner, catabolic enzymes encoded by genes on mobile genetic elements carried by an appropriate bacterial host are introduced to indigenous populations (Figure 1). Previous studies have demonstrated that conjugative catabolic plasmids of the incompatibility group P1 are frequently transferred to a broad range of proteobacterial recipients (Bathe et al., 2004a). In engineered systems such as wastewater treatment plants, cell densities and growth rates are usually high and the impact of HGT may be improved by a rapid division of transconjugants passing their newly acquired genetic information to the subsequent generations of bacteria (Wuertz et al., 2004). Similarly, transconjugants may serve as donor strains and enhance the spread of introduced genes throughout the indigenous microbial communities. Conjugation is likely to play a major role in spreading genetic information encoded on plasmids and can be exploited in bioaugmentation. Transfer of catabolic plasmids has been previously studied in model aerobic suspended growth microcosms (Bathe, 2004; Bathe et al.,

Reactor

Bacteria in wastewater Donor

(a)

Chromosomal DNA Contaminant molecule Degradative plasmid Single strand of plasmid DNA during conjugative plasmid transfer Transconjugant

(b)

Figure 1 Diagrammatic representation of plasmid-mediated bioaugmentation. (a) A normal (not bioaugmented) reactor where resident microorganisms degrade a portion of the contaminants present in the wastewater. (b) A bioaugmented reactor: a donor strain harboring a degradative plasmid is introduced into the reactor and subsequently transfers that plasmid via conjugation to resident bacteria, which become transconjugants and potential new plasmid donors. This results in the spread of the degradative plasmid throughout the existing microbial community, and therefore increased degradation of the contaminant.

Bioremediation: Plasmid-Mediated Bioaugmentation of Microbial Communities

2004b, 2005; Boon et al., 2000; Molin and Tolker-Nielsen, 2003) as well as in soils (Dejonghe et al., 2000; Newby et al., 2000a, 2000b; Pepper et al., 2002).

3.13.1.4 Monitoring of Gene Transfer: Detection of Donor Strains and Emerging Transconjugants Using Culture-Independent Techniques In order to interpret the success or failure of gene-mediated bioaugmentation, it is useful to monitor the survival or potential disappearance of the introduced strain and/or the emergence of transconjugants (upon successful conjugation) coupled to the performance of the reactor. Since not all transconjugants can be cultivated (Bathe et al., 2004b), cultivation-independent in situ techniques which employ fluorescent marker genes together with microscopy (Sorensen et al., 2005) are useful. Fluorescent proteins such as the green fluorescent protein (GFP) from Aequorea victoria (Chalfie et al., 1994; Tsien, 1998) or the red fluorescent protein DsRed (drFP583, commercially available as DsRed) isolated from a Discosoma sp. coral (Yarbrough et al., 2001) have been used as markers for in situ monitoring of plasmid transfer (Bathe et al., 2004b, 2005; Boon et al., 2000; Nancharaiah et al., 2003) in different environments, including agar surfaces, the phylloplane, microbial wastewater communities, or biofilms (Bathe et al., 2004b; Christensen et al., 1998; Normander et al., 1998).

3.13.1.5 Objectives In this chapter, we summarize our experience with bioaugmentation of microbial communities using donor strains carrying one of the following catabolic plasmids: plasmid pNB2 (Bathe, 2004; Bathe et al., 2005, 2009) which encodes genes necessary for the 3-chloroaniline (3-CA) degradation pathway, plasmid pJP4 (Bathe et al., 2004a, 2004b), encoding genes for the degradation of 2,4-dichlorophenoxyacetic acid (2,4-D), and the TOL plasmid pWW0 (Nancharaiah et al., 2003, 2008; Venkata Mohan et al., 2009), which carries genes encoding enzymes for toluene degradation and other related compounds. We will emphasize wastewater environments and, where appropriate, include examples from other microbial habitats.

3.13.2 Plasmid pWW0 3.13.2.1 Characteristics and Host Range There are various incompatibility group 9 (IncP-9) TOL plasmids which have been isolated; the most used in laboratory settings is plasmid pWW0. The conjugative IncP-9 TOL plasmid pWW0, initially isolated from Pseudomonas putida-mt2 is 116 580 bp in size (Greated et al., 2002; Williams and Murray, 1974; Wong and Dunn, 1974; GenBank Accession No. AJ344068) and contains genes that encode enzymes for the degradation of toluene/xylene in addition to containing, and constitutively expressing, genes which are necessary for its transfer from host to recipient (Greated et al., 2002; Worsey and Williams, 1975). Plasmid pWW0 seems primarily to transfer among the Pseudomonas species. However, other species, including Escherichia coli, Erwinia chrysanthemi, Hydrogenophaga palleronii, Serratia, and Burkholderia species, have

371

been shown to be recipients of pWW0 (Benson and Shapiro, 1978; Nakazawa et al., 1978; Nancharaiah et al., 2003; Ramos-Gonzalez et al., 1991; Sarand et al., 2000). However, the transfer of pWW0 depends not only on the donor strains and recipient pool but also on environmental factors such as nutrient concentration and temperature (Johnsen and Kroer, 2007; Fox et al., 2008). Conjugative transfer of plasmid pWW0 was accomplished efficiently on plates (Christensen et al., 1996), in biofilm or bioaggregates (Christensen et al., 1998; Nancharaiah et al., 2003), soils (Sarand et al., 2000), on bush bean leaves (Normander et al., 1998), and in activated sludge communities (Nancharaiah et al., 2003, 2008; Venkata Mohan et al., 2009). Previous studies demonstrated that transfer rates of pWW0 increase with donor growth rate (Smets et al., 1993) and substrate concentration (Smets et al., 1995). The transfer frequency for pWW0 can be as high as one transconjugant per donor if the conditions for the bacteria are optimal (Ramos-Gonzalez et al., 1991).

3.13.2.2 Plasmid pWW0 Transfer to Indigenous Microorganisms in Wastewater or Activated Sludge In our research, we use a green-fluorescent protein or redfluorescent protein labeled pWW0 for bioaugmentation studies. In this way, we can detect both donor cells and transconjugants microscopically. Previously, we investigated the transfer of pWW0 from its P. putida KT2442 donor to indigenous wastewater microbial communities (Nancharaiah et al., 2003; Venkata Mohan et al., 2009) in a laboratory scale (total volume of 1.6 l and a working volume of 1 l) and a pilot scale (total volume of 38 l and a working volume of 21 l) sequencing batch biofilm reactor (SBBR), using glass or clay beads, respectively, as carrier material for biofilm development. The laboratory-scale SBBR had a total volume of 1.6 l and a working volume of 1 l. The pilot scale SBBR had a total volume of 38 l and a working volume of 21 l. We showed that pWW0 was transferred to indigenous recipients in the laboratory-scale bioreactor with benzyl alcohol (BA) as the sole carbon source. We demonstrated that donor cells survived in the laboratory-scale bioreactor throughout the experimental period of 32 days. Concurrently, we were able to detect emerging transconjugants using confocal laser scanning microscopy (CLSM). The degradation rate of BA was enhanced from 0.98 mg BA min 1 prior to inoculation to 1.9 mg BA min 1 after donor inoculation. In contrast, in the pilot-scale bioreactor, donor cells disappeared 84 h after inoculation while transconjugants were not detected at all. The performance of the pilot-scale reactor with respect to BA removal was not affected by donor addition. Thus, the survival of a bioaugmented strain, conjugative plasmid transfer, and enhanced BA degradation was demonstrated in the laboratoryscale SBBR but not in the pilot-scale SBBR. We attributed the failure of bioaugmentation in the pilot-scale SBBR to insufficient selection pressure, since parallel experiments showed complete removal of BA even prior to donor addition. In another study (Nancharaiah et al., 2008), we showed successful bioaugmentation of aerobic granular sludge using P. putida KT2442 cells bearing plasmid pWW0. The study revealed both donor integration and transconjugant

372

Bioremediation: Plasmid-Mediated Bioaugmentation of Microbial Communities

proliferation, indicating successful plasmid transfer to indigenous microbial communities in the granules, coupled to a significant increase in degradation of BA (used as sole carbon source). In contrast, control microcosms (with non-augmented granules) did not show any noticeable increase in BA degradation. This study showed that bioaugmentation of aerobic granular sludge via donor colonization and plasmid transfer is feasible for enhanced biodegradation. A recent study by Pei and Gunsch (2009) investigated the transfer of pWW0 from P. putida BBC443 to microbial communities in batch reactors inoculated with activated sludge from a municipal wastewater treatment plant. The results confirmed plasmid transfer to recipient cells. A higher initial concentration of activated sludge cells (potential recipients) was associated with the highest number of conjugation events. However, successful gene transfer was not coupled to an increase in toluene degradation rates, perhaps due to the presence of a high number of degraders originally present in the activated sludge inoculum.

or dependent on yet-undefined environmental factors. In the case of wastewater communities, donor cell persistence and/or plasmid transfer was linked to enhanced BA degradation. However, it could not be differentiated whether accelerated degradation of BA was linked to the persistence of donor cells, the emergence of transconjugants or both. On the other hand, no enhancement of toluene degradation was observed upon successful transfer of pWW0 to activated sludge communities in batch cultures (Pei and Gunsch, 2009). Similarly, the addition of a pWW0-bearing donor strain to simulated rockfracture aquifers did not result in enhanced toluene or BA degradation. Therefore, the factors governing the success of pWW0-based bioaugmentation of microbial communities for enhanced toluene degradation (e.g., donor cell:recipient cell ratio, size of aggregates harboring potential recipients, and effect of toluene concentration) need to be further investigated.

3.13.3 Plasmid pJP4 3.13.2.3 Introduction of Plasmid pWW0 to Groundwater-Derived Microbial Communities While bioaugmentation via HGT has been successful in wastewater remediation, as well as in some applications to subsurface microbial communities (Smets et al., 2003) or to endogenous endophytic microbial communities from poplar to improve groundwater contaminated with organic solvents (Taghavi et al., 2005), a current study in our lab on the applicability of plasmid pWW0-mediated enhanced degradation of toluene or BA has not yielded similar results in a model of groundwater flow in a rock-fracture aquifer. These flowcell microcosms represent one of the first attempts to determine if HGT could be a suitable means of increasing the bioremediation capacity of subsurface biofilms exposed to groundwater flow. The introduction of donor strain P. putida SM1443 that harbors pWW0 to this model system has not accelerated the removal of toluene or BA; furthermore, no indication of HGT has been detected. These results are in contrast to concurrent conjugation experiments which have been performed as agar plate mating experiments with the same donor strain and recipient microbial community (Starek, 2010). These matings display a low, yet detectable, level of HGT. There are obvious contrasts between our system and previous studies – such as variety and concentration of nutrients, fewer potential recipients with regard to both cell numbers and bacterial species, concentrations of dissolved oxygen and a directional flow of bulk fluid – which focus on wastewater. Early results seem to support the notion that HGT depends not only on donor and recipient compatibility, but also on environmental factors (Lambertsen et al., 2004).

3.13.2.4 Utilization of Plasmid pWW0 for Bioaugmentation Purposes In summary, recent studies have shown that plasmid pWW0 can be successfully transferred to microorganisms within wastewater/activated sludge communities, but our current studies with groundwater-derived mixed cultures indicate that pWW0 transfer in other types of environments may be limited

3.13.3.1 Characteristics and Host Range Plasmid pJP4 is an 80-kb, IncPl, broad-host-range conjugative plasmid of Alcaligenes eutrophus, encoding resistance to mercuric chloride and phenyl mercury acetate and degradation of 2,4-dichlorophenoxyacetic acid, 2-methyl-4-chlorophenoxyacetic acid, and 3-chlorobenzoate (Don and Pemberton, 1985). It has been shown previously that conjugative catabolic plasmids of the incompatibility group P1 are transferred between bacterial cells at a high rate and possess a broad host range within the proteobacteria (Bathe et al., 2004a; Pukall et al., 1996). After introduction of bacteria carrying plasmid pJP4 into soil microcosms, conjugative transfer to indigenous bacteria and subsequent enhancement of 2,4-D degradation was observed (Newby et al., 2000a, 2000b). Our own studies with plasmid pJP4 (Bathe et al., 2004a) showed that it can be transferred to a variety of microbial genera of the a, b, and g classes of the Proteobacteria, mostly to the families Rhizobiaceae and Comamonadaceae and the genus Stenotrophomonas. However, only transconjugants identified as P. putida and Delftia spp. strains were able to grow on 2,4-D as the sole carbon source.

3.13.3.2 Plasmid pJP4 Transfer to Indigenous Microorganisms in Wastewater or Activated Sludge We investigated the possibility of enhancing degradation of the xenobiotic model compound 2,4-D in a model, laboratoryscale SBBR, inoculated with activated sludge from a municipal wastewater treatment plant, using plasmid pJP4 (Bathe et al., 2004b). After introduction of a plasmid donor strain to a labscale SBBR operated without 2,4-D, the number of plasmidcarrying cells first dropped, and then increased after switching to 2,4-D as the sole carbon source. The donor cells were unable to grow in the synthetic wastewater which was used in our experiment with 2,4-D as the sole carbon source, so the emergence of plasmid-carrying cells was attributed to successful plasmid transfer to members of the indigenous wastewater communities. Transconjugants could be detected both

Bioremediation: Plasmid-Mediated Bioaugmentation of Microbial Communities

by culture-dependent and culture-independent methods in the 2,4-D degrading bioaggregates. In contrast to 90% 2,4-D degradation in the bioaugmented reactor within 40 h, a control reactor that did not receive the plasmid donor still contained 60% of the initial 2,4-D concentration even after 90 h. Reactors consisted of 1.0 l of synthetic wastewater with an initial 2,4-D concentration of 2 mM. After 460 min of aerated reaction time, the entire liquid content of the reactor was drawn out. This experiment clearly demonstrates the introduction of 2,4-D degradative genes into a microbial biofilm and indicates that HGT is a promising tool for bioaugmentation of reactors treating wastewater. Although transconjugants could be cultured from the bioaugmented reactor, they did not belong to strains corresponding to sequences from dominant denaturing gradient gel electrophoresis (DGGE) (Muyzer et al., 1993) bands in the 2,4-D degrading biofilm sample. DGGE is a culture-independent, DNA-based fingerprinting technique which allows for the estimation of microbial diversity in a sample. It also permits partial identification of dominant microbial community members. In our study, it was not clear if dominant DGGE bands corresponded to transconjugant species which could not be isolated on agar plates or if the isolated Ralstonia strain was the dominant transconjugant species. However, the lack of a corresponding DGGE sequence indicated that the abundance of the isolated transconjugant stain within the activated sludge community was low. A closer relationship between phylogenetic identity and metabolic functions of specific bacteria may be established by the application of fluorescence in situ hybridization (FISH) probes to identify cells expressing plasmid-associated fluorescent proteins, by the use of microautoradiography combined with FISH or stable isotope probing as reviewed previously (Gray and Head, 2001).

373

Additionally, they may permit transfer of the plasmid to recipients which were not accessible by the initial donor strain (Bathe et al., 2004a).

3.13.4 Plasmid pNB2 3.13.4.1 Characteristics and Host Range Plasmid pNB2 is a broad host range IncP1 plasmid (88 kb) which carries genes for mercury resistance and for the degradation of aniline and 3-CA (Bathe, 2004) and was recently shown to be present in Comamonas testosteroni I2, a strain capable of aniline and 3-CA degradation (Boon et al., 2001). After conjugal transfer of this plasmid to Ralstonia eutropha JMP228, transconjugants gained the ability to degrade aniline, but not 3-CA. Therefore, pNB2 was described to be an anilinedegradative plasmid (Boon et al., 2001). Bathe (2004) investigated the involvement of the anilinedegradative plasmid pNB2 in degradation of 3-CA. Using plate matings of a P. putida strain containing pNB2 with a mixed bacterial culture derived from activated sludge and inoculation of the mating mixtures into batch cultures containing 3-CA, degradation of the compound was observed. A total of five different transconjugant strains were isolated from one of the batch cultures and two of them were able to degrade 3-CA. These two isolates were identified as C. testosteroni by partial 16S rDNA sequencing. Therefore, it was concluded that pNB2 carries a portion of the genes involved in the catabolism of 3-CA, but that the completion of the pathway must be provided by chromosomal genes in the host strain. The results suggest that pNB2 is a candidate plasmid which can be used in plasmid-mediated bioaugmentation of wastewater contaminated with chlorinated anilines.

3.13.3.3 Utilization of Plasmid pJP4 for Bioaugmentation Purposes

3.13.4.2 Plasmid pNB2 Transfer to Indigenous Microorganisms in Wastewater or Activated Sludge

In summary, our studies have shown that plasmid pJP4 can be successfully transferred to microorganisms within wastewater/ activated sludge communities and that plasmid transfer and the emergence of transconjugants is coupled to enhanced 2,4-D degradation. In the case of microbial wastewater communities, donor cell persistence and/or plasmid transfer was linked to enhanced BA degradation. However, it could not be differentiated whether accelerated degradation of BA was linked to the persistence of donor cells, the emergence of transconjugants or both. Even though transconjugants belonging to a number of different families were isolated, only a limited number of isolates, specifically those belonging to the Delftia and Pseudomonas genera were able to utilize 2,4-D as sole carbon source. The inability to utilize 2,4-D despite the presence of the plasmid may be caused by a lack of promoter recognition or translation of the produced mRNAs, failure to import the substrate, substrate toxicity and concentration, or incomplete assembly of the pathway if the host needs to provide additional catabolic genes. On the other hand, nondegrading transconjugants may serve as a plasmid reservoir, which might facilitate the adaptation of the community upon encountering the corresponding xenobiotic compound.

In a follow-up study, Bathe et al. (2005) tested the applicability of plasmid pNB2 for bioaugmentation of bacteria in model sequencing batch moving bed reactors (SBMBRs) containing wood chips as biofilm carrier material and receiving 3-CA. A setup of three biofilm reactors was studied, all initially inoculated with bacteria from activated sludge. One reactor (PB) received a P. putida pNB2 donor strain which, even though it carried the pNB2 plasmid, was not able to degrade 3-CA, due to a lack of chromosomally encoded genes required to complete the degradation pathway. A positive control reactor (P) received a 3-CA degrading C. testosteroni pNB2 transconjugant. A negative control reactor (N) received only the initial wastewater inocula but otherwise remained unchanged (i.e., no plasmid donors or 3-CA degrading strains were added). The results revealed that the positive control reactor (amended with a 3-CA degrading transconjugant) showed 3-CA degradation from the very beginning of the experiment. In contrast, degradation started after an initial lag period in reactor PB, amended with the plasmid donor (which was not able to degrade 3-CA, despite the presence of plasmid pNB2). No degradation was observed in the negative control reactor N, which received neither the

Table 1

Summary of plasmid-mediated bioaugmentation studies discussed in this contribution

Contaminant

Initial concentration (mg l 1)

Reactor type

Inoculum/ recipient community

Donor strain

Plasmid

Donor persistence

Transconjugant emergence

Enhanced degradation

Reference

BA

54

SBBR, pilot scale

Municipal WW

P. putida KT2442

pWW0

No

No

No

BA

62

Municipal WW

P. putida KT2442

pWW0

Yes

Yes

NDa

BA

108

Municipal WW

P. putida KT2442

pWW0

Yes

Yes

Yes

BA

541

Municipal WW

P. putida KT2442

pWW0

Yes

Yes

Yes

BA

2604

pWW0

No

No

No

173

Groundwaterderived mixed culture Municipal WW

P. putida SM1443

Toluene

SBBR, laboratory scale SBBR, laboratory scale SBR, aerobic granular sludge Continuous flowthrough chamber Flask

Venkata Mohan et al. (2009) Nancharaiah et al. (2003) Venkata Mohan et al. (2009) Nancharaiah et al. (2008) Starek (2010)

P. putida BBC443

pWW0

ND

Yes

No

Toluene

Continuous flowthrough chamber

Groundwaterderived mixed culture

P. putida SM1443

pWW0

No

No

No

2,4-D

303 415 (added as a slug of toluene to simulate a NAPLb spill) 442

Municipal WW

P. putida SM1443

pJP4

No

Yes

Yes

3-CA

80

Municipal WW

P. putida SM1443

pNB2

No

Yes

Yes

Bathe et al. (2004b) Bathe et al. (2005)

3-CA

100–400 (variable)

SBBR, laboratory scale SBMBR, laboratory scale Semicontinuous activated sludge, semicontinuous MBBR

Municipal WW

Comamonas testosteroni SB3

pNB2

ND

Yes

Yes, but variable

Bathe et al. (2009)

a

ND, not determined. NAPL, nonaqueous phase liquid.

b

Pei and Gunsch (2009) Starek (2010)

Bioremediation: Plasmid-Mediated Bioaugmentation of Microbial Communities

plasmid donor nor the 3-CA degrading transconjugant. Polymerase chain reaction (PCR) analysis showed that the P. putida donor abundance dropped in reactor PB (bioaugmented with the plasmid donor strain), but plasmid pNB2 abundance remained stable, indicating plasmid transfer to members of the indigenous wastewater communities. A number of different 3-CA degrading C. testosteroni strains carrying pNB2 were isolated from the plasmid donor bioaugmented reactor PB. In order to determine the most efficient bioaugmentation approach using plasmid pNB2, Bathe et al. (2009) investigated several strategies for achieving degradation of 3-CA in semicontinuous activated sludge reactors. The addition of a 3-CAdegrading C. testosteroni strain carrying the degradative plasmid pNB2 to a biofilm reactor resulted in complete 3-CA degradation together with spread of the plasmid within the indigenous biofilm population. A second set of reactors was bioaugmented with either a suspension of biofilm cells removed from the carrier material or with biofilm-containing carrier material (in this case, wood chips). 3-CA degradation was established rapidly in all bioaugmented reactors, followed by a slow adaptation of the nonbioaugmented control reactors. Variations in 3-CA concentration caused temporary performance breakdowns in all reactors. Duplicates of the control reactors deviated in their behavior but the bioaugmented reactors appeared more reproducible in their performance and population dynamics. The carrier-bioaugmented reactors showed an improved performance in the presence of high 3-CA influent concentrations in comparison the suspension-bioaugmented reactors. In contrast, degradation in one control reactor failed completely, but was rapidly established in the remaining control reactor.

3.13.4.3 Utilization of Plasmid pNB2 for Bioaugmentation Purposes Our studies using plasmid pNB2 clearly demonstrated that a successful plasmid-mediated bioaugmentation was achieved (Bathe et al., 2005, 2009). C. testosteroni was the dominant 3-CA degrading pNB2 transconjugant species isolated from the donor strain-augmented reactor PB. The study again demonstrated the potential of gene transfer to spread and establish xenobiotic degrading potential by dissemination of catabolic genes in particular.

3.13.5 Conclusions and Recommendations The studies discussed in this chapter are summarized in Table 1. Our studies have shown that plasmid transfer in bioreactors occurs with a subsequent formation and outgrowth of transconjugants, if a selective pressure in the form of the target compound is being applied. This was confirmed in our studies with three different degradative plasmids, namely the TOL plasmid pWW0, plasmid pJP4, and plasmid pNB2. On the other hand, we have also shown that successful bioaugmentation in one type of microbial community (i.e., wastewater) is not directly transferable to another type of microbial environments (i.e., groundwater-derived cultures). Further investigations are necessary to examine if plasmidmediated bioaugmentation can be applied to remove

375

xenobiotic compounds in industrial wastewater treatment systems and to reveal if a long-term bioaugmentation can be achieved using this approach. More research is required to analyze the role of fluctuations and threshold concentration of the target compound, the stability of the degradation potential, and thus the fate of plasmid-encoded catabolic genes in the absence of the target compound, the role and presence of other (possibly recalcitrant) carbon compounds, process conditions, and different ways of biomass retention specific or nonspecific for degrader populations. This should reveal if plasmid-mediated bioaugmentation is a generally applicable tool or if it is a technique that is best used to facilitate start-up and operation of specialized bioreactors receiving chemically defined and temporally stable wastewater.

Acknowledgments MH and MS acknowledge National Sciences and Engineering Council of Canada (NSERC) support (Discovery Grants Program– Individual; Grant No. 355606-2008 to MH).

References Adav SS, Lin JCT, Yang Z, Whitely CG, Lee DJ, and Peng XF (2010) Stereological assessment of extracellular polymeric substances, exo-enzymes, and specific bacterial strains in bioaggregates using fluorescence experiments. Biotechnology Advances 28(2): 255–280. Bathe S (2004) Conjugal transfer of plasmid pNB2 to activated sludge bacteria leads to 3-chloroaniline degradation in enrichment cultures. Letters in Applied Microbiology 38: 527--531. Bathe S, Lebuhn M, Ellwart JW, Wuertz S, and Hausner M (2004a) High phylogenetic diversity of transconjugants carrying plasmid pJP4 in an activated sludge-derived microbial community. FEMS Microbiology Letters 235: 215--219. Bathe S, Mohan TVK, Wuertz S, and Hausner M (2004b) Bioaugmentation of a sequencing batch biofilm reactor by horizontal gene transfer. Water Science and Technology 49: 337--344. Bathe S, Schwarzenbeck N, and Hausner M (2005) Plasmid-mediated bioaugmentation of activated sludge bacteria in a sequencing batch moving bed reactor using pNB2. Letters in Applied Microbiology 41: 242--247. Bathe S, Schwarzenbeck N, and Hausner M (2009) Bioaugmentation of activated sludge towards 3-chloroaniline removal with a mixed bacterial population carrying a degradative plasmid. Bioresource Technology 100: 2902--2909. Benson S and Shapiro J (1978) TOL is a broad-host-range plasmid. Journal of Bacteriology 135: 278--280. Boon N, Goris J, De Vos P, Verstraete W, and Top EM (2000) Bioaugmentation of activated sludge by an indigenous 3-chloroaniline-degrading Comamonas testosteroni strain I2. Applied and Environmental Microbiology 66: 2906--2913. Boon N, Goris J, De Vos P, Verstraete W, and Top EM (2001) Genetic diversity among 3-chloroaniline- and aniline-degrading strains of the comamonadaceae. Applied and Environmental Microbiology 67: 1107--1115. Bouchez T, Patureau D, Dabert P, et al. (2000a) Ecological study of a bioaugmentation failure. Environmental Microbiology 2: 179--190. Bouchez T, Patureau D, Dabert P, Wagner M, Delgenes JP, and Moletta R (2000b) Successful and unsuccessful bioaugmentation experiments monitored by fluorescent in situ hybridization. Water Science and Technology 41: 61--68. Bouwer EJ (1989) Transformation of xenobiotics in biofilms. In: Characklis WG and Wilderer PA (eds.) Structure and Function of Biofilms, pp. 251--267. New York: Wiley. Chalfie M, Tu Y, Euskirchen G, Ward WW, and Prasher DC (1994) Green fluorescent protein as a marker for gene-expression. Science 263: 802--805. Christensen BB, Sternberg C, Andersen JB, et al. (1998) Establishment of new genetic traits in a microbial biofilm community. Applied and Environmental Microbiology 64: 2247--2255. Christensen BB, Sternberg C, and Molin S (1996) Bacterial plasmid conjugation on semi-solid surfaces monitored with the green fluorescent protein (GFP) from Aequorea victoria as a marker. Gene 173: 59--65.

376

Bioremediation: Plasmid-Mediated Bioaugmentation of Microbial Communities

Dabert P, Delgenes JP, and Godon JJ (2005) Monitoring the impact of bioaugmentation on the start up of biological phosphorus removal in a laboratory scale activated sludge ecosystem. Applied Microbiology Biotechnology 66: 575--588. Dalton HM and March PE (1998) Molecular genetics of bacterial attachment and biofouling. Current Opinion in Biotechnology 9: 252--255. De Kreuk MK, Kishida N, and van Loosdrecht MCM (2007) Aerobic granular sludge – state of the art. Water and Science Technology 55: 75--81. Dejonghe W, Goris J, El Fantroussi S, et al. (2000) Effect of dissemination of 2,4-dichlorophenoxyacetic acid (2,4-D) degradation plasmids on 2,4-D degradation and on bacterial community structure in two different soil horizons. Applied and Environmental Microbiology 66: 3297--3304. Don RH and Pemberton JM (1985) Genetic and physical map of the 2,4-dichlorophenoxyacetic acid degradative plasmid pJP4. Journal of Bacteriology 161: 466--468. Flemming HC, Neu TR, and Wozniak DJ (2007) The EPS matrix: The ‘‘house of biofilm cells’’. Journal of Bacteriology 189(22): 7945--7947. Fox RE, Zhong X, Krone SM, and Top EM (2008) Spatial structure and nutrients promote invasion of IncP-1 plasmids in bacterial populations. International Society for Microbiology and Ecology 2: 1024--1039. Gray ND and Head IM (2001) Linking genetic identity and function in communities of uncultured bacteria. Environmental Microbiology 3: 481--492. Greated A, Lambertsen L, Williams PA, and Thomas CM (2002) Complete sequence of the IncP-9 TOL plasmid pWW0 from Pseudomonas putida. Environmental Microbiology 4: 856--871. Johnsen AR and Kroer N (2007) Effects of stress and other environmental factors on horizontal plasmid transfer assessed by direct quantification of discrete transfer events. FEMS Microbiology Ecology 59: 718--728. Kasai Y, Kodama Y, Takahata Y, Hoaki T, and Watanabe K (2007) Degradative capacities and bioaugmentation potential of an anaerobic benzene-degrading bacterium strain DN11. Environmental Science and Technology 41: 6222--6227. Lambertsen LM, Molin SM, Kroer N, and Thomas CM (2004) Transcriptional regulation of pWW0 transfer genes in Pseudomonas putida KT2440. Plasmid 52(3): 169–181. Molin S and Tolker-Nielsen T (2003) Gene transfer occurs with enhanced efficiency in biofilms and induces enhanced stabilisation of the biofilm structure. Current Opinion in Biotechnology 14: 255--261. Muyzer G, Dewaal EC, and Uitterlinden AG (1993) Profiling of complex microbialpopulations by denaturing gradient gel-electrophoresis analysis of polymerase chain reaction-amplified genes-coding for 16s ribosomal-RNA. Applied and Environmental Microbiology 59: 695--700. Nakazawa T, Hayashi E, Yokota T, Ebina Y, and Nakazawa A (1978) Isolation of TOL and RP4 recombinants by integrative suppression. Journal of Bacteriology 134: 270--277. Nancharaiah YV, Joshi HM, Hausner M, and Venugopalan VP (2008) Bioaugmentation of aerobic microbial granules with Pseudomonas putida carrying TOL plasmid. Chemosphere 71: 30--35. Nancharaiah YV, Wattiau P, Wuertz S, et al. (2003) Dual labeling of Pseudomonas putida with fluorescent proteins for in situ monitoring of conjugal transfer of the TOL plasmid. Applied and Environmental Microbiology 69: 4846--4852. Newby DT, Gentry TJ, and Pepper IL (2000a) Comparison of 2,4dichlorophenoxyacetic acid degradation and plasmid transfer in soil resulting from bioaugmentation with two different pJP4 donors. Applied and Environmental Microbiology 66: 3399--3407. Newby DT, Josephson KL, and Pepper IL (2000b) Detection and characterization of plasmid pJP4 transfer to indigenous soil bacteria. Applied and Environmental Microbiology 66: 290--296. Normander B, Christensen BB, Molin S, and Kroer N (1998) Effect of bacterial distribution and activity on conjugal gene transfer on the phylloplane of the bush bean (Phaseolus vulgaris). Applied and Environmental Microbiology 64: 1902--1909. Pandey J, Chuahan A, and Jain RK (2008) Integrative approaches for assessing the ecological sustainability of in situ bioremediation. FEMS Microbiology Reviews 33: 324--375. Pei R and Gunsch C (2009) Plasmid conjugation in an activated sludge microbial community. Environmental Engineering Science 26: 825--831. Pepper IL, Gentry TJ, Newby DT, Roane TM, and Josephson KL (2002) The role of cell bioaugmentation and gene bioaugmentation in the remediation of co-contaminated soils. Environmental Health Perspectives 110(supplement 6): 943--946.

Pukall R, Tschape H, and Smalla K (1996) Monitoring the spread of broad host and narrow host range plasmids in soil microcosms. FEMS Microbiology Ecology 20: 53--66. Ramos-Gonzalez MI, Duque E, and Ramos JL (1991) Conjugational transfer of recombinant DNA in cultures and in soils: Host range of Pseudomonas putida TOL plasmids. Applied and Environmental Microbiology 57: 3020--3027. Sarand II, Haario H, Jorgensen KS, and Romantschuk M (2000) Effect of inoculation of a TOL plasmid containing mycorrhizosphere bacterium on development of Scots pine seedlings, their mycorrhizosphere and the microbial flora in m-toluateamended soil. FEMS Microbiology Ecology 31: 127--141. Smets BF, Morrow JB, and Pinedo CA (2003) Plasmid introduction in metal-stressed, subsurface-derived microcosms: Plasmid fate and community response. Applied and Environmental Microbiology 69: 4087--4097. Smets BF, Rittmann BE, and Stahl DA (1993) The specific growth rate of Pseudomonas putida PAW1 influences the conjugal transfer rate of the TOL plasmid. Applied and Environmental Microbiology 59: 3430--3437. Smets BF, Rittmann BE, and Stahl DA (1995) Quantification of the effect of substrate concentration on the conjugal transfer rate of the TOL plasmid in short-term batch mating experiments. Letters in Applied Microbiology 21: 167--172. Sorensen SJ, Bailey M, Hansen LH, Kroer N, and Wuertz S (2005) Studying plasmid horizontal transfer in situ: A critical review. Nature Reviews Microbiology 3: 700--710. Starek M (2010) Evaluation of the Transfer of the TOL Plasmid from Pseudomonas putida to Groundwater-Derived Biofilms in a Model Rock-Fracture Aquifer. MSc Thesis, Ryerson University, Toronto, ON, Canada. Taghavi S, Barac T, Greenberg B, Borremans B, Vangronsveld J, and van der Lelie D (2005) Horizontal gene transfer to endogenous endophytic bacteria from poplar improves phytoremediation of toluene. Applied and Environmental Microbiology 71: 8500--8505. Tchelet R, Meckenstock R, Steinle P, and van der Meer JR (1999) Population dynamics of an introduced bacterium degrading chlorinated benzenes in a soil column and in sewage sludge. Biodegradation 10: 113--125. Top EM and Springael D (2003) The role of mobile genetic elements in bacterial adaptation to xenobiotic organic compounds. Current Opinion in Biotechnology 14: 262--269. Top EM, Springael D, and Boon N (2002) Catabolic mobile genetic elements and their potential use in bioaugmentation of polluted soils and waters. FEMS Microbiology Ecology 42: 199--208. Tsien RY (1998) The green fluorescent protein. Annual Review of Biochemistry 67: 509--544. Van Limbergen H, Top EM, and Verstraete W (1998) Bioaugmentation in activated sludge: Current features and future perspectives. Applied Microbiology and Biotechnology 50: 16--23. Venkata Mohan S, Falkentoft C, Venkata Nancharaiah Y, et al. (2009) Bioaugmentation of microbial communities in laboratory and pilot scale sequencing batch biofilm reactors using the TOL plasmid. Bioresource Technology 100: 1746--1753. Wilderer PA and McSwain BS (2004) The SBR and its biofilm application potentials. Water Science and Technology 50: 1--10. Williams PA and Murray K (1974) Metabolism of benzoate and the methylbenzoates by Pseudomonas putida (arvilla) mt-2: Evidence for the existence of a TOL plasmid. Journal of Bacteriology 120: 416--423. Wolfaardt GM, Lawrence JR, Headley JV, Robarts RD, and Caldwell DE (1994) Microbial exopolymers provide a mechanism for bioaccumulation of contaminants. Microbial Ecology 27: 279--291. Wong CL and Dunn NW (1974) Transmissible plasmid coding for the degradation of benzoate and m-toluate in Pseudomonas arvilla mt-2. Genetical Research 23: 227--232. Worsey MJ and Williams PA (1975) Metabolism of toluene and xylenes by Pseudomonas putida (arvilla) mt-2: Evidence for a new function of the TOL plasmid. Journal of Bacteriology 124: 7--13. Wuertz S, Okabe S, and Hausner M (2004) Microbial communities and their interactions in biofilm systems: An overview. Water Science and Technology 49: 327--336. Yarbrough D, Wachter RM, Kallio K, Matz MV, and Remington SJ (2001) Refined crystal structure of DsRed, a red fluorescent protein from coral, at 2.0-A resolution. Proceedings of the National Academy of Sciences of the United States of America 98: 462--467.

3.14 Drinking Water Toxicology in Its Regulatory Framework H Dieter, Federal Environment Agency (UBA), Dessau-Roßlau, Germany & 2011 Elsevier B.V. All rights reserved.

3.14.1 3.14.1.1 3.14.1.2 3.14.2 3.14.2.1 3.14.2.2 3.14.2.3 3.14.2.3.1 3.14.2.3.2 3.14.2.4 3.14.2.4.1 3.14.2.4.2 3.14.2.4.3 3.14.2.5 3.14.2.5.1 3.14.2.5.2 3.14.3 3.14.3.1 3.14.3.2 3.14.4 3.14.4.1 3.14.4.2 3.14.4.3 3.14.4.3.1 3.14.4.3.2 3.14.4.3.3 3.14.4.3.4 3.14.4.3.5 3.14.5 3.14.5.1 3.14.5.2 3.14.5.3 3.14.6 3.14.6.1 3.14.6.2 3.14.6.2.1 3.14.6.2.2 3.14.6.2.3 3.14.6.2.4 3.14.6.2.5 3.14.6.2.6 3.14.6.2.7 3.14.6.2.8 3.14.6.2.9 3.14.6.3 3.14.6.3.1 3.14.6.3.2 3.14.7 3.14.7.1 3.14.7.2 3.14.7.3 3.14.7.4

Introduction Drinking Water: A Unique Medium to Support Life and Personal Hygiene Acceptability and Tolerability of Chemicals in Drinking Water According to Their Functionality and the Potential to Avoid Them From Chemical Hazards to Chemical Standards Basic Facts Historical Landmarks Objectives and Goals of Protection Objectives of protection Goals of protection Timescales to Protect Goals of Protection Precautionary standards (enduring protection of human beings and drinking-water resources) Scientific standards (guide values for lifelong protection of human beings) Remedial standards (action values to protect from shorter-than-lifetime exposure) Special Aspects to be Considered when Setting Standards for Drinking Water Standards for accepted chemicals in drinking water Standards to protect from adverse effects of chemicals in drinking water Panels and Institutions for Setting Drinking Water Standards National International Defining Standards to Prevent Human Health Risks from Drinking Water Qualification of Risks – Critical Toxic Endpoints Groups of Compounds of Specific Interest for Drinking Water Risk Quantification Chemicals exhibiting systemic effects with threshold Chemicals assumed to exhibit threshold-free systemic effects Local effects on humans of chemicals in drinking water Effect combinations Exposure A Holistic Approach for Defining Quality Goals or Standards for Drinking Water Chemicals Whose Regulation Will Primarily Concentrate on Avoidance of Adverse Effects Chemicals Whose Regulation Should Concentrate on an Optimal Ratio of Functional Exposure to Functional Intention Chemicals Whose Regulation Should Concentrate on Remote Emission Control Practical Regulation of Drinking-Water Quality Quality Assurance and Surveillance of Raw Water, Finished Water, Tap Water Paths and Significance of Exposure to Chemicals in Drinking Water Acidification of raw water Agrochemicals Emissions from abandoned waste sites Disinfection by-products Small supplies and health risks Organics leaching from drinking water reservoir coatings and armatures Hygienic aspects of corrosion products from domestic pipes and metallic materials Hygienic aspects of domestic posttreatment of drinking water Hygienic aspects of domestic water saving Significance and Hygienic Assessment of Exposure at the Tap Inorganics Organics The Author’s Short Conclusions Unintended Exposure By-Products of Disinfection and Oxidation Risk Assessment and Management Derogations from Limit Values

377 377 378 379 379 379 380 380 381 381 381 382 382 383 383 383 384 384 384 384 384 385 385 385 387 387 388 389 390 390 390 391 391 391 393 393 393 393 394 395 396 396 396 397 397 397 403 407 407 408 408 408

377

378

Drinking Water Toxicology in Its Regulatory Framework

3.14.7.5 3.14.7.6 3.14.8 3.14.8.1 3.14.8.2 3.14.8.3 3.14.8.4 3.14.8.5 References

Drinking-Water Installations Surveillance of Drinking Water Perspectives on Perception of Drinking Water How Pure Is Pure? Erroneous Reasons to Ask for Absolute Purity Some Good Reasons for Not Asking for ‘Absolute’ Purity Many More Good Reasons for Not Exhausting Strictly Health-Based Levels How to Best Realize the Social Concept of an Esthetically Acceptable Drinking Water?

3.14.1 Introduction 3.14.1.1 Drinking Water: A Unique Medium to Support Life and Personal Hygiene Water, besides oxygen, is the most important medium to support life and physiological or hygienic needs. Apart from its conceptual designation of ‘water for human consumption/ drinking water’ it refers additionally to specific uses of water in private and public facilities for purposes of personal hygiene. Several technical and principal aspects of drinking water hygiene and the safety of its distribution and accessibility endorse the view that water that meets physiological or personal hygiene demands optimally should exhibit an unobjectionable chemical and microbiological quality as well. A list ranking countries for shortage of freshwater index reveals per capita availabilities between 10 m3 (Kuwait) and almost 100 000 m3 (Canada). Sufficient water supply is considered possible from 1600 m3 upwards in hydrogeologically disparate countries as Kenya (985 m3), Ethiopia (1749 m3), Germany (1878 m3), Afghanistan (2986 m3), Mexico (4624 m3), and Portugal (6859 m3) UNEP (2003a). The minimum amount of drinking water to meet daily needs of personal hygiene and physiology seems to vary between 20 l per person, if available within 1 km of the user’s dwelling (WHO, 2008), and up to about 100 l in case of household connections (Cairncross and Valdmanis, 2010). This discussion or controversy on the technical feasibility and hygienic advantages of two parallel systems to distribute ‘service water’ in one and ‘drinking water’ in the other is as old as the idea of central supply. In Europe, it was decided at the end of the nineteenth century to have one single system of pipes to deliver water of unobjectionable quality to serve all purposes of personal hygiene and physiology (Kluge and Schramm, 1986). The forced domestic use of greywater in private houses would very probably not reduce the respective chemical load, neither of the environment, nor of the sewage treatment plants, or of the reused water (Donner et al., 2010). Unobjectionable drinking water is not necessarily sterile, but devoid of infectious concentrations of pathogens. It should never be allowed to exhibit any property or potential to impair human health. Furthermore, according to current technical (national and supranational) norms, it should be appetizing, cool, colorless and odorless, and low in germs. It should have an inoffensive taste and stimulate consumption. Its content of dissolved substances has to be kept, as far as possible, within technically feasible limits. It should not lead to corrosion of technical materials within

408 408 409 409 409 409 410 410 410

the distribution system and households and be available at their hand-over points (‘taps’) in sufficient quantity and pressure. According to the general rule of hygiene, food production from raw materials should have, as far as possible, the same chemical and microbiological quality as the intended product, containing therefore, ideally, only technically inevitable contaminations or residues. The raw material of drinking water is the ‘raw water’ as taken or extracted from groundwater or (prepurified) surface water to transform it to drinking water by a more or less elaborate chemical or physical treatment. This leads to the present principle to rely, wherever possible, on raw water that requires only minimal or even no treatment – neither for health nor for technical or esthetic reasons. The same principle applies to any other life-supporting foods or principles we use or consume directly in their natural form, for example, sunlight, fruits, or air for breathing. A very important point when looking on our consumption of foods as biochemical energy carriers is the fact that, in contrast to foods, we do not consume or destroy (drinking) water to support our physiologic and hygienic needs. We only make use of it as a transporter of molecules, respectively, contaminants and their desired and undesired interactions. This is why in this context the central questions on the sustainability of human behavior toward ‘drinking water’ are not those dealing with technically available or absolute water quantities but with the tolerable degree and quality of technically or physiologically inevitable contamination of water after its intended use, its discharge as waste water, and its eventual reuse. The discussion on the question ‘which intensity of use and contamination of a regional water resource would correspond to optimal but neither maximal nor minimal quality claims’ has therefore to be negotiated on a societal level (Section 3.14.8). From the point of sustainability, ‘saving water for use’ means, therefore, preferentially ‘clean the used water’ as efficiently as reasonably achievable and reuse it after it has passed a sufficient number of (at least two to three) technical or natural safety barriers to retain infectious agents and chemical noxes. From several points of view (technical, hygienic, social, or psychological), the optimal fraction of water to be reused in moderately to highly populated areas is not qualified for being defined on the organizational level of small family homes or even single households. Instead, all qualitative and hydrogeological aspects of the natural regional water yield, its rate of renewal, and its technical availability must be considered to find the optimal technical structures and procedures to meet this important societal task to maintain the regional resources

Drinking Water Toxicology in Its Regulatory Framework

for drinking water as abundant, socially equitable, and as clean as possible. This purpose, ‘‘adequate sanitation of a standard which sufficiently protects human health and the environment,’’ shall particularly be ‘‘done through the establishment, improvement and maintenance of collective systems’’ (UN – Economic Commission for Europe, 1999). The only imperative and ‘absolute’ (scientific) quality requirement for any drinking water from any source consists in respecting health-based quality criteria as strictly as possible, whereas precautionary (not merely scientific) aspects of sustainability are best followed along the principle of keeping nonfunctional exposure ‘‘as low as reasonably (mostly technically) achievable’’ (ALARA). Within this discussion, the focus of what may be called achievable is not only on hard aspects such as human health but also on softer criteria such as esthetics or pureness of drinking water, taking into account, as WHO (2008) proposes, the socioeconomic, cultural, or ecologic contexts. It should include therefore criteria of technical functionality or efficiency of investments to minimize the inevitable preventable contamination right from the beginning to the end of the pipe. This holistic approach is also the best to minimize the risk that a consumer might refuse a subjectively unpleasant central supply and replace it by a seemingly more pleasant one being possibly present in his/her private backyard or in the hands of a street retailer. Taken together, limit values for substances in the drinking water system are not only a matter of science or human toxicology but also of sustainable drinking water hygiene (or pureness) and of safeguarding water resources as clean and as abundant as possible for future generations. The corresponding proactive limit values or standards are not to be defined purely by science. They need to be defined and fixed in a transparent societal decision process as well. Its stakeholders are bound to consider each reasonable option to define and quantify minimal maximal value(s) for the substance(s) under question, referring not only to toxicological but also to technical and hygienic advice, taking into account as many societal perspectives on drinking water as possible (see Section 3.14.8).

3.14.1.2 Acceptability and Tolerability of Chemicals in Drinking Water According to Their Functionality and the Potential to Avoid Them Proactive management of drinking water quality fits best with a three-dimensional rule of environmental hygiene, which is ‘‘Avoid useless, optimize functional and prevent harmful exposure’’ (Dieter, 1998). It leads directly to limit values that are functional from the management point of view. Only inevitable exposure, be it natural or technical/functional, needs to be limited or regulated at higher than zero (and sometimes even close to risky) levels. However, many substances functional on site find their way to distant environmental compartments off site. The aforementioned rule asks to regulate such useless and mostly anthropogenic exposure by means of the ALARA principle down to off site levels as close to a technical zero as possible, instead of allowing them to encroach on health-related maximal levels.

379

Such a concept of avoiding drinking water contamination by prevention is certainly a safer plane of consensus between experts and stakeholders than the too ambitious concepts of defining never-ending numbers of adverse effect levels on the one side and analytical zero exposure on the other – especially for contaminants devoid of any benefit at their off-site point of exposure. Exposure of raw and drinking water to contaminants such as pesticides, industrial chemicals, or pharmaceuticals is useless as per definition. It is not acceptable but just tolerable, and preventing them from entering the drinking water system is an everlasting challenge in minimizing the on-site emission. In contrast, exposure of consumers via drinking water to corrosion products, corrosion inhibitors, disinfectants, and their disinfection by-products (DBPs) is never devoid of a functional aspect. Therefore, it seems acceptable on site at ‘residual’ minimal, albeit functionally inevitable levels (see Section 3.14.2.3.2). However, they should never touch healthrelated values and if they ever do, their functionality or corresponding risk/benefit ratio has to be critically reexamined. This concept of distinguishing between acceptable and tolerable anthropogenic exposure is reflected exactly in the conceptual distinction between tolerable and acceptable daily intake (TDI and ADI) of a chemical as it was promulgated in 1987 by WHO (1987). The toxicological meaning – maximum daily intake of a chemical without any appreciable health risk, in mg per kg body mass (kg bm1) per day – of both is identical; however, the database from which a numerical value is extracted very often is much less complete in the case of a TDI than of an ADI. The reason behind is the mere absence of benefit from exposure to unaccepted contaminants and similarly, as a consequence, also the absence of a systematic gathering of toxicological data to assess their presence in the environment. By contrast, geogenic constituents of raw water are reduced to a technical limit of avoidance only if their natural levels appear to be harmful for either human health or the technical distribution system. Such levels, if not reduced by reasonably feasible treatment, are also the upper limit for the usability of a resource or respectively for the tolerability of a natural but useless constituent under question in case of nature-oriented treatments. The same holds true for biogenic constituents such as cyanotoxins in reservoirs for drinking water, most of them associated with a high toxic potential for humans. The only way to manage their reduction below nontoxic levels is by stopping over-fertilization of affluent waters. Most prominent examples of constituents with high human toxicity are fluoride, arsenic (III þ V), uranium 2þ (UO2þ 2 ), and selenium (Se ). Technical hurdles include Mn (IV), Fe(III), and hardness (Mg2þ þ Ca2þ), the latter one also for the possible health and nutritional benefit (Cotruvo and Bartram, 2009). Drinking water containing high levels of HCO3 may support reabsorption of Ca2þ and Mg2þ in the renal tubuli, thereby intensifying their potential to prevent cardiovascular disease (Rylander, 2008). On the other hand, natural constituents may also be judged as acceptable if these are essential nutrients or trace elements at levels of total intake that would not exceed the acceptable range of oral intake (AROI) (WHO, 1998).

380

Drinking Water Toxicology in Its Regulatory Framework

3.14.2 From Chemical Hazards to Chemical Standards 3.14.2.1 Basic Facts The water molecule is a strong dipole with an excited or oxygen atom with a partial negative charge in its center. Its distorted tetrahedral structure and high polarity are the preconditions for liquid water organizing itself in the form of dynamic molecular clusters sticking loosely together by permanently opening and closing H2O-hydrogen bonds, containing 4–40 kJ mol1 each. Its high polarity is also why it separates ionic molecular structures (salts) into cations and anions and is able to gather and transport easily an almost innumerable variety of polar compounds from its adjacent environments. Therefore, there is simply no way to get absolutely pure water out of a natural drinking water resource, a water works facility or the end (tap or faucet) of any distribution system, including domestic drinking water installations (Hopp, 2004). Natural waters (and their main ingredients) that can serve as possible raw water for drinking water production can be differentiated into

• •

• •

rain water (dust, dissolved natural and anthropogenic gases); spring and ground or subsoil water (up to 0.2% solids, mainly Ca- and Mg-, but also Fe- and Mn-compounds, either particular or containing most commonly carbonate, sulfate, fluoride, nitrate, and chloride as counter ions); water from lakes and rivers (freshwater with less than 0.02% solids); and seawater (with 3.5% solids, 3% being table salt, the rest comprising about 50 different geogenic ions, each in a typical and stable oceanic concentration).

Most living beings contain in their nonfat compartments more than 50% (up to 99%, up to 73% in humans) of water as a solvent and transport medium, reaction component for anabolic and catabolic processes, and source of hydrogen for hydrating reactions and of hydrogen bonding to stabilize biologic macromolecules for function. The renal filtration rate per adult person is about 30% of the total blood volume of 1700 l d1, whereas daily net use of water for transport of metabolic waste comprises only 2.6 l per adult person. Fifty percent or more of this amount is normally replaced or ingested in the form of drinking water and 40% or less as constituent of food, whereas 10% is generated in mitochondria as endogenous oxidation water. Daily excretion is renal (1.6 l), dermal (0.7 l), pulmonal (0.2 l), and fecal (0.1 l). Daily loss of water lasting several days and more than 10–20% of the daily need can be dangerous for health. Drinking water might become a significant source of some minerals and essential metals only in individuals whose regular diets are low in these metals (Deveau, 2010). Seen from the other side, consuming drinking water in normal amounts with low-to-zero mineral content does not hold any danger for health if combined with a balanced diet.

3.14.2.2 Historical Landmarks The Corpus Hippocratum by Hippocratus contains some ancient but still valuable guidelines on drinking-water quality,

recommending to use rain water, to prefer running to stagnating water, and to judge its quality according to local circumstances (geology, vegetation, staining of metals, and general state of the inhabitant’s health) (Garbrecht, 1986). Hippocrates also provides early advice on individual possibilities to improve water quality for own human consumption. Improvement of water quality for human consumption by boiling, straining, storing in copper vessels, or filtration over charcoal has been known to date back to as early as several hundred years BC (State of Alaska, 2010). Conflicts between intentional and unintentional strains on natural water resources and their protection or vulnerability, respectively, can be traced back as far as 2000 years from now when close to the Roman aqueduct providing what is today’s French city Lyon with freshwater, a plaque was put up against unintentional contamination, bearing the inscription ‘‘ex autoritate imp(eratoris) Caes(aris) Trajani Hadriani Aug(usti) nemini arandi, serendi pangendive jus est intra id spatium agri, quod ductus destinatum est’’ (‘‘At the behest of Emperor Caesar Trajanus Hadrianus Augustus nobody shall be permitted to plough, seed, or plant within the space dedicated to protect this aqueduct’’ (translation by the author from the original Latin; German translation given by Kolkmann (1990))). At these early times, it was still sufficient to protect just a central pipe and its content as the water transported therein came from catchments yet untouched by human settlements and economic activity (Kolkmann, 1990). In European medieval times, protection of wells from being poisoned by individuals with bad intentions was regulated by strict laws to prosecute and punish such individuals strictly, but there was no such diktat to prevent unintentional degradation or poisoning of a water source. Only in the course of the second half of the nineteenth century did it become clear that penalizing was not sufficient to protect public water sources from more or less unintentional contamination. Instead, this novel task obviously was to be solved normatively on societal levels of central administrative law and public health police (and policy). One of the concurrent consequences was to close the many private and public urban wells in favor of providing everybody with a centrally controlled drinking water supply at a reasonable cost and to create on 1 April 1901, in Prussia, a worldwide first ‘private–public partnership’ to support interdisciplinary research, expert advice, and teaching in the field of water hygiene under the leadership of Prussia’s Ko¨nigliche Versuchs- und Pru¨fungsanstalt fu¨r Wasserversorgung und Abwasserbeseitigung (Royal Research and Testing Institution for Water Supply and Sewage Disposal). Due to the pioneering microbiological and epidemiological work done by John Snow, Louis Pasteur, and Robert Koch, up to the first quarter of the nineteenth century, publicly supplied water was thought to exhibit a potential for spreading or transmitting infectious diseases, but not those of chemical origin. There was even some speculation on industrial wastewater and its chemicals to purify the accepting rivers of dangerous fecal germs by their precipitation and disinfection (Kluge and Schramm, 1986). Therefore, healthbased maximal values for a number of highly toxic pesticides could not be implemented as was hoped by the US Public Health Service (PHS) as late as 1967, because the legal basis

Drinking Water Toxicology in Its Regulatory Framework

called upon (prohibition of infectious diseases) was not affirmed on behalf of a private suit. Despite this unclear early legal situation, already from 1914 on technically based maximal values for some chemical parameters occurring typically in connection with central water supply and potentially detrimental for them were decreed, such as technical guide values for copper and zinc by the PHS in the US, whereas in Europe, especially Germany, principal weight was put not so much on maximal values but rather, like in the 2nd German Empire, officially, on guiding rules to safeguard high-quality installation, operation, and surveillance of public water facilities to serve not only technical purposes. In 1915, the German water hygienist from Jena university, August Ga¨rtner, published his yet modern and comprehensive, epoch-making lifework (Ga¨rtner, 1915). Ga¨rtner emphasized the need to consider local circumstances and not just chemical or bacteriological numbers when evaluating a water supply. His forward-looking definition of ‘drinking water’ which should exhibit the same high quality also for any other domestic purpose is valuable up to the present day. Later on, in 1942, in the US, health-related permissible values for arsenic and lead and even a very low one (1 mg l1) for phenols were decreed by PHS (Larson, 1990), the latter group of chemicals being the first known group of large-scale drinking water contaminants from industrial sources. They had attracted attention since the 1920s in Europe, not for any high toxicity but as precursors for a closely related group of chlorinated DBPs, the chlorophenols, arising from phenols during drinking-water disinfection by chlorination and giving the finished water a strange pharmacy taste at the mg l1 range itself (Kluge and Schramm, 1986). After 30 more years, chlorination of drinking water became a main milestone in developing a system of health-related parametric values for systemic contaminants of drinking water, since Rook (1974) in Europe and Bellar et al. (1974) in the US had discovered independently the emergence of disinfection by-products during drinking-water chlorination (see review Hrudey (2009)). Also from 1974 on, a new legal basis in the USA (Safe Drinking Water Act, SDWA) opened the possibility of starting and implementing an extensive list of chemical (and microbiological) parameters, containing health-based maximum contaminant level goals (MCLGs) and technical maximum contaminant levels (MCLs), the latter deemed to be as close to the corresponding MCLG as feasible, the final value depending on the result of the risk/benefit-approach (Cotruvo and Regelski, 1990). A milestone away from the concept of a strictly scientific risk/benefit approach when assessing the presence of constituents, residuals, and contaminants in the drinking water system was set by the World Health Organization (WHO) in its 1984 guidelines for drinking-water quality WHO decided then to depart from its ‘‘previous practice of prescribing international standards for drinking-water’’ in favor of the ‘‘desirability of adopting a risk-benefit-approach (y) to national standards and regulations’’ (WHO, 1984a). This was not to state that any reduction of a chemical agent should be thought of being accomplished when reaching an up-to-then lower health-based level. Rather the contrary is meant, as

381

WHO in 1984 stated this some pages later: ‘‘Although the guideline values describe a quality of water that is acceptable for lifelong consumption, the establishment of these guidelines should not be regarded as implying that the quality of drinking-water may be degraded to the recommended level. Indeed, a continuous effort should be made to maintain drinking-water quality at the highest possible level’’ (WHO, 1984b). This continued line of WHO to encourage resource protection and early surveillance instead of simply looking at compliance with strictly health-related values was further strengthened after that. This becomes evident when going over its 1993 published (especially part 4 of volume 1 of its most recent version) guidelines for drinking water (WHO, 2008). Its concept or plan of ‘water safety’ (WSP) takes compliance with any listed health-based guide values only as a minimum daily quality goal, thus bringing together the limit-value approach on the one hand and a preventive, sustainable risk-based water supply management approach with independent surveillance on the other hand. The concept followed by the author of this chapter to strictly differentiate between criteria of functionality, preventive avoidance, and adversity according to the origin of the agent when choosing one for chemical drinking-water standards fits perfectly with such precautionary approach.

3.14.2.3 Objectives and Goals of Protection 3.14.2.3.1 Objectives of protection In order to optimize legal compliance of drinking-water standards, it is good regulatory practice to allocate different parameters to different partial objectives of protection or points of surveillance in or at which the respective parameter(s) should not be exceeded: 1. The raw water ¼ drinking water before treatment. Parameters to control a. nontreated or untreatable geogenic constituents and b. environmental contaminants. 2. The finished water ¼ drinking water directly after treatment at the exit of the water-works facility. Parameters to control a. geogenic constituents possibly not properly eliminated by treatment, b. residuals from inevitable chemical treatment steps and their technical contaminants, and c. DBPs from drinking-water disinfection or oxidation in the water-works facility. 3. The delivered water ¼ drinking water at the handover point just before the water meter. Parameters to control a. delayed formation of DBPs in disinfected drinking water during its transport and distribution and b. inputs from transport and distribution (corrosion products, contaminants, etc.). 4. The tap water ¼ drinking water at the consumer’s faucet. Parameters to control a. corrosion products and contaminants from the domestic installation and its armatures.

382

Drinking Water Toxicology in Its Regulatory Framework

Summing up, limit values for chemical parameters in drinking water define and prescribe legally the minimum quality of an artificial or natural water serving as a source of raw water for drinking-water production, the maximally admissible change of quality during treatment and distribution of drinking water, or minimal improvements of quality by treating drinking water. Many countries apply or recommend competing quality goals on surface water in the form of scientifically based maximal contaminant or similarly defined regulatory levels. Very often they miss considering drinking-water as an objective of protection, even though in the precautionary context of drinking-water regulation, such standards could be set much lower than just based on ecotoxic potential. This is especially true for the ever-increasing number of relatively nontoxic, but highly polar and persistent environmental contaminants such as pharmaceuticals from human use, antibiotics, iodinated X-ray contrast media, veterinary pharmaceuticals, polar herbicides and metabolites, aminocarboxylate complexing agents, amines, and surfactants (Reemtsma and Jekel, 2002). Their special and environment-friendly affinity to the aquatic compartment goes in parallel with their potential to persist or accumulate continuously in reused water, although for economic, ecologic, and social ethics of water supply such reuse should gain continued significance and acceptance on a global scale (see Sections 3.14.1.1 and 3.14.8).

3.14.2.3.2 Goals of protection The legal concept limit value is characterized not only by its clear liability but also by the fact that different parametric values, depending on the chemical parameter under question, can be allocated to very different goals of protection. These correspond to different maximal values based on science, best technological treatment and avoidance of exposure, or just on acceptance of a contaminant in the object of protection, for example, drinking water. Human health is in fact an outstanding, but even within systems of drinking water supply, only one of several supposable goals of protection. Therefore, it seems reasonable to set limits on values or standards for chemicals in drinking water to support its original purity, given that original (or natural) purity of raw water for drinking-water production and use means at least its long-term compatibility with human health. Such standards should describe as exactly as possible the actual knowledge and technology to ‘‘maintain drinkingwater quality at the highest possible level’’ (WHO, 2008a). This means that standards for drinking water are optimally to be set as close as possible down to levels of best possible technical avoidance or natural background levels. Such levels normally are distinctly lower than those corresponding to human adverse effect thresholds or additional risks. Depending on parameters and protection goals is not only the option to a. protect human health, which is to be stressed for defining a numerical limit value, and also options for defining possibly lower parametrics than health-based values as there are b. indicators for optimal technical use and functioning of functional chemicals,

c. standards to ensure optimal technology for preventive avoidance of useless contaminants, d. indicators and goals of aesthetic quality (odor, taste, and purity), e. standards (indicators) to ensure resource protection and aquatic life, and f. standards (indicators) designed to facilitate or optimize quality surveillance. This chapter not only renders more precisely the regulatory framework for a toxicology of drinking water, but also intends to explain the scientific approach behind this expression to derive (a) health-based standards and to compare their criteria with those of stricter standards for (b) optimal technical function and (c) technical avoidance. A prominent example of a relatively high yet functionally minimal concentration is that of the disinfectant free chlorine (or of other less prominent disinfectants) in drinking-water. Chlorine-based disinfection of drinking water would not be sufficiently effective anymore below a minimum of 0.3 mg l1 free chlorine. However, if the raw water for drinking-water production and its distribution network are proven to exhibit steadily an unobjectionable microbiological and technical state, disinfection even for transport would not be required and the accepted minimal concentration of chlorine would accordingly be ‘zero’. Another example for accepted functional, and insofar inevitable, minimal concentration levels in the drinking-water system is DBPs, residuals from metal corrosion, or contaminants from materials in contact with drinking water, whose presence in drinking water is acceptable only if the processes, metals, and materials within the system appear technically indispensable or even irreplaceable. Levels of residuals and contaminants are deemed to encroach on health-based maximal levels only in technically inevitable or unfavorable situations. Otherwise, technical norms to ban their use, to constrict it on ALARA-compatible physicochemical or technical conditions to optimize a treatment process, should be implemented. Drinking water with a quality in accordance with standards as definable by options given earlier will always be of unobjectionable quality from the point of not only health but also (d) esthetics. However, they are only insofar hygienic (or precautionary) standards for drinking water as the criteria to derive them are a consequence of applying the earlier mentioned three-dimensional rule of environmental hygiene (Section 3.14.1.2). Standards for (e) resource protection and of aquatic (wild) life and (f) optimization of surveillance are not considered in this chapter.

3.14.2.4 Timescales to Protect Goals of Protection 3.14.2.4.1 Precautionary standards (enduring protection of human beings and drinking-water resources) Precautionary standards define drinking water as being ‘‘as pure as naturally possible or technically feasible.’’ They allow to compare desire and reality by reporting real concentrations on the more or less technical ALARA principle. The applicability of ALARA should be agreed upon by all stakeholders as

Drinking Water Toxicology in Its Regulatory Framework

the result of discussions on the basis of the environmental rule (Section 3.14.1.2). They stand for maintaining an optimal purity of drinking water far below health-based maximal values and for protecting the distribution system at a minimum of expenditure for continuous surveillance. In Germany’s drinking-water ordinance the ALARA principle is present in the form of a flexible rule, called ‘precept of minimization’. It requires keeping contamination of drinking water as low as possible and even below health-based maximal values, if this can be reasonably achieved. Precautionary standards for anthropogenic environmental contaminants may deviate upward from zero but be distinctly lower than health-based values if the transfer of the contaminants under question into aquatic environments was or seems to be technically inevitable. Examples for drinking water are the European Union (EU) standards for tetra-plus trichloroethylene (10 mg l1), four polycyclic aromatic hydrocarbons (0.2 mg l1) or 0.1 mg l1 per pesticide in drinking water.

3.14.2.4.2 Scientific standards (guide values for lifelong protection of human beings) Scientific standards are designed to protect

• • •

human health (for details see Section 3.14.4), or distribution network/installations from technically adverse chemical agents, or drinking water from tasting or smelling unusual/disgusting.

They are conceived by stressing on 1. either the basic dogma 1 on the presence of an effect threshold for unacceptable adverse effects (see Section 3.14.4.3.1), or 2. basic dogma 2 on the absence of an effect threshold with an accordingly allocated and accepted additional risk or annoyance over background (see Section 3.14.4.3.2). If a parameter was decided to be regulated under dogma 1 and the heading scientific standard, its numerical value is chosen in a way that it is virtually (quantitatively) identical with the starting point of the most sensitive positive dose/adverse response curve in persons or installations/distribution systems of distinctly higher-than-average sensitivity during repeated, preferentially, lifelong exposure. Regulatory levels under dogma 2 exist only for humans, but neither to protect materials nor to protect humans from taste and odor. In drinking water, such levels are generally accepted to correspond with additional risks of 105 106 for contracting an irreversible adverse effect over background incidence in the course of a 70-year life-span exposure.

3.14.2.4.3 Remedial standards (action values to protect from shorter-than-lifetime exposure) Remedial standards for drinking water are designed for being applied on shorter-than-lifetime exposure. They protect the goal of protection from adverse effects or from a higher-thanaccepted risk only during that shorter time period. In principle, they can be derived not only to safeguard human health but also to address the functional safety of materials and systems.

383

The rapid availability of remedial standards for drinking water is critical for informed decisions on whether the continuous access to a moderately or short-term contaminated drinking-water supply would create more or less risks for individual health and public hygiene than disconnecting or closing that supply. Even in those extremely rare situations where a drinking water might be contaminated up to an acutely toxic level, its use for most purposes of personal hygiene may not need to be excluded or delimited. Especially if weighing the risk from enhanced exposure to DBPs toward those from neglecting necessary disinfection of the same water, a correct decision will always vote in favor of continuing disinfection, hence supporting minimization of DBPs not by challenging disinfection but by its technical optimization. As far as human health goes, the concept of remedial standards means to allow exceeding standards for lifelong protection if sanitation, repair, or other remedy can definitely be expected to take place in less than 70 years. However, with actual levels of contamination being higher than regarded as safe for lifetime exposure, the risk for the health of the consumers must be reasonably excluded with equal dependability also in such situations. Compliance with any suggested protective measure in the form of formal water notices for the public during drinkingwater contamination incidents strongly depends on clear and pragmatic semantics of such notices. In a study with 107 undergraduate participants, warnings such as ‘do not drink’ or ‘boil and drink tap water’ revealed a number of partially risky behaviors and several strange correlations between independent behavioral patterns of one and the same person (Rundblad, 2008). Guidance to find remedial standards for acute to shortterm (r24 h up to a few weeks) exposure to spills from emergencies was recently published by WHO (2008). A remedial standard represents a defined aliquot in 2 l of drinking water of an acute reference dose (ARfD). An ARfD is the quantity on body mass basis of the amount of a spilled chemical (mixture) that can be ingested maximally within the exposure period without appreciable health risk. For pesticides, ARfDs have been published by the Joint FAO/WHO Expert Committee on Food Additives and short-term health advisories for chemical contaminants in drinking water were produced by US-EPA. A pragmatic approach to define longer-term allowable but distinctly less than lifelong exceedance may grant up to 100% of an ADI or TDI (lifelong tolerable or acceptable intake ¼ exposure per day (Section 3.14.1.2)) in the daily amount of drinking water while minimizing total exposure as soon as feasible below the ADI or TDI, respectively. Short-term exceedance of an ADI or TDI is considered not causing appreciable risk to health. This also applies to sensitive groups of the population as long as their exposure is not high enough to render toxic endpoints different from those considered for deriving an ADI or TDI more critical for the safety assessment than the latter ones. The allowance of an ADI or TDI for drinking water should be quantified by expert judgment on the basis of all modes and pathways of exposure to the same or similar compound(s). Such judgments would also have to consider the possibility of significant differences between intestinal

384

Drinking Water Toxicology in Its Regulatory Framework

resorption rates for chemicals, especially of dissolved metal ions, when comparing and combining their exposure from drinking water with that from food. A formalized and quantitative method or procedure to define allowable exceedance of lifetime tolerable exposure via drinking-water for up to several years, if ever necessary for not being strained to close a drinking-water supply, is practiced in Germany since 2003 (Federal Environmental Agency of Germany, 2010). The proposal for threshold compounds starts from quantifying a minimal lifelong unsafe exposure halfway between a lifelong (virtually safe) body dose (Bd) and the experimental or epidemiological no-adverse-effect level (NOAEL) which was chosen as the point of departure (PoD) to extrapolate on that Bd. The interpolation of a hazard-linked dose (HLD) between the Bd and the PoD harks back on widely accepted methods and conventions of toxicological risk quantification. The interpolation factor (IF) is quantitatively defined as the square root from exclusively those extrapolation factors (EFs) having been used as well to extrapolate experimental or epidemiologic data to the human target population for quantifying the Bd. The HLD equals IF  Bd. Both IF and EFs thus are equally conservative. They are always open for corrections in concert with improvements in quantifying the PoD. The larger the percentage by which the HLD encroaches on the safety margin between PoD and the Bd, the more completely the extrapolated data resemble human data. At the same time, IF decreases with data coming closer to data from humans (Dieter and Konietzka, 1995). An analogous procedure to define less than lifelong acceptable remedial standards was developed in Germany for nonthreshold carcinogens as well. It starts from proposing to accept over a lifetime up to 5-times higher than accepted additional cancer incidence IZ (see Section 3.14.4.3.2) per contaminant. The calculated IFs to allow for a 3- and a 10-year exceedance of a lifelong safe health-based maximal value (guide value, GV, see Section titled ‘Health-related chemical standards for drinking water’) are 6 and 17, respectively. The corresponding remedial less than lifelong tolerable exposure levels via drinking water are especially conservative insofar as they consider also the possibility of a higher sensitivity of small children (rapidly growing organisms) when compared to adults toward primarily genotoxic carcinogens (Schneider, 1999).

3.14.2.5 Special Aspects to be Considered when Setting Standards for Drinking Water 3.14.2.5.1 Standards for accepted chemicals in drinking water Legally defined and politically granted levels of exposure or contamination are not set arbitrarily but originate from accepted and realized proposals to benefit from the use of synthetic chemicals and natural resources with their constituents. There are many compounds whose presence in food or drinking water is linked with an intended functional benefit. They are accepted there in minimal values either

•

because below such minimal level they would not be functional any more or

•

because a raw water resource would fall out of favor for drinking-water production in case the contaminant is not accepted at a certain minimal level.

Examples are

•

•

•

Acceptance of geogenic constituents such as Pb, Cd, As, and F of raw water for drinking-water production up to healthbased maximal values but factually very often much lower natural or technically treated levels. Acceptance of anthropogenic residuals in drinking water up to a technical or health-based upper limit resulting from its transport through (or packaging in) pipes and armatures being more or less susceptible to corrosion. Acceptance of disinfectants in drinking water (and their DBPs) as far as their presence is linked indispensably to disinfection or other treatment to minimize microbial risks from drinking water and its corrosive properties.

3.14.2.5.2 Standards to protect from adverse effects of chemicals in drinking water If only functional but neither potentially dangerous nor technically avoidable contaminants of drinking water existed, the (technical) world of drinking water would not have the concept of risk society. However, any chemical load of drinking water provokes questions not only concerning its functionality, but also on its avoidance and drawbacks. Regarding the latter point, the following criteria serve to find the precise maximal numbers for tolerable contamination:

• • •

potential drawback of a chemical for the drinking water to adversely affect consumer’s health; potential of a chemical to damage certain functions (e.g., pumps) or materials in the course of drinking-water extraction, treatment, and distribution; and potential of some chemicals to influence negatively the esthetic (taste and odor) properties, perceived by the consumer not as an annoyance but as a threat to health.

There are no other goals of protection from adverse effects to be covered by standards for drinking water, especially not for the protection of natural aquatic communities from pesticides, despite the fact that many consumers would be glad if their drinking water is safe not only for its original purpose but also for aquatic communities in their private domestic aquariums, ponds, or backyard pools. Drinking-water standards are also not a good measure for standards for protecting groundwater from contaminants in leachates from abandoned waste sites. Health-related standards for drinking water. A health-related standard or health-based GV for a chemical in drinking water is the maximal concentration of a chemical agent in mg l–1 drinking water that would not give any reason to be concerned about the consumer’s health if he or she regularly ingests such water (Section 3.14.4). This definition is valuable for any duration of exposure between very short term (a few days) to lifelong (Z70 years) and for which the standard under question was factually quantified (see Section 3.14.2.4.3). When going over the limits of a GV, there is an increasing health concern and finally danger when exceeding the HLD (see Section 3.14.2.4.3) with risk of an adverse health effect to occur such as if ingestion paths such as air or food contribute

Drinking Water Toxicology in Its Regulatory Framework

significantly (80% and more) to regular intake of the same or toxicologically similar compounds. On the other hand, regular exposure to a contamination below its GV is not likely to give any reason for health concerns via drinking water. Standards to protect materials and technical functionality of drinking water systems. These standards are designed to protect the whole system of drinking-water extraction, treatment, and distribution or parts of it from damage by water loss from corrosion, pump damage, or any technical failure. Most common causative agents are geogenic constituents of raw water such as ionic/particular Al, Fe, or Mn, and salts and protons as well. Manganese in drinking-water systems, such as iron at more than 0.2 mg l1, may lead to the accumulation of deposits and occasional turbidity if permanently present at more than 0.05–0.1 mg l1. Unspecific chemical and physical parameters such as conductivity, dissolved organic carbon, absorbable organic halogens, or even temperature protect the functionality of the system insofar as they make it easier to retrieve points of technical failures or indicate suspicious changes in raw water quality or composition. Such parametric values are, very often, to be set distinctly lower than would be possible on the basis of human health considerations. The technical risk assessment on which they are based upon does not use default (safety) factors but quantifies any yet acceptable technical risks in the form of technical parametric levels. These are defined by combining results from technical small-scale models and worst cases but yet realistic predictions for technically adverse or undesired chemical behavior. The aim is to safeguard the usability of a resource or system at minimal risk from either (failures of) technical treatment or technical damage. Geogenic and biogenic constituents, although nonfunctional and also harmless to (components of) the technical system, seem tolerable up to health-based maximal levels, especially if treatment is difficult and a resource with a lower natural background is not readily available. Esthetic-sensory standards. Some of these standards (taste, odor, color, and turbidity) are surrogates or indicators for less evident but more critical quality characteristics of the consumers’ tap water. Their noncompliance may correspond to still more unfavorable-to-adverse consequences for the system or the consumer than their own exceedance. Other esthetic-sensory thresholds or parameters defend the taste and odor neutrality of drinking water and hence its acceptance by the consumer, not as surrogate parameters but by their own nature. Sulfate in the presence of magnesium from 250 mg l1 on gives a bitter taste to the water, whereas aluminum above 0.2 mg l1 leads to an astringent taste and causes turbidity and color in the presence of iron. Manganese stains sanitaryware and laundry if exceeding 0.05–0.1 mg l1. Sodium chloride may taste salty at more than 400 mg l1. More parameters and their thresholds from which they could impair odor and taste of an otherwise unobjectionable drinking water have been described by WHO (2008). The sensual perception of esthetic-sensory parameters has highly subjective denotations (Doria et al., 2009). Accordingly, standards can be derived only from experience with humans, not from animal experiments. The slope and range of the dose/response curve from which a standard may be derived should also include distinctly higher-than-average sensitivities.

385

The most common standards worldwide that impair the esthetic (odor and taste) quality of drinking water from 5 to 10 mg l1 on are alkylated aromatics (BTEX) and alkylated t-butylethers, such as methyl-t-butylether (MTBE) or its ethylanalog (ETBE). Simple expectations on (sensorially undetectable) purity of drinking water may be esthetically motivated as well. This pertains especially to contaminants possibly reaching the consumer with reused water and whose mere origin or known presence in his or her drinking water above a certain but by far not yet noxious level could merely be perceived as disgusting, for example, pharmaceuticals and their metabolites (see Section 3.14.8).

3.14.3 Panels and Institutions for Setting Drinking Water Standards 3.14.3.1 National National homepages or documents with extensive lists of parametric values for drinking water and criteria on which they are based are available from Australia (including the free quarterly newsletter Healthstream, responding to many actual questions relating to drinking water for an international expert readership), Health Canada, Japan, the New Zealand Ministry of Health, the European Union (EU-guideline 98/83/EC), and the US-EPA/Office of Water.

3.14.3.2 International The guidelines for drinking-water quality of the World Health Organization provide health-based guideline values for more than 100 chemicals (WHO, 2008). The guidelines are advisory in nature; for most countries, however, these guidelines provide the scientific point of departure in deriving national or supranational drinking-water standards. The guidelines, which are subject to rolling revision and ongoing update in response to new evidence, strongly encourage adapting standards to national priorities and to socioeconomic, cultural, and environmental contexts. For countries that do not have the resources to sustain their own regulatory-toxicologic capacities for deriving health-based drinking-water standards, the guidelines are the point of reference for standard derivation.

3.14.4 Defining Standards to Prevent Human Health Risks from Drinking Water 3.14.4.1 Qualification of Risks – Critical Toxic Endpoints The possibility of adverse effects of chemicals on human health is of highest regulatory interest if they are the result of permanent (chronic) exposure, for example, via drinking water as a medium for a steady lifelong consumption, often without the possibility or freedom of choice. Possible effects may occur either concomitantly with drinking water ingestion or only after a period of more or less irreversible substance or effect accumulation from earlier plus present exposure. Environmental contaminants, if using all paths of ingestion (oral with food or drinking water, dermal, and pulmonal), reach their target organ(s) by all these paths through the bloodstream, the so-called ‘central compartment’. They are

386

Drinking Water Toxicology in Its Regulatory Framework

then defined to act systemically, in contrast with those acting preferentially or exclusively along their port or route of entry into the body, then comprising the olfactorial pathway as well with its nerve endings being exposed directly to the environmental air. Critical toxic effects to be considered by proactive regulation of chemicals which may reach environmental compartments include

• • • • •

early irreversible damage (including cancer) to important organs, in their early functional and histopathologic manifestations; disturbance of normal growth and behavior; exposure-induced derailments of metabolism; teratogenic effects: impaired male and female fertility, sexual development and reproductive behavior with special consideration of acutely toxic exposure windows; and in general, any kind of early biochemical abnormalities (including endocrine disruption) as observed down to the lower end of a dose/response curve in an animal experiment or preferentially an epidemiologic setting.

Information on effect quality and potency is quantitatively exploitable only if exposure was performed experimentally (under reproducible conditions), hence preferably in animal experiments or in vitro tests which imitate the in vivo exposure as close as possible. Effect qualification and especially quantification by using human data is preferable if exposure is known to have caused the effect under question.

3.14.4.2 Groups of Compounds of Specific Interest for Drinking Water All these critical or toxicologically crucial endpoints of chemical toxicity are subject to different national and international procedures of proactive toxicological evaluation of chemicals during their examination for intended or functional but environmentally open use. Admissions are regulated or refused on the basis of their toxic potential or their potential to persist, accumulate, or dissipate in the environment. Hydrophilic properties are always weighted in favor of admission and there are, at first glance, good reasons to do so. By this way, however, drinking water not only remains the target compartment for more or less hydrophilic and toxic old chemicals but also turns more and more into the role as a sink for such environmental contaminants which slip through modern registration processes, because they arise only somewhere in the environment as metabolite(s) of an environmentally neutral and, accordingly, admitted parent compound (Schwarzenbach et al., 2006; Barnes et al., 2008). Such new analytes, albeit usually not very toxic, are often very hydrophilic and refractive to further environmental degradation. This supports their continuation in aquatic environments and drinking water and their circulation therein. Other new analytes, also called chemical transformation products, may even arise from different antecessor compounds only in the course of drinking-water chlorination or ozonation (Reemtsma and Jekel, 2002). A prominent but up-to-now unique proof of a very toxic transformation product is N-nitroso-dimethylamine (NDMA), a probable human carcinogen, which can be formed in raw water from dimethylsulfamide

(DMS), a so-called nonrelevant environmental metabolite (Dieter, 2010) of the agricultural fungicide tolylfluanide. NDMA was formed while treating DMS-containing groundwater with ozone for drinking-water production. The presence of catalytic traces of (geogenic) bromide is required for this surprising but highly undesired reaction (Arnold et al., 2010; Schmidt and Brauch, 2008). However, not only nonrelevant metabolites of pesticides on a long-term scale seem to be relevant for drinking water hygiene, but also pharmaceuticals, their metabolites, and transformation products. They occur worldwide at up to several mg l1 in surface water used for drinking-water production (Reemtsma and Jekel, 2002; Ternes and Joss, 2006; Bohannon, 2007; Cooper et al., 2008). Other groups of compounds with special relevance for drinking water are mainly naturally occurring organic constituents (as measured in the form of dissolved organic carbon (DOC)), environmental contaminants from industrial sources and human dwellings, biogenic toxins from cyanobacteria, residuals and DBPs of chemical water treatment, and corrosion products or contaminants from materials in contact with drinking water (WHO, 2008; Table 8.1). The experimental–toxicological material or database (see Section 3.14.4.3) to evaluate their toxic potential for humans such as by WHO is often missing or very limited. This very common regulatory toxicological situation asks for a remedy in the form of a toxicologically motivated surrogate approach to evaluate provisionally the presence of drinking-water-relevant chemicals on an incomplete-to-missing database under the aspect of preventive healthcare until the database improves (see Sections 3.14.5.3 and 3.14.6.3.2).

3.14.4.3 Risk Quantification The following is an extremely short basic introduction to what may be studied as regulatory toxicology in more detail and some variants either at WHO (2008) or with pertinent introductions into methods of human toxicological evaluation at Greim and Snyder (2008) and Leeuwen and Vermeire (2007) .

3.14.4.3.1 Chemicals exhibiting systemic effects with threshold In order to calculate a lifelong tolerable and health-related parametric concentration of a compound in drinking water, it is essential to know 1. the lifelong tolerable body dose Bd (discussed in this section), and 2. the percentage of Bd which would realistically be attributed to the daily intake of drinking water (see Section 3.14.4.3.5). In regulatory toxicology, the threshold concept is a rather pragmatic and not a scientific concept insofar as it refers not to the single cell as the smallest and most sensitive functional unit of life but to most sensitive subunits (cells) of population. In this sense, the Bd of the chemical under question is its population-based adverse effect threshold in the human target population as evaluated on the basis of scientific (human, animal, in vitro, and statistical) data by regulatory

Drinking Water Toxicology in Its Regulatory Framework

toxicologists. It is the regulatory-toxicological surrogate for the actual but scientifically mostly unknown adverse effect threshold in such a target group and is derived in a way to be never higher than the same threshold had it determined scientifically. In most cases, the Bd can even be assumed to be much lower, but in some unpredictable cases real threshold and Bd may be not only conceptually but also numerically identical. Lower doses than Bd by definition exert no observable adverse effect in any one of the target individuals (basic dogma 1). In contrast to a regulatory drinking-water level, given mostly in mg l1, the Bd is calculated or defined in mg of the chemical agent per kg body mass (mg kg1 bm). As such, and different from any environmental standard, the Bd is not directly accessible for regulatory interventions, except that each individual would be ready or forced to permanently equip himself or herself with a personal dosimeter. Toxicological information of sufficient quality to derive a Bd may be gained only in a few cases from an epidemiological (human) database, for example, from uptake of a chemical directly with drinking water. A main drawback of retrospective human studies is the difficulty to gain reliable and at the same time exact information on level and length of exposure due to its spatial and temporal variability (Legay et al., 2010). Therefore, the required information has mostly to be drawn from animal experiments controlled for exposure. Studies using water instead of food, air, or administration by gavage as exposure path are to be preferred. As far as the solubility, speciation, stability of the noxious agent, the frequency of daily exposure, and its resorption in the digestive tract are concerned, they reflect the real human exposure situation more closely. Even then, the uptake of drinking water and hence the noxious agent or chemical species must always be estimated as precisely as possible. Exposure paths different from water often exhibit distinctly lower rates of resorption than drinking water as has been shown for some heavy metals and/or their chemical speciation (Yokel et al., 2006). If there is no experiment available using water as a path of exposure, it may be allowed to change to systemic exposure via food or inhalation in exposure via water. It must be mentioned here that experiments, especially with exposure to lipophilic agents by gavage, are not suited as a basis to characterize and quantify their toxic potential in drinking water. Exposure via corn oil, the usual solvent in such study, is not continuous like from continuous drinking-water uptake, but fluctuates batch-wise between high and low. This may change, as observed with chloroform, the position and slope of the dose/response-curve for cancer due to precancerous or other cytoxicity so drastically, that the toxicokinetic and toxicodynamic behavior of the chemical agent should not be extrapolated from such studies on humans and their exposure by drinking water (see review Hrudey, 2009). The first section of the toxicological evaluation opens by defining an experimental NOAEL (in mg per kg of the experimental animal’s body mass). This NOAEL is the highest, yet ineffective, experimental dose of the chemical agent. The animal species or strain chosen for the experiment should not only exhibit a higher-than-average sensitivity but also be as similar as possible to conditions of exposure, physiology, and sensitivity of the human target population. It is also preferable

387

to define a NOAEL by using the benchmark approach, a standard procedure to assess, if ever possible, data from several studies exhibiting sufficient experimental quality and consistent toxicological information (Murrell et al., 1998). A guidance document containing recommendations for the application of the benchmark approach was developed and several default assumptions are proposed (Kalberlah and Hassauer, 2003). For the applications considered in this comprehensive report, the comparison of both approaches when using the concept proposed by the report generally does not show great differences between the benchmark and the NOAEL approach in terms of the resulting POD. An adverse effect, called ‘delayed toxicity’ then, may not always show up concurrent with exposure, but weeks to years after its onset. This phenomenon has been known for a long time as a characteristic of some organophosphates, but also needs to be addressed as a possible mode of action of endocrine disruptors, for example, if they are ingested at critical exposure windows during pregnancy or puberty. A biomimetic mode of action may set a hormonal switch to exhibit adverse results only much later as was hypothesized, for example, by Hens (2007). The second section of the total evaluation procedure starts from the NOAEL agreed upon either by expert judgment or from a benchmark procedure. Sometimes, an NOAEL needs to be extrapolated by means of a specific extrapolation factor EFa between 3 and 10 from an experimentally observed lowest-observed-effect level (LOAEL). Another EF (EFb) of mostly 3 may save subchronic data for an assessment by extrapolating them on chronic exposure. The NOAEL eventually chosen is the procedural PoD from which a Bd is then defined. This is done in two more steps c and d within this second section. They extrapolate, starting from the PoD, the experimental data on humans by dividing the NOAEL by two single extrapolation factors EFc and EFd, one for interspecific (animal-human) variability of experimental–toxicological results, the other for the intraspecific variability between normal and sensitive humans. The numerical amount of both EFs varies between 1 and 10. The final amount of each depends on the quality of the database and its similarity with toxicologically critical parameters of human physiology and metabolism. Each factor may be subdivided in partial factors of 2.5–4 to cope separately with extrapolation of the agent’s experimental kinetic (uptake, distribution, metabolism, accumulation, and excretion) and dynamic (mechanism and mode of action) characteristics on humans. A compact outline on motivations and designation of safety factors (SFs) and extrapolation factors, respectively, was published by Ritter et al. (2007) and completed with a commentary by Konietzka et al. (2008). The actual (but scientifically unknown) effect threshold within the sensitive (target) population is assumed to be positioned somewhere within this safety margin EFc  EFd below the NOAEL, whereas the Bd equals by definition its lower limit and was calculated to resemble this intended result as close as is scientifically traceable. Only in some unpredictable (worst but still realistic) cases, Bd would be numerically identical with the (unknown) actual and then very low effect threshold. Some expert groups propose a fixed additional SF of 10 to cope with incomplete or questionable experimental data,

388

Drinking Water Toxicology in Its Regulatory Framework

missing of tests on very critical endpoints, or doubts about the correctness of a performed extrapolation. However, as a rule, higher than 3000 total factors (as a product of all applied EFs and the SF) will signal such a high incompleteness of the database, that provisional evaluation by a default method (as outlined in Section 3.14.6.3.2) will become imperative. The scientific acceptability of the four EFs as conventionally amounting to 3–10 has been positively underlined by the screening of a large toxicological–experimental database worldwide being used for deriving regulatory-toxicological recommendations (Kalberlah et al., 2003). The only convention missing is how to answer the question as to what percentage of the subgroup of the population identified as sensitive should still be protected by not exceeding a Bd. Such extremely sensitive individuals may, if at all, be found among that subgroup of the population carrying congenital anomalies of differing severity and representing about 2–4% of the general population (Nicolopoulou-Stamati et al., 2007). As seen by science, the regulatory-toxicological statement that Bd ¼ effect threshold in the most sensitive target population represents only a working hypothesis lacking real probative force. However, the mode of deriving a Bd guarantees its scientific acceptability and consistency as long as it does not slip into contradiction with either scientifically proven human data or animal data, which might be extrapolated without major scientific limitation on humans. In fact, a Bd would be identical with the actual or real effect threshold in a conceptual as well as a numerical sense only if derived on a scientifically unobjectionable epidemiologic (human) data basis.

3.14.4.3.2 Chemicals assumed to exhibit threshold-free systemic effects The most extensively described effect ascribed to this regulatory-toxicological category of chemical agents is cancer. The individual risk of contracting cancer by primarily DNAreactive agents is considered as being governed by chance (basic dogma 2), while other carcinogens may be regulated as threshold compounds if following US-EPA (2005) or the SCOEL of the European Union (Bolt, 2008). If according to toxic mechanism, the absence of a threshold is to be assumed, the carcinogenic potential at a given exposure level can be represented or given the form of a probability number for an additional population risk. Therefore, there is no virtually safe dose attribute of a chemical in the form of a Bd if it acts along a nonthreshold probability and not a thresholded discontinuity. Instead, a socially accepted and/or politically tolerated incidence I ¼ additional number X of cancers per chemical exposure unit and predefined size Z of population can be calculated. The regulatory equivalent for I is that concentration of the chemical in the compartment of regulatory interest (e.g., drinking water), which in the given exposure scenario corresponds with the accepted or tolerated value of I. In order to calculate a concentration acceptable for lifelong exposure, it is inevitable to anticipate how many cancers X per mg exposure per day and kg bm1 might be expected in the course of 70 exposure years. X is known or calculated preferably from epidemiological studies or, if those are missing, from high-dose long-term animal experiments and

extrapolating them on real world ¼ low to very low dose human exposure scenarios. The resulting number gets hypothetical in the same rate as knowledge on the underlying time course and biochemical mechanism of irreversible molecular interaction between the chemical and DNA turns out to be speculative. The only inevitable precondition for such a number is the principal possibility of adverse effects at very low, if not molecular, doses. WHO (2008), in its drinking-water guidelines, uses for calculating X the most conservative of all possible models, the linearized multistage model (LMS). It delivers higher numbers for X than all other models and describes therefore an absolute upper limit of risk per exposure unit. It is hence correct to assume that any risk calculated by using the LMS model is probably higher than the real-life risk from daily exposure. However, similar to the situation with Bd and its hypothetical identity with the factual threshold, X is supposed by the regulatory-toxicological community to reflect the factual risk in some existent but unpredictable cases. This is why X is a regulatory-toxicological surrogate for the unknown factual risk as long as scientific data do not support a lower, more conservative number. X, the additional risk, is given per unit of exposure in (mg d1 kg bm)1 and is called slope factor by WHO (2008). If real exposure in (mg d1  kg bm) is called Y, the individual additional risk R to contract cancer equals R ¼ X  Y (R has no dimension). If a total of Z persons are exposed to the same daily amount X of a carcinogenic and primarily DNA-reactive chemical during the course of 70 years, the total number of persons to contract the prognosticated cancer within this population would be IZ ¼ X  Y  Z. For means of comparing risks from different exposure situations, Z is often set as Z ¼ 105. The worldwide accepted and politically tolerated additional risk IZ by ingesting one chemical contaminant during 70 years with drinking water seems to amount from 105 to 106. In order to have Y available in its unit (mg d1  kg bm 1) and hence to enable calculation of IZ from the chemical’s concentration c ( ¼ a (mg l1)) in drinking water, it is at first necessary to multiply c by V ¼ 2(l d1  70  kg bm 1), the body mass-normalized daily drinking-water consumption, so that Y ¼ c  V ¼a  2/70 (mg d1  kg bm 1). Table 1 lists a number of primarily DNA-reactive and at the same time polar (water-soluble) chemical carcinogens with unit risk values as applied by WHO (2008).

3.14.4.3.3 Local effects on humans of chemicals in drinking water Some chemicals possibly present in drinking water exert their adverse effects not systemically but locally, close to the port of entry (e.g., on skin, mucous membranes, lung, stomach, and intestine). Their regulatory-toxicological evaluation is not possible along the models of tolerated body dose or respectively accepted additional incidence, although they might be separable according to basic dogmas 1 and 2 by science. Their adverse effect potential depends not on whether a tolerable and weight-normalized body exposure (TDI or ADI) might or might not be reached, but solely on the concentration in the exposure medium (e.g., drinking water) at the point of

Drinking Water Toxicology in Its Regulatory Framework Table 1 Presumably DNA-reactive and at the same time polar (water-soluble) chemical carcinogens with values of X as applied (a) by WHO (2008) and (b) by UBA (2005) Carcinogen

Increase of risk per (mg d1  kg bm1) (slope factor, SF a)

Increase of riska per (mg l1) (oral risk unit, RUob)

(a) N-Nitrosodimethylamine (a) Acrylamide (a) Benzo(a)pyrene (a) 1,2-Dibromo-3chloropropane (b) 2,4-Dintrotoluene (a) Vinyl chloride (a) Benzene (b) 2,4,6-Trinitrotoluene (a) 1,2-Dibromoethane

SF ¼ 3.3

r10  105

SF ¼ 0.7 SF ¼ 0.46 SF ¼ 0.33

r2  105 r1.4  105 r1.0  105

SF ¼ 0.17 SF ¼ 0.066 SF ¼ 0.033 SF ¼ 0.033 SF ¼ 0.023– 0.83 SF ¼ 0.016 SF ¼ 0.011 SF ¼ 0.0056

r0.5  105 r0.2  105 r0.1  105 r0.087  105 0.07  105  2.5  105

(a) 1,3-Dichloropropene (a) 1,2-Dichloroethane (a) Bromodichloromethane

r0.05  105 r0.033  105 r0.017  105

389

water directive is its detection limit (0.1 mg l1). Its compliance presumes technical minimization of ECH leaching materials in contact with drinking water, according to ALARA . The most common parameters with GV for drinking water derived from or taking into account local effects on humans are copper and sulfate. The effect threshold of copper for triggering gastrointestinal complaints is positioned between 1 and 3 mg l1 (Araya et al., 2003), whereas the laxative effect of sulfate in unaccustomed persons seems to start at around 1000 mg l1 (for toxicity of copper as a systemic agent, see section titled ‘Copper’) The discussion on how to quantify a Bd to protect Nipresensitized persons from the systemic allergenic potential of systemic oral nickel exposure seems to have been concluded by WHO (2008) by deriving a systemic Bd based on the (assumed) reproductive toxicity of oral Ni2þ exposure. This Bd (5 mg kg1 bm) and its corresponding GVWHO (20 mg l1) are much lower than any earlier aspersed thresholds for the systemic allergenic potential of nickel in presensitized persons. Levels from 30 mg l1 on have been shown to be devoid of any health risk even for presensitized people (Alam et al., 2008).

a

SF ¼ Population-based probability to contract cancer during lifetime if he or she would permanently ingest 1 mg per day and kg body mass of the respective contaminant. The ratio between the highest and lowest SF is about 600. b Calculated from SF for a person with 70 kg bm1 and a lifelong consumption of 2 l drinking water per day

exposure. The main toxic endpoints to be discussed here are irritating and allergenic as well as local carcinogenic effects. Within the context of personal hygiene, including showering and bathing (target organs: skin and external mucous membranes), the concentration of a critical compound decides directly on the possibility of an adverse effect on up to 2.5 m2 skin surface. When using drinking water for preparing food (target organs: mucous membranes and epithelial cells in mouth and intestine) the initial concentration will be more or less diluted and/or the chemical be bound to or masked by body fluids and organic materials absent from drinking water. It is possible to examine in animal experiments whether local effects might be absent below a threshold concentration and what would be its numerical value. There are, however, huge differences in sensitivity between animals and humans when comparing different agents. It is therefore preferable to define the presence and numerical value of a possible threshold on the basis of retrospective epidemiologic studies, single-case observations, and prospective examination of voluntarily exposed persons. Epichlorohydrine (ECH), as an example, actually is a primarily genotoxic carcinogen, but due to its high reactivity it initiates tumors in mice only close to its port of entry, the forestomach, where it is also a strong irritant. As humans do not dispose on a forestomach, this information could not be used to derive a GV with the LMS model. Instead, WHO (2008) preferred to derive a Bd-based GV of 0.4 mg l1 by stressing basic dogma 1 and calculating a total safety margin of 10 000 between a human Bd for ECH and its NOAEL in mice for the precancerous endpoint forestomach hyperplasia. The parametric value for ECH of the present EU drinking

3.14.4.3.4 Effect combinations Drinking water is a potential sink for many environmental contaminants, especially the hydrophilic ones. It is not surprising to hear very often the fear their possible effects on humans via drinking water might combine in an additional, or even synergistic or (theoretically) antagonistic, manner. The possibility of effect combination, be it favorable or unfavorable for humans, is impossible to falsify or verify by experiment in each single case (Cassee et al., 1998). In cases where only retrospective evaluation is asked, the only way to proceed is to assess the exposure situation as it presents locally, considering and evaluating toxicological (kinetic and dynamic) similarity of all contaminants relevant for exposure and structure–activity relationships between them to help assess the mixture instead of its single components. Groups of contaminants being similar with respect to their adverse toxic mechanism and specific binding to (shared) targets can be assessed by means of the addition rule for ‘similar joint action’ or ’concentration additivity’ (Kortenkamp et al., 2009). It asks at first to compute the quotient between the measured and the tolerated or accepted concentration (the latter one being often a health-related indication value (HRIV) (see Section 3.14.6.3.2), but ideally a GV) for each component within similarly acting groups of the mixture. The groupspecific sums of these quotients are called risk index (RI). RI values lower than 1 indicate the absence of appreciable risk, but only regarding the group of compounds as evaluated by means of this RI under the precondition of similar joint action (WHO, 2008). The result of such group-specific assessments is distinctly more conservative than considering and evaluating just the presumably most toxic component of each group or mixture by neglecting all others (Kortenkamp et al., 2009). The parallel determination of several dissimilar groupspecific RI values should not indicate any risk if each RI remains lower than one since, under this condition, none of them would indicate trespassing of an effect threshold (Bd) or

390

Drinking Water Toxicology in Its Regulatory Framework

the possible exceedance of an accepted additional cancer incidence (IZ). An alternative to assess mixtures or combination of effects from a priori unknown chemical agents would be the integral experimental approach. It consists in identifying and comparing cytotoxic or genotoxic potentials from concentrated drinking water samples on the basis of results from standardized biological in vitro toxicity testing. An example of how to proceed was given for two treatment/distribution networks located near the source and at the mouth of River Po in Italy (Maffei et al., 2009). The short-term and long-time cell survival (endpoint cytotoxicity) and the degree of in vitro genotoxicity (as seen by the comet assay and the micronucleus test) of the water before and after disinfection by ClO2 varied considerably between both plants and depended on the drinking water’s residence time in each distribution network. However, the regulatory-toxicological evaluation of a contaminated drinking water with recourse to in vitro tests remains a matter of local or technical circumstances as long as conventions on the endpoints to be tested and how to evaluate the results are not standardized. There is undoubtedly an urgent need to define and standardize in vitro testing for more and new endpoints of crucial importance for the integrity of human health at different stages and conditions Anonymous (2010). Such tests would open the way to identify toxic potentials of hitherto unidentified chemicals and their mixtures on the basis of data from different human cells and tissues long before it is necessary to perform in vivo tests. The basic idea behind would be to define scientifically the presence of safety on a cellular level as long as scientific proof for the absence of toxicity for a population is not required or not yet possible (see Section 3.14.4.3.1, third paragraph). Support to assess mixtures by anticipating effect combinations on a molecular level could come in the near future from computational toxicology (Ekins, 2007).

3.14.4.3.5 Exposure Uptake of (contaminated) drinking water. The starting point for quantifying health-related tolerable exposure to a drinkingwater contaminant as specified by WHO (2008) is to assume lifelong consumption of 2 l of water per day and adult person, and possibly contaminated up to the contaminant’s GV. The amount of 2 l per day and person seems to have a reasonable epidemiological base as shown by data from Germany, where 50% of the 18- to 69-year-old population consumes up to nearly 700 ml drinking water per day and person and 98% up to nearly 2500 ml (Becker et al., 2001). Similar daily consumption was reported from other studies and countries. Variations may result mainly from climatic differences and gender-specific behavior underlying mean living times at home and work. An earlier approach proposes or calculates with only 1.4 l per day and person (WHO/IPCS, 1994). The difference between 1400 and 2000 ml per day per person is insignificant when looking on the considerable inexactness by which any toxicological evaluation is afflicted from which safe health-based standards may be derived. Allocation. There are, besides drinking water, some more important paths of exposure by which contaminants may

reach the human population, very often in much higher contingents of Bd than with drinking water. In order to regulate prospectively the possible appearance of environmental contaminants on virtually safe levels, it is not sufficient to agree upon the (maximal) daily uptake of water per person. It is also necessary to quantify the percentage of the contaminant’s Bd being allowed for in the daily amount of drinking water. Allocation is the term for attributing or allowing a certain percentage of the Bd (see Section 3.14.4.3.1) of a threshold contaminant on the daily drinking-water uptake. Allocation rate has to depend on guessed estimates on exposure fragmentation of the substance (group) under question among food, air, soil/dust, and drinking water; otherwise, there would be no way to control whether future exposure would stay within a predefined Bd. Allocation is not performed for carcinogenic contaminants characterized as primarily genotoxic and whose presence in the environment and drinking water is evaluated accordingly by comparing the additional exposure-controlled incidence I with an accepted IZ (see Section 3.14.4.3.2). The underlying reasoning is by far not that such allocation possibly could not be calculated. The reason is more fundamental in the sense that it is not advisable to assign any official allowances or allocations for primarily genotoxic chemicals because this would eventually legalize their appearance in environmental compartments. Instead the first, although not always successful, choice for managing these contaminants is to defeat their escape or development right where they are formed. Data on how exposure to different groups of environmental contaminants split over different exposure paths can be retrieved at best and with more or less preciseness from retrospective studies. They may in turn be used only conditionally for setting prospective standards. Depending on the exposure fragmentation to be expected factually, the acrossthe-board contribution of daily drinking-water is set usually between 10% and 50% of a Bd, dissolved in 2 l of water. As a rule, high default contributions are allowed for factually inevitable natural constituents of raw water (up to 50%), for DBPs exclusively from drinking-water disinfection and distribution, and for cyanotoxins occurring in drinking-water reservoirs (both up to 80%). Lower allowances (down to 10%) should apply on environmental contaminants such as pesticides or contaminants from industrial emissions. Alternatively, environmental contaminants may be regulated either by a health-related precautionary default, or by even a purely precautionary approach (see Section 3.14.6.3.2). On the other hand, if a Bd was derived from an experiment with intake by food instead of drinking water, the resorption of the chemical from drinking water might differ significantly from the experimental reference value. Especially for metals, resorption rates are often distinctly higher than from food. This may provisionally be accounted for by calculating the corresponding GV using an accordingly reduced allocation. Metals such as copper or manganese, if ingested by infants mainly with tea, juice, or formula prepared with drinking water from concentrates, may attain in the latter distinctly higher fractions than 10% of their Bd. It is advisable in these cases, in parallel with the higher allocation, to give special attention on toxic endpoints of possibly specific importance

Drinking Water Toxicology in Its Regulatory Framework

for weaned infants with regard to target organ(s), metabolism, and mode of action (World Health Organization/International Program on Chemical Safety, 1986) and to limit the content of the dry materials on these and other essential elements accordingly, if they are presented regularly and in significant amounts via drinking water. Concentrations representing more than 10% of TDI in 2 l of drinking water per day are often also observed with persons who, instead of drinking water from the source, consume regularly bottled water rich in minerals. Till date, there is no standard default assumption on the daily per capita consumption of bottled water and hence no international guideline as to how to derive health-based maximal values for bottled water. Such waters are regularly allowed to exhibit distinctly higher mineral contents than regulated drinking water, especially on fluoride, arsenic, and borate. For many minerals (e.g., sodium, chloride, sulfate, and manganese) there are no limitations at all. On the other hand, there are many mineral waters exhibiting much lower content of minerals than they often occur in drinking water.

3.14.5 A Holistic Approach for Defining Quality Goals or Standards for Drinking Water A holistic approach to define hygienically sustainable quality goals or standards for drinking water not only should include considerations dealing with adverse or annoying effects of chemicals on humans and their adverse effects on technical systems, but also should take into account optimizing functionality of intended exposure on site and technical or procedural avoidance of what can be called useless, albeit not necessarily risky, exposure off site. Such precautionary approach (Section 3.14.2.4.1) starts best with considering the origin of a chemical to be regulated. It turns out that the origin or (non)functional value of a chemical in a raw water/ drinking-water system is the factor that helps decide whether it should be regulated either by a health-based or a lower technical maximal or minimal concentration or rather by remote (on site) emission control long before health-based values can be exhausted off site in the drinking water. To make this clearer, Table 2 classifies a number of common (A) constituents of raw water, (B) residuals from functional use in the drinking-water system, or (C) contaminants of raw and possibly drinking water from remote emission, which are candidates for being regulated in drinking water according to their origin (A, B, or C) and (non-)functionality (B or C) in a raw water/drinking-water system. The criteria on why this was done in Table 2 are outlined in Sections 3.14.1.2 and 3.14.2.3.2.

3.14.5.1 Chemicals Whose Regulation Will Primarily Concentrate on Avoidance of Adverse Effects The natural constituents of group A are – almost by definition – difficult to avoid. Sometimes, if occurring at higher than not-adverse levels yet, it may be possible to treat raw water, otherwise it would be necessary to change the water source.

391

The steps or elements to decide on an optimal limit value for group A chemical parameters are as follows: 1. define at first a tolerable or acceptable (yet) nonadverse effect concentration (GV) not to exceed as based on human toxicology or annoyance (see Section 3.14.4) or a more sensitive technical endpoint of adversity (see Section 3.14.2.3.2.); 2. define a natural reference concentration (natural value, NV), for example, in the form of an upper percentile of current geogenic background values of the same compound in the final raw water to be processed as it is; 3. compare GV with NV; if GV rNV; 4. improve the water by treating it according to ALARA to lower the adverse constituent’s level to a tolerable level TV rGV. TV is a precautionary maximal value, it should be fixed as limit value to control technical treatment for failure; and 5. if TV rGV is not reasonably achievable, change the raw water or (some) of its source(s) or wells; if this is not possible keep the TV as a provisional TVp and continue to improve treatment or accept any residual risk as counterbalanced by continued use of the source. Advice for finding GVs as based on adversity for humans is internationally available at WHO (2008), and also at national and supranational bodies as mentioned in Section 3.14.4. Some additional or new information is given in Section 3.14.6.3. Representative information on background levels to define NV is often difficult to find. Existing levels of TV as based on technical adversity or efficiency to eliminate potentially adverse constituents from drinking water are described in appropriate technical norms for construction materials in contact with drinking water or for treatment of drinking water.

3.14.5.2 Chemicals Whose Regulation Should Concentrate on an Optimal Ratio of Functional Exposure to Functional Intention Residuals and by-products on site from functional additives (group B) are evitable down to minimal yet functional levels in the measure as their use gains in efficiency by or despite minimizing on-site exposure. The steps or elements to decide on an optimal limit value for group B chemical parameters are as follows: 1. Define at first a tolerable or acceptable (yet) nonadverse effect concentration (GV) not to exceed on site as based on human toxicology (see Section 3.14.4) or annoyance or a more sensitive technical endpoint of adversity (see Section 3.14.2.3.2). 2. Define the minimal yet reliably functional or residual concentration (FV1) on site of the same compound for the desired process or function. 3. Compare GV with FV1; if FV1 Z GV. 4. Improve the process or functional efficiency according to ALARA down to a lower yet functional on site level of FV2oGV. FV2 should later on be described in a technical norm and would be a precautionary maximal value. Any

392 Table 2

Drinking Water Toxicology in Its Regulatory Framework Assessment and management of chemicals in drinking water by origin A–C of inputa (see text)

A: Natural constituents of raw water

 Presence unintentional  May be functional

B: Anthropogenic residues and side/corrosion products in drinking water  Presence intentional  From functional use (treatment þ distribution)

 Could be adverse

 Could be adverse

Optimal regulation either at beginning or end of pipe in the water-works utility – Aluminumb Ammonium Antimony Arsenicd – – – Borate – Cadmium Chloride

Optimal regulation somewhere in the pipe between source and consumer Acrylamidec Aluminumd – Antimonyf Arsenicf Alkylated benzenesg Benzo(a)pyreneg – – Bromated Cadmiumf –

Chromium Conductivity – Cyanide Cyanotoxinsi – Fluoride Irond Lead Manganesed Mercury Nickel

– – Copperf –

–

–

– –

Nitritek PAHg

– Protonsd Selenium Sodiumd Sulfate – – Uranium

– – – – – Disinfectants and DBPsd Vinylchloridec –

a

Epichlorohydrinec – Ironf Leadf – – Nickelf

Bold: routine surveillance; others: surveillance on verified suspicion (author’s suggestion). In raw water from acidified soils. c Control preferably by specifying materials in contact with drinking water. d After treatment. e Plant protection products from agriculture. f From armatures and/or pipes. g From coatings. h Origin industrial or from abandoned waste sites. i Control preferably cyanobacteria by limiting P load of resource. j Industrial chemicals, nonrelevant metabolites from PPPe, pharmaceuticals. k From nitrate in anoxic water stagnating in galvanized steel pipes. b

C: Anthropogenic contaminants of drinking water  Presence unintentional  Nonfunctional, coming from diffuse or point sources/spills  Could be adverse Optimal regulation at beginning of pipe (remote emission control) – – Ammoniume – – – Benzeneh Borateh – – – Chlorinated solventsh – – – – – – – – – – – New analytesj Nitratee Nitroaromatic compoundsj Nitritee – Perfluorinated compounds (PFCs) Pesticides þ relevant metabolitese – – – Sulfateh – Vinylchlorideh –

Drinking Water Toxicology in Its Regulatory Framework

FV needs to be fixed as a limit value as under improper technical conditions it could exceed the respective GV. 5. If an FV25FV1ZGV is not reasonably achievable, rebalance the functional or hygienic benefit of the process against giving it up. Advice for finding GVs as based on adversity for humans are internationally available at WHO (2008), and also at national and supranational bodies as mentioned in Section 3.14.4. Some additional or new information is given in Section 3.14.6.3. Existing levels of FV2 and FV1 as based on technical functionality of the respective compounds in the drinking water system are described in appropriate technical norms for construction materials in contact with drinking water or its treatment.

3.14.5.3 Chemicals Whose Regulation Should Concentrate on Remote Emission Control Anthropogenic environmental contaminants (group C) by definition are chemicals that occur at places off site from their functional use. They are evitable there down to technical or analytical zero levels only in the measure as their functional use on-site gains in environmental neutrality by minimizing chemical emission into off-site compartments. The steps or elements to inform on an optimal limit value for a group C chemical parameter are as follows: 1. Define at first a tolerable or acceptable (yet) nonadverse effect concentration not to exceed a certain value (GV or a precautionary default surrogate, HRIV, see Section 3.14.6.3.2) as based on human toxicology (see Section 3.14.4) or annoyance or a more sensitive technical endpoint of adversity (see Section 3.14.2.3.2). 2. Define from the point of drinking-water hygiene a tolerable level TV1 for the contaminant in drinking water as far as reasonably justifiable below its GV (or precautionary default surrogate, HRIV), albeit not higher than actually feasible or observed levels. 3. Compare TV1 by means of prognostic models with emission on site of the contaminant into the environment and with observations on its occurrence off site from or by its use on site. 4. Improve the environmental neutrality of the chemical’s onsite functional use, if TV1 is significantly lower than prognosticated or observed emission levels, by reducing its emission according to ALARA down to loads or levels to comply off site with TV1. Such minimized loads or levels should be fixed in technical norms and by legal admissions. TV1 would be a precautionary minimal value. It may be fixed as limit value only if there is a chance to establish clear-cut causal relations between loads or levels admitted on site and possible exceedance of TV1 off site. 5. If TF1 is not reasonably achievable by limiting emission on site, there is, as WHO (2008) proposes, a choice of options for informed decisions, depending on the cultural, technical, hydrogeological, or economic context: a. treat the water to comply with TV1 which should then be set as a limit value to control performance of treatment, or

393

b. change to a less contaminated raw water source with a TVoTV1, or c. define a TV24TV1 as far as reasonably justifiable and achievable below the chemical’s GV (or HRIV) but not lower than prognosticated or observed levels; TV2 would only be a provisional precautionary value. Advice for finding GVs as based on adversity for humans is available at WHO (2008), and also at national and supranational bodies as mentioned in Section 3.14.4. Some additional or new information on a number of parameters is given in Section 3.14.6.3.

3.14.6 Practical Regulation of Drinking-Water Quality 3.14.6.1 Quality Assurance and Surveillance of Raw Water, Finished Water, Tap Water The quality of water bodies from which raw water is abstracted to produce drinking water should meet, as a minimum requirement, certain essential health and hygienic standards. Preferably, and wherever possible, its production should rely on raw water requiring only minimal or even no treatment – neither for health nor for technical or esthetic reasons (Section 3.14.1). If, traditionally, the conditions of abstraction and treatment processes are nature oriented, the raw water usually will meet these requirements and there will be no problem to ensure a drinking-water supply that is not only palatable and sound in health terms, but also acceptable in terms of environmental hygiene. In such drinking water, the actual concentrations of legal chemical parameters, apart from occasional geogenic constituents, are usually far below mandatory limit values. Regarding chemical parameters to which no formal limit values are ascribed, often an unspoken consensus between the precautionary healthcare sector, water supply companies, and the public is effective stating that unregulated environmental contaminants are to be kept off from drinking water, preferably by on-site emission control, and thus lend support at the end of (the drinking water) pipe to a highly defect-resistant error-friendly water abstraction and treatment technology. (An error-friendly system is designed to be highly resistant to technical failure and human error.) As an example, this basic consensus is reflected in the EU Drinking Water Directive 98/83/EC (1998) and their quantitative parametric values as being ‘‘minimum (quality) requirements,’’ and is also covered in essence by the wording of its Articles 4(1) and 5(3). For this purpose, it is the best to detect, quantify, and evaluate suspected contaminants already in the raw water. It would be capital to counteract the input of xenobiotics into the environment in a medium like drinking water just with human toxicological limit values even though such xenobiotics do not exhibit even a minimal benefit there. The highest priority is on closing the sources of xenobiotic input and on ‘‘effective protection of water resources used as sources of drinking water (y) from pollution from (y) agriculture, industry and other discharges and emissions of hazardous

394

Drinking Water Toxicology in Its Regulatory Framework

substances’’ (UN – Economic Commission for Europe, 1999). A comprehensive introduction into managing the quality of groundwater as a drinking-water source under very different aspects is provided in Schmoll et al. (2006). According to these and other sources of knowledge or management advice, it would seem the wrong way to impose at the end of the pipe the responsibility for failures on toxicologists whose scientific expertise to assess the consequences may, at best, be prepared to retrospectively assess single cases where prospective assessments would have been required. It should be no surprise if some of them occasionally pipe up with proposals on precautionary values which then in consequence are intentionally misunderstood by interested stakeholders as being scientifically sound and indicating imminent danger if trespassed. No toxicologist or water supplier ever pretended of being able to protect water bodies by means of human toxicological maximal values for drinking water. This is so because the cooperation required for sustainable development in this area goes far beyond the formal responsibility of the water utility. Any toxicological limit value at the end of (drinking water) pipe would be of no help to clear the underlying relationship between cause and contamination or effect. To achieve practically sustainable results, the preferable way is to agree upon voluntary cooperation. In this context, there exist two kinds of control cycles, number I being closed and number II being open (Figure 1). Unfortunately, in the open control cycle II, legal limit values do not by far have yet the same regulatory potency as in control cycle I. Much too often, the question arises whether the limit values in cycle I are (yet) safe enough or how better maximal (or minimal) values could be made available, whereas preferably the mechanisms of control cycle II should be strained for triggering causal or functional sanitation (for groups of contaminants: see Section 3.14.4.2).

Regulatory circuit II open

Another, at first glance, better-defined source of pollution is migration of substances from materials in contact with water used for human consumption. The control cycle in which these materials and their contaminants are regulated is more or less open when regarding the seemingly neverending number of materials and products that penetrate this worldwide market. The fact that substances originating from this source and entering the drinking water are not contaminants but residues from formally permitted (?) use makes the situation hardly better. In fact, the most significant fraction of drinking-water contamination originates from badly conceived domestic drinking-water installations and materials whose emissions and improper maintenance very often adulterate drinking water delivered in unobjectionable quality from a central supplier to the handover point. For matters of practical surveillance, it is reasonable to take samples for measuring parameters open to change during transport or stagnation of drinking water in such a way that the analytical result represents the quality at the consumer’s tap and not somewhere on the way to the handover point (see also Section 3.14.2.3.1). Parametric values to be checked therefore at either the outlet of the water-works facility or only at the consumer’s tap are listed in Table 3. A sampling procedure to represent the average weekly ingestion by the consumer of corrosion products from Cu, Pb, and Ni with tap water was developed in Germany (UBA, 2004) and conforms to a demand of the EU drinking-water guideline 98/83/EC. A similar conclusion on developing even more sampling procedures to quantitatively describe mean domestic instead of point exposure could be drawn, for example, when regarding impact of (hot) water stagnation and temperature on DBPs in household water (Dion-Fortier et al., 2009).

Resources of drinking water

???

Compliance Comparison with limit values

Finding

Noncompliance Drinking water

Measurements

Regulatory circuit I closed

Distribution Treatment Figure 1 In-plant and external regulatory circuits when operating a drinking-water supply.

Drinking Water Toxicology in Its Regulatory Framework Table 3 Selection of chemical parameters to be checked ideally only at either the outlet of the waterworks facility, or the consumer’s (domestic) tap Place I or II of surveillance I. Outlet of waterworks Chemical parameters whose concentration usually is unsuspicious to increase in the distribution network

II. Consumer’s tap Chemical parameters whose concentration may increase in distribution network including domestic installation

Acrylamide (AA) Benzene Borate ðBO4 3 Þ Bromate Volatile chlorinated solvents (VCSs) Chromium (Cr) Cyanide (CN–) Fluoride (F–) Mercury (Hg) Nitrate ðNO3  Þ

Antimony (Sb) Arsenic (As) Benzo(a)pyrene (B(a)P) Cadmium (Cd) Copper (Cu)

Plant protection products and their metabolites (ppp) Selenium (Se)

Epichlorohydrin (ECH) Lead (Pb) Nickel (Ni) Nitrite ðNO2  Þ Polycyclic aromatic hydrocarbons (PAHs) Trihalogenmethanes (THMs) and other disinfection by-products (DBPs) Vinylchloride (VC)

3.14.6.2 Paths and Significance of Exposure to Chemicals in Drinking Water 3.14.6.2.1 Acidification of raw water Soils under long-term exposure to protons from air may glide into the aluminum- or iron–/aluminum range of buffering situated at pH 4.2–3.8 and 3.8–3.0, respectively. The buffering takes place by dissolution of corresponding metal compounds and the destruction of three-layered clay minerals, which are important filters against the input of organic xenobiotics. The place of HCO 3 is taken over by the anions of strong acids (sulfate and nitrate), whereas on the cationic side strongly acidic ions such as Al3þ, Fe2þ, and Mn2þ are increasing, possibly together with increased mobilization of toxicologically critical heavy metals ions (Svensson et al., 1987). Aluminum in concentrations above 2 mg l1 may heavily compromise the functionality of pumping systems for drinking water (Lu¨kewille and Heuwinkel, 1990). The present regulations to stop or confine soil acidification in many parts of the world are insufficient to stop this risky process of threefold concern for adverse technological, ecological, and human health endpoints. Increasing proton concentrations mobilize toxic or undesired metals not only in the subsoil, but also in metallic distribution systems, if the distributed water is not neutralized according to existing technical norms.

3.14.6.2.2 Agrochemicals Agricultural activity affects water bodies mainly by its input of nutrients such as nitrogen and phosphate salts and pesticides and their metabolites. Nitrate/nitrite and their potential to

395

cause methemoglobinemia have attracted the interest of toxicologists for a long time (see Section 3.14.6.3.1). Overfertilization of lakes and reservoirs may also lead to massive growth of cyanobacteria and their corresponding contamination by highly hepato- and neurotoxic cyanotoxins up to hazardous levels in extracted drinking water, if not treated accordingly. Countries that do not regulate pesticides in drinking water by a precautionary limit value also deserve to be looked into by drinking-water toxicologists (see Section 3.14.6.3.2). Another toxicological observation is the microbial oxidation and mobilization of toxic heavy metals in the subsoil in connection with denitrification of nitrate from agriculture in reducing aquifers (Ko¨lle et al., 1983, 1987).

3.14.6.2.3 Emissions from abandoned waste sites Reams of chemicals are used in industrial processes or are present in products and wastes thereof, along with many more such degradation products. Their health hazard may vary considerably and it is necessary therefore to accordingly prioritize the hazard potential of those that are situated in water-catchment areas. Pilot surveys performed to detect highly mobile inorganic and organic parameters such as borate, sulfate, AOX, and GC-screening of abandoned waste sites or sources in drinkingwater catchment areas are able to discriminate potentially hazardous sources of underground emissions from the more harmless ones. Positive findings indicate the need for further explorations on the basis of an invariable list of contaminants with high toxicological priority (toxic potential) and, at the same time, unfavorable environmental behavior (mobility, persistence). Of the nine inorganic compounds from such a list, the first four were arsenic, nickel, chromium, and nitrite. The first four of its 15 organic compounds were benzene, vinyl chloride, trichloroethene, and tetrachloroethene. All eight compounds were also reliable indicators for the possibility that raw water for drinking-water production may be fed downstream by a contaminated aquifer (Schmoll et al., 2006)

3.14.6.2.4 Disinfection by-products Disinfection of drinking water renders the water not only in a desired microbiological state but, especially by chlorination, also creates chemical-toxicological risks. They result from the reaction of chlorine with chemical precursors naturally present in the water to be treated to hundreds of chlorinated compounds. Chloroform and, if the raw water contained bromide, (chloro-)bromoforms are the most prominent single compounds. The debate on whether chlorinated drinking water under practical conditions of exposure might or might not have a carcinogenic or other adverse health outcomes in humans has been ongoing for three decades. It discusses qualification and quantification of continued exposure mainly (see also Section 3.14.6.1, last paragraph). Neglect in considering spatial and temporal variations of DBP presence on their way to the consumer’s tap makes it very difficult to allocate adverse outcome and exposure and still more to analyze whether positive or negative results from different studies are consistent with

396

Drinking Water Toxicology in Its Regulatory Framework

others (Legay, et al. 2010). The author of a recent comprehensive review (Hrudey, 2009) states ‘‘the epidemiologic evidence has generally been found to be insufficient to declare chlorination DBPs to be carcinogenic in humans.’’ The evidence in favor of DBPs as urinary bladder carcinogens has been the least inconsistent, but ‘‘there are no data to indicate that any of these compounds can contribute to bladder cancer by any mechanism’’ (Bull et al., 2001, as cited in Hrudey, 2009). Only studies with positively proven control for such complex confounding factors as lifestyle and exposure, allowing even for the hypothetical presence of an extremely potent carcinogen in significant concentrations among all DBPs, could bring this discussion to an end. As an example, such studies would definitively have to include the genetic disposition of exposed persons for expressing a toxicologically critical isoform of the detoxifying enzyme glutathion transferase and their exposure to DBPs from swimming pool water (Zwiener et al., 2007). The regulatory-toxicological situation here is somewhat similar to the one with nitrite/nitrate and their presumed potential to initiate bladder cancer as precursors of carcinogenic nitrosamines. As to the another suspected endpoint for human toxicity of DBPs, a number of reproductive outcomes suspected to be adverse under the influence of DBPs, the total of available epidemiologic data were exhaustively considered by Hrudey (2009). The present author’s own conclusion and that of Hrudey’s are that these data are still less supportive in favor of an increased incidence under exposure to DBPs than for bladder cancer. A technically feasible way to cope with this regulatorytoxicological challenge is to minimize the necessary concentration of disinfectants. This would not only minimize formation of DBPs from precursors but also improve microbiological quality and stability of drinking water. This is possible by applying routine raw water-directed procedures of drinking water treatment (Zwiener, 2002; Haberer, 1994). Starting from raw water which is not a protected groundwater and whose microbiological quality may therefore not be unobjectionable, there are several ways to achieve both goals, namely

• • • • •

prefer raw water with low DOC, minimize DOC by flocculation and oxidation before chemical disinfection, use disinfective treatment in the water-works facility by alternative agents with lower DBP-forming potential as there are chlorine dioxide, ozone, UV-irradiation, carefully maintain a microbiologically clean distribution network for being able to minimize dose of disinfectant for transport even up to its complete abandonment, and apply, if feasible (despite high energy expenditure), several modes of micro- or nanofiltration.

In case of natural or quasi-natural resources exhibiting, by definition, doubtful microbiological quality (rivers, reservoirs, and natural lakes), a priori effective treatment (before disinfection) is the best way to produce a microbiologically safe drinking water with minimal levels of DBPs.

An established procedure to produce microbiological safety, while minimizing the DBP formation potential of a raw water simultaneously, is its ozonation after different grades or steps of treatment (von Gunten, 2003; Bonacquisti, 2006). Pre-ozonation oxidizes organic carbon to epoxides, aldehydes, ketones, or acids which then may be eliminated by flocculation. However, if formed after intermediate ozonation, such compounds are excellent substrates for microbiological growth and concurrent water contamination. Such secondary ozonation must therefore be followed by percolation on activated carbon. This step is effective not only by adsorption, as the filter bed will soon support stabile bacterial mats which easily degrade these partially oxidized carbon compounds. After percolation follows the final ozonation to disinfect the water before distribution. Before this finished water is fed into its distribution net (see Section 3.14.2.3.2) it needs to be microbiologically stabilized for its way to the consumer’s tap by dosing minimal but demonstrably effective amounts of chlorine or chlorine dioxide not only directly after treatment but also at selected dosing points on the way to the consumer. With a distribution system in optimal technical state, even transport disinfection may routinely be abandoned as demonstrated in the city of Berlin for several decades (Jekel and Gruenheid, 2008) The desired consequences of the method outlined here to separate (process) oxidation from disinfection (for transport) are a minimal requirement of disinfectants and equally minimal DBP formation in the range of maximally 10–20 mg l1 to be detected at the consumer’s tap. Under such conditions and keeping the conclusions by Hrudey (2009) in mind, any discussion on increased incidence of cancerous or other toxic endpoints in populations that are long-term consumers of chlorinated drinking water would not only appear obsolete but also cast superfluous doubt on the health benefit of drinking-water disinfection by chlorination. The only DBP formed in significant amounts (from bromide) as a consequence of ozonation is bromate (see Section 3.14.6.3.1).

3.14.6.2.5 Small supplies and health risks Many countries or their supranational organizations prescribe, dispose of, or provide technical norms to support the legal quality of drinking water from small supplies and private wells if constructed in consistence with such norms. Water from wells surrounded by catchment areas lacking such consistence and hence adequate protection may exhibit high nitrate levels and microbial growth, giving cause for health concerns if ingested. The water of such supplies may also be acidified (see Section 3.14.6.2.1) and, accordingly, support corrosion of domestic installations. Risky levels of copper, lead, nickel, and other metals may be the consequence. Additionally, the presence of geogenic constituents with a high toxic potential (arsenic, fluoride, heavy metals, etc.) should be excluded before approving such water as safe for consumption. As an example, uranium in well water was the cause of nephrotoxicity in a family (Magdo et al., 2007). Similar cases are documented for intoxication by geogenic fluoride from small supplies (Hardisson et al., 2001) and in the form of a mass intoxication

Drinking Water Toxicology in Its Regulatory Framework

by geogenic arsenic ingested with drinking water from wells in Bangladesh and West Bengal in India (Meharg, 2005). Biofilms in containers used for storage of well water may be a reservoir of toxic microcystins from cyanobacteria if the origin of water and the conditions of storage support their growth (Kanke-Fosso et al., 2008). Common geogenic parameters such as iron or manganese may be tolerated at higher levels on the basis of human health considerations than for technical reasons in larger systems (see Section 3.14.6.3.1). Surveillance and control of small community supplies (applicable also to private wells) have been extensively treated by WHO (1997) and in a report ‘Protecting Public Health in Small Water Systems’ at an International Colloquium held at Montana Water Center in May 2004 ( Ford et al., 2005; Committee on Environmental Health, 2009).

3.14.6.2.6 Organics leaching from drinking water reservoir coatings and armatures Construction of facilities for drinking-water supply and choice of materials should be organized in such a way that additional surface coatings to prevent corrosion are not necessary, although such wishful thinking is not always practicable. Exposure to coatings and paints may be of significance as well in the course of pipe sanitation, especially of those with old tar-based coatings which are an important source of polycyclic aromatic hydrocarbons. Pipes being sanitized on site by coating with epoxide resin may turn into a source of migrating chemicals if after coating, necessary waiting times or maximal temperatures to operate the sanitized system are not respected. Bitumen-based coatings may contain alkylated benzenes that are of interest more for esthetics than for toxicity. Besides this, they may offer organic carbon for microbial growth and impair disinfection by chlorine by forming DBPs. They may be perceived by odor when they are already a few mg l1, whereas human toxicological guide values may reach levels of 100 mg l1 and more (WHO, 2008). Contaminations of this source are common and occur mostly in connection with unprofessional (incomplete) drying of freshly coated construction elements or because contaminated exhaust-air had contact with drinking water. Defective mixer taps may be a source of leaching control liquids such as toluene, alcohols, and waxes. For adequate surveillance see last paragraph of Section 3.14.6.1.

3.14.6.2.7 Hygienic aspects of corrosion products from domestic pipes and metallic materials The intensity of mutual interaction between drinking water and metallic materials/pipes depends on several factors:

• • •

chemical properties and metallurgical state of the material, inorganic constituents and electrochemical properties of the drinking water in contact with that material, and operating conditions (temperature, stagnation, domestic treatment, etc.) of drinking-water supply and domestic installation.

The most common metals to reach drinking–water by this way are copper, zinc, iron, cadmium, and lead. Nitrate in

397

anaerobic drinking water may be chemically catalytically reduced to nitrite during stagnation in galvanized steel pipes. In order to prevent health and technical problems, it is necessary to avoid chemical corrosion of drinking-water pipes as far as possible. Technical norms to cope with this task are permanently created and updated in many countries. For reasons of practicability they cannot deal with the inorganic constituents and electrochemical properties of individual drinking waters, but only with commercial materials, products, and installations. The implementation of technical norms lies within the professional responsibility of the construction engineer. The alternative would be a normalized, chemically defined and all-round technically compatible hence uniform drinking water. Such a vision is neither desirable nor would it be feasible. The best predictions on the corrosive potential of drinking water within its technical environment can be given on the basis of its pH value. This is, at the same time, the most practical parameter to gain such information. For each metal there exists a specific, slightly alkaline pH interval in which its tendency to react with water crosses or finds a minimum. On the other hand, it is advisable not to exceed in real water its specific pHc of calcium carbonate saturation, since otherwise it would precipitate and gradually clog the pipes. Exceeding the pHc would therefore make it impossible to adjust the interval of minimal corrosion; the material under question would then be unemployable. Therefore, from the view of undesired chemical corrosion, a soft water with its own and relatively high pHc is always preferable to hard water and its lower pHc. Higher corrosive potentials are only to be found in unbuffered soft and acid waters with measured pH values of 6.5 and less. Depending on hardness and pHc of drinking water, the metallurgic properties of the used materials and the operating conditions, it is possible that some metals may exceed health-based maximal values if corrosion is not minimized by implementing the technical norms created for this purpose. A current example is copper. Its pH interval of technically and hygienically tolerable corrosion – depending on dissolved organic carbon and some other parameters – begins at the earliest at pH 7.4 (usually at pH 7.6). Since this pH is usually higher than the pHc of very hard water, the usability of uncovered (untinned) copper pipes to transport such water should be limited on pH values higher than 7.6. If the pH of the finished water exceeds a value of 7.8, control of copper content is required only in special situations (amendment of the German drinking-water ordinance as of 1 March 2010). The correct hygienic and exposure assessment of residues from metallic corrosion depends heavily on a correct representative sampling (see Section 3.14.6.1). As a rule, the mean concentration of a corrosion product at the consumer’s tap reaches its concentration maximally as measured in a 4-h stagnation sample.

3.14.6.2.8 Hygienic aspects of domestic posttreatment of drinking water Operating of small domestic devices for posttreating of drinking water (mainly by softening, reverse osmosis, or ultrafiltration) is, rarely, a reasonable alternative to (mis)trust a central supplier’s engagement to provide the consumer an

398

Drinking Water Toxicology in Its Regulatory Framework

unobjectionable product. Even for considerations unrelated to health, domestic devices, installations, and procedures to treat drinking water just before it reaches the tap are useful only in special situations. Unfortunately, the potential domestic operator or owner can only occasionally take the right functional and hygiene-related decision on his own but depends completely on the information provided to him/her by a plumber or the vendor/manufacturer of a treatment device. For different reasons, these parties are likely to overemphasize the advantages of their investment or devices. For the operator of a domestic installation, the only question of interest to be answered is whether the investment for a posttreatment device would save him from losses by later more expensive repairs that might occur from damage by increased corrosion. There is, however, no doubt that the professional design of a domestic installation and its correct completion later is the best warranty for its enduring functionality. Enduring functionality depends mainly on correct selection of construction materials as required by the characteristics of the water delivered by the local water supplier. Within the area of supply, the operator is the best addressee for questions regarding corrosiveness and other technical experience, whereas any question about its health compliance is answered best by the local health authority. An overall expert judgment of specific complaints, technical information, and possible desire for domestic comfort in an area of supply could reach the conclusion that a partial central softening of very hard water would be of advantage to all parties for economic, ecologic, and technical reasons; the only possible constraint for the supplier could be of keeping a minimal and nutritionally favorable concentration of Ca2þ instead of further increasing Naþ (Cotruvo and Bartram, 2009) (see also Section 3.14.1.2). Even when considering technical progress, there are always good hygienic reasons in favor of not interrupting the distribution pipe between the supplier and the consumer just for technical or comfort reasons. Each additional construction device, especially those that are the consumer’s private responsibility, creates additional risks because of human failure, faulty operation, and careless maintenance. Therefore, one single central (and partial) treatment for softening would always be the preferable alternative to many peripheral softening devices in single homes. Other technologies adapted for domestic use such as reverse osmosis, ultrafiltration, or ultraviolet (UV) irradiation are advisable for small supplies devoid of any reasonable possibility to benefit from central treatment and, in such cases, would be misclassified as posttreating methods.

3.14.6.2.9 Hygienic aspects of domestic water saving From the very inception of the existence of a central drinkingwater supply system, the question was raised whether it would be worthwhile to partially replace unobjectionable drinking water at home by water of minor quality (greywater) in the form of rain water collected from one’s private roof. The economic, ecologic, and technical benefit to expect from such replacement very often is heavily overestimated, and for hygienic risks the contrary is true. If not maintained properly and regularly or correctly stored and disinfected, the technical

equipment as well as the collected water itself is very likely to preserve a microbiologically objectionable condition (Schets et al., 2010). As far as chemical aspects go, such water is usually very soft and acidic since it is only occasionally adjusted to its pHc. As such, it will corrode the domestic installation (see Section 3.14.6.2.7) much more than if it is adjusted to its pHc. An important point from the perspective of personal hygiene and public health is that in moderately to densely populated areas none of the possible private investors would ever be able to renounce a central or piped supply completely, except they would have drinking water from a private well available to them to bridge periods of dryness. All others, in case of emergency, will always be tempted to reconnect themselves with the central supply, entailing all technical failures and adverse hygienic consequences which any hidden or unintentional technical default (e.g., back siphonage) might cause for the surrounding neighborhood or clients of the central supplier. Other hygienically questionable outcomes of decreased drinking-water turnover are increase of its stagnation in public pipes and that of wastewater and its concomitant fouling in sewerages. Any effort from the side of the supplier to clean the pipes and sewerages by flushing will undo such savings at the expense of those who anyhow were unable to afford it (see also Section 3.14.1.1). In fact, there are not many arguments in favor of an action to save, specifically, drinking water, besides some useful technical possibilities to limit superfluid domestic water use through regulation of flow or pressure by means of specific armatures. There should never be any argument at all against storage and use of rain water for agricultural or domestic irrigation. Matchless higher potentials to save (at inevitably much higher absolute needs of) water are hidden or slumbering in present modes and habits to produce certain foods and their consumption (UNEP, 2003b).

3.14.6.3 Significance and Hygienic Assessment of Exposure at the Tap In its monograph ‘Concern for Europe’s Tomorrow’ from 1995, WHO (1995) indicated the numbers of people and localities in Eastern Europe exposed to toxic chemicals via drinking water. The most important numbers relate to heavy metals (553 000 people from 123 localities), cyanides (10 000 people from 10 localities), phenolics (230 000 people from 57 localities), oil substances (1 million people from 1659 localities), and enhanced radioactivity (300 000 people from 100 localities). The following very short monographs come in alphabetical order. They concentrate on those parameters not dealt with in the rolling revision of the WHO (2008) drinkingwater guidelines or on those the author felt additional or more recent information than contained in the respective WHO background documents (the earliest dates from 2003) would be of interest for the reader.

3.14.6.3.1 Inorganics Arsenic (group A in Table 2). Inorganic arsenic (As) is a potent human carcinogen. WHO (2008) stated in its 2003 assessment

Drinking Water Toxicology in Its Regulatory Framework

that ‘‘although there is a substantial data base on the association between both internal and skin cancers and the consumption of arsenic in drinking-water, there remains considerable uncertainty over the actual risks at low concentrations’’ and therefore retained its earlier GVWHO of 10 mg l1 As for reasons of technical and analytical achievability. Applying the terminology of Section 3.14.5.1, this GV would be a TVp. It is frequently exceeded in sources for drinking water; in terms of exposed population (35–77 million people) the scale is greatest in Bangladesh (Smith et al., 2000; Meharg, 2005). On the other hand, a large ecologic study with 44 872 respondents revealed no association between As in 1 900 springs and groundwater wells containing 0.1–950 mg l1 As (of which15%410 mg l1 As) and incidence of As-related internal cancers, whereas other independent variables than As (race, smoking, sex, and overweight) showed some positive association. Most important, the study found no indication on an increased risk to contract cancer from exposure to inorganic As below 10 mg l1 As in drinking water (Han et al., 2009). A similar conclusion was also drawn earlier from large epidemiological cancer studies (Dieter, 1991) and is sustained by exhaustion of As detoxification only at regular daily ingestion of at least 200 mg As (Mazumder et al., 1988; Marcus et al., 1988; Rudel et al., 1996; Calderon et al., 1999). On the other hand, there is convincing evidence to assume monomethylarsonic acid (MMA) as a genotoxic intermediary product of As metabolism in mammals on the way to its excretion as dimethylarsonic (DMA) and trimethylarsonic acid (TMA) (Schuhmacher-Wolz et al., 2009). For the time being, it remains to be determined whether and, if ever, at which level the carcinogenicity of As might be mediated by one or more thresholded effects, the more as nutritional status, namely intake of methionine, protein, and cysteine may significantly affect the level of possible threshold(s) as recently shown by Heck et al. (2009) and earlier publications cited therein on other settings of As exposure. Nutritional, mechanistic, epidemiological, and other information was recently reviewed and used to evaluate the reliability of existing cancer risk assessments and to better quantify current assessments of noncancer health effects of As such as peripheral vascular disease, blackfoot disease, skin hyperpigmentation, and hyperkeratosis (Schuhmacher-Wolz et al., 2009). According to Schuhmacher-Volz et al. (2009), As at or below 100 mg l1 shows a broad spectrum of noncancer adverse effects on reproductivity, central nervous system (CNS; intellectual function), cardiovascular system, and skin. They provided consistent dose–response data from several new studies for skin hyperpigmentation and hyperkeratosis (thickened soles of feet). Based on the data for prevalence of such skin lesions from arsenic exposure in a population in Bangladesh and data on concurrent background arsenic exposure from food, the authors performed benchmark (BM) dose–response modeling and calculated a lower confidence limit of 109.2 mg d1 at a population effect level of 5% and set this LOAEL as PoD to derive a Bd. Their reference study (Ahsan et al., 2006) comprised a large rural population of more than 10 000 persons exhibiting a mean body mass of 50 kg for men as the most susceptible

399

group for the selected, most critical toxic endpoint. They deemed an extrapolation factor to cope for interindividual differences in susceptibility not necessary, whereas the 5% response level at the PoD was assessed as unacceptable. Therefore, they used an extrapolation factor of 5 to derive a tolerable daily intake for As from all sources of TDI ¼ 0.45 mg kg1 bm, corresponding to a Bd of 31.5 mg per 70 kg bm and 22.5 mg per 50 kg bm, respectively. In Western countries, where daily intake of As is estimated to amount to 14 mg (Hughes, 2006) this TDI would be exceeded if 2 l of daily drinking water contain more than 9 mg l1. Consequently, for these countries the GV of WHO would be supported by this evaluation of noncarcinogenic health effects, whereas for countries with higher personal drinking water intakes, higher intake of As from food and lower mean body mass it could easily be too high. Bromate (group B in Table 2). Bromate was last assessed for drinking water in 2005 by WHO (2008) in the form of a GV for reasonably feasible minimization of exposure to bromate from bromide as oxidized during ozonation of drinking water. The assessment was based on risk calculation with IZ values (see Section 3.14.4.3.2) of 106, 105, and 104 mainly for kidney cancer at 0.2, 2, and 20 mg bromate per liter. This GVWHO was eventually set for reasons of analytical and technical (bromate levels after drinking water ozonation) feasibility at 10 mg l1. Delker et al. (2006) published extensive biochemical data and arguments in favor of a nonlinear ( ¼ thresholded) dose– response relationship for the carcinogenic potential of bromate in 2006. Bromate triggers this potential by oxidative stress, mainly 1O2, which was assessed by the authors as being about 3 times higher in the rat kidney than its thyroid (in which anyhow no clear relationship between dose and response was observed). The intracellular capacity of the enzymatic and other defenses to cope internally with oxidative stress from bromate exposure seems to be exhausted at doses of more than 5 mg kg1 bm and their capacity to protect DNA from oxidative damage from doses of at least 20 mg kg1 bm on or higher (Dieter, 2003). The failure to protect DNA from damage by oxidation manifests itself intracellularly above background levels of excised 8-hydroxy-desoxyguanosin, the main reaction product of DNA with oxidative stress from bromate. The transcription of responsible enzymatic excision repair enzyme (8-glykosylase) correlates closely with increased oxidative stress long before mutations or even cancer can be observed (Dieter, 2003). The reason is that ROS-mediated mitogen-activated protein kinase (MAPK) activation involved in the molecular mechanisms of BrO 3 -induced cell cycle arrest occurs independently of 8-OH-dG production (Zhanga et al., 2010). An assessment different from the one proposed by WHO (2008) and starting from a thresholded PoD (see Section 3.14.4.3.1) seems therefore supported by strong biochemical arguments and may lead to distinctly higher lifelong tolerable GV than could be concluded from WHO (2008). This view is further supported at least for relatively low exposure by data demonstrating effective detoxification of oxidative stress from bromate by reducing agents, especially glutathione, in the extracellular space (Chipman et al., 2006) and the effective chemical reduction of bromate by typical

400

Drinking Water Toxicology in Its Regulatory Framework

sulfide-containing gastric juices to Z99% within 20–30 min (Keith et al., 2006). Residual levels of bromate distinctly below 10 mg l1 after treating drinking water with ozone seem routinely feasible (Bonacquisti, 2006). There is therefore no argument against limiting bromate levels in finished water maximally at 10 mg l1. Within the context of definitions given in Section 3.14.5 and the earlier arguments in favor of an effect threshold of bromate for precancerous cytotoxicity, it would appear that FV1 o GV is an acceptable precautionary level to avoid nonfunctional exposure. The nonthreshold point of view would come to the contrary conclusion, namely FV1 4 GV, and hence would define this FV1 not as a precautionary but an ALARA-based level of minimization of exposure without impeding the underlying function. From this view, FV1 is acceptable on the basis of indispensable benefit of ozonation for drinking water hygiene and its outweighing of more significant chemical risks from other oxidation/disinfection processes, not on the basis of precautionary minimization. The bromate example is a convincing example on how it is possible to regulate a chemical in drinking water, despite strongly differing views on its toxic potential, in a consensual manner on the basis of reasonable achievability and the corresponding three-dimensional rule of environmental hygiene as outlined in Section 3.14.1.2. Copper (group B in Table 2). Copper (in the form of its corrosion products in drinking water from copper pipes) was last assessed in 2004 by WHO (2008) in the form of a GVWHO of 2 mg l1 total Cu as based on local acute adverse health effects of Cu compounds from corrosion of copper pipes on the digestive tract and on their astringent taste (see Section 3.14.4.3.3). The WHO background document reports also on possible risk groups for liver toxicity by copper. A high exposure risk group for this so-called ‘idiopathic copper toxicosis’ (ICT) seems to be infants growing up in households providing themselves with acid water (Section 3.14.6.2.1) from a private well (Section 3.14.6.2.5) and copper plumbing (Section 3.14.6.2.7) (Dieter et al., 1999). Infants growing up under regulated conditions of drinking-water supply do not seem to be at risk (Zietz et al., 2003). Data collected directly in the customers, homes by Zietz et al. (2003) may allow some kind of rough risk calculation for a large city on the basis of the following estimations:

• • • •

0.6 ¼ frequency of copper installations, 0.1 ¼ frequency of (strongly buffered) drinking water with a pH r7.4, 0.03 ¼ frequency of weaning infants right from birth, and 0.01 ¼ frequency of average exposure to 4 2 mg l1 Cu in drinking water, a hypothetical threshold for early signs of hepatic copper toxicity (Dieter et al., 1999).

The total population risk or possible incidence IICT for an infant in a family with regulated central supply to contract early signs of ICT should equal the total product of these numbers, hence IICT ¼ 0.6  0.1 0.03  0.01 ¼1.8 105. WHO (2008) reflects on the existence of a yet undiscovered congenital condition of copper hypersensitivity, taking shape as ICT in infants only under the influence of very early and

distinctly higher than average environmental copper ingestion. This condition was discovered and extensively characterized since a long time in parts of India under the name of congenitally determined Indian childhood cirrhosis (ICC) (Tanner, 1998) and discovered later on as being very similar or identical to the ecogenetic copper storage disease endemic Tyrolean infantile cirrhosis (ETIC) as discovered by Mu¨ller et al. (1996). All three diseases, ICT, ICC, and ETIC, are clinically identical but could not be characterized by a gene defect common with Wilson’s disease, the most prominent and longknown genetic disorder of copper metabolism (Wijmenga et al., 1998; Mu¨ller et al., 2003). Despite a successful effort to establish an animal model to study the genetic basis and etiologic outcome of ICT or even of all three (then identical?) diseases themselves (Haywood et al., 2004), it was not possible yet to identify their genetic substrate although there are good arguments to support the existence of such substrate in the form of mild mutations or polymorphisms common to ICT, ICC, and ETIC, respectively, in variable extents (de Bie, 2007). This view was also supported by WHO (2008) in its actual background document on copper from 2004. Assuming as a worst case, a fraction of 5 102 for such congenital disposition in a genetically heterogeneous population as being exposed to high environmental copper, the IICT calculated above would reduce to less than 106. This risk would be of the same order as the one accepted for risks from exposure to nonthreshold carcinogens (see Section 3.14.4.3.2). Its management, however, differs distinctly insofar as with copper, each single case of an unfavorable or risky exposure is easily possible to trace back down to the exposed individual, whereas such risk management hardly ever seems possible regarding environmental exposure to nonthreshold carcinogens. Unfortunately, due to the low incidence of ICT and the strong homeostatic regulation of peripheral copper even an enhanced liver load, the possibilities to develop an early and easily accessible biomarker for this unexplained copper storage disease are very limited (Uauy et al., 2008). There was some speculation on a causative association between toxic free serum copper, as ingested in inorganic forms from drinking water or other sources, and incidence of Alzheimer’s disease (AD). The main argument given was oxidative stress by copper and its binding to ‘‘all the molecules involved with AD brain pathology’’ (Brewer, 2009), although it was clearly demonstrated in a strongly controlled clinical trial with mild Alzheimer’s patients that long-term oral intake of 8 mg Cu can be excluded as a risk factor for AD as based on biomarker analysis in cerebrospinal fluid (Kessler et al., 2009) Fluoride (group A in Table 2). Fluoride, like possibly copper or manganese in infants or subgroups thereof, has a rather steep dose–response curve in humans. It increases the metabolic turnover of the bone by directly inhibiting osteoblastic acid phosphatase activity (Krishnamachari, 1986; Lau et al., 1989). Like with geogenic arsenic and manganese or anthropogenic lead, it is very important to examine a new source for its fluoride content, since, as with geogenic arsenic, selenium, and uranium, levels relevant for health can easily be exceeded (Fawell and Nieuwenhuijsen, 2003; see also Section 3.14.6.2.5).

Drinking Water Toxicology in Its Regulatory Framework

A recent example for osteosclerosis due to endemic fluorosis from drinking water has been described from an area in southern Turkey with levels of F in drinking water above 1.2 mg l1. Skeletal fluorosis was evident in patients with mean urine fluoride levels of 1.27 mg l1 (0.22–3.99 mg l1), whereas a mean urine value of 0.6 mg l1 F (0.18–1.35) was not associated with this disease (Tamer et al., 2007). WHO (2008) in its 2004 background document, like in 1993, affirmed again its fluoride GVWHO of 1.5 mg l1 from 1984. According to WHO, exposure to moderately higher levels would result in dental fluorosis and higher levels would eventually lead to skeletal fluorosis. National standards worldwide should, as recommended by WHO, vary according to regional regular drinking water uptake und fluoride ingestion from other sources which in total should not exceed the Bd of 6 mg F per 60 kg person or 0.1 mg kg1 bm F (for children up to 10 years). There is an ongoing discussion all over the world on whether drinking water would be an appropriate communal carrier of artificially added 0.5–1.0 mg l1 fluoride in order to prevent dental caries on a societal level. For several reasons, many countries or communities do not practice drinking-water fluorination, although its health benefit seems to be proven. A socioethical reason says drinking water should never be misused as a carrier of pharmaceuticals. Environmental arguments against drinking-water fluorination are the environmental persistence and the high ecotoxic potential of this element; more so as at least 95% of the added fluoride would miss its target while nevertheless accumulating in regional water cycles. Treatment of drinking water for fluoride removal would be rather costly, as common technologies are not effective for this purpose. This is why raw water containing more than 1.5 mg l1 F usually is not exploited for drinking-water production, except if the source is irreplaceable and may then be treated by reverse osmosis or filtration over hydroxylapatite. In countries allowing fluorination of table salt (usually in the range of distinctly less than 1 mg NaF per 1000 mg NaCl) it is recommended not to drink either bottled water or drinking water containing more than 0.7 mg l1 F regularly. A comprehensive review on benefit and risks of artificial fluoride exposure has been published by the Australian National Health and Medical Research Council, covering English-language publications from 1996 to 2006 (NHMRC, 2007). Incidents from technical failure of F dosage in the water works have been described (WQRA, 2009; Gessner et al., 1994) and form an important argument against drinkingwater fluorination, at least in supplies without appropriate preventive control measures or safety plans. Detection of overdosing may be too late to inform overexposed and acutely intoxicated people in due time. Lead (group B in Table 2). Due to the general decline in atmospheric lead pollution in the course of the last few decades, drinking water from lead pipes became the most important single source of lead exposure. An initial estimate for the EU is that potentially 120 million people, corresponding to 25% of all domestic dwellings, are at risk from enhanced lead exposure via drinking water (Hayes and Skubala, 2009).

401

WHO evaluated lead last in 2003 (WHO, 2008) and set its GVWHO for drinking water at 10 mg l1 Pb. Its derivation related on Joint FAO/WHO Expert Committee on Food Additives (JECFA)’s provisional tolerable daily intake (PTDI) for lead which since 1986 is 3.5 mg kg1 bm for infants and small children. In 1993, JECFA extended it to all age groups. The allocation of PTDI corresponding with this GV is 50% in 0.75 l of drinking water as consumed daily by a formula-fed 5-kg infant. JECFA’S PTDI was considered by WHO (2008) as being sufficiently protective for the CNS, the most sensitive organ for lead toxicity. It impairs a number of CNS functions, namely intelligence, performance of attention and reaction, behavior, and level of hearing threshold. Protection from these effects should protect from systemic carcinogenicity of lead also; this was evaluated last in 2007 as exhibiting just limited evidence, even at much higher than its very low yet neurotoxic doses in humans (Deutsche Forschungsgemeinschaft, 2007). According to a recent review (Wilhelm et al., 2010a), the model of taking 100 mg l1 blood–lead (PbB) as a warning threshold below which adverse neurotoxic effects may not be expected was rejected recently ‘‘on the basis of linear and nonlinear effect extrapolation.’’ The authors state ‘‘it now appears to be certain that negative correlations between PbB and neuropsychological parameters also exist at PbB-levels below 100 mg/l.’’ Moreover, oral lead is suspected to affect onset of male and female puberty as indicated by several biological and anatomical parameters in three different populations already at PbB levels distinctly lower than 100 mg l1. Lead exposure through drinking water can easily be avoided. From the view of drinking-water hygiene, the conclusion is to prevent the risk groups – pregnant women, infants, and small children – from ingesting higher lead intakes than inevitable for the time being. They should consume food which had no contact with drinking water from lead pipes during its preparation. A reference PbB level to indicate, if exceeded, specific sources of lead exposure such as drinking water from lead pipes, has been set at 35 mg l1 Pb for infants and small children and 70 or 90 mg l1 for adult women and men, respectively (Umweltbundesamt, 2009b). The high toxicity of lead for humans was never in doubt in newer human history. Unfortunately, its toxic potential is too high despite exceptional technical performance for water plumbing. This was the reason to prevent its use for this purpose as early as 1790 (Duke of Wu¨rttemberg, Letter from 24 December 1790, on behalf of a new well order in which he insists on using iron or argil instead of lead because the latter could have adverse consequences for health) and to ban it formally in 1878 (Ko¨niglich-Wu¨rttembergisches Ministerium des Innern, 1878) both in the then South German Duke and later Kingdom of Wu¨rttemberg. Manganese (group A in Table 2). The health-based GV of WHO for manganese (Mn) is 0.4 mg l1. Its derivation dates from 2004 and is based on the upper limit of adult manganese intake of 11 mg d1, coming from dietary studies. This NOAEL was used by WHO (2008) as a PoD since ‘‘it is not believed that this amount of manganese in the diet represents an overexposure to the element.’’ Given this belief, 11 mg d1 would represent the upper limit of an AROI of Mn.

402

Drinking Water Toxicology in Its Regulatory Framework

The GV of WHO has the drawback to be based on a toxicologically unqualified AROI of mainly unsoluble Mn(IV) as it prevails in food, albeit qualified animal studies using drinking water to dose soluble Mn2þ are available. Moreover, an alternative approach to define a health-based GV for manganese should examine more closely the neurotoxic potential of Mn as known from workplace epidemiologic studies (Roels et al., 1999) and perhaps from a study with school children (Gongqiao et al., 1995). A 10-year-old boy, after having ingested a well water containing B1.1 mg l1 total Mn for at least 4 months and up to 5 years, exhibited strongly enhanced Mn levels in blood, urine, and hair. Neurological examinations revealed distinctly less than normal verbal and visual performance. The authors concluded on exposure via well water (for years?) to Mn as the main cause, but could not completely rule out exposure to other ‘toxic wastes’ (Woolf et al., 2002). Oral MnCl2, in a 240-day study with adult rats, changed their motor behavior at 10 mg kg1 bm Mn2þ and more (Bonilla, 1984). Young experimental animals are more sensitive to early intoxication by soluble Mn than adult ones. In a 2-year study with rats, early signs of Mn2þ intoxication were seen in young but not adult animals in the form of enhanced levels of the enzyme monoaminoxidase (Leung et al., 1982; Lai et al., 1982), a biochemical indicator fitting well with the systemic neurotoxicity of Mn in humans, called ‘manganism’ and exhibiting in late states symptoms similar to Parkinson’s disease. Higher sensitivity of young versus adult rats for the neurotoxic effects to orally ingested MnCl2 was confirmed in a short-time high-dose study (Dorman et al., 2000). Extensive information on the toxic and especially neurotoxic properties of Mn for primates (but without information on their dependence on chemical speciation) has been reviewed by Burton and Guilarte (2009). All neurological effects of Mn2þ as it is present in most (oxygen-poor) groundwaters were linked to minimal LOAELs of about 10 mg kg1 bm Mn2þ, corresponding to an estimated NOAEL of 1.4 mg kg1 bm Mn2þ seemingly suitable to be used as PoD (see Section 3.14.4.3.1). As Mn is an essential trace element with only a small margin between essentiality and toxicity, the intra-plus interspecific extrapolation should use a smaller than usual total factor (WHO, 2008). With a total factor of 10 to extrapolate from the PoD directly on a tolerable human Bd, its final number is Bd ¼ 10 mg 70 kg1 bm, a value identical with the earlier-mentioned upper limit of the AROI for soluble plus insoluble Mn but with the advantage of being based directly on toxicological

and not nutritional data. Moreover, this more toxicologically derived Bd opens the possibility of calculating a separate maximal value for Mn2þ in drinking water for preparing infants’ formula and for small children because the developing brain, similar to what is known from perinatal lead toxicity, is supposed to exhibit a special sensitivity for the neurotoxic potential of soluble Mn species like Mn2þ, the species to prevail in groundwater from single wells for drinking water, but not in de-manganized (oxygenated) water from public systems. When 20% of the present toxicological Bd are allocated on 0.75 l of drinking water for a child weighing 5 kg, a specially protective GV for infants of about 0.2 mg l1Mn2þ is obtained. The high potential for maternal and early life exposure to high levels of neurotoxic Mn2þ is underlined by findings in Bangladesh on a negative correlation between As and Mn with the consequence that waters seemingly safe with regard to As may actually be unsafe due to the presence of high soluble Mn which might even have provoked health effects erroneously ascribing them to arsenic (Ljung et al., 2009). This situation may by accentuated with weaned infants since they resorb Mn2þ from mother’s milk in distinctly smaller fractions than from liquefied dry milk. Natural radionuclides (group A in Table 2). In the third edition on radiological aspects of its guidelines for drinking-water quality, WHO (2008) proposes guidance for levels of radioactivity corresponding to a recommended reference dose level (RDL), that is, RDL ¼ 0.1 mSv a1 for drinking water as based on adult dose coefficients (DCs) for each radionuclide present in the specific sample. This RDL corresponds to an IZ (see Section 3.14.4.3.2) of about 104. If several nuclides are present in the same sample, the addition rule to cope with similar joint action would apply (see Section 3.14.4.3.4). WHO states that ‘‘the higher age-dependent dose-coefficients calculated for children (y) do not lead to significantly higher doses due to the lower mean volume of drinking water consumed by infants and children.’’ However, DCs for common natural nuclides in drinking water (Ra-228 and Ra226, and the decay products of Rn-222, namely Po-210 and Pb-210) exhibit much higher differences between adults and young children than their body mass-normalized intakes of drinking water (ICRP, 1996). The coefficients for two groups of children (0–1 and 1–2 years of age) and adults (417 years) are compiled in Table 4. The mean deviation of DCs between the one or the other children’s age group on the one side and adults on the other comprises factors of about 22 (0–1 a) and about 6 (1–2 a),

Table 4 Dose coefficients (DCs, (mSv kg1 bm)) as determined by ICRP (1996) for adults and two groups of children to calculate committed effective doses of some drinking-water-relevant natural radionuclides after oral uptake Age group

DC for Ra-228

DC for Ra-226

DC for Pb-210

DC for Po-210

0–1 a (infants) 1–2 a 417 a Ratio of highest to lowest DC

30 5.7 0.69 43.5

4.7 0.96 0.28 16.8

8.4 3.6 0.69 12.2

26 8.8 1.2 21.7

The mean deviation of DC between the one or the other children age group on the one side and adults on the other comprises factors of about 22 (0–1 a) and about 6 (1–2 a), whereas the respective drinking-water uptakes per person ( ¼ unit of exposure) vary only by maximal factors of 4 (0.5 l per 4-kg infant vs. 2 l per 70-kg adult) and 2 (1.0 l per 10-kg child vs. 2 l per 70-kg adult), respectively.

Drinking Water Toxicology in Its Regulatory Framework

whereas the respective drinking-water uptakes per person (¼ unit of exposure) vary only by maximal factors of 4 (0.5 l per 4-kg infant vs. 2 l per 70 kg adult) and 2 (1.0 l per 10-kg child vs. 2 l per 70-kg adult), respectively. It seems therefore justifiable to calculate effective doses from natural radioactivity in drinking water separately for children to see up to which levels of activity the RLD would yet be respected. The data not only from a comprehensive survey on natural radioactivity in drinking water from geologically suspicious areas but also from many unsuspicious ones in Germany support this view (Federal Office for Radiation Protection/Bundesamt fu¨r Strahlenschutz, 2009). If exclusively adult DCs are applied to transform measured activity values into effective doses, maximally 10% of the sampled public utility uses contain enough activity to ingest from Ra-228, Ra-226, Rn 222, Po-210, Pb-210, U-238, and U-234 to exceed the RDL of 0.1 mSv a1. When taking into account the higher sensitivity of infants for direct (genotoxic) DNA damage not only by chemicals (see Section 3.14.4.3.2) but also by irradiation, up to 22.5% of all sampled utilities would contain very high activity levels from these four radionuclides if the RDL of 0.1 mSv a1 should be used with respect to infants as well. The exposure scenarios assumed a yearly uptake of 170 l of drinking water by infants and of up to 730 l by adults, respectively. Nitrate (group C in Table 2). When evaluating nitrate in drinking water, one has to differentiate between its acute and chronic toxicity. Acute toxicity can be seen when a significant percentage (420%) of nitrate in the stomach is microbiologically reduced to nitrite and the latter then resorbed. Once in the blood, nitrite is able to oxidize Fe(II)hemoglobin (Hb) into Fe(III)Hb, the so-called methemoglobin (metHb). MetHb is unable bind oxygen and transport it from the lungs to peripheral tissues. If the metHb fraction becomes higher than 10%, the first signs of oxygen deficit may become apparent in peripheral tissues. The most probable target individuals for internal asphyxia from exposure to nitrite, called ‘methemoglobinemia’, are exposed infants since they exhibit, at the same time, low levels of physiological cytochrome b5-reductase in the blood and low gastric acidity, the latter allowing for survival of nitrate-reducing bacteria in the stomach. Although endogenous nitrate/nitrite may additionally arise in significant amounts also from internal oxidation of nitric oxide formed in macrophages and other cell types during infection (Speijers and van den Brandt, 2010), WHO (2008) in its 2007 background document on nitrate/nitrite assessed GVWHO ¼ 50 mg l1 nitrate as sufficiently protective for infants against methemoglobinemia. Continuous exposure to more than this GVWHO results in significantly higher than background (2% of Hb) metHb in the blood (Sadeq et al., 2008; Winton et al., 1971). If nitrate and nitrite are present simultaneously in drinking water, the addition rule for similar joint action applies (see Section 3.14.4.3), and it is recommended that nitrite should not exceed its GVWHO ¼ 3 mg l1 because its metHb-forming potential is estimated to exceed the one from nitrate by about 10-fold. Nitrate levels between 50 and 100 mg l1, according to WHO (2008), are safe only under close medical supervision and only if the nitrate-contaminated water is known as being microbiologically safe.

403

Chronic toxicity of nitrate has three aspects, its systemic toxicity, its goitrogenic properties, and its potential as indirect chemical precursor of carcinogenic N-nitroso-compounds as formed from amines and nitrite. Systemic chronic toxicity of orally ingested nitrate was detected in a long-term rat study in the form of growth inhibition at the earliest above a NOAEL of 370 mg kg1 bm nitrate. WHO (2008) confirmed this NOAEL as PoD to keep on its older ADI of 5 mg kg1 bm Na-nitrate or 3.7 mg kg1 bm nitrate ion, corresponding to a Bd of 259 mg d1 in a 70-kg person. With a relative source contribution of 10% in 2 l of daily drinking water, a health-based GV of 13 mg l1 for chronic toxicity of nitrate is obtained, hence just a quarter of the maximal value based on acute toxicity for babies. As nitrate 410 mg l1 is not normally a geogenic constituent of drinking water, its tolerability above this should be judged referring to its occurrence as an environmental contaminant (see Section 3.14.5). Its actual oral intake with drinking water due to agricultural over-fertilization in Europe amounts to 25% instead of 10% of the total daily intake (EFSA, 2008). Moreover, nitrate is the only environmental contaminant worldwide whose current GVWHO and even actual concentration in many ground- and drinking waters come very close to values not to be exceeded in order to protect a large group of the population (weaned babies) from acute health effects. The conclusion is that environmental contamination by this agricultural nutrient has exceeded since long time and in many parts of the world at any hygienically tolerable proportion or scale. The goitrogenic properties of nitrate were ascribed to its competition with iodine for transport into the thyroid. According to WHO (2008), this effect is weak or absent at or below its GV of 50 mg l1 nitrate if daily ingestion of iodine is sufficient (150–300 mg d1). However, already at 90 mg l1 nitrate exposure struma incidence was increased highly significantly in pregnant women and, possibly, in their children (Gatseva and Argirova, 2008). On the other hand, WHO in its Food Additives Report 50 states that nitrate, instead of inhibiting iodine transport, may stimulate it in humans even at threefold ADI exposure. A ‘‘further concern relating to the metabolism and toxic potential of dietary nitrate and nitrite is the potential formation in vivo of carcinogenic N-nitroso compounds from nitrite, or the derived nitrosating species, N2O3 and N2O4, and dietary amines’’ (Speijers and van den Brandt, 2010). Volunteers under a diet regimen rich in dietary amines excreted much more N-nitrosamines, especially N-nitroso-dimethylamine, if they ingested nitrate simultaneously at ADI levels (Vermeer et al. 1998). Due to complex interaction between exposure and cofactors such as meat intake (De Roos et al., 2003) or vitamin C and chewing gum (Mirvish et al., 1995), epidemiological studies on the relation between drinking-water nitrate and cancers provide, if any, only weak associations with either direction, positive or negative (Ward et al., 2005). A recent study by Chiu et al. (2007) on nitrate and bladder cancer is devoid of any individual information on the kind and source of nitrate exposure. Above all, it has none either for the preceding latency period for bladder cancer (20–40 years) or for the much more significant nitrate sources than drinking

404

Drinking Water Toxicology in Its Regulatory Framework

water, which in this study exhibited a mean difference between maximal and minimal nitrate values of only 0.4 mg l1. Furthermore, the study did not consider smoking and forgot to ask how frequently the test persons went to the restroom or how much drinking water they had been ingesting, albeit all three parameters could influence bladder cancer incidence dramatically in one or the other direction (Jiang et al., 2008). Again, like with regulation of DBPs in drinking water, the only reasonable regulatory-toxicological conclusion from this complex situation – on the one side, rather conclusive biochemical results and on the other side very inconclusive epidemiologic data – is to minimize exposure wherever this seems reasonably feasible but not to wait until nonambiguous scientific conclusions may or may not be drawn from this huge and ever-increasing data pool of very heterogeneous quality. Uranium (group A in Table 2). Water-soluble uranium (UO2 2þ ) was last assessed by WHO in the year 2005 (WHO, 2008) in the form of a health-based GVWHO referring to toxicological work published until 2002, including information published up to 2005, on how to eliminate uranium from drinking water. The assessment used an LOAEL from a rat study as a threshold-PoD (see Section 3.14.4.3.1) for renal toxicity. The WHO set its GVWHO at 15 mg l1 U, allocating 80% of the corresponding Bd of 36 mg kg1 bm on 2 l of drinking water per day for a 60-kg person. UO2 2þ predominates at low concentration in the presence of Ca2þ in the form of Ca2UO2(CO3)3 (Bernhard and Geipel, 2007), being distinctly less toxic than UO2(CO3)4 as shown with rat renal epithelial cells (Gouget et al., 2005). A recent assessment (Konietzka et al., 2005) with additional consideration of human data by Kurttio (2005) and experimental data from rabbits (Gilman et al., 1998b) leads to the derivation of a somewhat lower TDI than that proposed by WHO, mounting to 0.3 mg kg1 bm and corresponding to a Bd of 20 mg d1  70 kg bm. Its derivation starts from the rabbitPoD of 50 mg kg1 bm (instead of 60 mg kg1 bm from rats) and further assumes, as supported by kinetic data from animals as cited in Konietzka et al. (2005) and also by human data from Zamora et al. (2003), a higher resorption of uranium by a factor of 5 in humans (1.5%) as compared with rabbits (0.3%). The human data from Kurttio (2002) indicated a range of 6–60 mg d1 for a possible Bd when regarding the relative change of the creatinine-normalized calcium and phosphate levels (fractional excretion’) in blood and urine and their dependence on U uptake with drinking water. By means of 10 indicators for kidney toxicity of uranium, Kurttio et al. (2006) demonstrated a lowest no-observed effect level linked to uptake of U with drinking water as being possibly located around 100 mg l1, corresponding with a calculated human NOAEL of 200 mg d1 in a study with 193 adults. Prat et al. (2010) explain the ‘‘lack of significant health effects’’ of U in the Kurttio et al. (2006) study by low resorption rates of U from drinking water since UO22þ should prevail there are two calcium-dependent species, Ca2UO2(CO3)32 (Prat et al., 2010). Similar findings on a low resorption of these complexes from the matrix ‘mineral water’ had been reported already earlier by Bernhard and Geipel (2007). Nevertheless,

when extrapolating an NOAEL of 200 mg d1 U with EFd ¼ 10 on sensitive humans, a Bd of 20 mg d1 is obtained which corresponds well with a Bd as based on data from Zamora et al. (1998) on a slight increase of lactate dehydrogenase (LDH) and glucose in the urine of their most sensitive study subjects. The maximal contribution of uranium exposure from paths other than drinking water in the ecologic studies by Kurttio et al. (2002, 2005, 2006) and the one by Zamora et al. (1998) should not have been higher and, rather, was less than 50%. Under this assumption, the above estimations of a Bd from human data and the one from the rabbit study agree within a factor of 2. Given that, the human data seem more reliable to derive, eventually, a health-based GV of 10 mg l1, allowing for concurrent ingestion of another 20 mg d1 portion of uranium with food. Taken together, the final result of this assessment is not different from that by WHO (2008) since it similarly corresponds to a Bd of 40 mg d1 or a TDI of 0.6 mg kg1 bm applicable on total ingestion of uranium from food together with drinking water. The actual GVUBA of 10 mg l1 seems, however, somewhat better founded than the somewhat higher GV of WHO since the former mainly refers to recent human data (including higher resorption) which prevail under reallife exposure conditions and relies especially on a more realistic 50:50 instead of 80:20 splitting of uranium exposure between food and drinking water. In order to detect higher than average uranium exposure, such as from a private well, it is recommended to have an internal indicator of enhanced exposure. The Federal Environment of Germany proposes for this purpose a reference value of 30–60 ng l1 U in the urine as based on an upper normal level of 30 ng l1 U (Federal Environment Agency/ Umweltbundesamt, 2005). The upper limit of this reference range might be exceeded at permanent ingestion of 10 mg l1 U with drinking water plus normal U uptake from food (Kurttio et al., 2002).

3.14.6.3.2 Organics Perfluorinated compounds (PFCs; group C in Table 2). Linear perfluorinated tensides (PFTs) have not been assessed yet for drinking water in the form of a health-based GV by WHO (2008). The two most important PFTs, perfluorooctanoate (PFOA) and perfluorooctanesulfonate (PFOS), are used as surfactants in a variety of industrial processes and consumer products. Due to their physical–chemical properties, they spread and persist in environmental media, wildlife, and humans, in the latter mainly as a consequence of dietary intake (Fromme et al., 2008). PFOA and PFOS have elimination half-lives of several years in humans. Biomonitoring studies proved a clear relation between plasma load and drinking water contamination. Hepatotoxicity, developmental toxicity, immunotoxicity, hormonal effects, and also a weak carcinogenic potency in animal studies have been described as main endpoints of health concern. Their potential to contaminate raw water resources and drinking water up to health-related values has recently been summarized by Wilhelm et al. (2010b).

Drinking Water Toxicology in Its Regulatory Framework

Exposure to PFOA via contaminated drinking water was observed at water districts in Little Hocking, Ohio, USA at up to 3.5 mg l1 (Emmett et al., 2006), in the German city of Arnsberg at up to 0.64 mg l1 (Skutlarek et al., 2006; Ho¨lzer et al., 2008), and in Minnesota, USA, at up to several mg l1 (MDH, May 2009). PFOA can be effectively removed from drinking water by percolation over activated carbon (Wilhelm et al., 2008). PFTs leach from contaminated soils and firefighting foams under areas contaminated by fire brigades either by training or by fighting against conflagrations. Wastewater, either from chemical industry or from communal wastewater disposal plants, causes diffuse aquatic contamination. Sums of various PFTs in drinking water from bank filtration along the rivers Rhine and Ruhr amount mostly to between 0.05 and 0.15 mg l1 (Wilhelm et al., 2010b). In June 2006, the German Federal Advisory Board on drinking water (TWK) proposed the first worldwide healthbased GV for safe lifelong exposure at 0.3 mg l1 to sums from PFOA and PFOS, assuming effect additivity for both compounds by the mode of similar joint action (TWK, 2006). In accordance with this addition rule, the sum of all quotients from a measured concentration and the respective healthbased reference value (HRIV or GV, see Section 3.14.6.3.2 or 3.14.2.5.2) is deemed not being 41. For PFOA, the GV of the German TWK represents about 10% in 2 l of drinking water d1 70 kg bm of the TWK’s provisional Bd for PFOA which equals 7 mg d1. This Bd was derived from an estimated NOAEL of PFOA for reproductive toxicity in rats by applying two extrapolation factors (EFs) of 10 each for inter- and intraspecies biologic variability and an additional SF of 10 to cope with uncertainties due to the much longer elimination half-time of PFOA in humans than in rats. Regarding PFOS, TWK reverted to an NOAEL of 0.025 mg kg1 bm for proliferation of peroxisomes in chronically exposed rats. This toxic endpoint was considered by TWK to be sensitive enough for being extrapolated immediately (with an intraspecific EF ¼ 1) on sensitive humans. On the other hand, due to the exceptional persistence of PFOS in humans as compared to rats, TWK applied an exceptionally high interspecific EF ¼ 30 as supported by toxicokinetic results from animals and humans. The Bd calculated this way for PFOS was 0.083 mg d1 and eventually turned out to be very similar to Table 5 Compound

PFBA PFPA PFHxA PFHpA PFOA PFBS PFPS PFHxS PFHpS PFOS

405

the one (0.1 mg d1) for PFOA. More details have been published on behalf of the German Federal Environment Agency (UBA) by Dieter (2007). It should be mentioned that published GVs for PFOA in drinking water differ considerably, the lowest (0.04 mg l1) coming from the New Jersey (NJ) Department of Environmental Protection (Post et al., 2009), while Tardiff et al. (2009) proposed a lifetime safe drinking-water equivalent level (DWEL), whose 10% allocation on 2 l d1  70 kg bm would correspond to a GV of about 1 mg l1 of PFOS or PFOA, resulting in a maximal value comparable to those from US-EPA (2009), the German UBA, or the Drinking Water Inspectorate of England and Wales (DWI, 2007). The most significant drawback of the very low NJ maximal value for PFOA seems to be inadequate extrapolation of rat serum PFOA levels on humans (Tardiff, 2009). Moreover, if ever the NJ maximal value would be toxicologically sound, then exposure from diet would primarily need to be reduced (Zhang et al., 2010). Shorter-chained PFTs (C4–C7) are less persistent than the C8 analogs and hence are introduced as replacements of outphased PFOA and PFOS. Some of these will, possibly, be the main future contributors to total PFT levels in raw and drinking water, since they dissolve better in water and hence are more difficult to remove by percolation over activated charcoal. An approach to assess shorter-chained PFCs and their mixtures is explained by Wilhelm et al. (2010b). It refers to a proposal launched by the Federal Environment Agency of Germany (Umweltbundesamt, 2009a) to provisionally assess the toxic potential of any single PFC mainly on the basis of its measured or anticipated elimination half-life from the human body and to decide accordingly on a possible maximum level value for the same compound in drinking water (called HRIV in the case of a provisional health-related indication value, or GV in the case of a scientifically based guide value, see Section 3.14.6.3.2 or 3.14.2.5.2). The range of possibly tolerable concentrations for any single PFT in drinking water extends between a minimal GV of 0.3 mg l1, as derived by TWK for the very persistent PFOA and PFOA or their sums, and a maximal GV of 7 mg l1 applicable only on the easily excreted PFBA as derived in February 2008 by the Minnesota Department of Health (MDH, 2009) as explained by Wilhelm et al. (2010b). Table 5 lists all HRIVs and GVs for PFTs as applied currently in Germany (Umweltbundesamt, 2009a). Additionally,

Health-related indication value, guide values, and long-term precautionary quality goal in drinking water for PFTs Long-term precautionary value (PV; precautionary quality goal) for each single PFC and their sums (Section 3.14.2.4.1)

0.1 mg l1

Health-related indication value (HRIV) (Section 3.14.6.3.2)

Guide value (GV) (Section 3.14.4.3.1)

F 3 mg l1 1 mg l1 0.3 mg l1 F 3 mg l1 1 mg l1 0.3 mg l1 0.3 mg l1 F

7 mg l1

0.3 mg l1

0.3 mg l1

406

Drinking Water Toxicology in Its Regulatory Framework

the long-term precautionary quality goal for any PFT and their sums in drinking water is shown. All numbers are supported also by the German TWK. If shorter-chained PFTs than PFOA and PFOS are present concurrently in drinking water, the addition rule applies as well, assuming again the effect additivity by the mode of similar joint action (see earlier discussion). The health-based reference value on which the quotients to be summed up are based upon is either the GV (PFOS, PFOA, and PFBA) or a provisional HRIV (all others). Nitro- and nitroaminocompounds (NAC; group C in Table 2). Nitro- and nitroaminocompounds (NACs) are important industrial chemicals with a broad area of application. They serve mainly as readily available intermediates for dyes, pharmaceuticals, synthetic materials, and other applications. On their own, they are used as solvents, explosives, and pesticides. NACs are also emitted from many military areas or hot spots of abandoned explosives production sites. Due to their hydrophilic properties and relative stability, they may penetrate into aquatic systems and thus also into drinking water resources. The environmental fate of NACs is determined by biodegradation and photolytic degradation and results mainly in the formation of equally amino nitroaromatics and amino aromatic compounds. The WHO (2008) has not recommended drinking-water guideline values for any NAC except recently for nitrobenzene, for which a GV of GVWHO ¼ 8 mg l1 for cancerous and noncancerous endpoints was published in 2009. GVs for 19 NACs, including nitrobenzene, have been proposed by Wollin and Dieter (2005) and adopted as an official guide or advisory values in Germany (UBA, 2006). The threshold approach (Section 3.14.4.3.1) was used for 12 of them, resulting in corresponding GVs between 0.7 mg l1 (for nitrobenzene, similar to the 2009 GV from WHO) and 175 mg l1 (for octogen or HMX). Estimates of excess lifetime cancer risk (see Section 3.14.4.3.2) were used for evaluating five NCs, resulting in GVs between 0.05 mg l1 and 2 mg l1, depending on the individual NAC. GVs for the remaining 3- and 4-nitrotoluene were 10 mg l1 and 3 mg l1, respectively, referring to their organoleptic properties. Human exposure surveillance for those NACs which exhibit an aromatic moiety is routinely possible by monitoring hemoglobin adducts (HBAs) as proposed by Neumann et al. (1995), Sabbioni et al., (1996), and Thier et al. (2001). Although HBA levels do not immediately reflect the genotoxic/ carcinogenic potential or potency of a chemical under conditions of real exposure, they correlate usually much better with the internal biologically active dose than any assessment of external exposure. New analytes – and how to assess them? The evaluation in terms of environmental hygiene and human toxicology of substances which enter adjoining environmental compartments from (more or less) open applications or solid materials or formed there as metabolites is not legally regulated, except perhaps for certain metabolites from agricultural pesticides (Steinha¨user and Richter, 2002). Such new analytes very often have a high affinity to the aquatic environment, are very mobile there, and many of them are also persistent. The very polar ones, especially, may enter water sources by bank filtration and are difficult to retain

from drinking water, at least by nature-oriented treatment or by percolation over activated charcoal (Ternes and Joss, 2006). Their human toxicological potential can be prognosticated as being moderate to low, although the data to base such judgment when looking for single compounds are often sparse or missing completely (see Section 3.14.4.2). Sparseness of data for health-based evaluation of new analytes or emerging contaminants in drinking water may be concluded from a compilation by Schriks et al. (2010). Although Schriks et al. propose many GVs for drinking water, their rationales differ strongly by the quality of database. The underlying data originate very often from very old studies, short-time experiments, or exposure modes inapt to asses exposure from drinking water (gavage, inhalation, and injection). Extrapolation (rather: safety) factors to cope with such serious experimental and conceptual drawbacks are always more or less speculative and do not really help to describe sustainable drinking water for the future. On the other hand, missing or bad data do not automatically imply health risks, but they should, if possible, be completed in order to define or provide regulatorytoxicological certainty and safety. Up to this moment, it is necessary to have criteria available to evaluate the presence of substances in drinking water in terms of their toxic potential for humans even on a patchy database. These criteria must be allocated to concentrations that are at any rate equal to, but possibly lower than would otherwise be a sound health-based GV. To meet this goal, two systematic concepts exist to provisionally assess new analytes with patchy database: 1. an approach, called TTC concept, to provisionally define a threshold of toxicological concern in function of structural alerts or analogies as recently explained for application on nonrelevant metabolites from pesticides by MelchingKollmuss et al. (2010) and 2. a tiered experimental and exposure assessment approach (Dekant et al., 2010). The first way leads to a toxicologically safe, albeit provisional concentration (DWRTTC) for the analyte under question in drinking water; the second (and more laborious) one would open the possibility to assess, on a case-by-case basis, any new analyte above its DWRTTC and present in drinking water. Case-by-case risk assessments are the approach of choice to evaluate past anthropogenic contaminations of resources and environmental media during required sanitation (Section 3.14.2.4.3). In order not to risk such scenarios, quality of all waters for human use must strictly satisfy precautionary principles. A more effective (and less laborious) way to keep media and resources for the future in a sanitized (close to nature or unobjectionably safe) state is to avoid cause and need for case-specific assessments from the beginning. This situation in mind, a third approach may be recommended as a pragmatic and, at the same time, sustainable default approach (UBA, 2003).At its core are five provisional HRIVs (formerly called HPVs – health-related parametric values) to pragmatically assess the presence of new analytes on a patchy human toxicological database for purposes of healthrelated long-term chemical drinking water hygiene. These values eventually turned out to cover practically the same

Drinking Water Toxicology in Its Regulatory Framework concentration range 0.01–3 mg l1 as the four lower DWRTTC (ILSI, 2005) but the HRIV concept from UBA (2003) is simpler and therefore easier to handle. Depending on the completeness and informative value of the toxicological database for defined endpoints (Table 6), the HRIV of any new analyte with a patchy database is set for lifelong exposure at one of four concentration steps between 0.1 and 3.0 mg l1, each being a factor 3 away from the neighboring lower and higher one. The maximal HRIV is the same as the one derived by means of the TTC concept for Cramer’s toxicity class III, which according to Dekant et al. (2010) corresponds to 3 mg l1 instead of 4.5 mg l1 as was proposed yet in Barlow (2005). Table 6 Health-related indication values (HRIVs) as defined provisionally by the toxicological ‘default’ approach, called HRIV approach to assess presence of trace concentrations of new analytes in drinking water Designation

Numerical value (mg l1)

Explication

HRIV1

0.1

HRIV2

0.01 to o0.1

HRIV3

0.3

HRIV4

1.0

HRIV5

3.0

HRIVQSAR

0.1–3

HRIV6

43

For contaminants known to exhibit no genotoxic potential or if not tested as such in the absence of any other information For contaminants known to exhibit strong to weak genotoxic potential For contaminants known to be devoid of genotoxic potential, additional data on germ cell toxicity, immunotoxicity, and reproduction toxicity do not support any lower value Similar to HRIV3; additional data from at least one study on subchronic toxicity do not support any lower value Similar to HRIV4; additional data from the only available study on chronic toxicity do not support any lower value. An HRIV within this range may be set for a contaminant referring on known toxic potentials of similar structures (or elements thereof) as concluded by analogy or from QSAR considerations Similar to HRIV5; additional data on chronic toxicity do not support any value oHRIV5. Completeness of database may also allow to scientifically derive values4HRIV5. In most cases, an HRIV6 may be set as a scientific guide value (GV)

An HRIVx is lifelong tolerable and increases with completeness of database and decreasing severity of the toxic endpoint if not experimentally tested.

407

Another very low HRIV is reserved for analytes to exhibit primarily a strong in vitro genotoxic (DNA-damaging) potential as detectable in routine Ames tests. Such analytes are recommended by UBA (2003) for being tolerated depending on their in vitro genotoxic potential at 10–100 ng l1 only. HRIVs have the regulatory-toxicological function of healthrelated, not absolute precaution. To define them in a first step does not normally ask for analyzing heterogeneous expert systems or for screening any QSAR data files, as is the case when using the TTC approach: HRIVs are easier to define for a specific analyte than an appropriate DWRTTC. However, if ever an HRIV definitely appears as too low for practical purposes and, if in a second step, the structure or elements to be assessed are not found in pertinent QSAR data files, the first-step HRIV should always be examined for plausibility or even tested in vitro before taking irreversible regulatory measures. In case of doubt, it is more reasonable to refer such measures on the ALARA-principle than on a toxicological hazard referring to no hazard but data gaps (see also Section 3.14.4.3.4). To summarize, the HRIV approach is nothing else but a kind of pragmatic TTC concept as designed specifically for drinking water. Its regulatory-toxicological reliability is comparable or even better. The rationale for all five HRIV levels and their gradation is UBA’s own and literature-based international experience from assessing many well-described hydrophilic (highly polar) drinking-water contaminants of very different grades of toxicity, both support the continued assumption that such assessments also in future never would result in lower (but mostly higher) GVs than the respective HRIV for the same or similar toxic endpoint. It is an approach to combine knowledge on toxicological relevance with knowledge on the practical ‘relevance’ of hydrophilic environmental contaminants for drinking water before or after treatment. Such a pragmatic approach to bridge the gap between problem identification and problem solving was asked for lastly by Fawell (2008). Application of the HRIV-approach on side products of oxidation and disinfection. Drinking water, containing new analytes below endpoint-specific HRIVs, according to UBA (2003) may also be considered as being sufficiently safe for toxicologically relevant oxidative transformation products, possibly being formed from new analytes during oxidative treatment steps. However, even drinking water, where one or several new analytes may be detected above an endpoint-specific HRIV, is suited for human consumption without imminent health risk. Only in special cases (relatively high concentration of a new analyte), examining the finished water for transformation products after oxidative treatment could be suggested if there are no further treatment steps to eliminate, absorb, or degrade a possibly hazardous polar compound. Application of the HRIV approach on pharmaceuticals in drinking water. The HRIV approach is specially suited for being applied for assessing from the point of regulatory toxicology, the presence of residues from pharmaceuticals and their metabolites in drinking water. Their major source of input into aquatic environments and from there via wastewater into reused drinking water is their excretion after intended use by humans (Ternes, 2007; Focazio et al., 2008). The human database for this group of new analytes at first glance seems quite complete, as their toxic potential normally is assumed of

408

Drinking Water Toxicology in Its Regulatory Framework

being linked to therapeutic dose (Webb et al., 2003). The underlying assumptions on the informative value of therapeutic dose on chronic toxicity may or may not be true. In any case, clinical tests as linked to therapeutic doses do not inform on effects possibly to observe at lower doses than normally used in preclinical studies with experimental animals. There are many examples (mostly old compounds) for the absence of systematic toxicological examination of typical systemic endpoints such as chronic toxicity, neurotoxicity, reproductive and developmental outcome, or even cancer. Moreover, serious side effects (expected and unexpected ones) in patients, usually occurring in frequencies around 1%, may seem acceptable when weighed against the drug’s benefit and the possibility to exclude any involuntary risk by accordingly informing the therapeutic target population. Most probably, such populations are incomparably much smaller than any population exposed to a large supply of drinking water. When taking the possibility into account, that unobserved effects in fact may be present but just have been overlooked due to inadequate testing, possible Bd and maximal values to assess pharmaceuticals and their metabolites in drinking water would have to be derived by using very large margins of safety between a Bd and any experimental outcome, amounting easily to 1000, 3000 or more (Bull et al., in press; personal communication by J. Cotruvo), signaling then such a high incompleteness of the database that assessment of pharmaceuticals in drinking water should preferably be performed using a precautionary default-approach. Only exceedance of an HRIV would then have to result in measures of management to either complete the respective compound’s database or reduce exposure via drinking water. The basic idea behind is to preferably define scientifically the presence of safety as long as scientific proof for absence of toxicity is not required or not yet possible. Application of the HRIV approach on nonrelevant metabolites of pesticides in drinking water. A recommendation similar to UBA (2003) but specified for a group of unregulated new analytes, being called nonrelevant metabolites of pesticides in the European Union, has been published (UBA, 2008) and its motivation and regulatory context explained (Dieter, 2010). The parent pesticides and their relevant metabolites are subject to an extensive admission process, whereas the database of the remaining, so-called nonrelevant metabolites is more patchy. Nevertheless, across the board, their toxic potential can be judged as being either lower or better documented than in the case of new analytes from other sources. This is why only the two upper HRIVs (1 mg l1 and 3 mg l1) are recommended to assess the presence of these new analytes for lifelong exposure from the point of drinking-water hygiene. Pesticides and their relevant metabolites. Many countries all over the world, except members of the EU, regulate agricultural pesticides and their metabolites in drinking water as if these were present there for functional reasons by assigning them strictly health-related maximum values as was explained by Hamilton et al. (2003). They model themselves mainly on WHO, although WHO, since 1984, never ceased to qualify its GVs as describing a minimal and not an optimal drinking water quality for lifelong consumption (see Section 3.14.2.2). This view of WHO goes well with the EU members’ view on

pesticides and their similarly regulated relevant metabolites in drinking water as nonfunctional contaminants. Consequently, in the EU, they are regulated in drinking water, as outlined in Section 3.14.1.2, by an (agro-)technical maximal value, factually a precautionary limit value of PV ¼ 0.1 mg l1 per compound, being more or less but always lower than a lowest possible health-based value (some old polychlorinated pesticides being the only exceptions). The same concentration of 0.1 mg l1 is recommended by UBA (2003, 2009) as a general precautionary value – PV – for any nongenotoxic drinking-water contaminant with patchy or missing database (see also Section 3.14.8). Cyanotoxins. Cyanotoxins are hepatotoxins and neurotoxins produced by cyanobacteria, also known as blue-green algae and commonly found in over-fertilized surface water. The hepatotoxins are mostly microcystins. Their chemical structure includes two variable amino acids and an unusual aromatic amino acid. They differ by lipophilicity and polarity, affecting toxicity as well. Microcystin-LR has been associated with most of the incidents of toxicity involving microcystins in most countries. It is a cyclic heptapeptide with a molecular weight of about 1000 Da. Neurotoxins are not considered as widespread in water supplies, and they do not appear to pose the same degree of risk from chronic exposure as microcystins. Some of them, such as anatoxin-a and -a(s), are highly toxic nerve poisons but have short biological half-lives. Toxic cyanobacteria also produce cytotoxic alkaloids; the most recently described is a tricyclic guanidine linked to a hydroxymethyl uracil. These alkaloids have been implicated in a variety of health effects, ranging from gastroenteritis to kidney disease. WHO (2008) proposes a provisional GV of 1 mg l1 for only one microcystin (microcystin-LR) and derived it from a 13-week rat study for its liver pathology. This GV represents 80% of the Bd of 2.4 mg per 60-kg person for only this compound. New data for the toxicity of cyanobacterial toxins are being generated (WHO, 2008a). Other organics. Many more organic compounds, besides the few groups described here, occur occasionally in drinking water. They have, if of some importance at the consumer’s tap, been partially addressed in Section 3.14.6.2. Extensive toxicological information on any of these compounds can be found at WHO (2008) and on those relevant in groundwater in compact form in Schmoll et al., 2006.

3.14.7 The Author’s Short Conclusions 3.14.7.1 Unintended Exposure Unintended exposure is synonymous with useless exposure. Useless exposure may be present in drinking water in the form of either geogenic or biogenic constituents or anthropogenic contaminants. Constituents must often be accepted and hence controlled accordingly at the end of pipe, which here usually is identical to its beginning. Treatment of drinking water to eliminate natural constituents, except for manganese and iron, is indicated only if health-based values are exceeded and a better

Drinking Water Toxicology in Its Regulatory Framework

source is not available, be this for direct use or for blending. Most relevant (hazardous) geogenic constituents are arsenic, fluoride, and soluble manganese. Lower than health-based natural levels of undesired geogenic constituents are not a sound reason for treating water. Minimal level goals for desired natural mineral constituents, for example, within the context of central softening of drinking water, are a reasonable option to provide a minimal proportion of the respective daily physiologic need. However, the consumers’ acceptance for fixing such regulatory minimal levels could turn into regulatory misuse of drinking water as carrier of involuntary medication. Contaminants must not be accepted but may be tolerated in drinking water as long as precautionary health-based levels are not exceeded and emission is controlled at the beginning (on site) instead the end (off site) of pipe. This should be done in all three dimensions of the basic rule of environmental hygiene and be implemented technically in terms of the ALARA principle. The only anthropogenic contaminant of drinking water to which WHO, since several decades, had to assign a maximal value based on acute human toxicity (for babies) is nitrate. This situation deserves any precautionary effort to reduce exposure to nitrate via drinking water also, more so as the discussion on the role of nitrate/nitrite as precursors of some forms of cancer in humans is far from being conclusive in terms of congruence between scientific certainty and healthcentered safety.

409

the millions of people of very variable sensitivity and possibly exposed to hazardous chemicals, comprises a huge potential of irreversible and/or difficult to retrievable failure. This concerns both, technological equipment and health, if not protected as a whole and managed or surveyed accordingly. This is why drinking-water toxicology is not just a scientific method to allocate by the rule of three a certain proportion of a Bd on the daily amount of consumed drinking water and let the management go how and where it wants. Precautionary protection of drinking water and of its resources is best implemented if management and regulatory toxicology set their quality standards in close cooperation and knowledge of each other’s expertise in defining goals of protection and technical functionality. In order to achieve these goals, they convene on standards being at the same time not only scientifically stable and technically feasible, but also resistant to scientific error and technical failure. The most important part of regulatory toxicologists within this cooperation is their openness to intelligent guess and to formulate scientifically guided hypotheses on which events or changes in a given drinking-water system might give rise to toxicologically relevant change in future. By doing this, however, he or she should try to never glide into contradictions to basic toxicological science. They should also be always informed on most recent possibilities to theoretically identify and practically quantify and prognosticate adverse human health effects as early as possible, optimally on the cellular and molecular level.

3.14.7.2 By-Products of Disinfection and Oxidation By-products of DBP and oxidation (OBP) in many countries represent the highest proportion of xenobiotics present in drinking water. DBP formation is to be accepted if chemical disinfection of drinking water is indispensable in a given situation and need and choice of a disinfectant is optimized and minimized according to technical possibilities and expertise. The best way to minimize DBP without endangering disinfection is to minimize concentration of DBP precursors and to maintain the distribution net in a shape as unobjectionable as the drinking-water quality the consumers desire at their tap. Any discussions on health effects of DBP seem of relevance only when referring to trihalogenmethane (THM) values above about 50 mg l1 and even then only with exact (individual) knowledge of exposure right from the tap. Again, like with nitrate/nitrite, the best way to cope with this is precautionary reduction of exposure to DBP, while never impairing the efficiency of disinfection. OBP may be reasonably assumed to exhibit some toxicological relevance in finished water from waterworks where ozonation is not followed by percolation over activated charcoal. Some OBPs, however, might pass the filtration, and periodic surveillance of the filtrate or finished water for total mutagenic activity could help define safety instead of dealing with toxicological uncertainty.

3.14.7.3 Risk Assessment and Management The whole system of drinking water, from an ever-vulnerable resource to the technology of treatment and distribution up to

3.14.7.4 Derogations from Limit Values To decide on efficient and best-adapted measures to deal with exceedance of a maximal value, it is of crucial importance to know about the rationale behind such value in order to be able to adjust their management accordingly. Only in cases of acute incidents concurring with extreme exposure, it seems indicative to interrupt a contaminated supply for a short time (some hours to maximally 1 day or so) to retrieve exposure from the pipes. The overwhelming number of toxicologically relevant derogations relates on much lower maximal values designed for hygienic tolerance of lifelong exposure. Given this, a supply presumably never needs to be immediately interrupted for health reasons if such ‘low’ maximal value is exceeded. It could even be a substantial error to act against this rule, the more as such action would entrain significant microbiological health risks when impairing correct functioning of the waterborne sewage system.

3.14.7.5 Drinking-Water Installations Drinking-water installations, if not maintained and operated properly or if built with inappropriate construction materials, are a common reason for adulteration of the quality of finished water when it has passed the handover point. A general recommendation is to drink or prepare food only with a tap water of optimal quality, thus having no stagnation or (as a rule) at most 2–3 h of stagnation in domestic pipes behind it. Adulteration by stagnation routinely concerns only esthetic parameters or annoyance. However, specific toxic potentials for infants are always associated with lead from plumbing and

410

Drinking Water Toxicology in Its Regulatory Framework

sometimes possibly with copper in hard water coming from new copper plumbing.

3.14.7.6 Surveillance of Drinking Water Continued surveillance of chemical drinking-water quality should, if possible, concentrate on such parameters which in a given resource or technical context may be subject to change. Such concept of surveillance is the better justifiable, the better the characteristics of the resource are known (and accordingly protected) and the less technical equipment is involved in drinking water production. A resource where the main part of surveillance is done by nature itself is always preferable to a resource where such nature-oriented production of drinking water is not or only nearly achievable.

3.14.8 Perspectives on Perception of Drinking Water This subject needs to be treated in the context of perceptional aspects putting drinking water in relation to bottled water and wastewater.

3.14.8.1 How Pure Is Pure? Centrally supplied drinking water is deemed to be of a high quality but low-price medium to meet everybody’s essential daily fluid requirement, for preparing food, and to advance personal, domestic, and public hygiene. Its mode of central distribution and inherent difficulties to repair any mistake make it inevitable to keep resources and technical facilities in a microbiological and chemical state as clean and safe as possible on a short- and long-term scale. The mode of perception and hence acceptance of tap water by the consumer, however, follows not only such abstract health-related criteria but also esthetic ones such as color, smell, taste, turbidity, or – last but not least – purity. Such criteria are often much more ambitious than mandated by health considerations (Doria et al., 2009). On the other side, in highly populated regions, the main purpose of public surveillance and maintenance of drinking water cannot be an absolute purity unless the principle of Augias would be elated as authoritative to manage the eternal resource, in our domestic case called drinking water. In fact, water is an eternal source, at least where quantity and its potential to be cleaned, recycled, and reused forever is concerned. However, the answer to the question what degree of purity with regard to the contamination of a reused drinking water may seem socially acceptable below strictly health-based limits, should not be bargained from day to day. Instead, this question should rather be negotiated referring to criteria such as functionality and prevention of exposure, ecotoxicology, and overall rationality of water management, such as was proposed in Section 3.14.1.2, in the form of a three-dimensional rule of environmental hygiene to define inevitable and hence acceptable exposure between zero and adversity.

3.14.8.2 Erroneous Reasons to Ask for Absolute Purity Despite this basic intuition, erroneous demands in developed countries on an absolute quality of the daily water have

provoked a situation in which unique water resources all over the world are bottled and then are sold, although for horrendous prices, down the river (Arnold and Larsen, 2010). It is unlikely that the surge in bottled water consumption is about health benefits associated with such water. In most cases, bottled water is not of better quality than tap water (Parag and Timmons-Roberts, 2009). The majority of people from a recent study in Great Britain were satisfied with the quality of their tap water supply and believed it would not pose any adverse health risk, although they were consuming bottled water for some other reasons such as convenience, cost, or taste (Ward et al., 2009). One more reason may be that many people easily get insecure by news especially on pesticides or pharmaceuticals contaminating drinking water (IESK, 2008), without being able to rationalize their insecurity as is proposed in Section 3.14.6.3.2. Similarly, even an assumed higher mineral content of bottled versus tap water is perceived as an indicator of positive health benefit, although the majority of brands are quite poor in minerals. The more probable and irrational reason why so many people prefer bottled to tap water are the former’s increasing perception as an individual and ultrapure aquaceutical and the readiness of many consumers to pay (almost) any price to have their individual miracle water available everywhere (Petrie and Wessely, 2004). As a consequence, such globalization of many water treasures over usually huge distances asks for much energy, whereas one-way bottles increase the landfills. At the same time, publicly accessible rivers and lakes are deteriorating irretrievably in correlation with neglecting resource protection. There is no valid reason why people who do not drink bottled water should finance the handling and disposal of bottled water waste (Parag and Timmons-Roberts, 2009). While, on the one side, the production (and consumption) of 1 kg of meat asks for more than 10–15 m3 of water, there is, on the other side, only a minority of people who have the optimal daily amount of 100 l available; for others, even the daily minimum of 20 l is hardly affordable and often only in bad quality for horrendous prices from private dealers.

3.14.8.3 Some Good Reasons for Not Asking for ‘Absolute’ Purity In contrast to such discrepancy between private occupancy of and public need for high-quality drinking water, the principle of public drinking-water supply is an especially efficient variety of social ethics since it is sustainable and socially acceptable all around. Given the multibarrier source protection approach, it opens the way to use and reuse eternally the amount of publicly accessible freshwater while its available amount is limited by region. The sole condition is to measure and define the off-site acceptability of any residual contamination in all on-site dimensions of the environmental rule (Section 3.14.1.2) and the ALARA principle instead of exploiting more and more fresh (albeit geologically old) resources whose owners claim for or sell their ‘absolute’ purity. Many practical examples on how to recycle purified wastewater through multibarrier systems to finally produce an unobjectionable drinking-water quality meeting ALARA requirements have been collected in a valuable publication

Drinking Water Toxicology in Its Regulatory Framework

(Jime´nez and Asano, 2008). Such technical use of natural forces and facts unifies the economic, ecologic, and social ethic views on daily drinking water in a reasonable and practical manner. This holistic approach asks all stakeholders not to insist on their maximal (absolutely low, toxicologically exhaustive, economically maximally beneficial) quality criteria but to convene for defining an optimal drinking water, referring as exactly as possible on the (per capita) availability of regional freshwater yield. A proposal to put on a long-term scale such social contract into a number is to use 0.1 mg l1 for any anthropogenic contaminant of drinking water as a general precautionary and esthetic quality goal (see Section 3.14.6.3.2).

3.14.8.4 Many More Good Reasons for Not Exhausting Strictly Health-Based Levels Once the perception of drinking water by the consumer is confined to that of a solvent or even an abandoned sink for an unknown number of polar and persistent new analytes, it could quickly lose much of its present acceptance. Following Klo¨pffer and Wagner (2007), the adequate regulatory consequence is not releasing persistent chemicals into the environment – even if no negative effects are known. The longlasting efforts (tens of years since 1974) to avoid or to identify and quantify the risk potential of DBPs in drinking water may speak as an example in this context. Even these efforts are by far not finished yet. Once new analytes such as pharmaceuticals, pesticides, industrial chemicals, flagrants, or endocrine-disrupting compounds are present in the water cycle, they can hardly be retrieved technically. Therefore, the sociopolitical discussion on acceptable and tolerable contamination and risks caused by environmental contaminants in drinking water will go on. Public supply of drinking water is a task of social ethics. The innocence, called absolute purity, of a privately provided water or a transcendentally transfigured aquaceutical is an egoistic chimera. When talking about economic use of water and how to save it, it is indispensable to consider domestic drinking water strictly separated from water in general. Replacing drinking water by domestic graywater as separated from piped or bottled water for human consumption, in general, is a very inefficient, unsustainable concept to save water. The margins to reduce domestic flow rate are very small and below a certain per capita level very cost intensive (Sections 3.14.1.1 and 3.14.6.2.9), whereas huge margins to save water are to be observed worldwide in agriculture and industry.

3.14.8.5 How to Best Realize the Social Concept of an Esthetically Acceptable Drinking Water? Saving water in moderately to densely populated areas is best realized by perceiving wastewater not as a private but as a centrally collectable economic good. The esthetic concept called purity of the daily drinking water is best realized by strengthening esthetics of natural aquatic life together with the regional water cycle by artificial infiltration of groundwater and the use of purified, microbiologically safe wastewater for agricultural use. Stakeholders should convene to arrive at an

411

agreement to answer the societal question ‘‘How much too much is just about little enough?’’ The result then may be considered a socially acceptable zero level. This level can neither be the up-to-date lowest analytical zero nor a merely health-based upper one. The first alternative would make drinking-water regulation a hostage of day-to-day analytic certainty, the second one of a never completely certain toxicology. A level of r0.1 mg l1 per any nongenotoxic analyte as recommended by UBA (2003) appears to be safe, socially acceptable, and technically feasible under any aspect not only of regulatory toxicology but also of drinkingwater hygiene and esthetics.

References Ahsan H, Chen Y, Parvez F, et al. (2006) Arsenic exposure from drinking water and risk of premalignant skin lesions in Bangladesh: Baseline results from the Health Effects of Arsenic Longitudinal Study. American Journal of Epidemiology 163: 1138--1148. Alam N, Corbett SJ, and Ptolemy HC (2008) Environmental health risk assessment of nickel contamination of drinking water in a country town in NSW. NSW Public Health Bulletin 19: 9--10. Anonymous (2010) Committee on Toxicity Testing and Assessment of Environmental Agents: Toxicity testing in the 21st century – a vision and a strategy, Washington DC, 2007. Human Toxicology 29: 15--29. Araya M, Chen B, Klevay LM, et al. (2003) Confirmation of an acute no-observedadverse-effect and low-observed-adverse-effect level for copper in bottled drinking water in a multi-site international study. Regulatory Toxicology and Pharmacology 38: 389--399. Arnold E and Larsen J (2010) Bottled water: Pouring resources down the drain. Earth Policy Institute. http://www.earth-policy.org/index.php?/plan_b_updates/2006/ update51 (accessed April 2010). Arnold WA, Schmidt CK, Salhi E, and von Gunten U (2010) Kinetics and mechanisms of N-nitroso-dimethylamine (NDMA) formation upon ozonation of N,Ndimethylsulfamide (in preparation). Barlow S (2005) Threshold of Toxicological Concern (TTC) – a tool for assessing substances of unknown toxicity present at low levels in the diet. ILSI Europe Concise Monographs Series 2005: 1--31. Barnes KK, Kolpin DW, Furlong ET, Zaugg SD, Meyer MT, and Barber LB (2008) A national reconnaissance for pharmaceuticals and other organic wastewater contaminants in the United States Groundwater. Science of the Total Environment 4002: 192--200. Bellar TA, Lichtenberg JJ, and Kroner RC (1974) The occurrence of organohalides in chlorinated drinking waters. Journal of the American Water Works Association 66: 703--706. Becker K, Kaus S, Helm D, et al. (2001) Umwelt-Survey 1998, Band IV: Trinkwasser WaBoLu-Heft 02/01, Umweltbundesamt, Berlin und Dessau-Roßlau, ISSN 0175 4211. Bohannon J (2007) Hard data on hard drugs, grabbed from the environment. Science 316: 42--44. Bonilla E (1984) Chronic manganese intake induces changes in the motor activity of rats. Experimental Neurology 84: 696--700. Bernhard G and Geipel G (2007) Bestimmung der Bindungsform des Urans in Mineralwa¨ssern (Determination of the binding form of uranium in mineral waters). Vom Wasser 105: 7--10 (in German). Bolt H (2008) The concept of ‘‘Practical Thresholds’’ in the derivation of occupational exposure limits for carcinogens by the Scientific Committee on Occupational Exposure Limits (SCOEL) of the European Union. Genes and Environment 30: 114--119. Bonacquisti TA (2006) A drinking water utility’s perspective on bromide, bromite, and ozonation. Toxicology 221: 145--148. Brewer GJ (2009) The risks of copper toxicity contributing to cognitive decline in the aging population and to Alzheimer’s disease. Journal of the American College of Nutrition 28: 238--242. Bull RJ, Krasner SW, Daniel PA, and Bull RD (2001) Health Effects and Occurrence of Disinfection By-Products, 397 pp. Denver, CO: AWWA Research Foundation. Bull et al. (in press) Regulatory Toxicology and Pharmacology.

412

Drinking Water Toxicology in Its Regulatory Framework

Burton NC and Guilarte TR (2009) Manganese neurotoxicity: Lessons learned from longitudinal studies in nonhuman primates. Environmental Health Perspectives 117: 325--332. Cairncross S and Valdmanis V (2010) Water supply, sanitation and hygiene protection. http://www.ncbi.nlm.nih.gov/bookshelf/br.fcgi?book ¼ dcp2&part ¼ A5898 (accessed April 2010). Calderon RL, Hudgens E, Le XC, Schreinemacher D, and Thomas DJ (1999) Excretion of arsenic in urine as a function of exposure to arsenic in drinking water. Environmental Health Perspectives 107: 663--667. Cassee FR, Groten JP, van Bladeren PJ, and Feron VJ (1998) Toxicological evaluation and risk assessment of mixtures. Critical Reviews in Toxicology 28: 73--101. Chipman JK, Parsons JL, and Beddowes EJ (2006) The multiple influences of glutathione on bromate genotoxicity: Implications for the dose–response relationship. Toxicology 221: 187--189. Chiu HF, Tsai SS, and Yang CY (2007) Nitrate in drinking water and risk of death from bladder cancer: An ecological case-control study in Taiwan. Journal of Toxicology and Environmental Health – Part A 70: 1000--1004. Committee on Environmental Health (2009) Drinking water from private wells and risks to children. Pediatrics 123(6): 1599–605. http://www.pediatrics.org/cgi/content/ full/123/6/1599 (accessed April 2010). Committee on Toxicity Testing and Assessment of Environmental Agents (2007) Toxicity testing in the 21st Century: A Vision and a Strategy. Washington, DC: Committee on Toxicity Testing and Assessment of Environmental Agents. Cooper ER, Siewicki TC, and Phillips K (2008) Preliminary risk assessment database and risk ranking of pharmaceuticals in the environment. Science of the Total Environment 398: 26--33. Cotruvo J and Bartram J (eds.) (2009) Calcium and Magnesium in Drinking Water: Public Health Significance (ISBN 978 92 4 156355 0). Geneva: World Health Organization. Cotruvo AJ and Regelski M (1990) Overview of the current national primary drinking water regulations and regulation development process. In: Calabrese EJ, Gilbert CE, and Pastides H (eds.) Safe Drinking Water Act – Amendments, Regulations and Standards, ch. 3, pp. 17–28 (ISBN 0 87371 138 6). Chelsea, MI: Lewis. de Bie P (2007) Novel Insights in the Molecular Pathogenesis of Human Copper Homeostasis Disorders through Studies of Protein–Protein Interactions. Doctoral Thesis, University Medical Center, Utrecht, ISBN 978907867520 4, Ch. 4. Dekant W, Melching-Kollmuss S, and Kalberlah F (2010) Toxicity assessment strategies, data requirements, and risk assessment approaches to derive health based guidance values for non-relevant metabolites of plant protection products. Regulatory Toxicology and Pharmacology 56: 135--142. Delker D, Hatch G, Allen J, et al. (2006) Molecular biomarkers of oxidative stress associated with bromate carcinogenicity. Toxicology 221(2–3): 158--165. De Roos AJ, Ward MH, Lynch CF, and Cantor KP (2003) Nitrate in public water supplies and the risk of colon and rectum cancers. Epidemiology 14: 640--649. Deutsche Forschungsgemeinschaft (DFG) (2007) MAK- und BAT-Werte-Liste 2007. Weinheim: Wiley-VCH. Deveau M (2010) Contribution of drinking water to dietary requirements of essential metals. Journal of Toxicology and Environmental Health A 73: 235--241. Dieter HH (1991) Vorkommen und Toxikologie von Arsen und seine gesundheitliche Bedeutung im Trinkwasser. In: Die Trinkwasserverordnung, 3. Auflage, pp. 154– 171. Berlin: Erich Schmidt. Dieter HH (1998) Proactive limit values for responsible management of chemicals. Part I: Environmental hygiene and management of chemicals. ESPR – Environmental Science and Pollution Research 5: 51--54. Dieter HH (2003) Gesundheitliche Bewertung von Bromat, als Nebenprodukt der Aufbereitung mit Ozon (Health-based evaluation of bromate, as a side product of drinking water ozonation). In: Grohmann A, Ha¨sselbarth U, and Schwerdtfeger WK (eds.) Die Trinkwasserverordnung – Einfu¨hrung und Erla¨uterungen fu¨r Wasserversorgungsunternehmen und U¨berwachungsbeho¨rden 4, neu bearbeitete Auflage, pp. 613--623. Berlin: Erich Schmidt. Dieter HH (2007) Humantoxikologische Bewertung Perfluorierter Tenside (PFT) am Beispiel der Perfluoroctansa¨ure (PFOA) und der Perfluoroctansulfonsa¨ure (PFOS) (Human toxicological evaluation of perfluorinated tensides (PFT) with perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) as examples). Umweltmed Forsch Prax 12: 95--104 plus erratum in 12(3): 127. Dieter HH (2010) The relevance of ‘‘non-relevant metabolites’’ from plant protection products (PPPs) for drinking water: The German view. Regulatory Toxicology and Pharmacology 56: 121--125. Dieter HH and Konietzka R (1995) Which multiple of a safe body dose derived on the basis of default factors would probably be unsafe? Regulatory Toxicology and Pharmacology 22: 262--267. Dieter HH, Schimmelpfennig W, Meyer E, and Tabert M (1999) Early childhood cirrhoses (ECC) in Germany between 1982 and 1994 with special consideration of copper etiology. European Journal of Medical Research 4: 233--242.

Dion-Fortier A, Rodriguez MJ, Se´rodes J, and Proulx F (2009) Impact of water stagnation in residential cold and hot water plumbing on concentrations of trihalomethanes and haloacetic acids. Water Research 43: 3057--3066. Donner E, Eriksson E, Revitt DM, Scholes L, Lu¨tzhøft H-CH, and Ledin A (2010) Presence and fate of priority substances in domestic greywater treatment and reuse systems. Science of the Total Environment 408: 2444--2451. Doria MDf, Pidgeon N, and Hunter PR (2009) Perceptions of drinking water quality and risk and its effect on behaviour: A cross-national study. Science of the Total Environment 407: 5455--5464. Dorman DC, Struve MF, Vitarella D, Byerly FL, Goetz J, and Miller R (2000) Neurotoxicity of manganese chloride in neonatal and adult CD rats following subchronic (21-day) high-dose oral exposure. Journal of Applied Toxicology 20: 179--187. DWI (Drinking Water Inspectorate) (2007) Guidance on the Water Supply (Water Quality) Regulations 2000/01 specific to PFOS (perfluorooctane sulphonate) and PFOA (perfluoroctanoic acid) concentrations in drinking water. http://www.dwi.gov. uk/stakeholders/guidance-and-codes-of-practice/PFOS%20&%20PFOA%20in%20drinking%20water.pdf (accessed September 2010). EFSA (European Food Safety Authority) (2008) Nitrates in vegetables. EFSA Journal 689: 1–79. Ekins S (ed.) (2007) Computational Toxicology: Risk Assessment for Pharmaceutical and Environmental Chemicals. Hoboken, NJ: Wiley. Emmett EA, Shofer FS, Zhang H, Freeman D, Desai C, and Shaw LM (2006) Community exposure to perfluorooctanoate: Relationships between serum concentrations and exposure sources. Journal of Occupational and Environmental Medicine 48: 759--770. EU Drinking Water Directive 98/83/EC (1998) Council Directive 98/832/EC of November 1998 on the quality of water intended for human consumption. Official Journal of the European Communities 330: 32. Fawell J and Nieuwenhuijsen MJ (2003) Contaminants in drinking water: Environmental pollution and health. British Medical Bulletin 68: 199--208. Fawell JK (2008) Health risks of micropollutants – the need for a new approach. Water Science and Technology 57(2): 183--187. Federal Environment Agency/Umweltbundesamt (2005) Announcement on uranium and biomonitoring. http://www.umweltdaten.de/gesundheit-e/monitor/Uranium-inurine.pdf (accessed April 2010). Federal Environment Agency of Germany (2010) Action values for substances in drinking water applicable during periods of limit value exceedance pursuant to Article 9 paras. 6–8 of the Drinking Water Ordinance of 2001.pdf and commentary http://www.umweltdaten.de/wasser-e/action_values-recommendation.pdf (accessed April 2010). Federal Office for Radiation Protection/Bundesamt fu¨r Strahlenschutz (2009) Radiation exposure due to natural radionuclides in drinking water.pdf and commentary http:// www.umweltdaten.de/wasser-e/action_values-commentary-recommendation.pdf (accessed April 2010). Focazio M, Kolpin DW, Barnes KK, et al. (2008) A national reconnaissance for pharmaceuticals and other organic wastewater contaminants in the United States – II. Untreated drinking water sources. Science of the Total Environment, 402: 201--216. Ford T, Rupp G, Butterfield P, and Camper A (2005) Health protecting public health in small water systems. Report of an International Colloquium. Bozeman, Montana, USA, 9–12 May 2004. http://watercenter.montana.edu/pdfs/colloquium_report_ final.pdf (accessed April 2010). Fromme H, Tittlemier SA, Volkel W, Wilhelm M, and Twardella D (2008) Perfluorinated compounds: Exposure assessment for the general population in western countries. International Journal of Hygiene and Environmental Health 212: 239--270. Garbrecht G (1986) Die Wasserversorgung antiker Sta¨dte (The water supply of ancient vities) In:1901 WaBoLu 1986–85 Years of Protecting the Environment, pp. 25 and 45. Berlin: WaBoLu-Hefte 1/1986. Gatseva PD and Argirova MD (2008) High-nitrate levels in drinking water may be a risk factor for thyroid dysfunction in children and pregnant women living in rural Bulgarian areas. International Journal of Hygiene and Environmental Health 211: 555--559. Gessner BD, Beller M, Middaugh JP, and Whitford GM (1994) Acute fluoride poisoning from a public water system. New England Journal of Medicine 330: 95--99. Gilman AP, Villeneuve DC, Secours VE, et al. (1998b) Uranyl nitrate: 91-day studies in the New Zealand white rabbit. Toxicological Sciences 41: 129--137. Gongqiao Z, Dehua L, and Peng H (1995) Effects of manganese on learning abilities in school children. Chinese Journal of Preventive Veterinary Medicine 29: 156--158. Gouget CB, Gallien JP, Avoscan L, Gobin R, Verbavatz JM, and Khodja H (2005) Cellular distribution of uranium after acute exposure of renal epithelial cells: SEM,

Drinking Water Toxicology in Its Regulatory Framework

TEM and nuclear microscopy analysis. Nuclear Instruments and Methods in Physics Research B 231: 268. Greim H and Snyder R (eds.) (2008) Toxicology and Risk Assessment: A Comprehensive Introduction (ISBN 978 0 470 86893 5). Wiley. Haberer K (1994) Der Einsatz von Desinfektionsmitteln und das Auftreten von Desinfektionsnebenprodukten in deutschen Wasserwerken (The application of disinfectants and the occurrence of DBPs in German waterworks). GWF Wasser/ Abwasser 135: 409--417. Hardisson A, Rodrı´guez MI, Burgos A, Flores LD, Guttie´rrez R, and Wa´rela H (2001) Fluoride levels in publicly supplied and bottled drinking water in the Island of Tenerife, Spain. Bulletin of Environmental Contamination and Toxicology 67: 163--170. Hamilton DJ, Ambrus A, Dieterle RM, et al. (2003) Regulatory limits for pesticides in water. Pure and Applied Chemistry 75: 1123--1155. Han YY, Weissfeld JL, Davis DL, and Talbott EO (2009) Arsenic levels in ground water and cancer incidence in Idaho: An ecologic study. International Archives of Occupational and Environmental Health 82: 843--849. Hayes CR and Skubala ND (2009) Is there still a problem with lead in drinking water in the European Union? Journal of Water and Health 7(4): 569--580. Haywood S, Muller T, Mackenzie AM, et al. (2004) Copper-induced hepatotoxicosis with hepatic stellate cell activation and severe fibrosis in North Ronaldsay lambs: A model for non-Wilsonian hepatic copper toxicosis of infants. Journal of Comparative Pathology 130(4): 266--277. Heck JE, Nieves JW, Chen Y, et al. (2009) Dietary intake of methionine, cysteine, and urinary arsenic excretion in Bangladesh. Environmental Health Perspectives 117: 99--104. Hens L (2007) Environmental impacts on congenital anomalies – information for the non-expert professional. In: Nicolopoulou-Stamati P, Hens L, and Howard CV (eds.) Congenital Disease and the Environment, pp. 409–450 (ISBN-13 978 1 4020 4830 2). Springer. Ho¨lzer J, Midasch O, Rauchfuss K, et al. (2008) Biomonitoring of perfluorinated compounds in children and adults exposed to perfluorooctanoate-contaminated drinking water. Environmental Health Perspectives 116: 651--657. Hopp V (2004) WASSER – Krise?, chs. 2 and 3. Wiley-VCH. Hrudey StE (2009) Chlorination disinfection by-products, public health risks and me. Water Research 43: 2057--2092. Hughes MF (2006) Biomarkers of exposure: A case study with inorganic arsenic. Environmental Health Perspectives 114: 1790--1796. Human Toxicology 29(2010): 15–29. ICRP (1996), Age-Dependent Doses to Members of the Public from Intake of Radionuclides: Part 5: Compilation of Ingestion and Inhalation Dose Coefficients, Publication 72. Elsevier Health Sciences. IESK (Institut fu¨r empirische Sozial- und Kommunikationsforschung) (2008) Qualita¨t und Image von Trinkwasser in Deutschland: Ergebnisse einer online-Umfrage in Deutschland mit u¨ber 10.000 Befragten (Quality and image of drinking water in Germany. Results of an online inquiry with over 10.000 participants) in German. http://www.iesk.de/download/tw070508-2_shortversion.pdf (accessed April 2010). ILSI (International Life Sciences Institute) (2005) Threshold of toxicological Concern (TTC): A tool for assessing substances of unknown toxicity present at low levels in the diet. Jekel M and Gruenheid S (2008) Indirect water reuse for human consumption in Germany: The case of Berlin. In: Jime´nez B and Asano T (eds.) Water Reuse: An International Survey of Current Practice, Issues and Needs, ch. 22 (ISBN 13: 978 18433 9089 3). London: IWA Publishing. Jiang X, Castelao JE, Groshen S, et al. (2008) Water intake and bladder cancer risk in Los Angeles County. International Journal of Cancer 123: 1649--1656. Jime´nez B and Asano T (eds.) (2008) Water Reuse: An International Survey of Current Practice, Issues and Needs (ISBN 13: 978 18433 9089 3). London: IWA Publishing. Kalberlah F and Hassauer M (2003) Comparison of assessment methods to derive health based thresholds (Benchmark – NOAEL). Report UBA-FKZ 201 65 201/01 by order of the German Federal Environment Agency (UBA), November 15, 2003, http://www.apug.de/archiv/pdf/benchmark_endbericht.pdf (accessed September 2010). Kalberlah F, Schneider K, and Schuhmacher-Wolz U (2003) Uncertainty in toxicological risk assessment for non-carcinogenic health effects. Regulatory Toxicology and Pharmcology 37: 92--104. Kanke-Fosso E, Jagals P, and Du Preez H (2008) Exposure of rural households to toxic cyanobacteria in container-stored water. Water SA 34: 631--636. Keith JD, Pacey GE, Cotruvo JA, and Gordon G (2006) Experimental results from the reaction of bromate ion with synthetic and real gastric juices. Toxicology 221: 225--228.

413

Kessler H, Pajonk FG, Bach D, et al. (2009) Effect of copper intake on CSF parameters in patients with mild Alzheimer’s disease: A pilot phase 2 clinical trial. Journal of Neural Transmission 115: 1651--1659. Klo¨pffer W and Wagner BO (2007) Persistence revisited (editorial). ESPR – Environmental Science and Pollution Research 14: 141–142. Kluge Th and Schramm E (1986) Wasserno¨te – Umwelt- und Sozialgeschichte des Trinkwassers (Water Miseries – Environmental and Social History of Drinking Water), S. 62, p. 79/80, 103 (ISBN 3-924007-19-5). Aachen: Alano-Verlag (in German). Kolkmann J (1990) Die EG-Trinkwasserrichtline: Die Nitrat- und Pestizidgrenzwerte und ihre Umsetzung im deutschen Umweltrecht (The EU Drinking Water Guideline and Its Implementation in German Environmental Jurisdiction) (ISBN 3 503 03175 8). Berlin: Erich Schmidt Verlag (in German). Ko¨lle W, Werner P, Strebel O, and Bo¨ttcher J (1983) Denitrifikation in einem reduzierenden Grundwasserleiter. Vom Wasser 61: 125--147. Ko¨lle W, Werner P, Strebel O, and Bo¨ttcher J (1987) Reduced sulphur compounds in sandy aquifers and their interactions with groundwater. Groundwater Monitoring and Management – Proceedings of the Dresden Symposium (March 198). IAHS Publ. no. 173, 1990. Konietzka R, Dieter HH, and Schneider K (2008) Extrapolation beats uncertainty. Journal of Toxicology and Environmental Health – Part B 11: 609--611. Konietzka R, Dieter HH, and Voss JU (2005) Vorschlag fu¨r einen gesundheitlichen Leitwert fu¨r Uran im Trinkwasser (Proposal on a health-based guide value for uranium in drinking water). Umweltmed Forsch Prax 10: 133--143. Ko¨niglich-Wu¨rttembergisches Ministerium des Innern (1878) Amtsblatt Nr. 7 vom 29, April 1878, pp. 101–107, Stuttgart. Kortenkamp A, Backhaus T, and Faust M (2009) State of the Art Report on Mixture Toxicity: Findings of a Project on Mixture Toxicology and Ecotoxicology, Commissioned by the European Commission, DG Environment. http:// ec.europa.eu/environment/chemicals/pdf/report_Mixture%20toxicity.pdf (accessed September 2010). Krishnamachari KA (1986) Skeletal fluorosis in humans: A review of recent progress in the understanding of the disease. Progress in Food and Nutrition Science 10: 279--314. Kurttio P, Auvinen A, Salonen L, et al. (2002) Renal effects of uranium in drinking water. Environmental Health Perspectives 110: 337--342. Kurttio P, Harmoinen A, Saha H, et al. (2006) Kidney toxicity of ingested uranium from drinking water. American Journal of Kidney Diseases 47: 972--982. Kurttio P, Komulainen H, Leino A, Salonen L, Auvinen A, and Saha H (2005) Bone as a possible target of chemical toxicity of natural uranium in drinking water. Environmental Health Perspectives 113: 68--72. Lai JCK, Leung TKC, and Lim L (1982) Activities of the mitochondrial NAD-linked isocitric dehydrogenase in different regions of the rat brain. Changes in ageing and the effect of chronic manganese chloride administration. Gerontology 28: 81--85. Larson CD (1990) Historical development of the national primary drinking water regulations. In: Calabrese EJ, Gilbert CE, and Pastides H (eds.) Safe Drinking Water Act – Amendments, Regulations and Standards, ch. 2 (ISBN 0 87371 138 6). Lewis. Lau KH, Farley JR, Freeman TK, and Baylink DJ (1989) A proposed mechanism of the mitogenic action of fluoride on bone cells: Inhibition of the activity of an osteoblastic acid phosphatase. Metabolism 38: 858--868. Leeuwen CJ and Vermeire TG (eds.) (2007) Risk Assessment of Chemicals: An Introduction (ISBN 978 1 4020 6101 1). Springer. Legay C, Rodriguez MJ, Se´rodes JB, and Levallois P (2010) Estimation of chlorination by-products in drinking water in epidemiological studies on adverse reproductive outcomes: A review. Science of the Total Environment 408: 456--472. Leung TKC, Lai JKC, and Lim L (1981) The regional distribution of monoamine oxidase activities towards different substrates: Effects in rat brain of chronic administration of manganese chloride and of ageing. Journal of Neurochemistry 36: 2037--2043. Ljung KS, Kippler MJ, Goessler W, Grander GM, Nermell BM, and Vahter ME (2009) Maternal and early life exposure to manganese in rural Bangladesh. Environmental Science and Technology 43: 2595--2601. Lu¨kewille A and Heuwinkel B (1990) Grundwasserversauerung in der Senne am Teutoburger Wald – Trinkwasserschutz durch Immissionsagrenzwerte? In: Kommission Reinhaltung der Luft im VDI und DIN (Hrsg.) Wirkung von Luftverunreinigungen auf Bo¨den, Teil I. VDI-Berichte 837, pp. 761--781. Du¨sseldorf: VDI-Verlag. Maffei F, Carbone F, Forti CG, et al. (2009) Drinking water quality: An in vitro approach for the assessment of cytotoxic and genotoxic load on water sampled along distribution system. Environmental International 35(7): 1053--1061 (doi:10.1016/ j.envint.2009.05.007).

414

Drinking Water Toxicology in Its Regulatory Framework

Magdo HS, Forman J, Graber N, et al. (2007) Grand rounds: Nephrotoxicity in a young child exposed to uranium from contaminated well water. Environmental Health Perspectives 115: 1237--1241. Marcus WL and Rispin AS (1988) Threshold carcinogenicity using arsenic as an example. In: Cothern CR, Mehlman MA, and Marcus WL (eds.) Advances in Modern Environmental Toxicology, vol. XV, pp. 133--158. Princeton, NJ: Princeton Scientific Publishers. Mazumder DNG, Chakraborty AK, Ghose A, et al. (1988) Chronic arsenic toxicity from drinking tubewell water in rural West Bengal. Bulletin of the World Health Organization 66/4: 499--506. MDH (2009) Hazardous substances in Minnesota: Perfluorochemicals and health. http://www.health.state.mn.us/divs/eh/hazardous/topics/pfcshealth.html Meharg AA (2005) Venomous Earth. How Arsenic Caused the World’s Worst Mass Poisoning. (ISBN-13 978 1 4039 4499 3). Macmillan. Melching-Kollmuss S, Dekant W, and Kalberlah F (2010) Application of the ‘‘threshold of toxicological concern’’ to derive tolerable concentrations of ‘‘non-relevant metabolites’’ formed from plant protection products in ground and drinking water. Regulatory Toxicology and Pharmacology 56: 126--134. Mirvish SS, Grandjean AC, Reimers KJ, et al. (1995) Dosing time with ascorbic acid and nitrate, gum and tobacco chewing, and other factors affecting N-nitrosoproline formation in healthy subjects taking proline with a standard meal. Cancer Epidemiology, Biomarkers and Prevention 4: 775--782. Mu¨ller T, Feichtinger H, Berger H, and Mu¨ller W (1996) Endemic Tyrolean infantile cirrhosis: An ecogenetic disorder. Lancet 347: 887–880 Mu¨ller T, van de Sluis B, Zhernakova A, et al. (2003) The canine toxicosis gene MURR1 does not cause non-Wilsonian hepatic copper toxicosis. Journal of Hepatology 38: 164--168. Murrell JA, Portier CJ, and Morris RW (1998) Characterizing dose–response I: Critical assessment of the benchmark dose concept. Risk Analysis 18: 13--26. Neumann H-G, van Dorp C, and Zwirner-Baier I (1995) Implications for risk assessment of measuring the relative contribution to exposure from occupation, environment and lifestyle: Hemoglobin adducts from amino- and nitro-arenes. Toxicology Letters 82/83: 771--777. NHMRC (2007) A systematic review of the efficacy and safety of fluoridation. http:// www.nhmrc.gov.au/publications/synopses/eh14syn.htm (accessed April 2010). Nicolopoulou-Stamati P, Hens L, and Howard CV (eds.) (2007) Congenital Disease and the Environment, p. 412 (ISBN-13 978 1 4020 4830 2). Springer. Parag Y and Timmons-Roberts J (2009) A battle against bottles: Building, claiming, and regaining tap-water trustworthiness. Society and Natural Resources 22: 625--636. Petrie KP and Wessely S (2004) Getting well from water: Bottled water exploits our worries about what affects health in the modern world (editorial). British Medical Journal 329: 1417–1418. Prat O, Vercouter T, Ansoborlo E, et al. (2010) Uranium speciation in drinking water from drilled wells in southern Finland and its potential links to health effects. Environmental Science and Technology 43: 3941--3946. Post GB, Louis JB, Cooper KR, Boros-Russo BJ, and Lippincott RL (2009) Occurrence and potential significance of perfluorooctanoic acid (PFOA) detected in New Jersey public drinking water systems. Environmental Science and Technology 43: 4547--4554. Reemtsma T and Jekel M (eds.) (2002) Organic Pollutants in the Water Cycle: Properties, Occurrence, Analysis and Environmental Relevance of Polar Compounds (ISBN 978 3 527 31297 9). Wiley-VCH. Ritter L, Totman C, Krishnan K, Carrier R, Vezina A, and Morisset V (2007) Deriving uncertainty factors for threshold chemical contaminants in drinking water. Journal of Toxicology and Environmental Health Part B: Critical Reviews 10: 527--557. Roels HA, Ortega-Eslava MI, Ceulemans E, Robert A, and Lison D (1999) Prospective study on the reversibility of neurobehavioral effects in workers exposed to manganese dioxide. Neurotoxicology 20: 255--271. Rook JJ (1974) Formation of haloformes during chlorination of natural waters. Water Treatment and Examination 23: 234--243. Rudel R, Slayton TM, and Beck BD (1996) Implications of Arsenic genotoxicity for dose response of carcinogenic effects. Regulatory Toxicology and Pharmacology 23: 87--105. Rundblad G (2008) The semantics and pragmatics of water notices and the impact on public health. Journal of Water and Health 6(supplement 1): 77--86. Rylander R (2008) Drinking water constituents and disease. Journal of Nutrition 138: 423S--425S. Sabbioni G, Wie J, and Liu Y-Y (1996) Determination of hemoglobin adducts in workers exposed to 2,4,6-trinitrotoluene. Journal of Chromatography B 682: 243--248. Sadeq M, Moe CL, Attarassi B, Cherkaoui I, ElAouad R, and Idrissi L (2008) Drinking water nitrate and prevalence of methemoglobinemia among infants and children

aged 1–7 in Moroccan areas. International Journal of Hygiene and Environmental Health 211: 546--554. Schets FM, Italiaander R, van den Berg HJL, and de Roda Husman AM (2010) Rainwater harvesting: Quality assessment and utilization in the Netherlands. Journal of Water and Health 8(2): 224--235. Schmidt CK and Brauch HJ (2008) N,N-Dimethylsulfamide as precursor for Nnitrosodimethylamine (NDMA) formation upon ozonation and its fate during drinking water treatment. Environmental Science and Technology 42: 6340--6346. Schmoll O, Howard G, Chilton J, and Chorus I (eds.) (2006) Protecting Groundwater for Health: Managing the Quality of Drinking-Water Sources (ISBN 92 4 154668 9). IWA Publishing/World Health Organization. Schneider K (1999) Unterschiede in der Empfindlichkeit von Kindern gegenu¨ber krebserzeugenden Stoffen im Vergleich zu Erwachsenen (Differences in the sensitivity of children and adults to carcinogenic substances – literature study). Umweltmedizin in Forschung und Praxis 4: 155--162. Schriks M, Heringa M, Kooi MME, de Voogt P, and Wezel AP (2010) Toxicological relevance of emerging contaminants for drinking water quality. Water Research 44: 461--476. Schuhmacher-Wolz U, Dieter HH, Klein D, and Schneider K (2009) Oral exposure to inorganic arsenic: Evaluation of its carcinogenic and non-carcinogenic effects. Critical Reviews in Toxicology 39: 271--298. Schwarzenbach RP, Escher BE, Fenner K, et al. (2006) The challenge of micropollutants in aquatic systems. Science 313: 1072--1076. Skutlarek D, Exner M, and Farber H (2006) Perfluorinated surfactants in surface and drinking waters. Environmental Science and Pollution Research International 13: 299--307. Smith AH, Lingas EO, and Rahman M (2000) Contamination of drinking-water by arsenic in Bangladesh: A public health emergency. Bulletin of the WHO 78: 1093--1103. Speijers GJA and van den Brandt PA (2010) Nitrate (and potential Endogenous Formation of N-Nitroso Compounds), WHO Food Additives Series 50. http:// www.inchem.org/documents/jecfa/jecmono/v50je06.htm (accessed April 2010). State of Alaska (2010) Historic Information about Drinking Water. http:// www.dec.state.ak.us/eh/dw/publications/historic.html (accessed April 2010). Steinha¨user KG and Richter S (2002) Assessment and management of chemicals – how should persistent polar pollutants be regulated? In: Reemtsma T and Jekel M (eds.) Organic Pollutants in the Water Cycle: Properties, Occurrence, Analysis and Environmental Relevance of Polar Compounds, ch. 12 (ISBN 978 3 527 31297 9). Wiley-VCH. Svensson BG, Bjo¨rnham A, Schu¨tz A, Lettevall U, Nilsson A, and Skerfving S (1987) Acidic deposition and human exposure to toxic metals. Science of the Total Environment 67: 101--115. Tamer MN, Ko¨roglu BK, Arslan C, et al. (2007) Osteosclerosis due to endemic fluorosis. Science of the Total Environment 373: 43--48. Tanner MS (1998) The role of copper in Indian childhood cirrhosis. American Journal of Clinical Nutrition 76(supplement 5): 1074--1081. Tardiff RG, Carson ML, Sweeney LM, et al. (2009) Derivation of a drinking water equivalent level (DWEL) related to the maximum contaminant level goal for perfluorooctanoic acid (PFOA), a persistent water soluble compound. Food and Chemical Toxicology 47: 2557--2589. Ternes T (2007) The occurrence of micropollutants in the aquatic environment. Water Science and Technology 55: 327--332. Ternes TA and Joss A (eds.) (2006) Human Pharmaceuticals, Hormones and Fragrances: The Challenge in Urban Water Management. London: IWA Publishing. Thier R, Lewalter R, Selinski S, and Bolt HM (2001) Biological monitoring in workers in a nitrobenzene reduction plant: Haemoglobin versus albumin adducts. International Archives of Occupational and Environmental Health 74: 483--488. TWK (2006) Vorla¨ufige Bewertung von Perfluorierten Tensiden (PFT) im Trinkwasser am Beispiel ihrer Leitsubstanzen Perfluoroctansa¨ure (PFOA) und Perfluoroctansulfonsa¨ure (PFOS). http://www.umweltbundesamt.de/wasser/ themen/downloads/trinkwasser/pft-im-trinkwasser.pdf (accessed April 2010). Uauy R, Maass A, and Araya M (2008) Human populations at risk of copper excess: Estimating risk from copper excess in human populations. American Journal of Clinical Nutrition 88: 867S--871S. UBA (2003) Evaluation from the point of view of health of the presence in drinking water of substances that are not (yet) possible or only partially possible to evaluate. Recommendation of the Federal Environment Agency of Germany (UBA) after Hearing the Drinking Water Commission of the Federal Ministry of Health UBA (2004) Beurteilung der Trinkwasserqualita¨t hinsichtlich der Parameter Blei, Kupfer und Nickel (Assessment of drinking water quality regarding the parameters lead, copper and nickel). Empfehlung des Umweltbundesamtes nach Anho¨rung der Trinkwasserkommission des Bundesministeriums fu¨r Gesundheit. Bundesgesundheitsbl – Gesundheitsforsch – Gesundheitsschutz 47: 296--300.

Drinking Water Toxicology in Its Regulatory Framework

UBA (2006) Bewertung monocyclischer Nitroverbindungen und ihrer Abbauprodukte im Trinkwasser. Empfehlung des Umweltbundesamtes nach Anho¨rung der Trinkwasserkommission des Bundesministeriums fu¨r Gesundheit. Bundesgesundheitsbl-Gesundheitsforsch-Gesundheitsschutz 49: 701--703. UBA (2008) Hygiene-related assessment of ‘‘non-relevant’’ metabolites (as defined in chemicals law) of active substances of plant protection products in drinking water1. Recommendation of the Federal Environment Agency of Germany (UBA) after Hearing the Drinking Water Commission of the Federal Ministry of Health Umweltbundesamt (2009a) Grenzwerte, Leitwerte, Orientierungswerte, Maßnahmenwerte – Definitionen und Festlegungen mit Beispielen aus dem Umweltbundesamt (Limit values, guide values, indication values, action values – definitions and regulations with examples from UBA). Bundesgesundhbl 52: 1202--1206. Umweltbundesamt (2009b) 2. Addendum zur Soffmonographie Blei – Referenz- und ‘‘Human-Biomonitoring ‘‘Werte der Kommission ‘‘Human-Biomonitoring’’ (2nd Addendum to ‘‘Substance Monograph: Lead – Reference and Human Biomonitoring Values’’ of the Human Biomonitoring Commission of the Federal Environment Agency of Germany). Stellungnahme der Kommission ‘‘HumanBiomonitoring’’ des Umweltbundesamtes. Bundesgesundheitsbl Gesundheitsforsch – Gesundheitsschutz 52: 983--986. UN – Economic Commission for Europe (1999) Protocol on Water and Health to the 1992 Convention on the Protection and Use of Transboundary Watercourses and International Lakes, article 4, paragraphs 2 b and 2 c. London, 17 June 1999. http://www.unece.org/env/water (accessed April 2010). UNEP (2003a) Water for People – Water for Life: The 1st UN World Water Development Report, Part II/ch. 4, Table 4.2. http://www.unesco.org/water/wwap/wwdr/wwdr1/ pdf/chap4.pdf) (accessed September 2010). UNEP (2003b) Water for People – Water for Life: The 1st UN World Water Development Report, Part III/ch. 8, Table 8.3. http://www.unesco.org/water/wwap/ wwdr/wwdr1/pdf/chap8.pdf (accessed September 2010). US-EPA (2005) Guidelines for Carcinogen Risk Assessment. Risk Assessment Forum. Washington, DC: US Environmental Protection Agency. US-EPA (United States Environmental Protection Agency, Office of Water) (2009) Provisional Health Advisories for Perfluorooctanoic Acid (PFOA) and Perfluorooctane Sulfonate (PFOS). http://www.epa.gov/waterscience/criteria/ drinking/pha-PFOA_PFOS.pdf. (accessed September 2010). Vermeer ITM, Pachen DMFA, Dallinga JW, Kleinjans JCS, and van Maanen JMS (1998) Volatile N-nitrosamines formation after intake of nitrate at the ADI level in combination with an amine-rich diet. Environmental Health Perspectives 106: 459--463. von Gunten U (2003) Review ozonation of drinking water: Part I. Oxidation kinetics and product formation. Water Research 37: 1443--1467. Ward MH, de Kok TM, Levallois P, et al. (2005) Workgroup report: Drinking-water nitrate and health – recent findings and research needs. Environmental Health Perspectives 113: 1607--1614. Ward LA, Cain OL, Mullally RA, et al. (2009) Health beliefs about bottled water: A qualitative study. BMC Public Health 9: 196. Webb S, Ternes T, Gibert M, and Olejniczak K (2003) Indirect human exposure to pharmaceuticals via drinking water. Toxicology Letters 142: 157--167. WHO (1984a) Guidelines for Drinking-Water Quality, Vol. 1: Recommendations, Preface, p. x. Geneva: WHO. WHO (1984b) Guidelines for Drinking-Water Quality, Vol. 1: Recommendations, Introduction, p. 2, tier 1.3(d). Geneva: WHO. WHO (1987) Principles for the Safety Assessment of Food Additives and Contaminants in Food. Environmental Health Criteria, 70 p. 87. Geneva: WHO. WHO (1995) Concern for Europe’s Tomorrow. WHO European Center for Environmental Health, Copenhagen 1995. http://www.euro.who.int/document/wa3095co.pdf. (accessed September 2010). WHO (1997) Guidelines for Drinking-Water Quality, 2nd edn. http://www.who.int/ water_sanitation_health/dwq/gdwq2v1/en/index2.html WHO (1998) Environmental Health Criteria 200: Copper. Geneva: WHO. WHO (2008) Guidelines for Drinking Water Quality, Vol. 1 – Recommendations, 3rd edn., p. 2, 91, 157. http://www.who.int/water_sanitation_health/dwq/fulltext.pdf (accessed April 2010). WHO (2008a) Background Document from 1998 on cyanobacterial toxins microcystin. http://www.who.int/water_sanitation_health/dwq/chemicals/cyanobactoxins.pdf (accessed April 2010). WHO/IPCS (1994) Assessing Human Health Risks of Chemicals: Derivation of Guidance Values for Health-Based Exposure Limits. Environmental Health Criteria 170. Geneva: WHO. Wijmenga C, Mu¨ller T, Murli IS, et al. (1998) Endemic Tyrolean infantile cirrhosis is not an allelic variant of Wilson’s disease. European Journal of Human Genetics 6: 624--628.

415

Wilhelm M, Bergmann S, and Dieter HH (2010b) Occurrence of perfluorinated compounds (PFCs) in drinking water of North Rhine-Westphalia, Germany and new approach to assess drinking water contamination by shorter-chained C4–C7 PFCs. International Journal of Hygiene and Environmental Health 213: 224--232. Wilhelm M, Heinzow B, Angerer J, and Schulz C (2010a) Reassessment of critical lead effects by the German Human Biomonitoring Commission results in suspension of the human biomonitoring values (HBM I and HBM II) for lead in blood of children and adults. International Journal of Hygiene and Environmental Health 213: 265--269. Wilhelm M, Kraft M, Rauchfuss K, and Ho¨lzer J (2008) Assessment and management of the first German case of a contamination with perfluorinated compounds (PFC) in the Region Sauerland, North Rhine–Westphalia. Journal of Toxicology and Environmental Health – Part A 71: 725--733. Winton E, Tardiff RG, and McCabe LJ (1971) Nitrate in drinking water. Journal of the American Water Works Association 63: 95--98. Wollin KM and Dieter HH (2005) Toxicological guidelines for monocyclic nitro-, amino- and aminonitroaromatics, nitramines, and nitrate esters in drinking water. Archives of Environmental Contamination and Toxicology 49: 18--26. Woolf A, Wright R, Amarasiriwardena C, and Bellinger D (2002) A child with chronic manganese exposure from drinking water. Environmental Health Perspectives 110: 1--4. World Health Organization/International Program on Chemical Safety (WHO/IPCS) (1986) Principles for Evaluating Health Risks from Chemicals during Infancy and Early Childhood: The Need for a Special Approach, Environmental Health Criteria 59. Geneva: WHO/IPCS. WQRA (Water Quality Research Australia) (2009) Report on Brisbane Fluoride Incident. Health Stream 55 – September 2009. http://www.wqra.com.au/hsarch/HS55a.htm (accessed April 2010). Yokel RA, Lasley SM, and Dorman DC (2006) The speciation of metals in mammals influences their toxicokinetics and toxicodynamics and therefore human health risk assessment. Journal of Toxicology and Environmental Health, Part B 9: 63--85. Zamora ML, Tracy BL, Zielinski JM, Meyerhof DP, and Moss MA (1998) Chronic ingestion of uranium in drinking water: A study of kidney bioeffects in humans. Toxicological Sciences 43: 68--77. Zamora ML, Zielinski JM, Meyerhof D, Moodie G, Falconer R, and Tracy B (2003) Uranium gastrointestinal absorption: The f1 Faktor in humans. Radiation Protection Dosimetry 105: 55--60. Zhang T, Sun HW, Wu Q, Zhang XZ, Yun SH and Kannan K (2010) Perfluorochemicals in meat, eggs and indoor dust in China: Assessment of sources and pathways of human exposure to perfluorochemicals. Environmental Science and Technology (doi: 10.1021/es1000159). Zhanga X, De Silvaa D, Suna B, et al. (2010) Cellular and molecular mechanisms of bromate-induced cytotoxicity in human and rat kidney cells. Toxicology 269: 13--23. Zietz B, Dieter HH, Lakomek M, Schneider H, Kessler-Gaedtke B, and Dunkelberg H (2003) Epidemiological investigation on chronic copper toxicity to children exposed via the public drinking water supply. Science of the Total Environment 302: 127--144. Zwiener C (2002) Trihalomethanes (THMs), haloacetic acids (HAAs), and emerging disinfection by-products in drinking water. In: Reemtsma T and Jekel M (eds.) Organic Pollutants in the Water Cycle: Properties, Occurrence, Analysis and Environmental Relevance of Polar Compounds, pp. 251–286 (ISBN 978 3 527 31297 9). Weinhiem: Wiley-VCH. Zwiener C, Richardson SD, De Marini DM, Grummt T, Glauner Th, and Frimmel FH (2007) Drowning in disinfection byproducts? Assessing swimming pool water. Environmental Science and Technology 41(2): 363--372.

Relevant Websites http://scripts.oieau.fr Aqua-Lingua. http://www.umweltdaten.de Definition and derivation of health-related indication values (HRIV) for contaminants of drinking with patchy or nonexistent database, recommendation of Germany’s Federal Environment Agency (UBA) from March 2003, and the recommendation’s rationale in form of a commentary to it. http://www.epa.gov Drinking Water Health Advisories, Water Quality Criteria, EPA.

3.15 Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter SK Sharma, SK Maeng, and S-N Nam, UNESCO-IHE Institute for Water Education, Delft, The Netherlands G Amy, King Abdullah University of Science and Technology, Thuwal, Saudi Arabia & 2011 Elsevier B.V. All rights reserved.

3.15.1 3.15.1.1 3.15.1.2 3.15.2 3.15.3 3.15.3.1 3.15.3.2 3.15.3.3 3.15.3.4 3.15.3.5 3.15.3.6 3.15.4 3.15.5 3.15.5.1 3.15.5.2 3.15.5.3 3.15.5.4 3.15.6 References

Introduction 417 Relevance of Organic Matter Characterization in Wastewater Treatment and Reuse 417 NOM and EfOM: Definition and Sources 417 Advantages of Bulk Water Characterization over NOM/EfOM Isolates 418 Bulk Water Analysis Protocols 418 Dissolved Organic Carbon 419 Dissolved Organic Nitrogen 419 Specific UV Absorbance 419 XAD-8/-4 Adsorption Chromatography 419 Fluorescence Excitation-Emission Matrix 419 Liquid Chromatography with Online Organic Carbon Detection 420 EfOM versus NOM Differences in Bulk Water Parameters 420 Application of Protocols to Case Studies 420 Soil Column Studies Simulating SAT Using Primary and Secondary Effluent 420 Water Reclamation Case Study (China) 421 Impact of Wastewater Treatment Plant Effluent on River Water Quality (USA) 423 Comparison of Removal of Bulk Organic Fractions from River Water and Wastewater Impacted River Water during Soil Passage 423 Summary 425 425

3.15.1 Introduction 3.15.1.1 Relevance of Organic Matter Characterization in Wastewater Treatment and Reuse Wastewater reclamation and reuse offers a great opportunity and promise as an alternative source of water to meet the everincreasing different water demands (municipal, industrial, agricultural, and environmental) of the growing world population. Furthermore, wastewater reclamation and reuse has now been accepted as a necessity and attractive option to reduce water scarcity in different parts of the world. This is evident from the rapid increase in water reclamation and reuse projects worldwide in the last decade. However, proper selection of wastewater treatment plant (WWTP) effluent treatment or polishing techniques to satisfy the requirements (for direct or indirect reuse and nonpotable or potable reuse) of the relevant water quality guidelines and regulations as well as that of the users is a challenge and prerequisite for sustainability of water reuse projects. In this context, a better understanding of the different constituents in the WWTP effluent and their fate during different subsequent treatment processes are important to ensure proper planning, design, and operation of wastewater reclamation and reuse systems. The organic matter present in WWTP effluents, commonly known as effluent organic matter (EfOM), is mainly a mixture of (1) background natural organic matter (NOM) from the drinking water, (2) soluble microbial products (SMPs) added during biological wastewater treatment, and (3) trace levels of

effluent-derived organic micropollutants (OMPs) and disinfection by-products (DBPs). Figure 1 presents a schematic of the urban water cycle illustrating the different components of EfOM. Therefore, in the water cycle, the characteristics of the EfOM (the concentrations and proportions of each of these components in the effluent) depend on the source of drinking water, type of drinking water treatment, water use in the service area, and type of wastewater treatment. Removal of EfOM is one of the main concerns in the treatment of WWTP effluents for water reuse applications as it impacts the treatability of the water for the intended application. The refractory organic compounds remaining after advanced water treatment are of special concern. EfOM is a DBP precursor, exerts high coagulant and oxidant demands, and influences nitrification and denitrification processes as well as the removal of OMPs by biodegradation. Some components of EfOM, for example, protein-like organic matter, are also responsible for the fouling of membranes and adsorbents. Furthermore, some esthetic (color, taste, and odor) and operational problems (corrosion and regrowth) associated with NOM in drinking water also affect wastewater reuse applications. In general, EfOM affects essentially all chemicals and biological processes in aquatic environments (Shon et al., 2006). By systematic characterization, the problematic organic matter fractions can be targeted for removal and transformation. Insight into the behavior of different fractions or constituents of organic matter present in water sources will

417

418

Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter Natural organic matter (NOM)

Water treatment plant

Drinking water source

Biodegradable organic matter (BOM) Municipal (domestic) use

Soluble microbial products (SMPs) Wastewater effluent Wastewater treatment plant

Figure 1 Schematic of urban water cycle showing components of EfOM.

provide a better understanding of their fate, transport, and impact during water treatment and distribution. Therefore, proper characterization of the bulk organic matter present in WWTP effluents and after different treatment steps would be an important basis for the selection of the appropriate and cost-effective treatment processes and monitoring of the effectiveness of different treatment steps. A major constraint, however, is that much of our present knowledge on characterization protocols is based on drinking-water NOM, and there is a need to demonstrate their applicability to EfOM and to provide a basis for differentiating EfOM from NOM according to measured characterization protocols.

3.15.1.2 NOM and EfOM: Definition and Sources NOM is a complex heterogeneous matrix of organic compounds found in all natural waters. The type and amount of NOM in water depend on climatic conditions and hydrological regime as well as other environmental factors. NOM mainly consists of carbon, oxygen, and hydrogen. Depending on the source of NOM, nitrogen and sulfur can also be present. Furthermore, different cations and anions may be incorporated into NOM structure due to adsorption, complexation, and ion exchange. NOM consists of both humic (humic and fulvic acids) and nonhumic components. Humic/fulvic acids have molecular weight greater than 2000 Da, while that of fulvic acids ranges from 500 to 2000 Da. Fulvic acids represent the most watersoluble fraction of humic material. Humic molecules contain aromatic, carboxyl, carbonyl, methoxyl, and aliphatic units. In addition to humic substances, nonhumics such as hydrophilic acids, proteins, amino acids, polysaccharides, and other biopolymers also contribute to the NOM (Thurman, 1985; Owen et al., 1993; AWWARF, 2000; Drewes and Summers, 2002). Furthermore, organic matter in natural water sources may also contain many trace organic compounds contributed by human activities. Based on source or origin, NOM in water can generally be divided into three main types: 1. Allochthonous NOM. This type of NOM originates from the decay of terrestrial biomass or through soil leaching in the watershed, mainly from leaching of vegetative debris by runoff. It mainly consists of humic substances. The production and characteristics of this type of NOM are therefore related to vegetative patterns and to hydrologic and geological characteristics of the watershed.

2. Autochthonous NOM. This type of NOM originates from in situ sources, mainly algal organic matter (AOM), other phytoplankton, and macrophytes; components can be extracellular or intracellular organic matter consisting of macromolecules and cell fragments. The production of this type of NOM is therefore related to photosynthetic activity and decay products of algal matter. 3. EfOM. EfOM mainly consists of background drinking water NOM (dominated by humic substances) which is not removed during wastewater treatment plus SMPs formed during biological wastewater treatment. It also contains some OMPs introduced during domestic use and generated during water and wastewater treatment. The characteristics of EfOM therefore depend on the type of drinking water source and treatment as well as the type of wastewater treatment applied. NOM in lakes and reservoirs of moderate-to-high trophic status is often dominated by material generated in the water body (autochthonous material), whereas lower-order rivers and streams usually carry more NOM that is generated exterior to the water body (allochthonous NOM). Allochthonous NOM has a large C/N ratio (near 100:1), is highly colored, and has significant aromatic carbon content, whereas autochthonous NOM has relatively lower C/N ratios (near 10:1), is almost colorless, and has low aromatic carbon content (HDR Engineering Inc., 2001).

3.15.2 Advantages of Bulk Water Characterization over NOM/EfOM Isolates As organic matter present in water (NOM/EfOM) consists of thousands of distinct chemical species, its characterization based on analysis of individual compounds and their properties is not possible. The logical and well-accepted approach is to characterize NOM/EfOM by grouping constituents according to a set of fractions. Due to the diversity of the organic molecules in water and their relatively low concentrations compared to other solutes and ions in the inorganic matrix, methods are needed that can characterize NOM in dilute solutions containing a variety of other chemicals, or that can isolate NOM without altering their properties. Bulk water characterizations are more useful to water utilities and are relatively less analyst and time intensive. The results of bulk water characterization can be understood and

Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter

used by nonspecialists as well. Furthermore, they also provide opportunities for online monitoring and control of treatment processes. Isolation protocols (e.g., XAD-8 resin adsorption or reverse osmosis concentration), while producing purified and concentrated isolates, require processing of large volumes of water (e.g., hundreds of liters), do not provide 100% recovery of the organic matter, create possible artifacts, and are very analyst and time intensive (Gadmar et al., 2005). Some isolate protocols require pH adjustment which may also influence the NOM characteristics due to the possibility of degradation, decarboxylation, oxidation, and condensation reactions (Gaffney et al., 1996). Advanced instrumental techniques such as Fourier transform infrared spectroscopy (FTIR) and nuclear magnetic resonance (NMR) spectroscopy have been used to characterize NOM isolates/fractions. Even with these methods, it is often difficult to obtain detailed information. Infrared spectra yield very broadband with significant overlaps that often cannot be resolved with consequent loss of information. Band broadening can also occur with NMR techniques due to the presence of free radicals in the humic structure. The focus of this chapter is on characterization of bulk water samples with minimal pretreatment (e.g., filtration).

3.15.3 Bulk Water Analysis Protocols Several analytical methods and procedures have been developed for characterization of NOM/EfOM in bulk water, due to the complexity and heterogeneity in properties and structures of the compounds present. Some of the most commonly used methods for distinguishing EfOM from NOM are in drinking water treatment and wastewater reuse practices, which are elaborated below.

3.15.3.1 Dissolved Organic Carbon The concentration of total amount of organic matter molecules present in water is generally quantified as total organic carbon (TOC), given that carbon (C) is the building element of organic compounds. TOC consists of two fractions: (1) dissolved organic carbon (DOC) and (2) nondissolved particulate organic carbon (POC). DOC is the organic carbon passing through a 0.45-mm filter and represents the majority and the chemically reactive fraction of the organic matter present. DOC in natural water varies with the type of water ranging from as low as 0.5 mg l1 for some groundwaters and seawater to over 30 mg l1 for colored waters from swamps (Thurman, 1985). The DOC concentrations of primary, secondary, and tertiary effluents from WWTPs used for soil aquifer treatment (SAT; laboratory and field studies) were in the range of 9–35, 2–24, and 5–20 mg l1, respectively (Sharma et al., 2008). DOC, measured with TOC analyzers, provides a bulk measure of the amount of organic matter present. DOC can be measured using two types of analytical methods: combustion and wet oxidation. Generally, combustion methods are more accurate in DOC measurement than wet oxidation methods, specifically for seawater with high chloride concentration

419

(McKenna and Doering, 1995). The most commonly used DOC analysis methods include persulfate oxidation, ultraviolet (UV) irradiation, and a combination of the two (Sharp, 1993), specifically for the determination of low concentrations (Dafner and Wangersky, 2002).

3.15.3.2 Dissolved Organic Nitrogen Increasing use of nitrification–denitrification processes for municipal wastewater treatment leads to the presence of dissolved organic nitrogen (DON) as the main remaining form of nitrogen in the wastewater effluent. NOM also contains 1–5% of nitrogen by weight (Lee and Westerhoff, 2006). DON may act as nutrient and is a precursor of the carcinogenic DBP N-nitrosodimethylamine (NDMA), which is formed during disinfection using chlorine/chloramines. Measurement of DON is accomplished by a two-step process: first, involving elimination of dissolved inorganic nitrogen (DIN) by continuous dialysis through dialysis bags with a molecular weight cutoff (MWCO) of 100 Da, followed by direct measurement of DON as total nitrogen (TN) with a TN analyzer. Because the concentration of the DIN species is often considerably higher than that of the organic nitrogen species, DON measurements are potentially subject to substantial error. Pretreatment such as dialysis can improve the accuracy of DON measurement, separating DIN species (nitrate, nitrate, and ammonia) from DON (Lee, 2005); however, this method may be less useful for DON measurement in wastewater due to matrix effects.

3.15.3.3 Specific UV Absorbance UV absorbance (UVA) of a filtered (0.45 mm) sample at 254 nm (UVA254) is measured with a spectrophotometer to assess the humic contents or aromatic character (aromaticity) of the sample. The aromatic structure of NOM can absorb more UV light than an aliphatic structure. The UVA/DOC ratio, the specific UV absorbance (SUVA) expressed in l mg1 m1, is defined as the normalized UV absorbance of a water sample with respect to the DOC:

SUVA ¼

UV254 ðcm1 Þ  100 DOC ðmg l1 Þ

A SUVA value of 44 l mg1 m1 represents a water source dominated by humic substances and a SUVA o2 l mg1 m1 represents a source dominated by nonhumic material (Edzwald and Tobiason, 1999). Humic substances are more dominant in NOM, while nonhumic material is more dominant in EfOM. The SUVA of fulvic acid is higher than that of natural bulk waters (Westerhoff et al., 2001; Chow, 2006). In the case of comparing ozonated water and raw waters, ozonated water exhibits a lower SUVA value than the corresponding raw water because of the breakdown of aromatic structure by ozone. EfOM, compared to NOM, typically exhibits a lower SUVA.

3.15.3.4 XAD-8/-4 Adsorption Chromatography XAD-8 and XAD-4 resins (nonionic solid sorbents, Amberlite) have been used to isolate different organic matter from water.

420

Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter

The isolation procedure generally uses two types of XAD resin in series: XAD-8 resin first, followed by XAD-4 resin. This isolation technique, however, can also be used to determine polarity fractions of the DOC. An acidified water sample (pH 2.0) is first passed through a column of XAD-8 resin. The hydrophobic (HPO) fraction, mainly consisting of humic and fulvic acids, is quantitatively sorbed onto XAD-8 resin, with HPO-DOC calculated by the difference in initial DOC and the DOC of the XAD-8 effluent. The HPO fraction typically represents 50–65% of the DOC in water. The effluent from the XAD-8 column is then passed through a column of XAD-4 resin where the transphilic (TPI) fraction is adsorbed onto XAD-4. The hydrophilic (HPI) fraction is adsorbable neither on XAD-8 nor on XAD-4. These different fractions reflect differences in polarity, size, and charge density. HPO and TPI NOM components contain mainly acids and some neutrals, while hydrophilic NOM components contain mostly neutrals and bases (Thurman and Malcom, 1981; Malcolm and McCarthy, 1992; Labanowski and Feuillade, 2009). EfOM, compared to NOM, typically shows a greater proportion of polar organic matter.

3.15.3.5 Fluorescence Excitation-Emission Matrix Fluorescence spectroscopy is an extremely sensitive method that permits analysis of fluorescing organic matter fractions in water even at concentrations o1 mg l1. This method is useful to investigate the relative amounts of aromatic and nitrogen species, and to probe the origin of NOM (Chen et al., 2003). An excitation-emission matrix (EEM) is obtained by collecting the emission spectra over a series of excitation wavelengths. A fluorescence excitation-emission matrix (F-EEM) of a water sample is developed by scanning it over an excitation range of 240–450 nm by 10-nm increments and an emission range of 290–530 nm by 2-nm increments using spectrofluorometer. The result is a three-dimensional spectrum in which fluorescence intensity (arbitrary units) is represented as a function of excitation and emission wavelengths. There are three dominant peak areas observed in EEM: (1) humic/fulvic-like organic matter peak (at excitation ¼ 330–350 nm and emission ¼ 420–480 nm), (2) humic-like organic matter peak (at excitation ¼ 250–260 nm and emission ¼ 380–480 nm), and (3) protein-like organic matter peak (at excitation ¼ 250– 280 nm and emission ¼ 280–350 nm) (Baker and LamontBlack, 2001; Leenheer and Croue´, 2003). EfOM, compared to NOM, typically shows more protein-like and less humic-like organic matter. Based on an EEM, a fluorescence index (FI) can be calculated by the ratio of fluorescence intensities at emissions of 450 and 500 nm at an excitation 370 nm. A higher FI (B1.7 to B2.0) reflects organic matter of an autochthonous (microbial) origin, while a lower FI (B1.3 to B1.4) reflects organic matter of an allochthonous (terrestrial) origin. EfOM typically shows an autochthonous signature (higher FI), while NOM typically shows an allochthonous signature (lower FI) (Donahue et al., 1998; McKnight et al., 2001). Fluorescence is sensitive to factors such as pH, solvent polarity, temperature, redox potential of the medium, and interactions with metal ions and organic substances (Coble, 1996; Westerhoff et al., 2001; Leenheer and Croue´, 2003).

During F-EEM analysis of a sample, a uniform method of sample preparation is generally adopted. Filtered water samples are diluted to 1 mg l1 DOC with 0.01 N KCl, and adjusted to a pH of 3.0 before measurement. The EEMs of each sample are adjusted by subtracting an EEM of 0.01 N KCl (pH 3.0 adjusted with HCl) solution (set as a blank EEM) to remove Raman scatter peaks.

3.15.3.6 Liquid Chromatography with Online Organic Carbon Detection Liquid chromatography with online organic carbon detection (LC-OCD; also referred to as SEC-DOC, size-exclusion chromatography with DOC detection) is based on molecular size determination with gel permeation chromatography. The water sample is passed through a column of gel, and the extent to which fractions are retarded is a measure of their molecular size. Larger molecules that do not enter the gel pores pass through the columns, while smaller ones diffuse into the gel and take longer to pass through. This method thus provides an indication of the apparent molecular weight (MW) or molecular size (MS) of different classes of NOM fractions, including biopolymers (which include polysaccharides, organic colloids, and proteins), humic substances, building elements, low-molecular-weight acids (LMWAs), and low-molecularweight neutrals (LMWNs). An LC-OCD or SEC-DOC chromatogram can be represented either in terms of retention time or, if calibration chemicals are used, in terms of MW distribution – daltons (Amy and Her, 2004). A typical chromatogram of NOM present in surface water is shown in Figure 2 (Huber, 2007). The first fraction identified after approximately 25–45 min (first peak – largest molecular size) is the biopolymer peak with significant OCD only. The organic colloids and proteins present in this fraction provide responses in both OCD and UV detection (UVD). The second and third fraction responses in OCD and UVD are attributed to humic substances and building elements, respectively. The fourth response to OCD and UVD is attributed to LMWA. LMWNs comprise the last main fraction. LC-OCD can be used to effectively monitor polar NOM components with a lower SUVA, and has been successfully applied to monitoring changes in NOM associated with bank infiltration, SAT, ozone oxidation, coagulation, adsorption, bio-filtration, and membrane separation (Her et al., 2002). It has also been used to identify problematic NOM components in membrane fouling (Her et al., 2004a, 2004b).

3.15.4 EfOM versus NOM Differences in Bulk Water Parameters Previous studies have shown that there are some distinct differences between EfOM and NOM with respect to bulk water parameters, as listed below, which can be identified by using several bulk water analysis protocols (Frimmel and Abbt-Braun, 1999; Nam et al., 2008): 1. DOC and DON measurements show that EfOM has higher DOC, higher DON, and higher DON/DOC ratio compared to NOM. Organic matter from the wastewater-derived

Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter

421

12 Released from HSs after mid-oxidation Mw 350−500 g mol−1 Building blocks Mw 500−1200 g mol−1 Humics A) Comp. with IHSS B) Standards C) Ratio UV/DOC D) Retention time Acid hydrolysis and analysis of E) Peak form monosaccharides

10

Rel. signal response

8

Acids and LMW humics Mw < 350 g mol−1 Due to salt peak (nonbuffered sample)

OCD: organic carbon detection UVD: UV detection at 254 nm OND: organic nitrogen detection

Biopolymers Mw > 20.000 g mol−1

Retention time and tests with IEX resins LMW neutrals

6

Mw < 350 g mol−1 OCD

4

No OC response (even before BP fraction) By Raleigh scattering, no quantification Inorganic colloids

2

UVD Nitrate OND

0 20

30

40

50 Retention time in minutes

60

70

80

Figure 2 Typical NOM chromatogram of a surface water sample. From Huber S (2007) LC-OCD applications. DOC-Labor Dr. Huber, Germany. http:// www.doc-labor.de/english_pages/What_is_LC-OCD_About/What_is_LC-OCD_about_2007-2.pdf.

2. 3.

4.

5.

samples exhibits lower C and O contents and a significantly higher amount of H, N, and S. EfOM has lower SUVA compared to drinking water NOM (as it is low in humic substances and high in DOC). Compared to NOM, EfOM has a relatively higher amount of hydrophilic (polar) components than HPO (nonpolar) components (i.e., HPI-DOC4HPO-DOC). EfOM exhibits more dominant protein-like peaks in F-EEM as it is rich in nitrogenous organic matter of wastewater origin. Drinking water NOM normally does not show pronounced protein-like peaks unless impacted by AOM. Furthermore, EfOM exhibits higher FI values, indicating that the EfOM source is from autochthonous and microbially derived sources. Due to its high DOC concentration and high concentration of nonhumics, EfOM also exhibits a more dominant biopolymer peak (higher polysaccharide or protein content) in LC-OCD or SEC-DOC chromatograms.

These results from different studies clearly show that bulk water characterization is an effective method for differentiating EfOM from NOM. The following discussion illustrates these differences through application of the protocols within the context of several case studies.

3.15.5 Application of Protocols to Case Studies 3.15.5.1 Soil Column Studies Simulating SAT Using Primary and Secondary Effluent DOC and SUVA. Soil column studies were conducted using 5-m-long columns filled with bio-acclimated silica sand (0.8– 1.25 mm) and operated at a hydraulic loading rate of 1.25 m d1 under different conditions to simulate SAT. Primary and secondary effluents from a full-scale WWTP were applied to the soil columns and the removals of bulk organic compounds were monitored (after acclimation of the soil columns for about 60 days) using different analytical protocols. Table 1 summarizes the average DOC and SUVA of primary and secondary effluents before and after soil columns. Table 1 shows that for both primary and secondary effluents, DOC levels decreased and SUVA values for both effluents are in the range 2–4, indicating that the effluent is a mixture of both humic and nonhumic organic matter fractions (and mixture of HPO and hydrophilic compounds). SUVA values increased with soil passage, showing the preferential removal of nonhumics over humics in the soil columns. The amount of DOC removed in the column is also an indicator of biodegradable dissolved organic carbon (BDOC), which was confirmed by a strong correlation between the DOC removal profile and biomass profile along the depth of the column.

422

Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter

F-EEM. Figure 3 shows typical F-EEM spectra of primary and secondary effluents before and after soil column passage. Both influent and effluent showed three characteristic peaks of humic/fulvic-like, humic-like, and protein-like organic matter fractions. It was observed that with soil passage, there was a reduction in the intensities of all the characteristic peaks for both primary and secondary effluents. Table 2 presents the average reduction in intensities of the characteristic peaks of different organic matter fractions during SAT. It shows that SAT preferentially removes the protein-like organic matter fraction over humic-like organic

Table 1

matter fractions. More than 54% and 90% of the protein-like organic matter peak intensities were removed from primary and secondary effluents over a 5-m length of soil column. However, the reductions of intensities of other organic matter fractions are limited. This clearly supports the SUVA results in Table 1 that SAT preferentially removes non-humiclike organic matter fractions. This also shows that effluents from SAT systems have reduced concentrations of nonhumic and relatively high concentrations of humic matter, and thus more resembled characteristics of the natural water NOM.

Average DOC and SUVA values of primary and secondary effluents before and after simulated SAT

Effluent applied

Before SAT

After SAT

1

Primary effluent Secondary effluent

1

DOC (mg l )

SUVA (l mg

36.071.1 11.670.4

2.3070.2 2.9070.2

1

m )

DOC (mg l1)

SUVA (l mg1 m1)

15.571.5 9.770.2

2.9570.3 3.4070.2

PEO

PEI 440

440 60

420 400

400

50 40

360 340

1

30

320

20

340

1

30

280

10 3

2

240 300

350

(a)

400 Em (nm)

450

500

20

300

280

10

260 3 240

0

300

350

(b) 90

440

400 Em (nm)

2 450

500

0

90

440

420

80

420

80

400

70

400

70

380

60

380

60

360

50

360

50

340

40

320

Ex (nm)

Ex (nm)

40

360

320

300

340

40

320 30

300

20

260

10

260

0

240

300

350

400 Em (nm)

450

500

30

300

280

240

(c)

50

380 Ex (nm)

Ex (nm)

380

260

60

420

20

280

10 0 300

350

400 Em (nm)

450

500

(d)

Figure 3 F-EEM of primary and secondary effluents before and after SAT treatment (a) primary effluent; (b) primary effluent after SAT; (c) secondary effluent; and (d) secondary effluent after SAT.

Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter 3.15.5.2 Water Reclamation Case Study (China) Changes in the characteristics of bulk organic matter were monitored during a wastewater reclamation case study in China under the EU Reclaim Water project (RECLAIM WATER, 2008a, 2008b). Secondary effluent was pretreated with coagulation (polyaluminum chloride ¼ 10–30 mg l1) and then ozonated (ozone dose ¼ 10–15 mg l1) before infiltration. Table 3 compares the DOC and TN concentrations after different treatment steps. The DOC removal after infiltration was about 60%. Analysis showed that background groundwater had much higher DOC than that of secondary effluent and groundwater abstracted after infiltration. Furthermore, the TN concentrations of secondary effluent and background groundwater at this site were comparable. F-EEM spectra for samples were also measured and the intensities of characteristic peaks in the spectra are presented in Table 4. Three characteristic organic matter fraction peaks were visible in the F-EEM spectra of all the samples. The humic/fulvic-like peak intensity decreased by about 30%, whereas the (second) humic-like peak intensity increased after SAT, indicating a breakdown of larger humic/fulvic-like

423

organic matter fractions or a peak shift. There was an increase in the protein-like peak after infiltration, indicating some groundwater pollution, as background groundwater also has a high protein-like peak and TN. The FI values of the samples were 41.5, indicating a mixture of both autochthonous and allochthonous organic matter.

3.15.5.3 Impact of Wastewater Treatment Plant Effluent on River Water Quality (USA) The effects of WWTP effluent on characteristics of NOM present in river water were analyzed by conducting XAD fractionations, F-EEM, and SEC-DOC measurements of samples (1) upstream of a WWTP, (2) WWTP effluent, and (3) downstream of WWTP. Table 5 summarizes trends showing the characteristics of NOM, EfOM, and EfOM-impacted waters in the Northeast USA watershed. Compared to the upstream NOM sample, the WWTP effluent EfOM sample exhibited relatively low SUVA values (1.63–2.13 l mg1 m1, average: 1.90 l mg1 m1), increased fraction of hydrophilic organic matter (average: 34.8%), and higher FIs (average: 1.418). The

Table 2 Average % reduction in intensities of the peaks of different organic matter fractions and corresponding DOC removal for primary and secondary effluents during SAT Effluent type

Protein-like

Humic/fulvic-like

Humic-like

DOC removal

Primary effluent Secondary effluent

54.5 90.3

18.3 5.9

16.6 3.9

56.9 16.4

Table 3

TOC and TN measurement of water samples from a case study site

Parameter

Secondary effluent

After coagulation

After ozonation

After infiltration (SAT)

Background groundwater (local)

DOC (mg l1) TN (mg l1)

5.4 32.2

4.4 30.2

4.7 31.1

2.1 28.2

7.3 29.6

Table 4

Intensities of characteristic peaks in F-EEM spectra of samples from case study site

Sample

Humic/fulvic-like

Humic-like

Protein-like

FI

Secondary effluent After coagulation After ozonation After infiltration Background groundwater

19.2 17.3 10.9 12.0 11.9

25.9 22.9 15.2 43.5 23.4

12.2 13.2 13.7 24.8 19.7

1.57 1.63 1.60 1.45 1.55

Table 5

Characterization results of upstream water, wastewater, and downstream water

Sample

UVA254 (cm1)

DOC (mg l1)

SUVA l (mg1 m1)

HPO-DOC (%)

TPI-DOC (%)

HPI-DOC (%)

FI

Upstream of WWTP WWTP effluent Downstream of WWTP

0.100 0.120 0.122

3.60 6.39 4.31

2.76 1.90 2.82

52.1 41.3 56.2

22.7 23.9 21.1

25.2 34.8 22.8

1.229 1.418 1.261

Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter

low SUVA of WWTP effluent implies that the DOC of EfOM is comprised of more nonaromatic (or less aromatic) organic carbon than NOM. The FIs of EfOM were higher than those of NOM, which means that the properties of EfOM that distinguishes it from NOM are mainly microbial in origin. EEM measurements for upstream water, wastewater, and downstream of WWTP (influent of drinking water treatment plant) clearly showed different peaks between NOM and EfOM (Figure 4). The EEMs of wastewater and wastewaterimpacted water exhibited the presence of protein-like substances at the range of excitation wavelength 260–290 nm and emission wavelength 320–370 nm, which were at a similar location with other studies, and this protein-like peak likely originated from SMPs present in biologically treated wastewater. SEC-DOC of surface waters and wastewaters is separated into three main peaks according to their molecular-weight distributions, as shown in Figure 5: the zone 1 peak represents large molecules, such as polysaccharides, proteins, and colloids; the zone 2 peak is attributed to humic substances and

380

0.25

360

0.2

340

0.15

280 260 240 300

350

(a)

400 450 Em (nm)

1

Humic-like

0.8

380 Protein-like

0.1

300

Downstream of WWTP

400 Ex (nm)

Ex (nm)

420

0.3

320

The effects of source water quality matrix on the removal of different organic matter fractions during soil passage were investigated by conducting soil columns studies using a river water and a mixture of river water and secondary effluent (1:1), simulating wastewater-impacted surface water sources. The soil column depth was 5 m and the hydraulic loading rate was 0.56 m d1. Table 6 presents the DOC, UVA254, and SUVA values of two different types of water tested before and after the soil passage. After the ripening (bio-acclimation) of the soil column for 60 days, the average steady-state DOC removals were 18.3% and 33.5% for river water and mixture of river water and secondary effluent, respectively. Higher DOC removal in the case

440

0.35 400

3.15.5.4 Comparison of Removal of Bulk Organic Fractions from River Water and Wastewater Impacted River Water during Soil Passage

WW effluent (EFOM)

Upstream of WWTP (NOM) 440 420

building elements; and the zone 3 peak corresponds to lowmolecular-weight organic acids.

360

0.6

340 320

Ex (nm)

424

280

0

260 240 300

500 (b)

0.9

420

0.8

400

0.7

380

0.6

360

0.5

340

0.4

320

0.4

300

0.05

440

0.3

300 0.2 0 350

400 450 Em (nm)

280

0.2

260

0.1

240 300

500 (c)

0 350

400 450 Em (nm)

500

Figure 4 F-EEM spectra of samples collected upstream of WWTP (left), wastewater treatment plant effluent (middle), and downstream of WWTP (right) from a Northeast USA watershed (x-axis: emission wavelengths of 290–500 nm; y-axis: excitation wavelengths of 240–450 nm).

Log MW (Da) 5.3

4.3

3.3

2.3

1.3

0.3

0.25 Upstream of WWTP WW effluent Downstream of WWTP

Normalized intensity

0.2 0.15 0.1 0.05 0 0

1000

2000

3000

4000

5000

6000

7000

−0.05 Retention time (s) Figure 5 SEC of DOC fractions for upstream water, wastewater, and downstream water from a Northeast USA watershed, and their molecular-weight (MW) distributions.

Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter Table 6

Average DOC, UVA254, and SUVA values of river water and a mixture of river water and secondary effluent

Water type

Before soil passage

River water River water þ secondary effluent (1:1)

After soil passage

DOC (mg l1)

UVA254 (cm1)

SUVA (l mg1 m1)

DOC (mg l1)

UVA254 (cm1)

SUVA (l mg1 m1)

3.70 9.70

0.09 0.28

2.48 2.85

3.00 6.40

0.07 0.21

2.38 3.22

Lc1 inf 040209

Lc1 out 090209

90

440

90

440

420

80

420

80

400

70

400

70

380

60

380

60

50

360

50

360 340

40

320

Ex (nm)

Ex (nm)

425

30

300

340

20

260

10

260

0

240

300

350

(a)

400 Em (nm)

450

30

300

280 240

40

320

20

280

500

10 0 300

350

(b)

400 Em (nm)

450

500

Figure 6 F-EEM of river water during soil passage: (a) influent and (b) effluent.

Lc1 inf 040209

80

80

420

400

70

400

70

380

60

380

60

50

360

50

340

40

360 340

40

320

Ex (nm)

Ex (nm)

90

440

420

30

300

320

30

300 20

280 260 240 300 (a)

Lc2 out 090209

90

440

350

400 Em (nm)

450

20

280

10

260

0

240

500 (b)

10 0 300

350

400 Em (nm)

450

500

Figure 7 F-EEM of mixture of river water and secondary effluent during soil passage: (a) influent and (b) effluent.

of the mixture of river water and secondary effluent is likely due to higher concentration of biodegradable organic matter fractions present in secondary effluent. This shows that soil passage (bank filtration or artificial recharge) is effective in removal of bulk organic matter from wastewater-impacted water sources. For river water, SUVA decreased slightly during soil passage, while SUVA for the mixture of river water and secondary effluent increased with soil passage (Figures 6 and 7). Table 7 shows that reduction in F-EEM intensities for the mixture of river water and secondary effluent were 45.5% for

protein-like, 49.9% for humic/fulvic-like, and 51.4% for humic-like organic matter fractions. In comparison, for the river water, the reductions in peak intensities were 5.9% for protein-like, 2.8% for fulvic/humic-like, and 11.3% for humiclike organic matter fractions. Figure 8 shows the LC-OCD chromatograms for the river water and mixture of river water and secondary effluent before and after the soil passage. It was observed that there was preferential removal of biopolymer fractions while humics and other organic matter fractions were also removed to some

426

Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter

Table 7 Average % reduction in peaks of different organic matter fractions and corresponding DOC removal for river water and mixture of river water and secondary effluent during soil passage Water type

Protein-like

Humic/fulvic-like

Humic-like

DOC removal

River water River water þ secondary effluent

5.9 45.5

2.8 49.9

11.3 51.4

18.3 33.5

25 OCD UVD

Building blocks Humics (HS)

LMW acids and HS

20 Biopolymers Neutrals

Rel. signal response

RW+SE+OUT 15

RW+SE−IN 10

RW−OUT 5

RW−IN 0 0

20

40 60 Retention time (min)

80

100

Figure 8 LC-OCD chromatograms for the river water (RW) and mixture of river water and secondary effluent (RW þ SE) before and after the soil passage.

extent, which is in agreement with the results of F-EEM analysis. From LC-OCD chromatograms, it was estimated that biopolymer removals were 55% and 91% for river water and mixture of river water and secondary effluent, respectively. The corresponding removal efficiencies for humic fraction were 11% and 22%, respectively. For river water, the removal of building elements and neutrals were about 1% and 2%, respectively. Relatively higher removals of building elements (24%) and neutrals (36%) were observed for the mixture of river water and secondary effluent.

3.15.6 Summary Innovative NOM characterization tools, elucidating size, structure, and functionality are also applicable to samples containing EfOM and/or a mixture of NOM and EfOM. As

such, they provide a means of differentiating NOM from EfOM and elucidating wastewater impacts on drinking water sources through revelation of unique EfOM signatures.

References Amy G and Her N (2004) Size exclusion chromatography (SEC) with multiple detectors: A powerful tool in treatment process selection and performance monitoring. Water Science and Technology: Water Supply 4: 19--24. AWWARF (2000) Natural Organic Matter in Drinking Water: Recommendations to Water Utilities. Denver, CO: American Water Works Association Research Foundation. Baker A and Lamont-Black J (2001) Fluorescence of dissolved organic matter as natural tracer of ground water. Ground Water 39(5): 745--750. Chen J, LeBoeuf EJ, Dai S, and Gu B (2003) Fluorescence spectroscopic studies of natural organic matter fractions. Chemosphere 50: 639--647. Chow AT (2006) Disinfection byproduct reactivity of aquatic humic substances derived from soils. Water Research 40: 1426--1430.

Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter Coble PG (1996) Characterization of marine and terrestrial DOM in seawater using excitation–emission matrix spectroscopy. Marine Chemistry 51: 325--346. Dafner EV and Wangersky PJ (2002) A brief overview of modern directions in marine DOC studies. Part I. Methodological aspects. Journal of Environmental Monitoring 4: 48--54. Donahue WF, Schindler DW, Page SJ, and Stainton MP (1998) Acid-induced changes in DOC Quality in an experimental whole-lake manipulation. Environmental Science and Technology 32(19): 2954--2960. Drewes JE and Summers RS (2002) Natural organic matter removal during riverbank filtration: Current knowledge and research needs. In: Ray C, Melin G, and Linsky RB (eds.) Riverbank Filtration: Improving Source-Water Quality, ch. 15, pp. 303– 310. Dordrecht: Kluwer. Edzwald JK and Tobiason JE (1999) Enhanced Coagulation: USA requirements and a broader view. In: IAWQ-IWSA International Conference on Removal of Humic Substances from Water, IAWQ/IWSA Joint Specialist Group on Particle Separation, vol. 40, pp. 63–70. Trondheim, Norway, 24–26 June 1999. Oxford: Pergamon. Frimmel FH and Abbt-Braun G (1999) Basic characterization of reference NOM from Central Europe-similarities and differences. Environmental International 25(2–3): 191--207. Gadmar TC, Vogt RD, and Evje L (2005) Artefacts in XAD-8 NOM fractionation. International Journal of Environmental Analytical Chemistry 85(6): 365--376. Gaffney JS, Marley NA, and Clark SB (eds.) (1996) Humic and fulvic acids and organic colloidal materials in the environment. In: Humic and Fulvic Acids. Isolation, Structure and Environmental Role, ACS Symposium Series 651, ch. 1, pp. 2–16. Washington, DC: American Chemical Society. Her N, Amy G, Foss D, and Cho J (2002) Variations of molecular weight estimation by HP-size exclusion chromatography with UVA versus on-line DOC detection. Environmental Science and Technology 36: 3393--3399. Her N, Amy G, Park H, and Song M (2004a) Characterising algogenic organic matter (AOM) and evaluating associated NF membrane fouling. Water Research 38: 1427--1438. Her N, Amy G, Park H, and Von-Gunten U (2004b) UV absorbance ratio index with size exclusion chromatography (URI-SEC) as a NOM property indicator. Journal of Water Supply: Research and Technology – AQUA 57(1): 35--44. HDR Engineering Inc. (2001) Handbook of Public Water Systems, 2nd edn. New York, NY: Wiley. Huber S (2007) LC-OCD applications. DOC-Labor Dr. Huber, Germany. http:// www.doc-labor.de/english_pages/What_is_LC-OCD_About/What_is_LC-OCD_about_2007-2.pdf (accessed April 2010). Labanowski J and Feuillade G (2009) Combination of biodegradable organic matter quantification and XAD-fractionation as effective working parameter for the study of biodegradability in environmental and anthropic samples. Chemosphere 74(4): 605--611.

427

Lee W (2005). Occurrence, Molecular Weight and Treatability of Dissolved Organic Nitrogen. Doctoral Dissertation, Arizona State University, Tempe, AZ, USA. Lee W and Westerhoff P (2006) Dissolved organic nitrogen removal during water treatment by aluminum sulfate and cationic polymer coagulation. Water Research 40(20): 3767--3774. Leenheer JA and Croue´ J-P (2003) Characterizing dissolved aquatic organic matter. Environmental Science and Technology 37(1): 18A--26A. Malcolm RL and McCarthy P (1992) Quantitative evaluation of XAD 8 and XAD 4 resins used in tandem for removing organic solutes from water. Environment International 18: 597--607. McKenna JH and Doering PH (1995) Measurement of dissolved organic carbon by wet chemical oxidation with persulfate: Influence of chloride concentration and reagent volume. Marine Chemistry 48: 109--114. McKnight DM, Boyer EW, Doran PT, Westerhoff PK, Kulbe T, and Anderson DT (2001) Spectrofluorometric characterization of aquatic fulvic acid for determination of precursor organic material and general structural properties. Limnology and Oceanography 46(1): 38--48. Nam S-N, Krasner SW, and Amy GL (2008) Differentiating effluent organic matter (EfOM) from natural organic matter (NOM): Impact of EfOM on drinking water sources. In: Kim YJ and Platts U (eds.) Advanced Environmental Monitoring, ch. 20, pp. 259–270. Dordrecht: Springer. Owen DM, Amy GL, and Chowdhury ZK (1993) Characterization of Natural Organic Matter and Its Relationship to Treatability, 250p. Denver, CO: American Water Works Association Research Foundation. RECLAIM WATER (2008a) Monitoring of developed and adapted parameters. EU Reclaim Water Project. Milestone Report M 3.2. RECLAIM WATER (2008b) Water Reclamation Technologies. EU Reclaim Water Project. Deliverable No. D1.2/D5.2. Sharma SK, Harun CM, and Amy G (2008) Framework for assessment of performance of soil aquifer treatment systems. Water Science and Technology 57(6): 941--946. Sharp JH (1993) The dissolved organic carbon controversy: An update. Oceanography 6(2): 45--50. Shon HK, Vigneswaran S, and Snyder SA (2006) Effluent organic matter (EfOM) in wastewater: Constituents, effects, and treatment. Critical Reviews in Environmental Science and Technology 36(4): 327--374. Thurman EM (1985) Organic Geochemistry of Natural Waters. Dordrecht: Martinus Nijhoff/Dr Junk W Publishers. Thurman EM and Malcom R (1981) Preparative isolation of aquatic humic substances. Environmental Science and Technology 15: 463--466. Westerhoff P, Chen W, and Esparza M (2001) Organic compounds in the environment fluorescence analysis of a standard fulvic acid and tertiary treated wastewater. Journal of Environmental Quality 30: 2037--2046.

3.16 Chemical Basis for Water Technology P Huck, University of Waterloo, Waterloo, ON, Canada M Sozan´ski, Poznan´ University of Technology, Poznan´, Poland & 2011 Elsevier B.V. All rights reserved.

3.16.1 3.16.2 3.16.2.1 3.16.2.2 3.16.3 3.16.4 3.16.4.1 3.16.4.2 3.16.4.3 3.16.4.4 3.16.4.5 3.16.4.6 3.16.4.7 3.16.4.8 3.16.4.9 3.16.4.10 3.16.4.11 3.16.4.12 3.16.5 3.16.5.1 3.16.5.2 3.16.5.3 3.16.5.4 3.16.6 3.16.6.1 3.16.6.1.1 3.16.6.1.2 3.16.6.1.3 3.16.6.1.4 3.16.6.2 3.16.6.2.1 3.16.6.2.2 3.16.6.2.3 3.16.6.2.4 3.16.6.2.5 3.16.6.2.6 3.16.6.3 3.16.6.4 3.16.6.5 3.16.6.5.1 3.16.6.5.2 3.16.6.5.3 3.16.6.6 3.16.6.7 3.16.6.8 3.16.7 References

Introduction Goals and Processes for Water Treatment The Seven Goals Classification and Definition of Processes Used in Water Treatment Key Chemical and Physical Principles/Phenomena for Water Treatment Summary of Processes Used in Water Treatment Coagulation and Flocculation Sedimentation Flotation Filtration Membranes Disinfection Oxidation Gas–Liquid Transfer (Aeration and Air Stripping) Adsorption Biodegradation Ion Exchange (Including MIEXs) pH Correction The Evolving Nature of Water Treatment Increased Emphasis on Physical/Biological Processes The Evolving Role of Membranes Environmental Footprint Coping with Supply Constraints Addressing the Treatment Goals – From the Perspective of the Chemical, Physical, and Biological Processes Involved Particle Removal (Including Pathogens) Coagulation, flocculation, and sedimentation Flotation Filtration Membranes TOC Removal Enhanced coagulation MIEXs Biological treatment Adsorption Oxidation Nanofiltration Disinfection/Inactivation Maximizing Biological Stability Removal of Organic Chemical Contaminants Geosmin and MIB Pharmaceuticals and endocrine disrupting substances Volatile contaminants Inorganic Contaminants Maximizing Chemical Stability Maintaining Water Quality to the Consumer’s Tap Summary (Concluding Remarks)

429 429 429 430 431 432 432 434 435 435 436 437 438 438 439 440 442 442 442 443 443 443 443 444 444 444 444 444 445 445 445 445 446 447 447 447 447 447 448 448 458 460 460 463 463 464 466

429

430

Chemical Basis for Water Technology

3.16.1 Introduction This chapter reviews drinking water treatment from the perspective of the chemical phenomena and processes on which it is based. The framework is provided by the definition of seven goals for water treatment: removal of particles (including pathogens), total organic carbon (TOC) removal, disinfection/inactivation, maximizing biological stability, removal of chemical contaminants, maximizing chemical stability of the finished water, and maintaining quality to the point of use or consumption. The treatment required in a given situation is defined by the gap between the incoming or raw water quality, and the required final or finished water quality. The latter is typically defined by regulation or guideline. Achievement of the goals involves meeting defined values for a number of specific water-quality parameters. Many of these parameters have been discussed in detail in other chapters of this volume. In some situations treatment may not be required to address all of the seven goals, because the raw water quality may already be acceptable in this regard. As is evident later in the chapter, there are often a number of alternative processes that can be used to meet a specific goal, and some processes are capable of simultaneously addressing more than one goal. ‘Master variables’, such as pH and TOC, can both influence and be influenced by a number of treatment processes, and therefore constitute important linking factors among processes. Water treatment is based on a number of important chemical and physical (and biological) principles or phenomena. These include equilibria, kinetics, surface phenomena, and mass transfer. These are each reviewed briefly with regard to their importance for water treatment. The next section summarizes each of the major processes used in water treatment, from the perspective of both the basic phenomena on which they are based and also important considerations for their application. This section is followed by a discussion of the evolving nature of water treatment. Although drinking water treatment has traditionally been a relatively conservative field, with many of the processes having been in use for decades or longer, the rate of change is currently increasing. This is being driven by factors such as the development of new technology (e.g., the recent substantial reduction in costs and improvement in performance of membranes) and other developments such as minimizing the environmental impact (including energy requirements) of treatment. Considerable emphasis is given in the chapter to reviewing applications of the processes that can be used to achieve each of the treatment goals. This includes coverage of newer processes or newer applications of processes that may be less well described in the literature.

3.16.2 Goals and Processes for Water Treatment 3.16.2.1 The Seven Goals Seven goals can be defined for the treatment of drinking water. These goals are independent but linked, and the need to which each of them needs to be addressed in a given treatment situation depends on the gap between the raw water quality and the regulatory or otherwise desired finished water quality.

Where more than one treatment goal must be met, a treatment plant will invariably consist of several processes in series. As discussed in the next section, some goals may be met by a combination of treatment processes, and some treatment processes are capable of simultaneously addressing more than one goal. Each of the goals is described briefly below, in the general order in which they would be addressed in a treatment facility. 1. Removal of particles. This goal also includes the removal of colloidal material and the physical removal of pathogenic microorganisms. The removal of particulate matter is necessary for operational, public health, and esthetic reasons. Obviously, water that is turbid is not pleasant to drink and would be rejected by consumers. Operationally, particulate matter would settle out in the distribution system, leading to problems. However, the public health reason is paramount. Particulate matter both interferes with disinfection/ inactivation processes and may have microorganisms adsorbed to it. Therefore, it is necessary that particle concentrations (traditionally measured by light scattering in the form of turbidity) be reduced to low levels early in a treatment process. 2. Reduction in the concentration of TOC. The natural organic matter (NOM) present in all waters to varying degrees (discussed in detail in Chapter 3.15 Characterization Tools for Differentiating Natural Organic Matter from Effluent Organic Matter) can interfere with treatment objectives, lead to the creation of undesirable treatment byproducts, and be esthetically undesirable, that is, it may produce a noticeable color in the water. Depending on the TOC level in the raw water, the level may need to be reduced during treatment, although TOC does not need to be completely eliminated. The objectives for TOC removal are to reduce disinfectant demand and by-product formation, to reduce membrane fouling where that is relevant, and to improve the stability of disinfectant residuals in the distribution system in jurisdictions where these are employed. 3. Disinfection and inactivation. In virtually all cases disinfection or inactivation of microorganisms must be provided. In terms of immediate and short-term risks to health, achievement of this goal is invariably more important than meeting of the other goals. Only true groundwater can be reliably considered to contain no pathogenic microorganisms; however, this should be demonstrated on a case-by-case basis. To provide an increased level of public health protection, some jurisdictions require disinfection of all groundwater supplies, even though testing of regulated microorganisms may demonstrate their absence. Although various disinfectants are possible, the most practical ones are normally chlorine, chlorine dioxide, ozone, and ultraviolet (UV). Although chlorine has historically played a significant role in events of public health, issues involving by-product formation are seeing chlorine-based disinfectants lose ground to other approaches. Chlorine dioxide is used in some cases but does not offer some of the other process benefits of ozone nor the inactivation capability against Cryptosporidium of UV. 4. Removal of chemical contaminants. Jurisdictions regulate the levels of organic and inorganic chemical substances that

Chemical Basis for Water Technology

are considered acceptable in water, based in general on an assessment of risk to public health. In most cases, these substances are considered to represent a long-term health risk, often based on a cancer outcome. A number of these substances are discussed in detail in Chapter 3.02 Trace Metal(loid)s (As, Cd, Cu, Hg, Pb, PGE, Sb, and Zn) and Their Species and Chapter 3.04 Emerging Contaminants. When such substances are present in raw water above acceptable levels, they must be removed during the treatment process. In general, the options for doing this involve transferring the substance from the water to another phase (e.g., by volatilization, precipitation, or adsorption), oxidizing the substance either chemically or biologically, or physically removing it by a membrane process. There may be a link between achieving this goal and providing disinfection/inactivation because chemical disinfectants also function as oxidants. 5. Ensuring biological stability. Meeting this goal involves minimizing the opportunity for bacterial regrowth in the distribution system. Essentially, this involves introducing biological processes into the treatment train to remove sources of nutrients and energy for microorganisms, principally bacteria. In most drinking-water systems, biodegradable carbon is considered the limiting nutrient; however, there are systems where phosphorus has been found to be limiting. Achievement of this goal is more important in jurisdictions where either no or a minimal disinfectant residual is maintained in the distribution system. Where such a residual is maintained, it can suppress biological growth. In treatment processes involving membranes, maximizing the biological stability of the water upstream of the membrane will reduce the extent of biofouling on the membrane, which is an important operational issue. 6. Maximizing chemical stability. Raw water is invariably in chemical equilibrium before it enters the treatment plant; however, a number of treatment processes act to disturb this equilibrium. A prime example of this is changes in pH, for example, due to the addition of coagulants. One of the most important effects of this is to disturb the calcium carbonate equilibrium, which can lead to problems in the distribution system, in particular related to corrosion. It is also possible that dissolved coagulant residuals can precipitate in the distribution system. Depending on the extent of pH change during treatment, it may be necessary to correct pH at the end of the treatment process. 7. Ensuring esthetic quality. For consumers, the esthetic quality of the water (i.e., taste and odor) is one of the most important factors determining confidence and acceptance. In general, odor is a more common problem than taste, and most odor problems are related to excessive growth of algae or cyanobacteria in the raw water. Concentrations of odorous substances can be reduced to acceptable levels during treatment, and to some extent these substances can be considered a particular class of chemical contaminants. However, odors can also arise or increase in the distribution system, most likely in relation to biological instability of the distributed water. Although threshold odor concentrations for various common odorous substances have been determined, odor perception varies among individuals, and some individuals may detect low levels not

431

noticed by the majority of consumers. Although odorous substances themselves are not considered detrimental from a health perspective, meeting this goal is very important for drinking-water providers. It should however be noted that algae or cyanobacteria capable of producing odors may also be capable of producing toxins, which can have an adverse health effect.

3.16.2.2 Classification and Definition of Processes Used in Water Treatment Figure 1 shows the classification of both substances to be removed (contaminants) and processes in water treatment. At the top of the figure, the contaminants are classified by size: macroscopic, microscopic or colloidal, and truly dissolved. Simple treatment processes are classified into three categories according to their mechanism of action: physical, chemical, or biological. Complex processes, a category that applies to many of the processes actually used, involved two or all of these mechanisms. In the lower part of the figure, types of contaminants in each size range are linked to processes most commonly used to remove them, with the dominant mechanism(s) also identified. Thus, truly dissolved substances are removed by processes where chemical and/or biological mechanisms dominate. The most common treatment processes are described briefly in Section 3.16.4, and examples of their use are given in Section 3.16.6. Table 1 links processes to the achievement of each of the seven treatment goals identified previously. Although, for example, a number of processes can contribute to TOC removal, it is evident that really only a biological filtration process can improve the biological stability of the water. (Network or secondary disinfection is listed for completeness because it is used in a number of jurisdictions, although it is not really a treatment process. It does not of course remove substances causing biological instability, but rather masks or counteracts their effect (Huck and Gagnon 2004). Figure 2 shows the typical location in a treatment train where each of the goals would usually be met. (The figure is derived for surface waters requiring particle removal and for the more common case of a treatment train using granular media filtration rather than a membrane process.) The figure also gives a general indication of the number of processes that would typically contribute to the achievement of a given goal. Thus, removal of particles (and physical removal of pathogens) normally takes place early in the treatment train and a number of processes may contribute (e.g., coagulation, flocculation (sedimentation), and filtration). Disinfection or inactivation of pathogens would typically take place later in the treatment train and often is achieved in a single process (Figures 3 and 4).

3.16.3 Key Chemical and Physical Principles/ Phenomena for Water Treatment There are a number of well-known chemical principles or phenomena that are important for water technology. These are:

• • •

equilibrium, precipitation/dissolution, kinetics,

432

Chemical Basis for Water Technology

1. Classification of contaminants based on size Macroscopic: Microscopic (colloidal): Dissolved:

greater than 10−5 m from 10−5 to 10−9 m less than 10−9 m

2. Classification of simple processes based on mechanism Physical processes (PF) Chemical processes (PCh) Biological processes (PB) 3. Classification of complex processes based on integration of mechanisms Integration of processes

Physicochemical processes (PF-Ch)

PB

Biophysical processes (PB-F)

h -C

PB -F

PB

Biochemical processes (PB-Ch)

PB-Ch-F

Biophysicochemical processes (PB-Ch-F)

PB-Ch

PF

PCh

4. Classification of contaminants and removal processes Contaminant Suspensions Emulsions Algae Parasites Bacteria

Colloids Macromolecules Bacteria Viruses

Molecules Ions

Size classification Macroscopic

Processes

Mechanism

Screening Sedimentation Flotation Rapid filtration Micro filtration UV radiation

Physical processes dominate

Microscopic (colloidal)

Coagulation Rapid filtration Chemical oxidation Biological filtration Ultrafiltration Chemical stabilization

Integration of physical, chemical and biological processes

Dissolved

Oxidation/reduction Chemical precipitation Ion exchange Biological filtration Chemical stabilization

Chemical and biological processes dominate

Figure 1 Classification of contaminants and processes in water treatment.

• • •

oxidation/reduction, complexation, and surface phenomena.

The relevance of each of these is described briefly below, and some examples in relation to the goals defined in Section 3.16.2 are given. These principles are referred to in Section 3.16.6, where the various treatment goals are addressed from the perspective of the chemical (and physical/biological) processes involved. As in any process, equilibrium is important in defining the end state that the process can reach, given enough time. It thus provides an upper or lower limit on what may be achieved. In addition to homogeneous (i.e., single-phase) equilibria, heterogeneous equilibria, both gas–liquid and liquid–solid, are important in water treatment. The adsorption of contaminants on activated carbon provides an important example of liquid– solid equilibria. Gas–liquid equilibria are important in air stripping processes to remove volatile contaminants from water.

Precipitation and dissolution reactions are a special class of equilibria that are important in water treatment. One of the ways of removing dissolved contaminants or target substances is to cause them to precipitate or co-precipitate, and then physically remove the precipitate from the system. The feasibility of this depends on the solubility product, and an example of a precipitation process is the removal of phosphorus from wastewater using either calcium or iron. Dissolution is important in terms of chemical additions and also in terms of applications such as the use of solid calcium carbonate to dissolve gradually and add alkalinity to water. Kinetic phenomena are arguably among the most important in water treatment, because of the generally limited residence time of water in engineered treatment systems. Naturally based treatment systems can take advantage of much slower reactions; however, the retention time of water in individual treatment process steps typically ranges from a few seconds to perhaps several hours. Continuing the example of the adsorption processes mentioned above, true equilibrium is

Chemical Basis for Water Technology Table 1

433

Goals and Processes Goal

Process

Particle removal a

TOC removal

Coagulation/flocculation Sedimentation Flotation Rapid granular filtration Biological filtration Membranes (MF or UF) Membranes (NF) Disinfection/oxidationc Chlorine Ozone UV Ozone/UV or H2O2/UV Adsorption Air stripping Ion exchange (including MIEXs) Secondary disinfectiond pH correction

    ( ) 

()b () () ()

Disinfection/ inactivation

Biological stability

Chemical stability

() 

 () 

() ( )

Removal of chemical contaminants

 

() ()

()



   ()

()

Esthetic quality

()  ()    



()

  

  

a

Including physical removal of pathogens. Not necessarily the principal goal of this process. c When used as oxidant, may also provide (some) disinfection. d Provision of residual for distribution system. b

Number of processes

N

Removal of particles and pathogens

Removal of TOC

Removal of chemical contaminants

Removal of taste and odor

Chemical stability

Biological stability Disinfection/ inactivation 1 Beginning

Typical location in treatment train

(Secondary disinfection) End

Figure 2 Treatment goals and processes.

not reached because of the limited time the water is in contact with the activated carbon. In the case of this process the kinetics or rate of mass transfer of the absorbing molecule to the adsorption site is invariably limiting, rather than the kinetics of the actual adsorption itself. Another process where the kinetics are very important and determine the physical design is chemical disinfection or inactivation of microorganisms. For example, regulations in the USA (and similar ones have been adopted in other countries or jurisdictions) specify the contact times that must be provided, based on essentially a first-order kinetic model. Oxidation/reduction processes (primarily the former) are important in water treatment. A common way of removing contaminants is to oxidize them, in some cases to a form that

is more easily removed. Whether oxidation or reduction will occur depends on the redox potential of the system. Common oxidizing agents added to water include chlorine and ozone. To an important extent, the action of those substances as chemical disinfectants is based on oxidation reactions. Complexation. Complex formation, involving either organic or inorganic ligands, is important in certain aspects of water treatment. For example, trivalent metal ions (aluminum or iron) added as coagulants form complexes that are important in terms of reducing the charge on colloidal particles. Humic substances, often the major fraction of NOM in water, can form complexes with metal ions, affecting their solubility and treatability. Trace organic contaminants can also complex with humic substances.

434

Chemical Basis for Water Technology Pressure-driven membrane processes

Macromolecules (humics), viruses

Particles, sediment, algae, protozoa, bacteria

Multivalent ions & DOM

Monovalent ions (Na+, Cl−)

Water molecules

Pore size 0.1−1 μm

MF

1−100 nm (>1000 Da)

UF

~ 1 nm (200−1000 Da)

NF

<1nm (<200 Da)

RO

Figure 3 Removal capabilities of membrane types. From MWH (2005) Water Treatment: Principles and Design, 2nd edn., figure 12-2, p. 957. Hoboken, NJ: Wiley.

Diffused aeration

Spray towers

Percent removal

90.0 Cross-flow tower

99.0

Packed tower 99.9 99.99 Not feasible

99.999

CHCl3 PCE

NH3

CH4

99.9999

10−1

100

101

102

103

Mass transfer. Mass transfer is important in heterogeneous systems, and also in some cases in homogeneous systems. For example, von Gunten (2003a) has indicated that oxidation reactions involving hydroxyl radicals are with a few exceptions essentially diffusion controlled. Mass transfer is described mathematically by Fick’s first and second laws and the driving force for mass transport is the difference between the actual and equilibrium concentration in a given phase. In the case of adsorption of organic contaminants on activated carbon, either film or surface/pore diffusion, rather than the actual adsorption itself, is the rate-limiting step. In gas–liquid processes, such as air stripping of volatile contaminants from water or the transfer of ozone into water, maximizing interfacial mass transfer is crucial for successful process design. This involves both minimizing the thickness of the stagnant layers and maximizing the renewal of the surface.

104

Henry’s constant (atm) Figure 4 Gas-liquid separation process selection diagram. From MWH (2005) Water Treatment: Principles and Design, 2nd edn., figure 14-1, p. 1166. Hoboken, NJ: Wiley.

Surface phenomena. Because water treatment can involve all three phases (liquid, solid, and gaseous), surface phenomena can sometimes be important. For example, the behavior of colloidal particles is largely determined by the surface charge that they carry. Surface phenomena play an important role in the removal of particles by granular media filtration and the removal of various contaminants by various types of membrane processes.

3.16.4 Summary of Processes Used in Water Treatment 3.16.4.1 Coagulation and Flocculation Coagulation is one of the most important processes in water treatment and one where the chemistry can be extremely complex. Basically, coagulation involves either neutralizing the charges on colloidal particles so they can agglomerate in a subsequent flocculation step, or adding substances that can either bridge between like-charged particles or enmesh them. The flocculated particles can then be removed by a particleseparation process (sedimentation, flotation, granular media filtration, or low-pressure membranes). An important early article discussing the chemical aspects of coagulation is that of Stumm and O’Melia (1968).

Chemical Basis for Water Technology

Colloids or particles in water carry a charge, which for some species is influenced by pH. This charge leads to the creation of the well-known electrical double layer around the particles. Particle stability due to electrical double layer interactions is described by the well-known Derjaguin–Landau– Verwey–Overbeek (DLVO) theory, and the electrical forces prevent the particles coming close enough so that the physical attractive forces can draw and keep them together. Coagulants act either by forming charged intermediates that adsorb at the particle surface, thus reducing the charge and the thickness of the electrical double layer (charge neutralization), or by forming an amorphous hydroxide precipitate that can enmesh the particles (sweep coagulation). The most common coagulants are salts of aluminum or trivalent iron, with aluminum being more commonly used. As Edzwald (1993) points out, particles (both mineral and organic) may also be stable in water because of hydrophilic effects due to bound water or steric interactions from adsorbed macromolecules. He also notes that NOM rather than the particles in water can control the dosage and selection of coagulants, arguing that at neutral or acidic pH values, humic and fulvic acid organic ligands complex aluminum, resulting in an aluminum demand that must be satisfied before precipitation of aluminum hydroxide can occur. The removal of NOM with aluminum coagulants can involve aluminum hydrolysis reactions, complexation with aluminum, and precipitation. It can also involve adsorption on aluminum hydroxide that can precipitate once the aluminum demand referred to above is satisfied. Although alum (aluminum sulfate) and ferric chloride are the most widely used coagulants, the most commonly used among others include sodium aluminate, polyaluminum chloride (PACl), and cationic organic polymers. PACl has been introduced in recent decades, and is often more effective in cold waters. This is because some of the intermediates formed by alum are in fact polymeric species and in cold waters the kinetics of their formation are slower. The hydrolysis reactions of both alum and PACl have been examined by Van Benschoten and Edzwald (1990). The term ‘enhanced coagulation’ refers to a regulatory requirement in the US having the objective to remove TOC by coagulation to reduce the formation of disinfection byproducts. Edzwald and Tobiason (1999) introduced the term ‘multiple objective coagulation’. On this basis, the optimum coagulation conditions are those that maximize removals of pathogens in downstream processes, result in low turbidity values and particle counts, and minimize residual dissolved aluminum concentrations in the water. Sludge production and operating costs of course also play a role in optimizing coagulation. The physical process of flocculation involves the aggregation of the destabilized colloidal particles into loose macroscopic structures referred to as flocs. The rate of floc formation is determined by the rate of contact of the destabilized particles. Although perikinetic flocculation can take place by Brownian motion, this is not sufficiently rapid to be of practical importance in water treatment, and therefore orthokinetic flocculation is used. This occurs by the input of energy into the fluid, most commonly by mechanical mixing, although in some cases due to the turbulence created in the

435

water due to flow and the resulting loss of hydraulic head. The rate of flocculation, or rate of collision of primary particles per unit volume, is given by

Ji ¼ constant N2i d3i du=dz

ð1Þ

where Ji is the number of collisions of primary particles per unit time, Ni the number of primary particles per unit volume, di the diameter of the primary particles, and du/dz the local velocity gradient. Although Equation (1) indicates that increased time and energy input will reduce the number of primary particles, they also lead to breakup of flocs, which become more susceptible to shear forces as their size increases. Thus, there is an optimal Gt for flocculation, as well as optimal values for G and t, which are generally well established in practice. Often, flocculation will take place in several tanks in series, with progressively reducing Gt values to favor aggregation initially and minimize floc breakup in the later stages. Optimal G and t values will also be different depending on the downstream process, that is, whether sedimentation occurs or not and whether final particle separation is by classical granular media filtration or by a low-pressure membrane. In some cases, the so-called inline flocculation occurs without the use of a separate flocculation basin.

3.16.4.2 Sedimentation Sedimentation is one of the processes for which the development of theory has overtaken its application. In 1851, George G. Stokes analyzed the behavior of rigid, nonporous spherical particles in liquid under specific physical conditions. From a practical perspective, the important physical conditions he assumed were constant velocity under laminar flow conditions (Re r 0.4), and that the particles act individually. The force balance (the downward gravity force equals the sum of the upward buoyant and drag forces) leads to the wellknown Stokes’ law which allows calculation of the settling velocity for a given spherical particle under laminar flow conditions:

v ¼ gðrp  rÞd2p =18m

ð2Þ

where v is the settling velocity, g the acceleration due to gravity, rp the particle density, r the fluid (water) density, dp the particle diameter, and m the viscosity. The work of Stokes was subsequently verified experimentally, and supported the further development of the understanding of sedimentation through both theoretical and experimental work. The goal of the majority of this work was and remains attempting to develop mathematical relationships to describe the frictional resistance for a broad range of Reynolds numbers, and especially for nonspherical particles. In practice, Stokes’ law can only be used in a limited sense to describe sedimentation processes in water treatment. For example, the flocs developed as a result of coagulation do not follow Stokes’ law, because as they agglomerate their mass and size and therefore settling velocity increase. Also, actual suspensions are polydisperse, making their mathematical description more complicated. At higher concentrations of

436

Chemical Basis for Water Technology

particles or flocs, they interfere with one another, leading to a behavior known as zone settling. The practical value of Stokes’ law is in illustrating the major influence of particle diameter and density on settling behavior. Sedimentation is a well-established process in water treatment, and in practice design and operation are based on established ranges of important factors. Key parameters in this regard include the surface loading rate of the sedimentation basin (e.g., cubic meters per square meter per unit time, which has the units of velocity), residence time in the basin (related to the depth of the basin), and design of the inlet and outlet structures to minimize turbulence within the basin. So-called shallow depth sedimentation devices are now commonly used in water treatment. In these units the sedimentation basin contains a number of either closely spaced parallel plates or tubes on an angle. The water flows upward between the plates or within the tubes, thus minimizing the distance a particle or floc has to settle to be removed, and substantially increasing the surface loading rate at which the basin can be operated.

3.16.4.3 Flotation Dissolved air flotation (DAF) is an alternative to sedimentation that has found increased application in drinking-water treatment over the last several decades. In terms of criteria such as turbidity and particles, DAF can produce an effluent that is essentially equivalent to that from sedimentation. DAF basically involves attaching air bubbles (typically 10–100 mm diameter) to particles, although the fundamental mechanisms of this attachment are not well understood. The attachment of the bubbles causes particles in the water to float to the top of the tank. DAF is therefore especially suited for light particles such as algae, colloidal, or particulate NOM (causing color), for low to moderate turbidity waters that produce a light floc upon coagulation, and for cold waters. In some waters that vary seasonally, it is possible that operation of DAF may only be required during part of the year. The flotation process is accomplished by introducing air to the water under pressure prior to the flotation tank. Usually, the water that is pressurized is a portion of the effluent stream that has been recycled, and this pressurized water is mixed with the incoming flow prior to the flotation tank. When the water enters the tank near the bottom, the pressure is released causing the formation of fine bubbles that attach to solid particles as they rise to the surface. The float containing the solids that accumulates on the surface of the tank is then skimmed off mechanically or allowed to overflow the tank. The treated water, referred to as the subnatant, is withdrawn from the tank lower down. Principles and applications of DAF are discussed in detail in several references (e.g., Edzwald, 1995, 2010; MWH, 2005). One of the most important factors affecting DAF performance is proper coagulation (MWH, 2005). Also important are floc characteristics, bubble size and rise velocity, air loading, and floc-bubble attachment. Proper coagulation is important for a good attachment of particles to air bubbles and the same fundamental principles of coagulation discussed earlier apply here. DAF must be followed by a final particle-separation step. Normally, this would be granular media filtration, but it may

also be a low-pressure membrane (microfiltration (MF) or ultrafiltration (UF)).

3.16.4.4 Filtration Granular media filtration has historically been one of the main processes used in water treatment. It initially started as what is now referred to as slow sand filtration, however, rapid filtration has now been in use for more than 100 years. Additional processes where a filtration mechanism occurs are bank filtration and underground passage. Increasingly, rapid granular media filtration is being replaced by low-pressure membranes (Section 3.16.4.5) for particle removal. The theoretical background and practice of granular media filtration are well described in various standard environmental engineering textbooks (e.g., MWH, 2005). Despite its apparent simplicity, rapid granular filtration is one of the most complicated processes in water treatment, due to the large number of factors and parameters that determine its effectiveness, the complexity of what actually occurs in the filter bed and the resulting difficulty in unambiguously and accurately determining these parameters as well as mathematically describing the process. The fundamental variables and parameters of filtration change as a function of time and filter bed depth. The generally accepted filtration theory of Iwasaki (1937) was developed for filtration of a uniform suspension at constant filtration velocity, under isotropic and laminar conditions by a uniform bed. Further work on the development of filtration theory conducted by, among others, Ives (1960), Mints (1966), Hereit (1973), and Adin and Rebhun (1977) has led to important theoretical advances; however, the conditions investigated are generally not readily translatable to full-scale filtration practice. In addition, in the mathematical relationships developed by the authors mentioned above, the characteristics of the suspensions, filter bed and other parameters of the filtration process are often defined using constants and the parameters that are difficult to determine experimentally. Therefore, the existing theories of filtration cannot be directly applied for the design and operation of filters in practice. They are however helpful in planning more fundamentally based filtration investigations and interpreting the results of such studies. The particles (including pathogenic microorganisms) removed by granular media filtration are generally smaller than the channels through which they pass among the media grains. The first step of the removal process involves the transport of the particle from the bulk of the water to the surface of a media grain. In the second step, attachment to and retention on the media grain, particle or floc characteristics as well as media characteristics become important. There is thus a strong connection between pretreatment (coagulation/flocculation and possibly sedimentation) and the success of a downstream filtration step. O’Melia and Stumm (1967) published a seminal paper on the importance of physical and chemical factors in filtration. A detailed discussion is also provided in MHW (2005: ch. 11). Granular media filtration involves a well-defined cycle, beginning when the filter is placed back into service following the removal of previously accumulated particles through backwashing. Following this initial period there is normally

Chemical Basis for Water Technology

an extended period when filter effluent quality is at its best and pressure drop (headloss) is rising although relatively slowly because of particles being accumulated in the bed. At some point, often on the order of 24 h, the filter is ready for backwashing. Backwashing is typically initiated based on either a headloss trigger, a filter effluent turbidity (or particle count) trigger, or following a specified time. The first trigger is preferable to the second, because it ensures that filter effluent quality remains acceptable throughout the run. For example, even relatively modest increases in turbidity at the end of the filter cycle can correspond to a significant deterioration in the removal capability for Cryptosporidium (Huck et al., 2002). There are a number of factors that influence filtration performance: media (type, diameter), bed depth, hydraulic loading, chemical pretreatment, backwashing, and temperature. There are generally accepted ranges for these parameters, although temperature can of course not be controlled. It is generally desirable to base full-scale design on pilot testing. As indicated earlier, the removal of particles by filtration is a complex process. However, a useful fairly simple approximate mathematical representation is given by

CL =Co ¼ eZL=d

ð3Þ

where CL is the particle concentration at depth L, Co the influent particle concentration, Z the constant for a given filter bed, and d the media diameter. Thus, deeper beds and smaller media diameters will improve removals, although in a less-than-proportional way. Because headloss increases as particle diameter decreases, there is an inherent conflict between headloss and particle removal. The L/d ratio is sometimes used for filter design; a value of 1000 or greater is generally considered reasonable and MWH (2005: ch. 11) notes a range of 1000–2000. Most large filters are open to the atmosphere and flow is by gravity, although some small filters may be enclosed and operated under pressure. The choice of media is one of the most important decisions to be made. For rapid filtration there are essentially two variations: a relatively standard dual media design, typically with anthracite on top of sand, or a deeper bed monomedia (usually anthracite and known as filter adsorbers or GAC filter caps) design, which typically uses coarser media and operates at higher flow rates. Alternative media may include granular activated carbon (GAC; sometimes to replace only the anthracite) for simultaneous removal of dissolved contaminants by adsorption. Catalytic media have also been used for iron and manganese removal. In addition, some newer media have been developed that may be advantageous in specific applications.

3.16.4.5 Membranes In recent years, membranes have increasingly been used in drinking-water treatment, largely because of their substantial decrease in cost. On a life-cycle cost basis, low-pressure membranes (MF and UF) are sufficiently competitive with conventional particle removal processes (i.e., chemically assisted granular media filtration) that they are now normally given at least initial consideration in any plant upgrade or new

437

design. Additional reasons for adopting membranes can include improved and more robust treatment performance. Although hypothetically a membrane could eliminate the need for disinfection, such a single-barrier process may not be approved by regulatory authorities. A detailed description of membrane processes may be found elsewhere (e.g., MWH, 2005). Although there is currently limited application of ceramic membranes in drinking-water treatment, the overwhelming majority of membranes are made of polymeric material. The usual types of membranes that would be applied in drinking-water treatment from freshwater sources are MF, UF, and nanofiltration (NF). Reverse osmosis (RO) is used in desalination of seawater and brackish waters, and in water reuse applications, although NF may also be used in the latter. MF and UF are referred to as low-pressure membranes, whereas NF and RO are high-pressure membranes. A general indication of the classes of substances that can be removed by the common membrane types is given in Figure 3. Thus, low-pressure membranes are used for particle/pathogen removal, NF for organics (TOC and some micro-contaminant) removal and softening, and RO for micro-contaminant and inorganics removal. Low-pressure polymeric membranes are typically manufactured in a hollow fiber format, in which individual fibers are grouped together into a bundle or cassette. NF and RO membranes are configured as a spiral-wound flat sheet. As is evident in Figure 3, each membrane type has a range of pore sizes, and a given membrane could be classified as being in either of two categories (e.g., MF or UF). Nominal pore size can be used as an initial criterion in selecting a membrane type. However, because of the normal variability in pore size and other factors affecting rejection by membranes (see Section 3.16.6), the removal of a specific entity or substance in a specific water by a given membrane normally needs to be quantified by direct testing. Water is pushed/pulled through a membrane by a difference in pressure. Recovery is the percentage of the feedwater that passes through the membrane as useful product or permeate, and feasible recovery decreases with decreasing membrane pore size. However, membranes having a larger pore size require less pressure differential and therefore are cheaper to operate. Flux refers to the flow of a substance (either water or a specific contaminant) through the membrane, per unit area of membrane surface, in a given time. Units commonly used to express water flux are liters per square meter per hour (Lmh). The amount of membrane surface area and therefore the capital cost is directly related to the operating flux that can be maintained on an ongoing basis. In addition to water temperature (i.e., viscosity), the design of the membrane itself, and specifically its pore size or pore-size distribution, is a very important factor in determining achievable flux. However, in practice all membranes become fouled and this normally substantially decreases the flux for a given pressure differential. Fouling is caused by accumulation, on the surface of the membrane or within the pores, of material rejected by the membrane. The types of fouling that can occur are:

•

colloidal and particulate, caused by nondissolved inorganic and organic matter in the water, including particulate microbial cell components;

438

• • •

Chemical Basis for Water Technology

organic, caused by components of the background TOC, including dissolved microbial cell components; biofouling, caused by the actual growth of microorganisms on the membrane surface; and inorganic, caused by the deposition or precipitation of inorganic salts on the membrane surface.

For low-pressure membranes that are backpulsed, a distinction can be made between hydraulically reversible and irreversible fouling. Chemical cleaning is normally used periodically with all membranes to remove accumulated foulants. The factors controlling the extent of fouling in a particular situation are the type of membrane and its pore-size distribution, the membrane material, the quantity and nature of fouling material present, the pretreatment applied, the cleaning regime used for the membrane, and the operating conditions, especially the flux. Fouling is a complex and important phenomenon, and further detailed discussion is outside the scope of this chapter. Understanding and mitigating fouling is an extensive and very active research domain at the time of writing. One recent important finding has been the demonstration of the importance of organic fouling for low-pressure membranes. Particularly important in this regard are biopolymers, consisting primarily of polysaccharides and protein-like material. Recent bench and pilot-scale investigations have shown rapid biological filtration to be effective in reducing bioorganic fouling of low-pressure membranes (e.g., Halle´ et al., 2009). Hijnen et al. (2009) have demonstrated the important role played by easily biodegradable carbon in the biofouling of high-pressure membranes. The importance of fouling in the present context is that pretreatment steps to reduce fouling have an important influence on the overall treatment train.

3.16.4.6 Disinfection Achieving proper disinfection is one of the most important goals in water treatment. In general, this goal is the only one which, if not achieved, can create an acute as opposed to a long-term health risk. For this reason many jurisdictions specify the level of disinfection that must be achieved. Although required disinfection levels are generally similar around the world, the specific values can be different in different jurisdictions. Normally, disinfection requires application of a process to reduce the concentrations of pathogenic microorganisms present in the raw water to levels representing a negligible health risk. Thus, raw waters that are more heavily contaminated will require greater reductions. Quantitative microbial risk assessment (QMRA) has been used recently in some cases to assist in arriving at appropriate disinfection requirements. Disinfection does not imply complete sterilization of the water. The term inactivation is often used in recent years with regard to protozoan pathogens and also applies to viruses, and therefore the term disinfection/inactivation is often utilized. Pathogens can also be physically removed, for example, by granular media filtration or by membranes, contributing to the overall reduction required. Inactivations or removals are normally expressed in logarithmic units. For example, three-

log removal means that initial microbial concentrations have been reduced by three orders of magnitude (99.9%). Because disinfection is widely described in various standard and specialized texts, the topic is treated only briefly in this section. After providing some general comments, the emphasis is on how disinfection processes may impact or be impacted by the rest of the treatment train. The three types of pathogenic microorganisms of greatest concern in drinking water are bacteria, viruses, and protozoans, specifically Giardia and Cryptosporidium. In general, the latter are most difficult to disinfect/inactivate, and will often drive process design and operation. (In the case of UV however, some viruses may be the most difficult to inactivate.) Historically, chlorine was the major disinfectant used worldwide, and its introduction approximately 100 years ago or more represents one of the great public-health advances of all time. However with the development in the 1970s of the ability to measure potentially hazardous chlorination byproducts such as trihalomethanes in water, alternatives were more actively sought. Ozone, which had been in greater use in Europe, came to be considered as the generally best alternative. However when Cryptosporidium became a concern in drinking water field in the 1990s, it became evident that the doses and contact times of ozone, especially at low water temperatures, could make it economically unattractive. At about this time, it was demonstrated that UV radiation could be effective against Cryptosporidium at reasonable dosages, and for approximately the past decade that water industry globally has seen a major increase in the use of UV for disinfection. Other disinfection agents may be used in special cases and small systems; however, the three just mentioned are the workhorses of municipal water treatment. Some jurisdictions require the maintenance of a disinfectant residual in the distribution system, and in cases where chlorine would lead to unacceptably high levels of chlorination by-products, the distribution system residual may be provided by chloramination, either by the addition of preformed chloramines at the end of the treatment process, or by the addition of ammonia to react with chlorine already present. In the case of the chemical disinfectants, chlorine and ozone, practice or regulations involve maintaining a desired concentration of the disinfectant in contact with the water for a specified period of time during treatment. In some jurisdictions a CT value is specified, which is the product of the disinfection concentration (C) and contact time (T), calculated in a prescribed way. In the case of UV disinfection, a fluence is specified, being the product of UV intensity and time. Contact times for chlorine are generally on the order of tens of minutes, those for ozone on the order of a few minutes, and those for UV on the order of a few seconds. The major interactions between disinfection and other treatment processes or requirements can be summarized as follows: 1. Is oxidation also required, for example, for the removal of odorous compounds? 2. What level of physical removal can the process train achieve? 3. What is the TOC concentration in the water at the point of disinfection? This will affect both the level of by-product

Chemical Basis for Water Technology

formation, as well as the initial disinfectant demand or UV transmittance. If the TOC level is too high, some TOC removal may be required prior to the disinfection step. 4. What is the pH of the water? Lower pHs are more advantageous for disinfection with chlorine because more of the hypochlorite acid formed by reaction of the chlorine with the water is present in the undissocciated form, which is more effective for disinfection. In a full-scale study with ozone, Urfer et al. (1999) demonstrated that both the lower pH and lower TOC provided by enhanced coagulation prior to ozonation were beneficial in maintaining a higher ozone concentration in the disinfection step. 5. What is the range of water temperatures? Lower temperatures lead to the need for greater disinfectant concentrations and/or contact times to maintain the same level of disinfection. In summary, the disinfection requirements for a particular water often have a major impact on treatment and need to be considered together with the other goals to be met, in order to arrive at an optimally configured treatment train.

3.16.4.7 Oxidation Oxidation is another important process in water treatment, and is capable of addressing several treatment goals (Table 1). The discussion herein includes the interplay between oxidation and other processes. As indicated in the previous section, the major chemical disinfectants (chlorine and ozone) serve as oxidants. In addition, UV used for disinfection can carry out oxidation by direct photolysis, and in combination with hydrogen peroxide can act as an advanced oxidation process (AOP), due to the generation of hydroxyl radicals. However, it should be noted that the UV doses necessary for effective oxidation of microcontaminants are generally much higher, perhaps by even an order of magnitude, then the doses generally required for disinfection/inactivation. Ozone on its own can act through either the molecular ozone or hydroxyl radical pathway and in combination with hydrogen peroxide can also serve as an AOP, because the production of hydroxyl radicals is greatly increased. Other oxidants such as potassium permanganate can be used in specific applications, such as some taste and odor control situations. Biological oxidation is also important and is discussed later in Section 3.16.4. Although in previous decades chlorine was used as an oxidant, this is becoming less common. This is related directly and indirectly to the formation of chlorination by-products, the indirect relationship being that other disinfectants (primarily ozone and UV) are replacing chlorine and so it is less commonly present in treatment processes. A detailed discussion of the use of oxidation to reduce concentrations of the odorous compounds geosmin and 2-methylisoborneol (MIB) is provided in Section 3.16.6. Comprehensive reviews of the use of ozone in drinking water for both disinfection and oxidation are provided by von Gunten (2003a, 2003b). In terms of the interrelationship between oxidation and other treatment processes, the following points can be noted:

• •

oxidation can assist in coagulation and filtration, ozonation of NOM increases its biodegradability,

• •

439

oxidation increases the polarity of NOM, potentially reducing its adsorb ability on activated carbon, and oxidation early in the treatment process can lyse algal and cyanobacterial cells before they can be physically removed.

Oxidation reactions, including those from UV, can produce by-products, because the oxidized substances are rarely mineralized completely. The chlorination by-products have been most studied, and while those from ozonation and UV are either considered to be less harmful or are present at much lower concentrations, ongoing research need is the identification of by-products, particularly from oxidation of trace contaminants such as pharmaceuticals. When hydrogen peroxide is used as part of an AOP, consideration must be given to removal of any hydrogen peroxide residual exiting the process. Urfer and Huck (1997) have demonstrated the removals achievable by biological filtration. Chemical reducing agents can also be used, and in previous practice have sometimes been used where high chlorination dosages were used (referred to as superchlorination).

3.16.4.8 Gas–Liquid Transfer (Aeration and Air Stripping) Gas–liquid transfer processes are important in water treatment, especially for the treatment of groundwater. They can be used as aeration to add oxygen to water, which increases the redox potential and may be important for biological processes. In some cases, they can be used to add carbon dioxide. They can also be used to strip volatile components from water, including excess carbon dioxide, hydrogen sulfide or methane, and volatile organic contaminants. The discussion here does not consider gas–liquid transfer with chemical reaction (e.g., ozone addition), which is more complicated. Although there are some general similarities between aeration and air stripping (because in both cases good contact between the water and air must be assured), the equipment used for the two types of processes is generally different. The following discussion focuses primarily on air stripping. Gas (air) stripping has been a mature process in the chemical processing industry for a long time and its principles are well understood. Its wider application to drinking-water treatment began several decades ago when volatile organic contaminants began to be of wider concern. Several excellent references are available (e.g., MWH, 2005) and the design process, including the use of commercially available software, is well established. The most important parameter with regard to the feasibility of air stripping for a particular contaminant is the Henry constant, discussed below. Because air stripping does not destroy contaminants but simply transfers them to another phase, treatment of the offgases is normally required. Both equilibrium and kinetics (rate) considerations are important for air stripping. In terms of equilibrium, the driving force for gas transfer between the water and the air is the difference between the existing and equilibrium concentrations in the two phases. At equilibrium, the concentration (partial pressure) of the contaminant in the gas phase (air) is proportional to its concentration in the water. This relationship is known as Henry’s law. Most gases and vapors follow Henry’s law in the range of interest in water treatment (Kavanaugh and Trussell, 1980). Henry’s constant, which is a

440

Chemical Basis for Water Technology

measure of volatility for a particular substance, is given by the following equation:

H ¼ PT Y=X

ð4Þ

where H is Henry’s constant for a particular contaminant, PT the total pressure, Y the mole fraction (concentration) of the contaminant in the gas phase (air), and X the mole fraction (concentration) of the contaminant in the water. Because Henry’s constant is an intrinsic property of a particular gas, contaminants with a higher Henry’s constant will tend to partition more into the gas phase, that is, be more easily stripped from water. Conversely, a substance with a low Henry’s constant will be relatively soluble in water and therefore difficult to strip but easy to absorb into water. Henry’s constants for most contaminants of interest are tabulated in standard references. It should be noted that different units can be used to express Henry’s constant, which affects the numerical value. Therefore, when comparing Henry’s constants for different contaminants, one should check to be sure that they are reported in the same units. Henry’s constant is a strong function of temperature, increasing as the temperature increases. Although the temperature of groundwater is often relatively constant year-round, stripping towers are often located outside and therefore, depending on the local climatic conditions, the temperature of the process may vary considerably. Although the value of H decreases as the salt concentration of the water increases, for drinking water applications such changes can be ignored. Although pH does not have a direct effect on Henry’s constant, it can have an influence on the removal of some contaminants (such as hydrogen sulfide) that ionize, because only the unionized species is volatile. An example of the impact of pH on hydrogen sulfide removal is given by MWH (2005: 1175). The rate at which a volatile contaminant is transferred across the air–water interface depends on both the driving force (the difference between the actual and the equilibrium concentration) and the intensity of mixing at the interface. For the design of air stripping equipment, it is important to be able to determine the unit rate of gas transfer (i.e., contaminant removal per unit time and unit area of interface). There are three consecutive steps involved in gas transfer: transfer from the bulk water to the interface, transfer across the interface, and transfer away from the interface into the bulk of the gas (air). For most gas–liquid contacting systems of practical interest, flow of one or both phases is turbulent, and the rate of gas transfer must be estimated empirically. An overall mass-transfer coefficient can be calculated that incorporates the mass-transfer resistances in both phases. For most situations in water treatment, resistances in the water control the overall rate of mass transfer (James, 1985). The overall rate of mass transfer also depends on the interfacial area. Increasing this area (and ensuring that it is continuously renewed) requires energy, and therefore a balance must be struck in air-stripping design between increased mass transfer and increased energy costs. In turbulent flow, the specific interfacial area is difficult to determine. Therefore, the common practice is to measure the product of the masstransfer coefficient (KL or KG, the overall coefficient for the liquid or gas phase, respectively) and a (the total interfacial

area divided by the system volume, or specific interfacial area). This quantity KLa or KGa is known as the volumetric mass transfer coefficient and is the value typically provided by manufacturers for various air-stripping equipment. (It should be noted that the discussion above regarding the rate of gas transfer applies to situations in which the gas does not react in the water. Volatile contaminants to be removed by air stripping do not normally react in water.)

3.16.4.9 Adsorption Adsorption is a major process that can be used for the removal of chemical contaminants, and its use in water treatment is especially for the removal of organic contaminants. It is applicable to both groundwater and surface waters; however, it can be a relatively expensive process depending on the levels of contaminants to be removed and the presence of other substances in the water (such as background TOC) that also adsorb, reducing the capacity of the process for the contaminants of interest. By far, the most commonly used adsorbent in water treatment is activated carbon, which can be applied in two forms: GAC or powdered activated carbon (PAC). A detailed discussion of the theory of adsorption is outside the scope of this chapter. Basically, from the viewpoint of thermodynamics, the contaminant prefers to be on the surface of the carbon rather than in the water. The process of adsorption is not instantaneous however; therefore, the contact time between the activated carbon and the contaminant is important. Several excellent references (Sontheimer et al., 1988; MWH, 2005) provide more detail on the fundamentals of adsorption. Basically, activated carbon is produced by heating a carbon-based material such as coal or wood to high temperatures in the absence of oxygen. This modifies the surface properties of the carbon and creates a very high surface area (on the order of 1000 m2 g1). This large surface area is due to the extremely fine pores that are created in the carbon structure by the activation process. The smallest pores, referred to as micropores, are not much larger than the adsorbing contaminant molecules and are the ones responsible for the carbon’s capacity. The capacity for specific types of compounds is affected by the starting material and the manufacturing process, which in turn affect the chemistry of the carbon surface and the pore structure. Activated carbon is a heterogeneous material, meaning that there are different types of sites with different energies for adsorption. The properties of a given contaminant are also very important in determining the extent to which it will be adsorbed. For charged compounds, pH can play a major role. Important factors include molecular weight, chemical structure, and polarity. For practical purposes the effect of temperature on adsorption can be ignored in water treatment. Li et al. (2005) have developed a model to predict adsorption based on activated carbon and contaminant properties. The extent to which a given contaminant will adsorb at equilibrium is expressed by an isotherm. Although various equations can be used to describe isotherms, in water treatment practice the most commonly used one is the Freundlich equation:

Q ¼ KC 1=n

ð5Þ

Chemical Basis for Water Technology

where Q is the equilibrium adsorbed phase concentration of adsorbent, K the Freundlich adsorption capacity parameter, C the equilibrium liquid phase concentration of adsorbent, and 1/n the Freundlich adsorption intensity parameter. The literature contains a considerable amount of ‘purewater’ (e.g., distilled and/or deionized water) isotherm data for many common contaminants. Although an isotherm is specific to a given activated carbon, most of the available data have been obtained with carbons that are commonly available commercially. Because activated carbon is a very nonspecific adsorbent, it will also remove other substances in the water, primarily organics, which compete with the target compound(s) for adsorption sites. In practice, the most significant competition is from background TOC, which can reduce the capacity of the carbon substantially below that predicted from purewater isotherms (e.g., Graham et al., 2000a). For PAC, which is typically in contact with the water for at most on the order of an hour, direct competition is most important. For GAC, significant preloading of background TOC can also occur.The practical significance of competition and preloading is that pure-water isotherms can provide only an indication of relative adsorbability, and cannot be used for design. Because the contact times in water-treatment practice are rarely sufficient for even approximate equilibrium to be achieved, process kinetics and, in practice the available contact time, are important. The movement of a contaminant from the water to the surface of the carbon consists of three sequential steps: diffusion through the stagnant layer of water surrounding the carbon particle (film diffusion), diffusion within or along the surface of the carbon pore (pore or surface diffusion), and finally the actual adsorption at a site within the pore. Usually, one or other of the diffusion steps is rate limiting. This is true for both GAC and PAC. Although modeling and short-term laboratory testing of adsorption processes can be performed, they are especially complicated for GAC because of the preloading discussed above and because of potential biodegradation of some contaminants on the GAC. Yu et al. (2009a, 2009b) have addressed and modeled the significant impact of preloading on both kinetics and equilibrium for the adsorption of trace levels of selected pharmaceuticals and an endocrine disrupting compound. GAC is typically regenerated, whereas PAC is not. Regeneration involves heating the carbon under appropriate conditions and results in the restoration of most of its original adsorptive capacity. Detailed information regarding regeneration (reactivation) is provided by Sontheimer et al. (1988). Another important issue for GAC is the potential desorption of adsorbed contaminants. Because adsorption is an equilibrium process, a decrease in the influent concentration of a particular contaminant can cause some of it to be desorbed from the carbon and appear in the effluent. This can lead to the unusual situation in which the concentration of a contaminant is higher in the GAC effluent for a period of time than in the influent. In Section 3.16.6.5 the adsorption of odor-causing compounds is addressed in detail and the practical implications of issues considered in this section are discussed.

441

3.16.4.10 Biodegradation Biological treatment has played an important role historically in drinking-water preparation in processes such as slow sand filtration, bank filtration, and underground passage. Within the last several decades, rapid biological treatment has been implemented globally in a number of treatment facilities, in the form of the biological activated carbon (BAC) process. This process combines ozonation, one of whose effects is to increase the biodegradability of NOM in water, with a biologically active carbon filter. Rapid biological treatment has also been implemented with nonadsorbing filter media instead of GAC, and without prior ozonation. By reducing the concentration of biodegradable organic matter (BOM) in the water, biological treatment reduces the opportunity for bacterial regrowth in the distribution system. This can also lessen problems such as taste and odor and corrosion. Within the last few years, biological treatment has attracted increasing interest in North America, in part because of stricter regulations on disinfection by-products. By removing at least part of the biodegradable portion of NOM, biological treatment reduces the concentration of chlorination by-product precursors. Since chlorine demand is also reduced, a lower chlorine dosage can be applied to maintain a given residual in the distribution system. Although biological treatment can also remove various substances from drinking water such as ammonia, iron, and trace organic contaminants (Rittmann and Huck, 1989; Halle´, 2010), its major application is for BOM removal. Comprehensive reviews on biofiltration are provided by Urfer et al. (1997) and Huck and Sozan´ski (2008). The benefit derived from biological processes in drinkingwater treatment is based on the ability of bacteria to oxidize substances present in water that can be referred to as biological instability. Although bacteria can also reduce substances, the focus of this discussion is on oxidation, which is more common. For oxidation to take place, there must be an electron donor, an electron acceptor, appropriate nutrients and suitable environmental conditions such as temperature, pH, and the absence of toxic substances. Common electron donors in water are organic matter and ammonia. Iron and hydrogen sulfide are among other substances that can serve as electron donors. Since drinking water must remain aerobic, the electron acceptor normally of interest in drinking-water processes is oxygen. The major nutrients aside from carbon are nitrogen and phosphorus. The various other micro-nutrients required by bacteria are normally present in appropriate amounts. For most drinking waters, the limiting nutrient is organic carbon, although phosphorus may also be limiting in some waters (Miettinen et al., 1997). Bacteria which can only use organic forms of carbon are referred to as heterotrophs, and are therefore the bacteria of interest for BOM removal. Part of the organic carbon is converted to cell material and may exit the process eventually. The remainder of the organic carbon is either transformed to other soluble organics with a lower energy level (i.e., more oxidized) or mineralized to carbon dioxide. Since waters used for drinking-water preparation generally contain low levels of nutrients compared to wastewaters,

442

Chemical Basis for Water Technology

bacteria which dominate in this environment are referred to as oligotrophs. The bacteria of importance in biological drinking-water treatment exist as a biofilm, or agglomeration of bacterial cells and extracellular polymers on solid support surfaces such as filter media. Suspended or planktonic organisms are generally unimportant for removals because of their relatively low numbers and the relatively short process retention times. Biofilms in drinking-water treatment are generally much thinner than those in wastewater treatment, are normally not visible to the naked eye, and may be discontinuous or patchy. Any biological treatment process should be considered as an ecosystem, which may have additional levels of the food chain present, most commonly protozoa. The organic carbon oxidized by the bacteria is commonly referred to as the substrate. It is commonly assumed that bacterial growth is linearly related to substrate utilization through the yield coefficient. The most common models developed to describe drinkingwater biofiltration have been summarized by Urfer et al. (1997) and Huck and Sozan´ski (2008). For practical purposes, BOM removal can be approximated as a first-order process (Huck et al., 1994). The practical objective of any modeling is ultimately to provide a tool for process design and operation. A common design question is the required bed depth or contact time to achieve a given treatment objective. BOM is the most common electron donor and energy source of interest in drinking water. It is present in low concentrations, often less than 0.5 mg l1. It can be composed of many different organic substances and cannot readily be distinguished chemically from nonbiodegradable NOM. Therefore, some type of assay in which the BOM is biodegraded is required for its measurement. The biochemical oxygen demand (BOD) test used in wastewater treatment is far too insensitive to be used for drinking water. The initial BOM measurement method for drinking water was proposed by van der Kooij et al. (1982) and measures a parameter that he termed easily assimilable organic carbon (AOC). The AOC content of the sample is determined using a calibration curve and the measured bacterial growth. A subsequent modification (Haddix et al., 2004) uses strains that fluoresce. More recently, a different AOC method has been published by Hammes and Egli (2005). In critically reviewing the BOM methods available at that time, Huck (1990) divided the methods into those which measure bacterial growth and those which measure a change in dissolved organic carbon (DOC) concentration as a result of biodegradation. The parameter measured by these latter methods is termed biodegradable dissolved organic carbon (BDOC). Huck (1990) noted that the method chosen should relate to the purpose of the measurement: if the purpose is to control bacterial growth, then a method which measures bacterial growth should be used, whereas if the objective is to measure reduction in chlorine demand or chlorination byproduct precursors, then a method which measures DOC is more appropriate. A BDOC method which uses a column packed with a special type of glass beads was reported by Frias et al. (1992) and subsequently further refined (Kaplan et al., 1993). In the method, DOC is measured in the influent and effluent of the column. If a column is placed on-line at a given point in a

treatment plant or distribution system, the method can therefore give a BDOC value as quickly as DOC can be measured. van der Kooij et al. (1995) have developed the biofilm formation rate protocol and apparatus, which directly measures bacterial accumulation on rings (e.g., glass) simulating sections of pipe. Methods that provide only an overall measure of BOM may be insufficient for quantitative process design and optimization. Rather, knowledge and quantitation of the major BOM components may be required. In general, major components are carbohydrates, amino acids, biodegradable portions of humic substances, polysaccharides, protein-like material, aldehydes, oxoacids, and carboxylic acids. The latter three groups are expected to be important in ozonated waters. Some of the compounds in these groups are important in microbial metabolism and therefore may be expected to impact directly on bacterial growth. Polysaccharides and protein-like material have been shown to be important in fouling of low-pressure membranes (Haberkamp, 2008; Halle´ et al., 2009). There is ample evidence that ozone enhances the biodegradability of NOM. Therefore, increased levels of biological instability (e.g., AOC) will occur in the finished water if there is no biological treatment step following ozonation. However, to date, it has not been possible to develop generally applicable quantitative relationships between BOM levels and bacterial growth in distribution systems, because of the complexity of the processes involved.

3.16.4.11 Ion Exchange (Including MIEXs) Ion-exchange processes can be used for the removal of charged dissolved species from water. With the exception of the emerging use of magnetic ion-exchange (MIEXs) resins discussed below, they are less commonly used in municipal water treatment, and therefore are discussed in less detail. The principle of ion exchange involves exchanging target ions in the water for those on an ion-exchange resin. Resins are commonly in the form of beads, and the process functions as a packed bed reactor. Following exhaustion of the resin, it is regenerated with a concentrated solution containing the ion to be later exchanged into the water during the operating phase. Ion exchangers can be used for the removal of either cations or anions. In the case of the former, the process is most commonly used to remove hardness-causing calcium and magnesium, and is prevalent in home water softening units. Softening is also an important process in the treatment of boiler feedwater. Ion exchangers can also remove anions, for example, in water demineralization for advanced uses such as electronics or boiler feedwater, for the removal of nitrate in groundwater, and also for removal of organic ligands, as discussed below. A large number of commercial ion exchange resins have been developed. They are available with varying degrees of selectivity for specific items, since one of the limitations of the process is that ions other than the target one may also be removed in the process, sometimes preferentially, reducing its effectiveness. The impact of ion exchange on water quality also depends on the ion used to regenerate the resin. For example if an ion exchanger used for water softening is regenerated with sodium

Chemical Basis for Water Technology

chloride, as is commonly the case in home units, the sodium concentration of the water will increase. Regeneration of the resin with acid would increase the hydrogen ion concentration of the water being treated, lowering the pH. In terms of treatment process design, handling of the residues produced by ion exchange must be considered. For the removal of TOC (DOC) or some of its components, ion-exchange processes have also been investigated (e.g., Croue´ et al., 1999) and implemented (e.g., Hongve et al., 1999). The use of granulated iron hydroxide has also been investigated (Teermann and Jekel, 1999) and a review of the use of adsorptive/ion-exchange processes to remove humic substances has been written by Fettig (1999). However, in terms of wider application, the process of this type that is attracting increasing attention for TOC removal in recent years is the MIEXs process, originally developed in Australia. Pilotscale studies are being or have been conducted in a number of countries, and full-scale applications include a large plant in Perth, Australia, that has been operating for several years. In contrast to conventional ion-exchange processes, the MIEXs process is designed to be used in infrastructure that is very similar to that of conventional water-treatment plants. The theoretical basis for the ability of the MIEXs resin to remove DOC is that the resin has strong base functionality, and is therefore able to exchange (remove) weak organic acids (an important component of most DOC) at the neutral pH of most natural waters. The resin is highly selective and therefore can achieve effective removals. The resin is also resistant to physical attrition. Because the resin has a smaller particle size (mean diameter approximately 150 mm) but a comparable specific surface area (surface area per unit volume) compared to conventional ion-exchange resins, the MIEXs resin has considerably more external surface area. This leads to a higher rate of DOC removal. It also reduces fouling of the resin because less DOC is exchanged into the interior of the particles. A key feature of the MIEXs resin is that it contains a magnetic component. This allows the resin beads to agglomerate under the right hydraulic conditions and facilitates their separation from the water so the resin can be recycled.

3.16.4.12 pH Correction pH correction is done to ensure the chemical stability of the water. The chemical stability does not directly affect the quality of water for drinking or other uses; however, water that is not chemically stable leaving a treatment facility can undergo changes in the distribution system that would worsen its quality and cause problems (e.g., precipitation) for some uses. In principle, raw waters entering treatment are chemically stable, because of the generally long time that they have spent in the environment prior to entering a treatment plant. The addition of various chemicals during treatment may disturb this equilibrium, often by lowering the pH; therefore, in some cases it may be necessary to correct (raise) the pH at the end of the treatment process. Although the classical concern is to ensure equilibrium with respect to calcium carbonate, other issues may also be important, for example, the pH of the finished water can have a substantial impact on lead solubility in the distribution system (e.g., Lytle and Schock, 2005). An

443

additional aspect of chemical stability is to ensure that dissolved coagulants (e.g., aluminum) do not precipitate in the distribution system.

3.16.5 The Evolving Nature of Water Treatment The approach and philosophy of water treatment continues to evolve, particularly over the last one or two decades. This is driven by technical changes within the field itself (e.g., the availability of new technologies), an increasing and more sophisticated range of contaminants to be addressed, increased regulatory requirements and in some cases the changing regulatory philosophy, as well as broader issues such as environmental footprint, energy requirements, and climate change. Robustness is being given increased importance in water treatment design. For example, Hrudey and Hrudey (2004) have highlighted the risk associated with changes in raw water quality, in terms of microbial pathogen outbreak events. A robust process can be defined as one that is able to produce excellent water quality under normal conditions, and to deviate minimally from that when challenged (Huck and Coffey, 2004). The potential for more variable weather patterns associated with climate change will place increased importance on treatment process robustness. Risk assist procedures (e.g., QMRA – Quantitative Microbial Risk Assessment; Haas et al., 1999) allow ‘worst case’ scenarios to be explored as part of plant design or upgrading.

3.16.5.1 Increased Emphasis on Physical/Biological Processes Early water-treatment processes such as slow sand and bank filtration relied essentially exclusively on natural physical and biological processes. The first major use of a chemical technology was the introduction of chlorine for disinfection. The addition of metal salts for coagulation and flocculation represented another use of chemical processes. As additional man-made contaminants had to be addressed, chemical processes such as oxidation came more to be used. However, in recent years there has been a desire to move away from a reliance on chemical processes. The origins of this can be traced to the 1970s when the discovery of chlorination by-products (Rook, 1974; Bellar et al., 1974) placed essentially the first upper limit (other than cost) on disinfectant addition. Since that time, there has been a desire to minimize chemical additions because of cost and also because of the production of by-products and/or residuals that need further handling and disposal. In addition, in some cases a philosophical resistance to the addition of chemicals to water has arisen. The consequence of this is that physical and biological processes are now favored. To some extent, this is facilitated by advances in technology such as improved membranes and improvements in apparatus for UV disinfection/inactivation. However, biological processes are also seeing a resurgence, for example, at the time of writing for the last several years in the USA.

444

Chemical Basis for Water Technology

3.16.5.2 The Evolving Role of Membranes One of the most significant developments in water treatment in the last decade has been the much broader introduction of membrane processes. The initial application was of highpressure RO membranes for seawater desalination. However, membranes are now being routinely applied or at least evaluated for a much broader range of applications. Lowpressure membranes (MF and UF) are in many cases replacing conventional granular media filters for particle and pathogen removal. High-pressure (NF) membranes are being implemented for TOC and color removal, can be used for softening and can provide some removals of trace organic contaminants (e.g., Makdissy et al., 2007). This increasing use of membranes is consistent with the greater emphasis on physical processes noted in the previous section. In terms of particle removal, membranes are also more robust than granular media filtration, because they can provide an absolute barrier. However, membrane use is not entirely chemical free, because chemicals are required for cleaning. The use of biofiltration pre-treatment processes to reduce fouling in low-pressure (UF) membranes has however been successfully demonstrated (Halle´ et al., 2009; Huck et al., 2009). The robustness referred to above is being given increased importance in water-treatment design. For example, Hrudey and Hrudey (2004) have highlighted the risk associated with changes in raw water quality, in terms of microbial pathogen outbreak events. A robust process can be defined as one that is able to produce excellent water quality under normal conditions, and to deviate minimally from that when challenged (Huck and Coffey, 2004). The potential for more variable weather patterns associated with climate change will place increased importance on treatment process robustness.

3.16.5.3 Environmental Footprint As with other processes and activities in society, water treatment should seek to minimize its environmental footprint, consistent with providing good treatment. An important component of this is energy consumption, and the most obvious use of energy is in pumping, both the low-lift pumping to the treatment plant, possible pumping within the treatment process, and the high-lift pumping to the distribution system. While these are not directly water quality concerns, steps taken to minimize such energy use may have an impact on treatment (e.g., configuration of processes within the plant). One aspect of reducing energy use is to reduce water consumption, which may also delay required expansion of treatment facilities. Energy is also consumed during treatment to provide mixing, often associated with chemical additions. Therefore, minimization of chemical additions reduces energy requirements, including those associated with the manufacture and transport of the chemicals. Reduced addition of, for example, coagulant chemicals will also minimize solid residuals produced by the process.

3.16.5.4 Coping with Supply Constraints Increased population pressure and climate change will, in some cases, lead to the use of raw waters that may be more

difficult to treat, and to increase the need for water reuse. Seawater desalination will also increase. All of these will lead to more challenging treatment requirements, and will also favor the increased use of membrane processes. One option that may be pursued in some locations is the use of dual distribution systems, where only the water directly consumed is treated to meet all drinking-water requirements. While this approach can reduce treatment costs, it has of course its own associated health risks, which must be carefully evaluated and minimized. At a minimum, all water should be treated to be microbiologically safe. One option would be to partially treat all water centrally, and then provide a higher level of treatment for the water to be consumed, closer to the actual point of consumption, potentially at the neighborhood level. This approach, in addition to leading to a more robust system, helps to minimize problems of water quality deterioration in distribution systems. Although its costs would have to be evaluated carefully for a given situation, it is conceptually increasingly feasible because of the greater use of treatment processes such as membranes and UV disinfection that can be more amenable to automation and remote monitoring and control.

3.16.6 Addressing the Treatment Goals – From the Perspective of the Chemical, Physical, and Biological Processes Involved This section describes how the processes described in Section 3.16.4 can be used to address the various treatment goals identified in Section 3.16.2. The goal ensuring esthetic quality is addressed as part of the removal of chemical contaminants, because esthetic quality is often largely determined by odorous compounds that can be removed in treatment. A small section on maintaining water quality from the treatment facility to the consumer’s tap is also included. Because of the large area that has to be covered, this section can only provide general information on the processes that can be used to address specific goals. The focus is on the practical achievement of these goals, and where possible comprehensive review papers are cited that can provide much more specific detail. In this section, more detail is provided for some less common processes. The subsection on the removal of odorous compounds is quite detailed, both because there is less comprehensive review information in the literature, and also because the detailed description allows a number of important treatment factors and subtleties to be addressed, that are also relevant for the removal of other contaminants. It is said in water treatment that ‘‘every water is different’’ and thus resources such as the present chapter can fill an important role in suggesting processes that can be investigated on site, ideally at pilot scale, to define or improve the treatment process for given water.

3.16.6.1 Particle Removal (Including Pathogens) 3.16.6.1.1 Coagulation, flocculation, and sedimentation As indicated in Section 3.16.4, coagulation and flocculation are used as preparatory processes for solid separation. The final solids separation step will be either granular media

Chemical Basis for Water Technology

filtration or low-pressure membrane filtration (MF or UF). Except for raw waters having low suspended particle concentrations, sedimentation will normally be inserted prior to the final particle-separation step, to reduce the load on that step. Because of the preparatory nature of coagulation, flocculation, and sedimentation, specific applications are not described separately, but rather are discussed where appropriate in relation to other processes in the following sections. Flotation, as a newer and less widely practiced alternative to sedimentation, is discussed in the following section.

3.16.6.1.2 Flotation Although the same fundamental principles of coagulation discussed earlier apply, a small low-density floc is appropriate for DAF. Thus, optimal coagulation and flocculation conditions for DAF are likely to be different than those for sedimentation. Often flocculation time will be reduced, compared to those prior to sedimentation, and a higher flocculation energy may be advantageous. Polymers are rarely used (except possibly for the newer high-rate DAF processes) and the coagulant dose may be lower than for sedimentation. The coagulant dose may also need to be seasonally optimized, because of temperature and other factors. The rise velocity of a bubble depends upon its size and water temperature (i.e., viscosity). Because flow conditions should be laminar, it is possible to calculate the rise velocity of a bubble based on Stoke’s law. However, in practice, operating conditions such as hydraulic loading are determined based on optimization within known feasible ranges. The air loading (grams of air per cubic meter of influent water) is a function of the amount of air introduced into the recycle flow. The mass of air in the pressurized flow can be calculated using Henry’s law. A certain minimum air loading is required for successful operation. Above this minimum, an increase in air loading within a certain range will improve performance (i.e., reduce effluent turbidity). Beyond this point, further increases in air loading do not improve performance. The design basis of DAF is the hydraulic loading rate. Newer high-rate processes are capable of loadings in the range of 40 m h1 (MWH, 2005), compared to more traditional loadings of 10–20 m h1. Because of the number of factors that affect performance and because the fundamental mechanisms for floc-bubble attachment are not well understood, testing of DAF is required to establish its suitability in a given situation. Bench scale batch testing can be used to provide an initial indication of process feasibility and to aid in determining optimum coagulant and coagulant dosages. However, pilot-scale testing is required for the optimization of the various factors that affect process performance, and for reliable determination of process costs. DAF must be followed by a final particle-separation step. Normally, this would be granular media filtration, but it may also be a low-pressure membrane (MF or UF). Therefore, although the effluent turbidity and particle counts of the DAF effluent are very important, the optimum DAF conditions are those that will lead to optimum filtration or membrane performance. In addition to providing details regarding the design of DAF, MWH (2005) summarizes the advantages and disadvantages of

445

DAF in comparison with other clarification technologies (e.g., sedimentation). As indicated earlier, DAF is generally not a good process where higher raw water turbidities are experienced. Valade et al. (2009) also provide guidance regarding the selection of DAF versus sedimentation and direct filtration. They recommend DAF for waters of relatively high quality: river sources (i.e., mineral turbidity) with average turbidity below 10 NTU (NTU, number of transfer unit) or reservoir sources (i.e., nonmineral turbidity) with average turbidity below 100 NTU. They place no upper limit on organic content and note that maximum turbidities in both types of waters can be higher. In a comprehensive review of DAF, Edzwald (2010) notes that DAF is especially effective treating water from reservoirs, raw waters containing algae, color or NOM, and waters with low mineral turbidity. He also notes that DAF is more efficient than sedimentation in removing Giardia cysts and Cryptosporidium oocysts.

3.16.6.1.3 Filtration As discussed in Section 3.16.4, rapid filtration, whose goal is particle removal, can provide some physical removal of pathogens, especially the cysts and oocysts of Giardia and Cryptosporidium, respectively. Filtration is a mature technology and the general range of operating conditions and expected performance are well established in the literature and in textbooks. In a number of jurisdictions, filtration effluent quality is specified in order to allow the filtration process to receive removal credit for organisms such as Giardia and Cryptosporidium. Filter effluent quality is typically specified in terms of turbidity, and more recently in some situations in terms of particle counts. Optimum design and operation of a filtration process typically requires site-specific investigations. Among the various investigations that have quantified removals of Giardia and Cryptosporidium is an investigation by Nieminski and Ongerth (1995) that examined both conventional and direct filtration, at both pilot and full scale. In a review of water-treatment processes for the removal of Giardia and Cryptosporidium, Betancourt and Rose (2004) summarize removals obtained by filtration in various studies. They note that these studies affirm the importance of proper coagulation for cyst and oocyst removal through all stages of conventional treatment. In a comprehensive pilot-scale investigation, Huck et al. (2002) observed a substantial decrease in removal at the end of the filter cycle, before filter effluent turbidity had substantially increased. Weiss et al. (2005) indicated that riverbank filtration had the potential to provide substantial reductions in concentrations of various microorganisms, relative to levels in raw water. Their investigations involved more than a year of monitoring at three full-scale riverbank filtration facilities in the US. They note that accurate removals for Giardia and Cryptosporidium could not be directly determined because of the low and variable levels of these pathogens in the raw waters.

3.16.6.1.4 Membranes With low-pressure membranes (MF and UF) replacing granular media filtration, one of their functions is of course to provide physical removal of microbial pathogens. A number of jurisdictions allow specified log removals to be claimed for

446

Chemical Basis for Water Technology

these processes; however, monitoring and maintaining membrane integrity is critical for maintaining these removals. A well-cited study by Jacangelo et al. (1995a) evaluated the removal of protozoan cysts and oocysts and a model virus by six commercially available membranes (three MF and three UF). Three different source waters were used, and both benchand pilot-scale investigations were conducted. As long as the membranes remained intact, no cysts or oocysts were detected in the permeate, and physical straining appeared to be the primary removal mechanism. The extent of virus removal was membrane specific, ranging from o0.5 log to 46 logs. Phenomena contributing to removal appeared to be physical sieving or adsorption, cake layer formation, and the fouling state of the membrane.

3.16.6.2 TOC Removal The various processes available for TOC removal are discussed briefly in the following subsections. Their applicability in a particular water-treatment situation will depend on sitespecific circumstances, including the nature (i.e., composition) of the TOC. The optimum process for a particular water will also depend on the other treatment objectives being addressed, to which some or all of these processes may also contribute.

3.16.6.2.1 Enhanced coagulation In the US, enhanced coagulation requirements depend on raw water TOC concentration and alkalinity (e.g., Edzwald and Tobiason, 1999). Briefly, the required TOC percentage removals increase with increasing raw water TOC concentration and decrease with increasing alkalinity. The latter requirement is related to economic considerations. Raw waters having a TOC of 2 mg l1 or less are not required to practice enhanced coagulation. Edwards (1997) developed a model for predicting the DOC concentration remaining in the water after enhanced coagulation. Model inputs are coagulant dose, coagulation pH, and raw water UV254 and DOC. Vrijenhoek et al. (1998) investigated the removal of trihalomethane (THM) precursors and suspended particles from two surface waters by enhanced coagulation at pilot scale. They reported optimal removal of THM precursors at pH 5.5 and postulated that at this pH, humic substances were probably removed by formation of insoluble aluminum–humate complexes at a low alum dosage and adsorption to aluminum hydroxide precipitates at a high alum dosage. They noted that enhanced coagulation does not benefit raw waters with low specific UV absorbance (SUVA) values, but also noted that such waters may not form elevated concentrations of disinfection by-products (DBPs) because of the low humic fraction in the NOM. Yan et al. (2008) investigated enhanced coagulation of three well-characterized typical source waters in China with PACls. They found optimum NOM removal at pH 5.5–6.5 for all PACls, and recommended that basicity, speciation, and dosage of the coagulant should be optimized, based on the raw water alkalinity.

3.16.6.2.2 MIEXs The various process components required for an MIEXs installation are described by Slunjski et al. (2000). The first step involves contacting the water with the resin over a 10–30 min

detention time to allow the DOC to be exchanged on to the resin. Only a low-energy input is required to maintain the resin in suspension, because the magnetic attraction of the resin beads occurs over only a very short distance. Following this step, the resin–water suspension flows by gravity to the resin separation stage. The inlet of this vessel is designed to facilitate inter-particle collisions resulting in agglomeration of the magnetic resin beads. The agglomerates have a high settling velocity (Lange et al. (2001) report 25 m h1). This allows a high-upflow water velocity in the settling vessel, which prevents turbidity accumulation in the system. The settled resin is pumped back to the contractor for another DOC loading cycle while the treated water continues onward to the next process step. A small amount of used resin is continuously removed from the recycle line for regeneration and replaced with regenerated resin. The used resin is regenerated in a batch mode using a sodium chloride solution. The regenerant is reused a number of times although the chloride concentration of the regenerant must be restored to the original level prior to the next regeneration cycle (Lange et al., 2001). Slunjski et al. (2000) report that small additions of sodium hydroxide to the regenerant may also be beneficial for some waters and that periodic acid washes may be required when it is necessary to remove metal precipitates from the resin. The MIEXs process does not require pretreatment of the water (Slunjski et al., 2000). However in some cases, it might be advantageous to combine the process with coagulation for increased DOC removal. In testing for one treatment plant, the MIEXs process was shown to preferentially remove the lowermolecular-weight component of the DOC while alum coagulation removed the larger-molecular-weight DOC components (Lange et al., 2001). Those authors noted that MIEXs and coagulation would complement each other, regardless of the order of the two processes. In pilot testing for this plant, MIEXs (prior to coagulation) was shown to reduce the raw-water DOC (approximately 10 mg l1) by about 60%. Placing MIEXs ahead of coagulation would reduce the coagulant dose (and associated sludge production) needed for DOC removal and also reduce the need for pH correction following coagulation. Wert et al. (2005) evaluated MIEXs at pilot scale to remove DOC and bromide, and to assess the process as a pretreatment for ozonation. In this study treating Lake Mead water in the US, up to 30% DOC removal was obtained, leading to a measurable decrease in ozone decay rates. This led to a reduction in the transferred ozone dosages required for Cryptosporidium inactivation of 15–25%, and reduced bromate formation by 35%. In a bench-scale investigation, Boyer and Singer (2005) compared MIEXs with enhanced coagulation for the removal of DBP precursors and bromide. They studied raw waters from four drinking water facilities in California, having a range of raw water characteristics. They reported that MIEXs was more effective than coagulation in removing UV-absorbing substances and DOC. Using raw waters in the UK in a bench-scale investigation, Mergen et al. (2006) reported a substantial reduction in coagulant dose when coagulation was preceded by MIEXs. In pilot-scale investigations at four US locations, Singer et al. (2007) also examined the effectiveness of MIEXs for the

Chemical Basis for Water Technology

removal of DBP precursors, and found that it was generally more effective in this regard than enhanced coagulation. MIEXs removed substantial amounts of DOC and UVabsorbing substances, except in water that had a high concentration of total dissolved solids and a low SUVA value. Hydrophobic and transphilic DOC fractions were removed more effectively by MIEXs than was the hydrophilic fraction. The authors also noted that DBP speciation after MIEXs tended to shift from the fully chlorinated THM and haloacetic acid (HAA) components to their more brominated counterparts because bromide was less well removed than DOC. In further bench-scale investigations of three of the waters with quite different NOM character studied above (Mergen et al., 2006), Mergen et al. (2008) determined that the hydrophobicity of the DOC has an important effect on its removal using MIEXs. Although the water containing NOM with high hydrophobicity showed good DOC removal in the first cycle of contact with the resin, removal decreased substantially with subsequent use of the resin. This decrease was attributed to the blocking of resin sites by higher-molecularweight NOM present in the hydrophobic sample. For the water with more hydrophilic NOM, the DOC removal remained more consistent with subsequent use of the resin. In the water containing algogenic NOM, DOC was poorly removed and this was attributed to the greater presence of uncharged organics, considered likely to be carbohydrates and proteins. Son et al. (2005) conducted bench-scale investigations of MIEXs as a pretreatment for UF or MF membrane filtration. In considering MIEXs for a given application, total lifecycle costs should be determined and compared to other alternatives such as enhanced coagulation, BAC, or potentially NF. Costs would need to include residuals handling and disposal. Slunjski et al. (2000) presented cost estimates that showed that, for the Wanneroo plant in Perth, Australia, MIEXs was cheaper than either GAC adsorption or ozone combined with BAC. In summary, the MIEXs ion-exchange process represents a feasible technology for DOC removal. Although it does not have as long and extensive a full-scale operating history as alternative processes such as enhanced coagulation, it is beginning to be more widely applied at full scale. For a particular water where DOC removal is required, pilot testing is required for a reliable technical and cost comparison of MIEXs to other alternatives.

447

In terms of BOM removal, water temperature was extremely important if it remained below 5 1C for a significant period of time. The removal of easily biodegradable substances such as ozonation by-products was significantly impaired at these lower temperatures. Above 5 1C, temperature had some influence, but not a major effect, on performance. The shortest contact times (on the order of a few minutes) were required for the removal of easily biodegradable ozonation by-products. Somewhat longer contact times were typically necessary for the removal of parameters measuring the biological stability of the water (e.g., AOC and BDOC). The longest contact times are generally required for the removal of chlorination by-product precursors and chlorine demand. Acceptable removals of the easily biodegradable compounds, and perhaps AOC or BDOC, may be achievable within the common range of contact times used for conventional particle-removal filtration. At temperatures above about 10 1C, Huck et al. (2000) found that the choice of filtration media (anthracite vs. (exhausted) GAC) had no measurable impact on BOM removal. At lower temperatures, GAC performed better than anthracite. GAC filters also recovered more quickly from process perturbations. The difference between GAC and anthracite was also greater for individual BOM components such as oxalate than for overall parameters such as BDOC. A previously developed non-steady-state biofiltration model was refined in the project and could successfully predict trends seen in the experimental work. One of the most important practical predictions from the model was that less readily biodegradable substances were better removed in the presence of easily biodegradable compounds. This supports the common observation (mentioned above) that ozonation prior to a biofilter improves biological filtration performance by creating more easily biodegradable substances. In more recent work treating a challenging river water, Halle´ et al. (2009) demonstrated the ability of rapid biofiltration (without prior coagulation or ozonation) to measurably reduce both hydraulically reversible and irreversible fouling of UF membranes. This was due to the ability of the biofilters to reduce the concentrations of biopolymers (polysaccharides and protein-like substances) in the raw water. It is thus evident that biological filtration can be used to provide at least partial removal of TOC in water treatment. Its applicability in a given situation will depend on the specific treatment objectives.

3.16.6.2.3 Biological treatment

3.16.6.2.4 Adsorption

In examining the removal of humic substances by biological filtration, Huck (1999) noted that investigations reported in the literature have shown only very limited biodegradation of unozonated humic substances under water-treatment conditions. However, several investigations had demonstrated substantial removals of humic substances by biofiltration following ozonation. By cleaving and modifying humic molecules, ozonation creates product molecules with increased biodegradability and diffusivity. A later experimental investigation of single-stage biological filtration (i.e., particle and BOM removal in the same filtration step) was conducted by Huck et al. (2000), largely at full scale.

In principle, adsorption on GAC could be used for TOC removal. For example, Jacangelo et al. (1995b) compared GAC to membrane processes (NF) and enhanced coagulation for the removal of NOM. In a qualitative summary they rated GAC adsorption (with regeneration) as providing very good NOM removal and as being generally intermediate in cost between enhanced coagulation and NF. In general, because of the relatively short time to exhaustion of GAC for TOC removal (on the order of months), adsorption on GAC would not be a preferred option for TOC removal, unless other processes could not be used, or unless GAC was also addressing other treatment objectives.

448

Chemical Basis for Water Technology

3.16.6.2.5 Oxidation In addition to removing specific organic contaminants, oxidation can, in principle, also be used to reduce the concentration of TOC in water, and several example investigations are therefore cited. Carr and Baird (2000) examined mineralization of TOC at bench scale using ozone and ozone peroxide. They tested prepared standards, well waters from a groundwater recharge basin, and secondary and tertiary effluent from a wastewater reclamation plant. They found that the use of ozone alone or in combination with hydrogen peroxide may have practical limits for reducing concentrations of TOC. Thomson et al. (2004) used UV (low-pressure mercury vapor lamps) alone and in combination with hydrogen peroxide to investigate NOM removal from the highly colored surface water, high in TOC. They combined this process with biological treatment. They also found that although their preliminary study indicated that NOM could be removed, the UV doses used were thousands of times greater than those used for disinfection. It would therefore seem that, from a practical perspective, the use of oxidation processes for TOC removal may be restricted to specific cases where other processes cannot be used.

3.16.6.2.6 Nanofiltration Because of their pore size, NF membranes can remove at least a portion of the TOC. In a given situation, an important factor governing the extent of removal will be the pore-size distribution of the particular membrane being used. In terms of specific applications, Eriksson (1988) has cited the use of NF for the removal of color and TOC from surface water in Florida. In an article mentioned previously, Jacangelo et al. (1995b) compared NF to GAC adsorption and enhanced coagulation. In their qualitative comparison they rated NF as providing the best NOM removals, but also likely being the most expensive of the three processes. (They do note that, at the time of their article, costs for NF had been decreasing more rapidly than those for other processes and of course costs for NF have further decreased subsequently.) They note that because NF is not highly complex with regard to operation and maintenance, it is particularly attractive for small systems. Processes involving membranes (essentially NF pore size 1.5–5 nm) have been in use for TOC removal in small systems for several decades in Norway (Ødegaard et al., 1999).

3.16.6.3 Disinfection/Inactivation As indicated in Section 3.16.4, addressing the disinfection/ inactivation goal is arguably the most important in water treatment, in terms of minimizing acute health risk. Specific applications of processes to achieve this goal are not addressed herein, because they are well described in various standard reference works. In some jurisdictions, disinfection conditions, such as dose and contact time, are prescribed by regulation. As noted earlier, the major chemical disinfectants (chlorine, ozone, and potentially chlorine dioxide) also function as oxidants, as can UV. Together with the addition of substances such as hydrogen peroxide to promote the generation of hydroxyl radicals, ozone and UV can also function as AOPs. For this reason, as indicated in Section 3.16.4, process decisions with regard to disinfection should take into

consideration the need for oxidation in the process train to achieve other treatment goals. In addition to Section 3.16.6.2.5 on the use of oxidation to reduce concentrations of TOC, a detailed discussion is provided in Section 3.16.6.5 on the use of oxidation to reduce concentrations of odorous compounds.

3.16.6.4 Maximizing Biological Stability van der Kooij (2000) has provided a comprehensive discussion of biological stability. As indicated in Section 3.16.4.10, organic carbon is normally the limiting nutrient in drinking water, and maximizing biological stability therefore normally involves minimizing the concentration of BOM in the treated water. The necessary target BOM level may be sitespecific, and will be lower if no disinfectant residual is maintained in the distribution system. Because of their high surface area, filters make the best bioreactors to reduce BOM levels during treatment. Effective biofiltration can be obtained in first-stage filters that also provide particle removal (Huck et al., 2000) and in secondstage GAC filters (e.g., Pre´vost et al., 2005). Urfer et al. (1997) have critically reviewed the various factors affecting biofiltration. These include contact time, surface area, media, temperature, nature and concentration of the influent BOM, and biofilm disruption (due to backwashing). Another important concept (Huck et al., 2000) is whether biofiltration is managed or operated in the background. The former implies maximizing the performance of biofiltration without compromising other treatment objectives, whereas the latter implies focusing on other treatment objectives and accepting the degree of biological treatment that the filters thus operated are able to deliver. In general, biological filtration has classically been considered for four treatment objectives: the removal of easily biodegradable substances created during treatment (normally ozonation by-products) or present in the raw water, the removal of the biodegradable fraction of chlorination by-product precursors, the enhancement of biostability, and the removal of certain types of odorous compounds. More recently, biofiltration has been shown effective for the removal of specific trace contaminants (Halle´, 2010) and for the reduction of organic fouling in low-pressure membranes (Halle´ et al., 2009). Although contact time is a crucial variable, Zhang and Huck (1996) and Huck (1999) have shown the importance of considering a new parameter referred to as dimensionless contact time, which also includes reactor surface area and substrate parameters. It is, of course, important that disinfectant residuals not be present in biological filters.

3.16.6.5 Removal of Organic Chemical Contaminants As indicated in Figure 2, a number of different processes can be used to remove (trace) organic contaminants. The key to successful process selection and design is to take advantage of the most appropriate property of the substance to be removed. For example, the volatile contaminants in groundwater lend themselves to removal by some type of air stripping process, where they are transferred to the gas phase.

Chemical Basis for Water Technology

The processes for the removal of trace organic contaminants are basically oxidation, volatilization, adsorption, and membranes. Oxidation may, of course, be either chemical or biological. In a particular situation, more than one process may contribute to the removal of a specific contaminant. As will be evident, background TOC plays an important role in processes used for trace contaminant removal. The following section deals with the removal of the odorous compounds, geosmin and MIB. It is written in considerable detail, allowing us to address many important issues regarding the removal of these contaminants. As mentioned previously, it also illustrates the complexities and subtleties involved, many of which are directly relevant for the removal of other types of organic contaminants.

3.16.6.5.1 Geosmin and MIB Introduction. Problems of taste and odor continue to be perhaps the most difficult faced by drinking-water providers, and successful resolution of these problems is extremely important for consumer acceptance of the finished water. The complexity of taste and odor problems is due to the many possible sources of taste and odor in drinking water. Biological sources, such as cyanobacteria and other microorganisms, can produce a number of known and as-yet unidentified odor– causing compounds. Some of these have odor thresholds in the low nanogram per liter level. Anthropogenic or man– made chemicals can also cause taste and odor when they are discharged into drinking-water sources either in municipal/ industrial effluents or during spill events. Tastes and odors can also be created during treatment itself, in particular by the action of chlorine on some substances in the raw water. Tastes and odors can also arise in the distribution system as a result of microbiological activity, the reaction of disinfectants with organic matter, emissions from pipes and reservoir coatings, and possibly the diffusion of pollutants through plastic pipes. Home plumbing can also be a source of taste and odor. Temperature is an important factor in taste and odor episodes. Not only do higher temperatures lead to greater biological growth in distribution systems, but they generally lead to greater consumer perception of tastes and odors. Because of the complex nature of taste and odor, the successful resolution of a problem requires that the odor be properly characterized. Classifications proposed for tastes and odors are useful in this regard, as they give insights into the possible types of taste- or odor-causing compounds. Taste and odor can be analyzed both by sensory methods (i.e., using either the human nose or mouth as the detector) and by chemical techniques. In recent decades, the water industry has adapted the flavor profile analysis (FPA) method from the food and beverage industry. This is superior to the previously used threshold odor number (TON) technique, in that it provides both a characterization and an approximate quantitation of the taste or odor. Chemical analysis is however necessary for the identification of specific odorous compounds. The chromatographic sniffing technique provides a means of combining sensory and instrumental methods to identify individual odor components. Although the first line of defense against taste and odor problems is a high-quality source water, this represents an

449

unattainable ideal for many water utilities. Mitigation for taste and odor therefore consists of attempting to prevent formation of the offending substances in the raw water, typically by cyanobacteria and potentially algae control, removing these substances in the treatment plant, usually by adsorption, or transforming them during treatment by oxidation or by microbial activity (biodegradation) into less offensive substances. The complexity of odor treatment is due to the fact that the overall odor of a water may be composed of a number of odorous compounds of various odor types and intensities. The most information available in the literature regarding treatment is for naturally occurring taste and odor compounds, especially the most commonly occurring ones, geosmin and MIB. Although this section therefore focuses on these compounds, many of the considerations presented are also valid for other odorous compounds, and indeed for nonodorous trace contaminants. Aeration can be used for the removal of volatile compounds, especially the rotten egg odor caused by hydrogen sulfide. AWWARF-LE (1987) reported that air stripping is effective for compounds with a Henry’s law constant greater than 103 m3 atm mol–1. It is not effective for geosmin and MIB, which have Henry’s law constants approximately two orders of magnitude lower. Conventional processes (e.g., coagulation, flocculation, sedimentation, and granular media filtration) have very limited usefulness against taste and odor. At the doses typically used for disinfection, chemical disinfectants such as chlorine, chlorine dioxide, and ozone may achieve at least partial success against some types of odors. Membrane processes on their own, especially the low-pressure membrane processes MF and UF, are not effective, because the odorous molecules are small enough to pass through the membrane. Some removals may be achieved with NF membranes. If there is no predisinfection, coagulation/flocculation/ sedimentation can remove organisms such as cyanobacteria and actinomycetes before they can be lysed by disinfection and release odorous substances. Ashitani et al. (1988) reported that both geosmin and MIB were present in raw water in dissolved form and in suspended form associated with the cyanobacteria (blue–green algae) from which they originated. The geosmin and MIB in suspended form were well removed by coagulation and sedimentation alone. However, breakpoint prechlorination caused leakage of geosmin and MIB from the host cells into the water, as also reported for geosmin by Gammie (1987). Both geosmin and MIB were decomposed by sunlight in the presence of free residual chlorine. Decomposition for the conditions tested was on the order of 50%. The three processes most effective against taste and odor are oxidation, adsorption on activated carbon, and biodegradation – these are discussed in detail in the following. Except possibly for the addition of PAC, the installation of a process for taste and odor control can be relatively expensive. However, oxidation, adsorption, and biofiltration are also able to achieve other treatment objectives, and this may be important in assessing overall cost-effectiveness. It should be noted that if geosmin and MIB are present, toxic cyanobacterial metabolites may also be present. Consideration of

450

Chemical Basis for Water Technology

these is outside the scope of this section; however, treatment may be needed to deal with them as well, and some of the same processes discussed herein would apply. Oxidation. Oxidation is one of the key technologies that can be used for odor removal. As described earlier (Section 3.16.4) basically all of the common oxidizing substances and methods (i.e., UV) that can be used for disinfection can also be used for oxidation. Thus when taste and odor must be addressed as a treatment objective, this may influence the choice of disinfection method employed. Our understanding with respect to the use of oxidation processes for taste and odor control over the last 15–20 years has developed from the suspicion that the hydroxyl radical was important, to certainty that it is, and to the determination of kinetic parameters for hydroxyl radical oxidation, the socalled AOPs. Comparison of various oxidants. In laboratory studies, Glaze et al. (1990) evaluated a number of oxidants for the removal of six model taste and odor compounds (including geosmin and MIB) spiked into Colorado River water. In addition to conventional oxidants, the authors evaluated three AOPs: ozone with hydrogen peroxide, ozone with UV light, and hydrogen peroxide with UV light. (This was some of the first reported work for these AOPs.) The authors concluded that conventional oxidants (chlorine, chloramines, chlorine dioxide, potassium permanganate, and hydrogen peroxide) are unable to control taste and odor problems due to geosmin and MIB. AOPs were more successful, but the authors noted that they must be evaluated for factors such as cost and effectiveness. They found that ozone was able to oxidize geosmin and MIB without the addition of either hydrogen peroxide or UV radiation and attributed this effect to the fact that some constituents in natural waters were able to react with ozone to form highly reactive radicals, presumably the hydroxyl radical. Following on the work of Glaze et al. (1990), Ferguson et al. (1990) showed at pilot scale that PEROXONE (ozone in combination with hydrogen peroxide) required a significantly lower applied ozone dosage than ozone alone, to oxidize geosmin and MIB. Dietrich et al. (1995) examined the oxidation of six odorous and nonodorous algal metabolites by chlorine, chlorine dioxide, and permanganate. In some cases, the oxidants had little effect on the odor characteristic of the compounds (e.g., b-cyclocitral and phenethyl alcohol). In other cases, the oxidants created new odors from previously odor-free compounds (e.g., palmitic and linoleic acids), and in yet other cases the oxidants decreased, eliminated, or changed odors from odorous metabolites (e.g., linolenic acid (which produces a watermelon odor) and 2 t,6 c-nonadienal). There were many similarities in results obtained with the three oxidants. This research confirmed that elimination of one odor by oxidation can result in the formation of others. AWWARF-LE (1995: ch. 3) reported that, in comparison with chlorine, chloramination generally produces lower concentrations of odorous by-products such as iodoforms, chlorophenols, and aldehydes. However, chlorination is more efficient at removing a number of odors associated with anaerobic conditions, that may be described as septic, decaying

vegetation, swampy, and fishy. Chlorination is also more effective than chloramination at stopping the formation of medicinal iodoform. The use of potassium permanganate for taste and odor control appears to be specific to particular substances (AWWARF-LE, 1995). The authors note that in documented case studies involving KMnO4 and other treatments simultaneously, there has not been a careful evaluation to determine the controlling mechanism or mechanisms for odor removal. The authors also note that the effect of the background organic carbon matrix on taste and odor control by KMnO4 is also not clear. In evaluating the removal of odorous compounds by various oxidants, Jung et al. (2004) concluded that higher doses of ozone might be sufficient to control geosmin and MIB to below the odor threshold, which they identified as being 30 ng l1 for both compounds (i.e., higher than usual). The effect of several oxidants on MIB concentrations in the presence of cyanobacteria was investigated by Tung et al. (2004). Raw water samples were incubated to allow for the development of high concentrations of MIB (1000– 2000 ng l1), with as much as 70% of the MIB being contained within the algae cells. (Most of the algae were in fact cyanobacteria.) Under the conditions tested, ozone was much more effective than chlorine or potassium permanganate; however, all oxidants caused cell damage and release of intracellular MIB. Oestman et al. (2004) examined the ability of chlorine and chloramines to mask geosmin and MIB using two sensory analysis approaches – a statistical pairwise comparison test and FPA. The overall conclusion from this work was that neither chlorine nor chloramines are likely to be effective for reducing or masking other odors (such as geosmin and MIB) in drinking water. Bruchet and Duguet (2004) have provided a good summary on the role of various oxidants in removing, masking, and generating tastes and odors. They note that the main application of chlorine or chloramines as oxidants for removing odors that are either present in the raw water or that may develop in the distribution system is for the elimination of sulfide odors or related odors caused by anaerobic conditions. Although one or both oxidants can also remove several other odors, both are completely ineffective in removing most of the other odorous substances, including geosmin and MIB. The authors also note that chlorinous oxidants themselves impart a chlorinous odor at rather low concentrations. In contrast to the work by Oestman et al. (2004) reported previously, Bruchet and Duguet (2004) state that chlorine can mask other odors. They specifically cite the reduction in the musty taste caused by MIB as the concentration of free chlorine increased. They note that an important implication of this is that when the chlorine concentration decreases in the distribution system, other odors that were previously masked may reappear or increase in intensity. Bruchet and Duguet (2004) state that, at the time of writing of their article, the most efficient process for dealing with taste and odor problems was oxidation (ozone alone or in combination with hydrogen peroxide) in combination with adsorption on activated carbon. (It should be noted that recent work involving the combination of UV and hydrogen peroxide indicates that this process can be a viable alternative

Chemical Basis for Water Technology

to AOPs involving ozone.) In terms of the final disinfectant, Bruchet and Duguet (2004) note that it sometimes may be necessary to switch from chlorine to chlorine dioxide to minimize chlorinous or medicinal odors when phenols or iodides are present in the raw water. They also indicate that an AOP is highly desirable for dealing with taste and odor problems that may be caused by chemical spills in the source water. Finally, they also note that the best treatment processes for taste and odor removal are also the most efficient in dealing with cyanobacterial toxins, should those be present. Oxidation involving ozone. As noted previously (Section 3.16.4.7), ozone oxidation can occur either via molecular ozone or via hydroxyl radicals. This section includes studies where part of the oxidation may be due to the presence of hydroxyl radicals, even though their generation was not necessarily a deliberate objective. AWWARF-LE (1995) noted that additional work is needed to assess the health impacts of ozonation by-products, identify other by-products that create tastes and orders in treated water, and identify intermediate by-products that may be formed during the oxidation of taste and odor compounds. It is also noted that additional research is needed to identify intermediate by-products formed by the use of an AOP, as well as their potential health effects. Morioka et al. (1993) approximated the removal of geosmin or MIB by ozonation as a first-order reaction. Decomposition rates in natural waters were primarily influenced by the concentrations of carbonate ion and humic substances, attributable to scavenging effects on hydroxyl radicals. In synthetic waters, low concentrations of humic substances appeared to act as promoters. Ho et al. (2004) examined the effect of water quality and also the character of NOM on the ozonation of geosmin and MIB, using synthetic waters containing two hydrophobic NOM fractions isolated from two French surface waters and sodium bicarbonate. The NOM fraction having higher SUVA values (i.e., greater aromatic content) and higher average molecular weight showed higher ozone decomposition rates and higher apparent rate constants for the degradation of geosmin and MIB. This was attributed to the greater production of hydroxyl radicals. The authors also reported that the addition of bicarbonate stabilized the concentration of the concentration of ozone, thus reducing the removal of geosmin and MIB. The authors state that these results support the hypothesis that the destruction of geosmin and MIB is primarily through the OH radical since a lower ozone decomposition leads to less production of hydroxyl radicals. They noted that MIB was more resistant to ozonation than geosmin. An important implication of the work by Ho et al. (2004) is that for waters low in alkalinity, the advantage provided by an AOP compared to ozonation alone would be expected to be less than for waters higher in alkalinity. Because ozone decay is more rapid when radical scavenging occurs, in evaluating a given water, it may be possible to use the measured ozone decay rate as an indicator of the need for an AOP, compared to ozonation alone. Nerenberg et al. (2000) present preliminary full-scale ozonation/biofiltration results from a treatment plant on Lake Michigan, USA, whose source water contained musty/earthy odors. For MIB concentrations in the raw water ranging up to

451

about 40 ng l1, removals due to ozonation were between approximately 35% and 65%. As has been mentioned by others (e.g., Hrudey et al., 1995), AWWARF-LE (1995) note that ozone has the possibility to produce odors, sometimes characterized as fruity, attributable to aldehydes. It would be reasonable to expect that odors would be produced where the original odor had a complex character. Oxidation involving UV. The increasing use of UV for disinfection in drinking-water treatment in recent years has given an additional impetus to the evaluation of UV (alone or in combination with H2O2) for odor removal. Most of the research has focused on the traditional taste and odor compounds, geosmin and MIB. Rosenfeldt et al. (2005) reported on bench-scale investigations using UV and UV/H2O2 to remove geosmin and MIB from a surface water (Lake Michigan in USA). At a fluence of 1000 mJ cm2, low-pressure and medium-pressure direct UV photolysis (i.e., without hydrogen peroxide addition) removed 10% and 25–50% of the compounds, respectively. At the same fluence, the addition of hydrogen peroxide resulted in more than 70% removal. For MIB oxidation in the presence of H2O2, medium-pressure UV provided consistently faster oxidation than low-pressure UV. The authors also calculated the electrical energy per order (EEO), which is defined as the electrical energy (in kilowatt hours) required to decrease the concentration of a target compound by one order of magnitude in 1000 US gallons (3785 l) of water. Although this is a very useful value for process comparison, the authors note that it is very specific with respect to reactor geometry, UV lamp, the specific contaminant being investigated, and background water quality. They note that their calculated EEO values of less than 5 kWh for 90% oxidation of geosmin or MIB (with the addition of hydrogen peroxide) are lower than previously reported by others, and they attribute this difference to the important role played by background water quality in determining EEO. Although costs based on EEO values reflect only the cost of the electricity needed to run the UV lamps, this does give a general indication of process feasibility. It should be noted that the fluences for which the authors cited various percentage removals (1000 mJ cm2) are a factor of 25 larger than a value of 40 mJ cm2 frequently used for disinfection. At least one manufacturer of full-scale UV disinfection equipment offers a reactor system where additional UV lamps can be turned on and hydrogen peroxide can be fed when an odor event occurs in the raw water (see Trojan Technologies website). Gray (2006) also examined the removal of various odor compounds by UV photolysis, and by UV in combination with H2O2, in bench-scale experiments. In addition to geosmin and MIB, he also investigated 2-isopropyl-3-methoxypirazine (IPMP), two-isobutyl-3-methoxypirazine (IBMP), as well as four unsaturated aldehydes (E2-heptenal, E2,E4-nonadienal, E2,E4-decadienal, and E2,E4-heptadienal). Significant removal of all compounds was obtained with hydrogen peroxide doses as low as 1.5 mg l1 and a UV fluence greater than 500 mJ cm2 for a given water quality, and UV/H2O2 was able to provide up to twice the removal of geosmin, MIB, IPMP, and IBMP than UV alone. For the aldehydic compounds,

452

Chemical Basis for Water Technology

moderate removal was obtained with direct UV photolysis (i.e., UV alone), and geosmin, MIB, and the pyrazines were most resistant to treatment, regardless of whether or not hydrogen peroxide was present. The results indicate that, if odors in a given water were due only to the aldehydic compounds, an AOP might not be necessary. Experiments with different hydrogen peroxide doses and UV fluences that are typically used for disinfection (40– 200 mJ cm2) demonstrated that high concentrations of hydrogen peroxide would be necessary to treat any of the compounds investigated. The experiments with hydrogen peroxide verified that hydroxyl radical-driven reactions contribute only partially to the removal of the aldehydic compounds, whereas they play a very important role in the removal of geosmin, MIB, IBMP, and IPMP. In related pilotscale work, Gray (2006) reported that, as has been found by others, geosmin was consistently better removed than MIB with the combination UV/H2O2. The bench-scale investigations conducted by Gray (2006) provide quantitative information for the removal of these compounds, that could be useful in treatment process design. However, as has been stated elsewhere, the influence of background water quality on the performance of UV and UV/ H2O2 processes means that investigations must be conducted for a particular water. As has been suggested by others, Gray (2006) notes that his results point to the possibility of using a UV reactor designed for disinfection to also deal with seasonal taste and odor problems, by increasing the UV fluence and adding hydrogen peroxide. Oxidation summary. As with other treatment processes, many studies have focused on an individual water or location and the vast majority of the available literature is for the removal of geosmin and/or MIB. The major findings for oxidation processes can be summarized as follows:

• • •

•

•

Conventional oxidants (i.e., chlorine-based) have more limited success than ozone or AOPs for odor removal. The use of potassium permanganate for taste and odor control appears to be specific to certain substances. Ozone can be very effective for odor control, and provides oxidation by means of both molecular ozone and the hydroxyl radical. The relative concentrations of these two species are strongly influenced by the background characteristics of a particular water, especially the concentration of the carbonate ion and the level of humic substances. Although the carbonate ion acts as a scavenger of hydroxyl radicals, the impact of humic substances is more complex as they can be both a consumer of hydroxyl radicals and a promoter of their formation. An implication of the scavenging effect of carbonate ions is that, for low-alkalinity waters, the advantages of an AOP involving ozone (i.e., PEROXONE) versus ozone alone may be less than in higher alkalinity waters. In recent years, the usefulness of AOPs has increasingly been demonstrated for taste and odor control. The most common configurations are the combination of ozone and hydrogen peroxide (referred to as the PEROXONE process) and the combination of hydrogen peroxide and UV. The latter process has gained increasing popularity because of

the increasing use of UV for disinfection. However, it should be noted that the UV fluences required for effective odor removal are much higher than those conventionally used for disinfection. At least one manufacturer has developed a process where additional UV lamps can be turned on and hydrogen peroxide fed during an odor event. The following additional points are of note:

•

•

• •

• •

Oxidants can create odors (or change their character) as well as remove them. This is true for both chlorine-based oxidants and for ozone. The odors produced by ozone, sometimes characterized as fruity, are more likely to occur where the original odor is of complex character. The main application of chlorine or chloramines as oxidants to deal with odors is for the removal of sulfide odors and related odors caused by anaerobic conditions. Conflicting information appears in the literature regarding the ability of chlorine to mask odors. A switch from chlorine to chlorine dioxide may help minimize chlorinous or medicinal odors. Various workers have reported that geosmin is consistently better removed than MIB with UV/hydrogen peroxide. The parameter electrical energy per order (EEO) used for UV processes is useful but is very specific with respect to reactor geometry, UV lamp, the specific contaminant being investigated and background water quality. One study investigating the removal of aldehydic odorous compounds found moderate removals with direct UV photolysis (i.e., UV alone, without hydrogen peroxide edition). Not a great deal of work has been done on AOP by-products, to determine whether there may be compounds formed that are of health concern.

Adsorption. The overview on absorption and absorption of various compounds are given as follows. Introduction. As noted in Section 3.16.4.9, the adsorbent almost universally used in water treatment is activated carbon, applied as either PAC or GAC. PAC is particularly useful for periodic taste and odor episodes. GAC contactors as a separate treatment stage for taste and odor are relatively expensive. Filter adsorbers, combining filtration and adsorption in one step, are less expensive; however, the contact time for adsorption may be insufficient for effective removals, because of the shallower GAC layer. For GAC, biodegradation will inevitably also occur unless there is a disinfectant residual throughout the contactor or the duration of a pilot study is only several weeks, that is, too short for a biofilm to become properly established. Some authors investigating adsorption of odorous compounds on GAC mention biodegradation, while others do not. Biodegradation can essentially be neglected as a removal mechanism for PAC, because the contact time of the carbon with the water is typically at most a few hours, and therefore too short for biofilm development. AWWARF-LE (1995) provide good coverage of adsorption in chapter 4 of their book on taste and odor control, including a number of case studies, and discuss the effects of oxidants. They note that in all cases the most important aspect of activated carbon use is the competition with background organic matter or with other contaminants. As will be discussed

Chemical Basis for Water Technology

extensively later in this section, this competition reduces, often substantially, the capacity of activated carbon for specific odorcausing compounds. Although AWWARF-LE (1995) present extensive isotherm data, the quantitative use of this information for design purposes, as is the case with any isotherm data, must be done with extreme caution because of the previously mentioned competition effects. Isotherms can however provide a good indication of the relative adsorbabilities of different odorous compounds. The results of specific investigations on both PAC and GAC are reported below. In general, less literature is available for the removal of odor compounds using GAC than with PAC. Powdered activated carbon. Crozes et al. (1999) reported on a bench-scale program for rapid testing of various odor treatments (oxidation and adsorption). During the testing, the predominant raw water odor characteristic was earthy/musty. Using standard jar tests, the investigators evaluating PAC found that a wooded-based carbon was better than coal-based carbons for odor removal but indicated that economic analyses should also be performed. Gillogly et al. (1999) described a simplified method for determining PAC dose required to remove MIB. Using natural water with different initial concentrations of MIB, they showed that the percentage removal was independent of the initial concentration for a given PAC dose. They note that this relationship is specific for each type of PAC and natural water. They state that it is not valid at very high MIB concentrations, but they tested it at MIB concentrations up to 178 mg l1, which is far above raw water concentrations that normally would be encountered. They demonstrated the robustness of their approach by applying it to four different source waters and 14 different activated carbons. Graham et al. (2000b) also determined that the percentage removal of geosmin and MIB in natural waters by a given carbon dose was independent of initial concentration in the range of 40–180 ng l1. Cook et al. (2001) also confirmed that percentage removal for a given carbon dose was independent of starting concentration. Gillogly et al. (1999) note that their procedure gives a minimum dose and that the actual dose will be higher in a given water-treatment plant if equilibrium is not achieved. As the authors note, the value of their work is that it demonstrates that only a single isotherm test with a given carbon is necessary for a particular water. The authors also note that other investigators (Newcombe et al., 1994) have established that characteristics of the carbon such as surface area and micropore volume cannot be consistently used to predict a priori how well MIB would be removed. In another study, Gillogly et al. (1998) found that the ability of PAC to adsorb MIB was significantly reduced by free chlorine. They also found that, because chlorine can oxidize adsorption sites containing MIB, chlorine can cause the immediate release of adsorbed MIB back into the water. Graham et al. (2000b) found that equilibrium adsorption capacity for MIB and geosmin on PAC in natural waters was at least a factor of 10 lower than in laboratory water (i.e., without the presence of NOM). They also showed that modeling could be successful when a single equivalent background compound (EBC) was employed to simulate the competition from the background NOM. In natural waters, somewhat lower

453

adsorption of the odorous compounds was observed at pH 5.5 versus 7.5. It was hypothesized that this was probably because of the enhanced adsorption of NOM at the lower pH. Cook et al. (2001) note that the PAC dosage required is influenced strongly by the type of activated carbon, the presence of NOM, and the contact time. They also note that the effects of NOM concentration and character are not straightforward and could not be interpreted in terms of the parameters they measured. The addition of PAC with coagulant chemicals had no effect on geosmin or MIB removal; however, higher turbidity did reduce removal efficiency. The authors described an approach for predicting the required PAC dose, taking into account the kinetics of adsorption and were able to construct a PAC dose table. The important aspect of this work is that it assists in making realistic predictions for a specific treatment plant, where the actual contact time of the PAC is not long enough for equilibrium to be reached. However, they noted that full-scale confirmatory testing was required. Cook et al. (2001) also investigated adsorption isotherms for geosmin and MIB. They found that geosmin was more strongly adsorbed than MIB and that the adsorption was less sensitive to background water quality. They proposed that, for a given carbon, only one isotherm would be required to predict geosmin removal in a wide range of waters, whereas for MIB an isotherm would be required for each water. Jung et al. (2004) also found geosmin to be better adsorbed than MIB on PAC. Newcombe and Cook (2002) describe the various factors influencing the removal of tastes and odors by PAC, focusing on geosmin and MIB. Significant findings are as follows:

•

•

• •

• •

In general, good-quality microporous carbons (e.g., coconut or coal-based) will be better than wood-based carbons, although at shorter contact times the superior adsorption kinetics of the chemically activated wood-based carbons may be advantageous. The removal at contact times of 2 h or less is more important than the equilibrium removal capacity of the carbon. The adsorption kinetics for geosmin and MIB depend on both the equilibrium capacity of the carbon and the volume of larger transport pores that allow rapid access to the adsorption sites. The fact that geosmin has a slightly lower solubility than MIB and is flatter in structure helps to explain its observed greater adsorbability. Only some of the background NOM present in natural waters will compete for adsorption sites on carbon. The majority of the competitive effect is stated to be due to direct competition for adsorption sites by small, uncharged NOM compounds with low UV absorbance. In one study, the competing NOM was found to be less than 10% of the total NOM. Therefore, bulk characterization and concentration parameters for NOM would not be expected to be specific enough to directly predict the extent of competition for a particular water. A user-friendly computer program has been developed for application of the model at full scale. Prechlorination is not recommended when PAC is used for taste and odor control. This is consistent with findings reported by Gillogly et al. (1998).

454

•

Chemical Basis for Water Technology

In some cases, the simultaneous application of PAC and alum or other coagulants can increase the PAC dose required. In general, a higher alum dose because of a higher turbidity will likely lead to a higher required PAC dose than that predicted by the homogeneous surface diffusion model (HSDM).

Graham et al. (2000a) conducted a survey of full-scale watertreatment plants in the US and Canada using PAC for control of earthy–musty odors. The survey showed that most plants (60%) applied PAC at the rapid mix step. Bench-scale investigations conducted as part of the study showed that, under the operational constraints of a conventional treatment plant, the greatest removal of odorous compounds occurred when the PAC was added prior to coagulation. The adsorptive capacity of PAC for the odorants was reduced in the presence of oxidants. No consistent significant relationships were found in attempting to correlate performance with 11 different adsorption and physical properties for five PACs tested. None of the parameters typically used to characterize carbon adsorption performance (such as molasses number, iodine number, etc.) consistently predicted the effectiveness of a PAC for removing geosmin or MIB. GAC. Chen et al. (1997) examined GAC for removal of MIB to below the odor threshold concentration. Pure-water isotherms at realistic equilibrium concentrations (o10– 100 ng l1) showed that the bituminous carbon had the highest capacity, followed by the peat, lignite, and wood carbons. The authors also noted the reduction in capacity in natural water due to competition. In agreement with results reported earlier in this section, they found that the ranking of carbons according to traditional parameters such as the iodine or tannin number was inconsistent with the rankings obtained from the pure-water isotherms. Again, consistent with results reported above, the authors found that, under the conditions tested, the capacities were reduced by about an order of magnitude (factor of 10) in the natural water. It should be noted that isotherm tests with natural waters essentially only measure direct competition from NOM, and not the preloading effect discussed in Section 3.16.4.9. Kim et al. (1997) reported on pilot-scale investigations including GAC for the removal of taste and odor causing substances in a water-treatment plant in Korea. The compounds investigated were geosmin, MIB, IPMP (2-methoxy3-isopropyl-pyrazine), IBMP (2-methoxy-3-isobutyl-pyrazine), and TCA (2,3,6-trichloroanisole). The pilot plant consisted of coagulation/flocculation/sedimentation and rapid sand filtration followed by two parallel trains. Train 1 had ozonation followed by GAC, whereas train 2 had GAC only. The EBCTs in the GAC columns in both trains ranged from 10 to 15 min. The entire investigation lasted for 2 years. The raw water concentrations of the odorous compounds were between approximately 10 and 30 ng l1, except for IBMP and MIB, which were present at approximately 80 ng l1. The authors reported that the GAC filters following ozonation showed an increased BAC effect. The treatment process including ozonation was capable of reducing the five substances to less than the TON with an EBCT of more than 15 min, regardless of the iodine number of the GAC. This investigation, along with others, shows that while a multistep

process can be successful in treating odors, GAC alone will not necessarily provide sufficient removals. Ho (2004) studied the removal of cyanobacterial metabolites (MIB, geosmin, and the toxin microcystin) from drinking water using ozone and GAC. An important finding was that laboratory-scale minicolumn experiments, combined with the HSDM, were not effective in predicting GAC breakthrough at two different pilot plants. This was attributed primarily to the biological degradation taking place at the pilot plants that could not be modeled by the HSDM. Ndiongue et al. (2006) assessed the remaining odor removal capacity of GAC filter caps using four pilot-scale columns with the same bed depths as full-scale filters and fed with the same water received by the full-scale filters. Geosmin and MIB were spiked into the influent of the pilot-scale filters, at concentrations ranging up to several hundred nanograms per liter. The first experiments conducted by Ndiongue et al. (2006) were designed to assess the possible losses of the spiked compounds to the apparatus and therefore were conducted before the media was added to the columns. Such investigations are important because of the very low concentrations of the compounds being investigated, and the very high surface-to-volume ratio of pilot-scale equipment, in comparison with full-scale facilities. Similar losses were observed for both geosmin and MIB, ranging between 23% and 40%. Results obtained later in the study indicated that system losses may have decreased over time. Following the system loss experiments, media was added. Three columns contained different depths (25–95 cm) and types of used carbon while a fourth contained fresh carbon. The columns were operated continuously for approximately 2.5 months. Three experiments of several days’ duration each were conducted, in which the odor compounds were spiked. Under the conditions tested by Ndiongue et al. (2006), none of the pilot filters were able to reduce geosmin and MIB concentrations to below the commonly cited odor thresholds of 4 and 9 ng l1, respectively. The low adsorption capacity was attributed to fouling of the carbon by background TOC; although biodegradation may have played some role in this investigation, its contribution was considered minimal. The significance of this investigation is in demonstrating how quickly the capacity of GAC for target compounds is reduced by the adsorption of background TOC. Since GAC will never be completely fresh when an odor episode strikes, GAC alone may have limited success as a treatment strategy in dealing with such episodes. Newcombe et al. (1997) provide a good discussion of the impact of NOM on adsorption on GAC. They note the direct competition effect, as discussed earlier for PAC, and also the fouling or preloading effect, essentially only relevant for GAC. The authors report that the competitive effect between NOM fractions and MIB was largest for the smallest NOM fraction (having an UF molecular weight o500). Their explanation is that this fraction was the most similar in size to MIB and therefore could directly compete for the same adsorption sites. Hepplewhite et al. (2004) confirmed the main findings reported just above, including the fact that the most competition was provided by the low-molecular-weight NOM compounds. They also discuss the complexity of doing kinetic analysis. They

Chemical Basis for Water Technology

state that the rate of adsorption in the initial stages is strongly affected by the equilibrium capacity of the carbon, as well as by the carbon’s pore structure and competition from NOM. They also note that large NOM compounds can restrict, although not block, access to the adsorption sites in a carbon with substantial mesopore volume. For a carbon with primarily micropores, low-molecular-weight NOM components can restrict access to these pores (as well as compete for adsorption sites) and thus affect adsorption kinetics. MacKenzie et al. (2005) assessed the impact of different GAC reactivation procedures on both the ability of the carbon to remove MIB, and its ability to retain MIB once an odor episode had ended. The authors suggest that, for virgin carbons, the Brunauer–Emmet–Teller (BET) surface area is an important factor for MIB removal, although other GAC properties such as pore-size distribution and surface chemistry may also play a role. However, for the carbon reactivated using various procedures, there was no relationship between BET surface area and number of bed volumes to breakthrough. The authors indicate that this could suggest that surface chemistry is more important for reactivated carbons, although they also note that reactivation will have an impact on pore size. This information is significant for a water-treatment plant that may be planning to reactivate its carbon. Several investigations on the modeling of GAC performance, while not conducted using odorous compounds, show that background organic matter affects not only the capacity of GAC for target contaminants, but also the kinetics of adsorption. Carter and Weber (1994) conducted investigations with both a natural and a synthetic water to investigate the impact of preloading by background organic matter on the adsorption of TCE. They used an initial TCE concentration of approximately 50 mg l1. Although this is of course substantially higher than typical concentrations of odorous compounds, it likely does not preclude application of their findings to odor removal. They found that, for both waters, both equilibrium capacities and rates of adsorption decreased with increased periods of preloading. In contrast to the findings of several other investigators, they noted that in the natural water (from the Huron River) the effect of preloading reached a plateau after several weeks. Jarvie et al. (2005) used pilot- and field-scale data from 11 studies to attempt to improve the prediction of GAC adsorber performance. The data represented both surface and groundwater sources, and were for the removal of various synthetic organic compounds in the microgram per liter range. They found that the background organic matter in both ground and surface waters could significantly reduce both adsorption capacity and kinetics, and were able to obtain some trends regarding the way in which the diffusivity within the GAC particle changed as a function of time and depth of the bed. They considered that their modeling approach was sufficiently developed to make crude design calculations. They noted that their approach provided exceptionally better estimates of GAC use rates (compared to actual data) than the standard massbalance approach based on isotherm data that is generally used to obtain a quick determination of GAC requirements. More recently, Yu et al. (2007) reported on the reduction of both adsorption rate and capacity on GAC preloaded with background organic matter. In laboratory tests, they

455

investigated the impact on adsorption for the pharmaceutical naproxen. The initial naproxen concentration was 500 ng l1, which is at the upper end of the range at which odorous compounds such as geosmin and MIB may be observed. Pirbazari et al. (1993) published a GAC adsorber design protocol based on computer modeling for the removal of offflavors (taste and odor compounds). The model was used to simulate full-scale adsorbers, although the authors note that biodegradation was not considered and therefore their cost estimates would be conservative. The modeling results confirmed experimental data obtained in various investigations showing the major influence of background organic matter on GAC capacity. For the conditions chosen by Pirbazari et al. (1993), the breakthrough times for MIB were at most on the order of a few weeks. This calls into question the practical feasibility of using GAC in an adsorptive mode only (i.e., not as a support medium for a biofiltration process) for odor control. The value of the authors’ work lies both in the development of a protocol for predicting full-scale adsorber capacity (with recognized limitations) from small scale tests, and also in showing the complexity of such a task. Sagehashi et al. (2005) reported on a bench-scale investigation of a process involving both a high silica zeolite adsorbent (USY) and ozonation. They obtained good results for the removal of MIB. It will be interesting to watch for possible further development and larger-scale testing of this process. If such testing yields positive results, the process may represent a possible future alternative for taste and odor control. Adsorption summary. The major findings are as follows.

• •

•

•

In general, geosmin is better removed through adsorption than MIB. Both PAC and GAC can be effective for odor removal. PAC is generally well suited to episodic odor events and, in general, seems to be more frequently used than GAC. GAC may be more suitable if odor conditions last for much or all of the year; however, if GAC is installed, it is more likely to also have other treatment objectives. In this case, the optimal operating conditions, and even the choice of carbon type, may not be the same for all contaminants. A clear indication from the literature is the tremendous impact of background water quality, specifically NOM (TOC) on adsorption. TOC is present at the milligram per liter level, whereas odorous substances occur at the nanogram per liter level. Low-molecular-weight NOM components compete very effectively with the odorous compounds for adsorption sites and larger NOM molecules can block or hinder access to these sites, also reducing the kinetics of adsorption. NOM affects both PAC and GAC; however, in the latter case the effect will be more severe because of the extended exposure of GAC to NOM in the water, whether or not the odorous compounds are present. The equilibrium capacity of activated carbon for odorous compounds has been shown to be at least a factor of 10 lower in actual water than in laboratory water, due to the effect of NOM. It has also been demonstrated that measuring standard parameters such a water’s TOC is not helpful in predicting the impact on adsorption. For GAC, the original adsorptive capacity can be substantially reduced

456

•

•

•

•

Chemical Basis for Water Technology

relatively quickly (i.e., within a few weeks) by the background TOC. As a consequence of the effect of NOM, GAC alone may not be sufficient in eliminating odor problems; however, it can be a very effective component of an overall treatment strategy. In odor-removal investigations on GAC, biodegradation of the odorous compounds is not always considered. Biodegradation will lead to increased removals, and is discussed in detail in the next section. Biodegradation essentially does not contribute to removals when PAC is used, because the contact times are typically at most only a few hours. Several investigations have shown that for a given water and PAC dose, the percentage removal of an odorous compound is independent of its initial concentration, in the concentration range of the compounds normally expected in natural waters. This is very helpful for predicting performance in a given water where the influent concentration may be different than that tested. In the application of PAC for geosmin and MIB removal, the removal at contact times of 2 h or less (i.e., the time typically available in practice) is more important than the equilibrium removal capacity of the carbon. The adsorption kinetics for these compounds depend on both the equilibrium capacity of the carbon and the volume of larger transport pores in the carbon that allow rapid access to the adsorption sites. Adsorption isotherms (typically carried out much longer times) even with the actual water will typically overestimate the removals by PAC obtainable in practice. The ability of PAC to absorb MIB has been found to be significantly reduced by the presence of free chlorine. Therefore, chlorination upstream of PAC application is not recommended.

Additional findings are as follows:

•

•

• •

In terms of predicting a carbon’s capacity for odor removal, traditional parameters such as iodine and molasses numbers are not useful. Characteristics of the carbon such as surface area and micropore volume also cannot be consistently used to predict removals. In general, good-quality microporous carbons (e.g., coconut- or coal-based) will be better than wood-based carbons for the removal of geosmin and MIB. However, at shorter contact times (in PAC applications), the better adsorption kinetics of the chemically activated wood-based carbons may be advantageous. Differing results have been reported for the impact of coagulants or adsorption by PAC, with some investigations showing no impact. In conducting experiments, it is important to take into account the possible adsorption of odorous compounds onto the walls and surfaces of laboratory-scale apparatus, particularly when it is new.

Biological treatment. Introduction. The use of biological processes (i.e., biological filtration) has been increasingly investigated in recent years for odor removal. Most of these investigations have been conducted at laboratory or pilot scale and, as with other treatment

technologies, have focused largely on the removal of geosmin and MIB. As discussed in Section 3.16.4.10, important factors affecting the performance of biofiltration include contact time, temperature, the absence of a disinfectant or oxidant residual in the filter, and, to some extent, the choice of media. The degradation of substances such as odor compounds present at very low concentrations (typically at nanogram per liter levels) occurs by secondary utilization. This means that the amount of biomass in the filter is determined by the overall amount of biodegradable carbon present in the water, as measured by a parameter such as AOC or BDOC, which serves as the main carbon and energy source for the biofilm. In filters where the media is GAC, except for the initial operating period, the adsorptive capacity is essentially exhausted and therefore the removals taking place are due to biodegradation, and the media is essentially functioning as a biomass support. In some cases, GAC has been shown to be a better support medium than sand or anthracite; however, this is not always the case. Most investigations have been conducted for what is referred to as rapid biological filtration. Several investigations that have been conducted for bank filtration are also included at the end of the section. Identification of specific microorganisms responsible for biodegradation of geosmin and MIB. Tanaka et al. (1996) isolated two bacteria capable of degrading MIB from the backwash water of a pilot-scale biological filter at Lake Biwa in Japan. The identified bacteria belonged to the genera Pseudomonas and Enterobacter. Using lake water, Lauderdale et al. (2004) isolated and characterized a bacterium capable of aerobically degrading MIB. The bacterium was determined to be most closely related to Bacillus fusiformis and Bacillus sphaericus. Lauderdale et al. (2004) note that other authors, including Tanaka et al. (1996), have reported that oxidation is a significant microbial pathway for MIB transformation. In investigations involving bench-scale biological sand filters, Ho et al. (2006) identified a Sphingomonas species as a potential geosmin degrader. During the onset of geosmin degradation in a bench-scale biofiltration study, Hoefel et al. (2006) detected the predominance of three bacteria, most similar to previously cultured species of Sphingopyxis alaskensis, Novosphingobium stygiae, and Pseudomonas veronii. Subsequent investigation showed that degradation of geosmin, whether present as either the sole carbon source or when spiked into sterile water from a reservoir, occurred only when all three isolates were present. None of the isolates were demonstrated to be able to degrade geosmin either individually, or in any combination of two. In a full-scale water-treatment plant, the biofilm community (including nondegraders of odorous compounds) that will be sustainable over the longer term will be composed of those organisms best able to compete under the conditions present. Bench- and pilot-scale investigations. Saito et al. (1999) reported on the microbial degradation of geosmin obtained with the backwash water from a pilot-scale biological filter (and also with activated sludge from a wastewater-treatment plant), using batch-suspended growth experiments. They found geosmin to be more difficult to degrade microbiologically than was the case

Chemical Basis for Water Technology

with MIB, which the group had reported on previously (Tanaka et al., 1996). A reason for this is not given, nor is it immediately evident from a review of the two papers. However, it is possible that this difference could be attributed to differences, even subtle ones, in experimental conditions. The degradation potential of MIB and its producer via biofilm process was examined by Sugiura et al. (2003). The authors took biofilm from a full-scale biological treatment facility and performed suspended growth experiments. They found that degradation decreased as pH increased, in the range from 7.4 to 8.6. The rise in pH also corresponded to a decrease in ciliated species in the biofilm. They attributed the predation and degradation of the filamentous cyanobacterium Phormidium tenue (which was considered to be the exclusive producer of MIB), and its intracellular MIB, primarily to protozoa such as ciliates. In the full-scale facility, removal of the MIB released from P. tenue as a result of their degradation was attributed to bacteria in the biofilm. They concluded that abundant ciliates species in the biofilm and pH near neutral were important factors for the efficient removal of MIB and P. tenue. Huck et al. (1995) used a bench-scale biofilm reactor to attempt to determine kinetic parameters for the biodegradation of geosmin and 2,4,6-trichloroanisole. In experiments conducted over a relatively short period of time (the exact duration is not specified), the bioreactors did not effectively remove the compounds. It is important to note that the biofilm was established using amended river water without the odor compounds, and for the actual experiments the odor compounds were then fed in a synthetic solution. Based on their results, the authors do not recommend a biofilter as the primary treatment process for the removal of these compounds when it is not acclimated and when it receives an organic and hydraulic load in the normal range for rapid filters and GAC contactors. The acclimation aspect is important for intermittent odor events. However, the authors note that the poor removals they obtained may be due to nonacclimation of the system to either the carbon sources in the synthetic water used or the odor compounds themselves. An important aspect of the study by Huck et al. (1995) is that it documents the preliminary investigations undertaken to determine and minimize losses of the odorous compounds (by adsorption, degradation in the feed container, and possibly volatilization) in the experimental system. These investigations showed that such losses can be considerable, and must be taken into account for studies at realistic (i.e., low) concentrations in bench-scale systems, with their high surfaceto-volume ratios. (An evaluation of losses to the surfaces of experimental apparatus during adsorption experiments was reported in the previous section (Ndiongue et al., 2006).) Hrudey et al. (1995) described odor removal by biological treatment processes in a pilot plant containing conventional treatment and ozonation followed by three parallel two-stage biologically active filtration trains. The first-stage filters contained either anthracite/sand or GAC/sand and the second stage in all three trains contained a GAC contactor. The overall conclusion from this study was that a pilot-scale treatment process including ozonation and biologically active GAC was effective in removing a complex raw water odor to levels judged to be essentially odorless by FPA. It is however

457

considered likely that adsorption played some role in the overall removals. The authors stress the importance of the finding that ozonation alone, under the conditions used, was not effective for dealing with this odor event. Terauchi et al. (1995) reported on biological filtration for the removal of a musty odor, determined to be largely due to MIB. They conducted pilot-scale investigations over 2 years using a porous ceramic media with a diameter of about 5 mm. The depth of the media was 1.5 m, the filtration rate was 170 m d1 (approximately 7 m h1) and the contact time was about 13 min. During the season of natural odor, the total MIB concentration and the ratio of insoluble to total MIB varied substantially. Removal efficiencies of the filter were in the range of 60–80% for influent concentrations up to several hundred nanograms per liter of soluble MIB. Outside of the natural odor season, the investigators added MIB to the filter influent and the percentage removals were about the same. The authors report that as a result of their study, biofiltration was installed at full scale. Elhadi et al. (2004) reported on the startup phase of investigations on the removal of geosmin and MIB using small pilot-scale biofilters. Two columns were operated in parallel – each contained 50 cm of GAC (PICA P-830) over 25 cm of sand. One of the columns contained fresh GAC, while the other contained exhausted GAC from a water-treatment plant. The hydraulic loading rate was 7.5 m h1, corresponding to an empty bed contact time of 5.6 min. The columns were fed with dechlorinated tap water that was amended with the two odor compounds (each at target concentrations of 100 ng l1), several typical ozonation by-products, as well as nitrogen and phosphorus. The low-molecular-weight ozonation by-products, fed at a total concentration of approximately 800 mg l1, were added to simulate the effect of ozonation upstream of a biological filter. The investigations were conducted at room temperature (approximately 20 1C). In the study by Elhadi et al. (2004) an acclimation period close to 2 months was required for biological removals to be fully developed (this period did include a 3-week period during which the odor compounds were not fed, to simulate intermittent odor events). Once acclimation had occurred, biological filtration achieved 80–90% removal of geosmin and 50–60% removal of MIB. After several months (i.e., when the removals were essentially biological without a substantial contribution from adsorption), the removals of both odor compounds on exhausted GAC were only a few percentage points lower than on fresh GAC. The importance of an acclimation period suggests that biofiltration alone could not necessarily successfully handle a rapidly developing geosmin and MIB odor event, where the biomass did not have recent prior exposure to these compounds. Results of investigations carried out over several years using the same experimental setup were reported by Elhadi et al. (2006). Factorial experiments were designed to quantify the effects of the major factors expected to influence removals of geosmin and MIB during biological filtration: temperature, geosmin/MIB concentration, media type, and the concentration of easily BOM (to simulate the presence or absence of upstream ozonation). The experimental conditions were as described above (Elhadi et al., 2004) and were chosen to represent reasonable rapid filtration practice.

458

Chemical Basis for Water Technology

Although the geosmin and MIB were fed either at increasing concentrations (ramped) or at a constant maximum concentration of 100 ng l1 (not ramped), Elhadi et al. (2006) discuss in detail only results at 100 ng l1. The investigations showed that temperature, geosmin and MIB concentration (i.e., whether the concentration had been ramped or not), media type, and BOM level were all important factors affecting removals. The best removals (60% for geosmin and 40% for MIB) were obtained in the exhausted GAC filters at the higher temperature (20 1C) and in fact under these conditions, the BOM level was relatively unimportant. At influent concentrations of 25 ng l1 at both BOM levels, sufficient geosmin and MIB removal occurred in the GAC filters at 20 1C to produce an effluent that was within the low odor-threshold range. The better removals obtained by Elhadi et al. (2006) on exhausted GAC versus anthracite were observed at both BOM levels, but were more significant at the lower BOM level. For both geosmin and MIB, somewhat lower removals were obtained when the concentrations were ramped up to 100 ng l1 than when they were fed at that level from the beginning. Importantly, a quantitative relationship between biomass levels (as measured by the phospholipid method) and geosmin and MIB removals was not found, although there was a qualitative relationship. The results of the investigation by Elhadi et al. (2006) show that, once acclimated, biofiltration is capable of providing appreciable removals of geosmin and MIB under typical rapid filtration conditions. However, depending on the influent concentration, biofiltration alone will not necessarily reduce odor concentrations to below the threshold level. If biofiltration is coupled with upstream oxidation using ozone, additional easily biodegradable carbon compounds will be produced that will likely enhance the efficiency of the downstream biofiltration process for odor removal. In terms of media, GAC, even if exhausted for adsorption, would appear to be a better biomass support medium than anthracite. Uhl et al. (2006) investigated the removal of geosmin and MIB in pilot-scale biological filters receiving an untreated surface water in Sweden. At an empty bed contact time of 30 min and a temperature of 15 1C, biofiltration through either GAC or crushed expanded clay reduced an initial concentration of 20 ng l1 of geosmin and MIB by at least 97%. At a lower temperature (6–12 1C) where biomass concentrations were also lower, the GAC showed similar removal efficiency, but the efficiency was considerably lower in the expanded clay biofilter. Adsorption was shown to play an important role in removals in the GAC filter, even though the GAC had been in operation for nearly 4 years treating surface water. Summers et al. (2006) also reported investigations on the biodegradation of MIB in laboratory-scale filters having an EBCT of 7 min. Acclimated biologically active sand and activated carbon were obtained from the Greater Cincinnati Water Works in Cincinnati, USA. The GAC had been in use there for approximately 10 months and the sand had been in use for at least 2 years. After 4 months of operation, approximately 50% MIB removal was obtained by the acclimated biologically active sand filters at room temperature, and no removal was observed at 4 1C. (Nonacclimated biologically active sand filters at room temperature averaged only 7% removal after

4 months.) The biologically active GAC filters showed MIB removals of about 65% (whether this removal was also seen at the lower temperature is not specified). Ho et al. (2007) examined biodegradation of geosmin and MIB in biologically active laboratory sand filters and determined degradation rates in batch bioreactors inoculated with biofilm from one of the sand filters (no solid media was present in the bioreactor experiments). The batch experiments showed that the biodegradation of geosmin and MIB was pseudo-first-order, with rate constants ranging between 0.10 and 0.58 d1. The rate constants were dependent on the initial concentration of the microbial inoculum, but not on the initial concentration of geosmin or MIB when target concentrations of 50 and 200 ng l1 were used. Rate constants were demonstrated to increase when the biofilm was re-exposed to both taste and odor compounds. While these rate constants provide useful information, it should be noted that they cannot be directly applied to biofilters, where the concentration of biomass per unit reactor volume will be different. In another investigation with a laboratory sand biofilter, Ho et al. (2006) reported a lag period in excess of 75 days before greater than 95% removal was obtained for geosmin and MIB. Although the filter sand had been obtained from an operating water treatment plant, during the laboratory experiments it was fed with water from another location, which may help to explain the delay time. Initially, the EBCT was 15 min, and after significant initial breakthrough of both compounds, their removals steadily increased after the first 9 days. On day 79, the EBCT of the column was increased to 30 min, which corresponded to a further increase in the removal of geosmin and MIB to greater than 95%. For comparison, the authors note that geosmin and MIB appeared to be more difficult to degrade biologically than the nonodorous but toxic cyanobacterial metabolite, Microcystin-LR. As part of her investigations, Elhadi (2004) also determined rate constants for geosmin and MIB removal in pilot-scale biofilters. She found that zero-, first-, or second-order kinetics could describe her data equally well, and adopted a zero-order model for simplicity. Observed rate constants for geosmin and MIB ranged from approximately 2 to 8 ng l1 min1 at 20 1C. She also conducted experiments at 8 1C and was able to determine temperature coefficients for the rate constants. Full-scale investigations. AWWARF-LE (1995) includes a chapter on biological removals that provides information on full-scale practical experience in the Paris, France area. Nerenberg et al. (2000) reported preliminary investigations of ozone/biofiltration for MIB and geosmin removal at a full-scale plant. (The results relating to ozonation have been discussed in a previous section.) They reported that the ozonation/biofiltration combination could effectively remove MIB to below threshold concentrations, and that biofiltration played an important role in this reduction. During the period of investigation, the typical empty bed contact time was 17 min based on total bed height, or 11 min based on the height of the GAC only. The GAC had been in service (without regeneration) for 6 years, and therefore the role of adsorption in MIB removal was considered unlikely to be significant. The authors’ removals ranged from 26% to 64%. In full-scale biological sand filters in Cincinnati, USA, Metz et al. (2006) observed sustained biological removal of MIB

Chemical Basis for Water Technology

(80%) and geosmin (50%) over a 6-year period. During this time, the filter influent concentrations of TOC, geosmin, and MIB exhibited considerable variability. They noted that no disinfectant was added ahead of the filters and that the filter influent was not ozonated. They also noted that, during the investigation, the filter media was replaced in all filters over a period of several years and the hydraulic loading increased by 20%. Neither of these changes presented a problem with regard to reaching the utility’s taste and odor goals. Although there is little information available regarding odor removal for other (at least partly) biological processes such as slow sand or bank filtration, Ju¨ttner (1995, 1999) has reported on removals of various compounds, including monoterpenes and geosmin. In an investigation on the Ruhr River in Germany (Ju¨ttner, 1995), the aquifer where riverbank filtration occurred was anoxic, while the following slow sand filters were aerobic. Removals differed among compounds, and the schmutzdecke at the top of the slow sand filters played an important role. Another study on the Neckar River in Germany (Ju¨ttner, 1999) investigated removals of fragrance compounds through anoxic bank filtration. Efficient removals, attributed to microbial degradation, were observed after a relatively short distance from the river. Biofiltration summary. The major findings regarding the use of biological processes for taste and odor removal can be summarized as follows:

•

•

• • •

•

•

Biofiltration can be a useful process for the removal of the taste and odor compounds, geosmin and MIB. For relatively low levels of these compounds, biofiltration under typical rapid filtration operating conditions may be able to provide filter effluent values below the odor threshold. For higher influent levels, an effective process combination may be oxidation followed by biofiltration. A number of investigations have shown that biofiltration requires an acclimation period before reaching its full potential for removals. Thus, biofiltration alone may not be a suitable process when odor events occur intermittently. GAC, essentially exhausted with respect to its adsorption capacity, has been shown to be a somewhat better medium than anthracite for biofiltration for taste and odor removal. Conflicting results are reported in the literature with regard to the relative ease of biodegradation of geosmin and MIB. There is no immediate explanation for this difference. Several investigations have determined kinetic parameters for the biodegradation of geosmin and MIB. One investigation using suspended growth batch bioreactors determined the reaction to be pseudo-first-order. Although rate constants were determined, these are not directly applicable to biofilters. Another investigation using pilot-scale filter columns concluded that the removals could be equally well represented by zero-, first-, or second-order kinetics. The zero-order model was chosen for convenience and rate constants were determined. Although most studies have been conducted at bench or pilot scale, several full-scale investigations have demonstrated good levels of removal under rapid filtration conditions. Several investigations have identified specific bacteria capable of degrading geosmin or MIB. In one investigation,

•

459

degradation of geosmin occurred only when three specific bacterial species were present simultaneously. There is limited information available on the removal of taste and odor compounds by other processes with an important biological component, such as riverbank and slow sand filtration. However, several studies have shown effective removals by these processes.

3.16.6.5.2 Pharmaceuticals and endocrine disrupting substances Research in this area is relatively recent; however, there is sufficient information to provide reliable indications. One of the most important points is that this group of substances encompasses a large number of compounds having various chemical properties. Thus, blanket statements cannot be made, and process selection must be based on the types of compounds to be removed. Alternatively, several processes might be used to remove a range of compounds. Of course, these compounds are present at extremely low concentrations, and this has both process and analytical implications. As with other types of organic compounds, important considerations are whether the compound is destroyed or simply transferred to another phase as is (i.e., in the case of adsorption), and whether the treatment process (e.g., oxidation) transforms the original compound into something that may be of equal or greater concern. A number of researchers (e.g., Huber et al., 2003; Crosina et al., 2006; Zwiener and Frimmel, 2000) have examined the removal of PhACs and EDS by oxidation. Biodegradation has also been reported. For example, Halle´ (2010) found that drinking water biofilters removed some compounds effectively, whereas others were refractory to biodegradation. Yu et al. (2008, 2009a, 2009b) extensively investigated the adsorption of a small group of compounds. The background TOC substantially reduced the effectiveness of both PAC and GAC. Removals by membrane processes have been investigated by various researchers in recent years. Bellona et al. (2004) reviewed factors affecting the rejection of organic compounds by NF and RO membranes. In terms of solute properties the following were found to be most important: molecular weight, molecular size (length and width), acid dissociation constant (pKa), hydrophobicity/hydrophilicity (log Kow), and diffusion coefficient. Important membrane properties affecting rejection include molecular weight cutoff, pore size, surface charge (measured as zeta potential), hydrophobicity/hydrophilicity (measured as contact angle), and surface morphology (measured as roughness). Feed water characteristics such as pH, ionic strength, hardness, and the presence of organic matter were also found to have an influence. An immediate conclusion that can be drawn from this review is that obviously removals will be very compound and site specific, and no easy generalizations are possible. As part of their review, Bellona et al. (2004) developed rejection diagram, which is essentially a decision tree giving a general indication of the range of rejection to be expected for a particular compound as a function of certain solute and membrane properties. Kim et al. (2007) developed a transport model for trace organic compounds through NF and RO membranes. Both

460

Chemical Basis for Water Technology

neutral and charged compounds were considered. They found that convection was the dominant transport mechanism for most compounds; however, diffusion was important for more hydrophobic nonpolar compounds. Convection was more important for NF membranes as well. They included a useful graphical summary of their modeling results, showing combinations of membrane and compound properties where factors such as pore size or membrane zeta potential would contribute to improved rejection. Comerton et al. (2009) studied the influence of NOM and cations on NF rejection of pharmaceutically active compounds and endocrine disruptors. They used synthetic and natural waters, including a membrane bioreactor (MBR) effluent, and conducted experiments over a period of 48 h. They concluded that, in general, fouling and the presence of NOM increased rejection, whereas it was reduced by an increase in cation concentration. Verliefde et al. (2009a) developed a method for predicting the transport of uncharged organic compounds through NF and RO membranes. The method requires the following solute and membrane parameters: solute size, membrane pore size, and solute-membrane affinity (intermolecular free energy). They note that these parameters can be relatively simply determined experimentally. Solute-membrane affinity (measured by contact angle) reflects phenomena such as hydrophobic attraction, hydrogen bonding, and dielectric effects. The authors found good agreement between model predictions and experimental data and note that the influence of solutemembrane affinity on rejection is very high. In a second study Verliefde et al. (2009b) examined the influence of fouling by pretreated surface water on NF rejection of PhACs. Fouling-induced changes in membrane surface hydrophobicity altered the extent of partitioning and therefore rejection for both hydrophobic and hydrophilic pharmaceuticals. The authors note that the impact of fouling on rejection determined in their laboratory experiments cannot be quantitatively extrapolated to full-scale units, because of differences in flow and other conditions. Thus although they found some rejections to decrease by more than 40% and others to increase by more than 15% due to fouling, they state that in full-scale units fouling would probably not cause changes in rejection (before and after fouling) of more than 5–10%. Yangali-Quintanilla et al. (2009) also examined the impact of fouling on rejection of PhACs and endocrine disrupting substances by NF membranes. In these laboratory experiments, sodium alginate was added as a foulant. For the NF200 membrane, fouling either decreased or increased rejection, depending on the class of compound. For the NF-90 membrane, the rejections of hydrophobic compounds were not affected by fouling, although increases in rejections of hydrophilic neutral compounds were observed. Makdissy et al. (2007) examined the removal of endocrine disrupting and pharmaceutically active compounds by four commercially available polyamide NF membranes. Two waters were investigated – ultrapure water and a surface water with a DOC of approximately 6 mg l1 and a SUVA value of 2.8 l mgC1 m1. Both size exclusion and electrostatic repulsion played a role in rejection, depending on the compound and membrane. Lower removals were observed in the natural water compared to ultrapure water.

Another investigation by Yangali-Quintanilla et al. (2010) developed a quantitative structure activity relationship (QSAR) model to predict rejection of pharmaceuticals and endocrine disruptors by NF membranes. The developed model, using interactions between membrane characteristics, operating conditions, and compound properties, was able to satisfactorily predict rejection. It is evident from the preceding paragraphs that NF can be successfully used to achieve at least partial, and sometimes good, removal of pharmaceutically active and endocrine disrupting compounds. By extension, NF would be expected to achieve at least some removal of other trace organics that may have been less extensively studied. It is however also evident that removals are compound and membrane specific, and are influenced by fouling. Therefore, quantitative a priori predictions for a given situation cannot reliably be made at present. This is not surprising given the complexity of the removal mechanisms involved. Thus, approaches such as QSAR offer promise for developing tools that could provide at least semiquantitative predictions. It is also evident that, depending on the compound and degree of removal that might be required, NF might need to be used in combination with another technology such as oxidation, absorption, or potentially biodegradation.

3.16.6.5.3 Volatile contaminants Although volatile contaminants can potentially be removed by oxidation or adsorption on activated carbon, for substances with sufficiently high Henry’s constants, air stripping can be an effective technology for removal. This section therefore addresses air stripping, because the other technologies have been discussed previously in relation to other types of organic contaminants. For air stripping, the principal factors controlling process selection are Henry constant (Section 3.16.4.8) and the required degree of removal. Figure 4 shows the type of system that can be used under various combinations of Henry’s constant and required percent removal. Stripping often requires what is referred to as a water-in-air system in which the air is in contact with either droplets of water or a thin film of water. Packed towers or columns are one of the most efficient types of water-in-air systems and are commonly used for air stripping. The packing material creates turbulence, thus maximizing and renewing the contact between the air and water surfaces. Thus, the process is also known as a packed bed process. Various types of standard packing materials are available. The diameter of the column is a function of the air and water flowrates. The height of the column packing required to achieve the desired removal of a given contaminant is the product of two quantities: the height of a transfer unit (HTU) and the NTU. HTU is a function of the efficiency of mass transfer and therefore will change with seasonal temperature changes. NTU is a function of the difficulty of removing a contaminant from the water and is related to the difference between the actual and equilibrium concentrations. In cases in which a very high percent removal of a volatile contaminant is required, it is therefore often most efficient to provide most of

Chemical Basis for Water Technology

the removal using air stripping and final polishing using adsorption on activated carbon. MWH (2005: 1177) reported that packed towers commonly provide one to four NTUs. Both the fundamentals of packed tower air stripping and detailed design calculations, including those for determining HTU and NTU, are provided by MWH (2005). They noted that commercially available software is commonly used for design. In summary, air stripping can be an effective process for the removal of volatile contaminants. Initial design for a given situation can usually be done based on information in the literature and knowledge of the contaminant concentrations present. Therefore, only brief confirmatory pilot testing is normally required, unless the process is part of a multistep treatment train for which overall optimum conditions must be determined.

3.16.6.6 Inorganic Contaminants For reasons of space, this section focuses on the removal of several metals. Iron and manganese, which are discussed first, are common problems in groundwater. Although they are generally not considered to represent health issues, elevated levels of iron and/or manganese are important from an operational and esthetic point of view. One of the most common treatments applied to groundwater is the removal of iron and/ or manganese where their levels exceed prescribed values. Arsenic, which is also addressed, is a serious contaminant from a health perspective, and for this reason allowable limits for this metal are low and, in some jurisdictions, have been set even lower in the last few years. Iron and manganese. Iron and manganese are metals appearing in raw waters in many regions of the world. These waters (generally groundwaters) have a low redox potential, and therefore the metals are generally present as divalent ions – FE(II) and Mn(II). They can be present either as mineral compounds, aquo complexes, or complexed with organic matter. The mechanism for iron and manganese removal takes advantage of the low solubility of hydrated iron and manganese oxides: Fe2O3  nH2O and MnO2  nH2O, allowing these substances to be removed from water using simple technologies.

461

Therefore, an important component of treatment technologies is to oxidize iron and manganese to the Fe(III) and Mn(IV) states, respectively, so that these relatively insoluble oxides can form. Two major ways to achieve this transformation are:

• •

increasing the oxidation potential of the water by adding oxygen from the atmosphere and increasing the pH of the water by the addition of bases such as Ca(OH)2.

The choice of the optimal treatment in a given situation can be determined experimentally. Technology based on increasing the oxidation potential, Eh, using oxygen from the air is generally used when the iron and/or manganese are present as mineral compounds and the concentration of dissolved organic matter, especially humic substances, is minimal. The technology shown schematically in Figure 5 is based on aeration and rapid filtration through an active bed. For large concentrations of iron (above 10 mgFe l1), sedimentation is added to the process. In enclosed (pressure) systems the water entering the filters has divalent manganese (Mn(II)) and iron either in the form of Fe(OH)3 or unoxidized Fe(II). The effects of iron and manganese removal are governed by catalytic oxidation reactions on the active external layer of the media of the filter beds. In open systems, where more time is available for aeration, and CO2 is more easily removed, the water flow into the filters has iron mostly in the oxidized form (Fe(III)4Fe(II)), and divalent manganese. Iron(III) in water is present as either micro- or macroparticles and therefore the process of flocculation occurring in the filter beds is also important for removal. The oxidation of iron and especially manganese in filter beds has an autocatalytic character. This fact increases the opportunity to use technology for iron and manganese removal that does not require the addition of reagents. This is especially important for manganese removal, because at pH values typical for groundwater following aeration, manganese can be only oxidized catalytically by an active oxidizing filter bed after iron removal has taken place. The reactions for the catalytic oxidation of manganese are shown in Figure 6.

Processes

Oxidation of water by aeration

Hydrolysis of Fe2+, Mn2+ Oxidation Fe2+ → Fe3+

Hydrocomplexes and microflocs of Fe(OH)3

Flocculation Flocs of Fe(OH)3

Separation of Fe(OH)3 flocs in filter bed

Water with iron removed, manganese present as Mn(OH)2

Catalytic oxidation Mn2+ → Mn4+ and removal in active layer of filter bed

Water with Fe and Mn removed

System components Aerators and pipes transporting water to filter

Upper layer of filter bed, where iron removal takes place

Lower active layer of filter bed, where manganese removal takes place

Figure 5 Chemical reactions and processes for iron and manganese removal using aeration and rapid filtration.

462

Chemical Basis for Water Technology Adsorption + Oxidation Reduction

Mn(IV)

Mn(II) Stage I:

Mn(OH)2

MnO(OH)2

+

(Water)

Mn(III) =

Active coating in filter bed

Mn2O3 + 2H2O Growth and reduction of active coating

Oxidation Mn(III) Mn2O3

Stage II:

Mn(IV) +

1 O 2 2

Growth and reduction of active coating

+

2H2O

=

(Water)

2MnO(OH)2 Active coating in filter bed

Figure 6 Schema for catalytic oxidation of Mn(II) in an active filter bed.

Table 2

Reactions and selected process parameters for the oxidation of Fe(II) and Mn(II)

Oxidant

Reaction

Oxidant demand

Precipitate

Iron removal O2 O3 Cl2 ClO2 KMnO4

4Fe2þ þ O2 þ 10H2O-4Fe(OH)3k þ 8Hþ 2Fe2þ þ O3 þ 5H2O-2Fe(OH)3k þ O2 þ 4Hþ 2Fe2þ þ HOCl þ 5H2O-2Fe(OH)3k þ Cl þ 5Hþ Fe 2þ þ ClO2 þ 3H2 O-FeðOHÞ3 k þ ClO2  þ 3Hþ 3Fe 2þ þ MnO4  þ 2H2 O-3FeðOHÞ3 k þ MnO2 þ 5Hþ

mg/mgFe(II) 0.14 0.43 0.64 1.21 0.94

mg/mgFe(II) 1.9 1.9 1.9 1.9 2.4

Manganese removal O2 O3 Cl2 ClO2 KMnO4

2Mn2þ þ O2 þ 2H2O-2MnO2k þ 4Hþ Mn2þ þ O3 þ H2O-MnO2k þ O2 þ Hþ Mn2þ þ HOCl þ H2O-MnO2k þ Cl þ 3Hþ Mn 2þ þ ClO2 þ 2H2 O-MnO2 k þ 2ClO2  þ 4Hþ 3Mn 2þ þ 2MnO4  þ 2H2 O-5MnO2 k þ 4Hþ

mg/mgMn(II) 0.29 0.88 1.29 2.46 1.92

mg/mgMn(II) 1.58 1.58 1.58 1.58 2.64

In most practical situations, active oxidizing beds are filtration media (most frequently quartz) whose grains are coated with a permanent layer of MnO2. The compounds have amphoteric characteristics, and therefore at pH values lower than their iso-electric points (in the range from approximately 7.5 to 8.5) they take on a proton to become positively charged, and above that point they are negatively charged. In the usual pH range of aerated groundwaters, hydrated iron oxides are generally positively charged. Because the surface of quartz filter media in this pH range has a negative charge, iron oxides tend to adsorb and flocculate on these beds. In the filter beds used for iron and manganese removal, two characteristic zones are formed (Figure 5):

•

•

the upper iron removal zone, in which iron removal occurs through catalytic oxidation of Fe(II) and retention of the formed oxides through flocculation of the transported microparticles and particles of iron oxides, and the lower manganese removal zone where catalytic oxidation and retention of manganese through the development of a layer of MnO2 that forms on the media grams occur.

If the pH rises above 9.5 or Eh is increased by the addition of a chemical oxidant, the distinction between the zones disappears. A characteristic feature of the structure of Fe2O3 and MnO2

formed as a result of heterogeneic catalytic oxidation in oxidizing filter beds is their great hydrodynamic stability in comparison with oxides formed under homogeneous conditions (high pH and high Eh). This means that high filtration velocities can be used in these beds. Oxidation using oxygen from the air is determined and limited by its oxidation potential, whose value under atmospheric conditions is defined by

EhðO2 Þ ¼ 1:228  0:0591 pH

ð6Þ

This value is sufficient for the oxidation of divalent iron, whose oxidation under homogeneous conditions requires in general a potential not lower than that given by

EhðFeðIIÞÞ ¼ 1:057  0:1773 pH  0:0591 log½FeðIIÞ

ð7Þ

The value of Eh(O2) is a bit too low for the oxidation of divalent manganese, which under homogeneous conditions requires the addition of a stronger chemical oxidant, such as ozone, chlorine, chlorine dioxide, or potassium permanganate. The reactions for the oxidation of divalent iron and manganese with various oxidants are shown in Table 2 and include the oxidant demand and solids (precipitate) production. It is important to emphasize that the reactions shown

Chemical Basis for Water Technology

in Table 2 are simplified and do not represent the complex mechanism of these processes. The proposed use of chemical oxidation for either new or modernized water-treatment facilities should be evaluated experimentally, in investigations examining the choice of oxidant as well as the effect of process and water-quality parameters (oxidant dose, contact time, water flow rate, TOC, etc.). It is important to know in advance the effect on the process of additional reactions, the need for oxidant dosages above the stoichiometric requirement, the influence of pH and TOC on the rate of oxidation, and the possibility of formation of undesirable odors, especially if chlorine is used. Based on full-scale experience, ozone is the preferred oxidant. The successful use of ozone at the lowest dose depends on changes in the pH of the water, however, requires that the dosage be carefully controlled, because too high a dosage can lead to the oxidation of divalent manganese to permanganate. The presence of TOC, and especially humic substances, limits the use of potassium permanganate as an oxidizing agent. Experimental investigations have shown that the rate of oxidation of divalent iron and manganese is determined primarily by Eh, pH, TOC, concentrations of iron and manganese, temperature, and in waters having low alkalinity, also on the buffering capacity of the water. Technologies based on increasing the pH are often based on the premise that, in addition to iron and manganese removal, softening of the water is required using chemical precipitation. Iron and manganese are removed together with calcium and magnesium because of the high pH of the softening process. In waters where the iron and manganese are originally present in the form of Fe(HCO3)2 or Mn(HCO3)2, the iron and manganese can be removed as carbonate precipitates in the pH range of 8–8.5 by the addition of either lime or soda ash. Technologies based on increasing the pH are more costly than those based on increasing the redox potential, especially if the latter is achieved by aeration. In conclusion, with regard to technologies for iron and manganese removal, it is important to emphasize the great complexity and not completely understood mechanisms of many reactions and processes used to achieve this goal. The complexity is also due to the participation of microorganisms in the process, for example, Gallionella ferruginea, that can use divalent iron and manganese in their metabolism. The biological contribution to iron and manganese removal requires an acclimation time. Arsenic. Arsenic compounds are significant harmful substances that are not detectable by taste in water. In various parts of the world, for example, in eastern India and in Bangladesh, arsenic contamination of groundwater is a major problem. This compound is a semi-metal, appearing primarily in compounds in the oxidation state of As(III) and/or As(V). The natural sources of arsenic are primarily arsenopyrites. Processes for arsenic removal from water include:

• • • •

chemical and biological oxidation of As(III) to As(V), which is more easily removed from water; coagulation and chemical precipitation (co-precipitation); adsorption; and membranes.

463

In practice, chemical oxidation can be achieved using ozone, potassium permanganate, manganese dioxide, and Fenton’s reagent (H2O2/Fe(II)). Coagulation and chemical precipitation can be used successfully for the removal of As(V) by the addition of hydrolyzing iron coagulants (Fe(III)) at low pH values. Effective arsenic removal can also be achieved in groundwater by iron and manganese removal technologies based on aeration and rapid filtration. Increased effectiveness of arsenic removal at increased raw water concentrations of Fe(II) and Mn(II) has been confirmed. For adsorption, the following absorbents have been used: activated carbon, aluminum or iron oxides, and fly ash. pH has been demonstrated to be very important for arsenic removal, and removal is generally greater when pH is less than 7. The best adsorption effects on iron and aluminum oxides have been obtained. Activated carbon has been shown to successfully remove arsenic after being impregnated with copper or Fe(II). Promising results have also been obtained for the adsorption of arsenic on fly ash. Jekel (1994) has discussed arsenic removal and Jekel and co-workers have reported investigations with granular ferric hydroxide (Driehaus et al., 1998; Sperlich et al., 2008). Success has also been reported with NF (Moore et al., 2008), provided the arsenic is in the As(V) form.

3.16.6.7 Maximizing Chemical Stability The chemical stability of water leaving the treatment plant is important so that undesirable reactions do not take place in the distribution system. A major consideration with respect to chemical stability is the calcium carbonate equilibrium. The carbonate or bicarbonate concentrations of the water are typically expressed as the alkalinity. Raw waters, either groundwater or surface water, are normally at equilibrium with respect to calcium carbonate when they enter the treatment facility. This equilibrium can be disrupted by the addition of treatment chemicals such as alum that lower the pH and consume alkalinity. A pH that is too low will result in a water that is referred to as being aggressive. Such a water will dissolve metals and lead to corrosion in distribution system pipes. Lack of chemical stability can have other undesirable effects such as the precipitation of alum flocs (aluminum hydroxide) in the distribution system. Normally, regulations specify the pH range in which the water leaving the treatment plant must lie. (Therefore in some cases if the pH of the raw water is too low, it may need to be raised even if it has not been decreased during treatment.) Sometimes, the tendency of the water to either dissolve or precipitate calcium carbonate is also specified. Such tendency can be measured by a parameter such as the Langelier index, which relates the actual pH to the pH for equilibrium, although other approaches have also been investigated (MWH, 2005: ch. 11). Depending on the extent to which the pH has been reduced during treatment, it may be necessary to raise it again at the end of the process. (Although some treatment processes such as lime softening may increase the pH, such processes are much less commonly used and the usual situation is that the pH will be reduced rather than increased during treatment.) Another important aspect with respect to pH in the distribution system is its impact on the solubility of lead species

464

Chemical Basis for Water Technology

(e.g., Lytle and Schock, 2005). This may be another reason for raising pH during treatment. In general, the addition of chemicals to increase the pH prior to distribution of the water will be done at the end of the treatment process. Thus, this step will have no impact on other treatment steps with the possible exception of disinfectant addition to maintain a residual in the distribution system. In practice, there are only several standard chemicals that are used for pH correction (most commonly sodium hydroxide or lime). Typically, the choice of chemical will be made on the basis of cost and operational considerations. Treatment plant designers are normally able to make these choices on the basis of past experience. Therefore, it is unlikely that there would be a need to experimentally investigate these alternatives as part of treatment process investigations.

3.16.6.8 Maintaining Water Quality to the Consumer’s Tap Hydraulically, the role of the distribution system is to convey water from the treatment plant to the end-user. From the viewpoint of water quality it has been said, somewhat tonguein-cheek, that the distribution system is a large and complex reactor, whose sole purpose is to degrade water quality. While this statement is obviously an exaggeration, it does contain several important messages. The first of these is that it is important to view the distribution system as a reactor in which physical, chemical, and biological processes occur. Unfortunately, in almost all cases these processes have the effect of degrading, rather than the improving, water quality. Minimization of these effects therefore means making the distribution system as inefficient a reactor as possible. An important aspect of this is to make the water leaving the treatment plant as physically, chemically, and biologically stable as possible, to minimize the opportunity for further reaction in the distribution system. This is not defined as a separate goal in addition to the seven; however, it is the reason for the fifth and sixth goals addressed previously (maximizing biological and chemical stability), and is an important contributing reason for the first and second goals (particle removal and TOC removal, respectively). This section addresses the relationship of achieving these goals to maintain distribution system water quality. Several important physical aspects of the distribution system that do have a direct bearing on water quality are pipe materials and pipe diameter and the hydraulic residence time in various parts of the system. The hydraulic residence time or water age is affected by demand in relation to pipe size and by the extent to which dead ends are avoided by looping of pipes. (It should be recalled that a distribution system is probably never at steady state and that water flows at a given point may change direction several times a day. Therefore, it makes sense to speak only of average hydraulic residence times.) In terms of pipe materials, the presence of metallic pipes can mean that corrosion plays a substantial role in water quality. Of course, the presence of older lead pipes or lead connections means that lead levels in water can be unacceptably high. Pipe diameter is important because many reactions affecting water quality take place at the pipe surface. Smaller diameter pipes have a higher surface-to-volume ratio and therefore provide a greater opportunity for such reactions to take place. The fact

that smaller diameter pipes are often located at the ends of the system where water age may be high (and disinfectant residual low) means that growth of biofilms through bacterial utilization of BOM is likely to be highest in these areas. Physical stability of the water leaving the treatment plant is probably the easiest to obtain, compared to biological or chemical stability. If treated water is low in turbidity, this will minimize the deposition of particles in the distribution system. Similarly, minimizing the soluble concentrations of coagulant (e.g., aluminum) will minimize the opportunity for precipitation of alum floc in the system. Chemical stability with respect to the calcium carbonate equilibrium will minimize either precipitation, or dissolution of existing protective scale. As discussed in Section 3.16.6.4, biological stability means avoiding the regrowth of bacteria and the consequent formation of biofilm in the distribution system. As noted previously, various factors can influence the extent to which regrowth may occur. This is discussed in more detail by Huck and Gagnon (2004). In many parts of the world, a disinfectant residual (invariably involving chlorine in some form) is maintained in the distribution system. One of the effects of this is to control regrowth. If a residual is not maintained, or in parts of the system where the residual has declined to a low value, other factors such as the amount of BOM in the water become important with regard to regrowth (Huck and Gagnon, 2004). The degree of biological stability required in the finished water at a given treatment plant depends on the extent to which regrowth may occur in the plant’s distribution system and the method chosen to manage it (i.e., maintaining a residual or not). Maintaining a microbiologically safe water throughout the distribution system requires both adequate disinfection at the treatment plant and the absence of direct contamination of the distribution system. The maintenance of a disinfectant residual is often seen as a way of guarding against such potential contamination. The low residual typically maintained in a distribution system might provide limited protection against a massive contamination incident; however, a loss of residual would indicate that a problem had occurred. Rapid detection of such a problem requires of course that residuals be monitored at various points in the distribution system with sufficient frequency. In systems where a disinfectant residual is maintained, the issue becomes maintaining an appropriate residual to the end of the system without having an excessively high residual entering the system. A high initial residual could lead to unacceptable levels of disinfection by-products, may be esthetically unacceptable, and also carries additional costs. Therefore, the disinfectant (often chlorine) demand of the treated water and of the distribution system itself is important. The chlorine demand of the treated water will be reduced by measures taken to reduce TOC levels (goal 2). The chlorine demand of the system itself is a function of pipe material, pipe surface area, and the presence of deposits and/or biofilm on the pipes. Replacement of metallic with nonmetallic pipes and minimization of biofilm growth will reduce this demand. Reducing water age by appropriate hydraulic measures will facilitate maintaining a residual to the end of the system. In some jurisdictions, chloramines rather than free chlorine are used to provide a more stable distribution system residual.

Chemical Basis for Water Technology

For systems where a disinfectant residual is not maintained, microbial safety of the distributed water requires a system of high physical integrity, as well as vigilant monitoring and response to detect and address any intrusion of contamination. Corrosion of metallic pipes is a function of water quality. As stated previously, lead levels can be a problem where lead materials are present in the distribution system pipes, connections, or appurtenances. Both iron-based and copper pipes are susceptible to corrosion. For many waters, the issue of corrosion can be addressed by maintaining a slight precipitating tendency with respect to the calcium carbonate equilibrium. This is often expressed through an appropriate positive value of the Langelier index. For other waters, other indices may be required. pH is a master variable with respect to the various equilibria involved and therefore corrosion control involves maintaining a sufficiently high pH. These matters relate to the chemical stability of the water, which has been discussed in Section 3.16.6.7. The corrosion of metallic pipe material can also create esthetic problems because of the presence of iron precipitates. Other esthetic problems may arise because of the formation of taste and odor in the distribution system. In certain cases, pipe materials may impart a metallic taste. Odors may arise because of the growth of certain types of microorganisms. In general, this issue can be addressed by maintaining biological stability. Although the discussion above has been in the context of distribution system pipes, it is important to recognize that treated water-storage reservoirs are important distribution system components. It is necessary to ensure that there is no possibility for contamination of the water in these reservoirs, and also that they are operated in such a manner as to avoid an excessively high water age. Although distribution system maintenance issues are outside the scope of this chapter, it should be noted that regular distribution system flushing is important for maintaining water quality. In addition, an appropriate pipe replacement program is crucial for maintaining this important element of municipal infrastructure. In summary, in the absence of accidental or deliberate contamination incidents, maintaining water quality to the consumer’s tap primarily involves maximizing the physical, chemical, and biological stability of the water leaving the treatment plant.

3.16.7 Summary (Concluding Remarks) The development of water treatment as a self-standing discipline has taken place over approximately the past 100 years. This development has not taken place evenly, but rather has occurred as a series of waves. An example of a very recent period of rapid change is the emergence of the use of UV radiation for disinfection. The speed with which this technology has come into use in water treatment is almost unparalleled in what is by nature a conservative discipline. Advances in the discipline of water treatment depend on a sound knowledge of important fundamental principles and an ability to apply them to produce very practical outcomes.

465

Treatment is one element of the overall system required for the provision of safe drinking water. Within the context of an appropriate regulatory framework, this system can be considered to consist of five elements: a good source, effective treatment, secure distribution, appropriate monitoring, and an appropriate and timely response to an adverse monitoring result. To ensure public health protection, the system for supplying drinking water must be robust. The concept of robustness extends the so-called multibarrier principle to include the human and institutional elements of the system as well as the physical ones. It is useful to approach water treatment from the point of view of the goals that treatment must meet. It is possible to identify seven such goals: particle (and associated pathogen) removal, removal of TOC, disinfection/inactivation, removal of chemical contaminants, achievement of biological stability, achievement of chemical stability, and achievement of an appropriate aesthetic quality. While all of these goals are important, disinfection/inactivation is paramount for protecting public health from acute risk. The complexity of treatment required in a given situation depends on the raw water quality, that is, on the number of goals that treatment must address for a given water. Some goals (e.g., disinfection/inactivation) are normally achieved through a single treatment process, whereas others may be addressed by several processes in the treatment system or train. There is often a choice of processes to address a particular goal. Maintaining water quality in the distribution system to the consumer’s tap is important but is not an explicit treatment goal because it is as much dependent on the physical layout and quality of the distribution system as it is on treatment itself. Effective treatment should minimize the risk to the consumer, be optimized and be sustainable. The latter point implies a preference for physical and biological processes and the minimization of chemical additions and treatment residuals such as sludges. Investigations of water-treatment processes can range from very fundamental to very applied. As an example, in an applied investigation, the hypothesis being tested may be: will a membrane work for the treatment of this particular water? Investigations may be conducted at various scales: laboratory, pilot, and demonstration (effectively full) scale. For investigations, proper application of the general principles of experimental design is important, with some special considerations for water treatment, such as testing several processes in parallel to ensure that the impact of any changes in raw water quality will be experienced by all processes. In practical investigations, analysis of the results must include an estimate of the capital and operating costs of the processes or systems capable of meeting the treatment objectives. This chapter briefly discusses the seven previously identified goals for water treatment and important chemical and physical principles involved, and summarizes the major processes used, with a focus on municipal (public) water supplies. For a detailed discussion of the processes, the reader less familiar with the processes may wish to consult standard environmental engineering works. The second half of the chapter directly links goals and processes, providing a critical review of the processes that can be used to meet a particular goal. In particular situations, achieving the goal may require contributions from several processes in the treatment train, and of

466

Chemical Basis for Water Technology

course some processes (e.g., oxidation) can contribute to more than one goal. Achievement of the first goal, the removal of particles (including the physical removal of pathogenic microorganisms), is normally required for surface waters and for groundwater under the direct influence of surface water. Classically, particle removal has been achieved by a series of processes (coagulation, flocculation, and normally either sedimentation or flotation) culminating in granular media filtration. More recently, low-pressure membranes have come to replace granular media filtration even in large treatment plants. The optimal pretreatment required for membranes will probably be different than that for granular media filtration. In some cases, a low-pressure membrane used for particle removal may be able to operate without pretreatment. TOC removal may be required to reduce disinfection byproduct formation, to increase the stability of disinfectant residuals, and possibly to reduce membrane fouling. Although various approaches can be used to remove TOC, coagulation at low pH (approximately six or less) can be very useful, especially for waters of high humic content. Biological processes can reduce concentrations of biodegradable TOC. The biological stability of a water leaving the treatment plant is important to maintain water quality in the distribution system and minimize bacterial regrowth. Carbon is usually considered to be the limiting nutrient in drinking water and therefore biological stability is enhanced by the removal of BOM during treatment. Biological filtration, involving the growth of bacteria as a biofilm on the filter media, is the process normally used to reduce levels of BOM. Biological processes also play an important role in slow sand filtration, riverbank filtration, and underground passage of water following treatment. Biofiltration is not required for most groundwaters, which already contain lower levels of BOM. Biological processes must occur upstream of chlorination. Ozonation will increase the level of easily biodegradable material. Therefore although ozonation prior to biofiltration is helpful, it also means that if ozonation is used, it should be followed by biofiltration. Treatment should seek to minimize BOM levels (i.e., maximize biostability) of the finished water while meeting the other treatment objectives. The goal of reducing chemical contamination is water specific, in that the need to address this goal depends on the extent to which chemical contamination of the raw water exists. As examples of organic chemical contaminants that are important for the goal of addressing esthetic quality, the removal of the common odorous compounds, geosmin and MIB, is addressed in considerable detail to illustrate the three major processes that can be used (oxidation, adsorption on activated carbon, oxidation, and biodegradation) and to permit discussion of important water quality and other factors affecting the performance of these processes in specific situations. Many of these factors are relevant for the removal of other chemical contaminants by these processes. Removal technologies for the important inorganic substances, iron, manganese, and arsenic are also reviewed. The goal of ensuring chemical stability is important to minimize either precipitation in the distribution system or corrosion of metal distribution system piping and household

piping and fixtures. pH plays an important role in chemical stability and is also important from the perspective of controlling the concentration of substances such as lead at the consumer’s tap. The goal of esthetic quality is important because consumers want water that tastes and smells well. Odor is often more of a problem than taste. Although many odorous substances present in raw water are of natural origin and are not generally considered to have adverse health effects, it is difficult to convince people that the water is safe to drink if it tastes or smells bad. While not a specific treatment goal, the maintenance of water quality to the consumer’s tap is a very important issue. Water quality rarely, if ever, improves in the distribution system; rather, it normally degrades. Quality degradation is normally most evident at long residence times and is associated with the decay of the disinfectant residual. (The maintenance of a disinfectant residual is the approach used in much of the world to counter biological instability and maintain microbiological quality in the distribution system.) The extent of quality degradation is a function of the chemical, biological, and physical instability of the water leaving the treatment plant. However, the age, physical condition, and layout of the distribution system are also significant factors. From a treatment perspective, minimizing quality degradation in the distribution system involves making it as inefficient a reactor as possible, by maximizing the biological, chemical, and physical stability of the water leaving the treatment plant. Although some goals (e.g., disinfection) can be achieved by a single treatment process, others, such as TOC removal, may be accomplished by more than one process. For high-quality groundwater, the only treatment goal required may be disinfection. However for some groundwaters and for virtually all surface waters, more than one treatment goal must be addressed. Therefore, treatment processes must be combined into systems. In general, there is a logical sequence in which the processes would be combined, as discussed in the text. For example, particle removal of normally occurs early in the treatment train. In conclusion, the vital importance of high-quality and reliably available drinking water for the development and maintenance of society cannot be overemphasized. The design of robust and sustainable systems for the provision of such water must be based on the sound application of the relevant fundamental principles. Treatment is a vital component of such systems. The treatment goals that must be addressed for a particular water must be defined, and the processes required to meet those goals determined and optimally implemented.

References Adin A and Rebhun M (1977) A model to predict concentration and head-loss profiles in filtration. Journal of American Water Works Association 8: 444--453. Ashitani K, Hishida Y, and Fujiwara K (1988) Behaviour of musty odorous compounds during the process of water treatment. Water Science and Technology 20(8/9): 261. AWWARF/LE (American Water Works Association Research Foundation/Lyonnaise des Eaux) (1987) Identification of tastes and odors in drinking water. AWWA Research Foundation Report No. 90518, 292pp. Denver, CO: AWWA.

Chemical Basis for Water Technology

AWWARF/LE (American Water Works Association Research Foundation/Lyonnaise des Eaux) (1995) Advances in taste-and-odor treatment and control. AWWA Research Foundation Report, 385pp. (ISBN 0-89867-744-0). Denver, CO: AWWA. Bellar TA, Lichtenberg JJ, and Kroner RC (1974) The occurrence of organohalides in chlorinated drinking water. Journal of American Water Works Association 66: 703--706. Bellona C, Drewes JE, Xu P, and Amy G (2004) Factors affecting the rejection of organic solutes during NF/RO treatment – a literature review. Water Research 38: 2795--2809. Betancourt WQ and Rose JB (2004) Drinking water treatment processes for removal of Cryptosporidium and Giardia. Veterinary Parasitology 126(1–2): 219--234. Boyer TH and Singer PC (2005) Bench-scale testing of a magnetic ion exchange resin for removal of disinfection by-product precursors. Water Research 39: 1265--1276. Bruchet A and Duguet JP (2004) Role of oxidants and disinfectants on the removal, masking and generation of tastes and odours. Water Science and Technology 49(9): 297--306. Carr SA and Baird RB (2000) Mineralization as a mechanism for TOC removal: Study of ozone/ozone-peroxide oxidation using FT-IR. Water Research 34(16): 4036--4048. Carter MC and Weber WJ, Jr. (1994) Modeling adsorption of TCE by activated carbon preloaded by background organic matter. Environmental Science and Technology 28(4): 614--623. Chen G, Dussert BW, and Suffet IH (1997) Evaluation of granular activated carbons for removal of methylisoborneol to below odor threshold concentration in drinking water. Water Research 31(5): 1155--1163. Comerton AM, Andrews RC, and Bagley DM (2009) The influence of natural organic matter and cations on the rejection of endocrine disrupting and pharmaceutically active compounds by nanofiltration. Water Research 43(3): 613--622. Cook D, Newcombe G, and Sztajnbok P (2001) The application of powdered activated carbon for MIB and geosmin removal: Predicting PAC doses in four raw waters. Water Research 35(5): 1325--1333. Crosina Q, Peldszus S, and Huck PM (2006) The degradation efficiency of select PPCPs using UV and UV/H2O2 – degradation kinetics and application to pretreated water. In: Proceedings, American Water Works Association Water Quality Technology Conference and Exposition (WQTC), CD-ROM, paper Tues11-4. Denver, CO, USA. Croue´ JP, Violleau D, Bogaire C, and Legube B (1999) Removal of hydrophobic and hydrophilic constituents by anion exchange resin. In: Ødegaard H (ed.) Proceedings, International IAWQ–IWSA Joint Specialist Conference on Removal of Humic Substances from Water, p. 219. Trondheim, Norway, 24–26 June. Crozes G, Hagstrom J, Suffet IH, and Young C (1999) Bench-scale evaluation of adsorptive processes for taste and odors control using rapid small-scale column tests and flavor profile analysis. Water Science and Technology 40(6): 39--44. Dietrich AM, Hoehn RC, Dufresne LC, Buffin LW, Rashash DMC, and Parker BC (1995) Oxidation of odorous and nonodorous algal metabolites by permanganate, chlorine, and chlorine dioxide. Water Science and Technology 31(11): 223--228. Driehaus W, Jekel M, and Hildebrandt U (1998) Granular ferric hydroxide – a new adsorbent for the removal of arsenic from natural water. Journal of Water Supply: Research and Technology – AQUA 47(1): 30--35. Edwards M (1997) Predicting DOC removal during enhanced coagulation. Journal of American Water Works Association 89(5): 78--89. Edzwald JK (1993) Coagulation in drinking water treatment: Particles, organics and coagulants. Water Science and Technology 27(11): 21--35. Edzwald JK (1995) Principles and applications of dissolved air flotation. Water Science and Technology 31(3/4): 1--23. Edzwald JK (2010) Dissolved air flotation and me. Water Research 44(7): 2077--2106. Edzwald JK Tobiason JE (1999) Enhanced coagulation: US requirements and a broader view. Water Science and Technology 40(9): 63--70. Elhadi SLN (2004) Removal of earthy/musty odour compounds from drinking water by biological filtration. Doctoral thesis, Department of Civil Engineering, University of Waterloo. Elhadi SLN, Huck PM, and Slawson RM (2004) Removal of geosmin and 2-methylisoborneol by biological filtration. Water Science and Technology 49(9): 273--280. Elhadi SLN, Huck PM, and Slawson RM (2006) Factors affecting the removal of geosmin and MIB in drinking water biofilters. Journal of American Water Works Association 98(8): 108--118. Eriksson P (1988) Nanofiltration extends the range of membrane filtration. Environmental Progress 7(1): 58--62. Ferguson DW, McGuire MJ, Koch B, Wolfe RL, and Aieta EM (1990) Comparing PEROXONE and ozone for controlling taste and odor compounds, disinfection by-products, and microorganisms. Journal of American Water Works Association 82(4): 181--191.

467

Fettig J (1999) Removal of humic substances by adsorption/ion exchange. In: Ødegaard H (ed.) Proceedings, International IAWQ–IWSA Joint Specialist Conference on Removal of Humic Substances from Water, p. 185. Trondheim, Norway, 24–26 June. Frias J, Ribas F, and Lucena F (1992) A method for the measurement of biodegradable organic carbon in waters. Water Research 26(2): 255--258. Gammie L (1987) Personal communication. Buffalo Pound Water Administration Board, Moose Jaw, Saskatchewan. Gillogly TET, Snoeyink VL, Holthouse A, Wilson CM, and Royal EP (1998) Effect of chlorine on PAC’s ability to adsorb MIB. Journal of American Water Works Association 90(2): 107--114. Gillogly TET, Snoeyink VL, Newcombe G, and Elarde JR (1999) A simplified method to determine the powdered activated carbon dose required to remove methylisoborneol. Water Science and Technology 40(6): 59--64. Glaze WH, Schep R, Chauncey W, et al. (1990) Evaluating oxidants for the removal of model taste and odor compounds from a municipal water supply. Journal of American Water Works Association 82(5): 79--84. Graham M, Najm I, Simpson M, MacLeod B, Summers RS, and Cummings L (2000b) Optimization of powdered activate carbon application for geosmin and MIB removal. AWWARF Report, ISBN 1-58321-019-9. Denver, CO: AWWA. Graham MR, Summers RS, Simpson MR, and MacLeod BW (2000a) Modeling equilibrium adsorption of s2-methylisoborneol and geosmin in natural waters. Water Research 34(8): 2291--2300. Gray SL (2006) Removal of conventional & novel taste & odour compounds using UV photolysis & UV/H2O2 advanced oxidation. Master’s thesis, Department of Civil Engineering, University of Waterloo. Haas CN, Rose JB, and Gerba CP (1999) Quantitative Microbial Risk Assessment. New York, NY: Wiley. Haberkamp J (2008) Organisches Membranfouling bei der Ultrafiltration kommunaler Kla¨ranlagenabla¨ufe-Ursachen, Mechanismen und MaXnahmen zur Verringerung. Ph.D., Technische Universita¨t Berlin, Chair of Water Quality Control, Berlin, Germany. Haddix PL, Shaw NJ, and LeChevallier MW (2004) Characterization of bioluminescent derivatives of assimilable organic carbon test bacteria. Applied and Environmental Microbiology 70: 850--854. Halle´ C (2010) Biofiltration in drinking water treatment: Reduction of membrane fouling and biodegradation of organic trace contaminants. Doctoral Thesis, Department of Civil and Environmental Engineering, University of Waterloo. Halle´ C, Huck PM, Peldszus S, Haberkamp J, and Jekel M (2009) Assessing the performance of biological filtration as pretreatment to low pressure membranes for drinking water. Environmental Science and Technology 43(10): 3878--3884. Hammes FA and Egli T (2005) New method for assimilable organic carbon determination using flow-cytometric enumeration and a natural microbial consortium as inoculum. Environmental Science and Technology 39(9): 3289--3294. Hepplewhite C, Newcombe G, and Knappe DRU (2004) NOM and MIB, who wins in the competition for activated carbon adsorption sites? Water Science and Technology 49(9): 257--265. Hereit F (1973) Matematicky model provypocˇet filtracˇniho cyklu voda´renskych rychlofiltru. Vodni Hospodarˇstvi, ˇrada B 8: 193--196. Hijnen WAM, Biraud D, Cornelissen ER, and van der Kooij D (2009) Threshold concentration of easily assimilable organic carbon in feedwater for biofouling of spiral-wound membranes. Environmental Science and Technology 43(13): 4890--4895. Ho L, Croue´ J-P, and Newcombe G (2004) The effect of water quality and NOM character on the ozonation of MIB and geosmin. Water Science and Technology 49(9): 249--255. Ho L, Hoefel D, Aunkofer W, et al. (2006) Biological filtration for the removal of algal metabolites from drinking water. Water Science and Technology: Water Supply 6(2): 153--159. Ho L, Hoefel D, Bock F, Saint CP, and Newcombe G (2007) Biodegradation rates of 2-methylisoborneol (MIB) and geosmin through sand filters and in bioreactors. Chemosphere 66(11): 2210--2218. Ho LSW (2004). The removal of cyanobacterial metabolites from drinking water using ozone and granular activated carbon. Doctoral Thesis, University of South Australia. Hoefel D, Ho L, Aunkofer W, et al. (2006) Cooperative biodegradation of geosmin by a consortium comprising three gram-negative bacteria isolated from the biofilm of a sand filter column. Letters in Applied Microbiology 43: 417--423. Hongve D, Baan J, Becher G, and Beckman OA (1999) Experiences from operation and regeneration of an anionic exchanger for NOM removal. In: Ødegaard H (ed.) Proceedings, International IAWQ–IWSA Joint Specialist Conference on Removal of Humic Substances from Water, p. 227. Trondheim, Norway, 24–26 June.

468

Chemical Basis for Water Technology

Hrudey SE and Hrudey EJ (2004) Safe Drinking Water: Lessons from Recent Outbreaks in Affluent Nations. London: IWA. Hrudey SE, Huck PM, Mitton MJ, and Kenefick SL (1995) Evaluation of odour removal by pilot-scale biological treatment process trains during spring runoff in an icecovered river. Water Science and Technology 31(11): 195--201. Huber MM, Canonica S, Park GY, and Von Gunten U (2003) Oxidation of pharmaceuticals during ozonation and advanced oxidation processes. Environmental Science and Technology 37: 1016--1024. Huck PM (1990) Measurement of biodegradable organic matter and bacterial growth potential in drinking water. Journal of American Water Works Association 82(7): 78--86. Huck PM (1999) Development of a framework for quantifying the removal of humic substances by biological filtration. Water Science and Technology 40(9): 149--156. Huck PM and Coffey BM (2004) The importance of robustness in drinking-water systems. Journal of Toxicology and Environmental Health, Part A 67(20–22): 1581--1590. Huck PM, Coffey BM, Amirtharajah A, and Bouwer EJ (2000) Optimizing filtration in biological filters. In: Proceedings, American Water Works Association Research Foundation and American Water Works Association, 268pp. (ISBN 1-58321-0652). Denver, CO, USA. Huck PM, Coffey BM, Emelko MB, et al. (2002) Effects of filter operation on Cryptosporidium removal. Journal of American Water Works Association 94(6): 97--111. Huck PM and Gagnon GA (2004) Understanding the distribution system as a bioreactor: A framework for managing HPC levels. International Journal of Food Microbiology 92(3): 347--353. Huck PM, Kenefick SL, Hrudey SE, and Zhang S (1995) Bench-scale determination of the removal of odour compounds with biological treatment. Water Science and Technology 31(11): 203--209. Huck PM, Peldszus S, and Halle´ C, et al (2009) Pilot scale evaluation of biofiltration as an innovative pre-treatment for UF membranes for drinking water treatment. Journal of Water Science and Technology/Water Supply (accepted). Huck PM and Sozan´ski MM (2008) Biological filtration for membrane pre-treatment and other applications: Towards the development of a practically-oriented performance parameter. Journal of Water Supply: Research and Technology – AQUA 57(4): 203--224. Huck PM, Zhang S, and Price ML (1994) BOM removal during biological treatment: A first order model. Journal of the American Water Works Association 86(6): 61--71. Ives KJ (1960) Rational design of filters. Institution of Civil Engineers 16: 189--193. Iwasaki T (1937) Some notes on sand filtration. Journal of American Water Works Association 29: 1591--1602. Jacangelo JG, Adham SS, and Laine J-M (1995a) Mechanism of Cryptosporidium, Giardia, and MS2 virus removal by MF and UF. Journal of American Water Works Association 87(9): 107--121. Jacangelo JG, DeMarco J, Owen DM, and Randtke SJ (1995b) Selected processes for removing NOM: An overview. Journal of American Water Works Association 87(1): 64--77. Jarvie ME, Hand DW, Bhuvendralingam S, Crittenden JC, and Hokanson DR (2005) Simulating the performance of fixed-bed granular activated carbon adsorbers: Removal of synthetic organic chemicals in the presence of background organic matter. Water Research 39: 2407--2421. Jekel M (1994) Arsenic removal in water treatment. In: Nriagu J (ed.) Arsenic in the Environment, pp. 433–446. Jung S-W, Baek K-H, and Yu M-J (2004) Treatment of taste and odor material by oxidation and adsorption. Water Science and Technology 49(9): 289--295. Ju¨ttner F (1995) Elimination of terpenoid odorous compounds by slow sand and river bank filtration of the Ruhr River, Germany. Water Science and Technology 31(11): 211--217. Ju¨ttner F (1999) Efficacy of bank filtration for the removal of fragrance compounds and aromatic hydrocarbons. Water Science and Technology 40(6): 123--128. Kaplan LA, Ribas F, Joret JC, Volk C, Frias J, and Lucena F (1993) An immobilizedbiofilm reactor for the measurement of biodegradable organic matter in drinking water: Validation and field test. In: Proceedings, American Water Works Association Water Quality Technology Conference and Exposition (WQTC), pp. 1457–1475. Toronto, ON, USA. Kavanaugh MC and Trussell RR (1980) Design of aeration towers to strip volatile contaminants from drinking water. Journal of American Water Works Association 72(12): 684--692. Kim T-U, Drewes JE, Summers RS, and Amy GL (2007) Solute transport model for trace organic neutral and charged compounds through nanofiltration and reverse osmosis membranes. Water Research 41: 3977--3988. Kim Y, Lee Y, Gee CS, and Choi E (1997) Treatment of taste and odor causing substances in drinking water. Water Science and Technology 35(8): 29--36.

Lange R, Tattersall J, Botica C, and Smith P (2001) Design and construction of the world’s first large scale MIEXs water treatment plant. In: Proceedings, Australian Water Association Biennial Conference. Lauderdale CV, Aldrich HC, and Lindner AS (2004) Isolation and characterization of a bacterium capable of removing taste- and odor-causing 2-methylisoborneol from water. Water Research 38: 4135--4142. Li L, Quinlivan PA, and Knappe DRU (2005) Predicting adsorption isotherms for aqueous organic micropollutants from activated carbon and pollutant properties. Environmental Science and Technology 39(9): 3393--3400. Lytle DA and Schock MR (2005) Formation of Pb(IV) oxides in chlorinated water. Journal of American Water Works Association 97(11): 102--114. MacKenzie JA, Tennant MF, and Mazyck DW (2005) Tailored GAC for the effective control of 2-methylisoborneol. Journal of American Water Works Association 97(6): 76--87. Makdissy G, Peldszus S, McPhail R, and Huck PM (2007) Towards a mechanistic understanding of the impact of fouling on the removal of EDCs/PPCPs by nanofiltration membranes. In: Proceedings, American Water Works Association Water Quality Technology Conference and Exposition (WQTC), CD-ROM, extended abstract, Wed5-7, 5pp. Charlotte, NC. Mergen MRD, Jarvis P, Jefferson B, Nguyen H, and Parsons SA (2006) Impact of continuous MIEXs use on downstream water treatment processes. In: Proceedings, American Water Works Association Water Quality Technology Conference and Exposition (WQTC), 13pp. Denver, CO, USA. Mergen MRD, Jefferson B, Parsons SA, and Jarvis P (2008) Magnetic ion-exchange resin treatment: Impact of water type and resin use. Water Research 42: 1977--1988. Metz DH, Pohlman RC, Vogt J, and Summers RS (2006) Removal of MIB and geosmin by full-scale biological sand filters. In: Gimbel R, Graham NJD, Collins MR (eds.) Recent Progress in Slow Sand and Alternative Biofiltration Processes, pp. 352–359 (ISBN 9781843391203). London: IWA. Miettinen IT, Vartiainen T, and Martikainen PJ (1997) Phosphorus and bacterial growth in drinking water. Applied and Environmental Microbiology 63(8): 3242--3245. Mints DM (1966) Modern theory of filtration, special subject No. 10. In: Proceedings, International Water Supply Congress, W.1, 3-7. Barcelona. London: IWSA. Moore KW, Huck PM, and Siverns S (2008) Arsenic removal using oxidative media and nanofiltration. Journal of American Water Works Association 100(12): 74--83. Morioka T, Motoyama N, Hoshikawa H, Murakami A, Okada M, and Moniwa T (1993) Kinetic analysis on the effects of dissolved inorganic and organic substances in raw water on the ozonation of geosmin and 2-MIB. Ozone Science and Engineering 15: 1--18. MWH (2005) Water Treatment: Principles and Design, 2nd edn. Hoboken, NJ: Wiley. Ndiongue S, Anderson WB, Tadwalkar A, Rudnickas J, Lin M, and Huck PM (2006) Using pilot-scale investigations to estimate the remaining geosmin and MIB removal capacity of full-scale GAC-capped drinking water filters. Water Quality Research Journal of Canada 41(3): 296--306. Nerenberg R, Rittmann BE, and Soucie WJ (2000) Ozone/biofiltration for removing MIB and geosmin. Journal of American Water Works Association 92(12): 85--95. Newcombe G and Cook D (2002) Influences on the removal of tastes and odours by PAC. Journal of Water Supply: Research and Technology – AQUA 51(8): 463--474. Newcombe G, Donati C, Drikas M, and Hayes R (1994) Adsorption onto activated carbon: electrostatic and non-electrostatic interactions. Water Supply 14: 129. Newcombe G, Drikas M, and Hayes R (1997) Influence of characterised natural organic material on activated carbon adsorption: II. Effect on pore volume distribution and adsorption of 2-methylisoborneol. Water Research 31(5): 1065--1073. Nieminski EC and Ongerth JE (1995) Removing Giardia and Cryptosporidium by conventional treatment and direct filtration. Journal of American Water Works Association 87(9): 96--106. Ødegaard H, Eikebrokk B, and Storhaug R (1999) Processes for the removal of humic substances from water – an overview based on Norwegian experiences. Water Science and Technology 40(9): 37--46. Oestman E, Schweitzer L, Tomboulian P, Corado A, and Suffet IH (2004) Effects of chlorine and chloramines on earthy and musty odors in drinking water. Water Science and Technology 49(9): 153--159. O’Melia CR and Stumm W (1967) Theory of water filtration. Journal of American Water Works Association 59(11): 1393--1412. Pirbazari M, Ravindran V, Badriyha BN, Craig S, and McGuire MJ (1993) GAC adsorber design protocol for the removal of off-flavors. Water Research 27(7): 1153--1166. Pre´vost M, Laurent P, Servais P, and Joret J-C (eds.) (2005) Biodegradable Organic Matter in Drinking Water Treatment and Distribution, p. 300pp. Denver, CO: AWWA. Rittmann BC and Huck PM (1989) Biological treatment of public water supplies. CRC Critical Reviews in Environmental Control 19: 119--184.

Chemical Basis for Water Technology

Rook JJ (1974) Formation of haloforms during chlorination of natural waters. Journal of Water Treatment and Examination 23: 234--243. Rosenfeldt EJ, Melcher B, and Linden KG (2005) UV and UV/H2O2 treatment of methylisoborneol (MIB) and geosmin in water. Journal of Water Supply: Research and Technology – AQUA 54(7): 423--434. Sagehashi M, Shiraishi K, Fujita H, Fujii T, and Sakoda A (2005) Adsorptive ozonation of 2-methylisoborneol in natural water with preventing bromate formation. Water Research 39: 3900--3908. Saito A, Tokuyama T, Tanaka A, Oritani T, and Fuchigami K (1999) Microbiological degradation of (–)-geosmin. Water Research 33(13): 3033--3036. Singer PC, Schneider M, Edwards-Brandt J, and Budd GC (2007) MIEX for removal of DBP precursors: Pilot-plant findings. Journal of American Water Works Association 99(4): 128--139. Slunjski M, Cadee K, and Tattersal J (2000) MIEXs resin water treatment process. In: Proceedings, American Water Works Association Annual Conference. Denver, CO, USA, June. Son HJ, Hwang YD, Roh JS, et al. (2005) Application of MIEX pre-treatment for ultrafiltration membrane process for NOM removal and fouling reduction. Water Supply 5(5): 15--24. Sontheimer H, Summers RS, and Crittenden JC (1988) Activated carbon for water treatment. (Second edition in English of Adsorptionsverfahren zur Wasserreinigung.) DVGW –Forschungsstelle, distributed in United States by American Water Works Association Research Foundation, Denver, CO. Sperlich A, Werner A, Genz A, Amy G, Worch E, and Jekel M (2005) Breakthrough behavior of granular ferric hydroxide (GFH) fixed-bed adsorption filters: Modeling and experimental approaches. Water Research 39: 1190--1198. Stumm W and O’Melia CR (1968) Stoichiometry of coagulation. Journal of American Water Works Association 60(5): 514--539. Sugiura N, Isoda H, and Maekawa T (2003) Degradation potential of musty odour in a drinking water source by a biofilm method. Journal of Water Supply: Research and Technology – AQUA 52(3): 181--187. Summers RS, Chae S, Kim SM, and Ahn HW (2006) Biodegradation of MIB and geosmin in biological sand and BAC filters: Acclimation, steady-state and varying influent. In: Gimbel R, Graham NJD, Collins MR (eds.) Recent Progress in Slow Sand and Alternative Biofiltration Processes, pp. 369–372 (ISBN 9781843391203). London: IWA. Tanaka A, Oritani T, Uehara F, et al. (1996) Biodegradation of a musty odour component, 2-methylisoborneol. Water Research 30(3): 759--761. Teermann I and Jekel M (1999) Adsorption of humic substances onto a-FeOOH and its chemical regeneration. In: Ødegaard H (ed.) Proceedings, International IAWQ– IWSA Joint Specialist Conference on Removal of Humic Substances from Water, p. 211. Norway: Trondheim. 24–26 June. Terauchi N, Ohtani T, Yamanaka K, Tsuji T, Sudou T, and Ito K (1995) Studies on a biological filter for musty odor removal in drinking water treatment processes. Water Science and Technology 31(11): 229--235. Thomson J, Roddick F, and Drikas M (2004) Vacuum ultraviolet irradiation for natural organic matter removal. Journal of Water Supply: Research and Technology – AQUA 53(4): 193--206. Tung S-C, Lin T-F, Liu C-L, and Lai S-D (2004) The effect of oxidants on 2-MIB concentration with the presence of cyanobacteria. Water Science and Technology 49(9): 281--288. Uhl W, Persson F, Heinicke G, Hermansson M, and Hedberg T (2006) Removal of geosmin and MIB in biofilters – on the role of biodegradation and adsorption. In: Gimbel R, Graham NJD, Collins MR (eds.) Recent Progress in Slow Sand and Alternative Biofiltration Processes, pp. 360–368 (ISBN 9781843391203). London: IWA. Urfer D and Huck PM (1997) Effects of hydrogen peroxide residuals on biologically active filters. Ozone Science and Engineering 19(4): 371--386. Urfer D, Huck PM, Booth SDJ, and Coffey BM (1997) Biological filtration for BOM and particle removal: A critical review. Journal of American Water Works Association 89(12): 83--98. Urfer D, Huck PM, Gagnon GA, Mutti D, and Smith F (1999) Modeling enhanced coagulation to improve ozone disinfection. Journal of American Water Works Association 91(3): 59--73. Valade MT, Becker WC, and Edzwald JK (2009) Treatment selection guidelines for particle and NOM removal. Journal of Water Supply: Research and Technology – AQUA 58(6): 424--432. Van Benschoten JE and Edzwald JK (1990) Chemical aspects of coagulation using aluminum salts – I. Hydrolytic reactions of alum and polyaluminum chloride. Water Research 24(12): 1519--1526.

469

van der Kooij D (2000) Biological stability: A multidimensional quality aspect of treated water. Water, Air, and Soil Pollution 123(1–4): 25--34. van der Kooij D, Veenendaal HR, Baars-Lorist C, van der Klift DW, and Drost YC (1995) Biofilm formation on surfaces of glass and Teflon exposed to treated water. Water Research 29(7): 1655--1662. van der Kooij D, Visser A, and Hijnen WAM (1982) Determining the concentration of easily assimilable organic carbon in drinking water. Journal of American Water Works Association 74(10): 540--545. Verliefde ARD, Cornelissen ER, Heijman SGJ, et al. (2009a) Influence of solute – membrane affinity on rejection of unchanged organic solutes by nanofiltration membranes. Environmental Science and Technology 43: 2400--2406. Verliefde ARD, Cornelissen ER, Heijman SGJ, et al. (2009b) Influence of membrane fouling by (pretreated) surface water on rejection of pharmaceutically active compounds (PhACs) by nanofiltration membranes. Journal of Membrane Science 330: 90--103. Von Gunten U (2003a) Ozonation of drinking water: Part I. Oxidation kinetics and product formation. Water Research 37(7): 1443--1467. Von Gunten U (2003b) Ozonation of drinking water: Part II. Disinfection and byproduct formation in presence of bromide, iodide or chlorine. Water Research 37(7): 1469--1487. Vrijenhoek EM, Childress AE, Elimelech M, Tanaka TS, and Beuhler MD (1998) Removing particles THM precursors by enhanced coagulation. Journal of American Water Works Association 90(4): 139--150. Weiss WJ, Bouwer EJ, Aboytes R, et al. (2005) Riverbank filtration for control of microorganisms: Results from field monitoring. Water Research 39(10): 1990--2001. Wert EC, Edwards-Brandt JC, Singer PC, and Budd GC (2005) Evaluating magnetic ion exchange resin (MIEX)s pretreatment to increase ozone disinfection and reduce bromate formation. Ozone: Science and Engineering 27: 371--379. Yan M, Wang D, Yu J, Ni J, Edwards M, and Qu J (2008) Enhanced coagulation with polyaluminum chlorides: Role of pH/alkalinity and speciation. Chemosphere 71(9): 1665--1673. Yangali-Quintanilla V, Sadmani A, McConville M, Kennedy M, and Amy G (2009) Rejection of pharmaceutically active compounds and endocrine disrupting compounds by clean and fouled nanofiltration membranes. Water Research 43(9): 2349--2362. Yangali-Quintanilla V, Sadmani A, McConville M, Kennedy M, and Amy G (2010) A QSAR model for predicting rejection of emerging contaminants (pharmaceuticals, endocrine disruptors) by nanofiltration membranes. Water Research 44(2): 373--384. Yu Z, Peldszus S, and Huck PM (2007) Performance evaluation of granular activated carbon for the removal of selected pharmaceuticals and endocrine disrupting substances from drinking water. In: Chemical Water and Wastewater Treatment IX, Proceedings, 12th Gothenburg Symposium of 2007, pp. 343–352. Ljubljana, Slovenia. Yu Z, Peldszus S, and Huck PM (2008) Adsorption characteristics of selected pharmaceuticals and an endocrine disrupting compound – Naproxen, Carbamazepine and Nonylphenol – on activated carbon. Water Research 42(12): 2873--2882. Yu Z, Peldszus S, and Huck PM (2009a) Adsorption of selected pharmaceuticals and an endocrine disrupting compound by granular activated carbon. 1. Adsorption capacity and kinetics. Environmental Science and Technology 43(5): 1467--1473. Yu Z, Peldszus S, and Huck PM (2009b) Adsorption of selected pharmaceuticals and an endocrine disrupting compound by granular activated carbon. 2. Model prediction. Environmental Science and Technology 43(5): 1474--1479. Zhang S and Huck PM (1996) Removal of AOC in biological water treatment processes: A kinetic modeling approach. Water Research 30(5): 1195--1207. Zwiener C and Frimmel FH (2000) Oxidative treatment of pharmaceuticals in water. Water Research 34: 1881--1885.

Relevant Websites http://www.trojanuv.com Trojan Technologies.