This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
p/) and polysaccharides [141]. Alternatively, weak attractive interactions arise as a result of averaging over a multitude of individual specific chemical interactions between groups on the biopolymers. Examples include 1) ionic interactions in BSA-dextran sulfate [95, 123] or BSA-K-carrageenan [142]; 2) van der Waals interactions as in the case of lysozyme with different polysaccharides [143]; 3) hydrogen bonding as in gelatin complexed with
254
different polysaccharides [131]; and 4) hydrophobic interactions as in poly methacrylic acid [144] or in mixed gels of gelatin with sodium alginate [145]. The protein-polysaccharide interactions may change from net repulsive to net attractive, or vice-versa, on changing the temperature [144] or the solvent conditions (pH, ionic strength) [146]. Therefore, many polysaccharides carrying negatively charged carboxyl or sulfate groups, can form strong electrostatic association, so called "complex coacervation" [147] between the two kinds of polyelectrolytes of opposite net charge at pH values below the isoelectric point of proteins (typically pH~5). Nevertheless, the same polyelectrolytes fail to form strong electrostatic complexes at neutral pH. In this case local electrostatic interactions may still occur between anionic polysaccharide molecules and positively charged patches on proteins [105, 148]. Such interactions can be reinforced by non-ionic attraction (e.g. hydrogen bonding) leading to proteinpolysaccharide complex formation. 3.2. Microbial sources of bioemulsifiers Table 3 A and 3B list a number of bacterial and fungal strains, which have been shown to produce extracellular bioemulsifiers. The emulsification process causes droplet formation of one of the phases and a subsequent increase in emulsion turbidity, which is easily monitored. In addition, the bioemulsifiers also stabilize the emulsions by retarding droplet coalescence. 3.3. The emulsan paradigm Arguably, the most extensively studied polymeric microbial bioemulsifier is emulsan, the amphipathic, polyanionic polysaccharide produced by the oildegrading Acinetobacter venetianus RAG-1 [65, 149-151]. In addition, this product represents an interesting case study encompassing all stages of development, technology transfer and product development. Although many of the features of this model system have been presented in other reviews [6-8, 151-153], newer developments regarding the genetics, biosynthesis and regulation of its production are only now emerging, and will be discussed here. Similarly, the development of novel approaches to generating new amphipathic bioemulsifier formulations [140] has now been developed and will also be presented. 3.3.1. Physical and chemical characteristics Emulsan was initially isolated from the cell-free broth of a stationary phase culture of A. venetianus RAG-1 grown on a light crude oil. Subsequently, the non-dialyzable polymer was produced on a variety of water soluble carbon sources with optimum yields obtained on ethanol as sole source of carbon and energy [65, 75, 154-156]. The bioemulsifier (106 Da) consists of complex
255
between a polyanionic heteropolysaccharide (75%) and a non-covalently associated protein mixture (25%) [75]. As shown in Fig. 1, the linear backbone of the polysaccharide consists of a trisaccharide subunit consisting of a 1:1 ratio of the amino-sugars, Dgalactosamine, D-galactosamine uronic acid and 2,5-dideoxy, 2,5-diamino glucose. The amphipathic properties of the biopolymer arise from the presence of about 20% by weight fatty acids present in the form of amides and esters. The fatty acid composition can be modified by including fatty acids in the growth media [161-164], although the commercial product produced on ethanol minimal media consists of a mixture of 2, and 3 hydroxy dodecanoic acids, palmitic acid, 2-hydroxybutyric acid and acetic acids [7(55]. The apparent pK of the polymer is about 3.05 owing to the galactosamine uronic acid. In addition to the lipoheteropolysaccharide composition of the bioemulsifier, the extracellular emulsan complex also contains between 10 and 25% by weight protein. A portion of the protein mixture can be removed by any of a number of procedures including sepharose gel filtration chromatography, polymer precipitation in the presence of the quaternary ammonium detergent cetyl, trimethyl ammonium bromide [766], hot phenol extraction [75] and proteolysis. As discussed below, the deproteinized polymer, termed apoemulsan [154] retains a portion of the emulsifying activity towards some of the substrates, but is generally inactive in the emulsification of less polar materials such as pure n-alkanes such as hexadecane. The role of this protein in mediating emulsification by apoemulsan will be discussed in a different section below.
Fig. 1. The trisaccharide structure of the emulsan subunit.
256
Table 3A Production and surface activity of bioemulsifiers Organism C-source Compound Exopolysaccharides
Yield*
Acinetobacter calcoaceticus BD4
glucose
0.6
Acinetobacter calcoaceticus MM5
tetradecanc
0.06
ST
Reference [157]
\70]
Bacillus cereus IAF 346
sucrose
0.5-1.2
53
Bacillus sp. IAF343
sucrose
0.5-1.2
28
Halomonas eurihalina H28
crude oil
Alasan
Acinetobacter radioresistens K53
ethanol
2.2
[73]
Biodispersan
Acinetobacter calcoaceticus A2
ethanol
4
[74]
Emulsan
Acinetobacter venetianus RAG-1
ethanol
15
[158\
Liposan
Candida lypolytica ATCC 8662
hexadecane
Rhodococcus sp. Q15
glucose, acetate
36
[77]
Nocardia erythropolis ATCC 4277
29
[159\
Saccharomyces uvarum
hexadecane, kerosene n-dccane
20
[78]
Bacillus liqueniformis JF-2
glucose
25-34 0.35
\7l] [71] [72]
[76]
Lipids
Lipopeptides 30
[160]
glucose
27
[80]
L-broth
24
[56]
27
[34]
glucose
27
[80]
glucose
27
[S01
28
[38]
28
[42]
27
[43]
glucose
27
ISO]
glucose
26.5
[19\
glucose
27
[80]
Cory'nebacterium lepus
kerosene
Amphisin
Pseudomonas sp. DSS73
Arthrofactin
Pseudomonas sp. MIS 38
Bacitracin
Bacillus liqueniformis
Hodersin
Pseudomonas sp.
Lokisin
Pseudomonas sp. DSS41
Lychesin A
Bacillus liqueniformis BAS50
glucose
Scrrawettin
Serratia liquefaciens MG1
glucose
Surfactin
Bacillus subtilis ATCC 21332
glucose
Tcnsin
Pseudomonas fluorescens 96.578
Viscosin
Pseudomonas viscosa
Viscosinamide
Pseudomonas fluorescens DR54
ing/I. Surface Tension in mN/m.
[79]
0.16
3-4
257
The molecular weight of emulsan is about 10 Da [75]. Nevertheless, the polymer exhibits a rather low reduced viscosity in water of about 570 cc/gram [150] and appears to be a long rod of axial ratio about 50:1. The low viscosity of the biopolymer complex is due in part to the presence of the associated-protein. Emulsan is Newtonian in its flow properties, despite the rather high molecular weight, a feature which makes it easier to produce as a fermentation product. 3.3.2. Emulsan activity As an emulsifier emulsan does not dramatically affect either surface tension or interfacial tension between two immiscible phases. Nor does it form micelles, which is characteristic of detergents. Rather its activity appears to depend on its high affinity for the oil/water interface. Being a water-soluble polymer emulsan both forms and stabilizes oil-in-water emulsions resulting from the orientation of the biopolymer at the oil droplet surface [169]. This orientation was first inferred from the binding characteristics of the positively charged rhodamine cation, which bound emulsan in an emulsion, but did not bind to the polyanionic polymer when emulsan was in solution. This binding of cations to emulsan preferentially at the oil/water interface was observed with several metal ions such as C d \ Zn+ , UO2 , Mn , and a host of others [169170]. In many cases the binding was more extensive than anticipated from the charge density of the polymer suggesting that the polymer had assumed a different conformation when oriented at the interface. Under conditions of high energy input, when the formation of small oil droplets is the result of vigorous agitation, the high affinity of emulsan for the interface results in its coating of the droplets preventing coalescence due to charge repulsion of the negatively charged uronic acids. This stabilization does not require the presence of protein, and can be achieved with apoemulsan at correspondingly low ratios of emulsifier to oil between 1:100 to 1:1000 parts of emulsan to oil. The resulting emulsions are very stable, and withstand high-speed centrifugation. In fact, rather than breaking the emulsion into two phases, the centrifugation causes formation of a cream layer which itself is an oil-in-water emulsion consisting of a bulk aqueous phase which constitutes only about 30-50% of the cream, depending on the ratio of emulsan to oil. This cream layer, termed an emulsanosol, is itself an oil-in-water emulsion readily dispersible in the aqueous phase. Emulsanosols or apoemulsanosols can be prepared with a variety of pure and cruder oils, and are stable for years. Their potential applications will be discussed later in the Chapter. In addition to stabilization of oil/water emulsions, emulsan also forms emulsions at lower rates of agitation. In these systems the presence of proteins plays a major role since emulsion formation is quite specific for the hydrocarbon substrates [150], In general, emulsan has been shown to emulsify relatively polar mixtures of hydrocarbons such as those containing both an aliphatic and an
258
aromatic substrate. Hexadecane alone is a very poor substrate, and in the absence of protein, emulsan is completely ineffective with non-polar materials [139]. Interestingly, activity towards hexadecane and other non-polar waxes and sludges can be reconstituted in the presence of a single recombinant protein, the cell-surface esterase of A. venetianus RAG-1. The results in Table 4 illustrate the effect of addition of the cell surface esterase of RAG-1 on the emulsification of a variety of substrates, none of which is emulsified very well by emulsan itself. Interestingly, the addition of the recombinant esterase to emulsan itself had a dramatic effect on the emulsification towards very hydrophobic substrates [139]. Table 3B Production and surface activity of bioemulsifiers Compound Organism Glycolipids Rhodococcus erythropolis
Yield' 32 8 0.1
26
[24]
2.8
26
[27]
1
29
[167]
70
31
[168]
43
[31]
Flavolipid
Flavobacterium sp MTN11
Mihagol L Mihagol S glucose
Pentasaccharide
Nocardia corynebacteroides
nCi4_ 15
DSM 43215
Reference
C-source
ST
[25]
SMI
Rhamnolipids
Pseudomonasputida 21BN
Sophorose
Candida bombicola ATCC 22214 Torulopsis petrophilum Candida bogoriensis
hexadecane, glucose glucose, safflower oil glucose glucose
2
[SI]
Others biosurfactants Aerobactin Mannoprotein
Aerobacter aerogenes 621 Saccharomyces cerevisiae
glucose glucose
1 8"
Synthetic surfactants Cetyl Triethyl Ammonium Bromide (CTAB) Linear alkylbenzene sulfonate Sodium dodecyl sulphate Tween 20
30 47 37 30
Water
72
ing/1. Surface Tension in mN/m. ± in gr/ wet cell gr.
[23] [82]
259
Table 4 Enhancement of apoemulsan activity on different hydrophobic substrates by recombinant esterase Hydrophobic substrate
Emulsifying Activity Ratio
Anthracene Crude oil Dicyclohexane Diesel oil Eicosane Fluoranthene Heptadecane Immersion oil 2-Methyl Naphthalene Mineral oil Octadecane Petroleum refinery sludge Pyrene Soya oil Squalene Tetracosane
966 4.3 8.9 5.7 1800 593 28 10.4 1984 6.6 2250 2 420 1260 600 506
Apoemulsan emulsifying activity in presence of recombinant esterase: emulsifying activity.using emulsan alone.
3.3.3. Emulsan as a microbialproduct Physiology of production. As a fermentation product emulsan is produced primarily on a minimal medium containing ethanol as a sole source of carbon and energy [65]. During early exponential growth the emulsan capsule is present on the cell surface as a minicapsule, which is subsequently released from the cell surface as the cells approach stationary growth. Turnover experiments using radioactive substrate indicated that product release is accompanied by de-novo synthesis [10, 171-172]. Moreover, polymer release is accompanied by a prior release of a protein complex including the cell surface esterase mentioned above. When the cells were starved for carbon in the presence of chloramphenicol the esterase was released but the emulsan remained associated with the cell surface. Addition of fresh nutrients did not give rise to polymer release under these conditions, presumably because the esterase could not be made in the presence of chloramphenicol. The results strongly implicated cell surface esterase in the release process. Interestingly, when emulsan production was studied in a resting cell system in which the cells were immobilized on a celite column, emulsan was shown to accumulate on the surface of the cells in response to increased shear forces [173].
260
Emulsan biosynthesis. In order to study the biosynthesis of emulsan, an insertional plasmid was used to generate mutants-defective emulsan production. Such mutants are visualized either by virtue of their translucent colonial morphology (Fig. 2) or according to their resistance to particular RAG-1 specific phage, ap3, which uses the cell bound emulsan minicapsule as a phage receptor [174]. A mutant blocked in emulsan biosynthesis is thus resistant to phage ap3, but sensitive to a second phage, ap2, which does not require the emulsan receptor. In this regard, the phage ap3 does not bind to the released, water soluble emulsan complex, but does bind to the concentrated emulsanosol, again supporting the idea that the emulsan polymer undergoes a conformational change at the oil/water interface [174]. These data also support the important notion that the conformation of the biopolymer on the cell surface is similar to its conformation at the oil/water interface. Natural role for emulsan. Cells defective in emulsan do not grow on crude oil in liquid culture, although they do grow on these hydrocarbons when they are provided in the vapor phase [174]. Gutnick and Shabtai showed that the cell bound polymer protected cells of RAG-1 from the toxic effects of catalytic surfactants such as cetyl-trimethyl ammonium bromide [166]. In fact, among mutants resistant to the cationic detergent were those, which actually produced more biopolymer than the corresponding wild-type RAG-1. Based on findings that cells lacking emulsan on their surface more avidly to hydrophobic materials [175-177], it is possible that the natural role of the biopolymer is to aid in actually removing the cells from the oil, when utilizable carbon is no longer available to the organism. This would expand the versatility of the organism by allowing it to release itself and search for new more productive surfaces for metabolism. In support of this hypothesis was the finding that emulsan could be used to remove other organisms from hydrophobic surfaces [178]. Another potential role for emulsan relates to its role as a microbial capsule prior to its release from the cell surface. Ophir and Gutnick [179] showed that capsule producing organisms exhibit at least a ten fold higher resistance to desiccation than do isogenic strains lacking the capsule. This resistance depended on the adherence of the capsule to the cell surface and could not be restored by addition of fresh biopolymer from the outside. Emulsan producers were shown to exhibit such resistance to desiccation. Emulsan biosynthetic pathway. The genes encoding the biosynthetic pathway for apoemulsan have recently been localized to a 27 kbp cluster termed the wee regulon [180]. The entire cluster was sequenced and shown to encode 23 putative open reading frames arranged in two divergent operons separated by a non-translated region (Fig. 3). Mutations in any of the genes resulted in a defect in emulsan production. These defects could all be complemented by a wild-type allele. Interestingly, most of the genes for emulsan biosynthesis were homologous to genes discovered in the biosynthesis of most polysaccharides in
261
spite of the fact that A. venetianus RAG-1 is thus far the only natural isolate which has been shown to produce emulsan [180]. According to convention, the specific genes for emulsan biosynthesis were termed wee; the first letter signifies that the gene is a biosynthetic gene from a polysaccharide biosynthetic cluster, the second land third letters identifying the specific product (emulsan, exopolysaccharide). In accordance with convention, some gene products exhibit similar functions in all organisms, and thus are allowed to retain their original names. Figure 3 summarizes a hypothetical biosynthetic pathway for apoemulsan, with, the rightward operon encoding proteins involved in precursor synthesis and activation, aminoglycosyl transferases for assembling the trisaccharide subunit on the inner side of the cytoplasmic membrane, a polymerase, decorating enzymes for the acylation of the aminosugars, a translocase which moves the polymer from the cytoplasmic face of the membrane to the outer or periplasmie face, and enzymes involved in subsequent translocation of the polymer through a specific channel or porin to the outer surface of the cell [180]. Regulation and the production of viscoemulsan. Located in the intercistronic region are two putative d promoters. The leftward operon consists of three repeating frames wza, wzb and wzc, which encode a porin, a protein tyrosine phosphate phosphatase and a protein tyrosine kinase, respectively [180-181]. Knockout mutants in any of these genes resulted in defects in emulsan production. Both Wzc and Wzb proteins of RAG-1 were cloned and over-expressed in E. coli. The Wzc Ptk was shown to be an autophosphyorylase in which a tyrosine (s) in the C-terminal portion of the protein is phosphorylated and subsequently dephosphorylated by the phosphatase [182]. Similarly, the phosphotyrosine of Wzc from RAG-1 was shown to be dephosphorylated by Wzb [181]. According to other reports, the phosphorylated form of Wzc is expected to negatively regulate polymer export through the porin Wza. Elevated levels of extracellular biopolymer production would then be initiated with the activity of Wzb, the phosphatase, which removes the phosphates, permiting the enhanced export. Consistent with the hypothesis was the finding that knockout mutants in the phosphatase were also emulsan deficient. However, the results did not explain why knockouts in Wzc would be emulsan deficient as well. Apparently there is a requirement for the Wzc protein even in its non-phosphorylated state. The Wzc protein contains a series of five tyrosine residues in close proximity to each other at the C terminus. When these tyrosines were deleted, the resulting protein was made but could be phosphorylated and surprisingly, a high molecular mass polysaccharide, termed viscoemulsan, was produced [181]. This product appears to contain the same constituents as emulsan, but is not active as an emulsifier (Nakar, In preparation). The introduction of a wild-type allele of wzc gave rise to the production of a wild-type allele of emulsan suggesting that the protein
262
tyrosine kinase may act to control the size of the exported polymer. It is also of interest that the Wzc protein is required for viscoemulsan production even though it cannot be phosphorylated [181] suggesting that there is an additional role for the protein. A model to describe the role of phosphorylation and dephosphorylation is shown in Fig. 4 [181]. According to this model Wzc, Wzb, Wza proteins and others interact in a multienzyme complex to control the export of the exopolysaccharide. The process is initiated by dephosphorylation of the protein relaxing the control on the porin diameter and enabling larger amounts of polymer to be translocated to the external surface of the cell. Under conditions of rapid growth and high ATP, all of the tyrosine residues are phosphorylated and polymer production is low. In fact, emulsan production does not take place in rich media, although its biosynthesis has been shown to occur. The polymer can be detected immunologically. The manipulation of the export process coupled with the modifications of the biosynthetic genes offers new approaches to the generation of new and novel products and is currently in progress. 3.3.4. Engineering novel derivatives of emulsan The production of new viscous derivatives such as viscoemulsan represents one approach to engineering new biopolymers. Kaplan and coworkers have used nutritional modification to modify fatty acid composition and surface active properties of the resulting derivatives of emulsan [161-164]. Moreover, as described above, the surface activity of apoemulsan-containing formulations can be enhanced by the addition of a particular cell surface enzyme, the cell surface esterase of RAG-1 [139].
Fig. 2. Colonial morphology of parental emulsan-producing RAG-1 and a translucent, emulsan-defective mutant, TR3.
263
Fig. 3. The wee cluster for the biosynthesis of emulsan. The scale of the cluster size is in kilobases. The black arrows represent putative orf sequences. White arrows represent partially sequenced orf s. Putative promoter sites are indicated with thin black arrows. The names of the genes are shown below the corresponding orf s. Orf s labeled solely with capital letters are putative pathway specific genes encoding Wee A-K respectively.
Novel surface-active breakdown products. In order to stabilize water/oil emulsions emulsan must be a polymer [176]. The evidence supporting this comes from the activity of a particular emulsan depolymerase isolated from a bacterial isolate capable of using galactosamine as a sole source of nitrogen. When apoemulsan was incubated with this crude enzyme for different periods of time, it was found that cleavage of less than five percent of the glycosidic bonds were necessary to inactivate the biopolymer. The results support the idea that the interaction of biopolymer at the oil/water interface is a weak, and that the stabilization is brought about by many points of weak interactions at the oil/water interface. Interestingly, subjecting apoemulsan to exhaustive digestion by the depolymerase generated a series of small acylated aminooligosaccharides consisting of between three and six aminosugars [183]. These materials were all found to act as small molecular weight detergents, although they did not effectively stabilize emulsions. They were found to be active towards more hydrophobic substrates such as hexadecane. Emulsification enhancing proteins and peptides (EEPs). As described above, emulsification of oils by emulsan and apoemulsan is strongly enhanced by the addition of a cell surface recombinant esterase from RAG-1 [139]. Potential principles governing this type of interaction have been presented above. The cloning of the enzyme in E. coli was first accomplished by selecting recombinants of E. coli, which could grow on simple triglycerides as sole sources of carbon and energy [184-186]. The clone was required to generate metabolizable substrates such as glycerol and simple fatty acids such as acetate in order to grow. Cloning, sequencing, over-expression and mutagenesis experiments demonstrated that the esterase is a serine protease [187]. However, using a threading program in which primary sequence was related to predicted structures it was found that the protein in fact more resembled an a, P hydrolase enzyme such as acetyl cholinesterase [187]. Interestingly, this was also found
264
for an esterase from another member of the genus Acinetobacter, the strain A. calcoaceticus BD4 [188] and its miniencapsulated derivative BD413. While this enzyme shows strong sequence and structural homology to the RAG-1 enzyme, it did not display any emulsification enhancement when added to apoemulsan [139]. Specificity towards hydrocarbons. As shown in Table 4 the recombinant esterase protein enhances emulsification of apoemulsan towards a variety of pure and crude hydrophobic substrates. EEP activity was observed with mutants of the esterase defective in catalytic activity, suggesting a role for the protein other than as an enzyme. Esterase exhibits EEP activity towards other polysaccharides. Surprisingly, the interaction of the recombinant RAG-1 esterase with the water soluble, rhamnose-containing exopolysaccharide from A. calcoaceticus BD4 led to the formation of a new bioemulsifier complex. In sharp contrast, the esterase from BD4 did not enhance emulsifying activity of apoemulsan towards hydrophobic substrates [140]. Remarkably, the recombinant esterase from RAG1 exhibited EEP activity with over 25 different natural biopolymers, none of which exhibited any emulsifying activity in the absence of the protein. In these cases, the enhancement was not dependent on catalytic activity of the recombinant protein (Bach and Gutnick, in preparation). The results point to a new approach to generation of amphipathic emulsifiers, which is no longer dependent on fermentation to produce the polymer emulsifier. Among the inexpensive materials, which can be converted into bioemulsifiers using this unique formulation with the RAG-1 esterase are cellulose, dextran, starch, xanthan, alginic acid, and a variety of plant and bacterial polysaccharides including the inactive viscoemulsan described above (Table 5). The mode of action of the EEP remains to be elucidated although evidence is discussed below demonstrating that there is a unique motif in the RAG-1 esterase, which is missing from other homologues. Mapping the EEP domain. Initial observations showed that limited proteolysis of the recombinant esterase yielded a fragment of about 10 kDa, which retained the ability to enhance emulsification of hydrophobic substrates such as hexadecane. Accordingly, a series of site directed mutants were generated and over-expressed to produce different fragments of the esterase. Since the fragments were rapidly degraded even in strains of E. coli lacking Clp or Lon proteases, fragments were prepared which were fused in frame to the Cterminus of the maltose binding protein [189]. The various constructs are shown in Fig. 5. The each over-expressed fusion was tested with apoemulsan using the model hydrophobic substrate, hexadecane as a substrate for emulsification. Virtually all the enhancing activity was localized to the C-terminal third of the esterase. It was of interest that the maltose binding protein itself exhibited no EEP activity. Moreover the fusion protein containing the active polypeptide was
265
no less active than the intact enzyme. Removal of the terminal 15 amino acids from the C-terminus completely abolished the EEP activity. Sequence analysis showed that this 15 amino acid C-terminal peptide is unique to the RAG-1 esterase and probably accounts for the unique characteristics of this protein. However, as shown in Table 2, many organisms produce emulsifiers consisting of protein/polysaccharide complexes [7, 9, 190]. In most cases the protein requirement has yet to be clarified and it is possible that there are other proteins or peptides, which exhibit unique EEP activity. Regardless, EEP technology offers a new approach to bioemulsifier production and paves the way for new families of inexpensive, non-toxic, amphiphiles.
Fig. 4. Hypothetical model for the role of protein tyrosine kinase (Wzc) and protein tyrosine phosphatase (Wzb) in emulsan export. 1. Dephosphorylated Wzc allows for polymerization and translocation of emulsan. 2. Phosphorylation of Wzc halts the process, thereby determining the size of the exported polymer. 3. Emulsan release and beginning of a new round of polymerization, translocation and release. Wza-translocation channel; Wzb-protein tyrosine phosphatase; Wzc-Protein tyrsoine kinase; Wzx-polymerase; Wzy-translocase.
266
3.4. Other polymeric bioemulsifiers 3.4.1. Alasan Acinetobacter radioresistans radioresistens KA53 produces a bioemulsifier complex (10' kDa) consisting of three proteins and a polysaccharide [73]. The emulsifying activity was associated primarily with the AlnA protein. Interestingly, the N-terminal sequence of a recombinant form of the AlnA protein produced in E. coli showed strong homology to the outer membrane protein, OmpA [191]. The recombinant form of AlnA was more active as an emulsifier than the complex. It was also shown to solubilize polyaromatic hydrocarbons and at higher concentrations of the substrate, to form hexamers [192]. The crude alasan complex also formed alkane/water emulsions at an optimum pH of 5. This activity was significantly enhanced after heating at 100°C. Interestingly, the alasan producing strain does not grow on hydrocarbons or on oil substrates and the biological role of this complex remains to be elucidated.
Table 5 Enhancement of the emulsifying activity of different polysaccharides by recombinant esterase in the presence of hexadecane _ , , , Polysacchande Agarose Alginic acid Apoemulsan BD-4 exopolysaccharide Carrageenan Cellobiose Cellulose Chitin Colamc acid Dextran Emulsan Ficoll 400 Gum Arabic Pectin Polyvinyl Pyrrolydone Potato starch Pullulan Stewartan Xanthan Xylan
Emulsifying activity ,.., , , , . . (U/mg polysacchande/mg esterase) 963 496 5430 3396 3345 626 766 540 2050 583 6752 263 1895 1830 1950 544 3400 1196 2720 1854
267
Recently another strain of A. radioresistans SI3, which grows on aromatic substrates was analyzed for changes in membrane protein composition in response to changes in the growth substrates [193]. Two-dimensional gel electrophoresis of protein extracts from S13 revealed elevated levels in an Omp A-like alasan ortholog in response to growth on phenol. 3.4.2. Liposan Liposan is a polymeric bioemulsifier produced by the yeast Candida lipolytica ATCC 8662 [76]. The protein polysaccharide complex consists of 83% polysaccharides and 17% protein. When grown on hexadecane organism appeared to colonize the hexadecane droplets. Liposan emulsified alkanes with a chain length between C6 and C18 with the emulsifying activity increasing with increasing chain length. Liposan has also been shown to emulsify various crude oils such olive and corn oils, gas oil, kerosene, paraffin, halowax 1000 and a series of aliphatic and aromatic hydrocarbons. 3.4.3. Biodispersan This polymer is produced by Acinetobacter product exhibited a molecular mass of 51,400. This is a dispersant, which disperses limestone and aids for grinding limestone to form a powder, which is 194].
sp. A2. The extracellular polyanionic polysaccharide lowers the energy required an ingredient in paper [74,
Fig. 4. Generated esterase constructs fused in frame to the C-terminus of the maltose binding protein.
268
3.4.4. Exopolysaccharide-protein complex from Acinetobacter calcoaceticus BD4 A.calcoaceticus BD4 produces a thick rhamnose-containing exopolysaccharide protein complex [157, 195]. The same biopolymer is produced by a strain, which produces similar quantities of extracellular polysaccharide even though it accumulates lower amounts of the capsular polysaccharide on the cell surface. The protein-polysaccharide complex was shown to emulsify mixtures of alkanes and aromatic substrates, but was inactive against pure alkanes. Interestingly, the complex was not particularly active in the presence of crude oil. Of interest, however, was the finding, that unlike the emulsan complex, which retained partial emulsifying activity even after removal of the protein fraction, the BD4 product was completely dependent on both the protein and the polysaccharide fractions. The polysaccharide could be prepared without the protein by physically shearing it from the cell surface, and the activity restored by the addition of the protein fraction [196]. The protein fraction was specific for the BD4 polysaccharide, and was able to enhance the emulsifying activity of apoemulsan (H.Bach, in preparation). However, as was the case with a host of bacterial exopolysacchride [140], the recombinant esterase from RAG-1 was able to reconstitute emulsifying activity to the polysaccharide from A. calcoaceticus BD4. 4. POTENTIAL APPLICATIONS 4.1. General comments The petroleum industry consumes millions of tons of surfactants each year in a large number of applications (see chapters 4 and 19). Surfactants are used in oil field applications (drilling muds, enhanced oil recovery), environmental and equipment clean-up and maintenance, viscosity reduction and oil transportation, emulsion breakage and dewatering of crude oil prior to refining, and more recently, water/oil based fuels. Generally, the surfactant packages include a combination of surface-active agents, and frequently include compatible solvents and specialized chemicals depending on the quality of the specific oils and sludges. As described in this Chapter, biological products and processes can be employed in all of these applications. In this section we will consider only a few larger scale experiments using biosurfactants. The successful trials indicate that at least in terms of product efficacy, these materials have potential. However, their profitability has yet to be unequivocally demonstrated. In this section we will discuss a few large- scale field trials pointing to the possibility of employing biosurfactants and emulsifiers in the oil industry. Unlike biotechnology products for other sectors such as health care or Pharmaceuticals, where the cost of development and even the cost of obtaining
269
approval from the regulatory agencies, are offset by the high prices and profitability of the product, the cost of applications in the oil industry must be kept relatively competitive with products of the chemical industry. This is a particularly difficult constraint considering that production of many of the biotechnological products may involve large-scale fermentation processes, which exert a considerable impact on the cost of the product, particularly if extensive downstream processing is required. Several approaches may be used to enhance the cost effectiveness of biosurfactants. 4.1.1. Searching for the "world beater" Occasionally, the search for natural products yields a compound with unique properties unlike others either natural or developed by the chemical industry. Such products are termed "world beaters" because their properties are unique and unmatched by products of chemical synthesis. Antibiotics represent a classical example of natural materials of major chemotherapeutic importance whose activities are unmatched by chemical synthesis, although they are generally modified by various chemical transformations [ 197]. Another example of a microbial product with unique properties is the exopolysaccharide product of the plant pathogen Xanthomonas campestris, xanthan, is a major polymer whose sheer thinning properties and high reduced viscosity make it a major industrial product in foods as a thickening agent, in drilling muds, and as a gelling agent for use in oil field fracturing programs [5]. Moreover, xanthan viscosity is exploited in oil field flooding during enhanced oil recovery, since the extraordinary high viscosity enables it to actually "push" the released oil out of the well. In most cases, however, the natural biosurfactant exhibits characteristics, which are promising but not necessarily unique. 4.1.2. Cutting the cost of production The cost of production of biosurfactants is frequently a function of the cost of fermentation and subsequent downstream processing. This is particularly true for production of products in which the carbon source must be a hydrocarbon, which, is often the case with low molecular weight glycolipids [9, 16, 53, 198199]. A key approach in this system involves enhancing the product yield by upgrading and optimizing the fermentation [21]. In the latter case the fermentation of rhamnolipids has been upgraded such that lOOg/liter was produced from 160g of soybean oil as a carbon source, a remarkable conversion of substrate to product. In addition, production on various industrial waste products has also lowered the cost of production [200-202]. Assuming that the product biosurfactant is sufficiently active, this presents a way of upgrading the waste material by producing a product of higher added value. This approach may be particularly advantageous in the oil industry, since the crude product
270
need not be purified to any significant extent and may not require expensive downstream processing [203]. Enhanced productivity can in principle be obtained by transferring biosynthetic genes into an organism such E. coli K-12 which is easy to grow and which utilizes a "friendlier" source of carbon and energy [204-205]. At least in pilot scale, most biosurfactants are produced in batch fermentations. Cooper and co-workers developed a semi-continuous approach to producing biosurfactants via self-recycling system [206]. Similarly, emulsan was produced in a similar protocol in which the product was allowed to grow and accumulate in early stationary phase followed by the removal of 90% of the cells and emulsan, which was harvested downstream. The fermentor was filled with fresh media and the culture again entered exponential and early stationary growth, the major portion of the emulsan recovered and the cycles repeated in the same fermentor for several semi-continuous production runs. The cost of production is thus significantly reduced (Cooper, D., Personal communication). Another way to cut the cost of production is to upgrade the producing strain in order to enhance overall productivity [8, 199], This approach has been used in the case of the emulsan producing strain A. venetianus RAG-1 [166]. The positive selection for emulsan overproducers was based on the fact that the emulsan polyanion binds the toxic cation cetyl-trimethylammonium bromide (CTAB). Among the mutants of RAG-1 resistant to CTAB, were those such as strain A. venetianus CTR49, which overproduce the extracellular polyanion and are thus significantly more resistant to the CTAB than the parent. In the laboratory, mutants of this variety produced up to twice as much emulsan per gram of ethanol carbon source than the wild-type. 4.1.3. Upgrading the product Metabolic engineering of new products. Another approach to generating a viable technology employing biosurfactants is to employ physiology, formulation and/or recombinant DNA technology to generate modified products with improved properties. One such approach involved preparing emulsan from RAG-1 cells grown in the presence of various fatty acids [161-164] in order to modify the nature of the acyl groups present in the side chains of the bioemulsifier. We have carried out similar experiments and have found that the emulsan produced is significantly more active towards hydrophobic substrates such as hexadecane alone. Of course, the enhanced efficacy still needs to be weighed against the increased cost of the fermentation due to the inclusion of fatty acids in the media. Formulation packages. Emulsifiers and surfactants are generally incorporated into surfactant packages, which include a mixture of surfactants designed to lower interfacial tension between water and oil phases. Also, in cleaning applications, the formulations may also contain a biopolymer to
271
stabilize the emulsion [207-209] and prevent coalescence of the phases, and they can also contain a compatible solvent. The solvent need not be a water based solvent, but it should be able to dissolve both the low molecular weight surfactants as well as the biopolymer. Materials such as pine oil, liquid terpenoids, dimethyl sulfoxide, and various light crude oils have all been included in various surfactant packages [210]. In addition to solubilizing all of the components into a pumpable mixture, the solvent addition has also been shown to enhance the cleaning of oil contaminated tanks by removing the last remnants of sludge and other flammable materials from the walls of the container rendering the tank not only clean, but also gas-free. The choice of suitable components for various surfactant packages must also take into account other potential components, which must be included. For example, if routine cleaning operations include rinses with anticorrosive materials, the formulation package must be designed on the basis of compatibility with such components. Similarly, emulsion based fuels may need to be formulated together with materials which lower sulfur emissions. Specially designed surfactant formulations may also require compatibility with a variety of materials including flame retardants, biocides etc. 4.1.4. EEP technology In section 3.3.2 the ability of a recombinant cell surface esterase from RAG-1 to enhance the emulsification of apoemulsan towards a variety of hydrophobic substrates was described [139]. The remarkable feature of this system was the finding that the RAG-1 esterase and several of its derivatives were able to interact with a host of polysaccharides to generate a series of amphipathic complexes, which exhibited strong emulsifying activity [140]. This surprising activity, has paved the way for the generation of a whole suite of bioemulsifiers, in which the polymeric component need not be produced as a fermentation product. In fact, it may be possible to upgrade waste materials such as crude celluloses, starch, pectins, etc. to bioemulsifier when combined with a specific peptide derived from the esterase. This peptide can be produced as an over-expressed protein fusion following cloning in a suitable vector [Bach and Gutnick, in preparation], or it can be generated via proteolysis of the esterase itself. The feasibility of employing EEP technology for emulsification in oil industry applications will become clearer once larger scale field trials are conducted and evaluated. 4.2. Bioemulsification, cleaning and sludge recovery 4.2.1. Tank clean ing Oil storage containers accumulate enormous quantities of sludges and bottom sediments. Previous work from this and other laboratories have
272
described bioremediation techniques in which the container is basically turned into a fermentor and microbial biodegradation is used to clean the tanks [211212]. Bioremediation of oily wastes is covered in other chapters in this volume. Another approach is to generate stable oil/water emulsions using a surfactant or biosurfactant package in order to recover the sludge waste material in a homogenous, combustible form. With this approach the costs of the tank cleaning may be partially offset by the added value of the recovered emulsion. Petroferm U.S.A. developed an emulsan-based biosurfactant package designed to clean oil tanks and render them gas-free. Tanks in the order of several to tens of thousands of cubic meters were cleaned using this system after first designing several pumping and mixing devices. The oil-in-water emulsions generated in these large-scale field trials were stable and could withstand high-speed centrifugation to generate emulsanosols [207]. As will be discussed below, these crude emulsanosols containing up to 30% water can be used as an emulsionbased fuel. Another emulsan-based application was developed by Dr. Mary Ann Jones at the research station of the U.S. Navy in Washington. This application is designed to rapidly and efficiently clean the filters normally used in the engine rooms of ships for sludge removal from bilges. The emulsan-based formulation includes, in addition to crude emulsan, light crude oil as a solvent. The system is not designed for oil recovery, but rather for rapid filter cleaning. A rough estimate suggests that the emulsan-based cleaning process could cost as little as $300 per ship cleaning [Jones, Personal communication]. 4.2.1. Emulsion-based fuels Stable oil-in-water or water-in-oil emulsions, if sufficiently homogeneous can be burned for energy, provided that the water content does not exceed about 30%. In fact, this is the water content of a stable emulsanosol prepared on an oil such as a light Texas or Iranian crude oil. A large scale experiment was conducted in which about fifty barrels of an emulsanosol was prepared from a n emulsion consisting of about 70% high vacuum residuals emulsified in 30% by weight water. The combustion of this material was tested at an experimental laboratory in MIT. The combustion was identical to the combustion of a light fuel oil in terms of burn temperature, efficiency and light off. Interestingly, the emulsion of nitrogen gases was slightly lower suggesting that in the presence of water there may be less NOx emission. The results of this large-scale field trial [207-210, 213] have paved the way for a potential application in which sludge from storage tanks can be emulsified, and the emulsion (i.e. the emulsanosol) can be recycled as a source of energy. The system allows for the upgrading for otherwise unusable materials such as vacuum residuals to be recovered as a source of energy offsetting the cost of the tank cleaning. In addition, the fact that the oil phase consists of small droplets suggests that the burn efficiency may be
273
somewhat higher than with a regular hydrocarbon fuel since the surface area is larger. What is even more interesting is that the quality of the burn was indistinguishable from that of a light high quality fuel oil suggesting that emulsion based fuels can be a viable alternative for some applications [207210,213]. Interestingly, these larger scale burn experiments were performed with a biosurfactant containing formulation. However, at Petroferm U.S.A. specialty chemical formulations were developed some of which did not contain the emulsan, but was composed of chemical surfactants, a solvent and other components. Emulsion based fuels have become more and more popular, because they permit the efficient combustion of various hydrocarbons which are normally difficult to burn. Arguably, the best studied system is the waterbitumen emulsion system, termed Oriemulsion which has been commercialized and is currently exported from Venezuela throughout the world. The emulsion is chemical based, and resembles the initial Petroferm formulations. 4.3. Viscosity reduction and oil transportation As discussed above, the stability of emulsan-based oil-in-water emulsions results from the coating of the oil droplets with emulsan in an oriented conformation; hydrophobic moieties coming into contact with the oil surface, and the hydrophilic components oriented towards the aqueous phase. This results in a homogeneous suspension in which the viscosity of the oil component is significantly reduced at room temperature. The extent of the viscosity reduction is a function of the water composition of the bulk phase, which for an emulsanosol can be as high as 30%. Under these conditions the viscosity of a high viscosity oil from the Orinoco basin in Venezuela, (Boscan crude) was reduced from >50,000 Cp to about 85 Cp in the form of an emulsanosol generated with an emulsan based surfactant package [207-210, 213] at a ratio of 1 part surfactant to 500 parts oil. About fifty barrels of this emulsion was transported through an experimental pipeline of 1.25 inches for 96 h during which the mixture was subjected to over 500 pump transits. There was no effect on the low viscosity, although the shear forces on the emulsion might have been expected to produce an inversion from an oil-in-water to a water-in-oil emulsion. Moreover, even after the system was shut down and the emulsion allowed to stand undisturbed for 48 h, there was no breakage or inversion of the emulsion and the low viscosity was maintained through the pipeline for an additional 48 h. The results support the use of surfactant packages to generate oil-in-water emulsions for pipelining highly viscous oils. In fact, this is the basis of the Orimulsion technology.
274
5. CONCLUDING REMARKS There is little doubt that biosurfactants and bioemulsifiers exhibit characteristics and activities, which are applicable in the oil industry. As noted, in some cases product efficacy has been tested in large scale and successful trials have been recorded. However, the impact of such materials on the oil industry is likely to be far less than in other industrial sectors, where the price of the final product is high relative to the costs of production. Ongoing efforts to isolate new biosurfactants, genetically modify existing ones, enhance productivity of the producing organism and otherwise lower the cost of production, and formulate new and more effective biosurfactant packages should yield a host of products and applications in the future. Moreover, for some applications such as enhanced oil recovery or bioremediation, the biotechnology associated with in-situ biosurfactant production accompanying the growth of microorganisms, discussed elsewhere in this book, can be a useful and economically competitive strategy. Finally, the incorporation of biosurfactants in novel formulations suitable for sludge liquification and viscosity reduction leading to economically feasible techniques for waste recovery and recycling.
REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11 ] [13] [14]
C.A. Houston, (2002) Colin A. Houston & Associates, Inc, NY. J.L. Salanger, Surfactants-Types and Uses, Laboratorio FIRP, Merida, Venezuela, (1999). R. Modler and Y. Ishikawa, Surfactants, Specialty Chemicals Update Program - SRI Consulting: CA. (2001). G. Bognolo, Sustainable Surfactants, Renewable Feedstocks for the 21 st Century, Competitive Industrial Materials from Non-Food Crops, Sand Hutton, York, UK, (1998). I.W. Sutherland, Biotechnol. Adv., 12 (1994) 393. D.L. Gutnick and W. Minas, Biochem. Soc. Trans., 15(1987) Suppl. 22S. E. Rosenberg and E.Z. Ron, Appl. Microbiol. Biotechnol., 52 (1999) 154. Y. Shabtai, O. Pines and D.L. Gutnick, Develop. Indust. Microbiol., 26 (1985) 291. J.D. Desai and I.M. Banat, Microbiol. Molec. Biol. Rev., 61 (1997) 47. C. Rubinovitz, D.L. Gutnick and E. Rosenberg, J. Bacteriol., 152 (1982) 126. H.J.A. Asmer, S. Lang, F. Wagner and V. Wray, J. Am. Oil Chem. Soc, 65 (1988) 1460. [12] A.A. Bodour, K.P. Drees and R.M. Maier, Appl. Environ. Microbiol., 69 (2003) 3280. A. A. Bodour and R.M. Miller-Maier, Biosurfactants: types, screening methods, and applications., Encyclopedia of Environmental Microbiology, Wiley, NY, 2002, pp.75O770. N.H. Youssef, K.E. Duncan, D.P. Nagle, K.N. Savage, R.M. Knapp and M.J. Mcmerney, J. Microbiol. Meth., 56 (2004) 339.
275
[15] A. Fiechter, Trends in Biotechnol., 10 (1992) 208. [16] I.M. Banat, R.S. Makkar and S.S. Cameotra, Appl. Microbiol. Biotechnol., 53 (2000) 495. [17] K. Arima, A. Kakinuma and G. Tamura, Biochem. Biophys. Res. Commun., 31 (1968) 488. [18] T. Suzuki, K. Hayashi, K. Fujikawa, and K. Tsukamoto, J. Biol. Chem., 57 (1965) 226227. [19] D. Sorensen, Cyclic Lipopeptides from Pseudomonas spp., The Marine Chemistry Section, Department of Chemistry, Faculty of Science, University of Copenhagen, Copenhagen, 2002, pp. 1-61. [20] MA. Marahiel, T. Stachelhaus and H.D. Mootz, Chem. Rev., 97 (1997) 2651. [21] S. Lang and D. Wullbrandt, Appl. Microbiol. Biotechnol., 51 (1999) 22. [22] T. Suzuki, K. Tanaka, J. Matsubara and S. Kmoshita, Agric. Biol. Chem., 33 (1969) 1619. [23] F. Gibson and D.I. Magrath, Biochim. Biophys. Acta, (1969) 65. [24] A.A. Bodour, C. Guerrerobarajas, B.V. Jiorle, ME. Malcomson, A.K. Paull, A. Somogyi, L.N. Trinh, R.B. Bates and R.M. Maier, Appl. Environ. Microbiol., 70 (2004) 114. [25] A. Kretschmer, H. Bock and F. Wagner, Appl. Environ. Microbiol., 44 (1982) 864. [26] K. Brandenburg, B. Lindner, A. Schromm, M.H.J. Koch, J. Bauer, A. Merkli, C. Zbaeren, J. Gwynfor Davies and U. Seydel, Eur. J. Biochem., 267 (2000) 3370. [27] J. Kim, M. Powalla, S. Lang, F. Wagner, H. Lunsdorf and V. Wray, J. Biotechnol., 13 (1990)257. [28] S. Itoh and S. Inoue, Appl. Environ. Microbiol., 43 (1982) 1278. [29] P. Rapp and H.E. Gabriel-Jurgens, Microbiology, 149 (2003) 2879. [30] http://www.cyberlipid.org. [31] D.G. Cooper and D A Paddock, Appl. Environ. Microbiol., 46 (1983) 1426. [32] D. Sorensen, T.H. Nielsen, C. Christophersen, J. Sorensen and M. Gajhede, Acta Crys., C57 (2001) 1123. [33] M. Morikawa, H. Daido, T. Takao, S. Murata, Y. Shimonishi and T. Imanaka, J. Bacteriol., 175(1993)6459. [34] J.D. Epperson and L. Ming, Biochemistry, 39 (2000) 4037. [35] M. Moormann, U. Zahringer, H. Moll, R. Kaufinann, R. Schmid and K. Altendorf, J. Biol. Chem., 272 (1997) 10729. [36] M. Doi, S. Fujita, Y. Katsuya, M. Sasaki, T. Taniguchi and H. Hasegawa, Arch. Biochem. Biophys., 395 (2001) 85. [37] J.A. Tnschman, PR. Jensen and W. Fenical, Tetrahedron Lett., 35 (1994) 5571 [38] MM. Yakimov, K.N. Timmis, V. Wray and H.L. Frederickson, Appl. Environ. Microbiol, 61 (1995)1706. [39] M. Fridkin, I. Ofek, H. Tsubery and S. Cohen, http://www.weizmann.ac.il. [40] H. Ui, T. Miyake, H. Iinuma, M. Imoto, H. Naganawa, S. Hatton, M. Hamada, T. Takehuchi, S. Umezawaand K. Umezawa, J. Org. Chem., 62 (1997) 103. [41] I. Kuiper, L. Lagendijk, R. Pickford, J.P. Derrick, EM. Lamers, J.E. Thomas-Oates, JJ. Lugtenberg and G.V. Bloemberg, Molec. Microbiol., 51 (2004) 97. [42] P.W. Lindum, U. Anthoni, C. Christophersen, L. Eberl, S. Molin and M. Givskov, J. Bacteriol., 180(1998)6384. [43] M. Kowall, J. Vater, B. Kluge, T. Stein, P. Franke and D. Ziessow, J. Colloid Interface Sci., 204 (1998)1.
276
[44] A. Segre, R.C. Bachmann, A. Ballio, F. Bossa, I. Grgurina, N.S. Iacobellis, G. Marino, P. Pucci, M. Simmaco and J.Y. Takemoto, FEBS Lett., 255 (1989) 27. [45] T.H. Nielsen, C. Thrane, C. Christophersen, U. Anthoni, and J. Sorensen, J. Appl. Microbiol., 89 (2000) 992. [46] T.R. Burke, M. Knight and B. Chandrasekhar, Tetrahedron Lett, 30 (1989) 519. [47] T.H. Nielsen, C. Christophersen, U. Anthoni, and J. Sorensen, J. Appl. Microbiol., 86 (1999)80. [48] R.J. Mortishire-Smith, J.C. Nutkins, L.C. Packman, C.L. Brodey, P.B. Rainey, K. Johnstone and D.H. Williams, Tetrahedron, 47 (1991) 3645. [49] J.E. Zajic and W. Seffens, CRC Critical Rev. Biotechnol., 1 (1984) 87. [50] P.G. Carnllo, C. Mardaraz, S.J. Pitta-Alvarez and A.M. Giuletti, World J. Microbiol. Biotechnol., 12(1996)82. [51] I.M. Banat, Biotechnol. Lett, 15 (1993) 591. [52] H. Yonebayashi, S. Yoshida, K. Ono and H. Enomoto, Sekiyu Gakkaishi, 43 (2000) 59. [53] C.N. Mulligan, D.G. Cooper and R.J. Neufeld, J. Ferment. Technol., 62 (1984) 311. [54] D.K. Jain, D.L. Collins-Thompson, H. Lee and J.T. Trevors, J. Microbiol. Methods., 13 (1991)271. [55] A.A. Bodour and R.M. Miller-Maier, J. Microbiol. Meth., 32 (1998) 273. [56] M. Morikawa, Y. Hirata and T. Imanaka, Biochim. Biophys. Acta, 1488 (2000) 211. [57] E.Z. Ron and E. Rosenberg, Environ. Microbiol., 3 (2001) 229. [58] C. Fuqua, MR. Parsek and E.P. Greenberg, Ann. Rev. Gen., 35 (2001) 439. [59] D.G. Davies, MR. Parsek, J.P. Pearson, B.H. Iglewski, J.W. Costerton and E.P. Greenberg, Science, 280 (1998) 295. [60] G. Medina, K. Juarez, B. Valderrama and G. Soberon_Chavez, J. Bactenol., 185 (2003) 5976. [61] L. Eberl, S. Molin and M. Givskov, J. Bacteriol., 181 (1999) 1703. [62] E. Dickinson (ed.), Food Emulsions and Foams, The Royal Society of Chemistry, London, 1987, pp. 1-29. [63] G.O. Philips, D.J. Wedlock and P A Williams (eds), Gums and Stabilizers for the Food Industry, IRL Press, Oxford, 1988, pp. 249-263. [64] A. Stephen (ed.), Food Polysaccharides and their Applications, Marcel Dekker, N.Y. 1995, pp. 501-515. [65] A. Reisfeld, E. Rosenberg and D.L. Gutnick, Appl. Microbiol., 24 (1972) 363. [66] A. Oberbremer, R. Miller-Hurting and F. Wagner, Appl. Microbiol. Biotechnol., 32 (1990)485. [67] Y. Zhang and R.M. Miller, Appl. Environ. Microbiol., 58 (1992) 3276. [68] M.F. Van Dyke, P. Couture, M. Brauer, H. Lee and J.T. Trevors, Can. J. Microbiol., 39 (1993)1071. [69] M.A. Providenti, C.A. Flemming, H. Lee and J.T. Trevors, FEMS Microb. Ecol.,17 (1995) 15. [70] M. Marin, A. Pedregosa and F. Laborda, Appl. Microbiol. Biotechnol., 44 (1996) 660. [71] D.G. Cooper and B.G. Goldenberg, Appl. Environ. Microbiol., 53 (1987) 224. [72] C. Calvo, F. Martinez-Checa, F.L. Toledo, J. Porcel and E. Quesada, Appl. Microbiol. Biotechnol., 60 (2002) 347. [73] S. Navon-Venezia, Z. Zosim, A. Gottlieb, R. Legmann, S. Carmeli, E.Z. Ron and E. Rosenberg, Appl. Environ. Microbiol., 61 (1995) 3240. [74] E. Rosenberg, C. Rubinobitz, R. Legmann and E.Z. Ron, Appl. Environ. Microbiol., 54 (1988)323.
277
[75] A. Zuckerberg, A. Diver, Z. Peeri, D.L. Gutnick and E. Rosenberg, Appl. Environ. Microbiol.,37(1979)414. [76] M.C. Cirigliano and G.M. Carman, Appl. Environ. Microbiol., 50 (1985) 846. [77] G. Whyte, S.J. Slagman, F. Pietrantonio, L. Bourbonniere, S.F. Koval, J.R. Lawrence, W.E. Inniss and C.W. Greer, Appl. Environ. Microbiol., 65 (1999) 2961. [78] S.J. Hamza, A.S. Abbas and A.A. Halob, J. Islamic Academy Sci., 7 (1994) 1. [79] S. Lin, M.A. Minton, M.M. Sharma and G. Georgiou, Appl. Environ. Microbiol., 60 (1994)31. [80] T.H. Nielsen, D. Sorensen, C. Tobiasen, J.B. Andersen, C. Christophersen, M. Givskov and J. Sorensen, Appl. Environ. Microbiol., 68 (2002) 3416. [81] A.J. Cutler and R.J. Light, J. Biol. Chem., 254 (1979) 1944. [82] D.R. Cameron, D.G. Cooper and R.J. Neufeld, Appl. Environ. Microbiol., 54 (1988) 1420. [83] M. Shimazu, M. Saito and K. Yamauchi, Agric. Biol. Chem., 49 (1985) 189. [84] M. Hatton and K. Takahashi, Food Hydrocol., 7 (1993) 325. [85] C.W. Cumperand AE. Alexander, Trans. Faraday Soc, 46 (1950) 235. [86] E. Dickinson, B.S. Murray, G. Stamsby and D.M.W. Anderson, Food Hydrocol., 2 (1988)477. [87] M. Shimazu, T. Kamiya and K. Yamauchi, Agric. Biol. Chem., 45 (1981) 2491. [88] IE. Kinsella, Cnt Rev. Food Sci. Nutr, 21 (1984) 197. [89] R.D. Wamska and J.E. Kinsella, J. Agric. Food Chem., 33 (1985) 1143. [90] E. Dickinson and M. Lai, Adv. Molec. Rel. Intern. Processes, 17 (1980) 1. [91] J.A. Defeyter and J. Benjamins, J. Colloid Interf. Sci., 9 (1982) 289. [92] E. Dickinson and S.E. Euston, Molec. Phys., 68 (1989) 407. [93] E. Dickinson and S.E. Euston, J. Chem. Soc. Faraday Trans., 86 (1990) 805. [94] E. Dickinson (ed), Food Polymers, Gels and Colloids, The Royal Society of Chemistry, Cambridge, 1991, pp. 113-122. [95] G.O. Philips, D.J. Wedlock and P. A. Williams (eds.), Gums and Stabilizers for the Food Industry, Oxford University Press, Oxford, 1992, pp. 351-362. [96] N.S. Hettiarachchy and G.R. Ziegler (eds.), Protein, Functionality in Food System, Marcel Dekker, N.Y., 1994, pp. 225-259. [97] K. Watanabe, H. Toyokawa, A. Shimada and S. Arai, J. Food Sci., 46 (1981) 1467. [98] M. Saito, M. Ogasawara, K. Chikuni and M. Shimizu, Biosci. Biotechnol. Biochem., 59 (1995)388. [99] Z. Mozaffar and Z.U. Haque, Food Hydrocol., 5 (1994) 573. [100] S.W. Lee, M. Shimizu, S. Kaminogawa and K. Yamauchi, Agric. Biol. Chem., 51 (1987) 161. [101] M. Saito, K. Chikuni, M. Monma and M. Shimizu, Biosci. Biotechnol. Biochem., 57 (1993)952. [102] M. Watanabe, N. Fujn and S. Arai, Agric. Biol. Chem., 46 (1982) 1587. [103] E. Dickinson and D. Lorient (eds.), Food Macromolecules and Colloids, The Royal Society of Chemistry, Cambridge, 1995 pp. 1-19. [104] G. Williamson, C.B. Faulds, J.A. Matthews, V.J. Morris and GJ. Brownsey, Carbohyd. Polymers, 13(1990)387. [105] J.R. Mitchell and D.A. Ledward (eds), Functional properties of food macromolecules, Elsevier Applied Science, London, 1986, pp. 315-3 54. [106] Y.A. Antonov, N.P. Lashko, Y.K. Glotova, A. Malovikova and O. Markovich, Food Hydrocol., 10(1996)1. [107] V.B. Tolstoguzov and E.E. Braudo, J. Disp. Sci. Technol., 6 (1985) 575.
278
[108] K.D. Schwenke and R. Mothes (eds.), Food Proteins. Structure and Functionality, VCH, Weinheim, 1993, pp. 203-209. [109] G.O. Philips, P. A. Williams and D.J. Wedlock (eds.), Gums and Stabilisers for the Food Industry, IRL Press, Oxford, 1994, pp. 115-124. [110] V. Grinberg and V.B. Tolstoguzov, Food Hydrocol., 11 (1997) 145. [Ill] N. Garti, Y. Slavin and A. Aserin, Food Hydrocol., 13 (1998) 145. [112] N. Garti and M.E. Leser, Polym. Adv. Technol, 12 (2001) 123. [113] W. Liu and T. Sato, Carbohydr. Research, 160(1987)267. [114] J. Greener, B.A. Contestable and M.D. Bale, Macromolecules, 20 (1987) 2490. [115] E. Dickinson (ed.), Food Emulsions and Foams, The Royal Society of Chemistry, London, 1987, pp. 1-29. [116] G.O. Philips, P.A. Williams and D.J. Wedlock (eds.) Gums and Stabilizers for the Food Industry, IRL Press, Oxford, 1990, pp. 157-175. [117] G.L. Hasenhuettle and R.W. Hartel (eds.), Food Emulsifiers and their Applications, Thomson Publishing, 1997. [118] S.E. Hill, A.P. Ledward and J.R. Mitchell (eds.), Functional Properties of Food Macromolecules, Aspen Publisher, Maryland, 1998, pp. 252-277. [119] Y. Cao, E. Dickinson and D.J. Wedlock, Food Hydrocol., 5 (1991) 443. [120] E. Dickinson, V.B. Galazka and D.M.W. Anderson, Carbohyd. Polymers, 14 (1991) 373. [121] E. Dickinson and P. Walstra (eds.) Food colloids and polymers: stability and mechanical properties, The Royal Society of Chemistry, Cambridge, 1993, pp. 3-15. [122] V. Ya, V. Grinberg and V.B. Tolstoguzov, Food Hydrocol., 11 (1997) 145. [123] A. Kato, T. Sato and K. Kobayashi, Agric. Biol. Chem., 53 (1989) 2147. [124] A. Kato, Y. Sasaki, R. Furuta and K. Kobayashi, Agric. Biol. Chem., 54 (1990) 107. [125] S. Nakamura, A. Kato and K. Kobayashi, J. Agric. Food Chem., 40 (1992) 735. [126] M. Hattori, S. Imamura, K. Nagasawa and K. Takahashi, Biosci. Biotech. Biochem., 58 (1994) 174. [127] M. Hattori, S. Imamura, K. Nagasawa and K. Takahashi, J. Agric. Food Chem., 42 (1994)2120. [128] E. Dickinson and V.B. Galazka, Food Hydrocol., 5 (1995) 281. [129] K. Nagasawa, K. Takahashi and M. Hattori, Food Hydrocol., 10 (1996) 63. [130] J.M.V. Blanshard and JR. Mitchell (eds.) Polysaccharides in Food, Butterworths, London, 1979, pp. 205-217. [131] J.R. Mitchell and D.A. Ledward (eds.), Functional Properties of Food Macromolecules, Elsevier Applied Science, London, 1986, pp. 385-415. [132] M. Ahmed and E. Dickinson, Colloid Surface, 47 (1990) 353. [133] N. Garti and D. Reichman, Food Hydrocol., 8 (1994) 155. [134] R.C. Randall, G.O. Philips and P.A. Williams, Food Hydrocol., 2 (1988) 131. [135] J.E. Mckay, G. Stainsby and E.L. Wilson, Carbohyd. Polymers, 5 (1985) 223. [136] K. Nishinari and E. Doi (eds.) Food Hydrocolloid-structures, properties and functions,. Plenum Press, N.Y., 1982, pp. 211-224. [137] Y.A. Antonov, J. Lefevre and J. Doublier, Appl. Polymer Sci., 71 (1999) 471. [138] P. Harris (ed.), Food Gels, Elsevier Applied Science, London, 1990, pp. 291-359. [139] H. Bach, Y. Berdichevsky and D. Gutnick, Appl. Environ. Microbiol., 69 (2003) 2608. [140] D.L. Gutnick and H. Bach. US Patent No. 6 512 014 (2003). [141] S. Ohashi, F. Ura, M. Takeuchi, H. Lida, K. Sakaue, T. Ochi, S. Ukai and K. Hiramatsu, Food Hydrocol., 4 (1990) 323. [142] E. Dickinson and K. Pawlowsky, Food Hydrocol., 12 (1998) 417.
279
[143] K. Maenaka, G. Kawai, K. Watanabe, F. Sunada and I. Kumagai, J. Biol. Chem., 269 (1994)7070. [144] J.W. Goodwin (ED), Colloidal Dispersions, The Royal Society of Chemistry, London, 1982, pp. 99-128. [145] M.A. Muchkin, E.S. Wajnermann and V.B. Tolstoguzov, Nahrung, 20 (1976) 313. [146] R.G. Chilvers and V. J. Morris, Carbohyd. Polymers, 7 (1987) 111. [147] H.R Kruyt (ed.), Colloid Science: reversible systems, Elsevier, Amsterdam, 1949,pp. 232-258. [148] D.A. Imeson, A.P. Ledward and JR. Mitchell, J. Sci. Food Agnc, 28 (1977) 661. [149] E. Rosenberg, A. Zuckerberg, C. Rubinovitz and D.L. Gutnick, Appl. Environ. Microbiol.,37(1979)402. [150] E. Rosenberg, A. Perry, D.T. Gibson and D.L. Gutnick, Appl. Environ. Microbiol., 37 (1979)409. [151] D. Gutnick, Biopolymers, 26 (1987) S223. [152] N. Kosaric (ed), Biosurfactants and Biotechnology, Marcel Dekker, N.Y., 1987, pp. 211-246. [153] K.J. Towner, E. Bergogne-Berezin and C. A. Fewson (eds.) The Biology of Acinetobacter, Plenum Press, N.Y., 1991, pp. 411-441. [154] D.L. Gutnick, E. Rosenberg and Y. Shabtai. US Patent No. 4 234 689 (1980). [155] D.L. Gutnick, E. Rosenberg, I. Belsky and Z Zosim. US Patent No. 4 311 830 (1982). [156] D.L. Gutnick, E. Rosenberg, I. Belsky and Z. Zosim. US Patent No. 4 395 3 54 (1983). [157] N. Kaplan, E. Rosenberg, B. Jann and K. Jann, Eur. J. Biochem., 152 (1985) 453. [158] D.L. Gutnick, Personal Communication, (2004). [159] C.R. Macdonald, D.G. Cooper and J.E. Zajic, Appl. Environ. Microbiol., 41 (1981) 117. [160] D.G. Cooper, J.E. Zajic and D.F. Gerson, Appl. Environ. Microbiol., 37 (1979) 4. [161] A. Gorkovenko, J. Zhang, R.A. Gross, A.L. Allen and D.L. Kaplan, Can. J. Microbiol., 43(1997)384. [162] A. Gorkovenko, J. Zhang, R.A. Gross, D.L. Kaplan and A.L. Allen, Carbohydr. Polym., 39(1999)79. [163] A.K. Johri, W. Blank and D.L. Kaplan, Appl. Microbiol. Biotechnol., 59 (2002) 217. [164] J. Zhang, A. Gorkovenko, R.A. Gross, A.L. Allen and D. Kaplan, Int. J. Biol. Macromol., 20 (1997) 9. [165] I. Belsky, D.L. Gutnick and E. Rosenberg, FEBS Lett, 101 (1979) 175. [166] Y. Shabtai and D.L. Gutnick, Appl. Environ. Microbiol., 49 (1985) 192. [167] B.K. Tuleva, G.R Ivanov and N.E. Chnstova, Naturforsch, 57C (2002) 356. [168] A.M. Davila, R. Marchal and J.P. Vandecasteele, Appl. Microbiol. Biotechnol., 47 (1997)496. [169] Z. Zosim, D.L. Gutnick and E. Rosenberg, Biotechnol. Bioeng., 25 (1983) 1725. [170] D.L. Gutnick and H. Bach, Appl. Microbiol. Biotechnol., 54 (2000) 451. [171] S. Goldman, Y. Shabtai, C. Rubinobitz, E. Rosenberg and D. Gutnick, Appl. Environ. Microbiol., 44(1982)165. [172] Y. Shabtai and D.L. Gutnick, J Bacteriol., 161 (1985)1176. [173] S.D. Wang and D.I.C. Wang, Biotechnol. Bioeng., 36 (1990) 402. [174] O. Pines and D. Gutnick, J. Bacteriol., 157 (1984) 179. [175] O. Pines, Y. Shoham, E. Rosenberg and D.L. Gutnick, Appl. Microbiol. Biotechnol., 28 (1988)93. [176] Y. Shoham, M. Rosenberg and E. Rosenberg, Appl. Environ. Microbiol, 46 (1983) 573. [177] M. Rosenberg and E. Rosenberg, J. Bacteriol., 148 (1981) 51.
280
[178] M. Rosenberg, A. Perry, E.A. Bayer, D.L. Gutnick, E. Rosenberg and I. Ofek, Infect Immun., 33(1981)29. [179] T. Ophir and D. Gutnick, Appl. Environ. Microbiol., 60 (1994) 740. [180] D. Nakarand D.L. Gutnick, Microbiology, 147 (2001) 1937. [181] D. Nakarand D.L. Gutnick, J. Bactenol, 185 (2003) 1001. [182] Mijakovic, S. Poncet, G. Boel, A. Maze, S. Gillet, E. Jamet, P. Decottigmes, C. Grangeasse, P. Doublet, P. LeJVlarechal and J. Deutscher, Embo J., 22 (2003) 4709. [183] Y. Shoham, E. Rosenberg and D.L. Gutnick, US Patent No. 4 704 360 (1983). [184] P.G. Reddy, R. Allon, M. Mevarech, S. Mendelovitz, Y. Sato and D.L. Gutnick, Gene, 76(1989)145. [185] R. Alon, Esterase from the oil-degrading Acinetobacter lwoffii RAG-1: expression of the est gene and protein characterization. Ph.D. Thesis, Tel-Aviv University (1993). [186] R.N. Alon and D.L. Gutnick, FEMS Microbiol. Lett, 112 (1993) 275. [187] R.N. Alon, L. Mirny, J.L. Sussman and D.L. Gutnick, FEBS Lett, 371 (1995) 231. [188] R.G. Kok, J.J. Van Thor, I.M. NugterenJRoodzant, MB. Brouwer, M.R. Egmond, C.B. Nudel, B. Vosman and K.J. Hellingwerf, Mol. Microbiol., 15 (1995) 803. [189] H. Bach, Y. Mazor, S. Shaky, A. Shoham Lev, Y. Berdichevsky, D.L. Gutnick and I. Benhar, J. Mol. Biol., 312 (2001) 79. [190] N. Sar and E. Rosenberg, Curr. Microbiol., 9 (1983) 309. [191] A. Toren, E. Orr, Y. Paitan, E.Z. Ron and E. Rosenberg, J. Bactenol., 184 (2002) 165 [192] A. Toren, E.Z. Ron, R. Bekerman and E. Rosenberg, Appl. Microbiol. Biotechnol, 59 (2002) 580. [193] E. Pessione, M.G. Giuffrida, L. Prunotto, C. Barello, R. Mazzoli, D. Fortunate, A. Conti and C. Giunta, 3 (2003) 1070. [194] E. Rosenberg, C. Rubinobitz, A. Gottlieb, S. Rosenhak and E.Z. Ron, Appl. Environ. Microbiol., 54 (1988) 317. [195] W.H. Taylor and E. Jum, J. Bacteriol., 81 (1961) 688. [196] N. Kaplan, Z. Zosim and E. Rosenberg, Appl. Environ. Microbiol., 53 (1987) 440. [197] O. Abian, C. Mateo, G. FernandezLorente, J.M. Guisan and R. Fernandez_Lafuente, Biotechnol. Progr., 19 (2003) 1639. [198] F. Arendt, M. Hinsenveld, and W.J. van den Brink (eds), Contaminated soil '90, Kluwer Academic Publishers, Dordrecht, The Netherlands, 1990, pp. 491-492. [199] C.N. Mulligan, T.Y.K. Chowand B.F. Gibbs, Appl. Microbiol. Biotechnol, 31 (1989) 486. [200] GL. Ghurye, C. Vipulanandan and R.C. Willson, Biotechnol. Bioeng., 44 (1994) 661. [201] M.E. Mercade, M.A. Manresa, M. Robert, M.J. Espuny, C. De Andres and J. Guinea, Bioresource Technol., 43 (1993) 1. [202] M.E. Mercade and M.A. Manresa, J. Am. Oil Chem. Soc, 71 (1994)61. [203] H.J. Daniel, R.T Otto, M. Binder, M. Reuss, and C. Syldatk, Appl. Microbiol. Biotechnol., 51 (1999)40. [204] U.A. Ochsner, A. Fiechter and J. Reiser, J. Biol. Chem., 269 (1994) 19787. [205] U.A. Ochsner, J. Reiser, A. Fiechter and B. Witholt, Appl. Environ. Microbiol., 61 (1995)3503. [206] W.C. Mccaffrey and D.G. Cooper, J. Ferment. Bioeng., 79 (1995) 146. [207] M.E. Hayes, K.R. Hrebenar, P.L. Murphy, L.E. Futch Jr., J.F. Deal III and P.L. Bolden Jr. US Patent No. 4 684 372 (1987). [208] M.E. Hayes, K.R. Hrebenar, P.L. Murphy, L.E. Futch Jr., J.F. Deal m and P.L. Bolden Jr. US Patent No. 4 793 826 (1988).
281
[209] ME. Hayes, K.R Hrebenar, PL Murphy, L.E. Futch Jr., J.F. Deal in and PL. Bolden Jr.US Patent No. 4 821 757 (1989). [210] ME. Hayes, K.R. Hrebenar, J.L. Minor and L.M. Woodworth, US Patent No. 4 886 519 (1989) [211] D.L. Gutnick and E. Rosenberg, Ann. Rev. Microbiol, 31 (1977) 379. [212] I.M. Banat, N. Samarah, M. Murad, R. Home and S. Benerjee, World J. Microbiol. Biotechnol, 7(1991)80. [213]M.E. Hayes, K.R. Hrebenar, J.F. Deal in and PL. Bolden Jr. US Patent No. 4 666 457 (1987).
This page is intentionally left blank
Studies in Surface Science and Catalysis 151 R. Vazquez-Duhalt and R. Quintero-Ramirez (Editors) © 2004 Elsevier B .V. All rights reserved.
283
Chapter 10
Anaerobic hydrocarbon biodegradation and the prospects for microbial enhanced energy production J.M. Suflitaab, LA. Davidovaab, L.M. Giegab, M. Nanny ac and R.C. Princed a
Institute for Energy and the Environment, bDepartment of Botany and Microbiology, cSchool of Civil Engineering and Environmental Science, University of Oklahoma, Norman, OK 73019, USA. d
ExxonMobil Research and Engineering Co., Annandale, NJ 08801, USA.
1. INTRODUCTION Hydrocarbon-based energy underpins the economic, social, and political fabric of the world and demand for oil is expected to grow unabated for the foreseeable future. It is forecast that global oil consumption will increase annually by an average of more than 4 x 107 barrels per day to eventually reach 4.3 x 1010 barrels per year by 2020 [1]. This is projected to be about a 58% increase over current usage by 2025 [2]. The U.S. Geological Survey recently predicted that about 3 trillion barrels of oil remain to be recovered worldwide (with 1 trillion barrels already harvested), half from proven reserves and half from undeveloped or undiscovered sources [3]. However, as proven reserves get exploited and develop into mature fields, secondary and tertiary recovery technologies will increasingly be relied upon to obtain the remaining residual oil. This is particularly true for the U.S. and other oil-importing countries that tend to rely more heavily on mature, domestic energy sources. With demand far surpassing energy production in the U.S., there is heightened interest in diversifying energy sources, tapping unconventional energy supplies and the development of new technology to more fully exploit domestic reserves. Although oil is expected to remain the dominant energy fuel in the next 20 years, the use of natural gas as a substantial energy source has risen significantly in the past 10 years [4]. In fact, natural gas is projected to be the fastest growing primary energy source and an increasingly important alternative to oil [2]. Natural gas, consisting mainly of methane (>95%) but also with small amounts of other short-chain hydrocarbons (C2 to C4), can be harvested from
284
large gas fields, sometimes associated with oil reservoirs, or be obtained from unconventional sources such as shale, coalbeds, or tight sands. The use of natural gas is becoming increasingly popular due to its abundance across the globe. Further, the lower price of natural gas relative to oil makes it an attractive energy source [5]. Natural gas is also a cleaner energy source than oil or coal, and thus can help reduce greenhouse gas emissions [6]. Natural gas combustion produces only about 56% and 71% of the CO2 associated with the equivalent amount of energy produced from coal or oil, respectively [Energy Information Administration (1999). Natural Gas 1998: Issues and Trends (http://www.eia. doe.gov/oil_gas/natural_gas/analysis_publications/natural_gas_1998_issues_and _trends/it98.html). Moreover, methane use results in less NOX, SO2, and particulates per equivalent amount of energy generated, relative to other sources. Despite the increasing use of natural gas and its attendant environmental advantages, world reliance on oil is unlikely to wane in the near future given the existing energy infrastructure and the aforementioned dependence of many societies on this energy form. However, there is a biotechnological link between oil and natural gas that is the product of the relatively recent recognition that many hydrocarbons are susceptible to anaerobic biodegradation and can be converted to methane and carbon dioxide [7-9]. Unlike the well-documented patterns of aerobic oil biodegradation [10], anaerobic hydrocarbon metabolism was essentially dismissed as ecologically insignificant for many years. This view has been completely altered in recent years with the growing appreciation for the metabolism of hydrocarbons coupled with the consumption of electron acceptors other than oxygen. Not surprisingly then, the majority of world oil reserves are believed to be biodegraded to at least some degree, but it was generally accepted that aerobic oxidation processes were largely responsible for such alterations [11, 12]. Recent evaluations of many petroliferous formations have convincingly argued that it is actually anaerobic processes that predominate in oil and gas reservoirs, sometimes leading to the production of biogenic methane [12-14]. Geological evidence has suggested that such methanogenic processes occur very slowly over millennia, and are most important in reservoirs shallower than 4 km and at temperatures of less than 80°C [12, 15, 16]. Microbial decay of oils in deep subsurface reservoirs can clearly reduce oil quality, and a better understanding of the microbial principles behind such decay will be important to help distinguish between degraded, low-value oils and untouched, high-value oils [11]. However, if methanogenesis continues to be identified as an important process in deep reservoirs worldwide, the recovery of methane gas as an alternate form of energy from otherwise unrecoverable or biodegraded sources might have far-reaching economic and environmental implications. The purpose of this chapter is to review evidence for anaerobic hydrocarbon biodegradation and to provide an overview of some of the more
285
generalizing metabolic features. We will explore whether these reactions can be predicted and identify some of the implications for the ability of anaerobes to convert hydrocarbons into methane and and thereby generate useful energy.
2. ANAEROBIC HYDROGEN METABOLISM Crude oils are enormously complex mixtures containing tens of thousands of individual components [17]. Even condensates and refined products include a dizzying array of constituent hydrocarbons. Once released in the environment, the relative concentrations of these chemicals change over time reflecting individual susceptibilities to various fate processes such as sorption, volatilization, dispersion and biodegradation [18]. One way of assessing the susceptibility of such complex mixtures to anaerobic biodegradation is to consider individual chemical classes of hydrocarbons and determine how metabolism varies with structural complexity (see below). Two decades ago, it was generally considered that aliphatic and aromatic hydrocarbons could only be mineralized in the presence of oxygen. Oxygen served as both a respiratory electron acceptor and as a co-substrate for mono- and dioxygenases catalyzing initial hydrocarbon activation steps [19, 20]. We now know that microbial metabolism is much more diverse than this and many classes of hydrocarbons are amenable to microbial attack under a variety of anaerobic conditions. Substrates include BTEX (benzene, toluene, ethylbenzene and xylenes) compounds [21], polycyclic aromatic hydrocarbons [22], saturated and branched alkanes [23, 24] and alicyclic hydrocarbons [25, 26]. Knowledge of the mechanisms used by anaerobes to catalyze such transformations are summarized here. Toluene has historically served as a model substrate to study anaerobic alkylbenzene biodegradation. Initial work with denitrifying strains of Thauera and Azoarcus [27, 28] showed that the first step of toluene degradation occurred by the addition of the aryl methyl carbon to the double bond of fumarate to yield benzylsuccinic acid. This remarkable reaction is catalyzed by a novel glycyl radical-containing enzyme, benzylsuccinate synthase [29, 30]. Another denitrifying strain EbNl, [31], the sulfate-reducing isolates Desulfobacula toluolica and strain PRTOL1 [31, 32], the iron reducer Geobacter metallireducens [33], a defined methanogenic consortium [34] and an anaerobic phototroph [35] have also been shown to carry out this initial toluene transformation reaction. Subsequent transformations of benzylsuccinate lead to the formation of benzoyl-CoA and succinyl-CoA [28, 36, 37]. Benzylsuccinate gets thioesterified to a CoA derivative in a succinyl-CoA-dependent reaction and the resulting product is then oxidized to is-phenylitaconyl-CoA. The latter compound presumably undergoes modified [3-oxidation yielding benzoyl-CoA and regenerates succinyl-CoA. The succinyl-CoA-(i?)-benzylsuccinate CoA-
286
transferase and (7?)-benzylsuccinyl-CoA dehydrogenase responsible for these bioconversion steps have recently been purified and characterized [37, 38]. Benzoyl-CoA, a central metabolite in the anaerobic oxidation of numerous aromatic compounds, is presumably further oxidized to acetyl-CoA and CO2 as described by Harwood et al. [39]. Another addition reaction for anaerobic toluene decay has also been reported. Toluene activation by the denitrifying strain Azoarcus tolulyticus Tol-4 occurs in two consecutive steps with a 2-carbon fragment (presumably acetylCoA) to ultimately form benzylsuccinate [40]. Evidence supporting this mechanism includes the detection of cinnamic acid in cultures receiving toluene and the production of radiolabeled benzylsuccinic acid from incubations amended with trans-cvcmsmic acid and 14C-acetate. Benzylsuccinate formation is believed to be preceded by the formation of hydrocinnamoyl-CoA and cinnamoyl-CoA intermediates. Studies on this toluene pathway are rare so speculation on how common it might be is difficult. Biodegradation of m-xylene and o-xylene has been observed in enrichments under sulfate-reducing [41-43], nitrate-reducing [44] and methanogenic conditions [45]. The anaerobic biodegradation of p-xylene has only been demonstrated under sulfate- [41, 43] and nitrate-reducing conditions [46]. A handful of nitrate- and sulfate-reducing bacteria capable of m-xylene and o-xylene biodegradation have been isolated [47-51]. Under denitrifying conditions, m- and o-xylene are activated by fumarate addition reactions by Azoarcus sp. strain T [28, 52]. Tentative identification of £-(3-methylphenyl)itaconyl-CoA and 3-methylbenzoate from /w-xylene suggested a pathway analogous to that of anaerobic toluene oxidation [52]. Biochemical, molecular, and genetic studies of benzylsuccinate synthase from Azoarcus sp. strain T proved that this enzyme catalyzed the initial reaction in anaerobic biodegradation of both toluene and m-xylene [52, 53]. Further, partially purified benzylsuccinate synthase transformed all three xylene isomers to their methylbenzylsuccinate analogs [54]. Under sulfate-reducing conditions, wholecell suspensions of the toluene-grown sulfidogenic culture PRTOLl transformed o-xylene to (2-methylbenzyl)succinate. However, the cell could not grow on oxylene suggesting the involvement of benzylsuccinate synthase [55] in the transformation of the parent molecule. In sulfate-reducing enrichments, 2-, 3-, and 4-methybenzylsuccinates were positively identified as metabolites of 0-, m-, and p-xylene, respectively [43] indicating that parent molecules were anaerobically attacked in a comparable fashion when sulfate served as a terminal electron acceptor. There are multiple pathways for ethylbenzene metabolism under anaerobic conditions. Three pure denitrifying cultures were isolated for their ability to completely mineralize ethylbenzene [49, 56]. In these organisms, ethylbenzene is initially activated via dehydrogenation to yield 1-phenylethanol
287
[31,49] by a remarkable molybdenum enzyme that shows clear sequence similarities to the aerobic dimethyl sulfoxide reductase family of molybdopterincontaining enzymes [57]. In this reaction the oxygen atom in the hydroxyl group originates from water [56]. 1-Phenylethanol is then further transformed to acetophenone, and ultimately to benzoyl-CoA [36]. Under sulfate reducing conditions, ethylbenzene was converted to the corresponding ethylbenzylsuccinic acid (3-phenyl-l,2-butane-dicarboxylic acid) by a putative fumarate addition reaction [43]. Recent studies with a pure sulfate-reducing bacterium confirmed this mode of ethylbenzene metabolism [58], with the fumarate addition occurring at the methylene carbon of the side chain rather than at the terminal methyl group. Benzene is the least reactive of all aromatic hydrocarbons and was believed to be entirely recalcitrant under anaerobic conditions. This too has proven to be incorrect. Benzene can be degraded under nitrate- [59- 61], sulfate[62-65], and Fe(III)-reducing conditions [66-68] and with coupling to methane production [69-71]. To date only two pure benzene-degrading strains, both affiliated with the Dechloromonas genus, have been described [60]. Both strains can oxidize benzene with nitrate as an electron acceptor. Physiological studies and 13C-labeling data suggested benzene activation by an initial methylation reaction to form toluene [72]. However, alkylation reactions are not entirely consistent with previous information on anaerobic benzene decay. For example, 13 C-phenol and 13C-benzoate have been detected as intermediates in an enrichment culture incubated with 13C-benzene under sulfate-reducing conditions. 13C-Benzoate was also found in comparable methanogenic and Fe(III)-reducing enrichments [73], suggesting that the hydroxylation of benzene to phenol is one of the initial steps in anaerobic benzene decay. The conversion of phenol to benzoate could then occur by the carboxylation of phenol to form/>hydroxybenzoate followed by the reductive removal of the hydroxyl group to form benzoate. 13Carbon and deuterium labeling studies confirmed that the carboxyl carbon of the benzoate intermediate is derived from one of the carbon atoms of benzene [73, 74]. Therefore, it seems plausible that multiple mechanisms for anaerobic benzene decay also exist amongst the anaerobes. The anaerobic biodegradation of w-alkanes has also been demonstrated recently. Several sulfate-reducing and denitrifying bacterial strains [42, 75-78] as well enrichment cultures [44, 79] are capable of the complete conversion of «-alkanes to carbon dioxide. A broad range of «-alkanes may be susceptible to anaerobic biodegradation, including long-chain alkanes ranging from Ci5 to C34 [80]. All of the sulfate-reducing alkane-degrading strains isolated to date are short oval-shaped rods belonging to -subclass of the Proteobacteria. None of them has been fully characterized and the complete metabolic pathway for anaerobic alkane decay remains speculative. Recently, a product resulting from the initial metabolic activation of a model alkane was identified. When
288
deuterated dodecane was added to an alkane-degrading sulfate-reducing culture, alkylsuccinic acid derivatives with complete deuterium retention were formed, demonstrating that the primary attack on the parent substrate occurred by addition to the double bond of fumarate [79]. The same mechanism was subsequently demonstrated for a denitrifying Azoarcus-like strain [81]. This bioconversion represents a remarkable reaction that superficially resembles the anaerobic biodegradation of toluene. However, alkanes are not activated at a terminal methyl position like toluene or the xylene isomers. Rather the succinyl moiety is attached subterminally at the C2 (and less frequently the C3) position of alkanes [79, 81, 82]. The enzymology of this reaction is still under investigation, though EPR spectroscopy suggests a radical mechanism comparable to that for toluene decay [81]. Another reported alkane activation mechanism involves direct carboxylation with inorganic bicarbonate. In studies with the sulfate-reducer strain Hxd3, So et al. [83] showed that 13C-bicarbonate was added to alkanes at the C-3 position, followed by the elimination of the two adjacent terminal carbon atoms yielding a fatty acid one carbon shorter than the parent alkane. As of this writing, the fumarate addition mechanism seems to be more widespread for «-alkane activation as it has now been shown for three disparate anaerobic cultures [79, 81, 84, 85]. The metabolic steps following the formation of alkylsuccinates are not yet completely clear, but an important contribution has recently been published [86]. Studies performed with the denitrifying strain HxNl grown with deuterated «-hexane or deuterated fumarate revealed the formation of a suite of fatty acids [86]. The identification of 4methyloctanoic acid with deuterium in the C-3 position suggested that the product of fumarate addition, hexylsuccinate (or [l-methylpentyl]succinate), could possibly undergo rearrangement of the carbon skeleton prior to further oxidation. Based on the identification of other transient metabolites, such as 4methyloct-2-enoic and 3-hydroxy-4-methyloctanoic acids, a hypothetical pathway was proposed which allowed for the regeneration of fumarate. Recent studies with 13C-hexane transformation in a sulfate-reducing culture provided evidence supporting the proposed model [82]. Mass spectral and nuclear magnetic resonance (NMR) data indicate that multiple 13C nuclei originating from l-13C-hexane become incorporated into a variety of metabolites. The labeling patterns argue that 13C-fumarate is produced and recycled during hexane biodegradation. 4-Methyloctanoic acid, an important metabolite of the proposed pathway, was positively identified in the organic extracts of 12C- and 13 C-hexane-amended culture supernatants by both mass spectral analysis and 13 C-NMR. Similarly, 3-hydroxy-4-methyloctanoic acid, was also tentatively identified [82]. Common metabolites formed during anaerobic alkane utilization, as well as the presence of multiple 13C-carbons in the metabolites, strongly suggest that this sulfate-reducing culture and a denitrifying strain, HxNl, not
289
only initiate alkane degradation via fumarate addition, but most probably share the entire degradation pathway. Polycyclic aromatic hydrocarbons (PAHs) are also susceptible to anaerobic decay. Naphthalene can be completely mineralized by pure cultures of sulfate-reducing and denitrifying bacteria [87, 88]. Enrichments from coal-tar contaminated sediments and garden soil were reported to mineralize [14C]naphthalene with soluble Fe(III) and insoluble FeOOH, although not more than 15% of added radioactive substrate was recovered as 14CO2 [89]. Anaerobic degradation of phenanthrene was also demonstrated in sediments [22, 90] and by a sulfate-reducing enrichment culture [91]. Studies with marine sediments also indicated the loss of 2 to 5-ringed PAHs under anaerobic conditions, with the smaller PAHs degrading more rapidly than the heavier molecular weight counterparts [90]. It has been shown that unsubstituted PAHs, such as naphthalene and phenanthrene, are initially attacked by carboxylation to form 2naphthoic acid and phenanthrenecarboxylic acid, respectively [91, 92]. The carbon in both cases arises from inorganic CO2. 2-Methylnaphthalene is converted to 2-naphthoic acid following the anaerobic oxidation of the methyl group [93]. A mechanism for the activation of 2-methylnaphthalene is the addition of fumarate to the methyl group [92, 94]. The product of this reaction, naphthyl-2-methyl-succinic acid, is subsequently oxidized to 2-naphthoic acid which further decomposes by ring reduction reactions to form the fully saturated decalin-2-carboxylic acid prior to ring cleavage and ultimate mineralization [91, 92, 95]. Alicyclic hydrocarbons can comprise a substantial fraction (often up to ~12% wt/wt) of the organic molecules in petroleum mixtures. Despite this quantitative importance, little is known about the metabolic fate of this class of materials. Recently, a study of the anaerobic metabolism of a model alicyclic hydrocarbon, ethylcyclopentane, revealed that it too was initially activated by fumarate addition to form ethylcyclopentylsuccinic acid [25]. Wilkes et al. [84] recently observed that when the denitrifying strain HxNl was incubated with crude oil, a series of C4 to C8 «-alkanes as well as cyclic alkanes, were activated to their corresponding alkylsuccinates and methyl-branched fatty acids. Further, cyclopentane, cyclohexane, and their methyl-and ethyl substituted congeners were rapidly consumed in live incubations of sulfate-amended anoxic sediment enrichments from a gas condensate-contaminated aquifer [26]. Though alicyclic biodegradation was more extensive under sulfate-reducing conditions, there was biodegradation of simpler alicyclic compounds under methanogenic conditions. In parallel methanogenic incubations, 90% of cyclopentene and methylcyclopentene was lost in 100 days [26]. Thus, this class of materials is also susceptible to methanogenic biodegradation.
290
3. CHEMICAL THEORY AND THE SUSCEPTIBILITY OF HYDROCARBONS TO FUMARATE ADDITION REACTIONS The section above attests to the diversity of anaerobic hydrocarbon biodegradation reactions. Given the complexity of even simple hydrocarbon mixtures, it is clearly impractical, if not impossible, to assay all component molecules for their susceptibility to anaerobic biodegradation. While relatively little is known of the enzymatic processes and reaction mechanisms responsible for such bioconversions, insight can be gleaned from a consideration of the molecular structure of the reported metabolites. A common theme among many classes of hydrocarbons is the importance of fumarate addition reactions that lead to metabolites containing a succinic acid functional group. The site of fumarate addition to the hydrocarbon is likely a key determinant in the susceptibility of the parent molecules to anaerobic destruction and to accurate predictions of the metabolites that might reasonably be anticipated. Thus, accurate prediction of the fumarate addition site based upon chemical principles is essential. In a very general sense, fumarate addition to a hydrocarbon substrate can be thought of as a four-step process. The first step involves complexation of the hydrocarbon substrate and enzyme, and is controlled by factors such as diffusion of the hydrocarbon substrate to the enzyme and the thermodynamics of the enzyme-substrate complex. The second step involves oxidation of the hydrocarbon substrate, hypothesized by us to occur via a hydrogen atom transfer (HAT). This is a one-electron oxidation of the substrate through abstraction of a hydrogen radical (H) transforming the hydrocarbon substrate into a free radical intermediate [96]. In the case of alkyl aromatic compounds, a second oxidation mechanism is also possible, one that occurs through electron transfer (ET) [96]. ET involves the removal of an electron from the -orbitals of the aromatic ring, resulting in an aromatic radical cation. The aromatic radical cation then looses a proton (H+) to the surrounding matrix and is transformed into the hydrocarbon radical intermediate. It is hypothesized that the position of the radical in the hydrocarbon is controlled by the stability of the hydrocarbon radical intermediate, and therefore the site of fumarate addition can be readily predicted based upon chemical rules governing free radical stability. The third step involves addition of fumarate to the hydrocarbon radical, leading to the next step which is the release of the newly formed metabolite from the enzyme. The ensuing discussion focuses on the hypothesis that the structure of succinic acid metabolites suggests that anaerobic fumarate addition reactions proceed via a radical mechanism; either by a hydrogen abstraction transfer or an electron transfer mechanism. As a result, metabolite structure can be predicted for alkane, alicyclic and alkyl aromatic hydrocarbons based upon the chemical rules that govern free radical stability.
291
As noted, fumarate addition to «-alkanes predominantly occurs at the subterminal C2 position and to a lesser degree at the C3 position. Fumarate addition to these positions, rather than the less sterically-hindered terminal methyl position, strongly suggests a HAT mechanism. In the HAT mechanism, the site of homolytic cleavage {reaction 1) is a function of the C-H bond dissociation energy. RH->R+H
(1)
The bond dissociation energy (AH2g8) is a function of the stability of the free radical intermediate (R*). For alkane compounds, the relative stability of radical intermediates is of the order: tertiary > secondary > primary [97]. Thus, the ease of abstracting a hydrogen radical from carbon atoms in an alkane follows the same relative trend as the bond dissociation energy as seen for alkanes in Table 1. The relative stability pattern in alkanes results from hyperconjugation, that is, delocalization involving a bonds. The greater the number of hyperconjugative forms that can be generated for a free radical intermediate, the greater the stability of that intermediate [97]. The bond dissociation energy data in Table 1 presents a 1 to 3 kcal mol"1 difference between n-alkane methyl and methylene groups. More importantly, however, is the fact that a 1 kcal mol"1 difference exists between the terminal methyl group and the subterminal C2 carbon for both pentane and hexane, thus illustrating the favorable reactivity of the subterminal C2 carbon relative the terminal methyl group. Observation of fumarate addition to the C3 carbon is not surprising since the difference between the bond dissociation energies of the C3 and C2 methylene carbons is expected to be minimal, at least much less than 1 kcal mol"1. The site of fumarate addition to alicyclic compounds should follow the HAT mechanism similar to w-alkanes, although a decrease in ring strain due to the loss of a hydrogen atom from the alicyclic ring will slightly lower the bond dissociation energies relative to the w-alkane analog. This decrease in bond dissociation energy is observed for cyclopentane which is 3.6 kcal mol"1 less than that of the C2 carbon in «-pentane. Alkylation of cyclopentane to form methyl- and ethylpentane produces a tertiary carbon in the ring at the site of alkyl attachment. As predicted by the HAT mechanism, the tertiary carbon is more stable as a free radical than the secondary ring carbons, displaying bond dissociation energies of 93.7 kcal mol"1. Thus, based upon the bond dissociation energies, and assumption of a HAT mechanism, it is predicted that similar carbons, i.e., secondary and tertiary carbons, will be more reactive in alicyclic alkanes as compared to the corresponding «-alkane. Thus, the most favorable site for fumarate addition to an alkylated alicyclic compound will be at the tertiary carbon followed by secondary carbons on the alicyclic ring. Based upon
292
the relatively higher bond dissociation energies, fumarate addition to the alkyl side chain of an alkylated alicyclic hydrocarbon is therefore unexpected. Table 1 Bond Dissociation Energies (AH298) at 298 K for various hydrocarbons in the reaction RH -> R* + H* Bolded hydrogen atom represents abstracted hydrogen. zl//2»«(kcal mol"1)
reference
104.99 +/- 0.03 101.1+/-0.4 98.6 +/- 0.4 98.2 +/- 0.5 96.5 +/- 0.4 100.2 99.2 99.0 98.0
[98] [99] [99] [99] [99] [100] [100] [100] [100]
95.6 +/- 1 93.7 93.7
[101] [102] [102]
112.9+/-0.5 89.8 +/- 0.6 85.4+/- 1.5 87.5
[103] [104] [105] [106]
86.7 83.5 98.7
[107] [107] [107]
112.2+/-1.3 111.9+/-1.4
[108] [108]
85.1 +/- 1.5 85.6
[105] [107]
Alkanes CH3-H (methane) CH3CH2-H (ethane) (CH3)2CH-H (propane) CH3CH2CH2CH3 (H-butane) (CH3)3C-H (wo-butane) fl-CjHu-H («-pentane) CH3CH2(CH2)2CH3 («-pentane) «-C6H]3-H («-hexane) CH3CH2(CH2)3CH3 («-hexane) Alicvclic Alkanes CP-H (cyclopentane) CPH(CH3) (methylcyclopentane) CPH(CH2CH3) (ethylcyclopentane) Alkyl Aromatics C6H5-H (benzene) C6H5CH2-H (toluene) C6H5CH2 CH3 (ethylbenzene) C6H5CH2 CH2CH3 (n-propylbenzene) Y-C6H5CH(CH3)2 (zso-propylbenzene - substituted) Y = 2,5 dimethyl Y = 4-;-butyl C6H5C(CH3)2CH2-H(?-butylbenzene) Naphthalene-H (Ci position) (C2 position) Naphthalene-CH2-H (CH3 at Ci position) (CH3 at C2 position)
293
For alkyl aromatic compounds, the benzylic hydrogen atoms, i.e., hydrogen atoms bonded to the carbon atom directly attached to the aromatic ring, require the least energy to abstract due to resonance stabilization created by delocalization of the free radical within the aromatic 7t-orbital system of the aromatic ring. Table 1 illustrates how resonance stabilization decreases the bond dissociation energy of the benzylic hydrogen to a range of 83.5 - 89.8 kcal mol"1 for a variety of alkyl aromatic compounds relative to the bond dissociation energy for hydrogen atoms bonded directly to the aromatic ring (113 kcal mol"1) or to other carbons present in the alkyl functional group (e.g., 98.7 kcal mol"1 for /-butylbenzene). Therefore, in light of a radical mechanism, it is predicted that fumarate will add to the benzylic carbon atom (as long as the benzylic carbon is not quaternary and a benzylic hydrogen is available for abstraction) regardless of the alkyl functional. Moreover, in consideration of the stability afforded through hyperconjugation in the alkyl group, the C-H bond dissociation energy will be lower for a tertiary benzylic carbon compared to a secondary benzylic carbon. Such stabilization is demonstrated in the 0.8 to 4 kcal mol"1 decrease in the bond dissociation energy of various substituted z-propylbenzene compounds relative to n-propylbenzene. The metabolites of the TEX hydrocarbons have been observed to contain the succinic acid functional group at the benzylic carbon; no addition of fumarate to the methyl group of ethylbenzene has been detected. In light of a radical mechanism, the lower reactivity of benzene relative to the TEX compounds is supported by the relative bond dissociation energies. In fact, the lack of detection of succinic acid benzene metabolites and the detection of phenol and benzoate as intermediates (above), suggests that alternative mechanisms exist for oxidizing benzene for less energy than the 112.9 kcal mol"1 required for hydrogen radical abstraction from the aromatic ring. Similarly for naphthalene and alkylnaphthalene, the bond dissociation energy for abstracting a hydrogen radical from an unsubstituted naphthalene is relatively high, 111.9 to 112.2 kcal mol"1, while the comparable reaction from the methyl group of methylnaphthalene is 85 kcal mol"1. These differences in bond dissociation energies may account for the fact that naphthyl-2-methyl-succinic acid has been detected in cultures and in the field (above) but the succinic acid metabolites of naphthalene have not. 4. GEOCHEMICAL INDICATORS OF METHANOGENIC OIL BIODEGRADATION As discussed above, there is ample evidence that anaerobic microbial processes occur under reservoir conditions. There is even evidence, albeit indirect, that such processes are occurring in situ. The most widely used indicator for
294
biological methanogenesis comes from the carbon isotopic abundance signature of the methane in natural gas deposits (e.g. Hunt [109]). Most methane is thought to arise from thermogenic decomposition of biomass, kerogen, and oil [110], but biological processes are also important. Bacteria prefer the lighter 12C isotope over 13C, thus microbially-produced methane is isotopically lighter (C13 of -110 to -60 %o) than thermally-produced gas (C13 of -60 to -15 %o). The microbial process has classically been thought to occur in the relatively shallow subsurface, and to be from relatively recently buried biomass rather than from material that has undergone burial and catagenesis to petroleum. However, several reservoirs have now been found to have methane with isotopic signatures suggestive of a biogenic origin [14, 111], and this is certainly consistent with microbial methanogenesis from petroleum at depth. Unfortunately, ready interpretation of isotopic enrichment, already complicated by the likely mixing of thermogenic, biogenic, and abiogenic [112] sources, is further confounded by the discovery of anaerobic methane oxidation [113], a microbial activity in which the lighter methane isotope is clearly preferred [114]. It is thus clear that supporting evidence is needed to confirm a microbial origin for methane in many cases. This evidence might come from the oil itself. The consideration above indicates that anaerobes prefer to transform some hydrocarbons relative to others. For example, an anaerobic microbial consortium was able to degrade dimethyl-cyclopentanes and cyclohexanes under sulfate-reducing but not under methanogenic conditions and the activity under the former conditions was limited to specific isomers [26]. Perhaps the results of such preferences can be identified in oils from candidate reservoirs? Alternatively it may be possible to detect by-products of anaerobic biodegradation in waters associated with oil reservoirs, or in the oil itself. The former is proving very useful in identifying anaerobic biodegradation in contaminated aquifers, where succinate derivatives of w-alkanes, cyclic alkanes, and alkylaromatic hydrocarbons as well as naphthoic acids have been detected [115-117]. Detecting these compounds in produced waters would be good evidence that anaerobic hydrocarbon biodegradation was proceeding underground. Are there compounds in the oil that may act of fingerprints of biodegradation? Crude oils often contain naphthenic acids, carboxylic acids with one or more saturated ring structures, and at least some are believed to be the results of partial biodegradation of oil components [17]. Electrospray ionization mass spectrometry is proving to be an excellent tool for determining the molecular identity of naphthenic acids [118-120], and as more potential biodegradation intermediates are identified it will be important to see whether such compounds are present in crude oils. Dicarboxylic acids, such as the succinate derivatives indicative of anaerobic hydrocarbon metabolites, have not
295
yet been identified in oils, but they may be so polar that they primarily partition to the aqueous phase.
5. PROSPECTS FOR HYDROCABON METHANOGENESIS With notable exceptions, it is becoming increasingly clear that fumarate addition reactions represent an important mechanism for the initial activation of structurally-diverse hydrocarbons by anaerobic microorganisms. Indeed, recent surveys for such anaerobic metabolites at hydrocarbon-impacted sites identified a variety of alkylbenzylsuccinates and alkylsuccinates in situ, as well as putative PAH metabolites such as naphthoic acids and tetrahydronaphthoic acids [43, 115-117, 121, 122]. Based on such observations, one can envision that the same type of biochemical reactions might occur in oil reservoirs. However, it has long been accepted that the microbial food web in oil fields is based on aerobic hydrocarbon-oxidizing bacteria [109, 123-125]. According to this "aerobic" model, low molecular weight polar compounds such as fatty acids, organic acids, and alcohols resulting from aerobic hydrocarbon decay serve as substrates for fermentative, acetogenic, and sulfate-reducing bacteria. Further metabolic transformations of these compounds produce H2 and acetate, that can then be used by methanogenic bacteria to produce methane. While this aerobicanaerobic successional model of oil decomposition in reservoirs dominated popular thinking for many years, a reevaluation is needed in light of new knowledge. Recent geochemical considerations and microbiological data strongly indicate that oil biodegradation in the deep terrestrial subsurface proceeds mainly through anaerobic metabolism [11, 12, 16]. Biodegraded oils in deep anoxic horizons are often accompanied by hydrocarbon gasses of biological origin [14, 126]. Accordingly, isotopically light methane with 813C from -45% to -59 % indicative of a biological origin and in situ rates of methane production in the range from 1.3 to 80 nmol liter "' day"1 were observed in oil fields under various environmental conditions [127-130]. In the latter studies, methane precursors were considered to be low molecular weight compounds that originated from aerobic oil decomposition and migrated to anoxic layers. Recent studies have now shown that petroleum hydrocarbon biodegradation can be directly coupled to methane production. For example, the production of methane from the decay of toluene, o-xylene, benzene, alkanes, and some alicyclic compounds has been documented [26, 34, 45, 61, 70, 71] . In incubations of gas condensate-contaminated sediments amended with artificially weathered oil, the entire «-alkane fraction (Q3-C34 range) was completely consumed under both sulfate-reducing and methanogenic conditions. In the sulfate-free incubations, «-alkane degradation was accompanied by methane accumulation [9]. In other studies, individual alkanes such as hexadecane and
296
pentadecane were converted to CH4 by enrichment cultures and in sediment incubations [7, 8]. Based on our current understanding of methanogens, the conversion of hexadecane to CH4 might require as many as three groups of microorganisms: acetogenic (or syntrophic) bacteria converting hexadecane to acetate and H2, and acetoclastic and hydrogenotrophic archaea producing CH4 from acetate or H2 and CO2, respectively. Molecular characterization of a hexadecane-degrading methanogenic community confirmed this possible composition. It revealed three clones closely related to syntrophic bacteria of the genus Syntrophus, one clone closely related to the genus Methanosaeta, an acetoclastic methanogen, and two clones related to Methanospirillum and Methanoculleus, which comprise hydrogenotrophic methanogens [7]. Similarly, Watanabe et al. [131] found a substantial diversity of methanogens in the groundwater under an oil storage cavern in Japan. Though the most often described alkane-degrading bacteria are the sulfate-reducing bacteria, they can conceivably participate in methane production from hydrocarbons even in the absence of sulfate. These bacteria are known to couple with methanogens to form syntrophic associations wherein electron transfer occurs between the bacteria. In effect, the methanogen serves as the electron acceptor for the sulfate reducers. Thus, phylogenetic analysis of two alkane-degrading sulfate-reducing bacteria revealed that they were closely related to Syntrophobacter (from 92 to 95% identity), a genus that is known to degrade fatty acids in syntrophic co-culture with methanogens [132]. In consistent fashion, a defined co-culture of one of these organisms cultivated with Methanospirillum hungateii in the absence of sulfate could produce methane from dodecane (unpublished results). It is therefore not unreasonable to presume that similar microbial associations can exist in petroliferous subsurface formations and catalyze hydrocarbon conversions to methane and CO2. Of course, the rate of bioconversion is an extremely important when considering the prospects for microbial enhanced energy recovery. As noted, some researchers believe that such reactions, while clearly possible, take geologic time due to the limited diffusion of nutrients. While this may be true along oil migration paths, evidence to the contrary in other locales suggests that the rates need not be slow. For instance, it has been demonstrated that subsurface bacteria from oil-bearing sediments could convert hexadecane to methane quite rapidly and without a lag. Thus 10% of added 14C-hexadecane was converted to I4CH4 in about 15 d [8]. The in situ rates of methanogenesis can also be quite high in deep high temperature oil reservoirs. The rates of methanogenesis measured in formation waters of the Jurassic horizon (2299 m deep; 84°C) exceeded 80 nmol of CH4 liter "' day"1. Hybridization of 16S rRNA obtained from formation water with group-specific phylogenetic probes revealed the presence of thermophilic methanogens and heterotrophs [130]. Laboratory incubations of formation waters and raw production fluids from two deep high-
297
temperature petroleum reservoirs in California demonstrated active methane production at in situ temperatures (70-83°C). Total community DNA analysis revealed archaeal phylotypes closely related to thermophilic methanogens and sulfidogenic archaea as well as bacterial thermophiles such as Thermatoga sp., Thermococcus sp., Thermoanaerobacter sp. and Desulfothiovibrio sp. [133]. These findings contrast with the belief of low metabolic activity in the deep hot subsurface and the cessation of oil biodegradation due to the paleosterilization of formations that have at some time experienced temperatures greater than 80°C [12, 134]. The bulk of the accumulated microbiological evidence suggests that oildegrading subsurface microbial communities can be quite metabolically versatile. However, it is unreasonable to presume that the same community structure exists in all subsurface locales. The environmental conditions during oil diagenesis may have effectively eliminated critical bacterial components of obligate consortia responsible for oil methanogenesis. Clearly, the presence of hydrocarbons in the terrestrial subsurface attests to the fact that such consortia are far from ubiquitous in distribution. 6. HYDROCARBON METHANOGENESIS AND IMPLICATIONS FOR ENERGY RECOVERY Although oil is the dominant source of energy on a global scale, conventional oil production technologies are only able to recover about onethird of oil in reservoirs [135]. As a result, large quantities of residual oil remain trapped in reservoir rock pores, mainly due to capillary or subterranean forces in the vicinity of a well bore [135, 136]. Thus, enhanced oil recovery (EOR) methods have been developed to help overcome these forces and make oil move (see chapter 15). These technologies may be based on thermal, chemical, gasmiscible, or microbial technologies. It is estimated that EOR strategies can potentially add up to 60 billion barrels of oil in the near term though the increased use of existing domestic fields [137]. Understanding the multiphase flow properties of subsurface reservoir rocks and the forces that entrap oil is key for successful EOR and will help determine which technique may apply best for a given reservoir. The processes involved are complex and have been reviewed [135]. It has long been recognized that gasses dissolved in oil lower its viscosity and cause swelling. This is a major driving force for oil mobilization. In fact, gas-based EOR processes have been touted as the current, most profitable technology for recovering the large amounts of remaining oil in mature fields [135, 136]. Carbon dioxide has long been used effectively to drive enhanced oil recovery, and represents about 25% of EOR operations in the U.S. [6, 136, 138]. A secondary outcome in the use of CO2 to recover oil has far-reaching
298
environmental implications too; CO2 can be stored in reservoirs to help in the reduction of greenhouse gas emissions [6, 136, 138]. Since the combustion of fossil fuels is the largest contributor of greenhouse gas emissions, the recycling (capturing and subsequent sequestering) of anthropogenic CO2 into spent or even active reservoirs offers a promising way to both decrease the potential for global warming and increase oil recovery and profits. It has been estimated that fossil fuel reservoirs can store up to 900 billion metric tonnes of CO2 worldwide [138]. As outlined in the Introduction, natural gas is abundant worldwide, but like oil, natural gas fields can only be harvested to residual amounts or pressures making further gas unrecoverable. Carbon dioxide can also serve as an EOR gas for natural gas recovery by way of re-pressurization of reservoirs [139]. The use of CO2 as a cushion gas for natural gas storage is also being considered [140]. Of course, CO2 sequestration into natural gas fields for either recovery or as a cushion gas also offers the environmentally-friendly advantage of reducing greenhouse gas emissions [138] Although gas-based EOR with CO2 is best-understood and most widely used, the viscosity lowering of a crude by other gasses including nitrogen, flue gas, and dissolved methane and their relevance for EOR has also been considered [136, 141-143]. Indeed methane gas associated with oil can potentially help reduce its viscosity and thus enhance its recovery [141, 142]. In previous sections, we have discussed the prospect that methane gas found associated with oil reservoirs can be present as a by-product of anaerobic, microbial consumption of oil produced over millennia. In fact, there is evidence suggesting that many "dry gas" fields have arisen due to the microbial degradation of oil [13, 14]. In fields characterized by light hydrocarbons, C2 to C5 alkanes are presumably biodegraded to methane, helping to re-establish a "gas cap" [12, 15]. In theory, such "biogenic gas" could feasibly reduce oil viscosity to the point where it can be more easily recovered. In practice, gas pressure accumulations over geological time-scales have no doubt aided in conventional oil recovery but of course it remains unclear whether these gasses were thermally- or biologically-produced. Given the success of gas-based energy recovery, and the recent discovery that microorganisms can convert hydrocarbons into methane gas at substantial rates (i.e. faster than geological time scales), one could envision combining the principles of microbial- and gas-based-EOR to help recover residual oil in mature fields. Although not yet widely used in the oil industry, advances in microbial-EOR technologies have proved promising to recover residual oil (see chapter 15)[135]. Although too numerous to describe here, some MEOR technologies have explored the use of bacterial inoculation into wells to produce gaseous by-products which can help mobilize trapped oil [135]. By analogy, spent reservoirs might be inoculated with the appropriate microbial communities
299
to produce methane gas that could help decrease the viscosity of oil and aid in further recovery. What if such an inoculation procedure resulted in at least some fraction of the available energy being recovered as usable methane gas? Such speculative technology is quite far from being addressed or realized, especially from an economic point of view, but initial laboratory experimentation on this topic has been promising (Fig. 1). Samples (10 g) taken from a field in Nowata, OK that had undergone secondary oil recovery procedures (water flooding) were used to test the importance of a methane-producing oil-degrading inoculum enriched from a gas-condensate contaminated aquifer [9]. When residual oil core samples were ground or broken into small portions, the oil-degrading inoculum was effective in stimulating methanogenesis relative to a variety of controls. The latter included a heat-inactivated preparation, an oil-unamended control, and production water from the same field that received the inoculum ( Fig. 1). Interestingly, the rate of methanogenesis was much greater with the residual oil core samples than that observed when a standard oil or even when the formation (Nowata) crude alone served as a substrate for the inoculum. While the reasons for this result are under investigation, it is clear that such inocula may play a potential role for the enhanced recovery of methane from oil trapped in mature reservoirs.
Fig. 1. Methane production from residual oil in core samples inoculated with a methanogenic bacterial enrichment capable of anaerobic hydrocarbon metabolism. Symbols: Oil unamended control (•); Nowata crude oil (•); Production water (X); An artificially weathered Alaska north slope oil standard (A); Crushed core (o); Pebbled core (•). Heat inactivated and uninoculated controls are not depicted, but were uniformly negative.
300
7. MICROBIAL ENHANCED ENERGY RECOVERY AND CARBON DIOXIDE Figure 1 clearly indicates that a hydrocarbon-degrading methanogenic bacterial inoculum can attack oil deposited in rocks and covert it to natural gas. In fact, this metabolism is much faster than comparable incubations amended with an equivalent amount of oil from the same formation (estimated amount of oil in core was 0.0 lg oil/g rock based on 30-40% residual saturation). These observations lead to numerous questions that center on the rate and efficiency of oil bioconversion, the role of inocula in the process, the nutritional environment presented by petroliferous formations, the diversity of hydrocarbons susceptible to microbial attack, the biotechnological control of such bioconverstions and many other fundamental and practical considerations. Careful exploration of these issues in the future will help define the utility of enhanced energy recovery efforts at a time when the need for such considerations is particularly acute. Tomes have been written on the eventual transitioning of global energy use patterns and their potential impact on the environment. Yet, it seems clear that any energy form will have an impact on the environment and that fossil fuel use will remain the predominant energy form for decades to come. Global climate change concerns are forcing worldwide reductions in atmospheric CO2 emissions. Since methane consumption produces a fraction of the CO2 per BTU generated relative other fossil fuels, a greater reliance on methane will help reduce the rate of increase in global carbon dioxide emissions. The biotechnological link between the consumption of hydrocarbons for the production of methane may be a way of enhancing the recovery of energy in an environmentally responsible fashion, mostly from mature domestic reserves that are otherwise unprofitable or too technically difficult to exploit. It is our hope that this article helps spur such considerations. REFERENCES [1]
[2] [3] [4] [5] [6]
National Research Council (NRC). (2003). Committee on Oil in the Sea: Inputs, Fates, and Effects, Ocean Studies Board and Marine Board, Divisions of Earth and Life Studies and Transportation Research Board, NRC. In: Oil in the sea III: Inputs, fates, and effects. The National Academies Press, Washington, D.C. Energy Information Administration (EIA). (2004). www.eia.doe.gov/oiaf/ieo/index.html. C. Hall, P. Tharakan, J. Hallock, C. Cleveland, and M. Jefferson. Nature 426 (2003) 318. J.B. Curtis and S.L. Montgomery. AAPG Bull. 86 (2002) 1671. S.M. Al-Fattah and R.A. Startzman. SPE Journal, May (2000) 62-72. S. Bachu and S. Stewart. J. Can. Petrol. Technol. 41 (2002) 32.
301
[7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]
[22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]
Zengler, K., H.H. Richnow, R. Rosello-Mora, W. Michaelis, and F. Widdel. Nature 401 (1999)266. Anderson, R.T. and D.R. Lovley. . Nature 404 (2000) 722. Townsend, G.T., R.C. Prince, and J. M. Suflita. . Environ. Sci. Technol. 37 (2003) 5213. Prince, R. C. (2002). Petroleum and other hydrocarbons, biodegradation of. hi Encyclopedia of Environmental Microbiology; Bitton, G. Ed.; John Wiley, New York, pp. 2402-2416. Roling, W.F.M., I.M. Head, and S.R. Larter. Res. Microbiol. 154 (2003) 321. Head, I. M., D.M. Jones, and S.R. Larter. Nature 246 (2003) 344. Sakata, S., Y. Sano, Maekawa, T., and Igari, S.-I. Org. Geochem. 26 (1997) 399. Pallasser R.J. Org. Geochem. 31 (2000) 1363. Wenger, L.M., C.L. Davis, G.H. Isaksen. (2001). Multiple controls on petroleum biodegradation and impact on oil quality. SPE paper 71450. Larter, S., A. Wilhelms, I. Head, M. Koopmans, A. Aplin, R. DiPrimo, C. Zwach, M. Erdmann, N. Telnaes. Org. Geochem. 34 (2003) 601. Tissot, B.P. and D.H. Welte. 1984. Petroleum Formation and Occurrence. SpringerVerlag, Berlin. Prince, R. C , R.M. Garrett, R.E. Bare, M.J. Grossman, G.T. Townsend, J.M. Suflita, K. Lee, E.H. Owens, G.A. Sergy, J.F. Braddock, J.E. Lindstrom, and R.R. Lessard. Spill Sci. Technol. Bull. 8 (2003) 145. Britton, L. 1984. Microbial degradation of aliphatic hydrocarbons, hi D. T. Gibson (Ed.), Microbial degradation of organic compounds, Marcel Dekker, Inc., New York, pp. 89-129. Gibson, D.T. and V. Subramanian. 1984. Microbial degradation of aromatic hydrocarbons, hi T.D. Gibson (ed.), Microbial degradation of organic compounds, Marcel Dekker, Inc., New York, pp.181-252. Krumholz, L.R., M.E. Caldwell, and J. M. Suflita. (1996). Biodegradation of "BTEX" hydrocarbons under anaerobic conditions, hi Bioremediation Principles and Applications. (R.L. Crawford and D. L. Crawford, Eds.) Cambridge University Press, Great Britain, pp. 61-99. Coates, J.D., R.T. Anderson, and D.R. Lovley. Appl. Environ. Microbiol. 62 (1996) 1099. Widdel, F., and R. Rabus. Curr. Opin. Biotechnol. 12 (2001) 259. Spormann, A.M. and F. Widdel. Biodegradation 11 (2000) 85. Rios-Hernandez, L.A., L.M. Gieg, and J.M. Suflita. Appl. Environ. Microbiol. 69 (2003) 434. Townsend, G.T., R.C. Prince, and J.M. Suflita. FEMS Microbiol. Ecol. (2004). (in press). Biegert, T., G. Fuchs, and J. Heider. Eur. J. Biochem. 238 (1996) 661. Beller, H.R. and A.M. Spormann. J. Bacteriol. 179 (1997) 670-676. Leuthner, B., C. Leutwein, H. Schulz, P. Horth, W. Haehnel, E. Schiltz, H. Schagger, and J. Heider. Mol. Microbiol. 28 (1998) 615. Krieger C.J., W. Roseboom, S.P.J. Albracht, and A.M. Spormann. J. Biol. Chem. 276 (2001) 12924. Rabus, R. and J. Heider. Arch. Microbiol. 170 (1998) 377. Beller, H.R. and A.M. Spormann. Appl. Environ. Microbiol. 63 (1997) 3729. Kane, S.R., H.R. Beller, T.C. Legler, and R.T. Anderson. Biodegradation 13(2002) 149.
302
[34] Beller, H.R. and E.A. Edwards. Appl. Environ. Microbiol. 66 (2000) 5503. [35] Zengler, K., J. Heider, R. Rosello-Mora, and F. Widdel. Arch. Microbiol. 172 (1999) 204. [36] Heider, J., A.M. Spormann, H.R. Beller, and F. Widdel. FEMS Microbiol.Rev. 22 (1999) 459. [37] Leutwein C. and J. Heider. Arch. Microbiol. 178 (2002) 517. [38] Leutwein, C. and J. Heider. J. Bacteriol. 183 (2001) 4288. [39] Harwood, C.S., G. Burchhardt, H. Herrmann, and G. Fuchs. FEMS Microbiol.Rev. 22 (1999)439. [40] Chee-Sanford, J.C., J.W. Frost, M.R. Fries, J. Zhou, and J. Tiedje. Appl. Environ. Microbiol. 62 (1996) 964. [41] Edwards, E.A., L.E. Wills, M. Reinhard, and D. Grbic-Galic. Appl. Environ. Microbiol. 58(1992)794. [42] Rueter, P., R. Rabus, H. Wilkes, F. Aeckersberg, F. A. Rainey, H.W. Jannasch, and F. Widdel. Nature 372(1994)455. [43] Elshahed, M.S., L.M. Gieg, MJ. Mclnerney, and J.M. Suflita. Environ. Sci. Technol. 35(2001)682. [44] Rabus, R., H. Wilkes, A. Schramm, G. Harms, A. Behrends, R. Amann, and F. Widdel. Environ. Microbiol. 1 (1999) 145. [45] Edwards, E.A. and D. Grbic-Galic. Appl. Environ. Microbiol. 60 (1994) 313. [46] Haner, A., P. Hohener, and J. Zeyer. Appl. Environ. Microbiol. 61 (1995) 3185. [47] Dolfing J., J. Zeyer, P. Binder-Eicher, and R.P. Schwarzenbach. Arch. Microbiol. 154 (1990)336. [48] Fries, M.R., J. Zhou, J. Chee-Sanford, and J.M. Tiedje. Appl. Environ. Microbiol. 60 (1994)2802. [49] Rabus, R. and F. Widdel. Arch. Microbiol. 163 (1995) 96. [50] Hess, A., B. Zarda, D. Hahn, A. Haner, D. Stax,, P. Hohener, and J. Zeyer. Appl. Environ. Microbiol. 65 (1997) 2136. [51] Harms, G., K. Zengler, R. Rabus, F. Aeckersberg, D. Minz, R. Rossello-Mora, and F. Widdel. Appl. Environ. Microbiol. 65 (1999) 999. [52] Krieger, C.J., H.R. Beller, M. Reinhard, and A.M. Spormann. J. Bacteriol. 181 (1999) 6403. [53] Achong, G.R., A.M. Rodriguez, and A.M. Spormann. J. Bacteriol. 183 (2001) 6763. [54] Beller, H.R. and A.M. Spormann. (1999). FEMS Microbiol. Lett. 178 147. [55] Beller, H.R., A.M. Spormann, P.K. Sharma, J.R. Cole, and M.Reinhard. Appl. Environ. Microbiol. 62 (1996) 1188. [56] Ball, H.A., H.A. Johnson, M. Reinhard, and A.M. Spormann. J. Bacteriol. 178 (1996) 5755. [57] Johnson, H.A., D.A. Pelletier, and A.M. Spormann. J. Bacteriol. 183 (2001) 4536. [58] Kniemeyer, O., T. Fischer, H.Wilkes, F.O. Glockner, and F. Widdel. Appl. Environ. Microbiol. 69 (2003) 760. [59] Burland, S.M. and E.A. Edwards. (1999). Appl. Environ. Microbiol. 65 529. [60] Coates, J.D., R. Chakraborty, J.G. Lack, S.M. O'Connor, K.A. Cole, K.S. Bender, and L.A. Achenbach. Nature 411 (2001) 1039. [61] Ulrich, A C. and E.A. Edwards. Environ. Microbiol. 5 (2003) 92. [62] Edwards, E.A. and D. Grbic-Galic. Appl. Environ. Microbiol. 58 (1992) 2663. [63] Lovley, D.R., J.D. Coates, J.C. Woodward, and E.J.P. Phillips. Appl. Environ. Microbiol. 61 (1995)953.
303
[64] Coates, J.D., R.T. Anderson, J.C. Woodward, E.J.P. Phillips, and D.R. Lovley. Environ. Sci. Technol. 30 (1996) 2784. [65] Weiner, J. and D.R. Lovley. Appl. Environ. Microbiol. 64 (1998) 775. [66] Lovley, D.R., J.C. Woodward, and F.H. Chapelle. Nature 370 (1994) 128. [67] Anderson, R.T., J.N. Rooney-Varga, C.V. Gaw, and D.R. Lovley. Environ. Sci. Technol. 32(1998)1222. [68] Caldwell, M.E., R.S. Tanner, and J.M. Suflita. Anaerobe 5 (1999) 595. [69] Grbic-Galic, D. and T. Vogel. Appl. Environ. Microbiol. 53 (1987) 254. [70] Kazumi, J., M.E. Caldwell, J.M. Suflita, D.R. Lovley, and L.Y. Young. Environ. Sci. Technol. 31(1997)813. [71] Weiner, J. and D.R. Lovley. Appl. Environ. Microbiol. 64 (1998) 1937. [72] Coates, J.D., R. Chakraborty, and M.J. Mclnerney. Res. Microbiol. 153 (2002) 621. [73] Caldwell, M.E. and J.M. Suflita. Environ. Sci. Technol. 34 (2000) 1216. [74] Phelps, CD., X. Zhang, and L.Y. Young. Environ. Microbiol. 3 (2001) 600. [75] Aeckersberg, F., F. Bak, and F. Widdel. Arch. Microbiol. 156 (1991) 5-14. [76] Aeckersberg, F., F. Rainey, and F. Widdel. Arch. Microbiol. 170 (1998) 361. [77] So, CM. and L.Y. Young. Appl. Environ. Microbiol. 65 (1999) 2969. [78] Ehrenreich, P., A. Behrends, J. Harder, and F. Widdel. Arch. Microbiol. 173 (2000) 58. [79] Kropp, K.G., LA. Davidova, and J.M. Suflita. Appl. Environ Microbiol. 66 (2000) 5393. [80] Caldwell, M.E., R.M. Garrett, R.C. Prince, and J.M. Suflita. Environ. Sci. Technol. 32 (1998)2191. [81] Rabus, R., H. Wilkes, A. Behrends, A. Armstroff, T. Fischer, and F. Widdel. J. Bacteriol. 183(2001)1707. [82] Davidova, I., L. Gieg, K. Kropp, M. Nanny, J. Suflita. (2004) (submitted for publication) [83] So, CM., CD. Phelps, and L.Y. Young. Appl. Environ. Microbiol. 69 (2003) 3892. [84] Wilkes, H., S. Ktihner, C. Bolm, T. Fischer, A. Classen, F. Widdel, and R. Rabus. Org. Geochem. 34(2003)1313. [85] Callaghan, A.V., L.M. Gieg, K.G. Kropp, J.M. Suflita, and L.Y. Young. (2003). Fumarate addition during hexadecane degradation by the sulfate-reducer AK-01. American Society for Microbiology 103-rd General Meeting, Washington, D.C, Abstract Q-038, p. 521. [86] Wilkes, H., R. Rabus, T. Fischer, A. Armstroff, A. Behrends, and F. Widdel. Arch. Microbiol. 177(2002)235. [87] Galushko, A., D. Minz, B. Schink, and F. Widdel. Environ. Microbiol. 1 (1999) 415. [88] Rockne, K.J., J.C. Chee-Sanford, R.A. Sanford, B.P. Hedlund, J.T. Staley, and S.E. Strand. Appl. Environ. Microbiol. 66 (2000) 1595. [89] Ramsay, J.A., H. Li, R.S. Brown, and B. Ramsay. Biodegradation 14 (2003) 321. [90] Rothermich, M.M., L.A. Hayes, and D.R. Lovley. Environ. Sci. Technol. 36 (2002) 4811. [91] Zhang, X. and L.Y. Young. Appl. Environ. Microbiol. 63 (1997) 4759. [92] Annweiler, E., W. Michaelis, and R.U. Meckenstock. Appl. Environ. Microbiol. 68 (2002) 852. [93] Sullivan, E.R., X. Zhang, C. Phelps, and L.Y. Young. Appl. Environ. Microbiol. 67 (2001)4353. [94] Annweiler, E., A. Materna, M. Safinowski, A. Kappler, H.H. Richnow, W. Michaelis, and R.U. Meckenstock. Appl. Environ. Microbiol. 66 (2000) 5329. [95] Zhang X, Sullivan E. R, and L.Y Young. Biodegradation. 11 (2000) 117.
304
[96] Baciocchi, E., F. D'Acunzo, C. Galli, and O. Lanzalunga. J. Chem. Soc. Perkin Trans. 2.2(1996)133. [97] March, J. 1968. Advanced Organic Chemistry Reactions, Mechanisms, and Structure. McGraw-Hill Book Company, New York, New York. [98] Ruscic, B., M. Litorja, and R. Asher. J. Phys. Chem. A. 103 (1999) 8625. [99] Saekins, P.W., M.J. Pilling, J.T. Niiranen, D. Gutman, and L.N. Krasnoperov. J. Phys. Chem. 96(1992)9847. [100] Pedley, J.B, R.D. Naylor and S.P. Kirby. 1986. Thermodynamic Data of Organic Compounds. 2nd ed., Chapman and Hall, New York [101] Castelhano, A.L. and D. Griller. J. Amer. Chem. Soc. 104 (1982) 3655. [102] Tumanov, V.E. and E.T. Denisov. Neftekhimiya. 41 (2001) 109. [103] Wenthold, P.G., and R.R. Squires. J. Am. Chem. Soc. 116 (1994) 6401. [104] Ellison, G.B., G.E. Davico, V.M. Bierbaum, and C.H. DePuy. Int. J. Mass Spectrum. Ion Processes. 156 (1996) 109. [105] McMillen, D.F. and D. M. Golden. Ann. Rev. Chem. 33 (1982) 493. [106] Denisov, E.T. and T.G. Denosova. (2000). Handbook of Antioxidants. CRC Press, New York. [107] Kromkin, E.A., V.E. Tumanov, and E.T. Denisov. Neftekhimiya. 42 (2002) 3. [108] Reed, D.R. and S.R. Kass. J. Mass Spectrom. 35 (2000) 534. [109] Hunt, J.M. 1979. Petroleum Geochemistry and Geology, 2nd Ed. W.H. Freeman and Company, New York, pp. 413. [110] Domine, F., R. Bounaceur, G. Scacchi, P.-M. Marquaire, D. Dessort, B. Pradier, and O. Brevart. Org. Geochem. 33 (2002) 1487. [ I l l ] Sassen, R., A.V. Milkov, E. Ozgul, H.H. Roberts, J.L. Hunt, M.A. Beeunas, J.P. Chanton, D.A. DeFreitas, and S.T. Sweet. Org. Geochem. 34 (2003) 1455. [112] Sherwood Lollar, B., T.D.Westgate, J.A. Ward, G.F. Slater, and G. LacrampeCouloume. Nature 416 (2002) 522. [113] Nauhaus, K., A. Boetius, M. Kriiger, and F. Widdel. Environ. Microbiol. 4 (2002) 296. [114] Orphan, V.J., C. H. House, K.U. Hinrichs, K. D. McKeegan and E. F. DeLong. Science 293 (2001) 484. [115] Beller, H.R. Biodegradation. 11 (2000) 125. [116] Gieg, L.M. and J.M. Suflita. Environ. Sci. Technol. 36 (2002) 3755. [117] Griebler, C, M. Safinowski, A. Vieth, H.H. Richnow, and R.U. Meckenstock. Environ. Sci. Technol. 38 (2004) 617. [118] Barrow, M.P., L.A. McDonnell, X. Feng, J. Walker, and P.J. Derrick. Anal. Chem. 75 (2003) 860. [119] Gabryelski, W. and K.L. Froese. Anal. Chem. 75 (2003) 4612. [120] Lo, C.C., B.G. Brownlee, andNJ. Bunce. Anal. Chem. 75 (2003) 6394. [121] Phelps, C D , J. Battistelli, and L.Y. Young. Environ. Microbiol. 4 (2002) 532. [122] Martus, P. and W. Puttman. Sci. Total Environ. 307 (2003) 19. [123] Belyaev, S.S., K. Laurinavichus, A.Y. Obraztsova, S.N. Gorlatov, and M.V. Ivanov. Microbiologiya. 51 (1982) 997. [124] Nazina, T.N, E.P. Rozanova, and S.I. Kuznetsov. Geomicrobiol. J. 4 (1985) 103. [125] Palmer, S.E. (1993). Effect of biodegradation and water washing on crude oil composition. In Engel M.H, Macko S.A. (Eds.), Organic Geochemistry. Plenum Press, New York, pp. 511-533. [126] James, A.T. and B.J. Burns. Bull. Am. Assoc. Petrol. Geol. 68 (1984) 957.
305
[127] Ivanov, M.V., S.S. Belyaev, A.M. Zyakun, V.A. Bondar, and K.K. Laurinavichus. Geokhimiya. 11 (1983) 1647. [128] Borzenkov LA., S.S. Belyaev, Y.M. Miller, LA. Davydova, and M.V. Ivanov. Microbiology 66 (1997) 104. [129] Nazina, T,N., A.E. Ivanova, LA. Borzenkov, S.S. Belyaev, and M.V. Ivanov. Geomicrobiol. J. 13 (1995) 181. [130] Bonch-Osmolovskaya, E.A., M.L. Miroshnichenko, A.V. Lebedinsky, N.A. Chernyh, T.N. Nazina, V.S. Ivoilov, S.S. Belyaev, E.S. Boulygina, Y.P. Lysov, A.N. Perov, A.D. Mirzabekov, H. Hippe, E. Stackebrandt, S. L'Haridon, and C. Jeanthon. Appl. Environ. Microbiol. 69(2003)6143. [131] Watanabe, K., Y. Kodama, N. Hamamura, and N. Kaku. Appl. Environ. Microbiol. 68 (2002) 3899. [132] Davidova, I.A., K.G. Kiopp, K.E. Duncan, and J.M. Suflita. (2002). Anaerobic biodegradation of n-alkanes by sulfate-reducing bacterial cultures. Abstracts of the International Symposium on Subsurface Microbiology. Copenhagen, Denmark, September 8-13. [133] Orphan,, V.J., S.K. Goffredi, E.F. Delong, and J.R. Boles. Geomicrobiol. J. 20 (2003) 295. [134] Connan, J. 1984. Biodegradation of crude oils in reservoirs. In J. Brooks, and D.H. Welte (Eds.), Advances in Petroleum Geochemistry. Academic Press, London, pp. 89129. [135] Mclnerney, M.J. and D.W.S. Westlake. 1990. Microbial enhanced oil recovery. In: Microbial Mineral Recovery, H.H.L. Ehrlich and C.L. Brierley (Eds.), McGraw-Hill, NY, pp. 409-445. [136] Rao, D. J. Can. Petrol. Technol. 40 (2001) 11. [137] Aycaguer, A.-C, M. Lev-On, and A.M. Winer. Energy & Fuels 15 (2001) 303. [138] Oldenburg, CM., K. Pruess, and S.M. Benson. Energy & Fuels 15 (2001) 293. [139] Oldenburg, CM. Energy & Fuels 17 (2003) 240. [140] Killesreiter, H. Erdoel und Kohle, Erdgas, Petrochemie 38 (1985) 405. [141] Frauenfeld, T.W.J., R.K. Ridley, R.K., and D.M. Nguyen. J. Petrol. Technol. 40 (1988)333. [142] Mayne, CJ. and R.W. Pendleton. Soc. Petrol. Eng. AIME 1 (1986) 131. [143] Alvarez, M.R, M.F. Hilton, and H.L. Oil & Gas J. 82 (1984) 95.
This page is intentionally left blank
Studies in Surface Science and Catalysis 151 R. Vazquez-Duhalt and R. Quintero-Ramirez (Editors) © 2004 Elsevier B .V. All rights reserved.
307
Chapter 11
Using nitrate to control microbially-produced hydrogen sulfide in oil field waters R.E. Eckford and P.M. Fedorak Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2E9
1. INTRODUCTION The presence of hydrogen sulfide (H2S) in oil fields can be the result of abiotic or biotic processes. In the later case, sulfate-reducing bacteria (SRB) are the culprits that produce this nocuous gas, leading to "souring" that is defined as the process whereby petroleum reservoirs experience an increase in the production of H2S during the economic production life of the field [1]. The increase in H2S content leads to a decrease in the economic value of the gas and oil, as well as operational problems associated with the H2S. This microbial process in wastewaters and oil field waters can be controlled by another group of microbes, known as nitrate-reducing bacteria (NRB). Their metabolic activities stop sulfate reduction by SRB, and in many cases the NRB can actually consume sulfide, thus decreasing H2S concentration in the waters. Jenneman et al. [2] have referred to these sulfide-consuming bacteria as "sulfide bioscavengers". Hitzman and Sperl [3] used the term "biocompetitive exclusion" to describe the microbial process in which NRB use volatile fatty acids and out-complete SRB to prevent or decrease sulfide production, and enhance oil recovery. This chapter will review (a) H2S in the petroleum industry, (b) the metabolism of SRB leading to sulfide production, (c) the occurrence, types and activities of NRB that might be found in oil field waters, (d) some laboratory studies that have elucidated the mechanisms by which NRB control sulfide produced by SRB, (e) some oil field experiences with nitrate injection to control sulfide in wastewaters, surface waters and oil field waters, and (f) some of the U.S. patents that apply to this microbial process.
308
Although nitrite, rather than nitrate, addition has been studied, this chapter focuses solely on the use of nitrate to control sulfide in oil field waters. This is a proven biotechnology that is under-utilized by the petroleum industry. 2. H2S AND THE PETROLEUM INDUSTRY 2.1. Formation of H2S Kerogen is the organic source material from which petroleum is formed and released [4-5]. The formation of petroleum occurs in the deeper subsurfaces as burial continues and temperature and pressure increase [5]. First oil, then gas is expelled from kerogen as the maturation process continues. Significant oil generation occurs between 60° and 120°C, and significant gas generation occurs between 120° and 2v25°C [5]. During the maturation process, H2S is also released. Machel [6] wrote, "The association of dissolved sulfate and hydrocarbons are thermodynamically unstable in virtually all diagenetic environments. Hence, redox-reactions occur, whereby sulfate is reduced by hydrocarbons either bacterially (bacterial sulfate reduction) or inorganically (thermochemical sulfate reduction)." Temperature is the major factor determining which process occurs. The microbiological process is common at temperatures for 0 to 60 or 80°C, whereas, the thermochemical process occurs at temperatures greater that 100° to 140°C [6]. Because temperature increases with burial depth, H2S found at shallow depths is usually the result of bacterial sulfate reduction whereas, H2S found at greater depths is the result of thermochemical sulfate reduction [7]. However, there are shallow pools that contain higher than expected concentrations of thermochemically generated sulfide [8]. These are believed to be the result of thermochemical sulfate reduction occurring downdip and migrating upward to a shallow reservoir [8]. At the time of discovery, the H2S concentration in an oil field depends upon its maturation history and/or the migration of H2S into the oil field. However, during oil recovery from some oil fields, an increase in H2S concentration (souring) can occur as a result of pressurizing the formation by injecting water into the reservoir. This process, know as waterfiooding, is discussed in section 3. Three well-documented examples of oil field souring are given in the following paragraphs. Cochrane et al. [9] describe the souring of the Ninian field in the North Sea. This field was discovered in 1974, and after several years of operation, injection of sea water was used to maintain the production rate. This was followed by an increase in sulfide production attributed to bacterial sulfate reduction. The reservoir temperature was initially between 100° to 120°C, but in the areas adjacent to the injection well bores, the temperature was cooled to as low as 40°C, which was conducive to bacterial sulfate reduction.
309
Frazer and Boiling [10] described the souring of the Kuparuk River field on the North Slope of Alaska. The field was initially sweet, but after injection of Beaufort Sea water, detectable levels of H2S began to appear at the producing wells. The connate water contained essentially no sulfate. However, the sulfate in the sea water stimulated bacterial sulfate reduction in the reservoir that had a temperature of about 70°C. The Skjold oil field in the North Sea soured upon the onset of waterflooding [11]. Oil and gas production began from this field in 1982 and sea water injection began in April 1985. In September 1985, the first recorded H2S production was measured to be 1.8 ppm in the gas phase. In 2002, the concentrations varied from 10 to 1000 ppm [11]. In late 1999, this field produced 1150 kg H2S d"1. These examples clearly demonstrate that waterflooding can stimulate bacterial sulfate reduction, leading to souring. Although these examples refer to offshore oil fields, souring also occurs in land-based oil fields using waterflooding [12-15]. As a result of the bacterial production of toxic H2S, the value of the oil decreases as the oil field sours. 2.2. H2S toxicity and properties H2S is a very dangerous gas, even though it occurs in nature. Its characteristic rotten egg smell is generally obvious at 0.13 ppm by volume and quite noticeable at 4.6 ppm [16]. Unfortunately the smell sense becomes quickly fatigued and can fail to warn of higher concentrations. Collapse, coma and death from respiratory failure may occur within a few seconds after one or two inspirations of the undiluted H2S [17]. The U.S. Occupational Safety and Health Administration has established the acceptable ceiling concentration of 20 ppm (by volume) for H2S with an acceptable maximum peak above the acceptable ceiling concentration of 50 ppm for an 8-h shift [16]. The specific gravity of H2S is 1.19; therefore it will collect in low places and accumulate under poorly ventilated conditions [18]. H2S is soluble in water and oil. It is a weak acid existing in aqueous solutions as H2S, HS~, or S~ (pKa values of 7.04 and 11.96). Aqueous solutions of H2S absorb O2 leading to the formation of elemental sulfur [17]. 2.3. Detrimental effects of H2S Besides its toxicity, H2S is a nuisance in the petroleum industry because it contaminates gas and stored oil, it corrodes iron in the absence of air (anaerobic corrosion), and it precipitates as amorphous ferrous sulfide (FeS), plugging and diminishing the injectivity of water injection wells [18]. In addition, fluids with water and H2S, may cause sulfide stress cracking of susceptible metals. This is affected by metal composition, pH, H2S concentration, total pressure, total tensile stress, temperature and time [19].
310
Two types of cracking known to occur in wet H2S environments are sulfide stress corrosion cracking and hydrogen-induced cracking (see chapters 7 and 8). The former occurs in steels of relatively high strength and in welds of welded steel structures. A crack propagates under working stress or residual stress vertically to the stress axis [20]. This type of corrosion is most damaging to drillpipe and well production facilities [21]. Hydrogen-induced cracking occurs parallel to the surface when no external stress is applied. It is also known as hydrogen blistering because of the blisters that appear on the surface of the metal [20]. General corrosion attack by H2S is influenced by the presence of CO2, O2 and brine, [18, 21]. It is related to the alloy composition and strength of steel [21]. H2S forms FeS scale, which is cathodic to the metal, promoting localized attack under the scale, as well as the penetration of H2 into the metal [21-22]. Figure 1 shows the process whereby an anode and cathode pair are generated by the action of SRB acting on sulfates in the presence of iron. The cathode is depolarized as the SRB consume H2. At the anode, iron (Fe) is oxidized to Fe2+ which combines with H2S produced by the SRB, giving FeS. This process results in a loss of structural material. Heterotrophic SRB also play a role in the deposition of FeS (Fig. 1).
Fig. 1. Iron metal corrosion mediated by SRB in a biofilm. The process is caused by the consumption of H2 causing cathodic depolarization. Adapted from Ref. [18].
311
Removal of dissolved gases (O2, H2S and CO2) from drilling and produced fluids is necessary to minimize corrosion damage. H2S in oil base drilling fluid is removed by gas separators and vacuum degassers, and then neutralized. Controlling corrosion in H2S-containing environments requires proper selection of materials, including the use of low-hardness steels, application of inhibitors and complete exclusion and removal of O2 from water used in petroleum production [21]. Clearly, the presence of H2S greatly increases the cost of exploration for oil and natural gas, and the cost of production and storage of petroleum. Plugging (or biofouling) of injection wells is also caused by SRB. The sulfide they produce, precipitates soluble iron in the injection or formation water forming colloidal FeS [23]. This colloidal material becomes associated with bacterial cells and oil, forming a gummy mass that can clog reservoirs and plug injection wells. The activities of SRB can also produce calcite (CaCO3) that can add to the plugging problem. 3. OIL RECOVERY AND WATERFLOODING Under primary oil recovery, typically less than 30% of the original oil is produced, so that improved or enhanced methods are used to recover some of the remaining oil [24]. These processes, known as secondary and tertiary recovery methods, include the addition of energy into the reservoir and are accomplished by injecting some type of fluid through injection wells. This is referred to as enhanced oil recovery and involves water injection, gas injection, steam injection, combustion, miscible fluid displacement and polymer injection [24]. In this paper, only water injection or waterflooding will be discussed. Waterflooding involves pumping water into the reservoir to stimulate production. The injected water provides pressure to force the oil out of the rock and to sweep it toward producing wells as shown in Fig. 2. Waterflooding has been attempted in almost every type of reservoir, with its greatest success in relatively homogenous reservoirs having sufficient permeability to allow water injection at a reasonable rate [24]. Up to 60% of the oil can be recovered with waterflooding [5]. Water handling can become a major operational procedure. For example, in some western Canadian oil fields, the proportion of water in the oil-water emulsion brought to the surface can be 95% by volume [15]. That is, the volume of water handled is 19 times greater than the volume of oil produced. Water used as injection water can be of three types: formation water, sea water or fresh water. Formation water is subsurface brackish or brine water produced from a petroleum or non-petroleum producing formation. Sea water may also include water from a salty (non-potable) lake. Fresh water, containing
312
less than 2000 ppm dissolved solids, is primarily water that can be made potable by flocculation, filtration and chlorination [25]. Because oil field reservoir rocks are porous, they are susceptible to plugging by solids suspended in or precipitated from an injection fluid [26]. This makes water quality testing necessary to determine parameters such as: amount and composition of suspended solids, clay sensitivities, presence of bacteria, compatibility of two or more waters, and compatibility of the injection solution with reservoir rock. An example of incompatible waters occurs when sulfate scales, such as barium sulfate, calcium sulfate or strontium sulfate are formed by mixing waters containing sulfate with waters containing barium, calcium or strontium ions [26]. As well, the gases O2, H2S and CO2 found in injection waters and implicated in corrosion [25-26], must be monitored. Water quality testing, should be continued after the enhanced oil recovery operation hasstarted, to ensure that the system is maintained at optimum conditions [25]. Water treatment methods are outlined by Rose et al. [27].
Fig. 2. A simple waterflooding operation. Oil, gas and water are collected from the production wells and the produced water is separated from the oil and gas. The produced water is combined with source water and injected into the oil-bearing rock to pressurize the formation and sweep the oil to the producing wells.
313
Water should be free of bacteria that can cause corrosion [25-26], or plugging of equipment and injection wellbores [25]. The presence of bacteria can be problematic because they reproduce rapidly over wide ranges of pH, temperature, pressure and anoxia in the reservoir. Bacteria found in oil field injection waters that cause problems are SRB, iron-reducing bacteria and slimeformers [25, 27]. Of special concern are the SRB. Source waters used in waterflooding can increase the activities of SRB souring for several reasons [1]. The source water, especially sea water, may contain sulfate to serve as a terminal electron acceptor and may introduce SRB, nutrients such as short chain fatty acids and ammonium into the reservoir. Large volumes of source water may reduce the salinity and temperature in the formation near the injection well, providing an environment that is more conducive to the growth of SRB and oil field souring. 4. SULFATE-REDUCING BACTERIA Ask any person who works in the oil field or who is involved with the transport or storage of crude oil to name some bacteria, and most will immediately respond "sulfate-reducing bacteria" or "SRB". These bacteria are well-known, and in the oil field environment, they are a nuisance because their metabolic activities produce H2S that can sour reservoirs, create plugging through FeS formation and induce corrosion [28]. SRB have the unique ability to utilize sulfate as a terminal electron acceptor. This is an anaerobic respiratory process used to generate energy for the biosynthetic reactions involved in cell growth and maintenance [29]. The SRB are a diverse group of prokaryotes that are found in many anaerobic environments. These bacteria have been the subject of several books [30-33] and countless articles. The phylogeny of SRB has recently been reviewed [34], and based on rRNA sequences, they fall into four groups: Gramnegative mesophiles, Gram-positive endospore-formers, thermophilic bacteria, and thermophilic Archaea. 4.1. Overview of the metabolism of SRB The dissimilatory H2S-producing SRB have little energy available to them. The upper limits of energy conservation from sulfate reduction are set by thermodynamics. For example, if a potent electron donor like H2 is oxidized, the free energy change of the overall reaction, under standard conditions at neutral pH, is -38 kJ (mole H2)A (reaction 1), which is 6-fold lower than with O2 as a terminal electron acceptor (reaction 2) [35]. 4H2 + SOzf + 2H+ -> H2S + 4H2O
G°' = -38 kJ (mol H2)"1
(1)
314
4H2 + 202 -> 4H2O
G°' = -237 kJ mol H2)"1
(2)
As late as the 1970's, only a few genera of SRB were recognized, and these were known to use only a few growth substrates, most notably lactate, pyruvate or H2. Now it is apparent that SRB are capable of using various compounds for electron donors. Based on their metabolic capabilities, heterotrophic SRB fall into two groups: those that cannot oxidize acetate, and those that carry out complete oxidation of acetate to C0 2 [36]. Reaction (3) illustrates the overall reaction of lactate-utilizing SRB that cannot oxidize acetate. One mol of acetate accumulates for each mol of lactate that is consumed. 2CH3CHOHCOO" + S04 = + 2H+ -> 2CH3COO" + 2H2O + 2CO2 + H2S G°' = -77 kJ (mol lactate)"1
(3)
The complete oxidation of acetate is given by reaction (4), showing that less energy is available per mol of acetate than per mol of lactate (reaction 3). CH3COO" + SOzf + 3H+ -» 2CO2 + H2S + 2H2O G°' = -41 kJ (mol acetate)"1
(4)
Increased understanding of the metabolic diversity of SRB now indicates that nearly 100 organic compounds can be used by various SRB [37]. These substrates include fatty acids up to C2o; aromatic hydrocarbons such as toluene, xylenes, ethylbenzene, and naphthalene; «-alkanes from (C6 to C2o); and simple oxidation products of hydrocarbons such as benzoate, phenol, and cresol [3840]. These substrates are present in native crude oils or partially degraded crude oils. Thus, if there is an ample supply of sulfate in water contacting crude oil in an anaerobic environment, there is the potential for SRB to actively produce H2S, using many different organic compounds (or H2) as an energy source. The ability to reduce sulfate links this diverse group of bacteria. However, it is now apparent that various SRB can reduce other chemical species including Fe(III), nitrate, some chlorinated aromatics, sulfur oxyanions and O2 [37]. Molecular oxygen can be reduced by most SRB. In this case, the stoichiometry (for example, 2H2 consumed per O2 reduced) indicates that O2 can be completely reduced to water. SRB are also capable of fermentative growth or utilization of other electron acceptors, such as sulfite, thiosulfate and elemental sulfur [12, 35] and tetrathionate [12]. Many SRB are able to ferment organic substrates in the absence of sulfate. For example, Desulfotomaculum orientis can carry out
315
fermentation using homoacetate. Also many SRB can perform a unique fermentation of inorganic sulfur compounds which are disproportionated to sulfate (a more oxidized compound) and sulfide (a more reduced compound). For example, thiosulfate is transformed to equal amounts of sulfate and sulfide, and sulfite is disproportionated to 3/4 sulfate and 1/4 sulfide [35]. Some species of SRB are able to utilize nitrate as an electron acceptor. When nitrate is used as an electron acceptor, SRB produce ammonia, but not N2, as an end product. Nitrite is formed as an intermediate of nitrate reduction and can be reduced by many sulfate reducers unable to reduce nitrate. In the presence of both sulfate and nitrate some SRB will preferentially use one or the other as an electron acceptor, and some SRB will reduce both concomitantly [35]. 4.2. Activities of SRB in anaerobic environments When microorganisms get into stagnant or closed water systems, dissolved O2 is quickly and completely consumed. Despite the absence of O2, organic matter may undergo biological decomposition by microbial activities, including fermentation. The degradation reactions by which most fermentative bacteria gain energy are disproportionations of the organic matter, part converted to CO2, and part converted to reduced products, such as fatty acids, H2, and alcohols [18]. If sulfate is abundant in these anaerobic environments, the fermentation products are used by SRB. Sulfate serves as the terminal electron acceptor, and the reducing power from the decomposed organic matter results in the formation of H2S. SRB grow in anaerobic muds found in fresh water or sea water environments [41]. They are also indigenous members of the microbial community in ground waters, marine environments, coastal sediments, marine hydrothermal vents associated with volcanic or tectonic activity, and hot springs [42]. SRB can flourish in environments wherever decomposable organic matter gets into anaerobic, sulfate-containing waters. Here H2S is produced and evidenced by visible blackening of the sediment when FeS forms from iron minerals [18]. Marine and estuarine saltmarsh sediments, saline and hypersaline lakes and ponds, as well as oil field waters with high sulfate content are the most permanent and significant habitats of SRB [43]. Large amounts of sulfate are required for this process, so that the consequence resulting from the growth of SRB is the dissemination of massive quantities of H2S [29]. Many SRB use simple, low molecular weight compounds, and therefore depend on fermentative bacteria to cleave and ferment complex organic matter. SRB convert only about 10% of the total substrate carbon to cellular material, so that the bulk of the substrate has to be decomposed for providing energy. Thus,
316
SRB, make themselves conspicuous by the formation of their metabolic product, H2S, rather than by formed cell mass [18]. How do SRB become so closely linked to oil recovery processes? Some think that SRB are imported with surface or ground waters. This hypothesis is illustrated by a gradual increase of sulfide production after the beginning of operations in oil fields [18]. Azadpour et al. [42] reported that SRB were absent in thirteen core samples of petroliferous formations obtained from a wide variety of geographical locations, depths and types of formations. Produced waters from six of the wells were also tested and five were positive for SRB. Acetateutilizing SRB of the genus Desulfobacter were found in an oil field sea water injection system [44]. In culture, they produced extensive biofilm and exhibited high levels of hydrogenase activity, which suggests a sessile habit and a role in the cathodic depolarization mechanism of microbially influenced corrosion. Others have suggested that deep terrestrial subsurface reservoirs contain active and diverse populations of microorganisms including SRB [12]. Thermophilic SRB isolated from oil field waters in the Norwegian sector of the North Sea were thought to be indigenous to the reservoir [45]. See chapter 14 and Ref. [46] for discussion of microorganisms and oil reservoirs. 4.3. Controlling SRB in oil fields using biocides Virtually all oil field water systems contain some bacteria [27], and biocides are widely used to kill or inhibit the activities of these microorganisms, including SRB. There are two general types of biocides: oxidizing and nonoxidizing. Typically, oxidizing biocides (such as chlorine, sodium hypochlorite, chlorine dioxide, chloroamines and bromine) are used in fresh water systems, whereas non-oxidizing biocides (including aldehydes, quaternary amines, halogenated organics, organosulfur compounds, and quaternary phosphonium salts) are used in many different types of water systems [47]. Biocide application in large waterflooding systems presents problems such as high cost, environmental risks [18], and worker safety. The use of biocides is most successful in controlling unwanted activities in surface facilities. When used to eliminate bacteria in injection water or kill SRB in the formation, the degree of difficulty and expense increases significantly [12]. Nonetheless, application of biocides is the most common method of controlling microbial activities in the oil field. Jack and Westlake [48] reviewed the control of SRB in the petroleum industry. 5. NITRATE-REDUCING BACTERIA 5.1. Types of NRB There are two major groups of bacteria that could be stimulated by the presence of nitrate in anaerobic environments. These are chemoorganotrophs
317
(heterotrophs) that use organic compounds as electron donors and as their carbon source for growth (Fig. 3), and chemolithotrophs (autotrophs) that typically use reduced inorganic sulfur species as electron donors and CO2 as their carbon source for growth (Fig. 4). The latter group is also known as the "colorless sulfur bacteria". Figures 3 and 4 show some of the characteristics of these NRB and their end products from nitrate reduction. These figures broadly represent the types of bacteria that might be stimulated by nitrate, although some, such as Thiobacillus denitrificans, and Paracoccus pantotrophus, (Fig. 4) do not appear to have been described as oil field bacteria. Pseudomonas stutzeri is given as an example of a heterotrophic NRB that might be stimulated by nitrate (Fig. 3). A nitrate-respiring bacterium, that has a 100% similarity to P. stutzeri, was isolated from an enrichment from water injectors in a North Sea oil field [49]. Among the heterotrophs in Fig. 3 are facultative anaerobes (such as some Pseudomonas and Bacillus species), that prefer to grow using O2 as their terminal electron acceptors, but will grow using nitrate as their terminal electron acceptor in the absence of O2. These are known as denitrifying bacteria, yielding N2 as the major endproduct of nitrate respiration. There have been countless studies of denitrifying bacteria in soils and wastewater treatment, but these bacteria have been largely ignored in oil field studies. Denitrifying bacteria have been shown to degrade a variety of hydrocarbons (for review see Refs. [39-40]), and with the abundant supply of dissolved hydrocarbons in produced waters, these heterotrophs may be stimulated by nitrate injection into a reservoir. Another group of heterotrophic, facultative anaerobes is the ammoniumproducing, NRB, such as Citrobacter spp. (Fig. 3), other members of Enterobacteriaceae, and a few other genera [50]. We have found no investigations that have described ammonium production in oil field waters by this group of facultative anaerobes. However, Telang et al. [51] mentioned an oil field isolate (designated NH15b) that was tentatively identified as a Citrobacter sp. or Salmonella sp. These would have the potential to reduce nitrate to ammonium. Using a MPN method with medium that is selective for heterotrophic, ammonium-producing, NRB, we have observed that these NRB were detected, but not abundant, in western Canadian oil field waters nor were their numbers greatly increased when nitrate was added to laboratory incubations of produced waters [Eckford and Fedorak, unpublished data]. Recently, the strictly anaerobic ammonium-producing, nitrate-reducing bacterium, Denitrovibrio acetiphilus was isolated from an oil reservoir model column, and it was shown to produce ammonium in medium that contained acetate and nitrate [52]. Some SRB (Desulfovibrio spp.) have also been included as heterotrophs that might be stimulated by the addition of nitrate (Fig. 3) because a few of these
318
reduce nitrate to ammonium [53-56]. In the presence of nitrate, some SRB will preferentially use nitrate, and some will use both concomitantly [54]. Thiobacillus denitrificans is listed as one of the chemolithotrophs in Fig. 4. In general, this species is not tolerant to high sulfide concentrations, but Sublette and Woolsey [57] enriched Thiobacillus denitrificans strain F that initially tolerated up to 1.75 mM sulfide, and later up to 2.5 mM sulfide [58]. This strain has been used in studies to demonstrate its ability to reduce H2S concentrations in porous rock cores [59-60] and in sour produced waters [58,61]. Gevertz et al. [62] described two novel bacterial isolates that are obligate chemolithotrophs, using nitrate as a terminal electron acceptor, and sulfide as an energy source. Both grow under anaerobic conditions. One isolate is a denitrifier that closely resembles Thiomicrospria denitrificans, and it has been called Thiomicrospria strain CVO (Fig. 4). The other isolate was called Arcobacter strain FWKO B, and it reduces nitrate to nitrite.
Fig. 3. Examples of some heterotrophic bacteria that could be stimulated by the presence of nitrate in anaerobic environments that contain suitable organic substrates.
319
Fig. 4. Examples of some chemolithotrophic bacteria that could be stimulated by the presence of nitrate in anaerobic environments. See text for details.
Injection of nitrate into an oil field might also stimulate the activity of bacteria similar to P. pantotrophus [63] (formerly Paracoccus denitrificans [64] and Thiosphaera pantotropha strain GB17 [65]). This bacterium was isolated from a denitrifying effluent treatment system. It is a facultative anaerobe and facultative autotroph (Fig. 4) that uses nitrate as an electron acceptor. It grows autotrophically with sulfide as an electron donor, or heterotrophically with a variety of organic compounds (including acetate which is commonly found in produced waters [66-67]) as electron donors [65]. We are not aware of any research that has detected facultative chemolithotrophs in oil field waters. The bacteria shown in Fig. 4 all have the capability of oxidizing sulfide while reducing nitrate. These are referred to as nitrate-reducing, sulfideoxidizing bacteria (NR-SOB). Greene et al. [68] compared the sulfide tolerance of four species of NR-SOB. In their liquid medium, sulfide was oxidized by Thiobacillus denitrificans strain F at concentrations less than 0.5 mM, by Thiomicrospira denitrificans and Arcobacter sp. strain FWKO B at up to 3 mM, and by Thiomicrospira strain CVO at up to 15 mM.
320
Although only a few NR-SOB have been identified in oil field waters, Loka Bharathi et al. [69] isolated over 100 strains of anaerobic colorless NRSOB from sea water and a sulfide-rich creek. Their data showed that different isolates oxidized sulfide at different rates. For example, one isolate oxidized all of the sulfide in the medium within 9 days, whereas another isolate oxidized only 2.9% of the sulfide in the same time. Thus, it is likely that different NRSOB in the produced water from oil fields would oxidize sulfide at different rates. 5.2. NRB in oil field waters The presence of NRB in oil field waters has not be studied extensively. This group of microorganisms was not even mentioned in a review entitled "Microbiology of petroleum reservoirs" [46]. Several investigations have enumerated NRB in oil field waters using most probable number (MPN) methods with different media formulations. Some of the results are summarized in Table 1, in chronological order. One of the first enumeration studies [70] used molasses or sucrose as electron donors in the media to count heterotrophic NRB in samples taken as near the wellheads as possible. Very low numbers ( 4 L"1) were found in these samples. Most of the other media formulations preferentially, but not exclusively, cultured autotrophs. For example, the medium used by Davidova et al. [14] (Table 1) contained only inorganic compounds except for yeast extract, with thiosulfate serving as the electron donor. This would preferentially grow microorganisms that are similar to Thiobacillus denitrificans. Other investigations in Table 1 used sulfide as the electron donor with filter-sterilized produced water from the oil field that was being studied [51, 71]. The filtered produced water undoubtedly contained some dissolved organic compounds, so it would support the growth of heterotrophic NRB and autotrophic NRB. The medium used by Telang et al. [72] in Table 1, contained only inorganic compounds except for acetate, with sulfide serving as the electron donor. Telang et al. [72] in Table 1 described the isolation and characterization of two autotrophic NR-SOB from an oil field in Saskatchewan, Canada. One was designated Thiomicrospira strain CVO (formerly Campylobacter strain CVO, [51 ]) and the other was designated Arcobacter strain FWKO B. The DNA from these two isolates has been used extensively with a method known as reverse sample genome probing (RSGP), first described by Voordouw et al. [73]. Using RSGP, Telang et al. [51] (Table 1), demonstrated that the abundance of strain CVO increased after the waterflooded oil field was treated with nitrate. This molecular technique corroborated the increase in NR-SOB numbers determined by the MPN method. The high specificity of the RSGP for NR-SOB precluded the detection of other NRB in samples from four additional oil fields
321
from western Canada and west Texas [72], although culture methods detected NRB (Table 1). Eckford et al. [74], in Table 1, surveyed five oil fields in western Canada for various types of NRB. Different media formulations were used to selectively enumerate thiosulfate-oxidizing NRB, heterotrophic NRB, or NR-SOB. None of the 18 water samples contained detectable numbers of thiosulfate-oxidizing NRB. As was observed by Adkins et al. [70], the numbers of NRB were very low or non-detectable near the wellheads [74]. However, NRB were detected in source and preinjection waters, and in samples from water storage tanks and free water knock out units. Although much of the work on NRB in oil field waters has neglected the heterotrophic NRB, the numbers of heterotrophic NRB were greater than the numbers of autotrophic NRB in 12 of the 15 samples compared. In one oil field, heterotrophic NRB were found, but no autotrophic NRB were detected (Ref. 74, Table 1). NRB were detected in biofilms on coupons in the anaerobic part of the water injection system of the Veslefrikk field in the North Sea [75], (Table 1). The medium used to enumerate these attached bacteria contained organic acids as carbon sources, providing counts of heterotrophic NRB. These numbers increased dramatically after nitrate injection (Table 1). The literature surveyed in Table 1 represents 15 different oil fields that have been examined for NRB. Each of the oil fields contained detectable numbers of NRB at one or more sampling locations. Thus, each field had a microbial community containing NRB with the potential to be stimulated by nitrate amendment. 6. CONTROLLING MICROBIAL PRODUCTION OF SULFIDE WITH NITRATE ADDITION 6.1. Microbial mechanisms leading to the control of sulfide concentrations after nitrate addition There appear to be five mechanisms by which sulfide concentrations can be controlled in the presence of nitrate and sulfate. The first involves the competition between heterotrophic NRB and SRB for a common electron donor. For example, acetate serves as an electron donor for NRB [76] and for several genera of SRB [34]. Equations (5) and (6) illustrate that if acetate is available, nitrate reduction yields more energy per mol of electron donor or acceptor than does sulfate reduction [77].
322
Table 1 Detection and enumeration of NRB in oil field waters. Refs.
Oil
70
fields
Methods
Comments
Oklahoma, USA
MPN with molasses and sucrose as electron donor
Samples collected near wellheads. Medium would detect heterotrophic NRB. MPN values were 4mL"\
71
Saskatchewan, Canada
Single-bottle MPN using filter-sterilized oil field water supplemented with nitrate
Oil field water contained about 120 mg sulfide L"1. Method likely selected for NR-SOB. Initial count, 104 ml/ 1 , Count after nitrate injected into reservoir, 108 mL"1.
51
Saskatchewan, Canada
Single-bottle MPN using filter-sterilized oil field water supplemented with nitrate
Oil field water contained about 100 mg sulfide L"1. Method likely selected for NR-SOB. Initial counts as low as 0 mL"1. Counts after nitrate injected into reservoir, as high as 108 mL"1.
51
Saskatchewan, Canada
RSGP
NR-SOB strain CVO became dominant community member after nitrate injection into reservoir.
72
Western Canada Single-bottle MPN and west Texas, using medium with USA sulfide, acetate and nitrate
Method likely selected for NR-SOB, but may have grown heterotrophic NRB. Counts from 102 mL"1 to 106 mL"1 in five samples examined.
72
Western Canada RSGP & west Texas, USA
NR-SOB strains CVO and FWKO B detected in only one of five samples examined.
14
Oklahoma, USA and Alberta, Canada
MPN with inorganic salts, yeast extract, and thiosulfate as the electron donor.
Method likely selected for thiosulfateoxidizing NRB, but may have grown heterotrophic NRB. Counts were typically <500 mL"1.
74
Alberta and Saskatchewan, Canada
MPN with three different media. One selected for thiosulfate-oxidizing NRB, one selected for heterotrophic NRB, and one for NR-SOB.
No thiosulfate-oxidizing NRB detected in any of 18 samples. Other NRB detected in 16 samples. Number of heterotrophic NRB greater than number of NR-SOB in 12 of 15 samples.
75
Veslefrikk in the North Sea
MPN with acetate, butyrate, caproate and lactate as carbon sources.
Sampled biofilms on coupons in water injection system. Prior to nitrate injection, 103 NRB cm"2, after 18 months of nitrate injections, a 60,000fold increase in NRB occurred
323
5CH3COO" + 8NO3" + 3H+ -> IOHCO3" + 4N2 + 4H2O AG0' = -495 kJ (mol N O 3 ) " ' or AG°' = -792 kJ (mol acetate)"1 (5) CH3COO" + SO4= - • 2HCO3~ + HS" AG°' = -47 kJ (mol acetate or SO4= )"' (6) Thus, heterotrophic NRB out-compete heterotrophic SRB for electron donors, thereby suppressing sulfide production. Oil field waters contain dissolved organic compounds including short-chain fatty acid anions like acetate, propionate and butyrate [12, 46, 67], as well as aromatic compounds such as toluene and phenols that are substrates for heterotrophs. This mechanism would stop sulfide production, but it would not remove sulfide that is present in the reservoir or produced waters. A second mechanism results from the increased redox potential of an aqueous environment caused by the activities of denitrifying bacteria [78]. The production of N2O (and maybe NO), two oxidizing agents, raises the redox potential to above -100 mV, which is too high for the growth of SRB [31]. Nitrate reduction in laboratory experiments causes the redox indicator, resazurin, to turn from colorless to pink [78-79]. Resazurin is 50% oxidized at -51 mV [80]. This alteration of the redox potential in an aqueous environment inhibits sulfide production. A third mechanism results from the stimulation of NR-SOB in the presence of nitrate. Two processes come into play in this case. Some NR-SOB are denitrifiers and they produce N2O from nitrate, thereby elevating the redox potential of the medium [78]. In addition, the NR-SOB use sulfide as their electron donor, and oxidize it to elemental sulfur or sulfate [2]. Thus, these two processes combine to inhibit sulfate reduction and remove sulfide that is present in the aqueous environment. The activities of the NR-SOB have the potential to stop sulfide production, and to remove essentially all of the sulfide in the aqueous environment. A fourth mechanism is nitrate reduction by SRB. Some SRB reduce nitrate to ammonium [53-56]. The importance of this mechanism in controlling sulfide production is largely unexplored. Jenneman et al. [78] point out that when SRB reduce nitrate, ammonium is formed rather than N2O or N2. The formation of N2O would be detrimental to the SRB as discussed above. The fifth mechanism is the production and accumulation of nitrite during nitrate reduction. Myhr et al. [49] demonstrated that the activity of the dominant sulfate-reducing strain found in their laboratory experimental system was inhibited by 120 uM nitrite. However, some species of SRB contain nitrite
324
reductase which reduces nitrite to ammonium [81], thereby protecting these species from the nitrite produced by NRB [68]. 6.2. Control of sulfide in wastewaters Long before nitrate addition was considered for controlling sulfide in oil field waters, it was used to control odors in wastewaters and receiving surface waters. For example, in 1931, a combination of sodium nitrate and chlorinated lime was used to control odors from Coney Island Creek in New York [82]. This creek was described as "one of the vilest bodies of water in the United States" [82] as a result of receiving 6,000,000 gallons (23,000,000 L) of sewage and industrial wastewater. After the first day of chemical application, there was a marked decrease in odor. During the month-long treatment, 10 tons (9 Mg) of sodium nitrate were applied to the creek, and the sulfide concentrations in the water decreased sharply. Table 2 summarizes five studies, in chronological order, in which nitrate was used to control odors and sulfide production in wastewaters. The first four entries in Table 2 enhance the activities of native NRB by adding nitrate. Two of these were large scale projects that involved nitrate applications to a river [83] and to a sludge storage lagoon [84] for odor control. The other three studies were laboratory-scale investigations using sewage sludge [78], oily sludges from naval operations [85], and aqueous solutions of sulfide [58]. The latter report described work in which Thiobacillus denitrifwans was initially used to oxidize sulfide, and later Thiobacillus denitrificans strain F was used because of its tolerance to higher sulfide concentrations. Each of the attempts to control odor or sulfide production listed in Table 2 was successful. One of the studies [84] observed that nitrate amendment led to increased redox potential followed by a reduction in odor. The increased redox potential was observed in another study [78] and this was attributed to the microbial production of N2O. The increase in redox potential to above -100 mV would inhibit growth of SRB. 6.3. Laboratory studies using cores or columns Using nitrate to control sulfide production in a petroleum reservoir involves adding nitrate to the injection water and pumping it into the oil-bearing formation. To be effective, the nitrate must migrate into the reservoir and be consumed by NRB. The NRB may be present in the oil field or water handling system, or they might be deliberately added to the oil field to stimulate nitrate reduction. Several laboratory studies have been done to assess the effectiveness of this process using cores or a column of sand. Five of these studies are summarized in Table 3, in chronological order.
325
Table 2 Laboratory and field studies using nitrate to control sulfide production in wastewaters Ref.
Summary
83
Three pulp mills discharged sulfite wastes into the Androscoggin River in Maine U.S.A. This resulted in H2S production in the river and odor problems in nearby towns. In 1949, a total 641 tons (582 Mg) of NaNC>3 were added to the river. This controlled H2S production and odors. Most of the nitrate was reduced to ammonium.
84
To control odor, waste sodium nitrate liquor (containing both nitrate and nitrite) was added to a storage lagoon that held aerobically digested waste activated sludge. Initially, the redox potential of the water was near -lOOmv, but after several months of nitrate addition, it rose to near +300 mV. There was low odor potential when the redox was above +100. Acetate concentrations decrease in the lagoon, and N2 production from denitrification provided mixing within the sludge.
78
Laboratory studies were done with a 10-fold dilution of sewage sludge amended with 20 mM sulfate and one of three electron donors: glucose, acetate, or H2. The addition of 59 mM nitrate completely inhibited sulfide production. Nitrate, nitrite and N2O were detected in the inhibited samples, and the oxidation of the redox indicator, resazurin, was attributed to the presence of N2O. The numbers of SRB decreased with prolonged incubation of the oxidized medium.
85
Oily sludge from a settling tank at the U.S. Navy Craney Island Fuel Depot in Virginia was placed in serum bottles and amended with nitrate, stimulating indigenous NRB. Sulfate reduction was diminished with 50 mM nitrate, and sulfide accumulation was prevented with as little as 16 mM nitrate. Nitrite and nitrous oxide were products of nitrate reduction. Sulfide was oxidized to sulfur or sulfate. The results indicated that nitrate would be useful for preventing sulfide formation in oily wastes produced onboard marine vessels.
58
This paper reviewed bench-scale processes developed for the sulfide removal from gases and aqueous solutions by Thiobacillus denitrificans. When H2S was introduced to batch anoxic or aerobic cultures of T. denitrificans, the H2S was immediately metabolized. Oxidation of H2S to sulfate was accompanied by growth. T. denitrificans was immobilized by co-culture with floc-forming heterotrophs and this mixture was used to treat water that was contaminated with sulfide. The sulfide-active floe was stable for 5 months of operation with no external organic carbon required to support the growth of the heterotrophs. T. denitrificans strain F, which tolerates higher sulfide concentrations, was also used in some studies.
326
Table 3 Laboratory studies using nitrate to control sulfide production columns or cores Ref.
Summary
59
This study investigated the efficacy of nitrate and the sulfide-tolerant Thiobacillus denitrificans strain F in controlling H2S concentrations in cores of sandstone. Formation water from a gas storage facility in Redfield, Iowa, U.S.A. was injected into two core systems, with hydraulic retention times (HRTs) of 3.2 h and 16.7 h. With the addition of nitrate alone, no thiobacilli were cultured from the core system, but nitrate was consumed and the concentrations of sulfide in effluent decreased by about 40% in the core with the shorter HRT, and 98% with the longer HRT. Thus, an indigenous microbial community capable of oxidizing sulfide while using nitrate as the electron acceptor was present. Inoculation with strain F reduced the effluent sulfide by about 80% in the core with the shorter HRT.
60
The test materials for this study included core material from the St. Peter formation at Redfield, Iowa, U.S.A. and water from the same formation, supplemented with acetate and enriched with SRB to 107 cells ml/ 1 . The core material did not contain large numbers of organisms capable of using nitrate, and no strain F-like organisms were detected. When nitrate and strain F were injected into the core, sulfide concentrations decreased, demonstrating the ability of strain F to control sulfide in the core.
86
This work examined controlling microbial souring in anaerobic upflow columns containing crushed Beria sandstone maintained at 60°C. Produced waters from the Ninian North Sea and the Kuparuk North Slope oil fields were used as sources of microorganisms, and these gave similar results. A highly anaerobic medium that contained short-chain organic acids found in the produced waters was pumped through the columns. Nitrate injection stimulated indigenous microbes and inhibited souring at thermophilic temperatures. Initially, 3.6 mM nitrate was needed to inhibit souring but later 0.36 mM nitrate prevented further souring. Nitrate was reduced to nitrite, with no N2O, N2 or ammonium detected.
2
Brine from an oil field near Coleville, Saskatchewan, Canada was filtered, supplemented with phosphate and nitrate and pumped into a porous (1288 mD) ceramic core 19.1 cm long. When 5 mM nitrate was shut in the column, all of the sulfide was removed in 3 d and the numbers of NRB increased. Under various flow regimes, with sulfide-containing brine, sulfide removal was between 87 and 100%. Elemental sulfur, bacteria and CaCC>3 were produced, but there was no significant permeability changes across the core following all treatments.
49
Separate enrichments of aerobic oil-degrading bacteria, NRB, SRB and methanogens were inoculated into a 200-cm column packed with oil-soaked silica sand. The column was flooded with air-saturated synthetic sea water and operated under different influent regimes for nearly 1100 d. Injecting 0.5 mM nitrate led to the complete elimination of H2S. Inhibition of the SRB was attributed to the nitrite produced from nitrate reduction. Three strains of heterotrophic NRB were isolated from the column and none used H2S or S° as electron donor.
327
Four of the five studies in Table 3 detected NRB in the cores or produced waters used in the experimental systems. In the fifth study, [49] the investigators inoculated the column with a mixture of enrichment cultures, including NRB. Two of the studies, Refs. 59 and 60, focused on the activities of thiobacilli. None were detected in the cores or waters, similar to the findings of Eckford and Fedorak [74]. Inoculating these two cores with Thiobacillus denitrificans strain F stimulated sulfide reduction when nitrate was injected into the cores (Refs. 5960, Table 3). Two of the studies [2, 86], (Table 3) relied solely on the formation water as the source of NRB. One study supplemented the medium with short-chain organic acids [86], whereas the other study did not supplement with organic compounds [2]. Thus, these studies likely enriched for different nutritional types of NRB. Nonetheless, souring was inhibited in both studies. Indeed, sulfide production was controlled in each of the five studies summarized in Table 3. 6.4. Laboratory studies using natural microbial communities in produced waters Produced waters from various oil fields have been used as sources of planktonic microorganisms in studies of the ability of nitrate to control sulfide formation in these waters. Table 4 summarizes four of these investigations in chronological order. In each study, sulfide removal was stimulated by nitrate addition. In three of the reports, no organic supplementation was required to stimulate sulfide removal. However in one case [87], two of the four oil field waters did not respond to amendments with inorganic nutrients (nitrate and phosphate). Sulfide removal was only stimulated after the addition of acetate or formate plus vitamins or yeast extract, indicating that in some cases heterotrophic NRB play an important role in the process of sulfide removal. Eckford and Fedorak [15] demonstrated that heterotrophic NRB can be stimulated by simply adding nitrate. This is illustrated in Figs. 5 and 6. A produced water sample was collected from the free water knock out at the Coleville field that has a severe souring problem. This water was used for a serum-bottle microcosm study. Initially, the microcosm contained 2.7 mM sulfide which increased to 3.1 mM by day 1 and then dropped below detection by day 3 in the nitrate-amended microcosm (Fig. 5a). The sulfate concentration increased noticeably over the first 14 d of incubation, with a total increase of 3.5 mM by day 38, closely matching the 3.1 mM decrease in sulfide. The nitrite concentration was at a maximum of 1.8 mM on day 3 and then gradually decreased to 0.2 mM by day 38. Figure 5b shows the results of chemical analyses of a microcosm that was not supplemented with nitrate. The sulfide increased to 4 mM by day 5, and the sulfate remained fairly steady at from 0.68 mM to 0.54 mM throughout the testing period. Neither nitrate nor nitrite was detected in the microcosms. The huge increase in numbers of heterotrophic NRB
328
(Fig. 6a) during the time that sulfide was removed (Fig. 5a) suggests that these bacteria play a role in this process. However, their role has not be elucidated.
Table 4 Laboratory studies on controlling sulfide production in produced waters by adding nitrate to stimulate natural microbial communities. Ref.
Summary
71 & 2
Anaerobic enrichments were prepared by supplementing nitrate and phosphate to brine samples collected from an oil field near Coleville, Saskatchewan, Canada. Within 24 to 48 h after supplementation, complete oxidation of 3 to 4 mM sulfide was observed. Elemental sulfur was formed and the stoichiometry of the reaction was 5HS" + 2NO3~ + 7H+ -> 5S° + N2 + 6H2O.
87
Waters from four west Texas oil fields were used to determine which amendments were required to stimulate sulfide removal. In two of the samples, addition of 40 mM nitrate and phosphate was not sufficient to promote microbial removal of sulfide over a 28-d incubation. However, sulfide removal was observed when acetate or formate plus vitamins or yeast extract were added to these two waters that had been supplemented with nitrate and phosphate. These results illustrate the importance of heterotrophic activity in sulfide removal.
14
Two waterflooded, souring oil fields in Oklahoma, U.S.A. and Alberta, Canada were studied. SRB and NRB were found in produced waters from both oil fields. The majority of the sulfide production appeared to occur after the oil was pumped aboveground, rather than in the reservoir. Sulfide production was greatest in the water storage tanks in the Alberta field. Laboratory experiments showed that adding 5 and 10 mM nitrate to produced waters from the Oklahoma and Alberta oil fields, respectively, decreased the sulfide content to negligible levels and increased the numbers of NRB.
15
Produced waters from three sulfide-containing western Canadian oil fields were amended with nitrate only. In less than 4 d, the sulfide was removed from the waters from two of the oil fields (designated P and C), whereas nearly 27 d were required for sulfide removal from the water from the third oil field (designated N). Nitrate stimulated large increases in the numbers of the heterotrophic NRB and NR-SOB in the waters from oil fields P and C, but only the NR-SOB were stimulated in the water from oil field N. These data suggest that the stimulation of the heterotrophic NRB is required for rapid removal of sulfide from some oil field produced waters.
329
Fig. 5. Chemical analyses of microcosms that contained produced water from the Coleville oil field in Canada. Nitrate amended (a), unamended (b). From Ref. 15.
Bacterial enumerations were done on samples from the nitrate-amended and the unamended microcosm. The MPN results are shown in Fig. 6. Initially, the number of NR-SOB (2.1xlO5 ml/ 1 ) was much greater than the number of heterotrophic NRB (4.3x102 ml/ 1 ). There was no increase in the numbers of heterotrophic NRB (Fig. 6a) or NR-SOB (Fig. 6b) in the unamended microcosm. In contrast, there was a rapid increase in the numbers of heterotrophic NRB and NR-SOB by day 7 (Figs. 6a and 6b) in the nitrate-amended microcosm. The numbers of heterotrophic NRB and NR-SOB increased 22,000-fold and 440fold, respectively. These proliferations occurred during the time when nitrate consumption was the most rapid, and sulfide was depleted from the microcosm (Fig. 5a). At day 7, the numbers of heterotrophic NRB and NR-SOB were 9.3xl0 6 ml/ 1 and 9.3xl0 7 ml/ 1 , respectively. Over the remainder of the
330
incubation, the heterotrophic NRB numbers remained high, whereas the NRSOB numbers dropped to near their original count (Figs. 6a and 6b). The SRB numbers did not change in the nitrate-amended microcosm and showed a slight increase in the unamended microcosm with a maximum at day 7 (Fig. 6c).
Fig. 6. Heterotrophic NRB (a), NR-SOB (b) and SRB (c) counts is samples from microcosms that contained produced water from the Coleville oil field in Canada (Fig. 5). Error bars show 95% confidence intervals. From Ref. [15].
331
Laboratory studies have led to field application of nitrate or changes to field operations. For example, the work described in references [2, 71] (Table 4) preceded the experimental injection of nitrate into the Coleville field in Saskatchewan, Canada [13, 71, 88], and results from laboratory studies encouraged the implementation of nitrate injection in a North Sea oil field [75]. Based on laboratory investigations, Davidova et al. [14] observed that the rate of sulfide production was higher in aboveground samples than in samples collected from wellheads. At an Alberta oil field, they observed high sulfate reduction activity in water storage tanks that had retention times of 2 to 3 d, and they calculated that 80 kg of microbially-produced sulfide was injected into the reservoir daily from these storage tanks. Operators of this oil field have now eliminated the long retention time in the storage tanks, which has helped to reduce souring. 6.5. Laboratory studies using co-cultures of bacteria To assess the microbial dynamics and processes that occur when nitrate is added to communities containing NRB and SRB, Voordouw and co-workers have done several studies in which pure cultures of bacteria were mixed and monitored (Table 5). Their work focused on the activities of the autotrophic NR-SOB Thiomicrospira strain CVO and Arcobacter strain FWKO B. In all cases, the NR-SOB proliferated with the addition of nitrate, and in most cases, they removed sulfide from the medium and caused the cessation of sulfate reduction. However, sulfate reduction was not stopped in co-cultures in which the SRB produced nitrite reductase [68] (Table 5). Nitrite formed during nitrate reduction is inhibitory to SRB. However, the inhibition is only transient when SRB, that produce nitrite reductase, reduce nitrite to ammonium [81]. This work [68] (Table 5) clearly demonstrated that the activities of these NR-SOB cannot control sulfide production by all SRB, although the NR-SOB can oxidize the sulfide that is formed by the SRB. Rates of corrosion have also been studied in co-culture experiments (Table 5) [89]. The addition of strain CVO and nitrate to a culture of Desulfovibrio sp. strain Lac6 accelerated the corrosion rate to 0.07 mm y"1. Lacatena et al. [90] also measured corrosion rates, but they worked with an undefined, mixed enrichment culture in produced water. In the absence of nitrate in produced water, the corrosion rate was 0.46 mm y"1, but with nitrate in the produced water, the corrosion rate dropped sharply to 0.03 mm y"1. Data from nitrate injection into a North Sea oil field showed that prior to nitrate injection the corrosion rate was 0.7 mm y"1, but after 4 months of nitrate injection, the rate dropped to 0.2 mm y"1 [75]. Thus, the co-culture experiments (Table 5, Ref. 89) gave results that differed from those obtained with undefined mixed cultures [90] and full scale operations [75].
332
Table 5 Laboratory studies using co-cultures and nitrate to control sulfide production. Ref.
Summary
72
Mixtures of strains CVO and FWKO B were incubated in medium with different concentration of sulfide. Using RSGP, it was demonstrated that CVO dominated in co-cultures with low (1 mM) sulfide, but FWKO B dominated with high (15 mM) sulfide. CVO or FWKO B were co-cultured with Desulfovibrio strain Lac6. Sulfide drop from 1 mM to 0 mM in 24 h in the presence of CVO. Over a 277-h incubation, sulfide remained between 1 and 2 mM in the presence of FWKO B.
91
Strain CVO was added to cultures of Desulfovibrio strain Lac6 that were growing in various concentrations of nitrate or lactate. In pure culture, sulfate reduction by the Desulfovibrio sp. was unaffected by the nitrate concentrations up to 10 mM. Sulfide concentrations decreased rapidly after the addition of CVO. This effect was due to the increase in the redox potential of the medium, as indicated by the oxidation of resazurin.
89
The influence of nitrate-mediated control of sulfide production on metal corrosion was studied with strain CVO and a Desulfovibrio strain Lac6. The corrosion rate in cultures of the Desulfovibrio sp. without or with nitrate was 0.01 mm y"'. The addition of CVO to the nitrate-containing culture increased the corrosion rate to 0.07 mm y"1. The same trend was observed when CVO and nitrate were added to a consortium of SRB from a produced water. The increased rate of corrosion was attributed to the formation of thiosulfate and polysulfide during the oxidation of sulfide.
68
Strain CVO was grown in co-cultures with four different Desulfovibrio strains. Two of these did not have nitrite reductase, and their growth was stopped in the presence of CVO as it produced nitrite and elevated the redox potential of the medium. However, two of the strains had nitrite reductase, and they reduced the nitrite formed by strain CVO. The SRB decreased the redox potential and continued to produce sulfide. This illustrated that the action of strain CVO cannot inhibit SRB that possess nitrite reductase.
6.6. Oil field observations There have been few reports of field tests or full-scale application of nitrate injection to control sulfide. Six reports are summarized in Table 6 (in chronological order). Three of these focused on the extensive studies done on the Coleville oil field in Canada during two experimental injections [13, 51, 71, 88]. The microbial community in the Coleville oil field was extensively characterized using the RSGP method, and the produced water was the source of the well-studied NR-SOB, Thiomicrospira strain CVO and Arcobacter strain FWKO B. Results from nitrate addition to two oil fields in the North Sea have also been reported [11, 75] (Table 6). These include an 8-month study [11] and a
333
long-term application, with data reported after 32 months of operation [75]. Numbers of planktonic NRB were monitored in the first five studies listed in Table 6, and numbers of sessile NRB were reported in the last study given in Table 6. Three common observations were evident from the field studies summarized in Table 6. First, NRB were present in each of the oil field waters studied. Thus, no intentional inoculation of NRB was required to stimulate the beneficial activities of these bacteria. Second, nitrate injection stimulated the NRB and, in reports in which NRB were enumerated, their numbers increased 100- to 60,000-fold during the monitoring times. Third, nitrate injection controlled sulfide production. Each of these observations was completely predicable from laboratory studies summarized in Tables 3, 4, and 5. 6.7. U.S. Patents The ability to stop sulfide production in oil fields, or to remove sulfide from sour waters and petroleum are essential in petroleum recovery and processing. The inhibition of sulfate reduction decreases corrosion and other problems associated with SRB and provides huge cost saving to the oil field operators. Therefore, it is not surprising that several patents have been issued for the use of nitrate or NRB for sulfide removal or control. Table 7 lists some of the U.S. patents dealing with these processes. Patent no. 4,879,240 uses a mutant strain of Thiobacillus denitrificans that is tolerant to elevated concentrations of sulfide and glutaraldehyde (presumably strain F) to control sulfide in environments such as oil field injection waters, reservoirs, and waste treatment of materials that contain SRB. A sulfide-tolerant strain of Thiobacillus denitrificans is the microbial component of patent no. 4,880,542 used to remove H2S from sour waters originating from petroleum production, anaerobic sewage digestion or other industries. These autotrophic bacteria are co-immobilized with CaCO3 in alginate beads and placed in a column, through which the wastewater is pumped. Nitrate or O2 can serve as the terminal electron acceptor. The activities of heterotrophic denitrifying bacteria are stimulated by supplementing oil field waters (or other sulfide-containing waters) with nitrate and an organic compound, such as acetate (patent nos. 5,405,531 and 5,750,392; Table 7). This allows the NRB to out-compete the SRB for organic substrates. In addition, these patents include the addition of molybdate to further inhibit SRB. The use of the autotrophic NR-SOB Thiomicrospira (formerly Campylobacter sp.) strain CVO and Arcobacter strain FWKO B for the removal of sulfide from oil field brines is covered by patent nos. 5,686,293 and 5,789,236 (Table 7). The uses include aboveground treatment of sour waters or injection of these NR-SOB into subterranean formations. The waters are supplemented with nitrate and phosphate.
334
Table 6 Field studies and operations using nitrate to control sulfide production. Ref.
Summary
12
Ammonium nitrate (45 T) was injected into a souring oil field at the Southeast Vassar Verta Sand Unit in Oklahoma, U.S.A. At the time of injection, no nitrate was detected in three adjacent production wells. Forty-five days after injection, nitrate was detected at these wells, and the sulfide concentrations were reduced by 40 to 60%.
71
In 1994, a solution of NH4NO3 and NaH2PO4 was injected into three wells in the Coleville field in Saskatchewan, Canada. Prior to treatment, the produced waters from these wells contained between 52 and 160 mg sulfide L"1. After injection, there were shut-in periods of between 24 and 70 h before pumping resumed. The sulfide concentrations dropped by as much as 98% of the initial concentrations, with ranges between 40% and 60% being sustained for several hours. The numbers of NRB increased by 100- to 10,000-fold.
88 & 13
In 1996, a solution of NH4NO3 and NaH2PO4 was injected into two injection wells in the Coleville field for 50 d. Two producer wells were monitored for 90 d after the injection began. After 10 d, the sulfide in the producers decreased by as much as 50 to 60% of the initial concentrations of 60 and 40 mg L"1. The cumulative sulfide removal from the two producers were estimated to be 50 and 70 kg over the 90-d test period. The numbers of NRB increased at least 1,000-fold during the time of nitrate injection.
51
Samples were taken from the Coleville field in 1996. These were taken 8 d before and 20, 55, and 82 d after the injection of a solution of NH4NO3 and NaH2PO4 began. RSGP analyses, using 47 DNA standards, showed that strain CVO became the dominant community member immediately after injection. The abundance of CVO decreased within 30 d after completion of nitrate injection.
11
Studies were done in the Skjold oil field in the North Sea in 2000. Three injection strategies were used. In each case, the highest nitrate concentrations were used at the beginning of the treatment, then the concentration was decreased. First, nitrate (4.5 to 1.7 mM) was injected into one well for 1 month; second, nitrate (3.8 to 1.8 mM) was injected into this well plus another well for 2 months; third, nitrate (4.4 mM to a mean of 2.8 mM) was injected into all of the other wells for 3 months. Only one of the monitored production wells showed marked reduction in H2S. This well was in the highly fractured zone of the reservoir, and nitrate reached it within 24 h of the start of injection. The amount of H2S in the produced gas dropped from 240 ppm to between 30 to 60 ppm. After nitrate addition, the numbers of mesophilic NRB and NR-SOB increased about 10,000- and 1,000-fold, respectively.
75
Data were presented after 32 months of adding nitrate to water injected from the Veslefrikk platform in the North Sea. Glutaraldehyde injection was stopped in January 1999, and replaced by continuous 0.25 mM nitrate injection. Microbial counts in biofilms were monitored and corrosion was measured by weight loss from C-steel biocoupons. After 32 months, the numbers of SRB decreased 20,000-fold and after 18 months, the number of NRB increased 60,000-fold. Most of the NRB were heterotrophic facultative anaerobes. Sulfate-reducing activity (measured using 35S-sulfate) decrease 50-fold. Prior to nitrate treatment, the corrosion rate was 0.7 mm y"1. This fell to 0.02 mm y"1 after 4 months of nitrate injection.
335
Table 7 Examples of United States patents for the control of sulfide through the application of NRB. Patent no.
Inventors and year
Title
4,879,240
Sublette et al. 1989
Microbial control of hydrogen sulfide production by sulfate reducing bacteria
4,880,542
Sublette 1989
Biofilter for the treatment of sour water
5,405,531
Hitzman et al. 1995
Method for reducing the amount of and preventing the formation of hydrogen sulfide in an aqueous system
5,686,293
Jenneman et al. 1997
Sulfide-oxidizing bacteria
5,750,392
Hitzman et al. 1998
Composition for reducing the amount of and preventing the formation of hydrogen sulfide in an aqueous system, particularly in an aqueous system for oil field applications
5,789,236
Jenneman 1998
Process of using sulfide-oxidizing bacteria
6.8. Economics and advantages of using nitrate to control sulfide production Based on the trial injections at the Coleville oil field in Canada, Jenneman et al. [88] did a cost analysis for sulfide removal using different chemicals. They injected ammonium nitrate (cost US$0.31 kg"1) and monosodium phosphate (cost US$2.57 kg"1) to stimulate NRB in the reservoir. The combined cost of these chemicals was determined to be between US$0.76 and $1.19 kg"1 H2S removed. They compared this cost to reported costs for sulfide removal from wastewaters using hydrogen peroxide or sodium hypochlorite. With hydrogen peroxide, the estimated cost was between US$4.40 and $17.60 kg"1 H2S removed, and with sodium hypochlorite the estimated cost was between US$3.96 and $13.20 kg"1 H2S removed. With data from one well, Jenneman et al. [88] estimated the cost of using ammonium nitrate and monosodium phosphate to be $0,018 barrel"1, or $1.80 (100 barrels)"1, of produced water treated. Herbert [92] compared the costs of using nitrate with those of using the biocides glutaraldehyde or tetrakishydroxymethylphosphonium sulfate (THPS)
336
for offshore oil fields. The costs did not include the cost of transporting the chemicals. The estimated prices per litre of the chemicals were: US$0.25 for nitrate (as a 40% solution of CaNO3), $2.50 for glutaraldehyde (as a 50% solution), and $4.00 for THPS (as a 50% solution). Although the cost of nitrate was lower, the solution was continuously injected at a dose of 60 mg L"1. In contrast, the two biocides were injected for 1 h, twice per week at a dose of 500 mg L"1. Based on treating 200,000 barrels of produced water per d, the yearly costs for chemicals were US$575,000 for nitrate, $345,00 for glutaraldehyde, and $500,000 for THPS. Per 100 barrel of water treated, these costs become US$0.79, and $0.47, and $0.68, respectively. From these two cost analyses, the use of nitrate for sulfide control is competitive with other chemicals. The cost of treating 100 barrels of water calculated from the data given by Jenneman et al. [88] is higher than that reported by Herbert [92], because Jenneman et al. [88] also injected monosodium phosphate, which is 8 times as expensive as the ammonium nitrate. Herbert [92] used only calcium nitrate. The need to add a phosphate source to stimulate NRB would have to be evaluated for each oil field. Besides the cost, other factors must be considered when choosing chemicals for controlling sulfide in produced waters. Most notably, workers safety and potential environmental impact of spilled chemical must be considered. Nitrate salts are far less toxic than the biocides commonly used in oil fields, and therefore its use presents few safety issues for oil field workers. Spilled biocides have negative affects on the environment. In contrast, nitrate is widely used as an agricultural fertilizer, so spills on land present no major problem. Nitrate is listed as a substance that poses little or no risk to the marine environment [75]. However, caution must be used to avoid contamination of fresh surface waters or potable ground waters with nitrate (or any biocide). 7. CONCLUDING REMARKS The use of nitrate to control microbially-produced sulfide in oil fields is a proven biotechnology that is grossly under-used by the petroleum industry. Its effectiveness has been demonstrated in many laboratory investigations and in some field studies. The microbiology is adequately well-understood, although it is not clear whether heterotrophic or autotrophic NRB play the more important role. This may vary from oil field to oil field. Nonetheless, from the results in the literature, nitrate amendment (and in some cases phosphate or organic acid amendment) stimulates NRB in the oil field waters, and there appears to be little need to add an inoculum of NRB. Nitrate has replaced biocides in some of the oil fields in the North Sea, and the results have been very positive. Besides controlling sulfide levels, there is also preliminary evidence that corrosion rates are reduced [75]. In addition,
337
there are plans to use nitrate in the Gulf of Mexico when sea water injection begins in the near future (Stephen Maxwell, Commercial Microbiology Inc., personal communication). In contrast, there is little or no use of nitrate in landbased souring oil fields in North America. It is now very clear that land-based oil field operators should seriously consider using this proven biotechnology to control, and possibly eliminate, microbially-induced souring and the problems associated with H2S formation. REFERENCES [I] [2] [3] [4] [5] [6] [7] [8] [9] [10] [II] [12] [13] [14] [15] [ 16] [17] [18]
G.B. Farquhar, Corros. Prev. Contr., 45(2) (1998) 51. G.E. Jermeman, D. Gevertz and M. Wright, In Proceedings of the Third International Petroleum Environmental Conference, Vol. II, Albuquerque, NM, 1996, pp. 693-704. D.O. Hitzman and G.T. Sperl, Paper SPE 27752, presented at the 9th Symposium on Improved Oil Recovery Tulsa, Oklahoma, USA, April 17-20, 1994. B.P. Tissot and D.H. Welte, Petroleum Formation and Occurrence 2nd edition, Springer-Verlag, Berlin, 1984. R.C. Selley, Elements of Petroleum Geology, 2nd edition, Academic Press, NY, USA, 1998,296. H.G. Machel, Sediment. Geol, 140 (2001) 143. H.G. Machel and J.M. Foght, In R.E. Riding and S.M. Awramic (eds.), Microbial Sediments, Springer-Verlag, Berlin, 2000, pp. 105-120. B.K. Manzano, M.G. Fowler and H.G. Machel, Org. Geochem., 27 (1997) 507. WJ. Cochrane, P.S. Jones, P.F. Sanders, D.M. Holt and M.J. Mosley, Presented at the SPE European Petroleum Conference, London UK, October 1988, SPE paper 18368, Society of Petroleum Engineers, Richardson, Texas, USA. L.C. Frazer and J.D. Boiling, Presented at the International Arctic Technology Conference, Anchorage Alaska, May, 1991, SPE paper 22105, Society of Petroleum Engineers, Richardson, Texas, USA. J. Larsen, Paper 02025 Proceedings of the NACE Expo 2002 Annual Conference and Exposition, Denver, Colorado, USA, April 7-11, 2002. M.J. Mclneraey, K.L. Sublette, V.K. Bhupathiraju, J.D. Coates and R.M. Knapp, In E.T. Premuzic and A. Woodhead. (eds.), Microbial Enhancement of Oil Recovery-Recent Advances, Elsevier Science Publishing, BV. Amsterdam, 1993, pp. 363-371. G.E. Jenneman, P.D. Moffitt, G.A. Bala and R.H. Webb, SPE Prod. Facil., 63 (1999) 219. I. Davidova, M.S. Hicks, P.M. Fedorak and J.M. Suflita, J. Ind. Microbiol. Biotechnol., 27 (2001) 80. R.E. Eckford and P.M. Fedorak, J. Ind. Microbiol. Biotechnol., 29 (2002) 243. American Petroleum Institute, Recommended Practices for Oil and Gas Producing and Gas Processing Plant Operations Involving Hydrogen Sulfide, 2nd edition, API Recommended Practice 55, Washington DC, USA, 1995. P.G. Stecher, Hydrogen Sulfide Removal Processes, Noyes Data Corp., New Jersey, USA, 1972, p. 1. R. Cord-Ruwisch, W. Kleinitz and F. Widdel, J. Petrol. Techno!., 39 (1987) 97.
338
[19] R.N. Tuttle and R.D. Kane (eds.), H2S Corrosion in Oil and Gas Production - A Compilation of Classic Papers, National Association of Corrosion Engineers, Houston, Texas, USA, 1981. pp. 1018-1038. [20] A. Ikeda and M. Kowaka, Chem. Econ. Eng. Rev., 10 (1978) 12. [21] T.A. Bertness, G.V. Chilingarian and M. Al-Basson, In G.V. Chilingarian, J.O. Robertson, Jr. and S. Kumar (eds.), Surface Operations in Petroleum Production II, Elsevier Science Publishing Co. Inc., New York, USA, 1989, pp. 283-317. [22] I.B. Beech, In G. Bitton (ed.), Encyclopedia of Environmental Microbiology, Wiley and Sons, New York, USA, 2002, pp. 465-475. [23] W.P. Iverson and G.J. Olson, In R.M. Atlas (ed.), Petroleum Microbiology, Macmillan Publishing Company, New York, USA, 1984, pp. 633-683. [24] F.A. Giuliano (ed.), Introduction to Gas and Oil Technology, 3rd edition, Prentice Hall, Inc., Englewood Cliffs, New Jersey, USA, 1989. [25] A.G. Collins and C.C. Wright, In E.C. Donaldson, G.V. Chilingarian and T.F. Yen. (eds.), Enhanced Oil Recovery I: Fundamentals and Analysis, Elsevier Science Publishing Co. Inc., New York, USA, 1985, pp. 151-221. [26] C.C. Wright and G.V. Chilingarian, In G.V. Chilingarian, J.O. Robertson, Jr. and S. Kumar (eds.), Surface Operations in Petroleum Production II, Elsevier Science Publishing Co., Inc. New York, USA, 1989, pp. 319-372. [27] S.C. Rose, J.F. Buckwalter and R.J. Woodhall (eds.), The Design Engineering Aspects of Waterflooding, Society of Petroleum Engineers, Richardson, Texas, USA, 1989. [28] T.R. Jack, In F.T. Premuzic and A. Woodhead (eds.), Microbial Enhancement of Oil Recovery - Recent Advances, Elsevier Science Publishing, BV. Amsterdam, 1993, pp. 7-16. [29] J.M. Akagi, In L.L. Barton (ed.), Sulfate-Reducing Bacteria, Plenum Publishing Corp., New York, USA, 1995, pp. 89-112. [30] J.R. Postgate, The Sulphate-Reducing Bacteria, Cambridge University Press, Cambridge, UK. 1979. [31] J.R. Postgate, The Sulphate-Reducing Bacteria, 2nd edition, Cambridge University Press, Cambridge, UK, 1984. [32] J.M. Odom and R. Singleton, Jr. (eds.), The Sulfate-Reducing Bacteria: Contemporary Perspectives, Springer-Verlag, New York, USA, 1992. [33] L.L. Barton (ed.), Sulfate-Reducing Bacteria, Plenum Publishing Corp., New York, USA, 1995. [34] H.F. Castro, N.H. Williams and A. Orgam, FEMS Microbiol. Ecol., 31 (2000) 1. [35] H. Cypionka, In L.L. Barton (ed.), Sulfate-Reducing Bacteria, Plenum Publishing Corp., New York, USA, 1995, pp. 151-184. [36] N. Pfennig, F. Widdel and H.G. Triiper, In M.P. Starr, H. Stolp, H.G. Triiper, A. Balows and H.G. Schegel (eds.), The Prokaryotes, A Handbook of Habitats, Isolation, and Identification of Bacteria, Vol. 1, Springer-Verlag, Berlin, 1981, pp. 926-947. [37] L.L. Barton and F.A. Tomei, In L.L. Barton (ed.), Sulfate-Reducing Bacteria, Plenum Publishing Corp., New York, USA, 1995, pp. 1-32. [38] T.A. Hansen, In J.M. Odom and R. Singleton, Jr. (eds.), The Sulfate-Reducing Bacteria: Contemporary Perspectives, Springer-Verlag, New York, USA, 1993, pp. 21-40. [39] F. Widdel and R. Rabus, Curr. Opin. Biotechnol., 12 (2001) 259. [40] J. Heider, A.M. Spormann, H.R. Beller and F. Widdel, FEMS Microbiol. Rev., 22 (1999)459. [41] D. White, The Physiology and Biochemistry of Prokaryotes, Oxford University Press Inc., New York, USA, 1995.
339
[42] A. Azadpour, L.R. Brown and A.A. Vadie, J. Ind. Microbiol., 16 (1996) 263. [43] G.D. Faugue, In L.L. Barton (ed.), Sulfate-Reducing Bacteria, Plenum Publishing Corp., New York, USA, 1995, pp. 217-242. [44] D.E. Brink, I. Vance and D.C. White, Appl. Microbiol. Biotechnol., 42 (1994) 469. [45] J.T. Rosnes, T. Torsvik and T. Lien, Appl. Environ. Microbiol., 57 (1991) 2302. [46] M. Magot, B. Ollivier and B.K.C. Patel, Antonie van Leeuwenhoek, 77 (2000) 103. [47] J. Boivin, Mater. Perform., 34(2) (1995) 65. [48] T.R. Jack and D.W.S Westlake, In L.L. Barton (ed.), Sulfate-Reducing Bacteria, Plenum Publishing Corp., New York, USA, 1995, pp. 265-292. [49] S. Myhr, B.-L. Lilleb0, E. Sunde, J. Breeder, and T. Torsvik, Appl. Microbiol. Biotechnol., 58 (2002) 400. [50] J. Tiedje, In A.J.B. Zehnder (ed.), Biology of Anaerobic Microorganisms, Wiley, New York, USA, 1988, pp. 179-244. [51] A.J. Telang, S. Ebert, J.M. Foght, D.W.S. Westlake, G.E. Jenneman, D. Gevertz and G. Voordouw, Appl. Environ. Microbiol., 63 (1997) 1785. [52] S. Myhr and T. Torsvik, Int. J. Syst. Bacteriol., 50 (2000) 1611-1619. [53] R.G.L. McCready, W.D. Gould, and R.W. Barendregt, Can. J. Microbiol., 29 (1983) 231. [54] G.J. Mitchell, J.G. Jones and J.A. Cole, Arch. Microbiol., 144 (1986) 35. [55] H.-J. Seitz and H. Cypionka, Arch. Microbiol., 146 (1986) 63. [56] T. Dalsgaard and F. Bak, Appl. Environ. Microbiol., 60 (1994) 291. [57] K.L. Sublette and M.E. Woolsey, Biotechnol. Bioeng, 34 (1989) 565. [58] K.L. Sublette, MJ. Mclnerney, A.D. Montgomery and V. Bhupathiraju, In C.N. Alpers and D. W. Blowes (eds.), Environmental Geochemistry of Sulfide Oxidation, American Chemical Society, Washington, DC, USA, 1994, pp. 68-78. [59] M.J. Mclnerney, V.K. Bhupathiraju and K.L. Sublette, J. Ind. Microbiol., 11 (1992) 53. [60] M.J. Mclnerney, N.Q. Wofford and K.L. Sublette, Appl. Biochem. Biotechnol., 57/58 (1996) 933. [61] K.L. Sublette, D.E. Morse and K.T. Raterman, Presented at the 68th Annual Technical Conference and Exhibition of the Society of Petroleum Engineers Houston, Texas, SPE paper 26396, Society of Petroleum Engineers, Richardson, Texas, USA, October 3-6, 1993. [62] D. Gevertz, A.J. Telang, G. Voordouw and G.E. Jenneman, Appl. Environ. Microbiol., 66(2000)2491. [63] F.A. Rainey, D.P. Kelly, E. Stackebrandt, J. Burghardt, A. Hiraishi, Y. Katayama and A.P. Wood, Int. J. Syst. Bacteriol., 49 (1999) 645. [64] W. Ludwig, G. Mittenhuber and C.G. Friedrich, Int. J. Syst. Bacteriol., 43 (1993) 363. [65] L.A. Robertson and J.G. Kuenen, J. Gen. Microbiol., 129 (1983) 2847. [66] W.W. Carothers and Y.K. Kharaka, Am. Assoc. Petrol. Geol. Bull, 62 (1978) 2441. [67] T. Barth, Appl. Geochem, 6 (1991) 1. [68] E.A. Greene, C. Hubert, M.. Nemati, G.E. Jenneman and G. Voordouw, Environ. Microbiol, 5 (2003) 607. [69] P.A. Loka Bharatm, S. Nair and D. Chandramohan. J. Mar. Biotechnol, 5 (1997) 172. [70] J.P. Adkins, L.A. Cornell and R.S. Tanner, Geomicrobiol. J, 10 (1992) 87. [71] D. Gevertz , G.E. Jennemen, S. Zimmerman and J. Stevens, In R. Bryant (ed.), Proceedings of the Fifth International Conference on Microbial Enhanced Oil Recovery and Related Biotechnology for Solving Environmental Problems, Richardson, Texas, USA, 1995, pp. 295-309. [72] A.J. Telang, G.E. Jenneman and G. Voordouw, Can. J. Microbiol, 45 (1999) 905.
340
[73] G. Voordouw, J.K. Voordouw, R.R. Karkhoff-Schweizer, P.M. Fedorak and D.W.S. Westlake, Appl. Environ. Microbiol., 57 (1991) 3070. [74] R.E. Eckford and P.M. Fedorak, J. Ind. Microbiol. BiotechnoL, 29 (2002) 83. [75] T. Thorstenson, G. Bodtker, B.-L.P Lillebo, T. Torsvik, E. Sunde and J. Beeder, Paper 02033 Proceedings of the NACE Expo 2002 Annual Conference and Exposition, Denver, Colorado, USA, April 7-11, 2002. [76] E.G. Beauchamp, J.T. Trevors and J.W. Paul, Adv. Soil Sci., 10 (1989) 113. [77] R.K. Thauer, K. Jungermann and K. Decker, Bacteriol. Rev., 41 (1977) 100. [78] G.E. Jenneman, M.J. Mclnerney and R.M, Knapp, Appl. Environ. Microbiol., 51 (1986) 1205. [79] G.E. Jenneman, A.D. Montgomery and M.J. Mclnerney, Appl. Environ. Microbiol., 51 (1986)776. [80] H.-E. Jacob, In J.R.. Norris and D.W. Ribbons (eds.), Methods in Microbiology, Vol. 2, Academic Press Inc., New York, USA, 1970, 93-123. [81] S. Moura, C. Bursakov and J.J.G. Moura, Anaerobes, 3 (1997) 279. [82] W.T. Carpenter, Water Works Sewerage, 79 (1932) 175. [83] W.A. Lawrance, Sewage Ind. Wastes, 22 (1950) 820. [84] R.A. Poduska and B.D. Anderson, J. Water Pollut. Control Fed., 53 (1981) 299. [85] K.L. Londry and J.M. Suflita, J. Ind. Microbiol. BiotechnoL, 22 (1999) 582. [86] M.A. Reinsel, J.T. Sears, P.S. Stewart and M.J. Mclnerney, J. Ind. Microbiol., 17 (1996) 128. [87] M. Wright, G.E. Jenneman and D. Gevertz, Proceedings of the 4th International Petroleum Environmental Conference: Environmental Issues and Solutions in Exploration, Production and Refining. San Antonio, Texas, USA, (on CD-ROM), 1997. [88] G.E. Jenneman, P.D. Moffitt, G.A. Bala and R.H. Webb, Paper SPE 38768, presented at the 1997 SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA, October 5-8, 1997. [89] M. Nemati, G.E. Jennenman and G. Voordouw, BiotechnoL Prog., 17 (2001) 852. [90] R.M. Lacatena, S. Lunetto, A. Robertiello, M. Marzorati and F. de Ferra, Presented at 2nd International Conference on Petroleum Biotechnology, Mexico City, November 5-7, 2003. [91] M. Nemati, G.E. Jennenman and G. Voordouw, BiotechnoL Bioeng., 74 (2001) 424. [92] B. Herbert, Lecture presented at Reservoir Microbiology Forum 9: Use of Nitrates to Control Bacterial Problems, September 9, 2003, London, UK.
Studies in Surface Science and Catalysis 151 R. Vazquez-Duhalt and R. Quintero-Ramirez (Editors) © 2004 Elsevier B .V. All rights reserved.
341
Chapter 12
Regulation of toluene catabolic pathways and toluene efflux pump expression in bacteria of the genus Pseudomonas J.L. Ramos, E. Duque, M.T. Gallegos, A. Segura and S. Marques Estacion Experimental del Zaidin, CSIC, C / Profesor Albareda 1, 18008 Granada, Spain
1. TOLUENE EXTRUSION AND DEGRADATION PATHWAYS INFLUENCE SURVIVAL IN PSEUDOMONADS Aromatic hydrocarbons have been present in the environment for millions of years since they are the product of the natural pyrolysis of organic material [1] and are widely distributed in natural environments. One ring aromatic compounds such as benzene, xylenes, ethylbenzene and toluene, which have a logPow (logarithm of its partition coefficient in «-octanol and water) between 2.5 and 3.5, are toxic for microorganisms and other living cells because they partition preferentially in the cytoplasmic membrane, disorganizing its structure and impairing vital functions [2]. The toxicity of these compounds depends not only on the inherent toxicity of the solvent but also on the intrinsic tolerance of the species and strains. Because living organisms have been in contact with these chemicals through long evolutionary periods of time, it is not surprising that microbes have developed the capability to degrade them. Many of the aromatic compound-degrading organisms are bacteria that belong to the Pseudomonadaceae. All Pseudomonas strains that use toluene as a carbon source have a series of mechanisms that allow them to cope with the stress imposed by toluene itself. Nonetheless, most Pseudomonas strains are highly sensitive to aromatic hydrocarbons such as toluene (logPow 2.5), styrene (logPow 3.6) orp-xylene (logPow 3.2); however, independent laboratories have isolated Pseudomonas putida strains tolerant to these toxic compounds [3-7] . A common theme in toluene tolerance in Pseudomonas is the change in cis/trans isomerization of unsaturated fatty acids [8]. The increase in trans isomers (which are directly synthesized from the cis isomers with no shift in the
342
position of the double bond) causes rigidification of the membrane to counteract the increase in membrane fluidity caused by the organic solvent [9-12]. These changes probably represent a first response that allows the cells to prepare for the de novo biosynthesis of other components involved in tolerance towards organic solvents. One such response was thought to be the metabolism of organic solvents, but some of the tolerant strains are not able to degrade toluene. For example, P. putida S12 is a toluene-tolerant strain that cannot degrade it [13], and a P. putida DOT-TIE mutant deficient in the tod (toluenedioxygenase) degradation pathway is as tolerant as the wild type to a sudden toluene shock [14]. These two observations suggest that degradation of the toxic compound is not a key factor in solvent tolerance. All the efflux pumps for organic solvents identified so far in gramnegative bacteria belong to the Resistance-Nodulation-Cell Division (RND) family. The functioning of these efflux pumps seems to be coupled to the proton motive-force via the TonB system, although the intimate mechanism of energy transfer remains elusive [15, 16]. Bacterial RND efflux pumps work together with a membrane fusion protein (MFP) and an outer membrane protein (OMP). These three components form a structure that expand both the inner and outer membranes [17-19]. The efflux pump transporter AcrB of the E. coli RND multidrug efflux system AcrAB-TolC was recently crystalized. The AcrB component is as a trimer with a 50-A transmembrane region and a 70-A part located in the periplasm that is thought to be involved in substrate recognition [18, 20]. The crystal structure of the outer membrane protein TolC -which forms a trimeric channel that penetrates the periplasm and contacts the efflux pump transporter has also been reported [17]. Finally, a lipoprotein anchored to the inner membrane which expands into the periplasmic space may serve as a bracket for the other two components [19, 21]. This structural organization allows substrates to be extruded into the external medium by passing the periplasmic space [17, 18, 22]. In spite of its toxicity and thermodynamic stability, toluene can be degraded by many microbes using a common strategy to weaken the aromatic ring prior to its cleavage: the introduction of two hydroxyl groups that destabilize the chemically stable resonant structure. However, bacteria have developed different molecular mechanisms to produce this dihydroxylated compound. So far, five aerobic pathways have been described for the bacterial degradation of toluene, all of them leading to catechol, methylcatechols or protocatechuate: the so-called TOL, TOD, TMO, TOM, and TBU pathways. However, only three of them, the TOL, TOD and TMO pathways, have been found in Pseudomonas species. The enzymes that carry out the first reaction, i.e. the direct insertion of one or two oxygen atoms in the toluene molecule, largely determine the pathway that is followed for degradation. The key steps involved in the pathways are briefly described below, and they are summarized in Fig. 1.
343
Fig. 1. Pathways for the aerobic degradation of toluene.
1.1. Toluene degradation pathways The TOL pathway, coded by the archetypical plasmid pWWO [23, 24], is a very well characterized pathway for toluene degradation from a biochemical and genetic point of view. This pathway is composed of two segments, an upper and a lower pathway. Through the upper pathway, the methyl group of toluene is sequentially oxidized to render benzoate (Fig. 1). The first enzyme of this upper pathway is a toluene monooxygenase that oxidizes the methyl group of toluene to yield benzyl alcohol. Subsequent oxidation of the side chain is accomplished in two steps: first benzyl alcohol dehydrogenase renders benzaldehyde, which is further oxidized to benzoate by a benzaldehyde dehydrogenase. It is of interest to note that the enzymes of the upper pathway also accept as substrates the corresponding compounds substituted with a methyl group at the meta and/or para position and ethyl group at the meta position. These compounds are, therefore, oxidized to 3-, 4-methylbenzoate, 3,4dimethylbenzoate and 3-ethylbenzoate, respectively. The aromatic carboxylic acids are then further metabolized through the meto-cleavage pathway, in which the benzoate, alkylbenzoate(s) is(are) then oxidized and decarboxylated to produce the corresponding catechol(s) (Fig. 1).
344
These compounds undergo meta fission to yield 2-hydroxymuconic acid semialdehyde or the corresponding alkyl derivatives. Metabolism of the semialdehydes occurs via a branched pathway that rejoins at a common intermediate, 2-oxopent-4-enoate. The semialdehyde produced from w-toluate is hydrolyzed, whereas the semialdehyde from benzoate and p-methylbenzoate is metabolized via the oxalocrotonate branch, which involves at least three enzymatic steps [25]. 2-Oxopent-4-enoate is further converted in 2-oxo-4hydroxypentonate, which eventually renders Krebs cycle intermediates. For a more detailed description of the characteristics of the pathway enzymes, the reader is referred to earlier reviews and original papers [24, 26-30]. A second degradation pathway found in Pseudomonas is the so-called TOD pathway, which was first described in P. putida strain Fl [31]. In this pathway, probably the best known in terms of the biochemistry involved, the first step is carried out by a three-component enzyme complex, the toluene 2,3dioxygenase (TOD), which renders czs-toluene dihydrodiol (Fig. 1). This compound then undergoes dehydrogenation to yield 3-methylcatechol. Ethylbenzene is also a substrate of this enzyme being converted into 3ethylcatechol. The alkylcatechols are then the substrates for ring fission in the meta position in a set of reactions similar to those described for the TOL metacleavage pathway. Finally, a third pathway was described in P. mendocina KR1, where the first step in toluene metabolism is carried out by the toluene-4-monooxygenase (TMO), which hydroxylates the aromatic ring in the para position to render pcresol (Fig. 1 and [32, 33]). In the subsequent steps the methyl group is transformed by a methyl hydrolase, first rendering the alcohol and then the aldehyde derivative, which is finally oxidized to /?-hydroxybenzoate by a dehydrogenase. The ring is further oxidized to 3,4-dihydroxybenzoate, which is the substrate for ring cleavage in the ortho position to enter the P-ketoadipate pathway. As mentioned above, two additional pathways for toluene degradation with different initial reactions have been described in strains that were originally considered Pseudomonas: the so-called toluene-3-monooxygenase pathway (TBU) of Ralstonia picketti, where toluene was proposed to be first oxidized to ra-cresol and then to 3-methylcatechol [34]. However, recent work in Tom Wood's laboratory has suggested that toluene-3-monooxygenase indeed functions as a hydroxilating enzyme at the para position, and that 90% of toluene is oxidized to p-cresol, which is subsequently oxidized to yield 4methylcatechol (Fig. 1 and [35]). The second pathway, not present in Pseudomonas, is the one known as the toluene 2-monooxygenase pathway (TOM) of Burkholderia cepacia where the first oxygen is inserted in the ortho position to render o-cresol, which is then oxidized to 3-methylcatechol (Fig. 1). In addition, another pathway for the degradation of toluene was recently
345
described in P. stutzeri OX1, a strain able to degrade o-xylene through an initial step that involves two successive monooxygenations of the aromatic ring carried out by the same enzyme, toluene o-xylene monooxygenase (ToMO) [36-38]. This enzyme is interesting because of its broad substrate specificity and its relaxed regioselectivity, which make it able to hydroxylate more than one position of an aromatic substrate [39]. A common feature to all toluene pathways from different bacteria is that the genes involved in the different reactions are organized as operons, which are either independent for the different segments of the pathways (i.e., the TOL pathway) or transcribed as a single unit (i.e. the tod operon). In all cases, these pathways are under the control of regulatory mechanisms, which are ultimately modulated by toluene or intermediate substrates. Below we review the regulatory networks controlling the expression of the different pathways below and the mechanisms developed to regulate toluene efflux pumps.
2. REGULATION OF THE TOL PATHWAY In P. putida mt-2, the genetic information that determines growth on toluene, xylenes and other alkyl derivatives is encoded by the xyl operons of the TOL plasmid pWWO [24]. The xyl genes are organized in four transcriptional units: the upper and the meta operons and the xylS and xylR genes (Fig. 2). The upper operon xylUWCAMBN codes for the enzymes necessary for the oxidation of toluene to benzoate, whereas the meta-operon xylXYZLTEGFJQKIHencodes the enzymes for the oxidation of benzoate, the ensuing ring cleavage and the degradation to TCA cycle intermediates. The xylS and xylR genes, which are transcribed divergently, are located close to the meta operon 3' end, and their products regulate the expression of the meta and upper catabolic operons, respectively. This pathway is no doubt the most extensively characterized regulatory system among the aromatic degradation pathways, and, as a whole, the TOL regulatory network can be considered a paradigm of integrated transcriptional regulation in prokaryotes: there are two o54-dependent promoters, each with unique features, one regulator belonging to the NtrC, one to the AraC family of regulators, a o32/o38-dependent promoter, and several o70-dependent promoters, all of them under superimposed global control. 2.1. Overview of the regulatory network A scheme of the regulatory network that operates in the TOL pathway is presented in Fig. 2. The model summarizes experimental evidence collected since the 1970s in different laboratories to explain the expression of the pathway enzymes in the presence of toluene, benzoate or their derivatives. Two different regulatory circuits operate, depending on the nature of the aromatic compound present in the culture medium [40]. When cells are growing in the absence of
346
any aromatic substrate, the xylS gene is expressed at low levels from the a70dependent promoter Ps2, ensuring the presence of basal levels of XylS protein. This protein as such is not able to activate transcription. When a substrate of the meta pathway, e.g. 3-methylbenzoate, is present in the growth medium, the XylS protein interacts with it and becomes active to promote transcription from the Pm promoter, which controls expression of the meta pathway. Expression from Pm requires RNA polymerase with either a32 in the early exponential phase or a38 thereafter. The XylR protein, which regulates its own transcription from two o70-dependent promoters, is synthesized in sufficient amounts under all growth conditions. When a substrate of the upper pathway, e.g. toluene, is present in the culture medium, the binding of this effector to the protein triggers a series of molecular events that result in the activation of transcription from two o5 dependent promoters: Psl for the xylS gene, and Pu, which drives expression of the upper pathway. This latter activation requires the integration host factor (IHF). As a consequence of Psl activation, the XylS protein is overproduced, and even in the absence of a meta pathway effector, transcription from Pm occurs. The current knowledge of the molecular biology of each step on the regulatory pathway is reviewed in detail below.
Fig. 2. The TOL pathway regulatory network. Elliptical boxes indicate the inactive form of the regulatory proteins. Shaded square boxes indicate the active form of the regulatory proteins. Lines represent the connections between regulatory proteins and promoters, where (+) is activation of transcription and (-) is inhibition of transcription; GR, global regulation. The dotted line indicates transcription activation of overproduced XylS in the absence of effector. The sigma factor(s) involved in transcription initiation are indicated above each promoter. Aromatic substrates of the pathways that act as effectors of the regulatory proteins are indicated. The regulatory circuits are explained in the text.
347
2.2. Regulation of the meta pathway Two main players are responsible for the expression of the meta pathway: the tandem Pm promoter/RNA polymerase, and the tandem XylS/effector molecule. Extensive genetic data have been obtained to determine the architecture of Pm promoter and the mechanisms through which the regulator becomes active to promote transcription from Pm. As stated above, basal levels of the XylS protein are guaranteed under any growth conditions by the activity of the constitutive o70-dependent promoter Ps2 (Fig. 2). 2.2.1. The XylS regulator XylS belongs to the AraC/XylS family of transcriptional regulators, which includes at least 284 different proteins [41-44]. Members of this family present two domains: a 100 amino acid conserved domain involved in DNA binding (the C-terminal domain in most of the proteins of the family, including XylS), and a nonconserved domain (the N-terminal domain in XylS) involved in effector binding and dimerization. Interactions between XylS and its effector have been studied by analyzing the ability of the protein to activate transcription in the presence of a wide range of substituted benzoates, as well as by selecting XylS mutants with altered effector specificity [16, 43, 45, 46]. Recognition of ring substituents strongly depends on the position and nature of the chemical substituent, with meta being the most permissive position in the aromatic ring (-CH3, -CH2H5, and -OCH3 groups and F, Cl, Br and I atoms are permissible substituents), whereas positions ortho and para pose some restricitions to substituents (-CH3 and F and Cl atoms are allowed but not -C2H5 and I atoms) [45]. Although disubstitutions involving positions o- and m-, and m- and p-, are permissible, other combinations are usually non permissible, which suggests that interactions between the effector and the regulator are nonsymmetrical. Ramos et al. [48, 49] and Michan et al. [46] isolated and sequenced a series of mutant regulators able to recognize substituted benzoate effectors that are not recognized by the wild-type regulator. Key residues clustered in two noncontiguous segments in the N-teminus end of XylS. Mutations were found to be clustered at positions 37-45, 88-92, 151-155, and around residues 256 and 288. These finding suggest that the recognition pocket for XylS effectors may be composed of two or more noncontiguous segments of its primary sequence. Arg-41 seems to be a critical residue for interaction(s) with effectors, as changes in this position result in many different phenotypes. For example, XylSArg41Gly is a mutant regulator whose ability to recognize o- and p-methylbenzoate was lost, although it retained its capacity to be activated by w-methylbenzoate. Substitution of Arg41 by Leu resulted in a mutant that was unable to respond to benzoate effectors [46]. Therefore,
348
transition from the inactive to the active form may be mediated by effector binding. Recent studies further pinpointed residues Asp 137 and His 153 as crucial for interactions with the effector molecule [50]. In addition to influencing effector specificity, these two residues were shown to contact specific residues in the RNA polymerase a subunit carboxy terminal domain (aCTD) [50] . XylS mutants such as XylSArg41Cys, XylSPro37Gly XylSGly44Ser, XylSSer229Ile, XylSAsp274Val, and XylSAsp274Glu mediated transcription from Pm in the absence of effectors [46, 47]. These results support the hypothesis that XylS exists in vivo in a dynamic equilibrium between an inactive and an active form, with respect to transcriptional stimulation. Within the family, some regulators such as MarA are present in solution as monomers, whereas most of the members of the family are found as dimers [51-54]. XylS is likely active as a dimer and in vivo and in vitro assays have shown that Leu 193 and Leul94 in XylS play a crucial role in dimerization [55]. It is predicted that the DNA binding domain of XylS consists of seven ahelix units which fold to assemble two helix-turn-helix (HTH) motifs that interact with two neighboring major grooves on one face of the target DNA. Involvement of the XylS C-terminal domain in DNA binding was first predicted after the finding of mutations in this domain that rendered mutant regulators able to promote high transcription levels in the absence of effectors [47, 48]. Mutation analysis of the predicted conserved positions of the HTH motifs of XylS showed that the most conserved positions in the family seem to be essential to preserve the structure of this domain [56]. Deletion of the 209 Nterminal residues of XylS rendered a C-terminal domain-protein able to bind Pm promoter and, when overproduced, able to activate transcription in vivo to levels similar to those in the wild type protein. However, activity was clearly reduced when the C-terminal fragment was synthesized at physiological levels. As expected, the truncated protein was not responsive to effector-mediated control [57]. 2.2.2. The Pm promoter XylS-mediated transcription activation from Pm requires a DNA fragment extending to position -70 upstream from the transcription start site. The DNA in this region exhibits a 40° bend centered between positions -41 and -46 [58]. The XylS binding site in the Pm promoter was first defined through site-directed mutagenesis [59-63] and further confirmed by in vitro and in vivo footprint assays [60, 64, 65]. The XylS binding site in Pm consists of two directed repeats (5'-TGCAN6GGNTA-3') spanning positions -34 to -68, and overlapping the RNA polymerase biding site by 1 bp [58, 60]. This overlap with the RNA polymerase binding site is also observed in several other members of the family [66-68].
349
In vivo transcription from Pm is mediated by two different RNA polymerases, depending on the growth phase. In the early exponential phase, RNA polymerase with a32 is necessary for transcription, whereas in late exponential growth phase and in stationary phases, a38 is required [65, 69, 70]. Despite the alternation in RNA polymerase, the same transcription start site is detected along the growth curve, suggesting that the same promoter is used by both forms of RNA polymerase. Transcription of o32-dependent promoters requires the stabilization of this sigma factor, which takes place through a series of events known as heat-shock response [71]. In the TOL pathway regulatory network, aromatic effectors are required not only because of their direct role in XylS activation to promote transcription, but also to trigger the heat-shock response and provide the appropriate RNA polymerase for transcription in the exponential phase [70]. The direct involvement of the two sigma subunits in Pm transcription was further supported by the finding that a mutant Pm promoter with an altered XylS binding site, combined with the mutant regulator XylSGly44Ser, was able to overcome the requirement of a38 for transcription in the stationary phase [69]. Moreover, the mutants XylSAspl37Glu and XylSHisl53Gln were able to stimulate transcription from Pm in the absence of a38 [50]. 2.3. Regulation of the upper pathway In the presence of toluene or a substrate of the upper pathway, P. putida ensures the coordinated expression of the two catabolic operons, so that the aromatic compound is totally degraded to TCA intermediates. The key regulator in this process is XylR, which is responsible for the coordinated expression of the o54-dependent promoters Pu and Psl. Pu drives transcription of the upper pathway and Psl increases the synthesis of XylS, responsible for meta pathway expression (Fig. 2). 2.3.1. The XylR regulator XylR protein belongs to the NtrC family of enhancer-binding proteins (EBP) [72-74]. It contains the four distinctive domains of this family: i) An Nterminal A-domain responsible for signal reception, i.e., interaction with the effector molecule (see below), ii) the A-domain is linked to the central domain, called domain C by the short B-domain (Q-linker), iii) the C-domain is involved in ATP binding and hydrolysis, and plays a major role in the isomerization of the o54-dependent promoters from close to open complexes, and iv) the Cterminal D-domain contains the HTH motif for DNA binding. XylR is activated by aromatic compounds with a wide variety of substitutions such as alkyl groups of different length or oxidized intermediates of the toluene methyl group, such as benzyl alcohol, benzaldehyde and derivatives [75, 76]. Early evidence indicated that the A-domain was the signal
350
receptor from the environment and the direct sensor of the aromatic molecule. This was surmised from the ability of the protein to activate transcription from the Pu promoter in the presence of a wide range of toluene derivatives, and by experiments with XylR mutants with altered effector specificity [75, 76, 77, 78]. These data, obtained in the heterologous host E. coli, led to the conclusion that XylR was directly activated via interaction with the effector. XylR is closely related to the DmpR regulator for phenol degradation in Pseudomonas sp. CF600, which recognizes phenol and derivatives, but not toluene, as an effector [79]. Further evidence for the direct interaction of the A-domain of these proteins with the effector molecule came from the construction of a chimeric protein in which the receptor domain of DmpR was replaced by the corresponding domain of XylR, resulting in a hybrid regulator that responded to toluene for activation of the Vo promoter of the phenol degradation pathway [80]. DNA shuffling assays to create hybrid A-domains between DmpR and XylR confirmed that the residues 110 to 186 of both proteins were responsible for the effector profile of these regulators [80]. The A-domain operates as an intramolecular repressor of the central activating domain of the protein [81, 82]. In fact, a XylR derivative in which the A domain has been deleted is able to activate Pu in the absence of an aromatic effector. The truncated derivative of XylR depleted of the A domain and therefore unable to respond to effector-dependent modulation showed intrinsic ATP binding and hydrolysis activity, located in the central activation domain (C-domain). This activity was stimulated by the presence of a DNA fragment containing the native XylR binding site in Pu (UAS) [83]. Furthermore, binding of ATP to this truncated protein alone was able to induce conformational changes in the protein. Initially, a cyclic model to explain XylR activation of Pu was proposed by Perez-Martin and de Lorenzo [83], according to which ATP binding to the XylR central domain led to multimerization of the regulator bound to its UAS in Pu, followed by ATP hydrolysis. This in turn triggered a54dependent transcription initiation in Pu, allowing the system to return to its initial disassembled state [83]. Recently, Shingler and co-workers studied the analogous regulator DmpR, and suggested an alternative mechanism to explain effector-dependent activation of a54-dependent promoters. According to their model, DmpR dimers are activated after binding of the effector molecule to the A domain, followed by a conformational change that allows ATP binding to the central domain and oligomerization to a hexameric conformation, probably required to promote transcription initiation. Finally, ATP hydrolysis leads to dissociation of the hexameric structure and dissociation of the effector [83]. 2.3.2. The Pu promoter Pu promoter belongs to the class of promoters dependent on the alternative sigma factor o54 (Fig. 2), and shows the typical architectural
351
organization typical of the promoters of this class. A -12/-24 sequence is responsible for the recognition of o54-RNAP [84, 85], and upstream activator sequences (UAS) ensure XylR binding between positions -120 and -175 [86, 87]. An IHF binding site between these two elements (positions -52 to -79) [87] is required to bring the regulator into contact with the RNA polymerase by looping out the intermediate sequence. Direct evidence that IHF causes Pu to form an open bend was obtained by atomic force microscopy, where an angle of 123° was measured between the UAS and the RNA polymerase binding site [88]. However, Bertoni et al. [89] recently found that a second upstream element, reminiscent of the so-called a-CTD-binding UP elements of o70dependent promoters, was important in Pu recognition by a5 -RNAP, and that IHF-binding played an additional role in Pu. This role consisted of a5 -RNAP recruitment to its promoter, determined by the correct positioning of the UP-like element with respect to the -12/-24 binding site after IHF-dependent DNA bending [39]. That IHF-mediated o54-RNAP recruitment to Pu was reproduced in vitro with the XylR regulator acting from solution, i.e., in the absence of UAS [90]. These authors have shown that RNA polymerase binding is an important rate-limiting step in Pu activation, that could become crucial when the enzyme is present at a low concentration [90]. 2.4. Expression of the regulatory genes The level of the XylR and XylS proteins is finely modulated in vivo. This fine regulation takes place in the 300-bp intergenic region between the xylR and the xylS genes, which contains four promoters: the two a70-dependent tandem promoters of xylR, Prl and Pr2, divergent from the two that drive transcription of xylS, the a54-dependent promoter Psl and the a70-dependent promoter Ps2. The binding sites for the different regulatory proteins in this short region totally or partially overlap; thus XylR UASs in Psl partly cover the two RNA polymerase binding sites of Prl and Pr2. In addition, two sequences with different affinity for IHF are found which overlap the Psl -12/-24 RNA polymerase binding site and one of the UAS [91-93]. As a consequence, the levels of expression normally observed in the wild-type strain for each promoter are far below maximum values, suggesting the involvement of repressive element(s) in the maintenance of appropriate levels of expression. In fact, as expected from the promoter architecture described above, XylR strongly represses its own synthesis [75, 89, 94-97]. Activation of the Psl promoter and autoregulation of XylR expression seem to be the consequence of the binding of XylR to the UASs for Psl that overlap the Prl and Pr2 promoters. This is in agreement with the finding that XylR is consistently bound to target sequences [75]. Thus, xylR promoters may be subjected to two levels of repression depending on the mechanism of XylR activation discussed above: an ATP-independent level resulting from non-cooperative interaction of
352
nonactivated XylR with the Psl UAS [94], and an ATP-dependent repression level resulting from the cooperative oligomerization of activated XylR at the UAS in Psl [89]. As soon as the protein is activated and the UAS is strongly bound by the regulator, XylR expression is minimized, thus limiting the period of time during which the Ps 1 and Pu promoters of the TOL plasmid are in an activated state [89]. The role of IHF in Psl expression deserves special attention. Analysis of Psl activity in isogenic IHF-plus and minus backgrounds showed that in the presence of toluene, the highest levels of expression were achieved in the absence of IHF [97]. This may reflect a better access of either XylR to its binding site or of o54-RNA polymerase to the -12/-24 region of Psl, or both. On the other hand, it may be the consequence of structural hindrance, as the DNA bending induced by IHF bound to two sites may give rise to a highly ordered structure that restricts the access of regulatory proteins to the corresponding promoters. The high level of expression from Psl in the IHF-minus background in the presence of effectors contrasts with the diminished expression from the TOL plasmid Pu promoter for the upper pathway in an IHF-deficient background. The most noticeable difference between the two promoters is the position of the IHF binding site, which in Pu lies between the UASs and the 12/-24 box. In addition to affecting Psl expression, the close proximity of the regulatory sequences in the intergenic region results is a high expression level from Ps2 in the absence of a54, i.e., when RNA polymerase is unable to bind to the Psl promoter [97]. In general, the physiological consequence of this organization is that in the absence of any effector in the culture medium, Ps2, Prl and Pr2 promoters are slightly repressed. In the presence of toluene, activation of Psl causes a stronger repression of both xylR promoters. As a result, the level of XylR decreases at approximately 30 monomers per cell [98], which are apparently sufficient to promote high expression of both xylS and the upper pathway. Under these conditions the XylS protein is overproduced, which allows induction of expression from the Pm promoter even in the absence of meta pathway substrates. Therefore in the presence of toluene or a substituted derivative, both the upper and the meta pathways are coordinately expressed to optimize total degradation of the aromatic (Fig. 2). 2.5. Integration in the bacterial metabolism The expression of the TOL pathways is tightly regulated according to the carbon sources available for growth [99-104]. The regulation is exerted mainly at the level of the two o54-dependent promoters Pu and Psl, and was first observed as a delay in the induction of expression from these promoters when cells were induced in a rich complex medium [102, 103]. Because both promoters were silent in this medium during rapid exponential growth, and
353
expression appeared only at the end of the exponential phase, this behavior was named exponential silencing [105]. However, it is worth noting that exponential silencing is only observed in rich medium; in defined minimal media with succinate (for example) as a carbon source, expression of both Psl and Pu is observed immediately after induction [102, 103, 106]. Growth rate as a determinant of Pu and Psl expression was ruled out through a series of continuous culture experiments that compared different growth rates controlled by different limiting substrates. The results led to the conclusion that repressive conditions correlated with a high energy status of the cells [99, 100]. In other words, in all conditions tested where excess carbon was available, the system was repressed. However, when oxygen was the growth-limiting substrate, a situation where carbon was also present in excess, a certain degree of derepression was observed although activity never reached maximum values (Fig. 3).
Fig. 3. Integration of cell signals that lead to modulation of the expression of the Pm and Pu promoters.
354
In minimal medium batch cultures amended with casamino acids, carbon sources such as glucose, gluconate or a-ketoglutarate expression from Pu was inhibited. In general this phenomenon reflects the preferential and sequential use of the different carbon sources present in a mixture; hence it could be considered a typical case of catabolite repression. However, catabolite repression in Pseudomonas seems to be exerted through mechanisms that differ greatly from the classical CRP-dependent phenomenon observed in E. coli. Similar physiological regulations of other catabolic operons has been observed in Pseudomonas [79, 107]. The molecular basis for the observed repression of Pu and Ps 1 expression remains unknown. Several alternatives have been envisaged and the current picture is compatible with the partial involvement of different systems in the global regulation response. Originally, the phenomenon known as exponential silencing was shown to be due neither to a late activation of XylR by the aromatic effector nor to changes in the intracellular levels of IHF during growth [105]. However, recent findings obtained with in vivo UV laser footprint technology have shown that IHF occupancy of its target site in Pu increases upon entry into the stationary phase, in parallel with an increase in IHF concentration in the cell. Therefore, this could explain the increase in Pu activity with growth phase in batch cultures. Nevertheless, these results do not preclude the integration of physiological repressive signals through additional mechanisms [108]. The a54 factor of RNA polymerase has also been considered a possible target of global regulation. Overproduction of the a54 factor allowed Pu to partially overcome exponential silencing, although not carbon source-dependent repression [105]. Because a54 protein levels remained approximately constant during growth under physiological conditions [109], exponential silencing of Pu may be caused ultimately by changes in the activity of the sigma factor itself. The ATP-dependent physiological protease FtsH, a member of the so-called AAA family of ATPases responsible for the stability of various transcription factors such as a32 [110], has been shown to play a key role in Pu expression. FtsH is required for XylR-mediated Pu transcription in a process that is not related to XylR or IHF, but which is rather exerted through a mechanism that involves the loss of a54 activity (Fig. 3). In fact, overproduction of a54 restored about 60% of Pu activity in the absence of FtsH [111]. Furthermore, the overproduction of FtsH partially relieved exponential silencing of Pu expression. In this connection, the target of FtsH activity seems to be an additional factor that downregulates a54 post-translationally, or that hinders the contacts of o54RNAP with the promoter or with the activator. Interestingly, E. coli FtsH levels are controlled in response to physiological signals and its proteolytic function is stimulated by the proton-motive force [112].
355
Finally, exponential silencing and carbon source-dependent repression can be distinguished at a genetic level. The rpoN gene, which codes for the o54 sigma factor required for Pu activity, is the first gene of an operon found in gram-negative bacteria that includes 4 additional ORFs. Two genes of this cluster, ptsN and ptsO, code for two proteins, IIANtr and NPr respectively that show similarity to phosphotransferases belonging to the phosphoenol pyruvate: sugar phosphotransferase system (PTS) family. To understand the putative role of the ORFs in this cluster in the global control of Pu, knock-out mutants were generated and analyzed. A mutant in the ptsN gene (which encodes IIANtr) relieves C source inhibition, but not the exponential silencing of Pu [113]. The ptsO gene together with ptsN operates in Pu regulation, where phosphorylation of the pteOencoded protein NPr is necessary for the normal response of Pu to glucose [114, 115]. NPr probably modulates IIANtr activity, promoting its dephosphorylation. This increases the concentration of unphosphorylated IIANtr, and as a result inhibition of Pu disappears. Interestingly, a site-directed ptsN mutant in the conserved phospho-acceptor His-68 residue made Pu unresponsive to the presence or absence of glucose, thus supporting the notion that phosphorylation of IIANtr mediates the C source inhibition of the promoter [116]. The observations reported above suggest that the global regulation of TOL pathway expression responds to several complex mechanisms that act at the level of o54-dependent promoters (Fig. 3). 3. REGULATION OF THE TOD PATHWAY Pseudomonas putida strains Fl and DOT-TIE use toluene and ethylbenzene as the sole carbon source [6, 117, 118]. The catabolic genes for the complete conversion of toluene/ethylbenzene to TCA cycle intermediates are clustered in a single unit, the tod operon, as todXFClC2BADEGIH [14, 119-123] and these genes are coordinately induced by toluene/ethylbenzene (Fig. 4) [12, 124]. Wang and coworkers [121] and Mosqueda and coworkers [14] identified a single promoter upstream from the todX gene, whose -10 and -35 regions showed homology with P. putida o70-dependent promoters [125]. Expression of the tod catabolic operon is regulated by todST gene products, which are located as a separate transcriptional unit downstream of todH, the last gene of the operon. Translational coupling between todS and todT ensures the balanced transcription of both genes [119]. TodS and TodT proteins belong to the family of two-component signal transduction systems. The regulation mechanism of two-component control systems is based on a histidine-aspartate phosphorelay circuit working between the two components. One of them is a sensor that autophosphorylates in response to an external signal, and the other one is the socalled response regulator, which receives the phosphate from the former and
356
activates transcription. TodS, the 108-kDa sensor protein, is a member of the hybrid class of histidine kinases and possesses multiple protein domains. The N terminus of TodS contains a motif characteristic of the basic region leucine zirjper (bZIP), that consists of a region with several basic residues which probably contact DNA, and an adjacent region containing a heptad repeat of leucine, the leucine zipper. Indirect evidence suggests that in TodS, the leucine zipper mediates dimerization, which is required for DNA binding [119]. A duplicated histidine kinase motif, each element of which is characterized by five short sequence blocks that are highly conserved, flanks a receiver domain of the response regulator located at the center of the protein adjacent to one set of a PAS domain, known as a signal sensor and found in various redox, light and hydrocarbon sensor proteins [126, 127]. The todT gene encodes a 206-residue protein with an estimated mass of 23 kDa. Analysis of the amino acid sequence of TodT shows significant similarity with response regulators of two-component signal transduction systems [128, 129]. TodT consists of an N-terminal receiver domain to accept the phosphoryl group from TodS, and a helix-turn-helix (HTH) DNA-binding domain. The TodT protein was shown to specifically bind to the todX promoter region at a 6-bp inverted repeat located 105 bp upstream from the transcription start site, and known as the todbox [119].
Fig. 4. Organization of the tod genes and its regulatory circuit. Top. The tod genes are organized in two operons expressed from sigma-70 dependent promoters upstream from todX and todS. The TodS protein (O) is synthesized in an active form that in the presence of toluene phosphorylates TodT ( • ) , which functions as the activator of the todX...
357
This two-component signal transduction system positively regulates the tod operon [119]. TodS is predicted to function as a hybrid kinase that uses ATP to autophosphorylate a specific histidine residue in response to toluene. TodT as a response regulator probably receives the phosphate at a conserved aspartate (Asp-56) and then mediates transcription activation at the todX promoter. Although biochemical evidence supporting this model is consistent [119], in vivo studies confirming the role of each member of the tandem are scarce. Furthermore, the physiological behavior of this system under different growth conditions and the precise mechanism through which the presence of toluene triggers changes in tod gene expression in vivo are not completely understood. On the other hand, the role of TodS as a sensor for directly detecting inducers has not been clearly demonstrated, nor has the role of these proteins in selfregulation been clarified. Expression from the todX promoter occurred in response to toluene, ethylbenzene, styrene, xylenes and other aromatic hydrocarbons, although the greatest level of expression was obtained with toluene. Expression from the todS gene was constitutive regardless of which aromatic was tested [14]. It is interesting to note that both TodS and TodT proteins are required for chemotaxis to aromatic hydrocarbons in P. putida Fl [130]. This observation indicates that both catabolism and chemotaxis are coordinately regulated at the transcriptional level. 4. REGULATION OF THE TOLUENE-4-MONOOXYGENASE PATHWAY Genes involved in toluene degradation in Pseudomonas mendocina KR1 are organized in an unusual manner: the five proteins that make up the multicomponent enzyme toluene-4-monooxygenase, which carries out the primary oxidation of toluene to />-cresol (Fig. 5), are coded by the tmoXABCDEF gene cluster [33, 131], where tmoX is homologous to the outer membrane protein that codes todX from P. putida Fl [132]. Complete toluene oxidation to TCA cycle intermediates requires another operon, pcuCAXB for pcresol oxidation to ^-hydroxybenzoate, and the pobA gene, which presents two alleles, pobAl and pobA2, for the further transformation of this compound into protocatechuate. The pcuCAXB and pobA genes are not clustered with the tmo genes on the P. mendocina chromosome. These operons and the pobA genes are also independently regulated [133]. The regulatory elements involved in the toluene degradation pathway in Pseudomonas mendocina KR1 are summarized in Fig. 5.
358
Fig. 5. Regulation of toluene degradation in P. mendocina KR1. The upper part of the figure summarizes the oxidation of toluene to protocatechuate and the genes needed for each reaction. The lower part shows a scheme of the cluster of tmolpculpob genes in this strain. tmoABCDEF, toluene-4-monooxygenase genes; c, putative cytochrome c gene; pcuRCAXB, /?-cresol utilization genes; pobRl and pobAl, regulator and p-hydroxybenzoate hydroxylase, respectively; tmoST, two-component signal transduction system; p-HBOH, />-hydroxybenzyl alcohol; />-HBHO, p-hydroxybenzyl aldehyde; p-HBA, p-hydroxybenzoate; PCA, protocatechuate. Reproduced with permission from [132].
4.1. Regulation of tmo operon expression The transcription initiation point from the tmo operon has been mapped and the sequence upstream has revealed strong identity with the promoter of the tod operon of P. putida Fl, including an inverted repeat located at position -100 and an almost identical P. putida F1 tod box. This suggested the involvement of a regulatory system similar to TodS-TodT for the transcriptional control of the P. mendocina toluene degradation pathway. In fact, a novel two-component signal transduction system was recently described in P. mendocina KR1 [132]. Transcription from the VtmoX promoter, which directs the expression of the tmoXABCDEF gene operon, is induced in the presence of toluene or/>-cresol by a two-component system made up of TmoS and TmoT, which are 83% and 85% identical, respectively, to the TodS and TodT proteins described above. Furthermore, transcription from P,mox and P,O
359
tmoST as primary regulators. Surprisingly, in P. mendocina the tmoST genes are not located close to their target operon tmoXABCDE, as is the case in P. putida Fl, but are found approximately 700 bp downstream from the pobA 1 gene forphydroxybenzoate hydroxylase, in a different region in the chromosome. It is interesting to note that truncated sequences showing homology to transposases are found downstream from tmoST and upstream from the tmo operon (RamosGonzalez, M.I., unpublished). In fact, early mobilization experiments found good linkage between the pcu and pob operons together with their corresponding regulatory genes, but no linkage between this cluster and the tmoXABCDE operon [133]. Sequencing of the tmoST surrounding region recently confirmed this distribution [132]. 4.2. Regulation of the lower segments of the toluene degradation pathway In P. mendocina KR1, the further degradation of />-cresol to TCA cycle intermediates requires the catabolic operons pcuCAXB, pobA, and the gene necessary for ring opening and subsequent degradation. The pcuCAXB operon is required to transform the/?-cresol produced from toluene by toluene monooxygenase into /?-hydroxybenzoate. The pcuCAXB operon is regulated by the divergently transcribed pcuR gene, which belongs to the NtrC family. Expression analysis of the pcuCAXB operon using a reporter gene fused to the promoter showed that only substrates of the pathway, such as p-cresol, p-hydroxybenzyl alcohol or p-hydroxybenzyl aldehyde, were effectors of PcuR. Neither toluene nor />-hydroxybenzoate was able to induce expression of the pathway [134]. The pobA gene codes for/>-hydroxybenzoate hydroxylase, which converts /?-hydroxybenzoate into protocatechuate, the substrate for ring fission [135, 136]. It has been shown that in this strain, expression of the pobA 1 gene is under the control of the divergently transcribed pobRl gene. However, no data are available on the specific molecular mechanisms involved in this process. 5. OVERVIEW OF THE REGULATION OF THE TBU AND TOM PATHWAYS IN RALSTONIA PICKETTII AND BURKHOLDERIA CEPACIA. Ralstonia pickettii PK01 (formerly Pseudomonas pickettii PK01) is able to grow on toluene, phenol, and benzene as the sole carbon and energy source [34]. The pathway responsible for the degradation of these compounds to TCA cycle intermediates is called the toluene-3 -monooxygenase pathway (also tbu pathway for toluene benzene utilization) (Fig. 1). The genes that encode enzymes for the pathway are grouped in three operons. The tbuAlUBVA2C and tbuT operon encodes the initial toluene-p-monooxygenase and the transcriptional activator TbuT [137]. The tbuD operon encodes phenol/cresol hydroxylase [138, 139],
360
and the tbuWEFGKIHJ operon encodes enzymes of the /weta-cleavage pathway for the conversion of catechol and methylcatechols to tricarboxylic acid cycle intermediates [140]. In addition, tbuX codes for a putative facilitator of toluene entry into the cell [141], and is located immediately downstream from tbuT. In turn, TbuT, a member of NtrC family of transcriptional activators, controls transcription of each of these operons in response to aromatic effector compounds [137]. TbuT is activated by aromatic effectors (toluene, benzene and ethylbenzene) and trichloroethylene. Expression of tbuT auto-regulated, as suggested by the finding that it is linked to the expression of the tbuAlUBVA2C operon by read-through transcription of tbuT from the toluene-/>monooxygenase promoter. As a result, transcription of tbuT is low when the toluene-p-monooxygenase operon is not induced and high when expression of tbuAlUBVA2C is induced by toluene. Thus, the toluene-/>-monooxygenase promoter drives the cascade expression of both the toluene-p-monooxygenase operon and tbuT, resulting in a positive feedback circuit [137]. Upstream from the o54-dependent toluene-p-monooxygenase promoter (P/&«AI), a DNA region with dyad symmetry may serve as the TbuT-binding site [137, 142]. Two additional regulatory genes, TbuS and TbuR, are located upstream from tbuD. In the absence of effectors, TbuS represses transcription of tbu WEFGKIHJ. The current view of the regulatory mechanism suggests that in the presence of the effectors phenol or w-cresol, these compounds interact with TbuS and the effector-TbuS complex acts as a transcription activator of tbuWEFGKIHJ. In addition, the effectors that interact with TbuR form complexes able to activate transcription oftbuD [34]. Burkholderia cepacia G4, formerly Pseudomonas cepacia, degrades toluene through a unique initial step that involves toluene-o?t/j0-monooxygenase (TOM) (Fig. 1). This strain also contains a catechol 2,3-dioxygenase for the meta-cleavage of methylcatechol. Because of its unique substrate specificity, the biochemistry of the TOM enzyme has been studied extensively. However, little is known about the genetics and regulation of the pathway. The genes responsible for this pathway are located in a 70-100 kb megaplasmid present in this strain. It has been shown that expression of the pathway is constitutive [143]. Burkholderia sp. strain JS150 contains a plasmid that carries the genes encoding for a toluene-2-monooxygenase clustered in the operon tbmABCDEF and its NtrC-like regulator coding gene tmbR. Additionally, a toluene-4monooxygenase activity was assigned to an independent region of the same plasmid, which was shown to be regulated by TbmR [144]. Finally, crossactivation between toluene-3-monooxygenase and toluene-2-monoxygenase by regulators TbuT and TbmR has been reported [142].
361
6. TRANSCRIPTIONAL REGULATION OF THE ORGANIC SOLVENT EFFLUX PUMPS IN Pseudomonasputida In a recent study, several Pseudomonas putida strains were analyzed with regard to toluene tolerance [145]. Three of these strains have been classified as highly resistant (P. putida DOT-TIE, P. putida S12 and P. putida MTB6). In P. putida DOT-TIE, three efflux pumps are involved in solvent tolerance: TtgABC [146], TtgDEF [147] and TtgGHI [148]. The same three efflux pump operons are present in the P. putida MTB6 chromosome although their participation in organic solvent extrusion has not been studied in detail. Pseudomonas putida S12 contains two of these efflux pumps encoded by the arpABC genes (98% identical to ttgABQ [149], and the srpABC (99% identical to ttgGHI), although only one of these efflux pumps, SrpABC, has been involved in solvent tolerance in the S12 strain [151]. P. putida Fl has two efflux pumps ttgABC and sepABC [152] and is more tolerant to toluene that P. putida KT2440, which only has the ttgABC pump, but it is more sensitive than DOT-TIE. In P. putida DOT-TIE, solvent tolerance is an inducible process, as growth of P. putida DOT-TIE in the presence of toluene supplied in the gas phase has a clear effect on cell survival: the sudden addition of 0.3% (vol/vol) toluene to P. putida DOT-TIE pre-grown with toluene in the gas phase resulted in survival of almost 100% of the initial cell number, whereas only 0.01% of the cells pre-grown in the absence of toluene tolerated exposure to this aromatic hydrocarbon (Fig. 6) [146, 153]. The three efflux pumps in this strain should therefore work together in this strain to achieve the maximal level of solvent tolerance, and they are probably tightly regulated in order to produce an optimal response to solvent stress. Most of the regulatory genes that encode for proteins involved in control of the expression of the efflux pumps belonging to the RND family are located adjacent to the structural genes of the pump, divergently transcribed from the efflux pump operon [154]. 6.1. Regulation of the ttgABC efflux pump operon The TtgABC efflux pump was the first efflux pump identified in P. putida DOT-TIE as involved in solvent tolerance. Physiological experiments done with a ttgB knockout mutant suggested that this efflux pump was involved in the socalled intrinsic tolerance. This mutant (P. putida DOT-TIE-18) did not withstand the sudden toluene shock (0.3% vol/vol) at all, and only a small but significant fraction (about 1 out of 105 cells) survived if pre-exposed to low toluene concentrations [146]. On the basis of this observation the existence of other(s) efflux pump(s) involved in toluene extrusion was postulated. The fact that in P. putida DOT-TIE-18 cultures, no survival at all was observed after sudden toluene shock compared with the 0.01% cell survival observed in the
362
wild-type, pointed out to a high basal expression of the TtgABC pump responsible for this noninduced intrinsic resistance. This hypothesis was confirmed after studying the expression levels of the ttgABC operon. In fact, the operon is transcribed at relatively high levels in the presence and also in the absence of toluene in accordance with its physiological behavior [155, 156]. Interestingly, the analysis of the TtgABC efflux pump susbstrate range revealed that it not only extruded different organic solvents such as toluene, styrene or p-xylene, but also different antibiotics such as ampicillin, carbenicillin, tetracycline, nalidixic acid and chloramphenicol [146, 155, 157]. Gene fusion to lacZ and primer extension assays showed that some substrates of the TtgABC efflux pump, such as chloramphenicol or tetracycline, did increase (to a different extent) the expression of the pump operon, whereas others (nalidixic acid, streptomycin or carbenicillin) did not [156]. Therefore, the contribution of this solvent efflux pump to antibiotic resistance is probably a consequence of the broad substrate specificity of both the pump itself and the transcriptional regulator. Upstream of the ttgABC operon, there is an open reading frame (named ttgR) encoding a protein that shows 50-60% sequence similarity with a number of transcriptional repressors such as AcrR (repressor of the acrAB operon in Escherichia coli) or MtrR (regulator of the MtrCDE efflux pump in Neisseria gonorrhoeae [158, 159]). All these proteins belong to the TetR family of transcriptional regulators [160]. TtgR is a repressor of the TtgABC efflux pump: a ttgR knockout mutant (P. putida DOT-TIE-13) exhibited an increased expression (6-fold higher than the wild-type) of the ttgABC operon. However, this increase in the efflux pump transcription levels did not lead to a higher survival rate of the culture when shocked with 0.3% (vol/vol) toluene, but did increase resistance towards different antibiotics such as chloramphenicol, carbenicillin and tetracycline [155]. ttgR gene expression was very similar to that observed for its cognate efflux pump operon: both are induced by chloramphenicol and tetracycline but not by toluene [155, 156]. In the TtgR-deficient mutant strain the basal activity of the ttgR promoter was eight-fold higher than the wild-type background, indicating that TtgR also down regulates its own transcription. The transcription initiation points of ttgA and ttgR have been mapped. The -10 region, although not the -35 region, of both the ttgA and ttgR promoters exhibits a certain degree of similarity to promoters recognized by sigma-70 [155]. The location of the start sites indicates that the divergent promoter regions fully overlap (Fig. 6). This overlap could explain some of the great similarity in the expression pattern.
363
Fig. 6. Organization of the ttgABC (A), ttgDEF (B) and ttgGHI (C) operons and their respective regulatory genes. The regulatory regions of each gene cluster are zoomed. TtgR (A), TtgT (B) and TtgV (C) DNA binding regions, deduced from DNAsel footprinting, are shadowed. Putative palindromic (arrows) or symetric (bold and underlined) recognition sites for each repressor are indicated. The +1 and the direction of transcription are marked with small triangles for each promoter (except for ttgT one, which distance from indicated). The -10 and the -35 positions of each promoter are also shown.
364
A recombinant and functional His-tagged TtgR protein has been overexpressed, purified and used in several in vitro experiments carried out to elucidate its regulatory role in ttgABC and ttgR expression. Gel shift assays showed that TtgR binds specifically to a DNA fragment corresponding to the ttgR-ttgABC intergenic region. Teran and coworkers [156] demonstrated that the addition of chloramphenicol or tetracycline to the binding reaction led to a gradual dissociation of TtgR from the ttgR-ttgABC intergenic region, suggesting that TtgR is able to bind these two antibiotics, which triggers its release from the promoter regions with subsequent enhanced expression. DNAse I footprint assays allowed the identification of the TtgR operator within the ttgA-ttgR intergenic region: a 36-bp DNA segment that includes the -10 and -35 region of the ttgABC promoter and the -10 of the ttgR promoter. The TtgR binding site deduced from DNAse I footprint, revealed a particularly long inverted repeat (28 bp) comprising two 12-bp half sites separated by 4 bp [156]. This indicates that probably each half site would accept two monomers of TtgR, as reported for the TtgR homolog QacR (repressor of the qacA multidrug pump gene of Staphylococcus aureus), whose 3D structure bound to its operator was resolved [161]. The mepABC efflux pump of a toluene-resistant variant of P. putida KT2442 has also been implicated in solvent and antibiotic resistance, and its sequence is practically identical to those of the TtgABC and ArpABC efflux pumps. The corresponding regulatory protein is probably encoded by mepR, although its function as a regulator has not been investigated yet [162]. Although TtgABC-like efflux pumps are widespread in different Pseudomonas putida strains (P. putida MTB5, P. putida KT2440, P. putida SMO116, among others) with different levels of toluene tolerance [145]. The role of these efflux pumps in solvent or antibiotic resistance and their regulation remains unknown. 6.2. Regulation of the ttgDEF efflux pump operon As described above, solvent tolerance studies in a ttgB knockout mutant of P. putida DOT-TIE (P. putida DOT-TIE-18) suggested the existence of other(s) inducible solvent efflux pump(s). By sequencing downstream from the toluene dioxygenase (tod) operon of P. putida DOT-TIE, Mosqueda and Ramos [147] identified three open reading frames (ttgDEF) that encode for the three components of an efflux pump which shares homology with other efflux pumps of Pseudomonas. The transporter, named TtgE, shares 59% identity with the previously described TtgB and 75% identity with TtgH. The contribution of this efflux pump to solvent tolerance was studied in a ttgD knockout mutant (P. putida DOT-T1E-1). A culture of this mutant strain showed a survival rate identical to the wild type when shocked with 0.3% (vol/vol) toluene. However if the cells were pre-induced with toluene in the gas phase, the survival rate of the
365
mutant was 100 times lower than in the wild type. Expression studies of the ttgDEF operon at the transcriptional level revealed that this pump is not expressed during growth under normal laboratory conditions, and demonstrated its inducible character in the presence of organic solvents (toluene or styrene) (Table 1) [147]. The wild type multiple antibiotic resistance was not affected in a TtgDEFdeficient strain; moreover, no increase in antibiotic resistance was obtained by pre-inducing the culture with toluene [147], suggesting that the substrate specificity of this pump is limited to organic solvents. There was also no induction of the ttgDEF operon in the presence of several antibiotics in the culture media (Teran et al., unpublished). Upstream from the ttgDEF operon and divergently transcribed, there is an open reading frame whose product shares homology with several members of the IclR family of transcriptional regulators. This gene, called ttgT, encodes for a protein 70% identical to the SrpS-negative regulator of SrpABC solvent efflux pump of P. putida S12 (see below). A ttgT knockout mutant showed a small increase in ttgDEF expression under non-inducing conditions, suggesting its involvement in the negative regulation of this operon. The fact that in this mutant strain there was still a strong induction of the ttgDEF expression in the presence of organic solvents suggested that TtgT is not the only protein involved in the induction of this operon by organic solvents (Teran et al., unpublished). Differently from ttgR, ttgT gene expression remained unaltered regardless of the organic solvent present in the growth medium, which suggested that expression from ttgDEF and ttgT promoters was not coordinated. Moreover, in the TtgT-deficient mutant, the activity of the ttgT promoter was similar to that of the wild-type, indicating that TtgT does not regulate its own transcription (Teran et al., unpublished). Gel shift experiments showed that TtgT was able to specifically bind a DNA fragment containing the ttgT-ttgDEF intergenic region (Teran et al., unpublished). DNAse I footprint assays revealed a single binding site along the ttgT-ttgDEF intergenic region which covers only the ttgDEF promoter region (37 to +5 from the transcription start point) and not the ttgT one, consistent with the in vivo expression studies described above. Therefore TtgT is directly involved in ttgDEF operon repression, probably by competing with the RNA polymerase for access to the efflux pump promoter. Analysis of the operator sequence does not reveal the presence of a clear single inverted repeat. Recent results of our laboratory suggests the induction of this efflux pump in a ttgTdeficient background by organic solvents is mediated regulator by the TtgV regulator.
366
6.3. Regulation of the ttgGHI efflux pump operon The third efflux pump involved in solvent tolerance in P. putida DOTTIE is called TtgGHI. A knockout mutant in which the TtgGHI efflux pump of P. putida DOT-TIE is not functional (P. putida DOT-T1E-PS28) is not able to survive a sudden 0.3% (vol/vol) toluene shock regardless of the growth conditions [163]. In contrast, the TtgABC and TtgDEF knockout mutants were still able to survive the toluene shocks to a different extent. The extreme toluene sensitivity of the P. putida DOT-T1E-PS28 mutant strain (ttgH::Q.Sm) under induced or noninduced conditions suggested that this efflux pump is involved in intrinsic as well as inducible resistance to organic solvents. The ttgGHI operon is expressed from a single promoter PG2 at a certain basal level in the absence of solvents, and its expression increases several-fold in the presence of aromatic hydrocarbons such as toluene and styrene, aliphatic alcohols such as 1-octanol, but not in the presence of antibiotics [163, 164]. In P. putida DOT-TIE, two genes ttgV and ttgW were identified upstream from the ttgGHI operon. They are transcribed divergently from the efflux pump operon. TtgV showed an overall 50%-60% similarity with a number of transcriptional regulators belonging to the IclR family, whereas ttgW is probably a pseudogene and the protein encoded seems not to be functional. A TtgV knockout mutant was constructed and characterized [163]. This strain showed increased resistance toward toluene shocks under noninduced conditions when compared with the wild type. The fraction of cells that survived the sudden addition of 0.3% (vol/vol) toluene was the same (107 cells) under induced and noninduced conditions. Analysis of the expression of the ttgGHI operon in this genetic background showed that the level of transcription increased 4-fold in the absence of toluene. Taken together, these data clearly indicated that TtgV is a repressor that prevents expression of the ttgGHI operon. The transcription initiation point of the ttgVW operon was mapped in cells growing in the absence and in the presence of toluene. The operon was transcribed from a single promoter regardless of the growth conditions, but the level of expression in the presence of toluene was 3- to 4-fold higher than in the absence of the aromatic hydrocarbon. The ttgVW operon was also shown to be induced by several organic solvents but not by antibiotics. The ttgVW operon showed a pattern of inducibility similar to that of ttgGHI, probably because both promoters are regulated in the same way. Sequence analysis of the promoter region showed that the -10 and -35 boxes of the VG2 overlap with the -35 and 10 boxes of the ftgFpromoter (Fig. 6). P-galactosidase assays carried out with a transcriptional fusion of VtlgV promoter to lacZ in a TtgV-deficient background showed that expression was about 3-fold higher than in the wild type strain, also in the absence of toluene, indicating that TtgV negatively controls its own expression. Together, these results suggested that the TtgV protein is a repressor of its own synthesis as well as of ttgGHI operon expression [163]. It is
367
interesting to note that in TtgV-deficient background expression from ttgDEF operon promoter increased 3-fold suggesting the involvement of this regulator in the control of this operon. Sequence analysis of PHgG and VttgV showed that both promoters overlapped. The TtgV protein has been overexpressed and purified with an Nterminal histidine tag. In vitro gel mobility shift assays demonstrated the specific binding of TtgV to a 210-bp DNA fragment comprising the ttgG and ttgV intergenic region. The DNAse I-protected region extended a 40-bp covering the -10/-35 regions of the ttgG promoter and the divergently oriented ttgV promoter [163]. To gain insight into the mechanism of regulation of ttgGHI transcription by TtgV, in vitro transcription experiments were carried out using the purified protein and the ttgV-ttgG intergenic region on a supercoiled plasmid. When the TtgV protein and the plasmid were incubated before the addition of RNApolymerase, ttgGHI transcription was completely repressed. However, when TtgV was added after the formation of the RNA-polymerase-«gGif/ promoter open complex, the repression level became negligible [164]. These findings support the idea that TtgV binding to the intergenic region blocks the entry of RNA-polymerase to transcribe both operons. In vitro transcription assays were carried out in the absence and in the presence of increasing concentrations of 1-hexanol, a known inducer of ttgGHI operon. Addition of the inducer to the transcription reaction in the presence of TtgV led to transcriptional levels similar to those observed in the absence of TtgV repressor, resulting in VttgG expression. This suggests that 1-hexanol decreased TtgV binding to the intergenic region, as it has been demonstrated by gel shift assays [164]. Acknowledgments Work in our laboratory was supported by grants of the European Commission (QLK3-CT-1999-00041, QLK3-CT-2001-00435 and QLK3-CT2000-0170) and a grant from the Human Science Foundation (RGY0021/2001). We thank C. Lorente for reading the manuscript and improving the language. REFERENCES [1] [2] [3] [4] [5] [6] [7] [8]
S. Dagley, Adv. Microb. Physiol, 6 (1971) 1. J. Sikkema, J.A.M. de Bont, and B. Poolman, Microbiol. Rev. 59 (1995) 201. D.L. Cruden, J.H. Wolfram , R.D. Rogers and D.T. Gibson , Appl. Environ. Microbiol. 58 (1992) 2723. A. Inoue and K. Horikoshi, Nature 338 (1989) 264. K. Kim, S. Lee, K. Lee, and D. Lim, J. Bacteriol. 180 (1998) 3692. J.L. Ramos, E. Duque, MJ. Huertas and A. Haidour, J. Bacteriol. 177 (1995) 3911. F.J. Weber, S. Isken and J.A.M. de Bont, Microbiology 140 (1994) 2013. H. Keweloh and H.J. Heipieper, Lipids 31 (1996) 129.
368
[9] HJ. Heipieper and J.A.M. de Bont, Appl. Environ. Microbiol. 60 (1994) 4440. [10] HJ. Heipieper, G.Meulenbeld, Q. van Oirschot and J.A.M. de Bont, Appl Environ Microbiol. 62(1996)2773. [11] B. Loffeld and H. Keweloh , Lipids31 (1996)811. [12] A. von Wallbrun, N.H.Richnow,, Neumann, G., Meinhardt, F., and Heipieper, H.J., (2003), J. BacterioL, 185 1730. [13] F.J. Weber and J.A.M. de Bont, Biochim. Biophys. Acta 1286 (1996) 225. [14] G. Mosqueda, M. I. Ramos-Gonzalez and J. L. Ramos, Gene, 232 (1999) 69. [15] P. Godoy, M.I. Ramos-Gonzalez and J.L. Ramos, J. BacterioL 18 (2001) 5285. [16] Q. Zhao, X.Z. Li, A. Mistry, R. Srikumar, L. Zhang, O. Lomovskaya and K. Poole, Antimicrob. Agents. Chemother. 42 (1998) 2225. [17] V.Koronakis, A. Sharfl, E. Koronakis, B. Luisi and C. Hughes, Nature 405 (2000) 914. [18] S. Murakami, R. Nakashima, E. Yamashita and A. Yamaguchi, Nature 419 (2002) 587. [19] H. I. Zgurskaya and H. Nikaido J. Mol. Biol. 285 (1999) 409. [20] E.W. Yu, G. McDermott, H.I. Zgurskaya, H. Nikaido and D.E. Jr. Koshland, Science 300(2003) 976. [21] H.I. Zgurskaya and H. Nikaido, J. BacterioL 182 (2000) 4264. [22] C.A. Elkins and H. Nikaido, Drug Resist. Update 6 (2003) 9. [23] S. J. Assinder and P.A. Williams, Adv. Microb. Physiol., 31 (1990) 1. [24] P.A. Williams, R.M. Jones and G. Zylstra, (2004), Genomics of catabolic plasmids. In Pseudomonas, vol. I, Chapter 6, pp. 165-196, Ed. J.L. Ramos, Kluwer Academic/Plenum Publishers. [25] S. Harayama, N. Mermod, M. Rekik, P.R. Lehrbach and K.N. Timmis, J. BacterioL, 169 (1987) 558. [26] S. Harayama and K.N. Timmis (1989), Catabolism of aromatic hydrocarbons by Pseudomonas. In Genetics of Bacterial Diversity, ed. D. Hopwood, K. Chater, 151-174. New York Academic. [27] C. Nakai, H. Kagamiyama and M. Nozaki, J. Biol. Chem., 258 (1983) 2923. [28] A. Polissi and S. Harayam, EMBO J. 12 (1993) 3339. [29] J.P. Shaw and S. Harayama, Eur. J. Biochem., 191 (1990) 705. [30] M. Suzuki, T. Hayakawa, J.P. Shaw, M. Rekik and S. Harayama, J. BacterioL 173 (1991) 1690. [31] D. T. Gibson, J. R. Koch and R. E. Kallio, Biochemistry, 7 (1968) 2653. [32] G. M. Whited and D. T. Gibson, J. BacterioL, 173 (1991) 3010. [33] K. M. Yen, M. R. Karl, L. M. Blatt, M. J. Simon, R. B. Winter, P. R. Fausset, H. S. Lu, A.A. Harcourt and K. K. Chen, J. BacterioL, 173 (1991) 5315. [34] R. H. Olsen, J. J. Kukor and B. Kaphammer, J. BacterioL, 176 (1994) 3749. [35] A. Fishman, Y. Tao and T.K. Wood, J. BacterioL, (2004) in press. [36] G. Baggi, P. Barbieri, E. Galli and S. Tollari, Appl. Environ. Microbiol., 53 (1987) 2129. [37] G. Bertoni, F. Bolognese, E. Galli and P. Barbieri, Appl. Environ. Microbiol., 62 (1996)3704. [38] G. Bertoni, N. Fujita, A. Ishihama and V. de Lorenzo, EMBO J., 17 (1998) 5120. [39] G. Bertoni, S. Marques and V. de Lorenzo, Mol. Microbiol., 27 (1998) 651. [40] J.L. Ramos, E. Duque, J.J. Rodriguez-Herva, P. Godoy, A. Haidour and J.L. Ramos, J. Biol. Chem. 272 (1997) 3887. [41] EganS.M., (2002) J. BacterioL, 184 5529. [42] M. T. Gallegos, R. Schleif, A. Bairoch, K. Hofmann and J. L. Ramos, Microbiol. Mol. Biol. Rev., 61 (1997) 393.
369
[43] J. L. Ramos, C. Michan, F. Rojo, D. Dwyer and K. Timmis, J. Mol. Biol., 211 (1990) 373. [44] R.Tobes and J. L Ramos, Nucleic Acids Res., 30 (2002) 318. [45] J. L. Ramos, A. Stolz, W. Reineke and K. N. Timmis, Proc. Natl. Acad. Sci. USA, 83 (1986) 8467. [46] C. Michan, L. Zhou, M. T. Gallegos, K. N. Timmis and J. L. Ramos, J. Biol. Chem., 267 (1992) 22897. [47] L. M. Zhou, K. N. Timmis and J. L. Ramos, J. Bacteriol., 172 (1990) 3707. [48] J. L. Ramos, F. Rojo, L. Zhou and K. N. Timmis, Nucleic Acids Res., 18 (1990) 2149. [49] J.L. Ramos, A. Wasserfallen, K. Rose and K.N. Timmis, Science 235 (1987) 593. [50] R. Ruiz and J. L. Ramos, J. Biol. Chem, 277 (2002) 7282. [51] N. LaRonde-LeBlanc and C. Wolberger, Biochemistry, 39 (2000) 11593. [52] C. A. Poore, C. Coker, J.D. Dattelbaum and H. L. Mobley, J. Bacteriol, 183 (2001) 4526. [53] S. Rhee, R. G. Martin, J. L. Rosner and D. R. Davies, Proc. Natl. Acad. Sci. USA, 95 (1998) 10413. [54] S. M. Soisson, B. MacDougall-Shackleton, R. Schleif and C. Wolberger, Science, 276 (1997)421. [55] R. Ruiz, S. Marques and J. L. Ramos, J. Bacteriol, 185 (2003) 3036. [56] M. Manzanera, S. Marques and J. L. Ramos, FEBS Lett, 476 (2000) 312. [57] N. Kaldalu, U. Toots, V. de Lorenzo and M. Ustav, J. Bacteriol, 182 (2000) 1118. [58] M. M. Gonzalez-Perez, S. Marques, P. Dominguez-Cuevas and J. L. Ramos, FEBS Lett, 519(2002) 117. [59] M. T. Gallegos, S. Marques and J. L. Ramos, J. Bacteriol, 178 (1996) 6427. [60] M. M. Gonzalez-Perez, J. L. Ramos, M. T. Gallegos and S. Marques, J. Biol. Chem, 274(1999)2286. [61] B. Kessler, V. de Lorenzo and K. Timmis, J. Mol. Biol, 230 (1993) 699. [62] B. Kessler, M. Herrero, K. N. Timmis and V. de Lorenzo, J. Bacteriol, 176 (1994) 3171. [63] B. Kessler, S. Marques, T. Kohler, J. L. Ramos, K. N. Timmis and V. de Lorenzo, J. Bacteriol, 176(1994)5578. [64] N. Kaldalu, T. Mandel and M. Ustav, Mol. Microbiol, 20 (1996)569. [65] K. Miura, S. Inouye and A. Nakazawa, Mol. Gen. Genet, 259 (1998) 72. [66] I. Artsimovitch, K. Murakami, A. Ishihama and M. M. Howe, J. Biol. Chem, 271 (1996) 32343. [67] S. Busby and R. H. Ebright, Mol. Microbiol, 23 (1997) 853. [68] A. Dhiman and R. Schleif, J. Bacteriol, 182 (2000) 5076. [69] S. Marques, M. T. Gallegos and J. L. Ramos, Mol. Microbiol, 18 (1995) 851. [70] S. Marques, M. Manzanera, M. M. Gonzalez-Perez, M. T. Gallegos and J. L Ramos, Mol. Microbiol, 31 (1999) 1105. [71] T. Yura and K. Nakahigashi, Curr. Opin. Microbiol, 2 (1999) 153. [72] S. Inouye, A. Nakazawa and T. Nakazawa, Gene, 66 (1988)301. [73] S. Kustu, E. Santero, J. Keener, D. Popham and D. Weiss, Microbiol. Rev, 53 (1989) 367. [74] A. K. North, K. E. Klose, K. M. Stedman and S. Kustu, J. Bacteriol, 175 (1993) 4267. [75] M. A. Abril, C. Michan, K. N. Timmis and J. L. Ramos, J. Bacteriol, 171 (1989) 6782. [76] A. Delgado and J. L. Ramos, J. Biol. Chem, 269 (1994) 8059. [77] A. Delgado, R. Salto, S. Marques and J. L. Ramos, J. Biol. Chem, 270 (1995) 5144.
370
[78] J. Garmendia, D. Devos, A. Valencia and V. de Lorenzo, Mol. Microbiol., 42 (2001) 47. [79] V. Shingler,, (2004), Transcriptional regulation and catabolic strategies of phenol degradative pathways. In Pseudomonas, vol. II, Chapter 16, pp. 451-478, Ed. J.L. Ramos, Kluwer Academic/Plenum Publishers. [80] E. Skarfstad, E. O'Neill, J. Garmendia and V. Shingler, J. Bacteriol, 182 (2000) 3008. [81] S. Fernandez, V. de Lorenzo and J. Perez-Martin, Mol. Microbiol, 16 (1995) 205. [82] J. Perez-Martin and V. de Lorenzo, J. Mol. Biol, 258 (1996) 575. [83] J. Perez-MartinandV.de Lorenzo, Cell, 86(1996)331. [84] R. Dixon, Mol. Gen. Genet, 203 (1986) 129. [85] S. Inouye, A. Nakazawa and T. Nakazawa, Gene, 29 (1984) 323. [86] M. A. Abril, M. Buck and J. L. Ramos, J. Biol. Chem, 266 (1991) 15832. [87] V. de Lorenzo, M. Herrero, M. Metzke and K. N. Timmis, EMBO J, 10 (1991) 1159. [88] G. H. Seong, E. Kobatake, K. Miura, A. Nakazawa and M. Aizawa, Biochem. Biophys. Res. Commun, 291 (2002) 361. [89] G. Bertoni, M. Martino, E. Galli and P. Barbieri, Appl. Environ. Microbiol, 64 (1998) 3626. [90] M. Carmona, V. de Lorenzo and G. Bertoni, J. Biol. Chem, 274 (1999) 33790. [91] A. Holtel, M. A. Abril, S. Marques, K. N. Timmis and J. L. Ramos, Mol. Microbiol, 4 (1990) 1551. [92] A. Holtel, D. Goldenberg, H. Giladi, A. B. Oppenheim and K. N. Timmis, J. Bacteriol, 177(1995)3312. [93] A. Holtel, K. Timmis and J. L. Ramos, Nucleic Acids Res, 20 (1992) 1755. [94] G. Bertoni, J. Perez-Martin and V. de Lorenzo, Mol. Microbiol, 23 (1997) 1221. [95] M. Gomada, S. Inouye, H. Imaishi, A. Nakazawa and T. Nakazawa, Mol. Gen. Genet, 233(1992)419. [96] S. Inouye, A. Nakazawa and T. Nakazawa, J. Bacteriol, 163 (1985) 863. [97] S. Marques, M. T. Gallegos, M. Manzanera, A. Holtel, J. Bacteriol, 180 (1998) 2889. [98] S. Fraile, F. Roncal, L. A. Fernandez and V. de Lorenzo, J. Bacteriol, 183 (2001) 5571. [99] W. A. Duetz, S. Marques, C. de Jong, J. L. Ramos and J. G. van Andel, J. Bacteriol, 176(1994)2354. [100] W.A. Duetz, S. Marques, B. Wind, J.L. Ramos and J.G. Van Andel, Appl. Environ. Microbiol, 62(1996)601. [101] A. Holtel, S. Marques, I. Mohler, U. J akubzik and K. N. Timmis, J. Bacteriol, 176 (1994) 1773. [102] N. Hugouvieux-Cotte-Pattat, T. Kohler, M. Rekik and S. Harayama, J. Bacteriol, 172 (1990)6651. [103] S. Marques, A. Holtel, K. N. Timmis and J. L. Ramos, J. Bacteriol, 176 (1994) 2517. [104] M. J. Worsey, and P. A. Williams, J Bacteriol. 124 (1975) 7. [105] I. Cases, V. de Lorenzo and J. Perez-Martin, Mol. Microbiol, 19 (1996) 7. [106] M. Carmona, M. J. Rodriguez, O. Martinez-Costa and V. de Lorenzo, J. Bacteriol, 182 (2000)4711. [107] F. Rojo and M.A. Dinamarca, (2004), Catabolite repression and physiological control. In Pseudomonas, vol. II, Chapter 13, pp. 365-388, Ed. J.L. Ramos, Kluwer Academic/Plenum Publishers. [108] M. Vails, M. Buckle and V. de Lorenzo, J. Biol. Chem, 277 (2002) 2169. [109] P. Jurado, L. A. Fernandez and V. de Lorenzo, J. Bacteriol, 185 (2003) 3379. [110] T. Tomoyasu, K. Yamanaka, K. Murata, T. Suzaki, P. Bouloc, A. Kato, H. Niki, S. Hiraga and T. Ogura, J. Bacteriol, 175 (1993) 1352.
371
[111] M. Carmona and V. de Lorenzo, Mol. Microbiol, 31 (1999) 261. [112] Y. Akiyama, Proc. Natl. Acad. Sci. USA, 99 (2002) 8066. [113] I. Cases, and V. de Lorenzo, J. Bacteriol., 182 (2000) 956. [114] I. Cases, J. A. Lopez, J. P. Albar and V. de Lorenzo, J. Bacteriol., 183 (2001) 1032. [115] I. Cases, F. Velazquez and V. de Lorenzo, J. Bacteriol., 183 (2001) 5128. [116] I. Cases, J. Perez-Martin and V. de Lorenzo, J. Biol. Chem., 274 (1999) 15562. [117] B. A. Finette, V. Subramanian and D. T. Gibson, J. Bacteriol., 160 (1984) 1003. [118] D. T. Gibson, G. H. Zylstra and S. Chauhan, (1990), Biotransformations catalyzed by toluene dioxygenase from Pseudomonasputida Fl, p. 121-132. In S. Silver, A.M. [119] P. C. Lau, Y. Wang, A. Patel, D. Labbe, H. Bergeron, R. Brousseau, Y. Konishi and M. Rawlings, Proc. Natl. Acad. Sci. USA, 94 (1997) 1453. [120] F. M. Menn, G. J. Zylstra and D. T. Gibson, Gene, 104 (1991) 91. [121] Y. Wang, M. Rawlings, D. T. Gibson, D. Labbe, H. Bergeron, R. Brousseau and P. C. Lau Mol. Gen. Genet., 246 (1995) 570. [122] G. J. Zylstra and D. T. Gibson, J. Biol. Chem., 264 (1989) 14940. [123] G. J. Zylstra, W. R. McCombie, D. T. Gibson and B. A. Finette, Appl. Environ. Microbiol., 54 (1988)1498. [124] B. A. Finette and D. T. Gibson, Biocatalyst 2 (1988) 29. [125] P. Dominguez-Cuevas, and S. Marques, (2004), Compiling sigma-70 dependent promoters. In Pseudomonas, vol. II, Chapter 11, pp. 319-344. Ed. J.L. Ramos, Kluwer Academic/Plenum Publishers. [126] B. L. Taylor and I. B. Zhulin, Microbiol. Mol. Biol. Rev., 63 (1999) 479. [127] I. B. Zhulin, B. L. Taylor and R. Dixon, Trends Biochem. Sci., 22 (1997) 331. [128] J. A. Hoch and T. J. Silhavy, (1995), Two-component signal transduction. American Society for Microbiology, Washington DC. [129] J. Reizer and M. H. J. Saier, Curr. Opin. Struct. Biol., 7 (1997) 407. [130] R. E. Parales, J. L. Ditty and C. S. Harwood, Appl. Environ. Microbiol., 66 (2000) 4098. [131] K. M. Yen and M. R. Karl, J. Bacteriol., 174 (1992) 7253. [132] M. I. Ramos-Gonzalez, M. Olson, A. A. Gatenby, G. Mosqueda, M. Manzanera, M. J. Campos, S. Vichez and J. L. Ramos, J. Bacteriol., 184 (2002) 7062. [133] A. Wright and R. H. Olsen, Appl. Environ. Microbiol., 60 (1994) 235. [134] A. Ben-Bassat, M. Cattermole, A.A. Gatenby, K.J. Gibson, M.I. Ramos-Gonzalez, J.L. Ramos and S. Sariaslani, (2003), Method for the production ofp-hydroxybenzoate in species of Pseudomonas and Agrobacterium. US Patent 6,586,229 [135] B. Entsch, Y. Nan, K. Weaich and K. F. Scott, Gene, 71 (1988) 279. [136] C. M. Wong, M. J. Dilworth, and A. R. Glenn, Microbiology, 140 (1994) 2775. [137] A. M. Byrne and R. H. Olsen, J. Bacteriol., 178 (1996) 6327. [138] J. J. Kukor and R. H. Olsen, J. Bacteriol., 172 (1990) 4624. [139] J. J. Kukor and R. H. Olsen, J. Bacteriol., 174 (1992) 6518. [140] J. J. Kukor and R. H. Olsen, J. Bacteriol., 173 (1991) 4587. [141] H. Y. Kahng, A. M. Byrne, R. H. Olsen and J. J. Kukor, J. Bacteriol., 182 (2000) 1232. [142] J. G. Leahy, G. R. Johnson and R. H. Olsen, Appl. Environ. Microbiol., 63 (1997) 3736. [143] M. S. Shields, M. J. Reagin, R. R. Gerger, R. Campbell and C. Somerville, Appl. Environ. Microbiol., 61 (1995) 1352. [144] G.R. Johnson and R.H. Olsen, Appl. Environ. Microbiol., 63 (1997) 4047. [145] A. Segura, A. Rojas, A. Hurtado, M.J. Huertas and J.L. Ramos, Extremophiles 7 (2003)371. [146] J.L. Ramos, E. Duque, P. Godoy and A. Segura, J. Bacteriol. 180 (1998) 3323.
372
[147] G. Mosqueda and J.L. Ramos, J. Bacteriol. 182 (2000) 937. [148] A. Rojas, E. Duque, G. Mosqueda, G. Golden, A. Hurtado, J.L. Ramos and A. Segura, J. Bacteriol., 183(2001)3967. [149] J. Kieboom and J.A.M. de Bont, Microbiology 147 (2001) 43. [150] J. Kieboom, J.J. Dennis, J.A.M. de Bont and G.J. Zylstra, J. Biol. Chem. 273 (1998) 85. [151] J. Kieboom, J.J. Dennis, G.J. Zylstra and J.A.M. de Bont, J. Bacteriol. 180 (1998) 6769. [152] P. Phoenix, A. Keane, A. Patel, H. Bergeron, S. Ghoshal and P.C.K. Lau, Env. Microbiol. 5 (2003) 1309. [153] J. L. Ramos, S. Marques and K. N. Timmis, Annu. Rev. Microbiol., 51 (1997) 341. [154] A. Segura, E. Duque, G. Mosqueda, J.L. Ramos and F. Junker, Environ. Microbiol. 1 (1999)191. [155] E. Duque, A. Segura, G. Mosqueda and J.L. Ramos, Mol. Microbiol. 39 (2001) 1100. [156] W. Teran, A. Felipe, A. Segura, A. Rojas, J.L. Ramos and M.T. Gallegos, Antimicrob. Agents Chemother. (2003), in press. [157] J.L. Ramos, E. Duque, M.T. Gallegos, P. Godoy, M.I. Ramos-Gonzalez, A. Rojas, W. Teran and A. Segura, Ann. Rev. Microbiol. 56 (2002) 743. [158] D. Ma, M. Alberti, C. Lynch, H. Nikaido and J.E. Hearst, Mol. Microbiol. 19 (1996) 101. [159] W. Pan and B.G. Spratt, Mol. Microbiol. 11 (1994) 769. [160] W. Hillen and C. Berens, Annu. Rev. Microbiol. 48 (1994) 345. [161] M.A. Schumacher, M.C. Miller, S. Grkovic, M.H. Brown, R.A. Skurray and R.G. Brennan, EMBO J. 21 (2002) 1210. [162] F. Fukumori, H. Hirayama, H. Takami, A. Inoue and K. Horikoshi, Extremophiles 2 (1998)395. [163] A. Rojas, A. Segura, M.E. Guazzaroni, W. Teran, A. Hurtado, M.T. Gallegos and J.L. Ramos, J. Bacteriol., 185 (2003) 4755. [164] M.E. Guazzaroni, W. Teran, X. Zhang, M.T. Gallegos, and J.L. Ramos, J. Bacteriol., (2004) in press.
Studies in Surface Science and Catalysis 151 R. Vazquez-Duhalt and R. Quintero-Ramirez (Editors) © 2004 Elsevier B .V. All rights reserved.
373
Chapter 13
Bacterial hydrocarbon biosynthesis revisited B. Valderrama Departamento de Ingenieria Celular y Biocatalisis, Universidad Nacional Autonoma de Mexico, AP 510-3, Cuernavaca, Morelos, 62250, Mexico.
1. INTRODUCTION One of the greatest challenges faced by the modern world is the dissociation from the heavy dependency of the energy technologies upon the chemical bonds of hydrocarbons. The imminent exhaustion of conventional oil sources, ranging from a pessimistic ultimate recovery volume of 0.6 trillions of barrels to a highly optimistic volume of 3.9 trillions of barrels [4], results in a stringent requirement for the development of alternative technologies. It is important to note that the world is not to run out of hydrocarbons, given the substantial amount of lowgrade, hard-to-extract supplies such as the Canadian tar sands or the abundant heavy oil reservoirs in Venezuela and Mexico. Nevertheless, exploiting these reservoirs is likely to be much more expensive financially, energetically, politically and especially environmentally. Biotechnology has greatly impacted modern industry, from the now conventional production of goods by the use of fermentations to the novel synthesis of valuable fine chemicals using enzymes [5, 6, 7, 8]. Notwithstanding its enormous potential, the incorporation of biotechnological tools into the oil industry has faltered [9]. In particular, the search of alternative hydrocarbon sources through biotechnological media has not been assessed. Here, I compile information regarding the biological production of hydrocarbons by bacteria and explore its potential, not only as an environmentally-friendly fuel supply, but also as a renewable source for basic petrochemicals.
374
2. HYDROCARBON BIOSYNTHESIS IS A COMMON TRAIT AMONG BACTERIA The ability of organisms to synthesize hydrocarbons has been observed in all phyla [10]. Whereas alkanes are mainly involved in epicuticular wax biosynthesis in plants [11], in insects, their roles are more diverse, ranging from waterproofing of the cuticle to participation in sexual behaviors as aphrodisiac pheromones [12]. Furthermore, many marine animals, from invertebrates to whales, contain appreciable amounts of hydrocarbons as component of waxes, which appear to have a variety of functions, from serving as energy source to insulation, buoyancy and even echo-location [13]. The accumulation of nonvolatile hydrocarbons by microorganisms has been shown to occur in microalgae [14, 15, 16, 3], in bacteria [17, 18, 19] and in yeast [20, 21]. Originally, the ability of microorganisms to produce hydrocarbons was studied as part of the biogenic hypothesis for oil reservoir formation. This hypothesis was actively investigated between 1930 and 1960 by C.E. ZoBell, who proposed a significant bacterial role in the origin of petroleum [22, 23]. In 1950 he suggested that microbial modification of organic remains in sediments could contribute precursors for oil formation by lowering their oxygen and nitrogen content and increasing their carbon and hydrogen content, and by the direct production of methane and other higher hydrocarbons [24, 25]. ZoBell also suggested that hydrogenation of unsaturated fatty acids and their subsequent decarboxylation might have contributed to petroleum formation. Although a bacterial role in the initial processing of organic matter from which petroleum is derived has become generally accepted, there is no experimental evidence that bacteria were directly responsible for hydrocarbon production in significant quantities. Although ZoBell later abandoned the idea of direct microbial formation of hydrocarbons in large amounts from organic matter, he described several cases where cultures of marine bacteria presented substantial capacity for aliphatic hydrocarbon biosynthesis [26, 27, 23]. After the development of more powerful analytical procedures, the subject of bacterial hydrocarbon production resurged between 1960 and 1970 [1, 28, 19, 18]. Unequivocal evidence of hydrocarbon accumulation was observed in all the bacterial species tested (Table 1), including photosynthetic bacteria as well as in non-photosynthetic bacteria. The composition of bacterial hydrocarbons was complex, with length ranging from Ci5 up to C36, and including n-alkanes, alkenes, and branched hydrocarbons. In particular, nonphotosynthetic bacteria accumulate long-chain n-alkanes (C27-C29), whereas nalkanes with shorter chains (C|7-C2o) are more abundant in photosynthetic bacteria [29]. Photosynthetic bacteria, as well as anaerobic non-photosynthetic bacteria, are characterized by the presence of isoprenic units of pristane and phytane [30].
375
Table 1 Taxonomical distribution of eubacterial and archaeal species able to synthesize and accumulate hydrocarbons. Phyllum
Class
Genus
Reference
Proteobacteria
a Proteobacteria
Rhodopseudomonas sphaeroides Rhodospirillum rubrum Chromatium Escherichia coli Rhodimicrobium vannielii Vibrio furn issii Serratia marinorubrum Vibrio marinus Vibrio ponticus Vibrio furnissii Desulfovibrio desulfuricans Desulfovibrio Essex Desulfovibrio Hildenborough
[33,30] [33,30] [29] [33,30] [30] [17] [23] [34] [23] [17] [35] [30] [30]
Clostridium acidiurici Clostridium tetanomorphum Sarcinaflava Sarcina lutea Sarcina subflava Staphylococcus sp. Bacillus sp.
[30] [30] [18] [18,28] [18] [18] [36]
y Proteobacteria
8/E Proteobacteria
Firmicute
Clostridia
Bacilli
Actinobacteria
Actinobacteria
Micrococcus lysodeikticus Micrococcus sp. Mycobacterium sp. Corynebacterium sp. Arthrobacter sp.
[18,33,30] [36] [36] [36] [36]
Cyanobacteria
Nostococales Oscillatoriales
Nostoc muscorum Nostoc sp. Phormidium luridum
[30] [36] [30]
Chlorobi
Chlorobia
Chlorobium
[30]
Euryarchaeota
Thermoplasmata Methanomicrobia Halobacteria
Thermoplasma sp. Methanosarcina barkeri Halobacterium cutirubrum
[32] [32] [32]
Crenarchaeota
Thermoprotei
Sulfolobus sp.
[31]
This ability is not restricted to eubacteria. Some archeal species from the genus Sulfolobus, Thermoplasma, Methanosarcina and Halobacterium, have been demonstrated able to synthesize and accumulate hydrocarbons such as squalene and other acyclic isoprenoids (C20-C25). [31, 32]. Furthermore, individual species that produce hydrocarbons as major components have been isolated from mesophilic, thermophilic, psycrophilic, acidophilic, alkalinophilic,
376
and halophilic environments under aerobic or anaerobic, autotrophic or heterotrophic conditions. The environmental distribution of hydrocarbon producers follows no discernible pattern that can be used as a guide for finding prolific hydrocarbon producers. 3. BIOSYNTHETIC PATHWAYS Non-isoprenoid biological hydrocarbons were presumed to derive from fatty acids. Currently, there are two known pathways for the conversion of fatty acids to straight-chain hydrocarbons. The best known of them is the elongationdecarboxylation process (Fig.lA). In this case, a fatty acid precursor, such as oleic acid, is elongated by the continuous addition of a C2 unit derived from malonyl-CoA. The hydrocarbon produced is then cappedoff through a decarboxylation reaction when it reaches the designated length. The second mechanism involves the "head-to-head" condensation of two fatty acids (Fig. IB). In this path, one of the acid derivatives is specifically decarboxylated following the condensation step, while the total carbon chain of the other is incorporated into the hydrocarbon [3]. The commitment step in these pathways is the decarboxylation reaction. It has been well documented that CO2 elimination from carboxylic acids requires high energy and therefore has to be activated by a P-substituent able to stabilize the negative charge generated by CO2 release. Accordingly, it has generally been thought that activated fatty acid derivatives are the intermediates in the decarboxylation leading to hydrocarbons.
Fig. 1 Metabolic pathways for aliphatic hydrocarbon biosynthesis (Modified from [3] and [1]). LCFA - Long Chain Fatty Acids
377
Fatty acyl-CoA reductase activity has been identified in a variety of other organisms, from bacteria to animals [40, 41, 42, 38]. The gene encoding this enzyme has been recently cloned from the y-proteobacteria Acinetobacter calcoaceticus and Photobacterium leiognathi, as well as from seeds of jojoba (Simmondsia chinensis) [42, 43, 44]. As can be seen in Fig. 2, all of them harbor the residues conforming the catalytic triad observed in related dehydrogenases [45]. Interestingly, each one of these reference sequences represents independent groups, on the basis of sequence similarity (see Fig. 3). Groups I and II are rather selective, being all their members either plantsor bacteria, respectively. Group III has the sequence from Acinetobacter calcoaceticus as sole member. This sequence is similar to oxido-reductases with different substrate specificities from various sources. The aldehydes generated from fatty acid reduction in B. braunii are further reduced to hydrocarbons (alkanes). The initial observation that the resulting alkane had one less carbon than the aldehyde leaded to the proposal of a decarbonylation as the final step. Such an activity would yield one molecule of alkane and one molecule of CO as products. Two plant aldehyde decarbonylases (from pea and from B. braunii) have been studied in some detail [46, 47]. They are integral membrane proteins with the pea decarbonylase suggested to be located in the cuticular cell membrane and the algal decarbonylase in the microsomal membranes. Both use highly hydrophobic fatty aldehydes as substrate and need metal ions for their function. B. braunii decarbonylase is a cobalt-porphyrin enzyme able to convert a fatty aldehyde to hydrocarbon and CO without requiring any other cofactor under anaerobic conditions [48]. The partially purified decarbonylase from pea is merely known to depend on metal ions, probably copper, the activity being severely inhibited in the presence of metal ion chelators [47]. Unfortunately, none of their genes have been cloned to date. Genetic approaches produced mutants of Arabidopsis that have altered surface composition, including a decreased amount of hydrocarbons. One of these mutants, cerl, was postulated to be located in a decarbonylase because it is proportionally deficient only in alkanes (and alkane-derived metabolites) and accumulated fatty aldehydes [11]. The cerl gene was cloned and shown to bedeposits actively expressed in stem and in fruit tissue, which corresponded to the main of waxes. Interestingly, cerl affected pollen fertility. The CER1 protein has been related, on the basis of sequence similarity, to C-5 sterol desaturase and C-4 methyl oxidase, involved in cholesterol biosynthesis [49,50]. All these enzymes are integral membrane proteins containing many conserved histidine residues.
378
Fig. 2. Multiple alignment of fatty acyl CoA reductases from the proteobacteria Acinetobacter calcoaceticus [43] and Photobacterium leiognathi [44] and from the plant Simmondsia chinensis (jojoba) [42]. Residues conforming the predicted catalytic triad are highlighted.
The n-heptane tissue-specific biosynthetic pathway in Pinus jeffreyi proceeds through the polymerization of acetate via a tipical fatty acid synthase reaction sequence yielding a C8 thioester, which subsequently undergoes a twoelectron reduction to generate a free thiol molecule and octanal, the latter undergoes the direct loss of Ci to generate n-heptane [2] (Fig.4). Aside from the obvious generality of the aforementioned pathway, alternative reactions for alkane biosynthesis have been revealed recently. In insects, whereas aldehyde was the immediate precursor of alkanes, the enzyme involved in the deearbonylation step was proposed to be a P450 enzyme which required the presence of molecular oxygen and NADPH or NADH (less effectively) with the production of CO2 instead of CO [51]. In archaea, the
379
synthesis and cleavage of acetyl-CoA is catalyzed by the acetyl-CoA decarbonylase synthase (ACDS) complex, a completely different enzyme composed of five different subunits, [52, 53]. There is no information available regarding an equivalent acyl-CoA decarbonylation reaction in bacteria. 3.1. Bacterial biosynthesis In contrast to the heterogeneity of plant synthesized hydrocarbons [3], a very large proportion of bacterial hydrocarbons are less dispersed in length and structure. The best known example comes from the study of Sarcina lutea, whose most abundant hydrocarbon presented a branch methyl on both ends of the molecule, a double bond near the center, and contained a number of carbon atoms equal to one less than two times the average number of carbon atoms in the most abundant fatty acid [54] (Fig. 5). These structural characteristics, as well as the distribution of the carbon chains of iso-leucine, valine and acetate in the fatty acids and hydrocarbons synthesized in vivo, are consistent with a headto-head condensation of fatty acids mechanism [1, 54, 55]. A plausible precursor for this mechanism would be multiple methyl-branched fatty acyl-CoA molecules produced by the methyl-malonate driven elongation of fatty acid molecules, as has been observed in mycobacteria [56]. In the head-to-head condensation mechanism, one molecule of fatty acid undergoes decarboxylation. When S. lutea cells are grown with low acetate, approximately 70% of the incorporated palmitate molecules are decarboxylated, in contrast, when acetate was included in the growth medium, palmitate was incorporated without undergoing further decarboxylation, suggesting high specificity [57].
Fig. 3. Deduced Fatty Acyl CoA reductases are organized in three different groups on the basis of sequence similarity. Sequence identification numbers in parenthesis.
380
Fig. 4. Proposed pathway for formation of n-heptane from acetate in Pinusjeffreyi. Adapted from [2]
The in vitro incorporation of palmitate into hydrocarbons requires the addition of CoA, Mg++, ATP, NADPH and pyridoxal or pyridoxamine phosphate [58]. The requirement for the first three cofactors was consistent with the participation of acyl-CoA and this was confirmed by showing that in the absence of added CoA, palmitoyl-CoA was over 20 times better a precursor than the free acid. Under these conditions, palmitate was sometimes decarboxylated when added to the system. The specificity was originated from the source of palmitate, whether it was free or esterified. Approximately 30% of the free acid was decarboxylated while the methyl ester derivative was essentially 100% decarboxylated. Also, the CoA derivative of palmitate was extensively decarboxylated. Furthermore, the direct incorporation of palmitate required piridoxamine as a cofactor instead of CoA [58]. Although it is clear that biosynthesis occurs by a head-to-head condensation mechanism, the intermediates of the pathway have not yet been fully elucidated.
Fig. 5 Relationship between isoleucine and acetate, the anteiso C15 fatty acid and the C29 hydrocarbon with anteiso branch methyls in both ends. Modified from [1].
381
All this information leads to the idea that ability of bacteria to synthesize hydrocarbons is widespread and that it probably occurs by more than one mechanism which are significantly different compared to those described in plants. 4. DOWNSTREAM PROCESSING The isolation and refinement of bacterial hydrocarbons has not been approached at production scale. Nevertheless, there is abundant information regarding the operations developed for a similar procedure with B. braunii. The process of extracting hydrocarbons from these cells can be thought of as consisting of three major operations. The first is that of harvesting the cells from the growth medium. This involves the concentration or flocculation of cells from the liquid where it is grown. This operation can be achieved through a variety of means that include filtration, mechanical centrifugation or concentration, gravitational concentration or chemical flocculation. The most efficient method for largescale hydrocarbon recovery is chemical flocculation. For efficient extraction, the cells must be concentrated to a semi-dry paste. The second step is that of the actual physical extraction of the hydrocarbon fuel from the cells. Under suitable conditions, up to 70% of the total hydrocarbon content can be released by 30 min of contact with solvents. The selected solvent should be immiscible with water, with a density significantly different than water, should be non-toxic and reusable. In view of these considerations, hexane appears to be the solvent of choice [59]. Growth and hydrocarbon production are not affected by repeated extraction with hexane. In fact, a higher content of hydrocarbons has been observed in hexane-treated biomass relative to controls. Nevertheless, recovery yields are influenced by the physiological status of the culture. Scale up of the extraction can be difficult, given the algae propensity to aggregate. Extensive clumping shields a large fraction of the biomass from exposure to solvent. Alternative methods aimed to increase the oil extraction yield have been explored. Recovery yields are markedly increased relative to freely suspended controls when cells immobilized by adsorption in polyurethane foam were continuously extracted with hexane [60]. Supercritical fluid extraction is another technology that has been applied [61]. The third operation would be the collection and concentration of the hydrocarbon product. Although biosynthetic hydrocarbons can be directly used in internal combustion engines after extraction with hexane, its performance is improved by further modification. Cellular hydrocarbons can be converted to gasoline (60 to 70%), light cycle oil (10 to 15%), heavy cycle oil (2 to 8%) and coke (5 to 10%) after catalytic cracking [62]. The yield of gasoline obtained by
382
this method is comparable to the yield obtained from petroleum. Additionally, the gasoline produced has a sufficiently high octane number for direct use in automobiles. In principle, all or some of these operations might be directly applied for the extraction and upgrading of bacterial hydrocarbons. 5. FUTURE PROSPECTS One of the main features impelling the study of microalgae as an alternative source of hydrocarbons lies in their ability to fix carbon dioxide through photosynthesis. The possibility of reducing the atmospheric carbon load by direct recycling into fuels is appealing [63]. Despite all efforts, the production of algal hyodrocarbons is not competitive with petroleum derived fuels, mainly due to the slow growth rate of microalgae, the low extraction yields and the high viscosity of the cultures. The alternative approach of cloning the algal hydrocarbon synthesis genes into other microorganisms has proven to be difficult. Nevertheless, the necessity for an alternative source of hydrocarbons, not only as fuels, but also for the fine-chemicals industry is still there. As presented in this document, the ability to synthesize hydrocarbons is widely distributed among eubacteria. The biosynthetic pathways seem to be completely different compared to those observed in plants, involving a novel set of enzymatic activities. At the present time, two different strategies might be suggested in order to increase the accumulation of aliphatic hydrocarbons in bacteria. A direct one, involving the identification and cloning of the genes encoding the relevant activities involved, aimed at their subsequent expression in a suitable host might result in an important advance. However, this strategy might not be as straightforward as appears. It is well understood that the arbitrary modification of the cellular carbon fluxes might severely impair cell viability and performance. In this case, the natural deviation of the cellular carbon pool into a reserve compound (hydrocarbons) might not be gratuitous but the result of a finely tuned metabolic network. The deeper understanding of the physiology of the process might eventually provide the knowledge basis for the rational design of an imbalanced metabolism yielding the desired accumulation of hydrocarbons without compromising cell viability. The current availability of more powerful analytical as well as genetic tools enables us to face this challenge. Aknowledgements The author thanks Shirley Ainsworth for assistance during the bibliographical investigation. This work was supported by PEMEX grant 138.
383
REFERENCES [I] [2] [3] [4] [5] [6] [7] [8] [9] [10] II1] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]
P.W. Albro and J.C. Dittmer, Lipids, 5 (1970) 320. T.J. Savage, M.K. Hristova and R. Croteau, Plant PhysioL, 111 (1996) 1263. P. Metzger, C. Largeau and E. Casadevall in W. Herz, G.W. Kirby, W. Steglich and Ch. Tamm (eds.) Progress in the Chemistry of Organic Natural Products, Springer-Verlag, New York (1991). C. Hall, P. Tharakan, J. Hallock, C. Cleveland and M. Jefferson, Nature, 426 (2003) 318. A.L. Demain, Biotechnol. Adv., 18 (2000) 499. J.D. Rozzell, Bioorg. Med. Chem., 7 (1999) 2253. W. Gerhartz (eds.), Enzymes in industry: production and applications, VCH, Germany, (1990). A. Liese and M.V. Filho, Curr. Opin. Biotechnol., 10 (1999) 595. R. Vazquez-Duhalt, E. Torres, B. Valderrama and S. Le Borgne, Energy Fuels, 16 (2002) 1239. Chemistry and Biochemistry of Natural Waxes, Elsevier, Amsterdam, (1976). M.G. Aarts, C.J. Keijzer, WJ. Stiekema and A. Pereira, Plant Cell, 7 (1995) 2115. M. Ashburner (eds.) Drosophila. A laboratory handbook, Cold Spring Harbor Laboratory Press, Cold Spring Harbor (2003). RJ. Hamilton (eds.), Waxes: Chemistry, molecular biology and functions, The Oily Press, Dundee, (2003). T. Rezanka, J. Zahradnik and M. Podojil, Folia Microbiol, 27 (1982) 450. R. Vazquez-Duhalt, Current Topics in Phytochemistry, 14 (1995) 69. A. Banerjee, R. Sharma, Y. Chisti and U.C. Banerjee, Crit. Rev. Biotechnol., 22 (2002) 245. M.-O. Park, M. Tanabe, K. Hirata and K. Miyamoto, Appl. Microbiol. Biotechnol., 56 (2001) 448. T.G. Tornabene, S.J. Morrison and W.E. Kloos, Lipids, 5 (1970) 929. E.G. Dediukhina and V.K. Eroshin, Usp. Sovrem. Biol, 76 (1973) 351. J. Baraud, A. Maurice and C. Napias, Bull. Soc. Chim. Biol., 52 (1970) 421. E.G. Dediujina, V.P. Zhelifonova, L.V. Andreev and B.K. Eroshin, Prikl. Biokhim. Mikrobiol, 9(1973)813. H.L. Ehrlich in C.R. Bell, M. Brylinsky and P. Johnson-Green (eds.) Microbial Biosystems: New Frontiers, Atlantic Canada Society for Microbial Ecology, Halifax, Canada (1999). R.W. Stone and C.E. ZoBell, Ind. Eng. Chem., 44 (1952) 2564. C.E. ZoBell, World Oil, 130 (1950) 128. C.E. ZoBell, J. Sediment Petrol., 22 (1952) 42. G.J. Jankowski and C.E. ZoBell, J. Bacteriol., 47 (1944) 447. C.E. ZoBell, Science, 102 (1945) 364. P.W. Albro and C.K. Huston, J. Bacteriol., (1964) J.G. Jones and B.V. Young, Arch. Mikrobiol., 70 (1970) 82. J. Han and M. Calvin, Proc. Natl. Acad. Sci. U. S. A., 64 (1969) 436. T.G. Tornabene, J. Mol. Evol, 11 (1978) 253. T.G. Tornabene, T.A. Langworthy, G. Holzer and J. Oro, J. Mol. Evol., 13 (1979) 73. J. Han, E.D. McCarthy, W. Van Hoeven, M. Calvin and W.H. Bradley, Proc. Natl. Acad. Sci. U. S. A., 59 (1968) 29.
384
[34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63]
J. Oro, T.G. Tornabene, D.W. Nooner and E. Gelpi, J. Bacteriol, 93 (1967) 1811. J.B. Davis, Chem. Geol., 3 (1968) 155. J.G. Jones, J. Gen. MicrobioL, 59 (1969) 145. J. Templier, C. Largeau and E. Casadevall, Phytochemistry, 23 (1984) 1017. X. Wang and P.E. Kolattukudy, FEBS Lett., 370 (1995) 15. J. Vioque, T. Sirakova and P.E. Kolattukudy, J. Phycol., 35 (1999) 121. X. Wang and P.E. Kolattukudy, Biochem. Biophys. Res. Commun., 208 (1995) 210. J. Vioque and P.E. Kolattukudy, Arch. Biochem. Biophys., 340 (1997) 64. K.D. Lardizabal, J.G. Metz, T. Sakamoto, W.C. Hutton, M.R. Pollar and M.W. Lassner, Plant Physiol, 122 (2000) 645. S. Reiser and C. Sommerville, J. Bacteriol., 179 (1997) 2969. J.W. Lin, Y.F. Chao and S.F. Weng, Biochem. Biophys. Res. Commun., 191 (1993) 314. J. Benach, S. Atrian, R. Gonzalez-Duarte and R. Ladenstein, J. Mol. Biol., 282 (1998) 383. M.W. Dennis and P.E. Kolattukudy, Arch. Biochem. Biophys., 287 (1991) 268. F. Schneider-Belhaddad and P.E. Kolattukudy, Arch. Biochem. Biophys., 377 (2000) 341. M. Dennis and P.E. Kolattukudy, Proc. Natl. Acad. Sci. U. S. A., 89 (1992) 5306. M. Bard, D.A. Bruner, C.A. Pierson, N.D. Less, B. Biermann, L. Frye, C. Koegel and R. Barbuch, Proc. Natl. Acad. Sci. U. S. A., 93 (1996) 186. B.A. Arthington, L.G. Bennett, P.L. Skatrud, C.J. Guynn, R.J. Barbuch, CD. Ulbright and M. Bard, Gene, 102 (1991) 39. J.R. Reed, P. Hernandez, G.J. Blomquist, R. Feyereisen and R.C. Reitz, Insect Biochem. Mol. Biol., 26(1996)267. S. Gencic and D.A. Grahame, J. Biol. Chem., 278 (2003) 6101. E. Kocsis, M. Kessel, E. DeMoll and D.A. Grahame, J. Struct. Biol., 128 (1999) 165. P.W. Albro, J. Bacteriol., 108 (1971) 213. P.W. Albro, T.D. Meehan and J.C. Dittmer, Biochemistry, 9 (1970) 1893. S. Kikuchi, D.L. Rainwater and P.E. Kolattukudy, Arch. Biochem. Biophys., 295 (1992) 318. P.W. Albro and J.C. Dittmer, Biochemistry, 8 (1969) 1913. P.W. Albro and J.C. Dittmer, Biochemistry, 8 (1969) 3317. J. Frenz, C. Largeau, E. Casadevall, F. Kollerup and A.J. Daugulis, Biotehnol. Bioeng., 34(1989)755. J. Frenz, C. Largeau and E. Casadevall, Enzyme Microb. Technol., 11 (1989) 717. R.L. Mendes, J.P. Coelho, H.L. Fernandes, IJ. Marrucho, J.M.S. Cabral, J.M. Novais and A.F. Palavra, J. Chem. Technol. BiotechnoL, 62 (1995) 53. H. Kitazato, S. Asaoka and H. Iwamoto, Sekiyu Gakkaishi, 32 (1989) 28. P. Pedroni, J. Davison, H. Beckert, P. Bergman and J. Benemann, J. Energy Environ. Res., 1 (2001) 136.
Studies in Surface Science and Catalysis 151 R. Vazquez-Duhalt and R. Quintero-Ramirez (Editors) © 2004 Elsevier B .V. All rights reserved.
3 85
Chapter 14
The microbial diversity of deep subsurface oil reservoirs N.-K. Birkeland Department of Biology, University of Bergen, Box 7800, N-5020 Bergen, Norway
1. THE DISCOVERY OF MICROBIAL LIFE IN DEEP OIL WELLS "Souring" of oil reservoirs by the formation of hydrogen sulfide has been a problem since the beginning of commercial oil production (see chapter 11). Whether the sulfide is a result of abiotic chemical processes or due to microbiological activities has been debated for many decades. The first indications of an active role of sulfidogenic bacteria in this process were presented already in 1926 based on the observation that sulfate-reducing bacteria were widespread in oil-well production waters [1], Thermophilic sulfatereducing bacteria recovered from water produced from a North Sea oil-well was in 1991 found to be able to survive, multiply and actively produce sulfide under simulated reservoir conditions at temperatures and pressure up to 80°C and 4,500 psi, respectively [2, 3], thus demonstrating that sulfide formation can be caused by microorganisms even under the extreme physical conditions found in deep and hot petroleum reservoirs. Hyperthermophilic organisms able to grow at even higher temperatures were soon to be recovered from deep oil wells in Alaska [4] and the North Sea [5]. The question whether these bacteria were contaminants that had been introduced to the oil wells during drilling or through the repressurization by water injection into the oil-bearing strata, or whether they belonged to an indigenous community of subsurface microbes remained an open question. During the last decade, however, our perception of this has changed with the recovery from numerous oil wells of a large number of anaerobic microorganisms representing a wide range of different metabolic types, including sulfate- and iron-reducing bacteria, fermentative bacteria and methanogenic Archaea. A number of different thermophilic and hyperthermophilic organisms have now been recovered from both terrestrial and offshore
386
wells that have never been water-flooded [6, 7], strongly indicating an indigenous origin. The presence of an indigenous microbial community is further supported by the isolation of novel species that never has been recovered from any other sources, and by the physiological characteristics of some isolates indicating a close adaptation to the respective in situ reservoir conditions. Recovery of closely related strains from remote oil fields [7-9] also supports the existence of a widespread microbial biosphere in oil-bearing strata. However, the problems associated with recovery of biological samples from oil wells are extensive. Sampling from wellheads is the only way of collecting samples from petroleum reservoirs, and the possible sources of contamination are numerous. It is furthermore possible that exogenous mesophilic bacteria can propagate in top facilities of the oil field installations. Occasionally, aerobic and microaerophilic bacteria are recovered from produced oil-well water, but available chemical data suggest that oxygen is absent in oil reservoirs, and these isolates should thus not be considered as being truly indigenous to deep oil wells. In addition to SRB and fermentative bacteria, Voordouw et al. [10] detected several aero- and microaerophilic bacteria in a 600 m deep water flooded oil reservoir in Canada. Nitrate-reducing bacteria were recovered from a similar shallow oil field [11]. It is postulated that oxygen and nitrate is able to reach these shallow oil-bearing formations through diffusion or convection from surface layers, giving support to a limited community of bacteria respiring with nitrate or oxygen [10, 11]. The number of bacterial cells in water produced from oil reservoirs is highly variable. Total bacterial counts demonstrated the presence of more than 106 cells per ml in water from a non-water flooded reservoir in California [6], and from sulfide-rich production water from a German water-flooded petroleum reservoir up to 6.3 x 106 colony-forming units of sulfate-reducing bacteria per ml has been obtained [12]. Although only a few bacteria per ml have been detected in water produced from some oil wells, these results show that the bacterial density can be significant. In the present chapter our current knowledge of these microorganisms is reviewed. 2. SULFATE-REDUCING BACTERIA AND ARCHAEA (SRB) Sulfate-reducing prokaryotes constitute a diverse physiological group of sulfideproducing microorganisms able to carry out anaerobic respiration with sulfate as a terminal electron acceptor. They are widespread in anaerobic environments were sulfate is available. Typically, they oxidize organic acids either to acetate, or by complete oxidation of acetate or other acids to CO2, but autotrophic species using H2 and CO2 as energy and carbon-sources are also common. A wide range of organic acids, e.g. acetate, propionate, butyrate, pentanoate and hexanoate, at concentrations up to 20mM has been found in oil reservoirs [13, 14]. On the other hand, the concentration of sulfate seems to be rather low in
387
non-water-flooded reservoirs, and the abundance of SRB is therefore probably sulfate-limited. There are few reports on the quantification of SRB in produced water, but the few estimates that have been made show a considerable variation. As mentioned above, up to 6.3 x 106 colony-forming units of SRB per ml was obtained from wellhead water from a German sulfide-rich reservoir [11]. Acetate was used as a substrate. In a more recent survey, up to 4.5 x 104 SRB per ml were counted in wellhead water from low-temperature reservoirs using most probable number (MPN) technique and with lactate + acetate as substrates [15]. The number of SRB decreased with increasing in situ temperature, and from wells with an in situ temperature of 85° no SRB could be detected. These results are comparable to the results reported by Nazina et al. [16] and Rozanova and Nazina [17], who detected only very low populations of SRB in hightemperature reservoirs. A low number of SRB have also been detected in reservoirs of high salinity. In a study of sulfate reducers in a high-temperature oil field in the North Sea up to 2 x 104 thermophilic or hyperthermophilic SRB were detected directly in the produced water using fluorescent antibody technique with conjugated antibody directed against three specific SRB groups [18]. No correlation between the duration of seawater injection and the number of SRB in the water was observed, indicating that thermophilic and hyperthermophilic SRB are indigenous to this oil field. The salinity of oil-well water is very variable and is an important factor for the in situ microbial activity and diversity. Oil-well brines with salinity above 20% have been found, but most oil well formation waters have a moderate salinity (<6%). No extremely halophilic SRB has ever been recovered from oil wells, the most halophilic SRB being Desulfovibrio vietnamensis [19] and Desulfotomaculum halophilum [20], which both grow optimally at a salinity around 5%. Voordouw et al. [21] detected different communities of SRB in enrichments from oil wells with low and high salinity, demonstrating that salinity is an important discriminating factor. 2.1. Mesophilic SRB Mesophilic bacteria generally have growth optima in the 25-40°C temperature range. Mesophilic SRB have frequently been isolated from oil field production waters (Table 1). These organisms are believed proliferate in the top facilities of the petroleum installations. They also seem to proliferate in shallow low-temperature oil wells, but can also be found in deeper wells that have been subject to long-term water flooding. However, these organisms cannot be regarded as indigenous components of the deep oil well microbial community. Most of the mesophilic isolates are Gram-negative bacteria belonging to the delta subdivision of proteobacteria. They include species from the genera Desulfobacter, Desulfobacterium, Desulfobulbus, Desulfomicrobium and
388
Desulfovibrio. One Gram-positive mesophilic SRB has been recovered, Desulfotomaculum halophilum. Except for Desulfobacter vibrioformis and Desulfobacterium cetonicum, which oxidize their they substrates completely to CO2, they oxidize organic acids to the level of acetate. D. vibrioformis is restricted to oxidation of acetate only, while the other species can oxidize a range of substrates. Most of them can oxidize lactate and pyruvate, which are common substrates for SRB. Desulfovibrio gabonensis and Desulfovibrio vietnamensis can use a wide range of substrates, including formate, malate, fumarate and ethanol. D. cetonicum has a unusual substrate range, as it, in addition to common substrates like buturate, lactate, pyruvate, alcohols and fatty acids can oxidize ketones, benzoic acid, toluene and p/m-cresol. This property makes D. cetonicum an interesting organism from a bioremediation perspective. Many of the isolates can also utilize H2 as energy source when grown in presence of a carbon source such as formate and acetate, but only Desulfomicrobium apsheronum is able to grow with H2 autotrophically using CO2 as the sole carbon source. 2.2. Thermophilic SRB Thermophilic SRB isolated from oil well production waters also include members of the delta subdivision of proteobacteria and the Gram-positive genus Desulfotomaculum, but in addition, members of the genus Thermodesulfobacterium, a deeply branching lineage of thermophilic Gram-negative bacteria, have been recovered (Table 1). The most frequently isolated thermophilic SRB belong to the Gram-positive spore-forming genus, Desulfotomaculum. Desulfotomaculum nigrificans, isolated from Western Siberia in 1978, is a rather restricted organism oxidizing lactate and ethanol incompletely. Desulfotomaculum kuznetsovii was originally isolated from an underground thermal water system [22], and has later been recovered from wellheads in the Paris Basin [23]. D. kuznetsovii is nutritionally a very versatile organism. It is a methylotrophic bacterium able to oxidize methanol in addition to substrates like H2, formate, acetate, aliphatic fatty acids, ethanol, lactate, fumarate and malate. It oxidizes its substrates completely to CO2. One novel species, Desulfotomaculum thermocisternum, has been recovered from a North Sea oilwell. It has a more restricted substrate range than D. kuznetsovii, but shares its ability to grow on lactate, alcohols and aliphatic carboxylic acids. D. thermocisternum was isolated from produced oil reservoir water obtained before breakthrough of injected seawater, suggesting that it is a true indigenous oilfield organism. Thermodesulfobacterium thermophilum, originally described in 1974, was the first thermophilic organism to be isolated from water produced from an oil well. It was first described as Desulfovibrio thermophilus, but later renamed to Thermodesulfobacterium mobile before its final name, T. thermophilum, was
389
validated [24]. The genus Thermodesulfobacterium is the third deepestbranching phylogenetic lineage in the bacterial domain. Thermodesulfobacterium commune was originally isolated from a geothermal area in Yellowstone National Park [25], but later it has been recovered from a continental non-water flooded reservoir in the Paris Basin [6]. T. thermophilum has been recovered also from the Paris Basin and the North Sea [26]. This widespread occurrence indicates a cosmopolitan subterranean distribution of these species which obviously must be indigenous to oil reservoirs. Both species can utilize H2, formate, lactate and pyruvate as energy sources. The organic acids are oxidized incompletely to CO2. Two thermophilic SRB belonging to novel genera of the delta subdivision of proteobacteria, Desulfacium infernum and Thermodesulforhabdus norvegicus have been isolated from North Sea oil wells [27, 28]. Both species can utilize a range of organic acids, including acetate, which they oxidize completely to CO2.
Table 1 Sulfate-reducing prokaryotes recovered from oil field production waters Species Archaeoglobus fulgidus strain 7324 'Archaeoglobus lithotrophicusM Archaeoglobus profundus Desulfacinum infernum Desulfobacter vibrioformis Desulfobacterium cetonicum Desulfobulbus rhabdoformis Desulfomicrobium sp. Desulfomicrobium apsheronum Desulfotomaculum spp. Desulfotomaculum halophilum Desulfotomaculum kuznetsovii Desulfotomaculum nigrificans Desulfotomaculum thermocisternum Desulfovibrio gabonensis Desulfovibrio longus Desulfovibrio vietnamensis Thermodesulfobacterium thermophilum Thermodesulfobacterium commune Thermodesulforhabdus norvegicus a
Not yet validly described.
Temperature optimum [°C] 76 nd nd 60 33 30-35 31 25-35 25-30 65 35 60-65 60 62 30 35 37 65 70 60
Location of oil field North Sea North Sea North Sea North Sea North Sea Russia North Sea North Sea Apsheron North Sea Paris Basin, France Russia Russia North Sea Gabon, WestAfrica offshore Paris Basin, France Vietnam, offshore Caspian Sea North Sea Paris Basin, France North Sea
Complete oxidizer + nd + + + + + -
Ref.
-
80 19 26, 81 6 28
+
5 4 4 27 73 74 75 76 77 2 20 22 78 79 15
390
2.3. Hyperthermophilic SRB Hyperthermophilic microbes grow at temperatures above 80°C and most of them belong to the domain Archaea. Until recently, Archaeoglobus was the only known archaeal genus able to carry out dissimilatory sulfate reduction, but an additional archaeal sulfate reducer, Caldivirga maquilingensis, isolated from a hot spring in the Philippines, has now been described [29]. Only species belonging to the genus Archaeoglobus have so far been found in oil-field production waters. The type species of the genus, Archaeoglobus fulgidus (type strain VC-16), was first isolated from a shallow hydrothermal system in the Mediterranean [30, 31]. In 1994, A. fulgidus strain 7324 was recovered from an oil well in the Norwegian sector of the North Sea [5]. A. fulgidus has also been recovered from the East Shetland Basin of the North Sea along with its relatives, Archaeoglobus profundus and 'Archaeoglobus lithotrophicus' [4]. A. fulgidus strain 7324 has a lower optimal growth temperature (76°C) than the type strain, VC-16 [83°C], and unlike VC-16 it cannot grow autotropohically with H2 and CO2. Both strains can grow on lactate and pyruvate. In contrast to VC-16, strain 7324 is able to grow on starch [32]. A. profundus is a obligate mixotrophic species growing only on a mixture of H2 and an organic carbon source (e.g. acetate) [33]. A. litotrophicus can grow autotrophically on H2 and CO2 [5], but has not yet been validly described. 3. METHANOGENIC ARCHAEA Biological generation of methane is limited to a group of strict anaerobic archaeal organisms, the methanogens, which form methane as a product of anaerobic respiration. They are widespread in nature, and have been found in most natural anoxic environments. Their most frequently used energy source is hydrogen, which is usually oxidized with CO2 as electron acceptor. These hydrogenotrophic methanogens are also autotrophs, using CO2 as carbon source by assimilation of carbon via the acetyl-CoA pathway. Other possible energy sources include one-carbon compounds like formate, methanol and methylamines. Methanol is converted to methane, carbon dioxide and water. A few methanogens can utilize acetate, which is converted to methane and CO2 through an acetoclastic reaction. Ethanol and propanol can also be used by some species. Some commonly used methanogenic reactions and their standard changes in free energies are given in Table 2. Biological methane formation in oil-bearing strata has been well documented [34, 35, 36, 37], but few methanogens able to grow under in situ conditions have been isolated. Although hyperthermophilic methanogens are frequently isolated from geothermal environments like hot springs and hydrothermal vents, hyperthermophilic methanogens have never been isolated from oil wells. Methanobacterium thermoalcaliphilum, which grows optimally
391
at 65°C and has an upper growth-limit at 80°C, has been recovered from oilfields in Tataria and Western Siberia [38]. Two other thermophilic methanogens have been recovered from oil wells, which both share with M. thermoalcaliphilum, the ability to use hydrogen as energy source and CO2 as carbon source (Table 3). Although positive enrichments of acetate-utilizing methanogens at 60°C have been obtained, it has not yet been possible to obtain pure cultures of acetoclastic thermophilic methanogens from oil wells [37, 38, 39]. A plausible reason is that thermophilic methane production from acetate in these environments might be a result of interspecies hydrogen transfer [37]. Several mesophilic methanogens have been isolated, including hydrogenotrophic types as well as strains utilizing methylamines and acetate (Table 3). 4. FERMENTATIVE BACTERIA AND ARCHAEA Fermentative organisms are able to utilize organic substances such as carbohydrates and peptides for growth producing organic acids, ammonium and hydrogen as fermentation products. In contrast to SRB and methanogens, these organisms do not use any external electron acceptor in their energy-yielding reactions, thus maintaining an internal red-ox balance. Occasionally, fermentative microbes can transfer excess reduction power to sulfur compounds like thiosulfate or elemental sulfur, which both are reduced to sulfide, thereby improving growth rate and substrate utilization. However, the sulfur compounds only serve as electron "sinks", and the sulfur-reducing reactions do not appear to be linked to any energy-conserving mechanism. This sulfur-reducing ability can be an important step in the geochemical cycling of sulfur in these anaerobic thermal environments. Fermentative bacteria from a great variety of phylogenetic lineages have been isolated from oil reservoirs, especially during the recent years (Table 4).
Table 2 Examples of methanogenic reaction and their standard free energy changes. Reaction 4 H 2 + CO 2 -> CH 4 + 2H 2 O 4 Methanol - • 3CH 4 + CO 2 + 2H 2 O 4 Methylamine + 2H 2 O -> 3CH 4 + CO 2 + 4NH 4 + Acetate -> CH 4 + CO 2
AG 0 ' [KJ/mol CH 4 ] -135.6 -104.9 -75.0 -31.0
Table 3 Methanogens recovered from oil reservoirs Species Methanobacterium bryantii Methanobacterium ivanovii Methanobacterium thermoaggregans Methanobacterium thermoalcaliphilum Methanobacterium thermoautrophicum 'Methanocalculus halotolerans' Methanococcus thermolithotrophicus Methanohalophilus euhalobius ''Methanoplanus petrolearius' Methanosarcina mazei Methanosarcina siciliae
Temperature optimum [°C] 37 45 60 65
Location Tatarstan and Western Siberia Tatarstan California
60
Tatarstan Siberia Tatarstan
38
France
60
North Sea
and Western
28-37 37
Western Siberia Gulf of Guinea
37 40
Tatarstan Gulf of Mexico
Substrates used [methane produced from] H2
References
H2 H2
34,82
H2
38
H2
36
H2, formate
84
H2
39
methylamines H2 + CO2, formate, 2propanol + CO2 Acetate, methylamines methylamines
85 86
38
83
87 88
393
4.1. Order Thertnotogales Six novel species belonging to the genera Thermotoga, Petrotoga and Thermosipho, which all belong to the order Thermotogales have been isolated from oil reservoirs and described the last 4 years [40, 41, 42, 43]. Members of a related genus, Geotoga, were isolated earlier [44]. Bacteria belonging to this order are rod-shaped organisms possessing a characteristic outer sheath-like structure called a 'toga' [45] and represent one of the deepest phylogenetic branches in the bacterial line of evolutionary descent. Most members are thermophiles with optimal growth between 50 and 70°C, but some species of the genus Thermotoga are hyperthermophiles. In total, 13 different species of this order have been recovered from sources around the world, including oil reservoirs in France, Western Siberia, Japan, Africa, USA, the Gulf of Mexico and the North Sea. Most of these species have so far only been recovered from oil reservoir. The widespread distribution and uniqueness of these bacteria is clearly in support of their cosmopolitan nature as subterranean indigenous bacteria belonging to a natural microbial community in oil-bearing strata. Their temperature growth ranges also correlate well with the in situ temperature of the reservoirs. Nutritionally, these organisms are very versatile heterotrophs, fermenting a variety of organic substrates ranging from mono- and disaccharides, polysaccharides and protein hydrolysates. Most of the oil-well isolates belonging to the Thermotoga genus can ferment a large number of carbohydrates and grow also on peptide substrates like peptone and bio-Trypticase. Small amounts of yeast extract is usually required for growth on carbohydrates. Thermotoga subterranea, however, seems to be restricted to growth on complex media such as peptone and yeast extract and cannot grow on defined carbon sources. Thermosipho geolei is nutritionally very similar to the Thermotoga members, but is rather limited with regard to carbohydrate utilization, as it was found to only grow on glucose when carbohydrates were tested as substrates [42]. While members of the Thermotoga and Thermosipho genera have been isolated also from deep-sea hydrothermal vents, members from the Petrotoga and Geotoga genera have so far only been recovered from petroleum reservoirs. It is possible that these genera are unique to these special microbial habitats. An interesting characteristic of Thermotogales members is the widespread ability to degrade and utilize polymeric substrates such as starch, maltodextrin, xylan and peptides. Even a cellulolytic bacterium has been recovered (T. petrophila). Xylanases from thermophilic microbes has received much interest lately as it has a considerable biotechnological potential for the paper pulping industry. The common capacity of these organisms to utilize a wide range of polymeric substances for growth indicates a saprophytic life style and a role as consumers in this subsurface microbial ecosystem.
394
Table 4 Fermentative bacteria and Archaea recovered from oil reservoirs
Species
Acetoanaerobium romashkovii Anaerobaculum thermoterrenum Dethiosulfovibrio peptidovorans Fusibacter paucivorans Geotoga petraea Geotoga subterranea Haloanaerobium acetoethylicum Haloanaerobium congolense Haloanaerobium kushneri Haloanaerobium salsuginis Petrotoga mexicana Petrotoga miotherma Petrotoga mobilis Petrotoga olearia Petrotoga sibirica Spirochaeta smaragdinae Thermoanaerobacter acetoethylicus Thermoanaerobacter subterraneus Thermoanaerobacter brockii Thermococcus sp. Thermococcus sibericus Thermosipho geolei Thermotoga elfii Thermotoga hypogea Thermotoga maritima Ml2597 Thermotoga naphthophila Thermotoga petrophila Thermotoga subterranea
Optimal Temp.
37
55 42 37 50 45 34 42 35-40
40 55 55 58-60
55 55 37 Nd 65 55-60
85 78 70 66 70 nd 80 80 70
Location reservoir
of
oil
Reduction of sulfur compounds S2O3" S°
Western Siberia Utah Congo, offshore Congo, offshore Oklahoma/Texas Oklahoma/Texas Gulf of Mexico Congo, offshore Oklahoma Oklahoma Gulf of Mexico Oklahoma/Texas North Sea Western Siberia Western Siberia Congo, offshore Western Siberia
Nd + + + + + Nd + Nd + + + + + + Nd
France France Niigata, Japan Western Siberia Western Siberia Africa Cameroon Western Siberia Niigata, Japan Niigata, Japan Paris Basin, France
+ Nd + + Nd + + -
nd + + + nd nd nd + -? nd +
nd + + Nd + Nd nd + + Nd [weak +]
+ +
Ref.
59 56 58 89 44 44 48 49 51 50 40 44 90 41 41 53 68 55 54 46 47 42 95 97 68 43 43 98
4.2. Archaea Hyperthermophilic fermentative Archaea belonging to the genus Thermococcus were first isolated from Japanese oil reservoirs in 2000 [46]. Although the isolates were not described at the species level, they were nutritionally very similar to other thermococci, growing on proteinaceous substrates, yeast extract and amino acids. Although the in situ reservoir temperature ranged from 50 to 58°C, the optimal temperature of the isolates was above 80°C [43, 46]. The number of hyperthermophilic cocci that were present
395
in produced water from the oil wells was estimated to be up to 4.6 x 104 cells/ml. The organisms were not able to grow in produced water due to lack of required nutrients, but under starved conditions at 50°C the viable cell count was stable for 200 days, indicating that they have developed an amazing ability to survive prolonged periods under starved conditions. This feature is probably important for the continued existence in a hot subterranean oil reservoir where the supply of nutrients is limited [43]. A novel thermococcal species, Thermococcus sibiricus, has been isolated from an oil reservoir in Western Siberia [47], and a novel species has also recently been recovered from a North Sea oil well (Birkeland, unpublished). This indicates that the high-temperature oil reservoir biosphere is also inhabited by indigenous hyperthermophilic Archaea. There is also evidence for their presence in other oil wells [4, 7, 8]. 4.3. Halophilic bacteria Some oil reservoirs contain highly saline brines with salinity above 20%. Anaerobic fermentative bacteria that can grow at this high salinity have been isolated from such oil wells in Africa, the Gulf of Mexico and USA [48, 49, 50, 51]. These bacteria belong to the genus Haloanaerobium, one of a few genera of anaerobic fermentative bacteria that are adapted to high-salt conditions. These organisms have frequently been isolated from bottom sediments of hypersaline lakes and lagoons. They are typically saccharolytic bacteria, fermenting a range of carbohydrates. Only mesophilic members, with growth optima between 34 and 42°C, have so far been recovered from oil reservoirs, but related thermophilic bacteria have been isolated from other sources. In contrast to other halophilic bacteria, which accumulate organic osmotic solutes in order to maintain an osmotic balance between the surrounding medium and the cytoplasm, the members of the Haloanaerobium genus accumulate Na+, K+ and Cl as compatible solutes [48, 52]. Accumulation of inorganic ions is a property they share with extremely halophilic Archaea, which accumulate up to 3M KC1. Another unusual feature of the Haloanaerobium genus is the nature of their cellwall structure as compared to their phylogenetic position. They stain Gramnegative when subjected to the Gram-staining procedure, but electron micrographs show the presence of a typical Gram-negative cell wall. However, based on the sequence of the 16S rRNA they form a deep-branching cluster within the phylum of Gram-positive bacteria. This is taken as evidence that certain descendants of the ancestors of the Gram-positive bacteria maintained their Gram-negative cell wall structure, which is also the case with certain other strict anaerobic relatives. A moderately halophilic spirochete, Spirochaeta smaragdinae, growing optimally at a salinity of 5% has been isolated from an offshore oil well in Congo [53]. This bacterium is nutritionally very versatile, fermenting
396
carbohydrates, glycerol, fumarate, peptides and yeast extract, and is the only spirochete so far isolated from the deep subsurface. Evidence for the presence of a closely related spirochete in North Sea oil wells has, however, been obtained using direct molecular techniques (Birkeland, unpublished). 4.4. Other mesophilic and thermophilic fermentative bacteria Thermoanaerobacter subterraneus and Thermoanaerobacter brockii subsp. lactiethylicus are Gram-positive thermophilic carbohydrate-fermenting bacteria isolated from French oil wells [54, 55]. They grow optimally at 65 and 55-60°C, respectively. T. brockii forms endospores. Spores have not been observed in cultures of T. subterraneus, but because this organism can survive autoclaving for 45 minutes, the presence of heat-resistant forms has been suggested [55]. Anaerobaculum thermoterrenum is a Gram-positive bacterium isolated from an oil well in Utah [56]. It defines a novel moderately thermophilic genus, Anaerobaculum, phylogenetically related to Thermoanaerobacter. It is nutritionally versatile, growing on a wide range of carbohydrates including cellulose, as well as peptone and organic acids like citrate. It is able to utilize crotonate as electron acceptor, reducing it to butyrate [57]. Although it groups within the phylum of Gram-positive bacteria, A. thermoterrenum has a Gram-negative cell wall, a feature it shares with the halophilic genus Haloanaerobium. A mesophilic thiosulfate-reducing bacterium termed Dethiosulfovibrio peptidovorans, which can only utilize peptides and amino acids for growth, has been isolated from a corroding offshore oil well in Congo [58]. The isolate was shown to cause a strongly enhanced corroding activity of steel in the presence of thiosulfate, indicating that apart from SRB, thiosulfate-reducing bacteria can contribute to this process. Phylogenetically, this organism groups within Gram-positive bacteria of the clostridial group, but shares with Anaerobaculum a multilayered cell wall ultrastructure typical of Gram-negative bacteria. Together with Anaerobaculum and a few other small genera, Dethiosulfovibrio spp. form a separate phylogenetic cluster in the Clostridium group of Gram-positive bacteria. Another Gram-positive bacterium, Fusibacter paucivorans, isolated from an African oil well, defines a novel genus within the Clostridium phylum. F. paucivorans is a mesophilic and halotolerant Gram-positive bacterium fermenting a limited number of carbohydrates. Spores have never been observed. Acetoanaerobium romashkovii is a homoacetogenic mesophilic bacterium isolated from a Siberian oil field in 1992 [59]. Homoacetogenic bacteria is a group of anaerobes that can use CO2 as an electron sink and reduce it to acetate as a fermentation product via the carbon monoxide dehydrogenase pathway. Hydrogen can be used as energy source, but as is the case for A. romashkovii, various one-carbon compounds, amino acids and sugars can also be utilized.
397
Although the members of the genus Acetoanaerobium stain Gram-negative, they possess a Gram-positive cell-wall architecture [60]. 5. IRON REDUCERS It has been suggested that reduction of iron is an ancient and widespread mechanism for anaerobic respiration [61]. The contribution of Fe(III) reduction to the cycling of iron and organic matter in various anaerobic environments, including the subsurface [62, 63], is now well known, and geochemical evidence suggest that Fe(III) was the first electron acceptor of global significance during the early evolution of microbial energy metabolism [64]. Mesophilic Fe(III)reducing bacteria have been detected in oil field fluids [65, 66], some of which have been identified as Shewanella putrefaciens (formerly Alteromonas putrefaciens) [65, 66]. S. putrefaciens can use hydrogen and formate as electron donors during iron respiration, and can also reduce sulphur and sulphite. A moderately thermophilic Fe(III)-reducing bacterium, Deferribacter thermophilus, has been isolated from a North Sea oil well [67]. In addition to iron, it can use manganese and nitrate as electron acceptors, and use proteinacous substrates, hydrogen and organic acids as energy sources. Most iron reducers that have been tested can reduce manganese in addition to iron. In a survey of the iron-reducing capability of thermophilic and hyperthermophilic isolates from oil reservoirs in Western Siberia, 8 of nine strains were found to reduce Fe(III) using pepton or hydrogen as energy source [68]. These iron reducers included 5 strains belonging to the thermophilic bacterial genera Thermotoga and Thermoanaerobacter, and 3 hyperthermophilic archaeal strains belonging to genus Thermococcus. The isolates had not been subjected to an enrichment step in Fe(III)-containing media. In the same investigation it was shown that the major part of 25 samples taken from these oil reservoirs were positive for Fe(III) reduction in peptone or hydrogen enrichments. These results suggest that iron reduction is a common feature of thermophilic and hyperthermophilic microorganisms in deep subsurface petroleum reservoirs. This is further supported by the observation that the sulphate-reducing archaeon A. fulgidus and the methanogen M. thermolithotrophicus, which both have been found in hot water produced from North Sea oil wells, also can reduce Fe(III) [62]. Whereas Thermotoga maritima was previously considered to possess only a fermentative metabolism, it was found later to grow respiratory in the presence of Fe(III), coupling Fe(III)reduction with energy conservation [62]. The function of iron reduction in the energy metabolism of these bacteria needs further investigations.
398
6. CULTURE-INDEPENDENT APPROACHES Direct analyses of uncultured natural microbial communities based on fluorescence microscopy using fluorescent antibodies (FA) or oligonucleotide probes directed against specific bacterial groups, and amplification and analyses of genes from DNA extracted from environmental samples have contributed significantly to an improved understanding of the structural complexity of natural microbial communities during the last decade. The development of the polymerase chain reaction (PCR), automated DNA sequencing and DNA microchip technologies has provided efficient tools for culture-independent analyses of microbial diversity. Genus specific antibodies directed against the hyperthermophilic Archaeoglobus, and the thermophilic genera Desulfotomaculum and Thermodesulforhabdus have been used for analyzing the distribution of SRB in produced oil reservoir waters sampled at different dates and from different wells in the Gullfaks field in the North Sea [18]. Archaeoglobus and Thermodesulforhabdus strains were detected in 4 of 16 samples, but in most samples the numbers were below the detection limit. The number of cells varied from 400 to 2 x 104 per ml. Desulfotomaculum strains were only detected in one of the wells. In 3 wells only one of the three types could be detected. This investigation demonstrated that the distribution of these SRB in the Gullfaks field is subject to strong spatial and temporal variations. Oligonucleotide microchips containing specific 16S rRNA probes targeting selected microbial groups encompassing key genera of thermophilic bacteria and Archaea were used for probing the diversity in water samples from the Samotlor high-temperature oil reservoir in Western Siberia [37]. The results confirmed the presence of organisms identified by culture-based methods, but organisms that had not been identified by culture-dependent methods were also detected. These organisms included representatives of the aerobic genus Thermus and the microaerophilic Aquifwales group, as well as anaerobes belonging to the genera Desulfurobacterium and Thermovibrio. None of these groups have previously been detected in oil reservoirs. Orphan et al. [7] made 16S rDNA libraries from total DNA from water produced from hightemperature petroleum reservoirs in California, using either universal or archaeal primer sets, 83 unique clones were identified from the universal library, and sequence analysis revealed that the majority of the clones grouped within the bacterial domain, with only 8.8% of the library affiliated with the domain Archaea. The dominating bacterial phylotypes were close relatives of Grampositive fermentative genera Acidaminococcus and Thermoanaerobacter, and to the halophilic proteobacterium Halomonas. The genus Acidaminococcus has never been isolated from petroleum reservoirs. Clones related to aerobic bacteria were also identified. The archaeal library was dominated by methanogen-like
399
clones, with a lower percentage of clones belonging to the fermentative Thermococcales. Interestingly, archaeal sequence types related to the acetoclastic Methanosarcinales were identified, suggesting the presence of acetate-utilizing methanogens. The molecular analyses demonstrated a much higher diversity than indicated by cultivation-dependent methods. The same strategy was also used by Voordouw et al. [69] with samples from shallow lowtemperature oil field in Canada. A variety of Gram-negative SRB were detected, but a limited number of clones representing fermentative organisms were obtained. The results also indicated the presence of microaerophilic types. Molecular approaches have also been used for assessing the diversity in primary enrichments. Cloning and sequencing of 16S rDNA have been used for bacterial diversity analysis of SRB enrichments from North Sea oil field samples [70]. The library was dominated by Gram-positive SRB (Desulfotomaculum) and fermenters belonging to the Gram-positive genera Thermoanaerobacter and Clostridium. In addition, two clones, which probably represent a novel undescribed genus of deeply branching Gram-positive bacteria were obtained. Reverse sample genome probing (RSGP) is a microbial community fingerprinting technique that have been used successfully for both qualitative and quantitative analysis of SRB enrichments [21, 71, 72]. Using different carbon sources for enrichment of mesophilic SRB, 34 different types of SRB were detected by RSGP [21], giving a glimpse of the enormous diversity of SRB in low-temperature oil fields. 7. CONCLUSIONS Several lines of evidence indicate the existence of an indigenous deep subsurface microbial community in petroleum reservoirs; a) a large variety of prokaryotic groups have been isolated from produced waters around the world, b) many of the isolates belong to unique species or genera that have not been recovered from any other habitats, c) highly similar microorganisms have been isolated from geographically remote oil fields, d) several isolates are adapted to growth under the extreme in situ reservoir conditions, and e) using cultivationindependent approaches, molecular analyses have directly verified the presence of previously cultivated strains in produced water and have demonstrated the existence of a highly diverse community. A wide range of chemolitoauthotrophic and organotrophic types have been isolated from deep oil wells; hydrogenotrophic methanogens and sulfate-reducers, heterotrophic methanogens, fermentative bacteria and Archaea, heterotrophic sulfate-reducers, iron-reducing microbes, and saprophytes. Organisms with temperature optima from 25 to 85°C have been recovered, and optimal salinity ranges from freshwater conditions to more than 10% salinity. This enormous physiological diversity suggests that this microbial community constitutes a complex
400
ecosystem with an active biogeochemical cycling of carbon and minerals. Further characterization of this subsurface biosphere in order to understand the ecological role and significance of these organisms is a major geomicrobiological challenge.
REFERENCES [I] [2] [3] [4] [5] [6] [7] [8] [9] [10] [II] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26]
E.S. Bastin, Science, 63 (1926) 21. J.T. Rosnes, T. Torsvik and T. Lien, Appl. Environ. Microbiol., 57 (1991) 2302. J.T. Rosnes, A. Graue and T. Lien, SPE Production Engineering, May (1991) 217. K.O. Stetter, R. Huber, E. Blochl, M. Kurr, R.D. Eden, M. Fielder, H. Cash and I. Vance, Nature, 365 (1993) 743. J. Beeder, R.K. Nilsen, J.T. Rosnes, T. Torsvik and T. Lien, Appl. Environ. Microbiol., 60(1994)1227. S. L'Haridon, A.L. Reysenbach, P. Glenat, D. Prieur and P. Jeanthon, Nature, 377 (1995)223. V.J. Orphan, L.T. Taylor, D. Hafenbradl and E.F. Delong, Appl. Environ. Microbiol., 66 (2000) 700. G.S. Grassia, K.M. McLean, P. Glenat, J. Bauld and A.J. Sheehy, FEMS Microbiol. Ecol.,21(1996)47. M. Magot, Nature, 379 1996) 681. G. Voordouw, S.M. Armstrong, M.F. Reimer, B. Fouts, A.J. Telang, Y. Shen and D. Gevertz, Appl. Environ. Microbiol., 62 (1996) 1623. D. Gevertz, A.J. Telang, G. Voordouw and G.E. Jenneman, Appl. Environ. Microbiol., 66(2000)2491. R. Cord-Ruwisch, W. Kleinitz and F. Widdel, Erdol-Erdgas-Kohle, 102 (1986) 281. T. Barth, Appl. Geochem., 6 (1991) 1. T. Barth and M. Riis, Org. Geochem., 19 (1992) 455. C. Tardy-Jacquenod, P. Caumette, R. Matheron, C. Lanau, O. Arnauld and M. Magot, Can. J. Microbiol., 42 (1996) 259. T.N. Nazina, E.P. Rozanova and S.I. Kuznetsov, Geomicrobiol. J., 4 (1985) 103. E.P. Rozanova and T.N. Nazina, Microbiology (Engl. Tr.) 48 (1979) 907. R.K. Nilsen, J. Beeder, T. Thostenson and T. Torsvik, Appl. Environ. Microbiol., 62 (1996) 1793. P.N. Dang, T.C.H. Dang, T.H. Lai and H. Stan-Lotter, Anaerobe, 2 (1996) 385. C. Tardy-Jacquenod, M. Magot, B.K.C. Patel, R. Matheron and P. Caumette, hit. J. Syst. Bacteriol. 48 (1998) 333. G. Voordouw, J.K. Voordouw, J.K. Jack, J. Foght, P.M. Fedorak andD.W. Westlake, Appl. Environ. Microbiol., 58 (1992) 3542. T.N. Nazina, A.E. Ivanova, L.P. Kanchaveli and E.P. Rozanova, Microbiology [Engl. Tr.], 57(1988)823. M. Magot, B. Ollivier and B.K. Patel, Antonie Van Leeuwenhoek, 77 (2000) 103. H.G. Triiper, Int. J. Syst. Evol. Microbiol., 53 (2003), 927. J.G. Zeikus, M.A. Dawson, T.E. Thompson, K. Ingvorsen and E.C. Hatchikian, J. Gen. Microbiol., 129 (1983) 1159. B. Christensen, T. Torsvik and T. Lien, Appl. Environ. Microbiol. 58 (1992) 1244.
401
[27] G.N. Rees, G.S. Grassia, AJ. Sheehy, P.P. Dwivedi and B.K.C. Patel, Int. J. Syst. Bacteriol., 45 (1995) 85. [28] J. Beeder, T. Torsvik and T. Lien, Arch. Microbiol, 164 (1995) 331. [29] T. Itoh, K.-i. Suzuki, P.C. Sanchez and T. Nakase, Int. J. Syst. Bacteriol., 49 (1999) 1157. [30] K.O. Stetter, G. Laurer, M. Thomm and A. Neuner, Science, 236 (1987) 822. [31] K.O. Stetter, Syst. Appl. Microbiol., 10 (1988) 172. [32] A. Labes and P. Schonheit, Arch. Microbiol., 176 [2001) 329. [33] S. Burggraf, H.W. Jannasch, B. Nicolaus and K.O. Stetter, Syst. Appl. Microbiol., 13 (1990)24. [34] LA. Borzenkov, S.S. Belyaev, Y.M. Miller, LA. Davidova and M.V. Ivanov, Microbiology (Engl. Tr.), 66 (1997) 104. [35] S.S. Belyaev and M.V. Ivanov, Ecol. Bull, 35 (1983) 273. [36] M.V. Ivanov, S.S. Belyaev, A.M. Zyakun, V. Bondars and K. Laurinivicius, Geokhimiya 11 (1983) 1647. [37] E.A. Bonch-Osmolovskaya, M.L. Miroshnichenko, A.V. Lebedinsky, N. A. Chernyh, T.N. Nazina, V.S. Ivoilov, S.S. Belyaev, E.S. Boulygina, Y.P. Lysov, A.N. Perov, A.D. Mirzabekov , H. Hippe, E. Stackebrandt, S. L'Haridon and C. Jeanthon, Appl. Environ. Microbiol., 69 (2003) 6143. [38] LA. Davydova-Charakhch'yan, V.G. Kuznetsova, L.L. Mityushina and S.S. Belyaev, Microbiology [Engl. Tr.], 61 (1992) 299. [39] R.K. Nilsen and T. Torsvik, Appl. Environ. Microbiol., 62 (1996) 728. [40] E. Miranda-Tello, M.L. Fardeau, P. Thomas, F. Ramirez, L. Casalot, J.L. Cayol, J.L. Garcia and B. Ollivier, Int. J. Syst. Evol. Microbiol., 54 (2004) 169. [41] S. L'Haridon, M.L. Miroshnichenko, H. Hippe, M.L. Fardeau, E.A. BonchOsmolovskaya, E. Stackebrandt and C. Jeanthon, Int. J. Syst. Evol. Microbiol. 52 (2002) 1715. [42] S. L'Haridon, M.L. Miroshnichenko, H. Hippe, M.L. Fardeau, E. Bonch-Osmolovskaya, E. Stackebrandt and C. Jeanthon, Int. J. Syst. Evol. Microbiol., 51 (2001) 1327. [43] Y. Takahata, M. Nishijima, T. Hoaki and T. Maruyama, Int. J. Syst. Evol. Microbiol., 51 (2001)1901. [44] M.E. Davey, W.A. Wood, R. Key, K. Nakamura and D.A. Stahl, System. Appl. Microbiol. 16(1993)191. [45] R. Huber and K.O. Stetter, In: A. Balows, H.G. Triiper, M. Dworkin, W. Harder and K.H. Schleifer (eds.), The prokaryotes, 2nd edn., Springer, Berlin Heidelberg New York, (1992) 3809. [46] Y. Takahata, M. Nishijima, T. Hoaki and T. Maruyama, Appl. Environ. Microbiol., 66 (2000) 73. [47] M.L. Miroshnichenko, H Hippe, E. Stackebrandt, N.A. Kostrikina, N.A. Chernyh, C. Jeanthon, T.N. Nazina, S.S. Belyaev, E.A. Bonch- Osmolovskaya, Extremophiles, 5 (2001) 85. [48] S. Rengpipat, T.A. Langworthy and J.G. Zeikus, Syst. Appl. Microbiol., 11 (1988) 28. [49] G. Ravot, M. Magot, B. Ollivier, B.K.C. Patel, E. Ageron, P.A.D. Grimont, P. Thomas and J.L. Garcia, FEMS Microbiol. Lett., 147 (1997) 81. [50] V.K. Bhupatiraju, A. Oren, P.K. Sharma, R.S. Tanner, C.R. Woese and MJ. Mclnerney, Int. J. Syst. Bacteriol., 44 (1994) 565. [51] V.K. Bhupatiraju, and M.J. Mclnerney, C.R. Woese and R.S. Tanner, Int. J. Syst. Bacteriol., 49(1999)953. [52] A. Oren, Can. J. Microbiol., 32 (1986) 4.
402
[53] M. Magot, M.L. Fardeau, O. Arnauld, C. Lanau, B. Ollivier, P. Thomas and B.K.C Patel, FEMS Microbiol. Lett., 155 (1997) 185. [54] J.L. Cayol, B. Ollivier, B.K.C. Patel, G. Ravot, M. Magot, E. Ageron, P.A.D. Grimont and J.L. Garcia, Int. J. Syst. Bacteriol. 45 (1995) 783. [55] M.L. Fardeau, M. Magot, B.K. Patel, P. Thomas, J.L. Garcia, B. Ollivier, Int. J. Syst. Evol. Microbiol., 50 (2000) 2141. [56] G.N. Rees, B.K.C. Patel, G.S. Grassia and A.J. Sheehy, Int. J. Syst. Bacteriol. 47 (1997) 150. [57] R.J. Menes and L. Muxi, Int. J. Syst. Evol. Microbiol., 52 (2002) 157. [58] M. Magot, G. Ravot, X. Campaignolle, B. Ollivier, B.K.C. Patel, M.L.Fardeau, P. Thomas, J.L. Crolet and J.L. Garcia, Int. J. Syst. Bacteriol. 47 (1997) 818. [59] LA. Davydova-Charakhch'yan, A.N. Mileeva, L.L. Mityushina and S.S. Belyaev, Microbiology (Engl. Tr.), 61 (1992) 306. [60] R. Sleat, R.A. Mah and R. Robinson, Int. J. Syst. Bacteriol., 35 (1985) 10. [61] M. Vargas, K. Kashefi, E.L. Blunt-Harris and D.R. Lovley, Nature, 395 (1998), 65. [62] D.R. Boone, Y. Liu, Z.J. Zhao, D.L. Balkwill, G.R. Drake, T.O. Stevens, H.C. Aldrich, Int. J. Syst. Bacteriol., 45 (1995) 441. [63] S.V. Liu, J. Zhou, C. Zhang, D.R. Cole, M. Gajdarziska-Josifovska and T.L. Phelps, Science, 277(1997)1106. [64] A.G. Cairns-Smith, A.J. Hall and M.J. Russell, Orig. Life Evol. Biosph., 22 (1992) 161. [65] K.M. Semple and D.W.S Westlake, Can. J. Microbiol., 33 (1987) 366. [66] T.N. Nazina, A.E. Ivanova, O.V. Golubeva, R.R. Ibatulin, S.S. Belyaev and M.V. Ivanov, Microbiology (Engl. Tr.), 64 (1995) 203. [67] A.C. Greene, B.K.C. Patel and A.J. Sheehy, Int. J. Syst. Bacteriol., 47 (1997) 505. [68] A.I. Slobodkin, C. Jeanthon, S. L'Haridon, T. Nazina, M. Miroshnichenko, E. BonchOsmolovskaya, Curr. Microbiol., 39 (1999) 99. [69] G. Voordouw, S.M. Armstrong, M.F. Reimer, B. Fouts, A.J. Telang, Y. Shen and D. Gevertz, Appl. Environ. Microbiol., 62 (1996) 1623. [70] J.-Y. Leu, C.P. McGovern-Traa, A.J.R. Porter and W.A. Hamilton, Anaerobe, 4 (1998) 165. [71] G. Voordouw, J.K. Voordouw, R.R. Karkhoffschweizer, P.M. Fedorak and D.W.S. Westlake, Appl. Environ. Microbiol., 57 (1991) 3070. [72] G. Voordouw, Y. Shen, C.S. Harrington, A.J. Telang, T.R. Jack, D.W.S. Westlake, Appl. Environ. Microbiol., 59 (1993) 4101. [73] T. Lien and J. Beeder, Int. J. Syst. Bacteriol., 47 (1997) 1124. [74] A.S Galushko and E.P. Rozanova, Microbiology (Engl. Tr.), 60 (1991) 102. [75] T. Lien, M. Madsen, I.H. Steen and K. Gjerdevik, Int. J. Syst. Bacteriol., 48 (1998) 469. [76] J.-Y. Leu, C.P. McGovern-Traa, A.J. Porter, W.A. Hamilton, Lett. Appl. Microbiol., 29 (1999)246. [77] E.P. Rozanova, T.N. Nazina and A.S. Galushko, Microbiology (Engl. Tr.) 57 (1988) 634. [78] T.N. Nazina and E.P. Rozanova, Microbiology [Engl. Tr.], 47 (1978) 142. [79] R.K. Nilsen, T. Torsvik and T. Lien, Int. J. Syst. Bacteriol., 46 (1996) 397. [80] M. Magot, P. Caumette, J.M. Desperrier, R. Matheron, C. Dauga, F. Grimont and L. Carreau, Int. J. Syst. Microbiol., 42 (1992) 398. [81] E.P. Rozanova and A.I. Khudyakova, Microbiology [Engl. Tr.] 43 (1974) 1069. [82] S.S. Belyaev, R. Wolkin, W.R Kenealy, M.J. De Niro, S. Epstein and J.G. Zeikus, Appl. Environ. Microbiol., 45 (1983) 691. [83] T.K. Ng, P.J. Weimer and L.J. Gawel, Geomicrobiol. J., 7 (1989) 185.
403
[84] B. Ollivier, M.L. Fardeau, J.L. Cayol, M. Magot, B.K.C. Patel, G. Prensier and J.L. Garcia, Int. J. Syst. Bacteriol., 48 (1998) 821. [85] A.Y. Obraztsova, O.V. Shipin, L.V. Bezrukova and S.A. Belyaev, Microbiology [Engl. Tr.] 56 (1988) 523. [86] B. Ollivier, J.L. Cayol, B.K.C. Patel, M. Magot, M.L. Fardeau and J.L. Garcia, FEMS Microbiol. Lett., 147 (1997) 51. [87] A.Y. Obraztsova, V.E. Tsyban, K.S. Laurina Vichus, L.V. Bezrukova andS.S. Belyaev, Microbiology [Engl. Tr.], 56 (1987) 807. [88] S. Ni and D.R. Boone, Int. J. Syst. Bacteriol. 41 (1991) 410. [89] G. Ravot, M. Magot, M.-L. Fardeau, B.K.C. Patel, P. Thomas, J.-L. Garcia and B. Ollivier, Int. J. Syst. Bacteriol. 49 (1999) 1141. [90] T. Lien, M. Madsen, F.A. Rainey and N.-K. Birkeland, Int. J. Syst. Bacteriol., 48 (1998) 1007. [95] G. Ravot, M. Magot, M.L. Fardeau, B.K.C. Patel, G. Prensier, A. Egan, J.L. Garcia and B. Ollivier, Int. J. Syst. Bacteriol., 45 (1995) 308. [97] M.L. Fardeau, B. Ollivier, B.K.C. Patel, M. Magot, P. Thomas, A. Rimbault, F. Rocchiccioli and J.L. Garcia, Int. J. Syst. Bacteriol. 47 (1997) 1013. [98] C. Jeanthon C, A.L. Reysenbach, S. L'Haridon, A. Gambacorta, N.R. Pace, P. Glenat and D. Prieur, Arch. Microbiol., 164 (1995) 91.
This page is intentionally left blank
Studies in Surface Science and Catalysis 151 R. Vazquez-Duhalt and R. Quintero-Ramirez (Editors) © 2004 Elsevier B .V. All rights reserved.
405
Chapter 15
Biotechnological approach for development of microbial enhanced oil recovery technique K. Fujiwara3, Y. Sugaib, N. Yazawac, K. Ohnoc, C.X. Hong" and H. Enomotoe a
Chugai Technos Co. Ltd., 9-20 Yokogawa-Shinmachi Nisi-ku Hiroshima City 733-0013,Japan
b
Akita University Venture Business Laboratory, 1-1 Tegatagakuen-cho Akita City ,010-8502, Japan technology Research Center, Japan National Oil Corporation, 1-2-2 Hamada, Mihama-ku, Chiba 261-0025, Japan d
PetroChina Company Limited, Jilin Oilfield Company, Jilin province, China
e
Department of Geoscience and Technology, Graduate School of Environmental Studies, Tohoku University, Aramaki, Aoba-ku, Sendai 980-0845, Japan 1. INTRODUCTION
1.1. Microbial processes for oil recovery Limited opportunities for discovering major new oil accumulations have focused attention on processes which improve petroleum recovery and prolong the life of existing wells. The microbial processes for oil recovery are classified into three basic applications: well bore clean up, well stimulation and enhanced waterfloods. Well bore clean up is normally carried out when paraffin and scale are deposited on the well bore. Well stimulation and waterfloods enhancement, namely, microbial enhanced oil recovery (MEOR), are conducted when the target reservoir and oil production experience the following conditions: formation damage, pore damage, high water production, poor displacement efficiency, and/or poor sweep efficiency. The authors have focused on MEOR because MEOR is one of the techniques expected to be both economically feasible and environmentally friendly, while also considerably increasing oil production.
406
1.2. Development of MEOR technique Since Beckman proposed in 1926 [1] that bacterial metabolites assist in the release and transport of oil in geological structures, MEOR processes including well stimulation and waterflood enhancement have been attempted in numerous oil fields throughout the world [2]. Fig. 1 shows activity of MEOR that have the potential to enhance oil recovery.
Fig. 1. Mechanisms and effects expected for MEOR
407
According to previous research [3-4], reservoir conditions necessary for MEOR processes are as follows: lower reservoir temperatures (below 70 °C), higher permeability (above 50-75 md), higher porosity (above 20%), lower total salinity (below 15%), appropriate pH ranges (4 to 9) and lower oil viscosity (above 5-50mPa.s). 1.3. Types of MEOR processes There are three variations on the MEOR process. The first involves the injection of both microbes and nutrients. The microbes used here are selected for their ability to make products, such as gases, surfactants, polymers and biomass, that have the potential to increase oil recovery. This process often uses molasses, an inexpensive by-product produced in the sugar refining process, as the nutrients supply. The second variation involves simply the injection of nutrients in order to utilize indigenous microbes within the reservoir. These nutrients generally consist of molasses and/or plant fertilizers. The third process uses microbes which can utilize hydrocarbon. When applying these MEOR processes to a given reservoir, it is critical that the optimum process is chosen based on observed reservoir characteristics, such as geology and indigenous microbes. For example, the existence of useful indigenous microbes in the target reservoir is necessary for successful application of the first and second process variations described above. 1.4. Current stage of MEOR MEOR has not been commonly accepted by the petroleum industry and, for many, the question still remains as to whether MEOR can really be used in increasing oil recovery. One industry concern is that many of the processes simply "do not work." That is, MEOR is thought to be either based on unsupported theories or on isolated laboratory work, so that when MEOR is subjected to actual field tests it failed to generate substantial volumes of incremental oil production. Another reason for the dismissal of MEOR is that most people in the petroleum industry do not fully understand it. Part of this lack of understanding stems from the fact that many people do not realize that MEOR is a multiplicity of technologies, not a single process. 1.5. Current obstacles and breakthrough points The study of MEOR is entering a turning point and there are technical breakthrough points for overcoming the main obstacles to MEOR use. Current obstacles include the following: first, the need for a collection of solid data demonstrating the success of MEOR. Second, fundamental technologies in MEOR need to be more fully developed and tested. Breakthrough points for MEOR are listed below.
408
(1) Breakthrough Point 1: Most reports on field trials have been poorly documented for scientific acceptance. For example, many studies have demonstrated a lack of scientific knowledge about the fundamental microbiology related to MEOR and a lack of adequate control experiments. This is likely due to the fact that research into the fundamental microbiology related to MEOR and control experiments are added expenses and will not contribute directly to an increase in oil recovery. However, getting the data to scientifically document the success of MEOR technology is needed if MEOR is to become a respected tool of the oil industry. (2) Breakthrough Point 2: A technique for transplanting microbes into a formation has not yet been well established. (3) Breakthrough Point 3: So far, the metabolic function of target microbes in the reservoir has not been scientifically demonstrated. 1.6. Objectives of the present study The objectives of this research were twofold: the collection of valuable data proving MEOR's effectiveness, and the development of fundamental technologies in MEOR. These objectives were established for a collaborative research project between Technology Research Center (TRC) of Japan National Oil Corporation and PetroChina Jilin Oilfield Company, run from 1996 to 2002 [5-21]. The strategies of this research are as follows: (1) First step: development of biotechnological tools for estimating the behavior of microbes in the reservoir. (2) Second step: understanding of the MEOR test field. That is, the collection of scientific data about the fundamental microbiology and reservoir related to the MEOR field trial. (3) Third step: development of the following techniques for use in assessing the effectiveness of MEOR: a) Technique for transplanting microbes into the formation. b) Technique for demonstrating microbial metabolic function in the reservoir when needed.
409
2. TEST FIELD The test field, Fuyu oilfield, is located in the northeastern area of China (Fig. 2 A and B). Oil production in this area began in 1973, and waterflooding was instituted in 1983. Current production is conducted using sucker rod pumping.
Fig.2. Location of the test field (A) Location of Fuyu oilfield (Jilin province in China) (B) Wells map at the test field in Fuyu oilfield
410
Table 1 Reservoir data for test area Block Reservoir Areas
East 24-23 [km2]
East 24-26
0.562
0.219
Res. Depth
[m]
300-450
320-450
Res. Temperature
[°C]
28.0
28.0
Res. Thickness
[m]
15.2 (net)
14.8 (net)
[md]
240
240
[%]
27
27
[kg cm"2]
28.8
18.6
[%]
73.8
69.7
Permeability Porosity Current Pressure Water Cut (1995end)
* Present water cut of most wells is more than 90%. * There are heterogeneous fracture zones between each injection well and production well.
The reservoir data for test area is shown in Table 1. The target reservoir is sandstone and its depth is from 300 to 450 m. The temperature of the reservoir is approximately 30 °C. The average permeability is approximately 240 md and porosity is 27%. The present water cut of most wells averages more than 90% and there are heterogeneous fracture zones between each injection well and production well. 3. THE COLLECTION OF SCIENTIFIC KNOWLEDGE OF THE FUNDAMENTAL MICROBIOLOGY RELATED TO MEOR 3.1. Development of investigation technique of microbes related to MEOR. Fig. 3 shows the analytical protocol of microbes related to MEOR. The authors developed a biotechnological tool, a combination of plating and PCR-RFLP analysis [11] to estimate the behavior of microbes which are able to propagate in the reservoir using molasses. Those microbes unable to propagate using molasses were not relevant to our study and therefore did not need to be analyzed. The RFLP profile described in the present study refers to the profile of
411
Restriction Fragments Length Polymorphism based on the 16S rDNA sequence of bacteria. This PCR-RFLP analysis is able to discriminate microbes by comparing their RFLP profiles. Matching of the RFLP profiles of three restriction endonucleases also improved microbe identification. This method consists of the following stages: 1) Microbe extractions are incubated on the molasses agar plate. 2) 16S rDNA of each type of colony is individually amplified by PCR. 3) Amplified 16S rDNA are treated by restriction endonucleases (Hhal, Mspl, AM) 4) RFLP profiles obtained by electrophoresis of the digested 16S rDNA are compared with each other. 5) The microbes are classified based on their RFLP profiles. The PCR-RFLP analysis includes two major contrived conditions. One is the primer setting and PCR condition. Highly conserved regions of 16S rDNA are used as universal primer; amplifications of 16S rDNA of all microbes are accomplished using only one condition. Another is the selection of the three restriction endonucleases Mspl, Hhal and Alul. Through computer simulation, these endonucleases are shown to be particularly effective in generating many restriction fragments in various microbes, mainly resulting in moderate-size fragments of 100 to 1000 bp that are easily distinguished. 3.2. Investigation of microbes inhabiting the reservoir rock which have the ability to propagate using molasses. To obtain closed reservoir core samples, a #J15 well and #J16 well were drilled in the test area [11] (see Fig. 2(b)). A total of 8 and 4 samples of the reservoir rock were collected from, respectively, the #J15 and #J16 wells (see Fig. 4). Whole cores obtained from each layer were cut off using a hatchet; samples of reservoir rock were then collected by scratching the center of the whole core's cross section with a small sterilized pickax. Within two days, microbes inhabiting these samples were analyzed using the combination of plating and PCR-RFLP analysis described above. Within the #J15 reservoir rock, we observed and isolated a total of 177 species having the ability to propagate on molasses. Of these, 59 were determined to have unique RFLP profiles based on RFLP analysis. A total of 87 species having the ability to propagate on molasses were isolated from the r e s e r v o i r r o c k of # J 1 6 ; of t h e s e , 47 w e r e d e t e r m i n e d to have unique RFLP profiles based on RFLP analysis. Moreover, homology analysis (phylogenetic relationships) based on the RFLP profiles demonstrated that almost all of the microbes were different from the general soil bacteria, and some types of yeasts were detected in the high permeability zones and their surroundings.
412
Fig. 3. Analytical protocol of microbes related to the MEOR
In the comparison of RFLP profile of microbes from #J15 with those from #J16, 15 to 20% of RFLP profiles from #J16 matched those of #J15, and 20 to 25% of RFLP profiles from #J16 are closely related species (that is, two RFLP profiles were exact matches) to microbes from #J15. Almost all microbes of these microbes were detected at 102to 105 cfu g"1, and some grew to more than 108 cfu ml"1 using molasses. These results indicate that aerobes and facultative anaerobes isolated from the reservoir rock are trapped in high permeability zones and highly saturated water zones, such as fractures created by hydraulic fracturing operations. The aerobes and facultative anaerobes have also been carried into the reservoir from the surface by water flooding operations, and have accumulated over a long period of time. During these operations, the injection water, including these microbes, is apt to enter into the high permeability zones. Therefore, in the development of MEOR techniques, we must consider that microbes injected into the reservoir will need to co-exist with these indigenous microbes. It is necessary to monitor the indigenous microbes that make use of molasses, particularly those microbes' potential to suppress the growth of in situ microbes injected into the reservoir.
413
Fig.4. Sampling location of reservoir rock
4. ESTIMATION OF BEHAVIOR OF IN SITU MICROFLORA AT THE MOLASSES INJECTION TEST USING HUFF & PUFF PROCESS Four producing wells were selected for this experiment and the Huff and Puff molasses injection process shown in Fig. 5 [12]. The Huff and Puff process encompassed several steps. From the surface facilities, a 10% molasses solution was shipped in sterilized tank trucks and transported to the well site. At the well site, 200 kl per well of molasses solution was injected into the reservoir at an injection rate of 30 kl h"1. After injection, these wells were shut in for 20 days before production resumed.
414
Fig. 5. Injection process of molasses solution by Huff & Puff injection method
Table 2 shows the microbes detected inhabiting the ground water, molasses, and reservoir brine before the injection test. In the ground water, seven species were distinguished by their RFLP profile, and theconcentration of viable cells of each species was 10 to 103 cfu ml"1. Of these, three species grew to more than 107 cfu ml"1 using 4% molasses. On the other hand, in the molasses and reservoir brine, the number of species distinguished by their RFLP profile was two to eight, and the concentration of viable cells of each species was 10 to 106 cfu ml"1. Of these, almost no species grew to more than 107 cfu ml"1 using molasses.
415
Table 2 The number of species distinguished by RFLP profile (in ground water, molasses, reservoir brine) The number of species detected in the samples Ground water Molasses Reservoir brine from #26-231 #24-24, #20-28 #T-59
The number of species grown more than 107 cfu ml"1 using 4% molasses.
7 (10-103) 5 (102-103)
3 1
(10-104) (102-103) (104-105) (102-103)
0 1 0 0
2 8 4 6
( ) ; Concentration of viable cells [cfu ml"1]
Table 3 shows the microbes detected in the injected fluid before injection. These injected fluids were collected from tank trucks at each well site. Within the four injection fluid samples, the number of species distinguished by their RFLP profile was three to five, and the concentration of viable cells of each species was 104 to 107 cfu ml"'. Of these, almost all species grew to more than 107 cfu ml"1 using molasses. Moreover, based on their RFLP profiles, almost all microbes detected in the injected fluid matched microbes which were isolated from the ground water. Table 4 shows microbes detected in the production water after the injection test. In samples obtained from the four wells, three predominant species are distinguished by their RFLP profiles. These species, in viable cell concentrations of 103 to 107 cfu ml"1 were also detected in the 5 to 13 samples of production water collected daily throughout the 20-day test period. Moreover, based on their RFLP profiles, these microbes matched microbes which were isolated from the ground water. Fig. 6 (A, B, C and D) shows the production history of the producing wells in which the molasses solution was injected. It is apparent from these data that molasses injection into the target reservoir did not result in increased oil recovery.
416
The number of species detected in the samples Injected fluid for #26-23, #24-24, #20-28 #T-59
The number of species grown more than 107 cfu ml"1 using 4% molasses.
4 (104-107) 4 (104-106) 3 (106-107) 5 (106-107)
4 4 3 5
( ) ; Concentration of viable cells [cfu ml"1]
Analysis of the collected data indicates the following: (l)Microbes inhabiting the ground water thrive in the presence of molasses and may threaten the growth of other microbes injected into the reservoir. (2) Based on their RFLP profiles, those microbes are "Enterobacteriaceae" or that of a closely related species. (3) An injection of molasses alone is not expected to increase oil recovery in our target reservoir. (4) Microbes related to the molasses injection (such as those inhabiting the ground water, molasses, reservoir brine and reservoir rock) do not cause an increase in oil recovery when provided with molasses. Table 4 Predominant species in production water distinguished by RFLP profile (In production water) DCTB
n
RFLP profile A B C
Frequency of
/* . , . . detection (times)
5 (104-106) 8 (103-105) 13 (103-107)
( ) ; Concentration of viable cells [cfu ml"1]
417
Fig. 6. Result of oil production at the molasses injection tests • : Oil production
• : Total liquid
o : Water cut
Considering these results, it is clear that an increase in oil recovery will require the use of microbes selected for reservoir characteristics such as the development history of the reservoir, indigenous microbes, and geological features.
418
5. DEVELOPMENT OF TECHNIQUES VERIFYING MEOR EFFECTIVENESS In the initial stages of the experiment, oil field conditions were investigated in detail in order to determine what in situ metabolic processes can be supported in the reservoir. It was assumed that selective plugging of highly permeable zones would be effective for this reservoir, because there are many horizontal fractures near the production wells, caused by the hydraulic fracturing operations. Successful selective plugging field operations have been reported previously [22-24], and this methodology is regarded as an effective MEOR process. Fig. 7 shows the plugging mechanism in highly permeable zones. In conventional water flooding, water injected into the reservoir flows predominantly into large channels for a long period of time. When microbes are injected, they also enter primarily the large channels, growing and producing insoluble polymer in these places. As a result, insoluble polymer, including microbial cell mass, selectively plugs high permeability zones, and injection water is diverted from the large channels into previously un-swept areas of the reservoir.
Fig.7. Plugging mechanism of high permeable zones
419
5.1. Screening of microbes for injection in the reservoir 5.1.1. Screening of microbes [16] The essential parameters of microbes used for "selective plugging" are shown below. The microbes must: (1) Produce an insoluble polymer using relatively inexpensive molasses. (2) Propagate under both aerobic and anaerobic conditions. (3) Form biofilm at the surface of reservoir rock. (4)Propagate and produce insoluble polymer under the reservoir's unique conditions. Fig. 8 shows the screening protocol of microbes used for MEOR. This protocol consists of two stages: (1) Microbe extractions from reservoir samples (such as reservoir core and brine) are incubated on the molasses agar plate. (2) Each type of colony is incubated individually in the molasses liquid medium, and candidate microbes are selected visually based on their ability to produce the insoluble polymer. A strain CJF-002, which demonstrated the potential described above, was screened from reservoir rock in Fuyu oilfield. Based on 16S rDNA sequences, it was identified as a strain belonging to the Enterobacter species.
Fig. 8. Screening of microbes for injection in the reservoir
420
Fig. 9. Growth and production of insoluble polymer of strain CJF-002 (After 1 day incubation).
Fig. 9 shows the growth and production of insoluble polymer from the strain CJF-002. It is apparent from these data that one of the notable features of CJF-002 is its ability to grow and produce insoluble polymer when fed molasses. Moreover, the cells of the strain CJF-002 are small enough to pass through the pore throat of average sandstone. Fig. 10 shows the visual image (A) and SEM images (B and C) of insoluble polymer. Cellulase can degrade this insoluble polymer. Results of sugar composition (Table 5) and methylation analysis (Table 6) also indicated that the insoluble polymer produced by strain CJF-002 is a cellulose derivative. 5.1.2. Development of the monitoring technique of viable strain CJF-002 propagating in the reservoir Conventional culture-based bacteriological methods for detecting microbes in environment samples depend on their recovery from each sample and therefore on their culture conditions. However, these methods are not suitable for MEOR because their lower selectivity can not distinguish target microbes from the various microbes inhabiting the reservoir fluid. These methods also require several days to produce results. The process of MEOR
421
with flooding is a massive undertaking, involving the injection of a culture broth of microbes and other nutrients over a long period. Therefore, the technique for monitoring the strain CJF-002 must be not only effectively discriminating, but also rapid and simple.
Fig. 10. Visual image and SEM images of insoluble polymer by strain CJF-002 (A);Visual image, (B and C);SEM images (After 1 day incubation)
Table 5. Sugar composition of insoluble polymer Sugar Detected substance composition Glucose
97.31
Mannose
1.32
Arabinose
0.42
Galactose
0.95
Uronic acid
0.0
*Grown with 4% molasses
422
Table 6. Result of methylation analysis of insoluble polymer Peak No.
Retention time (MS)
partially methylated sugar alcohol
Binding location
Peak area
ratio
1
19'09"
1,5diO-acethyl-2,3,4,6-tetra-Omethylglucitol
nonreduced end Glc
2840
1.00
2
20' 50"
1,4,5-tri-O-acethyl-2,3,6-tri-Omethylglucitol
^4Glc
88917
29.22
3
20' 57"
l,5,6-tri-O-acethyl-2,3,4-tri-Omethylglucitol
-+6Glc
310
0.10
4
21' 27"
l,3,4,5-tetra-O-acethyl-2,6-diO-methylglucitol
^3,4Glc
706
0.22
5
21'38"
l,2,4,5-tetra-O-acethyl-3,6-diO-methylglucitol
—2,4Glc
1572
0.48
6
22' 01"
1,4,5,6-tetra-O-acethyl-2,3-diO-methylglucitol
^4,6Glc
1549
0.48
The authors developed a combination of plating and Direct-PCR analysis to estimate the behavior of strain CJF-002 in the reservoir. The Direct-PCR methodologies, which permit the rapid direct detection of a specific DNA sequence in a target microbe, have been studied for use in detecting a specific microbe in environmental samples including soil, foods and water [25-28]. These methodologies have the potential to identify a given target microbe because detection of DNA fragments indicates the presence of that microbe. Hence, it may be possible to apply these methodologies effectively in understanding the behavior of strain CJF-002 during the MEOR process. The PCR primer was designed using intergenetic spacer regions located between the 16S and 23S rDNA, which consist of highly species-specific sequences. Previous research has demonstrated that these intergenetic sequences are more available than 16S rDNA sequences when identifying bacteria by direct PCR method [29-31]. The sequences of these primers are 5'-AGGCCTACCAAATTTCAGCT-3' (CJF-2F, forward), and 5'-GAGACTCGCAGAACAGTTCG-3' (CJF-2R, reverse). Experimental specificity testing of the PCR primers was also performed on pure cultures of bacterial strains. Bacterial genomic DNA was extracted from each bacterium using an InstaGene matrix (Bio-Rad lab.) containing sterile
423
distilled water (SDW) and Chelex 100 resin, as per the manufacturer's instructions. PCR amplification was performed under the following conditions. PCR mixtures contained 2 ul of 10x PCR buffer (500 mM KC1, 100 mM Tris-HCl [pH 8.3], 15 mM MgCl2), 2 ul of 25 mM MgCl2, 2 ul of a deoxynucleotide triphosphate (dNTP) mixture (concentration of each dNTP, 2.5 mM), 10 pM of each primer, 5 ul of the extracted DNA sample and 0.4 U of Taq DNA Polymerase (Takara Shuzo Co., Ltd., Kyoto, Japan), in a total volume of 20 ul. After the solution was overlaid with 30 ul of mineral oil (Chill-out 14 Liquid Wax, MJ Research Inc., Watertown, MA), the PCR program was initiated with a preincubation at 94°C for 30 s. The amplification profile is 94°C for 45s, 58°C for 50 s, and 72°C for 60 s. PCR products were electrophoresed in a 1.5 % agarose gel and visualized by UV transillumination after being stained in ethidium bromide solution (5 ug ml"1). A primer annealing temperature close to the theoretical primer melting point, that is, 58°C, allowed amplification of a single 280 bp product only in strain CJF-002 (see Table 7), according to the direct PCR protocol. The band size of the amplificates matched the expected size. These results demonstrated that the 16S-23S spacer sequence of strain CJF-002 is sufficiently species-specific for the derivation of PCR primers used to identify strain CJF-002. 5.1.3. Biofilm formation test A biofilm formation test was performed as follows. Sliced reservoir rock was set vertically inside a bottle filled with the 4% molasses solution, synthetic brine, and strain CJF-002 (see Fig. 11). The components of the synthetic brine were established previously based on the components of reservoir brine obtained from the test field (see Table 8). The molasses medium with strain CJF-002 was then stirred 1.7cm per second and incubated at 30 °C. The medium predominantly affected one face of the rock during the test period. After one day, the surface of the sliced reservoir rock was covered with biofilm consisting of the insoluble polymer produced by strain CJF-002 (see Fig. 12). These results suggest that the insoluble polymer produced by strain CJF-002 will adhere to the surface of reservoir rock in the target reservoir.
424
Table 7. Summary of PCR amplification for various species by spacer primers Species CJF-002 Acinetobacter calcoacelis Aeromonas hydrophila Alcaligenes faecalis Azotobacter vinelandii Bacillus subtilis Rhodococcus erythropolis Staphylococcus aureus Pseudomonas fluorescens Enterobacter cloacae Citrobacter freundii Escherichia coli Erwinia carotovora Klebsiella pneumoniae Clostridium acetobutylicum Clostridium butyricum
Strain No.
Size of amplificats (bp) 280
IFO 12552 IFO 13286 IFO 14479 IFO 12018 IFO 3134 IFO 12320 IFO 12732 IFO 14160 IFO 13535 IFO 13546 IFO 13898 IFO 14082 IFO 13541 IFO 13948 IFO 13949
Fig. 11. Biofilm formation test by strain CJF-002
425
Teble 8. The components of reservoir brine Synthetic brine NaCl : KC13 NaHCO CaCl2 MgCl2 FeCl3 KH2PO4 NaHSO4
(1000ml) 1210 (mg) 23 2820 140 53 2 10 3
5.2. Investigation of availability of strain CJF-002 for the environments of given reservoir. Investigation into the availability of strain CJF-002 was carried out in order to determine whether the strain CJF-002 will survive and perform the desired metabolic functions in a given reservoir. A controllable factor in the reservoir, the effect of molasses concentration was studied first. Results showed that growth of the strain CJF-002 started in the presence of more than 0.1% molasses, and that production of insoluble polymer accelerated with 1% molasses. The effects of factors difficult to control in the real reservoir were then investigated. Growth of strain CJF-002 and production of insoluble polymer were observed at more than 15°C and at a pH higher than 5.4. Production of insoluble polymer also accelerated when NaHCO3 was present in the reservoir brine. Moreover, strain CJF-002 grew and produced insoluble polymer in co-existence with microbes inhabiting the reservoir brine, ground water, molasses and injection fluid. Table 9-(A) shows the results of competitive culture tests with indigenous microbes in the reservoir brine, and Table 9-(B) is with microbes in the injection water. Notably, in co-existence with the indigenous microbes in the reservoir brine, CJF-002's potential for propagation and survival are exceedingly high when compared with strains A and B. The strain CJF-002 was detected at approximately 108 cfu ml"1 until day 10 and detected at more than 104 cfu ml"1 by the 20th day. Thus it appeared that strain CJF-002 would have the ability to grow and survive in the environment of our target reservoir.
426
Table 9. Result of competitive vulture test (A) with indigenous microbes in the reservoir brine Incubation period
Screened place
initial
lday
3 days
5 days
lOdays
20days
Strain A
Japan
x 1O s
xlO6
xlO6
xlO 5
• OlO3
• OlO3
Strain B
Japan
xlO6
xlO7
xlO4
DOlO 3
•OlO 3
NT
CJF-002
China
xio 6
xlO8
xlO 8
xlO8
xlO 8
xlO4
(B) with microbes in the injection water Incubation period
Screened place
initial
lday
3 days
5 days
lOdays
20days
Strain A
Japan
xlO6
xlO8
xlO 8
xlO8
xlO8
• OlO3
Strain B
Japan
xlO5
xlO7
• OlO3
•OlO 3
•OlO 3
• OlO3
CJF-002
China
xlO6
xlO9
xlO8
xlO8
xlO8
• OlO3
The survivability of strain CJF-002 also depends on its initial concentration. More than 0 . 1 % of molasses and 105 cfu ml"1 of strain CJF-002 are vital for growth, insoluble polymer production and overcoming other microbes. Finally, the effect of uncontrollable factors in the reservoir was investigated. Growth of strain CJF-002 and the production of insoluble polymer were found to be unrelated to the presence of oxygen. In the presence of oil, however, the rate of insoluble polymer production decreased approximately 20%. That rate also decreased to approximately 1/3 of the original production under the reservoir pressure (30 atm). Growth of strain CJF-002 and the production of insoluble polymer were not affected by the micro culture environment.
427
Fig. 12. Result of biofilm formation test by strain CJF-002
5.3. Evaluation of microbial profile modification effect An experiment was performed to evaluate the effect of microbial profile modification at the laboratory [19]. Two sets of porous media were prepared (see Fig. 13): one high-perm sand pack, and one low-perm berea sandstone core. The two cores were first saturated with the synthetic brine described above. The absolute permeability of both cores was then measured by injecting brine through the pump. Permeability of the sand pack was determined to be 11,000 md, and the sandstone core permeability was found to be 900 md. Strain CJF-002 and 5% molasses were then injected. When the waterflooding was started after a 5-day shut in, core permeability was almost unchanged, while the permeability of the sand pack fell drastically (see Fig. 14). Also, prior to the injection of the strain CJF-002 and molasses, the fluid ratio, (the production rate of high permeability sand pack to the low permeability core), was 16, but it decreased to only 1.5 after the shut-in. This decrease indicates that the strain CJF-002 selectively plugs high permeability zones. In other words, strain CJF-002 has the ability to modify profiles.
428
Fig. 13. Laboratory experiment to confirm an ability of profile modification
5.4. Huff and Puff tests using production wells Huff and Puff tests using strain CJF-002 were performed at six producing wells in order to confirm the survivability of strain CJF-002 and the production of insoluble polymer [16-17]. Wells selected for the Huff and Puff tests featured a relatively high water cut (more than 70%) and high water production. Culture conditions of strain CJF-002 and the process of injection into the reservoir were as follows. In the laboratory, strain CJF-002 was incubated using a 1% molasses medium containing 0.01% of cellulase at 100 ml and 10L, in due order. The cellulase was added to the medium to prevent the production of insoluble polymer during the propagation of the strain CJF-002. At the surface facility, the culture scale increased to 1000 1. The 400 1 of final culture broth was mixed with water or molasses solution (final cone. 0.6 - 8%), then injected into the oil reservoir through six producing wells as described below: (1) 10 kl of CJF-002 culture broth & 80 kl of molasses were injected separately, followed by an injection of 20 kl of driving water. (2) 10 kl of CJF-002 culture broth & 80 kl of molasses were injected simultaneously, followed by an injection of 20 kl of driving water. After a 10-day shut-in period, concentrations of strain CJF-002 in production water drawn from producing wells were measured using the following method. 100 ul of each sample was placed on nutrient broth agar plates (Difco Co., Ltd., Detroit, MI); these plates were then incubated at 30°C under aerophilic conditions for two days. Concentration of the CJF-002-like colonies was measured by the colony forming units (CFU) method. Some of those colonies were selected from the agar plates to undergo the direct PCR analyses described
429
above, in order to demonstrate that the colonies distinguished by their characteristics were, in fact, strain CJF-002. The total number of non- CJF-002 microbes was also evaluated, using the plating method described above. All data from these experiments are presented in Table 10. The strain CJF-002 was detected at all six of the wells and a definite increase in oil production was observed at four wells. The results of oil production and monitoring of microbes at the 22-264 well are illustrated in Fig. 15 and 16, respectively. After shut-in, oil production increased remarkably and strain CJF-002 was detected in the production water at a relatively high concentration. It is apparent from these data that the strain CJF-002 has the ability to survive in the given reservoir and injection of strain CJF-002 into the target reservoir contributes to increased oil recovery. In the Huff and Puff tests, the concept of highly-permeable-zone plugging by insoluble polymer is as follows. If there are highly-permeable zones in the reservoir, injected fluid, including the strain CJF-002 and molasses, enters primarily into these zones. Insoluble polymer is produced selectively in the areas which contain the mixture of strain CJF-002 and molasses. A stream line of injected water is then modified, and residual crude oil in low-permeable zones (unswept oil) is newly recovered.
Fig. 14. Result of laboratory waterflooding experiment. • Permeability (Sand Pack) o Permeability (Core) D x Producing fluid ratio Injection pressure
430
Teble 10. Result of Huff & Puff Test by strain CJF-002 Before injection Well*1)'2)
1. #22-27 2. #24-23 4 3.#22-26 4 4. #26-231 5. #26-254 6. #24-24 j
Water R t cut [%] [t day"1] 99 97-98 98 98 97 99
0.2 0.1-0.2 0.2-0.3 0.2 0.2-0.3 o.l
Condition of injection Molasses cone. [%]
Injection system (CJF-002and molasses)
1.9-3.0 0.6-1.5 3.3-5.7 4.2 4.6-7.4 4.4-8.0
Separately Separately Separately Separately Simultaneously Simultaneously
After injection Microbe Oil WC*5) 3) 4) in Prod.* Prod.* r 0/1 L J [%] ° 83.3 85.7 88.9 57.1 62.5 85.7
?9
?9
*1) 1.-4. 10m3 of CJF-002 culture broth & 80m3 of molasses were injected separately. *2) 5. and 6. 10m3 of CJF-002 culture broth & 80m3 of molasses were injected simultaneously. *3) The ratio of samples with CJF-002 to samples of produced water analyzed. *4) Oil production; Increase , Same level *5) Water cut; Decrease , Same level
Fig. 15. Result of oil production at well 22-264 • Oil production A Water cut
431
Fig. 16. Result of microbial monitoring at well 22-264 • : Viable cell number before Injection O: Total viable cell number after resuming production • : CJF-002 viable cell number after resuming production
5.5. Continuous injection tests using two injection wells These tests were conducted to demonstrate the fundamental breakthroughs in MEOR technology [19-21]. Fig. 17 shows the location of the test area; there are 10 production wells and two injection wells. Production history of the test area is shown in Fig. 18. An increase in the monthly oil rate had not been observed for approximately 20 years. Fig. 19 shows a schematic of the surface facility where microbial treatment took place in the first trial. The microbial incubation facility is located 3 km away from the well site. The schedule of microbial treatments in the first trial is shown in Fig. 20. In these treatments, strain CJF-002 solution and molasses were injected separately to prevent the produced insoluble polymer from plugging the perforation holes in injection wells. The strain CJF-002 was incubated as it was for the Huff and Puff tests described above. The 400 1 of final culture broth was mixed continuously with 50 kl of water for one day, then injected into the reservoir through two injecting wells at a rate of 25 kl per day per well. The concentration of strain CJF-002 at the injection pump was approximately 106 cfu ml"1. The strain CJF-002 was injected for two weeks, followed by a one-week injection of 0.1% molasses solution. In addition, 1, 5 and 20% molasses solutions were injected subsequent to the strain CJF-002 injections. After the strain CJF-002 and molasses injections, water injection was continued as usual.
432
Fig. 17. Wells map at continuous
injection test aria
Fig. 18. Production history of East 24-26 Block Monthly oil rate,
Monthly water rate
433
Fig. 20. Schedule of microbial treatments (1st trial).
434
5.5.1. First Trial (2000) Throughout this experiment, the concentration of strain CJF-002 in both the injection fluid obtained from the manifold and the production water drawn from producing wells were measured by the plating and Direct-PCR analysis described above. The total number of non-CJF-002 microbes were also evaluated using this plating method. The microbial concentrations in injected water taken from the manifold are shown in Fig. 21. At the first injection of strain CJF-002 and molasses, strain CJF-002 and non-CJF-002 microbes are at similar concentration levels of 105 and 106 cfu ml"'. The competitiveness of strain CJF-002 under similar conditions had been confirmed in our previous laboratory study, however, at the second injection, the concentration of other microbes had become over 100 times higher than that of the strain CJF-002. By the third and final injection, non-CJF-002 microbes were completely dominant. Consequently, the strain CJF-002 was detected in only 60% of producing wells in this test area. In such conditions, the competitiveness of the strain CJF-002 is considered to be inferior to other microbes In addition, an increase in oil recovery, which should have been observed, was not confirmed in the first trial. These results indicate that it is crucial to inject the strain CJF-002 at high concentrations and keep its predominance in the injected fluid. The microbial monitoring of production water from well #J18 showed the first detection of strain CJF-002 was observed within four days, though this production well is located 60 m away from the injecting well. The strain CJF-002 detected in the production water was known to have come from the strain injected at the injection wells, because the strain CJF-002 was not detected in the pure molasses injection test described above. From this data, it appears that the strain CJF-002 injected into the reservoir flows at a high speed in the highly permeable regions. The detection period of the strain CJF-002 after the molasses injection is also a remarkable point. After the 0.1% molasses injection, the strain CJF-002 became undetectable in a week, while after a high molasses injection, a relatively high concentration of strain CJF-002 was detectable for a month (see Fig. 22). These results indicate that the strain CJF-002 has survivability in the reservoir environments when a desirable concentration of molasses, namely, more than 5%, is present.
435
5.5.2. Second Trial (2001) Based on the results of first trial, the following modifications to the injection method were determined: (1) Inject strain CJF-002 and molasses simultaneously to increase the chance of interaction between the CJF-002 and the molasses. (2) Inject 10% molasses to maximize the amount of insoluble polymer. (3)For one week preceding the molasses injection, inject only the strain CJF-002, so as to dominate the reservoir with strain CJF-002. (4)Reduce the other microbes in the pipeline by injecting the molasses closer to the well site. Fig. 23 shows a schematic of the surface facility used for microbial treatment in the second trial. The strain CJF-002 was injected through an injection pump at the microbial incubation facility, whereas the 10% molasses solution was injected from a pump near the injection wells. The schedule of microbial treatments in the second trial is shown in Fig. 24. The strain CJF-002 was injected along with injection water for one week.
Fig. 21. Results of microbial monitoring at manifold (1st trial) • CJF-002
o Other microbes
436
Fig. 22. Result of microbial monitoring of production water taken from well J-18 (2nd trial) • CJF-002
o other microbes
Fig. 23. Microbial treatment facility of 2nd trial.
437
Fig. 24. Schedule of microbial treatments (2nd trial).
A 10% molasses solution was then injected along with strain CJF-002. After two months of CJF-002 and molasses injections, water injection was continued as usual. The strain CJF-002 and other microbes were measured as in the first trial, described above. The microbial concentrations in injection water taken from the manifold are shown in Fig. 25. The strain CJF-002 and non-CJF-002 microbial concentrations were at almost same levels during injection of the strain CJF-002 and molasses. Theoretical concentration, that is, approximately 105 cfu ml" of strain CJF-002 was routinely detected in the injection water. In other words, results of this microbial monitoring indicate that the strain CJF-002 can increase in cell number and produce the insoluble polymer in the reservoir. Figure 26 shows the results of the microbial monitoring of production water from well #J18. Notably, the concentration of strain CJF-002 detected in the production water was relatively high, approximately 103 to 106cfu ml"1, for 20 days following the initial injection. Consequently, the strain CJF-002 was detected in all producing wells in this test area. Parts of insoluble polymer which may have come off the reservoir rock were also detected in all producing wells by HPLC analysis combined with cellulase degradation. This result proves that the strain CJF-002 has the ability to produce insoluble polymer in porous media in the reservoir. Results also show that the concentrations of the injected CJF-002 and molasses during the second trial are enough to sustain the strain CJF-002's competitiveness and survivability. The oil production of all wells in the test area is shown in Fig. 27. Eventually, the oil production increased by more than two times for at least one year, and incremental oil production reached 3,392 tons [approximately 24,521 bbls] after microbial injection. Total water cut also fell from 88% to 65%. Notably, a dramatic increase in oil production was observed approximately 20 days after beginning injections. Even after the final CJF-002 and molasses injection, the great improvements in oil production and water cut continued, and the oil recovery rate after one year was still doubled. These results indicate that
438
the increased oil production is related to the high concentration of the strain CJF-002 which was detected at the manifold in the 20 days following the initial injection. The authors believe that the principal cause of the sustained increase in oil recovery over the following year is the stability of the insoluble polymer in the reservoir. Degradation by strain CJF-002 and the other microbes inhabiting the reservoir is one factor that may affect the stability of the insoluble polymer. Some microbes producing cellulase, such as Clostridium sp., are generally well-known, and some researchers have reported that the microbes belonging to these species inhabit reservoirs. In our preliminary experiment, however, we confirmed that the insoluble polymer is not degradable by any microbes inhabiting the target reservoir or by the strain CJF-002 (data not shown). Another factor influencing insoluble polymer stability is the possible absorption of insoluble polymer into reservoir rocks. Considering the results of the biofilm formation test described above, it is presumed that the insoluble polymer produced by strain CJF-002 will adhere to the surface of reservoir rock in the target reservoir.
Fig. 25. Results of microbial monitoring at manifold (2nd trial) • CJF-002 A Other microbes
439
Fig. 26. Result of microbial monitoring at producing wellJ-18 (2nd trial) • CJF-002 o Other microbes A Insoluble polymer
Fig. 27. Behavior of total oil production at all test area (2nd trial) A Total liquid production • Oli production m Water cut
440
Figure 28 shows the carbon number of production oil from well #24-254' after microbial treatment. The carbon number of production oil shifted to a low molecular weight which can be recovered easily. This result indicates that the production oil after microbial treatment was produced from previously unswept zones. In addition, the effectiveness of the microbial treatment was evaluated through a comparison of tracer tests using NH4SCN solution. After the first and second microbial treatment, 30,000 ppm of NH4SCN solution was injected into the reservoir through two injecting wells. The cumulative quantity of that tracer was approximately 500 kg. In well #22-27, tracer could not be detected before the treatment, but could be detected after the treatment (see Fig. 29 (A)). In contrast, in well #26-25, tracer could be detected before, but not after the treatment (see Fig. 29 (B)). Furthermore, in the other wells, changes in tracer peak or time of detection were observed. These results show that the microbial treatment modified the sweep pattern of injection water. The results described above prove that the insoluble polymer increases the volumetric sweep efficiency by diverting injection water from the most highly permeable zones to previously unswept oil-bearing zones. Moreover, considering the results of the second trial, injection of strain CJF-002 and molasses for 20 days when the oil production rate decreases may be effective in continuing to increase oil recovery.
441
Fig. 29. Result of tracer test at well 22-27 and well 26-25 (2nd trial) O Before microbial treatment A After microbial treatment
Based on the present study, future development of the "selective plugging" technique is suggested as follows. The exploration of optimum conditions in field tests, such as the injection concentrations and the injecting periods of strain CJF-002 and molasses, should be continued based on biotechnological monitoring data. This research will contribute to improving the effectiveness of the plug, reliable plugging of the target location in the reservoir, and higher efficiency and durability of the selective plugging effect. A technique utilizing multiple diverse microbes should also be developed, in addition to using biotechnology to screen highly-useful microbes.
442
5.5.3. Economic feasibility o/MEOR Additional running costs for two months in the second trial were as follows: 236 tons of molasses (US$27,000), 40 tons of diesel oil (US$15,500) and the labor costs of two to three people. Additional facilities in the second trial included a molasses tank (5 kl), a microbial incubation tank (1 kl), and injection pumps for the molasses and strain CJF-002. In contrast, additional income over one year of the second trial was US$490,240 (US$20 per bbl). Previous studies report that operational costs of the MEOR process range from $2 to $4 per incremental barrel of oil. Based on the present results, the running costs (aside from labor costs and facility costs) for increasing oil recovery is 1.7 US$ per bbl. These results indicate that MEOR increases oil recovery in an economically attractive manner. 6. CONCLUSION 6.1. Conclusions of all the present studies Based on the results of laboratory experiments and MEOR field tests, the following conclusions were drawn: 1) Biotechnological tools for estimating the behavior of indigenous microbes in the reservoir (PCR-RFLP method), as well as injected microbes (Direct-PCR method), were successfully developed. 2) Biotechnological tools have improved our understanding of micro flora in the MEOR test field. a) Microbes isolated from reservoir rock have been carried into the target reservoir through the development of the oilfield. b) Fracture zones and their surroundings are susceptible to injected water, and various microbes included in the injected water are apt to propagate at those zones. c) Using microbes selected for reservoir characteristics (such as the development history of the reservoir, microbes inhabiting the reservoir environment, and geological features) is necessary for increasing oil recovery. 3) Transplanting microbes into the high permeability zones of the reservoir and demonstrating the resulting insoluble polymer production are best accomplished with operations which reflect and respond to data monitoring the behavior of the injected microbe. 4) MEOR is expected to be an economically feasible technique.
443
5) It is clear that the following points are very important for development of the MEOR technique: a) Understanding of microbes related to MEOR (including microbes inhabiting the ground water, molasses and reservoir environments, in addition to the injecting microbe). b) Monitoring of behavior of these microbes at field trials. c) Designing of field operations which reflect those monitoring data. d) Establishing techniques for transplanting microbes and demonstrating microbial metabolic function in the reservoir. In the present study, valuable data to demonstrate the MEOR effect were successfully collected and the results obtained from this research support the theory that MEOR can effectively increase oil recovery. Though the results are seen in only a few cases yet at most, the authors believe that this is an example of a successful application of MEOR to a broad range of reservoirs. 6.2. Contribution of biotechnology to the development of the MEOR technique. In the examination of microbes related to MEOR processes, PCR-RFLP analysis of the 16S rDNA of microbes was useful for investigating microbes inhabiting the ground water, molasses, reservoir brine and reservoir rock. This method is also useful for confirming a state of sterilization in equipment, such as incubation tanks, tank trucks, and fluid lines, during field operations. As demonstrated in this study, the Direct-PCR method, with primers including sequences in the ribosomal spacer region between the 16S and 23S rDNA, is a powerful tool for screening the injection microbe. The screening involves several tests, such as competitive culture tests, on multiple diverse microbes. The Direct-PCR method is also useful for monitoring injection microbes in the incubation tanks, tank trucks, manifolds and well sites (injection wells and production wells) during field operations. Therefore, biotechnologies contribute to promoting MEOR because the MEOR process must be designed based on the features of microbes and geological characteristics in each oil field or reservoir. Acknowledgements We would like to thank Japan National Oil Corporation, PetroChina Company Limited Jilin Oilfield Company and Chugai Technos. Company Limited for the permission to present this paper. We also thank Tohoku University and KRI Inc. for their constant support in microbial analysis at Fuyu oilfield. We are also very grateful to fellows for their continuous assistance during this collaborative research project.
444
REFERENCES [I] J. W. Beckman, Ind. Eng. Chem., November, 10 (1926) 3. [2] I. Lazar, and E. C. Donaldson (eds.), Microbial Enhancement of Oil Recovery - Recent Advances, MEOR Field trials carried out over the world during the last 35 years, Elsevier, 1991 [3] J. B. Clark, D. M. Munnecke, G. EJanneman, Dev. Ind. Microbiol., 22 (1981) 695. [4] R. S. Bryant, U. S. DOE Report, NIPER-478, August, DE91002208 (1990). [5] H. Yonebayashi, M. Taguchi, K. Fujiwara, S. Yoshida, H. Enomoto, J. Jap. Associ. Petrol. Technol, 61 (1996) 485. [6] H. Yonebayasi, H. Enomoto, T. Chida and K. Fujiwara, Proceeding of 17th Workshop of the International Energy Agency, Sydney Australia, September, 1996 [7] H. Yonebayasi, K. Ono, H. Enomoto, T. Chida, C-X. Hong and K. Fujiwara, Society of Petroleum Engineers 38070 (1997). [8] H. Yonebayasi, H. Enomoto, K. Fujiwara, T. Chida and C-X. Hong, Laboratory R & D leads to MEOR Field Pilot in Fuyu-oilfield, Chaina, Proceeding of 9th European Symposium on Improved Oil Recovery, The Hague, October 20-22, 1997 [9] K. Ono, S. Maezumi, H. K. Sarma, H. Enomoto, C-X. Hong, S-C. Zhou and K. Fujiwara, Society of Petroleum Engineers 54328 (1999) [10] S. Maezumi, H. K. Sarma, N. Yazawa, S-C. Zhou, K. Fujiwara, H. Enomoto and C-X. Hong, Proceeding of 20th Workshop of the International Energy Agency, Enghien-les-Bains (Paris), France, September 22-24, 1999 [II] K. Fujiwara, S. Tanaka, M. Ohtsuka, N. Ichimura, H. Yonebayashi, C. X. Hong and H. Enomoto, Sekiyu Gakkaishi (J. Jpn. Petrol. Inst.), 42 (1999) 342. [12] K. Fujiwara, S. Tanaka, M. Ohtsuka, K. Nakaya, S. Maezumi, N. Yazawa, C. X. Hong, T. Chida and H. Enomoto, Sekiyu Gakkaishi (J. Jpn. Petrol. Inst.), 43 (2000) 274. [13] K. Fujiwara, S. Tanaka, M. Ohtsuka, H. Yonebayashi and H. Enomoto, Sekiyu Gakkaishi (J. Jpn. Petrol. Inst.), 43 (2000) 43. [14] K. Fujiwara, Proceeding of International Symposium on Research and Education in the 21st Century, Sendai, Japan, August 18-25, 2000 [15] H. Enomoto, K. Fujiwara, H. Yonebayashi Sekiyu Gakkaishi (J. Jpn. Petrol. Inst.), 43 (2000)91. [16] K. Nagase, S. T. Zhang, H. Asami, N. Yazawa, K. Fujiwara, H. Enomoto, C. X. Hong and C. X. Liang, Society of Petroleum Engineers 68720 (2001) [17] K. Nagase, S. T. Zhang, H. Asami, N. Yazawa, K. Fujiwara, H. Enomoto, C. X. Hong and C. X. Liang, Proceeding of 22th Workshop of the International Energy Agency, Poland, September 10-14, 2001 [18] K. Fujiwara, Cellulose Commun., 8 (2001) 127. [19] K. Nagase, S. T. Zhang, H. Asami, N. Yazawa, K. Fujiwara, H. Enomoto, C. X. Hong and C. X. Liang, Society of Petroleum Engineer 75238 (2002) [20] K. Nagase, S. T. Zhang, H. Asami, N. Yazawa, K. Fujiwara, H. Enomoto and C. X. Hong, J. Jap. Associ. Petrol. Technol., 68 (2003) 271.
445
[21] K. Fujiwara and H. Enomoto, Proceeding of 2nd International Conference of Petroleum Biotechnology, Mexico City, Mexico, November 5-7, 2003 [24] A. A. Valie, J. O. Stephens and L. R. Brown, Society of Petroleum Engineers 35448 (1996) [22] L. R. Brown, Society of Petroleum Engineers 59306 (2000) [23] G. E. Jenneman, R. E. Lappan and R. H. Webb, Society of Petroleum Engineers 59307 (2000) [25] A. A. Khan, R. A. Jones and C. E. Cerniglia, J. Ind. Microbiol. Biotechnol, 20 (1988) 90. [26] Y. Tasi, and B. H. Olsen, Appl. Environ. Microbiol., 57 (1991) 1070. [27] R. J. Steffan, and R. M. Atlas, Appl. Environ. Microbiol., 54 (1992) 2185. [28] C. R. Kuske, K. L.Banton,, D. L. Adorada, P. C. Stark, K. K.Hill, and P. J.Jackson, Appl. Environ. Microbiol., 64 (1998) 2463. [29] C. D. Smart, B. Schneider, C. L. Blomquist, L. J. Guerra, N. A. Harrison, U. Ahrens, K. H. Lorenz, E. Seemuller and B. C. Kirkpatrick, Appl. Envir. Microbiol., 62 (1996) 2988. [30] N. P. Rijpens, G. Jannes, M. V. Asbroeck, R. Rossau and L. M. Herman, Appl. Envir. Microbiol., 62(1996) 1683. [31] J. Chun, A.Huq and R. R. Colwell, Appl. Envir. Microbiol., 65 (1999) 2202.
This page is intentionally left blank
Studies in Surface Science and Catalysis 151 R. Vazquez-Duhalt and R. Quintero-Ramirez (Editors) ©2004 Published by ElsevierB.V.
447
Chapter 16
Phytoremediation of hydrocarbon-contaminated soils: principles and applications R. Kamath, J. A. Rentz, J. L. Schnoor and P. J. J. Alvarez Department of Civil and Environmental Engineering, Seamans Center, University of Iowa, Iowa City, Iowa, U.S.A. - 52242
1. INTRODUCTION 1.1. Common Target Contaminants Total petroleum hydrocarbons (TPH) comprise a diverse mixture of hydrocarbons that occur at petrochemical sites and storage areas, waste disposal pits, refineries and oil spill sites. TPHs are considered persistent hazardous pollutants, and include compounds that can bioconcentrate and bioaccumulate in food chains [1], are acutely toxic [2], and some such as benzene [3] and benzo[a]pyrene are recognized mutagens and carcinogens [4]. Since this group includes chemicals that have physical and chemical characteristics that vary over orders of magnitude, TPHs are divided into two categories (Fig. 1). Gasoline range organics (GRO) corresponds to small chain alkanes (C6-Ci0) with low boiling point (60°-170 C) such as isopentane, 2,3-dimethyl butane, rc-butane and fl-pentane, and volatile aromatic compounds such as the monoaromatic hydrocarbons benzene, toluene, ethylbenzene, and xylenes (BTEX). Diesel range organics (DRO) includes longer chain alkanes (Cio-C4O) and hydrophobic chemicals such as polycyclic aromatic hydrocarbons (PAH). Whereas most of these contaminants do have natural sources, concentration and release of contaminants through anthropogenic activities has led to significant contamination of soil and groundwater. The extent of petroleum hydrocarbon contamination throughout the United States is reflected by the large number of Superfund sites and Leaking Underground Storage Tanks (LUST) sites that contain these contaminants (Fig. 2 and 3). These sites often contain high concentrations of contamination. However, individual contaminants behave differently. Some contaminants such as BTEX compounds are highly mobile in the environment, while others such as PAHs tend to bind
448
strongly to soil particles near the source or remain entrapped within an organic phase.
ii) Polycyclic Aromatic Hydrocarbons (PAH)
Fig. lb. Examples of common Diesel Range Orgamcs (DRO).
449
Since hydrocarbon spills at different sites represent different mixtures, it is very difficult to find a single, efficient method of cleanup. Current treatment techniques usually involve excavation and ex situ treatment of the source material and the contaminated soils. However, residual contamination often exceeds regulatory limits by a relatively small margin, and occurs over extensive areas [5]. The large volume of soil affected precludes ex-situ treatment due to economical constraints and requires the use of relatively inexpensive remediation schemes, such as phytoremediation.
Fig. 2. United States Superfund sites containing petroleum hydrocarbon contamination for FY1982 to FY1999 (834 total projects, [6]).
Fig. 3. Total United States underground storage tank corrective actions (FY 1992 to FY 2003, [7]).
450
Research and application of phytoremediation for treatment of petroleum hydrocarbon contamination over the past fifteen years has provided much useful information that can be used to design effective remediation systems and drive further improvement and innovation. This chapter will attempt to provide a strong foundation for understanding phytoremediation of petroleum hydrocarbon contaminated sites from principles to practice. 1.2. General Scope of Phytoremediation Phytoremediation is a biological technology process that utilizes natural plant processes to enhance degradation and removal of contaminants in contaminated soil or groundwater. Broadly, phytoremediation can be costeffective for: a) Large sites with shallow residual-levels of contamination by organic, nutrient, or metal pollutants, where contamination does not pose an imminent danger and only "polishing treatment" is required; and b) Where vegetation is used as a final cap and closure of the site [8]. Advantages of using phytoremediation include cost effectiveness, aesthetic advantages, and long-term applicability (Table 1). Furthermore, the use of phytoremediation as a secondary or polishing in situ treatment step minimizes land disturbance and eliminates transportation and liability costs associated with offsite treatment and disposal. Increasing public and regulatory acceptance are likely to extend the use of phytoremediation beyond current applications. 2. PHYTOREMEDIATION MECHANISMS Phytoremediation utilizes physical, chemical, and biological processes to remove, degrade, transform, or stabilize contaminants within soil and groundwater. Hydraulic control, uptake, transformation, volatilization, and rhizodegradation are important processes used during phytoremediation (Fig. 4) and are discussed below. 2.1. Hydraulic Control Phytoremediation applications can be designed to capture contaminated groundwater plumes to prevent off-site migration and/or decrease downward migration of contaminants, as illustrated in Fig. 5. Trees and grasses act as a solar "pump" removing water from soils and aquifers through transpiration. Contaminant plume capture relies on the formation of a cone of depression within an aquifer due to uptake of water by plants and subsequent transpiration.
451
Table 1 Advantages and disadvantages of phytoremediation over traditional technologies such as pump and treat of contaminated groundwater and soil excavation and above-ground treatment. Advantages Relatively low cost Easily implemented and maintained Several mechanisms for removal Environmentally friendly Aesthetically pleasing Reduces landfilled wastes Harvestable plant material
Disadvantages Longer remediation times Climate dependent Effects to food web might be unknown Ultimate contaminant fates might be unknown Results are variable
Figure 4. Schematic of different mechanisms of contaminant removal by plants [8].
452
The key to forming a successful barrier against plume migration is for trees to be rooted into a shallow water table aquifer. Phreatophytes, deep-rooted plants including hybrid poplars and willows are most often used for hydraulic control. When planted densely (more than 600 trees per acre), poplars and willows usually reach optimum working conditions after 3-4 years during canopy closure when almost all the direct sunlight is intercepted. The application of phytoremediation requires that the bottom of the aquifer be confined by materials of low hydraulic conductivity such as clay, shale, or rock (hydraulic conductivity < 10"6 cm/s) and does not "leak" water vertically down to another unit. However, plume capture is not limited to shallow aquifers, as poplar trees planted in well casings have been used to tap water tables at a depth of 10-m [10].
Fig. 5. Plan view of trees planted on a line (similar to an interdiction well field) to capture a shallow groundwater plume (Modified after [9]).
Downward migration of contaminants due to percolation of rainwater can also be controlled with phytoremediation. Within the upper region of an aquifer, grasses with dense, fibrous root systems are used to transpire water and limit percolation of contaminants through the vadose zone and to intercept rainwater that may discourage tree root penetration through the water table. 2.2. Uptake, translocation, and transformation Moderately hydrophobic (log KQW = 1.0 to 3.0) hydrocarbons, including BTEX, can be removed from soil and groundwater through direct plant uptake. The transpiration stream concentration factor (TSCF), an indirect measure of uptake efficiency, has been used to adequately predict whether contaminants will be taken up by plants (Fig. 6). Briggs [11] proposed a bell-shaped relationship between TSCF and contaminant hydrophobicity, indicated by the
453
logarithmic of the octanol-water partitioning coefficient (log Kow). This relationship was developed for pesticide uptake by barley plants, and is given by equation (1) below. Burken and Schnoor [12] adapted this equation to describe the uptake of a wide variety of organic contaminants (including BTEX) by hybrid poplar trees. This relationship is represented by equation (2) and is depicted in Figure 6. TSCF =0.784 exp {-(log K™ - 1.78)2/ 2.44}
(1)
TSCF = 0.756 exp{-(log K™ - 2.50)2 / 2.58}
(2)
The bell-shaped curve shown in Fig. 6 reflects poor plant uptake of hydrophilic compounds (log Kow < 1), which have little affinity for root membranes; high uptake efficiency of moderately hydrophobic hydrocarbons such as BTEX (1.5 < low Kow < 3.5); and poor uptake of hydrophobic hydrocarbons such as PAHs (log Kow > 4), which strongly sorb to soil and are therefore, not bioavailable. The rate of contaminant removal has been found to be a function of uptake efficiency (e.g., TSCF), transpiration rate, and the contaminant concentration in soil water, as discussed in section 5.1. Uptake efficiency varies with plant species, age, health, and physico-chemical properties of the root zone. Transpiration rate also varies dramatically and depends on the plant type, leaf area, nutrients, soil moisture, temperature, wind conditions, and relative humidity. Once the organic xenobiotic enters the plant system, it is partitioned to different plant parts through translocation. Unlike microbial species that metabolize organic contaminants to carbon dioxide and water, plants use detoxification mechanisms that transform parent chemicals to non-phytotoxic metabolites. The detoxification mechanism within plants is often described using the "green liver" concept [13, 14]. Once a contaminant enters the plant, any number of reactions within the following series may occur. • • •
Phase I - Conversion Phase II - Conjugation Phase III - Compartmentation
454
Fig. 6. Estimated transpiration stream concentration factors (TSCF) for BTEX using Eq. 2.
Conversion reactions include oxidations, reductions, or hydrolysis that the plant uses to begin detoxification. Conjugation reactions chemically link the Phase I products to glutathione, sugars, or amino acids and thus, the plant alters the solubility and toxicity of the contaminant. Once conjugated, xenobiotics can be removed as waste or compartmentalized. During compartmentation, chemicals are conjugated and segregated into vacuoles or bound to the cell wall material (hemicellulose or lignin). Phase III conjugates are often described as "bound residues" because chemical extraction methods do not recover the original contaminants. Trichloroethylene (TCE), which is not a hydrocarbon but is one of the more studied volatile organic compounds, has been shown to degrade to trichloroethanol, trichloroacetic acid, and dichloroacetic acid in hybrid poplars [15]. However, overall mass balances have been poor, indicating that other processes or further transformations that result in bound residues may be occurring [16]. Whereas Burken and Schnoor (1996) demonstrated that BTEX compounds translocate to the leaves, not much is known about the fate of BTEX compounds or other hydrocarbons in plants [17]. In general, the ultimate fate of phytotransformed contaminants with respect to C-cycling between a plant and its environment remains unclear. Concern centers on whether transformed contaminants will pose a threat to human or ecological health. Products of conversion reactions could be more
455
toxic than the parent contaminants when consumed by animals or potentially leached to the environment from fallen leaves [18]. Release of contaminants from conjugated complexes or compartmentalization could occur in the gut of a worm, snail, or butterfly [8]. This raises the potential of re-introducing the pollutant into the food chain. Therefore, a thorough understanding of pathways and end products of enzymatic processes within a plant is required if phytoremediation is to be applied successfully and accepted widely. 2.3. Phytovolatilization The natural ability of a plant to volatilize a contaminant that has been taken up through its roots can be exploited as a natural air-stripping pump system. Phytovolatilization is most applicable to those contaminants that are treated by conventional air-stripping i.e., contaminants with a Henry's constant KH > 10 atm m3 waterm"3 air, such as BTEX, TCE, vinyl chloride and carbon tetrachloride. Chemicals with KH < 10 atm m3 waterm"3 air such as phenol and PCP are not suitable for the air-stripping mechanism because of their relatively low volatility. Volatile pollutants diffuse from the plant into the atmosphere through open stomata in leaves. Radial diffusion through stem tissues has also been reported [19-21]. For example, methyl-tert-butyl ether (MTBE) can escape through leaves, stems, and the bark to the atmosphere [22-23]. Tree core samples of hybrid poplars exposed to TCE also showed radial diffusion from the stem [24] rather than transpiration from leaves [24, 25] as the main dissipation mechanism. Generally, the concentration of VOCs in the xylem decreases with increasing distance from the roots [24]. Once released into the atmosphere, compounds with double-bonds such as TCE and perchloroethylene (PCE) could be rapidly oxidized in the atmosphere by hydroxyl radicals. However, under certain circumstances (e.g., poor air circulation) phytovolatilization may not provide a terminal solution. For example, MTBE is long lived in the atmosphere and can pose a risk to shallow groundwater during precipitation [26]. In such cases, simple mass balance models can be utilized to determine if phytovolatilization poses a significant risk to humans and/or the environment [20, 24, 27]. Nevertheless, the rate of release of VOCs from plant tissues is generally small relative to other emissions [27]. Thus, phytovolatilization is a potentially viable remediation strategy for many volatile organic chemicals. 2.4. Rhizodegradation Microbial degradation in the rhizosphere might be the most significant mechanism for removal of diesel range organics in vegetated contaminated soils [28-34]. This occurs because contaminants such as PAHs are highly hydrophobic and their sorption to soil decreases their bioavailability for plant uptake and phytotransformation. Briggs (1982) first demonstrated that the
456
lipophilicity of a pesticide determines its fate in a barley plant [11]. High Kow values (an indicator of hydrophobicity) corresponded to a greater possibility that the compound would be retained in the roots (Eq. 3). Burken and Schnoor (1998) published similar results for the sorption of a wide range of organic contaminants to roots of hybrid poplar plants grown hydroponically (Eq. 4) [12]. log (RCF - 0.82) = 0.77 log Kow -1.52
(3)
log (RCF - 3.0) = 0.65 log Kow -1.57
(4)
Where the Root Concentration Factor (RCF) (L/kg dry roots) is the ratio of organic chemical sorbed on the root (mg/kg of fresh root tissue) to that in hydroponic solution (mg/L). This equilibrium partitioning coefficient has generally proved to be a good indicator of whether a plant retains a contaminant in the root, which increases the probability of microbial degradation (not withstanding significant bioavailability limitations). However, a few exceptions exist such as phenol and aniline, which bind irreversibly to the root (especially aniline) and are chemically transformed. They are not appreciably desorbed because they are covalently bound as metabolic products in plant tissue [35].
Fig. 7. Estimated Root concentration factors (RCF) for PAHs using Eq. 4.
457
Figure 7 uses Eq. 4 to estimate RCF values for a few common PAHs. The hydrophobic (high sorption) characteristics of PAHs and other DRO compounds result in high retention in the root zone. Fortunately, the rhizosphere of most plants promotes a wealth of microorganisms that can contribute significantly to the degradation of petroleum hydrocarbons during phytoremediation. Thus, though a plant may not directly act upon these contaminants, a plant can influence the microbial community within its root zone to a great extent. Potential rhizosphere interactions that may be important for phytoremediation of petroleum hydrocarbons include: 1. 2. 3. 4. 5.
Prolific microbial growth Repression/induction of catabolic enzymes Co-oxidation of contaminants Changes in bioavailability Chemotaxis of competent strains
Deposition of plant-derived carbon sources through root exudation, and/or root turnover provides rhizosphere bacteria with numerous organic substrates [36]. Rhizodeposition can account for release of 7 to 27 percent of the total carbon fixed during plant photosynthesis [37] and varies between plants. Commonly reported estimates are between 10 - 100 mg-C g-root material"1 [38] of which root exudation is reported to range between 0.4 - 27.7 mg-C g-root material"1 [39-41]. The composition and quantity of root-derived material released into the rhizosphere varies depending on the season [42], the age of plant [42] and the health of the plant [43] but generally contains sugars (15 65% total organic carbon), organic acids (9 - 33% total organic carbon), amino acids (2 - 31% total organic carbon) [34,39-40] and phenolics (0.3-4 mg-Cgroot material"1) [42-44]. Plant stress and age generally increase rhizodeposition. The availability of simple organic carbon sources that can be used for growth promotes rhizosphere microbial populations which have been reported to be 4- to 100- fold greater than that observed in surrounding bulk soils [33, 4548]. Selection of competent microorganisms during phytoremediation has been hypothesized. Miya and Firestone [28] observed greater percentages of phenanthrene degrading bacteria in rhizosphere soil than bulk soils and suggested the rhizosphere selected for PAH degraders. Siciliano et al. (2003) observed a higher frequency of catabolic genes in tall fescue rhizosphere than in bulk soil [49], suggesting that gene transfer or another mechanism of selection exists in the rhizosphere. However, the presence of contaminants in these experimental systems likely provided a strong selective pressure for competent strains [50]. Investigation of competent degraders within the rhizosphere of uncontaminated soil has not been reported; such studies are needed to provide conclusive evidence for selection of specific degraders through plant influence.
458
Induction of microbial aromatic degradation has also been hypothesized due to the deposition of phenolic compounds that are structurally analogous to known inducers of enzymes responsible for degradation of aromatic contaminants [51-52]. Gilbert and Crowley demonstrated induction of polychlorinated biphenyl (PCB) degradation in Arthrobacter sp. strain BIB, a gram-positive organism, using spearmint products and identified /-carvone as the compound responsible [52]. Interestingly, /-carvone was not a growth substrate tor Arthrobacter sp. strain BIB, and it inhibited growth of the bacteria on fructose. Induction of PAH degrading enzymes by plant root products has not been demonstrated in the literature. In a screening test of inducers of naphthalene dioxygenases potentially released by plants [53], none were detectable in root extracts at concentrations required for catabolic gene induction. Furthermore, Kamath et al., and Rentz et al. observed inhibition of catabolic enzyme activity on a per cell basis following exposure to environmentally relevant concentrations of plant root products (exudates and turnover) [53-54]. This was attributed to the presence of organic acids, carbohydrates, and amino acids, known repressors of aromatic catabolism within soil bacteria. However, both studies concluded that proliferation of competent genotypes through growth could compensate for the interference that labile substrates exert on the expression of PAH catabolic genes. Currently, little information concerning the expression of other catabolic enzymes during petroleum hydrocarbon phytoremediation is available. Several researchers have suggested that co-oxidation of high molecular weight (HMW) PAH within the rhizosphere [37,47-48] is an important mechanism for phytoremediation. Generally, HMW PAHs do not serve as carbon and energy source for microbial populations during degradation. The use of plants as a method to "inject" growth substrates to contaminated soil could overcome this limitation to degradation [28]. Soil experiments with plants and root exudates (pyrene, 4-rings) have shown degradation of HMW PAH and cooxidation was implied. However, oxidation or metabolism of HMW PAH has not been demonstrated using a well-defined system. Co-oxidation and cometabolism is likely an important process within the rhizosphere with the availability of a wide array of growth substrates, although no studies have assessed the importance of this mechanism compared to other processes. The bioavailabilitiy of hydrophobic contaminants may also be altered with the root zone environment. Exudation of organic acids could promote contaminant desorption from soil and solublization, but re-sorption to roots [55] may compete with microbial utilization. For carcinogenic and highly hydrophobic benzo[a]pyrene, sorption to roots could prove to be an acceptable end-point with respect to human and environmental risk. However, no studies have assessed the potential of this attenuation mechanism.
459
Chemotaxis of competent bacteria towards the rhizosphere may also enhance rhizoremediation. Ortega-Calvo et al. demonstrated chemotaxis of PAH-degrading rhizosphere bacteria towards root exudates [56]. Interestingly, these bacteria were also attracted to naphthalene and phenanthrene, but repelled by anthracene and pyrene. 4.5. Summary of mechanisms The different mechanisms discussed above could be utilized for the remediation of a wide variety of contaminants (Table 2). Phytoremediation could therefore be applied for the remediation of numerous contaminated sites. However, not much is known about contaminant fate and transformation pathways, including the identity of metabolites. Little data also exists on contaminant removal rates and efficiencies directly attributable to plants under field conditions. Therefore, further research is required before a tree can be designed as an engineered reactor system and optimized for efficiency at the field-scale. 3. PILOT STUDIES While numerous studies have been carried out at the lab-scale, very little has been published about field scale implementation of phytoremediation. Nedunuri et al. [57] investigated total petroleum hydrocarbon (TPH) removal at several field sites contaminated with crude oil, diesel fuel, or petroleum refinery wastes, at initial TPH concentrations of 1,700 to 16,000 mg/kg. Plant growth varied by species, but the presence of some species led to greater TPH disappearance than with other species or in unvegetated soil. At a crude oilcontaminated field site near the Gulf of Mexico, an annual rye-soybean rotation plot and a St. Augustine grass-cowpea rotation plot had significantly (P < 0.05) greater TPH disappearance than did sorghum-sudan grass or unvegetated control plots, at 21 months. At a diesel fuel-contaminated Craney Island field site in Norfolk, Virginia, the fescue plot had significantly (P < 0.10) greater TPH removal than did an unvegetated plot. At a refinery waste site, statistical analyses were not presented due to the short time since establishment of the plots, but Nedunuri et al. reported that qualitatively, the vegetated plots had greater TPH removal than the unvegetated control plots. After investigating the potential to use phytoremediation at a site contaminated with hydrocarbons, the Alabama Department of Environmental Management granted a site, which involved about 1500 cubic yards of soil of which 70% of the baseline samples contained over 100 ppm of total petroleum hydrocarbon (TPH). After 1 year of vegetative cover, approximately 83% of the samples contained less than 10ppmTPH[58].
460
Table 2 Potential clean-up mechanisms during phytoremediation of hydrocarboncontaminated sites based on physical properties of the target pollutants such as octanol-water partitioning coefficient (KoW) and Henry's dimensionless constant (KH). Contaminants
Sources
Gasoline Range Organics (GRO) Refineries, LUST, Fuel spills Gasoline Oxygenates
KoW* . ,
{ K H
,„
TITCT
Diesel Range Organics (DRO) Coal-gasification, TIATT Fpetroleum distillation, PAH . wood preservation, waste disposal
KH*
^1A4 >10
n m
,
ln-4
„ -s < 2 x l 1A O
Potential Removal Mechanisms Hydraulic Control Pnytovolatihzation Hydraulic Control Phytovolatilization
„, . ,. .. Rhizoremediation
4. FIELD SCALE CONSIDERATIONS Design of a phytoremediation system varies according to the contaminant/s, the conditions at the site, the level of clean up required, and the plant/s that are used. Nevertheless, it is possible to specify a few design considerations that are a part of most phytoremediation efforts. These include: • • • • • • •
Site Treatability Plant selection and planting density Irrigation, agronomic inputs and maintenance Cost Estimation Mathematical Modeling Clean-up time required Analysis of failure modes
4.1. Site Treatability 4.1.1. Source Removal For phytoremediation to succeed, it is very important to physically remove the source of contamination (e.g., excavation of highly-contaminated soil and/or extraction of free phase). The presence of a continuous source can be detrimental to the health of the plants and can extend the life of the phytoremediation project indefinitely.
461
4.1.2. Depth of Contamination Phytoremediation is most effective at sites with shallow (i.e., root accessible) contaminated soils where contaminants can be treated in the rhizosphere and/or by plant uptake. Roots of phreatophytic trees can be expected to grow at least 3 meters into a soil profile, and it is possible to encourage rooting to a depth of 5 meters or more using the tree-in-a-well concept [10]. On the other hand, roots of some grasses (alfalfa, switchgrass, tall fescue) can reach soil depths of only 0.25-0.4 m. Buffelgrass roots to a depth of 0.75 m but has been observed to have dense rooting pattern within 0.3 m from the topsoil layer. Hawaiian plants, Milo and Kou which were used to remediate saline soils contaminated with TPHs, rooted to a depth of more than 1.5 m by growing through the brackish water table into a zone of concentrated contaminants [59]. Optimizing irrigation patterns can also facilitate biodegradation of contaminants by creating an "expanded rhizosphere" due to translocation of organic root exudates and inorganic nutrients to relatively deep soil layers. Phytoremediation can therefore influence soils to the depth where irrigation water reaches, even if the roots are sparse in the contamination zone. 4.1.3. Soil composition and quality Soil quality is another important factor for determining successful germination, growth and health of plants. Heavily contaminated soils have a tendency towards poor physical conditioning which is unsuitable for vigorous growth of vegetation and rhizosphere bacteria. It is therefore critical to use amendments to improve the quality of soil before planting. Common limitations are poor moisture-holding capacity, insufficient aeration, low permeability and nutrient deficiencies. Agronomic soil analysis and preliminary greenhouse or pilot scale experiments can help identify these constraints. For example, nutrient analysis of contaminated soils from a site at the Unocal Bulk Storage Terminal at Superior, Wisconsin [54] indicated general deficiencies in nitrogen, phosphorus, potassium, and zinc. To decrease the soil pH, an addition of sulfur was also recommended. This information was subsequently used in greenhouse treatability studies, from which a formula of 50 lb/ac phosphorus, 225 lb/ac zinc, and 50 lb/ac potassium was identified as optimum for growth of native grasses. Organic amendments such as aged manure, sewage sludge, compost, straw, or mulch can be used to increase the water-holding capacity of a contaminated soil. Soil pH can be increased and decreased by the addition of lime and sulphur respectively. 4.1.4. Weather Phytoremediation might be best suited for tropical countries where plant growth occurs all year round. In temperate climates, the active contribution of phytoremediation is restricted to the growing period only. Winter operations
462
may pose problems for phytoremediation when deciduous vegetation loses its leaves, transformation and uptake cease, and soil water is no longer transpired. However, a combination of grasses can be used to prolong the growing period. 4.2. Plant Selection Criteria Plants should be selected according to the needs of the application, the contaminants of concern and their potential to thrive on contaminated soil. Design requirements should include the use of native plants to avoid introduction of invasive species. Apart from this, vegetation should be fast growing, hardy, easy to plant and maintain. The main aim is to ensure that roots expand throughout the entire contaminated zone. In temperate climates with shallow contaminated aquifers, phreatophytes, such as Populus sp. (hybrid poplar, cottonwood, aspen) and Salix sp. (willow) are often selected because of fast growth, deep rooting ability down to the surface of groundwater, large transpiration rates, and the fact that they are native throughout most of the country. Among tropical plants tested for use in Pacific Islands, three coastal trees, kou {Cordia subcordata), milo (Thespesia populnea), and kiawe (Prosopis pallida) and the native shrub beach naupaka (Scaevola sericd) tolerated field conditions and facilitated clean-up of soils contaminated with diesel fuel [59]. Grasses are often planted in tandem with trees at sites with organic contaminants as the primary remediation method. They provide a tremendous amount of fine roots in the surface soil, which is effective at binding and transforming hydrophobic contaminants such as TPH, BTEX, and PAHs. Grasses are often planted between rows of trees to provide for soil stabilization and protection against wind-blown dust that can move contaminants off-site. Legumes such as alfalfa (Medicago sativa), alsike clover (Trifolium hybridum ), and peas (can be used to restore nitrogen to poor soils. Fescue (Vulpia myuros), rye (Elymus sp.), clover {Trifolium sp.) and reed canary grass (Phalaris arundinacea) have been used successfully at several sites, especially petrochemical wastes. Once harvested, the grasses can be disposed off as compost or burned. Plant tolerance to high contaminant concentrations is also a very important factor to keep in mind. The phytotoxicity of petroleum hydrocarbons is a function of the specific contaminant composition, its concentration, and the plant species used. Major adverse effects typically include reduced germination and growth if contaminant concentrations are sufficiently high. In general, TPH values of 15 percent or greater can result in significant reductions in plant growth and in some cases mortality. Compared with uncontaminated soil, soils with 2% TPH reduced alfalfa yields by 32 percent [61]. Production of biomass by ryegrass was reduced 46 percent at a soil concentration of 0.5 percent (5000 mg/kg) hydrocarbons [47]. It was found that plants pre-grown in clean soil and subsequently transplanted to the contaminated soil grew nearly as well as the
463
control, showing that toxicity was associated with germination and/or early plant growth. Similarly, poor rooting of ryegrass compared to legumes appeared to adversely affect the removal of TPH from Gulf War-contaminated soils [62]. Also, although the germination of sunflower seeds and beans was greater than that of maize, vegetative growth was greater for maize than beans, demonstrating that germination and later plant growth may be affected differently [63]. Aged spills tend to be much less phytotoxic than fresh ones, possibly because of the lower bioavailability of toxic compounds in the aged spills. However, the speciation of petroleum hydrocarbons is also very important in determining phytotoxicity. A fuel oil with 30 percent aromatics resulted in LC50 germination (oil concentration lethal to 50 percent of test plants) values of 7 percent (70,000 mg/kg) for sunflower seeds. The volatile fraction can prove most toxic to plants. Aromatic volatile petroleum hydrocarbons such as benzene have been used as herbicides in the past years, illustrating their phytotoxicity when applied to plant leaves [64]. In contrast, no phytotoxic effects were observed in hybrid poplar trees exposed to a simulated groundwater containing a mixture of VOCs including BTEX, chlorinated aliphatics, and alcohols at a total concentration of 169mg/L [65]. Reduction of the volatile fraction may be accomplished through management, such as by tillage of the soil. If initial efforts at plant establishment at a site fail, replanting the area may ultimately lead to success as concentrations or bioavailability of the more phytotoxic components decline. Solution-phase concentrations of hydrocarbons are also important, particularly for aquifer remediation applications of phytoremediation. Additional components with phytotoxic effects include various unsaturated hydrocarbons and acidic hydrocarbons such as alicyclics with carboxylic acid groups (naphthenic acids) [64]. A screening test and knowledge from the literature of plant attributes is essential for selection of plants. Most experts recommend a mixture of grasses or legumes to address surface soils contaminated with petroleum hydrocarbons. However, design engineers should work in interdisciplinary teams that include a botanist and/or agricultural specialist to identify and select plants that will grow well at the site. Preliminary greenhouse studies should also be used to identify plants that can thrive and enhance transformation of contaminants of concern to non-toxic or less toxic products. The U.S. Department of Agriculture also provides two databases on plants (http://Plant-Materials.nrcs.usda.gov/ and http://plants.usda.gov/). For information specifically pertaining to plants used for phytoremediation of petroleum hydrocarbons, the Phytopet database compiled by the Department of Soil Science, University of Saskatchewan in co-operation with Environment Canada is available at http://www.phytopet.usask.ca.
464
4.2.1. Time scale of clean-up Degradation of organics may be limited by mass transfer, i.e., desorption and mass transport of chemicals from soil particles to the aqueous phase may become the rate determining step. Therefore, phytoremediation may require more time (see Section 4) to achieve clean-up standards than other more costly alternatives such as excavation or ex-situ treatment, especially for hydrophobic pollutants that are tightly bound to soil particles. In many cases, phytoremediation may serve as a final "polishing step" to close sites after more aggressive clean-up technologies have been used to treat the hot spots. 4.2.2. Plant Density Planting density depends on the application. Louis Licht, Ecolotree® Inc., (http://www.ecolotree.com), pioneered the use of hybrid poplar trees as riparian zone buffer strips, landfill caps, and at hazardous waste sites. For hybrid poplar trees, 1000-2000 trees per acre are typically planted with a conventional tree planter at 12-18 inches depth or in trenched rows 1-6 ft deep. The poplars are planted simply as "sticks", long cuttings that will root and grow rapidly in the first season. Several phreatophytes in the Salix family, such as willow and cottonwood, can be planted in a similar manner. Poplars have the ability to root along the entire buried depth. If a row conformation is used, the trees may be spaced with 2 ft between trees and 10 ft between rows. Hardwood trees and evergreens may require a lower planting density initially. Projects using hydraulic control are most effective at canopy closure, when transpiration is maximized (within 5-6 years). Theoretically, this can be determined based on the amount of energy received from the sun and that required to evaporate water. For mid-latitudes during the growing season, the earth receives an average 30 million Joules per square meter per day (30 x 106 J m"2 d"1) of solar insolation. It takes about 2.5 x 106 Joules to evaporate one liter of water. Thus, it is thermodynamically possible to evaporate 12 L m"2 d"1. But no plant is 100% efficient, and energy is required to lift the water from the groundwater to the atmosphere with friction. Typical crops, like corn, can evapotranspire about 4-5 L m"2 d"1 during their growth period. Poplars can perform about 30% more efficiently than corn if they are rooted in the groundwater table, but they actively transpire only about 4-6 months of the year (due to seasonal changes), depending on the geographic location. Thus, the best that can be expected from a phytoremediation effort where the trees have canopied and are rooted in the groundwater table is 4.5 L m" d"1 x 1.3 x 6/12 x 365 days per year x (lnrVlOOO L) = 1.07 m/yr, which is approximately one million gallons per acre per year. Typically, evapotranspiration rates range from about 0.4-1.0 million gallons per acre per year for a good phytoremediation effort using phreatophyte trees rooted into shallow groundwater.
465
A high initial planting density assures a significant amount of evapotranspiration in the first year which is normally desirable, but the trees will naturally thin themselves by competition to 600-800 trees per acre over the first six years. If desirable, hybrid poplars can be harvested on a six-year rotation and sold for fuel wood or pulp and paper, and the trees will grow back from the cutstump (coppicing trait). The dense, deep root system stays in place to sustain growth for the next year. The lifetime of hybrid poplars such as Populus deltoides x nigra DN-34 (Imperial Carolina) is on the order of 30 years which is usually sufficient as the design life of the project. 4.3. Agronomic Inputs 4.3.1. Irrigation Results suggest that irrigation can enhance bioremediation of certain diesel components. For terrestrial phytoremediation applications, it is often desirable to include irrigation costs on the order of 10-20 inches of water per year, in the design. Spray irrigation is less efficient than drip irrigation as it encourages the growth of weeds that compete for nutrients with plants and hinder their delivery to the contaminated zone. Irrigation of the plants is especially important during the start of the project. However, after the first year, hydrologic modeling can beused to estimate the rate of percolation to groundwater under irrigation conditions. Over time, irrigation can be withdrawn from the site, provided the area receives sufficient rainfall to sustain the plants. 4.3.2. Fertilizer Requirements Contaminated soils are usually deficient in macro- and micro-nutrients (Table 3) necessary for establishing healthy vigorously growing plants and stimulating microbial contaminant degradation. Nitrogen fertilization of motor oil-contaminated soils was found to increase the growth of corn and reduce what appeared to be nitrogen-deficient yellowing of the leaves [66]. The source of nutrients also appeared to affect the germination and growth of plants. Organic sources of nitrogen are better than inorganic sources. This is probably because organic nitrogen sources provide a slow release source of nitrogen, and also help to improve soil structure and soilwater relationships for plant growth. It was found that poultry manure increased the growth of corn in a soil containing 3 percent weight per volume crude oil more than an inorganic fertilizer containing nitrogen, phosphorus, and potassium [67]. The addition of sawdust alone improved germination by decreasing oil contact with seeds, but accentuated the adverse effect of the oil on later growth, apparently by further widening the carbon-to-nitrogen ratio [67].
466
Table 3 Macro- and Micro-nutrients required for healthy plant growth. Macronutrientsa (-100 ppm) Micronutrientsb (~1 ppm) Nitrogen (N) Iron (Fe) Phosphorus (P) Boron (B) Potassium (K) Zinc (Zn) Magnesium (Mg) Copper (Cu) Calcium (Ca) Manganese (Mn) Sulfur (S) Molybdenum (Mo) http://extension.oregonstate.edu/mg/botany/table3.html b http://extension.oregonstate.edu/mg/botany/table4.html
With respect to TPH degradation, nutrient addition during phytoremediation has yielded mixed results. Hutchinson et al. observed better degradation of TPH using grasses with N/P amendments than without inorganic amendments [68]. Joner et al. reported improved degradation of 3 and 4 ringed PAHs with the addition of N/P, but diminished degradation of 5 and 6 ringed PAHs [69]. Finally, Palmroth et al. observed no improved degradation of diesel fuel with nutrient amendments during phytoremediation with pine, poplar, or grasses [70]. Microbial bioremediation of TPH contaminants with nutrient addition also produced widely varying results. Diesel fuel degradation was stimulated with the addition of N/P using cold region soils [71] and P amendments stimulated creosote degradation [72]. Breedveld and Sparrevik observed improved degradation of 4 ringed PAHs with N/P addition, but no increased degradation of 3 ringed PAHs [73]. However, Graham et al. assessed an array of N/P amendments for hexadecane biodegradation and suggested amendments above stoichiometric requirements can lead to diminished rates of degradation [74]. This potentially occurs because addition of excessive nitrogen additions results in an increase in soil salinity and this increases the osmotic stress and suppresses the activity of hydrocarbon-degrading organisms [71]. Carmichael and Pfander observed slower degradation of 3 and 4 ringed PAHs with N addition and no effects for P addition [75]. Johnson and Scow reported similar results indicating N/P addition inhibited or did not change phenanthrene degradation (3 ringed PAH) [76]. Their results showed that soil with initial low concentrations of N or P is more likely to show decreased degradation with N/P addition. Many PAH-degrading organisms are adapted to low nutrient conditions and activity may decrease with the addition of soil amendments. Thus, addition of nutrients should be considered on a site-by-site basis and a balance should be considered between biodegradation and plant growth. Application of amendments exclusively for plant growth may result in diminished contaminant degradation, the ultimate goal of phytoremediation.
467
4.3.3. Oxygen requirements Soil oxygen is required for optimal aerobic microbial degradation of petroleum hydrocarbon contaminants. Similar to nutrient deficiencies, oxygen depletion is caused by natural microbial respiration of contaminants. Within phytoremediation, plants may be a net positive or negative oxygen source [77]. Plants may improve soil oxygen through two mechanisms. First, specially adapted plants use aerenchyma, channels of reduced air resistance, to transport oxygen to the root zone, enhancing aerobic biological degradation [37, 78]; although there are no reports of aerenchyma within hybrid poplars, the subject of this report. Second, soil dewatering and fracturing increases soil porosity, allowing increased diffusion of atmospheric oxygen [6]. Plant roots can also be a net oxygen sink within petroleum-contaminated soils. Rentz et al. [79], observed stimulation of hybrid poplar growth and increased poplar root density with the addition of Oxygen Release Compound® (ORC) when plants were grown in petroleum smear zone soils (high biochemical oxygen demand). Flux of oxygen into soil by plants could be offset by root turnover and root exudation that provides microbial populations with simple carbon sources that could deplete soil oxygen when metabolized [80]. Furthermore, plant roots are known to require oxygen [81]. For soils with a high biochemical oxygen demand, oxygen addition may be required to promote plant growth and stimulate microbial degradation. Passive methods of oxygen delivery are suggested to keep costs of phytoremediation low and include the following. Perforated aeration tubes, placed next to cuttings, can supply oxygen to roots along a vertical axis [82]. Perforated ADS tubing, placed at depth prior backfilling the planting trench provides oxygen on a horizontal plane. Gravel used to backfill planting trenchs allows permeation of oxygen on vertical and horizontal axis. Finally, the use of solid peroxides (e.g. Oxygen Release Compound®) can provide oxygen to soils when in contact with water [83]. 4.4. Cost Phytoremediation is usually less costly than competing alternatives such as soil excavation, pump-and-treat, soil washing, or enhanced extraction. Apart from costs incurred during installation of vegetation at the site, a field-scale phytoremediation project involves expenditure on design, site preparation, reporting, monitoring, and operation and maintenance. It would be prudent to include preliminary greenhouse experiments along with agronomic soil testing during the design phase to ensure vigorous plant growth at the field-site. Mathematical modeling may be necessary to demonstrate the effectiveness of the technology to regulatory agencies (See section 6).
468
Including all these costs, the start-up cost for phytoremediation at $10,000 - 25,000/ acre is still considerably less expensive than other competing technologies (Table 4). However, since phytoremediation usually requires five or more years, it is very important to make sure that funding for operation and maintenance is available during the life of the project. 4.5. Operation and Maintenance Issues Operation and maintenance (O & M) is vital to ensure vigorous growth of plants. Some of the major problems in the field have been weeds, killing frosts or drought, insect or disease infestation, beaver or deer browse, and damage by voles. It has been estimated that at least 30 percent of the plants may need to be replanted in the second or third year. Phreatophytic trees are also a source of concern since there is a potential for the expanding roots to enter and restrict flow of subsurface drains and sewers and break power and communication cables and small pipelines. Further, mowing, pruning, harvesting, monitoring vegetation for contaminants, irrigation and fertilizer costs should be included in the initial estimated costs. Jordahl, et al. (2002) provides a good summary of key siting and O&M issues that occur during the life of a field-scale project [85]. Table 4. Five-Year Cost Comparison of Phytoremediation by Hybrid Poplar Trees versus Conventional Pump and Treat [84] 1. Phytotransformation Design and Implementation $ 50,000 Monitoring Equipment Capital 10,000 Installation 10,000 Replacement 5,000 5-Year Monitoring Travel and administration 50,000 Data collection 50,000 Reports (annual) 25,000 Sample analysis 50,000 TOTAL $ 250,000 2. Pump and Treat (3 wells and Reverse Osmosis System) Equipment $ 100,000 Consulting 25,000 Installation/Construction 100,000 5-Year Costs Maintenance 105,000 Operation (electricity) 50,000 Waste disposal 180,000 Waste disposal liability 100,000 TOTAL $ 660,000
469
5. Mathematical Modeling 5.1. Groundwater Capture and Transpiration One must understand where the water is moving at a site in order to estimate contaminant fate and transport. For applications involving groundwater remediation, a simple capture-zone calculation [86] can be used to estimate whether the phytoremediation "pump" can be effective at intercepting and extracting the plume of contaminants. Trees can be grouped for consideration as average withdrawal points. The goal of such a phytoremediation effort is to create a water table depression where contaminants will flow to the vegetation for uptake and treatment or volatilization. It is important to realize that organic contaminants are not taken-up at the same concentrations that are present in the soil or groundwater. Rather, there is a transpiration stream concentration factor (a fractional efficiency of uptake) that accounts for the partial uptake of contaminant (due to membrane barriers at the root surface). U = (TSCF) (T) (C)
(5)
where: U = uptake rate of contaminant, mg/day TSCF = transpiration stream concentration factor, dimensionless T = transpiration rate of vegetation, L/day C = aqueous phase concentration in soil- or ground-water, mg/L A method for estimating the Transpiration Stream Concentration Factor (TSCF) for eq. (5) was given by eq. (1) and (2). If the contaminant plume is not completely taken-up by the vegetation, the plume that remains could be evapoconcentrated; i.e., the mass of contaminant in the plume will be less due to uptake by vegetation, but the concentration remaining will actually be greater due to preferential uptake of water over the contaminants. This is a potential concern for phytoremediation of groundwater plumes or created wetlands, where a relatively hydrophilic contaminant can be concentrated on the downstream side of the phytosystem. Mature phreatophyte trees (poplar, willow, cottonwood, aspen, ash, alder, eucalyptus, mesquite, bald cypress, birch and river cedar) typically can transpire 3-5 acre-ft of water per year (36-60 inches of water per year). This is equivalent to about 600-1000 gallons of water per tree per year for a mature species planted at 1500 trees per acre. Transpiration rates in the first two years would be somewhat less, about 200 gallons per tree per year, and hardwood trees would transpire about half the water of a phreatophyte. Two meters of water per year is a practical maximum for transpiration in a system with complete canopy coverage (a theoretical maximum would be 4 m/yr based on the solar energy
470
supplied at 40°N on a clear day that is required to evaporate water). If evapotranspiration of the system exceeds precipitation, it is possible to capture water that is moving vertically through soil. Areas that receive precipitation in the wintertime (dormant season for deciduous trees) must be modeled to determine if the soil will be sufficiently dry to hold water for the next spring's growth period. The Corps of Engineers Hydrologic Evaluation of Landfill Performance (HELP) model (Vicksburg, Mississippi) and other codes have been used to estimate vertical water movement and percolation to groundwater. 5.2. Contaminant Uptake and Clean-up Time From equation (5) above, it is possible to estimate the uptake rate of the contaminant/s. First order kinetics can be assumed as an approximation for clean-up time. The uptake rate should be divided by the mass of contaminant remaining in the soil: k = U/Mo
(6)
where: k = first order rate constant for uptake, yr"1 U = contaminant uptake rate, kg/yr Mo = mass of contaminant initially, kg Then, an estimate for mass remaining at any time is expressed by equation (7) below. M = Moe"kt where: M = mass remaining, kg t = time, yr Solving for the time required to achieve clean up of a known action level: t = -(lnM/M 0 )/k(8) where: t = time required for clean-up to action level, yr M = mass allowed at action level, kg Mo = initial mass of contaminant, kg
(7)
471
Equations (5-8) can be applied to most sites where soil clean-up regulations are known for metals or organic contaminants. 5.3. Rhizodegradation The Root Concentration Factor, which was previously described (eq. (3) and eq. (4)) is defined as the ratio of the contaminant in roots to the concentration dissolved in soil water (fig/kg root per |ag/L). It is important in estimating the mass of contaminant sorbed to roots in phytoremediation systems. While RCF is a simple indicator of whether a contaminant will be retained on the root surface, mathematical modeling of the removal of contaminants in the rhizosphere is highly complex. The most sophisticated rhizosphere fate model available is the Pesticide Root zone model (PRZM) available from the EPA (http://www.epa.gov/oppefedl/models/water/index.htm). It allows for the estimation of the fate of pesticides in the root zone through hydrologic and chemical transport simulation. The processes of plant uptake, surface run-off, erosion, decay, volatilization, advection, dispersion and adsorption are considered. However, for PAHs and other highly hydrophobic contaminants, factors such as microbial mobility, spatial variability, plant root growth and depth of root penetration, root turnover and rhizosphere volume are probably more important. Current models [87] are built on a conceptual framework in which the soil-plant contaminant system is compartmentalized into multiple zones: the root itself, a series of root influenced zones (the rhizosphere), a decaying root zone and a non-root-influenced zone (the bulk soil). The essence of the system conceptualization is that each of the modeled zones is treated as a variable volume, uniformly mixed continuous reactor. The change in each zone's volume over time is determined from a pair of forcing functions that describe the specific growth and senescence rates of the plant system. Thus, as the new roots penetrate the soil and the associated microbial community is established, bulk soil will be transformed into rhizosphere soil. Similarly, as root senescence occurs, the root and rhizosphere volume will be converted into decaying root zone that ultimately returns, through humification, to bulk soil. Different growth and senescence functions can be used to simulate various grass species growth and biomass production patterns throughout an annual cycle. The model also includes the idea that the rhizosphere bacteria will face a gradient of influencing factors as the distance from roots increases. The model is useful to identify important variables from those with only minor effects, and to extrapolate results for one geographic region to another, based on the patterns of interaction between physical and biological factors. However, it does not take into account the effect of temperature and availability of nutrients such as nitrogen, phosphorus, soil oxygen, moisture and pH on
472
degradation rates. Also, it cannot simulate growth of multiple plant species that might be used in field-scale applications. 6. REGULATORY ISSUES Compliance with regulatory concerns is a critical factor when considering remediation of a site. State and federal acceptance of the technology has been slow but is the product of input by the Interstate Technology and Regulatory Cooperation Work Group (ITRC), the Superfund Innovative Technology Evaluation (SITE) program and the Research Technologies Demonstration Forum (RTDF) program of EPA. The Phytotechnologies Work Team, a part of the ITRC (www.itrcweb.org), published a Decision Tree (1999) and a Guidance Document (2001) as a first approximation for whether phytoremediation is suitable for a particular site. The latter guidance document in combination with the USEPA document titled "Introduction to Phytoremediation " (EPA 600-R99-107) should be useful in guiding industrial site managers. Apart from the ITRC, the SITE program and RTDF were also designed to evaluate the potential of phytoremediation for field-scale purposes. Phytoremediation has been the subject of six SITE investigations and over 25 field trials by RTDF (http://www.rtdf.org). SITE is a formal program established by EPA's Office of Solid Waste and Emergency Response (OSWER) and the Office of Research and Development (ORD) in response to the Superfund Amendments and Reauthorization Act of 1986 (SARA). Consultants are responsible for operating the innovative system on site and are expected to pay the costs of the demonstration, together with site owners. EPA is responsible for project planning, sampling and analysis, quality assurance and quality control, preparing reports, disseminating information, and transporting and disposing of treated waste materials. Under Superfund laws, EPA (1998) [88] lists nine criteria for consideration: 1. Overall protection of human health and the environment 2. Compliance with Applicable and Relevant and Appropriate Requirements 3. Long-term effectiveness and permanence 4. Reduction of contaminant toxicity, mobility, or volume 5. Short-term effectiveness (including the length of time needed to implement the technology and associated risks to workers, residents, and the environment during that time) 6. Implementability (including availability of goods and services) 7. Cost including capital, operation and maintenance, and monitoring 8. State (and federal) acceptance of the technology and its evaluation of its performance
473
9. Community acceptance which is addressed in the Record of Decision (ROD) Amendment (including responsiveness summary that presents public comments and responses to those comments) Of these, phytoremediation addresses concerns about aesthetics, cost, ease of implementation and community acceptance. Phytoremediation also has an advantage over constructed remedies in the long run. Unlike other remediation technologies, the efficiency of phytoremediation increases with time until the system reaches its maximum during canopy closure. Further, since it is possible to monitor the effect of phytoremediation in mitigating vertical percolation of contaminants as well as soil erosion, it fulfills the criteria required by Risk Based Corrective Action (RBCA) as well as Monitored Natural Attenuation (MNA). For most other actions including Voluntary Programs, it is usually sufficient to show that the cover is lush and growing and that phytoremediation meets routine (quarterly to annual) groundwater monitoring requirements. There are certain regulatory limitations to applying phytoremediation to a site. Phytoremediation is passive technology. Meeting cleanup goals might be difficult and could require 10 years or more without a guarantee of reaching specific performance standards. Furthermore, if phytoremediation is to be used in conjunction with Monitored Natural Attenuation (MNA), it is necessary to demonstrate that the plume (or contaminated zone) is stable or shrinking and that it is not causing unacceptable risk to humans or the environment. In addition, proof that the contaminants are not in danger of moving off the site, and knowledge of the mechanism of degradation (metabolites, pathways, products) and/or immobilization/sequestration is required. The following is a list of environmental monitoring requirements that are often appropriate for phytoremediation efforts. • Tree survival rates and replacement requirements. • Plant (leaf area index) or root densities and replacement requirements. • Levels of contaminants and/or metabolites measured in leaves or grasses. Quarterly groundwater monitoring for applicable or relevant and appropriate requirements (ARAR). • Sap flow or evapotranspiration estimates to calculate volume of water treated. • Soil gas measurements and oxygen profiles with soil depth to demonstrate aerobic degradation of aromatic constituents or gradual improvement. • Soil corings to demonstrate that treatment is occurring at the site (heterogeneity makes this monitoring requirement imprecise and sometimes misleading)
474
Nevertheless, the fate of contaminants taken up by the plant or transformed in the rhizosphere is not well-understood and it can sometimes prove difficult to show that the technology reduces toxicity of the contaminants, prevents cross-media transfer of pollutants and/or reduces risks to human and ecological receptors. Furthermore, since the distribution and composition of contaminants in field-scale projects is very heterogeneous, it is almost impossible to prove that phytoremediation enhances the rate of contaminant removal at field sites. In summary, long-term monitoring and evaluation of phytoremediation technology is still needed to demonstrate efficacy, to further define suitable plants and applications, and to gain acceptance from regulatory agencies. 7. EMERGING ETHICAL ISSUES, OPPORTUNITIES AND CHALLENGES One emerging issue requiring consideration is the use of plants that could be genetically modified to exhibit beneficial traits for phytoremediation, such as increased water uptake for hydraulic control, drought and pest tolerance, and increased enzyme activity for faster and more complete phytotransformation of organic contaminants. A similar potential innovation is the inoculation of the rhizosphere with genetically modified organisms (GMOs) that overexpress catabolic enzymes for enhanced rhizoremediation. The use of (microbial and/or plant) GMOs represents a research frontier with broad implications. The potential benefits of using GMOs are significant, and extend beyond improved contaminant removal efficiency and lower O&M costs. For example, GMO's might facilitate coupling phytoremediation with the production of marketable non-food (cash) crops that could be used for energy production (e.g., biomass production for fuel wood, biodiesel, or fuel ethanol) or raw materials for commercial products (e.g., pulp for paper or feedstock for cosmetic or pharmaceutical industries). Nevertheless, although GMOs have been extensively used in agriculture, little research has been conducted to assess their long-term life cycle impacts, including the consequences of increased genetic drift across species on biodiversity and biological community structure. This gives rise to much speculation and polarization regarding the consequences of in vitro genetic manipulation, which represents a significant political barrier to the use of GMOs in phytoremediation. Furthermore, the need for GMOs may be questionable for many projects, considering that indigenous species often perform adequately and that we have not tapped the full potential of wild species due to our limited understanding of various phytoremediation mechanisms, including the regulation of enzyme systems that degrade pollutants. In summary, the potential benefits and risks associated with the use of genetic manipulation suggest that we need to be very cautious of GMOs, but not
475
necessarily rule out their application in phytoremediation yet. Additional scientific input will hopefully contribute to dissipating myths, discern the benefits and consequences of using GMOs, and ensure their safe use when their application is justified. 8. CONCLUSIONS Over the past 15 years, phytoremediation has developed into a more acceptable technology for the remediation of soils and groundwater polluted with residual concentrations of petroleum hydrocarbons. However, regulators as well as consumers are still wary about the efficiency, predictability and applications of the technique. The ITRC guidelines and decision tree has supported the use of phytoremediation for most field-scale applications. Yet, at this point there is an urgent need for strong evidence supporting the potential of phytoremediation in protecting human as well as ecological receptors from exposure to contaminants, using rigorous methods of risk analysis. For direct application to field projects, it would be desirable if more protocols for designing preliminary greenhouse experiments reflecting field-environments and cheap innovative methods of encouraging growth of healthy plants were published. Research examining the long-term fate of contaminants in the environment would be particularly relevant. Also important is the difficult task of evaluating acceptable endpoints (e.g., humification) using standard ecological toxicity or bioavailability assays that might support phytoremediation. Albeit, phytoremediation is an emerging technology that is based on sound ecological engineering principles. Phytoremediation is a practical and cost-effective approach with aesthetical and atmospheric-carbon-sequestration ancillary benefits, and is particularly attractive for rural areas with residual and shallow contamination. Phytoremediation also holds great potential to manage a wide variety of environmental pollution problems, including the cleanup of soils and groundwater contaminated with hydrocarbons and other hazardous substances, the attenuation of pollutants dispersing through the environment in agricultural drainage, landfill leachates, and other forms of surface runoff or sub-surface migration, and the assimilation of industrial wastewater effluents to support efforts to move towards a zero-discharge policy from industrial facilities (e.g., refineries). Although phytoremediation is not a panacea that would be universally applicable, it is rapidly achieving pedagogical maturity and it has already earned an important place in the menu of alternatives from where we select solutions for our environmental pollution problems.
476
REFERENCES [I] [2] [3] [4] [5] [6] [7] [8] [9] [10] [II] [12] [13] [14]
[15] [16] [17] [18] [19] [20] [21]
[22] [23] [24] [25]
E. McElroy, J. W. Farrington, and J. M. Teal, In: U. Varanasi (ed.) Metabolism of Polycyclic Aromatic Hydrocarbons in the Aquatic Environment, CRC Press Inc., Boca Raton, Florida, 1989. M. A. Heitkamp, and C. E. Cerniglia, Appl. Environ. Microbiol., 54 (1988) 1612. International Agency for Research on Cancer (IARC), Monographs on the Evaluation of the Carcinogenic Risk of Some Industrial Chemicals to Humans, Lyon, France, 2000. K. Mortelmans, S. Harworth, T. Lawlor, W. Speck, B. Tainer and, and E. Zeiger, Environ. Mutagen, 8 (1986) 1. National Research Council, Report of the National Research Council Committee on Groundwater Cleanup Alternatives, National Academy Press, Washington, DC, 1994. Environmental Protection Agency (EPA), Treatment Technologies for Site Cleanup: Annual Status Report (Tenth Edition), 2001. Office of Solid Waste and Emergency Response (OS WER), FY 2003 Semi-Annual (Mid-Year) Activity Report, 2003. J. L. Schnoor, L. A. Licht, S. C. McCutcheon, N. L. Wolfe, and L. H. Camera, Environ. Sci. Technol, 29 (1995) 318A. P. A. Domenico and F. W. Schwartz, Physical and Chemical Hydrogeology, Wiley, New York, 1998. E. G. Gatliff, Remediation, No. 8 (1994) 343-352. G. G. Briggs, R. H. Bromilow, and A. A. Evans, Pesticide Sci., 13 (1982) 495. J. G. Burken and J. L. Schnoor, Environ. Sci. Technol., 32 (1998) 3379. H. Sandermann Jr., Pharmacogenetics, 4 (1994) 225. H. Ohkawa, H. Imaishi, N. Shiota, T. Yamada, and H. Inui, (eds. G. T. Brooks and T. R. Roberts). In: Pesticide Chemistry and Bioscience-The Food Environment Challenge, Special Publication 233, The Royal Society of Chemistry, Cambridge, UK, 1999, pp. 259-264. L. A. Newman, S. E. Strand, N. Choe, J. Duffy, G. Ekuan, M. Ruszaj, B. B. Shurtleff, J. Wilmoth, P. Heilman, and M. P. Gordon, Environ. Sci. Technol., 31 (1997) 1062. W. E. Schnabel, A. C. Dietz, J. G. Burken, J. L. Schnoor, and P. J. J. Alvarez, Wat. Res., 31 (1997)816. J. G. Burken, J. L. Schnoor, J. Environ. Eng. ASCE, 122 (1996) 958. J.M. Yoon, B.T. Oh, C.L. Just and J.L. Schnoor, Environ. Sci. Technol. 36 (2002) 4649. Q. Zhang, L. C. Davis, and L. E. Erickson, Environ. Sci. Technol., 35 (2001) 725. M. Narayanan, L. E. Erickson, and L. C. Davis, Environ. Progress, 18 (1999) 231. L. C. Davis, D. Lupher, J. Hu, and L. E. Erickson, (eds. L. E. Erickson and M. M. Rankin), Proceedings of the 1999 Conference on Hazardous Waste Research, Kansas State University, Manhattan, Kansas, (1999) pp. 219-223, http://www.engg.ksu.edu/HSRC. M. S. Hong, W. F. Farmayan, I. J. Dortch, C. Y. Chiang, S. K. McMillan, and J. L. Schnoor, Environ. Sci. Technol., 35 (2001) 1231. S. Trapp, J. C. McFarlane (eds.), Plant Contamination: Modeling and Simulation of Organic Chemical Processes, Lewis Publishers, Boca Raton, Florida, 1994. X. Ma and J. G. Burken, Environ. Sci. Technol., 37 (2003) 2534. L. A. Newman, S. L. Doty, K. L. Gery, P. E. Heilman, I. Muiznieks, T. Q. Shang, S. T. Siemieniec, S. E. Strand, X. P. Wang, A. M. Wilson, and M. P. Gordon, J. Soil Contamination, 7 (1998) 531.
477
[26] J. F. Pankow, N. R. Thompson, R. L. Johnson, A. L. Baehr, and J. S. Zogorski, Environ. Sci. Technol., 31(1997)2821. [27] E. W. Aitchison, S. L. Kelley, P. J. J Alvarez, and J. L. Schnoor, Wat. Environ. Res., 72 (2000)313. [28] W. Aprill, and R. C. Sims, Chemosphere, 20 (1990) 253. [29] M. K. Banks, E. Lee, A. P. Schwab, J. Environ. Qual, 28 (1999) 294. [30] P. Binet, J. M. Portal, and C. Leyval, Soil Biol. Biochem., 32 (2000) 2011. [31] H. H Liste and M. Alexander, Chemosphere, 40 (2000) 11. [32] K. A. Reilley, M. K. Banks, and A. P. Schwab, J. Environ. Qual., 25 (1996) 212. [33] R. K. Miya and M. K. Firestone, J. Env. Qual, 29 (2000) 584. [34] R. K. Miya and M. K. Firestone, J. Env. Qual. 30 (2001) 1191. [35] S. Lang, M. S. Thesis, University of Iowa 1998. [36] M. B. Leigh, J. S. Fletcher, X.O. Fu, and F. J. Schmitz, Environ. Sci. Technol, 36 (2002) 1579. [37] J.F. Shimp, J.C. Tracy, L.C. Davis, W. Huang, L.E. Erickson, and J.L. Schnoor, Crit. Rev. Environ. Sci. Technol, 23 (1993) 41. [38] J. M. Whipps and J. M. Lynch, Ann. Proc. Phytochem. Soc, 26 (1985) 59. [39] W. Hiitsch, J. Augustin, and W. Merbach, J. Plant Nutr. Soil Sci, 165 (2002) 397. [40] Kraffczyk, G. Trolldenier, and H. Beringer, Soil Biol. Biochem. 16 (1984) 315. [41] J. A. Trofymow, D. C. Coleman, and C. Cambardella, Plant Soil, 97 (1987) 333. [42] R. S. Hegde and J. S. Fletcher, Chemosphere, 32 (1996) 2471. [43] Z. Rengel, Plant Soil, 245 (2002) 59. [44] H. Wu, T. Haig, J. Pratley, D. Lemerle, and M. An, J. Agric. Food Chem, 48 (2000) 5321. [45] H. Chaineau, J. L. Morel, and J. Oudot, J. Environ. Qual, 29 (2000) 569. [46] J. L. Jordahl, L. Foster, J. L. Schnoor, and P. J. J. Alvarez, Environ. Toxicol. Chem, 16 (1997)1318. [47] T. Gunther, U. Dornberger, and W. Fritsche, Chemosphere, 33 (1996) 203. [48] T. D. Nichols, D.C. Wolf, H.B. Rogers, C. A. Beyrouty, and C. M. Reynolds, Wat. Air Soil Poll, 95 (1997) 165. [49] S. D. Siciliano, J. J. Germida, K. Banks, and C. W. Greer. Appl. Environ. Microbiol, 69 (2003) 483. [50] S. D. Siciliano, N. Fortin, A. Mihoc, G. Wisse, S. Labelle, D. Beaumier, D. Ouellette, R. Roy, L. G. Whyte, M. K. Banks, P. Schwab, K. Lee, and C. W. Greer, Appl. Environ. Microbiol, 67 (2001) 2469. [51] J. S. Fletcher and R. S. Hegde, Chemosphere, 31 (1995) 3009. [52] S. Gilbert and D. E. Crowley, Appl. Environ. Microbiol, 63 (1997) 1933. [53] R. Kamath, J. L. Schnoor, and P. J. J. Alvarez, Environ. Sci. Technol, (In Press, 2004) [54] J. A. Rentz, P. J. J. Alvarez, and J. L. Schnoor, Environ. Microbiol, (hi Press, 2004). [55] A. P. Schwab, A. A. Al-Assi, and M. K. Banks, J. Environ. Qual, 27 (1998) 220. [56] J. J. Ortega-Calvo, A. I. Marchenko, A. V. Vorobyov, and R. V. Borovick. FEMS Microbiol. Ecol, 44 (2003) 373. [57] K. V. Nedunuri, R. S. Gouindaraju, M. K. Banks, A. P. Schwab, and Z. Chen, J. Environ. Eng, 126 (2000) 483. [58] D. Hecht and G. Badiane, New Internationalist, June (1998) 12. [59] U.S. Army Corps of Engineers, Agriculturally Based Bioremediation of PetroleumContaminated Soils and Shallow Groundwater in Pacific Island Ecosystems, May 2003. [60] K. Precht, M. S. Thesis, University of Iowa 2003.
478
[61] C.C. Wiltse, W. L. Rooney, Z. Chen, A. P. Schwab, and M. K. Banks, J. Environ. Qual., 27 (1998) 169. [62] Yateem, A., A.S. El-Nawawy, and N. Al-Awadhi,. Soil and Groundwater Cleanup, 1999, pp. 31-33. [63] C. H. Chaineau, J. L. Morel, and J. Oudot, J. Environ. Qual., 26 (1997) 1478. [64] Baker, J.M. 1970. Environ. Pollut., 1 (1970) 27. [65] Ferro, J. Kennedy, R. Kjelgren, J. Rieder, and S. Perrin, Int. J. Phytoremediation, 1 (1999) 9. [66] J. Giddens, J. Environ. Qual., 5 (1970) 179. [67] Amadi, A. A. Dickson, and G.O. Maate, Wat. Air Soil Pollut., 66 (1993) 59. [68] S. L. Hutchinson, N. K. Banks, and A. P. Schwab, J. Environ. Qual., 30 (2001) 395. [69] E. J. Joner, S. C. Corgie, N. Amellal, and C. Leyval, Soil Biol. Biochem., 34 (2002) 859. [70] M. R. T. Palmroth, J. Pichtel, and J. A. Puhakka, Bioresource Technol, 84 (2002) 221. [71] J. L. Walworth, C. R. Woolard, and K. C. Harris, Cold Regions Sci. Technol., 37 (2003) 81. [72] T. M. Phillips, A. G. Seech, D. Liu, H. Lee, and J. T. Trevors, Environ. Toxicology, 15 (2000) 99. [73] D. Breeveld, and M. Sparrevik, Biodegradation, 11 (2000) 391. [74] D. W. Graham, V. H. Smith, D. L. Cleland, and K. P. Law, Wat., Air, and Soil Poll., I l l (1999)1. [75] L. M. Carmichael, and F. K. Pfaender, Biodegradation, 8 (1997) 1. [76] C.R. Johnson and K.M. Scow, Biodegradation, 10 (1999) 43. [77] R.W. Lee, S.A. Jones, E.L. Kuniansky, G. Harvey, B.S. Lollar, and G.F. Slater, Int. J. Phytoremed., No. 2 (2000) 193. [78] L. E. Erickson, M. K. Banks, L. C. Davis, A. P. Schwab, N. Muralidharan, K. Reilley, and J.C. Tracy, Environ. Progress, No. 13 (1993) 226. [79] J. A. Rentz, B. Chapman, P. J. J. Alvarez, and J. L. Schnoor, Int. J. Phytoremed., No. 5 (2003) 57. [80] J. M. Lynch, The Rhizosphere, New York, Wiley, 1990. [81] D. S. Neuman, M. Wagner, J. H. Braatne, and J. Howe, Stress Physiology - abiotic, In: Biology of Populus and its implications for management and conservation, (eds. R. F. Stettler, H. D. Bradshaw, Jr., P. E. Heilman, and T. M. Hinckley), NRC Research Press, National Research Council of Canada, Ottawa, ON, 1996, pp. 423-458. [82] Ferro, J. Chard, R. Kjelgren, B. Chard, D. Turner, and T. Montague, Int. J. Phytoremed., No. 3 (2001) 105. [83] S. S. Koenigsberg and R. D. Norris, Accelerated Bioremediation Using Slow Release Compounds, Regenesis Bioremediation Products, San Clemente, CA, 1999. [84] E.G. Gatliff, Phytoremediation. Ground Water Monitoring Review, Winter/1996. [85] J. L. Jordahl, M. F. Madison, H. M. E. Smesrud, and M. Q. Motte, (Eds. S. C. McCutcheon and J. L. Schnoor), In: Phytoremediation- Degradation and Control of Contaminants, Wiley fnterscience, New York, NY, 2002. [86] P. A. Domenico and F. W. Schwartz, Physical and Chemical Hydrogeology, NY, John Wiley & Sons, Inc., 1998. [87] G. J. Thoma, T. B. Lam, and D. C. Wolf, Int. J. Phytoremed., 5 (2003) 47. [88] Environmental Protection Agency (EPA), Phytoremediation Handbook for Site Manager, 1998 Draft.
Studies in Surface Science and Catalysis 151 R. Vazquez-Duhalt and R. Quintero-Ramirez (Editors) © 2004 Elsevier B .V. All rights reserved.
479
Chapter 17
Biological treatment of polluted air emissions S. Revaha and R. Auriab a
Department of Process Engineering, Universidad Autonoma MetropolitanaIztapalapa (UAM-I). Apdo. Postal 55-534, 09340 Mexico D.F., Mexico b
Laboratoire IRD de Microbiologie, Universite de Provence, CESB/ESIL, Case 925, 163 Avenue de Luminy 13288, Marseille Cedex 9 France
1. METHODS OF ODOR AND VOC CONTROL 1.1. VOCs and odor definition. Volatile organic and inorganic compounds (VOCs and VICs) are emitted as gases from certain solids or liquids and have an impact on the health and the environment. VOCs are organic compounds having vapor pressure exceeding 0.1 millimeters of mercury (mrnHg) at standard conditions (20 °C and 760 mmHg). Diverse VOCs have shown short- and long-term adverse health effects. In the environment, VOCs have been identified as major contributors to atmospheric photochemical reactions and to smog, which can cause different damages to humans, plant, animal life and to many materials. Some common examples of the volatile inorganic gases (VICs) are hydrogen sulfide (H2S), sulfur dioxide (SO2) and ammonia (NH3). Odor-pollution problems are caused by mixtures of highly volatile compounds with very low threshold detection limit that are generally in small concentrations. Air pollution by the emission of VOCs, VICs and odors are widespread in the petroleum industry including extraction, refining, transport and distribution. The large volume emitted can result in significant health-related and ecological deterioration problems. 1.2. Methods for VOC, VIC and odor control from stationary sources. To select an appropriate control method it is essential to consider the physical, thermodynamic and reaction properties of the pollutant. These properties, the conditions of the stream and the extent of treatment required
480
determine the control method [1]. Figure 1 shows a classification of common technologies applied for VOC, VIC and odor control. Figure 2 was obtained from actual applications and shows that the initial choice can be made based on the stream flow and the pollutant concentration [2, 3]. Further selection of the biological system should include the biodegradability and other technical factors such as temperature, oxygen content of waste gas, stream composition, solubility, operating schedule, presence of particles, production of by-products, utility, and maintenance requirements. The overall selection must include the investment and maintenance costs and the secondary environmental impacts. 2. BIOLOGICAL METHODS 2.1. Introduction The basis of biological air treatment systems (BATS) is the competence of active microorganisms, including bacteria, yeast, and fungi, to transform certain organic and inorganic pollutants into compounds with lower health and environmental impact. These compounds result generally from oxidative reactions and include carbon dioxide, water, nitrate, and sulfate. Since microorganisms are unable to transform the pollutants directly in the gas phase, the first step is the solubilization in the biologically active aqueous phase. Microbes utilize these molecules as a source of nutrients and energy for growth, producing more biomass, which is partly recycled. The performance of the process is determined by the relative rates of the physical, chemical and biological processes. Several books have been published that cover different aspects of BATS [2,3,41.
Fig. 1. Methods for decontamination of emissions from stationary sources.
481
Fig. 2. Comparison of methods for decontamination of air-polluted emissions.
Although the basic-transfer and reaction mechanisms are the same for all the BATS, there are different equipment configurations. These can be grouped, as shown in Table 1, depending on the state of biomass and liquid phase in the reactor. 2.2. Types of reactors The principal reactors described in Table 1 are: 2.2.1. Biofllter (BF). In biofilters (Fig. 3), the polluted air percolates through a moist packed bed, which supports the microorganisms, that grow on its surface and crevices, forming a biofilm. Pollutant transformation rates depends on the microbial density and activity, its bioavailability and the environmental conditions, such as temperature, nutrients, pH and humidity. The biofilm humidity is one of the most critical condition to maintain a proper performance, since biological activity is highly dependent on the water activity (aw). Drying occurs by incoming air with low humidity and by the heat generated by the biological reaction [5]. Increased drying rates are obtained with dry air and high elimination capacity, therefore the air is generally pre-humidified and the support periodically water sprayed. Biofilters have generally a high void fraction to limit pressure drop and to reduce ventilation costs. The supports can be natural and bioactive, inert or a mixture. The natural bioactive supports are soil, peat, compost, bark etc. They are relatively
482
inexpensive and abundant, and have been used for many applications. They can retain water and generally contain an initial microbial population with enough mineral nutrients [6]. Suitable structural characteristics are obtained by mixing with a coarser fraction, including plastics or ceramics, to prevent high pressure drop and to limit bed compaction. The natural supports may degrade with time and lose their structure and water-retaining capacity, inducing channeling and performance loss [7]. In some cases, re-mixing the support with some fresh material and nutrients allows to recover the activity [8], but eventually, it will need to be replaced. With proper maintenance, the support can be used for several years. As mentioned above, biofilters can use inert, natural or synthetic supports such as activated carbon, ceramics, lava rock, polyurethane foam, vermiculite and perlite. On one hand, these supports lack the nutrients required to sustain the microbial activity, therefore it is necessary to intermittently add them. On the other hand, they are not degraded and, in theory, could be engineered to have optimal properties such as controlled head loss, porosity, adsorptive capacity, etc. This remains an area of active research [2, 3, 4]. The high surface and low water content favor degradation of the lesshydrophilic pollutants (Henry's constant, H >10). Empty bed retention time (EBRT) is generally between 30 seconds and 2 minutes. Due to the type of supports used, the height of the packed bed is generally about 0.8 to 1.2 m, making thus necessary to have a large footprint, which may be a disadvantage for situations where space is limited. 2.2.2. Biotricklingfilters (BT). In BT, the polluted air (Fig. 4) flows upflow or downflow through a packed column where liquid is continuously recirculated. The pollutant is first solubilized in the falling liquid film and then transferred to the biofilm developed on the support. The liquid provides moisture, nutrients, pH control to the biofilm and allows the removal of inhibiting products and excess biomass. Table 1. Classification of biological reactors Biomass Fixed on a support Fixed on a support
Liquid phase Stationary Flowing
Suspended Suspended or fixed Fixed on a membrane
Flowing Stationary Flowing
Reactor Biofilter, BF Biotrickling filter, BT Rotating contactors, RC Bioscrubber, BS Suspended growth, SR Membrane, MR
483
Fig. 3. Schematic representation of bio filter (BF).
Inert random supports or structured packing are used. Some examples include plastic corrugated structured PVC sheets, Raschig or Pall rings and saddles, lava rock and polyurethane foam [2]. To maintain low pressure drop and reduce clogging, the supports have low porosity and low specific surface (100 - 400 m2 m"3). EBRT are normally around 30 seconds but systems with EBRT as low as 2 seconds have been reported for low H2S concentration [9].
Fig. 4. Schematic representation of biotrickling filter (BT)
484
2.2.3. Bioscrubbers (BS). In bioscrubbers, the pollutant in the gas phase is first absorbed in a gasliquid contactor (Fig. 5). Subsequently, it is eliminated in a bioreactor and the liquid, containing the suspended microorganisms, is returned to the contactor. Nutrients and pH regulators can be added to maintain microbial activity and the excess of biomass and sub products can be controlled by purging. The gas-liquid contactors can be packed towers, venturi scrubbers or spray towers [10]. Bioscrubbers are designed to favor mass transfer with low pressure drop (< 3 cm H2O m"1). In the bioreactor, supplementary air is added to favor the oxidation of the pollutant. Water retention time in the reactor is calculated to eliminate the soluble pollutant, and the biomass concentration is generally about 5-8 g L"1 [11] to foster high volumetric rates while reducing clogging problems in the contactor. Bioscrubbers are used for hydrophilic pollutants (H<1) to avoid big absorbers or large water flows [12]. 2.2.4. Other configurations. Membrane bioreactors (MB). This is an emerging technology derived from the development of new porous materials [4, 12, 13]. The pollutant in the gas phase is transferred through a membrane (hollow fibers or flat sheets) to the biofilm, which is attached to the other side where oxygen and aqueous nutrients are fed. In hollow fibers, the gas is usually passed through the lumen of the tube and the biomass is on the shell side. These reactors have been used for other waste treatment applications where the conditions of the stream excluded the possibility of direct contact with the biomass. A distinct characteristic of the Membrane bioreactors is the physical separation of the polluted gaseous stream from the biomass, which allows the use of BATS in certain applications such as indoor air or in extreme case, for spaceship air treatment.
Fig. 5. Schematic representation of bioscrubber (BS).
485
Suspended cell bioreactor. In these reactors, the polluted air is bubbled directly in the bulk of the liquid containing suspended microorganisms. The most common reactor [14] is an activated-sludge aerator where the sparged polluted air is treated simultaneously with the municipal wastewater. The reactor parameters, such as biomass concentration, air feed and sparger design, are generally imposed by the requirements of the wastewater treatment. A review of the characteristics of several facilities is given in Ref. 15. In other cases, the reactors are designed to optimize mass transfer from the bubble to the bulk liquid where biodegradation occurs and to control the conditions that promote high microbial rates. These reactors include air- lift, external loop, split cylinder, etc. Rotating Biological contactors (RC). In these systems, initially developed to treat water, the polluted air flows through the headspace of a reactor containing discs that serve as support for a biofilm and are assembled on a rotating shaft. The shaft is slowly rotated (around 2 rpm) and the discs are partially wetted in water containing nutrients and other additives. Air can be fed tangentially to the disks or through perforations in a hollow shaft [16]. Intermittent wetting of the biomass favors mass transfer and biological activity. 2.3. Mechanism BATS involve complex physical, chemical and biological interactions, [2, 3,4] which will be shortly reviewed: 2.3.1. Gas-liquid phase equilibrium of the pollutant As the degradation of pollutants cannot occur directly in the gaseous phase, they have to be first transferred to the biofilm or to the liquid with the suspended cells. Gas-liquid resistance is often negligible and consequently interfacial concentrations can be considered in equilibrium. Moreover, because the concentration of pollutants involved in biological treatment is low, Henry's law is generally used to describe this equilibrium. Henry's law partition coefficient can be described as H = C g /C,
(1)
Where H: dimensionless Henry's constant Cg: gas phase concentration (g L"1) Q: liquid phase concentration (g L"1) In Table 2, some Henry's coefficients are reported for different pollutants. For high values of Henry's coefficient (H), the pollutant is slightly soluble in water. For example, with hexane concentration in air of 1 g m"3, corresponding
486
to those usually treated in biological processes, its concentration in water is lower than 30 ug L"1. Henry's coefficient increases with an higher salinity and temperature. Data is usually found for H in pure water but it may be different when biomass is present. A strong decrease of H value has been reported for toluene, benzene, and trichloroethylene when biomass is present in the aqueous mixture. In biofiltration, biomass, extra-cellular polymers (EPS) and support promote the solubilization of hydrophobic compounds [17]. A very important aspect of BATS performance is the solubilization of oxygen required for the oxidation of the pollutant. Oxygen limitation can be found when the pollutants are in high concentration or have very low Henry's coefficient. Gas- liquid phase equilibrium can be altered by reaction in the liquid. For example, soluble H2S dissociation depends on pH according to the following equilibrium reactions. At pH =10, sulfides are present mainly as HS" which is very soluble and consequently the apparent Henry's coefficient is 3 orders of magnitude lower than at pH =4.
2.3.2. Pollutant diffusion in the biofilm Once the pollutant and the oxygen are solubilized in the liquid, diffusion in the biolayer occurs under the influence of concentration gradients. Diffusion is described using Fick's law: Table 2 Henry's coefficient for some common compounds at 25 ° C (adapted from Ref. 18). Compound
Henry's coefficient (non-dimensional) Methane 41.3 U) Hexane 30.9 Oxygen 29.1 Hydrogen sulfide 0.92 Toluene 0.25 0.22 Benzene MTBE (methyl t- butyl ether) [19] 0.026 Ethanol 0.0012 Ammonia 0.0005 (1) http://www.epa.gov/regionl/measure/Natatten.pdf
487
J=-D^l
(2) ax
where: J: mass flux (g m"2 s"1) D :diffusion coefficient (m2 s"1) Cl: liquid concentration (g m"3) x: distance in the biolayer (m) In biofilms, diffusion coefficient may be strongly reduced by the presence of biomass and EPS. Most of the effective diffusion coefficients in biofilms are evaluated by either empirical or semi-empirical correlations. In terms of concentration gradient, two typical cases of biofiltration are found: 1) No diffusion limitation exists and the biofilm is fully active. In this case, the biological reaction is the limiting step. 2) Diffusion limitation occurs and the biofilm is not fully active, this condition is referred as diffusion limited. In this situation, a reaction-free zone is predicted. 2.3.3. Biodegradation of the pollutant in the biofilm In most of BATS, pollutants are degraded by microorganisms structured in biofilms mainly composed of microbial cells and EPS, which promote adhesion. EPS may account up to 90 % of the total. While the biofilm is often represented as an homogeneous mass with constant thickness, it has, in reality, a complex structure comprising diverse biological (presence of bacteria, yeasts and fungi, EPS,...) and physical (cavities, detachment of the biofilm, microbesupport interactions, less-dense zones, etc.) components. Biofilm structure and ecology are poorly known. In BATS, organic pollutants serve as carbon and energy sources, and the dissolved oxygen as electron acceptor. Degradation occurs not only during growth phase, but also when the net growth rate is zero. The energy released during degradation is used for growth and maintenance metabolism. The pollutant uptake rate can be defined as: (3) where: (a.: net cell growth (h"1) X: biomass concentration (gxL"') Yxs: biomass yield from the pollutant (gx g poiiutant"1) m: maintenance coefficient (g poiiutant gx' h"1)
488
Specific growth rate (a) depends on the concentration of the limiting substrate: (4) where: |imax: maximum cell growth (h"1) Ks: half saturation constant (g L"1) In BATS, mixed populations are generally found and their behavior can be described by a net growth rate as: *$=Mx
(5)
BATS are open biological systems as the air is never filtered. The large variety of microorganisms (fungi, yeast, and bacteria) come from the initial inoculum, from the biofilter packing material (compost, peat, etc..) and from the incoming air [2, 3, 4, 12]. The microorganisms present in the biofilters are similar to those found in the natural ecosystems or other biological processes such as compost or water treatment plants. Although bacteria are dominant in biofilters, fungi are frequently observed. Recent studies showed that some fungi [20, 21, 22, 23] can degrade toluene and hexane vapors at higher rates than bacteria. For a successful biological air treatment, the pollutants have to be biodegradable. In general, organic compounds with low molecular weight, highly soluble in water and simple-bond structures are the best candidates. Alcohols, aldehydes, ketones, and some simple aromatics have very good biodegradability, while phenols, aliphatic and chlorinated hydrocarbons show moderate to slow degradation (Table 4). Ethers, like diethyl and dimethyl ethers are generally easily degraded while MTBE (Methyl tert butyl ether) is reported to be very recalcitrant. Table 4 Biodegradability of pollutants [Ref. 24]. Rapidly
Alcohols**, aldehydes*, ketones**, esters**, ethers*, organic acids*, terpenes, amines*, thiols*, sulfides, ammonia Slowly Hydrocarbons**, phenols* Very slowly Halogenated hydrocarbons*T, polyaromatic hydrocarbons, *Branched molecules are less biodegradable | Removal in biofilters follows alcohols > esters > ketones > aromatics > alkanes t Biodegradability decreases with higher number of halogens
489
2.3.4. Performance parameters Three parameters are often used to compare the pollutant treatment efficiency in BATS. Inlet mass pollutant load (IMPL), is defined as the amount of pollutant introduced into the bioreactor normalized by its empty-bed volume. (6) Elimination capacity (EC), is the quantity of pollutant removed per bioreactor volume per time unit. (7) Removal efficiency (RE), is the fraction of the pollutant removed expressed as percentage. (8)
3. EXAMPLES OF TREATMENT OF VOLATILE PETROLEUM HYDROCARBONS BY BATS Petroleum products such as gasoline, fuel oils, and diesel fuels are among the most important water, soil and air pollutants. They are complex mixtures of organic compounds containing a significant volatile fraction. Hydrocarbons are composed of four main structural classes: 1) «-alkanes (linear saturated hydrocarbons), 2) isoalkanes (branched saturated hydrocarbons), 3) cycloalkanes (saturated cyclic alkanes) and 4) aromatics [25]. Moreover, in the case of gasoline, other oxygenated additives such as the ethers MTBE, ethyl tbutyl ether (ETDE) or /-amyl methyl ether (TAME) or alcohols such as ethanol, can be added to improve combustion and consequently air quality. Hydrocarbons may be released into the atmosphere by evaporation during production, transport and storage. Table 5 presents some examples of the degradation of volatile organic compounds from gasoline by BATS. The removal efficiency of volatile individual compound in the gasoline varies from 5 % to 99 %. Aromatics are generally removed more efficiently than n-alkanes. Light aliphatic compounds (
490
alkanes. Recent studies showed that biofilter performance was improved using filamentous fungi [22]. An elimination capacity of toluene of 258 g m"3 h"1 (RE = 98 %) was attained. This value was higher than those obtained for bacterial biofilters [27-28]. It is hypothesized that the aerial fungal mycelia, which are in direct contact with the gas, give a larger superficial area and allow a direct mass transfer of the volatile compound. Poor degradation of hydrophobic n-alkanes in biofilter could strongly be increased using filamentous fungi as shown in Ref. [23]. Very few reports have been published addressing MTBE biofiltration. Except for the results obtained by Fortin et al, 1999 [29], elimination capacities of MTBE are generally lower than 10 g m"3 h"1. MTBE is a highly soluble compound in water but the ether bond and the ter/-butyl moiety have been shown to limit MTBE biodegradability. Cometabolism has proved to be a good way to increase biomass production and MTBE degradation. Microbial consortia [30], can degrade MTBE by cometabolism with n-alkanes (hexane, pentane and heptane) present in gasoline. Cometabolism with pentane using a single bacteria (Pseudomonas aeruginosa) can be used to degrade MTBE in a biofilter packed with vermiculite [31]. Biofilters used to treat gasoline from a soil vapor extraction operation showed that higher molecular weight compounds were almost completely removed while lower molecular weight were less degraded [32]. The predominant compounds remaining in the outlet of the biofilter were tentatively identified as methyl-substituted alkanes and cycloalkanes in the C6 to C9 range. A pilot-scale biofilter system treating gasoline vapors presented total hydrocarbon-elimination capacities of 16 g m~3 h"1 [33]. Linear alkanes and aromatics were rapidly degraded, while branched alkanes had lower removal efficiencies. Pilot-scale biofilter elimination capacities for hexane, isooctane and toluene were 3.2 ghexane m"3 h"1, 3.1 g,so-octanc m"3 h"1, and 1.5 gtoiucne m"3 h"1, respectively. Removal efficiencies for toluene were the highest and the most stable. 4. CONCLUSIONS BATS are among the established technologies that can be applied to control VOC and odor emissions. For their choice, the characteristics of the stream (flow, temperature, presence of particles, humidity, etc.), the pollutant (composition, concentration, reactivity, solubility and biodegradability) and the required performance have to be considered. BATS are applicable for a wide range of volatile pollutants found in the petroleum industry and their applications are growing continually based on scientific and technological developments.
491
Table 5 Examples of treatment of volatile petroleum hydrocarbons by BATS Compounds
BATS
Packing
Cgi,,(gm-3) EBRT (min)
EC
Ref.
(grnV) Efficiency
Gasoline
BT
Gasoline
BT
Ethers MTBE
Compost /pine bark 70:30 (v/v) Compost/perlite
BF
Pall rings
MTBE a
BT
Vermiculite
MTBE
BT
Celite ™ R-635 b
MTBEC
BT
Compost /pine bark 70:30 (v/v)
Aliphatics Isopentane
BT
Peat moss
Pentane
BT
Vermiculite
Hexane
BT
Hexane
BT
Compost/perlite 50:50 (v/v) Granular expanded clay
Hexane
BF
Cyclohexane
BT
Polyamide structure wire mat Compost
BF
Fibrous sheet (cotton)
Aromatics Benzene Xylenes
BT
Peat
Toluene
BT
Toluene
BT
Vermicul ite/Granular activated carbon 85:15 (v/v) Peat
Toluene
BT
Peat
a) MTBE degraded in cometabolism with pentane b) Extruded diatomaceous earth pellet c) MTBE as a gasoline additive d) Methane equivalent gHC m"3 h"'
1000 ppmv 1 140-490 ppmv 2.3
16d 45
0.8 0.9 5 66 0.13 (35ppmv) 1 0.21 1
50 90 0.8 18 7.8 99 3.8 30
5 (1700 ppmv) 3 17.4 35 0.7 (200ppmv) 3 12 26 10 6.3 0.004 (1.2 ppmv) 2
50 40 12 40 21 99 150 >80 80 89 — 9
[33] [32]
89
0.5
3
Q O
5U
8.2 1.7 5 1.2 1.5 1 6.2 1.3
66 23 258 98 25 31 165 70
[29] [31] [38] [33]
[34] [31] [35] [23] [39] [36]
[37] [28] [22] [27] [28]
492
REFERENCES [I]
H. Rafson (ed.), Odor and VOC Control Handbook. McGraw Hill Professional. USA 1998.
[2]
J.S. Devinny, M. Deshusses and T. S. Webster, Biofiltration for air pollution control. CRC Press, Boca Raton, Fl. USA, 1999.
[3]
C. Kennes and M. C. Veiga MC (eds.), Bioreactors for waste gas treatment. Kluwer Academic Publishers, The Netherlands, 2001.
[4]
Z. Shareefdeen and A. Singh (eds.) Biotechnology for Odour and Air Pollution, Springer-Verlag, Germany (in press).
[5]
M. Morales, S. Hernandez, T. Cornabe, S. Revah and R. Auria Environ. Sci. Technol. 37 (2003) 985.
[6]
B. Cardenas-Gonzalez, S. Ergas, M. Switzenbaum and N. Phillibert, Environ. Progress 18(1999)205.
[7]
J. M. Morgan-Sagastume, S. Ergas, S. Revah and A. Noyola, J. Air Waste Manage. Assoc. 53(2003)1011.
[8]
R. Auria, G. Frere, M. Morales, M. E. Acuna and S. Revah S, Biotechnol Bioeng 68 (2000) 448.
[9]
D. Gabriel and M. Deshusses, PNAS 100 (2003) 6308.
[10] J. W. Van Groenestijn, In C. Kennes and M. C. Veiga MC (eds.), Bioreactors for waste gas treatment. Kluwer Academic Publishers, The Netherlands, (2001) 133. [II] S. P. P. Ottengraf In: Biotechnology 8, Rehm HJ and Reed G (eds), VCH Verlagsgesellschaft Weinheim, Germany, (1986) 426. [12] J. W. Van Groenestijn and P. G. Hesselink, Biodegradation 4 (1993) 283. [13] S. Ergas, In C. Kennes and M. C. Veiga MC (eds.), Bioreactors for waste gas treatment. Kluwer Academic Publishers, The Netherlands, (2001) 163. [14] A. Bielefeldt, In C. Kennes and M. C. Veiga MC (eds.), Bioreactors for waste gas treatment. Kluwer Academic Publishers, The Netherlands, (2001) 215. [15] R. P. Bowker, In H. Rafson (ed.), Odor and VOC Control Handbook. McGraw Hill Professional. USA (1998)8.192. [16] R. von Rohr and P. Ruediger In C. Kennes and M. C. Veiga MC (eds.), Bioreactors for waste gas treatment. Kluwer Academic Publishers, The Netherlands, (2001) 201. [17] B. Davison, J. Barton, T Klasson and, A. Francisco Biotechnol. Bioeng. 68, (2000) 279. [18] T. Card, In H. Rafson (ed.), Odor and VOC Control Handbook. McGraw Hill Professional. USA (1998)2.1. [19] A. Fischer, M. Muller and J. Klasmeier, Chemosphere 54 (2004) 689. [20] H. H. J. Cox, R. E. Moerman, S. Vanbaalen, W. N. Vanheiningen, H. J. Doddema and W. Harder, Biotechnol. Bioeng. 53 (1997) 259.
493
[21] J. Woertz, K. A. Kinney, P. Mclntosh and P. J. Szaniszlo, Biotechnol. Bioeng. 75 (2001) 550. [22] E. I. Garcia-Pena, S. Hernandez, E. Favela-Torres, R. Auria and S. Revah, 76 (2001) 61. [23] G. Spigno, C. Pagella, M. Daria Fumi, R. Molteni, and M. De Favieri, Chem. Eng. Sci. 58 (2003) 739. [24] S. Revah and J. M. Morgan Sagastume In Z. Shareefdeen and A. Singh (eds.) Biotechnology for Odour and Air Pollution, Springer-Verlag, Germany (in press). [25] R. Marchal, S. Penet, F. Solano-Serena and J. P. Vandecasteele, Oil Gas Rev. IFP. 58 (2003)441. [26] G. Leson and B. J. Smith, J. Environ. Eng. 123, (1997) 556. [27] M. Morales, S. Revah and R. Auria, Biotechnol. Bioeng. 60 (1998) 483. [28] H. Jorio, K. Kiared, R. Brzezinski, A. Leroux, G. Viel and M. Heitz, J. Chem. Technol. Biotechnol. 73(1998) 183. [29] N. Fortin and M. Deshusses, Environ. Sci. Technol. 33 (1999) 2980. [30] P. M. Gamier, R. Auria, C. Augur and S. Revah, J. Gen. Appl. Microbiol. 46 (2000) 79. [31] D. Dupasquier, S. Revah and R. Auria,. Environ. Sci. Technol. 36 (2002) 247. [32] W. F. Wright, Y. Davidova, E. D. Schroeder and D. P. Y. Chang,. Proceedings of Conference on Biofiltration. USC Los Angeles. (1995) 18. [33] A. Hernandez, M. Magana, B. Cardenas, S. Hernandez, S. Revah, S. Queney and R. Auria, Proc. 94th Annual AWMA Meeting Exhibition. (2001) Paper # 1037 AE-2a. [34] A. P. Togna and M. Singh, M Proc. 87th Annual AWMA Meeting Exhibit, (1994). [35] E. Morgenroth, E. D. Schroeder, D. P. Y. Chang and K. M. Scow, J. Air Waste Manage. Assoc. 46(1996)300. [36] D. X. Li, Proc. of Conference on Biofiltration. USC Los Angeles. (1995) 1. [37] Q. Zhou,, Y. L. Huang, D. H. Tseng, H. Shim and S. T. Yang S-T J. Chem. Technol. Biotechnol. 73(1998)359. [38] J. B. Eweis, E. D. Schroeder ED, D. P. Y. Chang and K. M. Scow, in G. Wickramanayake and R. Hinchee (eds) Remediation of Chlorinated and Recalcitrant Compounds, Physical chemical and thermal technologies, Battelle Press, Columbus Ohio USA (1998) 341. [39] J. W. Van Groesnetijn and M. E. Lake, hi: Arendt F, Hinsenveld M., Brink WJ van den (eds). Kluwer, Dordrecht. (1999) 151-155.
This page is intentionally left blank
Studies in Surface Science and Catalysis 151 R. Vazquez-Duhalt and R. Quintero-Ramirez (Editors) ©2004 Published by ElsevierB.V.
495
Chapter 18
Bioremediation of marine oil spills R. C. Prince1 and J. R. Clark2 'ExxonMobil Research & Engineering Co. Annandale, NJ 08801 2
ExxonMobil Research & Engineering Co. Fairfax, VA 22037
1. INTRODUCTION Crude oils are the liquid fossil residues of aquatic algae, sometimes with minor contributions from terrestrial plants, that grew in the distant past. Then as now, we can imagine that most of this material was biodegraded and recycled on an essentially annual timescale, but a small fraction became buried and underwent diagenesis and catagenesis to become oil [1]. This process usually took millions of years, and was dependent on the depth of burial and the temperature. Some oil dates from the Precambrian (>570 million years ago), but most is rather younger; the average age of commercially important oil is about 100 million years, with the majority being from the Jurassic and Cretaceous (180 to 85 million years ago) [2]. Commercially important oil has migrated from its source rock to a reservoir, and it is not unusual for these reservoirs to leak. If the leak reaches the surface, it is known as an oil seep. Humans have used material from such seeps for thousands of years. Early uses include hafting stone axes to their handles, as an embalming agent, and as a medical nostrum. Genesis (11,3) says that bitumen was used as the mortar for the Tower of Babel, and Exodus (2,3) that Moses' basket was made waterproof with a bitumen daub. It seems likely that several religions started around natural gas seeps, either as eternal flames or as sources of hallucinogenic vapors [3]. But these were only very minor uses, and it is only in the last century and a half that oil has come to play a truly central role in modern society [4]. Terrestrial seeps were the first locations to be drilled when oil production began in earnest in the nineteenth century, such as the 1860 Drake well in Pennsylvania, but marine seeps were drilled by the end of the century.
496
Marine seeps are widespread, as shown in Figure 1, and are a major source of oil into the World's oceans [5]. Even with today's enormous commerce in oil, seeps provide about 62% of the total releases into the coastal marine environment of North America, and 47% of the world. Such seeps must have been occurring for millions of years, providing an important input of degradable carbon for local ecosystems and perhaps even major fisheries [6]. A diverse group of microorganisms exploits this natural input. Oil-degrading microbes have been found in all marine environments where they have been looked for, and more than seventy genera of eubacteria and archaea, and a hundred genera of fungi, have been shown capable of degrading petroleum hydrocarbons [7]. It is these organisms that remove oil seepage and spilled oil from the marine and terrestrial environment, and underpin the bioremediation strategies for dealing with spilled oil that we will describe here. 2. ANTHROPOGENIC INPUT OF OIL INTO THE WORLD'S OCEANS Oil fuels the modern world on an enormous scale; annual consumption is of the order of 3.5 billion US gallons (1.2 x 1010 liters) per day [8]. Much of this is produced at sea; more than 25% of US production is from the Continental Shelf, and 25% of Saudi Arabia's production, 80% of Nigeria's, and 100% of Angola's, Australia's, Brazil's and Malaysia's production is offshore. This production is associated with some oil and grease discharges into the marine environment associated with the produced water; some 2,700 tonnes per year in the US, and more than ten times this in the rest of the world [5]. This input is dwarfed, however, by oil in municipal run-off from the land, and from the standard operation of marine shipping (Figure 2). Catastrophic spills from tankers and other ships are well known, but in fact their contribution to total oil input into the oceans is relatively small, some 8% of the global input, 2% of North American input. Nevertheless, since such spills are large on the local scale, they often require environmentally appropriate and cost-effective responses. Fortunately, despite the increasing volume of transported oil, the amount spilled from catastrophic spills has been generally decreasing over the last few decades [9]. The major exception to this decline was the appalling environmental crime in the Arabian Gulf in 1991, where Iraqi forces deliberately released more than a million tonnes (about 260 million gallons or a billion liters) of oil into the sea near Kuwait [10]. An additional 350 million gallons (1.2 xlO9 1) were deposited in the Gulf as fallout from the smoke plumes of the > 700 oil well fires in the Kuwait oil fields [11], making this by far the largest man-made release of oil into the marine environment to date.
497
3. COMPOSITION OF CRUDE OIL Crude oils are complex mixtures of hydrocarbons with significant quantities (typically about 15%) of compounds containing additional heteroatoms such as oxygen, sulfur or nitrogen. The hydrogen to carbon ratio of the hydrocarbons is typically between 1.5 and 2, indicating a mixture of aliphatic (predominant carbons are -CH 2 -, either as linear molecules, known as paraffins, or as rings known as naphthenes) and aromatic species (principal carbons are -HC=CH- in rings). Alkenes and alkynes, linear unsaturated molecules, are rare in crude oils, although they can be abundant in some refined products such as gasoline. Tissot and Welte [2] calculate the average composition of more than 525 crude oils as 58.2% saturates, 28.6% aromatics and 14.2% polar compounds, noting that the absolute values vary widely in different oils. On average, there is rough parity between paraffins, naphthenes and aromatic hydrocarbons in crude oils [2]. Paraffins in crude oils may start with methane, and extend to waxes with more than seventy carbons. Their total content varies widely, from essentially undetectable to as high as 35%, depending on source and reservoir conditions, but they typically make up 15-20% of an undegraded crude oil. There are also branched alkanes; especially in the C6 to C8 range, but pristane (C19H40) and phytane (C20H42), molecular relics of the phytol chains of chlorophylls and perhaps other biomolecules, are usually the most abundant individual branched alkanes. Pristane is thought to be the result of initial partial degradation of phytol in the presence of oxygen, while phytane is thought to be the result of diagenesis in the absence of oxygen [2].
Fig. 1. Map of the major oil seeps into the World's oceans. Data taken from reference [5].
498
Fig. 2. Oil input into the World's oceans. Data taken from reference [5].
The naphthenes include parent compounds, such as cyclopentane, cyclohexane and decalin, together with their alkylated congeners. Tissot and Welte [2] quote the average composition of the naphthene fraction of 299 crude oils as 54.9% one and two ring naphthenes, 20.4% tricyclic naphthenes, and 24.0% tetra and pentacyclic naphthenes. These latter molecules are amongst the better understood molecular biomarkers in crude oils, and they are used extensively in correlating reservoirs and source rocks [12], in assigning the depositional environment of source rocks [12], and more recently as conserved internal markers during biodegradation [13]. Because of the separation procedures used in the analysis of crude oils, any molecule containing at least one aromatic ring is included in the "aromatic" fraction, regardless of the presence of saturated rings and/or alkyl substituents. Aromatic heterocycles containing sulfur, such as thiophenes, benzothiophenes and dibenzothiophenes, or nitrogen, such as the indoles, carbazoles and quinolines, also fall into the aromatic category. Alkylated aromatic species are more abundant than their parent compounds, with mono-, di- and tri-methyl derivatives usually being most abundant. Nevertheless, the median aromatic structure probably has one or two methyl groups and a long-chain alkyl substituent [14]. The polar molecules are the most difficult to characterize because they typically cannot be analyzed by gas chromatography, the method of choice for the molecular characterization of hydrocarbons. Petroleum polar compounds contain heteroatoms such as nitrogen, oxygen and/or sulfur, and the category includes the porphyrins, typically with nickel or vanadium in the tetrapyrole,
499
naphthenic acids and large molecules known as asphaltenes. Some asphaltenes have molecular weights into the thousands and even higher, and many are suspended in the oil rather than dissolved in it [15]. The polar fraction of the oil contains the majority of the color centers in crude oil, and in isolation these materials are difficult to distinguish from more recent biological residues, such as the humic and fulvic acids [16], except with sophisticated tools such as isotope analysis. Recent developments in electrospray mass spectrometry promise significant progress in determining the molecular structure of these compounds [17, 18]. Several hundred different crude oils are being produced today, and their chemical composition and properties vary quite widely. They are typically classified by their density. The oil industry uses a unit known as API (American Petroleum Institute) gravity, which is defined as [142.5/(specific gravity)] 131.5, and expressed as degrees (°). Thus water has an API gravity of 10°, and denser fluids will have lower API gravities. For convenience, oils with API gravities greater than 40° are said to be light oils, while those with API gravities of less than about 17° are said to be heavy. Note that even these typically float on water, especially seawater. Light oils have higher proportions of hydrocarbons; heavy oils are rich in polars and asphaltenes. Viscosity is roughly inversely proportional to API gravity, but it is also dependent on the physical state of the polar compounds and longer alkanes in the oil, and is highly dependent on the temperature. Among petroleum products in commerce, crude oil is transported in the largest volumes, both in undersea pipelines and in tankers. Refined products are also shipped, and of course all ships contain large amounts of fuel for their own propulsion. Refining crude oils for commercial applications starts with distillation, and the simplest distinction of the various refined products can be related to this process. The most volatile liquid product is aviation gasoline, followed by automobile gasoline, jet fuels, diesel and heating oils, and then the heavy oils that are used for fueling ships and some electrical generation. Most of the molecules in gasoline have between four and ten carbons, most in diesel have between nine and twenty, and heavy fuel oils typically have very few molecules with less than fifteen carbon atoms except for some added as a diluent to achieve the appropriate viscosity to facilitate distribution. As an aside it is appropriate to note that fuels are valued based on their combustion properties, and not chemical composition. Fuels with the same name may have very different chemical compositions if they come from different refineries [19]. 4. PHYSICAL FATE OF SPILLED OIL When oil gets into the oceans from a seep, urban runoff or a spill from a production facility, pipeline or a tanker, it becomes subject to a group of phenomena that are usually grouped under the term "weathering" [20]. Almost all oils float, allowing the smallest molecules to evaporate [21, 22]. These
500
molecules are either degraded photochemically [23] or are washed from the atmosphere in rain and then biodegraded [24], likely far from the spill site. Under particularly aggressive aeration in water, such as in surf, evaporation may extend to molecules with >30 carbon atoms [25], but evaporation is more usually limited to molecules with less than about 15 carbon atoms [21]. Thus evaporation is the likely fate of most of a gasoline spill, three-quarters or more of a diesel spill, and perhaps 20-40% of a typical crude oil. Heavy fuels, such as the Bunker fuels used in ships, and bitumen emulsions (Orimulsion®, [26]) do not contain a significant volatile fraction. Two competing emulsification processes occur as water and oil mix; water can become entrained in the oil to form an emulsion known as mousse [27], or oil can disperse into the water column as a suspension of small droplets, as happened during the 1993 Braer spill off the Shetland Islands [28]. Mousses are remarkably persistent, and are thought to be the precursors of tarballs that can last for decades [29]. As we shall discuss below, chemical dispersants that break emulsions and stimulate the natural dispersion process are effective tools in the oil spill response "toolkit". Oil also interacts with small mineral particles in a process originally termed "Clay-oil flocculation" [30], and now termed "Oil-Mineral Fines Interactions" [31]. Like dispersion, this dramatically increases the surface area of the oil. Aliphatic hydrocarbons are remarkably insoluble, but small aromatics, particularly the notorious BTEX (benzene, toluene, ethylbenzene and the xylenes) and small polar molecules such as naphthenic acids dissolve from floating slicks or dispersed oil, and even from oils immobilized on shoreline sediments and particles [32]. Again, their eventual fate is biodegradation. Aromatic hydrocarbons can be photochemically oxidized [33], converting them to polar products that are probably polymerized species. The process is most effective on the larger and more alkylated forms, and although such hydrocarbons are only a minor component of crude oils [2], they have important toxicological properties [34], and are on the USEPA list of priority pollutants [35] and the EU list of priority substances in the field of water policy [36]. Since light cannot penetrate very far into a dark oil slick, photooxidation has little effect on the bulk properties of spilled oil. Nevertheless it may be important in generating a polymerized "skin" that enhances the stability of tarballs and "pavements" on beaches. Layers of immobile, hardened oil and sediment, termed pavements, form when oil reaches a shoreline as a heavy, thick slick. Oil becomes trapped in the sediment, and the oil and the sediment become saturated with each other [37]. Oil incorporated into such pavements is effectively preserved from weathering processes until this heavy, solidified material is physically disrupted, so a major goal of spill clean-up operations is to prevent the formation of pavements.
501
5. EVENTUAL FATE OF SPILLED OIL The weathering processes described above distribute and change the oil in various ways, but they do not actually remove oil from the environment. Only two processes, combustion and biodegradation, actually eliminate oil by converting it to carbon dioxide and water. Some spills do spontaneously ignite, as happened to the Haven spill in the Mediterranean [38], and deliberate ignition is an accepted response option in some situations, such as that of the New Carissa off the coast of Oregon [39]. Under optimal conditions burning may consume >90% of a spill, but there is usually only a small window of opportunity for success [40]. Far more generally, it is biodegradation that removes oil from the environment. As mentioned above, a diverse group of microorganisms has evolved to degrade hydrocarbons, many able to live with hydrocarbons as their sole source of carbon. They are probably ubiquitous, having been found in almost all natural environments where they have been searched for, and obviously very effective, since they have been consuming the vast majority of the oil entering the world's oceans from natural seeps for millions of years (600,000 tonnes, >600 million liters, per year). What separates these organisms from other heterotrophs is their ability to transform hydrocarbons into organic alcohols and acids that enter cellular metabolism. Under aerobic conditions the most common microbial activation of hydrocarbons involves the addition of one or both atoms of molecular oxygen. Alternatively, the activation may involve the addition of hydrogen peroxide. The activation of aromatic hydrocarbons is discussed by Foght in chapter 5, and many pathways are available in the University of Minnesota Biocatalysis/Biodegradation Database [41] and in a recent encyclopedia article [42]. Here it suffices to say that the vast majority of hydrocarbons are biodegradable under aerobic conditions. Thus refined products, such as gasoline, diesels and jet fuels, that are almost entirely hydrocarbons, are essentially completely biodegradable. McMillen et al. [43] examined the short-term biodegradability of 17 crude oils in soil microcosms, and found that the API gravity was the most useful predictor of biodegradability. At 0.5 wt% oil in soil with appropriate nutrients, moisture and aeration, more than 61 % of the most degradable oil (API = 46 ) was lost in four weeks, while only 10% of the least degradable oil (API =15 ) was consumed under the same conditions. Further degradation occurred on a longer timescale, and the literature reports biodegradation potentials as high as 97% for particularly light oils [44]. An important distinction between hydrocarbon-degrading microorganisms and animals and plants is that many microbes degrade polycyclic aromatic hydrocarbons to carbon dioxide, water and biomass. Animals and plants can also activate these molecules, but they do so with enzyme systems that form stable
502
permutations of the polycyclic hydrocarbons (see chapter 3). The enzyme systems are known as Cytochrome P450s because of their prominent absorption band when treated with carbon monoxide [45]. These enzymes generate epoxides that are excreted as adducts with sugars, anions, etc., but which may alternatively intercalate and form adducts with DNA [46]. It is thus clearly preferable that polycyclic aromatic hydrocarbons be degraded by bacteria rather than eukaryotes, and facilitating such a preference is one of the advantages of a successful bioremediation protocol. Aerobic biodegradation of hydrocarbons occurs over a wide range of environmental conditions [47]. Although no hyperthermophilic oil-degraders have yet been found, extreme thermophiles such as Thermus and Bacillus species degrade polycyclic aromatic hydrocarbons and long chain alkanes at 6070 °C [48]. Significant biodegradation occurs below 0°C [49] and extremely halophilic oil-degrading organisms have been described [50] that degrade hydrocarbons in the presence of several molar salt. In the last decade it has become clear that hydrocarbons are also degraded under anoxic anaerobic conditions. Small water-soluble aromatic compounds, such as benzene and toluene, have been shown to undergo biodegradation under sulfate-reducing, nitrate-reducing, perchlorate-reducing, ferric ion reducing, humic acid-reducing and methanogenic conditions [51], and this phenomenon is proving important in remediating terrestrial spills where these compounds have reached groundwater [52]. Larger hydrocarbons, such as n-alkanes up to «C34H7o [53] and two- three- and four-ring aromatic hydrocarbons [54] are also biodegraded under anaerobic conditions. This may be important if oil spills contaminate anaerobic environments, such as marshes, and in the degradation of the traces of oil that become entrained in sediments in harbors. Wherever oil is biodegraded, it is important to bear in mind the fact that crude oil and refined products provide a rather unusual "food" for heterotrophs. While hydrocarbons are excellent sources of carbon and energy, they do not provide any of the other nutrients essential for life; there are no significant amounts of biologically available nitrogen, phosphorus or other elements. Of course most environments have at least trace amounts of these essential nutrients, but most marine environments offer meager reserves to sustain new growth. It is thus likely that any significant input of hydrocarbon is likely to overwhelm the background levels of nutrients, and their availability soon limits that biodegradation. As we shall see below, alleviating this limitation forms the basis of the simplest forms of shoreline bioremediation.
503
6. SPILL RESPONSE 6.1. At sea: When oil is spilled at sea, deployment of mechanical equipment designed for containment and recovery is often a slow and inefficient, if not ineffective, response. The rapid spreading of the oil, the slow rate at which mechanical equipment (once deployed) can encounter spreading oil, and interference from waves and currents often limits recovery effectiveness to less than 20% of the oil spilled, significantly less under conditions of severe wind and weather [55, 56]. Unrecovered oil remains in the environment, and undergoes the weathering processes described above, with the most severe environmental consequences resulting when oil strands on shorelines [57, 55]. Beached oil increases the likelihood of contamination for habitats and animals found in subtidal, intertidal and supratidal areas, which include some of the most productive and diverse portions of the marine environment. Burning spilled oil in a contained and controlled manor, so as not to jeopardize the bulk of remaining cargo or other response assets, can rapidly remove bulk oil from the water surface. However, it is a logistical and operational challenge to contain the oil, arrange and control its placement out of the immediate area of spill response activity, and ensure sufficient oil thickness to sustain an efficient burn [40]. Many of the logistical and physical constraints working against efficient mechanical containment and recovery also confound attempts to collect and burn oil on water. When the oil does burn, the unburned residue is comprised mostly of the heavy, longer chain hydrocarbons, which are relatively resistant to ready microbial degradation [58]. Dispersants are widely recognized by many regulatory agencies as an effective at-sea response that provides a net environmental benefit when compared to reliance on mechanical recovery alone (see chapter 9). Application of chemical dispersants facilitates the breakup of the oil slick, moving oil from the water surface into the water column as neutrally buoyant oil droplets ranging from 1 to 100 micrometers in diameter, due to the mixing action of waves and currents. Subsequently, this plume of oil droplets rapidly distributes throughout the water column, mixing into lateral and deeper water masses and reducing oil concentrations below levels of concern for marine life. The rate and effectiveness of this process depends on the nature of the spilled oil (its API gravity and viscosity, degree of weathering, extent of emulsification, and pour point) and the ability of the dispersant formulation to mix with the oil. Dispersants have been an effective aspect of oil spill response over the past 30 years, with applications to major and smaller oil spill incidents in many of the world's oceans (Fig. 3). From 1970 through 1998, dispersants have been used on approximately 37% of oil spills covered in a worldwide database by the Oil Spill Intelligence Reporter [59]. In addition to countless small-scale tests
504
that have been conducted in laboratories around the world, critical assessments of dispersant performance have been organized by private and government research organizations, often cooperatively, using controlled releases of large volumes of oil and dispersant applications under real world conditions (Fig. 3). These studies have led to modern dispersant formulations with improved effectiveness and greater environmental safety. A range of dispersant products are stockpiled around the world for spill response, and a few have been shown to be effective over a broad range of oil types and environmental conditions [60]. An important environmental consideration associated with dispersant use is assessing the environmental tradeoff between intentionally exposing water column plants and animals to dispersed oil and the often significant effects of unrecovered oil left to drift at sea to potentially strand on a shoreline. In most cases, these considerations demonstrate a net environmental benefit to the use of dispersants because the short-term, transient exposure of water column communities has much less ecological effect than the prolonged, wide-spread contamination of oil reaching shorelines [57, 55, 61]. The environmental risks of dispersed oil are further decreased by the rapid degradation of the small, dispersed oil droplets moving through the water column, compared to the persistence observed for bulk oil stranded on shorelines and incorporated into sediments. The large surface to volume ratio characteristic of micron-sized dispersed oil droplets provides a colonizing substrate for oil degrading bacteria and a source of degradable hydrocarbon to support their growth. And, because the small oil droplets are widely dispersed in the water column, the supply of nitrogen and phosphorus nutrients needed to support bacterial degradation of the diluted oil is sufficient to maintain a viable degrader community in association with the oil droplet. Furthermore, laboratory studies have shown that some dispersants can enhance the initial rate of oil degradation due to the presence of constituents that serve as initial substrates for nascent bacterial growth [62, 63]. Laboratory studies of the fate of dispersed oil droplets have characterized the process by which it becomes a physical substrate for supporting a microbial community as well as a chemical substrate to support their growth. Within 2 to 4 days, the dispersed oil droplet becomes colonized by oil degrading microbes [63-65]. Subsequently, this can become a full floating heterotrophic community consisting of oil, bacteria, protozoa and even nematodes. Macnaughton et al. [65] reported that by day 16, the size of the clusters increased and sank in test microcosms, most likely the result of reduced buoyancy due to oil biodegradation and increased biomass associated with the droplets.
505
Fig. 3. Dispersant response to oil spills. Data taken from reference 59.
6.2. On shore: If oil reaches shore then the first response is to collect it [66]. Oil typically lands on only the upper portion of the intertidal zone, and on sandy beaches it may be possible to collect the oiled sand with mechanical equipment. This was done, for example, with the spill from the Sea Empress [67]. Particularly heavy oils may be best picked up by hand, as in the case of the spill from the Prestige [68]. On rocky beaches it may be possible to wash oil back into the sea where it can be collected with skimmers, as was done following the spill from the Exxon Valdez [69]. Once the bulk oil has been removed by physical techniques, residual oil is eventually naturally biodegraded. Bioremediation aims to stimulate the rate of natural biodegradation, without causing any additional adverse impact, by at least partially alleviating whatever is limiting microbial growth. In most porous, and therefore aerobic shorelines, the most likely limitation is biologically available nitrogen and phosphorus, and effective bioremediation protocols have applied various forms of fertilizers to deliver these nutrients. Research on this topic has been going on for decades in many parts of the world (Fig. 4; reviewed in 42, 44, 70-77). The simplest approach is to alleviate the nutrient limitation of oil biodegradation by adding fertilizers. This was the basis for the successful bioremediation of the spill from the Exxon Valdez [78-
506
80]. Two different fertilizers were used, an oleophilic fertilizer, Inipol EAP22®, designed to adhere to oil and deliver nutrients at the oil-water interface [81] and a slow-release granular agricultural product (Customblen®) that would release nutrients over many weeks through the beach gravel. Inipol EAP22 is a microemulsion with an external oil phase of oleic acid and trilaureth-4phosphate, containing an internal phase of urea in aqueous solution, cosolubilized with butoxy-ethanol to adjust the viscosity. It contains 7.4% nitrogen and 0.7% phosphorus by weight, and was applied with airless sprayers transported on small shallow-draft catamarans. Customblen® is a high quality agricultural fertilizer designed to release its nutrients over several weeks. It consists primarily of ammonium nitrate, calcium phosphate and ammonium phosphate, encapsulated in polymerized linseed oil. Customblen contains 28% nitrogen and 3.5% phosphorus by weight, and was applied by workers walking the beaches with broadcast spreaders. An extensive monitoring program demonstrated that the fertilizer applications were successful at stimulating the rate of biodegradation some two- to five-fold [78-80] A quite similar approach was used on a limited scale following the spill from the Sea Empress [82]. Much of this spill was treated with dispersants while at sea, and most of the residue that landed on shore was collected by work crews, but some oil landed on a relatively steep (gradient 10-12.5%) shingle and pebble beach at Bulwell Bay. Because the beach was so steep, slow-release fertilizer was placed in 1 m mesh bags, and secured to the beach with steel pegs. Again, the rate of biodegradation was stimulated more than two-fold on the fertilized part of the beach. To our knowledge, these are the only two occasions when bioremediation by the addition of relatively simple fertilizers was used following a spill, but there have been field and laboratory tests all over the world that have found similar results (see Figure 4). All sorts of fertilizers have been used, usually with success, including soluble and slow release forms of inorganic and organic nitrogen. Our most recent experiments were on Spitsbergen, the largest island of Svalbard, Norway, (approximately 78° N, 17' E.) [83, 84]. Slow release and soluble fertilizers were applied in much the same way they were in Alaska, and they led to an approximate doubling of the rate of biodegradation, even in this cold, Arctic environment. A slightly more complex approach has been championed by Rosenberg and colleagues [71, 85, 86]. In this case the fertilizer is an insoluble polymer of urea and formaldehyde, and it is applied together with an oil-degrading bacterial inoculum that can use this nitrogen source. The approach was apparently able to stimulate the biodegradation of a small spill (100 tons) of a heavy crude oil on a sandy beach between Haifa and Acre in Israel in the early 1990's [71, 86].
507
Fig. 4. Bioremediation response to oil spills. Data taken from references 70 - 80.
Others have suggested that what really limits oil biodegradation in the environment is the absence of effective oil degrading microorganisms, and they therefore seek to add such organisms. Most recently this has been attempted on heavy oil spilled by the Nakhodka in the Sea of Japan [87, 88]. Assessing this work is problematic. The published analyses of the field work rely on digital photography of representative oiled rocks, and no detailed chemical analyses have been presented that can be compared with what has been found in other spills. Earlier microbial inocula did not perform well in standardized tests [89]. An important corollary to any oil spill remediation is that it should have a net environmental benefit [90]. By aiming to stimulate natural processes, bioremediation is likely to have minimal adverse effects if carried out carefully and conscientiously, but there are obvious potential risks that must be evaluated. Potential adverse impacts include the possibility that the fertilizer applications might be acutely toxic to marine biota, might stimulate nearshore algal blooms, might cause the production of biosurfactants that could result in increased removal of oil from the shorelines by tidal flushing and lead to broader shoreline impacts, or might generate toxic by-products. Careful monitoring following the spill from the Exxon Valdez [91] and a field trial in the Arctic [92] failed to detect any adverse environmental impact from the careful application of fertilizers, while the rate of hydrocarbon biodegradation was stimulated two- to five-fold.
508
7. CONCLUSIONS Oil spill bioremediation technologies epitomize modern environmental techniques: working with natural processes to remove spilled oil from the environment while minimizing undesirable environmental impacts. If a floating oil slick cannot be collected or burnt, chemical dispersants will cause the oil to move into the water column as tiny droplets with a dramatically increased surface area that allows rapid biodegradation. If oil reaches a shoreline and cannot be removed physically, the careful addition of fertilizers will stimulate oil biodegradation without adverse environmental impact. These two tools are thus an important part of the toolkit for dealing with accidental and deliberate releases of oil into the marine environment.
REFERENCES [I] [2] [3] [4] [5] [6] [7]
[8] [9] [10] [II] [12] [13] [14] [15] [16]
J.M. Hunt, Petroleum Geochemistry and Geology, 2nd edition. W.H. Freeman, New York. 1996. B.P Tissot and D.H. Welte, Petroleum Formation and Occurrence. Springer-Verlag, Berlin. 1984. J.Z. De Boer, J.R. Hale, and J. Chanton, New evidence of the geological origins of the ancient Delphic oracle (Greece). Geology, 29 (2001) 707. D. Yergin, The Prize, the epic quest for oil, money and power. Simon and Schuster, New York. 1992. National Research Council, Oil in the Sea III: Inputs, Fates and Effects, National Academy Press, Washington DC, 2002. E.M Levy and K. Lee, Can. J. Fish. Aquat. Sci. 45 (1988) 349. R.C. Prince, Petroleum and other hydrocarbons, biodegradation of. hi Encyclopedia of Environmental Microbiology (G. Bitton, ed.) John Wiley, New York, 2002 pp. 24022416. Anonymous. Industry at a glance. World Oil 224 (2003) 75. D.S. Etkin, Historical overview of oil spills from all sources. In: Proceedings of the 1999 International Oil Spill Conference. American Petroleum Institute, Washington DC. 1999 pp. 1097-1102. H.J. Barth, Mar. Poll. Bull. 46 (2003) 1245. A.N. AlGhadban, F. Abdali, and M.S. Massoud, Environ. International 24 (1998) 23. K. Peters and J.M. Moldowan, The Biomarker Guide; Interpreting molecular fossils in petroleum and ancient sediments. Prentice-Hall, Englewood Cliffs, NJ. 1993. R.C. Prince, D.L. Elmendorf, J.R. Lute, C.S. Hsu, C.E. Haith, J.D. Senius, G.J. Dechert, G.S. Douglas, and E.L. Butler, Environ. Sci. Technol. 28 (1994) 142. W.K. Robbins and C.S. Hsu, Petroleum Composition. In: Kirk-Othmer Encyclopedia of Chemical Technology, Fourth Edition. John Wiley & Sons, New York. 1996 pp. 352370. Y.A. Ibrahim, M.A. Abdelhameed, T.A. Al-Sahhaf, and M.A. Fahim, Petrol. Sci. Technol. 21 (2003) 825. J Burdon, Soil Science, 166 (2001) 752.
509
[17] K. Qian, R.P. Rodgers, C.L. Hendrickson, M.R. Emmett and A.G. Marshall, Energy & Fuels, 15(2001)492. [18] K. Qian, W.K. Robbins, C.A. Hughey, HJ. Cooper, R.P. Rodgers and A.G. Marshall, Energy & Fuels, 15 (2001) 1505. [19] A.D. Uhler, S.A. Stout, K.J. McCarthy, S. Emsbo-Mattingly, G.S. Douglas and P.W. Beall, Contaminated Soil Sediment and Water, April/May (2002) 20. [20] R.C. Prince, R.M. Garrett, R.E. Bare, M.J. Grossman, T. Townsend, J.M. Suflita, K. Lee, E.H. Owens, G.A. Sergy, J.F. Braddock, J. E Lindstrom. and R.R. Lessard, Spill Sci. Technol. Bull, 8 (2003) 145. [21] M.F. Fingas, The evaporation of oil spills: development and implementation of new prediction methodology. In Proceedings of the 1999 International Oil Spill Conference, American Petroleum Institute, Washington DC, 1999 pp. 185-194. [22] Y. Wang and Y. Huang, Appl. Geochem. 18 (2003) 1641. [23] M.D. Hurley, O. Sokolov, T.J. Wallington, H. Takekawa, M. Karasawa, B. Klotz, I. Barnes, and K.H. Becker, Environ. Sci. Technol. 35 (2001) 1358. [24] K.M. Arzayus, R.M. Dickhut, and E.A. Canuel, Environ. Sci. Technol. 35 (2001) 2178. [25] R.C. Prince, R.T. Stibrany, J. Hardenstine, G.S. Douglas and E.H. Owens, Environ. Sci. Technol. 36 (2002) 2822. [26] Bitor Corporation (2004) www.orinutlsionfuel.com/ [27] M.F. Fingas, B. Fieldhouse, J. Lane and J.V. Mullin, What causes the formation of water-in-oil emulsions? In Proceedings of the 2001 International Oil Spill Conference, American Petroleum Institute, Washington DC, 2001 pp. 109-114. [28] R. Thomas and T. Lunel, The Braer Incident; dispersion in action. In Proceedings of the Sixteenth Arctic Marine Oilspill Program Technical Seminar, Edmonton, Alberta, 1993 pp. 843-859. [29] R. Goodman, Spill Sci. Technol. Bull. 8 (2003) 117. [30] J.R. Bragg and E.H. Owens. Clay-oil flocculation as a natural cleansing process following oil spills: Part 1-studies of shoreline sediments and residues from past spills. In: Proceedings of the Seventeenth Arctic and Marine Oilspill Program (AMOP) Technical Seminar. Ottawa, Canada. 1994 pp. 1-24. [31] E.H. Owens and K. Lee, Mar. Poll. Bull. 47 (2003) 397. [32] E. Lafargue and P. Le Thiez, Org. Geochem. 24 (1996) 1141. [33] R.M. Garrett, I.J. Pickering, C.E. Haith, and R.C. Prince, Environ. Sci. Technol. 32 (1998)3719. [34] International Agency for Research on Cancer. IARC Monograph on the evaluation of the carcinogenic risk of chemicals to humans. Polynuclear aromatic hydrocarbons, Part 1, Chemical, environmental and experimental data, Vol. 32. World Health Organization, Geneva. 1983 [35] L.H. Keith and W.A. Telliard, Environ. Sci. Technol. 13 (1979) 416. [36] European Commission, DECISION No 2455/2001/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 20 November 2001 establishing the list of priority substances in the field of water policy and amending Directive 2000/60/EC, Official Journal of the European Union L331, 1-5, 15 December 2001. [37] E.H. Owens, J.R. Harper, W. Robson and P.D. Boehm, Arctic 40 (1987) 109. [38] M. Martinelli, A. Luise, E. Tromellini, T.C. Sauer, J.M. Neff and G.S. Douglas, The M/C Haven oil spill: Environmental assessment of exposure pathways and resource injury. Proceedings of the 1995 Oil Spill Conference, American Petroleum Institute, Washington, D.C. 1995 pp. 679-685.
510
[39] J.J. Gallagher, H.B. Hile and J.A. Miller, The old New Carissa; a study in patience. Proceedings of the 2001 Oil Spill Conference, American Petroleum Institute, Washington, D.C. 2001 pp. 85-90. [40] I. Buist, Spill Sci. Technol. Bull. 8 (1993) 341. [41 ] The University of Minnesota Biocatalysis/Biodegradation Database (2004) http://umbbd.ahc.umn.edu/ [42] R.C. Prince, Crude oil biodegradation In The Encyclopedia of Environmental Analysis and Remediation, Volume 2, 1327-1342. John Wiley, New York. 1998. [43] S.J. McMillen, A.G. Requejo, G.N. Young, P.S. Davis, P.D. Cook, J.M. Kerr, and N.R. Gray, Bioremediation potential of crude oil spilled on soil. In: Microbial Processes for Bioremediation (R.E. Hinchee, CM. Vogel, and FJ. Brockman, eds.). Battelle Press, OH. 1995 pp, 91-99. [44] R.C. Prince, Crit. Rev. Microbiol. 19 (1993) 217. [45] C.J. Omiecinski, R.P. Remmel and V.P.. Hosagrahara, Toxicol. Sci. 48 (1999) 151. [46] V.J. Melendez-Colon, A. Luch, A. Seidel, and W.M. Baird, Cancer Res. 59 (1999)1412. [47] R. Margesin and F. Schinner, Appl. Microbiol. Biotechnol. 56 (2001) 650. [48] H. Feitkenhauer, R. Muller and H. Markl, Biodegradation 14 (2003) 367. [49] A.G. Rike, K.B. Haugen, M. Barresen, B. Engene. and P. Kolstad, Cold Regions Sci. Technol. 37 (2003) 97. [50] M.J. Gauthier, B. Lafay, R. Christen, L. Fernandez, M. Acquaviva, P., Bonin and J.C. Bertrand, Int. J. System. Bacteriol. 42 (1992) 568. [51] A.M. Spormann and F. Widdel, Biodegradation 11 (2000) 85. [52] J.A. Cunningham, H. Rahme, G.D. Hopkins, C. Lebron, and M. Reinhard, Environ. Sci. Technol. 35 (2000) 1663. [53] M.E. Caldwell, R.M. Garrett, R.C. Prince and J.M. Suflita, Environ. Sci. Technol. 32 (1998)2191. [54] J.D. Coates, J. Woodward, J. Allen, P. Philp and D.R. Lovley, Appl. Environ. Microbiol. 63(1997)3589. [55] A. Lewis and D. Aurand, Putting Dispersants to Work: Overcoming Obstacles. API Technical Report IOSC-004. 1997 International Oil Spill Conference Issue Paper. API. Washington, DC. 1997. [56] D.S. Etkin and P. Tebeau, Assessing progress and benefits of oil spill response technology development since Exxon Valdez. Proceedings of the 2003 International Oil Spill Conference, American Petroleum Institute, Washington, DC, 2003 pp. 843-850. [57] National Research Council, Using Oil Dispersants on the Sea. National Academy Press. Washington, DC. 1989. [58] R.M. Garrett, C.C. Guenette, C.E. Haith and R.C. Prince, Environ. Sci. Technol. 34 (2000) 1934. [59] D.S. Etkin, Factors in the Dispersant use decision-making process: historical overview and look to the future, Proceedings of the twenty-first Arctic and Marine Oil Spill Program (AMOP) Technical Seminar. Environment Canada. Ottawa. 1998 pp. 281-304 [60] R.R. Lessard and G. DeMarco. Spill Sci. Technol. Bull. 6 (2000) 59. [61] International Petroleum Industry Environmental Conservation Association (IPIECA), Oil Spill Preparedness and Response. Volume 5: Dispersants and their Role in Oil Spill Response. London. 2000. [62] R. Varadaraj, M.L. Robbins, J. Bock, S. Pace and D. MacDonald, Dispersion and biodegradation of oil spills on Water. In: Proceedings of the 1995 International Oil Spill Conference, American Petroleum Institute, Washington, DC. 1995 pp. 101-106.
511
[63] R. Swannell and F. Daniel, Effect of dispersants on oil biodegradation under simulated marine conditions. Proceedings of the 1999 International Oil Spill Conference, American Petroleum Institute, Washington, DC, 1999, pp. 169-176. [64] J.E. Lindstrom and J.F. Braddock, Mar. Poll. Bull. 44 (2002) 739. [65] S.J. Macnaughton, R. Swannell, F. Daniel and L. Bristow, Spill Sci. Technol. Bull. 8 (2003) 179. [66] B.E. Ornitz and M.A. Champ, Oil Spills First Principles. Elsevier, New York. 2002. [67] T. Lunel, K. Lee, R. Swannell, P. Wood, J. Rusin, N. Bailey, C. Halliwell, L. Davies, M. Sommerville, A. Dobie, D. Mitchell and M. McDonagh, Shoreline clean up during the Sea Empress incident: the role of surf washing (clay oil flocculation), dispersants and bioremediation. In Proceedings of the Nineteenth Arctic and Marine Oil Spill Program Seminar. Environment Canada, 1996 pp. 1521-1540. [68] J. Bohannon, X. Bosch and J. Withgott, Science 298 (2002) 1695. [69] S.A. Nauman, Shoreline clean-up: equipment and operations. In Proceedings of the 1991 International Oil Spill Conference, American Petroleum Institute, Washington DC, 1991 pp. 141-148. [70] R.M. Atlas, Crit. Rev. Microbiol. 5 (1977) 371. [71] E. Rosenberg, R. Legman, A. Kushmaro, R. Taube, E. Adler and E.Z. Ron, Biodegradation 3 (1992) 337. [72] R.P.J. Swannell, K. Lee and M, McDonagh, Microbiol. Rev. 60 (1996) 342. [73] Q. Lin, LA. Mendelssohn, C.B. Henry, P.O. Roberts, M.M. Walsh, E.B. Overton and R.J. Portier, Environ. Tech. 20 (1999) 825. [74] M. Mathew, J.P. Obbard, Y.P. Ting, Y.H. Gin and H.M. Tan, Acta Biotechnol. 19 (1999)225. [75] K. Lee and S. deMora, Environ. Tech. 20 (1999) 783. [76] I. Head and R.P.J. Swannell, Curr. Opin. Biotechnol. 10 (1999) 234. [77] S. Harayama, H. Kishira, Y. Kasai and K. Shutsubo, J Mol Microbiol Biotechnol. 1 (1999) 63. [78] J.R. Bragg, R.C. Prince, E.J. Harner, and R.M. Atlas, Nature 368 (1994) 413. [79] R.C. Prince, J.R. Clark, J.E Lindstrom, E.L. Butler, E.J. Brown, G. Winter, MJ. Grossman, R.R. Parrish, R.E. Bare, J.F. Braddock, W.G. Steinhauer, G.S. Douglas, J.M. Kennedy, P.J. Barter, J.R. Bragg, E.J. Harner and R.M. Atlas, Bioremediation of the Exxon Valdez oil spill: monitoring safety and efficacy, hi Hydrocarbon Remediation(R. E. Hinchee, B. C. Alleman, R. E. Hoeppel and R. N. Miller, eds.) Lewis Publishers, Boca Raton, FL., 1994 pp. 107-124. [80] R.C. Prince and J.R. Bragg, Bioremediation J. 1 (1997) 97. [81] A. Ladousse and B. Tramier, Results of 12 years of research in spilled oil bioremediation: Inipol EAP22. In: Proceedings of the 1991 International Oil Spill Conference, American Petroleum Institute, Washington DC, 1991 pp. 577-582. [82] R.P.J. Swannell, D. Mitchell, G. Lethbridge, D. Jones, D. Heath, M. Hagley, M. Jones, S. Petch, R. Milne, R. Croxford and K. Lee, Environmental Technol. 20 (1999) 863. [83] C.C. Guenette, G.A. Sergy, E.H. Owens, R.C. Prince and K. Lee, Spill Sci. Technol. Bull. 8 (2003), 245. [84] R.C. Prince, R.E. Bare, R.M. Garrett, M.J. Grossman, C.E. Haith, L.G. Keim, K Lee, G.J. Holtom, P. Lambert, G.A. Sergy, E.H. Owens and C.C. Guenette, Spill Sci. Technol. Bull. 8 (2003) 303. [85] E. Rosenberg, R. Legman, A. Kushmaro, E. Adler, H. Abir and E.Z. Ron, J Biotechnol. 51 (1996)273.
512
[86] E. Rosenberg and E.Z. Ron, Non-polluting compositions to degrade hydrocarbons and microorganisms for use thereof. US Patent 5780290 1998 [87] T. Hozumi, H. Tsutsumi, and M. Kono, Mar. Poll. Bull. 40 (2000) 308. [88] H. Tsutsumi, M. Kono, K. Takai, T. Manabe, M. Haraguchi, I. Yamamoto, and C. Oppenheimer, Mar. Poll. Bull. 40 (2000) 320. [89] A.D. Venosa, J.R. Haines and D.M. Allen, J. Ind. Microbiol. 10 (1992) 1. [90] J.M. Baker, Net environmental Benefit Analysis for oil spill response. In Proceedings of the 1995 International Oil Spill Conference, American Petroleum Institute, Washington DC, 1995 pp. 611-614. [91] R.C. Prince, J.R. Clark and K. Lee, Bioremediation effectiveness; removing hydrocarbons while minimizing environmental impact. In Proceedings of the Ninth International Petroleum Environmental Conference, Albuquerque, NM. 2002,available at: http://ipec.utulsa.edu/Ipec/Conf2002/prince dark lee 109.pdf [92] K. Lee, G. Wohlgeschaffen, G.H. Tremblay, B.T. Johnson, G.A. Sergy, R.C. Prince, C.C. Guenette and E.H. Owens, Spill Sci. Technol. Bull, 8 (2003) 273.
Studies in Surface Science and Catalysis 151 R. Vazquez-Duhalt and R. Quintero-Ramirez (Editors) © 2004 Elsevier B .V. All rights reserved.
513
Chapter 19
Biotreatment of water pollutants from the petroleum industry E. Razo-Flores,a>b P. Olguin-Lora," S. Alcantara3 and M. Morales-Ibarria" a
Instituto Mexicano del Petroleo, Programa de Biotecnologia. Eje Central Lazaro Cardenas 152, C.P. 07730, Mexico D.F. b
Instituto Potosino de Investigation Cientifica y Tecnologica,. Camino a la Presa San Jose 2055,. C.P. 78216, San Luis Potosi, SLP, Mexico. 1. INTRODUCTION Industrialization has resulted in the formation of waste products, which are released into the environment in the form of wastewater, gaseous emissions and solid residues leading to environmental pollution and deterioration. A good example of this situation is the petroleum industry (oil and gas, chemical and petrochemical). During decades, the production strategy aimed to maximize product outputs with minimum production costs. Therefore, a large water usage, soil contamination and energy wastage (oil by-products being lost into the environment) were a normal practice. Later, through the 1970s to 1980s, "end of the pipe" solutions were developed to control pollution. This approach was effective but it can not be affordable for much longer because the production and environmental protection costs are added together, rising global costs and wasting a great amount of energy and material resources (water, nutrients, metals, oil). A system approach that integrates human activities with the protection and restoration of the environment goes together with the sustainability concept in a world closely linked through communications and markets. More recently, during the last two decades, governmental regulatory actions changed profoundly the wastewater treatment in the petroleum industry, establishing effluent limitations for many specific organic and inorganic compounds. Nowadays, water can be considered as one of the main raw materials of the petroleum industry and its treatement and reuse with advanced treatment technology is being developed.
514
1.1. Characterizing petroleum industry wastewater Petroleum industry requires large water volumes for the oil and gas refining and processing. PEMEX, the state owned Mexican Oil Company, consumed 270.2 and 245.1 millions of cubic meters of water in the years 2001 and 2002, respectively, in its different processes. Besides, in 2002 the water input per unit of throughput was 0.17, 0.86, 1.73 and 10.8 m3 ton"1 for exploration and production, gas processing, refining, and petrochemical operations, respectively. At national level, the petroleum and chemical industry occupies the second place in industrial wastewater generation, both in volume and organic load, after sugarcane industry. Table 1 presents a general list of the main water pollutants in the petroleum industry. For a full review of the refinery and petrochemical effluents and the common treatments used (physicochemical and biological) [1-2]. 1.2. Biological reactions applied to petroleum wastewater treatment Biological processes are a cost-effective technology for the removal of organic, sulfur and nitrogen compounds from wastewaters. Table 2 shows the main transformations that can occur during biological petroleum wastewater treatment. Anaerobic processes are one of the most viable alternatives for the treatment of complex effluents like those produced in the petroleum industry. Up to date, methanogenic, sulfate-reduction and anoxic processes such as heterotrophic denitrification, have been used for the biodegradation of organic compounds. Most of these anaerobic processes were developed for the food industry, but recently have been successfully applied in the chemical and petrochemical industry wastewaters [3]. The most accepted high-rate process is that carried out in up-flow anaerobic sludge blanket reactor (UASB), where the hydraulic and biomass residence times are uncoupled, allowing a high biomass concentration inside it. The granule formation and stability are essential for the right operation of the UASB reactor. The methanogenic treatment of organic compounds (e.g. phenols, organic acids, etc.) is a complex microbial process involving many kinds of bacteria and several intermediate steps. Generally, the first step is the hydrolysis of the organic compounds producing simpler organics after which, they are fermented to volatile fatty acids by the acidogens. Furthermore, the acetogenic bacteria transform these compounds to acetate and hydrogen that are finally converted to biogas (methane and CO2) by the methanogens [4].
515
Table 1 Main water-soluble contaminants generated by the petroleum industry. Family Aromatic hydrocarbons
Oxygenated compounds
Sulfur compounds Nitrogen compounds
Compounds Benzene Toluene Ethylbenzene Xylene Phenols Organic acids Aldehydes Metyl tert-butyl ether Hydrogen sulfide Mercaptans Ammonium Amines Urea
Sulfur bearing wastewaters can be treated by using the biological reactions of the sulfur cycle. In the reductive side, both sulfate (SO42~) and elemental sulfur (S°) act as electron acceptors in the metabolism of a wide range of anaerobic bacteria, producing H2S. In the oxidative side of the cycle, sulfur reduced compounds are biologically oxidized to sulfate or elemental sulfur under either aerobic or anaerobic conditions by autotrophic bacteria. Table 2 Main transformations carried out during biological petroleum wastewater treatment. Biological Process (electron acceptor) Methanogenesis (CO 2 , HCO3") Organic compounds Sulfate-Reduction (SO 4 , S°) Organic compounds Metals Heterotrophic denitrification (NO3") Organic compounds Autotrophic denitrification (NO3") H2S Nitrification (O2) NH 4 + Aerobiosis (O2) H2S Organic compounds
Products —>
CH4 + CO2 + Biomass
—> H2S + CO2 + Biomass -» MeS —>
N2 + CO2 + Biomass N 2 + SO 4 (S°) + Biomass
->
NO3" + CO 2 + Biomass
-> -»
SO 4 (S°) + Biomass H2O + CO2 + Biomass
516
Nitrogen, mainly as ammonium (NH4+), is one of the most abundant contaminants in the petroleum industry wastewaters. Ammonium can be biologically eliminated by means of a double process, nitrification and denitrification, producing molecular nitrogen. Nitrification is a strict aerobic process, litoautotrophic, where ammonium is the electron and nitrogen sources, and it is oxidized to nitrate. Denitrification is a reductive process either heterotrophic or litoautotrophic process, where nitrate is reduced to elemental nitrogen. In this chapter it will be presented some of the more recent developments in biological wastewater treatment technology with application to the petroleum industry. The topics that will be covered are: a) Anaerobic biodegradation of aromatic compounds like phenol, alkylphenols and terephthalate. b) Biotransformation of S- and N-bearing inorganic compounds. c) Methyl-tert-butyl ether (MTBE) biodegradation. MTBE is a high recalcitrant compound and a potential water contaminant that only in few cases can be treated with technology originally developed for biological wastewater treatment. Conventional biological treatment, like activated sludge, is out of the scope of this chapter. 2. ANAEROBIC BIODEGRADATION AND BIOTRANSFORMATION OF AROMATIC COMPOUNDS The implementation of anaerobic wastewater treatment in the petroleum industry was initially limited due to the presumed toxicity and biodegradability of aromatic compounds present in these waste streams. However, the treatment of chemical and petrochemical wastewater has lately become a reality, due to a better understanding of the microbial biodegradation process and the discovery of the methanogenic granular sludge structure, which plays a key role in the development of the so called high rate anaerobic processes. The granular sludge is an aggregation of several metabolic groups of bacteria living in synergism. The granules have a diameter between 0.5 to 3 mm and a biomass concentration of approximately 100 g dry matter I'1. 2.1. Toxicity and biodegradability of phenolic compounds Spent caustic is one of the refineries waste streams, rich in phenolic compounds. This effluent is produced from nonregenerative desulfurization processes that use caustic soda scrubbing in combination with air oxidation. This process is used to remove H2S and CH3SH from gasoline and to remove H2S, CO2 and HCN from sour condensate gas [2]. The effluent, although involves very small volumes and its contain high concentration of phenols and sulfide.
517
The average phenol and alkyl phenols concentrations are 30.5 g 1" and 28.2 g 1' , respectively. There are several reports about the toxic effect produced by phenolic compounds on the acetoclastic methanogenic activity (AMA) of the granular sludge. Table 3 shows the inhibitory concentrations that reduce in 50% (IC50) the AMA. In general, the susceptibility of a granular sludge to the inhibitory effect of phenolic compounds is affected by the impact of its "acclimation history". The phenol-degrading acid-forming bacteria are more susceptible to phenol inhibition than the methanogens [5]. Most granules have a layered structure that protects bacteria, particularly methanogens. In the case of a phenol-acclimated granular sludge, it is possible that a phenol-degraders layer develops in the external zone of the granules, preventing the inward diffusion of the toxic compounds. This outer layer can prevent the methanogens deactivation either by reducing the exposure level or by a partial or complete biotransformation into nontoxic intermediates such as volatile fatty acids [5]. The selection and multiplication of an acetoclastic flora more resistant to those toxic compounds might be another protection mechanism. The inhibitory mechanism of the phenolic compounds is governed by their hydrophobicity that increases the ability of these compounds to solubilize into the lipid bacterial membranes, altering the membrane functions, such as ion transports causing cellular lysis. High linear correlations of methanogenic toxicity data to the logarithm of octanol-water partition coefficients of phenolic compounds (log P) have been proposed as shown in Fig. 1. This simple model adequately estimates the IC50 values for anaerobic granular sludge in the presence of phenolic compounds. Table 3 Inhibitory concentrations that reduce in 50% (IC50) the acetoclastic methanogenic activity of granular sludge (phenol-acclimated and non-acclimated) in batch assays [6-8]. Compound Phenol o-cresol m-cresol p-cresol 3,4-dimethylphenol 2-ethylphenol 4-methylphenol 4-ethylphenol
IC50 (mg I"1) 470 - 7802 433 - 844 443 - 919 389-1525 329-378 195-207 657 289
518
Fig. 1. Relationship between IC50 of phenolic compounds and the octanol/water partition coefficient (Log P). Synthetic "spent-caustic phenols mixture" (X), data from reference [6] (•), data from reference [7] (•) and data from reference [8]. (A). (1), phenol; (2), 4-methylphenol; (3), 4-ethylphenol; (4), o-cresol; (5), m-cresol; (6), /)-cresol; (7), 3,4-dimethylphenol; (8), 2-ethylphenol; (9), "synthetic phenols mixture". Log (l/ICso) = 0.77 Log P - 2.28, r2 = 0.90
Phenol is a compound easily biodegradable under anaerobic conditions. The biodegradation is initiated by phosphorylation of hydroxyl group followed by carboxylation of the ring in the para position (benzoyl-CoA pathway). In the case of the three cresol isomers (p-, m- and o-cresol) there are differences in their anaerobic biodegradability pathways. Methanogenic consortia are capable of /?-carboxylate the m-cresol and the o-cresol to their methylbenzoic acids. After carboxylation, the main degradation mechanism is the oxidation of the methyl group and in case of the /?-cresol it leads to the formation of a metabolic intermediate, the p-hydroxybenzaldehyde. It has been reported that p-cresol degradation also initiates by fumarate addition to the methyl group, forming benzyl-succinate. There are few reports about o-cresol biodegradation, since this compound is considered hard to be degraded under anaerobic conditions. Biodegradation rates of a mixture of phenol, p- and o-cresol obtained in batch experiments, with an adapted granular sludge, were approximately two-orders of magnitude higher than those observed with non-adapted sludge [9]. From evaluating the interaction of substrates, it was observed that p- and o-cresol did not affect phenol biodegradation, however, both phenol and o-cresol negatively affected /?-cresol biodegradation at the concentrations tested [9]. In other study, degradation of />-cresol ceased when phenol was depleted. This suggests that degradation of the most refractory p-cresol also requires phenol as a co-substrate. However, after a period of acclimation to the phenol-free environment, the biomass was able to degrade p-cresol without any co-substrate [10]. So far, both xylenols and ethylphenol biodegradation has not been reported
519
under methanogenic conditions. A reversible reaction from 2-ethylphenol to 3hydroxy-4-ethylphenol seems to take place, but no further degradation has been described. 2.1.2. Anaerobic treatment systems for the biodegradation of phenol Anaerobic treatment of phenolic-bearing wastewaters produced from the petroleum industry is a viable option. The bioreactor system most commonly used for the anaerobic treatment of phenolics is the UASB, operating to a certain organic loading rate (OLR), usually referred to chemical oxygen demand (COD). Lab scale UASB reactors have been applied to treat single phenolic compounds at OLR as high as 6 and 7.2 g COD I"1 d"1 for phenol and />-cresol, respectively, showing high compound removal efficiencies [11, 12]. However, effluents from the petroleum industry are expected to contain mixtures of phenol and cresols as the main COD bearing fractions. Thus, a successful treatment of these effluents would require a simultaneous degradation of the major phenolic substrates. Table 4 shows some results of anaerobic treatment of phenolic compounds mixtures. Table 4. Continuous anaerobic treatment results of mixtures of phenolic compounds treated in upflow anaerobic sludge bed reactors. Mixture
OLR (g COD I ' d 1 )
COD removal (%)
Reference
Phenol p-Cresol
7
94
[9]
7.1
91
[9]
2.95
81.8
[9]
8.12
85
[10]
0.66
85
[13]
4.3
-
[15]
Phenol /?-Cresol Phenol jt?-Cresol
o-Cresol Phenol p-Cresol Phenol />-Cresol o-Cresol Phenol m-Cresol
520
The operational parameters of the UASB reactors have important implications on the biodegradation efficiency of the phenolic compounds. In general, the effect of OLR is more drastic in reactors with increased phenolic concentrations, than in reactors with a constant phenolic concentration, and a decreased hydraulic retention time [10]. In the same way, the phenol/cresols ratio has to be controlled to avoid inhibitory or toxic effects to the living biomass. Cresol concentrations higher than 600 mg 1" can cause severe inhibition on the activity of the granular sludge [9, 10]. In a typical reactor wih phenol is used as sole source of carbon and energy, granulation is reported to initiate after 3 months of the start-up operation and develops for 6 months, to become fully mature. Granular sludge cultivated has an average diameter of 1.8 mm and is highly settable with a settlement volumetric index (SVI) of 14 ml g"1 [5]. The removal of phenolic mixtures can be improved in an UASB reactor using bioaugmentation. This method not only improves the start-up time, but also the COD removal. The bioaugmentation can be performed by simple adsorption of the specific bacterial consortium onto the granules, to protect it from being washed-out [13]. An increase of the enrichment from 2 to 5% improved considerably the start-up of the reactors treating phenolic compounds [14]. It was until 1981 that the two first full scale reactors treating chemical wastes were built by Celanese Company in USA. A third reactor was built three years later, and by 1989, 19 full-scale reactors were in operation treating wastewater from the chemical and petrochemical industry. Since 1990, the rate of digesters construction for that industrial sector increased from 2.1 to 4.6 reactors per year. Although an UASB reactor has been in operation since 1986 to treat phenol-bearing wastewater, no other reactor has been built to treat the same type of effluents since then. This UASB reactor of 1280 m3 is treating a 30.5 g COD I"1 with an OLR between 9 to 12 g COD l"'d"' and a COD removal of 95% has been achieved [3]. 2. 2. Toxicity and biodegradability of terephthalic acid Phthalic acid isomers (benzene-dicarboxylic acid) are important constituents of polyester fibers, films, polyethylene terephthalate (PET) bottles and other plastics. During production of phthalic acids, an important volume of wastewater is generated, approximately 3-10 m3 per ton of purified terephthalic acid (PTA) containing 5-20 kg COD m"3 [16]. The main components in the wastewater are terephthalic acid, acetic acid, benzoic acid and p-toluic acid in decreasing order of concentration. After neutralization with NaOH, all acids are present as sodium salts. Due to the characteristics of these wastes, anaerobic
521
pretreatment has been generally recognized as beneficial for wastewater treatment. 2.2.1. Toxicity and biodegradation Terephthalate concentration of 5 g COD I"1 does not produce any substrate inhibition on its biodegradation and methanogenic activity [17]. Hydrogenotrophic methanogenesis inhibition by 4-carboxybenzaldehyde, ptoluate and terephthalate generates IC50 values of 0.8 g I"1, 4.6 g I"1 and 16.6 g I"1, respectively. Nonetheless, methane production can be inhibited by un-ionized terephthalic acid, in near colloid state, using a settled terephthalic acid wastewater (pH 4.5) or purified terephthalic acid (0.183 g g'VSS) adjusted to pH6.15 [18]. The very low specific growth rate of terephthalate-biodegrading bacteria (0.04 h"1) explains the long-lasting acclimation period and low loading rate applied in UASB reactors [17]. The use of co-substrates like sucrose, benzoic and acetic acids inhibits the terephthalate and p-toluate biodegradation. The addition of benzoate delays the terephthalate biodegradation, which resembles a diauxic inhibition [17]. The generally accepted metabolic pathway of terephthalate biodegradation is the benzoyl CoA pathway after a probable decarboxylation leading to the formation of benzoate. The decarboxylation step is thermodynamically favorable under standard conditions, while the conversion of benzoate to acetate is a highly endergonic process. The global conversion of terephthalate to acetate and H2 becomes exergonic only when acetate and H2 are at very low concentrations. The fermentation of co-substrate by methanogenic granular sludge results usually in the production of H2 and acetate, generating an increase in the AG 0 ' which, in turn, may limit the terephthalate biodegradation. Analysis of specific activities of terephthalate and benzoate biodegradation demonstrated that terephthalate biodegradation activity was lower with a 33.6 mg COD g"'VSS d"1 value versus 117 mg COD g"1 VSS d'1 value for benzoate activity. Thus, the initial conversion of terephthalate to benzoate seems to be the limiting step of the microorganisms involved in terephthalate anaerobic biodegradation. Three bacterial populations were involved: 1) a syntrophic organism similar to that described for Syntrophus buswellii [19] able to convert terephthalate into acetate, CO2 and H2; 2) an acetoclastic methanogen; and 3) a hydrogenotrophic methanogen. 2.2.2. Terephthalate anaerobic treatment at full scale Crude terephthalic acid wastewaters must fulfill some conditions to be successfully pre-treated with anaerobic process: Certain grade of effluent neutralization, limiting concentration of other substrates than terephthalate or benzoate or acetate, high biomass retention rate and low volumetric loading rate.
522
The best operation is obtained with a plug-flow process or staged reactor system, because no substrate toxicity has been reported in normal operation with neutralized effluent. The company Amoco Petrochemicals Inc. operates a 15200 m3 full-scale downflow fixed film reactor with an OLR of 4.0 kg COD m3 d"1, which demonstrated the real feasibility of such pre-treatment [20]. The use of an expanded granular sludge bed-type bioreactor allowed a terephthalate removal higher than 80% and steady COD removal of 60% at an upflow velocity of 10 m d"1; however, in such conditions, p-toluate appeared to be recalcitrant to degradation [21]. Up to date, there are more than 10 full scale reactors treating terephthalic acid, indicating that the anaerobic treatment has become a conventional treatment for this kind of wastewater. The used bioreactor configurations are UASB, expanded granular sludge bed and hybrid reactors [22]. 3. BIOTRANSFORMATION OF S- AND N-BEARING INORGANIC COMPOUNDS FROM SOUR STREAMS The microbial treatment of sour wastewater resulting from either oil production or refining and other fossil fuels has been subject of intensive worldwide studies. The term "sour" was originated to describe those wastes contaminated with sulfide [23]. In refineries, sour wastewaters are generated from sour steam condensates that have been in contact with petroleum products, specifically from thermal or hydrogen cracking operations, where a carrier steam is used for injection or aeration [24]. Common total sulfur contents in sour water are around 1194 mg I"1. Because of the high sulfide, ammonium and phenols content, sour wastewater must be treated before its release into the environment. Both, aerobic and anaerobic processes have been reported to treat sour waste streams. 3.1. Aerobic processes Aerobic Thiobacilli species, which oxidize reduced sulfur compounds to obtain their growth energy, have been studied to promote the sulfur production from partial sulfide oxidation as shown in Eq. (1) [25, 26, 27]. These bacteria are gram-negative rods of about 0.3 um in diameter and 1 to 3 um long and belong to the colorless sulfur bacteria. An important characteristic is their capacity to excrete elemental sulfur, in contrast to filamentous colorless sulfur bacteria, as Thiotrix sp., which accumulate it intracellularly. Sulfur production from partial oxidation of sulfide instead of a complete oxidation to sulfate has a significant relevance because elemental sulfur can be recovered from the medium closing the sulfur cycle. Additionally, lower energy consumption is required because the oxidation to sulfur requires 4-fold less oxygen that the complete oxidation to sulfate, as shown in Eq. (2).
523
(1) (2) The reactor configuration, to promote both sulfur formation and accumulation, was evaluated and reported by Janssen et al. [27] and Alcantara et al. [28]. The configuration consisted mainly in the separation of aeration process from the bioreactor. Thus the liquid saturated with oxygen from the aerator vessel is sent to the reactor (reaction vessel) at a specific rate, which allows the control of stoichiometric molar ratio between oxygen and sulfide (theoretical molar ratio, Rmt, O2/S2~). When Rmt is close to 0.5, the sulfide oxidation is driven to elemental sulfur formation, while a Rmt close to 2 promotes sulfate as the main product. The performance of the system reported by Alcantara et al. [28] was inoculated with a sulfoxidizing consortium and it is shown in Fig. 2. Sulfide oxidation was studied under different dilution rates at steady state conditions of 0.5, 1, 1.5, 2 and 3 d"1 (zones A, B, C and D, respectively), maintaining a constant sulfide concentration in the feed solution at 4.0 g I"1. Elemental sulfur was produced at dilution rates of 0.5, 1, 1.5 and 2. The maximum sulfur formation occurred at Rmt of 0.5, where 85% of the total sulfur added to the reactor as sulfide was transformed to elemental sulfur and 92% of it was recovered from the bottom of the reactor.
Fig. 2. Performance of the recirculation reactor system under different culture conditions. Capital letters corresponds to the following dilution rates (d"1): A, 0.5; B, 1; C, 2 and D, 3. Subtitle letters show the Rmt evaluated: 2: b, c; 1.5: d; 1, e, k; 0.75: f, 1: m; 0.5: a, g; 0.35: h; 0.25: i; 0.15: j . Sulfide influent (—), sulfate (•), elemental sulfur (A), thiosulfate (o) and sulfide effluent (A).
524
Elemental sulfur production was affected by the dilution rate applied to the system. When the system operated at Rmt for sulfur production (0.5 and 0.75) and dilution rates of 0.5, 1 and 2, the elemental sulfur produced was higher than 60%, while washout conditions were observed when the dilution rate was increased from 2 to 3, at a Rmt of 0.75. The Thiobacilli species are strict autotrophic bacteria, thus organic compounds negatively affect their growth. However, sulfoxidizing consortia have shown an adequate metabolism to oxidize reduced sulfur and organic sulfur (CS2 for example) compounds [28], in presence of organic matter. According to Sublette et al. [23] and Alcantara et al. [28, 29], the oxidation of sulfur compounds is carried out by autotrophic bacteria while organic compounds are used as energy and carbon source by heterotrophic microorganisms. Phenol, o-, m- and />-cresol were degraded in a chemostat at various organic loading rates by the consortium. Under all conditions sulfide was completely oxidized to sulfate. Microcosm experiments showed that carbon dioxide production increased under presence of phenols, suggesting that these compounds were oxidized and they may be used as carbon and energy source by heterotrophic microorganisms present in the consortium [28]. The expanded bed reactor reported by Janssen et al. [27] is actually builtin to a family of processes called THIOPAQ, which are applied for the treatment of wastewater containing sulfide. Also, this technology has been proposed for the treatment of similar streams from petrochemical industries e.g. spent sulfidic caustics and from liquefied petroleum gas (LPG) scrubbers [30]. 3.2. Anaerobic processes Thiobacillus denitrificans is a gram-negative, chemoautotroph and facultative anaerobic bacteria, which oxidizes reduced sulfur compounds to obtain its growth energy and it is able to use nitrate as electron acceptor. According to Cadenhead and Sublette [31], this microorganism shows clear advantages to oxidize sulfide over other Thiobacilli, such as Thiobacillus thioparus, Thiobacillus versutus and Thiobacillus thiooxidans. Sulfide is commonly oxidized to sulfate (Eq. 3) or elemental sulfur (Eq. 4) under anoxic conditions, and where nitrate is used as a terminal electron acceptor being reduced to elemental nitrogen. 1.25 S2" + 2 NO3" + 2 H+ -> 1.25 SO42" + N2 + H2O
(3)
5 S2~ + 2 NO3" + 6 H2O - > 5 S ° + N 2 + 12 OH"
(4)
Sour waste streams, including sour water, sour gases and refinery spentsulfidic caustics, have been successfully treated using Thiobacillus denitrificans. For instance, the organic compounds such as benzene, toluene and phenol are
525
biodegraded by heterotrophic bacteria grown in co-culture with Thiobacillus denitrificans [23, 32]. Sublette [23] identified some technical limitations to apply this technology for the full-scale treatment of sour wastes. These include: substrate inhibition (sulfide), product inhibition (sulfate), the need for septic operation, biomass recycle and recovery, mixed waste issues, and the need for large-scale cultivation of the organism for the process start up. T. denitrificans strain F is sulfide tolerant [33] and it was used to treat oilfield produced water containing sulfides under full-scale field conditions at Amoco Production Co. in Salt Creek Field in Midwest, WY. More than 800 m3 d"1 of produced water containing 100 mg I'1 sulfide and total dissolved solids of 4800 mg I"1 were successfully biotreated in an earthen pit (3000 m3) over a sixmonth period. Based on an average flow of 795 m3 d"1, sulfide influx to the pit was about 80 kg d"'. Complete removal of sulfides and elimination of associated odors were clearly observed. More recently, there has been an increased interest about the oxidation of reduced sulfur compounds in presence of organics under denitrifying conditions [34, 35]. The novelty of this approach is the integration of biological processes that frequently were studied and applied separately. The coupling of carbon, nitrogen and sulfur cycles implicates the oxidation of reduced forms of sulfur, organic compounds, as well as the reduction of nitrate [36, 37]. According to Betlach and Tiedje [38], the heterotrophic denitrification process uses many organic compounds as carbon and energy source; thus organic transformations were coupled to nitrate reduction and further to molecular nitrogen. In the case of autotrophic denitrification, reduced sulfur compounds are oxidized to non-toxic compounds and nitrate, which is used as final electron acceptor, is reduced to molecular nitrogen. Reyes-Avila et al. [36] reported that the critical parameters to steer the nitrate reduction to molecular nitrogen are the C:N and N:S ratios for either heterotrophic or autotrophic processes, respectively. The same authors reported that biological denitrification was used to eliminate carbon, nitrogen and sulfur in an anaerobic continuous stirred tank reactor. Acetate and nitrate at a C:N ratio of 1.45 were fed at loading rates of 0.29 Kg C m"3 d"1 and 0.2 Kg N m"3 d"1, respectively. Under steady state denitrifying conditions, the carbon and nitrogen removal efficiencies were higher than 90%. Under these conditions, sulfide (S2") was fed to the reactor at several sulfide loading rates (0.044 to 0.295 Kg S2" m" 3 d~'). The high nitrate removal efficiency of the denitrification process was maintained along the whole process, whereas the carbon removal was 65%, even at sulfide loading rates of 0.295 Kg S2" m^d"1. The sulfide removal increased up to 99% via partial oxidation to insoluble elemental sulfur (S°) which accumulated inside the reactor.
526
In the same way, a denitrifying fluidized bed reactor for effectively remove sulfide, acetate and nitrate was proposed by Gommers et al., 1988 [39]. The authors reported that the rate-limiting step was the oxidation of sulfur to sulfate, nevertheless, the biomass showed an overcapacity to oxidize sulfide to sulfur and to degrade the acetate, under most tested loads. However, in order to develop a denitrifying technology to treat wastewaters from the petroleum industry, more studies are needed to elucidate the effect of phenolic compounds on both sulfide oxidation and nitrate reduction. 4. OXYGENATED FUEL ADDITIVES Oxygenated gasoline additives have been used since mid-1970s to substitute toxic lead compounds. The most common oxygenated used is methyl tert-hu\y\ ether (MTBE), that became the fourth chemical produced in USA [40] because of their mixing properties, high octane level, low cost and good results in reducing toxic emissions. MTBE is manufactured from isobutene (isobutylene or 2-methylpropene), a byproduct of petroleum refining, and methanol. Therefore MTBE can be easily and inexpensively produced at refineries. The MTBE presence in refinery effluents is due to discharges from facilities as a byproduct of the reprocessing of contaminated or "out of spec" product from the refinery. The volume and type of waste processed by refineries varies greatly over time, resulting in order-of-magnitude variations in the MTBE discharges. Few studies have evaluated the impact of this specific compound in complex wastewater in refineries [41]. MTBE has been present as a pollutant in numerous water resources mainly groundwater, The MTBE environmental impact is enhanced by the high solubility in water, low retention on organic matter, low detection threshold (2.5 and 2.0 jig I"1, for odor and taste, respectively) and low biodegradability. In 1996, the first case of contaminated aquifers by MTBE was reported in Santa Monica, CA. and 250,000 leaking underground fuel tank sites showed different levels of MTBE contamination [40]. In Germany, traces of MTBE were detected in rivers and influents and effluents of wastewater treatment plants [42]. The MTBE half-life in groundwater systems is several years [43]. There are few reports in Mexico about MTBE occurrence in the environment. Air concentrations of 11.5 ppb [44] and 4.4 ppb were monitored at a service station [45] and emissions of on-road vehicles, respectively. Additionally, concentrations between 100-1500 mg kg soil"1 were found in soils at fuels distribution and storage stations [46]. Concentrations in the range of 487 mg I"1 were found in groundwater at the surroundings of gas stations.
527
Fortunately, MTBE was detected in none of the nearby 33 monitored drinking water wells [47]. Increasing reports of MTBE in groundwater produced great concern about the toxicity and the carcinogenicity of this compound. Toxicological studies classified the compound as a potential carcinogen for humans [48] and regulations about the maximal concentration in groundwater were established. An extreme case was adopted in California where MTBE phase out by 2003 was ordered. However, as long as the use of MTBE continues, the risk of its presence in refinery effluents and water resources will be latent and treatments will be required. Due to its unique above-mentioned physicochemical properties, the clean up using common techniques like air injection, activated carbon filtering, etc. are inefficient for MTBE removal. Thus, biological techniques are of particular interest. In this section a review of MTBE biodegradation and biotreatments is done, in order to consider the experience adquired in this area for the eventual treatment of wastewater polluted with MTBE. 4.1 MTBE biodegradation MTBE has become a challenge for elucidation of its low biodegradability and the scarcity of MTBE-degrading microorganisms using it as carbon and energy source. The relatively recalcitrance of MTBE to microbial attack is intrinsic to its structure containing a combination of an ether link and the branched moiety. Alkyl ethers are stable molecules (AG° of the ether bond formation is 360 kJ mol"1 [49]). The high-energy demand for MTBE degradation is reflected by the low efficiency of biomass production on MTBE. Fortin et al, [50] pointed out the low MTBE biomass yield obtained analyzing different consortia. Salanitro [51] suggested that the slow growth on MTBE might also be due to considerable feedback regulation metabolites on the oxygenase responsible for the ether bond cleavage. The necessity of regenerating cofactors, such as NADH, could also have an influence on the rate of MTBE degradation, since reduced cofactors are required for several oxidation steps. Although initial works showed the high recalcitrance of this compound, some authors have reported the biodegradation of MTBE as sole carbon source. Moreover, cometabolism was shown to be an important mechanism for MTBE biodegradation by microorganisms able to grow mainly on short-chain alkanes. Anaerobic MTBE degradation has been recently observed under methanogenic [52], nitrate [53] and Fe(III) reducing conditions [54] with longer adaptation and degradation times. As far as we know, the highest value of MTBE heterotrophic degradation rate of 454 mg g protein"' h"1 was reported for a strain Hydrogenophaga flava ENV735 [55]. For cometabolism, the highest value was
528
obtained by the strain Mycobacterium vaccae JOB5 with a MTBE degradation rate of 111 mg g protein"1 h"1 when hexane was used as growth source [56]. A metabolic pathway for MTBE degradation has been proposed (Fig. 3), where the MTBE ether bond is enzymatically cleaved yielding tert-butyl alcohol (TBA) and formaldehyde as the main metabolic intermediates. TBA has been shown to further biodegrade to 2-methyl-2-hydroxy-l-propanol and 2hydroxyisobutyric acid [57]. Suspected further intermediates of the MTBE degradation metabolic pathway include 2-propanol, acetone and hydroxyacetone. The complete understanding of poor MTBE biodegradability would require the isolation of specialized microorganisms as will as the characterization of genes and enzymes involved in the degradation and regulation. Although microorganisms are able to grow using MTBE as a sole carbon and energy source, we are still far from understanding all causes for its low biodegradability. A number of excellent reviews are available on aerobic biodegradation of MTBE [10, 43, 58, 59].
Fig. 3. Proposed metabolic pathway for aerobic MTBE biodegradation Adapted from Fayolle et al. [49] and Steffan et al. [57].
529
4.2. MTBE removal biotreatments Although MTBE can be removed from groundwater by physical technologies such as activated carbon adsorption and air stripping, the costeffectiveness of these technologies in removal of MTBE is approximately 10 times higher than their application for removal of hydrocarbons, such as benzene and toluene, in groundwater. In December 2003, the USEPA established a database of 356 MTBE polluted sites and treatment technologies [60] including 111 full-scale completed cases. The main technologies used were: soil vapor extraction (18%), pump and treat (17%), in situ bioremediation (21%), air sparging (14%) and other technologies (30%). Bioremediation is a common technology and its cost has been estimated [61]. There are two engineering challenges associated with the in situ aerobic bioremediation of MTBE. First, groundwater polluted with MTBE has very low dissolved oxygen, thus in all cases the addition of air/oxygen is a requirement for the treatment; and the second, is the introduction of microorganisms able to degrade it. Table 5 shows some of the reported cases for in situ treatments. Field treatment includes the formation of a reactive zone named biobarrier by introducing to the subsurface MTBE-degrading microorganisms, which is placed to avoid the advance of the MTBE plume. Oxygen is supplied to the subsurface either by pulse injecting oxygen gas, air or any oxygen release compound. MTBE-contaminated water flowing through the biobarrier will contact the microbes and be degraded to CO2 and water. Biobarriers that have been applied successfully through biostimulation in some field studies, suggest that native microorganisms can degrade MTBE through amendments of nutrients and oxygen. However, the bioaugmentation, by adding microorganisms already adapted to MTBE degradation, has probed to be a more feasible option mainly when time-reduction in the treatment is required. In Salanitro's work, a comparison between biostimulation and bioaugmentation was performed [62]. The author found a notable MTBE reduction in both cases, but there was a difference of approximately 150 days in the lag phase between the treatments, achieving the total bioremediation of the site in approximately 200 days. Other example of biostimulation versus bioaugmentation was performed by Wilson et al. [63]. After six months a noticeable decrease in MTBE was achieved in both inoculated (with PM1 strain) and non-inoculated zones. Polymerase chain reactions techniques showed that in non-inoculated zone there was the presence of PM-1 like bacteria [64].
Table 5 Technology Performance for MTBE biological removal Treatment
Scale
Microorganism
MTBE initial concentration
Treatment period or removal rate
Reference
Field Field
MC-100 Native microorganisms PM-1 ENV425 Native microorganisms
7 mg I"1 1.5 mgT 1
150-200 days 4 days
[62] [63]
320 mg r 1 19.6 mgl- 1
90 days 60 days
[65] [66]
Hydrogenophaga flava Mixed culture Cytophaga-Flexibacter-Bacteroides
1000 mgl"1 5 mg T1
42 mg I"1 h"1 2.5 mg I"1 rf1
[67] [68]
ENV735
10 mg T1
10 mg 1"' in 15 min
[65]
10-50 mgT 1
29 mg r 1 h '
[69]
9 mg r 1 If1
In situ treatment
Field Field Ex situ treatment Laboratory Membrane Laboratory Fluidized bioreactor
Laboratory
Laboratory Mixed culture, cometabolism isopentane
Biotrickling filter Biofilter Biofilter N.S. not specified
Field Laboratory Field
Mixed culture Mixed bacterial culture N. S.
9.6 mg r 1 8.25 mg I"1 10 mgl"' and 15mgr'TBA
4.5 g rf' MTBE and 6.2 g hf' TBA
Laboratory Laboratory Laboratory
F-consortium PM-1 P. aeruginosa
0.8 mg I"1 100 mg 1"' 1.1-12.3 mgl"1
50 mg r1 h"1 58 mg r1 h"1 l.Smgr'h' 1
1
ls.smgr'h"
[69] [70] [71] [72] [73] [74]
531
On the other hand, the addition of a cosubstrate (propane) (US patent 5,814,514, Sept 29, 1998 and US patent 6,194,197 Feb 27,2001) to promote the cometabolic biodegradation of MTBE was useful for groundwater in situ bioremediation [65]. The authors inoculated a propane oxidizing strain ENV425 to cleanup the polluted site by installing biosparging and propane injection systems, and obtained a reduction of 90% in 90 days of treatment. This treatment should be preferred when polluted sites present hydraulic problems or for sites where groundwater extraction is required to stop the migration of contaminant plumes toward neighboring receptors. Bioreactors performance has been studied at lab scale and in some field applications (see Table 5). Most of the investigated bioreactors use immobilized microorganisms including membrane and fluidized reactors. Membrane technology retains high biomass levels improving the volumetric performance and reducing the area for treatment. However, limitations of this technology are the economic cost associated with the capital investment, low service-life and moderated operating costs associated with the pressure-driven mechanism of separation. Membrane fouling can also be a cost factor depending upon feed water conditions that might require pretreatment. Table 5 shows some works using this technology. In fluidized bioreactors, the biomass is immobilized in a support material (granular activated carbon, GAC, is commonly used) and this particles are in continuos movement using an upward water flow. Fluidization significantly increases the specific surface area available for biomass and thus degradation of contaminants. Besides the use of GAC as the fluidizing bed medium also increases specific surface area available for microbial colonization. These reactors avoid the bed plugging problems associated with a fixed bed bioreactor, but special care with operational flows should be taken to avoid washout the bed. However, this type of bioreactor requires a higher degree of operator maintenance and process control than the other readily available treatment processes. Some of the fluidized bioreactor studies are shown in Table 5, including two field experiences. MTBE treatment in vapor phase emissions is necessary when any stripping technology (soil vapor extraction, air stripping, etc.) is used for cleaning up groundwater containing MTBE (see chapter 17). Basically two configurations have been proved: Biofilters and biotrickling filters (Table 5). Biofilters use organic (diatomeaceous earth) or inert (vermiculate or granular activated carbon) packing material to support the microorganisms with non-addition or sporadic nutrient addition. Biotrickling filters are
532
similar to biofilters, but they have an aqueous phase trickling over the packed bed. The liquid contains essential nutrients and it is usually recycled. Biotrickling filters are more complex than biofilters but are usually more effective, especially for the treatment of compounds that generate acidic by-products (see chapter 17). 5. PERSPECTIVES It is expected that more stringent environmental regulatory actions will be taken by governments, worldwide. As water is the most important resource for human, animal and plant life, holistic environmental wastewater management will continue to gain in importance with time [75]. During the last decade significant efforts were devoted to the development of technologies for process integration targeting energy conservation and waste reduction. Great efforts have been done in industries in order to increase the water conservation and reduce wastewater [76]. However, these integrated technologies will produce less and more concentrated wastewater whose characteristic would lead to a complete redesign of the biological wastewater treatment processes that are currently applied on the process industry. Consequently, facility upgrading, innovative and sustainable treatment technologies would reshape the petroleum industry. The anaerobic processes for the treatment of organic compounds in industrial wastewater offer important advantages over conventional aerobic processes. To date, less than 15% of the nearly 1600 full-scale anaerobic wastewater treatment systems are used by the chemical and petrochemical industry. However, as the range of compounds that are found to be biodegraded under anaerobic conditions has increased enormously lately, a large potential expansion seems possible in the future [22]. Thanks to a combination of a simple construction and a high volumetric treatment capacity, the UASB reactor is the dominant concept in the industrial anaerobic wastewater treatment and it probably will keep reigning in the future. Nonetheless, higher loaded expanded granular sludge bed reactors will gradually replace at least part of the UASB applications. In the case of wastewater streams rich in reduced sulfur compounds, the new sulfur biotechnology has allowed the development of reactor systems to remove sulfide producing elemental sulfur. This technology has been adapted for the sweetening of natural gas [30] and more recently for liquefied petroleum gas (LPG), which contains predominantly sulfide and lower alkylthiols [77]. The latter process involves three steps: 1) extraction of the sulfur compounds from the liquefied hydrocarbon phase to a mild
533
carbonate solution in an absorption column; 2) anaerobic conversion of alkylthiols to sulfide and methane in an UASB reactor; and 3) partial oxidation of sulfide into elemental sulfur. Noteworthy, biological processes developed specifically for wastewater treatment will play a key role in the treatment of gas streams from the petroleum industry. Additionally, it is expected that the combination of the biological carbon, nitrogen and sulfur cycles under anaerobic conditions would be a potential technology for the removal of such contaminants in a single step. In conclusion, the application of biological wastewater treatment in the frame of a process integration treatment technology will hopefully close the water cycle allowing the "zero discharge" in the petroleum industry as shown in Fig. 4.
Fig. 4. Schematic representation of the close water cycle in the petroleum industry.
534
REFERENCES [I] [2] [3] [4] [5] [6] [7] [8] [9] [10] II1] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29]
M.R. Beychock, Aqueous Wastes from Petroleum and Petrochemical Plants, John Wiley & Sons, London (1967). F. Berne and J. Cordonnier, Industrial Water Treatment: Refining, Petrochemicals and Gas processing Techniques, Gulf Publishing, Houston (1995). H. Macarie, Water Sci. Technol., 42:5-6 (2000) 201. R.E. Speece, Anaerobic biotechnology for Industrial Wastewaters, Archaea Press, Nashville (1996) 25-68. J.H Tay, Y.X He and Y.G Yan, Water Environ. Res. 72 (2000) 189. P. Olguin-Lora, L. Puig-Grajales and E. Razo-Flores, Environ. Technol., 24 (2003) 999. R. Sierra-Alvarez and G. Lettinga, Appl. Microbiol. Biotech., 34 (1991) 544. B. Donlon, E. Razo-Flores, J. Field and G, Appl. Environ. Microbiol., 61 (1995) 3889. E. Razo-Flores, M. Iniestra-Gonzalez, J.A. Field, P. Olguin-Lora and L. PuigGrajales, J. Environ. Eng., 129 (2003) 999. H. Fang and G. Zhou, Water Sci. Technol., 42:5 (2000) 237. P. Hwang and S. Cheng, Water Sci. Technol., 24:5 (1991) 133. H. Fang, T. Chen, Y. Li and H. Chui, Water Res., 30 (1996) 1353. K. Tawfiki, K. Lepine, J. Bisaillon, R. Beaudet, J. Hawari and S. R Guiot, Biotechnol. Bioeng., 67 (2000) 419. S.R. Guiot, K. Tawfiki-Hajji and F. Lepine, Water Sci. Technol., 42:5-6 (2000) 245. G. Zhou and H. Fang, Bioresource Technol., 61 (1997) 47. R. Kleerebezem, J. Mortier, L.W. Hulshoff-Pol and G. Lettinga, Water Sci. Technol., 36:2-3 (1997) 237. R. Kleerebezem and G. Lettinga, Water Sci. Technol., 42:5-6 (2000) 259. H. Macarie, Ph. D. Thesis, Universite de Provence Marseille, France (in Freeh) 1992. D.O. Mountfort, W.J. Krumholz and M.P. Bryant, Int. J. System. Bacteriol, 134 (1984)216. S. Shelley, Chem. Eng., 98 (1991) 90. S.S. Cheng, C.Y. Ho and J.H. Wu. 8th Int. Conf. On Anaerobic Digestion. Sendai, Japan (1997) R. Kleerebezem and H. Macarie, Chem. Eng., April, (2003) 56. K. Sublette, R.M. Kolhatkar and K. Raterman, Biodegradation, 9 (1998) 259. C. Buisman, R. Post, P. Ijspeert, G. Geraats and G. Lettinga, Acta Biotechnol., 9 (1989)255. A. Janssen, R. Sleylter, C. van der Kaa, J. Jochemsen, J. Bontsema, S. Ma and G. Lettinga, Biotechnol. Bioeng., 47 (1995) 327. J. Visser, L. Robertson, H. Verseveld and J. Kuenen, Appl. Environ. Microbiol., 63 (1997)2300. A. Janssen, S. Ma, P. Lens and G. Lettinga, Biotechnol. Bioeng., 53 (1997) 32. S. Alcantara, A. Velasco, A. Munoz, J. Cid, S. Revah and E. Razo-Flores, Environ. Sci. Technol., 38 (2004) 918. S. Alcantara, I. Estrada, M. Vasquez and S. Revah, Biotechnol. Lett., 21 (1999) 81.
535
[30] B.J. Arena, H.N. Robson, A.L. de Vegt and C.J. Buisman, National Petroleum Refiners Asociation, Annual Meeting (1988). [31] P. Cadenhead and K. Sublette, Biotechnol. Bioeng., 35 (1990) 1150. [32] B. Rajganesh, M. Selvaraj, K. Sublette and C. Camp, Appl. Biochem. Biotechnol., 51/52(1995)735. [33] K. Sublette and M. Woolsey, Biotechnol. Bioeng., 34 (1989) 565. [34] E. W. Kim and J. H. Bae, Water Sci. Technol, 42:3-4 (2000) 233. [35] B. Krishnakumar and V. B. Manilal, Biotechnol. Lett., 21 (1999) 437. [36] J. Reyes-Avila, E. Razo-Flores and J. Gomez, Water Res., (2004) submitted. [37] F. Fdez-Polanco, M. Fdez-Polanco, N. Fernandez, M. Uruena, P. Garcia and S. Villaverde, Water Sci. Technol., 44: 4 (2001) 15. [38] M.R. Betlach and J.M. Tiedje, Appl. Environ. Microbiol., 42 (1981) 1074. [39] P.J. Gommers, W. Buleveld, F.J. Zuiderwijk and J.G. Kuenen, Water Res. 22 (1988) 1075. [40] R. Johnson, J. Pankow, D. Bender D., C. Price and J. Zogorsky, Environ. Sci. Technol., 32 (2000) 210. [41] J. Brown, S. Bay, D. Greenstein and W. Ray, Report Southern California Coastal Water Research Project. www.sccwrp.org/pubs/annrpt/99_00/abstl I_ar34.htm [42] C.Achten, A. Kolb and W. Puttmann, Environ. Sci.Technol, 36 (2002) 3652. [43] S. Fiorenza and H. Rifai, Bioremediation Journal, 7 (2003) 1. [44] G. Reyna, E. Vega, E. Reyes, V. Mugica, V. Chow, J. Watson and J. Arriaga. 94th Annual Conference and Exhibition of the Air and Waste Management Association. Orlando, FL (2001). [45] L. Manzanares, L. Mufioz, C. Romero, V. Nevarez, E. Ramirez, M. Delgado and A. Keer. 94th Annual Conference and Exhibition of the Air and Waste Management Association. Orlando, FL (2001). [46] R. Iturbe, R. Flores and L.Torres, Water Air Soil Poll., 14 (2003) 261. [47] C. Buenrostro and A. Dovali. DGCH. (2001) Study for determining and evaluating methyl tert-buryl ether (MTBE). Oficio GDF-SOS/01-523. In Spanish. [48] USEPA 1997, EPA822-F-97-009, Office of Water, 34. [49] F. Fayolle, J. Vandecasteele and F. Monot, Appl. Microbiol. Biotechnol., 56 (2001) 339. [50] N. Fortin, M. Morales, Y. Nakagawa, D. Focht and M. Deshusses, Environ. Microbiol., 3(2001)407. [51] J. Salanitro, Curr. Op. Biotechnol., 6 (1995) 337. [52] J. Wilson, J. Cho, B. Wilson and J. Vardy. U.S. Environmental Protection Agency. 2000. Natural attenuation of MTBE in the subsurface under methanogenic conditions. EPA/600/R-00/006. U.S. EPA Office of Research and Development: Washington, D.C. [53] P. Bradley, F. Chapelle and J. Landmeyer, Appl. Environ. Microbiol., 67 (2001) 1975. [54] K. Finneran and D. Lovley, Environ. Sci. Technol., 35 (2001) 1785. [55] R. Steffan, S. Vainberg, C. Condee, K. McClay and P. Hatzinger. In G. Wrickramayake, A. Gavaskar, B. Alleman, V. Magar (eds.) Bioremediation and phytoremediation of chlorinated and recalcitrant compounds. Batelle, Columbus, OH (2000) 165.
536
[56] M. Hyman and K. O'Reilly. In: Alleman B, Lesson A (eds.) In situ bioremediation of petroleum hydrocarbon and other organic compounds. Battelle, Columbus, OH. (1999)7. [57] R. Steffan, K. Me Clay, S. Vainberg, C. Condee and D. Zhang, Appl. Environ. MicrobioL, 63(1997)4216. [58] R. Deeb, K. Scow and L. Alvarez, Biodegradation, 11 (2000) 171. [59] R. Prince, Crit. Rev. MicrobioL, 26 (2002) 163. [60] USEPA 2003. MTBE treatment case study website, http://clu-in.org/products/mtbe/ December 2003. [61] B. Wilson and J. Wilson, Contaminated Soil Sediment and Water (2002) 47. [62] J. Salanitro, P. Johnson, G. Spinnler, P. Maner, H. Wisniewski and C. Bruce, Environ. Sci. TechnoL, 34 (2000) 4152. [63] R. Wilson R., K. Scow and D. Mackay, Environ. Sci. TechnoL, 36 (2001) 190. [64] K. Hrystova, B. Gebreyesus, D. Mackay and K. Scow, Appl.Environ. MicrobioL, 69(2003)2616. [65] R. Steffan, P. Hatzinger, Y. Farhan and S. Drew. NGWA/API Conference on Petroleum Hydrocarbons and Organic Chemicals in groundwater: Prevention, detection and Remediation. Westville, OH (2001). [66] J. Landmeyer, F. Chapelle, H. Herlong and P. Bradley, Environ. Sci. TechnoL, 35 (2001) 1118. [67] R. Steffan, J. Johnson and S. Drew. In Sublette K (eds.) Proceedings of the 7th International Petroleum Environmental Conference. IPEC, Albuquerque, NM (2000) 1-12. [68] J. Morrison, M. Suidan and A. Venosa, J. Environ. Eng., 128 (2002) 836. [69] W. Stringfellow and K. Oh, J. Environ. Eng., 128 (2002) 852. [70] S. Vainberg, A. Togna, P. Sutton and R. Steffan, J. Environ. Eng., 128 (2002) 852. [71] J. O'Connell and S. Zigan. In, E. Moyer, P. Kostecki (eds.) MTBE remediation Handbook. Amherst Scientific Publishers. Amherst Massachusets. (2003) [72] N. Fortin and M. Deshusses, Environ. Sci. TechnoL, 33 (1999) 2980. [73] J. Eweis, J. Scarano, B. Converse, D. Chang and E. Schroeder. Report American Petroleum Institute Contract # 97000-2577(1999). [74] D. Dupasquier, S. Revah and R. Auria, Environ.Sci.TechnoL, 36 (2002) 247. [75] W.W. Eckenfelder Jr. and A.J. Englande Jr., Water Sci. TechnoL, 34:10 (1996) 1. [76] R.F. Dunn and M.M. El-Halwagi, J. Chem. TechnoL BiotechnoL, 78 (2003) 1011. [77] J. Sipma, A.H. Janssen, L.W. Hulshoff Pol and G. Lettinga, BiotechnoL Bioeng., 82(2003)1.
537 STUDIES IN SURFACE SCIENCE AND CATALYSIS Advisory Editors: B. Delmon, Universite Catholique de Louvain, Louvain-la-Neuve, Belgium J.T. Yates, University of Pittsburgh, Pittsburgh, PA, U.S.A.
Volume 1
Volume 2
Volume 3
Volume 4
Volume 5
Volume 6
Volume 7
Volume 8 Volume 9
Volume 1 0
Volume 11
Volume 12
Volume 13 Volume 14
Volume 15
Preparation of Catalysts I. Scientific Bases for the Preparation of Heterogeneous Catalysts. Proceedings of the First International Symposium, Brussels, October 14-17,1975 edited by B. Delmon, P.A. Jacobs and G. Poncelet The Control of the Reactivity of Solids. A Critical Survey of the Factors that Influence the Reactivity of Solids, with Special Emphasis on the Control of the Chemical Processes in Relation to Practical Applications by V.V. Boldyrev, M. Bulens and B. Delmon Preparation of Catalysts ll.Scientific Bases for the Preparation of Heterogeneous Catalysts. Proceedings of the Second International Symposium, Louvain-laNeuve, September 4 - 7 , 1978 edited by B. Delmon, P. Grange, P. Jacobs and G. Poncelet Growth and Properties of Metal Clusters. Applications to Catalysis and the Photographic Process. Proceedings of the 32 nd International Meeting of the Societe de Chimie Physique, Villeurbanne, September 2 4 - 2 8 , 1979 edited by J. Bourdon Catalysis by Zeolites.Proceedings of an International Symposium, Ecully (Lyon), September 9 - 1 1 , 1980 edited by B. Imelik, C. Naccache,Y. BenTaarit, J.C.Vedrine, G. Coudurier and H. Praliaud Catalyst Deactivation. Proceedings of an International Symposium, Antwerp, October 1 3 - 15,1980 edited by B. Delmon and G.F. Froment New Horizons in Catalysis. Proceedings of the 7 th International Congress on Catalysis, Tokyo, June 30-July 4, 1980. Parts A and B edited by T. Seiyama and K.Tanabe Catalysis by Supported Complexes by Yu.l.Yermakov, B.N. Kuznetsov and V.A. Zakharov Physics of Solid Surfaces. Proceedings of a Symposium, Bechyne, September 29-October 3,1980 edited by M. Laznicka Adsorption at the Gas-Solid and Liquid-Solid Interface. Proceedings of an International Symposium, Aix-en-Provence, September 2 1 - 2 3 , 1981 edited by J. Rouquerol and K.S.W. Sing Metal-Support and Metal-Additive Effects in Catalysis. Proceedings of an International Symposium, Ecully (Lyon), September 1 4 - 1 6 , 1982 edited by B. Imelik, C. Naccache, G. Coudurier, H. Praliaud, P. Meriaudeau, P. Gallezot, G.A.Martin and J.C.Vedrine Metal Microstructures in Zeolites. Preparation - Properties - Applications. Proceedings of a Workshop, Bremen, September 2 2 - 2 4 , 1982 edited by P.A. Jacobs, N.I. Jaeger, P. Jirii and G. Schulz-Ekloff Adsorption on Metal Surfaces. An Integrated Approach edited by J. Benard Vibrations at Surfaces. Proceedings of the Third International Conference, Asilomar, CA, September 1-4, 1982 edited by C.R. Brundle and H.Morawitz Heterogeneous Catalytic Reactions Involving Molecular Oxygen by G.I. Golodets
538
Volume 1 6
Volume 17
Volume 1 8
Volume 19
Volume 20
Volume 21
Volume 22 Volume 23 Volume 24
Volume 25
Volume 26
Volume 27 Volume 28
Volume 29 Volume 30
Volume 31
Volume 32 Volume 33 Volume 34
Volume 35
Preparation of Catalysts III. Scientific Bases for the Preparation of Heterogeneous Catalysts. Proceedings of the Third International Symposium, Louvain-la-Neuve, September 6 - 9 , 1982 edited by G. Poncelet, P. Grange and P.A. Jacobs Spillover of Adsorbed Species. Proceedings of an International Symposium, Lyon-Villeurbanne, September 12-16, 1983 edited by G.M. Pajonk,S.J.Teichner and J.E. Germain Structure and Reactivity of Modified Zeolites. Proceedings of an International Conference, Prague, July 9 - 1 3 , 1984 edited by P.A. Jacobs, N.I. Jaeger, P. Jiru, V.B. Kazansky and G. Schulz-Ekloff Catalysis on the Energy Scene. Proceedings of the 9 th Canadian Symposium on Catalysis, Quebec, P.Q., September 30-October 3, 1984 edited by S. Kaliaguine and A.Mahay Catalysis by Acids and Bases. Proceedings of an International Symposium, Villeurbanne (Lyonl, September 2 5 - 2 7 , 1984 edited by B. Imelik, C. Naccache, G. Coudurier.Y. Ben Taarit and J.C.Vedrine Adsorption and Catalysis on Oxide Surfaces. Proceedings of a Symposium, Uxbridge, June 2 8 - 2 9 , 1984 edited by M. Che and G.C.Bond Unsteady Processes in Catalytic Reactors by Yu.Sh. Matros Physics of Solid Surfaces I984 edited by J.Koukal Zeolites:Synthesis, Structurejechnology and Application. Proceedings of an International Symposium, Portoroz-Portorose, September 3 - 8 , 1984 edited by B.Drzaj.S. Hocevar and S. Pejovnik Catalytic Polymerization of Olefins. Proceedings of the International Symposium on Future Aspects of Olefin Polymerization, Tokyo, July 4 - 6 , 1985 edited by T.Keii and K.Soga Vibrations at Surfaces 1985. Proceedings of the Fourth International Conference, Bowness-on-Windermere, September 15-19, 1985 edited by D.A. King, N.V. Richardson and S. Holloway Catalytic Hydrogenation edited by L. Cerveny New Developments in Zeolite Science and Technology. Proceedings of t h e 7th International Zeolite Conference, Tokyo, August 17-22, 1986 edited by Y.Murakami,A. lijima and J.W.Ward Metal Clusters in Catalysis edited by B.C. Gates, L. Guczi and H. Knozinger Catalysis and Automotive Pollution Control. Proceedings of t h e First International Symposium, Brussels, September 8 - 1 1 , 1986 edited by A. Crucq and A. Frennet Preparation of Catalysts IV. Scientific Bases for the Preparation of Heterogeneous Catalysts. Proceedings of the Fourth International Symposium, Louvain-laNeuve, September 1-4, 1986 edited by B. Delmon, P. Grange, P.A. Jacobs and G. Poncelet Thin Metal Films and Gas Chemisorption edited by P.Wissmann Synthesis of High-silica Aluminosilicate Zeolites edited by P.A. Jacobs and J.A.Martens Catalyst Deactivation 1987. Proceedings of the 4 t h International Symposium, Antwerp, September 29-October 1, 1987 edited by B. Delmon and G.F. Froment Keynotes in Energy-Related Catalysis edited by S. Kaliaguine
539
Volume 36
Volume 37
Volume 38
Volume 39
Volume 40
Volume 41
Volume 42 Volume 43 Volume 44
Volume 45 Volume 46
Volume 47 Volume 48
Volume 49
Volume 50
Volume 51 Volume 52
Volume 53
Volume 54
Methane Conversion. Proceedings of a Symposium on the Production of Fuels and Chemicals from Natural Gas, Auckland, April 2 7 - 3 0 , 1987 edited by D.M. Bibby, C.D.Chang, R.F.Howe and S.Yurchak Innovation in Zeolite Materials Science. Proceedings of an International Symposium, Nieuwpoort, September 1 3 - 1 7 , 1987 edited by P.J. Grobet, W.J.Mortier, E.F.Vansant and G. Schulz-Ekloff Catalysis l987.Proceedings of the 10 th North American Meeting of the Catalysis Society, San Diego, CA, May 17-22, 1987 edited by J.W.Ward Characterization of Porous Solids. Proceedings of the IUPAC Symposium (COPS I), Bad Soden a. Ts., April 26-29,1987 edited by K.K.Unger, j . Rouquerol, K.S.W.Sing and H. Krai Physics of Solid Surfaces I987. Proceedings of the Fourth Symposium on Surface Physics, Bechyne Castle, September 7 - 1 1 , 1987 edited by J.Koukal Heterogeneous Catalysis and Fine Chemicals. Proceedings of an International Symposium, Poitiers, March 1 5 - 1 7 , 1988 edited by M. Guisnet, |. Barrault, C. Bouchoule.D. Duprez, C. Montassier and G. Perot Laboratory Studies of Heterogeneous Catalytic Processes by E.G. Christoffel, revised and edited by Z. Paal Catalytic Processes under Unsteady-State Conditions by Yu. Sh. Matros Successful Design of Catalysts. Future Requirements and Development. Proceedings of the Worldwide Catalysis Seminars, July, 1 988, on the Occasion of the 30 th Anniversary of the Catalysis Society of Japan edited by T. Inui Transition Metal Oxides. Surface Chemistry and Catalysis by H.H.Kung Zeolites as Catalysts, Sorbents and Detergent Builders. Applications and Innovations. Proceedings of an International Symposium, Wurzburg, September 4 - 8 , 1 9 8 8 edited by H.G. Karge and j.Weitkamp Photochemistry on Solid Surfaces edited by M.Anpo and T. Matsuura Structure and Reactivity of Surfaces. Proceedings of a European Conference, Trieste, September 13-16, 1988 edited by C.Morterra.A. Zecchina and G. Costa Zeolites: Facts, Figures, Future. Proceedings of the 8th International Zeolite Conference, Amsterdam, July 10-14, 1989. Parts A and B edited by P.A. Jacobs and R.A. van Santen Hydrotreating Catalysts. Preparation, Characterization and Performance. Proceedings of the Annual International AlChE Meeting, Washington, DC, November 27-December 2, 1988 edited by M.L. Occelli and R.G.Anthony New Solid Acids and Bases.Their Catalytic Properties by KJanabe, M. Misono.Y. Ono and H. Hattori Recent Advances in Zeolite Science. Proceedings of the 1989 Meeting of the British Zeolite Association, Cambridge, April 1 7 - 1 9 , 1989 edited by J. Klinowsky and P.j. Barrie Catalyst in Petroleum Refining 1989. Proceedings of the First International Conference on Catalysts in Petroleum Refining, Kuwait, March 5 - 8 , 1989 edited by D.LTrimm.S.Akashah, M.Absi-Halabi and A. Bishara Future Opportunities in Catalytic and Separation Technology edited by M. Misono,Y.Moro-oka and S. Kimura
540
Volume 55
Volume 56
Volume 5 7A
Volume 57B
Volume 58 Volume 59
Volume 60
Volume 61
Volume 62
Volume 63
Volume 64 Volume 65
Volume 66
Volume 67
Volume 68
Volume 69
Volume 70
New Developments in Selective Oxidation. Proceedings of an International Symposium, Rimini, Italy, September 18-22, 1989 edited by G. Centi and F.Trifiro Olefin Polymerization Catalysts. Proceedings of the International Symposium on Recent Developments in Olefin Polymerization Catalysts, Tokyo, October 2 3 - 2 5 , 1989 edited by T.Keii and K.Soga Spectroscopic Analysis of Heterogeneous Catalysts. Part A: Methods of Surface Analysis edited by J.L.G. Fierro Spectroscopic Analysis of Heterogeneous Catalysts. Part B: Chemisorption of Probe Molecules edited by J.L.G. Fierro Introduction to Zeolite Science and Practice edited by H. van Bekkum, E.M. Flanigen and ].C. Jansen Heterogeneous Catalysis and Fine Chemicals II. Proceedings of t h e 2 n d International Symposium, Poitiers, October 2 - 6 , 1 990 edited by M. Guisnet, J. Barrault, C. Bouchoule.D. Duprez, G. Perot, R.Maurel and C. Montassier Chemistry of Microporous Crystals. Proceedings of the International Symposium on Chemistry of Microporous Crystals, Tokyo, June 2 6 - 2 9 , 1990 edited by T. Inui.S. Namba and T.Tatsumi Natural Gas Conversion. Proceedings of the Symposium on Natural Gas Conversion, Oslo, August 12-17, 1990 edited by A. Holmen, K.-j. Jens and S.Kolboe Characterization of Porous Solids II. Proceedings of t h e IUPAC Symposium (COPS II), Alicante, May 6 - 9 , 1990 edited by F. Rodriguez-Reinoso, j . Rouquerol, K.S.W. Sing and K.K.Unger Preparation of Catalysts V. Scientific Bases for the Preparation of Heterogeneous Catalysts. Proceedings of the Fifth International Symposium, Louvain-la-Neuve, September 3 - 6 , 1990 edited by G. Poncelet, P.A. Jacobs, P. Grange and B. Delmon New Trends in CO Activation edited by L. Guczi Catalysis and Adsorption by Zeolites. Proceedings of ZEOCAT 9 0 , Leipzig, August 20-23, 1990 edited by G. Ohlmann, H. Pfeifer and R. Fricke Dioxygen Activation and Homogeneous Catalytic Oxidation. Proceedings of t h e Fourth International Symposium on Dioxygen Activation and Homogeneous Catalytic Oxidation, Balatonf tired, September 10-14, 1990 edited by L.I. Simandi Structure-Activity and Selectivity Relationships in Heterogeneous Catalysis. Proceedings of the ACS Symposium on Structure-Activity Relationships in Heterogeneous Catalysis, Boston, MA, April 22-27', 1990 edited by R.K. Grasselli and A.W.SIeight Catalyst Deactivation 1991. Proceedings of the Fifth International Symposium, Evanston, IL, June 2 4 - 2 6 , 1991 edited by C.H. Bartholomew and J.B. Butt Zeolite Chemistry and Catalysis. Proceedings of an International Symposium, Prague, Czechoslovakia, September 8 - 1 3 , 1991 edited by P.A. Jacobs, N.I. Jaeger, L.Kubelkova and B.Wichterlova Poisoning and Promotion in Catalysis based on Surface Science Concepts and Experiments by M. Kiskinova
541
Volume 71
Volume 72
Volume 73
Volume 74 Volume 75
Volume 76 Volume 77
Volume 78
Volume 79
Volume 80
Volume 81
Volume 82
Volume 83
Volume 84
Volume 85 Volume 86 Volume 87
Catalysis and Automotive Pollution Control II. Proceedings of t h e 2 n d International Symposium (CAPoC 2), Brussels, Belgium, September 10-13, 1990 edited by A. Crucq New Developments in Selective Oxidation by Heterogeneous Catalysis. Proceedings of the 3rd European Workshop Meeting on New Developments in Selective Oxidation by Heterogeneous Catalysis, Louvain-la-Neuve, Belgium, April 8 - 1 0 , 1991 edited by P. Ruiz and B. Delmon Progress in Catalysis. Proceedings of the 12th Canadian Symposium on Catalysis, Banff, Alberta, Canada, May 2 5 - 2 8 , 1992 edited by K.J. Smith and EX. Sanford Angle-Resolved Photoemission.Theory and Current Applications edited by S.D. Kevan New Frontiers in Catalysis, Parts A-C. Proceedings of the 10 th International Congress on Catalysis, Budapest, Hungary, 19-24 July, 1992 edited by L. Guczi, F. Solymosi and P.Tetenyi Fluid Catalytic Cracking: Science and Technology edited by J.S.Magee and M.M. Mitchell, Jr. New Aspects of Spillover Effect in Catalysis. For Development of Highly Active Catalysts. Proceedings of the Third International Conference on Spillover, Kyoto, Japan, August 17-20, 1993 edited by T. Inui, K. Fujimoto.T.Uchijima and M. Masai Heterogeneous Catalysis and Fine Chemicals III. Proceedings of the 3rd International Symposium, Poitiers, April 5 - 8, 1993 edited by M. Guisnet, J. Barbier, J. Barrault, C. Bouchoule.D. Duprez, G. Perot and C. Montassier Catalysis: An Integrated Approach to Homogeneous, Heterogeneous and Industrial Catalysis edited by J.A. Moulijn, P.W.N.M. van Leeuwen and R.A. van Santen Fundamentals of Adsorption. Proceedings of the Fourth International Conference on Fundamentals of Adsorption, Kyoto, Japan, May 17-22, 1992 edited by M. Suzuki Natural Gas Conversion II. Proceedings of the Third Natural Gas Conversion Symposium, Sydney, July 4 - 9 , 1993 edited by H.E. Curry-Hyde and R.F.Howe New Developments in Selective Oxidation II. Proceedings of t h e Second World Congress and Fourth European Workshop Meeting, Benalmadena, Spain, September 2 0 - 2 4 , 1993 edited by V. Cortes Corberan and S.Vic Bellon Zeolites and Microporous Crystals. Proceedings of the International Symposium on Zeolites and Microporous Crystals, Nagoya, Japan, August 2 2 - 2 5 , 1993 edited byT. Hattori and T.Yashima Zeolites and Related Microporous Materials: State of the Art I994. Proceedings of the 10th International Zeolite Conference, Garmisch-Partenkirchen, Germany, July 17-22, 1994 edited by J.Weitkamp, H.G. Karge.H. Pfeifer and W. Holderich Advanced Zeolite Science and Applications edited by J.C. jansen, M. Stocker, H.G. Karge and J.Weitkamp Oscillating Heterogeneous Catalytic Systems by M.M. Slinko and N.I. Jaeger Characterization of Porous Solids III. Proceedings of t h e IUPAC Symposium (COPS III), Marseille, France, May 9 - 1 2 , 1993 edited by j.Rouquerol, F. Rodriguez-Reinoso, K.S.W. Sing and K.K.Unger
542
Volume 88
Volume 89
Volume 90
Volume 91
Volume 92
Volume 93 Volume 94
Volume 95 Volume 96
Volume 97
Volume 98
Volume 99 Volume 100
Volume 101
Volume 102 Volume 1 03 Volume 104 Volume 105
Catalyst Deactivation 1994. Proceedings of the 6 th International Symposium, Ostend, Belgium, October 3 - 5 , 1994 edited by B. Delmon and G.F. Froment Catalyst Design for Tailor-made Polyolefins. Proceedings of t h e International Symposium on Catalyst Design for Tailor-made Polyolefins, Kanazawa, Japan, March 10-12, 1994 edited by K.Soga and M.Terano Acid-Base Catalysis II. Proceedings of the International Symposium on Acid-Base Catalysis II, Sapporo, Japan, December 2 - 4 , 1993 edited by H. Hattori.M. Misono and Y.Ono Preparation of Catalysts VI. Scientific Bases for the Preparation of Heterogeneous Catalysts. Proceedings of the Sixth International Symposium, Louvain-La-Neuve, September 5-8, 1994 edited by G. Poncelet, J.Martens.B. Delmon, P.A. Jacobs and P. Grange Science and Technology in Catalysis I994. Proceedings of t h e Second Tokyo Conference on Advanced Catalytic Science and Technology, Tokyo, August 2 1 - 2 6 , 1994 edited by Y. Izumi, H.Arai and M. Iwamoto Characterization and Chemical Modification of the Silica Surface by E.F.Vansant, P.Van Der Voort and K.C.Vrancken Catalysis by Microporous Materials. Proceedings of ZEOCAT'95, Szombathely, Hungary, July 9-13, 1995 edited by H.K. Beyer, H.G.Karge, i. Kiricsi and J.B.Nagy Catalysis by Metals and Alloys by V. Ponec and G.C.Bond Catalysis and Automotive Pollution Control III. Proceedings of the Third International Symposium (CAPoC3), Brussels, Belgium, April 2 0 - 2 2 , 1994 edited by A. Frennet and J.-M. Bastin Zeolites:A Refined Tool for Designing Catalytic Sites. Proceedings of the International Symposium, Quebec, Canada, October 15-20, 1995 edited by L. Bonneviot and S. Kaliaguine Zeolite Science 1994: Recent Progress and Discussions. Supplementary Materials t o t h e 10th International Zeolite Conference, Garmisch-Partenkirchen, Germany, July 17-22, 1994 edited by H.G. Karge and J.Weitkamp Adsorption on New and Modified Inorganic Sorbents edited by A.Dabrowski and V.A.Tertykh Catalysts in Petroleum Refining and Petrochemical Industries 1995. Proceedings of the 2nd International Conference on Catalysts in Petroleum Refining and Petrochemical Industries, Kuwait, April 22-26, 1995 edited by M.Absi-Halabi, J. Beshara, H. Qabazard and A. Stanislaus II th International Congress on Catalysis -40 t h Anniversary. Proceedings of the 11 t h ICC, Baltimore, MD, USA, June 30-July 5, 1996 edited by J.W. High tower, W.N. Delgass, E. Iglesia and A.T. Bell Recent Advances and New Horizons in Zeolite Science and Technology edited by H. Chon.S.I.Woo and S. -E. Park Semiconductor Nanoclusters - Physical, Chemical, and Catalytic Aspects edited by P.V. Kamat and D. Meisel Equilibria and Dynamics of Gas Adsorption on Heterogeneous Solid Surfaces edited by W. Rudzihski.W.A. Steele and G. Zgrablich Progress in Zeolite and Microporous Materials Proceedings of the 11 t h International Zeolite Conference, Seoul, Korea, August 12-17, 1996 edited by H. Chon.S.-K. Ihm and Y.S.Uh
543
Volume 1 06
Volume 1 07
Volume 108
Volume 109
Volume 110
Volume 111
Hydrotreatment and Hydrocracking of Oil Fractions Proceedings of the 1 s t International Symposium / 6 th European Workshop, Oostende, Belgium, February 17-19, 1997 e d i t e d b y G.F. Froment.B. Delmon and P. Grange Natural Gas Conversion IV Proceedings of the 4 th International Natural Gas Conversion Symposium, Kruger Park, South Africa, November 19-23, 1995 e d i t e d b y M. de Pontes, R.L. Espinoza, C.P. Nicolaides, J.H. Scholtz and M.S. Scurrell Heterogeneous Catalysis and Fine Chemicals IV Proceedings of the 4 th International Symposium on Heterogeneous Catalysis and Fine Chemicals, Basel, Switzerland, September 8-12, 1996 e d i t e d by H.U. Blaser, A. Baiker and R. Prins Dynamics of Surfaces and Reaction Kinetics in Heterogeneous Catalysis. Proceedings of the International Symposium, Antwerp, Belgium, September 15-17, 1997
edited by G.F. Froment and K.C.Waugh Third World Congress on Oxidation Catalysis. Proceedings of the Third World Congress on Oxidation Catalysis, San Diego, CA, U.S.A., 21-26 September 1997 e d i t e d b y R.K. Grasselli,S.T.Oyama, A.M. Gaffney and J.E. Lyons Catalyst Deactivation I997. Proceedings of the 7th International Symposium, Cancun, Mexico, October 5-8,
1997 Volume 11 2
Volume 113
Volume 114
Volume 11 5
Volume 116
Volume 11 7
Volume 118
Volume 11 9
Volume 12 0 A
e d i t e d b y C.H. Bartholomew and G.A. Fuentes Spillover and Migration of Surface Species on Catalysts. Proceedings of the 4 th International Conference on Spillover, Dalian, China, September 15-18, 1997 e d i t e d b y Can Li and Qin Xin Recent Advances in Basic and Applied Aspects of Industrial Catalysis. Proceedings of the 13th National Symposium and Silver Jubilee Symposium of Catalysis of India, Dehradun, India, April 2-4, 1997 e d i t e d b y T.S.R. Prasada Rao and G.Murali Dhar Advances in Chemical Conversions for Mitigating Carbon Dioxide. Proceedings of the 4 th International Conference on Carbon Dioxide Utilization, Kyoto, Japan, September 7-11, 1997 e d i t e d b y T. Inui, M.Anpo.K. liui.S.Yanagida and T.Yamaguchi Methods for Monitoring and Diagnosing the Efficiency of Catalytic Converters. A patent-oriented survey by M. Sideris Catalysis and Automotive Pollution Control IV. Proceedings of the 4 th International Symposium (CAPoC4), Brussels, Belgium, April 9-11, 1997 e d i t e d b y N. Kruse, A. Frennet and J.-M. Bastin Mesoporous Molecular Sieves 1998 Proceedings of the 1 s t International Symposium, Baltimore, MD, U.S.A., July 10-12, 1998 e d i t e d b y L.Bonneviot, F. Beland, C.Danumah, S. Giasson and S. Kaliaguine Preparation of Catalysts VII Proceedings of the 7 th International Symposium on Scientific Bases for the Preparation of Heterogeneous Catalysts, Louvain-la-Neuve, Belgium, September 1-4, 1998 e d i t e d b y B. Delmon, P.A. Jacobs, R. Maggi, J.A.Martens, P. Grange and G. Poncelet Natural Gas Conversion V Proceedings of the 5th International Gas Conversion Symposium, Giardini-Naxos, Taormina, Italy, September 20-25, 1998 e d i t e d b y A. Parmaliana, D. Sanfilippo, F. Frusteri, A.Vaccari and F.Arena Adsorption and its Applications in Industry and Environmental Protection. Vol I: Applications in Industry e d i t e d b y A. Dijbrowski
544
Volume 1 2 0 B
Volume 121
Volume 122
Adsorption and its Applications in Industry and Environmental Protection. Vol II: Applications in Environmental Protection edited b y A. Dqbrowski Science and Technology in Catalysis 1998 Proceedings of the Third Tokyo Conference in Advanced Catalytic Science and Technology, Tokyo, July 19-24, 1998 edited b y H. Hattori and K. Otsuka Reaction Kinetics and the Development of Catalytic Processes Proceedings of the International Symposium, Brugge, Belgium, April 19-21,
1999 Volume 1 23
Volume 1 24 Volume 125
Volume 1 26
Volume 127
Volume 1 28
Volume 129
Volume 1 30
Volume 131
Volume 1 32
Volume 1 33
Volume 134
edited b y G.F. Froment and K.C.Waugh Catalysis:An Integrated Approach Second, Revised and Enlarged Edition edited b y R.A. van Santen, P.W.N.M. van Leeuwen, J.A. Moulijn and BAAveriil Experiments in Catalytic Reaction Engineering by J.M. Berty Porous Materials in Environmentally Friendly Processes Proceedings of the 1 s t International FEZA Conference, Eger, Hungary, September 1-4, 1999 edited b y I. Kiricsi, G. Pal-Borbely, J.B.Nagy and H.G. Karge Catalyst Deactivation 1999 Proceedings of the 8th International Symposium, Brugge, Belgium, October 10-13, 1999 edited b y B. Delmon and G.F. Froment Hydrotreatment and Hydrocracking of Oil Fractions Proceedings of the 2nd International Symposium/7th European Workshop, Antwerpen, Belgium, November 14-17, 1999 edited b y B. Delmon, G.F. Froment and P. Grange Characterisation of Porous Solids V Proceedings of the 5th International Symposium on the Characterisation of Porous Solids (COPS-V), Heidelberg, Germany, May 30- June 2, 1999 edited b y K.K.Unger,G.Kreysa and J.P. Baselt Nanoporous Materials II Proceedings of the 2nd Conference on Access in Nanoporous Materials, Banff, Alberta, Canada, May 25-30, 2000 edited byA. Sayari.M. jaroniec and T.J. Pinnavaia 12 th International Congress on Catalysis Proceedings of the 12 t h ICC, Granada, Spain, July 9-14, 2000 edited byA. Corma, F.V. Melo.S. Mendioroz and J.L.G. Fierro Catalytic Polymerization of Cycloolefins Ionic, Ziegler-Natta and Ring-Opening Metathesis Polymerization By V. Dragutan and R. Streck Proceedings of the International Conference on Colloid and Surface Science, Tokyo, Japan, November 5-8,2000 25 th Anniversary of the Division of Colloid and Surface Chemistry, The Chemical Society of Japan edited b y Y. Iwasawa, N.Oyama and H.Kunieda Reaction Kinetics and the Development and Operation of Catalytic Processes Proceedings of the 3rd International Symposium, Oostende, Belgium, April 2225, 2001 edited b y G.F. Froment and K.C.Waugh Fluid Catalytic Cracking V Materials and Technological Innovations edited b y M.L. Occelli and P. O'Connor
545
Volume 135
Volume 136
Volume 137
Volume 138 Volume 139
Volume 140
Volume 141
Volume 142
Volume 143
Volume 144
Volume 145
Volume 146
Volume 147
Volume 148
Volume 149
Volume 150
Zeolites and Mesoporous Materials at the Dawn of the 21 st Century. Proceedings of the 13lh International Zeolite Conference, Montpellier, France, 8-13 July 2001 edited by A. Galameau, F. di Renso, F. Fajula ans J. Vedrine Natural Gas Conversion VI Proceedings of the 6th Natural Gas Conversion Symposium, June 17-22, 2001, Alaska, USA. edited by J J . Spivey, E. Iglesia and T.H. Fleisch Introduction to Zeolite Science and Practice. 2nd completely revised and expanded edition edited by H. van Bekkum, E.M. Flanigen, P.A. Jacobs and J.C. Jansen Spillover and Mobility of Species on Solid Surfaces edited by A. Guerrero-Ruiz and I. Rodriquez-Ramos Catalyst Deactivation 2001 Proceedings of the 9th International Symposium, Lexington, KY, USA, October 2001 edited by J.J. Spivey, G.W. Roberts and B.H. Davis Oxide-based Systems at the Crossroads of Chemistry. Second International Workshop, October 8-11, 2000, Como, Italy. Edited by A. Gamba, C. Colella and S. Coluccia Nanoporous Materials III Proceedings of the 3rd International Symposium on Nanoporous Materials, Ottawa, Ontario, Canada, June 12-15, 2002 edited by A. Sayari and M. Jaroniec Impact of Zeolites and Other Porous Materials on the New Technologies at the Beginning of the New Millennium Proceedings of the 2nd International FEZA (Federation of the European Zeolite Associations) Conference, Taormina, Italy, September 1-5, 2002 edited by R. Aiello, G. Giordano and F.Testa Scientific Bases for the Preparation of Heterogeneous Catalysts Proceedings of the 8th International Symposium, Louvain-la-Neuve, Leuven, Belgium, September 9-12, 2002 edited by E. Gaigneaux, D.E. De Vos, P. Grange, P.A. Jacobs, J.A. Martens, P. Ruiz and G. Poncelet Characterization of Porous Solids VI Proceedings of the 6lh International Symposium on the Characterization of Porous Solids (COPS-VI), Alicante, Spain, May 8-11, 2002 edited by F. Rodriguez-Reinoso, B. McEnaney, J. Rouquerol and K. Unger Science and Technology in Catalysis 2002 Proceedings of the Fourth Tokyo Conference on Advanced Catalytic Science and Technology, Tokyo, July 14-19, 2002 edited by M. Anpo, M. Onaka and H. Yamashita Nanotechnology in Mesostructured Materials Proceedings of the 3rd International Mesostructured Materials Symposium, Jeju, Korea, July 8-11, 2002 edited by Sang-Eon Park, Ryong Ryoo, Wha-Seung Ahn, Chul Wee Lee and Jong-San Chang Natural Gas Conversion VII Proceedings of the 7lh Natural Gas Conversion Symposium, Dalian, China, June 6-10, 2004 edited by X. Bao and Y. Xu Mesoporous Crystals and Related Nano-Structured Materials Proceedings of the Meeting on Mesoporous Crystals and Related Nano-Structured Materials, Stockholm, Sweden, 1-5 June, 2004 edited by O. Terasaki Fluid Catalytic Cracking VI: Preparation and Characterization of Catalysts Proceedings of the 6th International Symposium on Advances in Fluid Cracking Catalysts (FCCs), New York, September 7 - 1 1 , 2003 Edited by M. Occelli Coal and Coal-Related Compounds Structures, Reactivity and Catalytic Reactions edited by T. Kabe, A. Ishihara, W. Qian, I.P. Sturisna and Y. Kabe
This page is intentionally left blank
