Persistent organic pollutants, Volume 3

Introduction Persistent Organic Pollutants are carbon-based chemicals that exhibit characteristics such as, for example...

Author: Heidelore Fiedler | V. Borja-Aburto

97 downloads 1681 Views 4MB Size Report

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!

Report copyright / DMCA form

DOWNLOAD PDF

Introduction

Persistent Organic Pollutants are carbon-based chemicals that exhibit characteristics such as, for example, that they do not break down under environmental conditions, are semi-volatile, have low solubility in water, and have an inherent toxicity. The combination of these chemical and physical properties results in long-range transport and in bioaccumulation of the substances. Consequently, POPs are found in regions far from where they have been used or released. Due to their lipophilicity and persistence, they accumulate in the food-chain and high concentrations have been detected in animals and humans. Acronyms such as PBTs (persistent bioaccumulative and toxic substances), PTS (persistent toxic substances) or PEPs (persistent environmental pollutants) have also been used interchangeably. In a narrower sense, the term “POPs” refers to twelve chemicals addressed in the Stockholm Convention on Persistent Organic Pollutants, a global treaty negotiated under the auspices of the United Nations Environment Programme (UNEP) in order to eliminate the production and use or release of POPs. The twelve “Stockholm” POPs are the ten intentionally produced chemicals aldrin, chlordane, DDT, dieldrin, endrin, heptachlor, mirex, toxaphene, hexachlorobenzene, and polychlorinated biphenyls (PCB), and the two unintentionally produced substances polychlorinated dibenzo-p-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF)1. The Stockholm Convention has as its objective the protection of human health and the environment from POPs. The process of developing the treaty was initiated in May 1995 by UNEP leading to the adoption of the convention in May 2001 in Stockholm. One year later, 151 countries are signatories of the Stockholm Convention. The Convention will come into force after 50 ratifications. There is a high level of interest among governments, international organizations, environmental and industrial non-governmental organizations, and academia in addressing POPs issues in a concerted way and searching for solutions to problems caused by POPs. With a global convention in place, we may move forward to eliminate POPs, considered by some to be the most toxic man-made substances. This volume introduces the history and obligations of the Stockholm Convention as well as its provisions for adding more POPs in the future. It also covers the POPs Protocol under the Convention on Long-range Transboundary 1

The Convention also contains provisions for PCB and HCB as unintentionally produced substances.

XII

Introduction

Air Pollution (LRTAP) of the United Nations Economic Commission for Europe. It addresses the Stockholm POPs and highlights their properties, toxicity, and occurrence in the environment and provides human data. Chapters are dedicated to global transport and the fate of POPs, inventories, and technical solutions for the reduction in POPs releases or their destruction. Case studies from three continents – Asia, Africa, and Central America – provide regional flavor and show that developing countries have been able to address this class of chemicals. The chapters were written by experts highly regarded for their knowledge of POPs issues; furthermore, they represent different perspectives. As POPs are global in their impact, this volume also attempts to cover a wider geographical range and different stages of industrial development. Châtelaine, September 2002

Heidelore Fiedler

CHAPTER 1

Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution on Persistent Organic Pollutants: The 1998 Agreement for the UNECE Region Keith Bull United Nations Economic Commission for Europe, Palais des Nations, 1211 Geneva 10, Switzerland E-mail: [email protected]

The 1979 Convention on Long-range Transboundary Air Pollution provides a framework for detailed agreements on particular substances through Protocols to the Convention. The Protocol on Persistent Organic Pollutants was adopted by 36 Parties in 1998. So far, 6 countries have ratified the Protocol, another 10 need to do so before it enters into force. The Protocol was the culmination of work under the Convention started in 1989 and led initially by Canada and Sweden.An ad hoc Working Group under the Convention provided the necessary information and draft text for the negotiations. The adopted Protocol covers 16 substances or groups of substances that were selected by a screening procedure followed by negotiations. For most of the substances (aldrin, chlordane, chlordecone, DDT, dieldrin, endrin, heptachlor, hexabromobiphenyl, hexachlorobenzene, mirex, PCB, toxaphene) the obligation of Parties is for elimination of production and use; the substances are mainly pesticides with well-demonstrated persistence and toxicity. For three substances (DDT, HCH, PCB) there are restrictions of use, and for another group of substances (PAHs, dioxins/furans, and hexachlorobenzene) there are obligations to reduce emissions from specified reference years.A mechanism for selecting substances to add to the Protocol, through an amendment procedure included in the Protocol, was agreed separately in a Decision by the Executive Body for the Convention. Amendments are possible once the Protocol enters into force. Review procedures to ascertain the sufficiency and effectiveness of the obligations are included in the Protocol, the first such review is to be within three years of the Protocol entering into force. Keywords: Persistent organic pollutants (POPs), Protocol, Convention on Long-range Transboundary Air Pollution

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2

. . . . . .

2

. .

3

Steps Towards Development of a Protocol on Persistent Organic Pollutants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5

5

The 1998 Aarhus Protocol on POPs . . . . . . . . . . . . . . . . .

8

6

The Status of the Protocol . . . . . . . . . . . . . . . . . . . . . .

9

7

The Future – Implementation and Revision of the Protocol . . . . 10

8

References

1

Introduction

2

The United Nations Economic Commission for Europe

3

The Convention on Long-Range Transboundary Air Pollution

4

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

2

K. Bull

1 Introduction The 1998 Protocol on persistent organic pollutants (POPs) [1] was adopted by the Executive Body for the Convention on Long-range Transboundary Air Pollution on 24 June 1998 in Aarhus (Denmark). It was a regional agreement reached as a result of much negotiating effort. However, it was a major step towards a global agreement and provided the basis for further steps in the regional control of POPs in the future. The Convention has been the focus for international air pollution controls for the UNECE (United Nations Economic Commission for Europe) area over many years and has played an important role in the emission decreases observed in Europe and North America in recent years. This paper describes the Protocol, its content and obligations, within the perspective of the Convention and the UNECE. It seeks to describe the history of the development of the Protocol and indicate the way it will operate in the future.

2 The United Nations Economic Commission for Europe The UNECE is one of the five regional commissions of the United Nations. It includes countries of Western and Eastern Europe, including the Newly Independent States (NIS) extending eastwards to countries such as Kazakhstan and Kyrgyzstan. The region also includes, despite its name, the United States of America and Canada. Created in 1947 as a United Nations regional organization it struggled in its early days to bring together “East” and “West” in a spirit of cooperation. After a difficult initial period the UNECE played a vital role in providing the forum for discussions and agreement between countries with very different political and economic systems. Such a framework provided an important platform for launching agreements such as the Convention on Long-range Transboundary Air Pollution (CLRTAP). In turn, the Convention played an important role in promoting a spirit of collaboration on a specific issue which was recognized by many as one requiring international cooperation to solve effects at a national level. The UNECE secretariat, based in Geneva, provides the secretariat support for a number of multi-lateral environmental conventions, including that on longrange transboundary air pollution. Indeed CLRTAP identifies the secretariat role of the Executive Secretary of UNECE in its text (Article 11). Members of the secretariat support the work of the Executive Body of the Convention and its subsidiary bodies primarily through the organization of meetings and the preparation of documents. They also provide the necessary links with international organizations with common goals and interests.

Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution

3

3 The Convention on Long-Range Transboundary Air Pollution The history of CLRTAP dates back to the 1960s when there was increasing concern about the effects of air pollution and it was suggested that long-range effects were possible. Links were drawn between emissions in Europe and effects on Scandinavian lakes. In 1972 a United Nations Conference on the Human Environment, held in Stockholm, initiated discussions aimed at international cooperation to address the effects of acidification. Throughout the 1970s scientific studies investigated the effects and their cause and clearly demonstrated that air pollutants could travel for considerable distances and could cause harmful effects thousands of kilometres away from the emission sources. The transboundary nature of the problem could only be solved by international agreement. It has been suggested that CLRTAP arose as the result of a specific “policy window” resulting from the convergence of Scandinavian concern for the issue and the will of the Soviet Union to find a more binding platform than that of the positive but less binding OECD [2]. The Convention [3] was adopted in Geneva in 1979 and entered into force in 1983 after ratification by 16 Parties. The Convention identifies the general principles for international cooperation on air pollution abatement and provides an institutional framework for bringing together science and policy. It provides no specific commitment to decrease or limit emissions of any air pollutant, but offers a binding legal framework within which specific agreements could be agreed. These agreements have taken the form of protocols to the Convention – enshrined in documents that have been separately adopted and ratified by Parties to CLRTAP. Thirty-three Parties (including the European Community) signed the Convention in 1979. Most of these signatories have now ratified the agreement and several other countries of the UNECE region have acceded to the Convention.As a result there are currently 48 Parties (and two signatories that are not Parties) to the Convention (listed in Table 1), from a total of 55 UNECE states. Non-Parties generally fall into two categories, namely some countries from Eastern Europe (e.g., Uzbekistan) and some of the smallest states in Europe (e.g.,Andorra). Since its entry into force CLRTAP has been extended by eight protocols. Five of these [3] have themselves already entered into force having received at least the necessary number (16) of ratifications: – The 1984 Protocol on Long-term Financing of the Cooperative Programme for Monitoring and Evaluation of the Long-range Transmission of Air Pollutants in Europe (EMEP); – The 1985 Protocol on the Reduction of Sulphur Emissions on their Transboundary Fluxes by at least 30 per cent; – The 1988 Protocol concerning the Control of Nitrogen Oxides or their Transboundary Fluxes; – The 1991 Protocol concerning the Control of Emissions of Volatile Organic Compounds or their Transboundary Fluxes; – The 1994 Protocol on Further Reduction of Sulphur Emissions.

4

K. Bull

Table 1. Status of the 1998 Aarhus Protocol on Persistent Organic Pollutants (POPs) – June 2002

Signature Armenia Austria Belarus Belgium Bosnia and Herzegovina Bulgaria Canada Croatia Cyprus Czech Republic Denmark Estonia Finland France Georgia Germany Greece Holy See Hungary Iceland Ireland Italy Kazakhstan Kyrgyzstan Latvia Liechtenstein Lithuania Luxembourg Malta Monaco Netherlands Norway Poland Portugal Republic of Moldovia Romania Russian Federation San Marino Slovakia Slovenia Spain Sweden Switzerland The former Yugoslav Republic of Macedonia Turkey Ukraine United Kingdom of Great Britain and Northern Ireland United States of America Yugoslavia European Community Total:

Ratification

18.12.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998

05.12.2001 18.12.1998

06.07.2001

24.06.1998 24.06.1998 24.06.1998 24.06.1998

25.04.2002

18.12.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 24.06.1998 25.06.1998 24.06.1998 24.06.1998

01.05.2000 23.06.2000 16.12.1999

19.01.2000 14.11.2000

24.06.1998 24.06.1998 24.06.1998 24.06.1998 36

9

Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution

5

The remaining three protocols, including the 1994 Protocol on Persistent Organic Pollutants, still await further ratifications before they enter into force: – The 1998 Protocol on Heavy Metals; – The 1998 Protocol on Persistent Organic Pollutants (POPs) and; – The 1999 Protocol to Abate Acidification, Eutrophication and Ground-level Ozone. Over the years, protocols have become increasingly complex, with increasing amounts of technical information provided (usually in the forms of annexes to the protocols) to guide Parties in their implementation of each protocol. Early protocols, whilst recognizing that environmental and human health effects were important and needed to be addressed, defined obligations in simple terms; percentage emission decreases, or a return to previous emission levels, sought to alleviate the effects of air pollutants, but did not link obligations to environmental goals. Through the 1990s, however, more attention has been paid to the effects of air pollutants. For two protocols, the 1994 Protocol on Further Reduction of Sulphur Emissions and the 1999 Protocol to Abate Acidification, Eutrophication and Ground-level Ozone, effects-based approaches, using critical loads [4] and integrated assessment models [5] have defined Parties obligations by taking account of national emissions, their effect on the environment, and the costs involved in achieving certain environmental goals. While neither the 1998 Protocols on Heavy Metals nor that on POPs take an effects-based approach, as is the case for “first Protocols” for other air pollutants, they both pay careful regard to the effects of the pollutants whose controls they encompass.

4 Steps Towards Development of a Protocol on Persistent Organic Pollutants This section outlines the history of the development of the Protocol on POPs. For a more comprehensive description of events and a discussion of the factors important for the development and final form of the protocol see Selin [6]. The Convention’s attention was first drawn to the issue of POPs by the Canadian delegation to the Working Group on Effects in 1989.An official of the Canadian Department of Indian and Northern Affairs presented a paper with the approval of Environment Canada, the body responsible for the Canadian delegation at CLRTAP meetings. The Swedish government had similar concerns and, in 1990, the Working Group on Effects received a further paper for its attention from Canada together with a proposal from Sweden for the creation of a Task Force on complex organic compounds. Subsequently, in November 1990, the Executive Body of the Convention agreed to establish an intergovernmental Task Force on Persistent Organic Pollutants to be jointly led by Canada and Sweden. The Task Force was to work under the auspices of the Working Group on Technology and in cooperation with the Working Group on Effects. The Task Force first met in March 1991.

6

K. Bull

The Task Force on POPs was given the job of assessing the effects of POPs and the possible need for action to control them. The Task Force set about compiling the necessary scientific information to prepare a substantive report on POPs. The lead countries were also able to take the opportunity to promote their interests and concerns to other Parties to the Convention. At the end of 1993, the Executive Body requested the Task Force to prepare a final report for its session at the end of 1994. The draft report was presented for discussion to a joint meeting of the Working Group on Technologies and the Working Group on Effects. It was received enthusiastically and recommended to the Executive Body. A similar report, from a Task Force on Heavy Metals, was treated similarly. The Executive Body deliberated upon future actions on both POPs and heavy metals and decided to set up an ad hoc Preparatory Working Group for each of the groups of pollutants. In this way the Executive Body did not commit itself to future action but provided the opportunity for further discussions and preparations to aid the decision-making process regarding future action. Further, the working groups were to report to the Working Group on Strategies which took the discussions into a more policy-oriented forum. The mandate of the Preparatory Working Group was to consider an initial list of substances and control options, consider and assess possible elements for a future protocol, and develop procedures for the future addition of substances to such a protocol. It was not a negotiating group but it did take steps to prepare documents for future consideration by the Working Group on Strategies that was charged with preparing a draft protocol and begin negotiations when there was a sound basis for doing so. The work-plan of the Preparatory Working Group, adopted in 1995, required it to prepare an annotated outline of a protocol and elements for a draft text, identify criteria for selecting an initial list of substances and developing proposals for adding substances to the list, as well as options for annexes and for a basic obligations article for a protocol. It was also expected to prepare, assemble and review information, hold technical workshops and assess implications of draft commitments. To arrive at a list of substances the Preparatory Working Group adopted a screening procedure. It considered a number of such procedures used by countries and by other bodies but also took into consideration the scope of the Convention regarding long-range transport. The three stages to the assessment involved: 1. Evidence of persistence (low vapour pressure, half life in the atmosphere of more than two days, low biodegradability (less than 30% in 28 days; OR monitoring evidence in remote regions); 2. Prioritization scoring based on bioconcentration factors or octanol/water partition coefficients and mammalian or aquatic toxicology; 3. Risk assessment. One hundred and seven POPs were considered and 87 remained after the first stage. For the second stage a scoring system was used to combine information on toxicity with that on bioaccumulation. For some substances data were not sufficient, but at this stage 36 substances were eliminated and 32 forwarded to stage 3. Through consideration of the scientific criteria, including properties such as the

Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution

7

Table 2. List of fourteen substances arrived at through the ad hoc Preparatory Working Group on POPs screening procedure, and the four borderline substances included for additional consideration (in parentheses)

Pesticides

Industrial chemicals

Unintentional by-products

Aldrin Chlordane DDT Dieldrin Endrin Hexachlorobenzene Mirex Toxaphene (Chlordecone) (Lindane) (Heptachlor)

Hexabromobiphenyl PCB Pentachlorophenol (Short-Chain Chlorinated Paraffins)

Dioxins Furans PAHs

risks from degradation products, together with more socio-economic issues such as use, production and emissions resulting from these, a risk assessment was made of each substance. The methodology resulted in a list of just 14 substances (Table 2). Following subsequent discussions and proposals in the Preparatory Working Group, the list of 14 together with another four “border-line” substances (SCCP – short-chain chlorinated paraffins, heptachlor, chlordecone and lindane) was put forward for negotiations to the Working Group on Strategies. In setting the work-plans for the Preparatory Working Group and for the Working Group on Strategies, the Executive Body stressed the importance of rapid agreement on a protocol of limited scope, particularly in view of the model that such a protocol would set for action beyond the region and/or at a global level. This was with a view to aiding the development of a global agreement on POPs that had been started under the auspices of UNEP. It was recognized that some mechanism was likely to be required for adding substances to the lists that would be annexed to the final protocol. For this it was decided that the Executive Body would make a formal decision on the process drawing upon the experience gained in the screening and evaluation exercise used for drawing up the original list. The option for a Decision provided greater flexibility than incorporating the technical assessment procedures for amendment in the protocol itself, though the formalities for proposing new substances were considered as an article for the protocol. Through the subsequent five negotiating meetings in 1997 and 1998, not only the text, but also the lists of substances to be covered by the protocol were re-considered and individual substances negotiated for inclusion or exclusion. While many substances received consensus for inclusion, others were the subject of negotiations throughout 1997 and into 1998. The final negotiating session took place in February 1998, and sought to resolve such issues as the formal definition of PCB, possible DDT and HCB exemptions and the inclusion or exclusion of pentachlorophenol and SCCP. It agreed on the final list of 16 substances, or

8

K. Bull

Table 3. Substances listed in the annexes to the 1998 Protocol on POPs

Annex I. Substances scheduled for elimination of production and use Aldrin Chlordane Chlordecone DDT (pp¢-dichlorodiphenyltrichloroethane) Dieldrin Endrin Heptachlor Hexabromobiphenyl Hexachlorobenzene Mirex PCB (polychlorinated biphenyls) Toxaphene Annex II. Substances scheduled for restrictions on use DDT HCH (hexachlorocyclohexane) PCB Annex III. Substances for which emissions must be reduced PAH (polycyclic aromatic hydrocarbons) Dioxins/furans Hexachlorobenzene

groups of substances (Table 3) comprising eleven pesticides, two industrial chemicals and three by-products/contaminants. This session also considered the final wording of the Executive Body decision on adding substances to the Protocol once it had entered into force. At a special meeting of the Executive Body in March 1998, Decision 1998/2 was formally taken on “information to be submitted and procedures for adding substances into the Protocol”. On 24 June 1998, at the fourth Environment for Europe Ministerial Conference in Aarhus, Denmark the Executive Body adopted the Protocol.

5 The 1998 Aarhus Protocol on POPs The Aarhus Protocol follows the pattern of many other such protocols. Following a preamble and definitions article, the objective and basic obligations are spelled out. Article 4 then specifies exemptions to the basic obligations. Articles 5 to 9 deal with the operational, technical and scientific issues: exchange of information and technology; public awareness; strategies, policies, programmes, measures and information; research, development and monitoring; and reporting. The introduction of public awareness into the Protocol was consistent with the adoption of the Aarhus Convention on Access to Information, Public Participation in Decision Making and Access to Justice in Environmental Matters at the same Ministerial Conference in 1998.Articles 11 and 12 deal with compliance and settlement of disputes, and Article 13 formally identifies the Annexes. Articles

Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution

9

14 to 20 identify the formalities of: amendments; signature, ratification, acceptance and accession; depository; entry into force; withdrawal; and authentic texts. The basic objective of the Protocol (Article 2) is to control, reduce or eliminate discharges, emissions and losses of persistent organic pollutants. For this, the Protocol clearly adopts the precautionary principle referring to Principle 15 of the Rio Declaration on Environment and Development (Agenda 21) where it was identified. Principle 15 states that “lack of full scientific consensus shall not be used as a reason for postponing cost-effective measures to prevent environment degradation” where “there are threats of serious or irreversible damage”. To meet the objective, basic obligations under the Protocol are detailed (Article 3). They specify the need to take effective measures to eliminate the production and use of substances listed in Annex I (Table 3). Some substances are banned outright (aldrin, chlordane, chlordecone, dieldrin, endrin, hexabromobiphenyl, mirex and toxaphene). Others have specified exemptions but are scheduled for elimination at a later stage (DDT, heptachlor, hexaclorobenzene, PCBs). Disposal or destruction of these substances, or their transboundary movement, must be undertaken in an environmentally sound manner. The obligations take note of the existing Basel Convention on the Control of Transboundary Movements of Hazardous Wastes and their Disposal. The Protocol severely restricts the use of DDT, HCH (including lindane) and PCBs, substances listed in Annex II of the Protocol (Table 3). It also obliges Parties to reduce their emissions of dioxins, furans, PAHs and HCB, substances listed in Annex III to the Protocol (Table 3), below their levels in 1990 (or an alternative year between 1985 and 1995). For the incineration of municipal, hazardous and medical waste, it lays down specific limit values for these substances. Other timescales, listed in Annex VI, determine the need for action to apply best available technology (BAT), achieve limit values, and take effective measures to control emissions from mobile sources as indicated in Annexes V, IV and VII, respectively.

6 The Status of the Protocol At the adoption of the Protocol, 33 Parties to the Convention, including the European Community, signed the agreement. Spain became the 34th Signatory one day later, and over the following 6 months when the Protocol was open for signature Armenia and Hungary also signed. Since 1998, the Protocol has been ratified by six signatories (Table 1). These have completed their necessary national procedures and have lodged their articles of ratification with the United Nations. Under Article 18 of the Protocol, entry into force takes place on the ninetieth day following the sixteenth ratification. The Protocol then becomes binding on those Parties that have ratified. It will also become binding on all those that subsequently ratify or accede (the ratification process for non-Signatories) to the Protocol. It is usually takes some period of time, two or more years, for a Party to complete its national procedures towards ratification. While procedures vary from country to country they usually involve detailed consideration of obligations,

10

K. Bull

often involving more than one government ministry, and a formal decision to ratify by government. Even so, the rate of ratification has been slow, and there are more ratifications of the Protocol on Heavy Metals that was adopted at the same time by the Executive Body. It is known that some countries have decided to delay ratification of the Protocol until completion of the negotiations on the global POPs Convention. There have been indications that the ratification of the two agreements would be considered in parallel to make most efficient use of national resources. This is likely to slow ratification and entry into force as the global Convention, drawn up under the auspices of UNEP, has only recently been adopted in May 2001 [7].

7 The Future – Implementation and Revision of the Protocol Once the Protocol enters into force there will be an obligation on Parties to the Protocol to report their usage and/or emissions of the substances listed. Already the Cooperative Programme on Monitoring and Assessment of the Long-Range Transmission of Air Pollutants (EMEP), that operates under the Convention, has taken steps to initiate reporting procedures. Some countries are already providing the required information. The Convention will monitor progress in implementation of the Protocol through the activities of its Implementation Committee in conjunction with the annual reporting procedures. In addition, Parties to the Convention are invited, every two years, to report on the strategies and policies they have adopted to meet obligations under the protocols to the Convention. This information too is scrutinized by the Implementation Committee, together with any direct submissions to the committee from individual countries. Under the terms of the Protocol a formal review of its obligations and consideration if they are still sufficient must be conducted within three years of entry into force. This provides the opportunity to update or amend the obligations, exemptions and annexes of substances. Amendments to the Protocol are dealt with under Article 14, though the procedure for adding substances is detailed in the Executive Body’s Decision 1998/2 [8]. In this, the technical screening procedure and provision of information to substantiate addition of a substance is described. While such a procedure can only take place after the Protocol has entered into force, the Executive Body in 1999 agreed to establish a group of experts on POPs, reporting to its Working Group on Strategies, that was charged with gathering and evaluating information on substances that might be added to the Protocol annexes. The expert group may also be able to provide advice for the review process when the Protocol enters into force. It remains to be seen what role the Protocol will play in the future controls of POPs now that the global POPs Convention has been adopted. While the substances and provisions of the two instruments are similar, the Protocol has additional substances and fewer exemptions. It might therefore be seen as leading the way for more stringent global controls in the future. In addition, it is likely

Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution

11

that the Protocol will enter into force sooner and provide the opportunity to add more substances at an earlier date. This again will set a precedent for the global instrument to follow.

8 References 1. United Nations (1998a) Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution on Persistent Organic Pollutants. ECE/EB.AIR/60 2. Castells N (1999) International environmental agreements: institutional innovation in European transboundary air pollution policies. PhD Thesis, University of Amsterdam 3. United Nations (1996) 1979 Convention on Long-Range Transboundary Air Pollution and its Protocols. United Nations: New York and Geneva 4. Bull KR (1995) Critical loads – possibilities and constraints. Water, Air and Soil Pollution 85:201–212 5. Hordijk L (1995) Integrated assessment models as the basis for air pollution negotiations. Water, Air and Soil Pollution 85:249–260 6. Selin H (2000) Towards International Chemical Safety: Taking Action on Persistent Organic Pollutants (POPs). Linköping. Linköping Studies in Arts and Science 211 7. UNEP (2001) Text of the Stockholm Convention on Persistent Organic Pollutants. UNEP/POPS/CONF/2. 8. United Nations (1998) Executive Body Decision 1998/2 on information to be submitted and the procedure for adding substances to Annexes I, II or II to the Protocol on Persistent Organic Pollutants. EB.AIR/WG.5/52, Annex II

CHAPTER 2

The Development of a Global Treaty on Persistent Organic Pollutants (POPs) John Buccini Chair, UNEP Intergovernmental Negotiating Committee on POPs, 31 Sycamore Drive, Ottawa, Ontario, Canada K2H 6R4 E-mail: [email protected]

This chapter summarizes the process involved in developing the recent United Nations Environment Programme (UNEP) treaty on POPs (the Stockholm Convention), summarizes the main provisions of the Convention, and briefly comments on the near-term prospects for further international developments on POPs. Keywords: Persistent organic pollutants, Stockholm Convention, Negotiation of convention

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

1

Introduction

2

Developing a Global UNEP Convention on POPs . . . . . . . . . . 15

2.1 2.2 2.3 2.4 2.5

The Mandate . . . . . . . . . . . . . . . . . Implementing the Mandate . . . . . . . . . Conclusions and Recommendations . . . . Actions in Support of the Negotiation Process The Negotiations . . . . . . . . . . . . . . .

3

The Stockholm Convention on POPs

3.1 3.2 3.2.1 3.2.1.1 3.2.1.2 3.2.2 3.2.3 3.3 3.4

General Provisions . . . . . . . . . . . . . . . . . . . . . Control Provisions . . . . . . . . . . . . . . . . . . . . . Intentionally Produced POPs . . . . . . . . . . . . . . . The Initial 10 POPs . . . . . . . . . . . . . . . . . . . . . Provisions Applicable to All Intentionally Produced POPs POPs that are not Intentionally Produced . . . . . . . . . POPs in Stockpiles or Wastes . . . . . . . . . . . . . . . Procedure for Adding New POPs . . . . . . . . . . . . . Financial and Technical Assistance . . . . . . . . . . . .

4

Future Actions on POPs

. . . . . . . . . . . . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

15 17 18 18 20

. . . . . . . . . . . . . . . . 21 . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

21 22 22 23 24 25 26 27 29

. . . . . . . . . . . . . . . . . . . . . . . 29

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

14

J. Buccini

Abbreviations POPs UNEP UN ECE LRTAP USA DDT PCBs HCB GC IOMC IPCS IFCS WHA PIC INC SADC CEG COP ppm BAT BEP CAS IUPAC logKOW ENGOs IPEN

persistent organic pollutants United Nations Environment Programme United Nations Economic Commission for Europe Long-Range Transboundary Air Pollution United States of America 1,1¢-(2,2,2-trichloroethylidene)bis(4-chlorobenzene) polychlorinated biphenyls hexachlorobenzene Governing Council Inter-Organisation Programme for the Sound Management of Chemicals International Programme on Chemical Safety Intergovernmental Forum on Chemical Safety World Health Assembly prior informed consent Intergovernmental Negotiating Committee South African Development Community Criteria Expert Group Conference of Parties parts per million best available techniques best environmental practices Chemical Abstracts Number International Union of Pure and Applied Chemistry logarithm of the octanol/water partition coefficient environmental non-government organizations International POPs Elimination Network

1 Introduction Persistent organic pollutants (POPs) are organic compounds of natural or anthropogenic origin that possess a particular combination of physical and chemical properties such that, once released into the environment, they remain intact for exceptionally long periods of time as they resist photolytic, chemical and biological degradation. POPs in the environment are transported at low concentrations by movement of fresh and marine waters and, as they are semi-volatile, are transported over long distances in the atmosphere. The result is widespread distribution of POPs across the globe, including regions where they have never been used. POPs are characterized by low water solubility and high lipid solubility, resulting in their bioaccumulation in fatty tissues of living organisms, including humans, and they are found at higher concentrations at higher levels in the food chain. Thus, both humans and environmental organisms are exposed to POPs around the world, in many cases for extended periods of time spanning generations, resulting in both acute and chronic toxic effects to both humans and wildlife.

The Development of a Global Treaty on Persistent Organic Pollutants (POPs)

15

In recent decades, the risks posed by POPs have become of increasing concern in many countries, resulting in actions to protect human health and the environment being taken or proposed at the national, regional and international levels. The following are some of the major regional and global initiatives that were underway prior to the formal commitment in 1997 to develop a global POPs treaty, and that were directed at identifying POPs and developing risk management measures to control the exposure of humans and the ecosystem to these substances. (a) The UNEP Global Programme of Action for the Protection of the Marine Environment from Land-based Activities was agreed to at a UNEP conference in Washington, D.C. (October 23–November 3, 1995) and POPs were identified as a priority for action under the plan. (b) The UN ECE Convention on Long-Range Transboundary Air Pollution (LRTAP) includes the Aarhus Protocol on POPs, which was signed on June 24, 1998, and calls for action on sixteen identified POPs. (c) The 1992 Convention on the Protection of the Marine Environment of the Baltic Sea (the Helsinki Convention). (d) The 1976 Convention for the Protection of the Mediterranean Sea Against Pollution (Barcelona Convention), as amended in 1995. (e) The North American Commission for Environmental Cooperation passed Resolution #95-5 on the Sound Management of Chemicals (October 13,1995) and gave immediate priority for Canada, Mexico and the USA to address persistent toxic substances and has resulted in the development and implementation of continental actions plans for DDT, chlordane and PCBs and a commitment to develop an action plan on dioxins, furans and hexachlorobenzene (HCB). (f) Canada-USA Great Lakes Water Quality Agreement (1972), including the Binational Toxics Strategy (April 1997), emphasizes action on POPs as well as other persistent toxic substances. While this list is not exhaustive, it does show that POPs were and will continue to be the subject of considerable attention for both scientists and policy makers.

2 Developing a Global UNEP Convention on POPs 2.1 The Mandate

At its May 1995 meeting, the UNEP Governing Council (GC) adopted Decision 18/32 on POPs, which invited the Inter-Organization Programme for the Sound Management of Chemicals (IOMC), working with the International Program on Chemical Safety (IPCS) and the Intergovernmental Forum on Chemical Safety (IFCS), to initiate an expeditious assessment process, initially beginning with twelve specified POPs [PCBs, dioxins, furans, aldrin, dieldrin, DDT, endrin, chlordane, hexachlorobenzene (HCB), mirex, toxaphene and heptachlor].As specified

16

J. Buccini

in the following text taken from Decision 18/32, this assessment process should, taking into account the circumstances of developing countries and countries with economies in transition: “a) consolidate existing information available from IPCS, UN ECE and other relevant sources, on the chemistry and toxicology of the substances concerned (particularly the impact on human, plant and animal health); b) analyze the relevant transport pathways and the origin, transport and deposition of these substances on a global scale; c) examine the sources, benefits, risks and other considerations relevant to production and use; d) evaluate the availability, including costs and effectiveness, of preferable substitutes, where applicable; and e) assess realistic response strategies, policies and mechanisms for reducing and/or eliminating emissions, discharges and losses of POPs.” Based on the results of this process, together with the outcome of the UNEP Intergovernmental Conference to Adopt a Global Programme of Action for the Protection of the Marine Environment from Land-based Activities (Washington, D.C., October 23–November 3, 1995), IFCS was invited to develop recommendations and information on international action, including any information that would be needed for a possible decision on an appropriate international legal mechanism on POPs, to be considered at the respective 1997 sessions of the UNEP GC and the World Health Assembly (WHA), the policy body of the World Health Organisation. At a UNEP meeting in Washington (Oct. 23–Nov. 3, 1995), countries adopted a Global Programme of Action for the Protection of the Marine Environment which, in part, recognized the importance of controlling releases of POPs, specified actions that should be taken on POPs, and encouraged countries to participate actively in implementing GC 18/32. The following paragraph from the Washington Declaration on Protection of the Marine Environment from Landbased Activities (November 2, 1995) was, therefore, taken into consideration in implementing Decision 18/32. “17. Acting to develop, in accordance with the provisions of the Global Programme of Action, a global, legally binding instrument for the reduction and/or elimination of emissions, discharges and, where appropriate, the elimination of the manufacture and use of the persistent organic pollutants identified in decision 18/32 of the Governing Council of the United Nations Environment Programme. The nature of the obligations undertaken must be developed recognizing the special circumstances of countries in need of assistance. Particular attention should be devoted to the potential need for the continued use of certain persistent organic pollutants to safeguard human health, sustain food production and to alleviate poverty in the absence of alternatives and the difficulty of acquiring substitutes and transferring of technology for the development and/or production of those substitutes.”

The Development of a Global Treaty on Persistent Organic Pollutants (POPs)

17

2.2 Implementing the Mandate

An international multi-stakeholder working group (Working Group) was established to implement Decision 18/32. Initially established as a Working Group under UNEP, the group was later adopted by IFCS in order to discharge IFCS obligations to provide recommendations to UNEP Governing Council as requested in Decision 18/32. This Working Group included representatives from intergovernmental organizations, governments, industry, public interest groups and scientific organizations from around the world. In initiating the assessment process requested by Decision 18/32, the Working Group took into account related international initiatives including: (a) UNEP GC Decision 18/12, which concerns the development of a legally binding instrument for the application of the Prior Informed Consent (PIC) procedure for certain hazardous chemicals in international trade, recognizing that some of the POPs specified in Decision 18/32 were covered by current voluntary PIC procedures; (b) UNEP GC Decision 18/31, which encouraged support for the Global Programme of Action for the Protection of the Marine Environment from Landbased Activities (wherein specific reference was made to POPs) that was subsequently accepted at the UNEP Intergovernmental Conference, as reflected in the Washington Declaration on Protection of the Marine Environment from Land-based Activities, and that involves countries in national, regional and international activities to implement the Plan; (c) the negotiations initiated in 1996 on a POPs protocol under the UN ECE Convention on LRTAP (concluded in 1998 and now referred to as the Aarhus POPs Protocol); and (d) the regional seas agreements, including conventions and protocols. The Working Group first met in Washington, D.C. (October 1995) and agreed to the development of a basic review of chemistry and toxicology of the 12 POPs. The result was a report prepared for the International Programme on Chemical Safety entitled Persistent Organic Pollutants, An Assessment Report on DDT, Aldrin, Dieldrin, Endrin, Chlordane, Heptachlor, Hexachlorobenzene, Mirex, Toxaphene, Polychlorinated biphenyls, Dioxins and Furans (December 1995). This report was considered at a meeting of the Working Group in Canberra,Australia (March 1996) and provided the basis for the Working Group to conclude that sufficient information was available on the chemistry, toxicology, transport pathways, origin, transport and deposition of the 12 specified POPs to demonstrate the need for immediate international action and to provide a basis for moving forward on realistic response strategies. The significance of achieving agreement on this conclusion cannot be understated as it marked a turning point in the consultations from a discussion of whether action was warranted to a discussion of what action to take and how to proceed. The Working Group met for the third and final time in Manila, Philippines (June 1996) and developed the Final Report of the Intergovernmental Forum on Chemical Safety ad hoc Working Group on Persistent Organic Pollutants (IFCS re-

18

J. Buccini

port) which included recommendations for future action on POPs. The IFCS report and recommendations were unanimously supported by all stakeholders and submitted to UNEP and WHA for consideration at their respective 1997 meetings. Several documents were developed in addition to the final IFCS report and most of these are available on the UNEP POPS website. 2.3 Conclusions and Recommendations

The IFCS conclusions and recommendations were approved by both UNEP Governing Council (Decision 19/13C, February 3, 1997) and the WHA (Resolution WHA50.13, May 12, 1997). The key IFCS conclusion was that sufficient scientific information was available on the 12 specified POPs to demonstrate the need for immediate international action and to provide a basis for moving forward on realistic response strategies. Based on acceptance of this key conclusion, UNEP and WHA also accepted the IFCS recommendation that immediate international action should be initiated to protect human health and the environment through measures which will “reduce and/or eliminate ... the emissions and discharges of the 12 POPs” and “where appropriate, eliminate production and subsequently the remaining use of those POPs that are intentionally produced”. UNEP GC agreed to begin negotiation of a global legally binding instrument “by early 1998” and to conclude this task “preferably by the year 2000”. It was noted that proposed action programs should take into account that the 12 specified POPs include pesticides, industrial chemicals, and unintentionally produced by-products and contaminants, and that, within the framework of overarching objectives that were to be negotiated by an intergovernmental negotiating committee (INC), different approaches were needed for each category of POPs. The negotiation mandate was now in hand, and it was time-bound. 2.4 Actions in Support of the Negotiation Process

UNEP Governing Council urged governments to initiate action on the recommendations in the IFCS report and to provide technical assistance, capacity building and funding to enable developing countries and countries with economies in transition to take appropriate action on POPs. Governing Council Decision 19/13C also requested UNEP to initiate immediate action on POPs in response to recommendations in the IFCS report including: (a) general awareness raising on the national, regional and global aspects of POPs; (b) information exchange, within and between countries and intergovernmental organizations; (c) promoting information exchange on alternative products and processes to reduce or eliminate POPs generation, use and release;

The Development of a Global Treaty on Persistent Organic Pollutants (POPs)

19

(d) assisting countries in identifying and developing inventories of PCBs and in identifying world-wide capacity to destroy PCBs; (e) developing inventories of information on dioxins and furans, including sources of releases and practices to manage these releases; and (f) collecting information that will be used in the INC negotiations to assist the development of criteria and a process for identifying additional POPs. At the second meeting of the IFCS (Ottawa, Canada, February 10–14, 1997), countries were informed of the progress of the ad hoc Working Group on POPs and agreed to continue the Working Group to assist in preparing for the UNEP negotiation process and to focus efforts of governments to take action on POPs. As a result, joint UNEP/IFCS regional and sub-regional workshops were held at the following locations to raise awareness of the many issues that had to be addressed in preparing nations for the commencement of the UNEP negotiation process in early 1998 and in identifying and taking action on POPs problems at the local, regional and international levels. – – – – – – – –

St. Petersburg, Russia (July 1–4, 1997); Bangkok, Thailand (November 25–28, 1997); Bamaco, Mali (December 15–18, 1997); Cartegena, Columbia (January 27–30, 1998); Lusaka, Zambia (March 17–20, 1998); Iguassu Falls, Argentina (April 1–3, 1998); Kranska Gora, Slovenia (May 11–14, 1998); and Abu Dhabi, United Arab Emirates (June 7–9, 1998).

The workshops attracted representatives of governments, industry, academia, labor and public interest groups from 138 countries and provided an opportunity to gather the views and concerns of countries in the regions with regard to the scientific, technical, social and economic challenges that needed to be addressed during the development and implementation of a global legally binding treaty to reduce the releases of POPs to the environment. Several other workshops were held during the conduct of the negotiations including the following: – Regional Workshop on Management of Persistent Organic Pollutants (Hanoi, Vietnam, March 16–19, 1999). – Workshop on the Management of Persistent Organic Pollutants (POPs) for the South African Development Community (SADC) region (Lusaka, Zambia, February 14–16, 2000). – Workshop on sustainable approaches for pest and vector management and opportunities for collaboration in replacing POPs pesticides (Bangkok, Thailand, March 6–10, 2000). Reports that include the papers presented at all eleven workshops are available on the UNEP POPs web site (http://chem.unep.ch/pops/) and in hard copy format from UNEP Chemicals in Geneva (Tel.: +41-22-917-1234; Fax: +41-22-797-3460; e-mail: [email protected]).

20

J. Buccini

2.5 The Negotiations

In June 1998, UNEP convened an INC with a mandate to prepare, preferably by the year 2000, an international legally binding instrument for implementing international action, initially beginning with the 12 specified POPs, and taking into account the mandate in UNEP GC Decision 19/13C and the conclusions and recommendations of the IFCS report. Provision was to be made for commitments at a national and regional level allowing for a higher level of protection than that afforded through the global instrument. In addition, consideration should be given to voluntary measures that may be implemented as a complement to, or independently of, a legally binding instrument. Participation in the INC was open to governments and relevant non-governmental and intergovernmental organizations. Coordination among different regional and international initiatives on POPs was recognized as essential to ensure harmonized environmental and health outcomes from mutually supportive and effective programs that result in the development of policies with complementary and non-conflicting objectives. The negotiations were well attended, with participation by numerous intergovernmental organizations, a wide range of non-governmental organizations and over 120 countries. There was a series of five negotiation meetings: – – – – –

Montreal, Canada (June 29–July 3, 1998), Nairobi, Kenya (January 25–29, 1999), Geneva, Switzerland (September 6–11, 1999), Bonn, Germany (March 20–25, 2000), and Johannesburg, South Africa (December 4–10, 2000).

One of the requirements of the negotiation process was the development of science-based criteria and a procedure for identifying chemicals, in addition to the 12 specified POPs, as candidates for future international action. The process of developing the screening procedure involved reviews of criteria pertaining to persistence, bioaccumulation, toxicity and exposure in different regions, and took into account dispersion mechanisms for the atmosphere and the hydrosphere, migratory species and the need to reflect possible influences of marine transport and tropical climates. An expert body, the Criteria Expert Group (CEG), was established at the first meeting of the INC to carry out this work. The CEG, which met in Bangkok, Thailand (October 26–30, 1998) and Vienna,Austria (June 14–18, 1999), included scientific and socio-economic expertise relevant to the POPs issue and was representative of countries in different stages of development and from different geographical regions, as well as participants from relevant non-governmental and intergovernmental organizations. The CEG considered the criteria and procedure being considered by the UN ECE in the development of the Aarhus Protocol on POPs and also took full account of varied ecosystems and the circumstances of developing countries and countries with economies in transition, as well as the need to conserve biodiversity and protect endangered species. The principles set out in the Rio Declaration on Environment and Development, es-

The Development of a Global Treaty on Persistent Organic Pollutants (POPs)

21

pecially Principle 15 that includes a reference to the precautionary approach, and the provisions of Chapter 19 of Agenda 21 were also taken into account. The result of this work was the inclusion in the Convention of a special provision (summarized below) concerning the evaluation and selection of chemicals that are candidates for addition to the Convention. One other issue arose during the negotiations that required special attention: the financial mechanism that would be established under the Convention to provide resources to developing countries and countries with economies in transition. This required a special meeting attended by representatives of eighteen countries in Vevey, Switzerland (June 19–21, 2000) to address the essential and desirable attributes of the financial mechanism. The results of this meeting proved to be of key importance in the final negotiation session in arriving at a consensus on this issue. All documents and reports relating to the negotiations are available on the UNEP POPs home page.

3 The Stockholm Convention on POPs The Stockholm Convention on POPs will enter into force when fifty countries will have ratified it. Once entry into force takes place, the countries that have ratified will meet as a Conference of the Parties (COP) to the Convention and will begin to take decisions on the operation of the Convention. The main provisions of the Stockholm Convention, other than those that relate simply to decision taking and administration of the Convention, are presented under the following headings: – – – –

General provisions; Control provisions; Procedure for adding new POPs; Financial and technical assistance.

3.1 General Provisions

The Preamble of the Convention acknowledges that due to their physical, chemical and toxicological properties, the release of POPs to the environment causes local and long-range impacts on ecosystems and human health, particularly on women and Arctic indigenous groups. It also recognizes that all sectors of society have a role to play in taking action to reduce and/or eliminate the releases and discharges of POPs to the environment, and recognizes the need to develop and implement environmentally sound products and processes as alternatives to the generation, use and/or release of POPs. The concept of “precaution” is introduced in the Preamble, reflected in a number of operative provisions, and is referenced prominently in the Objective of the convention which states: “Mindful of the precautionary approach as set forth in Principle 15 of the Rio Declaration on Environment and Development, the objective of this Convention

22

J. Buccini

is to protect human health and the environment from persistent organic pollutants.” Under the Convention, each Party has the following general obligations: (a) develop an implementation plan within two years of the Convention coming into force and endeavor to implement the plan, which must be reviewed and updated on a periodic basis: stakeholders should be involved in all these actions; (b) designate a National Focal Point to interact with other Parties and the UNEP secretariat to facilitate the exchange of a broad range of information on the production, use and release of POPs and on alternatives to POPs; (c) promote and facilitate the following as they relate to public information, awareness and education on POPs and their alternatives: – awareness among policy and decision makers, – provision of available and up-to-date information to the public, – development and implementation of educational, training and public awareness programs, – public participation in developing and implementing measures to address POPs, and – training and development programs for all stakeholders; (d) encourage and/or undertake research, development, monitoring and cooperation on all aspects of POPs and their alternatives, including aspects relating to their environmental releases, trends in levels in the environment and humans, transport, fate, transformation, effects, socio-economic impacts, and release reduction and/or elimination; and (e) report to the COP on measures that the Party has taken to implement the Convention and on the effectiveness of the measures. In the future, the COP must evaluate the effectiveness of the Convention in reducing and/or eliminating the releases of POPs. This will be done by establishing a mechanism to acquire comparable monitoring data on the presence, levels and trends of POPs in environmental and biological media, as well as on regional and global environmental transport of POPs. This mechanism will tap into existing national, regional and global networks and sources of information, and its design will be addressed at the first meeting of the COP. The Convention specifies that the COP will review the first effectiveness report four years after the Convention has come into force. 3.2 Control Provisions 3.2.1 Intentionally Produced POPs

For all intentionally produced POPs (i.e., industrial chemicals and pesticides), the goal of the Convention is elimination of production and use. To achieve this goal, the production and use of an intentionally produced POP will be either eliminated or restricted and, in each case, trade will be restricted.

The Development of a Global Treaty on Persistent Organic Pollutants (POPs)

23

3.2.1.1 The Initial 10 POPs

For nine of the initial ten intentionally produced substances that are included in the Convention, each Party will be required to “prohibit and/or take the legal and administrative measures to eliminate” the production and use of the POPs that are listed in Annex A (“Elimination”) of the Convention [i.e., aldrin, chlordane, dieldrin, endrin, heptachlor, hexachlorobenzene (HCB), mirex, polychlorinated biphenyls (PCBs) and toxaphene]. With regard to DDT, each Party will be required to restrict production and use to the “Acceptable Purposes” specified in Annex B (“Restriction”). With the exceptions of endrin and toxaphene, some “specific exemptions” with regard to production and/or use of the intentionally produced POPs have been included in Annex A and Annex B for each chemical. Any state on becoming a Party may register for one or more “specific exemptions” listed in Annexes A and B by informing the UNEP secretariat who will maintain a publicly available register as required by the Convention. Unless a Party specifies an earlier date, a specific exemption will be valid for a period of five years after the Convention comes into effect for a particular chemical. A Party may withdraw its exemption at any time, or may request an extension of five years to its exemption. Each request for an extension will be reviewed by the COP based on information submitted by the requesting Party justifying its continued need for the registered exemption. Parties that engage in activities involving intentional production or use of POPs under the “specific exemptions” or “acceptable purposes” provisions must take measures to prevent or minimize human exposure and releases to the environment. There are two high-profile POPs among the initial ten that warrant some detailed explanation: PCBs and DDT. With regard to PCBs, all Parties have agreed to cease production of PCBs and to eliminate the use of in-place equipment containing PCBs (transformers, capacitors, etc.) by 2025. The use of in-place equipment containing PCBs is a “specific exemption” under Annex A for all Parties that is subject to some conditions (e.g., use only in intact and non-leaking equipment) and restrictions (e.g., use is not permitted in food and feed processing areas).As all Parties are entitled to this exemption, Parties using the exemption will not be named in the Register as is the case for other intentionally produced POPs. Parties will make determined efforts to identify, label and remove from use equipment containing more than 0.005% (50 ppm) of PCBs, with higher priority given to equipment containing higher levels of PCBs. There is to be no trade in PCB equipment except for the purpose of environmentally sound waste management and, except for maintenance and servicing operations, there is to be no recovery for reuse in other equipment of liquids with more than 0.005% PCBs.Another goal is to achieve the environmentally sound management of PCB wastes as soon as possible but no later than 2028. Parties will report to the COP every five years on their progress in eliminating in-use equipment and the environmentally sound management of wastes, and the COP will review progress toward the 2025 and 2028 targets at fiveyear intervals, taking into account the Parties’ reports.

24

J. Buccini

A special regime has been agreed upon for DDT. The production and use of DDT will be eliminated except for Parties that notify the Secretariat of their intention to produce or use DDT in disease vector control programs, a specified “acceptable purpose” in Annex B. The production and/or use must be in accordance with WHO recommendations and guidelines on the use of DDT, and when locally safe, effective and affordable alternatives are not available to the Party. Such Parties will be included in a special publicly available DDT register maintained by the secretariat. Every three years, registered Parties will report on the quantities used, the conditions of use and the relevance to the Party’s disease management strategy. Each registered Party should develop national action plans to confine the use of DDT to disease vector management, explore alternatives to DDT, and take measures to strengthen health care and reduce the incidence of disease. All Parties will promote research and development to seek alternatives to DDT. The use of DDT will be allowed until such time as technically and economically feasible alternative products, practices or processes are available to countries that are currently reliant on DDT. The COP will review the situation at its first meeting and every three years thereafter to ascertain whether there is a continued need for DDT for disease vector control. There are two specific exemptions allowed for DDT related to its use as an intermediate in manufacturing other chemicals. 3.2.1.2 Provisions Applicable to All Intentionally Produced POPs

For all POPs in Annex A and Annex B, trade will be restricted. In general, imports and exports are limited to shipments intended either for environmentally sound disposal or to Parties with “specific exemptions” under Annex A or Annex B or with “acceptable purposes” under Annex B. Exports to Non-Parties may take place but there are conditions on both the Non-Party and the Party and accountability requirements for the use and disposal of POPs. The control provisions applicable to intentionally produced POPs listed in Annex A or B do not apply to those quantities of a chemical: (a) used for laboratory-scale research or as a reference standard; (b) occurring as unintentional trace contaminants in products and articles; or (c) occurring as constituents of articles manufactured or already in use before or on the date of entry into force of a relevant obligation concerning that chemical provided that a Party has notified the secretariat that a particular type of product remains within use within that Party, whereupon the secretariat will make the notification publicly available. One final exemption should be noted that is limited to two substances of the initial 10 (HCB and DDT) but that could be applicable to POPs that are added to the convention in the future. Provision has been made for a Party to produce or use these substances as closed-system site-limited intermediates that are chemically transformed in the manufacture of other chemicals that do not exhibit the properties of POPs. This requires notification to the secretariat of information on the

The Development of a Global Treaty on Persistent Organic Pollutants (POPs)

25

total amounts produced or used, the nature of the site-limited process, and the amount of the HCB or DDT present in the final product. The notifications will be made publicly available. Such production or use is not considered a “specific exemption” and will cease after a period of ten years unless a Party submits a new notification to the secretariat, in which case the period will be extended for another ten years unless the COP decides otherwise. Those Parties with regulatory and assessment schemes for new industrial chemicals or pesticides shall take “measures to regulate with the aim of preventing the production and use of ” new POPs. This measure is one of the precautionary measures included in the convention and is intended to prevent the commercial introduction of new substances that may have POPs properties. In addition, countries with regulatory and assessment schemes for industrial chemicals or pesticides shall, in conducting assessments of in-use substances, consider the screening criteria (in Annex D) for candidates for addition to the Convention. This will allow the identification of possible POPs as soon as possible in these assessment programs. 3.2.2 POPs that are not Intentionally Produced

POPs that are not intentionally produced are listed in Annex C of the convention (i.e., dioxins, furans, HCB, PCBs). For these chemicals, the intent is to reduce their total releases derived from anthropogenic sources, with the goal of “their continuing minimization and, where feasible, ultimate elimination”. In pursuit of this goal Parties are to: (a) develop action plans within 2 years of entry into force of the Convention for them, and implement their plans to identify, characterize and address the release of the chemicals in Annex C. The action plan shall include the following elements: – evaluate current and projected releases, including the development and maintenance of source inventories and release estimates, – evaluate efficacy of the Party’s laws and policies to manage such releases, – develop strategies to reduce releases, – promote education and training on the strategies, – review success of strategies every five years and report to the COP, and – develop a schedule for implementation of the action plan; (b) promote the application of available, feasible and practical measures to achieve realistic and meaningful levels of release reduction or source elimination; (c) promote the development and, where appropriate, require the use of substitute or modified materials, products and processes to prevent the formation and release of the POPs in Annex C; (d) promote, and as provided for in an action plan, require the use of best available techniques (BAT) for new sources within the following industrial source categories (as specified in Part II of Annex C) that have a potential for comparatively high formation and release of POPs to the environment: – waste incinerators (municipal, hazardous or medical waste; or sewage sludge),

26

J. Buccini

– cement kilns firing hazardous wastes, – pulp production involving chlorine, and – thermal processes used in the metallurgical industry (secondary production of aluminum, copper or zinc; sinter plants in the iron and steel industry) and phase in any BAT requirements for such sources as soon as practicable but no later than four years after the entry into force of the Convention for a Party. For these identified new source categories, Parties shall promote the use of best environmental practices (BEP). Guidance on both BAT and BEP will be developed by the COP; and (e) promote, as provided for in an action plan, the use of BAT and BEP for new sources within the following categories (as specified in Part III of Annex C) and for existing sources within all categories in Parts II and III of Annex C: – open burning of wastes (including landfill sites), – thermal processes in the metallurgical industry not specified in Part II, – residential combustion sources, – fossil-fuel fired utility and industrial boilers, – firing installations for wood and other biomass fuels, – chemical production processes releasing unintentionally produced POPs (e.g., production of chlorophenols and chloranil), – crematoria, – motor vehicles, especially those burning leaded gasoline, – destruction of animal carcasses, – textile and leather dying and finishing, – shredder plants for the treatment of end-of life vehicles, – smouldering of copper cables, and – waste oil refineries. Annex C of the Convention also contains guidance on BAT and BEP and on general measures to prevent the formation and release of unintentionally produced POPs. 3.2.3 POPs in Stockpiles or Wastes

For POPs that are in stockpiles or wastes, the goal is to ensure the environmentally sound management of stockpiles, wastes, and products and articles upon becoming wastes that consist of, contain or are contaminated by POPs. To this end Parties shall: (a) develop and implement strategies to identify the stockpiles, products and articles in use, and wastes containing POPs; (b) manage stockpiles in a safe, efficient and environmentally sound manner until they are deemed to be wastes; (c) take measures to handle, collect, transport and store wastes in an environmentally sound manner and dispose of these wastes in a way that destroys the POP content, or otherwise in an environmentally sound manner taking into account international rules, standards and guidelines;

The Development of a Global Treaty on Persistent Organic Pollutants (POPs)

27

(d) not allow recovery, recycle, reclamation, direct reuse or alternative uses of POPs; (e) not transport these materials across international boundaries without taking into account international rules (e.g., Basel Convention); and (f) develop strategies for identifying contaminated sites and, if remediation is attempted, do it in an environmentally sound manner. 3.3 Procedure for Adding New POPs

The Convention will be able to respond as new scientific and other information becomes available on other chemicals as provision has been made for the addition of new POPs through application of scientific criteria and an agreed process for evaluation of candidates that will be proposed by Parties in the future.A POPs Review Committee will be set up to advise the COP on the merits of proposals submitted by Parties that must address the following criteria contained in Annex D: – chemical identity (names, CAS number, IUPAC name, structure, etc.); – persistence: – evidence that the half life of the chemical in: – water is greater than 2 months, or – soil is greater than 6 months, or – sediment is greater than 6 months, or – evidence that the chemical is sufficiently persistent to warrant consideration under the Convention; – bio-accumulation: – evidence that the bioconcentration factor or the bioaccumulation factor in aquatic species for the chemical is greater than 5000, or absent such data, that the log KO/W is greater than 5, or – evidence that a chemical presents other reasons for concern (e.g., high bioaccumulation in other species, high toxicity or ecotoxicity), or – monitoring data in biota indicating that the bioaccumulation potential of a chemical is sufficient to warrant consideration under the Convention; – potential for long range transport: – measured levels of the chemical in locations distant from the sources of release that are of potential concern, or – monitoring data showing that long-range environmental transport of the chemical may have occurred, or – environmental fate properties and/or model results that show that the chemical has a potential for long-range environmental transport: for a chemical that migrates significantly through the air, its half life in air should be greater than 2 days; – adverse effects: – evidence of adverse effects to human health or the environment that justifies consideration under the Convention, or – toxicity or ecotoxicity data indicating the potential for damage to human health or the environment.

28

J. Buccini

In submitting a proposal, a Party must address these criteria and include a statement of the reasons for concern and the need for global control. The secretariat will review the proposals and submit complete ones to the POPs Review Committee for examination. If the Committee is not satisfied that the screening criteria have been fulfilled, the proposal is set aside, although there are provisions for a Party to resubmit a proposal. If the committee is satisfied that the screening criteria have been fulfilled, then the proposal and the committee’s report are made publicly available and all Parties and observers are invited to submit the following information specified in Annex E for development of a risk profile that further elaborates on and evaluates the information in Annex D and that submitted on Annex E: – sources (production data, uses, releases, etc.), – hazard assessment for the endpoint(s) of concern, – environmental fate (chemical and physical properties, persistence, environmental transport, degradation and transformation, etc.), – bioconcentration or bioaccumulation factor, – monitoring data, – exposure and bioavailability data, – national and international risk evaluations, assessments or profiles, – hazard classification and labeling information, and – status of the chemical under international conventions. If, on the basis of the risk profile, the Committee is not satisfied that the proposal should proceed, it will be set aside, although there are provisions for a Party to request reconsideration and ask for more information to be submitted within a period of one year. If the committee is satisfied that the proposal should proceed, then the proposal and the committee’s report are made publicly available and all Parties and observers are invited to submit the following information specified in Annex F for development of a risk management evaluation that includes an evaluation of possible control measures for the chemical, encompassing the full range of options, including management and elimination. Relevant information should be provided relating to socio-economic considerations associated with possible control measures to enable a decision to be taken by the COP: – efficacy and efficiency of possible control measures in meeting risk reduction goals (technical feasibility; and costs – including environmental and health costs); – alternative products and processes (technical feasibility; costs – including environmental and health costs; efficacy; risk; availability; and accessibility); – positive and/or negative impacts on society of implementing possible control measures [health – including public, environmental and occupational health; agriculture, including aquaculture and forestry; biota (biodiversity); economic aspects; movement towards sustainable development; and social costs]; – waste and disposal implications, in particular, obsolete stocks of pesticides and clean-up of contaminated sites (technical feasibility; and cost); – access to information and public education;

The Development of a Global Treaty on Persistent Organic Pollutants (POPs)

29

– status of control and monitoring capacity; and – any national or regional control actions taken, including information on alternatives, and other relevant risk management information. The Committee shall, based on the risk profile and the risk management evaluation, recommend whether the chemical should be considered by the COP for listing in Annexes A, B and/or C. The COP, taking due account of the Committee’s recommendations and any scientific uncertainty, shall decide, in a precautionary manner, whether to list the chemical and specify its related control measures in Annexes A, B and/or C. This process of evaluating nominations incorporates precaution in a number of ways to ensure that all possible candidates are thoroughly evaluated on the basis of available scientific data to see if they possess the properties that would indicate POPs behavior.As mentioned above, there are safeguards in the process to ensure that all Parties have the opportunity to get a full hearing on any nominated candidate. 3.4 Financial and Technical Assistance

One of the key features of the Convention is the recognition that developing countries and countries with economies in transition will need technical and financial assistance in order to meet their obligations as Parties to the Convention. Regional and subregional centres will be established for capacity building and transfer of technology to assist countries in need. The developed countries have undertaken to provide technical assistance and new and additional financial resources to meet the agreed full incremental implementation costs, and the Global Environment Facility has been named as an interim financial mechanism to handle the funding of capacity building and other related activities. It is expected that projects will begin immediately after the Stockholm conference in May 2001 to enable developing countries to prepare to meet their future convention requirements.

4 Future Actions on POPs The POPs Convention was adopted at a diplomatic conference that was held from May 22–23, 2001, in Stockholm. Since then 150 countries and the European Union have signed and 10 countries have ratified the convention. When 50 countries have ratified the convention, it will come into force. While this process will take a few years, that does not mean that action on POPs will not take place for years to come: far from it! At the Stockholm conference, countries agreed to continue with meetings of the Intergovernmental Negotiating Committee that developed the Convention to coordinate and promote activities on POPs reduction and elimination and lay the groundwork for the eventual first meeting of the COP. The process of negotiation stimulated widespread interest and concern about the effects of POPs and many

30

J. Buccini: The Development of a Global Treaty on Persistent Organic Pollutants (POPs)

actions have taken place or are underway. UNEP has developed a master list of actions that reports that over 108 countries have already taken or are taking some sort of action on POPs. Intergovernmental organizations and non-governmental organizations have similarly responded by taking actions to reduce and/or eliminate POPs generation and release. For example, during the negotiation process, environmental non-governmental organizations (ENGOs) established an International POPs Elimination Network (IPEN) that has now grown to include over 300 ENGO’s, and they are promoting action on POPs around the world. Thus, there is already underway a large effort by all sectors of society to deal with the POPs issue and to prevent the introduction of new POPs into commerce. Those interested in obtaining additional information on POPs activities should consult the UNEP POPs home page or the UNEP Chemicals POPs Team (11–13 chemin des Anemones, 1219 Chatelaine/Geneva, Switzerland, Tel.: +41-22-917-1234; Fax: +41-22-797-3460; e-mail: [email protected]).

CHAPTER 3

Criteria for Additional POPs Bo A. Wahlström UNEP Chemicals, 11-13, Chemin des Anémones, 1219 Châtelaine, Genève, Switzerland E-mail: [email protected]

The Intergovernmental Negotiating Committee for a global treaty on POPs developed sciencebased screening criteria for identifying additional substances as candidates for future international action. The screening criteria relate to properties, e.g., persistence, bio-accumulation, potential for long-range environmental transport and adverse effects. Numerical cut-off values have been agreed for persistence in different media and for bio-accumulation but not for longrange environmental transport potential and adverse effects. The cut-off values may be modified by certain external factors, e.g., very high toxicity. Data for individual nominated substances or groups of substances will be compared to the screening criteria as a first step in the assessment. At later steps an in-depth assessment will be undertaken. Keywords: Criteria, POPs, Persistence, Bio-accumulation, Long-range transport, Pollutants

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

1

Background

2

Recent Regional and International Initiatives

3

General Characteristics of the Criteria and their Application . . . 33

4 4.1 4.2 4.3 4.3.1 4.3.1.1 4.3.1.2 4.3.2 4.3.3 4.3.4 4.3.5 4.4 4.5

The POPs Criteria Process . . . . . . . . . . . . . Overriding Principles for the Criteria Work . . . Characteristics of POPs . . . . . . . . . . . . . . Parameters to Consider . . . . . . . . . . . . . . Persistence . . . . . . . . . . . . . . . . . . . . . Persistence in Air . . . . . . . . . . . . . . . . . . Persistence in Water, Soil and Sediments . . . . . Bio-Accumulation . . . . . . . . . . . . . . . . . Long-Range Environmental Transport . . . . . . Volatility . . . . . . . . . . . . . . . . . . . . . . Toxicity . . . . . . . . . . . . . . . . . . . . . . . Inherent Problems in the Application of Numerical and other Criteria . . . . . . . . . . . . . . . . . Bio-Diversity and Other Factors . . . . . . . . . .

5

Criteria in the Global Negotiating Process

6

Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . 42

. . . . . . . . . . . 33 . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

34 34 34 35 36 36 37 37 38 38 39

. . . . . . . . . 39 . . . . . . . . . 40

. . . . . . . . . . . . . 40

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

32

B.A. Wahlström

Annex D of the Stockholm Convention on POPs . . . . . . . . . . . . . . 42 Information Requirements and Screening Criteria . . . . . . . . . . . . . 42 Annex E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Information Requirements for the Risk Profile . . . . . . . . . . . . . . . 43 Annex F . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Information on Socio-Economic Considerations . . . . . . . . . . . . . . 44 7

References

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

1 Background In May 1995 the 18th meeting of the UNEP Governing Council (GC) adopted decision GC18/32 [1] inviting the Inter-Organization Programme for the Sound Management of Chemicals (IOMC), working with the International Programme on Chemical Safety (IPCS) and the Intergovernmental Forum on Chemicals Safety (IFCS) to initiate an expeditious assessment process on persistent organic pollutants, starting with an initial list of twelve substances. It further invited the IFCS to develop, based on the result of the assessment process and the outcome of the Washington Conference to Adopt a Global Programme of Action for the Protection of the Marine Environment from Land-based Activities,“recommendations and information on international action, including such information as would be needed for a possible decision regarding an appropriate international legal mechanism.” In response to this decision the IFCS established an ad hoc Working Group on Persistent Organic Pollutants (POPs), which met in Manila in June 1996. In the report of the meeting [2], the Working Group recommended that immediate international action should be initiated to protect human health and the environment through measures which would reduce and/or eliminate emissions and discharges of the 12 POPs specified in UNEP GC Decision 18/32. The Working Group also highlighted “the need to develop science-based criteria and a procedure for identifying additional POPs as candidates for future international action” and recommended “that the proposed INC should be directed to establish, at its first meeting, an expert group to carry out this work”. Based on the recommendations of the IFCS ad hoc Working Group the UNEP GC at its 19th meeting in its Decision 19/13C invited the Executive Director of UNEP to convene an intergovernmental negotiating committee to prepare an international legally binding instrument on POPs [3]. The GC also requested the INC to establish an expert group to develop criteria and a procedure for identifying additional POPs. The Criteria Expert Group (CEG) for POPs was established at the first session of the INC in Montreal in June 1998. The CEG met twice, in October 1998 in Bangkok [4] and in June 1999 in Vienna [5], and developed proposals for crite-

Criteria for Additional New POPs

33

ria and a procedure for identifying additional POP that were submitted to the third session of the INC in Geneva in September 1999.

2 Recent Regional and International Initiatives The Working Group on Strategies (WGS) under the 1979 Geneva Convention on Long-Range Transboundary Air Pollution (LRTAP) under the auspices of the UN Economic Commission for Europe (UNECE) started work in the early 1990s to prepare a protocol on persistent organic pollutants under the regional convention. Several working papers were developed in which criteria on persistence, bioaccumulation and toxicity were elaborated and tested on a set of chemicals. As a result the Executive Body (EB) of the LRTAP Convention adopted Decision 1998/2 on criteria to be used to identify additional substances [6]. The EB Decision listed numerical criteria for persistence in air, water, soil and sediments as well as for volatility and bio-accumulation. It also described data elements to be gathered for a more comprehensive assessment to develop a risk profile. The POPs Protocol was adopted in June 1998 in Aarhus, Denmark [7]. Some other initiatives to develop criteria for persistent organic pollutants should be mentioned: – In July 1998 the Society for Environmental Toxicology and Chemistry (SETAC) hosted a workshop in Fairmont Springs, Alberta, Canada, to develop a strategy for the selection and assessment of persistent toxic substances. The process developed did not contain any numerical figures but outlined a stepwise open and transparent procedure that would maximise the use of scientific data and allow for input from interested parties at all stages of the process [8]. – The Commission on Environmental Co-operation (CEC) established under the North American Agreement on Environmental Co-operation (NAAEC) similarly has developed criteria for identifying candidate substances for regional action including persistent toxic bio-accumulators [9]. – Canada in 1994 developed a Toxic Substances Management Policy (TSMP) which listed criteria, e.g., persistence and bio-accumulation for organic substances designated for virtual elimination [10]. – A Swedish Government Committee on Chemicals Policy has recently (Spring 2000) proposed criteria for identifying persistent and bio-accumulating substances with a view towards their elimination [11].

3 General Characteristics of the Criteria and their Application Existing systems for selecting and evaluating POPs generally consider properties such as persistence, bio-accumulation and potential for long-range transport at an early stage in the selection process. These properties are considered separately in a step-wise process, which in some cases starts with properties that facilitate transport, e.g., volatility and persistence in air (UNECE) while others start with screening data on persistence in all media (POPs INC). In addition, data on bioaccumulation and evidence of long-range transport is assessed together with ev-

34

B.A. Wahlström

idence of adverse effects on health and environment in the screening assessment step. At a later stage more comprehensive data on effects, uses etc. are used together with available exposure data to assess potential risks.

4 The POPs Criteria Process 4.1 Overriding Principles for the Criteria Work

In a paper presented to the 1st POPs Intergovernmental Negotiating Committee the author outlined some desired general characteristics of the criteria to be developed [12]. First of all, criteria should be based on sound science. Such criteria should be able to stand the test of time and not be influenced by changing social and political conditions. The stress on the need to establish science-based criteria also reflects the awareness that science in related fields is advancing with regard to our understanding the factors that determine the behaviour of chemicals in the environment and in organisms, including human beings. A better understanding will also in the long run lead to potentially better prediction of the behaviour of suspected chemicals that lack a comprehensive data base. Secondly, criteria should be open and transparent. In the present context this means that they should be understandable to the educated lay public and to policy makers. The procedure whereby a chemical is nominated, evaluated and accepted or rejected as a candidate POP should be easily understood. The process should also include steps for parties, to contribute new, relevant data to modify, clarify, reject or confirm decisions to move a chemical along its path towards its final designation by a Conference of Parties to the Stockholm Convention on Persistent Organic Pollutants (POP) as a Persistent Organic Pollutant of global concern. 4.2 Characteristics of POPs

Chemicals that are POPs are multi-compartment substances that typically move between the environmental compartments of air, water, soil, sediment and biota. In each compartment spread by diffusion needs to be considered, as well as bulk movement with moving media, e.g., air currents, ocean currents, rivers or aquifers. POPs may also move with biota. Migratory butterflies, birds or cetaceans might each constitute a quantitatively small route of dispersal of POPs, but could be important routes of exposure for other organisms. In the process to select candidate POPs all routes of environmental dispersal should be considered. Most potential POPs under discussion contain aromatic or alicyclic rings. Many, but not all, contain carbon-halogen bonds that, to a varying degree, resist degradation by physical, chemical or biological means. They may be characterized as semi-solids with low, but distinct, vapour pressures and the lipid solubility is usually several orders of magnitude higher than the water solubility.A typical POP would be released to air, directly from emissions, or indirectly through

Criteria for Additional New POPs

35

evaporation from water, soil or vegetation. Once in air it would travel a significant distance with existing winds, generally southwestern on the Northern Hemisphere and northwestern on the Southern Hemisphere. At higher latitudes, or higher altitudes, the substance would condense out of the air due to lower air temperature.When deposited on water, soil or vegetation, it would remain in that medium for a long time. It would redistribute itself between lipid and aqueous phases in media and biota with the net result being a high concentration in lipids and a low one in water. In water, soil and sediment, lipids are mainly found in biota. In cold regions thick fatty layers in warm-blooded animals protect against heat loss and also provide an important energy source. They also become significant storage sites for POPs. Predatory animals in cold regions, including man, need fat as an energy source because of its high specific energy content, thus enhancing the tendency of POPs to accumulate in the food chain. In warmer climates exposures may occur closer to the source, e.g., occupational exposure during use, or local exposure caused by run-off from use or leaking from stockpiles. Food, such as fish may be a major route of intake also in warmer climates and POPs may accumulate in the food chain and reach high levels in predatory species in these conditions. The scientific knowledge on the behaviour of POPs in the environment is increasing. Models for predicting the distribution of chemicals in environmental media are coming closer to agreement with environmental data. Consistent criteria for persistence based on scientific models may soon be developed for all environmental media and the environmental distribution and potential for long-range transport of chemicals, including POPs, scientifically evaluated by models. 4.3 Parameters to Consider

For the Stockholm Convention, the following parameters have generally been identified to be considered primarily for setting criteria to identify POPs: (a) Persistence: The ability of a substance to resist degradation in various media, such as air, soil, water and sediment, generally measured as the half-life of the substance in the medium, i.e., the time taken for the concentration of the substance to decrease by 50%; (b) Bio-accumulation: The ability of a substance to accumulate in living tissues to levels higher than those in the surrounding environment, expressed as the quotient between the concentration in the target tissue and the environmental concentration; (c) Toxicity: The ability of a substance to cause injury to humans or the environment; (d) Potential for long-range transport: The ability of a substance to travel long distances, hundreds and thousands of kilometers, from its point of origin. This property, that may be either evaluated by modeling or measured by sampling of biota in remote regions, is considered to be critical for identifying a chemical as a persistent organic pollutant of global concern;

36

B.A. Wahlström

In addition it has been proposed that chemicals which are biodegradable but which may give rise to persistent exposure to the environment due to continuous releases or emissions should also be considered as persistent. The issue of bioavailability has also been raised. Some of the above criteria lend themselves to the assignment of numerical cut-off values, while others need to be qualitatively assessed. In the actual situation of applying the criteria to an individual substance other factors, e.g., climate need to be considered. 4.3.1 Persistence

Persistence reflects the ability of the substance to resist degradation by physical, chemical or biological means. Persistence may be expressed in various ways. In laboratory testing of new and existing chemicals for the assessment of hazard and risk, the Organisation for Economic Cooperation and Development (OECD) Test Guidelines are often used together with the (OECD) Principles for Good Laboratory Practice, as adopted under the OECD Council Act on Mutual Acceptance Data [13–15]. There are guidelines for testing the easy biodegradability of substances as well as for measuring their inherent biodegradability. These, however, are primarily focused on differentiating between chemicals that are easily biodegradable and those that are not. Persistent chemicals in the present sense constitute a subset of those that are not easily biodegradable, and may thus not be adequately identified using existing methods. Discussions are underway to refine and develop existing methods to make them better suited for application to the evaluation of POPs. Persistence depends not only on the substance tested but also on the test medium. Therefore, even seemingly simple degradation tests may give results that are open to interpretation. Any single figure for half-life in a medium, without the corresponding information on the test conditions should be treated with extreme caution. Data from non-standard tests should be handled the same way. Comparisons with substances for which extensive data exist (bench-marking) may facilitate the evaluation. 4.3.1.1 Persistence in Air

The normally used persistence criterion for air is the half-life in air, expressed in days [6]. Modeling shows that substantial quantities of substances with a half-life of two days or more still remain in air after 8–10 days. During that time the substance may be transported several thousand kilometers. Substances may be transported in air absorbed to particles, which increases their half-life. Air persistence data should be considered together with the presence of the substance in remote regions, as demonstrated by the monitoring of biota or other media.

Criteria for Additional New POPs

37

4.3.1.2 Persistence in Water, Soil and Sediments

The half-life of a substance in soil, water or sediments has been proposed as a possible criterion to identify POPs [6]. Chemicals with long persistence half-lives in water, soil or sediments have a high potential for accumulation in the medium and also for uptake by living organisms. The pattern of environmental release or application is of great importance. Substances that are applied once annually, for example some pesticides, may have half-lives of several months and still not accumulate in soil in spite of long-term use. Most industrial chemicals, however, are not used in this way. Releases and emissions usually occur continuously from many sources, including diffuse releases from products or articles containing the substance. In practice, there is no clear demarcation line between persistent and non-persistent chemicals.Also laboratory data have to be applied with caution to real-life situations. Physical, chemical and biological factors such as temperature, pH and amount and content of biological fraction may greatly influence persistence under field conditions, as well as processes such as photolysis and hydrolysis. Substances that normally degrade in the environment may behave as persistent under some conditions, e.g., atrazine in ground water and vice versa degrade more quickly in the environment than indicated by laboratory tests, e.g., pyrethroids in microcosms or field trials. 4.3.2 Bio-Accumulation

Bio-accumulation is a measure of the potential for a chemical to concentrate in living tissues.While persistent substances are diluted by dispersion during longrange transport, bio-accumulation counteracts this process. Bio-accumulation can best be measured in intact organisms in the laboratory or in the field. It is usually expressed as the bio-concentration factor (BCF) or bio-accumulation factor (BAF). Bio-accumulation measured in this way confirms that uptake takes place and integrates accumulation with biodegradation by the organism. Values between 1,000 and 5,000 for BCF in fish have been proposed as cut-off criteria for identifying bio-accumulating substances [6, 8, 11].Among factors that influence BCF are choice of species, study design, lipid content of the organism and others. Systematic testing of existing chemicals is ongoing in several national, regional and international programmes, but, to date, only a fraction of all commercially available chemicals has been studied for bio-accumulating properties. In the absence of data from animal testing, the octanol-water partition coefficient (Kow) has been used as a surrogate. It can be measured relatively easily and even calculated on the basis of the molecular formula and structure and for many substance groups it correlates well with BCF. It must be noted, however, that it should primarily be used as a screening tool, since by itself it will not tell whether a chemical is actually taken up by the organism, or, if taken up, whether it is actually accumulated. Chemicals with molecular weights higher than 1,000 may have a high Kow, but such large molecules are, in general, not bioavailable because they

38

B.A. Wahlström

cannot pass biological membranes. There are also chemicals with high Kow values that are extensively metabolized into less persistent products. A high Kow (>10,000) should therefore always be confirmed by testing the BCF in an animal species. At present, methods are available for aquatic species only. 4.3.3 Long-Range Environmental Transport

The best way to establish whether or not long-range transport occurs is through direct measurement of POPs, e.g., in monitoring programmes in remote locations such as the Arctic or Antarctic, isolated islands, or mountain areas. Measurements in biota and human populations also provide data that may be used in risk assessments. Long-range transport may include regional transport within a continent, such as transport from a densely populated coastal area with intensive agriculture to inland mountainous regions. The potential for long-range transport may be assessed indirectly by persistence times in air, water or soil and by factors such as volatility. The behaviour of persistent chemicals in the environment is dependent, however, on a host of other factors, such as adsorption to particles, soil binding, etc., which makes prediction difficult. Long-range transport of POPs may also occur through migratory birds that accumulate substantial amounts of POPs because of local use in their tropical winter quarters. Some of these fall prey to predatory birds in their summer quarters, thus transferring their POPs content up the food chain but also from one part of the globe to another.Although rough calculations show this transport to be of minor quantitative importance in relation to air and water it might still be of importance as a route of exposure for specific predatory species. The marine transport through currents, or through repeated dissipation and condensation, as well as through migrating marine species is thought to an order of magnitude lower than the air transport. However, residence times for POPs in water are several orders of magnitude higher than in air, and the exchange mechanisms are slow enough to allow for continued releases from the oceans over several centuries or millennia. The recently evidenced presence of POPs in the deep sea at depths of 1000 meters and more are a cause of continued concern. The long-range environmental transport criterion is basically qualitative in character and needs to be assessed on a case-by-case basis for each POP substance. 4.3.4 Volatility

Volatility is sometimes considered as a separate criterion [6] and sometimes viewed as one important property together with other environmental fate properties (see below).Volatility is usually expressed as the vapour pressure of a substance.A volatility of less than 1,000 pascal has usually been assigned as a cut off value to distinguish between very volatile and semi-volatile substances. Chemicals that have a vapour pressure higher than 1,000 pascal are gases at normal tem-

Criteria for Additional New POPs

39

peratures, distribute exclusively to air and are not likely to be POPs. The volatility criterion should be applied together with persistence in air and/or data on presence in remote regions. Other ways to indicate a substance’s tendency to volatilize have been suggested, such as Henry’s law constants and fugacity calculations. It has been suggested that many factors, including volatility contribute to the fate of a substance in the environment. If the focus is on POPs in general and not specifically on long-range transboundary air transport, volatility becomes less critical as a property for criteria setting. 4.3.5 Toxicity

International schemes exist for identifying some acutely toxic chemicals, e.g., WHO [16]. However, for the long-term and chronic effects, some of which may appear at extremely low levels, there are at present no internationally agreed quantitative toxicity criteria for identifying POPs. Therefore, a qualitative approach needs to be applied taking into account a wide variety of toxicity and eco-toxicity endpoints. Chronic and irreversible effects should be assessed differently from acute and transient effects. Assessment of toxicity also requires an assessment of dose. Substances of high toxicity may cause concern even when they are present in the environment in very low concentrations, e.g., if they bioaccumulate to a significant degree. Adverse effect levels should, where possible, be compared to possible exposures. Toxicity is therefore essentially a qualitative parameter at the screening stage for presumptive POPs. 4.4 Inherent Problems in the Application of Numerical and other Criteria

Because of the inherent biological variation as well the normal statistical variation inherent in any physical or chemical measurement all numerical criteria must be applied with caution and judgement. Deficiencies in the scientific database for a substance may necessitate caution in using single numbers outside their proper context. The uncertainty related to any kind of scientific measurement is of particular importance in cases where the database consists of a single measurement. Similarly, numbers close to the cut-off values must be thoroughly assessed, regardless of whether they are slightly below or slightly above the cutoff. When several criteria are judged together one might consider a certain substance as fulfilling the criteria if two or more of the criteria are more than adequately fulfilled while one criterion may be just marginally fulfilled or not quite fulfilled. A more generic problem is the lack of appropriate methods for predicting what actually happens in the environment.While at present there is a lack of mutually agreed assessment procedures for chemicals this is something that needs to be developed in the process of applying the criteria. With regard to data quality and validity, data might have been generated under a wide variety of test methods and conditions. All available data should be considered, and scientific judgement should be applied by putting more empha-

40

B.A. Wahlström

sis on results generated under standardized conditions using widely accepted test methods and laboratory practices. Results obtained under non-standardized test conditions or without recognized test methods might also be considered, as they might sometimes be more appropriate to the issues of concern. Uncertainties in the database frequently give rise to different interpretations between countries or stakeholders.An important part of the criteria development process is to make sure that these divergences are eliminated to the fullest extent possible and that the reasons for any remaining differences are thoroughly understood. A prerequisite for this to happen is a clear understanding of how political considerations, particularly in the field of science policy, influence assessments of hazard and risk. Countries may make various assumptions in their interpretation of scientific data, for example, accepting one well-conducted lifetime study in one species as sufficient evidence of chronic effects, or always taking the “worst” value for an effect parameter. This might reflect the level of protection desired in a country. In the face of the number of possible candidates and the insurmountable costs of testing all substances to a satisfactory degree, predictive tools, e.g., accurate screening methods and fate modeling to identify probable POPs with a potential for long-range transport need to be developed. In applying criteria the strengths and weaknesses of existing methods should be identified and recommendations for method improvement made, as appropriate. The ultimate proof of long-range environmental transport is the presence of POPs in remote environments, such as the Arctic, as evidenced by monitoring data. There is a need for further development of existing monitoring programmes and of establishing such programmes in regions where they do not exist, including looking for new possible POPs. However, monitoring programmes will only reveal pollution that already has occurred and cannot substitute for taking proper preventative action at the source against substances with suspected or confirmed POPs properties. 4.5 Bio-Diversity and Other Factors

Special considerations may influence the application of criteria, such as the need to conserve bio-diversity and the protection of endangered species. For instance, it might be necessary to apply stricter cut-off values for POPs when the diversity of economically, socially and culturally important biota is at stake, or where individual species are threatened [12].

5 Criteria in the Global Negotiating Process The Criteria Expert Group established by the Intergovernmental Negotiating Committee quickly came to an agreement on the properties to consider for screening criteria [4, 5]. For an initial assessment data on persistence, bio-accu-

Criteria for Additional New POPs

41

mulation, potential for long-range transport and adverse effects should be considered. In view of the multi-compartment behaviour of these substances, persistence in water, soil and sediments were considered of equal importance. Persistence in air should be considered in conjunction with the potential for long-range transport for substances that signficantly migrate through the air and for such substances the half-life in air should be greater than 2 days. In addition, volatility was viewed as a property that should be considered together with other physical and chemical properties of importance for determining the environmental fate of a substance. For persistence in water, soil and sediments numerical cut-off values for halflives in these media were proposed. For half-life in water, two months was agreed, while for soil and sediment there was agreement to use a half-life of six months. Similarly, for bio-accumulation there was agreement to use a cut-off value of 5000 for BCF/BAF in aquatic species. In absence of BCF/BAF data, logKow should be greater than 5. It should be noted that high bio-accumulation in other species or high toxicity or eco-toxicity might be sufficient to fulfill the criteria for bio-accumulation. There is also agreement that when the assessment passes on to the more in-depth stage measured values of BCF/BAF are needed, i.e., logKow as a substitute will not be sufficient. For the other two criteria; potential for long-range transport and adverse effects, no numerical values have been suggested. With regard to environmental fate there are many environmental fate properties and data which are relevant for assessing long-range environmental transport. Those properties and data, many of which were relevant to several of those areas, may be grouped into those relevant for transport (vapour pressure, Henry’s Law constant; water solubility; studies relevant to local, regional, or global environmental transport; particle dispersion; density; etc.), transfer [log Kow, other partition coefficients, water solubility, molecular weight, molecular size, bio-concentration factor (BCF), bio-accumulation factor (BAF), etc.], and transformations (molecular structure, half-lives in various environmental media, and many of the properties and data noted above). In addition a broad interpretation of the terms “toxicity” and “eco-toxicity” should be used. These terms are intended to cover a broad scope of adverse endpoints as might be determined in a variety of controlled in vivo and in vitro laboratory studies, field studies of biota, and epidemiology studies. Furthermore, effects observed or reported could be associated with a variety of single, multiple, intermittent or continuous exposures, could be immediate or delayed, or could be short-term or chronic in their duration. The experts agreed that a certain substance may be considered as fulfilling the criteria if two or more of the criteria are more than adequately fulfilled while one criterion may be just marginally fulfilled or not quite fulfilled. It was also agreed that the assessment process should include the consideration of transformation products of the substance that possessed POPs characteristics. In this regard organic substances that are not in themselves POPs, but whose transformation products satisfy the criteria should be considered.

42

B.A. Wahlström

The final text, agreed to at the last meeting of the Intergovernmental Negotiating Committee, is presented in the Appendix to this paper.

6 Concluding Remarks A substantial scientific body of knowledge is available to guide us in designing a screening criteria process in which suspected substances may be judged in a step-wise procedure going from simple numerical values for individual properties to a full and comprehensive evaluation of all its impacts. Scientific developments in chemistry, biology, toxicology and related fields can only improve upon the present process to make it more robust and useful for the future. However, at present attempts at developing exact, science-based criteria for identifying with certitude all persistent organic pollutants warranting global action and assigning rigid numerical cut-off values for all selected properties are likely to fail because of the limited scientific knowledge of the underlying processes.

Annex D of the Stockholm Convention on POPs Information Requirements and Screening Criteria

1. A Party submitting a proposal to list a substance in Annexes A, B or C shall identify the chemical in the manner described in subparagraph (a) below and provide the information on the chemical, and its transformation products where relevant, relating to the criteria set out in subparagraphs (b) to (e): (a) Chemical identity, including: (i) Names: trade name or names, commercial name or names and synonyms, Chemical Abstracts Service (CAS) Registry number, International Union of Pure and Applied Chemistry (IUPAC) name; and (ii) Structure, including specification of isomers, where applicable, and the structure of the chemical class. (b) Persistence: (i) Evidence that the half-life of the chemical in water is greater than two months, or that its half-life in soil is greater than six months, or that its half-life in sediment is greater than six months; or (ii) Evidence that the chemical is otherwise sufficiently persistent to justify its consideration within the scope of this Convention; (c) Bio-accumulation: (i) Evidence that the bio-concentration factor or bio-accumulation factor in aquatic species for the chemical is greater than 5,000 or, in the absence of BCF and BAF data, that the log Kow is greater than 5; (ii) Evidence that a chemical presents other reasons for concern, such as high bio-accumulation in other species, high toxicity or ecotoxicity; or

Criteria for Additional New POPs

43

(iii) Monitoring data in biota indicating that the bio-accumulation potential of the chemical is sufficient to justify consideration within the scope of this Convention; (d) Potential for long-range environmental transport: (i) Measured levels of the chemical in locations distant from the sources of its release that are of potential concern; (ii) Monitoring data showing that long-range environmental transport of the chemical, with the potential for transfer to a receiving environment, may have occurred via air, water or migratory species; or (iii) Environmental fate properties and/or model results that demonstrate that the chemical has a potential for long-range environmental transport through air, water or migratory species, with the potential for transfer to a receiving environment in locations distant from the sources of its release. For a chemical that migrates significantly through the air, its half-life in air should be greater than two days; and (e) Adverse effects: (i) Evidence of adverse effects to human health or to the environment that justifies consideration of the chemical within the scope of this Convention; or (ii) Toxicity or ecotoxicity data that indicate the potential for damage to human health or to the environment. 2. The proposing Party shall provide a statement of the reasons for concern, including, where possible, a comparison of toxicity or ecotoxicity data with detected or predicted levels of a chemical resulting or anticipated from its long-range environmental transport, and a short statement indicating the need for global control. 3. The proposing Party shall, to the extent possible and taking into account its capabilities, provide additional information to support the review of the proposal referred to in paragraph 6 of Article 8. In developing such a proposal, a Party may draw on technical expertise from any source.

Annex E Information Requirements for the Risk Profile

The purpose of the review is to evaluate whether the chemical is likely, as a result of its long-range environmental transport, to lead to significant adverse human health and/or environmental effects, such that global action is warranted. For this purpose, a risk profile shall be developed which further elaborates on, and evaluates, the information referred to in Annex D and includes, as far as possible, the following types of information: (a) Sources, including as appropriate: (i) production data, including quantity and location; (ii) uses; and (iii) releases, such as discharges, losses and emissions;

44

B.A. Wahlström

(b) Hazard assessment for endpoint or endpoints of concern: the assessment should include a consideration of toxicological interactions involving multiple chemicals; (c) Environmental fate, including data and information on the chemical and physical properties and persistence of a chemical and how they are linked to its environmental transport, transfer within and between environmental compartments, degradation and transformation to other chemicals. A determination of BCF or BAF, based on measured values, shall be available, except when monitoring data are judged to meet this need; (d) Monitoring data; (e) Exposure in local areas and, in particular, as a result of long-range environmental transport, and including information regarding bio-availability; (f) National and international risk evaluations, assessments or profiles and labelling information and hazard classifications, as available; and (g) Status of the chemical under international conventions.

Annex F Information on Socio-Economic Considerations

An evaluation should be undertaken regarding control measures, for chemicals under consideration for inclusion in this Convention encompassing the full range of options, including management and elimination. For this purpose, relevant information should be provided relating to socio-economic considerations associated with control measures to enable a decision to be taken by the Conference of the Parties. Such information should reflect due regard for differing capabilities and conditions among Parties and should include consideration of the following indicative list of items: (a) Efficacy and efficiency of control measures in meeting risk reduction goals: (i) Technical feasibility; and (ii) Costs, including environmental and health costs; (b) Alternatives (products and processes): (i) Technical feasibility; (ii) Costs, including environmental and health costs; (iii) Efficacy; (iv) Risk; (v) Availability; (vi) Technical feasibility; and (vii) Accessibility; (c) Positive and/or negative impacts on society of implementing control measures: (i) Health, including public, environmental and occupational health; (ii) Agriculture, including aquaculture and forestry; (iii) Biota (biodiversity);

Criteria for Additional New POPs

45

(iv) Economic aspects; (v) Movement towards sustainable development; and (vi) Social costs; (d) Waste and disposal implications (in particular, obsolete stocks of pesticides and clean-up of contaminated sites): (i) Technical feasibility; and (ii) Cost; (e) Access to information and public education; (f) Status of control and monitoring capacity; and (g) Any national or regional control actions taken, including information on alternatives, and other relevant risk management information.

7 References 1. UNEP Governing Council (1995) Decision 18/32 2. IFCS ad hoc Working Group on Persistent Organic Pollutants (1996) Manila, The Philippines 3. UNEP Governing Council (1997) Decision 19/13C 4. UNEP/POPS/INC/CEG/1/3 (1998) Report of first meeting of Criteria Expert Group for POPs. Bangkok, Thailand 5. UNEP/POPS/INC/CEG/2/3 (1999) Report of second meeting of the Criteria Expert Group for POPs. Vienna, Austria 6. UNECE/LRTAP/EB (1998) Decision 2, Geneva, Switzerland 7. UNECE Protocol to the 1979 Convention on Long-range Transboundary Air Pollution on Persistent Organic Pollutants (1998) Aarhus, Denmark 8. Evaluation of Persistence and Long-Range Transport of Organic Chemicals in the Environment, Klecka GM, Boethling RS, Franklin J, Grady CPL Jr, Graham D, Howard PH, Kannan K, Larson RJ, Mackay D, Muir D,Van de Meent D (1998) Proceedings of a SETAC Pellston Workshop on Persistence and Long-Range Transport of Organic Chemicals in the Environment, 13–19 July, 1998, Fairmont Hot Springs, British Columbia, Canada. Society of Environmental Toxicology and Chemistry, Pensacola, FL, USA. 9. Process for Identifying Candidate Substances for Regional Action under the Sound Management of Chemicals Initiative (1997), Montréal, Canada 10. Toxic Substances Management Policy (1995) Ottawa, Canada 11. New Guidelines for Chemicals Policy (2000) Ministry of Environment Stockholm, Sweden 12. UNEP/POPS/INC.1/6 Note from the Secretariat (1998) Montreal, Canada 13. OECD Guidelines for the Testing of Chemicals (1981) Paris, France 14. OECD Principles of Good Laboratory Practice, as revised in 1997 (1998) Paris, France 15. OECD Council Decision on the Mutual Acceptance of Data in the Assessment of Chemicals C(81)30 (Final) (1981) Paris, France 16. WHO Recommended Classification of Pesticides by Hazard (2001) Geneva, Switzerland

CHAPTER 4

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex Vladimir Zitko 114 Reed Ave, St. Andrews, NB, E5B 1A1, Canada E-mail: [email protected]

This chapter provides an overview of history, chemistry, environmental fate and effects of DDT and of the cyclodiene pesticides aldrin, dieldrin, endrin, and mirex, and their metabolites and degradation products. These pesticides are, for the purpose of this review, summarily called the ‘classic organochlorines’ (COC). All cyclodiene COC have been practically phased out. DDT is still used in several countries, to some extent in agriculture, but primarily, to control malaria. Common structural features of COC are the presence of several chlorine atoms and a rigid shape of the molecule. Common properties include very low solubility in water and high solubility in lipids, low but significant vapor pressure, and a strong resistance to degradation. The consequences of this combination of properties include a wide distribution and persistence in the environment and accumulation in biota. A prolonged presence in biota results in a number of chronic effects, not anticipated from short-term toxicity studies. COC and similar compounds are not likely to be again released intentionally into the environment; thus this chapter documents a history which will not, hopefully, be repeated. The story should serve as example of difficulties in anticipating environmental effects of chemicals, unexpected consequences, and changing attitudes and terms of reference. Studies of COC contributed a great deal to the understanding of the behaviour of chemicals in the environment; however, nothing can be done about the trace concentrations of COC now present there. An important lesson has been learned and it is time to start paying attention to the detection, behaviour, and effects of many other chemicals released into the environment. Keywords: Production, Toxicity, Bioaccumulation, Trends

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

1

Introduction

2

Structure and Properties

3

Environmental Concentrations

3.1 3.2 3.3 3.4 3.5 3.6 3.6.1

Air . . . . . . . Soil . . . . . . Water . . . . . Sediment . . . Fauna and Flora Aquatic Fauna Molluscs . . .

. . . .

. . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . . . . . . . . . . . . . . . . . 51

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . . . . . . . . . . . . . . 54 . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

55 57 57 58 61 62 62

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

48

V. Zitko

3.6.2 3.6.3 3.7

Fish . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Marine Mammals . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Humans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

4

DDT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

4.1 4.2 4.3 4.3.1 4.3.2 4.3.3 4.3.4

History . . . . . . . . . . . . . Production . . . . . . . . . . . Effects . . . . . . . . . . . . . . Mode of Action . . . . . . . . . Structure-Activity Relationships Transformations . . . . . . . . Toxicology . . . . . . . . . . .

5

Aldrin and Dieldrin

5.1 5.2 5.3

History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 Mode of Action, Transformations, and Toxicology . . . . . . . . . 79

6

Endrin

6.1 6.2 6.3

History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Transformations and Toxicology . . . . . . . . . . . . . . . . . . 81

7

Mirex

7.1 7.2 7.3

History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 Toxicology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

8

Closing Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

9

References

. . . .

. . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

70 72 74 74 75 75 76

. . . . . . . . . . . . . . . . . . . . . . . . . 77

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

1 Introduction DDT was the first chemical found to be remarkably active against a number of insect pests. Its first large-scale use towards the end of World War II was such a success that the discoverer of the insecticidal properties of DDT, Paul Müller, received the Nobel Prize in 1948. One of the valued properties of DDT was its persistence. There was no need for frequent applications and, in addition, DDT has very low acute toxicity in mammals. For at least another 15 years nobody thought of food chains, bioaccumulation, chronic toxicity, and concentrations of residues in the µg/g (ppm) range. Gas chromatography was a curiosity, the electron capture detector was not invented until 1960 [37], and the analytical chemistry was comfortable with concentrations in the mg% (mg/100 g=10 ppm) range. Even in

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

49

1967, gas chromatography with electron capture detection, thin layer chromatography, and infrared spectrophotometry were the methods of choice to study environmental fate of DDT and metabolites [182]. The definitive confirmation of many pesticide residues was not possible before the introduction of a mass spectrometer as a detector in gas chromatography, in 1971 [18]. A very interesting history of residue analyses in wildlife, is provided by Keith [109]. It was thought that the presence of chlorine atoms in the molecule of DDT may be a source of its insecticidal activity, and attempts to prepare other such compounds were soon underway. Further impetus to this effort were cases of insect resistance to DDT that started to appear within a few years after large-scale DDT applications. The cyclodiene pesticides chlordane, aldrin, dieldrin, and endrin appeared on the scene and were, somewhat later, followed by mirex and its derivative, chlordecone. These insecticides also have a remarkable insecticidal activity and their typical agricultural application rate ranged from less than 0.5 kg/ha to 2 kg/ha, as compared to 5 kg/ha to 10 kg/ha, or even 40–50 kg/ha in northern orchards, for DDT. However, the cyclodienes have much higher mammalian toxicity than DDT and their use has been somewhat more limited. After a few years, the presence of DDT in human tissues was noticed and caused some concern. Between 1956 and 1961 a large number of bird kills was attributed to seeds treated with aldrin and dieldrin [169].At the same time, kills of birds, fish, and mammals were observed, not only in pesticide-treated areas but also at more distant sites. These incidents together with improvements in analytical techniques led to a better understanding of environmental effects and fate of pesticides. All these were summarized by Rachel Carson in her book Silent Spring in 1962. By the early 1970s, p,p′-DDE was discovered as the agent responsible for the eggshell thinning in a number of avian species [28]. Gradually, restrictions were imposed on the indiscriminate use of pesticides, as gas chromatography with the electron capture detector revealed a widespread contamination of humans and the environment by DDT and its main metabolites, DDE and DDD, and by dieldrin. A complication was encountered by the presence of a number of unknown peaks, some very close or overlapping, the peaks of DDT and its metabolites. Substances causing these peaks were identified in 1967 by Jensen as polychlorinated biphenyls (PCBs). It is likely that the presence of PCBs resulted in some erroneously high, early-reported concentrations, particularly of p,p′-DDE, but also of p,p′-DDD and p,p′-DDT. Even without considering PCBs, the separation of p,p′-DDE and dieldrin by gas chromatography on packed columns was difficult, and relatively low concentrations of the latter may have remained undetected. It took another ten years and the advent of gas chromatography on capillary columns and the mass analyzer detector, to achieve a reliable identification and confirmation of organochlorine pesticides. Many orders of magnitude increase in sensitivity, attained by the new analytical techniques, revealed a widespread presence of these compounds in the environment. Organochlorine pesticides were detected even in the most remote parts of the environment and their global circulation patterns were noticed. At this stage the term organochlorine pesticides covered the COC (Classic organochlorine compounds, an acronym introduced for convenience for the title pesticides of this

50

V. Zitko

chapter) and their metabolites, and, in addition, the hexachlorocyclohexanes and toxaphene. Hexachlorobenzene and PCBs were later included because of their equally widespread presence in the environment, and hexachlorobenzene had some applications as a fungicide. The group was referred to as the organochlorine compounds. Recently, this group was expanded and a new term was introduced, the Persistent Organic Pollutants (POPs). This new group is defined on the basis of environmental properties of its members – persistence, bioaccumulation, and adverse effects, and a global, legally-binding international action on POPs is promulgated. As of 1999, the international action for COC includes a complete phase-out of aldrin and endrin, and a partial phase-out of dieldrin and mirex. The case of DDT is still under discussion, mainly because of DDT’s importance in the control of malaria. In parallel with the persistence-based definition of POPs, another group of compounds, defined by their effects on the endocrine system, the Endocrine Disrupting Chemicals (EDCs), was formed. EDCs include COC and, in addition, branched-chain alkyl phenols and other chemicals grouped by their effects rather than by chemical structure or properties. In this connection it is interesting to note that the estrogenic activity of o,p′-DDT, a minor component of technical DDT, was noticed already in 1968, but did not receive much attention at that time. The endocrine disrupting activity of p,p′DDE, a major DDT metabolite and degradation product, displayed by the inhibition of the androgen receptor androgen binding, was discovered only recently [110]. For the last 30 years, about 800 papers per year were published on some aspects of DDT, followed by approximately 200, 130, 90, and 40 papers per year, on the subject of dieldrin, aldrin, endrin, and mirex, respectively (Table 1). The vast majority of the papers deals with concentrations of these compounds in various environmental compartments. The following frequent topic is toxicology, which describes chronic and, increasingly, subtle effects. Relatively few papers investigate time trends of the concentrations. Because of limited space, it is impossible to refer in this chapter to more than a few publications. Consequently, some statements are presented without references. When given, references were selected

Table 1. Average number of papers abstracted by Chemical Abstracts

Years

SDDT

Aldrin

Dieldrin

Endrin

Mirex

1967–1969 1970–1974 1975–1981 1982–1987 1988–1989 1990–1994 1995 1996 1997 1998 1999

494 835 742 536 717 643 853 873 977 846 1115

122 133 104 94 132 140 150 157 170 125 130

192 242 197 129 180 190 213 219 268 227 265

97 89 80 49 82 88 119 110 120 107 107

4 14 42 37 39 46 54 54 36 49 58

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

51

somewhat arbitrarily to cover the literature of the past 40–50 years and to provide a picture of increasing knowledge and of developing environmental awareness and insight. A more general discussion of the latter was published recently [216].

2 Structure and Properties Chemical and common names of COC, as well as their Chemical Abstracts Registry Numbers are in Table 2, properties of environmental interest are in Table 3 for COC as well as for some of their degradation products.‘Benchmark’ toxicity data, illustrated by toxicity to rats, are in Table 4. Structural formulae, metabolites and degradation products are mentioned in sections dealing with the individual pesticides. The solubility of COC in water is extremely low. The determination of such low solubilities is fraught with considerable difficulties. Its is therefore not surprising that the published estimates differ by as much as orders of magnitude. The values, provided in Table 3, are for the most part geometric means of values, colTable 2. Chemical and common names

p,p′-DDT p,p′-DDD Aldrin

Dieldrin

Endrin

Mirex

1,1,1-Trichloro-2,2-bis(4-chlorophenyl)ethane (IUPAC) 1,1′-(2,2,2-Trichloroethylidene)bis[4-chlorobenzene] (CA) CA RN 50-29-3 1,1-Dichloro-2,2-bis(4-chlorpohenyl)ethane (IUPAC) 1,1′-(2,2-Dichloroethylidene)bis[4-chlorobenzene] (CA) CA RN 72-54-8 (1R,4S,4aS,5S,8R,8aR)-1,2,3,4,10,10-Hexachloro-1,4,4a,5,8,8a-hexahydro1,4:5,8-dimethanonaphthalene (IUPAC) (1a, 4a,4ab,5a,8a,8ab)-1,2,3,4,10,10-Hexachloro-1,4,4a,5,8,8a-hexahydro1,4:5,8-dimethanonaphthalene (CA) HHDN (common, pure compound) Aldrin (material containing 95% HHDN) CA RN 309-00-2 (1R,4S,4aS,5R,6R,7S,8S,8aR)-1,2,3,4,10,10-Hexachloro-1,4,4a,5,6,7,8,8aoctahydro-6,7-epoxy-1,4:5,8-dimethanonaphthalene (IUPAC) (1aa,2β,2aa,3b,6β,6aa,7b,7aa)-3,4,5,6,9,9-Hexachloro-1a,2,2a,3,6,6a,7,7aoctahydri-2,7:3,6-dimethanonaphth[2,3-b]oxirene HEOD (common, pure compound) Dieldrin (material containing >85% HEOD) CA RN 60-57-1 (1R,4S,4aS,5S,6S,7R,8R,8aR)-1,2,3,4,10,10-Hexachloro-1,4,4a,5,6,7,8,8aoctahydro-6,7-epoxy-1,4:5,8-dimethanonaphthalele (IUPAC) (1aa,2b,2ab,3a,6a,6ab,7b,7aa)-3,4,5,6,9,9-Hexachloro-1a,2,2a,3,6,6a,7,7aoctahydro-2,7:3,6-dimethanonaphth[2,3-b]oxirene (CA) CA RN 72-20-8 Dodecachloropentacyclo[5.3.0.02,6.03,9.04,8]decane (IUPAC) 1,1a,2,2,3,3a,4,5,5,5a,5b,6-dodecachlorooctahydro-1,3,4-metheno-1Hcyclobuta[cd]pentalene (CA) CA RN 2385-85-5

2.39E –04

7.52E–01

6.79

1.41E –05

Henry’s Law constant, 1.80E+00 Pa.m3/mole

6.19

5.39

Solubility in water, mole/l

logKow

logKoc

6.62

8.5 E –02

1.90E–07

1.35E–06

355

5.00E –03

109

mp

Solubility in water, mg/l

355

MW

789-02-6

o,p-DDT

vp mmHg, 20 °C

50-29-3

CA RN

p,p-DDT

Table 3. Properties of environmental interest

6.64

7.00

2.50E +01

3.46E–05

1.10E–02

6.50E –06

89

318

72-55-9

p,p-DDE

5.78

6.00

2.02E +01

4.09E–05

1.40E–01

6.20E –06

318

3424-82-6

o,p-DDE

5.89

6.02

1.28E +00

1.41E–04

4.50E–02

1.35E –06

109

320

72-54-8

DDD

4.98

5.30

1.07E +01

8.22E–05

3.00E–02

6.60E –06

104–104.5

365

309-00-2

Aldrin

3.62E–04

2.50E–01

3.00E –06

228

381

72-20-8

Endrin

3.23

3.50

4.93

5.20

1.13E +00 1.10E +00

3.54E–04

1.35E–01

3.00E –06

175–176

381

60-57-1

Dieldrin

Chlordecone

491 350

7.38

6.89

5.15

5.41

3.64E+00 5.45E–03

1.10E–05 5.50E–03

6.00E–03 2.70E+00

3.00E–07 2.25E–07

485

546

2385-85-5 143-50-0

Mirex

52 V. Zitko

53

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex Table 4. Oral toxicity to rats [72]

Aldrin p,p′-DDT p,p′-DDE Dieldrin Endrin Mirex Chlordecone

LD50 (mg/kg)

LD1(mg/kg)

Lowest dose to kill (mg/kg)

Male

Female

Male

Female

Male

Female

39 113 880 46 18 740 125

60 118 1240 46 8 600 125

18 52 360 25 5 200 92

27 80 460 25 5 270 92

25 75 750 30 10 400 100

40 100 500 30 6 500 125

lated in [179]. There do not seem to be accurate values for the aqueous solubility of mirex and chlordecone. The vapor pressures of COC are also low, but the ratio of a low vapor pressure to an extremely low solubility in water yields high values of the Henry’s law constants, which determine the distribution of organic chemicals in environmental compartments (see, for example, [129]). Similarly, the values of the octanol/water distribution coefficients are, with the exception of dieldrin, within the ‘optimal range’ (log Kow 5–7) for bioaccumulation (see, for example, [158]).At the same time, the values of adsorption coefficient on organic carbon are also high and indicate that these compounds are strongly bound by organic matter in soil and aquatic sediments. This strong binding extends to dissolved high-molecular weight organic materials, described as fulvic acid, humic acid, or natural organic matter [184]. Such substances may influence strongly the solubility of COC in water. For example, the solubility of p,p′-DDT increases from 5 mg/l in the absence to 35 mg/l in the presence of soil humic acid at 100 mg/l [46]. At the same time, the bioavailability may decrease because of binding of the pesticides to the natural organic matter. However, this subject has not been studied in great detail [26], although it has been suggested as an explanation of decreased bioaccumulation of very highly lipophilic compounds [209]. Estimates of the half-life of COC in the environment range from 365 days for aldrin to 4300 days for endrin. In environmental context, half-lives are difficult to define and the mentioned values are only an indication of the extreme persistence of COC. Models, such as those developed by Mackay [129], present a better picture of the environmental distribution and behaviour. As can be seen from the example of acute toxicity to rats, the toxicity of aldrin, dieldrin, and endrin is considerably higher than that of DDT. The toxicity of mirex approximates that of DDT. On the other hand, the metabolite of mirex, chlordecone, also used as a pesticide on its own, is as toxic as DDT. The toxicity of all the pesticides to aquatic biota is very high. The distribution of toxicity values to fish and crustaceans, expressed in terms of LC50 or IC50, re-drawn from data compiled in [210], is presented in Fig. 1. With the exception of the toxicity of aldrin to crustaceans, the distributions are nearly parallel, with endrin being the most toxic. COC are acutely toxic to practically all species of fish and crustaceans at concentrations below 100 mg/l.

54

V. Zitko

Fig. 1. A summary of aquatic toxicity data, adapted from [210]. A, ac, d, dc, e, ec, ddt, ddtc in-

dicate toxicities of aldrin, dieldrin, endrin and DDT to fish and crustaceans, respectively

According to the toxicity ranking factor [69], the reciprocal of the product of Henry’s Law constant and the acute toxicity LC50, the aquatic toxicity hazard decreases in the order endrin dieldrin>DDT>aldrin.

3 Environmental Concentrations The most frequently reported concentrations are those of p,p′-DDE and p,p′DDT, followed by dieldrin and mirex. The o,p′-isomers of the DDT group, aldrin and endrin are reported much less frequently.Aldrin is seldom found because of its rapid conversion into dieldrin. Endrin has not attracted much attention because of its relatively small-scale use and very low concentrations, yielding a signal possibly hidden in the background of the sample’s analytical signal. The scatter of the measured concentrations is very large, partly due to the variability, typical of environmental matrices, partly because of the complicated analytical techniques, involving in most cases a lengthy cleanup. In addition, the measurement of concentrations in the part-per trillion (pg/g) through part-perbillion (ng/g) to part-per-million (µg/g) is, according to the Horwitz principle, associated with standard deviations ranging from 10–100% [9]. It is important to confirm findings of unexpected presence or of unusually high concentrations. Thus, for example, endrin in concentrations of 1–141 ng/g wet weight was reported in the muscle of a cyprinid fish Barbus xanthopterus, and in concentrations of 11–236 ng/g wet weight in the muscle of Indian shad (Tenualosa ilisha)

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

55

from Iraq [63]. These concentrations are in the same range as those of SDDT, measured in the same fish. Similarly, o,p′-DDD was reported in average concentrations ranging from 111 to 148 ng/g wet weight in eggs of cormorants (Phalacrocorax carbo sinensis) in Greece [116]. For comparison, the average concentrations of p,p′-DDE ranged from not detectable to 0.42 ng/g wet weight. The authors’ explanation is that the elevated concentrations of o,p′-DDD “are due to its presence in zooplankton and in the water column”. It is more likely that an error was propagated through the data evaluation and publication process and, probably, the concentrations of these two compounds were transposed. This emphasizes the need for a careful quality control of the whole process of residue analysis. The results may be reported in µg, ng, or, sometimes, pg, per g or kg wet, dry, or lipid weight, which adds to the confusion, particularly when the content of dry matter or lipid is not stated. In addition, several different methods may be used to determine the lipid content and yield different results. Methylsulfonyl metabolites of p,p′-DDE are increasingly being reported, particularly in marine mammals and aquatic birds. Tris(4-chlorophenyl)methane (TCPM) and tris(4-chlorophenyl)methanol (TCPMOH) have also been found recently in the environment. There is some evidence that they may have originated from technical DDT and some authors found a correlation between the concentrations of TCPM/TCPMOH and DDT. However, there are also excellent correlations between the concentrations of PCBs and DDT, which, of course mean that both substances behave similarly in the environment, not that PCBs originate from DDT. The determination of time trends of concentrations is very difficult and costly since it requires large number of analyses over many years. The results are affected by many factors [25]. These include biological variability, sampling protocol, frequency and duration of sampling, and, last but not least, changes in the measurement techniques. Thus, for example, the ratio of SDDT concentrations in the same samples of mussels, measured in 1999 to the concentrations obtained in 1977 ranged from 0.03 to 7.1, with a mean of 0.6 [122]. 3.1 Air

The concentrations of COC are usually in tens of pg/m3, but may reach hundreds or thousands in areas of heavy agricultural use of pesticides. In 1971 Woodwell et al. [225] estimated the concentration of DDT in the atmosphere, based on the United States production to that date (Fig. 2). The solid line assumes DDT production in the United States declining to zero in 1974, the dotted line corresponds to continuing production. Concentrations actually measured over the oceans are up to several orders of magnitude below the predicted ones, but approach them in localized areas of high DDT use. The measurement of extremely low concentrations is difficult. Consequently, several ‘surrogate’ matrices, such as pine needles, moss, and tree bark were suggested. The concentrations of SDDT (in this chapter, the term ΣDDT means the sum of the concentrations of p,p′-DDT and its metabolites, most frequently p,p′-DDE and p,p′-DDD. Relatively seldom it also includes some of the o,p′-isomers) in pine needle wax from Sweden and Norway

56

V. Zitko

Fig. 2. Concentration of DDT in the troposphere [101, 171, 225]. Solid and dotted lines are model estimates, developed in 1971. The former assumes DDT production ending in 1974, the dotted line, continuing production of DDT. Vertical lines indicate ranges of concentrations actually measured [8, 64, 101, 22, 23]

ranged from 0.12–0.43 ng/g of fresh needle and the concentrations declined from 1984 to 1986. The concentration range was 0.25–0.65 ng/g for Switzerland, Germany, and Denmark and the concentrations declined similarly to those in Sweden. In Poland, in 1985 and 1986 the concentrations ranged from 0.3–2.48 ng/g [65]. In the Czech republic during 1988–1994, moss (Hypnum cupressiforme L. ex Hedw.) accumulated higher concentrations of COC than pine needles, with a range from <0.1 to 0.4 and a mean of 0.15 ng/g for p,p′-DDE and a range from <0.1 to 6.8, and a mean of 1.14 ng/g for p,p′-DDT, while the concentration of both pesticides in pine needles was <0.1 ng/g [93]. SDDT was measured in the bark of trees from a number of countries. The concentrations ranged from 2 ng/g in Costa Rica to 4310 ng/g in India and 27,700 ng/g in Romania [180]. Somewhat surprisingly, the bark contains relatively high concentrations of p,p′-DDD. The second COC, most frequently encountered and abundant in air, is dieldrin. However, very high concentrations of aldrin and dieldrin (geometric means 945 and 1122 pg/m3, respectively) were detected in Belize in 1995–1996 [11]. In North America, p,p′-DDT is deposited from the atmosphere into the Great Lakes. On the other hand, p,p′-DDE and dieldrin are transported from the lakes to the atmosphere. Unfortunately, the largest measurement uncertainty is associated with the gas transfer process [94].

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

57

3.2 Soil

At the peak of DDT use in the 1960s, the concentrations of DDT in the soil of agricultural areas were in the range of several to tens of mg/g. Thus, for example, in southwestern Ontario (Canada), the concentrations ranged from 0.4 µg/g in sugar beet fields, through 9.5 mg/g in vegetables fields, to 62 mg/g in orchards [88]. Cyclodienes were not used in orchards, but their concentrations ranged from 0.3 to 1.6 mg/g in fields. Recently, in agricultural soils, the concentrations of SDDT ranged from traces in Queensland, Australia [44] to hundreds of ng/g in heavily farmed areas of India [5], the Russian Federation and Uzbekistan [73]. In these areas, p,p′-DDT is frequently the most abundant component of ΣDDT. The halflife of SDDT in soils in the temperate zones is about 2–10 years and may be much shorter in the tropics [31]. On the other hand, the half-life appears closer to 30 years in some forest soils in the United States [60]. Aldrin, dieldrin, and mirex are also found in areas of former application. The half-lives of dieldrin and mirex in soil are comparable to that of SDDT. The disappearance of aldrin is much faster, partly because of conversion to dieldrin, partly because of relatively rapid volatilization. There is some evidence that COC in soil become, with time, less bioavailable [146, 168]. On the other hand, even after almost 30 years, orchard soils are a significant source of DDT residues found in American robins (Turdus migratorius) [91]. Insufficient information is available on the enantiomeric composition of o,p′DDT in soil. One study reports that soils containing racemates and enantiomerenriched mixtures are about equally represented among the north-American soils investigated [150]. 3.3 Water

The 1971 predictions of Woodwell et al. [225] for the concentration of DDT in seawater are shown in Fig. 3. Only in the Baltic and possibly in other relatively enclosed areas may the concentrations approach the predictions. With the exception of some developing countries, the concentrations of most pesticides in surface waters are probably well below the 1 ng/l level.At the end of the extensive use some 20 years ago, the concentrations were in the range from not detectable to about 100 ng/l for aldrin, to about 10 ng/l for p,p′-DDT, 5 ng/l for dieldrin, 10 ng/l for endrin, and 4 ng/l for mirex. In Egypt in 1997–1998, in lake water draining into the Mediterranean, the concentrations of p,p′-DDE, p,p′-DDT, and p,p′-DDD were approximately 30–40 ng/l, and 10–20 ng/l, respectively [1]. The concentrations were somewhat higher in the Nile estuaries [2]. On the other hand, in the 1990s in the agricultural area of the Ganga (India) watershed, the concentrations of SDDT and, on one occasion, of aldrin, were up to several hundred ng/l, and that of dieldrin was several tens of ng/l [142].

58

V. Zitko

Fig. 3. Concentration of DDT in sea water [225]. Solid and dotted lines are model estimates, de-

veloped in 1971. The former assumes DDT production ending in 1974, the dotted line, continuous production of DDT. Vertical lines indicate ranges of concentrations actually measured [101, 171]

3.4 Sediment

The United States National Status and Trends Program for Marine Environmental Quality contains a large compilation of COC concentrations in USA coastal sediments. Mean, median and maximum values are summarized in Table 5, together with comparative data from Kattegat and Skagerrak [55], sediments off the Baltic South Coast [67], sediments from Guba Pechenga on the Kola peninsula (Russia) [174], and sediments from coastal lagoons in the southeastern Gulf of Mexico [208]. The distribution of SDDT concentrations in the USA coastal sediments is in Fig. 4. The median concentration of p,p′-DDE in percent of SDDT is 37%. The relative concentration of p,p′-DDE ranged from 44 to 64% in sediments from several Portuguese estuaries [78]. Lower relative concentration of p,p′-DDE may indicate a more recent input of DDT. For example, in sediments of two Brazilian watersheds, the relative concentrations of p,p′-DDE were 10 and 32% [205]. Similarly, in the Pearl River (China) estuary in 1996–1997, the relative concentrations of p,p′-DDE were 17 and 20% in surface sediments and suspended particulate matter, respectively [95]. Interestingly, no p,p′-DDE, only p,p′-DDT, o,p′-DDT, and p,p′-DDD were reported in sediments in the Macao estuary in southern China [229].

a

0.00 2.35 319.1

o,p-DDD

0.15

0.66 12.29 2244 0.58 20.79 1.89

p,p-DDD

1.8

0.00 0.38 49.45

o,p-DDT

1.44

1.16

0.04 5.80 850.0 0.26 9.24 2.46

p,p-DDT

Geometric mean, used when only range of values is reported in the literature.

0.60

Gmean a

0.83 16.85 4300 0.47 8.47 0.28

0.06

5.0

0.00 4.03 1100

p,p-DDE

Median

Median Mean Max Mean Mean Median

o,p-DDE

1.35

0.00 0.19 5.89 0.01

Aldrin

2.47

0.11 0.88 26.00 0.25

Dieldrin

4.59

0.17

Endrin

0.00 0.26 9.50

Mirex

Kattegat [55] S. Baltic [67] Guba Pechenga (Russia) [174] Kola Peninsula (Russia) [174] G of Mexico [208]

USA

Ref

Table 5. Concentrations (ng/g dry sediment) of the DDT group of compounds, aldrin, dieldrin, and mirex in the U.S.A. coastal sediments (from the US National State and trend data (http://ccmaserver.nos.noaa.gov/)) and some reference sites

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

59

60

V. Zitko

Fig. 4. Distribution of SDDT concentrations along the coast of the United States in years 1986–1991 (from raw data, downloaded from http://ccmaserver.nos.noaa.gov/)

The absence of p,p′-DDE appears strange and could be an analytical artifact. On the other hand, coastal sediments off the coast of India contained primarily o,p′-DDE and p,p′-DDE, with median concentrations of 75 and 25 ng/g dry weight (recalculated from reported data by assuming 20% moisture and a detection limit of 0.005 mg/g wet weight [173]). This may indicate a faster degradation of the p,p′-compounds in warmer waters, but again, analytical artifacts or different composition of technical DDT cannot be excluded. In support of the degradation assumption, p,p′-DDT predominates in the bottom sediments of the Barents Sea fjords. On the other hand, no difference in the rate of photodegradation was observed between p,p′-DDT and o,p′-DDT [74]. In any case, the degradation of DDT appears to proceed more slowly at low temperatures [174]. There appears to be a rising trend in the proportion of p,p′-DDT in the Macao estuary, as well as in two sediment cores from the Baltic [157]. The data for USA sediments are within the range of those from the Baltic (Fig. 5), but their time span is too limited for trend determination. The distribution of DDTs in sediments as a function of grain size has received relatively little attention. According to a recent report, p,p′-DDE was enriched in coarser (>63 mm) particles in sediments from the Atoya river (Nicaragua) [42]. Data on the concentrations of the cyclodienes in sediments are quite limited. It appears that levels are low, except for some localized areas of heavy use. The presence of aldrin is somewhat surprising and one must wonder whether the reported values are not due to interference by elemental sulfur.

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

61

Fig. 5. DDT in % of SDDT in sediments from the Macao estuary (filled squares) [229], the

Stockholm archipelago (filled triangles), northwestern Baltic (open squares) [157], and from the USA NST program (asterisks)

3.5 Fauna and Flora

The concentration of COC in flora is generally low and, in comparison to fauna, COC in flora have received little attention. With the exception of birds, wild terrestrial fauna was also studied relatively much less than aquatic fauna. For example partridges [92], and, more recently, lizards [41], were suggested as bioindicators, and DDE was monitored in terrestrial biota in the Rio Grande-Rio Bravo basin [144]. Considerable attention was paid to COC in birds because of frequent kills and reproductive failure caused primarily by eggshell thinning [155, 222]. Numerous studies were devoted to the determination of tissue concentrations related to mortality. Relatively few systematic studies are available for other terrestrial fauna. Concentrations of COC in livestock and marketed livestock products are regularly monitored by food-inspection agencies in most countries, and, occasionally, also reported in the literature (see, for example, [6, 71, 83]). The results of the food-inspection agencies are usually reported as numbers of samples analyzed and numbers of those exceeding some guidelines. Limited space does not allow us to do justice to the wealth of data, accumulated in the literature, on the levels of COC in fauna. An extensive summary of concentrations in birds is presented in [29], and concentrations in freshwater fish, in [175]. A review of concentrations detected in marine organisms is published annually [165].

62

V. Zitko

There is substantial evidence that, in northern Europe, the concentrations of SDDT in freshwater and marine fish, as well as in guillemot eggs, decreased from 1967 to 1995 by 3–10% annually [24].A ‘transient’ increase around the mid 1980s in some locations may have been caused by the use of DDT in East Germany at that time. In marine fish from seas bordering member states of the European Union, the concentrations of SDDT decreased from 1100 mg/kg in 1979 to 260 mg/kg in 1993 [51]. DDT is still used in many tropical countries, but its volatilization there and transport by air to colder climates [121] do not seem to have a trend-changing influence. 3.6 Aquatic Fauna

The concentrations, expressed on a lipid basis, of COC in aquatic fauna are generally in the ng/g to µg/g range, and increase with trophic status. The biomagnification along the aquatic food webs is not continuous, but increases abruptly for example from marine fish to marine mammals or to aquatic birds. In general, predatory species have higher concentrations of COC than their prey. Of the SDDT complex, p,p′-DDE is practically always the most abundant component. Only in areas of recent DDT input, or in metabolically relatively inactive compartments, may p,p′-DDT remain in considerable concentrations. Dieldrin is much less frequently found and aldrin and endrin are seldom reported. However, there is at least one early publication [115], and a few recent ones, that report concentrations of endrin. It may be that the recent reports result from refined analytical techniques and that the presence of endrin was overlooked in the past. Thus, for example, endrin was detected in concentrations from 0.14 to 2.3 ng/g dry weight in mussels from the north-west Mediterranean coast, collected in 1988–1989 [212], and in concentrations from 1.5 to 10.6 ng/g dry weight in the American oyster (Crassostrea virginica) from the southeastern Gulf of Mexico [208]. The oysters also contained aldrin in concentrations from 2.6 to 6.7 ng/g dry weight. Mirex is found particularly in biota from Lake Ontario, which was contaminated by industrial discharges.Again, this compound might have been missed in the past, since it co-elutes with p,p′-DDD on some packed columns. 3.6.1 Molluscs

Distributions of SDDT concentrations in blue mussels (Mytilus edulis), California mussels (M. californianus), and oysters (Crassostrea virginica), obtained by the NST program, and in green mussels (Perna viridis), obtained by the Marine Pollution Monitoring in Asian Waters program [196], are shown in Fig. 6. The distributions are remarkably similar. However, the green mussel data are derived from only 53 measurement of composite samples (30 individual specimens each) [196]. The green mussels originated in coastal waters of India, Philippines, and Thailand. In green mussels from India and Thailand, the average relative concentration of p,p′-DDE is 49.4%. In the NST blue mussels, California mussels, and oysters, the median concentrations of p,p′-DDE are 47.4, 76.8, and 59.3%, re-

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

63

Fig. 6. Distribution of ΣDDT concentrations in mussels along the coast of India, Thailand, and

the Philippines (filled squares) [196–199] and from the U.S. National Status and Trends Program, calculated from data available at http://ccmaserver. nos.noaa.gov/

Fig. 7. Relative concentration of p,p′-DDE in % of SDDT in blue mussels (Mytilus edulis) (filled

triangles), California mussels (M. californianus) (open squares), and oysters (Crassostrea virginica) (filled diamonds), obtained by the NST program

64

V. Zitko

Fig. 8. Concentrations of PCB and SDDT in green (India, the Philippines and Thailand

[196–199]) and blue (Korea [111] and Baltic [85]) mussels

Fig. 9. Distribution of SDDT concentrations in fish from the US NST sites [76] and from fish from Cambodia [143]

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

65

spectively. The high relative concentration of p,p′-DDE in California mussels remained practically constant throughout the motoring period from 1986 to 1998. It seems that the difference in p,p′-DDE concentrations between blue mussels and oysters appeared only after 1992 (Fig. 7). Figure 8 shows concentrations of PCB and SDDT in green mussels (Perna viridis) from India, the Philippines, and Thailand [197–199], blue mussels from Korea [111], clams (Meretrix meretrix) from Vietnam [151], and, for comparison, in blue mussels from the Baltic [85]. The proportions of PCB and SDDT are similar in mussels from the Philippines and Korea, and in mussels from India and Thailand, with those from the Baltic practically separating the two groups. 3.6.2 Fish

The distribution of SDDT concentrations in fish from the United States NST sites [76], and, for comparison, of fish from Cambodia [143] are shown in Fig. 9. The concentrations are lower in Cambodian fish, but in several locations the median relative concentration of p,p′-DDE ranges from 23–48% in marine, and 17–20% in freshwater fish, thus indicating a recent input of DDT. 3.6.3 Marine Mammals

Probably the highest recorded concentrations of COC are in marine mammals. Most of marine mammals are at the top of the food chains, contain high concentrations of lipids, have relatively long life spans, and, in comparison with terrestrial mammals, a lower capability to metabolize foreign compounds. The concentration of COC in marine mammals depends on many factors, including species, sex, age, location, migratory patterns, etc., and ranges over several orders of magnitude (Fig. 10). Concentrations of PCB and SDDT were compiled for dolphins from [103, 105, 130, 131, 192, 193, 197–199], seals from [21, 40, 47, 86, 99, 113, 117, 148, 211], porpoises from [36, 75, 84, 108, 113, 114, 141, 145, 221], and whales from )[17, 77, 132, 133, 147, 194, 197, 203, 218–220]. Concentrations of SDDT and PCB in some cetacean market products from Japan [87] are included in the insert. The relative concentration of p,p′-DDE in marine mammals is usually about 80%, except for the Ganges River dolphins, where p,p′-DDE in blubber constituted only about 40% of the SDDT complex [105, 231], obviously because of recent and possibly continuing, input of p,p′-DDT. In the liver, the relative concentration of p,p′-DDE reaches 73%. These dolphins also contain the much lower concentrations of PCB than dolphins from other areas. In the Mediterranean the concentration of SDDT (and also of other organochlorine compounds) in the blubber and other tissues of striped dolphins (Stenella coeruleoalba) increased dramatically in 1990 [130, 131]. This may have been a factor in the large mortality of these dolphins in 1990–1991. Of the other COC, dieldrin and mirex at 67 and 21 ng/g lipid, respectively, were found in the blubber of one G. griseus from

66

V. Zitko

Fig. 10. Concentrations of PCB and SDDT in dolphins, seals, porpoises, and whales [49, 50,

177]. Insert: Cetacean products from Japanese market [87]

the west coast of North America [103]. The specimen contained SDDT at 16,700 ng/g lipid. The average concentrations of dieldrin and mirex in the blubber of six white-sided dolphins (Lagenorhynchus acutus ) from the east coast of North America (Gulf of Maine) were considerably higher, 2300 and 102 ng/g lipid, respectively, at an average SDDT concentration of 32,000 ng/g lipid [220]. Interestingly, aldrin at 32 and 93 ng/g lipid, in addition to dieldrin (158 and 177 ng/g lipid), was found in the blubber of a neonatal and a one-year-old male Ganges River dolphin (Platanista gangetica), containing SDDT at 13,800 and 29,300 ng/g lipid, respectively. Concentrations of aldrin and dieldrin (7 and 17.6, and 2 and 79 ng/g lipid, respectively) were considerably lower in blubber of 1- and 30-yearsold female dolphins. The same animals contained S DDT at 29,000 and 17,000 ng/g lipid, respectively [105]. The lowest concentrations of SDDT and PCB are in Caspian seals (Phoca caspica) from the Caspian sea [86] and in South African Fur seals (Arctocephalus pusillus pusillus) from Cape Cross (Namibia) [211]. Interestingly, the concentration of dieldrin in some of these seals was approximately equal to that of p,p′DDT. The highest concentrations of SDDT and PCB were quoted for ringed seals (Phoca (Pusa) hispida) from Lake Saimaa (Finland) and the Gulf of Finland [99]. The lowest relative concentration of p,p′-DDE, indicative of a recent input of DDT, was found in ringed seals (Phoca hispida) from Lake Ladoga [117]. Higher concentrations of SDDT were found in ringed seals from the east coast of Greenland than in seals from the west coast [47]. Mirex in concentrations over 100 ng/g blubber lipid was present in harbour seals (Phoca vitulina), who are permanent

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

67

residents of the St. Lawrence estuary (Canada). On the other hand, its concentration in grey seals (Halichoerus grypus), who may reside for shorter time periods in the estuary, was generally lower [21]. Some of the highest concentrations of SDDT on record in harbour porpoises (Phocoena phocoena), were found in 1969/1970 in porpoises from the Bay of Fundy (Canada). The relative concentration of p,p′-DDE was 36–47% [75]. These data cannot be shown in Fig. 10 since PCBs were separated but not quantified. The high levels of SDDT in the porpoises from the Bay of Fundy may reflect in part large amounts of DDT, used in those years in forest spraying against spruce budworm in New Brunswick (Canada). The concentrations of SDDT and PCB in porpoises and whales occupy the ‘middle ground’ in Fig. 10, with high concentrations present particularly in porpoises in the west Atlantic, and in beluga whales from the St. Lawrence estuary (Canada). Porpoises from Cardigan Bay, West Wales contained, in 1987, dieldrin in concentrations ranging from 3 to 9 mg/g blubber wet weight, as compared to a concentrations range of 2 to 7 mg/g for p,p′-DDE [145]. Porpoises caught in Scandinavian waters from 1987 to 1991 contained dieldrin and endrin, both in the ‘several’ µg/g blubber lipid range [114]. Methylsulfonyl-p,p′-DDE was measured in porpoises, captured incidentally in Swedish waters in 1996. The concentrations were 1–4 ng/g lipid in the blubber and 0.8 to 21 ng/g lipid in liver, muscle, and brain. The concentration ranges of p,p′-DDE in the same compartments were 1300–1700, and 130–2200 ng/g lipid, respectively [108]. Whales spending summers in the Gulf of St. Lawrence (Atlantic minke (Balaenoptera acurostrata), fin (B. physalus), blue (B. musculus), and humpback (Megaptera novaeangliae)), sampled by biopsy in 1991, contained dieldrin and mirex in concentrations ranging from <1 to 1500 ng/g lipid, and from <0.5 to 80 ng/g lipid, respectively [77]. The concentration ranges of these pesticides in the blubber of male and female beluga whales (Delphinapterus lucas) were 0.81–1.06 and 0.21–0.87 mg/g blubber wet weight, and 0.19–1.54 and 0.38–2.66 mg/g blubber wet weight, in male and female Gulf of St. Lawrence belugas, respectively [17]. As will be discussed later, mirex originates in Lake Ontario and is practically absent from belugas from the Arctic. It is also interesting to note that, in contrast to the DDT group of compounds, which is at considerably lower concentrations in female than in male cetaceans, there is practically no sex-related difference in the concentrations of mirex, and little difference in the concentrations of dieldrin. Somewhat surprisingly, aldrin was found in the blubber of right whales (Eubalaena glacialis), sampled by biopsy between 1996 and 1997 in the Bay of Fundy (Canada), but not in those from the Cape Cod Bay (USA), and off the southern coast of Georgia (USA), in average concentrations of 513, 93, and 105 ng/g lipid [219]. The concentrations of tris(4-chlorophenyl)methane (TCPM) and tris(4chlorophenyl)methanol (TCPMOH) in ringed, baikal, larga, and Caspian seals were in ranges 0.4–0.9, 0.5–9.3, 1.6–17, and 2.0–88 ng/g lipid, respectively [215]. The average concentration of TCPMOH in two killer whales was 1600 ng/g lipid. In addition, lipids of the killer whales contained an unidentified isomer of TCMPOH, not present in other samples, at an estimated concentration of 1 ug/g lipid [215].

68

V. Zitko

3.7 Humans

Concentrations of p,p′-DDT in the several to tens of µg/g range were found in the 1950s in human adipose tissue by, as considered today, crude analytical methods and the detection of dieldrin in the µg/g range soon followed. Later, more reliable data were obtained by gas chromatography. Thus, for example, in the U.S.A. in 1967, by using gas chromatography with an electron-capture detector, but no cleanup, DDT, DDE, DDD, and dieldrin were found in human adipose tissue of accident victims at concentrations of 2.77, 6.69, 0.28, and 0.21 µg/g, respectively [56].Average concentrations of these compounds in tissues of subjects who died from portal cirrhosis, carcinoma, and hypertension, were 4.84, 12.21, 0.30, and 0.46 mg/g, respectively. Neither of the groups of subjects was exposed to the pesticides occupationally, and the authors concluded that domestic use of pesticides must be an important source of the residues. A review of concentrations in humans and of results of feeding experiments in livestock concluded in 1967 that there is no evidence that the concentrations of SDDT and dieldrin have reached a steady state [163]. A potential interference by polychlorinated biphenyls (PCBs) was recognised in the late 1960s and largely eliminated in analyses reported from then on. For example in Germany, Acker and Schulte [3, 4] reported p,p′-DDE and p,p′-DDT at 0.3–26 mg/g and 0.2–7 mg/g of human adipose tissue, respectively. Dieldrin, p,p′DDD, and o,p′-DDT were found in concentrations of about 0.1, 0.02, and 0.05 mg/g, respectively. The concentrations of p,p′-DDE, p,p′-DDT and dieldrin reported at that time for human adipose tissue from other countries fall within the

Fig. 11. SDDT concentrations in human adipose tissue. From data summarized by Kutz and

others [120]

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

69

same concentration range [61]. In the 1990s the concentration of p,p′-DDE remains in the same range; however, the concentration of p,p′-DDT dropped below 0.2 mg/g in most countries [58]. As a result of restrictions or bans on the use of DDT, its concentrations in adipose tissue appeared to have stabilized in most countries at levels below 10 mg/g (Fig. 11). On the other hand, even in the 1980s, in some, particularly the developing countries, the concentrations were very high. One of the highest on record appears to be 62.45 mg/g, reported for 1993 in Zaire in 1982/83. However, in the same survey, the minimum and mean concentrations were 2.57 and 15.38 mg/g, respectively [156]. In the adipose tissue of humans from most industrialized countries, the concentrations of SDDT and dieldrin have decreased over the past 15–20 years. Thus, for example, in 1982, in Canada, the concentrations of p,p′DDT, p,p′-DDE, and dieldrin were 0.31, 1.72, and 0.05 mg/g in the adipose tissue of accident victims, and 0.42, 2.79, and 0.05 mg/g in adipose tissues of humans who died of various diseases [139]. These data demonstrate the considerable decrease in the concentrations, resulting from restrictions on the use of DDT and dieldrin. In the adipose tissue of the general population in the USA, the concentrations of SDDT and dieldrin decreased between 1970 and 1983 at about 5% of their 1970 concentrations, per year (Fig. 12). In Sweden, the concentration of p,p′-DDT, p,p′-DDE, MeSO2-DDE, and dieldrin decreased exponentially since the 1970s, as illustrated in Fig. 13 for p,p′-DDE in human milk [153, 154]. Concentrations of p,p′-DDE reported from other countries are included for comparison.As can be seen, the United Kingdom, Germany,

Fig. 12. Decrease of concentrations of SDDT and dieldrin in the adipose tissue in the USA

(Kutz [120, 124, 176])

70

V. Zitko

Fig. 13. Concentrations of p,p′-DDE in human milk. Solid squares are data from Sweden [154]. Data for the UK, Germany, Japan, Canada, Ukraine, Russia, and Brazil are from [3, 4, 61, 58, 59, 79, 88–91, 100, 159, 161, 178, 206]

Japan, and southern Canada follow the same trend. The concentrations are higher in countries which probably used DDT more recently (India, Brazil, Ukraine, Russia), and in the Canadian Arctic, which receives p,p′-DDE by long-range transport. Concentrations of o,p′-DDT are reported less frequently than those of p,p′-DDT and it has been shown only recently that the more estrogenically active enantiomer, (–)-o,p′-DDE [136], is depleted in human adipose tissue [39]. The concentrations of TCPM and TCPMOH in human adipose tissue from Japan range from 2.5 to 21 ng/g lipid and from 1.1 to 18 ng/g lipid, respectively [141]. The toxicological significance of these values cannot be assessed at this time.

4 DDT 4.1 History

In 1874, Zeidler [228] studied the reaction of bromo- and chlorobenzene with chloral and prepared dibromophenyl trichloroethane (DBT) by treating one part of bromobenzene with two parts of anhydrous chloral in 4–5 volumes of concentrated sulfuric acid, with occasional warming on a water bath. The compound yielded dibromophenyldichloroethylene by boiling with alcoholic

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

71

Fig. 14. Formulae of p,p′-DDT, p,p′-DDE, and p,p′-DDD

potassium hydroxide and a dinitro compound by dissolving in fuming nitric acid. Under similar conditions Zeidler also prepared dichlorophenyltrichloroethane and noted that the reaction proceeds somewhat more slowly. The compound crystallized from diethyl ether- alcohol in white felt-like needles, similar to crystals of quinine sulfate. The compound melted at 105 °C and had solubility properties similar to those of the bromo compound. This is rather interesting, since, according to Zeidler, DDT was insoluble in benzene, slightly soluble in cold alcohol and glacial acetic acid, better soluble in chloroform and ether, and well soluble in carbon disulfide. However, DDT is well soluble in benzene (78 g/100 ml) and, consequently there is a bit of mystery about Zeidler’s DDT. However, its melting point is in good agreement with that of p,p′-DDT. Zeidler also prepared the respective ethylene (m.p. 89 °C) and dinitro (m.p. 143 °C) compounds. The formulae of the p,p′-isomers of DDT, DDE, and DDD are presented in Fig. 14.

72

V. Zitko

DDT was synthesized again in 1939 by Müller, among sulfur, oxygen, nitrogen, or carbon compounds containing two chlorophenyl rings, in a search for compounds against moths and carpet beetles. The history of the search is described in detail by Brooks [34, 35]. The remarkable insecticidal activity of DDT was noticed and its production soon reached industrial scale. Müller was awarded the Nobel price in 1948. Large-scale industrial production started in 1944 in England and in the United States.A brief summary of the history of DDT was prepared by Metcalf [140]. 4.2 Production

DDT is prepared by the Bayer condensation of chlorobenzene (CA RN 108-90-7) with trichloroacetaldehyde (chloral CA RN 75-87-6) in oleum (fuming sulfuric acid, CA RN 8014-95-7), and the reaction is carried out with an excess of chlorobenzene (recommended molar ratio 3:1). In the exothermic reaction, oleum is added to the mixture of chlorobenzene and trichloroacetaldehyde under cooling, to maintain the temperature below 30 °C. Approximately 4 moles of oleum are added over 4–7 h, and the mixture is agitated for another 0.5–4 h. Crude DDT is filtered off, washed with alkali, and dried [181]. Technical DDT contains about 85% of p,p′-DDT, 15% of o,p′-DDT, and traces of o,o′-DDT and other compounds, including three isomeric tris(chlorophenyl)methanes (TCPM), recently detected in the environment, together with their oxidation product, tris(chlorophenyl)methanol [38]. Some of the minor impurities, as well as o,p′-DDT, are chiral and are present as racemates [39]. The overall world production of DDT is estimated at 2.6 megatonnes [213]. It is difficult to arrive at a precise amount, because of reporting problems in different countries. The amounts produced in the USA, China and Mexico are shown in Fig. 15. It appears that India is the last country which still produces DDT. The data on DDT consumption in various countries are also difficult to find. In Fig. 16 are the amounts of COC used in Canada [202]. The discontinuities in some of the curves are caused by unavailable data. In Thailand, from 23 tonnes to over 73 tonnes of DTT were used annually from 1988 to 1997 [30]. In Costa Rica, 128 tonnes of DDT and an additional 147 tonnes of an unspecified mixture of DDT and toxaphene were imported from 1977 to 1985 [43]. China, India, and Japan used, in total, 270, 330, and 30 kilotonnes of DDT, respectively [123]. The former USSR used about 10 kilotonnes annually from 1950 to at least 1970 and the amount dropped to 300 tonnes in 1980 [119].

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

Fig. 15. Production of DDT. Data are from [123, 125, 225]

Fig. 16. Consumption of COC in Canada [202] (Solid squares-DDT)

73

74

V. Zitko

4.3 Effects 4.3.1 Mode of Action

DDT affects the transport of sodium and potassium in the nerve membranes and the resulting excitatory effects on the axons can be readily demonstrated electrophysiologically [149]. DDT inhibits the closure of voltage-gated sodium channels, but the molecular mechanism of the action is not known. The formation of a charge-transfer complex with components of the nervous system [134], and the inhibition of adenosine triphosphatase [135] have been suggested as possible mechanisms. It was demonstrated later that spectral changes associated with the formation of charge-transfer complexes also take place during the ageing of DDT solutions in aqueous isopropanol [224]. However, DDT and related compounds also form charge-transfer complexes with tetracyanoethylene, but of much lower intensity than those of chlorobenzene and of bis(p-chlorophenyl)-methane. Consequently, the role of charge-transfer complexes in the toxicity of DDT is not straightforward. Nuclear magnetic resonance indicates that DDT binds to lecithin and that the benzylic proton plays a significant role in this process [204]. This may explain the difference between the toxicity of DDT and DDE, with the latter not containing a benzylic hydrogen. There is little doubt that lipophilic compounds incorporate themselves into the lipid-rich membrane and are likely to interfere with normal functioning of the membrane. Calcium is involved in the process, as demonstrated recently by studies with membranes of the thermophilic bacterium Bacillus stearothermophilus [62]. Both DDT and DDE inhibit the growth of this bacterium, but the inhibition by DDE was eliminated by the presence of calcium. DDT and DDE also exhibited different effects on the membrane phospholipids. In the presence of DDT, the relative proportion of branched and iso- acids was decreased considerably. On the other hand, DDE decreased the proportion of diphosphatidyl glycerol. In fasting human plasma, almost 80% of p,p′-DDE is present in the lipoprotein depleted, primarily albumin fraction [153, 154]. p,p′-DDD is a very potent toxicant for the non-cilliated bronchiolar cells [152]. The mechanism, suggested earlier by studies in mice [81], involves oxidation at the C1 carbon atom which may lead to the formation of a reactive acyl chloride, a part of which binds covalently with components of the tissue, while another part hydrolyzes to p,p′-DDA. The estrogenic activity, measured by binding to calf uterine estrogen receptor, decreases in the order o,p′-DDT (racemate)> o,p′-DDEo,p′-DDD, with the respective IC50 (log nmol/l) concentrations of 4.94, 5.77, not active, respectively, and 0.837 for a positive control by 17b-estradiol [118]. The antiandrogenic activity of p,p′-DDE was not demonstrated at concentrations in the range found in the Great Lakes area (Canada/United States) by studying the feminization of snapping turtles (Chelydra serpentina serpentina) [162]. In utero exposure of rats to a very high dose of p,p′-DDE affected the response of prostate in adult offsprings to further exposure to p,p′-DDE and demonstrated that, in principle, such

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

75

interactions may have to be taken into consideration, even though this particular one is not likely to occur with the current environmental concentrations of p,p′-DDE [227]. 4.3.2 Structure-Activity Relationships

A large number of DDT analogues have been prepared, partly in search for a more biodegradable insecticide, partly to explain the mechanism of action. The biodegradability is enhanced by replacing the chlorine substituents on the benzene rings by more polar groups such as methoxy- or methylthio-. This substitution also increases the water solubility of the compounds and the increased biodegradability is probably a result of the combination of better solubility and a change in the electronic structure of the molecule [107]. The oxidation of the more polar groups may provide another opening for a more rapid biodegradation. On the other hand, steric factors seem to play the most important role in terms of toxicity to insects [66]. 4.3.3 Transformations

Transformations of DDT, whether by metabolism or by degradation, occur almost exclusively on the trichloroethane moiety [7]. The most common reaction is dehydrochlorination to DDE (see Table 6 for formulae of DDT metabolites and transformation products). This reaction plays an important role in the development of insect resistance and an enzyme catalyzing it was isolated. The mechanism of the reaction in other fauna does not appear to have been studied in detail. However, it seems that the rate of this reaction is related to some general detoxification processes and increases from invertebrates to fish, birds, and mam-

Table 6. Formulae and abbreviations for various DDT-related compounds

Formula (only the ethane moiety is shown)

Abbreviation

CH–CCl3 CH–CHCl2 C=CCl2 C=CHCl CH–CH2Cl C=CH2 CH–CH3 CH–CH2OH CH–COOH CH–OH C=O COH–CCl3

DDT DDD DDE DDMU DDMS DDNU DDNS DDOH DDA DBH DBP dicofol

The chlorine atoms on the benzene rings may be either in p,p′ or in o,p′ positions.

76

V. Zitko

mals. It is possible that this reaction in higher animals may proceed via dicofol, with a subsequent loss of HOCl. Under anaerobic conditions the dichloroethylene structure of DDE is hydrogenated and yields DDD, which itself is an insecticide. In rat hepatic microsomes, at low levels of oxygen, this reaction proceeds via a free radical intermediate, which binds to microsomal lipids [16]. In the environment, the loss of chlorine may continue and yields, in series a dichloro-, and chloroethane, and, finally ethane. In the presence of anaerobic sewage sludge, DDT yields bis(p-chlorophenyl)acetonitrile [10]. Bis(p-chlorophenyl) ethane may, in turn, be oxidized to diphenyl ethanol, acetic acid, and ultimately to dibenzophenone. In an opposite process, DDT may be oxidised to dicofol, in which trichloroethane is replaced by trichloroethanol. Dicofol is also an insecticide. The benzene rings of DDT are much less reactive. Those of p,p′-DDE and o,p′DDT can be hydroxylated, at least in some mammalian species [195], and also form methylsulfonyl-substituted (MeSO2–) compounds [20]. There is some evidence that mixed bacterial cultures are able to cleave at least one benzene ring of DDT [70]. Recently, 2-, and 4-hydroxy derivatives were produced from p,p′-DDT by a Gram-negative bacterium, isolated from activated sludge [132, 133]. The benzene rings of p,p′-DDE, which was for long considered the final degradation product, may be hydroxylated or metabolized to yield 2-, 3-, and 4-hydroxy-p,p′-DDE [195], as well as 3-methylsulfonyl 4,4′-DDE (MeSO2-DDE). 4.3.4 Toxicology

There is a voluminous literature of the toxicological properties of DDT and its derivatives and degradation products and to go into details would be beyond the scope of this chapter (see for example [15, 89, 90, 185, 189–191]). Effects continue to be detected, at times by using very high concentrations [19]. In contrast to the other COC, there is a strong point for the continuing use of DDT in public health actions in the tropics. DDT has an impressive record in this direction, which is not to be dismissed lightly [164, 187].At the same time, work on sublethal effects of DDT continues and, while no link between p,p′-DDE and cancer of male reproductive tract could be established [48], the compound was found to affect progesterone synthesis [53]. Similarly, a link between exposure to DDT and PCBs, and increased risk of endometrial cancer could not be established [217]. As often happens in epidemiological studies, the results of studies of connections between the DDT compounds and cancer are not clear cut. For a recent review see [104] and also Hoyer’s study [97] implicating dieldrin. MeSO2-DDE, mentioned above, is formed after conjugation with glutathion and metabolism in the mercapturic acid pathway, and is highly toxic [102]. The concentrations of MeSO2-DDE in biota are only a fraction of those of p,p′-DDE In seals, the concentrations increase in the order blubber
Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

77

single dose of 25 mg/kg. The adrenocortical mitochondrial P450s is involved in the formation of the bound products [126]. In contrast, p,p′-DDD and o,p′DDD also induce necrosis, but at five times higher doses, and in both, zona fasciculata and zona reticularis . Feeding a mixture of methyl-sulfonyl-chlorinated biphenyls and MeSO2-DDE to mink (Mustela vison) affected reproduction and resulted in increased litter sizes, decreased birth weight and survival [127]. The environmental significance of these findings is not clear. The effects may be overshadowed by the effects of other toxicants, present at much higher concentrations.

5 Aldrin and Dieldrin 5.1 History

Aldrin (a product containing 95% of 1,2,3,4,10,10-hexachloro-1,4,4a,5,8,8a-hexahydro-exo-1,4-endo-5,8-dimethanonaphthalene, HHDN) was first prepared in 1948 and was used primarily against insects in soil, where it was effective at rates considerably lower than those of DDT. Dieldrin (a product containing 85% of 1,2,3,4,10,10-hexachloro-6,7-epoxy-1,4,4a,5,6,7,8,8a-octahydro-exo-1,4-endo-5,8dimethanonaphthalene, HEOD) resulted from an attempt to make aldrin less volatile. The formulae outlining the stereochemistry of aldrin, dieldrin and, for comparison, endrin, are presented in Fig. 17. Because of its lower volatility, dieldrin was used on crops and their foliage, in addition to soil and seed dressing applications, both at considerably lower rates than those of DDT. Dieldrin was also very effective against ectoparasites of cattle and sheep, as well as for mothproofing fabrics. In Britain the extensive use of aldrin and dieldrin started in 1956. By the early 1960s, mortalities of birds associated with seeds treated with aldrin or

Fig. 17. The formation of the aldrin and isodrin series of compounds. Reprinted from [27] with permission from IOS Press, Amsterdam, The Netherlands

78

V. Zitko

dieldrin were observed and confirmed in laboratory experiments [160, 189–191]. In addition, dieldrin was detected in human tissues. As a result of these observations and of the potential carcinogenicity of dieldrin, concerns were raised about the safety of aldrin and dieldrin, and their use was gradually limited. In 1962, their application to spring-sown cereals was voluntarily restricted, in 1965 the inclusion of aldrin in fertilizers ceased in Britain, and in 1966 the use of dieldrin in sheep dips ended [52]. In the United States, cancellation proceedings were initiated in 1971, except for subsurface ground insertion for termite control, dipping of plant roots, and moth-proofing in closed systems. The latter appears to have been discontinued by 1986. These changes were reflected in the concentration of dieldrin in shags’ (Phalacrocorax aristotelis) eggs from northeast England and southeast Scotland. For the years 1964 to 1971 the geometric mean concentration of dieldrin, expressed as a function of the year, was log (dieldrin, µg/g)=–1.0124+0.3060 (year-1960)–0.0236 (year-1960)2 [52]. Dieldrin contamination of the Humber estuary in the UK then decreased from dieldrin concentrations of 5–6 mg/l in 1976 to 1–3 ng/l in 1995–1997 [137].Approximately equal concentrations were found in rivers draining primarily agricultural areas.Aldrin and dieldrin have not been produced in the United States since 1974; their application as termiticides has been voluntary canceled in 1987 by the Shell Chemical Company and it appears that now they are not produced and used anywhere in the world. 5.2 Production

Cyclopentadiene (CA RN 542-92-7) and acetylene (CA RN 74-86-2) are condensed to produce bicyclo(2.2.1)heptadiene (I, norbornadiene, CA RN 121-46-0). Hexachlorocyclopentadiene (II, CA RN 77-47-4) is prepared either by reacting pentane (CA RN 109-66-0) and chlorine, or by the reaction of cyclopentadiene (CA RN 542-92-7) and chlorine. The former is a free radical chlorination, catalyzed by illumination or by free radical initiators. The chlorination is carried out until hexa- to octa-chloropentanes are obtained. The chlorination is then continued at 400–425 °C. II is a very toxic compound [138]. The chlorination of cyclopentadiene is also carried in two stages, first at a low temperature and then at 350–500 °C. Alternatively, a hypochlorite may be used at temperatures between –5 and 50 °C. Finally, aldrin results from the reaction of I with II, at about 120 °C, without solvent. After about 18 h (batch process), excess I is distilled off, aldrin is recovered by vacuum distillation and may also be purified by crystallization [181]. Technical aldrin is approximately 82–85% pure and contains 1–3% each of polychlorinated hexahydrodimethanonaphthalenes, di-adducts, bicycloheptadiene, hexachlorocyclopentadiene, hexachlorobutadiene, hexachloroethane, and octachlorocyclopentene [34, 35]. Epoxidation of aldrin yields dieldrin. The epoxidation may be carried out either in hydrogen peroxide in the presence of tungstic oxide as catalyst, in tertbutanol, dioxane, dimethyl formamide, or sulfolane, at 20–75 °C for 1–6 h. Alternatively, the epoxidation may be accomplished by a peracid, such as peracetic acid. In this case, aldrin is dissolved in an organic solvent, such as benzene,

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

79

and the peracid in water [181]. Technical dieldrin is 83% pure and contains, in addition to impurities usually present in technical aldrin, carbonyl compounds and traces of benzene and acetic acid. A detailed description of the chemistry of the cyclodiene insecticides is given in [167]. 5.3 Mode of Action, Transformations, and Toxicology

Aldrin and dieldrin are nerve poisons, but in contrast to DDT seem to attack ganglia. Cyclodiene insecticides interfere with the functioning of the chloride channel by blocking its activation by γ-aminobutyric acid (GABA). The permeability by chloride is decreased, which leads to hyperexcitation [27]. However, as for DDT, the molecular mechanism of their action is not known. In contrast to DDT, fewer cyclodiene analogues have been prepared and tested. However, it appears that the action mechanism depends on the closeness of fit into a receptor cavity, and the presence of two electronegative centers [186]. The metabolism and transformation of aldrin and dieldrin in the environment are very complex. In biota, aldrin is rapidly converted to dieldrin, whose epoxide is opened, yielding dihydroxydieldrin, which can be conjugated and excreted. However, Clostridium sporogenes, Escherichia coli, Streptococcus faecalis, and Lactobacillus rhamnosus, when incubated under nitrogen, convert dieldrin to aldrin [112], with the extent of conversion decreasing in the above order of species. By light, both aldrin and dieldrin are converted to the corresponding ‘photo-’compounds. These are more toxic to aquatic biota than the parent pesticides [80] as well as to mammals, but in view of their relatively low concentration in food, the toxicity of photodieldrin was not considered to change significantly the risk assessment of dietary dieldrin residues [214]. Dieldrin may also be converted to dieldrin ketone or to hydroxy dieldrin, which may then form a glucuronide conjugate. Chemical formulae metabolites and transformation products are given in Fig. 18. The environmentally most important metabolite is 12-ketoendrin. A detailed description of the symptoms of poisoning by aldrin and dieldrin is given by Brooks [34, 35]. Summaries of toxicity data and of environmental concentrations are in [13, 14, 96]. In Britain, the mean intake of dieldrin from 1961 to 1965 was 12.9 mg/day and started to decline somewhat after 1965 [169, 170]. For comparison, the intake of dieldrin by occupationally exposed workers was estimated at 770 mg/day. Concerns about the carcinogenicity of aldrin and dieldrin in humans were not confirmed by a retrospective cohort study of workers involved in the production of aldrin and dieldrin [57]. According to a re-examination of epidemiological data and additional experimental studies, it was suggested that aldrin and dieldrin be classified as ‘not likely a human carcinogen’ compounds [188]. However, a recent epidemiological study suggests that dieldrin has an adverse effect on breast cancer survival [97].

80

V. Zitko

Fig. 18. Metabolites and transformation products of aldrin, dieldrin and endrin. Reprinted

from [27] with permission from IOS Press, Amsterdam, The Netherlands

6 Endrin 6.1 History

Endrin was first used in 1951. Its applications were similar to those of aldrin and dieldrin, but endrin was not used for termite-proofing. Endrin turned out to be particularly effective against pests on cotton and in the tropics. Endrin is much more toxic to non-target species than aldrin and dieldrin and, consequently, its uses in the United States, where endrin was used almost entirely in the southeast, were voluntarily canceled in 1986. It seems that endrin is no longer produced anywhere in the world. However, a Material Safety Data Sheet ‘Endrin-methyl parathion 1.6–1.6 emulsive cotton spray insecticide’, revised on September 16, 1999 is in the ChemWeb.com database [45].

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

81

6.2 Production

Hexachlorocylopentadiene (CA RN 77-47-4) is condensed with acetylene at 150–175 °C and 2000–4000 psi, and the product is condensed with cyclopentadiene (CA RN 542-92-7) at 50–90 °C and atmospheric pressure and, finally epoxidized by peracetic acid. The resulting mixture is steam-distilled to remove excess reagents and solvents and endrin is extracted with ether and dried. The addition of dipicolinic acid (2,6-pyridine dicarboxylic acid, CA RN 499-83-2) during the epoxidation step results in a much purer (>95%) final product, by controlling metal impurities [181]. Technical endrin is typically 96.6% pure and contains traces of aldrin, dieldrin, endrin halfcage ketone, and other minor impurities. 6.3 Transformations and Toxicology

Endrin is a nerve poison, but, in contrast to aldrin and dieldrin, its accumulation in biota is relatively limited because of its relatively rapid biodegradation. The molecular mechanism of endrin’s toxicity is not known. Endrin is metabolized rapidly to 9-hydroxy-endrin and excreted as such or in the form of glucuronide or sulfate. In addition, there are at least four other polar metabolites, which are relatively rapidly excreted. Residues of endrin were not detectable in the blood of occupationally exposed workers and the estimated half-life of endrin in blood is about 24 h. Endrin is much more acutely toxic than dieldrin. For example in a 120-day feeding study with mice, 5 mg/g of endrin or dieldrin in the diet resulted in 33% and 6% mortality, respectively, and both groups of mice produced significantly smaller litters [82]. The toxicology of endrin is described in detail in [12, 207].

7 Mirex 7.1 History

Mirex (Fig. 19) was introduced in 1959, primarily as a stomach poison against ants. Mirex is the only cyclodiene pesticide that also has an industrial application as the flame retardant Dechlorane.A considerable portion of environmental contamination was caused by this application. A derivative of mirex, chlordecone (Fig. 19), is also a pesticide. Although large amounts of mirex were used to control fire ants in the southeastern United States, the amount of mirex used as flame retardant was four times larger [98]. Over 90% of mirex produced in the United Sates was exported to Latin America, Europe, and Africa [166].The production of mirex in the United Sates was discontinued in 1976.

82

V. Zitko

Fig. 19. Formulae of mirex, kepone

7.2 Production

Mirex is prepared from hexachlorocyclopentadiene (CA RN 77-47-4) by dimerization in hexachlorobutadiene (CA RN 87-68-3), catalyzed by anhydrous aluminum chloride at 110 °C. The product is washed with water, excess reactants are removed by vacuum distillation, and mirex is recrystallized from benzene [181]. Chlordecone is produced by a reaction of hexachlorocyclopentadiene with sulfur trioxide in the presence of antimony pentachloride [166]. In the United States, considerable environmental contamination resulted from discharges from a plant manufacturing chlordecone, in Hopewell, Virginia [128], and a plant manufacturing mirex in Niagara Falls, New York [106]. Costa Rica imported 277 tonnes of mirex between 1977 and 1991, of which 166 tonnes were imported from 1989 to 1991 [43]. 7.3 Toxicology

The mechanism of the toxic action is not known. Mirex has a very low acute oral toxicity in vertebrates, but has considerable chronic toxicity. Mirex was first detected in the blubber of a seal (Phoca vitulina) from the North Sea [200] and, shortly after, in fish from Lake Ontario [106]. Mirex is very persistent and has a large bioaccumulation potential. The environmental and toxicological properties of mirex and chlordecone are summarized in [54, 68, 189–191]. In contrast to workers exposed occupationally to aldrin and dieldrin, workers with occupational exposure to mirex and chlordecone developed illnesses related to their exposure. One such serious incident took place in the chlordecone manufacturing plant near Hopewell (Virginia, USA) in 1975 and resulted in poisoning of plant

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

83

workers and contamination of the James river [54]. Similar to other organochlorine compounds such as polychlorinated biphenyls, exposure to mirex resulted in fatty livers, hyperexcitability, and inhibition of reproduction. Chlordecone interferes with estrogen-controlled processes. In the aquatic environment, mirex is not acutely toxic to freshwater invertebrates in concentrations below 1 mg/l. On the other hand, the 48-h EC50 values for daphnids and midges are 260 and 350 mg/l, respectively. The 96-h-LC50 concentration for amphipods is 180 mg/l [172]. There also appears to be a considerable difference between these two pesticides in their uptake and excretion. Bluegills (Lepomis macrochirus), exposed in separate experiments to mirex and chlordecone from food (Daphnia magna) and water, accumulated both compounds linearly with time, for 28 days, at rates of 0.070 and 0.056 mg/(kgday), respectively. After the exposure, chlordecone was eliminated linearly with time, for 28 days, at a rate of 0.03 mg/(kgday), but the concentration of mirex remained practically constant during this period [183].

8 Closing Remarks There are probably more data on the concentration and effects of COC, than on the concentration and effects of any other organic chemical. These data formed the basis for the understanding of the transport, effects and fate of persistent, lipophilic compounds in the environment. Many unanswered questions remain, particularly about the mechanism of COC action at the molecular level, chronic toxicity in individual organisms, and effects at the population level. Many of these questions may never be answered satisfactorily. As long as there is no acceptable alternative, the use of DDT to control malaria is likely to continue.As for the other COC, the only remedial action possible is the prevention of further releases into the environment. Except for ‘hot spots’, nothing economically acceptable can be done about COC, already in the environment, just as it is not feasible to recover, for example, platinum, present in the environment in similar concentrations. COC left in the environment their trace – their residues, relatively easily measurable with the benefit of hindsight. The presence of chlorine atoms in COC molecules is a blessing for the analytical chemist. Unfortunately, many more pesticides and chemicals in general are reaching the environment now than there were when COC were introduced. For example, ethyl parathion, methyl parathion, ethion, mocap, naled, malathion, coumaphos, and zolone were found in the Atoya river (Nicaragua) [42], and methyl parathion, methamidophos, mevinphos, permethrin, carbofuran, and dicofol were present in water and soil of Thailand’s agricultural areas [201]. Many of these compounds have a higher acute toxicity than COC and, at the same time, do not leave a trace and are much more difficult to measure. It is time to move on and concentrate more on currently used and prospective compounds. The 1972 consensus of 100 experts about the hazards of DDT concluded that data on mutagenicity in man are inconclusive, data on carcinogenicity cannot be assessed, and epidemiological studies found no adverse effects, but there is evidence of adverse effects on aquatic fauna and birds

84

V. Zitko

[226]. The opinion has not changed much since then. The extensive use of COC in the ‘early years’ seems inconceivable from today’s perspective. The use still continues in Africa [223] and, everywhere, many other pesticides enter the environment. What will be thought of today’s situation 30 years from now?

9 References 1. 2. 3. 4. 5. 6. 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. 34. 35.

Abbassy MS (2000) Bull Environ Contam Toxicol 65:508 Abbassy MS, Ibrahim HZ, Abu El-Amayem MM (1999) J Environ Sci Health B 34:255 Acker A, Schulte E (1970) Naturwissenschaften 57:497 Acker A, Schulte E (1974) Naturwissenschaften 61:32 Agnihotri NP, Kulshrestha G, Gajbhiye VY, Mohapatra SP, Singh SB (1996) Environ Monit Assessm 40:279 Ahmed FE (ed) (1991) Seafood safety. Committee on Evaluation of the Safety of Fishery Products. National Academy Press, Washington DC Aizawa H (1982) Metabolic maps of pesticides. Academic Press Alegria HA, Bidleman TF, Shaw TJ (2000) Environ Sci Technol 34:1953 Albert R, Horwitz W (1997) Anal Chem 69:789 Albone ES, Eglinton G, Evans NC, Rhead MM (1972) Nature 240:420 Alegria HA, Bidleman TF, Shaw TJ (2000) Environ Sci Tech 34:1953 Anon (1991) Chemical review: endrin, dangerous properties. Ind Mater Rep 11:202 Anon (1988) Hazardous materials: aldrin, dangerous properties. Industr Materials Rpt 8:23 Anon (1986) Chemical review: dieldrin, dangerous properties. Industr Materials Rpt 6:9 Anon (1985) Chem review: DDT, dangerous properties. Industr Materials Rpt 5:12 Baker MT, van Dyke RA (1984) Biochem Pharmacol 33:255 Béland P, DeGuise S, Girard C, Lagacé A, Martineau D, Michaud R, Muir DCG, Norstrom RJ, Pelletier E, Ray S, Shugart LR (1993) J Great Lakes Res 19:766 Bellman SW, Barry TL (1971) JAOAC 54:499 Benguira S, Hontela A (2000) Environ Toxicol Chem 19:842 Bergman Å, Wachtmeister A (1977) Acta Chem Scand B 31:90 Bernt KE, Hammill MO, Lebeuf M, Kovacs KM (1999) Sci Total Environ 243/244:243 Bidleman TF, Christensen EJ, Billings WN, Leonard R (1981) J Mar Res 39:443 Bidleman TF (1999) Water Air Soil Pollut 115:115 Bignert A, Olsson, M, Persson W, Jensen S, Zakrisson S, Litzén K, Eriksson U, Häggberg L, Alsberg T (1998) Environ Pollut 99:177 Bignert A, Göthberg A, Jensen S, Litzén K, Odsjö M, Reutergårdh L (1993) Sci Total Environ 128:121 Björk M (1995) Ann Zool Fennici 32:237 Bloomquist JR (1998) Rev Toxicol 2:333 Blus LJ, Gish CD, Belisle AA, Prouty RM (1972) Nature 235:376 Blus LJ (1996) DDT, DDD, and DDE in birds. In: Beyer WN, Heinz GH, Redmon-Norwood AW (eds) Environmental contaminants in wildlife. SETAC Special Publication Series. CRC Lewis Publishers, Boca Raton, p 49 Boonyatumanond R, Tabucanon MS, Thongklieng S, Boonchlaermkit S (2000) Chem Ecol 17:31 Boul HL (1994) NZ J Agric Res 38:257 Brandt I, Bergman Å (1987) Chemosphere 16:1671 Brandt I, Jönsson CJ, Lund BO (1992) Ambio 21:602 Brooks GT (1974) Chlorinated insecticides, vol I: technology and application. CRC Press Brooks GT (1974) Chlorinated insecticides, vol II: biological and environmental aspects. CRC Press

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

85

36. Bruhn R, Kannan N, Petrick G, Schulz-Bull DE, Duinker JC (1999) Sci Total Environ 237/238:351 37. Burke JA (1971) Residue Revs 34:59 38. Buser HR (1995) Environ Sci Technol 29:2133 39. Buser HR, Müller MD (1995) Anal Chem 67:2691 40. Cameron ME, Metcalfe TL, Metcalfe CD, MacDonald CR (1997) Mar Environ Res 43:99 41. Campbell KR, Campbell TS (2000) Rev Environ Contam Toxicol 165:39 42. Castilho JAA, Fenzl N, Guillen SM, Nascimento FS (2000) Environ Pollut 110:523 43. Castillo LE (1998) Persistent organic pesticides in Costa Rica, http://www.unep.ch/pops/ POPs_Inc/proceedings/ cartagena/CASTILLO.html 44. Cavanagh JE, Burns KA, Brunskill GJ, Coventry RJ (1999) Mar Pollut Bull 39:367 45. ChemWeb.com (2000) Endrin-methyl parathion 1.6–1.6 emulsive cotton spray insecticide. OHS Material safety data sheet 46. Chiou CT, Malcolm RL, Brinton TI, Kile DE (1986) Environ Sci Technol 20:502 47. Cleemann M, Riget F, Paulsen GB, de Boer J, Dietz R (2000) Sci Total Environ 245:103 48. Cocco P, Benichou J (1998) Oncology 55:334 49. Cockroft VG (1999) J Cetacean Res Manage (Spec Issue 1) 169 50. Colborn T, Smolen MJ (1996) Rev Environ Contam Toxicol 146:91 51. Comber S, Gardner M (1999) Sci Total Environ 243/244:193 52. Coulson JC, Deans IR, Potts GR, Robinson J, Crabtree AN (1972) Nature 236:454 53. Crellin NK, Rodway MR, Swan CL, Gillio-Meina C, Chedrese PJ (1999) Biol Reprod 61:1099 54. Dai D, Rose RL, Hodgson E (1998) Rev Toxicol 2:477 55. Dave G, Nilsson E (1999) Aquat Ecosyst Health Manag 2:347 56. Deichmann WB, Radomski JL (1968) Ind Med Surgery 37:218 57. De Jong G, Swaen GMH, Slangen JJM (1997) Occupat Environ Med 54:702 58. Dewailly É, Mulvad G, Pedersen HS, Ayotte P, Demers A, Weber J-P, Hansen JC (1999) Environ Health Perspect 107:823 59. Dewailly É,Ayotte P, Bruneau S, Laliberté C, Muir DCG, Norstrom R (1993) Environ Health Perspect 101:618 60. Dimond JB, Owen RB (1996) Environ Pollut 92:227 61. Doguchi M (1973) Chlorinated hydrocarbons in the environment in the Kanto Plain and Tokyo Bay, as reflected in fishes, birds and man. In: Coulston F, Korte F, Goto M (eds) New methods in environmental chemistry and toxicology. Int Symp Ecol Chem, Susono, International Academic Printing, Totsuka, Tokyo, p 270 62. Donato MM, Jurado AS, Antunes-Madeira MC, Madeira VMC (2000) Arch Environ Contam Toxicol 39:145 63. DouAbul AAZ, Al-Omar M, Al-Obaidy S, Al-Ogaily N (1987) Bull Environ Contam Toxicol 38:674 64. Duce RA, Liss PS, Merrill JT, Atlas EL, Buat-Menard P, Hicks BB, Miller JM, Prospero JM, Arimoto R, Church TM, Ellis W, Galloway JN, Hansen L, Jickells TD, Knap AH, Reinhardt KH, Schneider B, Soudine A, Tokos JJ, Tsunogai S, Wollast R, Zhou M (1991) Global Biogeochem Cycles 5:193 65. Eriksson G, Jensen S, Kylin H, Strachan W (1989) Nature 341:42 66. Fahmy MAH, Fukuto TR, Metcalf RL, Holmstead RL (1973) J Agric Food Chem 21:585 67. Falandysz J, Strandberg B, Strandberg L, Rappe C (1999) Environ Sci Technol 33; 517 68. Faroon O, Kueberuwa S, Smith L, Derosa C (1995) Toxicol Industr Health 11:1 69. Fiedler H, Lau C (1998) Environmental fate of chlorinated organics. In: Schüürmann G, Markert B (eds) Ecotoxicology. Wiley, p 317 70. Focht DD, Alexander M (1970) Science 170:91 71. Frank R, Braun HE, Fleming G (1983) J Food Protection 46:893 72. Gaines TB (1969) Toxicol Appl Pharmacol 14:515 73. Galiulin RV, Bashkin VN (1996) Water Air Soil Pollut 89:247 74. Garrison AW, Nzengung VA,Avants JK, Ellington JJ, Jones WJ, Rennels D,Wolfe NL (2000) Environ Sci Technol 34:1663

86

V. Zitko

75. Gaskin DE, Holdrinet M, Frank R (1971) Nature 233:499 76. Gottholm BW, Turgeon DD (1992) Toxic contaminants in the Gulf of Maine. National Oceanic and Atmospheric Administration, Rockville, MD 20852, USA 77. Gauthier JM, Metcalfe CD, Sears R (1997) Marine Environ Res 44:201 78. Gil O, Vale C (1999) Aquat Ecol 33:263 79. Gladen BC, Monaghan SC, Lukyanova EM, Hulchiy OP, Shkyryak-Nyzhnyk ZA, Sericano JL, Little RE (1999) Environ Health Perspect 107:459 80. Georgacakis E, Khan MAQ (1971) Nature 233:120 81. Gold B, Brunk G (1984) Biochem Pharmacol 33:979 82. Good EE, Ware GW (1969) Toxicol Appl Pharmacol 14:201 83. Glynn AW, Wernroth L, Atuma S, Linder CE, Aune M, Nilsson I, Darnerud PO (2000) Sci Total Environ 246:195 84. Granby K, Kinze CC (1991) Marine Pollut Bull 22:458 85. Gustavson K, Jonsson P (1999) Mar Pollut Bull 38:723 86. Hall AJ, Duck CD, Law RJ, Allchin CR, Wilson S, Eybator T (1999) Environ Pollut 106:203 87. Haraguchi K, Endo T, Sakata M, Masuda Y, Simmonds M (2000) J Food Hygienic Soc Japan 41:287 88. Harris CR, Miles JRW (1975) Residue Rev 57:27 89. Harris CA, O’Hagan S, Merson GHJ (1999) Human Exp Toxicol 18:602 90. Harris O, McLellan W (1994) Toxicological profile for 4,4′-DDT, 4′4′-DDE, 4,4′-DDD (update). US Department of Health and Human Services TP-93/05 91. Harris ML, Wilson LK, Elliott JE, Bishop CA, Tomlin AD, Henning KV (2000) Arch Environ Contam Toxicol 39:205 92. Herrera A, Ariòo A, Conchello MP, Lazaro R, Bayarri S, Yagüe C, Peiro JM, Aranda S, Simon MD (2000) Arch Environ Contam Toxicol 38:114 93. Holoubek I, Korˇínek P, Sˇeda Z, Schneiderová E, Holoubková I, Pacl A, Trˇíska J, Cudlín P, Cˇáslavskyˇ J (2000) Environ Pollut 109:283 94. Hoff RM, Strachan WMJ, Sweet CW, Chan CH, Shackleton M, Bidleman TF, Brice KA, Burniston DA, Cussion S, Gatz DF, Harlin K, Schroeder WH (1996) Atmosph Environ 30:3505 95. Hong H, Chen W, Xu L, Wang X, Zhang L (1999) Marine Pollut Bull 39:376 96. Howard PH (1991) Handbook of environmental fate and exposure data for organic chemicals. Lewis Publishers, Chelsea MI USA 97. Høyer AP, Jørgensen T, Brock JW, Grandjean P (2000) J Clin Epidemiol 53:323 98. Huckins JN, Stalling DL, Petty JD, Buckler DR, Johnson BT (1982) J Agric Food Chem 30:1020 99. Hutchinson JD, Simmonids MP (1994) Rev Environ Contam Toxicol 136:123 100. Iida T, Hirakawa H, Matsueda T, Nakagawa R, Hori T, Nagayama J (1999) Organohalogen Compds 44:123 101. Iwata H, Tanabe S, Sakai N, Tatsukawa R (1993) Environ Sci Technol 27:1080 102. Janák K, Becker G, Colmsjö A, Östman C, Athanasiadou M, Valters K, Bergman Å (1998) Environ Toxicol Chem 17:1046 103. Jarman WM, Norstrom RJ, Muir DCG, Rosenberg B, Simon M, Baird RW (1996) Marine Pollut Bull 32:426 104. Juberg DR (2000) Ecotoxicol Environ Safety 45:93 105. Kannan K, Tanabe S, Tatsukawa R (1994) Toxicol Environ Chem 42:249 106. Kaiser KLE (1974) Science 185:523 107. Kapoor IP, Metcalf RL, Hirwe AS, Coats JR, Khalsa MS (1973) J Agric Food Chem 21:310 108. Karlson K, Ishaq R, Becker G, Berggren P, Broman D, Colmsjö A (2000) Environ Pollut 110:29 109. Keith JO (1996) Residue analyses: how they were used to assess the hazards of contaminants to wildlife. In: Beyer WN, Heinz GH, Redmon-Norwood AW (eds) Environmental contaminants in wildlife. SETAC Special Publication Series. CRC Lewis Publishers, Boca Raton 110. Kelce WR, Stone CR, Laws SC, Gray LE, Kemppainen JA, Wilson EM (1995) Nature 375:581

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

87

111. Khim JS,Villeneuve DL, Kannan K, Hu WY, Giesy JP, Kang S-G, Song K-J, Koh C-H (2000) Arch Environ Contam Toxicol 39:360 112. Kitamura S, Mita M, Shimizu Y, Sugihara K, Ohta S (1999) Biol Pharm Bull 22:880 113. Kleivane L, Severinsen T, Skaare JU (2000) Marine Environ Res 49:343 114. Kleivane L, Skaare JU, Bjørge, de Ruiter E, Reijners PJH (1995) Environ Pollut 89:137 115. Koeman JH, van Genderen H (1970) Tissue levels in animals and effects caused by chlorinated hydrocarbon insecticides, biphenyls and mercury in the marine environment along the Netherlands’ coast, FAO Tech Conf Marine Pollut Effects Living Resources Fishing, Rome, FIR:MP/70/E-21 116. Konstantinou IK, Goutner V, Albanis TA (2000) Sci Total Environ 257:61 117. Kostamo A, Medvedev N, Pellinen J, Hyvärinen H, Kukkonen JVK (2000) Environ Toxicol Chem 19:848 118. Kramer VJ, Giesy JP (1999) Sci Total Environ 233:141 119. Kundiev YI (2000) State of affairs in Ukraine and other CIS countries. http://www.chem.unep.ch/pops/POPs_Inc/proceedings/stpetbrg/kundiev09.10.00 120. Kutz FW, Wood PH, Bottimore DP (1991) Rev Environ Contam Toxicol 120:1 121. Larsson P, Berglund O, Backe C, Bremle G, Eklöv A, Järnmark C, Persson A (1995) Naturwiss 82:559 122. Lauenstein GG (1995) Marine Pollut Bull 30:826 123. Li YF, Cai DJ, Singh A (1999) Adv Environ Res 2:497 124. Lordo RA, Khoan TD, Schwemberger JG (1996) Am J Publ Health 86:1253 125. López-Carillo L, Torres-Arreola L, Torres-Sánchez L, Espinosa-Torres F, Jiménez C, Cebrián M, Waliszewski S, Saldate O (1996) Environ Health Perspect 104:584 126. Lund BO, Lund J (1995) J Biol Chem 270:20,895 127. Lund BO, Örberg J, Bergman Å, Larsson C, Bergman A, Bäcklin BM, Håkansson H, Madej A Brouwer A, Brunström B (1999) Environ Toxicol Chem 18:292 128. Lunsford CA (1981) Pestic Monit J 14:119 129. Mackay D, Paterson S (1981) Environ Sci Technol 15:1006 130. Marsili L, Focardi S (1997) Environ Monit Assess 45:129 131. Marsili L, Casini C, Marini L, Regoli A, Focardi S (1997) Mar Ecol Prog Ser 151:273 132. Massé R, Lalanne D, Messier F, Sylvestre M (1989) Biomed Environ Mass Spec 18:741 133. Massé R, Martineau D, Tremblay L, Béland P (1986) Arch Environ Contam Toxicol 15:567 134. Matsumura F, O’Brien RD (1966) J Agric Food Chem 14:36 135. Matsumura F, Patil KC (1969) Science 166:121 136. McBlain WA, Lewin V, Wolfe FH (1976) Can J Physiol Pharmacol 54:629 137. Meharg AA, Wright J, Leeks GJL, Wass P, Osborn D (2000) Sci Total Environ 251/252:255 138. Melnikov NN (1971) Chemistry of pesticides. Springer, Berlin Heidelberg New York 139. Mes J, Davies DJ, Turton D (1985) Ecotoxicol Environ Safety 10:70 140. Metcalf RL (1973) J Agric Food Chem 21:511 141. Minh TB,Watanabe M, Nakata H, Tanabe S, Jefferson TA (1999) Marine Pollut Bull 39:383 142. Mohapatra SP, Kumar M, Gajbhiye VT, Agnihotri NP (1995) Environ Monit Asses 35:155 143. Monirith I, Nakata H, Tanabe S, Tana TS (1999) Mar Pollut Bull 38:604 144. Morra MM, Wainwright SE (1998) Rev Environ Contam Toxicol 158:1 145. Morris RJ, Law RJ, Allchin CR, Kelly CA, Fileman CF (1989) Marine Pollut Bull 20:512 146. Morrison DE, Robertson BK, Alexander M (2000) Environ Sci Technol 34:709 147. Muir DCG,Wageman R, Grift NP, Norstrom RJ, Simon M, Lien J (1988) Arch Environ Contam Toxicol 17:613 148. Nakata H, Tanabe S, Tatsukawa R, Koyama Y, Miyazaki N, Belikov S, Boltunov A (1998) Environ Toxicol Chem 17:1745 149. Narahashi T (1969) Residue Revs 25:275 150. Ngabe B, Bidleman TF, Leone AD, Falconer RL,Wiberg K, Harner T (1999) Am Chem Soc, Div Environ Chem, Preprints Extend Abstracts 39:213 151. Nhan DD, Am NM, Carcalho FP, Villeneuve JP, Cattini C (1999) Sci Total Environ 237/238:363

88

V. Zitko

152. Nichols WK, Terry CM, Cutler NS, Appleton ML, Jesthi PK, Yost GS (1995) Drug Metab Dispos 23:595 153. Norén K, Meironyté D (2000) Chemosphere 40:1111 154. Norén K, Weistrand C, Karpe F (1999) Arch Environ Contam Toxicol 37:408 155. Ohlendorf HM, Risebrough RW,Vermeer K (1978) Exposure of marine birds to environmental pollutants. US Department of the Interior, Fish and Wildlife Service, Wildlife Res Rept 9, Washington, DC 156. On’Eponga O, de Lavaur E, Sech JL, Lich NP, Moan GL (1984) Ann Fals Exp Chim 77:531 157. Olsson M, Bignert A, Echhéll J, Jonsson P (2000) Ambio 29:195 158. Paterson S, Mackay D (1989) Ecological Modelling 47:85 159. Paumgartten FJR Cruz CM, Chahoud I, Palavinkas R, Mathar W (2000) Environ Res A 83:293 160. Peakall DB (1996) Dieldrin and other cyclodiene pesticides in wildlife. In: Beyer WN, Heinz GH, Redmon-Norwood AW (eds) Environmental contaminants in wildlife. SETAC Special Publication Series. CRC Lewis Publishers, Boca Raton 161. Polder A, Becher G, Savinova TN, Skaare JU (1998) Chemosphere 37:1795 162. Portelli MJ, de Solla SR, Brooks RJ, Bishop CA (1999) Ecotoxicol Environ Safety 43:284 163. Quaife ML, Winbush JS, Fitzhugh OG (1967) Food Cosmet Toxicol 5:39 164. Raloff J (2000) Sci News Online 158:1 165. Reish DJ, Oshida PS, Mearns AJ, Ginn TC, Buchman M (1999) Water Environ Res 71:1100 166. Research Triangle Institute (1995) Toxicological profile for mirex and chlordecone. US Dept Health Human Services Contract No 205-93-506 167. Riemschneider R (1963) World Rev Pest Control 2:29 168. Robertson BK, Alexander M (1998) Environ Toxicol Chem 17:1034 169. Robinson J (1968) Chem Br 4:158 170. Robinson J, Roberts M (1969) Food Cosmet Toxicol 7:501 171. Roots O (1996) Toxic chlororganic compounds in the ecosystem of the Baltic Sea. Estonian Environment Information Centre (EEIC) Tallinn 172. Sanders HO, Huckins J, Johnson BT, Skaar D (1981) Arch Environ Contam Toxicol 10:531 173. Sarkar A, Gupta RS (1988) Bull Environ Contam Toxicol 41:664 174. Savinov V, Savinova T, Dahle S, Matishov G, Iljin G (1999) Organohal Compd 43:409 175. Schmitt CJ, Zajicek JL, May TW, Cowman DF (1999) Rev Environ Contam Toxicol 162:43 176. Schoula R, Hajslova J, Gregor P, Kocourek V, Bencko V (1998) Toxicol Environ Chem 67:263 177. Senthilkumar K, Kannan K, Sinha RK, Tanabe S, Giesy JP (1999) Environ Toxicol Chem 18:1511 178. Shaw I, Burke E, Suharyanto F, Sihombing G (2000) Environ Sci Pollut Res 7:75 179. Shiu WY, Ma KC, MacKay D, Seiber JN, Wauchope RD (1990) Rev Environ Contam Toxicol 116:1 180. Simonich SL, Hites RA (1997) Environ Sci Tech 31:999 181. Sittig M (1977) Pesticides process encyclopedia. Noyes Data Corporation, Park Ridge 182. Siewierski M, Helrich K (1967) JAOAC 50:627 183. Skaar DR, Johnson BT, Jones JR, Huckins JN (1981) Can J Fish Aquat Sci 38:931 184. Smith JA, Witkowski PJ, Chiou CT (1988) Revs Environ Contam Toxicol 103:127 185. Snyder NFR, Snyder HA, Lincer JL, Reynolds RT (1973) BioScience 23:300 186. Soloway SB (1965) Correlation between biological activity and molecular structure of the cyclodiene insecticides. In: Metcalf RL (ed) Advances in pest control research. Interscience Publishers, New York, p 85 187. Spindler M (1983) Residue Rev 90:1 188. Stevenson DE, Walborg EF Jr, North DW, Sielken RL Jr, Ross CE, Wright AS, Xu Y, Kamendulis LM, Klaunig JE (1999) Toxicol Let 109:123 189. Stickel L, Stickel W (1969) Ind Med 38:44 190. Stickel WH, Stickel LF, Spann JW (1969) Tissue residues of dieldrin in relation to mortality in birds and mammals. In: Miller MW, Berg GG (eds). Chemical fallout, current research

Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

191.

192. 193. 194. 195. 196. 197. 198. 199. 200. 201. 202. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 213. 214. 215. 216. 217. 218. 219. 220. 221. 222. 223. 224. 225.

89

on persistent pesticides. Proc 1st Rochester Conf Toxic, Charles C Thomas, Springfield, p 174 Stickel WH, Galyen JA, Dyrland RA, Hughes DL (1973) Toxicity and persistence of mirex. In: Deichmann WB (ed), Pesticides and the environment: a continuing controversy.... Selected papers presented and papers reviewed at the 8th Inter-American Conference on Toxicology and Occupational Medicine, University of Miami, School of Medicine, Miami, Florida, vol 8, p 437 Storelli MM, Stuffler RG, Ceci E, Marcotrigiano GO (1999) In: Marine pollution. IEAETECDOC-1094, p 409 Storelli MM, Marcotrigiano GO (2000) Marine Pollut Bull 40:555 Subramanian A, Tanabe S, Tatsukawa R (1988) Mar Pollut Bull 19:284 Sundström G, Jansson B, Jensen S (1975) Nature 255:627 Tanabe S (ed) (2000) Mussel watch marine pollution monitoring in Asian waters. Center for Marine Environmental Studies, Ehime University Tanabe S, Loganathan BG, Subramanian A, Tatsukawa R (1987) Marine Pollut Bull 18:561 Tanabe S, Tatsukawa R, Tanaka H, Maruyama K, Miyazaki N, Fujiyama T (1981) Agric Biol Chem 45:2569 Tanabe S, Gondaira F, Subramanian A, Ramesh A, Mohan D, Kumaran P,Venugopalan VK, Tatsukawa R (1990) J Agric Food Chem 38:899 Ten Noever de Brauw MC, van Ingen C (1973) Sci Total Environ 2:196 Thapinta A, Hudak PF (2000) Environ Monit Assess 60:103 Thompson JN (1973) Statistics on the use of pest control products in Canada. Environmental contaminants inventory study No. 1. Environment Canada, Rportt Series No 35. Information Canada, Ottawa, Cat No En 36–508/35 Tilbury KL, Adams NG, Krone CA, Meador JP, Early G, Varanasi U (1999) Arch Environ Contam Toxicol 37:125 Tinsley IJ, Haque R, Schmedding D (1971) Science 174:145 Torres JPM, Malm O, Vieira EDR, Japenga J, Koopmans G (1999) J Brazilian Assoc Advancement Science 51:54 Tuomisto J, Hagmar L (1999) Scand J Environ Health 25 (suppl) 3:65 Van Esch GT, van Heemstra-Lequin EAH (1992) Environmental Health Criteria 130. Endrin,World Health Organization, Geneva Vazquez-Botello A, Diaz-Gonzales G, Rueda-Quintana L (1999) Marine pollution. Proceedings of a symposium held in Monaco, 5–9 October 1998, IAEA-TEDOC-1094:135 Verhaar HJM, De Jongh J, Hermens JLM (1999) Environ Sci Technol 33:4069 Verschueren K (1996) Handbook of environmental data on organic chemicals. Van Norstrand Reinhold, New York Vetter W, Weichbrodt M, Scholz E, Luckas B, Oelschläger H (1999) Mar Pollut Bull 38:830 Villeneuve JP, Carvalho FP, Fowler SW, Cattini C (1999) Sci Total Environ 237/238:57 Voldner EC, Li YF (1995) Sci Total Environ 160/161:201 Walton MS, Bastone VB, Baron RL (1971) Toxicol Appl Pharmacol 20:82 Watanabe M, Tanabe S, Miyazaki N, Petrov A, Jarman WM (1999) Marine Pollut Bull 39:393 Wauchope RD (2000) Chem Innovation 30(9):33 Weiderpass E, Adami HO, Baron JA, Wicklund-Glynn A, Aune M, Atuma S, Persson I (2000) Cancer Epidemiol Biomarkers Prev 9:487 Weisbrod AV, Shea D, Moore MJ Stegeman JJ (2000) Environ Toxicol Chem 19:667 Weisbrod AV, Shea D, Moore MJ Stegeman JJ (2000) Environ Toxicol Chem 19:654 Weisbrod AV, Shea D, Moore MJ, Stegeman JJ (2000) Mar Environ Res 50:435 Westgate AJ, Tolley KA (1999) Mar Ecol Prog Ser 177:255 Wiemeyer SN, Schmeling SK, Anderson A (1987) J Wildl Dis 23:279 Wiktelius S, Edwards CA (1997) Rev Environ Contam Toxicol 151:1 Wilson WE, Fishbein L, Clements ST (1971) Science 171:180 Woodwell GM, Craig PP, Johnson HA (1975) DDT in the biosphere: where does it go? In: Singer SF (ed .) The changing global environment, D Reidel Publishing, Dordrecht, Holland, 295–309

90

V. Zitko: Chlorinated Pesticides: Aldrin, DDT, Endrin, Dieldrin, Mirex

226. World Health Organization (1972) Health hazards of the human environment. WHO Geneva 227. You L, Brenneman KA, Heck HïA (1999) Toxicol Appl Pharmacol 161:258 228. Zeidler O (1874) Ber Deutsch Chem Gesell 7:1180 229. Zhang G, Min YS, Mai BX, Sheng GY, Fu JM, Wang ZS (1999) Marine Pollut Bull 39:326

CHAPTER 5

Hexachlorobenzene Vladimir Zitko 114 Reed Ave, St. Andrews, NB, E5B 1A1, Canada E-mail: [email protected]

Hexachlorobenzene (HCB) was used as a fungicide, and as an intermediate in the production of pentachlorophenol and other pesticides. In addition, it is generated unintentionally in a number of industrial processes involving chlorine and organic compounds, such as the production of chlorinated solvents, or processes involving chlorine and carbon, such as the production of magnesium. HCB is very lipophilic, persistent, and has Henry’s law constant that favors a wide dispersion in the environment and accumulation in biota. HCB has a low acute toxicity, but considerable chronic toxicity and is classified as a probable human carcinogen. HCB caused a mass poisoning of humans in Turkey and several localized high contamination problems. It appears that all intentional uses of HCB now have been canceled and steps are underway to eliminate environmental releases of HCB from industrial processes as much as achievable by current technology. Keywords: Fungicide, Properties, Sources, Concentration, Toxicity, Risk, Incidents

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

1

Introduction

2

Chemical and Physical Properties

3

Production

4

Sources of Environmental Releases . . . . . . . . . . . . . . . . . 95

5

Concentrations in the Environment

5.1 5.2 5.3 5.4 5.4.1 5.4.2 5.4.3 5.5 5.6 5.7

Air . . . . . . . . . . . . . . . . . . Water . . . . . . . . . . . . . . . . Soil and Sediment . . . . . . . . . Aquatic Biota . . . . . . . . . . . . Molluscs . . . . . . . . . . . . . . Zooplankton, Amphipods, and Fish Marine Mammals . . . . . . . . . Birds . . . . . . . . . . . . . . . . Terrestrial Mammals . . . . . . . . Humans . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . 93

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

. . . . . . . . . . . . . . . . 95 . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

95 96 96 98 98 100 102 103 105 105

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

92

V. Zitko

6

Toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

6.1 6.2 6.3 6.4 6.5

Aquatic Biota . . . . . Birds and Mammals . Pharmacodynamics . Mechanism of Toxicity Risk Assessment . . .

7

Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . 117

8

References

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

110 110 111 113 115

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

1 Introduction Hexachlorobenzene (HCB) was prepared for the first time by Faraday in 1824. Its first use as a fungicide started in the 1940s, primarily for the seed treatment of wheat against smut. This application was expanded briefly in the early 1970s to replace mercury-based fungicides. At about the same time, a widespread HCB contamination of the environment was detected, starting with the detection of HCB in fish, facilitated by a procedure developed by Holden and Marsden [72], that separated HCB from the hexachlorocyclohexanes during the sample cleanup. It was soon realized that considerable environmental contamination was also caused by unintentional HCB discharges from various industrial processes and from applications of products that contain HCB as an impurity. The consumption of HCB-treated wheat in Turkey resulted in the poisoning of a large number of people and a high mortality of children in 1956–1959. Later,‘hot spots’ generated primarily by handling chlorinated wastes and the production of pentachlorophenol were discovered in the United States (Louisiana), in Brazil, and in Spain. HCB is on the list of persistent organic pollutants (POPs) subject to an international ban. A search of Chemical Abstracts showed that the number of papers published annually on HCB has increased about ten times from the 1960s to the 1990s (Fig. 1). Chemical, toxicological, and environmental properties of HCB, as well as its amounts released into the environment, are the subject of a number of excellent reviews (see for example [10, 19, 68, 73, 129]), and of an international symposium [109]. A large file of annotated bibliography is also available on the National Library of Medicine TOXNET system [151]. Consequently, there is no point in writing yet another review and this chapter is intended as a brief overview, which mentions only a fraction of the work available in the open literature, the ‘gray’ literature, and in the electronic media.

Hexachlorobenzene

93

Fig. 1. Number of references in Chemical Abstracts, retrieved by a search for ‘hexachloro-

benzene’

2 Chemical and Physical Properties HCB is a stable, hydrophobic compound with low vapor pressure and a very low solubility in water (Table 1). However, the value of their ratio (Henry’s Law constant), and the chemical stability of HCB, lead to a widespread distribution of HCB in the environment. In terms of ‘spatial range’, which is a characteristic calculated from the distribution coefficients and the decay constants of chemicals [114], HCB ranks fourth, after trichlorofluoromethane, 1-chloro-1,1-difluoroethane, and carbon tetrachloride. In spite of its stability, HCB reacts relatively easily with nucleophiles, particularly in pyridine [130]. The technical HCB may contain almost 2% of pentachlorobenzene (QCB), 0.2% 1,2,4,5-tetrachlorobenzene, hepta- and octachlorodibenzofurans (0.45 and 2.83 mg/kg respectively) [10], hepta- and octachlorodibenzo-p-dioxin (0.47 and 6.7 mg/kg, respectively) [10], decachlorobiphenyl, and other, only tentatively identified compounds [73, 157]. Pentachlorobenzene has been seldom measured in environmental samples. Literature contains some data on HCB and other chlorobenzenes in UK watersheds [101].

94

V. Zitko

Table 1. Properties

Formula

C6Cl6

Molecular weight Melting point, °C Boiling point, °C Flash point, °C Density @ 20°C, g/cm3 Aqueous solubility, g/m3 @ 5°C Aqueous solubility, g/m3 @ 15°C Aqueous solubility, g/m3 @ 25°C Aqueous solubility, g/m3 @ 35°C Aqueous solubility, g/m3 @ 45°C Octanol/water partition coeff, log Kow @ 5°C Octanol/water partition coeff, log Kow @ 15°C Octanol/water partition coeff, log Kow @ 25°C Octanol/water partition coeff, log Kow @ 35°C Octanol/water partition coeff, log Kow @ 45°C Vapor pressure, mm Hg @ 15°C log p=12.94–5279/T Vapor pressure, mm Hg @ 25°C Vapor pressure, mm Hg @ 35°C Vapor pressure, mm Hg @ 45°C Henry’s law constant, atm m3/mol UV absorption maximum, nm Specific absorbance @ 300 nm, l/ml/cm Specific absorbance @ 320 nm, l/ml/cm Sediment/water partition coefficient, log Koc

284.79 229 322 242 2.075 0.0022 0.0035 0.00544 0.00853 0.014

Dissolved organic carbon/water partition coefficient, log Kdoc

5.74

226 332

230

[139]

[17]

5.6 5.46 5.3 5.17 4.07E–06

[50]

1.68E–05 6.32E–05 0.000218 6.80E–04 1.30E–03 1.20E–03 [10] 291 230 23 5.53 4.99

[78] [83]

5.49

[87]

3 Production It appears that HCB is not produced intentionally anywhere in the world. In the past it was prepared by chlorination of either benzene, less chlorinated chlorobenzenes, or non-pesticidal isomers of hexachlorocyclohexane. The latter could also be converted into HCB by oxidation with air [102]. Almost theoretical yields of HCB are attained by treating chlorinated cyclohexanes with sulfuric acid chloride or anhydride at 130–200 °C in the presence of ferric or aluminum chloride [141].

Hexachlorobenzene

95

4 Sources of Environmental Releases As a fungicide and a seed disinfectant, HCB, formulated as powders or suspensions, was used at a rate of 0.6–1.0 kg/ton seed, usually in combination with other active ingredients. HCB supplemented and replaced partially, and later totally, organomercury compounds in seed disinfectants.An ‘interim’ product contained 1% ethylmercuric chloride, 20% HCB, and 20% lindane [102]. Products registered in Canada in 1974 contained HCB, captan, and aldrin in concentrations 1–2, 1.6–2.0, and 2.5 kg/l, respectively. Other products contained lindane at 3 kg/l instead of aldrin and, in addition, maneb at 1.2–1.5 kg/l [9]. Some formulations also contained mancozeb, methoxyethyl mercury silicate, or fuberidazol (2-(2-furanyl)-1H-benzimidazole, CA RN 3878-19-1) [10]. In addition to the use as a fungicide, HCB was used in the industry for the treatment of carbon anodes, for the degassing of aluminum, for synthetic rubber peptizing by the conversion to pentachlorothiophenol, and in the manufacture of pesticides, such as pentachlorophenol, the hexachlorocyclohexanes, pentachloronitrobenzene, dimethyl-tetrachloroterephthalate, picloram, and chlorothalonil. Such pesticides may have been considerable sources of HCB. For example, the annual consumption of quintozene in California in 1963 was about 180 tons [33]. A minor use, based on the volatilization of HCB in the form of white smoke, or its decomposition to yield chloride ions that intensify light, include military smoke and light signals [10]. HCB is also formed unintentionally during the preparation or use of other organochlorine compounds, such as chlorinated solvents, or during the degassing of aluminum by hexachloroethane. Bailey [19] estimates that total HCB emissions are 12–92 tons/year, not considering emissions in developing countries. Increased control over combustion processes and phasing out the uses of many organochlorine compounds are likely to decrease the emissions of HCB.

5 Concentrations in the Environment The concentration of HCB in the environment, more than the concentrations of many other organochlorine compounds, is quite closely related to the level of industrial development. This was demonstrated by correlating the concentrations of organochlorine compounds in the vegetation or in the bark of trees with the gross national product or the human development index of different countries [32, 140]. In this respect, HCB is similar to PCB [172]. 5.1 Air

Practically all HCB in the air is present in the gas phase [125]. The concentration of HCB in the air generally ranges from non-detectable to about 0.1 ng/m3 in pristine areas, 0.19 ng/m3 in the Arctic [51], from 0.1 to 0.2 ng/m3 in populated areas, and may be considerably higher in industrial locations. In 1979, the concentra-

96

V. Zitko

tions of HCB were 0.10, 0.15, 0.20, and 0.12 at the Enewetak Atoll in the North Pacific, in the North Atlantic, in College Station, Texas, and in Pigeon Key, Florida, respectively [13]. In the Great Lakes area, the concentration of HCB in the air decreased from about 0.1 ng/m3 in 1992 to approximately 0.05–0.06 ng/m3 in 1997 [18]. These values are only rough benchmarks of the typical orders of magnitude. Considerably higher concentrations of HCB, ranging from 0.317 to 0.594 ng/m3 were reported in samples collected in 1993/94 in France [132]. A much wider range of concentrations, from 0.099 to 4.74 ng/m3, was reported in samples collected in 1995/96 in Slovakia [86]. These high concentrations of HCB indicate a considerable HCB contamination of Slovakia, also indicated by relatively high concentrations of HCB in the population (see Figs. 11 and 14). 5.2 Water

The concentrations of HCB water range from non-detectable to about 0.04 ng/l in pristine, and up to about 2 ng/l in populated areas. Industrial locations may contain considerably higher concentrations. For example, the concentrations of HCB in the Elbe (Germany) ranged from about 2 to 20 ng/l between February 1990 and April 1991 [59]. Surprisingly high concentrations were reported from Egypt [1, 2]. In the Rosetta and Damietta regions of the Nile the concentrations were around 200 ng/l in 1995–1997. In the Alexandria region, the concentrations were approximately 300 ng/l in 1997–1998. The concentrations were quite constant throughout the years and approximately equal to the concentrations of lindane at two locations in the Alexandria region. However, at one station, the HCB concentration was about 800 ng/l, twice the concentration of lindane. The concentrations of HCB in the Atlantic Ocean are reported from non-detectable to about 0.028 ng/l [51, 90]. The concentrations in the surface film were approximately 10 times higher [90]. QCB (pentachlorobenzene) concentrations are seldom reported and, if given, they are much lower than those of HCB. One exception is a report of QCB concentrations at least one order of magnitude higher than those of HCB in sea water off Java, Indonesia [71]. The Maximum Permissible Concentration (MPC) and the Environmental Quality Standard (EQS), used in The Netherlands [36] for HCB, are 2400 ng/l and 9 ng/l, respectively. The corresponding values for pentachlorobenzene (QCB) are 7500 and 300 ng/l. The MPC is a concentration above which the risk of adverse effects is unacceptable. EQS is used as a standard in the assessment of environmental quality [36]. 5.3 Soil and Sediment

The concentration of HCB in soil ranges from non-detectable to about 5 ng/g dry weight in pristine, to about 70 ng/g dry weight in populated, and, possibly, to hundreds of ng/g dry weight in industrial areas. In 1992 samples of Canadian agri-

Hexachlorobenzene

97

cultural soils the concentrations ranged from <0.5 to 1.6 ng/g, and from <0.5 to 2.1 ng/g in municipal sludge treated soil [162]. In the Merseyside, United Kingdom, area, the chlorination of benzene started in about 1900 and, specifically, the HCB production in the 1930s. The maximum concentration of HCB in the local salt marshes reached 9–22 ng/g. The dated sediment profiles have maxima in 1913, 1926, 1946, and 1965 [56]. The concentrations of HCB in the Elbe (Germany) sediments were 200–400 ng/g between February 1990 and April 1991 [59]. Large differences in environmental concentrations of HCB in sediments are illustrated in Fig. 2 by the data of the US National Status and Trends (NST) program [153] (see its URL in the References). There is a five orders of magnitude range of HCB and PCB concentrations in the coastal sediments of the United States. Limited data from Scotland show higher concentrations of both HCB and PCB (insert, Fig. 2). The concentrations of HCB are in most cases several orders of magnitude lower than the concentrations of PCB.An interesting exception is agricultural soil in the Philippines, which contains HCB at 6.6 ng/g dry soil, as compared to the average PCB concentration of <1.0 ng/g [96]. The concentration of HCB and PCB in sewage sludge was quite closely correlated (r2 =0.7, n=9) [113] and the HCB concentration ranged from 6 to 125 ng/g of dry sludge in the early 1980s. At the same time, the application of such sludge to agricultural land did not result in an appreciable accumulation of HCB [113]. Over a period of about 10 years the concentrations have not changed and in 1989, they were from <0.5 ng/g to 118 ng/g of dry sludge, with a median concentration

Fig. 2. Concentration of HCB plotted against the concentration of PCB in US coastal sediments (extracted from data of the US National Status and Trends program [153], available at http://ccmaserver.nos.noaa.gov/. Insert: concentrations of HCB and PCB in sediments off Scotland [11] in 1990 (filled squares) and in 1991 (filled triangles)

98

V. Zitko

of <0.5 ng/g in municipal sewage sludge. The average concentration was 21 ng/g when the non-detectable concentrations were set to zero, and 39 ng/g, when the non-detectable concentrations were set to the detection limit [58]. For industrial sewage sludge, the maximum, median, and average concentration was 303, 11, and 51 ng/g, respectively. As would be expected, HCB is adsorbed strongly by organic matter and the bioconcentration factor from a compost, containing a large concentration of organic matter (60% weight loss on ignition), to earthworms is only about 1.5 [159]. The desorption of HCB from sediments is slow, with a half-life of approximately 330 days [79]. The MPC for soil are 1300 and 300 ng/g for HCB and QCB, respectively. For sediments the MPC values are 1300 and 3000 ng/g. The EQS values for both substrates are 5 and 100 ng/g, respectively [36]. 5.4 Aquatic Biota 5.4.1 Molluscs

As can be seen from Fig. 3, there is no correlation between the concentration of HCB and PCB in molluscs. The concentrations of HCB span the same range as those of HCB in the sediment, thus indicating little accumulation from sediments to molluscs. As Figs. 4 and 5, derived from the NST program, show, particularly

Fig. 3. Concentration of HCB plotted against the concentration of PCB in molluscs from US coastal sediments See caption of Fig. 2 for source

Hexachlorobenzene

99

Fig. 4. Concentration of HCB in molluscs from US coastal sediments, extracted from data of

the US National Status and Trends program [153] (see caption of Fig. 2 for source), plotted against station numbers (see Fig. 5 for approximate station locations)

Fig. 5. Location of molluscs sampling stations (see Fig. 4)

100

V. Zitko

elevated concentrations of HCB are found in molluscs off the northeastern United States (stations #1–40), northwestern United States (stations #220–258), and in the Great Lakes (stations >#260). 5.4.2 Zooplankton, Amphipods, and Fish

Examples of concentrations, reported between 1992 and 1995 for specimens from the Canadian Arctic, for fish collected between 1989 and 1998 in Asian waters and in the mid-1990s in the North Sea and Chile are shown in Fig. 6. Lines show ranges of values reported for zooplankton and amphipods from the Canadian Arctic. The range of concentrations found in fish from the Arctic is at the high end of the zooplankton line and is delimited by asterisks. Whole fish from Asian waters were analyzed, as opposed to livers of fish from the North Sea. Since the concentrations are expressed on a lipid basis, a comparison is possible. It appears that fish from the North Sea contain higher concentrations of both HCB and PCB than fish from the Pacific Ocean, off southeastern Asia, Japan, and the Solomon Islands. On the other hand, the concentrations of HCB and PCB in fish off Australia are similar to those found in fish from the North Sea.

Fig. 6. Concentration of HCB and PCB in aquatic biota from the Canadian Arctic [112], North Sea [34, 144], Chile [55], Korea [84], south-east Asia, Polynesia, Australia, and Japan [82, 107]. Fish livers from the North Sea and whole fish from Asia were analyzed. Insert: HCB concentration in male (asterisks) and female (open squares) perch [120]. Dotted and dashed lines span the range of concentrations reported for Arctic zooplankton and amphipods, respectively. Points for Arctic fish (asterisks) span the range of reported concentrations

Hexachlorobenzene

101

In a well-defined population of perch, the concentration of HCB changes only slightly with the size and, consequently, the trophic status of the fish (Fig. 6, insert). In a more complex food chain of an arctic Polynya, the slopes of the regression lines of the ln(concentration) vs trophic level were 1.41 and 1.25 for HCB and QCB, respectively [53]. For a comparison, the slopes of lindane, p,p′-DDE, and the chlorobiphenyl CB #153 were –0.08, 2.62, and 2.27, respectively. The biomagnification factors from Arctic cod to ring seal were 0.2, 0.5, 0.3, 7.0, and 17.9, for HCB, QCB, lindane, p,p′-DDE, and CB #153, respectively. An overview of the time trends of HCB concentrations is presented in Fig. 7. According to some of the earliest reports, American eel, chain pickerel, Atlantic salmon, and herring of average weight 16, 59, and 222 g contained HCB at 12, 3, 2, and 3, 4, and 6 ng/g wet weight, respectively [170]. A USA survey of freshwater fish, not shown in Fig. 7, reported a geometric mean of 10 ng/g wet weight in 1976–1977, and less than 10 ng/g in 1978–1986 [135]. The HCB concentrations appear constant through the 1970s and then slowly decrease (Fig. 7). In fish landed in 1995 in waters off southern and eastern Asia, HCB levels were below the detection limit of 0.5 ng/g wet weight. Spanish mackerel, with HCB at 1 ng/g, and a sea eel with HCB at 2 ng/g were exceptions. HCB above the detection limit was not found in fish landed in the same year in the Norwegian and North Sea and in the Gulf of Alaska [93]. In the early 1990s in the USA, HCB was detected in fish in about one half of sampled sites (Fig. 7, insert). In 95% of the sites, the concentration of HCB was below 10 ng/g dry weight.

Fig. 7. Concentration of HCB in fish in the years 1971–1995. Prepared from the data from [18,

21, 48, 49, 170]. Insert: Distribution of HCB and QCB in freshwater and marine fish in the USA in 1990 [47]

102

V. Zitko

The concentration of pentachlorobenzene (QCB) in fish is seldom reported. When it is, it is always much lower than that of HCB. In the USA survey (Fig. 7, insert), QCB was detected at only 20% of the sites, in approximately 10 times lower concentrations. Recent data for fish from the Gulf of Gdansk shows no correlation between the concentrations of QCB and HCB [49]. The main input of HCB over the open oceans appears to be the atmosphere. This is indicated by the concentration of HCB in mesopelagic myctophid fishes that have different vertical migration pattern in the water column. The lowest concentrations of HCB are in non-migrant fish that live permanently at depth of 500 – 700 m, and highest in fish that migrate during the night to shallow depth. This concentration pattern is similar to that of the hexachlorocyclohexanes, and opposite to the patterns observed for PCBs, DDT, and chlordanes [148]. 5.4.3 Marine Mammals

Organochlorine compounds are present in high concentrations in marine mammals and Fig. 8 gives a general outline of the concentrations of HCB and PCB in the blubber of different species. It appears that porpoises and whales (see also the right insert for white whales (belugas)), contain, relatively to PCB, higher con-

Fig. 8. Concentration of HCB in the blubber of marine mammals [60, 63, 75, 77, 81, 95, 98, 100, 106, 111, 131, 137, 145, 146, 149, 156]. The solid and dotted lines indicate means±1 STD for common seals off northeastern United States in 1992 and 1980, respectively [94]. Left insert: distribution (%) of HCB, DDT, and PCB in tissues and organs of dolphins [145], right insert: HCB vs PCB in belugas (white whales) [8]

Hexachlorobenzene

103

centrations of HCB than do seals and walruses. Data on time trends are limited; the solid and the dotted line in Fig. 8 show the mean and one standard deviation of concentrations in the blubber of common seals off the northeastern coast of the United States in 1992 and in 1980, respectively. It seems that HCB is relatively enriched in the lung and brain of dolphins (Fig. 8, left insert). The concentrations of HCB in the blubber of grey seals (Halichoerus grypus) from Sable Island, Nova Scotia, Canada, remained practically constant between 1984 and 1994 at about 19 ng/g lipid [5]. 5.5 Birds

HCB in birds has received relatively little attention. HCB was overshadowed by other organochlorine compounds, particularly p,p′-DDE, associated with eggshell thinning, and dieldrin, associated with a high acute toxicity. A few examples of the concentrations of HCB, plotted against the concentrations of PCB, both on a lipid basis, are presented in Fig. 9. As expected, aquatic birds from in-

Fig. 9. Concentration of HCB and PCB in aquatic and terrestrial birds from Iceland (livers of glaucous and Icelandic gulls [34], breast muscle of ptarmigans, mallards, tufted ducks, golden plovers, purple sandpipers, black guillemots, common eiders, and gyrfalcons [118]), Gulf of Mexico (eggs of neotropic cormorants, black-crowned night herons, and great egrets [57]), British Columbia, Canada, orchards (eggs of swallows, wrens, mountain bluebirds, and robins [46]), the Danube delta (eggs of cormorants, egrets, herons, ibis, mallards, greylag geese, mute swans, and coots [15]), Italy (eggs of kestrels, sparrowhawks, herons, and mallards [126]), Poland (whole sparrows [123]), and the Falklands (brains, livers, and muscles of gentoo penguins [40])

104

V. Zitko

dustrialized areas such as the Texas coast contained high concentrations of HCB and PCB. Surprisingly, in the Danube delta, a high concentration of HCB was found in the eggs of the little egret (Egretta garzeta) and low concentrations of HCB were reported in the eggs of the glossy ibis (Plegadis falcinellus) and coot (Fulica atra). In Iceland, glaucus gulls (Larus hypereboreus) contained higher concentrations of HCB than Icelandic gulls (Larus glaucoides), and, in both species, the concentrations were higher on the south and east coasts, than on the west coast [34]. The concentrations of HCB in Gentoo penguins (Pygoscelis papua) were quite low and the fish to penguin bioaccumulation factor was about 18. The concentrations of HCB in kestrel (Falco tinnunculus), sparrowhawk (Accipiter nisus), heron (Ardea cinerea), and mallard (Anas platyrhyncos) eggs from Calabria (Southern Italy) [126] are in the mid-range of values in Fig. 9 and illustrate the food chain differences between the sparrowhawk (worm/insect-sparrowhawk) and kestrel (bird-kestrel). The HCB concentrations in heron and mallard eggs from the same area are relatively low. In Poland, the 1994 concentrations of HCB in sparrows (Passer domesticus and P. montanus) are about 100 times lower than those found in 1988 [123]. A few examples of the trends of HCB concentrations in eggs of aquatic birds are presented in Fig. 10. The data include eggs of double-crested cormorants (Phalacrocorax auritus) from the Bay of Fundy [171], of yellow-headed herring

Fig. 10. Trends of HCB concentrations in eggs of herring gulls [18, 67], guillemots [21], yellowheaded herring gulls [54], and double-crested cormorants [171]. The lines were fitted to many points in the original publications

Hexachlorobenzene

105

gulls ( Larus cachinnans) from the Tuscan archipelago [54], of herring gulls (Larus argentatus) from Lake Ontario, Lake Superior [18], and from a colony on the Scotch Bonnet Island, eastern Lake Ontario [67], and of guillemot (Uria aalge) from the Baltic [21]. The lines without indicated points were fitted to a large number of points in the original publications. The trend lines correspond to half-lives of 2–5 years. The concentration of HCB in eggs of little terns (Sterna albifrons) in the western Baltic Sea declined from 1978 to 1996 with a half-life of about 7 years, almost twice as fast as the concentration of PCB, DDT, and the hexachlorocyclohexanes [152]. In this monitoring program, individual female birds could be identified. Interestingly, in some cases, the concentrations of organochlorine compounds in individual birds did not follow the overall trend. 5.6 Terrestrial Mammals

Data on the concentration of HCB in terrestrial mammals are limited. It seems that the concentrations are low enough not to attract the attention given the marine mammals. HCB is the predominant organochlorine compound in the caribou in the Arctic, with concentration from about 30 to over100 ng/g lipid, increasing from west to east [24]. According to Bacon et al. [16], sea otters from the Aleutian Islands, southeast Alaska, and California contain HCB at about 1–2 ng/g liver wet weight, or 33–66 ng/g lipid. In Sweden, the concentration of HCB in the adipose tissue of bovines and swine was in 1991, 4 and 1.5 ng/g lipid, respectively [61, 62]. Between 1991 and 1997, both concentrations decreased at a rate of approximately 0.15 ng/(g year). 5.7 Humans

The concentration of HCB in human adipose tissue is low in the USA, Canada, and Mexico, and high in the Czech and Slovak republics. The latter may be a consequence of an unusually high concentration of HCB in the environment of the Slovak Republic (see Air above). In other countries, the concentrations are intermediate or decreasing (Fig. 11). The ranges of concentrations, shown for Canada and Turkey, are too wide to make conclusions about trends. Considering all countries, there is no discernible trend in the HCB concentration in human adipose tissue. Attempts to determine the trend are complicated further by the tendency of HCB to accumulate with age (Fig. 12). The concentrations of HCB in serum (insert, Fig. 12) may be increasing with age but, again, the ranges are too wide to detect a trend. On the other hand, HCB concentrations in breast milk lipids are declining with a half-life of about 13 years in many western countries (Fig. 13, dotted line), the trend established by data from Sweden. The levels in the Americas appear more or less constant and below the trend line. HCB concentrations in Germany, Spain, Russia, Poland, and, in particular, Indonesia, are above the line and, except for data from Germany, do not follow a trend. As can be seen from Fig. 14, HCB concentrations in serum or blood in various countries are within approximately two orders of magnitude and do not show any trend.

106

V. Zitko

Fig. 11. Trends in HCB concentrations in human adipose tissue. Data are from Mexico [160], The Netherlands [31], Canada [42, 103], Turkey [27], New Zealand [143], Japan [31], Czech and Slovak Republics [136], Greenland [42], Sweden [42], USA [42, 97, 133], Spain [42], Germany [4], and from various other countries, quoted by Kutz et al. [91]

Fig. 12. Trend of HCB concentration in human adipose tissue with age. Data are from Jordan [6], Japan [105, 108], Canada [103], and USA [97]. Insert: HCB concentration in serum as a function of age [122]

Hexachlorobenzene

107

Fig. 13. Trends of HCB concentrations in breast milk lipids. Dotted line fitted to data from

Sweden [116], with one point added from [14]. Data from other countries are from [3, 25, 35, 37–39, 41, 66, 73, 74, 124, 134, 138, 142, 167]

6 Toxicity The acute toxicity of HCB is low and ranges from 1 g to over 10 g/kg body weight for most species. At first, HCB was considered to have limited or no toxicity: – “Another important advantage of HCB preparations is that they are nonpoisonous. This is indeed a major consideration, not only in the seed treating process, but also in the possible disposition of seed treated in excess of that needed for seeding purposes ... . The question of toxicity to animal life has been raised ... . This material is said to be nontoxic, but so far as we know, little is known yet of its effects on livestock in actual feeding tests” [52]. It is an irony that soon after the publication of the quoted book, HCB-treated seed caused a large scale poisoning of population in Turkey. HCB became probably the first chemical associated with a named ‘disease’, the Pembe Yara (pink sore), caused by the consumption of HCB-treated wheat in Turkey between 1956 and 1959. This disease affected an estimated 3000–5000 people, had an overall mortality rate of 3–11%, with a very a high mortality among children up to the age of 4. Numerous symptoms included skin lesions, particularly in areas exposed to sunlight, porphyria cutanea tarda, hyperpigmentation of the skin, and hepatic pathology. The dose of HCB, causing this disease, is not known. The

108

V. Zitko

Fig. 14. Concentration of HCB in human blood or serum. Louisiana line: HCB contamination from chlorinated solvent production waste in Louisiana [29]. Points from top, process technician for 5 years, instrument technician for 4 years, laboratory technician, 3 years, exposed, and control population. Brazil line: contamination from pentachlorophenol production in Brazil [39], points correspond to an exposed and a control group of workers, respectively. Flix line: contamination in the vicinity of an organochlorine compounds production plant in Spain [70]. Extreme points: factory workers and non-factory workers, respectively, 2nd highest point males, 2nd lowest point females of the whole examined populations which had an overall mean of 39.8 ng/ml. Data for Spain, USA, and Germany are cited in [70]. Data for Slovakia, Holland, Italy, Germany, and India are from [85], data from Sweden, Croatia, Mexico, Greenland, and the Faroe Islands are from [23, 30, 61, 89, 161], respectively.Additional data for USA are from [88] and for Canada, Spain and USA from [73]. The ‘Syndrome’ line spans the range of HCB concentrations in patients with diagnosed chronic fatigue syndrome, with known and not known exposure to toxic chemicals (the upper two points, respectively) and in a control group (the lowest point) [45]

estimates range from 50 to 200 mg/day, taken over a period of months. The symptoms developed slowly and persisted for a long time after the cessation of the HCB consumption in 1959. Mercury compounds, also present in some HCB fungicide formulations, have a synergistic effect, as shown for mercuric chloride in Fig. 15 [128]. It is not known whether mercury compounds played a role in Pembe Yara. Another contamination incident was discovered in Louisiana in 1972 and some of the earliest measurements of HCB concentration in human blood were obtained during this investigation. The source of HCB was the transportation and disposal of waste from the production of carbon tetrachloride, tetrachloroethylene, and other chlorinated solvents [29]. The contamination was first detected in cattle in December 1972. Subsequently, factory workers and the local general pop-

Hexachlorobenzene

109

Fig. 15. Synergistic effect of mercuric chloride on the toxicity of HCB. Plotted from data

in [128]

ulation were examined. No signs of cutaneous porphyria, so characteristic of the previous incident in Turkey, were found. Factory workers had elevated concentrations of HCB in blood, but levels in the general population of the area were within the range found later in other populations (Fig. 14).A similar incident occurred in Brazil in connection with the production of chlorinated solvents, as well as pentachlorophenol [39]. The pentachlorophenol plant near Cubatão City was closed in 1978, but the carbon tetrachloride and tetrachloroethylene plant was remained on line until 1993. It seems that detailed investigations were carried out only in the 1990s.A third occurrence, also associated with the production of chlorinated solvents, occurred in Flix, Spain [70]. Also in this incident no cutaneous porphyria was found, even at an HCB concentration of 1616 ng/ml serum, and no correlation between HCB concentration in serum and total porphyrins in urine was found. Herrero and co-authors did not exclude genetic factors, but suggested that exposure to airborne HCB may not have been sufficient to alter the heme biosynthesis. To-Figueras et al. [150] studied the general population in Flix and also concluded that an airborne route could not provide hepatotoxic concentrations of HCB.A close correlation between the concentration of HCB in serum and in feces confirmed the inhalation route of exposure, resulting in the storage of HCB in the adipose tissue and its slow excretion in feces as HCB, and in urine in the form of the metabolites pentachlorothiophenol and pentachlorophenol. The extent of medical investigation of the Cubatão City incident is not known. It is unfortunate that the levels of HCB exposure and associated concentrations in tissues and organs during the mass poisoning in Turkey are not known. The exposure occurred probably exclusively from food, although the fate of HCB during cooking or baking with contaminated flour is uncertain.

110

V. Zitko

Toxic effects of organochlorine compounds, including HCB, were demonstrated or suggested in a number of other investigations. The concentration of HCB in serum women who survived Pembe Yara, as well as those in the serum of a control population are significantly correlated with the rate of abortion [76]. This indicates the possible effect of HCB on steroidogenesis or immunochemistry, also demonstrated in animal studies. For example, massive doses of HCB (1 g/kg body weight for 30 days) reduced the circulating concentrations of oestradiol and prolactin in Wistar rats [7]. In addition, HCB has immunotoxic properties, indicated for example in polar bears [20] and in occupationally exposed workers [104], by elevated serum immunoglobulin levels. Studies in animals established that HCB is a carcinogen in animals; however, the mechanism of the tumor induction is not known [73] and the evidence of HCB carcinogenicity in humans is inconclusive. For example, there was no difference in HCB concentration in the serum of control patients and patients with endometrial cancer (0.37 and 0.39 ng/ml, respectively) [163]. Consequently, HCB is classified as only a probable human carcinogen [151]. Dunstan et al. [45] suggested an association of HCB, heptachlor epoxide, DDE, and dieldrin with chronic fatigue syndrome (CFS), as both HCB (Fig. 14) and DDE were elevated in subjects with diagnosed CFS. 6.1 Aquatic Biota

The low solubility in water limits the availability of HCB to aquatic biota and HCB is not acutely toxic. However, aquatic biota takes up and accumulates HCB from water, with the bioconcentration factors (BCF) of the order of thousands [26, 28, 110, 119, 121]. There seem to be no data on chronic toxicity of HCB to aquatic fauna. Recently, Zhan et al. [169] found an increase in estrogen concentration in the serum of female crucian carp (Carassius auratus gibello) exposed to HCB at 50 mg/l for two weeks. The concentration of HCB in the liver, associated with the estrogen increase, was approximately 600 mg/g. 6.2 Birds and Mammals

An overview of the toxicity of HCB is presented in Fig. 16.As mentioned already, the acute toxicity is very low. Chronic toxicity occurs at a range of concentrations and exposure regimes and details are beyond the scope of this chapter. A recent review of the toxicity of HCB to birds [164] quotes only eight publications, the most recent published in 1984. In laboratory experiments, concentrations of HCB in tissues or organs of animals with observable effects are usually in the tens to hundreds of mg/g lipid. It is impossible to judge environmental toxicity of HCB, since HCB is practically never present alone, but it is only one of a number of organochlorine compounds, many of them as, or more, toxic than HCB, and most present in much higher concentrations.

Hexachlorobenzene

111

Fig. 16. An overview of the toxicity of HCB. Data are from [73]. In the inserts are data for Japan-

ese quail [158]. S=Swiss mice, SD-rats=Sprague-Dawley rats. TAO inhibits cytochrome P-450IIIA1/2 and, consequently, the oxidative transformation of HCB. The ‘Rats, porphyria’ box indicates the area of doses and exposure time resulting in the induction of porphyria in rats. SER=smooth endoplasmic reticulum. Left insert: the excretion of coproporphyrin in the Japanese quail, maintained on diets containing HCB at 0–80 mg/g. Right insert: Japanese quail liver weight at 90 days of feeding

6.3 Pharmacodynamics

HCB is degraded reductively by bacteria under methanogenic conditions to QCB, which is further degraded to tetra-, tri-, and di-chlorobenzenes, and to chlorobenzene. Of the two tetrachlorobenzenes with isolated free carbons, 1,2,4,5-tetrachlorobenzene is, but the 1,2,3,5- isomer is not, reductively dechlorinated under these conditions. Similarly, 1,3,5-trichlorobenzene is not dechlorinated [127]. Under the specified conditions, the half-lives of HCB and QCB were 31 and 24 days, respectively. In mammals, HCB is metabolized and excreted extremely slowly. For example, in rats, after a single oral dose of 1.3–2 mg/kg, administered in corn oil, only about 0.08% of the dose was excreted in urine, and about 10% in feces, both as the original HCB [168]. The concentrations in the blood and organs of the animals reached a maximum in about 1–2 days, and then slowly declined. The halflives were 12.4, 39.7, 15.0, 24.5, and 19.7 days in the adipose tissue, blood, brain, kidney, and liver, respectively. For whole animals, the half-life was about 15 days.

112

V. Zitko

The major HCB metabolites are QCB and pentachlorophenol [128]; however, in the population of Flix, the major metabolite is pentachlorothiophenol (see below) [150]. The first step in HCB metabolism is the formation of glutathion conjugates, which are further converted to cysteine conjugates and mercapturic acids. These are transformed further into pentachlorothiophenol, which, in turn may be converted into pentachlorophenol or QCB. Further dechlorination may be taking place during the process.Altogether, 35 metabolites of HCB were identified, mostly in the rat. In a follow-up study of the general population of Flix, Spain, the average amounts of pentachlorothiophenol and pentachlorophenol excreted in urine were 8.8 (range 0.5–86.9) and 3.8 (range 0.6–18) mg/24 h, respectively). These amounts are very small compared to a mean total adipose burden of 300 (range 5.9–3653) mg and would result in a whole-body half-life of about 6 years [150].

Fig. 17. The effect of chlorobenzenes on drug-metabolizing enzymes and hepatic constituents in the rat. Data from [12], processed by principal component analysis [173] and projected on the plane of the principal components 1 and 2. Fractions of the original variance are indicated on the axes. C1 and C2 are controls, numbers indicate the numbers of chlorine atoms in the benzene molecule. The projections of similar enzymes and constituents profiles are close to each other (see for example C1 and C2, or di- and trichlorobenzene, 2 and 3)

Hexachlorobenzene

113

6.4 Mechanism of Toxicity

The molecular mechanism of HCB toxicity is not known in detail. As other organochlorine compounds, HCB affects drug-metabolizing enzymes and the composition of the liver. The changes caused by HCB are different from those induced by QCB and other, less chlorinated benzenes [12]. This is demonstrated in Figs. 17 and 18, which summarise, by principal component projections (PCA, see for example [173]), the differences in 24 variables, after the administration of chloro-, p-dichloro-, 1,3,5-trichloro-, 1,2,4,5-tetrachlorobenzene, QCB, and HCB, respectively, to female Wistar rats. The compounds were administered in three oral daily doses of 250 mg/kg, as suspensions in 2% tragacanth gum solutions. The variables measured include body weight, liver weight and composition, protein and phospholipid fatty acids in liver microsomes, and the activities of cytochromes P-450 and b3, aniline hydroxylase (AH), aminopyrine demethylase (AD), d-aminolevulinic acid synthase (ALA), and the spectral change aniline-cytochrome P-450. As can be seen from Figs. 17 and 18, the changes caused by the di- and the trichlorobenzene are similar. Changes caused by the other chlorobenzenes are all different. In particular, QCB, under the used experimental conditions, is a stronger inducer of AH, AD, and ALA.

Fig. 18. The effect of chlorobenzenes on drug-metabolizing enzymes and hepatic constituents in the rat. Data from [12], processed by principal component analysis [173] and projected on the plane of the principal components 1 and 3. See Fig. 18 for additional details

114

V. Zitko

Fig. 19. The induction of hepatic benzo[a]pyrene (BaP) and testosterone hydroxylases by phenobarbital, HCB, and organochlorine pesticides. The data are from [65], presented by principal component projections [173] on the plane of principal components 1 and 2. The amount of original variance in indicates on the axes. Projections of similar product profiles are located close to each other (for example PB and d). CO, L, dde, d, hep, chlord, tox, and hcb stand for control, lindane, DDE, dieldrin, heptachlor, chlordane, toxaphene, and HCB, respectively

HCB differs in terms of hepatic benzo[a]pyrene (BaP) and testosterone hydroxylases induction from phenobarbital and organochlorine pesticides [65]. This can be seen from the PCA projections of the hydroxylation products profiles (Figs. 19 and 20). HCB-induced BaP hydroxylases are considerably more effective than hydroxylases induced by the other studied compounds, in producing most of the BaP hydroxy compounds and the 3,6-quinone, than the other organochlorine pesticides. The mechanism of the induction of porphyria cutanea tarda is not known. HCB inhibits the uroporhyrinogen decarboxylase (URO-D), an enzyme in the porphyrin metabolic pathway [44]. In rats treated by massive doses of HCB (1 g/kg body weight, 5 days a week) for 1–4 weeks, the URO-D activity was significantly lower at 20 weeks after the treatment than at the end of each treatment [22]. In addition, the formation and a lasting presence of an URO-D inhibitor was demonstrated. It appears that HCB must be metabolised to exhibit its porphyrinogenic activity. The metabolite, which binds covalently to protein is tetrachlorohydroquinone [154, 155]. How these processes take place in the structured environment of the cells is not known. It has been

Hexachlorobenzene

115

Fig. 20. The induction of hepatic benzo[a]pyrene (BaP) and testosterone hydroxylases by phenobarbital, HCB, and organochlorine pesticides. The data are from [65], presented by principal component projections [173] on the plane of principal components 1 and 3. For additional details see Fig. 19

shown recently that HCB changes the morphology of human erythrocyte membranes [147]. 6.5 Risk Assessment

Many toxicological studies were initiated after the discovery that HCB is the cause of Pembe Yara. In 1969, the first FAO/WHO meeting that evaluated HCB based its decision on acute oral toxicity and animal feeding studies. The meeting derived a ‘tentative negligible daily intake’ (conditional Acceptable Daily Intake, ADI) of 0.0006 mg/kg body weight, and established ‘temporary tolerance’ or ‘practical residue limits’ (Table 2). During the second evaluation in 1973 it was realized that HCB is reaching the environment also from industrial wastes and from pesticides such as pentachloronitrobenzene (quintozene) and tetrachloronitrobenzene (technazine), and dimethyl tetrachloroterephthalate (chlorothal-dimethyl), in addition to its applications as a fungicide. At the same time, a number of reports documented the wide-ranging contamination of the environment by HCB. Several countries established legal limits or action levels, particularly for meat at 0.5 mg/g lipid, and for milk at 0.3–0.5 mg/g lipid. The meeting decided that HCB

116

V. Zitko

Table 2. WHO/FAO tolerances and practical residue limits (ppm)

1969 Raw cereal (wheat) Raw grain Cereal products from wheat Flour and milled cereal products Fat of cattle, sheep, goats, pigs, and poultry Milk Milk products Milk and milk products (fat basis) Eggs (shell free)

1973

1974

0.05

0.05

0.01 1

0.01 1

0.5 1

0.5 1

0.05 0.01 1 0.012 0.3 1

Fig. 21. HCB intake, mg/day by a 60-kg person, plotted against years. Data for Japan, Vietnam, Thailand, USA, Germany, Finland, and Australia are from [92], for India from [80], data for Spain from [69], for ‘USA 1995’ from [43], with HCB concentrations set at 0 for non-detectables at the bottom of the range and to 0.5 of the detection limit the top of the range. The dotted line at 1982 spans the estimated actual range for UK, USA, and Japan [166], the solid line at 1998 (early 1990s estimated range, offset for clarity to 1998 in the graph) is the IOMC [73] range estimate. The dash-dot line shows the range from the cancer benchmark concentration (bottom, cancer risk 1 in 1E6) to a high probability of no adverse health effects (top [43])

Hexachlorobenzene

117

should be kept under annual review. By 1974, additional data on the concentration of HCB in the environment became available, The Netherlands restricted the concentration of HCB and QCB in quintozene to no more than 0.1% and 1% respectively, but the residue limits have not been changed. In response to the discovery of liver tumors in mice on a diet containing HCB at 100–200 mg/g, the conditional ADI was withdrawn at the 1978 FAO/WHO meeting. In 1985, at the last meeting dealing with HCB, FAO/WHO noted that HCB concentrations in food commodities appeared to be decreasing, but concluded that data are not sufficient to change the existing recommendations [165, 166]. Estimated daily intakes of HCB in various countries are summarized in Fig. 21. In a recent surveys of food baskets in six cities in Canada [115], HCB at an average concentration of 0.5 ng/g was present in 4.8% of samples. It appears that, except for some pollution accidents, the risks of HCB to humans and wildlife are negligible.

7 Concluding Remarks HCB is an environmentally undesirable compound that has no essential uses. All intentional uses have been canceled. The remaining high priority task is the prevention of unintentional formation and discharges of HCB into the environment. The limiting factors for this task are both technical and economic and are being considered in a number of studies (see for example [64, 117]). Toxicity studies of HCB alone, often using massive doses, are interesting to elucidate mechanisms of HCB actions. However, in the author’s opinion, such studies consume resources that would be more appropriately spent on hazard identification and risk assessment of the great variety of currently used pesticides and other chemicals that are released to the environment and have not yet received much publicity. Acknowledgements. I thank Ms Joanne Cleghorn for extremely efficient help in obtaining publications, Dr. R.E. Bailey for preprints of his papers and reports, Dr. J.L. Herrman and Amelia W. Tejada for WHO/FAO materials, and Drs. S.C. Billi de Catabbi,A. Renzoni, Ott Roots, E.Weiderpass, A. Wicklund-Glynn for additional information.

8 References 1. 2. 3. 4. 5. 6. 7.

Abbassy MS (2000) Bull Environ Contam Toxicol 64:508 Abbassy MS, Ibrahim HZ, Abu El-Amayem MM (1999) J Environ Sci Health B34:255 Acker A, Schulte E (1970) Dtsch Lebensm-Rundsch 66:385 Acker A, Schulte E (1974) Naturwissenschaften 57:497 Addison RF, Stobo WT (2001) Environ Pollut 112:505 Alawi MA, Tamimi S, Jaghabir M (1999) Chemosphere 38:2865 Alvarez L, Randi A, Alvarez P, Piroli G, Chamson-Reig A, Lux-Lantos V, Kleiman de Pisarev D (2000) J Appl Toxicol 20:81 8. Andersen G, Kovacs KM, Lydersen C, Skaare JU, Gjertz I, Jenssen BM (2001) Sci Total Environ 264:267 9. Anon (1972) Pesticides listed by active ingredient registered for use in Canada for1972. Control Products Section, Plant Products Division, Canada Department of Agriculture, Ottawa, Ontario

118

V. Zitko

10. Anon (1995) Stoffbericht Hexachlorobenzol (HCB), www.uvm.baden-wuerttemberg.de/ alfaweb/bericht/tba 18–95/hcb.html 11. Anon (1996) Environmental monitoring of the seas around Scotland 1970–1993. SOAEFD Marine Laboratory, PO Box 101, Victoria Rd, Aberdeen, AB11 9DB, ISSN:1361–763X 12. Ariyoshi T, Ideguchi K, Ishizuka Y, Iwasaki K,Arakaki M (1975) Chem Pharm Bull 23:817 13. Atlas E, Giam CS (1981) Science 211:163 14. Aune M, Atuma S, Darnerud PO, Wicklund-Glynn A, Cnattigius S (1999) Organohal Compd 40:87 15. Aurigi S, Focardi S, Hulea D, Renzoni A (2000) Environ Pollut 109:61 16. Bacon CE, Jarman WM, Estes JA, Simon M, Norstrom RJ (1999) Environ Toxicol Chem 18:452 17. Bahadur NP, Shiu WY, Boocock DGB, Mackay D (1997) J Chem Eng Data 42:685 18. Bailey RE (2000) Hexachlorobenzene concentration trends in the Great Lakes. Report to the HCB/BaP Work Group of the Binational Toxics Strategy, 5 May 2000. 21p. Bailey Associates, 4115 Elm Court, Midland, MI 48642, Phone/fax 517–835–3410, e-mail [email protected] 19. Bailey RE (2001) Chemosphere 43:177 20. Bernhoft A, Skaare JU, Wiig Ø, Derocher AE, Larsen HJS (2000) J Toxicol Environ Health A 59:561 21. Bignert A, Olsson M, Persson W, Jensen S, Zakrisson S, Litzén K, Eriksson U, Häggberg L, Alsberg T (1998) Environ Pollut 99:177 22. Billi de Catabbi SC, Aldonatti C, San Martin de Viale LC (2000) Comp Biochem Physiol C 127:165 23. Bjerregaard P, Hansen JC (2000) Sci Total Environ 245:195 24. Braune B, Muir D, DeMarch B, Gamberg M, Poole K, Currie R, Dodd M, Duschenko W, Eamer J, Elkin B, Evans M, Grundy S, Hebert C, Johnstone R, Kidd K, Koenig B, Lockhart L, Marshall H, Reimer K, Sanderson J, Shutt L (1999) Sci Total Environ 230:145 25. Brevik EM, Bjerk JE (1978) Acta Pharmacol Toxicol (1978) 43:59 26. Bro-Rasmussen F (1986) IARC Sci Publ 77:231 27. Burgaz S, Afkham BL, Karakaya AE (1994) Bull Environ Contam Toxicol 53:501 28. Burkhard LP, Sheedy BR, McCauley DJ, DeGraeve GM (1997) Environ Toxicol Chem 16:1677 29. Burns JE, Miller FM (1975) Arch Environ Health 30:44 30. Burse VW, Najam AR,Williams CC, Korver MP, Smith BF Jr, Sam PM,Young SL, Needham LL (2000) J Exposure Anal Environ Epidemiol 10:776 31. Burton MAS, Bennett BG (1987) Sci Total Environ 66:137 32. Calamari D, Tremolada P, Notarianni V (1995) Environ Sci Technol 29:2267 33. Caseley JC (1968) Bull Environ Contam Toxicol 3:180 34. Cleeman M, Riget F, Paulsen GB, Dietz R (2000) Sci Total Environ 245:117 35. Conde C, Maluenda C, Arrabal C (1993) Bull Environ Contam Toxicol 51:827 36. Crommentuijn T, Sijm D, de Bruijn J, van Leeuwen K, van de Plassche E (2000) J Environ Manag 58:297 37. Currie RA, Kadis VW, Breitkreitz WE, Cunningham GB, Bruns GW (1979) Pestic Monit J 13:52 38. Czaja K, Ludwicki JK, Góralczyk K, Strucinski P (1997) Bull Environ Contam Toxicol 58:769 39. Da Silva AS (2000) Environmental contamination and occupational and urban exposure to hexachlorobenzene at Baixada Santista, SP, Brazil http://www.chem.unep.ch/pops/ POPs_Inc/proceedings/Iguazu/Agnes.html 40. De Boer J, Wester P (1991) Mar Pollut Bull 22:441 41. Dewailly É,Ayotte P, Bruneau S, Laliberté C, Muir DCG, Norstrom R (1993) Environ Health Perspect 101:618 42. Dewailly É, Mulvad G, Pedersen HS, Ayotte P, Demers A, Weber JP, Hansen JC (1999) Environ Health Perspect 107:823 43. Dougherty CP, Holtz SH, Reinert JC, Panyacosit L, Axelrad DA, Woodruff TJ (2000) Environ Res Sect A 84:170

Hexachlorobenzene

119

44. Downey DC (1999) Med Hypotheses 53:166 45. Dunstan RH, Donohoe M, Taylor W, Roberts TK, Murdoch RN,Watkins JA, McGregor NR (1995) Med J Australia 163:294 46. Elliott JE, Martin PA,Arnold TW, Sinclair PH (1994) Arch Environ Contam Toxicol 26:435 47. EPA (1992) National study of chemical residues in fish. Vol I. EPA 823-R-92–008a 48. Falandysz J (2000) Polish J Environ Studies 9:377 49. Falandysz J, Strandberg L, Strandberg B, Bergqvist PA, Rappe C (2000) Polish J Environ Studies 9:129 50. Farmer WJ, Yang MS, Letey J, Spencer WF (1980) Soil Sci Soc Am J 44:676 51. Fiedler H, Lau C (1998) Environmental fate of chlorinated organics. In: Schüürmann G, Markert B (eds) Ecotoxicology. Wiley, p 317 52. Fisher GW, Holton CS (1957) Biology and control of the smut fungi. Ronald Press Company, New York, pp 431, 444 53. Fisk AT, Hobson KA, Norstrom RJ (2001) Environ Sci Technol 35:732 54. Focardi S, Fossi C, Lambertini M, Leonzio C, Massi A (1988) Environ Monit Assess 10:43 55. Focardi S, Fossi C, Leonzio C, Corsolini S, Parra O (1996) Environ Monit Assess 43:73 56. Fox WM, Connor L, Copplestone D, Johnson MS, Leah RT (2001) Mar Environ Res 51:213 57. Frank DS, Mora MA, Sericano JL, Blankenship AL, Kannan K, Giesy JP (2001) Environ Toxicol Chem 20:608 58. Frost P, Camenzind R, Mägert A, Bonjour R, Karlaganis G (1993) J Chromatog 643:379 59. Gaumert T (1993) Schadstoffüberwachung der Elbe bei Schnackenburg mit der Dreikantmuschel. Wassergütestelle Elbe, Nessdeich 120–121, 21129 Hamburg 60. Gauthier JM, Metcalfe CD, Sears R (1997) Mar Environ Res 44:201 61. Glynn AW, Wernroth L, Atuma S, Linder CE, Aune M, Nilsson I, Darnerud PO (2000) Sci Total Environ 246:195 62. Glynn AW, Wolk A, Aune M, Atuma S, Zettermark S, Mæhle-Schmid M, Darnerud PO, Becker W, Vessby B, Adami HO (2000) Sci Total Environ 263:197 63. Granby K, Kinze CC (1991) Marine Pollut Bull 22:458 64. Great Lakes Binational Toxics Strategy (2000) Hexachlorobenzene (HCB): reduction options. US EPA Great Lakes National Program Office Contract # 68-W-99–033 65. Haake J, Kelley M, Keys B, Safe S (1987) Gen Pharm 18:165 66. Harris CA, O’Hagan S, Merson GHJ (1999) Human Exp Toxicol 18:602 67. Hebert CE, Norstrom RJ, Weseloh DVC (1999) Environ Rev 7:147 68. Heinisch, E (1978) Biogeochemische Kreisläufe persistenter organischer Verbindungen: dargestellt am Hexachlorobezol. Sitzungsberichte der Akademie des Wissenschaften der DDR, 202–100/164/79, Akademie-Verlag Berlin 69. Herrera A, Ariòo A, Conchello P, Lázaro R, Bayarri S, Pérez-Arquillué C, Garrido MD, Jordal M, Pozo R (1996) Bull Environ Contam Toxicol 56:173 70. Herrero C, Ozalla D, Sala M, Otero R, Santiago-Silva M, Lecha M, To-Figueras J, Deulofeu R, Mascaró JM, Grimalt J, Sunyer J (1999) Arch Dermatol 135:400 71. Hillebrand MTHJ, Everaarts JM, Razak H, Moelyadi Moelyo D, Stolwijk L, Boon JP (1989) Netherlands J Sea Res 23:369 72. Holden AV, Marsden K (1969) J Chromatog 44:481 73. IOMC (1997) Hexachlorobenzene. Environmental Health Criteria 195. World Health Organization, Geneva 74. Ip HMH, Phillips DJH (1989) Arch Environ Contam Toxicol 18:490 75. Jarman WM, Norstrom RJ, Muir DCG, Rosenberg B, Simon M, Baird RW (1996) Marine Pollut Bull 32:426 76. Jarrell J, Gocmen A (2000) Pure Appl Chem 72:1015 77. Jenssen BM, Skaare JU, Ekker M,Vongraven D, Lorentsen SH (1996) Chemosphere 32:2115 78. Jepsen R, Borglin S, Lick W, Swackhamer DL (1995) Environ Toxicol Chem 14:1487 79. Kan AT, Chen W, Towson MB (2001) Resistant desorption kinetics of chlorinated organic compounds from contaminated soil and sediment. In: Lipnick RL, Hermens JLM, Jones KC, Muir DCG (eds), Persistent, bioaccumulative, and toxic chemicals. I. Fate

120

80. 81. 82. 83.

84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117.

V. Zitko and exposure. ACS Symposium Series 772, American Chemical Society, Washington DC, p 112 Kannan K, Tanabe S, Ramesh A, Subramanian A, Tatsukawa R (1992) J Agric Food Chem 40:518 Kannan K, Tanabe S, Tatsukawa R (1994) Toxicol Environ Chem 42:249 Kannan K, Tanabe S, Tatsukawa R (1995) Environ Sci Technol 29:2673 Kenaga EE (1983) Correlations of physical and chemical properties of compounds with their toxicities and distributions in the environment. In: Francis CW,Auerbach SI, Jacobs VA (eds) Environmental solid wastes: characterization, treatment disposal. Butterworths, Boston, p 105 Khim JS, Villeneuve DL, Kannan K, Hu WY, Giesy JP, Kang SG, Song KJ, Koh CH (2000) Arch Environ Contam Toxicol 39:360 Kočan A, Petrík J, Drobná B, Chovancová J (1994) Chemosphere 29:2315 Kočan A, Petrík J, Uhrinová H, Drobná B, Chovancová J (1999) Toxicol Environ Chem 68:481 Koelmans AA, Heugens EHW (1998) Water Sci Tech 37:67 Korrick SA, Altshul LM, Tolbert PE, Burse VW, Needham LL, Monson RM (2000) J Exposure Anal Environ Epidemiol 10:743 Krauthacker B (1993) Bull Environ Contam Toxicol 50:8 Krämer W, Ballschmiter K (1988) Fresenius Z Anal Chem 330:524 Kutz FW, Wood PH, Bottimore DP (1991) Rev Environ Contam Toxicol 120:1 Kuwabara K, Matsumoto H, Murakami Y, Hori S (1997) J Food Hyg Soc Japan 38:286 Kuwabara K, Harada A, Matsumoto H, Hori S (1999) Toxicol Environ Chem 73:93 Lake CA, Lake JL, Haebler R, McKinney R, Boothman WS, Sadove SS (1995) Arch Environ Contam Toxicol 29:128 Law RJ, Allchin CR, Harwood J (1989) Mar Pollut Bull 20:110 Lee DB, Prudente MS, Tanabe S, Tatsukawa R (1997) Toxicol Environ Chem 60:171 Lordo RA, Dinh KT, Schwemberger JG (1996) Am J Public Health 86:1253 Luckas B, Vetter W, Fisher P, Heidemann G, Plötz J (1990) Chemosphere 21:13 Marsili L, Focardi S (1997) Environ Monit Assess 45:129 Marsili L, Casini C, Marini L, Regoli A, Focardi S (1997) Mar Ecol Prog Ser 151:273 Meharg AA, Wright J, Osborn D (2000) Sci Total Environ 251/252:243 Melnikov NN (1971) Chemistry of pesticides. Springer, Berlin Heidelberg New York Mes J, Davies DJ, Turton D (1982) Bull Environ Contam Toxicol 28:97 Michielsen CCPPC, van Loveren H, Vos JG (1999) Environ Health Perspect 107:783 Minh TB,Watanabe M, Tanabe S,Yamada T, Hata J,Watanabe S (2000) Environ Health Perspect 108:599 Mitchell SH, Kennedy S (1992) Sci Total Environ 115:163 Monirith I, Nakata H,Watanabe M, Takahashi S, Tanabe S, Tana TS (2000) Water Sci Technol 42:241 Mori Y, Kikuta M, Okinaga E, Okura T (1983) Bull Environ Contam Toxicol 30:74 Morris CR, Capbal JRP (eds) (1986) Hexachlorobenzene: Proceedings of an International Symposium. International Agency for Research on Cancer, Lyon Mortimer MR, Connell DW (1993) Aust J Mar Freshwater Res 44:565 Muir DCG, Segstro MD, Hobson KA, Ford CA, Stewart REA, Olpinski S (1995) Environ Pollut 90:335 Muir D, Braune B, DeMarch B, Norstrom R, Wagemann R, Lockhart L, Hargrave B, Bright D, Addison R, Payne J, Reimer K (1999) Sci Total Environ 230:83 Müller MD (1982) Chimia 36:437 Müller-Herold U, Nickel G (2000) Ecol Model 126:191 Newsome WH, Doucet J, Davies D, Sun WF (2000) Food Addit Contam 17:847 Norén K, Meironyté D (2000) Chemosphere 40:1111 North American Commission for Environmental Cooperation (1998) Nomination Dossier for Hexachlorobenzene. www.cec.org/programs_projects/pollutants_health/ smoc/hcbcan.cmf?varlan=english

Hexachlorobenzene

121

118. Ólafsdóttir K, Petersen Æ, Magnúsdóttir EV, Björnsson T, Jóhannesson T (2001) Environ Pollut 112:245 119. Oliver BG, Niimi AJ (1983) Environ Sci Technol 17:287 120. Olsson A, Valters K, Burreau S (2000) Environ Sci Technol 34:4878 121. Parrish PR, Cook GH, Patrick JM Jr (1974) Proc 28th Ann Conf Southeast Assoc Game Fish Commissioners, p 179 122. Pellini GF, Dick T, Rodriguez MTR (1999) Revista Brasileria de Toxicologia 12:27 123. Pinowski J, Niewiadöwska A, Juøicová Z, Literák I, Romanowski J (1999) Bull Environ Contam Toxicol 63:736 124. Polder A, Becher G, Savinova TN, Skaare JU (1998) Chemosphere 37:1795 125. Popp P, Brüggemann L, Keil P, Thuss U, Weiss H (2000) Chemosphere 41:849 126. Provini A, Galassi S (1999) Ecotoxicol Environ Safety 43:91 127. Ramanand K, Balba MT, Duffy J (1993) Appl Environ Microbiol 59:3266 128. Renner G (1988) Toxicol Environ Chem 18:51 129. Research Triangle Institute (1996) Toxicological profile for hexachlorobenzene. US Department of Health and Human Services. Contract No. 205-93-0606 130. Rocklin AL (1956) J Org Chem 21:1478 131. Roots O (1994) J Ecol Chem 3:35 132. Sanusi A, Millet M, Mirable P, Wortham H (2000) Sci Total Environ 263:263 133. Schaefer WR, Hermann T, Meinhold-Heerlein I, Deppert WR, Zahradnik HP (2000) Fertil Steril 74:558 134. Schecter A, Ryan JJ, Päpke O (1998) Chemosphere 37:1807 135. Schmitt CJ, Zajicek JL, May TW, Cowman DF (1999) Rev Environ Contam Toxicol 162:43 136. Schoula R, Haj_lová J, Gregor P, Kocourek V, Bencko V Toxicol Environ Chem 67:263 137. Senthilkumar K, Kannan K, Sinha RK, Tanabe S, Giesy JP (1999) Environ Toxicol Chem 18:1511 138. Shaw I, Burke E, Suharyanto F, Sihombing G (2000) Environ Sci Pollut Res 7:75 139. Shiu WY, Wania F, Hayley H, Mackay D (1997) J Chem Eng Data 42:293 140. Simonich SL, Hites RA (1997) Environ Sci Technol 31:999 141. Sittig M (1977) Pesticide process encyclopedia. Noyes Data Corp, Park Ridge, USA, p 276 142. Skaare JU, Tuveng JM, Sande HA (1988) Arch Environ Contam Toxicol 17:55 143. Solly SRB, Shanks V (1974) NZ J Sci 17:535 144. Stange K, Maage A, Klungsoyr J (1996) Contaminants in fish and sediments in the North Atlantic Ocean. TemaNord 522. Nordic Council of Ministers, Copenhagen 145. Storelli MM, Stuffler RG, Ceci E, Marcotrigiano GO (1999) In: Marine pollution. IEAETECDOC-1094, p 409 146. Storelli MM, Marcotrigiano GO (2000) Marine Pollut Bull 40:555 147. Suwalsky M, Rodríguez C,Villena F,Aguilar F, Sotomayor CP (1999) Pestic Biochem Physiol 65:205 148. Takahashi S, Tanabe S, Kawaguchi K (2000) Environ Sci Technol 34:5129 149. Tanabe S, Tatsukawa R, Tanaka H, Maruyama K, Miyazaki N, Fujiyama T (1981) Agric Biol Chem 45:2569 150. To-Figueras J, Barrot C, Sala M, Otero R, Silva M, Ozalla MD, Herrero C, Corbella J, Grimalt J, Sunyer J (2000) Environ Health Perspect 108:595 151. TOXNET http://toxnet.nlm.nih.gov/cgi-bin/sis 152. Thyen S, Becker PH, Behmann H (2000) Environ Pollut 108:225 153. U.S. National Status and Trends program, http://ccmaserver. nos.noaa.gov 154. Van Ommen B, Adang AEP, Brader L, Posthumus MA, Müller F, van Bladeren PJ (1986) Biochem Pharmacol 35:3233 155. Van Ommen B, Hendriks W, Bessems JGM, Geesink G, Müller F, van Bladeren PJ (1989) Toxicol Appl Pharmacol 100:517 156. Vetter W, Hummert K, Luckas B, Skírnisson (1995) Sci Total Environ 170:159 157. Villanueva EC, Jennings RW, Burse VW, Kimbrough RD (1974) J Agric Food Chem 22:916 158. Vos JG, van der Maas HL, Musch A, Ram E (1971) Toxicol Appl Pharmacol 18:944

122

V. Zitko

159. Wågman N, Strandberg B, van Bavel B, Bergqvist PA, Öberg L, Rappe C (1999) Environ Toxicol Chem 18:1157 160. Waliszewski SM, Aguirre-Gutiérrez ÁA, Infanzón-Ruiz RM (2000) Rev Int Contam Ambient 16:13 161. Waliszewski SM,Aguirre AA, Infanzon RM, López-Carrillo L, Torres-Sánchez L (2000) Bull Environ Contam Toxicol 64:8 162. Webber MD, Wang C (1995) Can J Soil Sci 75:513 163. Weiderpass E, Adami HO, Baron JA, Wicklund-Glynn A, Aune M, Atuma S, Persson I (2000) Cancer Epidemiol Biomarkers Prev 9:487 164. Wiemeyer SN (1996) Other organochlorine pesticides in birds. In: Beyer WN, Heinz GH, Redomon-Norwood (eds) Environmental contaminants in wildlife: interpreting tissue concentrations. CRC Press, Boca Raton, p 99 165. WHO/FAO Reports (1969, 1973, 1974, 1978) Pesticide residues in food. Provided by Dr. JL Herrman, WHO Joint Secretary, International Program on Chemical Safety, WHO, CH-1211 Geneva 27, Switzerland 166. WHO/FAO (1985) http://www.fao.org/WAICENT/FAOINFO/AGRICULT/AGP/AGPP/Pesticid/Default.htm 167. Wickström K, Pyysalo H, Siimes MA (1983) Bull Environ Contam Toxicol 31:251 168. Yamaguchi Y, Kawano M, Tatsukawa R (1986) Chemosphere 15:453 169. Zhan W, Xu Y, Li AH, Zhang J, Schramm KW, Kettrup A (2000) Bull Environ Contam Toxicol 65:560 170. Zitko V (1971) Bull Environ Contam Toxicol 6:464 171. Zitko V (1976) Bull Environ Contam Toxicol 16:399 172. Zitko V (1979) Chemosphere 8:45 173. Zitko V (1994) Mar Pollut Bull 28:718

CHAPTER 6

Dioxins and Furans (PCDD/PCDF) Heidelore Fiedler UNEP Chemicals, 11–13, chemin des Anémones, 1219 Châtelaine (GE), Switzerland E-mail: [email protected]

Polychlorinated dibenzo-p-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF) are environmental contaminants detectable in almost all compartments of the global ecosystem in trace amounts. In contrast to other chemicals of environmental concern such as polychlorinated biphenyls (PCB), polychlorinated naphthalenes (PCN), and polychlorinated pesticides like DDT, pentachlorophenol (PCP) or others, PCDD/PCDF were never produced intentionally and do not serve any useful purpose. They are formed as by-products of numerous industrial activities and all combustion processes. Besides the anthropogenic sources of PCDD/PCDF, an enzyme-mediated formation of PCDD and PCDF from 2,4,5- and 3,4,5-trichlorophenol has been demonstrated to be responsible for their biogenic formation. The most toxic of the PCDD/PCDF congeners, 2,3,7,8-tetrachlorodibenzo-para-dioxin (2,3,7,8-Cl4DD or 2,3,7,8-TCDD) is classified to be carcinogenic to humans (=Group 1 carcinogen) according to IARC. Excluding occupational or accidental exposures, most human exposure to PCDD/PCDF occurs as a result of dietary intake, mainly by eating meat, milk, eggs, fish, and related products. PCDD/PCDF are persistent in the environment and accumulate in animal fat. Occupational exposures to PCDD/PCDF at higher levels have occurred since the 1940s as a result of production and use of chlorophenols and chlorophenoxy herbicides. Even higher exposures have occurred sporadically in relation to accidents in these industries. Many data are available for PCDD/PCDF concentrations in various compartments of the environment such as soils, sediments and air but also in biota, e.g., vegetation, wildlife, domestic animals and animals for human consumption, and finally in humans. Generation of PCDD/PCDF occurs in chemical industrial and in combustion processes, resulting in dioxin contamination of exhaust gases, solid and liquid residues, of effluents, and products. The mechanisms which lead to the formation of PCDD/PCDF have been investigated and largely understood although open questions remain. The identification of PCDD/PCDF generating activities and the quantification of the releases of PCDD/PCDF from these activities has resulted in national emission inventories. Some international conventions, such as the Stockholm Convention on Persistent Organic Pollutants (POPs) or the UN-ECE Aarhus Protocol on POPs under the LRTAP Convention require countries to report their annual emissions of PCDD/PCDF. Attempts to cover all sources and accurately quantify the releases in reporting are underway. Keywords: Polychlorinated dibenzo-p-dioxins, Polychlorinated dibenzofurans, Toxicity, En-

vironmental concentrations, Inventory

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

124

H. Fiedler

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126

1

Introduction

2

Physical and Chemical Properties . . . . . . . . . . . . . . . . . . 127

3

Environmental Fate

3.1 3.2

General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 Carry-Over Rates: Environment-to-Food . . . . . . . . . . . . . . 129

4

Concentrations in the Environment, in Foodstuffs and in Humans 132

4.1 4.1.1 4.1.2 4.1.3 4.1.4 4.1.5 4.2 4.2.1 4.2.2 4.2.3

Environment . . . . . . . . . . . . . . Soil . . . . . . . . . . . . . . . . . . . . Sediment . . . . . . . . . . . . . . . . Ambient Air . . . . . . . . . . . . . . . Biomonitors . . . . . . . . . . . . . . . Sewage Sludge . . . . . . . . . . . . . . Human Exposure and Levels in Humans Feedstuffs and Foods . . . . . . . . . . Results from Individual Studies . . . . Humans . . . . . . . . . . . . . . . . .

5

Toxicity of PCDD/PCDF

5.1 5.1.1 5.1.2 5.1.3 5.1.4

Toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mode of Action . . . . . . . . . . . . . . . . . . . . . . . . . . . Carcinogenicity . . . . . . . . . . . . . . . . . . . . . . . . . . . Toxic Effects in Humans Other than Carcinogenicity . . . . . . . Toxic Effects in Laboratory Animals Other than Carcinogenicity

6

Risk Assessment and Risk Management

6.1 6.2 6.3 6.3.1 6.3.2 6.3.3 6.3.4 6.3.5 6.3.6

Risk Assessment and the TEF Approach Risk Assessment by US-EPA . . . . . . Risk Management . . . . . . . . . . . . Tolerable Intakes . . . . . . . . . . . . Regulation of Chemicals . . . . . . . . Incineration and Combustion . . . . . Water Discharges and Solid Residues . Environmental Media . . . . . . . . . Food and Feedstuff Regulations . . . .

7

Sources of PCDD/PCDF

7.1 7.2 7.2.1 7.2.2 7.3 7.4

Overview . . . . . . . . . . . . . . Primary Sources of PCDD/PCDF . Industrial-Chemical Processes . . . Thermal Processes . . . . . . . . . Secondary Sources of PCDD/PCDF Natural Sources . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . 128

. . . . . .

. . . . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

132 133 134 135 140 141 141 141 142 145

. . . . . . . . . . . . . . . . . . . . . . . 145 . . . .

146 146 147 149 150

. . . . . . . . . . . . . . 151 . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

151 152 153 153 156 157 160 161 163

. . . . . . . . . . . . . . . . . . . . . . . 164 . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

164 165 165 166 172 173

125

Dioxins and Furans (PCDD/PCDF)

8

PCDD/PCDF Inventories . . . . . . . . . . . . . . . . . . . . . . . 174

8.1 8.1.1 8.2 8.3 8.3.1 8.3.2 8.3.3 8.3.4 8.3.5 8.3.6 8.4

Methods to Establish Dioxin Inventories . . . . . . . . General . . . . . . . . . . . . . . . . . . . . . . . . . . Existing Inventories until 1999 . . . . . . . . . . . . . . PCDD/PCDF Inventories after 1999 . . . . . . . . . . . Canada . . . . . . . . . . . . . . . . . . . . . . . . . . Denmark . . . . . . . . . . . . . . . . . . . . . . . . . Japan . . . . . . . . . . . . . . . . . . . . . . . . . . . Hong Kong SAR . . . . . . . . . . . . . . . . . . . . . . New Zealand . . . . . . . . . . . . . . . . . . . . . . . European Union Member States . . . . . . . . . . . . . Outlook: PCDD/PCDF Inventories under the Stockholm Convention on POPs . . . . . . . . . . . . . . . . . . .

9

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

174 174 174 177 177 178 181 182 183 184

. . . . . . 193

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195

Abbreviations 2,4,5-T AhR BAT bgvv

2,4,5-Trichloroacetic acid Aryl hydrocarbon receptor Best available techniques Bundesinstitut für gesundheitlichen Verbraucherschutz und Veterinärmedizin (Germany) [Federal Institute for Health Protection and of Consumers and Veterinary Medicine] bw Body weight °C Degrees Celsius DG Dirección Générale d.m. Dry matter EC European Commission EPA Environmental Protection Agency EU European Union (Member States: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, The Netherlands, Portugal, Spain, Sweden, United Kingdom) FAO Food and Agricultural Organization h Hour HCl Hydrochloric acid, hydrogen chloride I-TEQ International Toxic Equivalents IARC International Agency for Research on Cancer IPCS International Programme on Chemical Safety Partition coefficient: octanol/carbon KOC Partition coefficient: octanol/water KOW LOQ Limit of quantification LRTAP [Convention on] Long-range Transboundary Air Pollution M(S)WI Municipal (solid) waste incineration NATO/CCMS North Atlantic Treaty Organization/Challenges of Changes in Modern Society

126

H. Fiedler

Nm3 ND PCB PCDD PCDF PCN PCP POPs ppt PUF PVC t TCDD TDI TEF TEQ TMI TWI UN-ECE UNEP WHO mg µg ng pg fg

Normal cubic meter at 11% O2 , 101.3 kPa, 273 K, dry gas Not determined Polychlorinated biphenyls Polychlorinated dibenzo-p-dioxins Polychlorinated dibenzofurans Polychlorinated naphthalenes Pentachlorophenol Persistent Organic Pollutants Parts per trillion Polyurethane foam Polyvinyl chloride Ton (metric) Tetrachlorodibenzo-p-dioxin Tolerable daily intake Toxicity Equivalency Factor Toxicity Equivalent Tolerable monthly intake Tolerable weekly intake United Nations Economic Commission for Europe United Nations Environment Programme World Health Organization

Milligram Microgram Nanogram Picogram Femtogram

10—3 g 10—6 g 10—9 g 10—12 g 10—15 g

1 Introduction Polychlorinated dibenzo-p-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF) are environmental contaminants detectable in almost all compartments of the global ecosystem in trace amounts. These compound classes in particular have caused major environmental concern. In contrast to other chemicals of environmental concern such as polychlorinated biphenyls (PCB), polychlorinated naphthalenes (PCN), and polychlorinated pesticides like DDT, pentachlorophenol (PCP) or others, PCDD/PCDF were never produced intentionally. They are formed as by-products of numerous industrial activities and all combustion processes [1]. The term “dioxins” 1 is frequently used and refers to 75 congeners of polychlorinated dibenzo-p-dioxins (PCDD) and 135 congeners of polychlorinated dibenzofurans (PCDF). These are two groups of planar, tricyclic ethers, which 1

In this paper, where the term “dioxin” or “dioxins and furans” is used alone, it should be interpreted as including all polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans, unless specifically stated otherwise.

127

Dioxins and Furans (PCDD/PCDF)

PCDD

PCDF

Fig. 1. Structural formula of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans and numbering of carbon atoms

have up to eight chlorine atoms attached at carbon atoms 1 to 4 and 6 to 9 (Fig. 1). Amongst these 210 compounds, 17 congeners can have chlorine atoms at least in the positions 2, 3, 7, and 8 of the parent molecule. These seventeen 2,3,7,8substituted congeners are toxic to many laboratory animals, resistant towards chemical, biological, and physical attack, and thus many accumulate in the environment and in organisms, such as animals and humans. The 2,3,7,8-TCDD (2,3,7,8-Cl4DD) also named “Seveso dioxin” is considered to be the most toxic man-made compound. Besides the anthropogenic sources, an enzyme-mediated formation of PCDD and PCDF from 2,4,5- and 3,4,5-trichlorophenol has been demonstrated in vitro being responsible for biogenic formation to occur, e.g., in sewage sludge, compost, etc. [2, 3]. In recent years dioxins have been detected in natural formations of clay in different parts of the world.

2 Physical and Chemical Properties PCDD and PCDF each can have between one and eight chlorine atoms bound to the dibenzo-p-dioxin or dibenzofuran molecule, respectively. This substitution pattern results in eight homologues and 75 congeners for PCDD and 135 congeners for PCDF. Table 1 summarizes the number of possible isomers within each group of homologue. Starting in the 1970s, PCDD and PCDF congeners have been characterized with most information available for 2,3,7,8-Cl4DD. Today, all 17 2,3,7,8-substituted congeners are available commercially, either individually or as mixtures. Knowledge of the numeric values of certain parameters characterizing the properties of individual PCDD/PCDF is necessary in order to predict the behavior of the mixtures found in the environment. However, measured values for PCDD/PCDF congeners are scarce. The physical and chemical properties, which are measures of, or control the behavior of dioxins and furans are: – their low vapor pressure (ranging from 4.0 ¥10 – 8 mm Hg for 2,3,7,8-Cl4DF to 8.2¥10 –1 3 mm Hg for Cl8DD);

128

H. Fiedler

Table 1. Possible number of isomers within the homologue groups for PCDD and PCDF

Number of Chlorine Atoms=Homologues

Possible number of isomers Dibenzo-p-dioxins (PCDD)

Acronym

MonochloroDichloroTrichloroTetrachloroPentachloroHexachloroHeptachloroOctachloro-

2 10 14 22 14 10 2 1

Cl1DD Cl2DD Cl3DD Cl4DD Cl5DD Cl6DD Cl7DD Cl8DD

Total

75

Dibenzofurans (PCDF) 4 16 28 38 28 16 4 1

Acronym Cl1DF Cl2DF Cl3DF Cl4DF Cl5DF Cl6DF Cl7DF Cl8DF

135

– their extremely low solubility in water (ranging from 419 ng L–1 for 2,3,7,8Cl4DF, 7.9 and 19.3 ng L–1 for 2,3,7,8-Cl4DD to 0.074 ng L–1); – their solubility in organic/fatty matrices (log KOW range from 5.6 for Cl4DF and 6.1/7.1 for Cl4DD to 8.2 for Cl8DD); – their preference to bind to organic matter in soil and sediments (log KOC values for 2,3,7,8-Cl4DD are between 6.4 and 7.6). Detailed information on the physical and chemical characteristics of PCDD/PCDF can be found at Mackay et al. [4].

3 Environmental Fate 3.1 General

The environmental processes by which PCDD/PCDF move through the environment are reasonably well known. PCDD/PCDF are multimedia pollutants and, once released to the environment, become distributed between environmental compartments [5]. Due to their high lipophilicity and low water solubility, PCDD/PCDF are primarily bound to particulate and organic matter in soil and sediment, and in biota, they are concentrated in fatty tissues. In air, as semi-volatile compounds PCDD/PCDF can exist in both the gaseous phase and bound to particles. The two key parameters, the congener’s vapor pressure, and the ambient air temperature govern the partitioning between gaseous phase and particle. Especially during the warmer (on the northern hemisphere summer) months the lower chlorinated PCDD/PCDF congeners tend to be found predominantly in the vapor phase. PCDD/PCDF in the vapor phase can undergo photochemical transformation with dechlorination process leading to more toxic congeners if octa- and hep-

Dioxins and Furans (PCDD/PCDF)

129

tachlorinated congeners degrade to tetra- and pentachlorinated and finally to non-toxic compounds with only three or less chlorine atoms. PCDD/PCDF attached to particulate matter seem to be resistant to degradation. In the terrestrial foodchain (air Æ grass Æ cattle Æ milk/meat Æ man) PCDD/PCDF can be deposited on plant surfaces via wet deposition, via dry deposition of chemicals bound to atmospheric particles, or via diffusive transport of gaseous chemicals in the air to the plant surfaces. Each of these processes is governed by a different set of plant properties, environmental parameters, and atmospheric concentrations. Investigations with native grassland cultures showed that dry gaseous deposition plays the dominant role for the accumulation of the lower chlorinated PCDD/PCDF, whereas dry particle-bound deposition played an important role in the uptake of the PCDD/PCDF with six and more chlorine atoms. There was also some evidence indicating an input of the higher chlorinated PCDD/PCDF from wet deposition [6]. Levels in, e.g., grass, reflect recent exposure to PCDD/PCDF, as vegetation is only exposed for a relatively short time, with new growth replacing old and crops being harvested. For agricultural leaf crops the main source of contamination is direct deposition from the atmosphere and soil splash. Root uptake and translocation of dioxin contamination into the crop has been confirmed for zucchini and cucumber only. Grazing animals are exposed to dioxins by ingesting contaminated pasture crops and PCDD/PCDF are found to accumulate primarily in the fatty tissues and milk [7]. For agricultural soils an additional source of PCDD/PCDF can be the application of sewage sludge. Small amounts of PCDD/PCDF deposited onto soil can be returned to the atmosphere by the resuspension of previously deposited material, or revolatilization of the less chlorinated congeners. Because of their chemical characteristics and very low solubility PCDD/PCDF accumulate in most soil types, with very little water leaching and negligible degradation of the 2,3,7,8substituted PCDD/PCDF congeners. PCDD/PCDF partition quickly to organic matter and so accumulate in sediments. They accumulate in aquatic fauna as a result of the ingestion of contaminated organic matter. The concentration of PCDD/PCDF in fish tissue is found to increase up the foodweb (biomagnification) as a result of the progressive ingestion of contaminated prey [5]. 3.2 Carry-Over Rates: Environment-to-Food

The transfer of PCDD/PCDF from grass into cattle has been studied and carryover rates have been determined. In general, carry-over rates decrease with increasing degree of chlorination of the chemical, indicating that absorption through the gut also decreases. This decrease in absorption is attributed to the greater hydrophobicity of the higher chlorinated PCDD/PCDF, which inhibits their transport across aqueous films in the digestive tract of the cow. In studies at background concentrations, the highest transfer was determined for two lower chlorinated dibenzo-p-dioxins and one dibenzofuran, namely 2,3,7,8-Cl4DD (2,3,7,8-tetrachlorodibenzo-p-dioxin), 1,2,3,7,8-Cl5DD (1,2,3,7,8pentachlorodibenzo-p-dioxin), and 2,3,4,7,8-Cl5DF (2,3,4,7,8-pentachlorodiben-

130

H. Fiedler

zofuran). For these three congeners about 30–40% are transferred from feed to cow’s milk. About 20% are transferred for the 2,3,7,8-substituted Cl6DD (hexachlordibenzo-p-dioxins) and Cl6DF (hexachlorodibenzofurans) homologues. For the hepta- and octachlorinated PCDD and PCDF not more than 4% of the ingested congeners find their way into the milk.Although highly dependent on the characteristics of each congeners, the overall transfer on a TEQ basis is about 30%; in other words: about 30% of the most toxic PCDD/PCDF congeners, which are ingested by the cow are excreted via the milk [8]. The numeric values for the carry-over rates from feed to cow’s milk are summarized in Table 2. Generally, Table 2 shows that the carry-over rates decrease with increasing degree of chlorination, indicating that absorption through the gut also decreases. This decrease in absorption was attributed to the greater hydrophobicity of the higher chlorinated PCDD/PCDF, which inhibits their transport across aqueous films in the digestive tract of the cow [8]. Comprehensive investigations have shown that the type of housing for the laying hens results in differences in the PCDD/PCDF contamination of their eggs. As can be seen from Table 3 the majority of PCDD/PCDF concentrations in eggs from chickens housed in elevated wire cages is below 2 ng I-TEQ/kg fat; only a small number of samples ranged up to 2.3 ng I-TEQ/kg fat. In contrast, eggs from laying hens kept on ground and from foraging chickens raised on fields show a broader range of contamination; of these a considerable number of samples revealed PCDD/PCDF levels above 2 ng TEQ/kg fat. The highest contamination was found to be 23.4 ng I-TEQ/kg fat. Moreover, these studies revealed that the concentrations and congener profiles of PCDD/PCDF in eggs of chickens appear to be related to the soil on which they are raised. As a consequence, chicken eggs,

Table 2. Carry-over rates for PCDD/PCDF from feed into cow milk

Welsch-Pausch and McLachlan [8]

McLachlan and Richter [112]

2,3,7,8-Cl4DD 1,2,3,7,8-Cl5DD 1,2,3,4,7,8-Cl6DD 1,2,3,6,7,8-Cl6DD 1,2,3,7,8,9-Cl6DD 1,2,3,4,6,7,8-Cl7DD Cl8DD

0.34 0.31 0.127 0.21 0.11 0.028 0.0121

0.38 0.39 0.33 0.33 0.16 0.034 0.0068

2,3,7,8-Cl4DF 1,2,3,7,8-Cl5DF 2,3,4,7,8-Cl5DF 1,2,3,4,7,8-Cl6DF 1,2,3,6,7,8-Cl6DF 2,3,4,6,7,8-Cl6DF 1,2,3,4,6,7,8-Cl7DF 1,2,3,4,7,8,9-Cl7DF Cl8DF

0.0083 0.0107 0.26 0.094 0.098 0.089 0.0146 0.023 0.0053

0.40 0.24 0.187 0.189 0.034

131

Dioxins and Furans (PCDD/PCDF)

Table 3. Germany: PCDD/PCDF concentrations in eggs [9]. Concentrations in pg I-TEQ/g lipid

Method keeping

n

Min

Max

Mean

Caged, housing in elevated wire cages

20 69 11 32 23 31

0.56 0.23 1.03 0.19 0.38 0.49

2.30 6.04 23.4 5.57 11.4 22.8

1.16 a 1.36 1.81 a 1.63 1.91 a 4.58

Chicken kept on ground Free foraging a

Median.

especially from laying hens kept on contaminated ground or from free foraging chickens may contribute considerably to human dioxin body burden. Consequently, egg samples very often are characterized by high concentrations of Cl8DD indicating that the soil contamination may be transferred into the animal. In cases of very high dioxin contamination, concentrations up to several hundred pg I-TEQ/g lipid (e.g., 300 pg I-TEQ/g lipid in Baden-Württemberg, Germany, and 219 pg I-TEQ/g in Hamburg) have been reported [8]. The transfer of PCDD/PCDF from soil into chicken eggs was investigated in an exposure study by Petreas et al. [10]. There was only little variation in PCDD/PCDF concentrations in eggs after 30, 60, and 80 days, respectively. Thus,

Table 4. PCDD/PCDF transfer from soil to eggs

Petreas et al. [10] [egg concentration (pg/g fat)/ soil concentration (pg/g)]

Schuler et al. [12] [egg concentration (pg/g fat)/ soil concentration (pg/g)]

2,3,7,8-Cl4DD 1,2,3,7,8-Cl5DD 1,2,3,4,7,8-Cl6DD 1,2,3,6,7,8-Cl6DD 1,2,3,7,8,9-Cl6DD 1,2,3,4,6,7,8-Cl7DD Cl8DD

n.c. 0.41 0.52 0.53 0.36 0.30 0.14

1.2 2.4 1.5 1.6 0.8 0.4 0.1

2,3,7,8-Cl4DF 1,2,3,7,8-Cl5DF 2,3,4,7,8-Cl5DF 1,2,3,4,7,8-Cl6DF 1,2,3,6,7,8-Cl6DF 1,2,3,7,8,9-Cl6DF 2,3,4,6,7,8-Cl6DF 1,2,3,4,6,7,8-Cl7DF 1,2,3,4,7,8,9-Cl7DF Cl8DF

0.25 1.37 0.67 0.61 0.53 n.c. 0.26 0.22 0.16 0.09

3.3 4.4 0.8 0.9 1.0 0.1 0.6 0.2 0.1 0.1

n. c. not calculated.

132

H. Fiedler

it can be assumed that a relatively fast equilibrium between the concentration in the soil ingested with feed and the concentration in the eggs is reached. The results were confirmed by a second study, which clearly showed that the accumulation factors were inversely correlated with the degree of chlorination [11]. Table 4 summarizes the egg/soil ratios (mean of samples collected on days 30, 60, and 80) reported by Petreas et al. [8], and compares them to the soil-to-egg transfer rates determined by Schuler et al. [12]. Although there is are several differences between these two studies (i.e., Petreas et al. performed laboratory experiments, whereas Schuler et al. conducted a field study), common trends can be recognized as follows: Table 4 shows that the quotients (egg/soil) decrease by more than one order of magnitude from the lower chlorinated to the higher chlorinated congeners. Similar observations were reported for the transfer of PCDD/PCDF from feed into dairy milk (see Sect. 3.2).

4 Concentrations in the Environment, in Foodstuffs and in Humans 4.1 Environment

Many data are available for PCDD/PCDF concentrations in soils, sediments, and air. Biomonitors, such as vegetation or cows’ milk, have been successfully applied to identify or monitor ambient air concentrations in the neighborhood of potential point sources, although a linear correlation between PCDD/PCDF concentrations in vegetation and air samples cannot be established. Due to public concern regarding dioxins and furans, many studies have been aimed at identifying potential ‘hotspots’ of contamination. As a result, the overall presentation of data is often biased towards contaminated samples and higher concentrations, rather than baseline information. When evaluating concentrations of PCDD/PCDF in the environment, it should be taken into account that some matrices are sensitive to short-term inputs, e.g., ambient air or short-lived vegetation, whereas other matrices, such as sediments and soils, are relatively insensitive to temporal variation. Further important factors for the interpretation of results are season (e.g., in winter PCDD/PCDF concentrations in air may be higher by a factor of ten on TEQ basis than in summer), length of the sampling or exposure (e.g., few hours vs. weeks), location (e.g., urban vs. rural), the sampling method (e.g., high-volume sampling vs. particulate deposition), sampling depth (e.g., surface vs. core), etc. [5]. Soils are natural sinks for persistent and lipophilic compounds such as PCDD/PCDF, which adsorb to the organic carbon of the soil and, once adsorbed, remain relatively immobile. Soil is a typical accumulating matrix with a long memory; in other words, dioxin inputs received in the past will remain and, due to the very long half-lives of PCDD/PCDF in soils, there is hardly any clearance. Soils can receive inputs of environmental pollutants via different pathways of which the most important are: atmospheric deposition, application of sewage sludge or composts, spills, erosion from nearby contaminated areas. Sed-

Dioxins and Furans (PCDD/PCDF)

133

iments are the ultimate sink for PCDD/PCDF (and other persistent and lipophilic organic substances). As with soils, sediment samples are accumulating matrices for lipophilic substances and can receive inputs via different pathways: atmospheric deposition, industrial and domestic effluents, stormwater, spills, etc. Today, PCDD/PCDF can be detected ubiquitously and have been measured in the Arctic, where almost no dioxin sources are present. It became clear that the lipophilic pollutants, such as PCDD/PCDF, at the North and the South Pole originated from lower (warmer) latitudes. Emission of most PCDD/PCDF from combustion sources into the atmosphere occurs in the moderate climate zones; PCDD/PCDF then undergo long-range transport towards the North Pole, condensing in the cooler zones when the temperatures drop. This process of alternating re-volatilization and condensing, also named the “grasshopper effect”, can carry pollutants thousands of kilometers in a few days. Thus, the air is an important transport medium for PCDD/PCDF. An indirect method of determining ambient air concentrations is the use of biomonitors, such as vegetation. The outer waxy surfaces of pine needles, kale or grass absorb atmospheric lipophilic pollutants and serve as an excellent monitoring system for PCDD/PCDF [5]. The European Commission has commissioned a project to collect and evaluate PCDD/PCDF results from the fifteen Member States in order to have a better overview of existing data and to provide a basis for a common policy for these substances (EC 1999). In most countries a broad range of PCDD/PCDF concentrations has been detected in all media.As illustrated in the subsequent Tables for all matrices (Table 5 to Table 14) lowest concentrations are always close to the limit of determination whereas highest concentrations are more than 1,000-fold higher [5]. 4.1.1 Soil

Sampling depth and use patterns play an important role when reporting soil concentrations. In many sampling programs, agricultural soils are sampled to a depth of 30 cm in cases of arable land and 2–10 cm in cases of pastureland. Contained soils are sampled according to their composition in layers (on optical inspection). Forest soils are usually separated into litter and the various horizons of the mineral soil. Within the EU Member States, the largest databases existed for PCDD/PCDF concentrations in soil. As shown in Table 5 at contaminated locations measured concentrations range from several hundred to around 100,000 ng I-TEQ/kg d.m. The highest concentrations were found in Finland at sites contaminated with wood preservatives and the Netherlands close to a scrap car and scrap wire incinerator [5, 13]. The EPA Dioxin Reassessment document estimate mean TEQ values for background urban and rural soils to be 13.4 and 4.1 ng I-TEQ/kg of soil, respectively [14].

134

H. Fiedler

Table 5. Summary of PCDD/PCDF concentrations in soil from EU Member States. Concentrations in ng TEQ/kg d.m. [13]

Any type Austria Belgium Finland Germany Greece Ireland Italy Luxembourg The Netherlands Spain Sweden United Kingdom

Forest 0.01–64

Pasture

Arable

Rural

1.6–14

332

2.7–8.9 0.1–42 2–45 0.15–8.6 0.057–0.12 1.8–20 2–55 0.63–8.4

Contamin.

2.1–2.7 10–30 4.8

0.004–30

0.03–25

0.8–13 0.1–43

1.9–3.1

6.0

85,000 30,000 1,144

1

1.4 2.2–17 0.1–8.4 0.11 0.78–20

0.78–87

98,000 11,446 1,585

4.1.2 Sediment

PCDD/PCDF concentrations in sediments from EU Member States are summarized in Table 6. Normally, the ranges are from a few ng TEQ/kg d.m. to ca. 50 ng TEQ/kg d.m. However, hotspots were identified in many countries where concentrations exceed 1,000 ng TEQ/kg d.m.: up to 80,000 ng I-TEQ/kg d.m. were reported from Finland, downstream from a wood preservative producing site (Table 6) [13]. US-EPA conducted a study on sediment cores from 11 U.S. lakes/reservoirs [15]. The lakes were located in various geographic locations throughout the United States and were selected to represent background conditions (i.e., no known PCDD/PCDF sources nearby). Based on the most recently deposited sediments, the I-TEQ concentrations ranged from 0.11 ng I-TEQ/kg to 15.6 ng I-TEQ/kg with a mean of 5.3 ng I-TEQ/kg (when concentrations below the limit of quantification were set to one-half the quantification limit) (Table 7). Chandler Lake, an Arctic lake located in North Slope,Alaska, had the lowest concentration, and Canandaigua Lake in New York and Santeetlah Reservoir in North Carolina, both eastern lakes, had the highest concentrations.

Table 6. Summary of PCDD/PCDF concentrations in sediments from EU Member States. Concentrations in ng TEQ/kg d.m. [13]

Finland

Germany Italy

Background 0.7–100 1.2–19 Urban 12–73 Contaminated 80,000 >1500

Lux.

Netherl. Spain

0.07–10 1–10 0.5–23 2.4–16 570 4000

Sweden

UK

0.8–207 0.2–57 1692

2–123 7,410

135

Dioxins and Furans (PCDD/PCDF)

Table 7. PCDD/PCDF concentrations in the most recent layer of sediment cores from 11 U.S. lakes [15]; ND=1/2 LOQ

Lake

Conc. (ng I-TEQ/kg d.m.)

Range of Dates

Chandler Lake, AK Canandaigua Lake, NY Skaneateles Lake, NY Great Sacanaga Reservoir, NY Santeetlah Reservoir, NC Blue Ridge Reservoir, GA Deer Creek Reservoir, UT Echo Lake, UT Panguitch Lake, UT Ozette Lake, WA Beaver Lake, WA Mean

0.11 15.0 10.1 6.4 15.6 5.6 1.2 0.82 0.91 1.2 0.98 5.3

1956–1993 1981–1991 1984–1991 1974–1983 1974–1983 1973–1983 1973–1982 1973–1982 1976–1985 1977–1985 1974–1985

4.1.3 Ambient Air

Results for air samples were available for only eight countries (Table 8). There are three basic approaches to determine the PCDD/PCDF concentrations in air: highvolume samplers which will collect particle-bound and gas-phase PCDD/PCDF, Bergerhoff or similar samplers which will collect dry and wet deposition and biomonitors such as kale, spruce needles or grass which preferentially absorb the gas-phase dioxins and furans. Table 8 shows that, once again, the concentrations in ambient air range from 1 to several hundred fg I-TEQ/m3 and in deposition, a similar range was found for the concentrations in pg TEQ/m2 · d. The extremely high concentration of 14,800 fg I-TEQ/m3 in ambient air was measured in 1992/93 at the Pontyfelin House site, in the Panteg area of Pontypool in South Wales, which is very close (~150 m) to an industrial waste incinerator [5, 13]. Table 8. Summary of air concentrations from EU Member States. Concentrations of ambient

air samples in fg TEQ/m3 and deposition in pg TEQ/m2 · d [13] Ambient Air Unspecified Austria 1.3–587 Belgium Germany 1–705 Italy 85 Luxembourg The Netherlands 4–99 Sweden United Kingdom

Deposition Urban

Rural

Urban

Rural

86–129

70–125

0.9–12 0.5–464

0.7–3.1

47–277 54–77 0.2–54 17–103

30–64 9–63 6–12

Contaminated

6–140 0.4–312

0–517

14,800

136

H. Fiedler

A one-year sampling of ambient air in Leoben, Austria, was performed from November 1999 until October 2000. The results (Fig. 2) show a pronounced seasonal trend with the highest PCDD/PCDF concentrations in winter and lower concentrations in summer. It can also be seen that the PCB concentrations – expressed as the sum of the six Ballschmiter congeners – show have the maximum during summer (7,066 fg/m3 in May 2000; minimum=1,490 fg/m3 in February 2000). The concentrations of dioxin-like PCB were low and ranged from 3 fg WHO-TEQ/m3 in September 2000 to 25.3 fg WHO-TEQ/m3 in January 2000; a seasonal trend could not be recognized for these PCB. The annual average was 194 fg I-TEQ/m3 (minimum=82 fg I-TEQ/m3 in June 2000 and September 2000; maximum=490 fg I-TEQ/m3 in January 2000). This mean value is higher than the concentrations in other urban areas in Austria, e.g., 80 fg I-TEQ/m3 in Vienna or 120 fg I-TEQ/m3 in Graz [16]. The strong seasonal trends in ambient air were demonstrated in an evaluation of the data of the German Dioxin Database, jointly maintained by the Federal Environment Agency (Umweltbundesamt=UBA) and the Federal Institute for Consumer Protection (Bundesinstitut für gesundheitlichen Verbraucherschutz und Veterinärmedizin=bgvv). The monthly averages from 847 samples collected between April 1989 and March 1999 at sites without special impact from nearby point sources are shown in Fig. 3 (seven outliers, which had concentrations above 250 fg I-TEQ/m3 were not included). As can be seen, strong seasonal trends have been observed with a variation of the monthly means by a factor of 7–30 and semi-annual means by a factor of 2–5. Further, a downward trend was observed: the winter maxima of 1998/99 were about half of the winter maxima of the year 1989/90 [17]. The median, minimum, maximum and 90th percentiles of these 847 samples show that highest concentrations were found in urban and agglomeration areas

Fig. 2. PCDD/PCDF (in fg I-TEQ/m3 and WHO-TEQ/m3) and PCB (in fg WHO-TEQ/m3 and for the sum of the 6 Ballschmiter PCB) in ambient air around a sinter plant in Austria [16]

Dioxins and Furans (PCDD/PCDF)

137

Fig. 3. PCDD/PCDF (fg I-TEQ/m3) in ambient air – monthly means from 847 samples collected in Germany between April 1989 and March 1999 [17]

Fig. 4. PCDD/PCDF (fg I-TEQ/m3) in ambient air – median, minimum, maximum and 90th percentiles from 847 samples collected in Germany between April 1989 and March 1999 [17]

138

H. Fiedler

in winter (90th p=130–160 fg I-TEQ/m3, median=50 fg I-TEQ/m3), whereas the concentrations in rural areas in winter and at all locations in summer were very similar (90th p=50 fg I-TEQ/m3; median =20 fg I-TEQ/m3) (Fig. 4). Many countries have utilized vegetation to monitor ambient air concentrations. The use of these biomonitors was found useful for both routine programs on a long-term basis or to identify potential hotspots around potential point sources. The use of kale was successfully implemented around a steel producing plant in Luxembourg where mean concentrations up to 10 ng I-TEQ/kg d.m. were detected; in Germany 12.6 ng I-TEQ/kg d.m. were determined close to combustion sources. In Austria, spruce needles are utilized as biomonitors: the background concentrations were in a very narrow range between 0.3 and 1.9 ng I-TEQ/kg d.m. Normally, baseline concentrations were around 0.5 ng I-TEQ/kg d.m. in rural areas and around 1–1.7 ng I-TEQ/kg d.m. in urban areas. Studies from Bavaria and Hesse in Germany reported that mean PCDD/PCDF concentrations in pine needle ranged from 0.53 to 1.64 pg I-TEQ/g d.m. However, in the neighborhood of the Brixlegg copper reclamation plant between 51 and 86 ng I-TEQ/kg were determined. In Welsh Rye grass, which is typically exposed for four weeks during the summer, concentrations normally are between 0.5 and 1 ng I-TEQ/kg d.m. [5]. Atmospheric concentrations have been determined by parallel sampling in Mississippi when in winter 1995/96 (December 95/January 96) and in summer 1996 (June/July 1996) ambient air samples were taken with the a high-volume sampler (glass-fiber filter+PUF), deposition samples were taken according to the Bergerhoff method, and pine needles were collected and analyzed (Fiedler et al. 1997). The sampling sites represented rural areas in the southern United States; the concentrations at both sampling sites were comparable. The concentrations of PCDD/PCDF, expressed as I-TEQ and as SPCDD/PCDF (not shown in Table 9) were by a factor of approximately 3 higher in winter than in summer. In the deposition samples (Table 10), the winter exposure was approximately 4-times higher than in summer in both counties. The results for the pine need-

Table 9. PCDD/PCDF in ambient air and in deposition samples from Mississippi [18]

Location – Week

Lamar Co. – 1 Lamar Co. – 2 Lamar Co. – 3 Lamar Co. – 4 George Co. – 1 George Co. – 2 George Co. – 3 George Co. – 4 Mean a 1/ 2

I-TEQ (fg/m3)

I-TEQ (pg/m2)

Winter

Summer

Winter

Summer

17 12 13 6.0 5.6 13 15 7.0 10.4

6.1 5.1 2.6

3.1

0.42

2.0

0.73

2.6

0.58

a

2.6 a

2.3 a

3.7

the LOQ for non-quantifiable congeners was used to calculate the I-TEQ.

139

Dioxins and Furans (PCDD/PCDF) Table 10. PCDD/PCDF in pine needle samples from Mississippi [18]

Location

Shoot

Lamar Co. George Co. Mean

1995 1995

9 9 9

0.29 0.23 0.26

Lamar Co. George Co. Mean

1994 1994

21 21 21

0.56 0.40 0.48

Lamar Co. George Co. Mean

1995 1995

15 15 15

0.55 0.79 0.67

Lamar Co. George Co. Mean

1996 1996

3 3 3

0.30 0.16 0.23

a 1/ 2

Exposure Time (months)

I-TEQ (ng/kg d.m.)

the LOQ for non-quantifiable congeners was used to calculate the I-TEQ.

Table 11. PCDD/PCDF concentrations in spruce needles from Bavaria (ng I-TEQ/kg d.m.) [19]

n

Fall 92 Spring 93 Fall 93 Spring 94 Fall 95 Spring 96 Fall 97 Spring 98 26 26 15 15 20 20 21 20

Min Max Median

0.18 1.2 0.53

0.27 3.45 1.12

0.14 1.1 0.5

0.18 1.33 0.74

0.14 1.53 0.56

0.52 1.91 1.01

0.12 0.52 0.31

0.24 0.65 0.46

les (Table 10) showed the trend towards higher concentrations with increasing exposure times; however, a linear correlation could not be established [18]. The German dioxin database also contains 163 results from spruce needle monitoring performed in the State of Bavaria between 1992 and 1998. Samples have been taken in the fall and in the spring of the following year. This means that the needles collected in spring had a month longer exposure time compared to the needles collected in fall. Table 11 shows the minimum, maximum and median concentrations of for the various seasons. The concentrations range from 0.12 ng I-TEQ/kg d.m. to 3.45 ng I-TEQ/kg d.m. with a mean concentration of 0.67 ng I-TEQ/kg d.m. for all samples. It can be seen that the concentrations in spring are always higher than in the preceding fall [17]. In an extensive biomonitoring program in Hesse (Germany) in the year 1992/93, spruce trees and kale have been exposed in standardized soil at 24 monitoring stations; in addition, a special program was performed at Frankfurt Rhein-Main airport.At four of these stations, the local soil has been used to grow the plants. PCDD/PCDF concentrations in these soils were between 1.4 and 23.9 ng I-TEQ/kg d.m. For each matrix, two “clean” exposures have been included in the program for comparison: an open-top chamber and a background station

140

H. Fiedler

Table 12. PCDD/PCDF concentrations in spruce needles and kale in Hesse (ng I-TEQ/kg d.m.)

[13] Station

Min

Max

Mean

Median

Background station

Open-top chamber

1.71 0.91

2.14 2.44

1.25 0.89

Stations 1–24 (standardized soil) (n=24) Spruce Kale

1.02 0.59

2.5 1.51

1.72 0.91

Stations 2a,9a, 15a, 17a using local soil (n=4) Spruce Kale

1.45 0.67

2.67 0.96

2.09 0.85

1.83 0.88

Table 13. PCDD/PCDF concentrations in kale at Frankfurt Rhein-Main airport (n=10) [13]

PCDD/PCDF (ng I-TEQ/kg) PCB (#52, 101, 105, 138, 153,180) (mg/kg) PCB (3–10 Cl) (mg/kg) a

Minimum

Maximum

Mean

Median

0.64 3.04 8.07

1.47 43.99 120.28

0.95 4.57 a 11.90 a

0.92 4.61 a 11.99 a

Without maximum concentration at sampling station FAG 7.

with no known dioxin source nearby. The results are shown in Table 12 and Table 13. The concentrations in spruce needles and kale in Hesse are comparable with concentrations typically found in industrially impacted areas in Germany in the early 1990s. The program at the Frankfurt airport did not give higher concentrations of PCDD/PCDF indicating that the air traffic did not have an impact on the PCDD/PCDF concentrations. But extremely high concentrations of PCB were found in one sample at a station close to the terminal. This result is an indicator that PCB-containing equipment was still in use at the end of 1992 [20]. 4.1.4 Biomonitors

Fish and shellfish were frequently used as biomonitors for the aquatic environment. As can be seen from Table 14, fish are highly bioaccumulative for PCDD/ PCDF so that several hundred pg TEQ/g fat were detected in these animals. These concentrations are much higher than those found in terrestrial animals, such as cattle, pig, or chicken. Top-predators like sea eagles or guillemots also showed high concentrations of PCDD/PCDF: as an example in Finland, 830 to 66,000 pg TEQ/g fat were found in white-tailed sea eagles. The Swedish Dioxin Database reported a wide range of dioxin concentrations in the blubber of ringed seal: 6.3 to 217 pg TEQ/g fresh weight [5].

141

Dioxins and Furans (PCDD/PCDF)

Table 14. Summary of fish concentrations from EU Member States. Concentrations in pg TEQ/g

fat [13]

Concentration

Finland

Germany

Sweden

United Kingdom

75–200

40–51

9.1–420

16–700

Table 15. Summary of sewage sludge concentrations from EU Member States (ng TEQ/kg d.m.)

[13] Country

Austria

Denmark

Germany

Spain

Sweden

UK

Range Average

8.1–38 13.1

0.7–55 9.1

0.7–1,207 20–40

64

0.02–115 20

9–192

4.1.5 Sewage Sludge

In Austria and Germany, sewage sludge for application in agriculture has to be analyzed for PCDD/PCDF and comply with legal limit values. Both countries have established a maximum permissible concentration of 100 ng I-TEQ per kg dry matter for sewage sludge applied to agricultural land.Additional data were available from Denmark, Spain, and the UK. As can be seen from Table 15, in general, the concentrations ranged from below 1 ng I-TEQ/kg d.m. to around 200 ng I-TEQ/kg d.m, with levels in Germany reaching over 1,000 ng TEQ/kg d.m. Average concentrations of PCDD/PCDF in sewage sludge are quite similar for each country, lying between 10 and 40 ng I-TEQ/ kg d.m. These findings indicate that similar sources are responsible for the contamination in sludges from industrialized countries. Results, mainly from Germany and Sweden, revealed that “normal” effluents from households, especially from washing machines, could explain these results. Additional inputs can originate from dishwashers but also run-off from streets and from roofs. Industrial inputs, where untreated effluents enter the municipal sewer systems, can cause very high contamination in sewage sludges and, in such cases, more than 1,000 ng I-TEQ/kg d.m. have been detected [5]. 4.2 Human Exposure and Levels in Humans 4.2.1 Feedstuffs and Foods

Human exposure to background contamination with PCDD/PCDF is possible via several routes: – Inhalation of air and intake of particles from air, – Ingestion of contaminated soil,

142

H. Fiedler

– Dermal absorption, – Food consumption. In 1990, a WHO working group concluded that 90% of the daily dioxin intake (from background contamination) results from ingestion. Especially, foodstuffs of animal origin are responsible for the daily intake of approximately 2 pg TEQ/(kg bw · d).All other foodstuffs, especially the “non-fatty” ones, are of minor importance in terms of PCDD/PCDF intake. They are either of plant origin or do not have a high potential for bioaccumulation of lipophilic compounds. Due to many measures to reduce emissions of PCDD/PCDF into the environment, reduction of PCDD/PCDF contamination in food was observed.As a consequence, the daily intake via food decreased. Whereas in Germany in 1991, the average daily intake was 127.3 pg TEQ/d, the present daily intake for an average German adult is estimated to 69.6 pg TEQ/d. The strongest decline was observed for fish. In 1991, fish contributed for ca. 30% of the daily intake (same percentage as for dairy and meat products): today only 10% of the daily intake is due to fish. In 2001, the EU conducted a scientific assessment of PCDD/PCDF in food. A summary of the evaluation is presented in Table 16 [21]. 4.2.2 Results from Individual Studies

The UK Total Diet Study (TDS) provides information on dietary exposures of the general UK population to PCDD/PCDF. It covers a total of 121 categories of food and drink, which are assigned to one of twenty broad food groups. Besides providing concentrations of PCDD/PCDF in food, it also allows one to determine

Table 16. Summary of EU food data – means and 99% Confidence Intervals (CI). Concentra-

tions in I-TEQ. Animal food data are on fat basis; others are on fresh weight basis [21] Food group

Mean

CI (99%)

Range

Cereals and cereal products Eggs Fish and fish products Wild fish (marine, freshwater; and some farmed salmon) Freshwater fish (culture) Fruit and vegetables Meat and meat products Poultry Beef and veal Pork Game Others: liver Mixed meat Milk and milk products Milk as such Others

0.019 1.19 9.80 9.92

0.004–0.081 0.895–1.57 6.57 –14.6 6.34 –16.2

0.010–0.020 0.460–7.32 0.125–225 0.125–225

8.84 0.029 0.525 0.524 0.681 0.258 1.81 2.27 0.540 0.882 0.972 0.612

5.54 –14.1 0.014–0.063 0.387–0.712 0.355–0.774 0.499–0.929 0.174–0.381 0.403–8.15 1.12 –4.59 0.043–6.76 0.720–1.08 0.749–1.26 0.555–0.675

2.33 –27.9 0.004–0.090 0.130–3.80 0.370–1.40 0.380–1.10 0.130–3.80 0.970–1.97 0.950–3.29 0.270–0.760 0.260–3.57 0.260–3.57 0.300–1.50

143

Dioxins and Furans (PCDD/PCDF)

Table 17. Summary of estimated upper bound mean dietary exposures of all age groups to diox-

ins and dioxin-like PCBs in 1982, 1992 and 1997 (pg WHO-TEQ/kg bw · d) [22] Age Group

Estimated mean dietary exposure (pg WHO-TEQ/kg bw · day) 1982 1992 1997 PCDD/ PCB PCDD/ PCDD/ PCB PCDD/ PCDD/ PCB PCDD/ PCDF PCDF+ PCDF PCDF+ PCDF PCDF+ PCB PCB PCB

Toddlers (age): 1.5–2.5 2.5–3.5 3.5–4.5 (boys) 3.5–4.5 (girls)

15 12 11 11

7.9 6.6 5.9 5.8

23 19 17 17

5.0 4.2 3.7 3.7

2.6 2.1 1.9 1.9

7.5 6.3 5.6 5.6

2.6 2.3 2.1 2.1

2.6 2.2 1.9 1.9

5.1 4.4 4.0 4.0

Schoolchildren

5.6

3.0

8.6

2.0

1.0

3.0

1.2

1.0

2.2

Adults

4.6

2.6

7.2

1.6

0.9

2.5

0.9

0.9

1.8

Population average

4.7

2.7

7.5

1.5

0.9

2.4

1.0

0.8

1.7

Table 18. Summary of estimated upper bound high level dietary exposure of all age groups to

dioxins and dioxin-like PCBs in 1982, 1992 and 1997 (pg WHO-TEQ/kg bw · d) [22] Age group

Estimated high level dietary exposure (pg WHO-TEQ/kg bw · d) 1982 1992 1997 PCDD/ PCB PCDD/ PCDD/ PCB PCDD/ PCDD/ PCB PCDD/ PCDF PCDF+ PCDF PCDF+ PCDF PCDF+ PCB PCB PCB

Toddlers (age): 1.5–2.5 2.5–3.5 3.5–4.5 (boys) 3.5–4.5 (girls)

34 27 22 24

Schoolchildren

10

Adults

8.3

16 14 11 11

49 41 33 34

8.9 7.5 6.0 6.6

5.0 4.0 3.3 3.2

14 11 9.2 9.6

5.2 4.3 3.6 3.8

4.9 4.1 3.4 3.4

10 8.4 6.9 7.2

5.2 15

3.2

1.6

4.7

1.9

1.7

3.5

4.6 13

2.8

1.6

4.3

1.6

1.6

3.1

age-dependent exposures. The most recent study showed that, for the reference year 1997, the estimated average and high level dietary exposures of adults and schoolchildren via the total diet in 1997 were within the recommended WHO TDI of 1–4 pg TEQ/kg bw/day, but that toddlers are all at or above the upper end of the range (28% above for the average toddler and 2-fold exceedance for highlevel toddlers). The TDI also showed that exposures for all age groups have declined substantially since 1982 (Table 17 and Table 18) [22]. Santillo et al. analyzed butter samples randomly collected from 24 countries; results are given for WHO- and I-TEQ including the data for mono-ortho-sub-

144

H. Fiedler

Table 19. PCDD/PCDF and dioxin-like PCB in butter from countries [23]

Country

WHO-TEQ

I-TEQ PCDD/PCDF

PCDD/PCDF

Mono-o-PCB

co-PCB

S TEQ

Philippines New Zealand South Africa USA (G. Lakes) Thailand Japan Canada Brazil Mexico Australia USA (East Coast) Sweden Denmark UK Israel Argentina Austria China India Germany Italy Netherlands Czech Republic Tunisia Spain

0.11 0.06 0.21 0.27 0.22 0.40 0.35 0.28 0.50 0.56 0.54 0.20 0.52 0.75 0.50 0.36 0.55 1.01 0.79 0.58 1.03 1.46 0.66 0.91 4.80

0.01 0.07 0.17 0.13 0.21 0.20 0.21 0.39 0.21 0.19 0.23 0.51 0.37 0.35 0.49 0.92 0.99 0.65 1.01 1.51 1.14 0.85 1.80 2.58 0.74

0.06 0.06 0.08 0.08 0.11 0.09 0.14 0.10 0.09 0.09 0.11 0.22 0.16 0.15 0.26 0.21 0.22 0.13 0.25 0.28 0.26 0.39 0.46 0.28 0.17

0.18 0.19 0.46 0.48 0.54 0.69 0.70 0.77 0.80 0.84 0.88 0.93 1.05 1.25 1.25 1.49 1.76 1.79 2.05 2.37 2.43 2.70 2.92 3.77 5.71

0.11 0.05 0.17 0.22 0.19 0.34 0.28 0.23 0.43 0.44 0.45 0.18 0.44 0.57 0.43 0.31 0.45 0.90 0.69 0.51 0.87 1.25 0.59 0.75 4.61

Mean – this study Median – this study

0.70 0.52

0.64 0.39

0.18 0.15

1.52 1.05

0.62 0.44

5 Germany 1998: range, 4 samples Netherlands 1998 – single sample Spain 1998: mean of 8 brands Germany 1993–1996, mean, 222 samples

1.18–1.67

1.00–1.41

2.29

1.97 1.09

0.74

0.64

stituted and coplanar PCB (Table 19) [23]. The median of this survey was 0.51 pg WHO-TEQ for PCDD/PCDF only and 1.02 pg WHO-TEQ for the 28 dioxin-like PCDD/PCDF/PCB. Whereas the share from the coplanar and mono-orthosubstituted PCB in median was approximately 50%, the share varies from country to country; e.g., whereas in Sweden and the Czech Republic the share of PCDD/PCDF to the WHO-TEQ is “only” 22% or 23%, respectively, the PCDD/PCDF play a major role in countries such as Mexico (63%), Australia

Dioxins and Furans (PCDD/PCDF)

145

Fig. 5. Concentrations of PCDD/PCDF (pg I-TEQ/g fat) in breast milk from mothers in Ger-

many, years 1985–1998 (n=2438) [25]

(67%), USA-East coast (61%), and United Kingdom (60%). The unusually high concentration for the Spanish samples would give a share of 84% for the PCDD/PCDF. 4.2.3 Humans

The PCDD/PCDF pattern in humans may yield information as to different sources.Also, people from certain geographic regions may have specific patterns because of predominant exposures from different sources, e.g., Europeans have higher 2,3,4,7,8-Cl5DF concentrations compared to U.S. residents [24]. Breast milk is a frequently used monitor for human exposure and trends of PCDD/PCDF concentrations have been established in several countries (see chapters by Päpke and Fürst, this volume). In cooperation with the German Länder, the Federal Institute for Health Protection and of Consumers and Veterinary Medicine (bgvv) has established a database to compile results for organochlorine pesticides, PCB, and PCDD/PCDF. The annual means obtained from a total of 2,438 samples during the year 1985–1998 are shown in Fig. 5. It can be seen that the mean concentrations dropped by approximately 60% during ten years from ca. 30 pg I-TEQ/g fat to 12.9 pg I-TEQ/g fat [25].

5 Toxicity of PCDD/PCDF First risk assessments only focused on the most toxic congener, the 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-Cl4DD = 2,3,7,8-TCDD). Soon it was recognized, though, that all PCDD/PCDF substituted at least in position 2, 3, 7, or 8 are highly toxic and thus, major contributors to the overall toxicity of the dioxin mixture. In addition, despite the complex composition of many PCDD/PCDF con-

146

H. Fiedler

taining “sources”, only congeners with substitutions in the lateral positions of the aromatic ring, namely the carbon atoms 2, 3, 7, and 8, persist in the environment and accumulate in food-chains. 5.1 Toxicity

PCDD and PCDF produce a spectrum of toxic effects in animals; however, most information is available on 2,3,7,8-Cl4DD only. Most toxicity data on 2,3,7,8Cl4DD result from high-dose oral exposures to animals. There is a wide range of difference in sensitivity to PCDD lethality in animals. The signs and symptoms of poisoning with chemicals contaminated with Cl4DD in humans are similar to those observed in animals. Dioxin exposures to humans are associated with an increased risk of severe skin lesions (chloracne and hyperpigmentation), altered liver function and lipid metabolism, general weakness associated with drastic weight loss, changes in activity of various liver enzymes, depression of the immune system, and endocrine and nervous system abnormalities. 2,3,7,8-Cl4DD is a potent teratogenic and fetotoxic chemical in animals and a potent promoter in rat liver carcinogenesis. 2,3,7,8-Cl4DD also causes cancers of the liver and other organs in animals (see below). 5.1.1 Mode of Action

The toxicity of 2,3,7,8-Cl4DD segregates with the cytosolic aryl (aromatic) hydrocarbon receptor (AhR), and the relative toxicities of other PCDD and PCDF congeners are associated with their ability to bind to the receptor, which occurs in all rodent and human tissues. The AhR binding affinities of 2,3,7,8Cl4DF, 1,2,3,7,8- and 2,3,4,7,8-Cl5DF are in the same order of magnitude as that observed for 2,3,7,8-Cl4DD. PCDDs with at least three lateral chlorine atoms bind with some affinity to the AhR. Current evidence is that most, if not all, biological effects of 2,3,7,8-Cl4DD and other PCDDs arise from an initial high affinity interaction with the AhR and it appears that the biochemical and toxicological consequences of PCDF exposure are the result of a similar mode of action. It is generally believed that the toxic effects of 2,3,7,8-substituted PCDD and 2,3,7,8-substituted PCDF exhibit the same pattern of toxicity. The toxic responses are initiated at the cellular level, by the binding of PCDD/PCDF to a specific protein in the cytoplasm of the body cells, the aryl hydrocarbon receptor (AhR). 2,3,7,8-substituted PCDD/PCDF bind to the Ah receptor and induce CYP1A1 (cytochrome P450 1A1) and CYP1A2 (cytochrome P450 1A1) gene expression. The binding to the Ah receptor constitutes a first and necessary step to initiate the toxic and biochemical effects dioxins, although it is not sufficient alone to explain the full toxic effects. This mechanism of action of 2,3,7,8-Cl4DD parallels in many ways that of the steroid hormones, which have a broad spectrum of effects throughout the body and where the effects are caused primarily by the parent compound. However, TCDD and steroid hormone receptors (e.g., estro-

Dioxins and Furans (PCDD/PCDF)

147

gen, androgen, glucocorticoid, thyroid hormone, vitamin D3, and retinoic acid receptors) do not belong to the same family. Ah-receptor-binding affinities of 2,3,7,8-Cl4DF, 1,2,3,7,8- and 2,3,4,7,8-Cl5DF are of the same order of magnitude as that observed for 2,3,7,8-TCDD. With increasing chlorination, receptor-binding affinity decreases. The induction of the cytochrome P450 1A1 enzyme is frequently used as a convenient biomarker for PCDD/PCDF and other dioxin-like compounds. 5.1.2 Carcinogenicity

A Working Group for IARC (International Agency for Research on Cancer, Lyon, France) classified 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-Cl4DD or 2,3,7,8TCDD) as a multi-site carcinogen in animals as well as in humans [26]. For the classification of carcinogenicity in humans, the most important epidemiological studies for the evaluation of 2,3,7,8-Cl4DD are four cohort studies of herbicide manufacturers (one each in the United States and the Netherlands, two in Germany) and for the cohort of residents from Seveso, Italy. The relative risk for all cancers combined in the most highly exposed and longer-latency sub-cohorts is 1.4. In these cohorts, the blood lipid 2,3,7,8-Cl4DD levels estimated to the last time of exposure were 2,000 ng/kg (mean) (up to 32,000 ng/kg) in the United States cohort, of-the-art 434 ng/kg geometric mean (range, 301–3,683 ng/kg) among accident workers in the Dutch cohort, of-the-art 008 ng/kg geometric mean in the group of workers with severe chloracne in the BASF accident cohort in Germany and measurements up to 2,252 ng/kg in the Boehringer cohort in Germany. These calculated blood 2,3,7,8-Cl4DD levels in workers at time of exposure were in the same range as the estimated blood levels in a two-year rat carcinogenicity study. These studies involve the highest exposures to 2,3,7,8-Cl4DD. Although better known, the exposures at Seveso were lower and the follow-up shorter than those in the industrial settings. Most of the four industrial cohorts include analyses of sub-cohorts considered to have the highest exposure and/or longest latency. Overall, the strongest evidence for the carcinogenicity of 2,3,7,8-Cl4DD was for all cancers combined, rather than for any specific site and for more than one year exposure and with a 20 year latency period. On the basis of these studies, IARC concluded that there is limited evidence in humans for the carcinogenicity of 2,3,7,8-Cl4DD. There was inadequate evidence in humans for the carcinogenicity of PCDD other than 2,3,7,8-Cl4DD. For PCDF, two incidents, each involving about 2,000 cases occurred in which people were exposed to sufficient PCB and PCDF to produce symptoms (Yucheng and Yusho accidents). Fatal liver disease was 2–3 times more frequent than national rates in both cohorts. In the Yusho cohort from Japan, after 22 years, there was a three-fold excess of liver cancer mortality in men, which was already detectable and even higher at 15 years of follow-up. In the Yucheng cohort, Taiwan, after 12 years of follow-up, there was no excess of liver cancer mortality. Based upon these data, IARC concluded that there was inadequate evidence in humans for the carcinogenicity of PCDF.

148

H. Fiedler

2,3,7,8-Cl4DD causes liver tumors in animals at lower concentrations than any other man-made chemical. Dioxins and furans are not genotoxic (i.e., do not initiate cancer development), but 2,3,7,8-Cl4DD and other PCDD and PCDF are strong promoters of tumor development. 2,3,7,8-Cl4DD interferes with several functions that probably influence the tumor promotion process, such as growth factors, hormone systems, oxidative damage, intercellular communication, cell proliferation (division and growth), apoptosis (cell death), immune surveillance, and cytotoxicity (cellular toxicity). In laboratory animals, male and female rats and mice constantly showed an increase in the incidence of liver tumors after administration of 2,3,7,8-Cl4DD. In rats exposed to 100 ng/kg bw 2,3,7,8-Cl4DD per day, hepatocellular carcinomas and squamous-cell carcinomas of the lung were observed. Estimated blood levels were 5,000–10,000 ng/kg 2,3,7,8-Cl4DD. In addition, tumors were increased at several other sites in rats, mice and Syrian hamsters, but these effects were dependent upon the species, sex, and route of administration of 2,3,7,8-Cl4DD. Although the doses resulting in increased tumor incidence in rodents are extremely low, they are very close to doses that are toxic in the same species. These data led to the conclusion that there is sufficient evidence in experimental animals for the carcinogenicity of 2,3,7,8-Cl4DD. Further, evaluation of much smaller databases led to the conclusion that there is limited evidence in experimental animals for the carcinogenicity of a mixture of 1,2,3,6,7,8- and 1,2,3,7,8,9-Cl6DD and that there was inadequate evidence for the carcinogenicity in experimental animals of 2,7-Cl2DD, 1,2,3,7,8-Cl5DD and 1,2,3,4,6,7,8-Cl6DD [26]. There are no long-term carcinogenicity studies on PCDF, but some tumor promotion studies were evaluated in which rats and mice were exposed to some of the congeners following short duration exposure to known carcinogens. IARC concluded that there is inadequate evidence in experimental animals for the carcinogenicity of 2,3,7,8-Cl4DF, but there is limited evidence in experimental animals for the carcinogenicity of 2,3,4,7,8-Cl5DF and 1,2,3,4,7,8-Cl6DD. The limited carcinogenicity data available for congeners other than 2,3,7,8Cl4DD indicate that carcinogenic potency is also proportional to AhR affinity. Based on this evidence, all PCDD and PCDF are concluded to act through a similar mechanism and require an initial binding to the AhR. Binding of 2,3,7,8Cl4DD to the AhR results in transcriptional activation of a battery of 2,3,7,8Cl4DD-responsive genes, but currently no responsive gene has been proven to have a definitive role in its mechanism of carcinogenesis. Taking all of the evidence into consideration, the following evaluations were made by IARC in 1997 [26]: – 2,3,7,8-Tetrachlorodibenzo-p-dioxin (2,3,7,8-Cl4DD) is carcinogenic to humans (Group 1). – Other polychlorinated dibenzo-p-dioxins are not classifiable as to their carcinogenicity to humans (Group 3). – Dibenzo-p-dioxin is not classifiable as to its carcinogenicity to humans (Group 3). – Polychlorinated dibenzofurans are not classifiable as to their carcinogenicity to humans (Group 3).

Dioxins and Furans (PCDD/PCDF)

149

In its recent Dioxin Reassessment, US-EPA basically follows the IARC classifications and concludes that “under EPA’s current approach, TCDD is best characterized as a “human carcinogen.” [27]. In view of the results mentioned above, it should be noted that the present background levels of 2,3,7,8-Cl4DD in human populations (2–3 ng/kg) are 100to 1,000-times lower than those observed in this rat carcinogenicity study. Evaluation of the relationship between the magnitude of the exposure in experimental systems and the magnitude of the response (i.e., dose-response relationships) does not permit conclusions to be drawn on the human health risks from background exposures to 2,3,7,8-Cl4DD [26]. 5.1.3 Toxic Effects in Humans Other than Carcinogenicity

In humans, effects associated with exposure to dioxins are mainly observed in accidental and occupational exposure situations. There is a number of cohorts with high exposure to PCDD/PCDF (and PCB), e.g., NIOSH (National Institute of Occupational Health and Safety, USA) and Boehringer occupational studies, veterans of Operation Ranch Hand in Vietnam, residents of Seveso, etc. Seveso residents had high levels of 2,3,7,8-Cl4DD and one effect that has been observed recently were that significantly more girls were born than boys (change in normal sex ratio). Although the number of births was relatively few for seven years post-exposure, the sex ratio is altered; other sites, e.g., Ufa in Russia, have been examined for the effect of 2,3,7,8-Cl4DD exposures on sex ratio with mixed results, but with smaller numbers of offspring [27]. Other toxic effects include an increased prevalence of diabetes (Ranch Hands cohort) and increased mortality due to diabetes and cardiovascular diseases have been reported. In children exposed to PCDD/PCDF and/or PCB in the womb, effects on neurodevelopment and neurobehavior (object learning) and effects on thyroid hormone status have been observed at exposures at or near background levels. At higher exposures, children exposed transplacentally to PCDD/PCDF and PCB show skin defects, developmental delays, low birth-weight, behavior disorders, decrease in penile length at puberty, reduced height among girls at puberty and hearing loss. It is not totally clear to what extent dioxin-like compounds are responsible for these effects, when considering the complex chemical mixtures to which human individuals are exposed. However, it has been recognized that subtle effects might already be occurring in the general population in developed countries, at current background levels of exposure to dioxins and dioxin-like compounds and, due to the high persistence of the dioxin-like compounds, the concentrations in the environment, as well as in food, will only decrease slowly. For humans, chronic effects are of greater concern than acute toxicity. Amongst the most sensitive endpoints are reproductive, developmental, immunotoxic and neurotoxic effects. From these results obtained in high-exposure groups, it seems unlikely that clinically observable health effects will be found in the general adult population [24].

150

H. Fiedler

5.1.4 Toxic Effects in Laboratory Animals Other than Carcinogenicity

The extraordinary potency of 2,3,7,8-Cl4DD (tetrachlorodibenzo-p-dioxin) and related 2,3,7,8-substituted PCDD and PCDF has been demonstrated in many animal species. They elicit a broad spectrum of responses in experimental animals such as: liver damage (hepatoxicity); suppression of the immune system (immunotoxicity); formation and development of cancers (carcinogenesis); abnormalities in fetal development (teratogenicity); developmental and reproductive toxicity; skin defects (dermal toxicity); diverse effects on hormones and growth factors; and induction of metabolizing enzyme activities (which increases the risk of metabolizing precursor chemicals to produce others which are more biologically active). In all mammalian species tested so far, lethal doses of 2,3,7,8-Cl4DD result in delayed death preceded by excessive body weight loss (“wasting”). Other signs of 2,3,7,8-Cl4DD intoxication include thymic atrophy, hypertrophy/hyperplasia of hepatic, gastrointestinal, urogenital and cutaneous epithelia, atrophy of the gonads, subcutaneous oedema and systemic hemorrhage. The lethal dose of 2,3,7,8Cl4DD varies more than 5,000-fold between the guinea-pig (LD50 =1 mg/kg bw), the most sensitive, and the hamster, the least sensitive species. In tissue culture, 2,3,7,8-Cl4DD affects growth and differentiation of keratinocytes, hepatocytes and cells derived from other target organs. Toxicity of 2,3,7,8-Cl4DD segregates with the Ah receptor, and relative toxicity of other PCDD congeners is associated with their ability to bind to this receptor. PCDD cause suppression of both cell-mediated and humoral immunity in several species at low doses. PCDD have the potential to suppress resistance to bacterial, viral and parasitic challenges in mice. Kinetics: In most vertebrate species, the 2,3,7,8-substituted PCDD and PCDF congeners are predominantly retained, in other words, if chlorine atoms are present on all 2, 3, 7, and 8 positions; the biotransformation rate of PCDD/PCDF is strongly reduced, resulting in significant bioaccumulation. In most species the liver and adipose tissue are the major storage sites. Although the parent PCDD/PCDF congeners cause the biological effects, biotransformation to more polar metabolites should be considered to be a detoxification process. Oxidation by cytochrome P450 primarily occurs at the 4 and 6 positions in the molecule and the presence of chlorine atoms at these positions reduces metabolism more than substitution at the 1 and 9 positions. The half-lives of especially the PCDF in humans are much longer than those in experimental animals. 2,3,7,8-Cl4DD is both a developmental and reproductive toxicant in experimental animals. The developing embryo/fetus appears to display enhanced sensitivity to the adverse effects. Whereas perturbations of the reproductive system in adult animals require high toxic doses, effects on the developing organism occur at doses >100-times lower that those required in the mother. Sensitive targets include the developing reproductive, nervous and immune systems. Perturbation of multiple hormonal systems and their metabolism due to PCDD exposure may play a role in these events.

151

Dioxins and Furans (PCDD/PCDF)

6 Risk Assessment and Risk Management 6.1 Risk Assessment and the TEF Approach

First risk assessments only focused on the most toxic congener, the 2,3,7,8-Cl4DD. Soon it was recognized, though, that all PCDD/PCDF substituted at least in position 2, 3, 7, or 8 are highly toxic and thus, major contributors to the overall toxicity of the dioxin mixture. In addition, despite the complex composition of many PCDD/PCDF containing “sources”, only congeners with substitutions in the lateral positions of the aromatic ring, namely the carbon atoms 2, 3, 7, and 8, persist in the environment and accumulate in food-chains. For regulatory purposes so-called Toxicity Equivalency Factors (TEF) have been developed for risk assessment of complex mixtures of PCDD/PCDF [28]. The TEF are based on acute toxicity values from in vivo and in vitro studies. This approach is based on the evidence that there is a common, receptor-mediated mechanism of action for these compounds. Although the scientific basis cannot be considered as solid, the TEF approach has been adopted as an administrative tool by many agencies and allows converting quantitative analytical data for individual PCDD/PCDF congeners into a single Toxic Equivalent (TEQ). As TEFs are interim values and administrative tools, they are based on present state of knowledge and should be revised, as new data becomes available. Today’s most commonly applied TEFs were established by a NATO/CCMS Working Group on Table 20. International Toxicity Equivalency Factors (I-TEFs) for PCDD/PCDF (NATO/CCMS 1988) and WHO-TEFs for PCDD/PCDF [29, 31]

Congener

2,3,7,8-Cl4DD 1,2,3,7,8-Cl5DD 1,2,3,4,7,8-Cl6DD 1,2,3,7,8,9-Cl6DD 1,2,3,6,7,8-Cl6DD 1,2,3,4,6,7,8-Cl7DD Cl8DD 2,3,7,8-Cl4DF 1,2,3,7,8-Cl5DF 2,3,4,7,8-Cl5DF 1,2,3,4,7,8-Cl6DF 1,2,3,7,8,9-Cl6DF 1,2,3,6,7,8-Cl6DF 2,3,4,6,7,8-Cl6DF 1,2,3,4,6,7,8-Cl7DF 1,2,3,4,7,8,9-Cl7DF Cl8DF

I-TEF

1 0.5 0.1 0.1 0.1 0.01 0.001 0.1 0.05 0.5 0.1 0.1 0.1 0.1 0.01 0.01 0.001

WHO-TEF Humans/Mammals

Fish

Birds

1 1 0.1 0.1 0.1 0.01 0.0001 0.1 0.05 0.5 0.1 0.1 0.1 0.1 0.01 0.01 0.0001

1 1 0.5 0.01 0.01 0.001 – 0.05 0.05 0.5 0.1 0.1 0.1 0.1 0.01 0.01 0.0001

1 1 0.05 0.01 0.1 <0.001 – 1 0.1 1 0.1 0.1 0.1 0.1 0.01 0.01 0.0001

For all non-2,3,7,8-substituted congeners, no TEF has been assigned.

152

H. Fiedler

Table 21. TEFs for PCB [29, 31]

Congener

Humans/mammals

3,4,4¢,5-TCB (81) 3,3¢,4,4¢-TCB (77) 3,3¢,4,4¢,5-PeCB (126) 3,3¢,4,4¢,5,5¢-HxCB (169) 2,3,3¢,4,4¢-PeCB (105) 2,3,4,4¢,5-PeCB (114) 2,3¢,4,4¢,5-PeCB (118) 2¢,3,4,4¢,5-PeCB (123) 2,3,3¢,4,4¢,5-HxCB (156) 2,3,3¢,4,4¢,5¢-HxCB (157) 2,3¢,4,4¢,5,5¢-HxCB (167) 2,3,3¢,4,4¢,5,5¢-HpCB (189)

0.0001 0.0001 0.1 0.01 0.0001 0.0005 0.0001 0.0001 0.0005 0.0005 0.00001 0.0001

Fish 0.0005 0.0001 0.005 0.00005 <0.000005 <0.000005 <0.000005 <0.000005 <0.000005 <0.000005 <0.000005 <0.000005

Birds 0.1 0.05 0.1 0.001 0.0001 0.0001 0.00001 0.00001 0.0001 0.0001 0.00001 0.00001

Dioxins and Related Compounds as International Toxicity Equivalency Factors (I-TEF) [28]. However, in 1997, a WHO/IPCS (World Health Organization/Intergovernmental Programme on Chemical Safety) working group re-evaluated the I-TEFs and established a scheme, which besides human mammalian TEFs also established TEFs for birds and fish (Table 20). The same expert group also assessed the dioxin-like toxicity of PCB and assigned TEF values for 12 coplanar and mono-ortho-substituted PCB congeners (Table 21) [29]. It should be noted that most existing legislation and most assessments still use the I-TEF scheme. However, the recently agreed Stockholm Convention on POPs (persistent organic pollutants, see reference [30]) refers to the combined WHOTEFs as the starting point as a reference. 6.2 Risk Assessment by US-EPA

The final assessment of risks posed by dioxins – release date supposed to be early 2002 – may include the following (BNA 2001): – 2,3,7,8-tetrachlorodibenzo-p-dioxin is a human carcinogen; – mixtures of dioxin-like compounds are likely human carcinogens; – the general public’s exposure to ambient levels of dioxins may cause up to one case of cancer for 1000 people exposed; however, the true risks are likely less than that and may be zero; – there does not appear to be a “threshold” or safe dose, of dioxins that does not cause toxic (non-cancer) effects; – U.S. residents are exposed daily to about one picogram dioxins per kilogram body weight, meaning their exposure is close to the level that caused biological changes in animals; – emissions of dioxins have declined more than 80 percent since 1987; and – open burning of household waste is one of the largest known, non-regulated sources of dioxins.

Dioxins and Furans (PCDD/PCDF)

153

6.3 Risk Management

As PCDD and PCDF have never been produced intentionally, their production and use cannot be regulated by chemical legislation and a prohibition of production. Indirect measures have to be taken by, e.g., ban of production and use of chemicals that are known to be contaminated with PCDD/PCDF and measures to reduce emissions into the environment from known sources of dioxins and furans. All efforts aim to minimize exposures of the environment and humans to PCDD/PCDF. There are several options for such action and most of them have been used in the past: legally binding instruments, guidelines, and recommendations. In addition, industries have committed to changes in processes, to use cleaner input materials or to set maximum concentrations of PCDD/PCDF for their materials. International organizations such as WHO to establish a TDI (Tolerable Daily Intake) or intergovernmental institutions such as the EC (European Community) hold expert consultations and came out with recommendations or legislation (e.g., on food and feedstuffs). For sensitive uses such as soil and important foodstuffs, such as dairy products, guidelines and recommendations have been established by various authorities. The proposed measures and guidelines are recommendations for action and in most cases are not legally binding. Nevertheless, they are a basis for political decisions to protect human health and the environment. In some cases, e.g., accidents such as a fire at a plastic store, or for marketing purposes, these recommendations for actions were used for decision making. A different situation exists in cases of emergency responses, e.g., some of the feed and food contaminations, which occurred during the last years especially in Europe. 6.3.1 Tolerable Intakes

Scientists today agree that the major pathway of human exposure to PCDD/PCDF accounting for >95% of the human intake is via ingestion of food. Uptake of water and soil (toddlers!), inhalation, and dermal contact are of minor concern. Different international expert groups have performed health risk assessments of dioxins and related compounds.A Nordic expert group (for five Scandinavian countries) proposed a tolerable daily intake (TDI) for 2,3,7,8-Cl4DD and structurally similar chlorinated PCDD and PCDF of 5 pg/kg body weight (bw), based on experimental studies on cancer, reproduction and immunotoxicity. Germany used a tried approach and recommended 1 pg I-TEQ/kg bw·d as a desirable target to be achieved in the long-term and that actions should be taken if the daily exposure exceeds 10 pg I-TEQ/kg bw and day. In the USA,ATSDR (Agency for Toxic Substances and Disease Register) has set a minimal risk level (MRL2) for dioxins pf 1 picogram per kilogram body weight 2

An MRL is an estimate of the daily human exposure to a hazardous substance that ATSDR thinks would not cause harm.

154

H. Fiedler

[33]. The U.S. federal government has made the following recommendations to protect human health [34]: – The EPA has set a limit of 0.00003 micrograms of 2,3,7,8 Cl4DD per liter of drinking water (0.00003 mg/L). – Discharges, spills, or accidental releases of 1 pound or more of 2,3,7,8-Cl4DD must be reported to EPA. – The Food and Drug Administration (FDA) recommends against eating fish and shellfish with levels of 2,3,7,8-Cl4DD greater than 50 parts per trillion (50 ppt). A first World Health Organization (WHO) meeting, in 1990, established a TDI of 10 pg/kg bw for 2,3,7,8-Cl4DD, based on liver toxicity, reproductive effects and immunotoxicity, and making use of kinetic data in humans and experimental animals. Since then new epidemiological and toxicological data have emerged, in particular with respect to neuro-developmental and endocrinological effects. In May 1998, a joint WHO-ECEH (World Health Organization-European Centre for Environmental Health) and International Programme on Chemical Safety (IPCS) expert group re-evaluated the old TDI and came out with a new TDI (which is a range) of 1–4 pg TEQ/kg bw, which includes all 2,3,7,8-substituted PCDD and PCDF as well as dioxin-like PCB (for reference see the twelve PCB in Table 21). The TDI is based on the most sensitive adverse effects, especially hormonal, reproductive and developmental effects, which occur at low doses in animal studies; e.g., in rats and monkeys at body burdens in the range of 10–50 ng/kg bw. Human daily intakes corresponding with body burdens similar to those associated with adverse effects in animals were estimated to be in the range of 10–40 pg/kg bw·d. The 1998 WHO-TDI does not apply an uncertainty factor to account for interspecies differences in toxicokinetics since body burdens have been used to scale doses across species. However, the estimated human intake was based on Lowest Observed Adverse Effect Levels (LOAELs) and not on No Observed Adverse Effect Levels (NOAELs). For many endpoints humans might be less sensitive than animals, uncertainty still remains regarding animal to human extrapolations. Further, differences between animals and humans exist in the half-lives for the different PCDD/PCDF congeners. To account for all these uncertainties, a composite uncertainty factor of 10 was recommended. As subtle effects might already be occurring in the general population in developed countries at current background levels of exposure to dioxins and related compounds, the WHO expert group recommended that every effort should be made to reduce exposure to below 1 pg TEQ/kg bw · d [35, 36]. Since the WHO expert consultation has established the new TDI of 1–4 pg WHO-TEQ 3/kg bw · d, countries started to move towards this recommendation; for example, Japan established a TDI of 4 pg WHO-TEQ/kg bw · d as its environmental standard. In November 2000, the Scientific Committee on Food (SCF) of the European Commission established a target taking into account the current exposure situ3

Note: This TEQ includes seven 2,3,7,8-substituted PCDD, ten 2,3,7,8-substituted PCDF, four coplanar PCB and twelve mono-ortho-substituted PCB.

155

Dioxins and Furans (PCDD/PCDF)

ation and a recommended temporary tolerable weekly intake (t-TWI) of 7 pg 2,3,7,8-Cl4DD/kg body weight using the body weight approach was established. It was also concluded that the TEQ approach should be applied to include all 2,3,7,8-substituted PCDD/PCDF and dioxin-like PCB. Thus, the t-TWI of 7 pg TEQ/kg bw wk is applicable for these compounds (seven PCDD, ten PCDF and twelve dioxin-like PCB). The t-TWI is based on the most sensitive endpoints from animal studies, e.g., developmental and reproductive effects in rats and monkeys and endometriosis in monkeys [37]. It can reasonably be assumed that the exposure of the majority of the European population will be within the temporary tolerable weekly intake (t-TWI) for dioxins, furans and dioxin-like PCB. In order to achieve this target, the Council of the European Commission adopted legally binding limits for the presence of PCDD/PCDF in animal feed. Any feed or feed material exceeding these limits is excluded from the feed and food-chain. The Directive and the limits as displayed in Table 22 will enter into force on 1 July 2002. Table 22. Action and target concentrations for food and feedstuffs [38] (All concentrations are

for the sum of polychlorinated dibenzo-para-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF) expressed in World Health Organisation (WHO) toxic equivalents, using the WHO-TEFs 1997) Feedings stuffs

Maximum content relative to a feeding stuff with a moisture content of 12%

All feed materials of plant origin including vegetable oils and by-products

0.75 ng WHOPCDD/PCDF-TEQ/kg a, b

Minerals

1.0 ng WHOPCDD/PCDF-TEQ/kg a, b

Animal fat, including milk fat and egg fat

2.0 ng WHOPCDD/PCDF-TEQ/kg a, b

Other land animal products including milk and milk products and eggs and egg products

0.75 ng WHOPCDD/PCDF-TEQ/kg a, b

Fish oil

6–0 ng WHOPCDD/PCDF-TEQ/kg a, b

Fish, other aquatic animals, their products and by-products with the exception of fish oil c

1.25 ng WHOPCDD/PCDF-TEQ/kg a, b

Compound feeding stuffs, with the exception of feeding stuffs for fur animals, of feeding stuffs for fish and of feeding stuffs for pet animals

0.75 ng WHOPCDD/PCDF-TEQ/kg a, b

Feeding stuffs for fish, feeding stuffs for pet animals

2.25 ng WHOPCDD/PCDF-TEQ/kg a, b

a

b

c

Upper-bound concentrations; upper-bound concentrations are calculated assuming that all values of the different congeners less than the limit of detection are equal to the limit of determination. These maximum limits shall be reviewed for the first time before 31 December 2004 in the light of new data on the presence of dioxins and dioxin-like PCB, in particular with a view to the inclusion of dioxin-like PCB in the levels to be set and will be further reviewed before 31 December 2006 with the aim of significantly reducing of the maximum levels. Fresh fish directly delivered and used without intermediate processing for the production of feeding stuffs for fur animals is exempted from the maximum limit. The products, processed animals proteins produced from these fur animals cannot enter the food chain and the feeding thereof is prohibited to farmed animals which are kept, fattened or bred for the production of food.

156

H. Fiedler

In May 2001, the Scientific Committee on Food (SCF) of the European Commission revised their earlier recommendation based on new scientific information and instead recommended a tolerable weekly intake (TWI) of 14 pg WHOTEQ/kg · bw for PCDD, PCDF and dioxin-like PCB [39]. The earlier designation “temporal” was removed. The SCF stressed that, given the average dietary intakes of dioxins and dioxin-like PCBs in the European countries of 1.2–4.0 pg/kg · bw · d, a considerable proportion of the European population would still exceed the TWI derived by the Committee. Therefore, it was concluded that the considerations on risk characterization, risk management strategies and recommendations of the previous assessment of November 2000 were still valid (SCF 2000). Most recently, on 4–14 June 2001, the Joint FAO/WHO Expert Committee on Food Additives (JECFA) held its 57th meeting in Rome [40]. The Committee decided to express tolerable intakes as monthly values due to the long half-lives of PCDD, PCDF, and dioxin-like PCB. Thus, a monthly-based period would be a much more appropriate period to better reflect the average intakes as daily ingestion has a small or even negligible effect on overall exposure. A provisional tolerable monthly intake (PTMI ) of 70 pg/kg · bw · month was finally chosen as midpoint of two studies: data by Ohsako et al. [41] would results in a TMI of 100 pg/kg · bw · month, whereas the data by Faqi et al. [42] would result in a TMI of 40 pg/kg · bw · month. Similar to the other evaluation, the TEQ includes PCDD, PCDF, and dioxin-like PCB [40]. When compared to adults, breast-fed infants are exposed to higher intakes of PCDD, PCDF and PCB on a body weight basis, although for a limited time only. Although no adverse health effects could be causally linked so far with background exposures of PCDD/PCDF in human milk, for reasons of preventive health care, the relatively high exposure of breast-fed infants must still be considered a matter of concern. Analyses of more than 1000 individual human milk samples from nursing mothers in North-Rhine Westphalia (Germany) revealed that the mean PCDD/PCDF concentration decreased from 34 pg I-TEQ/g milkfat in 1989 to 14.2 pg I-TEQ/g milkfat in 1996. Despite this decline of 60%, the PCDD/PCDF daily intake for babies is 68 pg I-TEQ/kg bw · d, which is almost 70-fold above the TDI of 1 pg TEQ/(kg bw · d) for an adult. Despite the higher exposure to contaminants, WHO as well as many other agencies noted the beneficial effects associated with breast-feeding and therefore promote and support breast-feeding. Further, the subtle effects detected in infants were associated with transplacental rather than lactational exposure [35]. 6.3.2 Regulation of Chemicals

As polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans have never been produced intentionally on an industrial scale or for any commercial application, the production of PCDD/PCDF cannot be prohibited or phased out by law. Therefore, indirect measures have been taken to reduce new inputs of PCDD/PCDF into the environment. The first laws addressed the ban of chemicals known to be contaminated with PCDD/PCDF (ppb to ppm-range I-TEQ): as a consequence most industrialized countries banned the production and use of

157

Dioxins and Furans (PCDD/PCDF)

Table 23. Limit values of PCDD/PCDF as given by the German Chemicals Law [43] Note: the

concentrations given in Table 23 are absolute values, not I-TEQ! Congeners

Maximum concentrations

No. 1

2,3,7,8-Cl4DD, 1,2,3,7,8-Cl5DD, 2,3,7,8-Cl4DF, 2,3,4,7,8-Cl5DF

Sum of congeners under No. 1: 1 mg/kg

No. 2

1,2,3,4,7,8-Cl6DD, 1,2,3,7,8,9-Cl6DD, 1,2,3,6,7,8-Cl6DD, 1,2,3,7,8-Cl5DF, 1,2,3,4,7,8-Cl6DF, 1,2,3,7,8,9-Cl6DF, 1,2,3,6,7,8-Cl6DF, 2,3,4,6,7,8-Cl6DF

Sum of congeners under Nos. 1 and 2: 5 mg/kg

No. 3

1,2,3,4,6,7,8-Cl7DD, 1,2,3,4,6,7,8,9-Cl8DD, 1,2,3,4,6,7,8-Cl7DF, 1,2,3,4,7,8,9-Cl7DF, 1,2,3,4,6,7,8,9-Cl8DF

Sum of congeners under Nos. 1, 2 and 3: <100 mg/kg

polychlorinated biphenyls (PCB) and pentachlorophenol (PCP).Whereas PCB is banned in most industrialized countries, PCP is still being produced and used in several countries worldwide. In Germany, the First Ordinance on the Prohibition of Certain Chemicals has set stringent limit values for eight PCDD/PCDF in substances, preparations and articles placed on the market. This regulation was amended in 1994 and 1996 and the present ordinance sets limit values for all seventeen 2,3,7,8-chlorine substituted congeners as well as for eight of the lower brominated 2,3,7,8-bromine substituted dibenzo-p-dioxins and dibenzofurans (PBDD/PBDF). According to the law, substances, preparations, and/or articles are not allowed to be placed on the market [43]: – if the sum of the concentrations of the congeners listed under No. 1 of column 1 (Table 23) exceeds a value of 1 mg/kg, or – if the sum of the concentrations of the congeners listed in column 1 under Nos. 1 and 2 exceeds a value of 5 mg/kg, or – if the sum of the concentrations of the congeners listed in column 1 under Nos. 1, 2 and 3 (= all congeners substituted in 2,3,7,8-position), exceeds a value of 100 mg/kg. Similar regulations exist in the United States where in the Toxic Substances Control Act; maximum permissible concentrations for 2,3,7,8-Cl4DD were set. To stop the entry of dioxins and furans into the environment from use of socalled scavengers (e.g., dichloroethane or dibromoethane) as additives in leaded gasoline, a ban of the use of such scavengers was implemented in 1992 in Germany [44]. 6.3.3 Incineration and Combustion

As incineration of wastes was considered to be a major source of PCDD/PCDF emissions into the environment, legally binding concentrations for stack emissions have been established in several countries. Whereas first regulations only

158

H. Fiedler

included a requirement to minimize PCDD/PCDF emissions by applying Best Available Techniques (BAT), countries have moved towards defined limit values. Very often, these laws also include requirements on sampling, analysis, and reporting methods. In general, new incinerators have to comply immediately with these limits. For existing incinerators there are transient times between three and six years to comply with regulations. The Protocol on Persistent Organic Pollutants (=Aarhus Protocol) within the framework of the United Nations Economic Commission for Europe (UN-ECE) Convention on Long-range Transboundary Air Pollution (LRTAP) sets legally binding limit values for the emission of dioxins and furans of 0.1 ng I-TEQ/m3 for installations burning more than three tons per hour of municipal solid waste, 0.5 ng I-TEQ/m3 for installations burning more than 1 ton per hour of medical waste, and 0.2 ng I-TEQ/m3 for installations burning more than 1 ton per hour of hazardous waste [45]. For the 15 Member States of the European Union 4, a legally binding limit value of 0.1 ng I-TEQ/m3 for PCDD/PCDF was first set in 1994 for hazardous waste incinerators by Directive 94/67/EC [46]. At that time, countries like Austria, Belgium, Germany [47], Finland, France, Italy, the Netherlands, and Sweden went beyond this requirement and included municipal waste incinerators, sewage sludge, and hospital waste incinerators into their national laws. The remaining countries have the 0.1 ng I-TEQ/m3 limit as a guideline concentration. In December 2000, the European Commission adopted Directive EU Directive 2000/76/EC on the incineration on waste, which replaced the earlier Directive 94/67/EC and now established stringent operational conditions, technical requirements, and emission limit values for plants incinerating or co-incinerating waste within the Community [48]. The limit values set should prevent or limit as far as practicable negative effects on the environment and the resulting risks to human health; for PCDD/PCDF a legally binding limit value of 0.1 ng I-TEQ/Nm3 for PCDD/PCDF emissions to air was established. In addition, emission limit values for the discharge of wastewater from the cleaning of exhaust gases from incineration and co-incineration plants were established to limit a transfer of pollutants from the air into water. For PCDD/PCDF, the limit value is 0.3 ng I-TEQ/L. Pilot plants that treat less than 50 tons of waste per year are exempted from this regulation. Under the Directive, the term “incineration plant” means any stationary or mobile technical unit and equipment dedicated to the thermal treatment of wastes with or without recovery of the combustion heat generated. This includes the incineration by oxidation of waste as well as other thermal treatment processes such as pyrolysis, gasification or plasma processes in so far as the substances resulting from the treatment are subsequently incinerated. For municipal solid wastes, a minimum combustion temperature of 850 °C measured near the inner wall and for hazardous waste, which are all wastes than contain more than 1% of halogenated organics, a minimum combustion temperature of of-the-art 100 °C should be maintained for at least 2 seconds. The air limit values for PCDD/PCDF have to be controlled at least twice a year and every 3 months after start-up of a 4

Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain, Sweden, United Kingdom.

159

Dioxins and Furans (PCDD/PCDF)

new plant. Within two years, the member states of the European Union have to translate the requirements of this directive into national law. For new plants, the provisions of Directive 2000/76/EC shall enter into force on December 28, 2002; for existing plants the provisions shall apply as from December 28, 2005 [48]. In the USA, there is a limit value of 0.2 ng TEQ/m3 for new municipal and of 0.15 ng I-TEQ/m3 for new hazardous waste incinerators. In its “Law Concerning Special Measures against Dioxins”, which was approved and promulgated in July 1999 and went into effect in January 2000, Japan took a tiered approach (Table 24): new incinerators with a capacity greater than 4 t/h have to comply with a limit concentration of 0.1 ng I-TEQ/m3 (12% O2) whereas the limits for MSWI <2 t/h and MSWIs with a capacity between 2–4 t/h are 5 ng I-TEQ/m3 and 1 ng I-TEQ/m3, respectively [49]. In addition, measures to reduce PCDD/PCDF concentrations in incinerator residues, like fly ash and bottom ash, are addressed in the guideline as well. Some countries have issued additional laws: Austria: a limit value of 0.4 ng I-TEQ/m3 was set for sinter plants built after 1 January 2001 [50]. For iron and steel plants, there is a limit value of 0.25 ng ITEQ/m3 until 31 December 2005. Limit value of 0.1 ng I-TEQ/m3 from 1 January 2006. For electric arc furnaces and induction ovens the limit is at 0.4 ng ITEQ/m3. For existing plants, the latest date to comply with is five years after publication of the law [51].

Table 24. Japan – Emission standards for waste incinerators for emissions to air in ng WHOTEQ/m3 [49]

Waste incinerators (>50 kg/h) Incineration capacity

New facilities

Existing facilities Jan. 15, 2001– Nov. 30, 2002

More than 4 t/h 2 t/h – 4 t/h Less than 2 t/h

0.1 1 5

80

Dec. 1, 2002 1 5 10

Table 25. Japan – PCDD/PCDF limit values for installations other than municipal solid waste

[49] Kind of specified facility

Electric steel-making furnaces Steel industry: sintering processes Zinc recovery industries Aluminum production

New facilities

0.5 0.1 1 1

Existing facilities Jan. 15, 2001– Nov. 30, 2002

Dec. 1, 2002

20 2 40 20

5 1 10 5

160

H. Fiedler

Germany: in 1997, a dioxin limit value of 0.1 ng I-TEQ/m3 and a minimum temperature of 850 °C for crematories were set by law [52]. In Japan, emissions from the metal sector are regulated through the Air Pollution Control Law (Table 25) [49]. 6.3.4 Water Discharges and Solid Residues

Water discharges are regulated in Japan for various sectors (Table 26) and in the European Union [48]. Whereas in Japan the limit value in water discharges is set at 10 pg WHO-TEQ/L [49], the EU limit for water discharges from waste incineration plants is 0.3 ng I-TEQ/L 5 [48].

Table 26. Japan – Emission Standards for effluent (in pg WHO-TEQ/L) [49]

Specified Facilities a

New Existing facilities facilities b

– Facilities for bleaching using chlorine or chlorinated compounds for use in the manufacture of sulfate pulp (Kraft pulp) or sulfite pulp – Decomposition facilities for spent PCB or treated PCB – Washing facilities for PCB polluted materials or treated PCB

10

10

– Waste gas washing facilities or wet dust collectors of aluminum and aluminum alloy roasting furnaces, dissolving furnaces or drying furnaces – Washing facilities for ethylene dichloride used in the vinyl chloride monomer manufacturing

10 (20) c

– Waste gas washing facilities, wet dust collectors or ash landfill facilities for discharging polluted effluent of municipal solid waste incinerators (only those with capacity equal or higher than 50 kg/h) – Waste gas washing facilities, wet dust collectors or ash landfill facilities for discharging polluted effluent of industrial waste incinerators (only those with capacity equal or higher than 50 kg/h)

10 (50) c

– Sewage treatment plants that treat effluents from the facilities above – Facilities for treating effluents from the business establishments that set up the facilities above

10

a b c

5

The standard for effluent spillage from final waste disposal sites is 10 pg TEQ/L as provided in the order setting standards for maintenance and management of final waste disposal sites. Existing facilities will become subject to regulation from January 15, 2001. Standards in parentheses indicate a provisional effluent standard in use for 3 years after the enforcement of the Law.

Note: the Directive says “0.3 mg TEQ/L”, which has been corrected to “0.3 ng TEQ/L!

Dioxins and Furans (PCDD/PCDF)

161

6.3.5 Environmental Media 6.3.5.1 Soil and Soil Amendments

It is known that sewage sludge is contaminated with PCDD/PCDF. To reduce input of PCDD/PCDF through application of sewage sludge, Germany established a limit value of 100 ng I-TEQ/kg dry matter for sewage sludges used as fertilizer in agriculture, horticulture or forestry [53]. In addition, the law sets a freight limit for 5 tons of dry matter of sewage sludge per hectare once within three years. Law forbids application of sewage sludge on pasture. Similarly, Austria established maximum concentration of 50 ng I-TEQ/kg in fertilizers, soil additives, culture substrates, and plant additives. In addition, products containing 20–50 ng TEQ/kg have to be labeled with a warning sign “Attention contains dioxins/ furans” (forbidden for use on children’s playgrounds). Culture substrates are not allowed to contain more than 20 ng TEQ/kg [54]. In addition, the States of Oberösterreich (o.Ö.) and Niederösterreich (N.Ö.) published ordinances with a limit value of 100 ng I-TEQ/kg d.m. for sewage sludge. The guideline concentrations established by Germany, the Netherlands, and Sweden are compiled in Table 27 [5]. In Germany in 1992, reference values and recommended action for agricultural and horticultural land uses have been issued, which have been translated into governmental decrees in a number of Länder (Federal States in Germany) (Table 27 – upper part [55]): – For preventive reasons and as a long-term objective, the dioxin concentrations of soil used for agricultural purposes should be reduced to below 5 ng TEQ per kg;

Table 27. Soil guideline concentrations (concentrations in ng I-TEQ/kg d.m.)

Germany <5 5–40 >100 >1000 >10,000 The Netherlands 1 10

Target concentration Control of products if dioxin transfer Soil exchange on children playgrounds Soil exchange in residential areas Soil exchange independent of the location Agricultural farming Dairy farming

Sweden 10 250 Japan 1000

Sensitive uses Non-sensitive uses

162

H. Fiedler

– Cultivation of foodstuffs is not restricted in the case where the soil contains 5–40 ng per kg. However, critical land uses, e.g., grazing management, should be avoided if increased dioxin levels were found in foodstuffs grown on such soils; – Limitations on the cultivation of certain feedstuffs and foodstuffs might be necessary if the dioxin contamination were above 40 ng TEQ per kg soil. However, unlimited cultivation is allowed for plants with minimum dioxin transfer, e.g., corn. Guideline values were established for measures to be taken on children’s playgrounds and in residential areas [55]: – Remediation of contaminated soil is required in playgrounds if the soil contains more than 100 ng TEQ per kg. Remediation means sealing, decontamination or exchange of soil; – In residential areas, such action should be taken if the soil is contaminated with more than 1,000 ng TEQ per kg; – In industrial areas, the limit value was set to 10,000 ng TEQ per kg. In 1988, the Netherlands published a guideline for classification and remediation of soil [56]. Since its revision in 1994, there is a guideline level for PCDD/PCDF of 1 ng I-TEQ/kg d.m. for agricultural use and of 10 ng I-TEQ/kg d.m. for dairy farming (Table 27). In Sweden, there is a guideline concentration of 10 ng I-TEQ/kg d.m. for sensitive uses and of 250 ng I-TEQ/kg d.m. for less sensitive uses (Table 27) [5]. In Japan the Law Concerning Special Measures against Dioxins sets an environmental standard of 1,000 pg WHO-TEQ/g [49]. ATSDR has established a decision framework for sites contaminated with dioxins and dioxin-like substances. A screening level of 50 ppt TEQ is used to determine whether further site-specific evaluation is needed. A concentration of 1 ppb TEQ (=1,000 ng TEQ/kg soil) is used to determine the potential need for public health actions on a site-specific basis. Exceeding this concentration may result to interdict/prevent occurrence of exposure, such as surveillance, research, health studies, community education, physician education, or exposure investigations. Alternatively, based on the evaluation by the health assessor, none of these actions may be necessary. The concentration range greater than 50 ppt and smaller than 1,000 ppt includes the evaluation levels, where site-specific factors such as bioavailability, ingestion rates, pathway analysis, soil cover, climate, other contaminants, background exposure, etc. are being taken into account [34]. 6.3.5.2 Air and Water

Japan set an environmental air quality standard of <0.6 pg WHO-TEQ/m3 for ambient air and of <1 pg WHO-TEQ/L for water. Both standards represent annual averages [49].

Dioxins and Furans (PCDD/PCDF)

163

6.3.6 Food and Feedstuff Regulations

The Scientific Committee on Food concluded that although dioxin source reduction has been successfully accomplished in many European countries, a considerable proportion of the European population still exceeds the t-TWI. Therefore, further measures are needed to limit environmental releases of PCDD/PCDF and dioxin-like compounds [37]. The recent incidents of food and feed contamination have shown that present regulation is not existing or inadequate and a root cause analysis is required to develop appropriate monitoring, prevention and management. Setting feed and food limits alone will not prevent further accidents and there is no way to exclude the possibility of similar incidents to occur in the future unless specific measures are taken. However, regulatory levels would build the legal basis at least to eliminate products with extraordinary contamination levels from the market. Monitoring of the animal feed production chain could mitigate impacts and identify causes. In contrast to former dioxin cases, which mainly originated from high emissions of individual sources, the recent incidents have been caused by entry of contaminants more directly into the human food chain. Dealing with these accidents, there are mainly three distinct objectives to address. These require different approaches for assessment, prevention, monitoring and regulatory response [24]: – Identification and response to an emergency situation of an acute contamination (e.g., Belgian case); – Identification and seizure of products with exceptionally high levels (e.g., citrus pellets, choline chloride and Brandenburg cases), which can even effect the general population if used to a large extent in the feed and food chains. – Measures aiming to reduce exposure of the general population by ceasing feed ingredients, which are higher contaminated than comparable components (e.g., fish meal and fish oil from the Northern hemisphere). Each case should be carefully addressed and it should be recognized that solutions for one case will not necessarily provide an effective means for the others. 6.3.6.1 Guidelines for Milk and Milk Products

Germany, Ireland, The Netherlands, and the United Kingdom established guidelines for PCDD/PCDF concentrations and recommendations in milk. The guideline concentrations are summarized in Table 28. For Germany, the second report of the Joint Working Group contained guidelines and maximum values for milk and dairy products together with recommendations for action [57]. The limit values as given in Table 28 were derived as follows: – based on a TDI (total daily intake) value of 10 pg 2,3,7,8-Cl4DD/kg body weight and day, the maximum dioxin concentration in milk should not exceed 5.0 pg TEQ/g milkfat. Thus, milk and dairy products should not be out on the mar-

164

H. Fiedler

Table 28. Recommendation values and action levels for PCDD/PCDF in milk and milk products.

(Concentrations in ng I-TEQ/kg milk fat) [5, 57] Germany <0.9 >3.0 >5.0

Target concentration (minimum of dioxin input) 1. Identification and reduction of sources. If not possible within a short time Æ stop dairy farming 2. Recommendation not to market milk to end-user Milk and milk products are not allowed to be marketed

Ireland 5.0

Maximum level

The Netherlands 6.0

Maximum level

United Kingdom 0.7 16.6

Milk and milk products with less than 2% fat Maximum concentration for milk and milk products

ket if the dioxin contamination exceeds this value. However, from the existing data it is obvious, that a limit value of 5 pg TEQ/g milkfat would cut off only a few cases of extreme contamination, this limit value was considered not to have any consequences at all and would not improve the consumer’s situation; – to reduce the human impact via consumption of dairy products, a limit value of 3 pg TEQ/g fat was set. If such concentration were exceeded, the dioxin source should be identified and emission reducing measures taken. Moreover, it was recommended that direct supply to the consumer be stopped for milk and dairy products containing more than 3 pg TEQ/g fat; and – finally, an orientation value of 0.9 pg TEQ/g milkfat, based on the principle of precaution, was set. This concentration was derived from a TDI of 1 pg TEQ/kg body weight and day. The value of 0.9 pg TEQ/g milkfat can only be regarded as target value to be achieved as the data from Germany and other European countries showed that more than 50% of all dioxin concentrations in milk would exceed this value. To reach this target, it is necessary to further reduce the dioxin release into the environment.

7 Sources of PCDD/PCDF 7.1 Overview

Since the first overview on formation and sources of PCDD/PCDF was published in 1980 [58], several updates are available in the international literature. The findings can be summarized as follows [59]: – PCDD/PCDF were never produced intentionally but occur as trace contaminants in a variety of industrial and thermal processes;

165

Dioxins and Furans (PCDD/PCDF)

– Due to their chemical, physical and biological stability PCDD/PCDF are able to remain in the environment for long times. As a consequence dioxins from so-called “primary sources” (once formed in industrial or combustion processes) can be transferred to other matrices and enter the environment. Such “secondary” sources are sewage sludge/biosludge, compost, or contaminated soils and sediments; – Enzymatic reactions can dimerize chlorophenols to PCDD/PCDF. However, compared to chemical-industrial and combustion sources, biological formation seems to be negligible. 7.2 Primary Sources of PCDD/PCDF 7.2.1 Industrial-Chemical Processes

Primary sources of environmental contamination with PCDD/PCDF in the past were due to production and use of chloroorganic chemicals, including the pulp and paper industry. In wet-chemical processes the propensity to generate PCDD/PCDF during synthesis of chemical compounds decreases in the following order [1, 59]: chlorophenols>chlorobenzenes>aliphatic chlorinated compounds>inorganic chlorinated compounds. Factors favorable for the formation of PCDD/PCDF are high temperatures, alkaline media, presence of UV-light, and presence of radicals in the reaction mixture/chemical process [59, 60]. An overview on dioxin concentrations in chemicals is given in Table 29. As can be seen the concentrations can vary by several orders of magnitude. Changes in the industrial processes have resulted in reduction of PCDD/PCDF concentrations in the products: e.g. an estimate for Germany says that until 1990 about 105 g I-TEQ have been introduced through use of the dye pigment Violet 23

Table 29. PCDD/PCDF concentrations in chemical products

Substance

Concentration mg I-TEQ/kg

PCP PCP-Na PCB – Clophen A30 PCB – Clophen A60 2,4,6-Trichlorophenol Trichlorobenzene p-Chloranil (old process via chlorination of phenol) o-Chloranil (old process via chlorination of phenol) Hostaperm Violet RL Violet 23 Blue 106

up to 2,320 up to 450 11 2,179 680 0.023 376 63 1.2 19 56

166

H. Fiedler

(chloranil produced by old process as intermediate).Application of a new process via hydroquinone will reduce the annual input to about 3 g I-TEQ [61]. Emissions of PCDD/PCDF into the environment via water and to soils occur from kraft pulp and paper mills. The US-EPA inventory estimates annual emissions from kraft pulp and paper mills in the range of 20 g I-TEQ. In addition, PCDD/PCDF were detected in the final product (pulp, paper) as well as in the pulp and paper sludges. With advanced bleaching technology, the PCDD/PCDF contamination in effluents, products, and sludges was reduced [61]. In Germany there exist exclusively sulfite mills, which presently do not use molecular chlorine. Dioxin levels detected in German pulp were below 0.1 ng TEQ/kg d.m. The analysis of imported sulfate (Kraft) pulps gave concentrations in the range between 0.2 and 1.3 ng TEQ/kg d.m. Presently the import of Kraft pulps to Germany stands at three million tons, so that the total import of dioxins via kraft pulp is between 0.6 and 3.9 g I-TEQ. Dioxin levels in paper products from fresh fibers generally are less than 1 ng TEQ/kg d.m. In recycling paper, however, average dioxin concentrations are between 5 to 10 ng TEQ/kg [61]. As a consequence, products containing any of the above-mentioned chemicals are contaminated with PCDD/PCDF as well. Amongst these, wood treated with pentachlorophenol (PCP) or other chlorinated preservatives as well as PCB-based electric fluids are amongst those with the highest contaminations. Other PCPtreated materials include textiles, leather goods, and cork products. 7.2.2 Thermal Processes

Whereas in the past, the chemical industry and to a lesser extent the pulp and paper industry were considered to be the main source of dioxins and also the cause of today’s contaminated sites in many industrialized countries, today’s dioxin input is mainly due to thermal processes. The presence of PCDD/PCDF in the emissions and residues from municipal solid waste incinerators were detected first in 1997 in the MSWI in Amsterdam [62]. There is still a considerable focus on waste incineration but based on the requirements for dioxin reduction in stack gases set by several national authorities, the importance of this source category has declined during the last years. Examples can be seen especially in the European emission inventories (Table 30). For example, in Germany the annual input from municipal solid waste incineration (MSWI) via exhaust gases of about 400 g TEQ per year in 1988/89 was reduced to less than 4 g TEQ since 1997 [63].

Table 30. PCDD/PCDF trends in emissions from municipal solid waste incineration (MSWI)

MSWI of the 1970s MSWI around 1990 Modern MSWI

Concentration (ng I-TEQ/m3)

Emission factor (mg I-TEQ/t of waste burned)

Flux (mg I-TEQ/h)

50 5 0.1

2500 250 0.5

5 0.5 0.01

Dioxins and Furans (PCDD/PCDF)

167

The process by which PCDD/PCDF are formed during incineration are not completely understood nor agreed upon. Three possibilities have been proposed to explain the presence of dioxins and furans in incinerator emissions [64, 65]: – PCDD/PCDF are already present in the incoming waste – in Germany representative measurements gave about 50 ng I-TEQ/kg waste – and are incompletely destroyed or transformed during combustion. Not relevant for modern MSWIs. – PCDD/PCDF furans are produced from related chlorinated precursors (=predioxins) such as PCB, chlorinated phenols and chlorinated benzenes. – PCDD/PCDF are formed via de novo synthesis. That is, they are formed from the pyrolysis of chemically unrelated compounds such as polyvinyl chloride (PVC) or other chlorocarbons, and/or the burning of non-chlorinated organic matter such as polystyrene, cellulose, lignin, coal, and particulate carbon in the presence of chlorine donors. From the knowledge gained from MSWIs it can be concluded that PCDD/PCDF can be formed in other thermal processes in which chlorine-containing substances are burnt together with carbon and a suitable catalyst (preferably copper) at temperatures above 300 °C in the presence of excess air or oxygen. Preferentially PCDD/PCDF formation takes place in the zone when combustion gases cool down from about 450 °C to 250 °C (de novo synthesis) and not in the combustion chamber. Possible sources of the chlorine input are PVC residues as well as chloroparaffins in waste oils or even inorganic chloride [65]. These basic findings led to the establishment of the “Trace Chemistries of Fires” and later on it was verified in a variety of thermal processes that PCDD/PCDF were present in all emissions – (1) flue gases, (2) bottom ashes, fly ashes, (3) scrubber water [66]. Although all three of the above-mentioned possibilities can occur in large-scale operations, results showed that options (2) and (3) dominate over option (1). The smaller probability as indicated in option (1) is due to the fact that with modern combustion and flue gas cleaning technologies and due to thermodynamic reasons, PCDD/PCDF are destroyed when incinerated at temperatures above 800 °C and sufficient residence times (e.g., 2 s as legally required by some countries for MSW combustion). Today there is agreement that the most important pathway for formation of PCDD/PCDF is when the flue gases are transported down the cooling zone at temperatures between 250 and 450 °C [61, 66, 67]. Both, fly ash with its constituents – organic carbon, chlorides of alkali and earth alkali metals, metal activators, and catalysts – and dioxin/furan precursors in the gas phase play a role in the formation mechanism of PCDD/PCDF [68]. In addition, parameters such as oxygen, water vapor, and temperature have to be taken into account. In MSW incinerators the preferred location to generate PCDD/PCDF are economizer and equipment for dedusting, especially electrostatic precipitators [69–71]. Although much research was performed to study the formation of PCDD/PCDF in combustion processes, there is still no clear evidence, which mechanism is dominating and which parameters are important. There is some evidence that both, homogeneous reactions in the gas phase and heterogeneous reactions on surfaces of particles, play a role to form these thermodynamically

168

H. Fiedler

stable compounds [60, 64]. In addition, there are several indications that the mechanisms to generate PCDD/PCDF in the gas phase and in the particle phase are different. Within the following paragraphs, some key-parameters are briefly summarized, which have been identified to influence the formation of PCDD/PCDF in combustion processes [65]. Role of Temperature: Some early experiments were performed at high temperatures, when, e.g., Rubey et al. studied the thermal stability of PCB and the formation of PCDF. The experiments clearly showed that PCB (here: 2,3,4,4,5-pentachlorobiphenyl=2,3,4,4´,5-CB) are stable up to temperatures around 700 °C. With increasing temperatures, there is a decrease in the PCB concentration and an increase in PCDF formation. Preferentially, lower chlorinated PCDF (Cl4DF) were formed with a maximum at about 750 °C. Further increase of the temperatures results in destruction of the newly formed PCDF [72]. Experiments to study the temperature dependence typically range from 180–550 °C and the formation of PCDD/PCDF in the heterogeneous phase at long residence times. Using heated fly ash in a stream of air, Vogg and Stieglitz determined an optimum window for the de novo formation of PCDD/PCDF at temperatures 280–320 °C (Fig. 6) [68, 69, 73]. In subsequent experiments, Schwarz et al. found a second maximum around 400 °C for especially PCDF; for the PCDD the maximum was less pronounced (Fig. 7) [74]. As can be seen from Fig. 7, PCDD were less stable than PCDF at higher temperatures. Such results are confirmed from large scale operations, e.g., municipal waste incinerators, where more PCDF are present than PCDD [65]. Role of Temperature and Residence Time: For gas-phase reactions, temperature is not the single limiting factor and the combination of two, e.g., tempera-

Fig. 6. Temperature dependence of formation of Cl8DD, Cl4DD, Cl5DF, and Cl4DF on fly ash (2 h annealing time) [68, 73]

Dioxins and Furans (PCDD/PCDF)

169

Fig. 7. Temperature dependence of PCDD/PCDF formation (2 h annealing time) [74]

ture and residence, is an important parameter for determining the efficiency of how organic substances are being destroyed. As a general rule: higher temperatures need shorter residence times of the gaseous molecules. Thus, it is an engineering question how to build and operate a plant. Role of Precursors: From experiments to condense pentachlorophenol (PCP) over fly ash by Karasek and Dickson (1987) in a temperature range from 250 to 350 °C the precursor theory was established. The authors concluded that metallic constituents in the fly ash act as catalysts for the formation of PCDD. In more recent works, Milligan and Altwicker found that gas-phase 2,3,4,6-tetrachlorophenol was the most efficient precursor in PCDD formation [75, 76]. In addition, the newly formed PCDD were found desorbed in the gas phase and not adsorbed on fly ash. The measured conversions of chlorophenols to PCDD were in agreement with a model suggesting that two adjacent adsorbed precursor molecules dimerize to the product. Role of Sulfur/Chlorine Ratio: In 1986, Griffin established a hypothesis to explain the formation of PCDD/PCDF as a result of the sulfur-to-chlorine ratio in the feed [77]. It is well known that combustion of fossil fuels like coal generates much less PCDD/PCDF than combustion of municipal solid waste. The hypothesis states that in coal there is a sulfur-to-chlorine ratio of 5/1 whereas in municipal waste the ratio S/Cl is 1/3. The latter ratio allows formation of molecular chlorine according to the Deacon process catalytically driven by metals, e.g., copper. The molecular chlorine is considered to be responsible for the de novo dioxin formation according to the following Eq. (1) [77]: CuCl2 +1/2 O2 Æ CuO +Cl2 CuO +2 HCl Æ CuCl2 +H2O 969979 2 HCl +1/2O2 Æ H2O +Cl2

(1)

170

H. Fiedler

However, in fossil fuel with a surplus of sulfur over chlorine, molecular chlorine (Cl2) will be “captured” according to Eq. (2) and formation of chlorinated aromatics does occur. In cases of fossil fuels, such as coal, crude oil, and gas, reaction (2) dominates over reaction (1). Cl2 +SO2 +H2O Æ SO3 +2 HCl

(2)

Similar S-to-Cl ratios as in coal are found in wood and sewage sludge. In addition to the above-mentioned theoretical considerations and observations from large-scale operations, Lindbauer et al. found lower PCDD/PCDF concentrations when high-sulfur coal was added to the fuel [78]. Later, results from Raghunathan and Gullett showed that in the presence of HCl relatively more PCDD/PCDF were formed; however, upon addition of SO2, the formation rate of PCDD/PCDF decreased. The authors determined a critical S/Cl ratio of 0.64. Further increase of S did not result in less dioxins and furans [79]. As no congeneror homologue-specific correlation for the inhibition of dioxin formation could be established, the authors concluded that the depletion of molecular chlorine Cl2, the active chlorinating agent, by SO2 through a gas-phase reaction appears to dominate over the deactivation of the copper catalysts in fly ash (= inhibition mechanism) as previously reported [77]. Role of Chlorine Species: The influence of the chlorine species can be summarized in that chlorination of aromatic compounds readily occurs in the presence of Cl2. Such substitution reactions do occur in the presence of fly ash (heterogeneous phase, probably surface-catalyzed) as well as in the gas phase (homogeneous phase). At temperatures up to 250 °C, HCl does chlorinate chlorine-free dibenzodioxin, 1,2,3,4-Cl4DD or toluene when adsorbed to fly ash.Without fly ash, Cl2 was 4-times more efficient than HCl in chlorinating these compounds [80]. Gaseous chlorine (Cl2) was found to be the most efficient chlorinating agent [81]. Role of Oxygen: From laboratory, pilot-scale, and large-scale experiments it was concluded that increasing oxygen concentrations from 0 to 10% resulted in increasing formation of PCDD/PCDF. The O2 content pushes the Deacon reaction towards Cl2-production and subsequently to formation of organochlorine compounds [82]. Under pyrolytic conditions (oxygen deficiency), dechlorination of PCDD/PCDF occurs at temperatures above 300 °C. Role of Metals: When testing the efficiency of metals to catalyze formation of PCDD/PCDF, copper was found to be the most efficient compound [68]. Further studies have shown that small hydrocarbons such as acetylene and ethylene are readily chlorinated in the presence of cupric chloride or cupric oxide and HCl [83, 84]. The mechanism to reduce Cu(II) to Cu(I) and the oxychlorination of the newly formed Cu(I)Cl to reconvert to Cu(II)Cl2 completes the catalytic cycle. The mechanism is very similar to the copper catalyzed Deacon reaction that converts HCl into Cl2.With acetylene, however, the reaction is accelerated as the activation energy for the formation of Cu(I)Cl is reduced. Role of Deposits and Other Parameters: Results from Kanters and Louw showed that in the absence of fly ash, deposits in the cooler ends of a municipal solid waste incinerator favor the formation of PCDD/PCDF and other PICs (products of incomplete combustion). The authors showed that catalytic pro-

171

Dioxins and Furans (PCDD/PCDF)

cesses caused by conditioned walls played an important role in the formation of PCDD/PCDF via oxychlorination at temperatures above 600 °C [85]. To complete this survey, some additional parameters should be mentioned that were reported to favor the formation of PCDD/PCDF in combustion processes. However, quantitative information is not available. From MSWI incineration it is known that humidity in the feed leads to poorer combustion conditions resulting in a poorer burn-out and higher concentrations of organic carbon in the fly ashes, thus favoring PCDD/PCDF formation. High copper concentrations in fly ashes generate higher PCDD/PCDF levels. An interesting finding is that the HCl concentration in the raw gases seems to be less important for the formation of dioxins and furans than the content of inorganic chloride in the fly ashes [68–71]. Whereas the chlorine concentration in the gas phase is a result of the chlorine in the input, a saturation of the fly ashes seems to occur at relatively low chlorine concentrations; in other words: once a saturation with Cl is reached in the fly ashes (occurring at relatively low chlorine input), formation of PCDD/PCDF occurs. As a result, high chlorine levels in the input do not increase the PCDD/PCDF concentrations in the emissions as the Cl concentrations on the fly ashes are independent of the chlorine in the feed. Most findings obtained in MSWI combustion can be transferred to other thermal processes. Combustion processes generate solid residues as bottom and fly ashes. The quality of the combustion process and especially the burn-out will determine the concentration in the ashes. Typically, higher concentrations are found in the fly ash whereas bottom ash has lower concentrations. If both ashes are mixed, the combined residues will be more contaminated as the bottom ashes alone (which constitute the larger mass) [86].“Typical” ranges of PCDD/PCDF concentrations are shown in Table 31.

Table 31. PCDD/PCDF in residues from waste incineration and other combustion processes

[86] Matrix Municipal solid waste incineration Fly ash Bottom ash Wood combustion Fly ash Bottom ash Home heating systems Soot

Concentration ng I-TEQ/kg

Remarks

13,000 <1,000 50 5–20

Mean, Germany, late 1980s New technology, Germany Mean Germany, late 1980s New technology, Germany

5,800 2.5 820 5.3

Mean waste wood, Switzerland Mean natural wood, Switzerland Mean waste wood, Switzerland Mean natural wood, Switzerland

4–42,048

Range, wood and coal, Germany

172

H. Fiedler

7.3 Secondary Sources of PCDD/PCDF

Dioxin reservoirs are those matrices where PCDD/PCDF are already present, either in the environment or as products. The PCDD/PCDF found in these reservoirs are not newly generated but concentrated from other sources. A characteristic of the reservoir sources is that they have the potential to allow re-entrainment of PCDD/PCDF into the environment. Product reservoirs include PCP-treated wood, PCB-containing transformers and sewage sludge, compost, and liquid manure, which can be used as fertilizers in agriculture and gardens. Reservoirs in the environment are, for example, landfills and waste dumps, contaminated soils (mainly from former chemical production or handling sites), and contaminated sediments (especially in harbors and rivers with industries discharging directly to the waterways). A compilation of German sewage sludge and compost data is given in Table 32. A first survey of German sewage sludges where potentially contaminated sludges should be targeted gave a mean concentration of 202 ng TEQ/kg d.m.: in 1990, most sludges were in the range 50–60 ng TEQ/kg d.m. The legal limit concentration for application on agricultural land is 100 ng I-TEQ/kg d.m. Composting of the total organic fraction from municipal waste collection results in a highly contaminated compost, not suitable for application in house gardens or in agriculture (mean concentration: 38 ng TEQ/kg d.m.). Compost from biowaste, kitchen wastes, or green wastes gives better qualities of approximately 14 ng I-TEQ/kg d.m. Such a mean value, however, is close to the guideline concentration of 17 ng I-TEQ/kg d.m. [5] So far, hardly any country has done a reservoir inventory for PCDD/PCDF. First attempts were made by the city of Hamburg where stationary reservoirs were calculated (see Table 33) [87]. In other words, there is almost no knowledge about the total amounts of PCDD/PCDF present in sinks such as sediments of harbors, rivers, lakes, and oceans, landfills or contaminated soils from (chemical) production sites. Reference is given within the country reports when data are available. It should be noted that similar contaminations were found or can be expected in many other locations. As an example, She and Hagenmaier found almost 4000 ng I-TEQ/kg in the sludges from a former chloralkali plant using graphite electrodes. However, so far, no quantification is available [88]. Although these reservoirs may be highly contaminated with PCDD/PCDF, the chemical-physical properties of these compounds imply that dioxins and Table 32. PCDD/PCDF in sewage sludge and compost from Germany

Sewage Sludge:

Limit value: 1986/87: 1990:

100 ng I-TEQ/kg d.m. 202 ng I-TEQ/kg d.m. 50–60 ng I-TEQ/kg d.m.

Compost:

Guideline value All Wastes: Biowaste

17 ng I-TEQ/kg d.m. 38±22 ng I-TEQ/kg d.m. 14±9 ng I-TEQ/kg d.m.

173

Dioxins and Furans (PCDD/PCDF) Table 33. PCDD/PCDF reservoirs in Hamburg; reference year 1992 [87]

Compartment

Reservoir (g I-TEQ) Percentage of Total

Soil (without landfills and contaminated areas) Sediments of the river Elbe and the harbor Water of the river Elbe and the harbor Air Vegetation

4340 1980 1.1 0.04 1

68.7% 31.3% <0.01% 0.01% <0.01%

Total

6322

100%

furans will stay absorbed to organic carbon of soils or other particles. On the other hand, mobilization can occur in the presence of lipophilic solvents (ﬁ leaching into deeper layers of soils and/or groundwater) or in cases of erosion or run-off by rain from topsoil (ﬁ translocation into the neighborhood). Experience has shown that PCDD/PCDF transport due to soil erosion and run-off does not play a major role for environmental contamination and human exposure [89]. 7.4 Natural Sources

Biological formation of PCDD/PCDF from chlorinated precursors was discussed for compost and sewage sludge and questions on the possibility of a biogenic formation did arise for sediments and soils (especially forest soils). Based on the results of Öberg and Rappe the turnover to convert pentachlorophenol (PCP, the most suitable precursor) to PCDD is in the low ppm-range [2, 3, 90, 91]. Consequently, a chlorinated precursor present in an environmental matrix, such as soil or sediment, at ppm-concentrations should be converted to not more than pptlevels of high-chlorinated PCDD (Cl7DD and Cl8DD). In other words, ppm-concentrations of chlorophenols would generate ppt-levels of Cl7DD and Cl8DD or ppq-concentrations in TEQ. Thus, based on present knowledge, biological formation of PCDD from chlorinated phenols under environmental conditions is negligible [89]. High concentrations of mainly PCDD were found in mined ball clay from the United States of America, kaolinitic clay from Germany, deep soil samples from Great Britain, in dated marine sediment cores from Queensland, Australia, and in man-made lake sediment cores from Mississippi, USA. Typical for all samples is the almost total absence of PCDF and the nearly identical congener and isomer distribution throughout all geographies. All studies provide a strong indication that PCDD/PCDF were formed by natural processes [92–94].

174

H. Fiedler

8 PCDD/PCDF Inventories 8.1 Methods to Establish Dioxin Inventories 8.1.1 General

Since the 1980s, countries made attempts to estimate the emissions of PCDD/PCDF from all sources in their territories. Early inventories were made for Canada [95] and later for Germany [96, 97], and Sweden [98] or the USA [99]. Since the mid-1990s, national agencies or ministries established emission inventories also due to reporting requirements under international conventions such as the POPs (or Aarhus) Protocol under the UN-ECE Convention on LongRange Transport of Air Pollution (LRTAP) [45] or under the Stockholm Convention on Persistent Organic Pollutants (POPs) [30]. National inventories, so far, report emissions of PCDD/PCDF in Toxicity Equivalents (TEQ) most of them using the International Toxicity Equivalency Factors (I-TEF) as established by the NATO/CCMS Working Group on Dioxins and Related Compounds in 1988 (the list of the I-TEFs can be found in Table 20) [28]. The annual emission of a given source can be calculated as [94]: Emission of source=Emission factor x “Active rate”

(1)

or Emission of source=Concentration in emission x. Operational hours x flue gas volume per hour (2) Usually, the annual PCDD/PCDF emission is given in grams TEQ per year. According to Eq. (3), the annual flux is calculated by multiplying the release of PCDD/PCDF (e.g., in µg I-TEQ) per unit of feed material processed or produced (e.g., ton or liter) with the amount of feed material processed or produced (in tons per year). A second method – Eq. (4) – calculates the annual emission of a source by multiplying the measured emissions (e.g., in ng I-TEQ/m3) with the operational hours per year and the flue gas volume per hour [94]. 8.2 Existing Inventories until 1999

In 1999, UNEP Chemicals published a report summarizing existing dioxin and furan inventories worldwide. At this time, 15 countries had performed PCDD/PCDF inventories and estimated releases from known sources. The data on emissions of dioxins and furans compiled in this report were based on a multitude of sources. This variety is also reflected in the depth of the information available. For some countries (e.g., Canada, Denmark, the Netherlands, Belgium, Australia, and Germany) full reports were available given in-depth information on generation of data and aggregation. Unfortunately, some reports were only

175

Dioxins and Furans (PCDD/PCDF)

available in the original language (e.g., Belgium, the Netherlands). Sometimes, the emission inventories (or the most recent updates) were taken from references found in the scientific literature (e.g., Korea and the United Kingdom). In such cases, less background information was available on how data were generated and how extrapolation to a nationwide basis was performed. The information for Austria and France was found in the Internet on the homepage of the respective ministries or agencies. Extracts of national emission inventories were provided by the Czech Republic (data together with the Slovak Republic), Hungary, and the Slovak Republic.Very often not more than the “pure” data were available. For two countries, USA and Sweden, the national emission inventories were in a draft stage. Nevertheless, the USA has distributed several hundred pages of background information (including an electronic database) of the draft report. In addition, data were based on personal communications as some reports are not yet finished, e.g., Sweden) [100]. This UNEP report concluded that today’s major emissions of PCDD/PCDF into the environment came from combustion processes. Based on the available data and a reference year around 1995, the central estimate of total annual PCDD/PCDF emissions was approximately 10,500 g I-TEQ with a lower estimate around 8,300 g I-TEQ/a and an upper estimate of approximately 36,000 g I-TEQ/a. The high PCDD/PCDF emissions reported by a few countries for the early 1990s mainly drive these numbers. It should be noted that Japan has updated its dioxin inventory; for the year 1998, a total emission of 5,300 g I-TEQ was estimated. This new number will add another 1,300 g I-TEQ to the “global” inventory. The estimate for a year around 1995 is shown in Table 34. For France, the estimate for 1998 was included [100].

Table 34. National dioxin and furan inventories: PCDD/PCDF emissions to air for a reference year around 1995 [100]

Country

Annual emission (g TEQ/a)

Austria Australia Belgium Switzerland Canada Germany Denmark France Hungary Japan The Netherlands Sweden Slovak Republic United Kingdom United States

29 150 861 181 290 334 39 873 112 3,981 486 22 42 569 2,744

Total

10,713

176

H. Fiedler

The 1999 situation of PCDD/PCDF inventories in terms of geographic coverage and methods used was summarized as follows [100]: – The present number of national PCDD/PCDF emission inventories is very small (15 based on national data). – Most data is available for countries from Western Europe and Northern Am erica. However, the inventory from the United States is still in a draft stage. From Asia, there is only one inventory for Japan covering a few sectors and the estimate for MSWIs for South Korea. From the Southern Hemisphere, so far, only Australia has estimated annual emissions based on emission factors from the literature. From Africa, Central and Southern America, there are no data at all. – Some countries have based their inventories on emission factors generated outside the own country. For some sources, there are no emission factors determined, e.g., open garbage burning, landfill fires, etc. – Most inventories cover emissions to air only; there are few estimates on emissions to water and land or with products. If existing, the estimates have a high degree of uncertainty. From the existing inventories the following conclusions can be drawn: – Amongst the source sectors, the best coverage exists for municipal solid waste incineration for both stack emission measurements and activity rates. This sector also undergoes the most dramatic changes in technology and thus, emission factors and PCDD/PCDF emissions change rapidly. As a consequence, strong downward trends are recognized in countries with modern technology or stringent legislation. – The sector of hazardous waste incineration is relatively homogeneous and does not present a major source in any country. However, it should be taken into account that the such evaluation is based on data from industrialized countries and such results do not necessarily apply for less developed countries. – There is only limited information available from the iron and steel-producing sector. Some European countries have identified this sector as the most major contributor to national dioxin inventories. The United States and Canada are aware of these sources, but so far, no measurements have been performed. Here, generation of reliable data is urgently required. – From the few examples on dioxin and furan emissions to water, land and with products, it can be concluded that emissions to water only cover wastewaters from the pulp and paper industry. Contamination in products largely is limited to pentachlorophenol (PCP) and the PCDF in polychlorinated biphenyls (PCB) are being ignored. – For some countries, the inventory should be updated to improve estimates of the present situation, especially where more stringent regulation has been established since the current inventory. – Presently there exist no harmonized methods for generating and evaluating data for national PCDD/PCDF inventories. In addition, the coverage of sources varies from country to country. Some countries such as the United Kingdom and the United States to a certain extent include releases of PCDD/PCDF to

Dioxins and Furans (PCDD/PCDF)

177

landfills and land. Some countries give ranges of lower and upper estimates whereas other countries use mean/median values to calculate the annual dioxin emissions for a given source. Harmonization of data acquisition and evaluation is an obvious need and will help to better compare national dioxin inventories. – Finally, harmonization of protocols for sampling stack emissions, water, soil, etc. and for analyzing these samples is highly recommended. The present report should only be seen as a snapshot on PCDD/PCDF emissions and estimates of total releases of these compounds into the environment. Only major sectors of PCDD/PCDF releases into the air were identified. Further, PCDD/PCDF sources may exist which have not yet been identified nor quantified, especially in geographic areas with no existing data. Presently, the coverage is not sufficient to estimate accurately global emissions of PCDD/PCDF. Nevertheless, there are several efforts underway to identify dioxin sources better in parts of the world where so far, there is no information available. In addition, existing inventories will be updated, as it is obvious that measures were taken by many countries to reduce emissions of PCDD/PCDF into the environment. For some industrialized countries in Europe and North America, strong downward trends were observed during the last years. Implementation of dioxin abatement technologies in industrial sectors and advanced combustion technology will help to reinforce such trends. 8.3 PCDD/PCDF Inventories after 1999

Since the UNEP report of 1999, some more information has become available, e.g., for Hong Kong SAR, New Zealand, or existing information has been updated, e.g., for Canada, Denmark, Slovak Republic, and the United States (Table 35). Some of the new information will be detailed in the following section: 8.3.1 Canada

The 2001 report is an up-date of the “Inventory of Releases of PCDDs/PCDFs” report published in January 1999 [101]. The total emissions to air were 164 g I-TEQ per year (Table 36). In addition to jurisdictions, the report is intended to encourage all facilities generating PCDD/PCDF to review this draft and to submit relevant comments and corrections where appropriate, especially as it relates to missing information on concentration where testing has been carried out. In Canada, PCDD/PCDF have been added to the National Pollutants Release Inventory (NPRI) for the year 2000.A number of sectors have been identified as priority sectors for which Canada-Wide Standards are being developed. These sectors are: conical burners, medical, municipal, hazardous waste, and sewage sludge incinerators, electric arc furnaces of the steel manufacturing industry, sinter plants, and pulp and paper industries burning salt-laden wood. The largest individual sources to air in 1999 were identified and listed as shown in Table 37.

178

H. Fiedler

Table 35. National dioxin and furan inventories: PCDD/PCDF emissions to air (UNEP 1999 up-

dated) g TEQ/a Best Austria (A) Australia (AUS) Belgium (B) Switzerland (CH) Canada (CDN) Croatia (HRO) Czech Republic Germany (D) Denmark (DK) Finland (FIN) France (F) Hong Kong SAR (HGK) Hungary (HUN) Japan (JPN) The Netherlands (NL) New Zealand (NZ) Norway (N) Sweden (S) Slovak Republic (SR) United Kingdom (UK) United States of America (USA) Global Flux

29 150 661 181 290 95.5 650 323 19 98.3 873 23 103 2,260 486 14 9.15 22 616 560 2,501 9,964

Reference Year

Reference

1994 1998 1995 1995 1999 ca. 1997 1998 1994 1998/99 ca. 1997 1998 1997 1998 1999/1996 1991 1998 ca. 1997 1993 1996 1993 1995

100 100 100 100 101 100 113 100 102 100 100 105 113 104 100 106, 107 100 100 113 100 14

Max 2,300

170 198 2,737 33 6,370 51 88 1,099 4,901 21,391

8.3.2 Denmark

In the year 2000, an updated picture of the dioxin circulation in the Danish society was published on behalf of the Danish Environmental Protection Agency [102]. The formation of PCDD/PCDF in Denmark in the years 1998–99 has been estimated at 90–830 g I-TEQ per year with emissions to the various compartments of the environment as follows: Air: Water: Soil: Depots:

19–170 g I-TEQ/a 0.3–1.4 g I-TEQ/a 1.3–54 g I-TEQ/a 38–420 g I-TEQ/a

The formation of PCDD/PCDF in Denmark is almost entirely related to combustion processes (Table 38). The main conclusions of the study are [102]: – The total Danish formation of PCDD/PCDF in 1998–99 is estimated at 90–830 g I-TEQ/a. The dominant source is municipal waste incineration. Other sources to mention include coal and biomass combustion and fires, both

179

Dioxins and Furans (PCDD/PCDF) Table 36. Canada: PCDD/PCDF releases to air. Fluxes in g I-TEQ/a [101]

Source

Facilities I-TEQ/a Comments tested 1990 1997 1999

Conical burners factors Waste incineration (total) Medical Municipal Hazardous Sewage sludge Barrel burn factors

4/4 4/9

Steel manufacturing: 6/13 electric arc furnaces Fuel combustion – diesel (traffic) Fuel combustion (agriculture/residential) Iron manufacturing: 2/2 sintering plants Pulp & paper: burning 8/10 salt laden wood Electric power generation 6/29

44 282 130

44 127 36

44 41 25

143 9

83 8

9 7

0 20

0 20

0 20

9

10

11

9

9

9

7

7

7

numbers may be revised

25

25

6

10

10

5

3

5

5

3

3

CWS & SOP CWS expected for 2001 CWS 2000 range (2.7–7.6) with average 5.1 g I-TEQ/a; reductions achieved 89 multi-pollutant initiative underway 3 CWS 2001 decreased from 36 [revised fuel consumption and factor based on test] SOP release info to be developed: lead, zinc, copper, nickel smelters 96 new data was provided by the association, more results will follow SOP SOP

Residential wood combustion

CWS 2001; range: 10–75 g I-TEQ/a total for incineration CWS 2000 numbers increased due to use of revised factors CWS 2000 CWS 2000 numbers increased due to new test results included CWS 2000 not included previously, range 15–25 g I-TEQ/a CWS & SOP CWS expected ¢01; 6 plants tested/reported

Base metals smelting

2/29

3

3

3

Cement kilns

13/23

3

3

2

In service – utility poles Wood preserving plants Wood waste comb. (saw mills/P & P mills) Steel Foundries EAFs Pulp & paper: kraft liquor boilers Magnesium production

n.a. 0/12 1

2 2 1

2 2 1

2 2

0 42

1 1

1 1

Release info to be developed

1/1

0

0

0

new listing; +Magnola to start production in 2000

2/6

2

2

0

427

274

164

Chemical Production Aluminum – secondary smelting Petroleum refineries Sub-total

180

H. Fiedler

Table 37. Largest sources of PCDD/PCDF emissions into the air (>0.5 g I-TEQ/a) – Canada

1999 [101] Category

Facility name

City, province

I-teq/a

Hazardous waste incinerators

Ontario Hydro

Tiverton Ontario

6.80

Iron-sintering

Stelco

Hamilton Ontario

6.00

Steel-electric arc furnace

Sidbec-Dosco

Contrecoeur Québec

3.69

Base metals smelting

Falconbridge

Sudbury Ontario

2.90

Municipal incinerator

Hamilton-Wentworth Solid Waste

Hamilton Ontario

1.93

Steel-electric arc furnace

IPSCO Inc.

Regina Saskatchewan

1.40

Steel-electric arc furnace

Stelco-Master Ltee

Contrecoeur Québec

1.13

Steel-electric arc furnace

Vaco Inc

L’Orignal Ontario

0.92

Steel-electric arc furnace

Slater Steel Hamilton Specialty Bar

Hamilton Ontario

0.82

Steel-electric arc furnace

Co-Steel Inc. (LASCO)

Whitby Ontario

0.79

P & p: salt laden wood

Howe Sound P & P

Pt. Mellon British Columbia

0.66

Steel-electric arc furnace

Gerdau MRM

Selkirk Manitoba

0.63

Hazardous waste incinerators

Safety-kleen (prev. Laidlaw) Corruna Ontario

0.59

Total of above

28.26

accidental fires and others. Most PCDD/PCDF formed in these processes are released to the environment, i.e., air, water, in deposits or products. A minor part is exported with residues like coal fly ash and filter dust out of the country. – Denmark also receives PCDD/PCDF through imported products Denmark and by raw materials extracted from nature. The import by products is estimated at 3.4–106 g I-TEQ/a and is partly related to import of products like wood, leather, and textiles treated with pentachlorophenol (PCP). Other imports include clay, paper/cardboard, and feedstuff. Raw materials extracted from nature in Denmark account for 5–1010 g I-TEQ/a predominantly via clay but also via fish, grass, and animals used for food and feedstuff. – The total Danish emission of PCDD/PCDF to air in 1998–99 is estimated at 19–170 g I-TEQ/a. The dominant sources include municipal waste incineration, biomass combustion in small units without flue gas cleaning like wood stoves and farm boilers, evaporation from PCP-treated wood in use in Denmark, fires, steel and aluminum reclamation. Other sources of air emission are cable scrap reclamation, lime and cement manufacturing, traffic and landfills. In 1999, incineration of chemical waste was a significant source as well, but the contribution from this source is likely to be heavily reduced in 2000 due to redesign of kilns and installation of dioxin filters.

181

Dioxins and Furans (PCDD/PCDF) Table 38. Sources of PCDD/PCDF in Denmark; reference year 1998/99 [102]

Activity

Emissions (g I-TEQ/a)

Miscellaneous Cement and lime Other high temperature materials Steel and aluminum reclamation Other metal manufacture Other manufacturing processes

0.045–3.5 0.006–0.46 1.3–5.6 0.06–0.5 0.004–0.08

Energy generation Coal combustion Other fossil fuels Biomass combustion

0.4–2.3 0.14–0.4 0.73–41

Use of products PCP-treated wood Other PCP-treated materials Bleached processes and bleaching agents Miscellaneous other human and natural activities Fires – accidental Fires – others Traffic Cremation Other activities Waste treatment and disposal Cable scrap reclamation Chemical waste incineration Municipal waste incineration Landfills Waste and storm water Sewage sludge disposal Other activities Total (rounded)

0.5–26 <0.05 <0.5 0.5–20 0.03–6.5 1.3–1.7 0.01 0.09–0.22 0.005–5 2.2–2.7 11–42 0.25–10 0.3–1.4 0.07–0.15 0.08–0.2 19–170

8.3.3 Japan

The Law concerning Special Measures against Dioxins in Japan was legislated by House members as a special law of Air Pollution Control Law and Water Pollution Control Law. It is a basic framework and includes environmental quality standards and emission standards. It should be noted that the term “dioxins” in the Law includes the seven 2,3,7,8-substituted PCDD, ten 2,3,7,8-substituted PCDF, four non-ortho- (or coplanar) PCB and eight mono-ortho-substituted PCB. In total there TEQ includes 29 dioxin-like PCDD, PCDF and PCB; in other words, all compounds, which have a WHO-TEF assigned. For Japan, the emission inventory is prepared annually. The first inventories were for the years 1997 and 1998 (both published in June 1999), the I-TEF from

182

H. Fiedler

Table 39. Japan – Dioxin inventories to air, 1997–1999 (annual releases in g WHO-TEQ/a) [104]

Source

Annual Emissions 1997

1998

1999

Municipal waste incinerators Industrial waste incinerators Small-scale waste incinerators Crematoriums Industrial sources Electric steel-making furnaces Steel industry: sintering facilities Zinc recovery facilities Aluminum production facilities Others Cigarette smoke

5,000/0.037 a 1,500/0.51 a 340–591 2.1–4.6 228 135 42.3 21.3 26.3 –0.2 1.12

1,550/0.037 a 1,100/0.51 a 340–591 2.2–4.8 139.9 113.8 20.4 19.4 25.7 –0.2 1.12

1,350/0.028 a 690/0.50 a 279–481 2.2– 4.8 141.5 101.3 18.4 13.6 17.6 –0.2 1.12

Total

7,300–7,550

3,310–3,570

2,620–2,820

Automobile exhaust Final waste disposal sites

0.093 a

0.093 a

0.093 a

a

Indicates emission to aquatic environment.

1988 was utilized to estimate the emissions of PCDD+PCDF.With the new law in 2000, all inventories now report the total PCDD/PCDF/PCDF releases utilizing the WHO-TEFs. The results for the three past years are shown in Table 39. As can be seen from Table 39, reductions of about 60% have been achieved between 1997 and 1999 [103, 104]. 8.3.4 Hong Kong SAR

In March 2000, the Environmental Protection Department of the Hong Kong Special Administrative Region published a report “An Assessment of Dioxin Emissions in Hong Kong” [105]. For the first time, PCDD/PCDF emissions to air have been estimated for Hong Kong. Almost all estimates are based on U.K. emission factors. The estimates for the year 1997 and the predicted emissions for the year 2007 are summarized in Table 40. According to these estimates, the PCDD/PCDF emissions into air were between 23 and 33 g I-TEQ in the year 1997 and only 2–4 g I-TEQ in the year 2007. According to Table 40, the 1997 inventory is dominated by the emissions from the incineration of municipal solid waste. Coal combustion for power generation was the second largest source. Other known emitters of PCDD/PCDF such as sinter plants do not exist in Hong Kong.Within ten years, a drastic decrease has been predicted mainly due to the fact that the old small MSWIs will be replaced by state-of-the-art plants within the next ten years. Natural and accidental fires have not been quantified.

183

Dioxins and Furans (PCDD/PCDF)

Table 40. Hong Kong SAR – PCDD/PCDF emissions to air; reference years 1997 and 2007 [105]

Sources

Industrial Sources Coal combust. (power) Landfill gas combustion migrating gas flared gas combustion gas Non-ferrous metal Cement manufacture a MSW combustion Chem waste combust. Clinical waste combustion Sewage sludge comb Asphalt mixing Non-Industrial Sources Crematoria humans animals Cars leaded unleaded (with cat) diesel LPG Light GVs (diesel) Heavy GVs (diesel) Buses (diesel) Motorcycles Total a b c

d e

Activity 1997

Inventory 1997 (g I-TEQ)

Activity 2007

Inventory 2007 (g I-TEQ)

6.1 MT

0.4–2.0

5.6 MT

0.3–1.8

254,773 t CH4 17,662 t CH4 NA 27,450 1,514,838 t clinker 116,508 t (old) 10,198 t (CWTC) 3,650 t (old plant) – 84,050 t

0.2–0.3 0.001 NA 0.1–1.0 0.32 21–27 0.004 b 0.4–1.8

145,000 t CH4 10,052 t CH4 NA 27,450 1,514,838 t clinker 1000,000 t (new) 10,198 t (CWTC) 5,290 t (CWTC) c 259,000 dry t 84,050 t

0.13–0.15 0.001 NA 0.1–1.0 0.32 0.5 0.024c 0.2 0.004

16,250 bodies –

0.024

20,750 bodies 7,300 t d

0.031 0.015

2049 M km 2237 M km 2515 M km – 2000 M km 2288 M km 612 M km 287 M km

0.002–0.45 0.001–0.03 0.002–0.03

0.004

c

– 7250 M km 2515 M km 2600 M km 0.001–0.02 2400 M km 0.06–0.09 2557 M km 0.016–0.023 620 M km 0.0001–0.006 469 M km

0.003–0.09 0.002–0.03 – 0.002–0.04 0.07–0.1 0.016–0.023 0.0002–0.01

23–33

2–4

Assuming maximum operational conditions at 0.1 ng I-TEQ/m3 limit, 7680 h/year operation, 7000 m3/min flow rate and is not based on activity data. According to CWTC monitoring data, 4.3 mg I-TEQ of PCDD/F was released in 1997. Assuming maximum operational conditions at CWTC at 0.1 ng I-TEQ/m3 limit, 8000 h/year operation, 30,000 m3/h flow rate (SEIA for CWTC, 29/3/99) and includes BOTH chemical and clinical waste incineration at the CWTC. Assuming average body weight of 70 kg, 7,300 tª100,000 bodies. For new plant, an emission factor of 0.15 fg I-TEQ body –1, corresponding to an emission of 0.1 ng I-TEQ m–3, is used. GV stands for goods vehicles.

8.3.5 New Zealand

In 1995, New Zealand’s Ministry for the Environment initiated the Organochlorines Programme to assess the level of risk posed by POPs. Part of the project was an inventory of dioxin and furan releases. In this section, the findings concerning releases to air are briefly summarized [106, 107]. The reference year for the

184

H. Fiedler

inventory is 1998 and the total releases to air, water and land were estimated in the range 41–109 g I-TEQ/a. Of these, 14–51 g I-TEQ/a were released to air (see Table 41). The emissions to air from waste burning, either in hospital waste incinerators or from uncontrolled fires at landfills, coal and wood combustion, and secondary non-ferrous metal production were the most abundant. Lower emissions came from the manufacture of cement and lime, pulp and paper, iron and steel production, power generation, and crematoria. Overall, the inventory estimates that approximately 60% of PCDD/PCDF emissions to air were from industrial sources and 40% from non-industrial sources such as domestic burning of wood and coal for home heating, backyard waste burning and incidental fires in buildings or from the use of gasoline and diesel for land transport.Among the natural sources considered in this inventory were uncontrolled forest, scrub, and grass fires. One notable difference between the New Zealand inventory and the inventories from most other countries is that New Zealand does not have any municipal solid waste incinerators. 8.3.6 European Union Member States

In 1997, the dioxin release inventory for 17 European countries (15 Member States of the EU plus Norway and Switzerland) has been published [108]. Analyses of basic documents obtained from the 17 European countries gave total PCDD/PCDF emissions into the air from known sources of approximately 3300 g I-TEQ/a (see Table 42). However, a re-evaluation of the data using emission factors for a given process were applied taking into account parameters such as abatement technologies. The results of the re-evaluation procedure yielded PCDD/PCDF emissions to the air from the most relevant sources of 5,800 g ITEQ/a which are thought to represent 90% of the total emissions. Thus, it was estimated that the annual air emissions of PCDD/PCDF from all known sources in the 17 countries are 6500 g I-TEQ/a. A comparison of the two estimates for air emissions on a country basis is shown in Table 43. It can be seen that large differences could be found. This was an expected finding as for some countries almost no or only limited information was available. In addition, the sources of information varied from comprehensive reports based on large numbers of emissions measurements to short estimates based on literature data. A further complication is given by the fact that within the last years considerable technical improvements were implemented in many areas, so that the emission estimates undergo rapid changes. Nevertheless, the re-estimate still has to be considered as considerably uncertain due to the lack of many basic data or low quality of data [108]. Based on the present European emission inventory as compiled by Landesumweltamt Northrhine-Westphalia [108]: 62% of the PCDD/PCDF emissions are due to: – Incinerators for municipal solid waste, – Iron ore sinter plants, – Incineration for clinical waste, and – Facilities of non-ferrous metal industry.

185

Dioxins and Furans (PCDD/PCDF)

Table 41. New Zealand – PCDD/PCDF emissions to air for the year 1998 (g I-TEQ/a) [106, 107]

Source Incineration and combustion processes Clinical, pathological and quarantine waste incineration Hazardous waste incineration Wastewater solids incineration Crematoria

Annual emission g I-TEQ/a

Certainty

0.38–3.5 0.00054–0.0039 0.009 0.0080–0.45

M, M H, H H, L H, L

0.059–0.11

H, M

0.032–3.8 0.0017–0.20 0.36–0.59

H, M H, L H, L

0.28–1.2 0.57–1.2 0.71–8.7 0.54–6.4

M, M M, L H, L L, L

Land transport Unleaded petrol Diesel

0.010–0.59 0.10–0.57

H, L H, L

Uncontrolled fires Forest, scrub and grass fires Structure fires Vehicle fires

0.080–1.1 0.27–2.7 0.10–0.14

L, L L, L L, L

0.10–0.65 0.0030–0.16

H, M H, L

0.10 0.017–0.063 0.10–1.3

H, H H, H L, L

0.0091–1.8 0.00024–0.0038 0.033–0.045

M, L H, L H, M 22

0.00029–0.0084 0.00068–0.024

H, L L, L

Power generation Industrial, commercial and agricultural coal combustion Industrial and commercial appliances Agricultural appliances Domestic coal burning Industrial wood combustion Wood processing wastes Contaminated wood wastes Domestic wood burning Domestic waste burning

Manufacturing and production processes Cement and lime manufacture Cement manufacture Lime manufacture Iron and steel production Primary steel production Secondary steel production Non-ferrous metal production Aluminum production Secondary aluminum production Glass production Pulp and paper production (black liquor recovery boilers) Miscellaneous activities Cigarette smoking Used oil use and disposal Landfills Landfill gas (fugitive emissions) Landfill gas (flared and combusted in engines) Landfill fires

0.077–0.086 0.0013–0.077 10–15

Total annual estimate of emissions to air for 1998

14–51

H=High certainty; M=Medium certainty; L=Low certainty. See Sect. 3.3.

L, L 26 M, L L, L

186

H. Fiedler

Table 42. PCDD/PCDF air emissions in the European Community – Comparison of results from

national reports and LUA re-estimate; reference years 1993–1995 (g I-TEQ/a) [109] Country

National reports

LUA re-estimate

Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg The Netherlands Norway Portugal Spain Sweden Switzerland United Kingdom Total

29 727 43 25 621 600 No data No data No data 29 89 45 No data 134 36 182 715 3,273

121 484 50 69 1,119 840 122 33 1,050 50 117 41 127 327 89 183 928 5,750

Table 43. PCDD/PCDF air emissions in the European Community – Most important sources

for reference years 1993–1995, LUA re-estimate (g I-TEQ/a) [109]; EF=Emission factor, AR=Activity rate Source type

PCDD/PCDF Remarks (g I-TEQ/a)

Uncertainty EF/AR

Municipal waste incineration

1,467+174

Decreasing trend Illegal burning of domestic waste

Low/Low

Sinter plants

1,010+115

Sinter plants for recycling materials

Medium/Low

Use of contaminated wood uncertain

Medium/High

Residential wood combustion 945 Clinical waste incineration

816

Few plant data and statistics

High/High

Wood preservation

381

From PCP-treated materials

v. high/v. high

Fires

380

Based on one reference only

v. high/v. high

Non-ferrous metals

136

Cu, Al, Zn

Medium/Low

Road transport

111

Mainly leaded fuel; decreasing trend

Low/Low

Dioxins and Furans (PCDD/PCDF)

187

38% of the PCDD/PCDF emissions are dominated by non-industrial sources, such as: – Domestic heating, especially wood combustion, – Accidental fires, and – Traffic (mainly if leaded gasoline is used). An evaluation of the re-estimated emissions revealed that the European dioxin inventory is dominated by a few source categories. In the reference period 1993–1995, the largest contributor to the inventory was municipal waste incineration followed by sinter plants (see Table 43). Strong decreasing trends were identified for two source categories: Many countries are in the process to phase out leaded gasoline what will reduce dioxin emissions from this sector. The strongest decline in the present inventory can be expected for municipal waste incineration: Whereas for the reference period 1993–1995, the re-estimate was calculated to be 1,467 g I-TEQ/a, only 20 g I-TEQ/a will be emitted if all MSWIs in Europe will comply with a limit value of 0.1 ng I-TEQ/m3 as required by the 2000 Directive on the incineration of waste [48]. Based on the results of the European Dioxin Inventory [108], Stage II of the European Dioxin Project was conducted between the years 1997 and 2000, which comprised a number of sub-projects aimed to fill several data gaps identified in Stage I. First conclusions to be drawn are that to date, that most Western European countries have established PCDD/PCDF inventories to air. Most of them are based on actual measurements; however, emission measurements are still missing for Greece and – except for municipal solid waste incineration – for Spain and Italy. Ireland, which may have minor PCDD/PCDF releases compared to the other more industrialized European countries, will start a multi-media dioxin project soon. A compilation of the PCDD/PCDF emissions to air for the years 1995 (recalculated from the 1997 inventory with the new results obtained in Stage II) and 2000 as well as the emission reduction achieved is shown in Table 44 [110]. Within the individual source sectors, a rapid decrease in emissions was observed for municipal solid waste incineration (MSWI). Therefore, MSWI is no longer the largest single source in the European emission inventory. Of the Western European countries, only France still has a number of plants with considerable emission. There are some countries in the EU (Ireland, Portugal, Greece) which are going to build up their first incineration plants in the near future. It is expected that these plants will have state-of-the-art flue gas cleaning and therefore will not create significant new emissions of PCDD/PCDF to air [110]. 8.3.6.1 Waste Incineration

The incineration of municipal solid wastes has experienced a rapid decrease of PCDD/PCDF emissions to air due to abatement measures and plant closures in the 1990s. For EU Member States, municipal solid waste incineration no longer is the most important source type of dioxin releases to air. Of the Western Euro-

188

H. Fiedler

Table 44. European PCDD/PCDF emission inventory per sector (concentrations in g I-TEQ/a)

[110] Sector Power plants – fossil fuel Resident. comb. – wood Resident. comb. – coal, lignite Industrial boilers, stationary engines Sinter plants Sec. zinc production Sec. copper production Sec. aluminum production Cement Metal cable reclamation Electric arc furnace steel plants Non-ferrous metal foundries Sintering of spec. materials, drossing facilities Preservation of wood Road transport MSWI (legal) MSWI (illegal) Incineration of industrial waste Incineration of hospital waste Crematoria

1995 (min-max)

2000 (min-max)

Change

59–122 544–989 92–408 32–83 671–864 242–245 31–33 41–50 14–50 42–52 115–162 36–78 115–200

55–72 532–971 86–370 34–81 447–554 22–25 15–17 13–49 13–49 40–50 120–153 40–74 1–86

–30% – 2% – 9% 0% –35% –90% –50% – 2% – 2% – 3% – 1% 0% –72%

145–388 57–138 973–1,213 129–221 149–183 133–530 11–46

131–349 37–82 412–506 126–200 131–166 96–392 9–19

–10% –39% –58% – 7% –10% –27% –51%

Total

3,685–6,470

2,435–4,660

–30%

– Industrial sources

2,793–4,165

1,589–2,516

–41%

892–2,305

846–2,144

– 6%

– Non-industrial sources

pean countries only France still has considerable emissions from these facilities, but here, too, the emissions are declining. It should be noted that waste disposal plans in some countries, like Ireland, Portugal, and Greece, envisage to build their first incineration plants in the near future.According to EU Directive 2000/76/EC [48], these plants will have to comply with the 0.1 ng I-TEQ/Nm3 emission standard and thus be equipped with state-of-the-art flue gas cleaning technology. Such plants will not create contribute significantly to a nation’s air emission inventory. Nevertheless, depending on the type of abatement measures installed dioxin emissions via solid residues and waste water streams are likely to occur. These emissions (to land and water) may further increase in future if the capacities of municipal solid waste incineration were enlarged as a result of decreasing space suitable for landfilling. So far, there did hardly exist any measured data from European hospital waste incinerators as many countries did close down small hospital waste incinerators once they became aware of a dioxin problem in the early 1990s. Within Stage II of the European Dioxin Inventory, emission measurements were carried out in Portugal and in Greece. PCDD/PCDF flue gas concentrations at these on-site fa-

Dioxins and Furans (PCDD/PCDF)

189

cilities ranged from below 0.1 ng I-TEQ/m3 (for a facility equipped with active carbon injection) to up to 300 ng I-TEQ/m3; most results were between 10 and 50 ng I-TEQ/m3. From national data in Belgium, an emission factor of 2250 mg I-TEQ/ton waste was derived based on an average of 150 ng I-TEQ/m3 from four measurements and a specific flue gas volume of 15,000 m3/ton of waste. This clearly shows that these plants may still constitute relevant emission sources. In particular, the small on-site hospital waste incinerators have proven to have inadequate technology with respect to PCDD/PCDF prevention and abatement. The total number of small on-site incinerators for hospital wastes in the Member States of the European Union is presently unknown. Considerable numbers may exist in Greece, Portugal, Italy, and Spain; they are also known to be found in Poland and presumably in the other accession countries [110]. Co-incineration of hospital waste in municipal solid waste incinerators is the most common way of health care waste treatment in countries like France, Germany, and Denmark, where sufficient capacity exist. In large, well-managed stateof-the-art MSWI, the co-incineration of a few percent of health care waste together with “normal” municipal solid waste did not increase the emission from the MSWIs and the emissions were below 0.1 ng I-TEQ/Nm3. For Denmark, however, the results from six measurements of co-incineration of hospital waste in municipal waste incinerators was slightly above the 0.1 ng I-TEQ/Nm3 limit value (average co-incineration=0.29 ng I-TEQ/m3; average MSWI only=0.36 ng I-TEQ/m3). 8.3.6.2 Iron Ore Sintering

Iron ore sintering was still one of the major sources for PCDD/PCDF emissions to air. Measurements performed in the Walloon part of Belgium as well as tests from Flemish and French plants confirmed quite considerable variation of emissions between different plants. PCDD/PCDF concentrations in the flue gas ranged from below 1 to 20 ng I TEQ/m3. Based on these data, the annual emissions for these ten plants yielded nearly 200 g I-TEQ/a. From the Czech Republic measured emission factors between 0.05 and 20 g I-TEQ/t of sinter have been determined during the years 1997–1999; these are well within the range found in the EU Member States. Data from the Spanish, Italian, and Portuguese sinter plants are still missing. With these emissions, iron ore sintering will become the most relevant industrial sector in the European inventories to air. The importance of this source will be further enhanced by facilities located in accession countries in Central Europe, namely in the Czech Republic and in Poland [110]. The results of the measurements performed at the two Walloon plants during Stage II of the European Emission Inventory are shown in Table 45.As can be seen from Table 45, there is almost a factor of 10 between the flue gas concentrations of the two plants resulting in a difference in emission factors by a factor of 5. Whether the difference in dust emissions, 29–61 mg/m3 at the Liège plant vs. 61–165 mg/m3 at Charleroi and the PCDD/PCDF emissions are caused by the same parameters needs to be examined.

190

H. Fiedler

Table 45. PCDD/PCDF emissions from the two Belgian sinter plants measured within the Stage II project [110]

Sample

PCDD/PCDF Flow rate Mass flow Annual mg I-TEQ/h emission ng I-TEQ/m3 Nm3/h g I-TEQ/a

Liège 1 Liège 2 Liège 3

0.80 0.59 0.74

1,504,934 1.2 1,369,811 0.8 1,470,864 1.1

Liège mean

0.71

1,448,536 1.0

9.1

Charleroi 1 Charleroi 2 Charleroi 3

9.50 5.83 5.15

366,608 3.5 383,271 2.2 389,948 2.0

30.5 19.6 17.6

Charleroi Mean 6.82

379,942 2.6

22.5

Production Emission rate t/h factor mg I-TEQ/t

10.5 7.1 9.6 581

1.8

289

8.9

8.3.6.3 Electric Arc Furnaces in the Steel Industry and Iron Foundries

Electric arc furnaces (EAFs) used for steel production from scrap were the second largest source for PCDD/PCDF emissions to air. Although the annual emissions per plant are much lower than those for sinter plants; EAF emissions play an important role as the number of installations is increasing in EU Member States; in 1995, there were 250 EAFs in European Member States. The emissions from this sector have not exhibited the same downward trends as were shown for other sectors as emission reductions through introduction of advanced techniques and technology are being compensated by increased production numbers. Overall, this source type might be the only one among the industrial source with constant or even increasing emissions to air. Nevertheless, it has been shown that through application of suitable abatement technologies, flue gas concentrations below 0.1 ng I-TEQ/m3 can be reached. A release pathway to be considered is the dioxin-contaminated filter dusts: if disposed in landfills or mines they cause a release to land; if transported to zinc recovery plants – a common practice EU-wide – fugitive emissions during handling and transport may occur; in addition, they present an input of PCDD/PCDF into the secondary zinc industry. Although cold-air cupola furnaces used for production of iron and steel castings had quite low PCDD/PCDF emission concentrations (up to ~0.2 ng I-TEQ/m3), analyses of filter dusts exhibited a wide range of PCDD/PCDF contamination with the highest value of about 12,000 ng I-TEQ/kg dust. Therefore, if the dioxin concentrations of filter dust mirror stack releases, this source may play a more important role in national inventories than anticipated before. This assumption is supported by the results reported from the French dioxin emission program [110].

Dioxins and Furans (PCDD/PCDF)

191

8.3.6.4 Secondary Zinc Industry

Zinc recovery plants, e.g., which utilize filter dusts from electric arc furnace (EAF), have been identified as major PCDD/PCDF sources in some countries, e.g., Japan. In Europe, secondary zinc plants exist in France, Germany, Spain, and Italy. However, it is not known whether they all apply the same processes. Stack emissions have been measured at a French zinc recovery plant: initial analyses showed concentrations of 135 ng I-TEQ/m3, which by 1999 have been reduced through installation of abatement technology to 5 ng I-TEQ/m3; with the third and last step of abatement finalized, emissions below 1 ng I-TEQ/m3 should be achieved. Further, these types of plants may not only have stack emissions but may also have fugitive emissions, which may cause PCDD/PCDF deposition in the vicinity of the plant; such impact has been experienced close to a German facility. These fugitive emissions are at least in part due to open-air handling of EAF filter dusts. Since these dusts are shipped from the steel plants to the recycling installations, further emissions may occur during transport in case of big leakage [110]. 8.3.6.5 Miscellaneous Sources

The European emission inventory has shown that there is a vast number of miscellaneous industrial installations, which per plant only have small PCDD/PCDF releases but together they contribute considerably to the annual dioxin and furan emissions in Europe. Among these are secondary smelters for non-ferrous metals (i.e., aluminum, copper), iron foundries (cupola furnaces), cement production (particularly when the wet technology is used and co-incineration of hazardous waste takes place in such plants) [110]. 8.3.6.6 Non-Industrial Sources and Uncertainties

As can be seen from Table 44, non-industrial sources in the past made and still make a large part of the PCDD/PCDF emissions in Europe. Although there exist data from the combustion of fossil fuels and wood in domestic ovens, chimneys, and furnaces, better knowledge is needed especially for the PCDD/F releases from domestic solid fuel combustion. A result of the recent European emission inventory is that the emissions from coal combustion may have been underestimated previously. Another field of uncertainty or contradictory results exist with respect to domestic wood combustion since the fraction of wood contaminated with chlorine containing compounds or in the worst case wood preservatives is hard to assess. Especially in countries like Norway, where wood is used for heating purposes, and where the air contains high concentrations of inorganic chlorine, the burning of “clean” wood may pose a dioxin emission problem.

192

H. Fiedler

Better understanding is also needed to quantify the PCDD/PCDF releases caused by accidental fires and releases of PCDD/PCDF from reservoir sources containing pentachlorophenol, such as PCP-treated wood [110]. 8.3.6.7 Trends in the EU 1985 – 2005

In its 5th action program The EU aimed at a 90% reduction of PCDD/PCDF releases between the years 1985 and 2005. Based on the data from the Stage II of he European Dioxin Inventory, this goal could be achieved for major PCDD/PCDF sources in the industrial sectors. The successful realization was due to fact that already by 1985/1990 targets of PCDD/PCDF reduction were set by the policy, e.g., national limit values for waste incinerators, close-down of hospital waste incinerators, prohibition of open air cable burning, prohibition of halogenated scavengers, etc. Individual plants with high PCDD/PCDF emissions were retrofitted with appropriate abatement technology, e.g., zinc recovery plants in Germany and France, a hot briquetting special sinter plant in Germany. The success of these reduction measures is reflected by decreasing PCDD/PCDF concentrations in ambient air and in deposition (see Sect. 4.1.3), in foodstuffs (Sect. 4.2.1), human blood, and mother’s milk (Sect. 4.2.3). However, with respect to the remaining industrial sources, which predominantly belong to metallurgical industries, considerable effort is required to further minimize PCDD/PCDF and the 90% reduction target was not met.Also for the domestic combustion of solid fuels this goal could not be achieved. The authors of the LUA study attribute this to two reasons [110]: – PCDD/PDCF emissions from non-industrial sources are much more difficult to assess and to regulate than emissions from industrial sources; – not all European countries have characterized these types of emissions sources. The inventories of releases to land and water are even more difficult to estimate and are still incomplete; at the present stage, a conclusion towards an upward or downward trend cannot be made. However, some conclusions can be drawn such as that dioxin-containing waste materials from industrial production, which formerly were disposed off in – very often – inadequate landfills, today mostly are incinerated. Old production processes, which were identified to generate PCDD/PCDF, such as manufacture of chlorinated pesticides and some dyestuffs, have either been abandoned or have been changed and thus, the release of PCDD/PCDF in products and with residues was either eliminated or reduced. Also, through engineering measures to make the combustion process more efficient and achieve a better burn-out, the concentrations in the solid residues, i.e., bottom ashes and fly ashes, decreased for modern incinerators. Nevertheless, there is one important consideration to be taken into account: the adsorption and absorption measures, when applied properly, are capable to reduce emissions to air and/or to water, but they do not destroy PCDD/PCDF; they simply transfer the PCDD/PCDF contamination to another compartment. In these cases, the sinks for these PCDD/PCDF taken out of the stack or the discharge waters are translocated into solid residues, which then have to be treated or otherwise disposed of.

Dioxins and Furans (PCDD/PCDF)

193

8.4 Outlook: PCDD/PCDF Inventories under the Stockholm Convention on POPs

On May 22–23, 2001, the Stockholm Convention on Persistent Organic Pollutants (POPs) (see Chapter by Buccini in this volume) was adopted. One year later, the Convention has 151 signatures and eleven ratifications (Canada, Fiji, Germany, Iceland, Lesotho, Liberia, Nauru, the Netherlands, Rwanda, Samoa, and Sweden). The objective of the Convention is to protect humans and the environment from the adverse effects of these chemicals. Article 5 of this Convention in its subparagraph (a) requires Parties to develop an action plan designed to identify, characterize and address the release of the chemicals listed in Annex C 6 and promote implementation of subparagraphs (b) to (e). Subparagraph (a) also lists the elements that are to be included in such action plans, including an evaluation of current and project releases, i.e., the development and maintenance of source inventories and release estimates, taking into account source categories identified mainly in western industrialized countries and listed in Annex C, Part II. Therefore, dioxin sources must be quantified and the methodology used to assess sources must be consistent in order to follow or monitor dioxin releases over time and between countries [30]. Provisional guidance on the evaluation of current and projected releases of PCDD/PCDF – and as requested by the Convention for polychlorinated biphenyls (PCB) and hexachlorobenzene (HCB) as well – includes the following steps: a. Identifying sources of releases: the Convention requires that all anthropogenic sources be considered; b. Quantifying releases from the identified sources. Principally, quantification for a given source or activity can be done by measuring all individual sources or by applying validated emission factors; and c. Combining individual sources and source categories to develop a nation-wide release inventory. In order to assist countries to identify sources of dioxins and furans and to estimate the amount of their releases into the environment, UNEP Chemicals has developed a “Standardized Toolkit for Identification and Quantification of Dioxin and Furan Releases” (the “Toolkit” for short) [94]. The “Toolkit” was assembled using the accumulated experience of inventory compilation for countries, regions and in depth assessments of source categories. It consists of a report (approximately 190 pages) and a database with default emission factors (in Microsoft EXCEL). The database of default emission factors, which differentiate between the individual classes of technology or performance, have been developed in such a way that the releases differ by order of magnitude. The Toolkit is flexible and applicable to all countries. Countries with no PCDD/PCDF data at all may use the Toolkit to screen industrial and other activities to make first estimates of the scale of potential PCDD/PCDF sources and releases. Countries with measured 6

These are polychlorinated: dibenzo-p-dioxins (PCDD), polychlorinated dibenzofurans (PCDF) plus dioxin-like PCB to be reported in WHO-TEQ as well as polychlorinated biphenyls (PCB) and hexachlorobenzene (HCB).

194

H. Fiedler

1. Apply Screening Matrix to identify Main Source Categories in the Country 2. Check subcategories to identify existing sources and get initial estimate of activity 3. Gather detailed information on processes – e.g., by applying Standard Questionnaire to obtain information on sources – to choose the characteristic parameters for emission factors 4. Quantify identified sources with default/measured emission factors 5. Apply nation-wide to establish full inventory following guidance on presentation of findings Fig. 8. The recommended five step approach to establish standardized PCDD/PCDF source inventories [94]

data may use the Toolkit to review and update the coverage of their inventory, as well as seek agreement between their data and data provided in the Toolkit [94]. The Toolkit was released for use in January 2001 and named a “draft. As with any methodology, the Toolkit needs live testing and validation; such testing is presently underway 7 and first results can be expected towards the end of the year 2002. With the feedback from that test phase UNEP will update the Toolkit and complement the existing Dioxin and Furan Release Inventory of 1999. The inventories received will be published and made available on the UNEP’s POPs Clearinghouse (http://www.chem.unep.ch/pops). The methodology of the Toolkit is summarized in five basic steps and shown in Fig. 8. The screening matrix (Table 46) indicates the 10 major source categories and includes industrial and non-industrial sources as well as reservoirs and contaminated sites. For each main category a listing of subcategories indicates the detailed process activities. Within each process type key parameters or process characteristics are provided, which enable to assign emission factors and to quantify the releases to all media, namely to air, to water, to land, in products, and with residues. For example, within the category 2 “metal processing”, in the subcategory of “aluminum production from scrap” three classes of processes are to be distinguished based on the technology applied and the controls in place – the emissions from these three categories are very different. For each of the classes, emission default factors are provided in the Toolkit. The intention is that the end user can use relatively easily accessible plant and process information to adequately and simply select an emission factor, which will describe the releases from that broad technology. Application of the Toolkit in a first stage does not involve any sampling or analysis. However, where measured data are available or national estimates have been made the Toolkit is designed to allow for their inclusion alongside the estimates made by application of the default emission factors. This should help to rapidly show where there are data gaps or uncertainties and differences between 7

e.g., in the “Asia Toolkit Project on Inventories of Dioxin and Furan Releases” with five Asian countries (Brunei Darussalam, Jordan, Lebanon, Philippines, and Vietnam) [111].

195

Dioxins and Furans (PCDD/PCDF) Table 46. Screening Matrix – Main Source Categories [94]

No. Categories and Subcategories

Air

1 2 3 4 5 6 7 8 9 10

¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ Probably registration only to be followed by site-specific evaluation

Waste Incineration Ferrous and Non-Ferrous Metal Production Power Generation and Heating Production of Mineral Products Transport Uncontrolled Combustion Processes Production of Chemicals and Conumer Goods Miscellaneous Disposal Identification of Potential Hot-Sports

Water

Land Product Residue

processes in one country and emission factors generated from the international literature. The final country PCDD/PCDF release inventory will clearly show that all potential sources have been addressed (even if it is to conclude that the activity does not exist or is insignificant). For each source within a country there will be an estimate of releases to all media where data are sufficient and an indication of likely magnitude if full data are unavailable. Additional information may be included such as plans for upgrading of processes or imminent closure of plants – all of which should help in the interpretation of the results and the prioritization of future actions.

9 References 1. Fiedler H, Hutzinger O, Timms C (1990) Dioxins: Sources of Environmental Load and Human Exposure. Toxicol Environ Chem 29:157–234 2. Öberg LG, Glas B, Swanson SE, Rappe C, Paul KG (1990) Peroxidase-Catalyzed Oxidation of Chlorophenols to Polychlorinated Dibenzo-p-dioxins and Dibenzofurans. Arch Environ Contam Toxicol 19:930–938 3. Wagner HC, Schramm K-W, Hutzinger O (1990) Biogenes polychloriertes Dioxin aus Trichlorphenol. UWSF – Z Umweltchem Ökotox 2:63–65 4. Mackay D, Shiu WY, Ma KC (1992) Illustrated Handbook of Physical-chemical Properties and Environmental Fate for Organic Chemicals, vol II. Lewis Publishers, Boca Raton, FL, USA 5. EC (1999) Compilation of EU Dioxin Exposure and Health Data. Report produced for European Commission DG Environment and UK Department of the Environment Transport and the Regions (DETR), by D. Buckley-Golder, P. Coleman, M. Davies, K. King,A.Petersen, J. Watterson, M. Woodfield, Fiedler H, A. Hanberg, October 1999. Full report at http://europa.eu.int/comm/environment/dioxin/download.htm. Compilation of EU Dioxin exposure and health data 6. Welsch-Pausch K, McLachlan MS, Umlauf G (1995) Determination of the Principal Pathways of Polychlorinated Dibenzo-p-dioxins and Dibenzofurans to Lolium multiflorum (Welsh Rye Grass). Environ Sci Technol 29:1090–1098 7. Fiedler H, Hutzinger O, Welsch-Pausch K, Schmiedinger A (2000) Evaluation of the Occurrence of PCDD/PCDF and POPs in Wastes and Their Potential to Enter the Foodchain.

196

8. 9. 10. 11. 12. 13. 14.

15.

16. 17.

18. 19. 20.

21.

22. 23. 24.

H. Fiedler Co-ordination by the Joint Research Centre, Environment Institute, Soil & Waste Unit, Dr. Gunther Umlauf, on behalf of DG ENV E.1. Study on behalf of the European Commission, DG Environment, September 2000 Welsch-Pausch K, McLachlan MS (1998) Fate of Airborne Polychlorinated Dibenzo-pdioxins and Dibenzofurans in an Agricultural Ecosystem. Environ Pollution 102:129–137 Fürst P (1998) Dioxine in Lebensmitteln. In: (Oehme M, ed), Handbuch Dioxine – Quellen, Vorkommen, Analytik, pp 227–266. Spektrum Akademischer Verlag, Heidelberg Petreas M, Ruble R, Visista P, Mok M, McKinney M, She J, Stephens R (1996) Bioaccumulation of PCDD/Fs from Soil by Foraging Chickens. Organohalogen Compd 29:51–53 Petreas MX, Goldman LR, Hayward DG, Chang RR, Flattery JJ,Wiesmüller T, Stephens RD (1991) Biotransfer and Bioaccumulation of PCDD/PCDFs from Soil: Controlled Exposure Studies of Chickens. Chemosphere 23:1731–1741 Schuler F, Schmid P, Schlatter C (1997) The Transfer of Polychlorinated Dibenzo-p-dioxins and Dibenzofurans from Soil into Eggs of Foraging Chicken. Chemosphere 34:711–718 Fiedler H, Buckley-Golder D, Coleman P, King K, Petersen A (1999) Compilation of EU Dioxin Exposure and Health Data: Environmental Levels. Organohalogen Compd 43:151–154 US-EPA (2000) Exposure and Human Health Reassessment of 2,3,7,8-Tetrachlorodibenzop-dioxin (TCDD) and Related Compounds. Part I: Estimating Exposure to Dioxin-Like Compounds – Volume 2: Sources of Dioxin-Like Compounds in the United States. EPA/600/P-00/001Bb, September 2000. Draft Final Report, Washington, D.C., USA. See WebSite http://www.epa.gov/ncea/pdfs/dioxin/dioxreass.htm Cleverly D, Monetti M, Phillips L, Cramer P, Hert M, McCarthy S, O’Rourke K, Stanley L, Winters D (1996) A Time-trends Study of the Occurrences and Levels of CDDs/CDFs and Dioxin-like PCBs in Sediment Cores from 11 Geographically Distributed Lakes in the United States. Organohalogen Compd 28:77–82 Moche W, Thanner G (2001) One Year Continuous Monitoring of PCDD/F and PCB Ambient Air Concentrations in the Vicinity of Steelworks in Austria. Organohalogen Compd 51:81–83 BLAG (2001) Daten zur Dioxinbelastung der Umwelt. 3. Bericht der Bund/Länder-Arbeitsgruppe DIOXINE. Bundesministerium für Umwelt, Naturschutz und Reaktorsicherheit, Bonn, 30.08.01. Redaktion: Prof. Dr.Armin Basler, Gerlinde Knetsch, Marianne Rappolder, Dr. Heidi Fiedler Fiedler H, Lau C, Cooper K, Andersson R, Hjelt M, Rappe C, Bonner M, Howell F (1997) PCDD/PCDF in the Atmosphere of Southern Mississippi, USA. Organohalogen Compd 33:122–127 Fiedler H, Rottler H, Peichl L, Knetsch G, Basler A (2000) Concentrations of PCDD/PCDF in Atmospheric Samples in Germany. Organohalogen Compd 45:264–268 Fricke W, Zimmermann T, Frank M, Bender M, Gasch G, Hanewald K, Ott W, Grünhage L, Jäger H-J, Fiedler H, Gaßner G, Hutzinger O, Lau C, Weidner E (1996) Ökosystemares Biomonitoring-Programm in der Region Biebesheim 1992–1994. Schriftenreihe der Hessischen Landesanstalt für Umwelt, Umweltplanung,Arbeits- und Umweltschutz, Heft 193, November 1996 SCOOP (2000) Assessment of dietary intake of dioxins and related PCBs by the population of EU Member States, 7 June 2000. European Commission, Reports on tasks for scientific cooperation, Task 3.2.5. http://europa.eu.int/comm/dgs/health_consumer/library/pub/pub08_en.pdf Rose M, White S, Harrison N, Gleadle A (2001) PCDD/Fs (Dioxins) and PCBs in the UK Diet: 1997 Total Diet Study Samples. Organohalogen Compd 52:143–146 Santillo D, Fernandes A, Stringer R, Johnston P, Rose M, White S (2001) Concentrations of PCDDs, PCDFs and PCBs in Samples of Butter from 24 Countries. Organohalogen Compd 51:275–278 Büchert A, Cederberg T, Dyke P, Fiedler H, Fürst P, Hanberg A, Hosseinpour J, Hutzinger O, Kuenen JG, Malisch R, Needham LL, Olie K, Päpke O, Rivera Aranda J, Thanner G, Um-

Dioxins and Furans (PCDD/PCDF)

25. 26. 27.

28.

29. 30. 31.

32.

33. 34. 35. 36. 37.

38.

39.

40. 41.

197

lauf G,Vartiainen T, van Holst C (2001) ESF Workshop on Dioxin Contamination in Food. Environ Sci & Pollut. Res. 8:84–88 bgvv (1999) Bundesinstitut für gesundheitlichen Verbraucherschutz und Veterinärmedizin, Frauenmilch- und Dioxin-Human-Datenbank, 1999 IARC (1997) Polychlorinated dibenzo-para-dioxins and polychlorinated dibenzofurans. IARC monographs on the evaluation of carcinogenic risks to humans.Volume 69. WHO, IARC, Lyon, France US-EPA (2000) Exposure and Human Health Reassessment of 2,3,7,8-Tetrachlorodibenzop-dioxin (TCDD) and Related Compounds. Part II: Health Assessment of 2,3,7,8-Tetrachlorodibenzo-p-dioxin (PCDD) and Related Compounds. Chapters 1–7, EPA/600/P00/001Be, September 2000. Draft Final Report, Washington, D.C., USA. See WebSite http://www.epa.gov/ncea/pdfs/dioxin/dioxreass.htm NATO/CCMS (1988) International Toxicity Equivalency Factor (I-TEF) Method of Risk Assessment for Complex Mixtures of Dioxins and Related Compounds. Pilot Study on International Information Exchange on Dioxins and Related Compounds, Report Number 176,August 1988, North Atlantic Treaty Organization, Committee on Challenges of Modern Society WHO (1997) WHO Toxic Equivalency Factors (TEFs) for Dioxin-like Compounds for Humans and Wildlife. 15–18 June 1997, Stockholm, Sweden UNEP (2001) The text of the Stockholm Convention on POPs can be found at http://www.chem.unep.ch/pops Van den Berg M, Birnbaum L, Bosveld ATC, Brunström B, Cook P, Feeley M, Giesy JP, Hanberg A, Hasegawa R, Kennedy SW, Kubiak T, Larsen JC, Van Leeuwen R, Liem AKD, Nolt C, Petersen RE, Poellinger L, Safe S, Schrenk D, Tillitt D, Tysklind M, Younes M, Wærn F, Zacharewski T (1998) Toxic Equivalency Factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environ Health Perspect 106:775–792 BNA (2001) “Dioxins – Final EPA Reassessment of Risks to be Issued ‘Early’ 2002, Farland Says” and “Dioxins – Science Panel Urged to Resis Pressure to Review EPA Dioxin Risk Reassessment”. Chemical Regulation 25 (No. 51), 1822–1823, December 31, 2001. Bureau of National Affairs, Inc. Washington, DC, USA ATSDR (no date): http://www.atsdr.cdc.gov/dioxindt.html ATSDR (1998) Toxicological profile for chlorinated dibenzo-p-dioxins (CDDs).Agency for Toxic Substances and Disease Registry Atlanta, GA; U.S. Department of Health and Human Services, Public Health Service WHO (1998) WHO Experts Re-evaluate Health Risks from Dioxins. World Health Organization, Press Release WHO/45, 3 June 1998 van Leeuwen FXR, Younes M (1998) WHO Revises the Tolerable Daily Intake (TDI) for Dioxins. Organohalogen Compd 38:295–298 SCF (2000) Opinion of the SCF on the Risk Assessment of Dioxins and Dioxin-like PCBs in Food.Adopted on 22 November 2000. European Commission, Health & Consumer Protection Directorate-General, Scientific Committee on Food. SCF/CS/CNTM/DIOXIN/ 8 Final. 23 November 2000 http://europa.eu.int/comm/food/fs/sc/scf/out78_en.pdf EC DG Health & Consumer Protection (2000) Assessment of Dietary Intake of Dioxins and Related PCBs by the Population of EU Member States. Reports on tasks for scientific cooperation; Report of experts participating in Task 3.2.5. European Commission: Health & Consumer Protection Directorate-General, Brussels, Belgium, 7 June 2000 SCF (2001) Opinion of the SCF on the Risk Assessment of Dioxins and Dioxin-like PCBs, Update based on new scientific information, 30 May 2001. European Commission, Scientific Committee on Food. http://europa.eu.int/comm/food/fs/sc/scf/ out90_en.pdf JECFA (2001) Joint FAO/WHO Expert Committee on Food Additives, 57th Meeting, Rome, 5–14 June 2001 http://www.fao.org/es/esn/jecfa/jecfa57c.pdf Ohsako S., Y. Miyabara, N. Nishimura, S. Kurosawa, M. Sakaue, R. Ishimura, M. Sato, K. Takeda,Y.Aoki, H. Soni, C. Tohyama, J.Yonemoto (2001) Maternal Exposure to a Low Dose of 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) Suppressed the Development of Repro-

198

42.

43.

44.

45.

46. 47.

48. 49. 50. 51. 52. 53. 54. 55.

56. 57. 58.

H. Fiedler ductive Organs of Male Rats: Dose-dependent Increase of mRNA Levels of 5-alpha-Reductase Type 2 in Contrast to Decrease of Androgen Receptor in the Pubertal Ventral Prostate. Toxicol Sci 60:132–143 Faqi AS, Dalsenter PR, Merker HJ, Chahoud I (1998) Reproductive Toxicity and Tissue Concentrations of Low Doses of 2,3,7,8-Tetrachlorodibenzo-p-dioxin in Male Offspring Rats Exposed throughout Pregnancy and Lactation. Toxicol Appl Pharmacol 150:383–392 ChemVerbotsV (1996) Verordnung über Verbote und Beschränkungen des Inverkehrbringens gefährlicher Stoffe, Zubereitungen und Erzeugnisse nach dem Chemikaliengesetz (Chemikalien-Verbotsverordnung – ChemVerbotsV) vom 19. Juli 1996. BGBl. I 1996, S. 1151, BGBl. I S. 1498 (Chemical’s Law) BImSchV (1992) 19. Verordnung zur Durchführung des Bundesimmissionsschutzgesetzes vom 24.07.1992 (Verordnung über Chlor- und Bromverbindungen als Kraftstoffzusatz-19. BImSchV). Bundesgesetzblatt Teil 1, Jahrgang 1992, 75 (Ordinance on ban of halogenated scavengers, Germany) UN-ECE (1998) Protocol to the 1979 Convention on Long-range Transboundary Air Pollution on Persistent Organic Pollutants and Executive Body deciion 1998/2 on Information to be Submitted and the Procedure for adding Substances to Annexes I, II or III to the Protocol on Persistent Organic Pollutants. United Nations Economic Commission for Europe, New York and Geneva, 1998 EC (1994) Council Directive 94/67/EC of 16 December 1994 on the incineration of hazardous waste. Official Journal of the European Community L 365/34 BImSchV (1990) 17. Verordnung zur Durchführung des Bundesimmissionsschutzgesetzes vom 23.1.1990 (Verordnung über Verbrennungsanlagen für Abfälle und ähnliche brennbare Stoffe – 17. BImSchV). Bundesgesetzblatt Teil I, Jahrgang 1990, 2832. (Ordinance for waste incinerators, Germany) EC (2000) Directive 2000/76/EC of the European Parliament and of the Council of 4 December 2000 on the incineration of waste. Official Journal of the European Communities L 332/91–111 Law Concerning Special Measures against Dioxins (Law No. 105 of 1999. Promulgated on July 16, 1999). Environment Agency, Japan Austria (1997) Verordnung: Begrenzung der Emission von luftverunreinigenden Stoffen aus Anlagen zum Sintern von Eisenerzen. BGBl. II, Jahrgang 1997, 20.01.1997, Nr. 163 (Ordinance for sinter plants) Austria (1997) Verordnung: Begrenzung der Emission von luftverunreinigenden Stoffen aus Anlagen zur Erzeugung von Eisen und Stahl. BGBl. Teil II, Jahrgang 1997, 17.06.1997, Nr. 160 (Ordinance for iron and steel plants) BImSchV (1997) 27.Verordnung zur Durchführung des Bundes-Immissionsschutzgesetzes (Verordnung über Anlagen zur Feuerbestattung – 27. BImSchV) vom 19. März 1997. BGBl. I, S. 545. (Ordinance for crematories) AbfKlärV (1992) Klärschlammverordnung (AbfKlärV) vom 15.04.1992. Bundesgesetzblatt, Jahrgang 1992, Teil 1, 912–934 (Sewage Sludge Ordinance, Germany) Austria (1994) Verordnung des Bundesministers für Land- und Forstwirtschaft, mit der Bestimmungen zur Durchführung des Düngemittelgesetzes erlassen werden (Düngemittelverordnung 1994). BGBl. Nr. 1007/1994 BLAG (1992) Umweltpolitik: Bericht der Bund/Länder-Arbeitsgruppe DIOXINE. Rechtsnormen, Richtwerte, Handlungsempfehlungen, Meßprogramme, Meßwerte und Forschungsprogramme. Bundesminister für Umwelt, Naturschutz und Reaktorsicherheit (Hrsg.), Bonn, Januar 1992 Leidraad bodemsanering (1988) Prüftabelle für Boden und Grundwasser, afl. 4 November 1988 Staatsuitgeverij’s – Gravenhage. In: Landesamt für Wasser und Abfall NordrheinWestfalen (1989): Materialien zur Ermittlung und Sanierung von Altlasten, Band 2 BLAG (1993) Umweltpolitik: 2. Bericht der Bund/Länder-Arbeitsgrupe DIOXINE. November 1993, Bonn Esposito MP, Tiernan TO, Dryden FE (1980) Dioxins. US-EPA/600/-80–197

Dioxins and Furans (PCDD/PCDF)

199

59. Hutzinger O, Fiedler H (1993) From Source to Exposure: Some Open Questions. Chemosphere 27:121–129 60. Hutzinger O, Fiedler H (1991) Formation of Dioxins and Related Compounds from Combustion and Incineration Processes. In: Bretthauer EW, Kraus HW, di Domenico A (eds) Dioxin Perspectives – A Pilot Study on International Information Exchange on Dioxins and Related Compounds, Chapter 3, NATO – Challenges of Modern Society, Volume 16, Plenum Press, New York, pp 263–434 61. BGA/UBA (1993) Dioxine und Furane – ihr Einfluß auf Umwelt und Gesundheit. Erste Auswertung des 2. Internationalen Dioxin-Symposium und der fachöffentlichen Anhörung des Bundesgesundheitsamtes und des Umweltbundesamtes in Berlin vom 09. bis 13.11.1992. Bundesgesundheitsblatt Sonderheft/93 (36. Jahrgang, Mai 1993) 62. Olie K, Vermeulen PL, Hutzinger O (1977) Chlorodibenzo-p-dioxins and Chlorodibenzofurans are Trace Components of Fly Ash and Flue Gas of Some Municipal Waste Incinerators in the Netherlands. Chemosphere 6:445–459 63. Johnke B (1998) Situation and Aspects of Waste Incineration in Germany. UTA Technology & Environment 2/98:98–103 (GIT Verlag, Darmstadt) 64. NATO/CCMS (1988) Hutzinger O, Fiedler H: Emissions of Dioxins and Related Compounds from Combustion and Incineration Sources. Pilot Study on International Information Ex-change on Dioxins and Related Compounds, NATO/CCMS Report No. 172 65. Fiedler H (1998) Thermal Formation of PCDD/PCDF – A Survey. Environ Eng Sci 15/1:49–58 66. Bumb RR, Crummett WB,Artie SS, Gledhill JR, Hummel RH, Kagel RO, Lamparski LL, Luoma EV, Miller DL, Nestrick TJ, Shadoff LA, Stehl RH,Woods JS (1980) Trace Chemistries of Fire: A Source of Chlorinated Dioxins. Science 210:385–390 67. Hutzinger O, Blümich MJ, van den Berg M, Olie K (1985) Sources and Fate of PCDDs and PCDFs: An Overview. Chemosphere 14:581–600 68. Stieglitz L, Zwick G, Beck J, Roth W, Vogg H (1989) On the De novo-Synthesis of PCDD/PCDF on Fly Ash of Municipal Waste Incinerators. Chemosphere 18:1219–1226 69. Vogg H (1991) Technische Minderungsmaßnahmen für Dioxine/Furane in Abfallverbrennungsanlagen. Organohalogen Compd 6:279–296 70. Vogg H (1993) Formation Mechanisms and Technical Reduction Measures for Dioxins/Furans in Waste Incineration Plants. In: Current Views on the Impact of Dioxins and Furans on Human Health and the Environment, Berlin, Germany, November 9–11, 1992. The Toxicology Forum, pp 459–465 71. Vogg H (1995) PCDD/PCDF und Abfallverbrennung. Organohalogen Compd 22 : 31–48 72. Rubey W, Dellinger B, Hall DL, Mazer SL (1985) High-Temperature Gas-phase Formation and Destruction of Polychlorinated Dibenzofurans. Chemosphere 14:1483–1494 73. Vogg H, Stieglitz L (1986) Thermal Behavior of PCDD/PCDF in Fly Ash from Municipal Incinerators. Chemosphere 15:1373–1378 74. Schwarz G, Stieglitz L, Roth W (1990) Formation Conditions of Several Polychlorinated Compound Classes on Fly Ash of a Municipal Waste Incinerator. Organohalogen Compd 3:169–172 75. Milligan MS, Altwicker ER (1996) Chlorophenol Reactions on Fly Ash. 1. Adsorption/Desorption Equilibria and Conversion to Polychlorinated Dibenzo-p-dioxins. Environ Sci Technol 30:225–229 76. Milligan MS,Altwicker ER (1996) Chlorophenol Reactions on Fly Ash. 2. Equilibrium Surface Coverage and Global Kinetics. Environ Sci Technol 30:230–236 77. Griffin RD (1986) A New Theory of Dioxin Formation in Municipal Solid Waste Combustion. Chemosphere 15:1987–1990 78. Lindbauer RL,Wurst F, Prey T (1992) Combustion Dioxin Suppression in Municipal Solid Waste Incineration with Sulfur Additives. Chemosphere 25:1409–1414 79. Raghunathan K, Gullett BK (1996) Role of Sulfur in Reducing PCDD and PCDF Formation. Environ Sci Technol 30:1827–1834

200

H. Fiedler

80. Gullett BK, Lemieux PM, Dunn JE (1994) Role of Combustion and Sorbent Parameters in Prevention of Polychlorinated Dibenzo-p-dioxin and Polychlorinated Dibenzofuran Formation during Waste Combustion. Environ Sci Technol 28:107–118 81. Gullett BK, Bruce KR, Beach LO (1990) Formation of Chlorinated Organics during Solid Waste Combustion. Waste Manage Res 8:203–214 82. Vogg H, Metzger M, Stieglitz L (1987) Recent Findings on the Decomposition of PCDD/PCDF in Municipal Waste Incineration. Waste Manage Res 5:285–294 83. Froese KL, Hutzinger O (1993) Polychlorinated Benzene and Polychlorinated Phenol in Heterogeneous Combustion Reactions of Ethylene and Ethane. Environ Sci Technol 27:121–129 84. Froese KL, Hutzinger O (1996) Polychlorinated Benzene, Phenol, Dibenzo-p-dioxin and Dibenzofuran in Heterogeneous Combustion Reactions of Acetylene. Environ Sci Technol 30:998–1008 85. Kanters J, Louw R (1996) Thermal and Catalysed Halogenation in Combustion Reactions. Chemosphere 32:89–97 86. Fiedler H (1996) Dioxine in Produkten und Abfällen. In: Dioxine-Vorkommen, Minderungsmaßnahmen, Meßtechnik. VDI Band 1298, 231–247. VDI-Verlag, Düsseldorf 87. FHH (1995) Dioxin-Bilanz für Hamburg. Hutzinger O, Fiedler H, Lau C, Rippen G, Blotenberg U, Wesp H, Sievers S, Friesel P, Gras B, Reich T, Schacht U, Schwörer R. Hamburger Umweltberichte 51/95. Freie und Hansestadt Hamburg, Umweltbehörde (eds). Hamburg September 1995 88. She J, Hagenmaier H (1994) PCDDs and PCDFs with Chloralkali Pattern in Soil and Sludge Samples. Organohalogen Compd 20:261–264 89. Fiedler H (1995) EPA DIOXIN-Reassessment: Implications for Germany. Organohalogen Compd 22:209–228 90. Öberg L, Rappe C (1992) Biochemical Formation of PCDD/F from Chlorophenols. Chemosphere 25:49–52 91. Öberg LG, Andersson R, Rappe C (1992) De novo Formation of Hepta- and Octachlorodibenzo-p-dioxins from Pentachlorophenol in Municipal Sewage Sludge. Organohalogen Compd 9:351–354 92. Rappe C, Bergek S,Andersson R, Cooper K, Fiedler H, Bopp R, Howell F, Bonner M (1999) PCDDs in Naturally-Formed Lake Sediment Cores from Southern Mississippi, USA. Organohalogen Compd 43, 111–116 93. Rappe C, Andersson R, Cooper K, Bopp R, Fiedler H, Howell F, Bonner M (2000) PCDDs in Naturally-Formed and Man-Made Lake Sediment Cores from Southern Mississippi. Organohalogen Compd 46:19–22 94. UNEP (2001) Standardized Toolkit for Identification and Quantification of Dioxin and Furan Releases. UNEP Chemicals, Geneva, Draft January 2001, at http:// www.chem.unep.ch/pops/newlayout/prodocas.htm 95. Sheffield A (1985) Sources and Releases of PCDD’s and PCDF’s to the Canadian Environment. Chemosphere 14:118 96. Fiedler H, Schramm KW, Hutzinger O (1990) Dioxin Emissions to the Air: Mass Balance for Germany Today and in the Year 2000. Organohalogen Compd 4:395–400 97. Wintermeyer D, Rotard W (1994) Dioxin-Emission und -Deposition in der Bundesrepublik Deutschland – Versuch einer Bilanzierung. Staub, Reinhaltung der Luft 54:81–86 98. De Wit C., Lexén K, Strandell M (1996) The Swedish Dioxin Inventory: Levels, Sources and Trends of Dioxins and Dioxin-Like Substances in the Swedish Environment, Part 2 (Draft Version) 99. Thomas VM, Spiro TG (1996) The U.S. Dioxin Inventory: Are there Missing Sources. Environ Sci Technol 30:82A–85A 100. UNEP (1999) Dioxin and Furan Inventories – National and Regional Emissions of PCDD/PCDF. Fiedler H; Report by UNEP Chemicals, Geneva, Switzerland. May 1999 101. Environment Canada (2001) Inventory of Releases – Updated Edition. Prepared by Environment Canada, February 2001

Dioxins and Furans (PCDD/PCDF)

201

102. COWI (2001) Substance Flow Analysis for dioxins in Denmark. Erik Hansen, COWI, Environmental Project No. 570 2000 103. EA Japan (2000) Results presented by S. Sakai “Formation Mechanism and Emission Reduction of PCDDs in Municipal Waste Incinerators” at UNEP Workshop on Training and Management of Dioxins, Furans, PCBs. Seoul, Republic of Korea, July 24–28, 2000. http://www.chem.unep.ch/pops/newlayout/prodocas.htm 104. Konda C (2001) Emission Inventory of Dioxins, Furans and Coplanar PCBs in Japan. Organohalogen Compd 50:297–300 105. Hong Kong (2000) An Assessment of Dioxin Emissions in Hong Kong: Final Report, March 2000. Environmental Protection Department, Hong Kong, SAR 106. NZ (2000) New Zealand Inventory of Dioxin Emissions to Air, Land and Water, Reservoir Sources. S.J. Buckland, H.K. Ellis, P.H. Dyke. Organochlorines Programme, Ministry for the Environment, Wellington, NZ, March 2000 107. Dyke PH, Buckland S, Ellis H (2000) Multi-media Inventory of PCDD/F Releases for New Zealand. Organohalogen Compd 46:43–46 108. LUA (1997) Identification of Relevant Industrial Sources of Dioxins and Furans in Europe. Materialien No. 43. Landesumweltamt Nordrhein-Westfalen, Essen, 1997 109. Quaß U, Fermann MW, Bröker G (1998) Steps Towards an European Emission Inventory. Organohalogen Compd 36:7–10 110. LUA (2000) The European Dioxin Emission Inventory – Stage II. Final Report December 2000. Materialien No. 59. Landesumweltamt Nordrhein-Westfalen, Essen 2001 111. UNEP (2001) Report: Kick-off Workshop of the Asia Toolkit Project on Inventories of Dioxin and Furan Releases, Hanoi, S.R. Vietnam, October 1 – 4, 2001 http://www.chem.unep.ch/pops/POPs_Inc/proceedings/coverpgs/procovers.htm 112. McLachlan, Michael, Richter (1998) Uptake and Transfer of PCDD/Fs by Cattle Fed Naturally Contaminated Feedstuffs and Feed Contaminated as a Result of Sewage Sludge Application. 1. Lactating Cows. J. Agric Food Chem 46:1166–1172 113. UNEP (2000) Proceedings of the Subregional Workshop on Identification and Management of PCBs and Dioxins/Furans, Cavtat, Croatia, 29 May–June 2000. UNEP Chemicals, Geneva, Switzerland; http://www.chem.unep.ch/pops/POPs_Inc/proceedings/coverpgs/ procovers.htm

CHAPTER 7

Releases of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans to Land and Water and with Products Patrick H. Dyke PD Consulting, Magdalen, Brobury, Herefordshire HR3 6DX, UK E-mail: [email protected]

This chapter of the Handbook of Environmental Chemistry addresses the release of polychlorinated dibenzo-p-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF) to water and to land as well as their presence in products and residues. Whilst a great deal of attention has been focused in recent times on the release of PCDD/F to air and detailed inventories of sources have been developed for a number of countries, there has not been a similar systematic approach used to compile releases to water and land or to quantify the amount of PCDD/F present in products and residues. The chapter discusses work that has been done in this area and uses a selection of examples to support the author’s contention that releases of PCDD/F to all media, as well as their presence in products and residues, need to be included in a comprehensive assessment of pollution by these chemicals.A comprehensive understanding of the presence of PCDD/F in releases to the environment as well as in products and residues will enhance the development of the most appropriate and effective policies to reduce and control exposure and consequent adverse effects. Keywords: PCDD, PCDF, Effluent, Inventory, Waste, Residue

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204

1

Introduction

1.1 1.2 1.3 1.4

Aims of this Chapter . . . . . . . . . . . . . . . . Sources of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans . . . . . . . . . . System Boundaries and Relative Impacts . . . . . Toxic Equivalency Schemes . . . . . . . . . . . .

2

Why Consider Releases to Land, Water and PCDD/F in Products?

3

Historical Perspective

4

Information in Existing Inventories

. . . . . . . . . . . . . . . . 209

5

Special Cases and Double Counting

. . . . . . . . . . . . . . . . 214

5.1 5.2

Sewage Sludge . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214 Wastes and Waste Management . . . . . . . . . . . . . . . . . . . 215

6

Importance of Managing Intermediates, Wastes and Effluents Effectively . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216

. . . . . . . . . 204 . . . . . . . . . 205 . . . . . . . . . 205 . . . . . . . . . 206 207

. . . . . . . . . . . . . . . . . . . . . . . . 208

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

204

P.H. Dyke

6.1 6.2

PVC Production and Related Activities . . . . . . . . . . . . . . . 217 Magnesium and Nickel Production in Norway . . . . . . . . . . . 218

7

Clay, Citrus Pulp Pellets and PCB Oils

8

Conclusions

9

References

. . . . . . . . . . . . . . . 219

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221

1 Introduction There has been significant concern over the possible effects of persistent organic pollutants (POPs) on human and ecosystem health for some time. It has been recognised that POPs may travel large distances and cause problems at a considerable distance from the source and therefore that concerted international action is required to address the problems. The United Nations Environment Programme facilitated the negotiation of an international legally binding instrument the Stockholm Convention for the reduction and elimination of releases of twelve persistent organic pollutants (POPs). Two of these POPs identified for action are classed as “by-products” which are produced by a range of processes – the polychlorinated dibenzo-p-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF). PCDD and PCDF form a group of 210 closely related compounds and are widely distributed in the environment. Of the 210 possible congeners only 17 of the congeners are thought to exhibit appreciable toxicity. For an in depth discussion of the properties, fate, transport and effects of these compounds the reader is directed to other reviews. There is considerable concern in many areas of the world about the potential adverse effects of exposure to low levels of these pollutants. 1.1 Aims of this Chapter

This chapter of this edition of the Handbook of Environmental Chemistry considers the release of PCDD and PCDF to land, in the form of waste products, to water and in products where PCDD and PCDF may be present as an impurity. Over the past few years the science and policy of controlling PCDD and PCDF has concentrated on reducing or eliminating releases to air and releases to land, water and in products have received comparatively little attention. Cursory review of recent work in many countries seems to indicate only limited releases to water and a relatively low priority for releases to land and in products. However, these countries typically have well developed programmes for assessment and control of PCDD/F and it cannot be assumed that the same situation will exist in other countries. Assumptions need to be reviewed in the light of changing circumstances and new information to ensure that appropriate and effective measures are taken.

Releases of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans

205

This chapter aims to indicate the importance of including an assessment of releases and flows of PCDD/F to water and land as well as in products and residues as part of an integrated approach to PCDD/F. 1.2 Sources of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans

PCDD and PCDF have never been produced intentionally, other than for research, but may be found as a trace by-product in a large number of industrial and non-industrial processes. Detailed studies on the magnitude of sources are available such as those carried out for the UK Government [1, 2], in the US [3] and for the European Commission [4]. There is also a valuable overview of inventories available world-wide in 1999 produced by the United Nations [5]. Other work describes in more detail the nature of processes that can form PCDD/F [6]. The main groups of processes that produce PCDD and PCDF can be grouped as: – Chemical production processes – for example the production of 2,4,5-T, chlorinated phenols and the oxychlorination of mixed feeds to make certain chlorinated solvents – these can generally be addressed and controlled by modifications to the process or by product substitution – Thermal and combustion processes – including incineration of wastes, the combustion of solid and liquid fuels and the thermal processing of metals – Reservoir sources such as historic dumps of contaminated wastes and soils and sediments which have accumulated the chemicals over extended periods – there may also be substantial natural deposits of PCDD/F – Biochemical processes which may form PCDD/F from precursors – there is some evidence of this occurring in compost PCDD/F can be released directly to the environment in emissions to air and water; they can also be present in products, wastes and residues which may in turn be introduced to the environment. 1.3 System Boundaries and Relative Impacts

It is conceptually easy to consider releases of PCDD/F to air from identified processes. Releases to land and water and in products are more complex to address and understand. However, it is relatively simple to consider a single process – take for example a scrap metal melting furnace – and to distinguish possible release routes as follows: to air in flue gases and fugitive emissions, to residues and, possibly, in a liquid effluent. In general, emissions to air after the final stage of cleaning are released directly to the environment. Classifying the final releases to the environment for PCDD/F in products, residues and liquid effluents may be more complex. A variety of residues may be produced by a process. Some of these may be disposed of directly – they may be landfilled untreated, treated off-site (including chemical and thermal treatments which may alter the PCDD/F present) or they

206

P.H. Dyke

may find a beneficial use on land – for example for their nutrient or mineral content; they may even be used as raw materials for further processes. Liquid effluents may be released to open waters (e.g. rivers, lakes, the ocean) or to sewer. Once in the sewerage system the final fate of the PCDD/F will depend on the sewage treatment system and the fate of the residues and products from that system. There is clearly a complex web of possible routes by which PCDD/F formed or released from one process may enter others and for which the eventual fate will be determined by a combination of factors. It is difficult therefore to ensure that no “double counting” of PCDD/F occurs. It is useful to set some system boundaries at the outset.When considered at the process level we consider releases to be those points at which PCDD/F passes outside the process unit. It is possible to quantify the amount of PCDD/F passing by different routes. It is useful to be able to identify the fate of the PCDD/F in broad terms to assist in the first stages of risk assessment. Therefore for the purposes of this chapter a release in residues would be quantified and the fate considered where possible. The primary distinctions are between residues applied to land directly (e.g. some sewage sludge), landfilled, or treated in a way that will affect levels of PCDD/F (e.g. incineration). For liquid effluents the primary distinction is between a release to the environment (e.g. to a river) and a release to sewer where additional treatment is expected. For products the fate of PCDD/F will depend on the use of the product; this can be the crucial factor in determining the magnitude and time-scale of releases. It is beyond the scope of this chapter to carry out a risk assessment of any of the releases. It should be noted that there are enormous differences between the potential for causing exposure to PCDD/F depending on the way the pollutants are released. Therefore the mass of PCDD/F alone can be a poor guide to assessing relative risk. 1.4 Toxic Equivalency Schemes

Much data relating to PCDD/F is presented in terms of toxic equivalent concentrations (TEQ) and although a discussion of the toxicity of PCDD/F is beyond the scope of this chapter it is important to appreciate some aspects of this approach and the use of toxic equivalency factors (TEFs). A number of schemes have been developed to provide a simple system for expressing the overall toxicity of a complex mixture of PCDD/F. These systems work by assigning TEFs to each congener and the TEQ is then the sum of each congener multiplied by the TEF for that congener. This gives a single result broadly equivalent to the amount of 2,3,7,8-TCDD which would have the same effect as the mixture. Two TEF schemes are of primary interest here. The International Toxic Equivalency Factors (I-TEF) and the scheme proposed by the World Health Organization (WHO) in 1997 [7]. Much data on levels of PCDD/F has been expressed as International Toxic Equivalents over the past decade and this is the standard metric, used here and denoted TEQ. The changes to the TEF scheme proposed by

Releases of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans

207

WHO are relatively few but can make a significant difference to the overall TEQ values. This is especially so for sources dominated by the octachlorinated dioxin or furan congeners. In general most data are expressed as International Toxic Equivalents (I-TEQ) and the process of recalculating a WHO-TEQ is hindered if congener specific results are not presented.

2 Why Consider Releases to Land, Water and PCDD/F in Products? As noted above, over recent years the majority of the attention focused on releases of PCDD/F has been on releases to air. There are good reasons for this – there are many sources which emit PCDD/F to air and once in the air the pollutants become widely dispersed and can enter the food chain and thereby contribute to human exposure. Less attention has been paid to releases to water, although significant reductions have been made to releases from pulp bleaching processes used in papermaking. The amount of PCDD/F in residues and in products has received much less attention. In order to ensure that a well balanced approach is taken to reduce and control threats from PCDD/F it is valuable to consider all potential flows. As circumstances and knowledge change the relative importance of different releases to different media can change. In the extreme, PCDD/F present in residues or wastes or even in natural products can cause unacceptable human exposure. A key approach used to reduce releases of PCDD/F to air is the application of pollution control technology. In many cases systems such as wet scrubbers, electrostatic precipitators and fabric filters are effective at removing particulate material from flue gas streams and with it adsorbed PCDD/F. There is therefore a potential to shift PCDD/F from one receiving medium to another, often from air to water or residues. It is also becoming more popular to consider the possibility of recycling and reusing residual materials and wastes. Where materials are used as raw materials in other processes or where they are recovered for reuse it is important to consider whether there is any possibility of inadvertently increasing the probability of exposure to pollutants such as PCDD/F. In many places levels of PCDD/F have been reduced; however there has also been increased concern about the potential effects of long term exposure to low levels of PCDD/F grows (as expressed by a recent reduction in the WHO tolerable daily intake). In these circumstances it is appropriate to review and augment the information on PCDD/F in releases to land, water and in products and residues to determine whether there is cause for concern. It is also worth considering whether, on a holistic basis, the overall potential for the production of persistent pollutants such as PCDD/F could be a factor in comparisons between processes. At the present time the geographical interest in PCDD/F is expanding rapidly away from the relatively small group of developed countries in which most research and action has taken place and into the wider world; in particular this process has been driven by global negotiations on a legally binding agreement on

208

P.H. Dyke

Persistent Organic Pollutants (POPs). As the focus shifts to countries where PCDD/F have not been systematically assessed and addressed it is important to consider all the possible sources and not to assume that the spread or impact of sources will reflect the latest situations in developed countries.

3 Historical Perspective It is worth noting at this point that the current preoccupation with releases of PCDD/F to air is a relatively recent phenomenon. One of the main areas of concern in the early stages of the “dioxin story” was over levels of PCDD/F in products and in particular in chemical products derived from chlorophenols – for example the herbicide 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) and the infamous Agent Orange. Significant quantities of some of these potentially highly contaminated products were used in the past and large amounts of PCDD/F were introduced into the environment by this route. However, a series of controls were introduced both by regulation and by voluntary actions that reduced inputs of PCDD/F into the environment. There are also sites where a large quantity of PCDD/F has been deposited with waste products and residues. For example, in a mass balance study of PCDD/F in Hamburg an estimate was made that more than 378 kg TEQ were present in waste dumps, much arising from historic chlorinated chemical production [8]. In some cases uses were restricted or products banned – for example all uses of 2,4,5-T were banned in the USA in 1983 and in many other countries between 1970 (Italy) and 1985 (Federal Republic of Germany) [6]. In other cases changes could be made to the production process which significantly reduced levels of PCDD/F in the final products. For example, in the USA high levels of PCDD/F (up to 3065 mg kg–1 TEQ) were found in samples of chloranil which was used in dye manufacture; action was taken to ensure an industry wide switch to the use of “low dioxin chloranil” which had levels of PCDD/F more than two orders of magnitude lower (<20 mg kg–1 TEQ) by virtue of feedstock and process changes [3]. The quantities of PCDD/F entering the environment from the use of chemicals will depend on the use and the eventual fate of the product or items treated with the product. Broad-brush estimates made in the late 1980s for the use of tetra- and penta-chlorophenols in the UK suggested an annual input of 4.5 kg TEQ with a further 3.8 kg TEQ of PCDF in PCB used in the UK [9]. Such estimates were made before detailed inventories of PCDD/F releases could be established so were not easily put into context. However, some perspective can be gained by considering that these estimates for PCDD/F in just three chemical products are much larger than estimates of total releases to air from industrial and non-industrial processes in the UK where estimated releases from 28 process categories were in the range 560–1100 g TEQ in 1993 [1]. As the more obvious sources of PCDD/F to the environment are identified and controlled the remaining sources assume greater importance. At the same time our knowledge of the science is developing and, increasingly, special situations are arising where exposure to PCDD/F may result from specific localised prac-

Releases of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans

209

tices rather than as a general background. In this context it is important to take a more structured approach to releases of PCDD/F to water, to land and in products.

4 Information in Existing Inventories There is rapidly increasing interest in many countries of the world in assembling inventories of sources of PCDD/F. A recent study of available inventories [5] showed that despite the high profile of these pollutants in many parts of the world only 15 countries had assembled detailed inventories by the late 1990s. All of these addressed releases to air, only six addressed one or more of releases to water, to land, in products and residues. In each case problems were encountered due to a lack of data. Large differences are apparent in the estimated releases to land and in residues. In part these result from differences in the setting of system boundaries (e.g. the inclusion or exclusion of PCDD/F in material that may be landfilled) and the lack of relevant data. The summary data from the six inventories including data on releases to water, land, in products and residues is shown in Table 1. Some features are prominent from the summary of national inventories. For recent estimates the releases to water are low although much higher releases were estimated in previous years for some countries. Releases to land, in residues and in products tend to be higher and it is clear that that estimating releases is difficult. In part because releases in this form are a “grey area” in terms of inventory compilation, there is no established methodology for assessing when PCDD/F present in wastes, residues and products should be included since it is not always clear when the PCDD/F may actually be released. It is helpful to consider the major reductions that are shown in some countries’ estimates of releases to water. In particular the US and Canada show very large reductions, from 454 to just 5 g TEQ per year for Canada between 1990 and 1997 and from 356 to 20 g TEQ per year for the USA between 1987 and 1995. In both these cases the values are driven by releases from pulp and paper production. In the USA this was the only source with a quantified release to water; in Canada releases from pulp and paper and chemical production were quantified. In both the US and Canada the reductions in the release of PCDD/F to water from pulp and paper production have resulted from the change in process technology for bleaching. A major shift has occurred with a change from the use of elemental chlorine in the bleaching process to the use of chlorine dioxide and, in some cases, total chlorine-free bleaching. Regulations and monitoring of effluents from pulp and paper mills are largely based on measuring and controlling amounts of just two congeners – 2,3,7,8-TCDD and 2,3,7,8-TCDF – since these were found to dominate the contribution to TEQ [3]. Estimated releases are based on analysis of these two congeners only. It is assumed therefore that process changes have not significantly altered the congener profile of the releases and that releases based on these two congeners alone will underestimate actual releases by an unquantified amount.

210

P.H. Dyke

Table 1. Summary releases of PCDD/F to all media for national inventories Country

Year

Annual release – g TEQ per year

Notes

Air

Water

Land

Residue Product

1985

850

6.21

N/D

347

N/D

1990 1995

892 662

6.87 3.77

N/D N/D

447 485

N/D N/D

Denmark [5] 1995

50

0.1

2

175

N/D

The Netherlands [29]

1991

484

3

N/D

1050

N/D

Sweden [5]

1990 1993

32–115 22–88

2–5 2–5

N/D N/D

35–52 35–52

5–11 5–11

Belgium [5]

Residue PCDD/F denoted “to waste”

Soil release is from sewage sludge and compost only

United King- 1993 dom [1, 2]

560–1,100 N/E

1500– 12000

See land N/D

Release to land includes PCDD/F present in residues residues and wastes that are landfilled. Insufficient data to estimate overall release to water

Canada [30]

1990

353

454

173

N/D

226

Water release based on pulp and paper and chemical production only, soil release from utility poles and railroad ties

1997 1999

290 200

5 5

173 173

N/D N/D

226 226

1987

12,000

356

221

36,600

1995

2,745

20

208

25,050

USA [3]

Release in products is dominated by PCDD/F in PCP

It is clear that estimates of releases in residues, wastes or to land (in the case of the UK where the legal definition of a release to land includes material that is sent to landfill) are usually significantly higher than estimates of releases to water and are, in many cases, comparable to or higher than estimated releases to air. Wide variations in the methodology and coverage of the estimates are found between different country inventories which may explain some of the discrepancies. In the UK the system of regulation of industrial processes addresses releases of pollutants to air, land and water in an integrated way designed to minimise overall adverse impacts, to simplify regulation and to ensure that a pollutant is not simply transferred from one medium to another. Such an approach has now

Releases of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans

211

been extended to a European level with the introduction of Integrated Pollution Prevention and Control. Having assembled information on the releases of PCDD/F to air [1, 10] there was a need to consider releases of PCDD/F to land and water in order to make informed decisions about overall impacts of different processes. The UK review of releases to land and water was published in 1997 and addressed releases from a wide range of processes [2, 11]. The broad definition of a release to land to include materials that are landfilled increases the number of potential sources of release compared to the alternative, narrower, definition of releases to land in the sense of the open environment. Even though the intention in the UK study was to make estimates for as many processes as possible, even where there was very limited data, there were several sources which could not be quantified even approximately.A ranking system was used to indicate the quality of the estimates.A total of 50 process categories were considered and estimates made where possible. Important sources of PCDD/F to land included the incineration of municipal waste (520–2400 g TEQ per year), per- and trichloroethylene production (350–630 g TEQ per year), the use of chemicals (100–3200 g TEQ per year) and the non-ferrous metal industry (150–480 g TEQ per year). Other potentially significant sources which clearly required further assessment were accidental fires (7.5–2400 g TEQ per year) and pesticides production (8.9–2000 g TEQ per year); for both of these the large uncertainty is reflected in the wide range of estimated releases. It is interesting to consider each of these sources in a little more detail. The incineration of municipal waste has long been identified as a potentially significant source of PCDD/F and considerable research has been dedicated to developing both detailed emissions data as well as technologies and techniques to reduce and control releases. In the UK in 1993, the year which the estimates of releases refer to, the incineration industry was entering a period of profound change as new emission standards resulting from the European Commission Directives on emissions from municipal incineration plants were being imposed. In 1993 most UK plants were based on old designs and had only basic flue gas cleaning systems which were designed to reduce emissions of dust but not specifically acid gases or PCDD/F. There was a single new plant which had been designed to meet new plant standards and stringent emissions limits. The bulk of the estimated release to land was in the form of fly ash from old plants (500–2300 g TEQ per year), with grate ash from the same plants accounting for only 9–54 g TEQ per year. Releases from the new plant were estimated at 1.5–9.1 g TEQ per year in grate ash and 13–29 g TEQ per year in the flue gas cleaning residues. To compare these figures it is useful to express them on the basis of release per unit waste burned. Table 2 shows the relevant figures. The major impact of measures to reduce both formation of PCDD/F (indicated by the sum of the emission factors) and the improvements in capture efficiency of the gas cleaning equipment (indicated by the change in emission factor to air) are clear. The estimate of releases of PCDD/F from the production of per- and trichloroethylene are based on figures from a single manufacturing facility in the UK. This process was part of an integrated chemical production site where ethylene dichloride (EDC) is made for the production of PVC and the residues

212

P.H. Dyke

Table 2. Estimated release of PCDD/F for UK incineration plants in 1993 per unit of waste

burned Air – mg Grate ash – mg TEQ per tonne TEQ per tonne Old plants a 180–230 0.8 New plant b a b

4–20 4–20

Fly ash – mg Gas cleaning residue Sum TEQ per tonne – mg TEQ per tonne 200–900 –

– 30–70

380–1200 35–90

Old plants typically were fitted with electrostatic precipitators only. New plant fitted with spray-dryer fabric filter system with activated carbon for PCDD/F control.

from the oxychlorinator were treated in a further oxychlorination step to yield chlorinated solvents. The residues from this process were taken to secure disposal in underground salt caverns and were estimated to contain 350–630 g TEQ per year of PCDD/F. Since the PCDD/F are largely contained in residues from this process the key issue here in preventing inadvertent exposure to the PCDD/F has been careful control over the handling and disposal of the residues. High levels of PCDD/F formed in the production of these solvents has been linked to the by-products arising from the oxychlorination step in the production of EDC and process changes are expected to reduce or eliminate this production. Furthermore, improvements in the management of the wastes from the process are expected to include incineration of contaminated residues rather than their disposal to land; consequently the source should now be reduced to near zero. Making estimates of the PCDD/F introduced to the environment from the use of chemicals is not simple. The UK estimates were based on the amount of PCDD/F estimated to be present in pentachlorophenol (PCP) used for wood treatment in the UK and PCDD/F that might be present in other pesticides. Little information is available on the levels of PCDD/F in formulations in use in the UK; it is known that some production processes can give rise to contamination with PCDD/F and it is hoped that more analytical data will become available so that this potential source can be better quantified. The largest flow of PCDD/F in chemicals in use that could be addressed related to the use of PCP. The use of PCP has been decreasing in the UK and the main remaining use is for commercial timber treatment. UK use of PCP for this purpose decreased from 1290 tonnes in 1974 to 290 tonnes in 1992. However, it has been estimated that the main flow of PCP into the UK is with wood treated with PCP outside the UK and that this adds a further 1000 tonnes per year of PCP to the annual input to the UK [12]. In recognition of the potential for contamination, controls have been placed on the maximum levels of contamination in PCP that can be used. In Europe a limit of four parts per million (ppm) of total hexachlorodibenzodioxin (HxCDD) is set, and in the US a limit of 2 ppm of total HxCDD was imposed in 1987 expressed as a monthly average with a batch limit of 4 ppm [13]. If imported PCP comes from the US it is likely to have comparatively low levels of PCDD/F in it; levels in PCP from other parts of the world may be different and could be higher.

Releases of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans

213

When looking at the use of PCP the issues of system boundaries and what constitutes a release become very important. Three routes of release should be considered – releases at the point of application, gradual release while wood is in use and potential releases when treated timber comes to the end of its service life. Typically the PCDD/F in PCP used to treat wood is expected to remain relatively tightly bound to the wood. Various estimates have been made for rates of evaporation – usually in an attempt to include a value in an emissions inventory, and studies are now underway in the US which aim to assess the fate of PCDD/F in treated utility poles [14]. Large variations in estimates of the importance of PCDD/F in PCP will result from the assumptions used to estimate the release from the “reservoir” in treated wood. At this stage it seems to be useful to quantify the amount of PCDD/F that is contained in PCP and to note the potential for this to be released to air, land and water perhaps at some future time. It is possible that the presence of PCDD/F as a contaminant and the presence of PCP itself will impact on decisions about the best way to handle and dispose of waste wood. In the US inventory [3] an estimation of the amount of PCDD/F in PCP used in the US suggests that 8400 tonnes was used in 1994 down from 12,000 tonnes in 1987. Using an average concentration of 3 mg kg–1 TEQ an estimated 25,000 g TEQ of PCDD/F were present in the PCP used in 1994 and 36,000 g TEQ in the PCP used in 1987. EPA carefully distinguish this flow from other categories in recognition that it is not a direct release to the environment. Clearly, given the large scale of this estimated annual flow of PCDD/F and the amount of PCDD/F estimated to be present in wood in service (468 kg TEQ in 1996 [14]) it is important to study the fate further. It may also be necessary to assess levels of PCDD/F in PCP from a variety of sources and to consider the quantity and fate of PCDD/F present in products treated with PCP before import (see the discussion on PCDD/F in sewage sludge below). Unregulated combustion sources could also contribute significantly to releases of PCDD/F to land and water. Unfortunately releases from events such as accidental fires in buildings, factories, production facilities, chemical stores, vehicles and open burning of wastes are notoriously difficult to quantify. The processes are highly variable and it is likely that formation of PCDD/F will vary considerably from one event to another; the poor combustion conditions and the wide range of materials involved in such fires can give rise to high concentrations of PCDD/F in emissions and residues. As well as releases to air it is likely, in many cases, that PCDD/F will be released to land and water. A release to land can result from the disposal or dispersion of residues from such fires and a release to water can occur with run-off of rainwater on open sites or run-off of water used to fight a fire. Once again the UK study took the approach of attempting to make an estimate of possible releases using the limited information available. This is an area in which the data available relating to accidental fires are not immediately applicable to estimating pollutant releases to the environment. Information was found indicating the number of fires attended by the fire service but this gave no indication of the size, severity or of the materials involved. Estimates of the average amount of residues had to be made. Similarly broad assumptions had to be made regarding the levels of contamination of the residues from such fires.Available test data tends to be focused on PCDD/F formation from individual com-

214

P.H. Dyke

ponents of fires – for example laboratory tests on the combustion of PVC [15] or on individual chemical combustion [16]. Real fire test data often relate to the amount of PCDD/F deposited on surfaces which is relevant in making clean-up and remediation decisions but is of limited usefulness in estimating total amounts of PCDD/F formed.

5 Special Cases and Double Counting As noted in the introduction there are a number of grey areas in the compilation of inventories that consider releases of PCDD/F to water, land, in products and in residues.An important one is where PCDD/F may not be formed in the process under consideration but simply enter the process from other sources, for example contained in raw materials. Two examples are discussed to illustrate the importance of assessing the flows of PCDD/F despite the likelihood that most was not formed within such processes. 5.1 Sewage Sludge

The presence of PCDD/F in sludges arising from the treatment of waste waters is well established. There is a particular interest in the levels of PCDD/F in sewage sludges because of the attraction of using some sludge as a soil amendment or fertiliser and the possibility that this practice could contribute to human exposure to PCDD/F. The application of sludge containing PCDD/F to agricultural land opens up the possibility for PCDD/F to enter the foodchain via transfer to plants or to grazing animals. Surveys of levels of PCDD/F in sludges have been carried out in a number of countries. Results have indicated that a wide range of contamination can be expected, in some surveys levels generally seem to be falling over time and that the total amount of PCDD/F in sludge applied to land can represent one of the major flows of PCDD/F released to land. In the UK 14–56 g TEQ per year were estimated to be released [2] and in the US an estimated 208 g TEQ per year of PCDD/F are present in sludge which is not incinerated and of this 106 g TEQ is applied to land [3]. The release of PCDD/F to water from sewage treatment may also represent a significant source and has received relatively little attention. In the US preliminary estimates suggested that between 13 and 170 g TEQ may be released to water by sewage treatment, making this potentially a very major source of PCDD/F to water [3]. The sources of PCDD/F in the sludges are not completely understood although a wide range of sources has been identified which could contribute to the presence of PCDD/F in sludge. The main categories of sources can be grouped under the headings: industrial effluents; surface run-off (including atmospheric deposition); household waste water; and formation within a treatment works [17]. The mix of sources is likely to vary from one sewage treatment plant to another, making it difficult both to assess the main causes of contamination and to take measures to reduce levels of PCDD/F in sludge. Over time it is likely that both the

Releases of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans

215

scale and the balance of sources of PCDD/F in sludge will change. Measures impacting on sludge levels include regulations controlling PCDD/F in products, such as PCP, that may enter the sewage system, direct controls on effluents released to sewer and indirect effects from limits on emissions to air which can, in turn, reduce deposition and run-off into the sewage system. An interesting series of studies carried out in Germany showed that household waste water was contributing significant amounts of PCDD/F to the effluents studied and that the ultimate source was the laundering of textiles [18]. Further work indicated that PCP use was the major source of PCDD/F in textiles although other sources were also found. For one treatment works the load of PCDD/F from textiles could account for all of the PCDD/F entering the works. In Germany where comprehensive measures have been taken to address sources of PCDD/F it was estimated that textiles could account for all the higher chlorinated congeners of PCDD/F found in sewage sludge; other sources were thought to be responsible for levels of the lower chlorinated congeners [17]. As well as controls which reduce levels of PCDD/F in sludge the other approach that has been used to minimise potential exposure via this route has been to develop controls over the way that sludge may be applied or setting maximum acceptable levels of PCDD/F in sludge. Management measures for land application such as defining acceptable crop types, timing and rates of applications can be highly effective in reducing the likelihood of any significant effect on levels of PCDD/F in agricultural produce from land where sludge is applied. There is a good argument to be made that PCDD/F present in sewage sludge and released to water from sludge treatment is largely derived from other processes which release PCDD/F in effluents or to air and via run-off to sewer. Whilst it is important to appreciate that this is likely to be the case and it is useful to note it, to exclude consideration of releases via sewage sludge in an inventory would be misleading and could hinder effective management decisions. 5.2 Wastes and Waste Management

It can be easy to associate PCDD/F with wastes although the link is often limited to waste incineration and some wastes from chemical production. It can also be important to consider the PCDD/F content of other wastes and the fate of PCDD/F in waste management operations.Where PCDD/F are present in a waste or residue it may be important to consider the fate and impact of this contamination in downstream processes using the waste. Examples are used here of municipal waste and some residues from the metal processing industry to illustrate aspects of the potential for transfer of PCDD/F with wastes. Although there is no indication that municipal waste itself can be classed as a source of PCDD/F in terms of forming the pollutants, there may be a significant quantity of PCDD/F in the mixed waste. A detailed test programme in the UK gave levels of PCDD/F in various waste fractions and from three locations. Selected results from this study are shown in Table 3 [19]. The level of PCDD/F in mixed municipal waste may have implications for releases from waste treatment operations and in some cases may lead to a level of contamination in other

216

P.H. Dyke

Table 3. Levels of PCDD/F in UK household waste fractions

Waste fraction

Proportion of UK household waste (mid-1990s), % by weight

Paper and cardboard Plastics (all) Textiles Combustibles Misc. non-combustibles Glass Total metals Putrescibles Fines Total waste

31.6 9.2 2.4 7.7 3.7 7.4 7.2 19.6 11.1 –

PCDD/F mg TEQ per tonne wet weight 4.5 (12 samples) 8.9 (15 samples)

}

7.9 (2 samples)

}

3.4 (3 samples) 9.5 (23 samples) 14.7 (13 samples) 6.3 (13 samples)

processes. Quite significant amounts of PCDD/F may be present in household wastes taken in totality – for example in the UK an estimated 150 g TEQ per year is present in municipal waste which is landfilled [2]. In many waste management systems at the moment there is pressure to increase rates of product reuse and recycling. Whilst in many cases this will bring benefits there may be examples where certain wastes will introduce PCDD/F into a process and where the fate and impact of these should be assessed as part of the overall assessment of the desirability of the process. The metals industry has examples of situations where residues from one process may be used as raw material for another process. For example dusts from bag filters used on electric arc furnaces are sometimes processed to recover zinc. In the UK an estimated 59 g TEQ of PCDD/F is contained in the dusts from electric arc furnaces [2] and it is important to consider the fate of this contamination in subsequent processing steps. Such potentially inadvertent transfers of PCDD/F may lead to high releases from the subsequent processing and it may be that the process itself has not been identified as likely to lead to high levels of PCDD/F and regulations may not be applied. This area is complex and may change as new processes are developed and more recycling and recovery takes place. Note the examples of contamination of citrus pulp pellets and Belgian produce in Sect. 7.

6 Importance of Managing Intermediates, Wastes and Effluents Effectively The overall impact of a process which forms or has PCDD/F present within it (e.g. as a contaminant in an intermediate or in a raw material or residue) will depend to a large degree on the techniques used to control and prevent releases to the environment. Two examples can illustrate this point and these are discussed below. From the point of view of quantifying releases from processes – i.e. compiling an inventory – it is crucially important that the way that intermediates, effluents and wastes are managed is understood in order that appropriate emission factors may be applied.

217

Releases of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans

6.1 PVC Production and Related Activities

The PVC production industry in the US has been the subject of a major study to identify and quantify releases of PCDD/F. The study was initiated in response to concern expressed that the potential existed for the formation of high levels of PCDD/F within process steps and that these might lead to unacceptable releases to the environment. For the purposes of this study PCDD/F was identified and quantified where it left the control of the manufacturing facility – therefore in products, wastes and releases to air. PCDD/F that was present in materials entirely within the facilities and therefore under control was not considered a release and was not quantified. Such an approach can be entirely appropriate where high standards of internal management of materials can be ensured; it allows the assessment of releases both to the environment and transfers in materials, wastes and products that pass to the control of others. However, in other circumstances, if there is significant concern that control over materials within such a process may be lost then the PCDD/F in them should be considered also. Therefore it cannot be assumed without evidence that the releases from a well controlled process will be representative of all processes for the purposes of estimating releases. The flows evaluated for the production of EDC, vinyl chloride monomer (VCM) and PVC are shown in Table 4 [20].

Table 4. Estimated annual releases from US EDC/VCM/PVC manufacture

Estimated mass (g I-TEQ per year) Emissions Water – treated waste water from PVC only production sites Water – treated waste water from integrated sites Air – PVC only plants, vent gas incinerators Air – EDC/VCM liquid and liquid/vent incinerators Air – EDC/VCM vent gas incinerators Air – from third party incinerators (taking wastes from processes) Air – suspension resin dryers

0.15 0.17 0.0014 3.7 6.9 0.65 <0.2

Secure landfill Waste water treatment plant solids Spent catalyst

12.1 6.7

Products Suspension and mass PVC resin Dispersion PVC resin EDC HCl

3 0.1 0.29 0.0044

Other

0.005

Total annual release

34.0

218

P.H. Dyke

It is interesting to compare the process that was carried out in the UK and described above in the discussion of releases from per- and trichloroethylene production with the processes examined in the US study on the PVC industry. In the UK the plant producing per- and trichloroethylene, as assessed in 1993, was linked to the plant producing EDC for PVC production. After the oxychlorinator which made EDC the heavy ends were passed to the oxychlorinator for the production of per- and trichloroethylene. Levels of PCDD/F measured in these heavy ends were in the order of 3100–7500 ng I-TEQ kg–1 and carried 12–30 g I-TEQ per year of PCDD/F into the oxychlorinator for per- and trichloroethylene production [21]. Two residue streams from the second oxychlorinator had concentrations of 31,000–38,000 ng I-TEQ kg–1 and 43,000–90,000 ng I-TEQ kg–1 and transferred 63–75 g I-TEQ per year and 220–450 g I-TEQ per year to secure containment disposal [21]. Clearly the oxychlorination step in the process to make PVC has the potential to lead to high levels of PCDD/F (apparently primarily associated with the heavy ends residue stream) in terms of a release to the environment; the way in which this stream is handled and treated is crucial. In the US typical practice is high temperature incineration whereas in the plant in the UK the use of this residue as raw material for per- and trichloroethylene production in an oxychlorinator and with subsequent disposal of the residue stream to containment as opposed to destruction led to high flows of PCDD/F. Therefore it is important to consider carefully the handling and treatment of such potentially contaminated flows. 6.2 Magnesium and Nickel Production in Norway

The production of metals has been associated with the formation and release of PCDD/F for some time and increasingly this sector is having to address releases. Once again most attention is being focused on releases of PCDD/F to air since there is a wide range of thermal processes which can lead to air emissions. It is nevertheless important to consider the possibility of significant releases to land (especially with residues) and to water. Two examples of releases to water which were identified are used as a reminder to consider these processes carefully and for releases to all media. Studies in Norway conducted in the late 1980s examined releases from the production of magnesium and nickel [22]. Magnesium was produced from the electrolysis of magnesium chloride which in turn was formed by heating pellets of magnesium oxide and coke in an atmosphere of chlorine to 700–800 °C. The chlorine gas was scrubbed with sea water scrubbers and the liquid effluent released to the sea. An estimated 500 g TEQ per year (Nordic TEF scheme) was released to water until the plant was improved and water treatment installed [23]. High levels of environmental contamination were found in the region of a nickel production process; in this case the contamination was linked to a process which was no longer used. This process converted nickel chloride to nickel oxide in a fluidised bed reactor at 800 °C with air and propane; sea water scrubbing cleaned the flue gases. Again the implications are that the potential for releases to water from such processes should not be ignored and the presence of water treatment can make a large

Releases of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans

219

difference to the amount of PCDD/F actually released (the fate of the residues from the treatment should be considered).

7 Clay, Citrus Pulp Pellets and PCB Oils Perhaps the clearest illustration of the need to consider the presence of PCDD/F in wastes, residues and products is seen when specific incidents of contamination occur. These incidents provide good reason to start with a broad approach to the assessment of flows and transfers of PCDD/F. The potential for problems to arise from unforeseen circumstances can be reduced by full consideration of the flows of PCDD/F. Three recent examples will be touched on briefly to illustrate the point here – the unexpected discovery of PCDD/F in clays, high levels of PCDD/F in citrus pulp pellets and the entry of PCDD/F into the food chain in Belgium in 1999. In all three cases a relatively small amount of PCDD/F has a disproportionate effect by virtue of effectively short-circuiting the route between source and exposure, in these cases by introduction to animal feeds. Studies on PCDD/F in poultry and catfish found elevated levels of PCDD/F in some samples which triggered an investigation to identify the source [24]. By examining and sampling for PCDD/F contamination at the farms where the poultry were produced, contamination was found to be associated with the feed. In turn studies showed that the elevated levels of PCDD/F were due to the use of “ball clay” as an anti-caking agent. The patterns of contamination in the clay were dominated by PCDD and high levels of 2,3,7,8-TCDD [25]. No anthropogenic source of the contamination has been identified and the tentative conclusion is that large quantities of PCDD/F have been deposited in the clay as a result of natural processes. Other recent work has found high levels of hexa- and octachlorinated dibenzodioxin in sediment from lakes in the southern Mississippi area and suggested that the evidence points to a natural source [26]. Evidence has also been gathered over recent months that PCDD/F may be found in clay in Europe and the European Commission has moved to ensure that no contamination from this source should enter animal feeds and hence the human food supply. Although a great deal more work is required to understand fully the findings that PCDD/F can be present at elevated levels in clays there may be real implications for human exposure in certain situations. It will be useful to assess the flows of PCDD/F in such situations to ensure that potential releases to the environment or other possible routes of exposure are adequately accounted for. In 1998 monitoring of food products in Baden-Würtemberg in south-western Germany showed elevated levels of PCDD/F. Investigations traced these high levels to contamination of citrus pulp pellets coming from Brazil which were used as animal feed [27]. Citrus pulp pellets are a by-product from orange juice production and a significant market had built up with approximately 1.5 million tonnes of pellets being sold per year worth US $100–150 million [27]. Studies to trace the source of contamination in the pellets indicated that lime originating as a by-product from a production process used in the manufacture of the pellets was introducing the PCDD/F and that contamination was not present in all lime supplies.Action to restrict and prevent any contamination included the im-

220

P.H. Dyke

position of a limit on the level of PCDD/F in citrus pulp pellets being imported to the European Union (500 pg TEQ kg–1) and a requirement that only “virgin” lime be used in manufacture of the pellets. Further strict controls were introduced to ensure that aspects of the lime production would not lead to contamination [27]. In Belgium in January 1999 a stock of recycled fat was inadvertently contaminated by PCB and PCDD/F (predominantly PCDF in this case). The contaminated fat was used in the manufacture of animal feeds and the first signs of problems were seen in dropping egg production, followed by increased chick mortality, reduced weight and reduced hatching success for eggs. Several hundred farms were affected. In order to control and minimise the spread of contamination stringent restrictions were placed on the trade in selected agricultural products from Belgium and a system of monitoring on farms and of food products was introduced. A series of Belgian Government and European Commission actions were put in place to respond to the crisis and to restore confidence in the agricultural system. The total amount of PCB and PCDD/F involved in the contamination incident has been estimated at 50 kg PCB and 1 g TEQ of PCDD/F [28]. Despite the small amount of PCB and PCDD/F involved there were far reaching impacts including very large direct and indirect costs as well as measures taken affecting agriculture, the animal feed and vegetable oil recycling industries. Consumer confidence in Belgium and beyond has been damaged and the effects will be felt for many years to come. The suggestion has been made that a relatively small batch of a PCB oil containing PCDF was illegally placed into a collection vessel designed to collect vegetable oils for recycling. If this proves to be the case it illustrates again the dangers of improper management of materials containing PCDD/F and the need for systems to be suitably designed to prevent or at the least respond rapidly to such events.

8 Conclusions The systematic assessment of releases of PCDD/F to water, land, in residues and in products has not been as well developed as the assessment and estimation of releases to air. Whilst there are good reasons to assess and address releases of PCDD/F to air, releases to water, land and the presence of PCDD/F in residues and products should not be overlooked. In many countries a series of measures have been implemented to control and reduce releases of PCDD/F to air. In some countries substantial reductions have also been achieved in releases to water although releases to water, in products and in residues often appear to have been addressed on a case-by-case basis rather than considered as a whole across a country. In countries where significant reductions of releases of PCDD/F to air have been achieved, the focus of attention can shift to less obvious or more localised sources of PCDD/F. There may be a need to reconsider the potential for releases to water, land and in products and residues to contribute to exposure. In addition, exposure to PCDD/F is increasingly of global interest and many countries which have not systematically addressed PCDD/F are now beginning

Releases of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans

221

to assess sources, exposure and to develop appropriate control strategies. It is important that all relevant sources are assessed in order to have a sound base from which to develop effective strategies and it is likely that the balance and impact of different sources will not be the same as those developed countries which have assessed PCDD/F releases. At the present time there is no widely accepted methodology for establishing inventories of releases of PCDD/F to water, land, residues and in products. It would be valuable to promote the development of a suitable, simple but comprehensive approach to identify and quantify such releases. This should be sufficiently broad to ensure that transfers of PCDD/F into and out of processes are caught even where the processes themselves may not form PCDD/F. It should also cater for transfers which may result from materials like clays. The impact of PCDD/F can be greatly influenced by the ultimate fate and handling of materials containing PCDD/F. The relevant characteristics of processes should be assessed and the fate indicated when quantifying flows of PCDD/F for an inventory. Clearly further work is needed to assess the potential for exposure to result from releases of PCDD/F to media other than air. In particular a framework is required to assist in the assessment of the potential for adverse effects to result from the presence of significant amounts of PCDD/F in products such as PCP and dumps of wastes such as manufacturing wastes from 2,4,5-T production. The examples of contamination of the human food supply as a result of entry of PCB oil in Belgium and clay with high levels of PCDD/F to the animal feed supply chain serve to illustrate the need to both track transfers of PCDD/F in products and residues and show the need to go beyond simply quantifying the amounts of PCDD/F in particular flows.

9 References 1. Environment Agency (1995) A review of dioxin emissions in the UK. DOE/HMIP/RR/ 95/004, Environment Agency, London 2. Environment Agency (1997) A review of dioxin releases to land and water in the UK. Environment Agency R&D Publication 3, Environment Agency, London 3. US EPA (1998) The inventory of sources of dioxin in the United States. EPA 600/600/P98/002Aa, Review Draft 4. Landesumweltamt Nordrhein-Westfalen (1997) Identification of relevant industrial sources of dioxins and furans in Europe. Report No 43, Essen. 5. UNEP Chemicals (1999) Dioxin and furan inventories, national and regional emissions of PCDD/PCDF. UNEP Chemicals, May 1999 6. Fiedler H, Hutzinger O (1990) Toxicol Environ Chem 29:157 7. Van den Berg, Birnbaum LS, Bosveld ATC, Brunström B, Cook P, Feeley M, Giesy JP, Hanberg A, Hasegawa R, Kennedy SW, Kubiak T, Larsen JC, van Leeuwen FXR, Liem AKD, Nolt C, Peterson RE, Poellinger L, Safe SH, Schrenk D, Tillit D, Tysklind M, Younes M, Wærn F, Zacharewski T (1998) Toxic equivalency factors for PCBs, PCDDs and PCDFs for humans and wildlife. Environ Health Perspect 106(12): 775–792 8. Freie und Hansestadt (1995) Dioxin-Bilanz für Hamburg, Hamburger Umweltberichte 51/95 Freie und Hansestadt Hamburg, ISSN 0179–8510 9. Eduljee GH (1988) Chem Br 24:1223 10. Eduljee GH, Dyke PH (1996) Sci Total Environ 177:303

222 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

22. 23. 24. 25. 26. 27. 28. 29. 30.

P.H. Dyke Dyke PH, Foan C, Wenborn M, Coleman PJ (1997) Sci Total Environ 207:119 Wild SR, Harrad SJ, Jones KC (1992) Chemosphere 24:833 Eduljee GH (1999) Sci Total Environ 232:193 Winters DL, Barton RG, Boggess KE, Davis M, Alburty DS, Lorber MN (1999) Organohal Compd 41:35 Theisen J, Funcke W, Balfanz E, König J (1989) Chemosphere 19:423 Vikelsøe J, Johansen E (2000) Chemosphere 40:165 McLachlan MS, Horstmann M, Hinkel M (1996) Sci Total Environ 185:109 Horstmann M, McLachlan MS (1995) Chemosphere 31:2887 Eduljee GH, Dyke P, Cains PW (1997) Chemosphere 34:1615 Carroll WF Jr, Berger TC, Borelli FE, Jacobs RA, Ledvina J, Lewis JW, McCreedy RL, Smith TP, Tuhovak DR, Weston AF (1999) Organohal Compd 41:31 Pumphrey NWJ (1994) Formation of dioxins in oxychlorination, significance for human health and monitoring proposals. Report to the Chief Inspector HMIP Authorization AK6039 Improvement Condition Part 8, Table 8.1 Item 2. Chemicals and Polymers Ltd, Safety and Environment Department, Runcorn Oehme M, Manø S, Bjerke B (1989) Chemosphere 18:1379 Musdalslien UI, Nøkleby PH, Wallevik O (1998) Organohal Compd 36:81 Winters DL (2000) US Environmental Protection Agency. Personal communication Ferrario J, McDaniel D, Byrne C (1999) Organohal Compd 40:95 Rappe C, Bergek S, Andersson R, Cooper K, Fiedler H, Bopp R, Howell F, Bonner M (1999) Organohal Compd 43:111 Malisch R, Berger B, Verstraete F (1999) Organohal Compd 41:51 Bernard A, Hermans C, Broeckhaert F, De Poorter G, De Cock A, Houlns G (1999) Nature 401:231 Bremmer HJ, Troost LM, Kuipers G, de Koning J, Sein AA (1994) Emissions of dioxins in The Netherlands. Rep 770501018, TNO, The Netherlands Environment Canada (1999) Dioxins and furans and hexachlorobenzene inventory of releases. Environment Canada

CHAPTER 8

Toxicology and Risk Assessment of POPs S. Safe Department of Veterinary Physiology and Pharmacology, Texas A&M University, 4466 TAMU, College Station, TX 77843-4466, USA E-mail: [email protected]

Polyhalogenated industrial compounds and combustion by-products including the polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) have been identified in almost every component of the global ecosystem including fish, wildlife and humans. Many of these compounds such as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) induce a common pattern of biochemical and toxic responses that are mediated through initial binding to the aryl hydrocarbon receptor (AhR). These compounds primarily differ in their potencies, and risk/hazard assessment for these AhR-active contaminants can be determined using a toxic equivalency factor (TEF) approach where TCDD equivalents (TEQ) of a mixture is equal to the sum of the concentrations of individual components times their potency (TEF) relative to TCDD.A number of studies also demonstrate that AhR-inactive POPs including orthosubstituted PCBs and hydroxy-PCBs also induce multiple toxic and biochemical responses; however, methods for risk assessment of these chemicals have not yet been developed. Keywords: Toxicology, Risk Assessment, POPs

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224

1

Introduction

2

Toxicology of Different Classes of POPs

2.1 2.1.1 2.1.2 2.1.3 2.2 2.2.1 2.2.2 2.2.3

Aryl Hydrocarbon Receptor (AhR)-Active Compounds PCDDs and PCDFs . . . . . . . . . . . . . . . . . . . PCBs and Related Compounds . . . . . . . . . . . . Problems Associated with TEF/TEQ Approach for Risk Assessment . . . . . . . . . . . . . . . . . . . . . . . AhR-Inactive Compounds . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . PCBs . . . . . . . . . . . . . . . . . . . . . . . . . . Hydroxy-PCBs . . . . . . . . . . . . . . . . . . . . .

3

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232

4

References

. . . . . . . . . . . . . . 226 . . . . . . 226 . . . . . . . 226 . . . . . . . 228 . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

229 230 230 231 231

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

224

S. Safe

List of Abbreviations 2,4,5-T 2,4-D AhR Arnt DREs ERα MC PAHs PB PBBs PBDEs PCB PCDDs PCDEs PCDFs PCNs PCTs POPs TCDD TEF TEQs

2,4,5-trichlorophenoxyacetic acid 2,4-dichlorophenoxyacetic acid aryl hydrocarbon receptor AhR nuclear translocator dioxin responsive elements estrogen receptor α 3-methylcholanthrene polynuclear aromatic hydrocarbons phenobarbital polybrominated biphenyls polybrominated diphenylethers polychlorinated biphenyl polychlorinated dibenzo-p-dioxins polychlorinated diphenyl ethers polychlorinated dibenzofurans polychlorinated naphthalenes polychlorinated terphenyls persistent organic pollutants 2,3,7,8-tetrachlorodibenzo-p-dioxin toxic equivalency factor toxic equivalents

1 Introduction Synthetic or petroleum-derived organic chemicals are major precursors for production of the multitude of plastic products, drugs, pesticides, specialty chemicals for computational/informational equipment and a host of other materials that contribute to the high standard of living and health in developed countries. Some of these organic chemicals, particularly those that contain chlorine or other halogen substituents, have been identified in the environment and are now classified as persistent organic pollutants (POPs) [1]. The pesticide DDT and its major metabolite DDE were among the first POPs identified in environmental samples and, in 1996, Soren Jensen, a Swedish environmental scientist, identified polychlorinated biphenyl (PCB) mixtures in extracts from environmental, wildlife and human samples [2]. Other POPs are also routinely detected in these extracts [3–5]. POPs that have been produced for various industrial applications include: DDT (a pesticide); PCBs; polychlorinated naphthalenes (PCNs), terphenyls (PCTs) and diphenyl ethers (PCDEs); polybrominated biphenyls (PBBs) and diphenylethers (PBDEs); organochlorine pesticides such as toxaphene; and chlorinated cyclodiene-derived compounds (e.g., dieldrin, endrin, and endosulfan); and lindane isomers. Moreover, combustion of organic material containing chlorine also results in formation of polychlorinated dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) and benzenes, and their brominated and mixed

Toxicology and Risk Assessment of POPs

225

bromo/chloro analogs have also been detected in some samples. Combustion of municipal wastes and various fuels are major sources for release of POPs (particularly PCDDs/PCDFs) into the environment [6–8]. PCDDs are also by-products formed during the production of chlorinated phenols and their derived products, including the herbicides 2,4-dichlorophenoxyacetic acid (2,4-D) and 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) [9–13]. A major problem associated with environmental and human exposures to PCDDs and PCDFs is the high toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related 2,3,7,8-substituted PCDDs and PCDFs as well as other POPs that exhibit TCDD-like activity (Fig. 1). Environmental problems associated with POPs are related to their physicochemical properties, namely chemical stability and lipophilicity. After their introduction into the environment, most POPs are stable and resistant to chemical and biodegradation and due to their fat-soluble properties, POPs also bioaccumulate in the food chain [13]. Moreover, POPs that are introduced into the atmosphere can be transported from sites of origin to more distal locations and patterns of atmospheric transport of POPs have been well-documented [14]. Thus, residues of PCBs, PCDDs, PCDFs, DDT, and other POPs are routinely detected in fish, wildlife, and human adipose tissue, serum, and milk [3, 4, 15, 16]. The identification of DDT and other POPs as environmental contaminants resulted in the restricted use or banning of production of many POPs in most developed nations; however, DDT is still used as an insecticide in several less developed or underdeveloped countries. Not surprisingly, levels of these environmental contaminants have decreased in many locations [5, 17–20] and, in the Great Lakes which was heavily contaminated in the 1960s/1970s, decreased levels of POPs (particularly TCDD-like compounds) is paralleled by increased wildlife reproduction in many areas [21].A recent paper by Norén and Meironyté [5] summarized levels of different structural classes of POPs in Swedish human milk samples. Overall levels of POPs have decreased by approximately 90% from 4410 ng/g lipid in 1972 to 483 ng/g lipid in 1997, and this is paralleled by decreases

Fig. 1. Structures of important classes of POPs

226

S. Safe

1972 1997

Fig. 2. Time-dependent changes in levels of different classes of POPs in Swedish breast milk samples [5]

in most classes of POPs (Fig. 2). The only exception are the PBDEs which constituted about 1% of the total POPs in 1997, but levels of these compounds in Swedish milk have been increasing over the 1972–1997 time period.

2 Toxicology of Different Classes of POPs 2.1 Aryl Hydrocarbon Receptor (AhR)-Active Compounds 2.1.1 PCDDs and PCDFs

2,3,7,8-TCDD was identified as a by-product in the preparation of trichlorophenol and derived products such as the herbicide 2,4,5-T and shown to be one of the most acutely lethal synthetic compounds with an LD50 in the range of 1 µg/kg for the male/female guinea pig (reviewed in [22–28]). The LD50 values for TCDD are highly variable between animal species/strains since LD50 values in the least resistant Syrian hamster and Hann Wistar rat are greater than 1000 times higher than observed for the guinea pig. In most species acutely or subchronically exposed to TCDD, several common responses are generally observed and these include a wasting syndrome, thymic atrophy, skin toxicity (including chloracne), tissue-specific hypo- and hyperplastic effects, and induction of CYP1A1, other CYP1 gene and genes encoding several phase II drug-metabolizing enzymes. Other well-characterized responses associated with exposure to TCDD include hepatic porphyria, developmental and reproductive toxicity, neurotoxicity, tumor promoter activity (particularly in skin and liver), immunotoxicity, and disruption of multiple endocrine signaling pathways that can lead to reproductive tract dys-

Toxicology and Risk Assessment of POPs

227

function in both male and female rodents. One of the hallmarks of TCDD-induced responses are the remarkable differences that are dependent on the species, sex and strain of animal, age of the animal, and timing of exposure. For example, humans, rabbits, and certain strains of mice develop chloracne-like responses after exposure to TCDD; however, most other animals including the guinea pig are resistant to this response. TCDD induces hepatocellular carcinomas in female Sprague-Dawley rats, whereas male rats are resistant to this response [29]. It was also observed that although TCDD was the most toxic individual compound among the 75 PCDD and 135 PCDF congeners, several other 2,3,7,8-substituted compounds induced similar toxic responses in laboratory animal studies but were less potent than TCDD [15, 16, 22]. Poland and coworkers [30] subsequently identified a protein in mouse hepatic cytosol that bound [3H]TCDD with high affinity but low capacity, and competitive binding studies among PCDD and PCDF congeners showed that there was a rank order correlation between their protein binding affinities and their toxic/biochemical potencies. This protein also bound other aromatic hydrocarbons such as benzo[a]pyrene and 3methylcholanthrene and was named the aryl hydrocarbon receptor (AhR). Subsequent studies in several laboratories have identified the AhR protein in many mammalian species including humans and the AhR gene has also been cloned (reviewed in [25–28, 31]). The mechanism of AhR action by TCDD and related compounds has been most extensively studied using the CYP1A1 gene as a model and the following pathway is consistent with results of most studies [23, 32, 33]. TCDD or related compounds initially bind the cytosolic AhR complex in the target cell/tissue and rapidly forms a nuclear heterodimer with the AhR nuclear translocator (Arnt) protein. The resulting nuclear AhR complex binds 5′-promoter dioxin responsive elements (DREs) in target genes and interactions of this complex with the basal transcriptional machinery results in increased gene expression. Based on the common AhR-mediated mechanism of action for TCDD and related compounds, hazard/risk assessment of complex mixtures of the POPs can be carried out by using a toxic equivalency factor (TEF) approach [34–41]: TCDD or toxic equivalents (TEQs)=S[PCDD]i ¥TEFi +S[PCDF]i ¥TEFi where [PCDD]i and [PCDF]i are the concentrations of individual 2,3,7,8-substituted congeners in any mixture and TEFi represents their individual fractional potencies relative to TCDD (TEF=1.0). Table 1 summarizes TEFs for the most environmentally relevant 2,3,7,8-substituted PCDDs/PCDFs, and these individual values have been derived from a range of response/species-specific TEFs reported from multiple studies [34–39]. For some compounds, the ranges of experimentally-derived TEFs vary by over 100-fold; however, several studies have demonstrated that calculated TEQs for mixtures of PCDDs and PCDFs are similar to experimentally-derived TEQs. These results suggest that TEQs can be used to estimate the potential TCDD-like toxicity of PCDD/PCDF mixtures and this approach is now extensively used by scientists and regulatory agencies.

228

S. Safe

Table 1. TEFs for PCB, PCDD and PCDF congeners [39]

Congener

TEF value

Dibenzo-p-dioxins 2,3,7,8-TCDD 1,2,3,7,8-PnCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD OCDD

TEF value

Non-ortho PCBs 1 1 0.1 0.1 0.1 0.01 0.0001

Dibenzofurans 2,3,7,8-TCDF 1,2,3,7,8-PnCDF 2,3,4,7,8-PnCDF 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDF

Congener

PCB 77 PCB 81 PCB 126 PCB 169

0.0001 0.0001 0.1 0.01

Mono-ortho PCBs 0.1 0.05 0.5 0.1 0.1 0.1 0.1 0.01 0.01 0.0001

PCB 105 PCB 114 PCB 118 PCB 123 PCB 156 PCB 157 PCB 167 PCB 189

0.0001 0.0005 0.0001 0.0001 0.0005 0.0005 0.00001 0.0001

2.1.2 PCBs and Related Compounds

Initial research on PCBs focused on the biochemical and toxic effects of commercial mixtures such as Aroclors 1242, 1248, and 1254, and results showed some of their effects resembled those observed for TCDD and related 2,3,7,8-substituted PCDDs and PCDFs [42–46]. For example,Aroclor 1254 induced rat hepatic cytochrome P450-dependent N-dealkylase activity commonly observed after treatment with phenobarbital (PB) as well as benzo[a]pyrene hydroxylase activity. This latter response is observed in rodents after treatment with TCDD and carcinogenic polynuclear aromatic hydrocarbons (PAHs) such as 3-methylcholanthrene (MC) [42–46]. It was also shown that at least four PCB congeners, namely 3,4,4′,5-tetrachlorobiphenyl (tetraCB), 3,3′,4,4′-tetraCB, 3,3′,4,4′,5-pentaCB, and 3,3′,4,4′,5,5′-hexaCB, most closely resembled TCDD in their biochemical and toxic responses [47–49]. These compounds were substituted in both para (4,4′) and two or more meta (3,3′/5,5′) positions, contained no ortho substituents, and were isosteric with TCDD in their planar conformation. These compounds also competitively bound the AhR. However, levels of coplanar PCBs in environmental extracts and commercial mixtures are low to nondetectable, suggesting that coplanar PCBs may not account for the AhR-mediated activities of these mixtures. Moreover, the CYP1A1 induction activity of a reconstituted mixture of 14 PCB congeners identified in breast milk was higher than observed for Aroclor 1254, and the reconstituted mixture did not contain any coplanar PCB

Toxicology and Risk Assessment of POPs

229

congeners [50]. Subsequent studies showed that all possible eight mono-ortho coplanar PCB derivatives of the four coplanar PCBs induced CYP1A1-dependent activity in vivo and in vitro and also competitively bound the rat cytosolic AhR [51–57]. These mono-ortho-substituted PCBs were much less active than coplanar PCBs in bioassays for AhR agonists; however, concentrations of some mono-ortho substituted PCBs in commercial PCBs and environmental samples are relatively high, and results of enzyme induction assays in rodents suggest that they contribute a major component of the activity of commercial PCBs [58]. For example, the following congeners: 2,3,3′,4,4′-pentaCB (#105) 1.51%; 2,3,4,4′,5-pentaCB (#114) 0.3%; 2,3′,4,4′,5-pentaCB (#118) 7.12%; 2′,3,4,4′,5pentaCB (#123) nondetectable; 2,3,3′,4,4′,5-hexaCB (#156) 0.89%; 2,3,3′,4,4′,5′hexaCB (#157) trace; 2,3′,4,4′,5,5′-hexaCB (#167) 0.65%; 2,3,3′,4,4′,5,5′-hepta CB (#189) trace, constitute 10.47 wt% of Aroclor 1254 [59]. In contrast, concentrations of 3,3′,4,4′-tetraCB (PCB #77), 3,3′,4,4′,5-pentaCB (PCB #126), and 3,3′,4,4′,5,5′-hexaCB (PCB #169) (coplanar PCBs) in Aroclor 1254 were <0.07 wt% [60, 61]. Di-ortho-substituted analogs of the coplanar PCBs have also been investigated [53, 62] and, with the exception of 2,2′,4,4′,5,5′-hexaCB (#153), most of the di-ortho-substituted analogs induced CYP1A1-dependent activity or other AhR-mediated responses. Two of the more AhR-active di-ortho-substituted analogs, namely 2,2′,3,3′,4,4′5-heptaCB (#170) and 2,2′,3,4,4′,5,5′-heptaCB (#180), are important environmental contaminants and represent 1.37 and 14.86 wt% of Aroclor 1254 and Aroclor 1260, respectively [59]. Tanabe and coworkers first recognized that TEQs could also be used for PCB mixtures [60, 63–65] by using TEFs derived from induction of CYP1A1dependent responses in rat hepatoma H4IIE cells [56]. It was concluded that PCB-TEQs in some environmental samples exceeded TEQs calculated for PCDDs/PCDFs. TEF values for PCBs were initially proposed by Safe [36], and these values have since been updated (Table 1) as new data has become available [39–41]. Relative contributions of PCB-TEQs compared to PCDD/PCDF-TEQs is dependent on the specific environmental or food extract. A recent report on PCDDs/PCDFs/PCBs in Swedish breast milk [5] showed that PCB-TEQs were >50% of the total TEQs and the major contributors were PCBs #126, #118, and #156. 2.1.3 Problems Associated with TEF/TEQ Approach for Risk Assessment

TEFs for PCDDs/PCDFs and PCBs are routinely used by scientists for calculating TEQs; however, these values have not been officially adopted by regulatory agencies. The TEF approach for PCB, PCDD, and PCDF mixtures has been critically examined in laboratory animal studies where calculated TEQs can be compared to experimentally-derived TEQs for various AhR-mediated endpoints. As noted above, PCDD and PCDF mixtures essentially give additive responses [66–68]; however, PCB mixtures alone or in combination with one or more PCDD/PCDF (usually TCDD) give both additive and nonadditive responses that are species- and response-dependent (reviewed in [69, 70]). Induction of

230

S. Safe

CYP1A1-dependent hepatic microsomal enzyme activity by Aroclors 1242, 1254, and 1260 in immature male Wistar rats gave responses that were consistent with additive contributions by the major AhR-active mono- and di-ortho substituted congeners, and interactions with high levels of AhR-inactive congeners were minimal [58]. In contrast, the calculated immunotoxic potencies for commercial Aroclors in mice (e.g., S[PCB]i ¥TEFi) overestimated this response compared to the observed immunotoxic effects, indicating that interactions with AhR-inactive compounds in the PCB mixtures were antagonistic [71]. Research in several laboratories have investigated interactions between weak AhR agonists, such as PCB #153 (2,2′,4,4′,5,5′-hexaCB) and Aroclor mixtures, and potent AhR-active compounds including PCB #126 and TCDD. Inhibition of TCDD- or PCB #126-induced responses by these compounds/mixtures have been observed for cleft palate formation, hydronephrosis, porphyria, and immunotoxicity in mice [71–77]; embryotoxicity in Japanese medaka [78]; chick embryo malformations, edema and liver lesions [79, 80], induction of CYP1A1-dependent activity (or reporter gene activity) in rat, mouse, and chick embryo-derived cell lines [81–84].Antagonistic interactions between AhR agonists and PCBs are generally observed at PCB/AhR agonist ratios of >1000 which are commonly observed in many environmental samples. The ratio of the total PCBs (324 ng/g lipid)/PCB #126 (0.076 ng/g lipid) or SPCDDs/PCDFs in Swedish breast milk samples taken in 1997 was >1000 [5]. To complicate further application of TEQs to dietary intakes of PCDDs, PCDFs, and PCBs, many phytochemicals, such as flavonoids, indole-3-carbinol and related compounds, and carotenoids, also exhibit weak AhR agonist activity, and some are also partial AhR antagonists [85–88]. Thus, a comprehensive risk assessment of POP-TEQs associated with contaminants in food should take into account the high dietary intakes of phytochemical-derived AhR agonists/antagonists, coupled with decreasing environmental levels of PCDDs, PCDFs, and PCBs. 2.2 AhR-Inactive Compounds 2.2.1 Introduction

Although the AhR-active POPs have been extensively investigated and have been linked to wildlife problems and toxic effects in humans, the toxic responses associated with POPs that act through AhR-independent pathways are unclear. For example, there has been a correlation between in utero exposure to total PCBs and neurodevelopmental deficits in infants and young children, and there is evidence from laboratory animal studies that both AhR-active and inactive POPs induce neurotoxic responses (reviewed in [89–92]). Thus, although the correlation (or causation) of adverse responses in wildlife and humans have not been directly linked with AhR-inactive POPs (with the exception of DDE), it is likely that some of these contaminants may significantly contribute to toxicity of POP mixtures in the environment. Some of the AhR-independent activities associated with different classes of POPs are indicated below.

Toxicology and Risk Assessment of POPs

231

2.2.2 PCBs

Structurally-diverse PCB congeners that contain two or more ortho-chloro substituents induce CYP2B and other CYP isozymes, and some of these compounds are also tumor promoters in rodent bioassays [53, 93–95]. 2,2′,4,4′,5,5′-HexaCB (PCB #153) is the major persistent PCB congener in most environmental samples and typifies the di-tri-ortho substituted PCB congeners that induce CYP2B activity and also resembles phenobarbital as a CYP2 inducer and tumor promoter. Seegal and coworkers [96, 97] have extensively investigated the neurotoxicity of PCB mixtures and congeners and showed that several ortho-substituted AhRinactive congeners affect brain neurochemistry and decreased dopamine levels in PC-12 cells. Many of these same compounds induce other potentially neurotoxic responses including disruption of calcium homeostasis in preparations from brain and other tissues/subcellular fractions [98–103]. The potential roles of these structurally-diverse AhR-inactive compounds on the hypothesized link between in utero exposure to PCBs and neurodevelopmental deficits in offspring is unproven; however, the laboratory animal/cell data provides some support for this hypothesis [104]. It could also be argued that the relative neurotoxic potencies of the environmentally persistent ortho-substituted PCB congeners is low as is their tissue levels, and more research is required to resolve this issue. 2.2.3 Hydroxy-PCBs

The oxidative metabolism of PCBs and the subsequent formation and excretion of hydroxy-PCB metabolites has been extensively investigated and represents a major pathway for removal of these compounds from contaminated biota [105–107]. Surprisingly, Bergmann and coworkers [108] first showed that serum from wildlife, laboratory animals administered PCBs, and humans contained substantial levels of higher chlorinated hydroxy-PCBs that were metabolites of some of the more environmentally persistent PCB congeners [108–111]. Early studies demonstrated that hydroxy-PCBs can act like other phenolics and uncouple mitochondrial oxidative phosphorylation [112, 113] as well as induce estrogenic activities in cell culture and in vivo models [114]. Brouwer and coworkers [115–117] also demonstrated that hydroxy-PCBs including some of those identified in human serum bind high affinity to thyroid hormone transport proteins (e.g., transthyretin in rodents), and this interaction can effectively lower circulating thyroid hormone levels. This interaction may also be important for neurodevelopmental processes in the fetus where thyroid hormone levels are important [118, 119]. The endocrine disruptor activities of hydroxy-PCBs have also been investigated with a major focus on their estrogenic/antiestrogenic responses [120–124]. Serum hydroxy-PCBs weakly interact with estrogen receptor α (ERα) and ERβ, and the most active compounds were 2′,4′,6-trichloro- and 2′,3′,4′,5-tetrachloro4-biphenylol. These congeners were approximately 103 to 104 times less potent than 17b-estradiol but exhibited full ER agonist activities in some bioassays

232

S. Safe

[124]. In contrast, the hydroxy-PCBs identified in human serum were partial ER agonist/antagonists (or inactive) [122, 123] suggesting that their potential as estrogenic endocrine disruptors were minimal. A recent paper reported that hydroxy-PCBs inhibited estrogen sulfotransferase activity; the IC50 value for the most active compound, 2,3,3′,4′,5-pentachloro-4-biphenylol, was as low as 0.15 to 0.25 nmol/l and this compound has been identified in human serum [125]. Other hydroxy-PCBs identified in wildlife and human serum also exhibited IC50 values in the low nmol/l range. It has been suggested that inhibition of estrogen sulfotransferase activity by hydroxy-PCBs in some cell types will increase free estrogen levels and indirectly induce estrogenic responses. This is an area of research that should be further investigated.

3 Summary POPs in the environment are a continuing concern for regulatory agencies, scientists, and citizens, and there is evidence that environmental levels of many of these contaminants are decreasing although some regions may still be “hotspots”. The TEF/TEQ approach for risk assessment has been used to regulate AhR-active compounds. AhR-inactive POPs such as ortho-substituted PCBs and hydroxy-PCBs exhibit an increasing number of toxic, genotoxic, and biochemical responses; however, their role in POP-induced responses in wildlife and humans is unclear. Moreover, methods for hazard and risk assessment of AhR-inactive compounds have not been developed, and this is an area that requires further study. Acknowledgements. The financial assistance of the National Institutes of Health (ES04917 and

ES09106), the Electric Power Research Institute, and the Texas Agricultural Experiment Station is gratefully acknowledged.

4 References 1. Vallack HW, Bakker DJ, Brandt I, Brostrom-Lunden E, Brouwer A, Bull KR, Gough C, Guardans R, Holoubek I, Jansson B, Koch R, Kuylenstierna J, Lecloux A, Mackay D, McCutcheon P, Mocarelli P, Taalman RDF (1998) Environ Toxicol Pharmacol 6:143 2. Anonymous (1966) New Sci 32:621 3. Kuehl DW, Butterworth B (1994) Chemosphere 29:523 4. Tanabe S, Kannan N, Subramanian A,Watanabe S, Tatsukawa R (1987) Environ Pollut 47:147 5. Norén K, Meironyté D (2000) Chemosphere 40:1111 6. Marklund S,Wikström E, Lofvenius G, Fangmark I, Rappe C (1994) Chemosphere 28:1895 7. Fiedler H (1996) Chemosphere 32:55 8. Olie K, Vermeulen PL, Hutzinger O (1977) Chemosphere 6:455 9. Hutzinger O, Fiedler H (1993) Chemosphere 27:121 10. Hagenmaier H, Brunner H (1990) Chemosphere 20:2425 11. Christmann W, Kloppel KD, Partscht H, Rotard W (1989) Chemosphere 18:861 12. Norstrom A, Rappe C, Lindahl R, Buser HR (1979) Scand J Work Environ Health 5:375 13. Safe S (1991) Environ Carcin Ecotox Rev C9:261 14. Buckley EH (1982) Science 216:520 15. Safe S (1986) Annu Rev Pharmacol Toxicol 26:371

Toxicology and Risk Assessment of POPs

233

16. Goldstein JA, Safe S (1989) Mechanism of action and structure-activity relationships for the chlorinated dibenzo-p-dioxins and related compounds. In: Kimbrough RD, Jensen AA (eds) Halogenated biphenyls, naphthalenes, dibenzodioxins and related compounds, 2nd edn. Elsevier-North Holland, Amsterdam, p 239 17. Turle R, Norstrom RJ, Collins B (1991) Chemosphere 22:201 18. Schmitt CJ, Zajicek JL, Peterman PH (1990) Arch Environ Contam Toxicol 19:748 19. Lake CA, Lake JL, Haebler R, McKinney R, Boothman WS, Sadove SS (1995) Arch Environ Contam Toxicol 29:128 20. Williams LL, Giesy JP,Verbrugge DA, Jurzysta S, Heinz G, Stromborg KL (1995) Arch Environ Contam Toxicol 29:52 21. Tremblay NW, Gilman AP (1995) Environ Health Perspect 103:3 22. Poland A, Knutson JC (1982) Annu Rev Pharmacol Toxicol 22:517 23. Whitlock JP Jr (1993) Chem Res Toxicol 6:754 24. Lucier GW, Portier CJ, Gallo MA (1993) Environ Health Perspect 101:36 25. Gilman A, Newhook R (1991) Chemosphere 23:1661 26. Swanson HI, Bradfield CA (1993) Pharmacogenetics 3:213 27. Landers JP, Bunce NJ (1991) Biochem J 276:273 28. Okey AB, Riddick DS, Harper PA (1994) Toxicol Lett 70:1 29. Kociba RJ, Keyes DG, Beger JE, Carreon RM,Wade CE, Dittenber DA, Kalnins RP, Frauson LE, Park CL, Barnard SD, Hummel RA, Humiston CG (1978) Toxicol Appl Pharmacol 46:279 30. Poland A, Glover E, Kende AS (1976) J Biol Chem 251:4936 31. Wilson CL, Safe S (1998) Toxicologic Pathol 26:657 32. Whitlock JP (1996) FASEB J 10:809 33. Whitlock JP (1999) Annu Rev Pharmacol Toxicol 39:103 34. Ahlborg UG, Brouwer A, Fingerhut MA, Jacobson JL, Jacobson SW, Kennedy SW, Kettrup AAF, Koeman JH, Poiger H, Rappe C, Safe SH, Seegal RF, Tuomisto J, Van den Berg M (1992) Eur J Pharmacol 228:179 35. North Atlantic Treaty Organization (NATO) (1988) Scientific basis for the development of international toxicity equivalency factor (I-TEF), method of risk assessment for risk assessment for complex mixtures of dioxins and related compounds. Committee on the Challenges of Modern Society, Rep 178 36. Safe S (1990) CRC Crit Rev Toxicol 21:51 37. Safe S (1994) CRC Crit Rev Toxicol 24:87 38. Birnbaum LS, DeVito MJ (1995) Toxicology 105:391 39. van Leeuwen FXR, Feeley M, Schrenk D, Larsen JC, Farland WH, Younes M (2000) Chemosphere 40:1095 40. Van den Berg M, Birnbaum L, Bosveld ATC, Brunstrom B, Cook P, Feeley M, Giesy JP, Hanberg A, Hasegawa R, Kennedy SW, Kubiak T, Larsen JC, van Leeuwen FX, Liem AK, Nolt C, Peterson RE, Poellinger L, Safe S, Schrenk D, Tillitt D, Tysklind M, Younes M, Waern F, Zacharewski T (1998) Environ Health Perspect 106:775 41. Ahlborg UG, Becking GC, Birnbaum LS, Brouwer A, Derks HJGM, Feeley M, Golor G, Hanberg A, Larsen JC, Liem AKD, Safe S, Schlatter C,Wærn F,Younes M,Yrjänheikki E (1994) Chemosphere 28:1049 42. Alvares AP (1977) Stimulatory effects of polychlorinated biphenyls (PCB) on cytochromes P-450 and P-448 mediated microsomal oxidations. In: Ullrich V, Hildebrandt A, Roots I, Estabrook RW, Conney AH (eds) Microsomes and drug oxidations. Pergamon Press, Oxford, p 476 43. Alvares AP, Bickers DR, Kappas A (1973) Proc Natl Acad Sci USA 70:1321 44. Alvares AP, Kappas A (1977) J Biol Chem 252:6373 45. Yoshimura H, Yoshihara S, Ozawa N, Miki M (1979) Annu NY Acad Sci 320:179 46. Ryan DE, Thomas PE, Korzeniowski D, Levin W (1979) J Biol Chem 254:1365 47. Stonard MD, Grieg JB (1976) Chem Biol Interact 15:365 48. Goldstein JA, Hickman P, Bergman H, McKinney JD, Walker MP (1977) Chem Biol Interact 17:69

234

S. Safe

49. 50. 51. 52. 53.

Poland A, Glover E (1977) Mol Pharmacol 13:924 Parkinson A, Robertson L, Safe S (1980) Biochem Biophys Res Commun 96:882 Parkinson A, Robertson L, Safe L, Safe S (1981) Chem Biol Interact 31:1 Parkinson A, Robertson L, Uhlig L, Campbell MA, Safe S (1982) Biochem Pharmacol 31:2830 Parkinson A, Safe S, Robertson L, Thomas PE, Ryan DE, Levin W (1983) J Biol Chem 258:5967 Parkinson A, Cockerline R, Safe S (1980) Biochem Pharmacol 29:259 Parkinson A, Cockerline R, Safe S (1980) Chem Biol Interact 29:277 Sawyer T, Safe S (1982) Toxicol Lett 13:87 Bandiera S, Safe S, Okey AB (1982) Chem Biol Interact 39:259 Harris M, Zacharewski T, Safe S (1993) Fund Appl Toxicol 20:456 Frame GM,Wagner RE, Carnahan JC, Brown JF Jr, May RJ, Smullen LA, Bedard DL (1996) Chemosphere 33:603 Kannan N, Tanabe S, Tatsukawa R (1988) Bull Environ Contam Toxicol 41:267 Schulz DE, Petrick G, Duinker JC (1989) Environ Sci Technol 23:852 Parkinson A, Robertson L, Safe S (1980) Life Sci 27:2333 Tanabe S, Kannan N, Ono M, Tatsukawa R (1989) Chemosphere 18:485 Tanabe S, Kannan N,Wakimoto T, Tatsukawa R, Okamoto T, Masuda Y (1989) Toxicol Environ Chem 24:215 Kannan N, Tanabe S, Tatsukawa R (1988) Arch Environ Health 43:11 Eadon G, Kaminsky L, Silkworth J,Aldous K, Hilker D, O-Keefe P, Smith R, Gierthy JF, Hawley J, Kim N, DeCaprio A (1986) Environ Health Perspect 70:221 Schrenk D, Lipp HP, Wiesmuller T, Hagenmaier H, Bock KW (1991) Arch Toxicol 65:114 Stahl BU, Kettrup A, Rozman K (1992) Arch Toxicol 66:471 Safe S (1998) J Animal Sci 76:134 Safe S (1998) Teratogen Carcinogen Mutagen 17:285 Harper N, Connor K, Steinberg M, Safe S (1995) Fund Appl Toxicol 27:131 Bannister R, Davis D, Zacharewski T, Tizard I, Safe S (1987) Toxicology 46:29 Haake JM, Safe S, Mayura K, Phillips TD (1987) Toxicol Lett 38:299 Biegel L, Harris M, Davis D, Rosengren R, Safe L, Safe S (1989) Toxicol Appl Pharmacol 97:561 Davis D, Safe S (1989) Toxicol Lett 48:35 Davis D, Safe S (1990) Toxicology 63:97 Morrissey RE, Harris MW, Diliberto JJ, Birnbaum LS (1992) Toxicol Lett 60:19 Harris GE, Metcalfe TL, Metcalfe CD, Huestis SY (1995) Environ Toxicol Chem 13:1393 Zhao F, Mayura K, Harper N, Safe SH, Phillips TD (1997) Chemosphere 34:1605 Zhao F, Mayura K, Kocurek N, Edwards JF, Kubena LF, Safe SH, Phillips TD (1997) Fundam Appl Toxicol 35:1 Keys B, Piskorska-Pliszczynska J, Safe S (1986) Toxicology Letters 31:151 Aarts JMMJG, Denison MS, Cox MA, Schalk MAC, Garrison PM, Tullis K, De Haan LHJ, Brouwer A (1995) Eur J Pharmacol 293:463 Bosveld ATC, Verhallen E, Seinen W, Van den Berg M (1995) Organohal Compd 25:309 Tysklind M, Bosveld ATC,Andersson P,Verhallen E, Sinnige T, Seinen W, Rappe C,Van den Berg M (1995) Environ Sci Pollut Res 4:211 Chen I, Safe S, Bjeldanes L (1996) Biochem Pharmacol 51:1069 Ciolino HP, Wang TT, Yeh GC (1998) Cancer Res 58:2754 Bjeldanes LF, Kim JY, Grose KR, Bartholomew JC, Bradfield CA (1991) Proc Natl Acad Sci USA 88:9543 Gradelet S, Astorg P, Pineau T, Canivenc MC, Siess MH, Leclerc J, Lesca P (1997) Biochem Pharmacol 54:307 Brouwer A, Ahlborg UG,Van den Berg M, Birnbaum LS, Boersma ER, Bosveld B, Denison MS, Gray LE, Hagmar L, Holene E, Huisman M, Jacobson SW, Jacobson JL, Koopman-Esseboom C, Koppe JG, Kulig BM, Morse DC, Muckle G, Peterson RE, Sauer PJJ, Seegal RF, Smits-van Prooije AE, Touwen BCL, Weisglas-Kuperus N, Winneke G (1995) Eur J Pharmacol 293:1

54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89.

Toxicology and Risk Assessment of POPs

90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125.

235

Tilson HA, Kodavanti PRS (1998) Neurotoxicology 19:517 Jacobson JL, Jacobson SW (1996) N Engl J Med 335:783 Jacobson JL, Jacobson SW (1997) Teratology 55:338 Connor K, Safe S, Jefcoate CR, Larsen M (1995) Biochem Pharmacol 50:1913 Buchmann A, Kunz W, Wolf CR, Oesch F, Robertson LW (1986) Cancer Lett 32:243 Laib RM, Rose N, Bunn H (1991) Toxicol Environ Chem 34:19 Seegal RF, Bush B, Shain W (1990) Toxicol Appl Pharmacol 106:136 Shain W, Bush B, Seegal R (1991) Toxicol Appl Pharmacol 111:33 Kodavanti PRS, Shafer TJ, Ward TR, Mundy WR, Freudenrich T, Harry GJ, Tilson HA (1994) Brain Res 662:75 Kodavanti PRS, Shin D, Tilson HA, Harry GJ (1993) Toxicol Appl Pharmacol 123:97 Kodavanti PR, Ward TR, McKinney JD, Waller CL, Tilson HA (1996) Toxicol Appl Pharmacol 138:251 Shafer TJ, Mundy WR, Tilson HA, Kodavanti PRS (1996) Toxicol Appl Pharmacol 141:448 Wong PW, Pessah IN (1996) Mol Pharmacol 49:740 Wong PW, Pessah IN (1997) Mol Pharmacol 51:693 Schantz SL, Moshtaghian J, Ness DK (1995) Fundam Appl Toxicol 26:117 Safe S (1989) Polyhalogenated aromatics: uptake, disposition and metabolism. In: Kimbrough RD, Jensen AA (eds) Halogenated biphenyls, naphthalene, dibenzodioxins and related compounds, 2nd edn. Elsevier-North Holland, Amsterdam, p 51 Hutzinger O, Nash DM, Safe S, Norstrom RJ, Wildesh DJ, Zitko V (1972) Science 178:312 Sipes IG, Schnellmann RG (1987) Biotransformation of PCBs: metabolic pathways and mechanisms. In: Safe S, Hutzinger O (eds) Polychlorinated biphenyls (PCBs): mammalian and environmental toxicology. Springer, Berlin Heidelberg New York, p 97 Bergman Å, Klasson-Wehler E, Kuroki H (1994) Environ Health Perspect 102:464 Sjodin A, Tullsten AK, Klasson-Wehler E (1998) Organochlorine Compd 37:365 Newsome WH, Davies D (1996) Chemosphere 33:559 Jansson B, Jensen S, Olsson M, Renberg L, Sundström G, Vaz R (1975) Ambio 4:93 Nishihara Y, Utsumi K (1987) Biochem Pharmacol 36:3453 Narasimhan TR, Kim HL, Safe S (1991) J Biochem Toxicol 6:229 Korach KS, Sarver P, Chae K, McLachlan JA, McKinney JD (1988) Mol Pharmacol 33:120 Brouwer A, Van den Berg KJ (1986) Toxicol Appl Pharmacol 85:301 Lans MC, Klasson-Wehler E,Willemsen M, Meussen E, Safe S, Brouwer A (1993) Chem Biol Interact 88:7 Brouwer A, Klasson-Wehler E, Bokdam M, Morse DC, Traag WA (1990) Chemosphere 20:1257 Morse DC, Klasson-Wehler E, van de Pas M, de Bie AT,Van Bladeren PJ, Brouwer A (1995) Chem Biol Interact 95:41 Morse DC, Klasson-Wehler E, Wesseling W, Koeman JH, Brouwer A (1996) Toxicol Appl Pharmacol 136:269 Connor K, Ramamoorthy K, Moore M, Mustain M, Chen I, Safe S, Zacharewski T, Gillesby B, Joyeux A, Belaguer P (1997) Toxicol Appl Pharmacol 145:111 Kuiper GG, Lemmen JG, Carlsson B, Corton JC, Safe SH,Van der Saag PT,Van der Burg B, Gustafsson J-Å (1998) Endocrinology 139:4252 Kramer VJ, Helferich WG, Bergman Å, Klasson-Wehler E, Giesy JP (1997) Toxicol Appl Pharmacol 144:363 Moore M, Mustain M, Daniel K, Chen I, Safe S, Zacharewski T, Gillesby B, Joyeux A, Balaguer P (1997) Toxicol Appl Pharmacol 142:160 Ramamoorthy K,Vyhlidal C, Wang F, Chen I-C, Safe S, McDonnell DP, Leonard LS, Gaido KW (1997) Toxicol Appl Pharmacol 147:93 Kester MHA, Bulduk S, Tibboel D, Meinl W, Glatt H, Falany CN, Coughtrie MWH, Bergman A, Safe S, Kuiper GG, Schuur AG, Brouwer A, Visser TJ (2000) Endocrinology 141:1897

CHAPTER 9

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants Martin Scheringer 1 · Frank Wania 2 1

2

Laboratory of Technical Chemistry, Swiss Federal Institute of Technology, ETH Hönggerberg, 8093 Zürich, Switzerland E-mail: [email protected] Division of Physical Sciences, University of Toronto at Scarborough, Toronto, Ontario, Canada M1C 1A4

Persistent organic pollutants (POPs) are long-lived multimedia chemicals that cycle between the different environmental compartments and are capable of traveling long distances, thus becoming ubiquitous global pollutants. With the help of multimedia fate and transport models, the interplay of release, phase partitioning, degradation and transport of these pollutants on the global scale can be systematically described and analyzed. Currently, multimedia box models and models based on atmospheric dispersion models are used to investigate the global behavior of POPs. Box models are simpler to construct and use, yet have only low spatial and temporal resolution. Dispersion models have high resolution but also require high computational effort. In this chapter, both types of models are described and results from their application to selected POPs (a-HCH, polychlorinated biphenyls, DDT) are summarized. The model results show, e.g., that only small fractions of the total global inventory of a POP are sufficient to cause contamination of polar regions and that long-range transport may occur via both air and ocean water. Due to incomplete or uncertain data on POP emissions, physical-chemical properties and degradation rate constants, as well as a lack of process understanding in regions other than the temperate North, there is significant uncertainty associated with the model results. Since also data from field studies are scarce, quantitative comparison of model results with measured data is difficult. Sensitivity and uncertainty analyses as well as model comparison studies are important means for the evaluation of global POPs models. Keywords: Persistent organic pollutants, Multimedia models, Persistence, Long-range transport, Cold condensation, Exposure assessment, a-HCH, PCBs, DDT

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238

1

Introduction

2

Modeling Approaches

3

Data Requirements and Modeling Results

3.1 3.1.1 3.1.2 3.1.3 3.2 3.2.1 3.2.2

Model Input Data . . . . Chemical Properties . . . Environmental Parameters Emission Data . . . . . . Model Results . . . . . . Basic Model Output . . . Aggregated Indicators . .

. . . . . . . . . . . . . . . . . . . . . . . . 240

. . . . . . . . . . . . . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . . . . . . . . 242 . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

242 242 245 246 247 247 247

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

238

M. Scheringer · F. Wania

. . . . . . . . . . . . . . . . . . . . . . . . . 249

4

Model Calculations

4.1 4.2 4.2.1 4.2.2 4.3 4.3.1 4.3.2

Description of Selected Models . . . . . . . . . . . Model Calculations with Idealized Release Scenarios Spatially Uniform Multimedia Box Models . . . . . Multimedia Box Models with Climatic Zones . . . . Calculations with Realistic Release Scenarios . . . Multimedia Box Models . . . . . . . . . . . . . . . Atmospheric Dispersion Models . . . . . . . . . .

5

Model Reliability and Validation

6

Open Questions and Further Research

6.1 6.2 6.3

Input Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . 265 Modeling Approaches . . . . . . . . . . . . . . . . . . . . . . . . 265 Assessment Criteria . . . . . . . . . . . . . . . . . . . . . . . . . 266

7

References

. . . . . . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

249 255 255 256 259 259 261

. . . . . . . . . . . . . . . . . . 262 . . . . . . . . . . . . . . . 264

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

1 Introduction The term persistent organic pollutants (POPs) has been introduced to identify chemicals that are persistent in the environment and cause toxic effects to humans and other organisms. Typical chemicals of this kind are the organochlorine pesticides such as p,p¢-DDT while volatile and non-toxic chemicals such as the chlorofluorocarbons (CFCs) are normally excluded from the POPs category.1 Nine organochlorine pesticides – aldrin, chlordane, DDT, dieldrin, endrin, heptachlor, hexachlorobenzene, mirex, toxaphene – together with the polychlorinated biphenyls (PCBs) and dibenzodioxins (PCDDs) and -furans (PCDFs) form the list of the first 12 POPs designated in the course of the international negotiations leading to the Stockholm Convention on Persistent Organic Pollutants [1]. The POPs Protocol of the UN-ECE Convention on Long-Range Transboundary Air Pollution lists, in addition to the 12 compounds or groups of compounds of the Stockholm Convention, chlordecone, the hexachlorocyclohexane (HCH) isomers, hexabromobiphenyl, and the polycyclic aromatic hydrocarbons (PAHs). (Note that HCH is not a POP according to the UNEP list.) Besides their persistence and mobility, the (mainly chronic) toxicity of many POPs is a source of concern. The different POPs cause a variety of toxic effects such as reproductive and developmental impairments in numerous species, including mammals and humans (e.g., reduced fertility, abortion, stillbirths, eggshell thinning, low birth weight, impaired intellectual development of chil1

In principle, CFCs are persistent organic pollutants, too. However, the term POPs is commonly used for chemicals that cause toxic effects within the biosphere. This specific definition excludes CFCs, which lead to adverse chemical changes in the stratosphere.

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

239

dren); estrogenic activity; immuno-suppression; induction of liver enzymes; contribution to cancer development; reduced feed intake and weight loss. The complex toxicity pattern of POPs will not be discussed here. For more detail, we refer to [2–5]. Important problems and open questions in the context of the environmental fate and distribution of POPs are: – to understand the mechanisms of multimedia partitioning, long-range transport and contamination of remote areas; – to identify key properties of POPs and to characterize POPs by suitable criteria for persistence, mobility, bioaccumulation, and toxicity, in particular with respect to the requirements of the Stockholm Convention; – to investigate typical time scales and recovery rates of POPs contamination as well as scenarios of future developments. Aims of research into these areas are, among others, the support of decisions concerning the designation of additional chemicals under the Stockholm Convention and the improvement of the risk management for present and future POPs. The investigation of environmental contamination through POPs comprises three main elements: determining emission patterns, field measurements, and modeling. Here we focus on multimedia models describing the global transport and fate of POPs. Such models mainly serve two purposes: (i) understanding the environmental processes leading to global contamination through POPs and developing indicators that would allow a clear definition of what constitutes a POP (this can be achieved with idealized release scenarios such as a single point source); (ii) assessment of the actual risk to environmental and human health, requiring the determination of environmental concentrations, based on reliable emission estimates and in relation to toxicity thresholds. A particular challenge associated with describing the environmental fate of POPs is that they are multimedia chemicals, i.e., relevant amounts of POPs are present in different environmental media such as soil, surface and ocean water, vegetation, air (in gaseous and particle-bound form), and that they are reversibly exchanged between these different media. This implies that their environmental fate is influenced by the different degradation and transport processes occurring in all these media.Accordingly, computer models of the multimedia fate of POPs are helpful tools because they make it possible to combine theoretical descriptions of different processes in a consistent mathematical framework that is open to expansion and adaptation if new processes are to be included. Models are thus useful for testing hypotheses about the system under consideration. By means of a sensitivity analysis, models show what factors are most influential and should be investigated in more detail. Finally, models guide our understanding of complex systems which are not fully accessible [6]. In this chapter, we briefly review basic approaches to modeling the environmental fate of POPs (Sect. 2), describe the input data required for and results obtained from POPs multimedia models and summarize results from some modeling studies performed for selected POPs (Sects. 3 and 4). We conclude with a

240

M. Scheringer · F. Wania

discussion of model validation issues and an overview of open questions and needs for further research (Sects. 5 and 6).

2 Modeling Approaches POPs have some characteristics that should be reflected in the design of models that seek to simulate their global environmental fate [7]: As mentioned in the introduction, POPs are multimedia chemicals with notable amounts in several environmental media and reversible exchange processes between these media. Therefore, single-media approaches tend to be insufficient when aiming for a comprehensive understanding of the long-term fate of POPs. Secondly, POPs degrade slowly or very slowly in the environment so that they become relatively well-mixed in the environment and their distribution between different media is governed by the tendency to establish thermodynamic partitioning equilibria. In addition to the typical local or regional contamination patterns developing around a strong source, POPs also reveal large-scale or even global contamination patterns that are not strongly influenced by single sources and short-time events. Thirdly, POPs are difficult to measure in the environment, which precludes frequent and spatially highly resolved measurements. This, in turn, limits the availability of measurement data for use in model calibration and confirmation. Finally, emission inventories for POPs are often incomplete and highly uncertain. These factors imply that models for POPs have to account for the multimedia processes but do not necessarily demand a very high spatial and temporal resolution. The basic processes that have to be covered by global models for POPs are: (1) (2) (3) (4)

chemical releases (idealized or reflecting actual emission patterns), degradation and transformation, reversible intermedia exchange, and transport in atmosphere and oceans.

A systematic combination of these processes leads to mass-balance equations that balance the input and output of a chemical in each environmental medium and spatial compartment. Global models constructed that way belong to the type of multimedia box models that have been used widely for various purposes [8]. These models have the advantages that they are relatively easy to construct and use and that the computational effort required for the model solution is relatively low. The simplest global multimedia box models contain one soil, water, and air compartment. More comprehensive and detailed descriptions of certain media and processes lead to models comprising different types of soil, surface and ocean water, vegetation, different atmospheric layers, and the corresponding exchange processes [7]. Nevertheless, the spatial resolution of these models tends to be rather low, i.e., a single box represents an area of several thousand square kilometers. The averaging character of these models makes them particularly suitable for investigating the long-term development of environmental contamination through long-lived chemicals.

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

241

Another approach to modeling the global or at least large-scale fate of POPs builds upon atmospheric dispersion models that have been developed and used for calculating the atmospheric transport and deposition of pollutants such as NOx and SOx , aerosol particles, or heavy metals. They are based on a dynamic description of the air flow in the atmosphere and use a relatively high spatial and temporal resolution, e.g., a grid of 50 km¥50 km and time steps in the range of hours [9–11]. The Earth’s surface constitutes normally a more or less complex boundary in such models and concentrations and processes in environmental surface media such as water, vegetation, and soil are not described explicitly. If such atmospheric dispersion models are adapted to the needs of POPs modeling, these media and the processes taking place within and between them have to be incorporated. In this context, the modeled spatial resolution of the surface media becomes an important question because it determines the need for environmental input parameters, the number and types of processes that have to be calibrated, and the computational effort. When these models are adapted to POP-specific needs or when simpler box models are refined by introducing a higher number of spatially separated compartments, an intermediate type of model may emerge [12, 13]. Since both types of models, multimedia box models and atmospheric dispersion models, are based on the principle of mass conservation, this correspondence is plausible. However, while both model types are based on the same principles, it should be kept in mind that they have been constructed for different purposes and, accordingly, have different advantages and limitations. Multimedia fate box models: – are easier to understand, test, and use; their results are easier to interpret, – require less computational effort in terms of computer code, computer power, and computation time, – are more flexible for adjustments and inclusion of new processes, – have only low resolution and are not capable of investigating specific situations; model parameters are based on temporal and spatial averages, and thus have only limited capability of reflecting environmental variability. Atmospheric dispersion models: – have higher spatial and temporal resolution, and are capable of calculating single episodes of pollutant release and transport, – require higher computational effort, which makes long-term calculations of decades or centuries not possible or very time consuming (only a factor of about 10 to 100 faster than real time), and – require large numbers of model parameters and are difficult to calibrate due to the lack of high-resolution data for environmental parameters. As mentioned before, there are two primary applications of both types of models. The first is the evaluative use that aims to explore the basic principles of global environmental transport and accumulation. This use does not require actual emission data but can be based on idealized emission scenarios such as single pulse or steady-state emissions. This approach provides insights without being impeded by the lack of emission data. Here, mainly multimedia box models are used because they are easier to adapt to different scenarios.

242

M. Scheringer · F. Wania

The second use is the estimation of real concentrations, fluxes and amounts, based on actual and estimated emission data. Only this model use can be evaluated by comparison with observations and is applicable in a risk assessment in terms of exposure vs. effect concentrations. Both types of models are used in this way. Because of their low spatial resolution, multimedia box models are especially suitable for long-term calculations. Dispersion models, on the other hand, can be used for tracking the pathways of specific, i.e. temporally and spatially well defined, releases.

3 Data Requirements and Modeling Results In this section, we describe the types of parameters that are required for performing global multimedia modeling studies and the results that can be derived. 3.1 Model Input Data

Required input data comprise (1) physical-chemical properties of the chemicals, (2) environmental parameters, and (3) emission data. Many of these parameters are not always easily accessible and have to be estimated by extrapolation methods. Those values that are available for a given parameter often exhibit significant variability and uncertainty. Therefore, careful data selection and documentation is an important task when developing and using a global fate model. 3.1.1 Chemical Properties

Most models require the input of two of the three partitioning coefficients between the vapor phase, the aqueous phase and the n-octanol phase, namely the Henry’s law constant (H, in Pa m3/mol, or KAW , dimensionless), the octanol-water partitioning coefficient (KOW , often as log KOW), and the octanol-air partitioning coefficient (KOA).Alternatively, a model may require the input of the three “solubilities” in air (vapor pressure PL), water (CW) and octanol. Degradation rate constants for air, water and soil also belong to the minimum set of chemical properties. Measured or estimated values for the partitioning properties and “typical” half-lives in soil, water, and air are available from compilations such as those by Howard et al. [14] and Mackay et al. [15]. If a model accounts for variable temperatures in either space or time, thermodynamic quantities describing the temperature dependence of partitioning coefficients (energies of phase transfer) and degradation rate constants (activation energies) need to be known as well. Such data for POPs have been compiled by Shiu and Ma [16, 17] and others [18, 19 and references therein]. The property values listed for POPs in these compilations often show high variability. Pontolillo and Eganhouse [20] recently documented this variability using the aqueous solubility and the KOW of DDT and DDE as an example, suggesting that even after three decades of measurements these properties remain

243

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

uncertain by approximately two to three orders of magnitude. One reason for this variability is that these properties are difficult to determine experimentally. Similarly, the reaction rate constants of many POPs in the gas phase have not been determined experimentally because their low vapor pressure leads to experimental difficulties in controlling gas phase concentration.Values estimated from quantitative structure-property relationships are often used instead. The degradation rate constants of POPs are additionally variable because they are influenced by environmental conditions. For example, atmospheric degradation rate constants depend on temperature, OH radical concentrations and the concentration and nature of aerosol particles to which the chemicals may sorb [21–23]. As a result of their uncertainty, the measured partitioning coefficients and the associated energies of phase transfer are often not internally consistent. Beyer et al. [19] have proposed a method of selecting the parameter values such that they are consistent with each other, yet deviate minimally from the reported measurements. The 12 or 16 POPs exhibit considerable differences in their properties (Table 1 and Fig. 1): Their KOW values spread over four to five orders of magnitude, the

Table 1. Partitioning properties (log KOW , log KAW , log KOA) and degradation half-lives in soil,

water and air at 25 °C. Except for heptachlor and heptachlor epoxide, the partitioning properties are taken directly from Beyer et al. [19] or, in the case of dieldrin and aldrin, were obtained by using a similar approach. They are based on measured data reported in the literature and have been adjusted to be internally consistent. Values for heptachlor and heptachlor epoxide are geometric means of data given in Mackay et al. [15]. Chemical Name

CAS no.

t1/2 soil a (h)

t1/2 water a t1/2 air b (h) (h)

log KAW (–)

log KOW log KOA (–) (–)

aldrin chlordane c p,p¢-DDT p,p¢-DDE dieldrin heptachlor heptachlor epoxide hexachlorobenzene a-HCH g-HCH PCB 28 PCB 101 PCB 153 2,3,7,8-TCDD

309-00-2 57-74-9 50-29-3 72-55-9 60-57-1 76-44-8 1024-57-3

3.72·103 1.71·104 3.36·104 3.89·104 1.49·104 5.06·103 3.24·103

2.68·103 1.95·104 4.62·103 8.76·103 1.04·104 1.00·102 2.07·103

2.86 1.64·101 5.60·101 5.60·101 1.27·101 3.13 2.35·101

–1.66 –2.51 –3.33 –2.84 –2.74 –1.22 –2.02

5.84 5.94 6.16 5.95 5.55 5.00 3.25

8.01 8.99 10.11 9.33 8.69 6.23 5.27

118-74-1

3.41·104

3.41·104

1.19·104

–1.40

5.48

7.25

319-84-6 58-89-9 7012-37-5 37,680-72-3 35,065-27-1 1746-01-6

1.03·103 5.00·103 5.50·104 5.50·104 5.50·104 2.90·104

1.03·103 5.27·102 5.50·104 5.50·104 5.50·104 2.31·103

3.61·101 6.24·101 2.06·102 5.50·102 1.10·103 3.68·101

–3.56 –4.17 –1.90 –1.97 –2.04 –2.31

3.98 3.95 5.71 6.35 6.75 6.83

7.37 7.95 8.06 9.00 9.62 10.00

a b

c

Geometric mean of values from Howard et al. [14] and Mackay et al. [15]. Geometric mean of values from Howard et al. [14] and Mackay et al. [15]. Estimates for gas phase reaction with OH radicals, based on [53]. Effects of aerosol particles not included. At least in some cases, structurally related metabolites are formed. Partition coefficients for trans chlordane; half-lives not specified.

244

M. Scheringer · F. Wania

Fig. 1. Diagram showing the location of selected POPs in a space delineated by the log KOA and log KAW of a compound. Reflecting their multimedia partitioning characteristics, POPs tend to fall in the center of such plots when compared to the “universe” of potentially occurring property combinations. Values of log KOA and log KAW were taken from Table 1

KAW values over more than four orders of magnitude, and the estimated half-lives in air (gas phase) range from a few hours (aldrin, dieldrin, heptachlor) to several months (hexachlorobenzene, chlordecone, mirex). Data-related problems faced when modeling the global fate of POPs are: – Often the physical-chemical properties of POPs are not known as a function of temperature, preventing the reliable prediction of partitioning equilibria at the temperature extremes encountered in tropical and polar latitudes. – In most models, partitioning equilibria involving environmental phases, e.g., for water-suspended solids, air-soil, air-vegetation and air-aerosol, are estimated from empirical relationships with KOW , KOA , PL or CW . These empirical equations were typically derived using aerosols, soils and vegetation sampled at temperate latitudes. It is unknown whether these relationships are applicable to the conditions, soils and species of other world regions. – Some fate processes may be unique to non-temperate latitudes and the capability to treat them in chemical fate models may not yet exist. An example is the description of the effect of ice and snow in high latitudes and altitudes or the treatment of arid environments where low humidity and very low levels of living and dead organic matter may cause mineral surfaces to play a much larger role as an environmental reservoir than in vegetated regions.

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

245

– Degradation half-lives are often known only for typical laboratory conditions (pressure, temperature, compositions of water and soil used in experiments) or for field conditions typical of temperate regions.Values for different situations are not known. In the models, the limited amount of available data is used, either in a hypothetical scenario of averaged environmental properties so that more extreme conditions are not reflected by the model, or, with more complex models, under the assumption that the parameters can be extrapolated to more extreme conditions than covered by the measurements. This is sufficient to gain some first insight into the global behavior of POPs, but obviously introduces significant uncertainties. In view of the great variability of conditions experienced by a chemical on the global scale, we do not have sufficient and sufficiently reliable physicalchemical data and knowledge of chemical fate processes to completely describe and assess the global fate of POPs. Chemical properties and transformation rate constants of POPs should be determined under a much broader set of conditions than is currently done. Further, it is desirable that relevant POP transformation products are identified and that the properties of these transformation products are determined as well. Typically, POPs are not mineralized in a single reaction but are transformed into structurally related compounds which then degrade further. Such transformation products can be toxic and/or persistent themselves. Examples of relevant POP transformation products are DDE for DDT, dieldrin for aldrin, and heptachlor epoxide for heptachlor. The final aim should be to include the dynamic formation of transformation products into the global modeling of POP fate [24]. 3.1.2 Environmental Parameters

Since typical multimedia fate models consist of boxes that represent large sections of the environment, the environmental parameters used in the models are typically averaged across space and time. Relevant parameters are precipitation rates, aerosol deposition velocities, wind speed and eddy diffusion coefficients in water and air, the composition of soil, water and air compartments, as well as their temperature and pressure. If a model has spatial and temporal resolution, a more differentiated pattern of parameter values may be appropriate for selected parameters by, e.g., defining temperatures and wind speeds that vary in time and between different climatic zones. The number of spatially and temporally resolved parameters may be increased if justified by the relevance of a parameter and the extent of its environmental variability. Examples are the soil type and composition, or the extent and type of vegetation present in a certain area. For parameters that are to be resolved to a greater extent, a sufficiently extensive and reliable empirical basis or a plausible estimation technique has to be found so that their environmental variability can be represented by a suitable type of distribution, e.g., log-normal or uniform. Models with sufficiently high spatial res-

246

M. Scheringer · F. Wania

olution might be based on information from Geographical Information Systems [25, 26]. Atmospheric transport models utilize meteorological data for calculating atmospheric processes with high resolution. For the other media, the same approaches and data can be used as for multimedia box models. It might be difficult, however, to find data with sufficient spatial and temporal resolution for all relevant parameters. 3.1.3 Emission Data

For exploring the basic behavior of POPs in a multimedia fate model, a simplified or idealized emission scenario such as a single point source (or several point sources at different points and into different media) is sufficient. Such an emission scenario provides results for all mass-independent quantities such as the chemical’s persistence or the relative magnitude of amounts in and fluxes between environmental compartments. If absolute concentration estimates are to be obtained, information on the actual emission pattern of a POP is required. This is a particular challenge for global scale POP models because of the large spatial and temporal scale of the relevant emissions. If the chemical is sufficiently mobile and persistent, emissions in any part of the global environment may affect concentrations in any other part. Further, environmental concentrations of many POPs are not controlled by current use and release, but by emissions that may have occurred decades ago. Reliable historical emission data for the entire globe are thus required if the present concentration pattern is to be determined. An additional complexity arises from the fact that several POPs, most notably the PCBs, PCDD/Fs and the PAHs, are not pure substances, but mixtures of a large number of individual compounds with often highly divergent characteristics. Divergent properties necessitate separate modeling efforts, and thus substance-specific emission estimates. Although reporting of emissions is required by some international conventions [e.g., the Convention on Long-Range Transboundary Air Pollution (LRTAP) and the Stockholm Convention], for the majority of POPs it is difficult, if not outright impossible, to assemble global inventories that cover a time scale reflective of the persistence of POPs, which may be as high as several decades. For many POPs, therefore, the information on global releases is fragmentary. Common problems are: – Emission estimates are available for some parts of the world (e.g., for North America, Europe) but not for others, necessitating some sort of extrapolation to fill the gaps. – Production and usage figures are reported for relatively large geographical units (e.g., for continents or states), and have to be allocated to the smaller spatial and temporal units finding use in the models. Surrogate measures such crop area, population distribution or Gross National Product (GNP) have to be employed in such allocations.

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

247

Currently available and/or under development are global emission inventories for HCHs [27, 28], PCDD/Fs [29, 30], and PCBs [31]. Fragmentary data are available for other POPs.2 Information on the uncertainty of these estimates is mostly lacking although first efforts are being made to address this issue [31]. 3.2 Model Results 3.2.1 Basic Model Output

In principle, a multimedia model provides a complete mass balance for each individual compartment of the model at each point in time. If necessary, each single mass flow and mass content can be extracted from the model calculations. In most cases, however, it is useful to concentrate on some principal quantities such as the mass distribution among different media or different geographical regions, the elimination rates in different media (also in comparison to the amounts stored in the media so that storage media and elimination media can be identified), the horizontal mass flows in water and air in comparison, the mobile fraction leaving the place of release, or the direction of the net flow between water and air. 3.2.2 Aggregated Indicators

From the masses, concentrations and flows, aggregated quantities characterizing the chemical’s behavior on a less detailed level can be calculated. Such quantities are helpful indicators for the comparison and assessment of different chemicals. Typical indicators of this type are: – The overall persistence [32–35], which can be obtained from the model results as the ratio of the mass present in the model system and the mass flow through the system. For POPs, it is most appropriate to use a closed global model and to determine the persistence from the entire mass present in the model. Calculated that way, the overall persistence integrates all different degradation rates in the different media according to a chemical’s partitioning behavior. – The spatial range [33, 36] measures the spatial extent of a concentration distribution by quantifying the distance (or area) that contains 95% of the weight 3 of this concentration distribution. It indicates the potential for longrange transport of a chemical. A related quantity is the characteristic travel 2

3

See the data provided by the Co-Operative Programme for Monitoring and Evaluation of the Long Range Transmission of Air Pollutants in Europe (EMEP): http://www.emep.int/emissions.html and http://www.msceast.org/pops/emission.html and by the Global Emissions Inventory Activity (GEIA): http://weather.engin.umich.edu/geia/. The weight is the sum of all concentration values of a discrete distribution or the integral of a continuous concentration function c(x).

248

M. Scheringer · F. Wania

distance [37, 38]. For a comparison of the two concepts, see [39]. Persistence and spatial range are not directly related because they are influenced by a chemical’s partitioning between the environmental media, which differ with respect to phase mobility and the chemicals’ degradation rate constants. Thus, different spatial ranges can be observed for a given overall persistence. For example, if released to soil, the overall persistence of a POP is much larger than the atmospheric residence time that would be required for its long-range atmospheric transport. Two-dimensional plots of spatial range vs. persistence such as shown in Fig. 2 help to visualize the different transport potential and persistence of different chemicals. – The cold condensation potential [40] and the arctic accumulation potential [41] intend to describe a chemical’s tendency to move to and accumulate in polar regions. The cold condensation potential indicates whether a chemical’s

Fig. 2. Spatial range in air, R, vs. overall persistence t of selected POPs calculated with the

ChemRange model after release to the soil [1-butanol and F-11 (CCl3F) are shown for comparison]. Chemical-specific input data are from Table 1. The lines indicate the intervals spanned by the scenarios “all adsorbed” (stars) and “all gaseous” (circles), see text and the discussion given in references [36, 81]. With modification reproduced from Scheringer [81, p 187]

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

249

concentration profile increases towards the poles. It is calculated as the ratio of the concentration in the polar zone and the minimum concentration between equator and polar zone. The arctic accumulation potential expresses the fraction of the total global amount of a chemical that has accumulated in the Arctic zone, based on a generic emission scenario. These indicators are suitable to characterize, compare, rank, and assess POPs as it is, for example, required by the assessment procedure delineated in the Annexes D, E, and F of the Stockholm Convention.

4 Model Calculations 4.1 Description of Selected Models

In this section, we summarize some of the results on the global chemical fate of POPs that have been obtained with multimedia box models and atmospheric dispersion models. Global multimedia box models considered here are: – the global model by Strand and Hov [42], which is an atmospheric transport model originally developed for atmospheric tracers [43] and adapted for application to a- and g-HCH (Fig. 3).

Fig. 3. Structure of the global model by Strand and Hov [42]

250

M. Scheringer · F. Wania

Fig. 4. Structure of the Globo-POP model by Wania and Mackay [44]

– the climatic zone model Globo-POP developed by Wania and Mackay [44], which evolved from an earlier global multimedia model [45, 46] by adapting a two-dimensional representation 4 of the global atmosphere presented by Strand and Hov [43] (Fig. 4). A steady-state version of this model was presented by Klein [47]. – the spatially uniform circular and spherical models presented by Scheringer [33] and Held [48] (Fig. 5), and – the climatic zone model “CliMoChem” by Wegmann [49] and Scheringer et al. [40] (Fig. 6). These four multimedia box models are listed and compared in Table 2. They represent several different conceptual and technical approaches to calculating the global fate of POPs. Atmospheric dispersion models adapted for POPs have been presented by Koziol and Pudykiewicz [10], Pekar et al. [9], Sofiev and Galperin [11], and Lammel et al. [50]. 4

Note on the usage of “dimensions” of a model: In atmospheric modeling, models of a column of air at a certain location are called one-dimensional, models with vertical resolution and transport in north-south direction are two-dimensional, and models with vertical, northsouth and east-west resolution are three-dimensional. Most multimedia models, on the other hand, do not have several atmospheric layers so that the vertical dimension does not exist. Thus, models with explicit transport in north-south direction are called one-dimensional.

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

251

Fig. 5. Structure of the ChemRange model as formulated by Scheringer [33], and Held [48]. Reproduced with permission from Scheringer [81, p 138]

Fig. 6. Structure of the CliMoChem model by Scheringer et al. [40]. Reprinted with permission

from Scheringer et al. [40]. Copyright (2000) American Chemical Society

Comparing the results obtained with several models of the same type as well as comparing results from multimedia box models and dispersion models helps to identify similarities and differences, to understand the effects of mathematically and conceptually different modeling approaches; it also contributes to the evaluation of the models. First, we give an overview of the characteristics of the some of the models. In Sect. 4.2, we describe some findings obtained with idealized release scenarios, illustrating some general aspects of the global transport of POPs. In Sect. 4.3, we present results of model calculations aimed at estimating actual environmental concentrations, fluxes, and mass budgets based on historical emission estimates for selected POPs.

Globo-POP

CliMoChem

KOA-based as described [76]

None

User-specified gas phase reaction rate constant

Diffusive gas exchange and rain dissolution; no particle bound deposition

Treatment of gasparticle partitioning

Treatment of atmospheric degradation

Treatment of atmosphere-surface exchange

One macro-diffusion coefficient based on average of values given in [56]

Gas phase reaction rate cal- User-specified gas phase reaction rate culated as a function of constant; reaction rate constant of sorbed spatially and temporally fraction can bespecified by user; default: variable temperature; no reaction of sorbed fraction reaction rate constant of sorbed fraction can be specified by user; default: no reaction of sorbed fraction

Sorbed fraction can be chosen as a user-defined chemical input parameter or is derived from KOA [76]

zonally averaged macrodiffusion using diffusion coefficients from [56]

None

Diffusive gas exchange, rain dissolution and wet and dry particle deposition

Gas phase reaction rate constant calculated as a function of spatially and tempo rally variable OH con centration and temperature; no reaction of sorbed fraction

zonally averaged advection and macro-diffusion coefficients interpolated from GCM [43]

4 atmospheric layers (stratosphere, middle/upper None troposphere, lower/middle troposphere, atmospheric boundary layer)

Treatment of atmospheric transport

Vertical atmospheric resolution

Atmosphere

Atmosphere, surface Atmosphere, surface atmosphere, surface ocean, atmosphere, surface ocean, soil ocean, several deep ocean ocean, fresh water, fresh soil layers, 2 soil types water sediment, 2 soil types

Flexible number n of zones Boxes are not zones, but discretization for corresponding to latinumerical solution; analytically solved tudinal range of (180°/n) version has no zones

Compartments

Ten zones based on climate zones defined in [75]

Closed loop

ChemRange

Six zones of 30° latitude

Zonally averaged units that differ in size and environmental parameters

Bergen model

Zonal subdivision

Principle design

Model name

Table 2. Comparison of the major characteristics of the four multimedia box models described in the text and depicted in Figs. 3 to 6

252 M. Scheringer · F. Wania

User specified half-life

Air, water, and organic carbon with fractions of 0.2, 0.3, and 0.0125

constant soil depth of 0.15 m

Composition of the soil types

Depth/vertical resolution

Zonally variable soil depth that can be changed by user

Air, water, and organic carbon; relative size of these constituents is zonally variable and can be changed by user

Cultivated (receives emissions) and uncultivated soil (receives no emissions)

constant soil depth of 0.1 m

Air, water, and organic carbon with fractions of 0.2, 0.3, and 0.02

one soil type (receives emissions)

User-specified reaction rate constant

Sorption to particles and settling was not included in the versions described in the references. In the present versions of the models, sorption and settling have been included.

One macro-diffusion coefficient based on data from [55]

Reaction rate constant calculated as function of spatially and temporally variable ocean water temperature

Soil types

Terrestrial Environment

Treatment of oceanic degradation

Sorption based on empirical relationship between KOC and KOW [78]; settling based on carbon export rates by Falkowski et al. [79].

Based on vertically rezonally averaged macrosolved HILDA model by diffusion coefficients Siegenthaler and Joos [77] based on average water residence time in various marine areas from various literature sources

Treatment of sorption None to suspended solids and vertical particle settling

Treatment of oceanic transport

Marine Environment

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

253

Turbo Pascal

Not available

Strand & Hov, 1996 [42]

Software

Availability

Reference

Wania & Mackay, 1999b, 2000 [44, 63]

www.utsc.utoronto.ca/ ~wania

MS Visual Basic

a-HCH, seven PCB congeners

a- and g-HCH

Application using realistic emission scenarios

Linear interpolation of monthly values for temperature, OH radical concentration, sea ice cover and atmospheric transport parameters

Globo-POP

90 linear mass balance equations in fugacity notation solved stepwise using finite difference approximation

Four seasonal values for air temperature, precipitation and atmospheric transport parameters

Bergen model

Mathematical solution Step-wise solution of zonally averaged continuity equation using finite difference approximation

Other Treatment of innerannual variability

Model name

Table 2 (continued)

http://ltcmail.ethz.ch/hungerb/research/ product/product.html

MS Excel and Visual Basic

Not applicable

System of 3 · n linear mass balance equations solved by matrix inversion [33] or analytically derived continuous functions c(x) for the three media [48]

No innerannual variation

ChemRange

Wegmann 1999 [49] Scheringer, 1996, 1997, Scheringer et al., 2000 [40] Held, 2002 [33, 36, 48]

Not yet available

Mathematica

a-HCH, p,p¢-DDT

3· n linear mass balance equations solved analytically for each time period with constant conditions

user-defined number (max. 12) of “seasons” per year; temperatures based on data from [80]

CliMoChem

254 M. Scheringer · F. Wania

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

255

4.2 Model Calculations with Idealized Release Scenarios 4.2.1 Spatially Uniform Multimedia Box Models

Idealized release scenarios such as single point releases make it possible to investigate the general behavior of POPs in a model system. For this purpose, mainly multimedia box models are used because different scenarios can be run and compared rather easily. Since the models are linear, all relevant characteristics can be determined when a single pulse release is used; solutions of complex release scenarios are simply superpositions of the results obtained for single point releases [51]. The simplest global model is the circular model “Chemrange” [33, 36, 52]. It consists of a circular tube containing soil, oceanic surface water, and tropospheric air (Fig. 5). Processes covered are first-order degradation, phase-partitioning with globally and annually averaged parameters, and macro-diffusive transport in air and water. The simplest release scenario is continuous release from a single point source; the results obtained with the model are the steady-state concentrations in all media, and therefrom the overall persistence and the spatial range of a chemical are derived. A relevant finding for POPs is that the spatial range in air as a measure of long range transport potential (LRTP) depends strongly on the effective removal rate constant in air, which, in turn, is based on degradation, washout, particle deposition, gaseous deposition and revolatilization. For some POPs such as dieldrin, aldrin, chlordane, or endrin, very different spatial ranges from less than 10% of the circumference of the earth up to more than 30% are found if the degradation rate constant in air is varied between theoretical minimum and maximum values. The maximum value is determined by degradation rate constants for the gasphase chemicals as listed by Howard et al. [14], which are based on estimates according to Atkinson [53]. If the degradation rate constant is lowered according to the assumption that the chemical adsorbs to aerosol particles and that the particle-bound fraction is precluded from degradation, the spatial range increases significantly. Reduced degradability of particle-bound material has been reported in the literature [21–23] but the second extreme of this evaluative scenario, complete adsorption and no degradation, is certainly not a realistic assumption under most conditions. Nevertheless, such explorative calculations with a simple model illustrate the possible effect of adsorption to or inclusion into particles and of low temperatures (increasing particle-bound fractions and decreasing OH radical rate constants). They indicate that the effective degradation rate constants in air are likely to be lower than suggested by the currently assumed half-lives of only a few hours for chemicals such as dieldin, chlordane, and DDT. 5 Stated differently, the ob-served extent of long-range transport of 5

This model result corresponds to a finding recently reported by Lohmann et al. [54]. They state for polychlorinated dibenzodioxins and -furans measured along a North-South transect over the Atlantic Ocean that “the influence of OH radical initiated depletion reactions of gaseous PCDD/Fs was not as evident as expected” (p 4052).

256

M. Scheringer · F. Wania

these chemicals cannot be reconciled with the estimates of their atmospheric degradation rates constants. This leads to the recommendation that the influence of temperature on the gas-particle distribution and the effect of both on the atmospheric lifetime of POPs be explored in more detail. Held [48] developed an analytical solution of the circular global model and solved also the two-dimensional (spherical) analogon of this model. He demonstrated that the numerical solution used by Scheringer [33, 36] is sufficiently accurate and that the results obtained with the one-dimensional and the two-dimensional version of the model are quite similar, i.e., the restriction to one dimension does not introduce substantial errors when compared to the spherical geometry. 4.2.2 Multimedia Box Models with Climatic Zones

Multimedia box models with climatic zones simulate the annual temperature course in the different climatic zones and calculate the influence of these temperature changes on the degradation rate constants and partitioning coefficients of a chemical. The climatic zones are represented by well-mixed latitudinal bands with volumes increasing from the terminal polar zones to the large tropical zones. This kind of model represents the spherical geometry of the earth’s surface but includes only an averaged transport dynamics in the meridional direction, represented by large-scale eddy diffusion coefficients for water [55] and air [56]. The assumption of zonally averaged concentrations is based on the observation that atmospheric mixing in the east-west direction is much faster than in the north-south direction. It can, however, introduce errors when the distribution of emissions within a zone is highly non-uniform and mixing in the east-west direction is not sufficiently efficient to smooth out the associated spatial heterogeneity.Another error associated with zonal averaging stems from the inter-zonal crossing of large rivers, such as occurs in Siberia. The number of zones is 10 in Globo-POP (Fig. 4) and varies between 10 and 120 in CliMoChem (Fig. 6). Each zone consists of several compartments representing soil (different types), air (different layers), freshwater, and oceanic surface water. The effect of temperature on degradation rate constants and partitioning constants is included (see Sect. 3). Main findings from calculations with these models are: – The lower temperatures at the poles lead to reduced degradability and a higher tendency of the chemicals to partition into soil and water. The increased affinity to water and soil is a thermodynamic effect, caused by the temperature dependence of the partitioning constants, and is observed for POPs as well as for volatile chemicals such as CCl4 [40, 57] or CFCs. In addition, the deposition processes of POPs are affected by the lower temperature: particle adsorption is increased as is scavenging by rain water (which is a proxy for snow scavenging, which might be highly effective). The model results show some agreement with concentration increases towards the poles observed for a-HCH and CCl4.

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

257

– The mobile fraction of POPs, calculated as the fraction that evaporates from the soil in the zone of release (after emission to the soil), depends to some extent on the vapor pressure but is generally below 5% (mirex, a-HCH, PCBs, DDT). For DDT in CliMoChem, e.g., the non-mobile fraction decreases by about 8% per year through degradation and by 0.08% through dispersion (after a pulse release into the soil at the equator), i.e., the main part remains at the place of release and is degraded there. – The maximum of the concentration in polar soils is reached after 1.5 years (a-HCH), 10 years (PCBs), or 20 years (mirex, DDT). – Persistence and spatial range are dependent on the place of release. The persistence of a-HCH is about 100 days if release occurs near the equator, yet increases to around 250 days for release in a northern temperate zone. If the persistence is calculated as a function of time as the ratio of the total amount in the environment (tonnes) and removal rate (tonnes/year), it increases because the fraction remaining in colder regions becomes more important (and shows a seasonal variation with higher values in winter and lower values in summer). – Cold condensation potential: for a pulse release of a-HCH, the latitudinal concentration distribution changes from one showing a peak around the point of release to one dominated by slowly degrading residues in cold regions (concentration inversion, see Fig. 7). In the case of highly particle adsorbing substances such as mirex or DDT, this concentration inversion is suppressed by a strong concentration gradient from the point of release (equator) towards the poles, which is caused by the very efficient wet and dry deposition of such particle-sorbed chemicals. Note that the temperature dependence of the partition coefficients alone would also lead to a “condensation” (concentration inver-

Fig. 7. Spatial concentration distribution of α-HCH in ocean water, calculated with the

CliMoChem model. After five years, the peaked concentration profile has turned into an “inverted” profile with residues in the polar regions (right). Reprinted with permission from Scheringer et al. [40]. Copyright (2000) American Chemical Society

258

M. Scheringer · F. Wania

sion) in water and soil of these chemicals.A higher number of zones increases the temperature effects because the spatial averages are over smaller regions and more extreme temperatures occur in the polar zones. – Arctic Amplification Potential (AAP): The Globo-POP model can be employed to identify chemical properties and emission situations that favor accumulation in Arctic ecosystems. The criterion defined to quantify this AAP is the fraction of the total amount in global surface media that is in the Arctic zone after a certain time period [41]. The simulations use hypothetical emission estimates with a generic spatial and temporal distribution. When used to assess the AAP of a multi-dimensional “space” of hypothetical chemical property combinations (Fig. 8), Globo-POP indicates that the potential of a perfectly persistent organic chemical to accumulate in the Arctic is determined by a complex set of processes, but tends to be higher for substances with high volatility (low log KOA) that neither sorb to organic phases (low log KOW) nor are very water soluble (high log KAW). The model simulations also suggested that very hydrophobic compounds (log KOW >8) with intermediate volatility (log KOA =7) may have an extremely high potential for accumulation in polar regions, if persistent and emitted to air. Efforts should be directed to identify chemicals with such properties.

Fig. 8. The Arctic Amplification Potential of a perfectly persistent organic chemical with a particular log KOA/log KAW property combination after ten years of steady emissions into the atmosphere as obtained with the Globo-POP model [41]. Property combinations shown in dark indicate a very large potential for accumulation in the Arctic, those in lightshades a very low potential. A comparison with Fig. 1 reveals the location of typical POP property combinations in that diagram and suggests intermediate potential for accumulation in the Arctic

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

259

– Further, the AAP was found to be lower when the chemical is emitted to surface media. This is especially important for soil-emitted chemicals of intermediate volatility (log KOA =6 to 9). Even if fairly volatile (log KOA <8), a chemical with a high log KOW has no significant AAP, when emitted to water. Such chemicals sorb to suspended particles in water and settle to sediments and the deep sea. – Additional AAP calculations quantified the potential for delayed transport to the Arctic, i.e., accumulation in high latitudes well after emissions have ceased. Such behavior, caused by slow oceanic advection, was predicted for very persistent, involatile, yet water-soluble substance (low log KAW). Finally, variable persistence of a chemical in air and surface media can either enhance or decrease its potential to become relatively enriched in Arctic ecosystems. – The mass flow in ocean water is relevant for meridional transport, in particular in high latitudes. Both air and water contribute to the transfer of POPs to polar regions. – The gas-phase degradation rate, kOH , is a very important and at the same time highly uncertain parameter. In general, degradation rate constants have a strong influence on the model results. 4.3 Calculations with Realistic Release Scenarios

The second type of global model calculations for POPs are those that are based on realistic emission inventories, providing model results that can be compared with measured concentration data.6 This type of model application has been impeded by the lack of reliable, sufficiently comprehensive, and spatially and temporally resolved emission inventories for POPs (see above, Sect. 3.1.3). Most calculations of this type focused on α-HCH or several HCH isomers because relatively comprehensive emission data exist for these substances [27, 28]. Strand and Hov [42] as well as Wania et al. [58] performed such studies using multimedia box models, whereas Koziol and Pudykiewicz [10], Sofiev and Galperin [11], and Pekar et al. [9] used a variety of global or hemispheric scale atmospheric dispersion models. Additional studies are being conducted for the PCBs [59–61] and for DDT and DDE [62]. 4.3.1 Multimedia Box Models

Wania et al. [58] and Wania and Mackay [63] investigated the global fate of a-HCH with the Globo-POP model. a-HCH, which is a component of a technical pesticide formulation, was released in massive quantities in Asia in the 1970s, has seen a steady decline in use since then and is now virtually out of use worldwide [28]. Compared to most other POPs, it is relatively water soluble, relatively 6

Such a comparison requires that model-predicted and measured concentrations represent comparable conditions, i.e. refer to similar spatial domains and time periods.

260

M. Scheringer · F. Wania

volatile and less persistent (see Table 1). It is of particular interest from an Arctic perspective because by the 1990s the concentration levels in the Canadian Arctic Ocean were higher than anywhere else in the global marine environment, notably an order of magnitude higher than in the Indian Ocean and the Western Pacific which are much closer to the major source regions [64, 65]. The calculations with the Globo-POP model suggested that this inverted concentration profile developed during the two decades of declining emissions and was the result of the a-HCH being trapped and preserved in the cold Arctic Ocean [59], whereas the HCH levels in source regions declined as a result of degradation and volatilization. The Arctic Ocean thus constitutes the last global refuge of a-HCH, a conclusion from the modeling studies confirmed by other investigations [66]. This demonstrates that inverted concentration profiles can be established even if only a small percentage of the globally emitted amount is transferred to Arctic latitudes. Similar to the results from pulse release calculations, the model predicted that the bulk of the a-HCH never left the agricultural systems in which it had been applied. Nevertheless, relatively small amounts can result in high concentrations in the Arctic because of the relatively small size of the Arctic as a whole and of the environmental phases with high capacity for hydrophobic organic chemicals (organic soils, vegetation, organic sediments) within the Arctic. By binding substances such as HCH, these phases have the ability to reduce the amount present in the atmosphere and in aquatic systems. The model also suggested that oceanic currents are as important in the northbound transport of a-HCH as the atmosphere. Most of the a-HCH in the Arctic originated in the northern temperate zone which also was a net exporter of a-HCH to low latitudes. However, whereas the chemical accumulated in the north, it was rapidly degraded under tropical conditions. Comparisons showed that calculated and measured a-HCH concentrations in the atmosphere and sea water generally agree within one order of magnitude [58, 67]. Deviations are due to the zonal averaging characteristics of the model and uncertainties associated with the environmental degradation rate constants of a-HCH. Interestingly, the agreement between modeled and measured a-HCH concentrations is not necessarily higher if the atmosphere compartment is modeled with a higher vertical resolution. An earlier version of the Globo-POP model, which did not have a vertically resolved troposphere, reproduced measured data to a higher degree than the model by Strand and Hov [42], which has a vertical resolution of the troposphere [68]. The Globo-POP model has also been used to investigate the global fate of polychlorinated biphenyls (PCBs) over a time scale of several decades [59–61]. PCBs have been used as mixtures consisting of individual substances which differ substantially in their physical-chemical characteristics and persistence. It is also likely that the temporal and spatial pattern of release into the environment has been different for different congeners [31]. Large differences in the simulated fate of the various PCB congeners reaffirm the need to perform calculations for individual chemicals rather than chemical mixtures. Calculations were thus performed for a selection of congeners that vary in the number of chlorine substitutions.

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

261

One of the motivations for modeling PCBs globally is to identify the major global loss processes in order to assess the likely rate of future concentration decline in the Arctic environment and elsewhere [61]. The model calculations showed that historically atmospheric degradation and transfer to the deep sea contributed the most to the loss of PCBs from the global environment, whereas burial in fresh water sediments is of little significance on a global scale. Reaction of the gaseous compound with OH radicals is the loss process of primary importance to the lighter congeners, whereas deep sea transfer increases in relevance with the degree of chlorination. The model further predicts that the relative importance of the various loss processes has been changing in time, with degradation in soils taking over as a major loss process in the past twenty years. As primary emissions decreased, the concentrations (and thus the loss rates) of PCBs in atmosphere and ocean water have declined quickly, whereas soils retained a high pollutant load due to their slow response time (large capacity, but slow evaporation and degradation). Unfortunately, this implies that the future rate of purification of the global environment will be determined by the slow and poorly quantified degradation rate in the soil environment. The results from the Globo-POP model also demonstrate compositional shifts among the PCB congeners between compartments, zones and different time periods. Due to their wide range of physical-chemical properties, PCBs have played an important role in the derivation of the concept of global fractionation, which results in compositional shifts of compound mixtures with latitude [65, 69]. The model reproduced shifts towards lighter PCB congeners with increasing latitude [59], as have been observed in various measurement campaigns [70, 71]. The model further suggests that these shifts are rather complex, namely differ between various environmental media, have sometimes surprising anomalies, and change in time. It should be stressed that these global fractionation patterns have been established with only minor fractions of the global inventory of PCBs being transferred northward. 4.3.2 Atmospheric Dispersion Models

With atmospheric transport models, the transport in and deposition from air of a pollutant can be calculated with considerably higher spatial and temporal resolution than that of box models. As these models are typically applied to certain periods of time for which measurement data are available, e.g., one or two years, the background concentrations in the different media at the beginning of a particular model simulation have to be incorporated into the initial conditions. Calculations with atmospheric dispersion models have so far concentrated on the HCH isomers because, on the global scale, the most reliable, spatially resolved global emission inventories are available for these chemicals. The main results obtained from such studies are: – Maps of spatially resolved concentration and deposition patterns as well as mass balances for the region of investigation.

262

M. Scheringer · F. Wania

– Trajectories of pollutant transport during certain episodes, calculated forward (predicting future development) or backward (tracing the origin of pollutants deposited at a certain point). – The spatial distribution of deposition fluxes in comparison to the spatial distribution of releases, indicating the transport directions and efficiencies. – Traveling times and half-lives of the airborne fraction of a chemical. – Pathways of atmospheric transport to the Arctic, which occurs by episodic air exchange between mid and high latitudes. Such exchange takes place most frequently over the northeast Atlantic and, somewhat less frequently, over the northeast Pacific [72]. Agreement between modeled and measured air concentrations on the order of one magnitude or better has been noted (but such agreement is not necessarily higher than that obtained with multimedia models). In some instances, there are systematic deviations that are attributed to certain model assumptions (e.g., incorrect estimates of precipitation patterns or parameterization of air-soil exchange) or to the highly uncertain emission data. Several authors state that the uncertainty in the emission data contributes significantly to the uncertainty of the calculated air concentrations [10, 50, 73].

5 Model Reliability and Validation There are several aspects under which the validity and reliability of global models for POPs can be discussed. Especially with global models, which reflect selected features of an extremely complex system, one should keep in mind that a “straightforward” verification or validation of models by comparison of model outcome and observational data is not possible [6]: First, natural systems are never closed so that unknown factors not covered by a certain model can always influence observational data – and this also applies to each new version of a model that has been designed to include such newly discovered factors. Second, model results consistent with observational data can always be obtained with different models, i.e., the algorithm leading to a certain outcome is not unique. Of course, consistency between model predictions and observations will lend credibility to a model (Oreskes et al.[6] suggest the term “confirmation”), and an increasing number of such consistencies will increase the degree of confirmation of the model but this process is never completed, so confirmation does not imply “verification”. In addition, not all global models for POPs aim at providing concentration estimates that could be compared to measured environmental concentrations. For models designed to investigate basic mechanisms of POPs fate without relying on realistic emission scenarios (Sect. 4.2), other methods of evaluation are necessary. Here, we distinguish, following Oreskes et al. [6], model validation, calibration, and confirmation; in addition, we illustrate the application of sensitivity and uncertainty analysis to global models for POPs. – Validation: A model can be considered valid if it is logically consistent, if the model algorithm is based on sound scientific concepts and if it has been con-

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

263

verted into a correct and appropriate computer code. Although this seems to be obvious, fulfillment of these criteria is not easy to prove when a model has reached a certain level of complexity. Thus, global models for POPs should be checked for these aspects of validity in all stages of their development. – Calibration: Often there are some model parameters that have to be given appropriate numerical values, and this can be accomplished by comparing the model results to experimental data providing a suitable benchmark. An example is the macroscopic (eddy) diffusion coefficient of the circular model ChemRange which was tuned in a way that the model yields a uniform global distribution of an atmospheric tracer within one year (which is the known global mixing time of trace gases in the atmosphere). – Confirmation: This refers to the degree of consistency between model results and experimental observations. For multimedia models with boxes representing large environmental areas, it has to be checked whether the measurement conditions reflect the assumptions of the model, e.g., zonal averaging. If the model uses actual emission data, absolute concentration and mass flux values can be compared with measured ones. Other model results that can be compared with observations are temporal and spatial trends of concentrations in air and water; fugacity ratios, mass distributions among different environmental media, or the relative composition of mixtures such as the PCBs. To some extent, these aspects of a chemical’s global fate can also be evaluated if idealized emission data or more generic models are employed. It is more difficult to evaluate a calculated potential for LRT because it is presently unclear how such a potential could be quantified empirically. The occurrence of a chemical in remote areas does not necessarily imply that the chemical is subject to LRT or exhibits a high spatial range, because it may be due to a very high emission rate or an extremely low detection limit. Sensitivity and uncertainty analyses indicate the influence of uncertain model parameters and model assumptions on the model results. The main purposes of these techniques are to identify the major uncertainties and variabilities in the input data as well as in the model construction and to determine the influence of these uncertainties on the model results (sensitivity analysis, comparison of different modeling scenarios, (formal) uncertainty analysis, e.g., Monte Carlo (MC) simulation, comparison of different models for the same or similar purposes). Results obtained by such analyses for POPs comprise the following points: – A model comparison including MC calculations for the ChemRange model and the advective model used for calculating the characteristic travel distance [39] demonstrated that both models yield similar rankings of different chemicals according to the spatial range (both models are suitable for a screening level POPs assessment). Furthermore, the uncertainty associated with different models for particle adsorption and the influence of this uncertainty as compared to other parameter and model uncertainties was investigated, showing the high importance of gas-particle partitioning for the fate and transport of POPs. – A recent model comparison between ChemRange [52] and TaPL3, an LRT assessment model developed by Beyer et al. [38], focused on the relative poten-

264

M. Scheringer · F. Wania

tial for long-range transport of PCB congeners of different degree of chlorination. It showed that the LRT potential of highly chlorinated PCBs is affected by irreversible removal of these compounds by deposition from the water column to the sediment (TaPL3) or to the deep sea (ChemRange). The first version of ChemRange (1.0) did not include transfer to the deep sea and suggested that the LRT potential of PCBs increases continuously with the degree of chlorination.7 In TaPL3, in contrast, the LRT potential first increases with chlorination but then decreases for highly chlorinated congeners. This discrepancy leads to the conclusion that the LRT potential of POPs strongly sorbing to suspended particles and being removed from the surface water with these particles might have a reduced long-range transport potential. – Comparison of different model algorithms has shown that the numerical and analytical solutions of the ChemRange model agree well; furthermore, the model behaves similarly in one and two dimensions [48]. – Another method that shows the response of a model to varying input parameters is to calculate a model result such as persistence or fractions in different media for a broad range of KAW –KOW combinations. This provides contour plots of the chosen model result in a two dimensional KAW –KOW space. Regions of closely spaced contour lines in such plots indicate regions of high sensitivity of the model result to the physical-chemical partitioning characteristics of the chemical. Examples are the investigation of the partitioning behavior of POPs (see Fig. 1) [74] and the calculation of the AAP ([41], see Fig. 8). – To some extent, sensitivity analyses have been performed for global POPs models [11, 73], showing the reaction of the models to changes in degradation rate constants, boundary conditions etc.. However, due to the complexity of many global models and the high numbers of input parameters and model outcomes, it is difficult to carry out systematic sensitivity or uncertainty analyses for such models. Extensive model confirmation by comparison of modeling results and measurement data is impeded by the uncertain physical-chemical properties and degradation rate constants of most POPs, by the insufficient emission data (two factors on the model input side), and by limited measurements on the global scale (a factor on the data side). This means that several elements of the two sides to be compared are unknown at the same time and sources of disagreement cannot be identified unequivocally.

6 Open Questions and Further Research For the future development and application of global multimedia models for POPs, research in several fields is required. These include the improvement of input parameters and of the methods used for parameter estimation, the under7

The current version of Chemrange (2.1) includes suspended particles and deposition to the deep sea [52].

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

265

standing and inclusion of additional processes and mechanisms, the relationship between modeling and measurement strategies, and the further development of assessment criteria for POPs. 6.1 Input Parameters

To increase the reliability of model calculations aimed at actual concentration estimates and risk assessment, more comprehensive and consistent emission data of all POPs (and possible future POPs) are imperative. Furthermore, the assessment of the future development of global POP contamination requires forecasts of usage and emissions of those POPs that are listed for continued application, for example DDT. For POPs with several different release pathways, such as the PCBs, methods describing these pathways and the fractions entering each pathway need to be developed. For many POPs and other environmental contaminants (and also their environmental transformation products), the physical-chemical properties, the partitioning coefficients describing environmentally relevant phase equilibria (e.g., air-vegetation) as well as the degradation rate constants are still not sufficiently well known. Of particular importance is the influence of temperature, which should be included in the quantification of thermodynamic and kinetic parameters whenever possible. In addition, uncertainty and variability estimates should be developed for all these substance-specific parameters. Another category of relevant model input comprises the various environmental parameters describing media composition, inter-media exchange rates, etc. For these parameters, estimates of their variability (in specific regions as well as on a global scale) will be helpful for integrating a more realistic spectrum of environmental conditions into the models. Relevant points are the absolute range of a quantity, the type of distribution that quantifies a parameter’s variability, and a link between information on parameter variability and the spatial structure of a model or a geographical information system underlying a model. 6.2 Modeling Approaches

Several environmental media and processes included in current multimedia fate models are represented by highly simplified sub-models. If a more appropriate description of environmental media and a more detailed understanding of relevant processes can be gained, this will also improve the overall model reliability. Examples are different types of vegetation or ice and snow, the inclusion of which requires thermodynamic coefficients describing the capacity of these media for POP-type chemicals and also kinetic models representing the exchange processes with other media (e.g., air-vegetation exchange; snow scavenging). Of particular importance for the global distribution of POPs is their fate in the troposphere where there are complex interactions between gasphase and aerosol-borne fractions, both being chemically transformed at – presumably – different rates. This area requires significant research entailing

266

M. Scheringer · F. Wania

laboratory experiments, field studies, and modeling of reactivity and transformation processes. On a next level of model evaluation, the basic assumptions and limitations of multimedia box models with large spatial compartments have to be discussed and the modeling concept has to be compared with more detailed approaches. One point is the question of artifacts introduced by the zonal averaging [63]. In reality, mixing in a latitudinal band is less efficient and fast than assumed in the box models so that significantly different concentrations can be found at different places in the same zone.Another point is efficient coupling of adjacent zones by large rivers carrying significant amounts of (water soluble and particle adsorbed) POPs, which is not considered in present multimedia models. In discussing such inadequacies, however, it should be kept in mind that a trade-off is necessary between model complexity (spatial and temporal resolution as well as “functional resolution”, i.e., number of processes and level of detail), on the one hand, and model applicability (computational effort in terms of computer power and code length; complexity and amount of model output to be analyzed), on the other hand. This also means that multimedia box models and atmospheric transport models should be used in a complementary way with model comparison as one important exercise that helps improve both types of models. Because of the many uncertainties of emission data, substance properties and environmental parameters as well as the extreme complexity of the global system, more complex models are not necessarily more adequate than highly simplified multimedia box models. 6.3 Assessment Criteria

The Stockholm convention lists criteria for the identification of further POPs (Annexes D and E). These criteria are given in terms of single-media half-lives (6 months for soil, 2 months for water, 2 days for air), and KOW or bioconcentration factors (BCF). (Additional sources of evidence that a chemical might be a POP are also mentioned.) Up to now, it is not sufficiently clear how these criteria relate to more aggregated characteristics of POPs such as overall persistence, potential for long-range transport, or accumulation in polar regions, and how additional factors such as the temperature dependence of partitioning coefficients and degradation rate constants, the adsorption of POPs to aerosols and suspended particles, the time and place of release influence the overall POP character of a chemical. Therefore, the relationship between the POPs criteria listed in the Stockholm convention and the more aggregated POPs characteristics obtained from global models should be investigated. By doing so, it will be possible to understand whether the criteria of the convention capture the key properties of POPs and, on this basis, to confirm or refine the criteria.

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

267

Acknowledgements. We thank F. Wegmann, ETH Zürich, for helpful discussions and support with model calculations. Thanks go also to H. Fiedler, UNEP chemicals, and H. Held, Potsdam Institute of Climate Impact Research, for their helpful comments. M.S. gratefully acknowledges financial support by the Swiss Agency for the Environment, Forests and Landscape. F.W. is grateful to the Northern Contaminants Program of the Canadian Department of Indian Affairs and Northern Development for continued financial support of his research into modeling the global fate of POPs.

7 References 1. Vallack HW, Bakker DJ, Brandt I, Broström-Lundén E, Brouwer A, Bull KR, Gough C, Guardans R, Holoubek I, Jansson B, Koch R, Kuylenstierna J, Lecloux A, Mackay D, McCutcheon P, Mocarelli P, Taalman RDF (1998) Environ Toxicol Pharmacol 6:143 2. Golub MS, Donald JM, Reyes JA (1991) Environ Health Perspect 94:245 3. Stone R (1992) Science 225:798 4. Blus LJ (1995) Organochlorine Pesticides. In: Hoffman DJ, Rattner BA, Burton GA, Cairns J (eds), Handbook of Ecotoxicology. Lewis Publishers, Chelsea, p 275 5. Jacobson JL, Jacobson LW (1996) N Engl J Med 335:783 6. Oreskes N, Shrader-Frechette K, Belitz K (1994) Science 263:641 7. Wania F, Mackay D (1999) Environ. Pollut 100:223 8. Cowan CE, Mackay D, Feijtel TCJ, van de Meent D, di Guardo A, Davies J, Mackay N (ed) (1995) The Multimedia Fate Model: A Vital Tool for Predicting the Fate of Chemicals. SETAC Press, Pensacola 9. Pekar M, Pavlova N, Erdman L, Ilyin I, Strukov B, Gusev A, Dutchak S (1998) Longrange transport of selected Persistent Organic Pollutants. EMEP/MSC-E report 2/98, Moscow 10. Koziol AS, Pudykiewicz JA (2001) Chemosphere 45:1181 11. Sofiev MA, Galperin MV (2002) Numerical modeling of the atmospheric transport of toxic pollutants in the northern hemisphere. In: Sportisse B (ed) Air Pollution Modelling and Simulation. Proceedings of the Second Conference on Air Pollution Modelling and Simulation, APMS ’01. Springer, Berlin, p 101 12. Wania F (1999). Differences, similarities, and complementarity of various approaches to modelling persistent organic pollutant distribution in the environment. WMO/ EMEP/UNEP Workshop on Modelling of Atmospheric Transport and Deposition of Persistent Organic Pollutants and Heavy Metals (Volume I), Geneva, Switzerland, 16–19 November 1999. World Meteorological Organisation, Global Atmospheric Watch No. 136, 115–140 13. MacLeod M,Woodfine DG, Mackay D, McKone TE, Bennett DH, Maddalena R (2001) ESPR – Environ Sci & Pollut Res 8:156 14. Howard PH, Boethling RS, Jarvis WF, Meylan WM, Michalenko EM (eds) (1991) Handbook of environmental degradation rates. Lewis Publishers, Chelsea 15. Mackay D, Shiu WY, Ma KC (2000) Illustrated handbook of physical-chemical properties and environmental fate. (CD-ROM) Chapman & Hall/CRCnetBase, Boca Raton, FL 16. Shiu WY, Ma KC (2000) J Phys Chem Ref Data 29:41 17. Shiu WY, Ma KC (2000) J Phys Chem Ref Data 29:387 18. Sander R (1999) Compilation of Henry’s Law Constants for Inorganic and Organic Species of Potential Importance in Environmental Chemistry (Version 3) http://www.mpchmainz.mpg.de/~sander/res/henry.html 19. Beyer A,Wania F, Gouin T, Mackay D, Matthies M (2002) Environ Toxicol Chem 21:941–953 20. Pontolillo J, Eganhouse RP (2001) The Search for Reliable SW and KOW Data for Hydrophobic Organic Compounds: DDT and DDE as a Case Study. US Geological Survey, Reston, VA 21. Atkinson R (1991) Sci Tot Environ 104:17

268

M. Scheringer · F. Wania

22. Koester CJ, Hites RA (1992) Environ Sci Technol 24:502 23. Harrad SJ (1998) Dioxins, Dibenzofurans and PCBs in Atmospheric Aerosols. In: Harrison RH, van Grieken RE (eds), Atmospheric Particles. Wiley, New York 24. Fenner K, Scheringer M, Hungerbühler K (2000) Environ Sci Technol 34:3809 25. Wania F, Persson J, Di Guardo A, McLachlan MS (1999). The POPCYCLING-Baltic model. A non-steady state multicompartment mass balance model of the fate of persistent organic pollutants in the Baltic Sea environment. Norwegian Institute for Air Research (NILU), Technical Report and Computer Program 76 pages 26. Woodfine DG, MacLeod M, Mackay D, Brimacombe JR (2001) ESPR – Environ Sci & Pollut Res 8:164 27. Li YF, McMillan A, Scholtz MT (1996) Environ Sci Technol 30:3525 28. Li YF, Bidleman TF, Barrie LA, McConnell LL (1996) Geophys Res Lett 25:39 29. Fiedler H (2001) UWSF – Z Umweltchem Ökotox 13:88 30. Fiedler H, chapter in this volume 31. Breivik K, Sweetman A, Pacyna JM, Jones KC (2002) Sci Total Environ 290:199–224 32. Klöpffer W (1994) ESPR – Environ Sci & Pollut Res 1:108 33. Scheringer M (1996) Environ Sci Technol 30:1652 34. Webster E, Mackay D, Wania F (1998) Environ Toxicol Chem 17:2148 35. Bennett DH, Kastenberg WE, McKone TE (1999) Environ Sci Technol 33:503 36. Scheringer M (1997) Environ Sci Technol 31:2891 37. Bennett DH, Matthies M, McKone TE, Kastenberg WE (1998) Environ Sci Technol 32:4023 38. Beyer A, Mackay D, Matthies M, Wania F, Webster E (2000) Environ Sci Technol 34:699 39. Bennett DH, Scheringer M, McKone TE, Hungerbühler K (2001) Environ Sci Technol 35:1181 40. Scheringer M,Wegmann F, Fenner K, Hungerbühler K (2000) Environ Sci Technol 34:1842 41. Wania F, Lei YD, Wong A (2001) Quantifying and reducing uncertainty in model calculations of global pollutant fate. In: Kalhok, S. (ed), Synopsis of research conducted under the 2000–2001 Northern Contaminants Program. Minister of Indian Affairs and Northern Development, Ottawa, p 138 42. Strand A, Hov Ø (1996) Water Air Soil Pollut 86:283 43. Strand A, Hov Ø (1993) J Geophys Res 98:9023 44. Wania F, Mackay D (2000) The global distribution model. A non-steady state multicompartmental mass balance model of the fate of persistent organic pollutants in the global environment. Technical Report and Computer Program on CD-ROM, 21 pages (www.utsc.utoronto.ca/~wania) 45. Wania F, Mackay D (1993) Chemosphere 27:2079 46. Wania F, Mackay D (1995) Sci Tot Environ 160/161:211 47. Klein M (1999) Bestimmung von Gesamthalbwertszeiten von POPs (Persistente Organische Stoffe) in der Umwelt. Fraunhofer Institut für Umweltchemie und Ökotoxikologie, Schmallenberg 48. Held H (2002) Stoch Env Res Risk Assess 16: in press 49. Wegmann F (1999) Diploma thesis, Department of Chemistry, ETH Zürich 50. Lammel G, Feicher J, Leip A (2001) Long-Range Transport and Multimedia Partitioning of Semivolatile Organic Compounds: A Case Study with Two Modern Agrochemicals, MPI für Meteorologie, Report No. 324, Hamburg 51. Stiver W, Mackay D (1989) Chemosphere 19:1187 52. Scheringer M, Held H, Stroebe M (2002) Chemrange 2.1 http://ltcmail.ethz.ch/hungerb/research/product/product.html 53. Atkinson R (1987) Int J Chem Kinet 19:799 54. Lohmann R, Ockenden WA, Shears J, Jones KC (2001) Environ Sci Technol 35:4046 55. Okubo A (1971) Deep Sea Res 18:789 56. Keeling CD, Heimann M (1986) J Geophys Res 91 D7:7782 57. Hunter-Smith RJ, Balls PW, Liss PS (1983) Tellus 35B:170 58. Wania F, Mackay D, Li YF, Bidleman TF, Strand A (1999) Environ Toxicol Chem 18:1390 59. Wania F (1999) Global Modelling of Polychlorinated Biphenyls. WECC-Report 1/99, June 1999, 22 pages

Multimedia Models of Global Transport and Fate of Persistent Organic Pollutants

269

60. Wania F, Mackay D, McLachlan MS, Sweetman A, Jones KC (1999) Global modeling of polychlorinated biphenyls. In: Kalhok S (ed), Synopsis of research conducted under the 1998–1999 Northern Contaminants Program. Indian and Northern Affairs Canada, Ottawa, p 61 61. Wania F, Lei YD, Daly G, McLachlan MS (2000) Quantifying and reducing uncertainty in model calculations of global pollutant fate. In: Kalhok S (ed), Synopsis of research conducted under the 1999–2000 Northern Contaminants Program. Indian and Northern Affairs Canada, Ottawa, p 142 62. Wegmann F, Scheringer M, Hungerbühler K (2002) Modeling the global fate of DDT and other POPs. Project Report for the Swiss Agency for the Environment, Forests and Landscape, Swiss Federal Institute of Technology, Zürich 63. Wania F, Mackay D (1999) Environ Toxicol Chem 18:1400 64. Iwata H, Tanabe S, Sakai N, Tatsukawa R (1993) Environ Sci Technol 27:1080 65. Wania F, Mackay D (1996) Environ Sci Technol 30:390 A 66. MacDonald RW, McLaughlin FA, Adamson L (1997) Canadian Chemical News 1997:28 67. Lakaschus S,Weber F,Wania F, Bruhn R, Schrems O (2001) Environ Sci Technol 35: in press. 68. MacDonald RW, Barrie LA, Bidleman TF, Diamond ML, Gregor DJ, Semkin RG, Strachan WMJ, Li YF,Wania F,Alaee M,Alexeeva LB, Backus SM, Bailey R, Bewers JM, Gobeil C, Halsall CJ, Harner T, Hoff JT, Jantunen LMM, Lockhart WL, Mackay D, Muir DCG, Pudykiewicz J, Reimer KJ, Smith JN, Stern GA, Schroeder WH, Wagemann R, Yunker MB (2000) Sci Total Environ 254:93 69. Wania F, Mackay D (1993) Ambio 22:10 70. Muir DCG, Omelchenko A, Grift NP, Savoie DA, Lockhart WL, Wilkinson P, Brunskill GJ (1996) Environ Sci Technol 30:3609 71. Ockenden WA, Sweetman AJ, Prest HF, Steinnes E, Jones KC (1998) Environ Sci Technol 32:2795 72. Pudykiewich J, Dastoor AP (1996) Study of the global scale transport of sulphur and persistent organic pollutants with special emphasis on Arctic regions. In: Murray JL, Shearer RG, Han SL (eds), Synopsis of Research conducted under the 1994/5 Northern Contaminants Program. Indian and Northern Affairs Canada, Ottawa, p 7 73. van Jaarsveld JA, van Pul WAJ, de Leeuw FAAM (1997) Atmos Environ 31:1011 74. Gouin T, Mackay D, Webster E, Wania F (2000) Environ Sci Technol 34:881 75. Troll C (1966) Seasonal Climates of the Earth. The seasonal course of phenomena in the different climatic zones of the Earth. In: Rodenwaldt E and Jusatz H (ed), World Maps of Climatology by HE Landsberg, H Lippmann, KH Pfaffen, C Troll. Springer, Berlin 76. Finizio A, Mackay D, Bidleman TF, Harner T (1997) Atmos Environ 31:2289 77. Siegenthaler U, Joos F (1992) Tellus 44B:186 78. Seth R, Mackay D, Muncke J (1999) Environ Sci Technol 33:2390 79. Falkowski PG, Barber RT, Smetacek V (1998) Science 281:200 80. Legates DR, Wilmott C (1990) Theoret Appl Climatol 41:11 81. Scheringer M (2002) Persistence and Spatial Range of Environmental Chemicals.Wiley-VCH, Weinheim

CHAPTER 10

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and Other POPs Olaf Päpke 1 · Peter Fürst 2 1 2

Ergo Forschungsgesellschaft mbH, Geierstrasse 1, 22305 Hamburg, Germany E-mail: [email protected] Chemisches Landes- und Staatliches Veterinäruntersuchungsamt, Sperlichstrasse 19, 48151 Münster, Germany E-mail: [email protected]

Since the first finding of the insecticide DDT in human tissues, the problem of persistent organic pollutants (POPs) has become a growing issue of special public and toxicological concern. In the past few decades numerous reports have assessed the global distribution of contaminants, such as organochlorine pesticides (OCPs), polychlorinated biphenyls (PCBs), polychlorinated dibenzo-p-dioxins (PCDDs), dibenzofurans (PCDFs) and polybrominated flame retardants. Consumption of food has been shown to be the major pathway of human background exposure to POPs. Because humans are at the top of the food chain, it is obvious that human tissues may contain relatively high concentrations of those lipophilic residues and contaminants that tend to bioaccumulate in the food web. Human body burden can best be determined by analysis of human adipose tissue, blood or breast milk. For a meaningful risk assessment, it is important to precisely characterise the specimens, because it was demonstrated that factors such as age, consumption habits, changes in body weight or nursing status might have a severe influence on human body burden. Analyses of human tissue samples from industrialized countries for organochlorine pesticides show a significant decrease within the past 20 years. This demonstrates that the ban of these pesticides in the Western World in the early 1970s had a beneficial effect on human body burden. On the other hand, human samples from countries with an ongoing application of persistent insecticides for vector control, such as DDT, may still contain elevated pesticide levels. A decrease can also be seen for polychlorinated biphenyls in human samples from those countries which early banned the use of technical PCB mixtures in open systems and strictly regulated their use in closed systems as well as their disposal. In general, human samples from industrialized countries show higher PCDD/PCDF levels than corresponding specimens from developing regions.An exception represents, however, certain Inuit cohorts living in Greenland and Northern Quebec who show the highest “background” contaminant values world-wide. Because of numerous measures which were taken to reduce PCDD/PCDF emissions into the environment, a significant decline of these compounds in humans living in industrialised countries was observed between the late 1980s and the middle of the 1990s. Since then, however, this declining trend seems to have come to a stop. Results on dioxin-like PCBs in former and present human specimens are still scarce. Those data available indicate, however, that these compounds may contribute significantly to human body burden, expressed as total TEQ values. In contrast to most other persistent organic pollutants, polybrominated diphenyl ethers (PBDE) widely used as flame retardants do not show a comparable decreasing trend in human samples. Recent analyses rather demonstrate an ongoing human exposure to these contaminants. Keywords: POPs, Dioxins, PCDDs, PCDFs, PCBs, PBDEs, Organochlorine pesticides, Human ex-

posure

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

272

O. Päpke · P. Fürst

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272

1

Introduction

2

Organochlorine Pesticides

3

Dioxins and PCBs

3.1 3.2 3.3 3.4 3.5

Correlation Between Different Matrices . . . . . . . Dioxin and PCB Pattern in Human Tissues . . . . . . Contribution of PCDDs/PCDFs and Dioxin-Like PCBs to Total Body Burden . . . . . . . . . . . . . . . . . . Parameters Influencing Human Body Burden . . . . Dioxins and PCBs in Humans in the Course of Time .

4

Polybrominated Diphenyl Ethers . . . . . . . . . . . . . . . . . . 290

5

References

. . . . . . . . . . . . . . . . . . . . . 273

. . . . . . . . . . . . . . . . . . . . . . . . . . 275 . . . . . . . 278 . . . . . . . 279 . . . . . . . 279 . . . . . . . 280 . . . . . . . 288

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292

1 Introduction The widespread application of organochlorine pesticides in agriculture started with the discovery of the insecticide properties of dichlorodiphenyltrichloroethane (DDT) in 1939 by the Swiss chemist Paul Müller. Being inexpensive and easy to produce, this broad spectrum pesticide quickly spread over the globe. Besides increased yields of treated crops, DDT was also effective to bring diseases such as malaria under control. Due to its properties DDT was also used for a number of non-agricultural applications. In 1944 Müller was awarded the Nobel Prize for Medicine for his discovery. At the same time, several other pesticides with similar properties were produced. Compounds such as chlordane, aldrin, toxaphene, mirex, dieldrine, hexachlorocyclohexane (HCH), heptachlor and others then became major agents for pest control. The enthusiasm about these compounds came to a certain stop after the publication of Rachel Carson’s best selling book “Silent Spring” in which she warned that pesticides poison the food supply of animals and kill large numbers of birds and fish. Her demand to stop the “rain of chemicals” and her warnings of a massive destruction of the fragile planet’s ecosystem finally lead to a ban of DDT and other persistent pesticides in almost all countries of the Western World in the early 1970s. It took another 30 years before more than 100 countries signed the Stockholm Convention in May 2001. This treaty sets out control measures covering the production, import, export, disposal and use of DDT and 11 other persistent organic pollutants (POPs) with the intention to completely phase out these chemicals. However, a health-related exemption of an immediate ban was granted to DDT which is still needed in many developing countries to control malaria transmitting mosquitoes. Polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) are often termed simplified as “dioxins”. Depending on the degree of chlorination (1–8 chlorine molecules) and the substitution pattern, one can distinguish between 75 PCDDs and 135 PCDFs, called “congeners”. Although

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and other POPs

273

dioxins do not have any benefits and therefore, with the exception of research and analytical purposes, were not produced specifically, these contaminants have meanwhile found a ubiquitous distribution due to their formation as unwanted and often unavoidable by-products in a number of industrial and thermal processes. Polychlorinated biphenyls (PCBs) belong to the group of chlorinated hydrocarbons which are synthesized by direct chlorination of biphenyl. Depending on the number of chlorine substituents (1–10) and their position at the two rings, there are 209 different compounds, also called “congeners”, theoretically possible. Due to their physical and chemical properties, such as non-flammability, chemical stability, high boiling point, low heat conductivity and high dielectric constants, technical PCB mixtures were widely used in a number of industrial and commercial open and closed applications. It is estimated that more than 1 million tons of technical PCB mixtures were produced and marketed worldwide since their first commercial use in the late 1920s. From a toxicological point of view, 12 non-ortho- and mono-ortho-PCB congeners are of special interest because these show toxicological properties that are similar to dioxins. Therefore, these congeners are often termed “dioxin-like PCBs”. Polybrominated diphenyl ethers (PBDEs) are extensively used as flame retardants and added to a wide range of products such as electronics, plastics and textiles. PBDEs have been found to accumulate in lipids.While the benefits of flame retardants are obvious, their impact on the environment and human health cannot be ignored. PBDEs which share some similarity to POPs have recently been shown to occur in both environmental and biotic media in significant amounts as reported by De Boer et al. [1]. Humans are contaminated with POPs through environmental (background), occupational or accidental exposure. This overview focuses only on environmental or background exposure.Assessments of human body burden can be performed by analysing adipose tissue, human milk or blood. Blood is easy to obtain from humans, unlike breast milk, it is not limited to a specific gender and therefore represents the preferred tissue to be investigated.

2 Organochlorine Pesticides Figure 1 shows the mean values for b-HCH, hexachlorobenzene (HCB) and p,p′DDE in 1,839 individual human milk samples collected in North Rhine-Westphalia/Germany and analysed between 1984 and 2000. DDE represents the persistent metabolite of DDT, formerly widely applied as insecticide around the world. The results for DDE include also the parent compound. While samples from the 1980s contained measurable levels of p,p′-DDT, this pesticide can nowadays only be detected occasionally at levels near the detection limit. As can be clearly seen, the levels for the above compounds which represent the predominant OCP residues in human milk have decreased by 80–90% within the past 16 years. The results demonstrate that the ban of these pesticides in the Western World in the early 1970s have had a beneficial effect on the body burden of humans.

274

O. Päpke · P. Fürst

Fig. 1. Time trend for organochlorine pesticides in human milk from North Rhine Westphalia/Germany, Fürst [2]

Table 1. Organochlorine pesticides in human samples from developing countries

Country

Compound

mg/kg fat

Sample type

Reference

Mexico

SDDT b-HCH SDDT SDDT

5.9 0.16 7.8 15.0

Adipose tissue Adipose tissue Human milk Human milk

Waliszewski et al. [3]

SDDT SDDT b-HCH DDE SDDT SDDT SDDT SDDT SDDT b-HCH HCB SDDT b-HCH HCB

0.83 1.7 0.27 2.5 6.5 (mean) 25.3 (Kariba) 7.0 3.2 6.3 0.23 0.31 2.4 0.40 0.29

Human milk Human milk Human milk Human milk Human milk Human milk Human milk Human milk Adipose tissue Adipose tissue Adipose tissue Human milk Human milk Human milk

Saeed et al. [6] Paumgartten et al. [7]

Northern Thailand Kuwait Brazil Ukraine Zimbabwe Kenya Uganda Poland Jordan

Pardio et al. [4] Stuetz et al. [5]

Gladen et al. [8] Chikuni et al. [9] Chikuni et al. [9] Ejobi et al. [10] Ludwicki et al. [11] Alawi et al. [12]

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and other POPs

275

A comparison of the residue situation in different countries is somewhat difficult due to differences in the sampling strategy (occupationally, accidentally or environmentally exposed mothers), low numbers of specimens analysed and sometimes poor analytical sensitivity. Table 1 shows some selected recent results of pesticide analyses in human milk from developing countries. Although not representative, the results seem to be typical in the respective regions of the world. In contrast to samples from the Western World, the samples from developing countries often show elevated levels, especially for total DDT (DDT plus its metabolites DDE and DDD=SDDT) and b-HCH, which is an indication of a longer application of these pesticides in the respective countries.

3 Dioxins and PCBs Dioxins have been found in environmental samples of early or even of ancient origin by Kjeller et al. [13] and Hartmann et al. [14], respectively. For ancient human samples, first dioxin findings have been reported by Schecter et al. [15], and by Tong et al. [16]. In all these human samples the concentrations were found to be quite low, mostly in the range of blank samples analysed in parallel. The beginning of the chlorine industry induced a strong increase of these substances in the environment as presented by Hagenmaier et al. [17] in sediment cores from Lake Constance. The beginning of the chlorine industry in the 1930s is indicated by an increase of the dioxin concentrations in the sediment samples. The peak of the contamination in the core samples was reached between 1960 and 1970. Actual levels found nowadays are about 80% lower compared to the 1970s values.

Fig. 2. TEQs for PCDDs, PCDFs and non-ortho(co)-PCBs in humans from European countries

276

O. Päpke · P. Fürst

212

152

Fig. 3. TEQs for PCDDs, PCDFs and non-ortho(co)-PCBs in Humans, world-wide

In contrast to environmental samples, only a limited number of PCDDs/PCDFs congeners are found in humans. First analyses of human tissue were performed by Baughman and Meselson [18], by Miyata et al. [19], and by Langhorst and Shadoff [20]. Rappe et al. [21] reported on the finding of only 2,3,7,8-chlorine substituted PCDDs/PCDFs in human milk samples originating from Europe. These results indicated for the first time the existence of a dioxin background contamination for the general population. As reported by various authors, such as Beck et al. [22], Fürst et al. [23], [24] and Grün et al. [25], a daily intake of 0.5–3 pg I-TEQ/kg body weight via food results in PCDD/PCDF-environmental or background contamination of humans. In the first years of dioxin analysis of human samples, adipose tissue was used almost exclusively for exposure estimation. Because of the low lipid content of blood (about 0.5%) it was not possible to analyse blood from the background contaminated general population with adequate detection limits 25 years ago.

Table 2. I-TEQ-values for PCDDs/PCDFs and PCBs in adipose tissue samples from Greenland

Inuit, Ryan et al. [26]. and plasma samples from Canadian Inuit from Nunavic and Southern Quebec, Ayotte et al. [27], all values in pg/g lipid Components

Greenland (mean, n=13)

Nunavik (mean, n=20)

S. Quebec (mean, n=3)

PCDDs/PCDFs non-ortho-PCBs mono- and di-ortho-PCBs

151.9 212.2 386.0

39.6 26.3 118.2

14.6 5.2 6.3

Total TEQ

750.2

184.2

26.1

277

Fig. 4. TEQs for PCDDs, PCDFs and coplanar PCBs in humans, ranking world-wide

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and other POPs

278

O. Päpke · P. Fürst

However, in the past two decades the increasing sensitivity of mass spectrometers opened the possibility of analysing serum or whole blood. In the following Figs. 2–4 relatively new measurements for dioxins and coplanar PCBs are given for European and others countries. The shadowed bars represent the TEQs for the non-ortho-PCBs while the nonshadowed are representative for PCDD/PCDF-TEQ (Fig. 2). The map indicates relatively similar values for most of the industrialised countries, on the other hand, less developed countries like Albania and Croatia show the lowest values for dioxins and PCBs. The view on the bars in the world map gives some surprises. It has to be realised that the scale in Fig. 3 – compared to the map of Europe – is different for these bars. The highest values found for the general normal population worldwide are reported from Ryan et al. [26] for Inuit from Greenland (Table 2). The values found are about 20 times higher than values found in other parts of the word. The main reason for these high values seems to be the intake of polluted fish and seal. Somewhat lower values were found by Ayotte et al. [27] for Inuits from Northern Quebec, Canada. Figure 4 shows a comprehensive overview on the body burden with PCDDs/ PCDFs and coplanar PCBs determined in various countries. Coming from the high end, high concentrations were found in the background contaminated population in countries like Greenland (Inuit), Canada (Inuit) and Japan. Countries with mean concentrations in humans for dioxins and PCBs are, e.g., Germany, United Kingdom and Russia. The lowest concentration for these components are observed in lower industrialised countries like Albania, Pakistan and Northern Vietnam. 3.1 Correlation Between Different Matrices

The correlation between serum and adipose tissue for 2,3,7,8-TCDD was demonstrated at first by Patterson et al. [28] who analysed paired serum and adipose tissue samples from 50 persons. The high correlation found between TCDD levels in adipose tissue and serum, indicated that the determination of TCDD in serum is a valid measurement of TCDD body burden, if the levels are calculated on lipid basis. The correlation between whole blood and adipose tissue for other PCDDs/ PCDFs were reported by Schecter et al. [29]. In both matrices, the levels for I-TEQ values and most of the congeners were quite similar whereas the hepta- and octacongeners revealed some differences. To demonstrate the correlation of PCDDs/PCDFs between human blood and human milk, the results from breast milk analyses performed by Fürst et al. [30] are compared to blood levels of an age-matched German background group (dominated by male individuals) from 1994 of Päpke et al. [31] (Fig. 5). The I-TEQ values for both matrices are quite similar at 17.2 pg/g for milk fat and 15.7 pg/g for blood fat, respectively. In contrast, the hepta-CDD/CDF and octaCDD/CDF are significantly higher in blood.

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and other POPs

Human Milk, n=50 Fürst et al. [30]

279

Whole Blood, n=37 Päpke et al. [37]

Fig. 5. Comparison of PCDDs/PCDFs in human blood and milk, from Päpke [31]

3.2 Dioxin and PCB Pattern In Human Tissues

Human milk from The Netherlands was analysed for various compounds by Liem and Theelen [32]. In Fig. 6, the pattern of human milk is given for dioxins and PCBs. On the left side the concentration for dioxins and the coplanar PCBs #77, 126 and 169 are shown. OCDD and PCB 126 and 169 are found to have the highest concentrations. On the right side the concentrations for other non-ortho- and di-ortho-PCBs are given. PCB #138, 153 and 180 were found to be the dominating congeners. It has to be considered that the concentrations on the left graph are given in pg/g lipid while the others on the right graph are given in ng/g lipid. 3.3 Contribution of PCDDs/PCDFs and Dioxin-Like PCBs to Total Body Burden

Liem and Theelen reported for the contribution of PCDDs/PCDFs, non-ortho (coplanar) PCBs (77, 126, 169) and other dioxin-like PCBs (105, 118, 156, 157, 167, 180, and 189) to the total level of 2,3,7,8-TCDD equivalents (TEQs) 53%, 20% and

280

O. Päpke · P. Fürst

Fig. 6. Pattern of PCDDs/PCDFs and PCBs in human milk, from Liem and Theelen [32]

27%, respectively.When looking at the contribution of the different congeners to the total TEQ, 2,3,4,7,8-penta-CDF, PCB 126 and PCB 156 are of special importance for human milk. These three congeners amount to more than 50% to the total TEQ value (Fig. 7). 3.4 Parameters Influencing Human Body Burden

Human body burden with PCDDs/PCDFs and PCBs can be influenced by a number of factors, such as: – – – – –

consumption habits, body weight changes, age, nursing status, time of sample collection and analysis.

In order to study the influence of diet on the body burden of PCDDs/PCDFs, blood from three different Swedish groups was analysed by Svensson et al. [33]. Group 1 had no fish consumption, group 2 had normal (around 50 g/day) and group 3 had high fish consumption (>100 g/day). Group 3 revealed a body burden, calculated as TEQ, about three times higher than group 2. Group 1, however, had only slightly lower blood levels of PCDD/PCDF than group 2. The dominating congener among the tetra- and penta-chlorinated congeners was 2,3,4,7,8-

Fig. 7. Contribution of groups and single congeners (in %) to total TEQs in human milk (n=78), from Liem and Theelen [32]

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and other POPs

281

282

O. Päpke · P. Fürst

PeCDF. The difference between the three groups was also highest for this particular congener, which is also the major congener in fish from the Baltic Sea. The mean PCDD/PCDF values for groups 1, 2 and 3 were 17.5, 25.8 and 63.5 pg ITEQ/g, respectively. Different dietary habits may be expected for vegetarians and non-vegetarians. Welge et al. [34] studied this kind of different consumption habits. They analysed blood of 24 vegetarians and 24 non-vegetarians. No significant difference between both groups could be observed in the mean levels of the measured PCDDs/PCDFs. While most vegetarians normally consume milk and milk products and eggs, in contrast, vegans follow a strict vegetarian diet resulting in absolutely no consumption of food of animal origin. The influence of vegan diet on human dioxin body burden was studied by analysing blood samples originating from two strict vegans from Oregon/USA for PCDDs/PCDFs reported by Päpke et al. [35]. In Fig. 8 the I-TEQs of the vegans and the control groups are compared. The values of American (pools) and German (individual analyses) non-vegetarians were found to be quite similar while the values for the vegans are considerably lower. The I-TEQ

Fig. 8. Comparison of total I-TEQ values of persons with vegan and non-vegetarian diet (common food consumption), from Päpke et al. [34]

283

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and other POPs

value of 1.6 pg/g lipid for vegan 1 was the lowest dioxin level measured in blood of adult persons at the time of measurement in the dioxin laboratory. This value can be compared to the minimum I-TEQ value for the age-matched group (mean age=49 years) of the German background population found at 9.6 pg/g lipids. A pronounced influence on the dioxin concentration of humans may result from a body weight reduction. This effect has been reported firstly by Zober and Päpke [36] who analysed blood samples from cancer patients a few months before they died. A second analysis – post-mortem collected – resulted in much higher values. In connection with a weight reduction of 20–25 kg much higher values were found in the second sample compared to the first analysis. Because of the reduction of body lipids a concentrating of dioxins resulted in the remaining lipids. The so-called concentrating factors ranged between 4 and 25 in this case. The influence of age on the PCDD/PCDF body burden of humans has been reported by some different authors. In Table 3 a summary of the I-TEQ increase in pg/year of age observed by different authors and analysed in different matrices is given. The values range between 0.4 and 0.8 pg I-TEQ per year of age. The original situation for the age-related dependency for all congeners is demonstrated when looking at the data of dioxin background investigation for 1996 by Päpke et al. [31]. For comparison with unknown samples, three age groups were formed which serve as individual age-dependent reference groups. Depending on the half-life, most of the congeners show an increase with age. The mean I-TEQ for these three groups range at values between 13.1; 16.3 and 19.1 pg/g lipids while the mean age of the groups is 26, 37 and 49 years. In 1995 Päpke et al. [40] firstly reported the age-dependency for the coplanar PCBs. The annual increase in concentration for the PCBs #126 and 169 is quite similar at values of 2.9 and 3.0 pg/g (lipid based) per year of age. An important factor of influence on dioxin levels in human milk is the total length of the nursing periods and the number of breast-fed children.As reported firstly by Fürst et al.[41], the dioxin levels decrease with increasing number of

Table 3. Age dependency of TEQ a/I-TEQ b in humans

Author

Matrix

Beck et al. [22] Beck et al. [22] Beck et al. [22]] Sagunski et al. [37] Schrey et al. [38] Päpke et al. [39] Päpke et al. [40]

Mother’s milk Fat Mother’s milk/fat Blood/fat Blood Blood Blood

a b

n 34 20 59 67 95 101 134

Increase in pg per year of age

Observed age range (years)

0.44 TEQ 0.39 TEQ 0.50 TEQ 0.6 I-TEQ 0.8 I-TEQ 0.5 I-TEQ 0.4 I-TEQ

20–40 20–70 0–70 0–65 12–82 6–60 22–69

TEQ=Toxicity Equivalents (Germany). I-TEQ=International Toxicity Equivalents (NATO CCMS).

284

O. Päpke · P. Fürst

breast-fed children and the length of the nursing period. The mean decrease per nursing week results at about 0.1 pg I-TEQ/g lipids. The influence of breast feeding on the I-TEQ-body burden of infants has been investigated by different groups: Beck et al. [42], Abraham et al. [43] and Fürst et al. [44]. In Fig. 9 the PCDD/PCDF levels in blood from nursed and non-nursed 11month-old infants are compared. As reported by Abraham, almost all congeners including the I-TEQ are found at about 10 to 15 times higher in the nursed infant. In order to examine whether contaminants probably influence the risk for the sudden infant death syndrome (SIDS) various tissues (subcutaneous fat, liver, kidney and spleen) from 27 infants who died suddenly and unexpectedly were analysed for dioxins, Fürst et al. [44]. Table 4 shows the PCDD/PCDF levels determined in various tissues besides some basic data, such as year of sampling, length of breast feeding as well as sex and age of the victim. While in 1991/1992 kidney was analysed in addition to adipose tissue and liver, the 1996/1997 survey comprised the matrices adipose tissue, liver and spleen. All PCDD/PCDF levels are given as ng I-TEQ/kg fat. As can be seen from Table 4, tissues from infants who were partly or exclusively breast fed contain higher PCDD/PCDF levels than tissues from non-breast fed victims. The duration of breast feeding was directly and the birth order of the victim was indirectly proportional to the PCDD/PCDF body burden. In single cases of the 1991/1992 survey, the dioxin concentration in adipose tissue distinctly exceeded the mean PCDD/PCDF level measured at that time in hu-

149

Fig. 9. Influence of nursing and not nursing on the I-TEQ values in blood of infants, from Abra-

ham et al. [43]

285

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and other POPs Table 4. I-TEQ levels (ng/kg fat) in different tissues from SIDS cases, Fürst et al. [44]

I-TEQ Levels in Different Tissues from SIDS-Cases No.

Year

Breast fed (weeks)

age (weeks)

sex

exclusively partly 349 356 66 55 17 378 354 297 379 374 311 288 366 12 389 381 26 388 58 133 127 80 211 416 91 311 345

96 96 97 97 97 96 96 96 96 96 96 96 92 92 91 91 92 91 92 92 92 92 92 91 91 91 91

0 0 0 1 4 6 13 0 2 16 3 11 0 0 0 0 0 0 0 0 0 25 30 17 24 12 5

0 0 5 1 2 0 0 8 0 8 0 0 0 0 0 0 0 14 4 10 5 0 0 0 0 0 0

13 31 27 10 10 33 13 37 14 42 3 11 59 10 14 23 17 17 38 32 18 31 35 18 28 12 5

m f m f m f m f m m f m m m m m m m m f m m f m m m m

ng I-TEQ/kg fat adipose

liver

kidney

spleen

1.8 0.6 1.7 2.0 1.9 2.8 6.7 0.7 4.2 14.9 9.2 23.8 1.2 15.6 2.5 3.8 2.4 17.2 4.0 27.6 3.8 37.3 28.6 43.9 87.4 29.1 29.0

2.5 1.8 4.0 2.1 1.8 3.8 6.9 1.7 8.1 24.4 9.8 25.8 6.1 15.6 6.2 11.1 8.8 31.2 18.2 71.3 17.5 215.8 114.3 n.a. 192.0 n.a. 26.3

2.1 1.2 2.9 1.6 3.1 1.9 7.5 1.4 4.3 14.8 10.5 29.8 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 1.9 17.2 3.5 3.6 3.3 27.9 6.4 55.4 5.2 84.2 40.4 39.0 n.a. 29.1 n.a.

m=male; f=female; n.a.=not analysed.

man milk samples from nursing mothers in the same region. Compared to 1991/1992, the body burden of partly and exclusively breast-fed infants analysed in 1996/1997 is remarkably lower. This can be explained by the fact that the PCDD/PCDF levels in human milk decreased during the same period by approximately 50%, resulting in a significantly lower daily intake of these contaminants via breast feeding. It is striking that in almost all cases the PCDD/PCDF levels, based on fat weight, in liver, kidney and spleen are higher than those in subcutaneous fatty tissue. This is especially true for OCDD, 1,2,3,4,6,7,8-HpCDD, OCDF, 1,2,3,4,6,7,8HpCDF, 1,2,3,4,7,8-HxCDF, 1,2,3,6,7,8-HxCDF and 2,3,4,6,7,8-HxCDF which are significantly accumulated in liver fat (Fig. 10). Similar observations were also made by Beck et al. [42, 45] and Wuthe et al. [46]. The data summarised in Fig. 10 even seem to indicate that the accumulation of specific congeners in liver tissue

286

O. Päpke · P. Fürst

Fig. 10. Lipid-based PCDD/F ratios of liver/adipose tissue levels in breast-fed and non-breast-

fed SIDS-cases, from Fürst et al. [44]

from non-breast fed infants is somewhat higher than that from partly or exclusively breast fed victims, although the absolute levels in the former cases are generally much lower. Whether these observations have a pathophysiological significance cannot be decided at present. In any case, it has to be mentioned that according to the results of some epidemiological studies, e.g., the Westphalian crib death study where breast feeding could be characterised to be a protective factor for SIDS, it can be concluded that PCDD/PCDF do not seem to influence the risk for sudden infant death, Jorch et al. [47]. Because of the importance of breast feeding for infants, contamination of human milk is of specific public concern. In 1984, the first measurements for background contaminated milk were reported for Sweden and Germany. The World Health Organisation (WHO) induced world-wide large measurement campaigns in 1988 and 1993 for dioxins and PCBs. In Table 5 the results for PCDDs, PCDFs and PCBs are given for both collection periods. As can be seen from the table, the most important finding was a decline of these components in almost all countries. In Germany more than 2000 individual human milk samples were analysed by different groups until now. Resulting from this, in Table 6 the intake situation of nursed infants is shown by results from Fürst [2]. This table demonstrates that fully breast-fed infants have an average daily dioxin intake of about 66 pg I-TEQ/kg body weight and day. Consequently, the average daily PCDD/PCDF-intake for a breast-fed baby is approximately 50 times higher than the average daily PCDD/PCDF-intake for an adult. For most of the PCDDs/PCDFs and PCBs, long half-lives have been observed in humans. The half-life of 2,3,7,8-TCDD has been studied most comprehensively.

287

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and other POPs

Table 5. Comparison of I-TEQs values in human milk from different counties, collected in 1988 and 1993, values in pg TEQ/g milk fat, Liem et al. [48]

Origin

Range

Tirana Librazhd Austria Wien (urban) Tulln (rural) Belgium Brabant Liege Brüssel Canada All Provinces 1981 All Provinces 1992 Maritimes Quebec Ontario Prärien British Columbia Croatia Krk Zagreb Czechoslovakia Kladno Uherske Hradiste Denmark Various areas Finland Helsinki Kuopio Germany Berlin Nordrhein-Westfalen Hungary Budapest Scentes Netherlands Rural areas Urban areas All areas Norway Tromsø (Cost) Hamar (rural) Skien/Porsgrunn (Industry) Lithuania Palanga (Cost) Anykshchiai (rural) Vilnius (urban) Pakistan Lahore

TEQ, Nordic

I-TEQ

1987/88

1992/93

n

Albania

17.1 18.6 33.7 40.2 38.8

54 51

15.6 18.1 17.6 19.4 23.0 12.0 11.8

19 34 76 31 23 14 41

17.8 18.0 15.5 32.0 31.6 9.1 11.3 37.4 39.6 34.2 18.9 15.0 19.4

42 38 31 40 79 100 50 13 13 10 11 10 10

n

4.8 3.8 10.7 10.9 20.8 27.1 26.6 28.6 14.5 10.8 13.4 18.1 14.6 15.7 8.4 13.5 12.1 18.4 15.2 21.5 12.0 16.5 20.7 8.5 7.8

10 10 13 21 8 20 6 200 100 20 20 20 20 20 10 13 11 11 48 10 24 10 –34 20 10

22.4 10.1 9.3 12.5

17 10 10 10

16.6 14.4 13.3 3.9

12 12 12 14

Difference in %

–37 –41 –38 –33 –31

–31 –25 +3 –25 –32 –30 +14

–15 +19 –23 –48 –7 –31

–35 –47 –38 –36

For this congener a half-live at about 7 years was found. Geusau et al. [49] recently reported on a case of poisoning of two women with 2,3,7,8-TCDD. The first measurements – probably 5 to 6 months after the poisoning, revealed values of 144,000 and 25,000 pg 2,3,7,8-TCDD/g blood lipid, the highest values ever measured in adults. The half-live studies performed on the blood of both persons resulted in the first year after the exposure at a value of less than 1 year.

288

O. Päpke · P. Fürst

Table 6. I-TEQ values in human milk, North Rhine-Westphalia/Germany 1999, and daily intake by fully breast-fed infants, a Fürst [2]

Human Milk, n=23

Daily Intake

pg/g lipid-based

Total pg

pg/kg BW/day

Med.

Min.

Max.

Med.

Min.

Max.

Med.

Min.

Max.

13.8

6.4

29.3

331

154

703

66

31

141

a

Basis: body weight (BW): 5 kg, milk amount: 800 ml, lipid content: 3%.

Table 7. Elimination half-lives of 2,3,7,8-TCDD and related compounds in adipose tissues or blood from humans. Half-lives are expressed in years. (Liem and Theelen, [32])

2,3,7,8-TCDD 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD OCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,7,8,9-HpCDF OCDF 3,3′,4,4′-TeCB (PCB #77) 3,3′,4,4′,5-PeCB (PCB #126) 3,3′,4,4′,5,5′-HxCB (PCB #169)

6.2 8.6 19 >70 8.5 6.6 5.6 0.4 0.9 9.9 5.7 6.2 2.4 2.6 <0.2 0.1 2.7 13

Elimination half-lives for all PCDD/F-congeners and for PCBs are given in Table 7. As can be seen, the elimination half-lives for the different compounds show quite a wide range, most of them in the range of several years. On the other hand, some congeners, such as 2,3,7,8-TCDF, OCDF and PCB No 77 have much shorter half-lives at only some weeks. 3.5 Dioxins and PCBs in Humans in the Course of Time

First measurements for PCDD/PCDF-background contamination in adipose tissue were performed in Germany by Beck et al. [50]. The mean value from 20 individual analyses was found at 59 pg I-TEQ/g lipid. First background data

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and other POPs

289

in blood for Germany originated from 10 individuals with no known exposure except food were found at a mean of 46 pg I-TEQ/g lipid by Päpke et al. [39]. A declining trend for PCDD/PCDFs in humans in the course of time was first observed for human milk in Germany by Fürst et al. [41].These data are part of Fig. 11. The figure shows a comparison of mean values found in human milk and human blood regularly analysed since 1987. Between the end of the 1980s and 1999, the decline ranged between 50 and 70% on an I-TEQ basis. Similar time trends have been observed in The Netherlands, Denmark, the United States and the United Kingdom. These results seem to indicate that efforts to reduce emissions from known sources have had notable and beneficial effects on human body burden. Despite declining PCDD/F trends, the exposure of babies during the breastfeeding period is still a matter of concern and justifies further measures to reduce PCDD/F emissions into the environment. A declining trend can also be observed for the dominating PCBs (indicator PCBs) in humans, as was demonstrated by Fürst [2]. Figure12 shows the mean values for the PCB congeners #138, 153 and 180 in more than 1800 individual milk samples from Germany collected and analysed between 1984 and 2000. While the levels for these three compounds were almost equal in the 1980s, the results from samples analysed in the 1990s also show a declining trend. Compared to 1984, the levels in samples from 2000 are approximately 60–70% lower. Obviously, this decrease is the result of the ban of PCB usage in open systems and the strict regulations on the use of PCBs in closed systems as well as on their disposal.

Fig. 11. Time trend for PCDDs/PCDFs in humans from Germany, measured in adipose tissue, human milk and blood

290

O. Päpke · P. Fürst

Fig. 12. Time trend for indicator PCBs in human milk from North Rhine-Westphalia/Germany, from Fürst [2]

Similar declining trends were also reported for other industrialized countries which early took measures to strictly regulate PCBs, SCOOP [51].

4 Polybrominated Diphenyl Ethers The extensive use of products containing flame retardants like polybrominated diphenyl ethers (PBDEs) has resulted in the release of these components into the environment. Due to their lipophilic and persistent character PBDEs accumulate in the human body. For Swedish human milk decreasing levels of organochlorine compounds have been found by Norén and Mayronyté [52, 53]. In contrast, levels for PBDEs have increased continuously since 1972. Only a slight increase of PBDE levels was observed between 1985 and 1999 in Germany when Schröter-Kermani et al. [54] reported results for blood collected in the German environmental specimen bank. Recently, Mayronyté and Norén [55] reported after the peak level found for Swedish human milk in 1997, a decline for the years 1998 to 2000 as presented in Fig. 13. Meanwhile human data are available for quite a few countries. Typical values found for human samples like breast milk, blood and adipose tissue collected in various regions are shown in Table 8. Due to the unexpectedly high PBDE concentration found in a pooled US milk sample, the material was analyzed in three laboratories with long experience in

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and other POPs

291

Fig. 13. Mean concentration of PBDEs (ng/g lipid) in Swedish human milk from 1972 to 2000, from Mayronyté and Norén [55]

Table 8. Total PBDE concentration in different tissues from various countries

Country

Year of coll.

Sample type

Total PBDE concentration a Mean

Sweden Sweden Finland Belgium Spain Germany

1999/2000 1996–99 1994–98 –b –b 1999

milk milk milk adipose adipose blood

Germany Canada USA USA

2000 1992 1998 2000

milk milk adipose milk

a b

Values are given in ng/g, lipid-based. No information available.

Author

20/20 93 11 20 13 20

Mayronyté et al. [55] Lind et al. [56] Strandmann et al. [57] Covaci et al. [58] Meneses et al. [59] Schröter – Kermani et al. [54] Fürst [2] Ryan et al. [60] She et al. [61] Päpke et al. [62]

Range

3.5/2.6 (pools) 4 0.9–28.2 2.3 0.7–6.5 3.7 1.7–10.1 1.4 0.2–5.8 5.8 0.9–12.6 2.3 5.8 25.1 200

n

1.5–3.4 0.8–28.5 11.6–37.0 (pool)

7 10 5 ca. 20

292

O. Päpke · P. Fürst

Fig. 14. PBDEs in human milk (values in ng/g lipid). Comparison of data from three laboratories with different analytical methods, from Päpke et al. [62]

determination of halogenated contaminants in human samples: ERGO in Hamburg (Päpke), Chemical and Veterinary Control Laboratory in Münster (Fürst) and Karolinska Institute in Stockholm (Norén). Comparing the HRMS results from the three laboratories – presented in Fig. 14 – the differences in concentrations for most congeners are quite low and the total PBDE concentrations of 204, 196 and 217 ng/g lipid are similar. Due to the relatively difficult determination of compounds with high boiling points like PBDEs, the results of this comparison are quite satisfying. The congener PBDE-47 occurred at the highest level, followed by PBDE-99, PBDE-100 and PBDE-153. These compounds contributed approximately 61–69%, 11–17%, 10–13% and 5–9%, respectively, to the total PBDEs in the pooled US milk sample. It is striking that a concentration of PBDEs as high as approx. 200 ng/g lipid weight is indicated for the US human breast milk. This concentration is about 10 to 40 times higher than levels reported by other authors for mean total PBDEs values for human samples collected recently. It is notable that the PBDE levels in the US human milk pool are at least close to the levels of PCBs in Swedish mothers’ milk, Norén and Meironyté [49]. It cannot be stated that the values found in the US breast milk sample are representative with respect to origin and collection time. The result calls for some in-depth studies on the situation in milk from US women. Despite declining trends for most chlorinated contaminants, the ongoing exposure with polybrominated diphenyl ethers as well as other persistent organic pollutants, especially during the breast-feeding period, is still a matter of concern and justifies further measures to reduce the releases of harmful contaminants to levels as low as technically achievable.

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and other POPs

293

5 References 1. De Boer J, De Boer K, Boon JP (2000) In: Passivirta J (ed), Handbook Environmental Chemistry. Springer, Berlin Heidelberg New York, Ch.4:61–95 2. Fürst P (2001) Organochlorine pesticides, dioxins, PCB and polybrominated biphenyl ethers in human milk from Germany in the course of time. Organohalogen Compounds 52:185–188 3. Waliszewski SM, Aguirre AA, Infanzon RM, Lopez-Carrillo L, Torres-Sanchez L (2000) Comparison of organochlorine pesticide levels in adipose tissue and blood serum from mothers living in Veracruz, Mexico. Bull Environ Contam Toxicol 64:8–15 4. Pardio VT, Waliszewski SM, Aguirre AA, Coronel H, Burelo GV, Infanzon RM, Rivera J (1998) DDT and its metabolites in human milk collected in Veracruz City and Suburban Areas (Mexico). Bull Environ Contam Toxicol 60:852–857 5. Stuetz W, Prapamontol T, Erhardt JG, Classen HG (2001) Organochlorine pesticide residues in human milk of a Hmong hill tribe living in Northern Thailand. Sci Total Environ 273:53–60 6. Saeed T, Sawaja WN, Ahmad N, Rajagopal S, Dashti B, al-Awadhi S (2000) Assessment of the levels of chlorinated pesticides in breast milk in Kuwait. Food Addit Contam 17: 1013–1018 7. Paumgartten FJ, Cruz CM, Chahoud I, Palavinskas R, Mathar W (2000) PCDDs, PCDFs, PCBs and other organochlorine compounds in human milk from Rio de Janeiro, Brazil. Environ Res 83:293–297 8. Gladen BC, Monaghan SC, Lukyanova EM, Hulchiy OP, Shkyryak-Nyzhnyk ZA, Sericano JL, Little RE (1999) Organochlorines in breast milk from two cities in Ukraine. Environm Health Perspect 107:459–462 9. Chikuni O, Polder A, Skaare JU, Nhachi CFB (1997) An evaluation of DDT and DDT residues in human breast milk in the Kariba valley of Zimbabwe. Bull Environ Contam Toxicol 58:776–778 10. Ejobi F, Kanja LW, Kyule MN, Müller P, Krüger J, Latigo AAR (1996) Organochlorine pesticide residues in mother’s milk in Uganda. Bull Environ Contam Toxicol 56:873–880 11. Ludwicki JK, Goralczyk K (1994) Organochlorine pesticides and PCBs in human adipose tissues in Poland. Bull Environ Contam Toxicol 52:400–403 12. Alawi M A,Ammari N, al-Shuraiki Y (1992) Organochlorine pesticide contamination in human milk samples from women living in Amman, Jordan. Arch Environ Contam Toxicol 235–239 13. Kjeller L-O, Rappe C, Jones KC, Johnston AE (1990) Evidence for increases in the environmental burden of PCDD/PCDF over the last century. Organohalogen Compounds 1:433–436 14. Hartmann P, Grupe A, Neupert M (1992) UWSF-Z Umweltchem. Ökotox 4:197–201 15. Schecter A, Dekin A,Weerasinghe N,Arghestani S, Gross M (1988) Sources of dioxins in the environment: A study of PCDDs and PCDFs in ancient, frozen Eskimo tissue. Chemosphere 17:627–631 16. Tong H, Gross M, Schecter A, Monson S, Dekin A (1990) Sources of dioxins in the environment: Second stage study of PCDD/PCDFs in ancient human tissues and environmental samples. Chemosphere 20:987–992 17. Hagenmaier HP, Walczok M (1996) Time trends in levels, patterns and profiles for PCDD/PCDF in sediment cores of Lake Constance. Organohalogen Compounds 28:101–104 18. Baughman R, Meselson R (1973) An analytical method for detecting TCDD (dioxin): levels of TCDD in samples from Vietnam. Environ Health Perspect 5:27–35 19. Miyata H, Kashimoto T, Kunita N (1977) Detection and determination of polychlorodibenzofurans in normal human tissues and kanemi rice oils caused “kanemi Yusho”. J Food Hyg Soc Jpn 18:260–265 20. Langhorst M, Shadoff (1980) Determination of tetra-, hexa-, hepta- and octachlorodibenzo-p-dioxins in human milk. Anal Chem 52:2037–2044

294

O. Päpke · P. Fürst

21. Rappe C, Nygren M, Gustafsson G (1983) Human exposure to polychlorinated dibenzo-pdioxins and dibenzofurans. In: Choudhary G, Keith L, Rappe C (eds), Chlorinated Dioxins and Dibenzofurans in the Total Environment I. Boston, Butterworth, pp 355–365 22. Beck H, Mathar W (1992) Abschluβbericht zum BMFT-Forschungsvorhaben 0765001.PCDD/PCDF in Lebensmittel- und Humanproben incl. Frauenmilch 23. Fürst P, Fürst C, Groebel W (1990) Levels of PCDDs and PCDFs in food-stuffs from the Federal Republic of Germany. Chemosphere 20:787–792 24. Fürst P, Wilmers K (1997) Decline of human PCDD/F intake via food between 1989 and 1996. Organohalogen Compounds 33:116–121 25. Grün M, Päpke O, Weiβbrodt M, Lis A, Schubert A (1995) PCDD/PCDF Intake of humans – a duplicate study in a contaminated area. Organohalogen Compounds 26: 151–154 26. Ryan J, Dewailly E, Ayotte P, Pedersen H, Mulvad G, Hansen J (1996) Inuit Greenland exposure to dioxin-like compounds. Organohalogen Compounds 30:247–250 27. Ayotte P, Dewailly E, Ryan JJ, Bruneau S, Lebel G, Ferron L (1995) PCBs and dioxin-like components in pooled plasma samples from Inuit adults living in Nunavik (Arctic Quebec). Organohalogen Compounds 26:181–185 28. Patterson DG Jr, Needham LL, Pirkle JL, Robert DW, Bagby JR, Garret WA, Andrews JS Jr, Falk H, Bernert JT, Sampson EJ, Houk VN (1988) Correlation between serum and adipose tissue levels of 2,3,7,8-tetrachloro-p-dioxin in 50 persons from Missouri.Arch Environ Toxicol 17:139–143 29. Schecter A, Ryan JJ, Päpke O, Ball M (1991) Comparisons of dioxin and dibenzofuran levels on whole blood, blood plasma and adipose tissue, on a lipid basis. Chemosphere 23:1913–1999 30. Fürst P (1994) In: Jahresbericht des Chemischen Landesuntersuchungsamtes Münster 31. Päpke O (1998) PCDD/PCDF: Human background data for Germany, a 10-year experience. Environmental Health Perspectives 106 (Suppl 2):723–731 32. Liem AKD and Theelen RMC (1997) (Thesis) Dioxins: Chemical Analysis, Exposure and Risk Assessment, ISBN 90–393–2012–8, Den Haag 33. Svensson B-G, Nielsson A, Hansson M, Rappe C, Akesson B, Skerfving (1991) Exposure to dioxins and dibenzofurans through the consumption of fish. N Engl J Med 324: 8–12 34. Welge P,Wittsiepe J, Schrey P, Ewers U, Exner M, Selenka F (1993) PCDD/F-levels in human blood of vegetarians compared to those of non-vegetarians. Organohalogen Compounds 13:13–17 35. Päpke O, Schecter A (1999) Influence of various food consumption habits on the human PCDD/F body burden; 217th ACS National Meting,Anaheim, Div. of Environm. Chemistry, Extended Abstracts 39:108–110 36. Zober M, Päpke O (1993) Concentrations of PCDDs and PCDFs in human tissue 36 years after accidental dioxin exposure. Chemosphere 27:413–418 37. Sagunski H, Csicsaky M, Fertmann R, Roller M, Schümann M (1992) Age dependent levels of PCDD and PCDF in human blood samples. Organohalogen Compounds 9: 211–214 38. Schrey P, Wittsiepe J, Ewers U, Exner M, Selenka F (1992) Age-related increase of PCDD/PCDF-levels in human blood – a study with 95 unexposed persons from Germany. Organohalogen Compounds 9:261–267 39. Päpke O, Ball M, Lis A, Scheunert K (1989) PCDD and PCDF in whole blood samples of unexposed persons. Chemosphere 19:941–948 40. Päpke O, Ball M, Lis A (1995) PCDD/PCDF und coplanare PCB in Humanproben – Aktualisierung der Hintergrundbelastung, Deutschland 1994. Organohalogen Compounds 22:275–284 41. Fürst P, Fürst C, Wilmers K (1992) PCDDs and PCDFs in human milk – statistical evaluation of a 6 years survey. Chemosphere 25:1029–1038 42. Beck H, Dross A, Kleemann WJ, Mathar W (1990) PCDD/PCDF concentrations in different organs from infants. Chemosphere 20:903–910

Background Contamination of Humans with Dioxins, Dioxin-Like PCBs and other POPs

295

43. Abraham K, Päpke O, Ball M, Lis A, Helge H (1994) Concentrations of PCDDs, PCDFs and coplanar PCBs in blood fat of a breast-fed and a formula-fed infant. Organohalogen Compounds 21:163–165 44. Fürst P, Bajanowski T,Wilmers K, Köhler H, Brinkmann B (1999) PCDD/PCDF levels in various tissues from Sudden Infant Death Syndrome (SIDS) cases. Organohalogen Compounds 44:104–110 45. Beck, H, Kleemann WJ, Mathar W, Palavinskas R (1994) Organohalogen Compounds 21:259–264 46. Wuthe J, Link B, Filser J, Kreuzer PE, Piechotowski I, Päpke O (1995) Organohalogen Compounds 26:209–212 47. Jorch G, Schmidt-Troschke S, Bajanowski T, Heinecke A, Findeisen M, Nowack C, Rabe G, Freislederer A, Brinkmann B, Harms E (1994) Epidemiologische Risikofaktoren des plötzlichen Kindstodes. Ergebnisse der westfälischen Kindstodstudie 1990–1992. Monatsschr Kinderheilkd 142:45–51 48. Liem AKD,Ahlborg UG, Beck H, Haschke F, Nygren M,Younes M,Yrjänheikki E (1996) Levels of PCBs PCDDs and PCDFs in human milk. Results from the second round of a WHOco-ordinated exposure study; Organohalogen Compounds 30:268–273 49. Geusau A, Tschachler E, Meixner M, Sandermann S, Päpke O, Wolf C, Valic E, Stingl G, McLachlan M (1999) Olestra increases faecal excretion of 2,3,7,8-tetrachlorodibenzo-pdioxin. The Lancet 354 (No 9186):1266–1267 50. Beck H, Eckart K, Mathar W,Wittkowski R (1989) Levels of PCDD and PCDF in adipose tissue of occupationally exposed workers. Chemosphere 18:507–517 51. SCOOP, Scientific co-operation on questions relating to food. “Assessment of dietary intake of dioxins and related PCBs by the population of EU Member States” Task 3.2.5 – Final Report – 7 June 2000 52. Norén K, Mayronyté K (1998) Contaminants in Swedish human milk. Decreasing levels of organochlorine and increasing levels of organobromine compounds. Organohalogen Compounds 38:1–4 53. Norén K, Meironyté D (2000) Certain organochlorine and organobromine contaminants in Swedish human milk in perspective of past 20–30 years. Chemosphere 40:1111–1123 54. Schröter-Kermani C, Helm D, Herrmann T, Päpke O (2000) The German environmental specimen bank – Application in trend monitoring of polybrominated biphenyl ethers in human blood. Organohalogen Compounds 47:49–52 55. Mayronyté K, Norén BFR (2001) The second international workshop on brominated flame retardants, Swedish Chemical Society, Stockholm 2001, pp 303–305 56. Lind Y,Atuma S,Aune M, Bjerselius R, Darnerud P, Cnattingius S, Glynn A (2001) BFR 2001, The second international workshop on brominated flame retardants, Swedish Chemical Society, Stockholm 2001, pp 117–120 57. Strandmann T, Koistinen J,Vartiainen T (2000) Polybrominated diphenyl ethers (PBDEs) in placenta and human milk, Organohalogen Compounds 47:61–64 58. Covaci A, De Boer J, Ryan JJ, Schepens P (2001) BFR 2001, The second international workshop on brominated flame retardants, Swedish Chemical Society, Stockholm 2001, pp 171–175 59. Meneses M, Wingfors H, Schumacher M, Domingo JI, Lindström G, Bavel B (1999) Polybrominated diphenyl ethers detected in human adipose tissue from Spain. Chemosphere 39:2271–2278 60. Ryan J, Patry B (2000) Determination of brominated diphenyl ethers (BDEs) and levels in Canadian human milk. Organohalogen Compounds 47:57–60 62. She J, Winkler J,Visita P, McKinney M, Petreas M (2000) Analysis of PBDEs in seal blubber and human breast adipose tissue samples. Organohalogen Compounds 47:53–56 62. Päpke O, Bathe L, Bergman Å, Fürst P, Meironyté Guvenius D, Herrmann T, Norén K (2001) Determination of PBDEs in human milk from the United States – Comparison of results from three laboratories. Organohalogen Compounds 52:147Y–150

CHAPTER 11

POPs in Southern Africa H. Bouwman School for Environmental Sciences and Development, Potchefstroom University for Christian Higher Education, P. Bag X 6001, Potchefstroom 2520, South Africa E-mail: [email protected]

The current POPs negotiations have brought to the fore the lack of data and information on the environmental chemistry and ecotoxicology of POPs in developing countries. South Africa has sources and uses of POPs comparable with developed countries, but also has conditions and considerations that are distinctly different. To understand the POPs issue from a South African perspective, I describe aspects of the geography, climate, society, economy, development, regulation and the biota that are relevant to POPs. Natural fires are a possible source of POPs, but these fires have been part of the ecology for more than a million years. The composition of the gases, relating to chlorine are also described. Some of the larger commercial companies have already taken action to reduce their use of PCB-contaminated oils.Very little data are available on environmental levels of dioxins, dibenzofurans and PCBs, but they have been found. Waste burning was identified as a potential major, but unknown, contributor. Stocks of obsolete POPs pesticides have been reduced, and good legislation is in place that has eliminated the registrations of all POPs pesticides, save for chlordane use to protect buildings against termites and DDT for malaria control. Malaria control remains a serious issue, and I illustrate from data that the conditions of application and exposure are very much different from those used in agriculture, and that these considerations must be taken into account when evaluating alternatives. The lack of data could hamper the power of the negotiation positions of developing countries, when compared with developed countries that have more data and information to motivate their agendas. Keywords: Persistent organic pollutants, South Africa, Disease vector control,Vegetation burn-

ing, Environmental conditions

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298

1

Introduction

2

Conditions in South Africa Regarding POPs . . . . . . . . . . . . 299

2.1 2.2 2.3 2.4

Geography and Climate . . . . . . . . . . Environment and Biota . . . . . . . . . . Society, Economy and Development . . . Regulation and Monitoring . . . . . . . .

3

Sources of POPs

3.1 3.1.1 3.1.2

Burning of Vegetation . . . . . . . . . . . . . . . . . . . . . . . . 303 Natural Fires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303 Deliberate Burning of Vegetation . . . . . . . . . . . . . . . . . . 304

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

299 300 301 302

. . . . . . . . . . . . . . . . . . . . . . . . . . . 302

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

298

H. Bouwman

3.1.3 3.1.4 3.2 3.2.1 3.2.2 3.3 3.3.1 3.3.2 3.3.3 3.4 3.4.1 3.4.2 3.4.3 3.4.3.1 3.4.3.2 3.4.3.3 3.4.3.4 3.4.3.5 3.5

Composition of Gases Released During Burning of Vegetation Other Sources of Ignition of Vegetation . . . . . . . . . . . . Industrial Use and Sources . . . . . . . . . . . . . . . . . . . PCBs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Smokestack Emissions . . . . . . . . . . . . . . . . . . . . . Unintentional Combustion . . . . . . . . . . . . . . . . . . . Accidents . . . . . . . . . . . . . . . . . . . . . . . . . . . . Waste Burning . . . . . . . . . . . . . . . . . . . . . . . . . Environmental Levels of Dioxins and Dibenzofurans . . . . . Agricultural and Non-Agricultural Sources of POPs Pesticides Obsolete Stocks . . . . . . . . . . . . . . . . . . . . . . . . . Heptachlor . . . . . . . . . . . . . . . . . . . . . . . . . . . DDT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Human Levels of DDT from Malaria Control . . . . . . . . . Environmental Levels of DDT from Malaria Control . . . . . Alternatives for DDT in Malaria Control . . . . . . . . . . . Methylsulphone Derivatives . . . . . . . . . . . . . . . . . . Human Levels of DDT from Tsetse Fly Control . . . . . . . . Sources of PTS . . . . . . . . . . . . . . . . . . . . . . . . .

4

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316

5

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

304 305 306 306 306 307 307 307 307 308 308 308 309 309 312 313 314 315 315

1 Introduction The negotiations preceding the signing and implementation of the Stockholm Convention to limit or ban the release of Persistent Organic Pollutants (POPs), thereby protecting human health and environment, has focused attention on the available knowledge of the environmental chemistry and ecotoxicology of these compounds. The effects of these compounds are due to the combination of being persistent, the ability to be transported over long distances, the potential to be bio-accumulated, and their toxicity to biological systems. The combination of characteristics of POPs means that the biological effects can be concentrated in areas where they have not been used. The levels of POPs in Inuit Indians in the Arctic is a good example of this [1]. It is also true that the increased awareness, and subsequent action taken to address the POPs problem, was due to the work done in the Arctic [1]. Current consensus is therefore that POPs release by whichever means in warmer areas, does affect biological systems at colder climes, due to the ability of these compounds to be transported over long distances. During the negotiations of the Consultative Expert Group on POPs (during 1998 and 1999), the Intergovernmental Negotiating Committee on POPs (1998–2000), as well as the results from the Global Environment Facility sponsored workshops to prepare for a Regionally-Based Assessment of Persistent Toxic Substances [2], it became clear that there are significant data and information gaps regarding POPs for developing countries, as well as for regions with moderate and tropical

POPs in Southern Africa

299

climes. The extent and importance of these gaps, as well as the future implications regarding the obligations of developing countries as parties to the Stockholm Convention, once these data gaps have been addressed, are largely unknown. The conditions under which POPs are released to the environment in South Africa are similar to those in developed countries, but there are also particular circumstances and considerations that must be taken into account. The aim of this paper is to describe the known and possible sources and release patterns of POPs in South Africa and neighbouring countries, supported by some data from the literature. I will then examine the conditions and constraints regarding actions to be taken to limit the release of and reliance on POPs, as well as those conditions that could favour persistence or break-down of compounds that are less obvious, when compared with conditions from moderate and colder regions under which the current POPs characteristics are assumed. I will conclude with a synthesis, followed by a brief analysis of the importance of the data and information gaps to developing countries during international negotiations.

2 Conditions in South Africa Regarding POPs The combination of environmental conditions and some aspects of the development in South Africa need to be understood to be able to derive a framework within which POPs and required action can be assessed. The principle of common but differentiated responsibility (based on countries as parties to the Stockholm Convention) regarding minimisation of POPs release therefore needs to be taken into account when interpreting the “differentiated” part of the principle. It is obvious that there are already differences in the levels of social and economic development between South Africa (and similar countries), compared with those more developed. The same reasons can also be advanced to explain the paucity of data regarding POPs, as referred to in the introduction. It could be argued that these data gaps can be filled by modelling and extrapolation, but I will indicate that the underlying assumptions of use and release patterns, as well as biology and sociology differ so much that the results from indirect approaches should be treated with caution. These approaches need verification by direct measurements. The conditions described below will not be dealt with exhaustively, but serve to provide a reference within which the POPs issues can be evaluated. 2.1 Geography and Climate

Southern Africa contains a diversity of habitats that is arguably unsurpassed by any other region of comparable size [3]. This diversity is the result of a combination of geography and climate. There are two major oceanic currents – the warmer Agulhas Current from the Indian Ocean brings tropical water from the north to the east coast, while the colder Benguela Current brings colder water from the south, flowing northwards along the Atlantic (west) coast [4]. The cold Benguela Current does not bring rain, and the western part of the subcontinent is therefore mainly arid, while the highest precipitation occurs mainly in the east-

300

H. Bouwman

ern parts. Most of southern Africa receives less than a mean of 500 mm per year, which is also seasonal and very unpredictable; many parts experience extended drought periods and occasional floods. Rain is cyclical, with most of the region receiving rain during the austral summer, while the south-western tip experiences winter rainfall. Few major rivers are present, with the Orange and Vaal Rivers being the major system. This is also the system that supplies water to most of the industrial areas of the Vaal Triangle, but flow and levels of water in associated reservoirs can be significantly reduced during prolonged droughts. Many of the other minor rivers also experience large variations in flow, depending on rainfall and season. Wetlands are one of the most threatened ecosystem types in South Africa. In many places more than half have been lost [5]. Current landcover data shows almost 600,000 ha of wetlands, or 0.48% of the total land surface [6]. The growing seasons are generally short, motivating the POPs negotiation position of South Africa that the half-life criterion for POPs in water be set at two months, and not six months. Temperature varies seasonally, geographically and diurnally. Snow falls every winter on the Drakensberg. In the winter in the interior, the daily temperature can change by 20 to 25 °C between day and night. Large areas experience long sunny periods, with temperatures that can exceed 40 °C during daytime in the arid central and western regions. The coastal regions experience the moderating influence of the oceans, but large day to day fluctuations do occur [3]. The combination of climate and topography of the eastern coastal plain of South Africa (and Mozambique), create conditions suitable for malaria, as well as sleeping sickness. Variable climate (especially wet seasons and floods) also causes sporadic and serious outbreaks of these diseases in more arid areas where they do not normally occur, and the health infrastructure cannot always cope with these type of emergencies. 2.2 Environment and Biota

The geography, rainfall and temperature are the crucial determinants of the biogeographical patterns in southern Africa. Amongst regions of comparable size (2,570,000,000 km2), southern Africa has the richest flora in the world. In this subcontinent, there are seven major biomes, ranging from forest to desert (excluding the freshwater and marine habitats), covering 68 vegetation types. Five of the biomes are rich in species, and also comprises one of the world’s floral kingdoms; the Cape Floristic Region that includes the fynbos vegetation type referred to later. This is also one of the 10 floristic “hot-spots” in the world. Together they comprise 50,000 species (20% of the world total) on 0.5% of the world’s land surface. More than 6000 of these species occur only in the Cape Floristic Region, making it the “hottest” of the “hot-spots” [7]. There is also a high level of plant species endemism (80%) in southern Africa [8]. Mammalian diversity is also high, with 291 species of terrestrial and land living animals [9]. There are more than 108 species of frogs [10], 245 species of freshwater fish, many of which are endemic [11], and an estimated 80,000 species of insects (including 10,000 to 12,000 species of moths and butterflies) [12]. Bird

POPs in Southern Africa

301

diversity is also exceptional with 928 species on the list for southern Africa, of which 161 are endemic or near-endemic to the region [13]. Of these 928 bird species, 123 are taken up in the Red Data Book [14]. Many of the other groups of biota mentioned above have high proportions of species that are endangered, mainly due to habitat destruction, including those species associated with the endangered wetlands. The combination of habitat and species diversity forms the basis of a growing tourism and eco-tourism industry, that needs to be protected from the effects of a number of threats, including possible pollution from current and possible future POPs. 2.3 Society, Economy and Development

The more than 41 million people of South Africa populate 1,221 million km2 at an average density of 34 people km–2. For 1997 the World Bank estimated the life expectancy of South Africans at 62 and 68 years for males and females respectively [15], but the 2000 estimation from the WHO reduced this to 39 and 41 years, respectively [16], mainly due to AIDS. On GNP South Africa was ranked 32nd in the world in 1998, with agriculture contributing only 4% and industry 38%. More than 50% of the population were rated below the poverty gap at 2 $ purchasing power parity per day [15]. South Africa produced an estimated total of 292.7 million tons of CO2 from industry during 1996 [15]. Domestic combustion of coal, wood, crop residues and dung are the most serious sources of air pollution in South Africa. The Development Bank of Southern Africa estimates that 20–25 million people breathe unsafe air [17]. People relying on wood and coal as domestic fuel experience total suspended particulates at rates 3–8 times higher than guidelines. Children between the ages of 8–12, exposed to domestic wood and coal smoke, experience an increased risk of 290% to develop upper respiratory tract infections during winter, compared with the summer. Wood is also the prime fuel source for the rural population, contributing to deforestation, erosion, decreased surface water retention and siltation, as well as contributing to greenhouse gas production [17]. No data on the extent of this are available. The major electricity supplier (Eskom) is, however, committed to increase connections, and managed to connect 1,750,000 homes between 1994 and 2000. This will eventually contribute towards the decrease in inefficient domestic fuel combustion. During 1998 Eskom used 87.2 million tons of coal to produce 165,473 GWh of electricity. Eskom estimates that the chlorine content of the coal is very low, and Eskom have no POPs data [Pers. com. Gericke]. Past mining activities though (by various companies), have left a legacy of a 1000 million tons of discard coal in dumps [17]. Slow-burning fires in these dumps, as well as underground fires in coal seams, also contribute unknown amounts towards air pollution.

302

H. Bouwman

2.4 Regulation and Monitoring

No regulations specific to dioxins, dibenzofurans or PCBs exist in South Africa. POPs pesticides are covered as pesticides under a specific act (Act 36 of 1947) that allows the government to limit or ban the use of specific products. Under this act the registration of the following products were withdrawn [18]: – Aldrin: All uses withdrawn voluntarily in 1992. – Chlordane: All uses withdrawn, except for protecting constructions against termites. – Dieldrin: Registration withdrawn in 1980. All subsequent resale or use of existing stocks prohibited in 1983. – Endrin: All uses withdrawn voluntarily in 1980. – Heptachlor: Registration withdrawn in 1976. – DDT: Registration withdrawn on all agricultural uses in 1976. All subsequent resale or use of existing stocks prohibited in 1983. Use is only allowed for malaria control by the government. – Hexachlorobenzene: Registration withdrawn; use and resale prohibited in 1983. (HCB might be used in the production of other chemicals.) – 2,4,5-T: Although not a POPs pesticide per se, the production process of this herbicide allowed the formation of 2,3,7,8-TCDD [19]. All registrations were withdrawn in 1989. The other two POPs pesticides, mirex and toxaphene, have never been registered in South Africa. The registrar of the pesticide registration act (Act 36/1947) has an inspectorate that investigates all aspects of the act, including any use of prohibited pesticides. There are no national monitoring schemes for POPs. The data presented below are mainly from independent studies, formal and informal reports, personal communications or limited monitoring efforts. Obsolete stocks will be dealt with in Sect. 3.4.4. Agricultural export products are being monitored for many pesticide residues, including the POPs pesticides, by the Forensic Chemistry Laboratory of the National Department of Health. In the last few years, however, very few POPs pesticides have been detected. No DDT has been detected, and only occasional dieldrin residues in certain types of produce (squash, butternuts, etc.). All of these were below any levels of concern or action.

3 Sources of POPs Volcanic activity and vegetation fires are two possible natural sources of dioxins and dibenzofurans [20]. There are no active volcanoes in southern Africa, nor on the continental shelf. The closest volcanoes are located in central Africa. The other sources, both natural and anthropogenic, will be discussed below, but some attention will be given to sources or uses particular to southern and South Africa. One of these sources are fires, both natural, accidental and managed burning of vegetation.

POPs in Southern Africa

303

3.1 Burning of Vegetation

The next section discusses natural fires in some detail, so as to understand management options and constraints regarding suggestions to reduce POPs emissions from this source. 3.1.1 Natural Fires

An estimated 562–1736 million tonnes of dry biomass is burned every year (including natural and accidental) in the southern hemisphere part of Africa [21], but other data indicate less (177±87 million tonnes) [22]. For the purpose of this discussion I will include under natural fires also those fires used for vegetation management in conservation areas. Otherwise, natural fires are normally started by lightning strikes, estimated as 1–10% of all fires in southern Africa [21]. Since plant biomass must be available for combustion (a minimum of 50–100 g m–2 is needed), few fires occur in the more arid regions, except after very wet seasons [21]. Also, insect and mammalian herbivory can remove much of the vegetation in certain areas before fires occur, depending on circumstances [23]. Fires are also mainly seasonal, consuming mainly dry vegetation during the end of the dry season, or beginning of the rain season when convective storms occur, with enough dry vegetation still present. The fynbos vegetation of the Cape (in the winter rainfall region) is predominantly evergreen, but accumulation of dead material results in high intensity fires, because of the long intervals between these events (5–45 years). Volatile organic emissions are therefore expected to be high [23]. Because of the natural occurrence of fires in the African ecosystems as a whole, evolutionary processes resulted in many plant and animal species having adapted to the fire cycles to a significant extent. Flowering, pollination, seed dispersal, germination, and even senescence and death of many plant species, as well as the breeding, feeding, behaviour and nesting cycles of many animals are regulated by fires [23, 24]. In the savanna and grassland systems, the competitive balance between the grasses and trees is also maintained by fire [23]. Specific associations have developed between animals and plants (especially in the fynbos), to such an extent that through a strategy called myrmecochory, ants pick up specifically adapted, protein-rich seeds from certain plants, and then store them 5 cm below ground, protected from the effect of fire. The protein-rich part of the seed is consumed by the ants, but the seed will germinate after fire. Since these ants make their nests in open sunny sites, the seedlings experience less competition [24]. The above illustrates that for the management and protection of fire adapted and dependent bio-diversity, natural fires are a necessity, and cannot be reduced. The regime of natural fires must be maintained, but accidental fires seem to increase this frequency in certain areas, and could be considered a threat to the systems affected. In some other areas, because of diligent fire prevention and management, the current burning frequency and intensity seems to be less

304

H. Bouwman

than in historical times, favouring trees over grass [21], and could even contribute towards reducing grassland area available for pasture, or grazing by the smaller herbivores such as antelope. If the burning of vegetation is considered a source of POPs subject to the Stockholm convention, then natural fires need to be specifically exempted from action under the still to be finalised Article D [25]. 3.1.2 Deliberate Burning of Vegetation

Humans have used fire in southern Africa for abut a million years, but the relationship between human and naturally induced fires and grassland remains unclear [27]. Although there are now many more sources of ignition of vegetation fires than before the recent period of large-scale human development of southern Africa, the resultant fragmentation of the landscape by roads and habitation, as well as current fire management practices, have probably resulted in comparable or even less areas being burned than previously [26]. Human induced burning of vegetation, especially grasslands, is due to many reasons: clearing of areas, soil fertilisation by rapid pyro-mineralization of the plant-contained nutrients, improved hunting opportunities (green shoots attracting grazers), and elimination of pests, to name a few. Large-scale sugarcane burning prior to harvesting is also deliberate [21, 27]. It is not possible at present to separate the influence of human and naturally induced fire regimes, and therefore the POPs contribution from these types of events. It is clear though that fire remains a part of bio-diversity maintenance and ecological management. The basis on which fires will be managed in conservation areas have recently changed from a more planned and regulated fire regime, controlled on area basis, to that of allowing natural fires to run their course. This is due to the recognition that irregular events have an important role to play in the maintenance of diversity, system structure, system composition and ecosystem function in space and time. 3.1.3 Composition of Gases Released During Burning of Vegetation

The composition of the gases released during vegetation burning in southern Africa, has been the subject of a large integrated study, conducted during the early 1990s. This was called the Southern African Fire-Atmosphere Research Initiative (SAFARI) [28]. This study did not collect data on chlorinated dioxins or any other chlorinated POPs, but some of the chlorine and furan related measurements could give some indication of possible POPs formation. One of the pyrogenic trace gasses found in plumes, namely methyl chloride, is also involved in the stratospheric ozone cycle, and therefore regularly measured. Globally, biomass burning contributes almost 100% of the methyl chloride emissions (1.1 gigatons), and about 66% of the particulate organic carbon (60 gigatons) [28]. The African savannah fires are estimated to contribute between 14–27% of global biomass burning emissions of CO2 and NOx, from 2.5 gigatons of biomass [28].

POPs in Southern Africa

305

The SAFARI data were summarised by Andraea in 1997 [27], from which the following data were obtained. The furans that were measured were normalised according to the CO2 released during controlled fires in the Kruger National Park. The emission ratios for furan, 2-methylfuran, and 3-methylfuran were 0.22, 0.080–0.11, and 0.018, respectively. The chlorine and bromine content of the fuel (the biomass before burning) was 1,260 (±310) and 7.0 (±1.6) mg kg–1, respectively [27]. Of this content, 83% and 81%, respectively, were emitted in the smoke (calculated from the remaining content in the unburned and ash fractions). Of the halogens emitted it seems that methyl chloride accounts for only 3% of the chlorine, and methyl bromide only 5% of the bromine emitted [27]. No analyses on other halogen species were done, and these could have been hydrogen halides that will combine with alkaline substances in the aerosol, or with gaseous ammonia to give halide particles. From these data it was calculated that African savannahs contribute roughly 220 million tons of methyl chloride, compared with the total of 1.1 gigatons emitted globally, all through burning of vegetation, per year [27]. From the above it is therefore difficult to establish any contribution of POPs from African biomass fires. The studies referred to concentrated on mass-balance ratios and budgets, not on the, presumably, extremely small proportion of chlorine that might be captured in POPs compounds during combustion. The halogen content of African biomass is, however, larger than initially expected. These aspects therefore need further study. As the Stockholm Convention will eventually reduce the major anthropogenic sources of POPs, the minor and natural sources (such as biomass burning) could receive more attention. 3.1.4 Other Sources of Ignition of Vegetation

A unique and possible additional source of POPs is the deliberate use of explosives to control Quelea, a pest bird species (Quelea quelea) in South Africa. There are an estimated 1,500,000,000 birds in Africa, causing serious losses to small-grain crops. Because of the toxic nature of organophosphates (normally fenthion) to aquatic life, chemical control is not allowed on or near water bodies [29]. Whenever a flock roosts in a wetland, or in certain cases in trees, use is made of a fuel-air explosion. In most cases most of the vegetation is burned. Data available from the National Department of Agriculture show that 171 control operations were carried out in habitats described as reeds, wetlands and rivers between 28/10/1993 and 28/05/1999. A total of 327.8 ha was affected, killing an estimated 48,000,000 Quelea. The smallest area was 0.04 ha, and the largest was 35 ha. The explosive used is pentolyte, with gasoline or paraffin as fuel. The flame lasts for a couple of seconds, reaching a temperature of about 600 °C. The surface area treated is small, however, when compared with that involved in other vegetation fires, and would therefore contribute little to POPs releases.

306

H. Bouwman

3.2 Industrial Use and Sources

No POPs pesticides are currently produced in South Africa, although DDT has been manufactured before. 3.2.1 PCBs

The main electricity supplier in South Africa is Eskom, which generates 95% of the countries’ requirements, and also for export. Eskom does have transformers and capacitors with PCBs, but a programme with the aim of a PCB-free system is in place [30]. There are a few pure PCB transformers and capacitors still in place, but they are being replaced as they become redundant. Only a small percentage of the remaining equipment contains PCBs in the oils at above 50 mg l–1, and procedures are in place to prevent further contamination during transfers and maintenance. Eskom has also shipped some PCB-contaminated oils overseas for incineration [30]. Sasol, which is a very large chemical concern, are in the process of screening their own transformer oils for PCBs. Iskor, a major steel producer, has also embarked on a similar process, and have already sent most, if not all, of their PCBcontaminated oils overseas for incineration [31]. Not much is known about environmental levels of PCBs in South Africa. In a recent study, fish samples from only one site (Olifants River in Mpumalanga) showed detectable levels of PCBs (41.0 mg kg wet weight). Fish, water and sediment samples from other sites of the same catchment had no detectable levels of PCBs [32]. 3.2.2 Smokestack Emissions

No information on POPs releases from smokestacks is in the public domain. I know of reports done by consultants, for companies in South Africa, but the information remains proprietary. The environmental reports for Eskom and Sasol do not contain this information, nor are they measured at present. Both companies are aware of POPs. Sasol (the largest chemical industry in southern Africa), operates gasifiers (using high ash-content coal) and incinerators, and have embarked during 2000 on an initial survey to analyse for dioxin and dibenzofuran from these plants. If anything significant is found, further analyses will be conducted [33].

POPs in Southern Africa

307

3.3 Unintentional Combustion 3.3.1 Accidents

Little is known about POPs from accidental fires in South Africa. There are unconfirmed reports of transformer fires and transformer oil spillages. I know of instances of soil pollution with PCB-contaminated oils, but the extent and location is not known. There have been a number of industrial fires, but I have been unable to obtain any reports in the public domain that contain POPs information. From the materials involved, and the type of accidents, these incidents are likely to have generated POPs. Again the extent of formation is not known. 3.3.2 Waste Burning

Uncontrolled burning of waste is a problem throughout South Africa. Uncontrolled and informal waste dumps are particularly at risk. Waste dumps are also used as informal resources by poor people. Materials are scavenged for resale or own use. The burning of rubber tyres to harvest the wire reinforcement for resale is a well known, albeit very inefficient and dangerous, source of income. Waste incineration is a regulated form of waste management in SA.According to the National Waste Management Strategy [34], the general waste incinerators, which are currently in operation, are small and generally inefficiently operated. It is envisaged that these incinerators will either be upgraded (using technology appropriate to South African conditions) to meet new national air emission standards (to be revised based on international standards), before 2002, or they will be decommissioned. The government also considers that incineration of general waste with efficient air pollution control as an expensive treatment method, and it will not promote this type of treatment. For hazardous waste from domestic activities, industry and manufacture, including stockpiled waste, the current treatment practice of landfilling (codisposal in class H sites in many cases) will be phased out by two envisaged thermal treatment facilities operating at international standards, by 2008. For hazardous medical wastes, those dedicated incinerators currently in operation will be upgraded to comply with the revised standards, or else decommissioned and replaced by new facilities [34]. It is not stated whether POPs criteria will be part of these revised standards. I could, however, trace no data on POPs emissions from any formal or informal waste treatment sources. 3.3.3 Environmental Levels of Dioxins and Dibenzofurans

The only published environmental data on chlorinated dioxins and dibenzofurans were from analysis done on breast milk [35]. For black rural South Africans

308

H. Bouwman

(n=6) and white urban South Africans (n=18), the mean dioxin levels were 279.8 and 355.6 ng kg–1 in milk fat, respectively. For the dibenzofurans the levels were 20.4 and 23.1 ng kg–1 in milk fat, respectively. The respective TEQs were 8.3 and 12.6. The same article also looked at breast milk samples from the USA (rural Tennessee, n=9) and Vietnam (Tay Ninh, n=4). The dioxin levels were 350.7 and 577.5 ng kg–1 in milk fat, respectively. For dibenzofurans the levels were 22.2 and 48.3 ng kg–1 in milk fat, respectively. The respective TEQs were 14.6 and 22.9 [35]. This shows that these compounds are present in South Africa, and that different population groups have different exposure profiles, probably related to diet [35]. The numbers sampled are small, however, and do not for instance reflect exposure of black urban women to smoke from coal and wood fires, as reported in Sect. 2.3. 3.4 Agricultural and Non-Agricultural Sources of POPs Pesticides

Not all of the identified POPs pesticides have been registered (Sect. 2.4), and of the remainder, only DDT and heptachlor retain restricted use status. In this section I will discuss obsolete pesticides and their management, as well as the remaining uses of POPs pesticides, particularly the use of DDT in malaria control. 3.4.1 Obsolete Stocks

Although most of the compounds have been banned for quite some time now, here, as elsewhere in Africa and other parts of the world, stocks of obsolete pesticides have been accumulating for various reasons. The FAO estimates about 20,000 tonnes in Africa alone [36]. South Africa addressed this problem during 1997/99, with a country-wide retrieval scheme, funded by central government. Together with the crop protection industry association (AVCASA), 32 collection points throughout the country were set up, where users could dump their stock. Containers with unknown contents were sampled and analysed by the Agricultural Research Council – Plant Protection Research Institute. The containers were then repacked and shipped overseas for incineration. More then 700 tonnes, including many of the POPs pesticides, were collected and destroyed in this way [37]. It is unknown how much obsolete stock remains. A programme to prevent accumulation of obsolete stock is being developed. 3.4.2 Heptachlor

South Africa did not request any exemptions for heptachlor under the POPs negotiations, and will probably phase out its registration after acceding to the Stockholm Convention, since alternatives, although more expensive, are available. Presently about 200–300 tons of active ingredient are imported into South Africa every year.

POPs in Southern Africa

309

3.4.3 DDT

The more important POPs compound remains DDT. The context of its importance is the use in disease vector control. Currently only malaria control is done with DDT, but previously the two major vectors for sleeping sickness have been eliminated, also with DDT. The importance of the use of DDT in these two diseases will be discussed next. 3.4.3.1 Human Levels of DDT from Malaria Control

Malaria is one of the most important causes of death, ranked second after HIV/AIDS, in Africa [16]. In 1999 an estimated 953,000 people died in Africa, and more than 400 million suffered acute malaria, costing the African countries south of the Sahara more than $ 2 billion (1997 data) [16]. One of the mainstays of combating malaria, is the interruption of transmission, using pesticides applied on the walls of dwellings in affected areas. In 1934, weekly application of pyrethrum was introduced with good results [38]. DDT has been used for malaria control in South Africa since 1946, for both larviciding and house spraying. Larviciding with DDT was discontinued in 1956 [38]. Between 1957 and 1977, DDT was applied twice a year; after 1977 it was reduced to only once a year [39]. DDT is normally applied at 2 g m–2. During 1988, for instance, in the province of KwaZuluNatal alone, 98,912 dwellings, housing more than 300,000 people, were treated with 26,018 kg of 75% emulsifiable DDT [39]. Two other provinces in South Africa are also malaria endemic, namely Mpumalanga and the Northern Province, where the application regime is the same as in KwaZulu-Natal. Currently between 4–5 million people in South Africa live in malaria endemic areas, and are protected from infection by almost a 1000 malaria control personnel applying insecticides to the insides of the dwellings [40]. DDT proved to be very successful in South Africa, keeping the rate of infection down. The malaria data for South Africa are given in Table 1, which will also illustrate the problems South Africa encountered with the introduction of alternatives (Sect. 3.4.3.3). Because of public pressure over the continued use of DDT, research was initiated to investigate the human and environmental levels and possible impacts of DDT in KwaZulu-Natal in the middle 1980s. This represents a period of typical DDT application and use pattern, still continuing today, in an area where DDT had not been used, other than for tsetse-fly control between 1946 and 1952 by intensive aerial application, dust and smoke generators (see Sect. 3.4.3.5) [41]. The following discussion will investigate some of the findings of this research programme. One of the immediate questions that comes to mind when applying DDT to the indoor surfaces of living dwellings, is how much of the DDT is taken up by the inhabitants. Table 2 shows the results from analysis of serum from an exposed and control group. All DDT variables were significantly (p<0.05) higher in the exposed group. The safety of DDT used in malaria control for subjects aged three and more, was confirmed by the levels of DDT in serum when compared with

310

H. Bouwman

Table 1. Number of malaria cases and deaths in South Africa [40]. The introduction of

pyrethroids as replacement for DDT was started in 1996 Year

Cases

1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985

364 1792 331 1623 1821 1747 3513 7103 2022 3109 2343 2184 2130 4642 11358

Deaths 5 23 12 16 4 6 1 36 12 10 9 13 13 19 32

Year

Cases

Deaths

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000

7491 10374 9317 7055 6822 4693 2872 13285 10289 8750 27035 23121 26445 51535 37279

20 10 48 30 35 19 14 45 12 44 163 104 198 402 280

Table 2. Summary statistics of DDT levels (mg L–1) in serum from a DDT exposed group

(n =71), and a non-exposed group (n =77). The ages ranged from 3 to more than 80 years old [39] Exposed

DDE DDD DDT ΣDDT b a b

Non-exposed

Mean

SD a

% Positive

Mean

SD a

% Positive

103.4 0.21 37.3 140.9

85.1 0.7 27.2 108.3

100 14 100

5.95 0.015 0.077 6.04

7.98 0.13 0.31 8.19

92.2 1.3 6.5

Standard deviation. SDDT=DDT+DDE+DDD.

other studies available in 1991, which showed lack of any negative effects associated with these levels in adults. Liver function in the exposed and control groups were also comparable and not indicating any effects from exposure [39]. The effects of these levels regarding endocrine disruption [42] in humans have, however, not yet been quantified. The levels present in the group protected from malaria is rather high, and any effects would more likely be found here, rather than in developed countries where these levels are much lower. Since the DDT is applied once a year, the next issue is how these levels change over 12 months in the exposed group measured above. Household members exposed to DDT were sampled four times over 12 months and longitudinal changes measured. In the exposed group the increase in DDT, DDE and SDDT in the older

311

POPs in Southern Africa

age group (21+) following application was significant, but not for the group aged between 3–20 [43]. We found a reduction in serum levels in the younger age group in contrast to the increase in levels in the older age group. This increase, however, we considered at that stage not to present additional risk. More significantly though was that we hypothesised that two concurrent processes govern the increase and decrease in levels of DDE and DDT in the serum of the exposed group. The relative contributions of these two processes transpose in prominence with age. This was confirmed by a significantly faster rate of change (decrease) for DDT than for DDE in the younger age group, but not in the older age group [43]. From this we concluded that any risk assessment concerning vector control chemicals must therefore consider all sub-groups of a population, since we found that children experience conditions different from their parents, and very much different from children in developed countries. I will return to this aspect later. The level of DDT in breast milk, and subsequent infant exposure, is also a contentious issue.We analysed breast milk from the same populations as represented in Table 2, and looked at the factors that determined these levels [44]. The results are presented in Table 3. The difference between the two groups were again highly significant (p<0.001). Primiparous mothers of the exposed group had significantly more SDDT and metabolites (24.82 mg kg–1) than multiparous mothers (12.21 mg kg–1). We found parity to be the best predictor of DDT in breast milk of the exposed group. Another interesting finding was that the percentage DDT of the SDDT increased significantly with an increase in parity. This was explained by an increase in percentage DDT occurring with higher parities, due to the uptake of DDT and elimination via milk. This process then, was faster than the uptake and endogenous formation of DDE. Based on information available in 1990, we found these levels not to present an appreciable health risk to the mothers [44]. During one breast milk collection event in the exposed area, we sampled infant blood, using a capillary tube (44.7 mL), from toe pricks of 23 infants, as well as breast milk from the corresponding mothers [45]. The mean SDDT in the whole infant whole blood was 127.03 mg L–1 (min– max 29.4–316.5). These values should at least be doubled to compare with the serum levels in Table 2. The mean SDDT in the breast milk fat was 15.06 mg kg–1. The percentage DDT was significantly higher (p<0.05) in the infant blood than in the breast milk. MultiTable 3. Summary statistics of DDT levels (mg kg–1 in milk fat) in breast milk from a DDT ex-

posed group (n=129), and a non-exposed group (n=88) [44]

DDE DDD DDT SDDT b a b c

Mean

SD a

Min–Max

Mean

SD a

Min–Max

8.65 0.400 6.77 15.83

7.67 0.398 4.31 11.62

0.5–46.9 ND c-2.14 0.42–28.8 1.05–59.3

0.65 0 0.04 0.69

0.665 0 0.05 0.68

ND-4.73 ND-0.03 ND-0.36 ND-4.8

Standard deviation. SDDT=DDT+DDE+DDD. ND=Not detected.

312

H. Bouwman

plicative regression showed a significant increase in SDDT (p<0.01) in infant whole blood with infant age.We therefore concluded that the DDT and DDE were mainly transferred from the mother, with DDT in the environment of the infant only a secondary source [45]. Combined with the data presented in Table 2, it was clear that blood levels dropped dramatically, after cessation of breast feeding (mothers regularly breast feed for up to two years in rural areas), and then increased again after the age of about 20–30 years. We did a risk assessment based on the infant data and other published studies, and found that there exists a well founded risk (neurological) to the infants, particularly to the firstborns, under malaria control conditions [39]. The clinical significance, as well as the permanency of the possible effects were however, not known, due to the very few clinical studies done on infants. Again the data must be reviewed, taking the known endocrine disrupting effects of DDT [42] into account. 3.4.3.2 Environmental Levels of DDT from Malaria Control

It is to be expected that the application of large amounts of DDT for malaria control, albeit very much different from agricultural application, will result in release to the environment.We investigated levels in fish and birds from the DDT-treated areas of KwaZulu-Natal. Fish was investigated because it is also a significant source of dietary protein to the inhabitants [46]. Some of the data are presented in Table 4. In general we found low levels of DDT and metabolites.We also found no significant variations in levels before and after malaria application of DDT.As such, these levels we considered not to pose a health hazard to the local population. These levels also contributed very little to their body burden. Possible deleterious effects to fish-eating animal species could not be ruled out [46]. Following these findings, we determined the whole blood levels of DDT in a bird called the Pied Kingfisher Ceryle rudis [47], using the same techniques as developed for infants. The highest levels were found in Kosi Bay, on the KwaZuluNatal coast (189.09, 5.55, 47.14 and 241.78 mg L–1 of DDE, DDD, DDT and SDDT, respectively, in whole blood). For birds these values should be multiplied by about 1.8 to derive serum data, comparable with those in Table 2. This also meant that this bird is an excellent indicator of DDT pollution associated with aquatic systems, wherever these birds occur [47]. We could find no eggs of these birds (a drought interrupted the breeding cycle), but from extrapolations based on studTable 4. Levels of DDT and metabolites (mg kg–1 wet weight) in two fish species from the Pon-

golo Flood Plain – KwaZulu-Natal [46]. Standard deviations are given in brackets Species

n

DDE

DDD

DDT

SDDT

Tigerfish a Tilapia b

15 26

54.2 (66.9) 12.2 (17.0)

24.6 (32.5) 16.8 (30.8)

14.9 (17.4) 7.1 (17.4)

91.8 (113.4) 33.4 (60.2)

a b

Hydrocynus vittatus. Oreochromis mossambicus.

POPs in Southern Africa

313

ies of other birds, we calculated mean and maximum SDDT egg levels of 2.91 and 4.79 mg kg–1, respectively. This is in the same range as for the eggs of the Brown Pelican Pelecanus occidentalis, which showed impaired reproductive success at levels as low as 2.5–3 mg kg–1 [48]. We therefore concluded that at these levels, there may be reproductive risk, keeping in mind the problems associated with extrapolations [47]. There have been a number of other studies on levels of DDT; but since many of them have been conducted during the time of agricultural use of DDT, or outside the malaria areas, they are not reported here. 3.4.3.3 Alternatives for DDT in Malaria Control

South Africa recently took the decision to phase out DDT and to switch to pyrethroids for malaria control. This was after a lengthy research and development phase on laboratory and pilot trials in the field. Large-scale phase-in of pyrethroids started during the 1996/1997 season, and was completed by the beginning of 1999. During this phase-in period, however, it became apparent that the number of malaria cases was on the increase (Table 1). By the end of 1999, this number stood at 50,000 for South Africa – far higher than in recent decades (Table 1).Also, a mosquito species (that was eradicated by DDT 30 years ago) has made a comeback. This species is pyrethroid-resistant and also transmits malaria throughout the year. It has subsequently been found in Mpumalanga and the Northern Province. Based on this the National Department of Health decided to return to DDT in all three provinces, in an effort to combat this species. An estimated 215 tons of DDT will be used in the 2000/2001 season, which is comparable to amounts used before. In the mean time, organophosphate resistance has also been detected, severely limiting the classes of insecticides available for malaria control. The return to DDT is perhaps a serious setback (or a timely warning) to malaria control programmes and the Roll Back Malaria campaign. It has also had implications for the Stockholm Convention. One of the issues that I have indicated from the beginning is the difference in conditions and use patterns regarding POPs in southern Africa, when compared with regions where more research and data have been done. In other developing regions such as in South America and Asia, successful alternative methods have been introduced. In southern Africa we have different mosquito vector complexes, disease epidemiology, parasites, environmental conditions and infrastructure constraints, amongst others.A working Integrated Vector Management system (IVM; including bednets, environmental management and biological control) still needs a long development period, and also needs to be acceptable to the communities. A working malaria vaccine has also not yet been developed. Alternative pesticides also cost more, making it unaffordable to many countries. As I indicated in Sect. 3.4.3.1, the exposure of people to insecticides applied in houses, differs significantly from the human exposure profile for which they were developed, namely agricultural crop protection. Therefore the human health risk assessment under malaria control conditions should by different from that of agricultural exposure. Inhabitants of these houses, including infants, children and the

314

H. Bouwman

elderly, are exposed for much longer compared with the exposure profile of agricultural workers. It follows that health risk assessments of alternatives should take these differences into account, including combined exposure to DDT (DDT still present from previous applications) and the alternative(s), as well as other aspects such as endocrine disruption. Also, the use of alternative insecticides should be sustainable and predictable, and the risk of mosquito resistance development must be minimised and managed, primarily by involving agriculture in the malaria areas, as water contamination with crop protection chemicals poses the most probable risk of such development. Even if alternatives are employed that have less of an impact, environmental monitoring must still be done to determine the risks to biota, since malaria control often occurs in bio-diversity rich areas (or close to conservation areas), and the application methods differ very much from that used in agriculture. 3.4.3.4 Methylsulphone Derivatives

One metabolite of DDT, that was not well known during the studies mentioned above, is MeSO2DDE. Normally, lipophilic substances are metabolised to more hydrophilic compounds (hydroxylation) that can then be excreted. The mercapturic acid pathway, on the other hand, produces a hydrophilic metabolite of DDE – a methyl sulphone metabolite (MeSO2DDE) that is retained in the body after formation [49]. MeSO2DDE was first identified from the environment in seal blubber in 1976 [50]. MeSO2DDE has subsequently also been found in breast milk from Swedish mothers [49]. The levels of MeSO2DDE found were low, but MeSO2DDE is a potent toxicant of the adrenal cortex in mice [50]. Animal studies have been based on single doses (IP), but no information about chronic exposure and effects is available. MeSO2DDE has also been shown to be transferred (via milk) from dams to pups of mice, where it was accumulated in the adrenal glands. A single dose of 25 mg kg–1 MeSO2DDE to the dam resulted in lesions in the adrenals of the receiving pup via milk, at the higher dose. Seven days after dosing, the labelling of the pup adrenals was 5.3 and 9.3 times higher when compared with the dam. Because it seems that DDT will be used for malaria control for some time, the DDE residues in the environment, as well as those in the human bodies, will also remain. We have also shown that there is a significant difference in the rates of change of DDE and DDT in serum between younger (3–20 years) and older (21+ years) age groups [43]. DDT was reduced at a much faster rate than DDE in the younger group, while slight increases of both were observed for the older group. This could imply that whereas DDT was reduced, the DDE was changing much slower, presenting a store of DDE available for further metabolism – including to MeSO2DDE. Extrapolated from the Swedish data we have calculated the daily intake of infants from breast milk (Table 3) at 0.11, 0.19 and 0.59 mg kg–1 MeSO2DDE at 8.56, 15.06 (mean for primiparous mothers), and 46.9 mg kg–1 (maximum) DDE in milk fat in breast milk. Again these are extrapolations, indicating the lack of information and data for the region. PCBs also have this derivative, but again, no data for South Africa.

POPs in Southern Africa

315

3.4.3.5 DDT in Tsetse Fly Control

Large amounts of DDT and BHC have been used to interrupt tsetse fly transmission of trypanosomiasis (sleeping sickness) to cattle and humans between 1946 and 1952 [41]. The two major vector species in the area were eliminated with a combination of aerial and ground based applications. No environmental data could be found for this period, except the possibility that certain orchids in KwaZulu-Natal may have declined because of the loss of their insect pollinators [52]. In South Africa, the main area of past sleeping sickness occurrence is congruent with the malaria endemic area of KwaZulu-Natal; where most of the data on DDT was presented above, was derived from. There remain, however, two minor tsetse fly vectors of the disease in the area, and minor outbreaks have occurred in 1980, 1987 and 1990. Currently pyrethroid dips and veterinary medicines are used to protect and treat cattle, and the situation is actively monitored [41]. 3.5 Sources of PTS

Although the Stockholm Convention covers the twelve current POPs, it also makes provision for adding more compounds, under certain procedures [25]. The concept of Persistent Toxic Substances (PTS) was created to accommodate compounds that have characteristics similar to POPs, but are not classified as such [2]. It also accommodates a situation for persistent release (therefore persistent presence) of chemicals in the environment. In this context I would like to discuss an issue concerning the environmental conditions typical to southern Africa, and possibly other regions with similar climates. It is generally assumed that persistence is a function of the conditions experienced by the molecule; generally speaking the more benign the conditions, the more persistent, especially in terms of temperature, UV and pH. For the Stockholm Convention, the criterion of a half-life of six months in soil and sediment is currently accepted [25]. There are substances though, for which the major route of breakdown is not chemical or physical, but biological. Persistent herbicides (typically substituted ureas, uracils and triazines) are used under certain conditions for total vegetation control, and are resistant to photolysis and chemical degradation [53]. Under conditions that are typical for use in Europe or elsewhere, the soil bacterial activity metabolises these compounds, and the soil organic fraction prevents mobility. The microbial activity in soils is normally related to its organic content. There are unpublished data suggesting that the generally low organic content (typically <1%), bacterial activity, low moisture content (from arid areas) and the aged nature of our soils, contributes to much longer half-lives of the compounds (measured in years, rather than months) [53]. The low soil organic content means that the compounds are more mobile (less binding sites), and that plant toxicity occurs at soil concentrations lower than for soils with higher organic content, because of higher bio-availability. The increased mobility also means that the compounds can be leached from soils to rivers, and can therefore be transported, as

316

H. Bouwman

well as being persistently present. These preliminary finding obviously need further interpretation, but show that arid conditions can promote persistence. It again points towards the lack of data, and the underlying assumptions of degradation models, more suited towards moderate and polar conditions. It has the further implication that current use compounds, with no POPs characteristics in moderate climates, could conceivably be classified as POPs, given that long-range transport and toxicity is also shown.

4 Conclusions Persistent organic pollutants are clearly an international issue. The long-range transport characteristics of these compounds are a cause for concern and concerted international action. Part of the negotiations of the Stockholm Convention is the international assistance to be provided by the developed countries, so that developing countries can meet their obligations. In general there seems to be four issues: 1) use of pesticide POPs in agriculture; 2) elimination of PCBs from transformers; 3) reduction in the unintentional production of chlorinated dioxins, dibenzofurans and PCBs; and 4) the use of pesticide POPs in disease vector control. Of these four issues, South Africa in particular does not seem to have a problem with the eventual elimination of the remaining registration of chlordane for agricultural use.Although alternatives are expensive, they are available. The regulatory framework for registration and inspection exists and has been operating for quite a while. The Act has already eliminated the uses of all the other POPs pesticides, and a retrieval scheme has been very successful in reducing the overall stocks of obsolete pesticides.Although it is likely that illegal stocks are still being used, the little data from monitoring suggest that this is not problematic at this stage, although the monitoring issue needs much more attention.A National Agricultural Residue Monitoring Scheme is being developed between government departments and research organisations [54]. Similarly, the PCB issue is being addressed by industry, despite no formal driver to do so. The principles of Responsible Care and Product Stewardship are accepted by organised industry. It is obvious that the larger concerns are aware and have action plans in various stages of completeness, but the smaller users might not be aware of the urgency. The soil pollution problems, although not formally recognised, are also being addressed with bio-remediation and other initiatives.Very little data are available on environmental PCB levels, and this clearly needs to be addressed, also in the light of the obligations of South Africa under the Stockholm Convention. On the issue of unintentional POPs by-product formation, much less is known. No specific legislation exists (at the time of writing) to address this issue, or to enforce compliance. There is also no set standards of emissions or other forms of release. The larger companies are aware of this issue, and some have already taken action, at least to determine the potential of such formation. The presence of dioxins and dibenzofurans in breast milk does indicate that there are sources, but it is not currently possible to identify these. Much more needs to be done to determine human and environmental levels, as there are conditions of possible

POPs in Southern Africa

317

exposure via coal and wood smoke, produced during combustion for domestic heating of a large proportion of South Africans. Other sources of incineration, both formal and informal, also need attention, as they are potentially large sources of POPs. The burning of vegetation as a possible POPs source, also needs attention. The vegetation in southern Africa seems to have relative high levels of chlorine, most of which is released in the smoke. Depending on the type of vegetation and fire, and from what is already known from the composition of the smoke, it is not inconceivable that POPs can be formed. It needs to recognised that this is a natural process. From the malaria control point of view, it seems probable that DDT will remain, if not as the prime agent, then certainly as an option to manage vector resistance, for the foreseeable future. This issue clearly needs urgent attention from a southern African perspective. The Roll Back Malaria campaign and the concerted WHO/UNEP attention will keep the focus on this issue. The unknown MeSO2DDE implications, as well as the realistic risk assessments of alternatives should always be part of research and implementation. It serves no purpose to replace a reasonably well known set of risks (for example that associated with DDT), with another set of risks (due to alternative measures), if these risks are not well quantified. Most of all, from the recent South African experience, the sustainability of alternative compounds and strategies must be kept in mind. I have set out to describe the conditions under which South Africans experience POPs. These conditions are very much different from those in the more developed countries, as the DDT data clearly indicate. We have use patterns and considerations, that, because many of the qualifiers of these on POPs production and exposure are so badly known since a reduction of these under the broad assumptions of models geared towards the conditions and exposures of biota under more moderate climates, can only be interpreted with the utmost caution, unless verified or suitably adapted. Furthermore, the combination of topography and climate has blessed South Africa with rich, and in many cases unique bio-diversity and ecological processes. This is part of our natural capital and needs protection, not only from habitat destruction, pollution and other threats, but also from lack of knowledge and international awareness. South Africa and, I assume, many other developing countries have a distinct disadvantage when it comes to negotiations and obligations under international treaties. The principle of common responsibility is clear, but to negotiate from a base that lacks knowledge, could presumably oblige countries to accept (albeit conditionally) the positions or arguments from others (possibly at their own disadvantage), because these positions and arguments cannot be countered from facts relevant to their own conditions. The only option left is to advance arguments as I have done above, but these lack power due to their hypothetical or speculative nature. The only way forward is to fill the knowledge gaps, especially those based on the experience and needs of the countries requiring knowledge, and not primarily dictated from perceptions from elsewhere. From the perspective of the Stockholm Convention, this clearly indicates much effort concerning the environmental chemistry and ecotoxicology of POPs and their alternatives where relevant.

318

H. Bouwman

Acknowledgement. I would like to thank the following, in no particular order: Sarel Cilliers, Pieter Theron, Koos Kotze, Kobus Pienaar, Renier Terblanche, Dr E.M. Nevill, Peter Nelson, Gerhard Gerricke, Rob Wood, Egmont Rohwer, Leon Smuts, Marthie Coetzee, Johan van Vuuren, Steve Jenkins, Lianda Lötter, the personnel from Act 36 of 1947, Nompelo Daniel, Buti Mathebula, Fred Goede, Dr Brückner, Huib van Hamburg, Peter Apps, Steven Evans, Kobus le Roux, Dr K.D. Hearshaw, Azel Swemmer, BiBi Bouwman and Louis van Dyk. The opinions expressed above are entirely my own and do not necessarily reflect those that have assisted me, the government of South Africa, industry, my employer, nor any other organisation. It also does not assume or presume any current or future positions of South Africa regarding POPs.

5 References 1. Crane K, Galasso JL (1999) Arctic environmental atlas. Office of Naval Research, Naval Research Laboratory, Hunter College 2. United Nations Environment Programme (1999) Regionally based assessment of persistent toxic substances. UNEP Chemicals, Geneva 3. Harrison JA, Allan DG, Underhill LG, Herremans M, Tree AJ, Parker V, Brown CJ (eds) (1997) The atlas of southern African birds, vol 1. Non passerines. Birdlife South Africa, Johannesburg 4. Payne AIL, Craford RJM (1989) Oceans of life of southern Africa. Vlaeberg publishers, Halfway House, South Africa 5. Kotze DG, Breen CM, Quinn N (1995) Wetland losses in South Africa. In: Cowan GI (ed), Wetlands of South Africa. Department of Environmental Affairs and Tourism, Pretoria, South Africa, p 263 6. Fairbanks DHK, Thompson MW,Vink DE, Newby TS, van den Berg HM, Everard DA (2000) The South African land-cover characteristics database: a synopsis of the landscape. S Afr J Science 96:69 7. Cowling RM, Hilton-Tyler C (1994) Patterns of plant diversity and endemism in southern Africa: an overview. In: Huntley BJ (ed), Botanical diversity in southern Africa. National Botanical Institute, Pretoria, South Africa, p 31 8. Cowling RM, Gibbs Russel GE, Hoffman MT, Hilton-Tyler C (1989) Patterns of plant species diversity in southern Africa. In: Huntley BJ (ed), Biotic diversity in southern Africa. Oxford University Press, Cape Town, South Africa, p 19 9. Smithers RHN (1992) Land mammals of southern Africa: a field guide, 2nd edn. Southern Book Publishers, Halfway House, South Africa 10. Channing A, van Dyk DE (1995) Amphibia. In: Cowan GI (ed), Wetlands of South Africa. Department of Environmental Affairs and Tourism, Pretoria, South Africa, p 193 11. Skelton P (1993) ‘n Volledige gids van varswatervisse van suider Afrika. Southern Book Publishers, Halfway House, South Africa 12. Scholtz CH, Holm E (1986) Insects of southern Africa. Butterworths, Durban, South Africa 13. Maclean GL (1999) Southern African endemic birds: Their distribution and conservation. In: Adams NJ, Slotow RH (eds), Proc 22 Int Ornithol Congr. Durban. BirdLife South Africa, Johannesburg, South Africa 14. Barnes KN (2000) The Eskom red data book of birds of South Africa, Lesotho and Swaziland. BirdLife South Africa, Johannesburg, South Africa 15. The World Bank (2000) Entering the 21st century: World development report 1999/2000. Oxford University Press, New York 16. World Health Organization (2000) World health report 2000: Health systems: Improving performance. World Health Organization, Geneva 17. Development Bank of Southern Africa (1998) Infrastructure: a foundation for development. Development Bank of Southern Africa, Midrand, South Africa

POPs in Southern Africa

319

18. Nel A, Krause M, Ramautar N, van Zyl K (1999) A guide for the control of plant pests. National Department of Agriculture, Pretoria, South Africa 19. Carey J, Cook P, Giesy J, Hodson P, Muir D, Owens W, Solomon K (1998) Ecotoxicological risk assessment of the chlorinated organic chemicals. Setac Press, Pensacola, chap 1 20. Niimi AJ (1994) PCBs, PCDDs and PCDFs. in: Calow P (ed), Handbook of ecotoxicology. Blackwell Science, Oxford, p 668 21. van Wilgen BW, Scholes RJ (1997) The vegetation and fire regimes of southern Africa. In: van Wilgen BW, Andreae MO, Goldammer JG, Lindesay JA (eds), Fire in southern African savannas. Witwatersrand University Press, Johannesburg, South Africa, p 27 22. Kendall JD, Justice CO, Dowty, PR, Elvidge CD, Goldammer JG (1997) Remote sensing of fires in southern Africa during the SAFARI 1992 campaign. In: van Wilgen BW, Andreae MO, Goldammer JG, Lindesay JA (eds), Fire in southern African savannas. Witwatersrand University Press, Johannesburg, South Africa, p 89 23. Fishman J (1997) The larger perspective: Trace-A. In: van Wilgen BW, Andreae MO, Goldammer JG, Lindesay JA (eds), Fire in southern African savannas. Witwatersrand University Press, Johannesburg, South Africa, p 17 24. Cowling R, Richarson D (1995) Fynbos: South Africa’s unique floral kingdom. Fernwood Press, Vlaeberg, South Africa, chap 5 25. United Nations Environment Programme (2000) Intergovernmental negotiating committee for an international legally binding instrument for implementing international action on certain persistent organic pollutants: Fourth session, Bonn. UNEP/POPS/INC.4/5. United Nations Environment Programme 26. Goldammer JG (1997) A synthesis of the results of the SAFARI campaign. In: van Wilgen BW, Andreae MO, Goldammer JG, Lindesay JA (eds), Fire in southern African savannas. Witwatersrand University Press, Johannesburg, South Africa, p 239 27. Andreae MO (1997) Emissions of trace gasses and aerosols from southern African savanna fires. In: van Wilgen BW, Andreae MO, Goldammer JG, Lindesay JA (eds), Fire in southern African savannas. Witwatersrand University Press, Johannesburg, South Africa, p 161 28. Lindesay JA (1997) African savanna fires, global atmospheric chemistry and the southern tropical Atlantic regional experiment. In: van Wilgen BW, Andreae MO, Goldammer JG, Lindesay JA (eds), Fire in southern African savannas. Witwatersrand University Press, Johannesburg, South Africa, p 1 29. Bouwman H, Lötter L (1998) Redbilled Quelea control in South Africa: 1 January – 30 September 1998. Bird Numbers 7:20 30. Lombard AE (1999) Polychlorinated biphenyl (PCB) management and disposal in South Africa. PCBs and dioxins – Seminar. Stellenbosh, South Africa 31. van Vuuren J (2000) SASOL. Personal communication 32. Grobler DF (1994) A note on PCBs and chlorinated hydrocarbon pesticide residues in water, fish and sediment from the Olifants River, Eastern Transvaal, South Africa. Water SA 20:187 33. Goede F (2000) SASOL. Personal communication 34. Department of Environmental Affairs and Tourism (1999) National waste management strategy: Version D. Danced Ref No M123–0136. Department of Environmental Affairs and Tourism, Pretoria, South Africa 35. Schecter A, Startin JR, Rose M, Wright C, Parker I, Woods D, Hansen H (1990) Chlorinated dioxin and dibenzofuran levels in human milk from Africa, Pakistan, southern Vietnam, the southern U.S. and England. Chemosphere 20:919 36. Food and Agriculture Organization (1999) Obsolete pesticides: Problems, prevention and disposal. Food and Agriculture Organization, CD ROM 37. van der Merwe MJ (1999) The role of PPRI in the safe disposal of toxic waste in South Africa. Plant Protection News (55):7 38. Sharp BL, Ngxongo S, Botha MJ, Ridl F, le Sueur, D (1988) An analysis of 10 years of retrospective malaria data from the KwaZulu areas of Natal. S Afr J Sci 84:102

320

H. Bouwman

39. Bouwman H, Cooppan RM, Reinecke AJ, Becker PJ (1990) Levels of DDT and metabolites in breast milk from KwaZulu mothers after DDT application for malaria control. Bull WHO 68:761 40. Maharaj R (2000) National Department of Health. Personal communication 41. Nevill EM (1999) Agricultural research Council – Onderstepoort Veterinary Institute. Personal communication 42. Kelce WR, Gray LE, Wilson EM (1998) Antiandrogens as environmental endocrine disruptors. Reprod Fertil Dev 10:105 43. Bouwman H, Becker PJ, Schutte CHJ (1994) Malaria control and longitudinal changes in levels of DDT and metabolites in human serum from KwaZulu. Bull WHO 72:921 44. Bouwman H, Reinecke AJ, Cooppan RM, Becker PJ (1990) Factors affecting levels of DDT and metabolites in human breast milk from KwaZulu. J Tox Environ Health 31:93 45. Bouwman H, Becker PJ, Cooppan RM, Reinecke AJ (1992) Transfer of DDT used in malaria control to infants via breast milk. Bull WHO 70:241 46. Bouwman H, Coetzee A, Schutte CHJ (1990) Environmental and health implications of DDT-contaminated fish from the Pongolo Flood Plain. Afr J Zoo 104:275 47. Evans SW, Bouwman H (2000) The geographic variation and potential risk of DDT in the blood of Pied Kingfishers from the Northern KwaZulu-Natal, South Africa. Ostrich 71:351 48. Blus LJ (1982) Further interpretation of the relation of organochlorine residues in Brown Pelican eggs to reproductive success. Environ Pollut Ser A 28:15 49. Noren K, Lunden A, Pettersson E, Bergman A (1996) Methylsulfonyl metabolites of PCBs and DDE in human milk in Sweden. Environ Health Persp 104:766 50. Jensen S, Jansson B (1976) Methyl sulfone metabolites of PCB and DDE. Ambio 5:257 51. Lund BO (1994) In vitro adrenal bioactivation and effects on steroid metabolism of DDT, PCBs and their metabolites in the grey seal (Halichoerus grypus). Environ Tox Chem 13:911 52. Bond WJ (1989) The dynamic nature of biotic diversity. In: Huntley BJ (ed), Biotic diversity in southern Africa. Oxford University Press, Cape Town, South Africa, p 1 53. Meinhardt HR, Nouwman, H, Cloete, MM (1998) Herbicide usage in agriculture and industry – a need to know more. National Rivers Initiative, Pietermaritzburg, South Africa 54. Brückner GK, Odendaal L (2000) National Department of Agriculture. Personal communication

CHAPTER 12

Organochlorines in Nigeria and Africa Oladele Osibanjo Department of Chemistry, University of Ibadan, Ibadan, Nigeria E-mail: [email protected]

Many organochlorine compounds including DDT, lindane, HCB and PCBs which are persistent and bioaccumulative, have been used in Africa including Nigeria for almost half a century without environmental considerations. The main uses are in crop protection, animal and public health as well as industries. Nonetheless, there is gross misuse and abuse of these ecologically harmful chemicals including unorthodox pathways of human exposure. Despite the paucity of data on organochlorine use and contamination levels in the continent, the residue levels in Nigeria are used as a microcosm of the African environment contamination problems by these xenobiotics. Degradation studies on aldrin, DDT and lindane show that these organochlorines degrade much faster with short half-lives of a few weeks in Nigerian soils compared to several weeks and even years reported in cold temperate soils. This raises the problem of transposition data from developed countries to African countries and underscores the need for local research to obtain true persistence data derived from empirical models. Environmental contamination by organochlorines to different extents in the aquatic environment, land, wildlife, foodstuffs, human diets, human blood and breast milk in Nigeria was established from a literature review, and the data compared with those from other sub-regions in the African continent, Asia and the developed countries. The estimated maximum daily intake (EMDI) of aldrin + dieldrin in some Nigerian foodstuffs was greater than the FAO’s acceptable daily intake (ADI) which calls for caution and control action. Compared to more industrialized countries and with the exception of some hot-spots sites, the concentrations of organochlorines were low. Nonetheless, the gross contamination of human blood and mothers’ milk place children especially and the human population at risk from exposure to organochlorines. The foregoing situation underscores the fact that contamination of the environment and humans is a problem of both developed and developing countries. With increasing population, greater demand for food and industrialization, organochlorines use may increase. Priority attention should therefore be given in Africa to capacity building in organochlorine residue analysis and regulatory control of toxic chemicals including the tracking of organochlorines from cradle to grave. Keywords: Organochlorine residues, Half-life, Breast milk, Contamination, Nigeria

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322

1

Introduction

2

Production and Uses . . . . . . . . . . . . . . . . . . . . . . . . . 323

3

Sources of Contamination and Exposure to Organochlorines

4

Degradation of Organochlorines in Nigerian Soils

. . 324

. . . . . . . . 326

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

322

O. Osibanjo

5

Organochlorines Levels in Nigeria and the African Environment

5.1 5.2 5.3 5.4 5.5 5.6 5.7

Concentration of Organochlorines in Water . . . . . . . . . . Concentrations of Organochlorines in Sediments . . . . . . . Concentrations of Organochlorines in Fish . . . . . . . . . . . Concentrations of Organochlorines in Foods . . . . . . . . . Concentrations of Organochlorines in Nigerian Wildlife . . . Organochlorine Concentrations in Human Breast Milk from Nigeria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Organochlorine Concentrations in Human Blood from Nigeria

6

Issues of Concern

7

References

. . . . .

328 . . . . .

329 334 337 342 345

. . 345 . . 350

. . . . . . . . . . . . . . . . . . . . . . . . . . 352

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353

1 Introduction Organochlorines (OCs) are synthetic chlorinated hydrocarbons and include pesticides (e.g., DDT and derivatives, hexachlorocyclohexane (HCH), aldrin, dieldrin, heptachlor and endosulfan) as well as some commercial or industrial chemicals and by-products of combustion such as polychlorinated biphenyl (PCBs), hexachlorobenzene (HCB), and polychlorinated dibenzo-p-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF). The peculiar characteristics of these chemicals are that they have a relatively high chemical stability and persist in the environment for long periods; are transboundary pollutants travelling large distances from their point of origin in the environment and bioaccumulate in humans and other organisms through the food chain. Consequently, most developed countries have banned or severely restricted the use of most of the OC pesticides replacing them with the less persistent organophosphates and carbamates. In spite of these measures some of the organochlorines still persist in the environment at dangerous levels threatening humans, ecosystems and biodiversity. In Africa, population explosion, rapid urbanization and industrialization as well as severe pest problems including insects, plant diseases, weeds, rodents as well as most of the devastating plagues known to man, especially the African armyworm, locusts and grain-eating birds have increased reliance on the use of pesticides for almost half a century in the African region [1]. Organochlorine pesticides are widely used in the region because of their relatively cheaper cost and also being first generation insecticides. The environmental fate as well as the adverse ecotoxicological and human health effects of OCs and other persistent organic pollutants (POPs) have been well investigated and documented in the developed countries while such environmental data are scanty in developing nations. Furthermore, public awareness on the health hazards and risks associated with the use of OCs is low in Africa. The need for global action to control organochlorines, especially the POPs and other persistent toxic substances (PTS), has been recognised and would require the co-operation and integrated efforts of all governments.

Organochlorines in Nigeria and Africa

323

An overview of organochlorines sources, environmental fate and contamination status is presented below in the African region, with Nigeria as the most populous country in Africa and the second largest economy in the continent after South Africa as a case study.

2 Production and Uses The history of organochlorine pesticides use in the continent date from the 1940s for agricultural production of food for the continent’s rapidly growing population (3–4% annually) and cash crops for economic buoyancy as well as malaria vector control activities. But there is a paucity of data on the production and use of these chemicals in Africa while such data are readily available in developed countries [1]. This is a reflection of the low level of understanding of the hazards of pesticide use. Most of the pesticides used in Africa are imported from the developed countries, especially Europe,America and Japan; but there are a few production/formulation facilities in some countries, e.g., Nigeria, Senegal, South Africa, Ivory Coast and Egypt. It is estimated that Africa uses about 100,0000 metric tons of pesticides annually out of the world annual total use of 2.5 million metric tons [2]. The largest pesticides users in Sub-Sahara Africa based on expenditure on agrochemicals are Sudan, Tanzania, Zimbabwe Cameroon, Ivory Coast, Kenya, Nigeria and South Africa [1]. The bulk of the pesticides are used on the following cash crops: cotton, coffee, cocoa, maize, tobacco, bananas, sugar, and rice. About 50,000 tonnes of OCs are consumed annually in the region while regulatory control is lacking. The types, quantities and application mode of OCs vary across the continent. A survey on pesticide usage in Nigeria [3] indicated that about 15,000 metric tons annually of pesticides comprising about 135 pesticide chemicals marketed locally under 200 different product brands and formulations were imported during 1983–1990. About 2,500 tonnes of OCPs mostly DDT, toxaphene and endosulfan were used annually on cotton plantations in the 1970s in Sudan [3]. More than 3500 tonnes of DDT were used on cotton plantations in Uganda between 1965 and 1972 [4]. In Ivory Coast in 1976, about 600 tonnes of lindane were used for cocoa and 320 tonnes of DDT were applied on cotton. In 1981, about 350 tonnes of lindane, dieldrin, heptachlor and endrin were used for timber protection. In Zimbabwe, about 300 tonnes of DDT applied at the rate of 2–3 kg/ha were used in agriculture between 1981 and 1982. In Burkina Faso, in the HounddDddougou region, 30 tonnes of DDT and 30 tonnes of endosulfan annually were used on cotton. In the areas north of the Sahara (Mauritania, Mali, Niger and Sudan), the control of desert locusts for several years was by OCs (dieldrin and lindane). Limited amounts of dieldrin were still used during 1986–1988 while presently organophosphates are used which are less persistent than the OCs [5]. Ground spraying and aerial application of OCs, especially DDT, dieldrin and endosulfan to control vectors for human and livestock diseases are also an important source of contamination of aquatic ecosystems. Since 1944, DDT had been used largely for black fly larvae control (Simuiiidae) in many regional pro-

324

O. Osibanjo

grammes in Africa.About 60 tonnes of DDT were used annually in the continent to control Simulildae from 1966 to 1970 (Dejoux, 1988). Tsetse fly control and eradication programmes involving the spraying of DDT, dieldrin and endosulfan have taken place in different parts of the region over the last 20 to 30 years as well [4]. The pesticide usage rate and quantum are likely to increase substantially in third millennium in view of the commitment of most African governments to green revolution, local sourcing of raw materials, food security and improved health care for the increasing population.

3 Sources of Contamination and Exposure to Organochlorines Anthropogenic activities provide the primary point source of OCs into the environment either through deliberate application or accidental release into the environment. The main anthropogenic sources of these substances into the African environment are [6]: (i)

Deliberate application, e.g., spraying of pesticides to control aquatic weeds, snails and insects; or air-craft spraying of large farms for control of birds, disease vectors, tsetse fly, etc. (ii) Dumping of obsolete stock of OCs, wastes/containers from public health, agriculture and industrial usage on land and water. Release of OCs in this way had caused massive fish kills in many countries such as Senegal, Nigeria and Kenya. According to there are about 20,000 tonnes of stock of obsolete industrial chemicals and pesticides all over Africa. (iii) Accidental spillage from agricultural, industrial sites and electrical transformer sites, road and rail vehicles and ships. (iv) Drainage and run-off (including leachates) from treated farmlands, garbage and industrial solid wastes dump. (v) Domestic and industrial effluents – effluents from pesticide manufacturing or formulating industries or industries using OCs, e.g., food industry. (vi) Dumping of sewage sludge, municipal, agricultural and industrial solid wastes on land and inland waters. (vii) Atmospheric input, e.g., in dry deposition and wet precipitation; burning or uncontrolled incineration of OCs. (viii) Domestic, municipal or agricultural solid wastes; industrial emissions, e.g., through evaporation/vapourisation of pesticides stored in the sun for long periods or leaks/seepages from containers. Organochlorines are known to cause acute poisoning in humans and biota. The important routes of human body exposure are through the respiratory, oral or dermal routes. Dermal exposure and absorption are the most important routes of entry under occupational exposure situations [7]. Occupational poisoning episodes occur largely during spraying, mixing, and diluting of pesticides. The use of malfunctioning or defective equipment is also an important factor contributing to accidental acute poisoning among agricultural workers. Recent data indicate that about 11 million cases of pesticide intoxication occur annually in

325

Organochlorines in Nigeria and Africa

Africa, including minor cases that did not require hospitalization [8]. Table 1 provides an overview of pesticides poisoning in some African countries. Misuse and abuse of pesticides in the continent as well as unorthodox pathways of human exposure have also been identified. In Nigeria. for example, g-HCH (Gammalin 20) has been used to kill fish in streams and rivers, to treat smoked fish against weevils, to control pests of stored kolanuts, to kill head lice, to relieve/treat tooth aches and in diluted form, it is ingested as a worm expeller [9]. Most common occurrences of pesticides intake and human exposure in the African region due to lack of regard for safety and the use of Personal Protection Equipment (PPE) are: (i) Dusts and sprays entering mouth during agricultural application, (ii) Drinking pesticides by mistake from unlabelled or contaminated sources, (iii) Drinking pesticides in improperly stored food containers like soft drink or beer bottles; drinking pesticides for committing suicide; using the mouth for siphoning pesticides concentrates; eating contaminated food and the transfer of pesticides to the mouth from contaminated hands and cuffs spillage on the skin during pouring and mixing, entering the treated farms unprotected, (iv) Spillage on the skin during pouring and mixing, entering the treated farms unprotected, (v) Inhaling droplets or particles of pesticides during application. The relative importance of each of these point sources depends on the quantity or concentration of OCs discharge, abiotic and biotic conditions. Food and water contamination, and the issue of contaminated soil and derelict land are some of the problems arising from the discharge of OCs into the environment through the foregoing routes . The accidental release of OCs in large quantities had caused massive fish kills in many countries such as Senegal, Nigeria and Kenya.

Table 1. Number of poisonings in some African countries in the 1980s Country

Population (millions)

% Agricultural labor force

Cases of pesticide poisoning per year

Sudan Tanzania Kenya Uganda Mozambique Cameroon Zimbabwe Ivory Coast Malawi Senegal Mauritius

24 23 22 17 15 11 10 10 8 7 2

80% 85% 80% 80% 70% 80% 80% 80% 85% 80% 75%

384000 368000 350000 272000 240000 175000 160000 160000 128000 112000 3200

326

O. Osibanjo

The principal land-based sources of OCs into the aquatic environment are soils sprayed with pesticides for agricultural practices or vector control operations, land fills, production sites and contaminated soils. Co-disposal of industrial, municipal, domestic and medical wastes in open dumps, or non-sanitary landfills or open burning is commonly practiced. Though these waste disposal methods may be cheap and convenient, they are not environmentally safe and sound and therefore not acceptable. Leachate from open dumps and landfills are therefore recognised also as sources of OCs pollution of surface and ground water in Africa [10]. Land-based sources especially agricultural run-off, rivers and direct discharge of industrial and municipal wastes have been estimated to contribute a total organochlorine pesticide load of about 90 tonnes/annum to the Mediterranean Sea [11]. No information has been published yet for the load into the coastal and marine environment in the African region. Nonetheless, similar exercises are in progress in the West and Central Africa (WACAF) and East Africa (EAF) subregions especially as some OCs such as PCBs, DDT and dieldrin have become global pollutants due to (a) their volatilisation from the sites of application; (b) atmospheric transport and deposition and (c) transport via rivers and ocean currents [12, 13]. According to FAO there are about 20,000 tonnes of stock of obsolete industrial chemicals and pesticides all over Africa with potential hazard to the environment. While data for individual countries are not generally available Nigeria has about 22 tonnes of obsolete stocks [14] while Gambia had 25 tonnes of obsolete/banned pesticides [15]. which include contaminated soils and packaging materials which were to be exported to the United Kingdom for incineration in August last year. Electrical transformer oils are the principal source of PCBs, which are nonagrochemicals, release into the environment. Most of the electrical transformers purchased by the power generating agencies and companies in the continent during the 1970s and 1980s contain polychlorinated biphenyls (PCBs). Some of the transformers are leaking PCB-containing oil without bounding walls around them thereby creating a problem of PCBs contaminated soil. Nigeria has unquantified large stocks of PCB-containing transformer oil from power stations and manufacturing industries. Generally, the disposal practice for spilled PCBs or PCBs-bearing equipment was to dig pits and bury underground which practice is environmentally unsustainable.A regional study on the inventory of PCBs-contaminated transformer oil and disposal as well as PCBs-containing equipment is desirable.

4 Degradation of Organochlorines in Nigerian Soils The half-life of a pesticide is a measure of its persistence in the environment and has been recognised as one of the key chemical properties apart from the vapour pressure at 25 °C, the octanol-water partition coefficient (Kow) and the octanolair partition coefficient (KoA) as generic criteria in the selection of candidate OCs for international control and phase-out.

327

Organochlorines in Nigeria and Africa

There is, however, a paucity of data on the half-lives of OCs in soils of Africa. Transposing OCs degradation data from developed countries with temperate weather will not be realistic in the tropical weather conditions of the African region.A review of the pathways of OCs in the atmosphere suggested volatilisation as a major pathway. Scientific literature has shown that volatilisation and degradation of pesticides are more rapid in the tropics than in the temperate zones [16, 17]. In view of the long history of OCs use in Nigeria, the persistence of some commonly used OCs namely g-HCH (lindane), aldrin and DDT in cropped and uncropped soils from Southern Nigeria had been investigated [18]. Table 2 indicates the half-life values obtained for these organochlorine pesticides in Nigerian soils compared to values reported in literature for temperate, African and Asian countries. These OCPs degrade much faster with shorter halflife t1/2 , of a few weeks in the tropical Nigerian soil compared to t1/2 of several weeks in the cold temperate soils. The trend of the persistence of the organochlorines was aldrinaldrin>DDT similar to the water solubility trend of the chemicals. This suggests a higher potential for lindane to leach into ground water. In spite of the relatively short half-lives of OCs in Nigerian soils, there is wide spread contamination of Nigerian soils. The contamination trend in soils (Table 3) is private farms
Table 2. Persistence of some organochlorine pesticides (OCPs) in Nigerian soil compared to

soils in other countries OCP

DDT

Aldrin Lindane

Half life, t1/2 (weeks) Cropped Soils from other countries

Nigeria (uncropped)

Nigeria (cropped)

146 (Temperate climate) 161 (Temperate climate) 8.6–17.1 (India) 28.6 (Kenya) 2.6 (Sudan) 56 (Taiwan)

7.9

8.7

2.7

3.5

15.6 (Temperate climate) 52–208 (Temperate climate) 62.4 (Temperate climate) 104 (Temperate climate) 4.3–6.4 (India)

7.1

328

O. Osibanjo

Table 3. Chlorinated hydrocarbon levels in some Nigerian farm land, industrial and refuse

dump soils Pollutant

Lindane Aldrin p,p-DDE Total DDT Heptachlor Dieldrin Endosulfan PCBs

Concentration (ng/g) dry weight Farm land soils

Industrial soils

Refuse dump soils

8.7 (ND-0.5) ND 7.9 (ND-60) 2 (ND-11) 6 (3–43) – – ND

8.6 (ND-13.7) ND 32 (ND-127) 195 (4–774) 8 (ND-56) 11 (ND-28) – 122 (ND-740)

135 (ND-712) 104 (9–630) 57 (4–204) 201 (ND-530) 84 (ND-352) 41 (7.4–159) 16 (ND-60) 16 (ND-60)

N.B: – implies no data, ND=non-detectable.

dump soils had the highest concentrations (ng g–1) of OCs with the mean and range as follows: lindane 135 (ND-712), aldrin 104 (9–630), DDE 57 (4–204), total DDT 201 (ND-530), heptachlor 84 (ND-352), dieldrin 41 (7.4–159), endosulphan 16 (ND-60) and PCBs 1,141 (353–3311).

5 Organochlorines Levels in Nigeria and the African Environment Until the late 1980s data on organochlorine contaminants in environmental media in Africa were lacking. In the GEMS/WATER programme report [24]; the absence of data on organochlorines levels in African waters was emphasised. The lack of data on the levels of these substances in coastal and marine fish was also underscored at a pollution monitoring workshop [25]. However UNEP’s Regional Seas Programme for the West and Central Africa (WACAF 2) and East Africa (EAF) provided the vehicle for baseline data generation in the Africa region in the 1980s up to early 1990s on OCs in marine biota and sediments. Many chlorinated insecticides have been used in the Africa region for over half century in agriculture, vector control and public health as earlier mentioned. A review of chlorinated hydrocarbon substances in the African aquatic environment was published in 1994 by FAO under the aegis of its Committee for Inland Fisheries of Africa (CIFA) [6]. Most of the monitoring studies have been carried out to assess the ecological impact of specific groups of OCPs which have been used, in particular DDT and isomers, lindane, endosulfan, dieldrin, aldrin and heptachlor. Only in a few isolated cases have industrial chemicals such as PCBs and HCB been determined. Most of the data available indicate that OCs have been detected and quantified in different compartments, i.e., water, sediments, plants and fish, only in a few countries namely Nigeria, Ivory Coast, Kenya, Egypt and South Africa. In other countries of the region studies of OCs have focused on fish and sediments. Nigeria is the most populous country in Africa, with fairly active agricultural and vector disease control activities involving the use of OCs. It is one of the

Organochlorines in Nigeria and Africa

329

most industrialised countries in the continent with about 70% of the manufacturing industries including electric power generating stations, petroleum refineries, pulp and paper mills, located in the coastal zone. Nigeria participated actively in the contaminants monitoring program. under WACAF 2. Opportunistic studies were carried out on the levels of OCs in surface waters, soils, agricultural food crops, mosses, mango leaves, oil seeds, foodstuffs of animal origin, wildlife, fish, human blood and mothers’ breast milk to generate much needed baseline data in the country. 5.1 Concentration of Organochlorines in Water

Generally, African inland waters are contaminated by a broad spectrum of OCs. However, HCH, aldrin, endosulfan, p,p-DDE, p,p-DDT and total DDT were not detectable in many Nigerian inland water samples as shown in Table 4 despite fairly large scale use of these chemicals. Perhaps in the hot tropical climate, volatilisation of these chemicals and atmospheric transport are responsible for the trace concentration of these chemicals in water. The concentrations in water can also be correlated with water solubility and persistence in water of these chemicals. The occurrence and levels of chlorinated hydrocarbons in water of 17 rivers, 2 lakes and 1 dam in southern Nigeria had been studied [26, 27]. The overall range of values (ng/L) of the major OCs found were lindane ND-1 67, aldrin ND-1 90, endosulfan ND-750, HCB ND-9.2, heptachlor ND-96 and PCBs ND-8,991, respectively. DDT and metabolites have not been detected. Table 4 indicates the levels of these substances in some specific rivers. For example, concentrations (ng L–1) in River Ogun which traverses three states and discharges into Lagos Lagoon were: lindane 1.4–41.9 (13.3), aldrin 5.1–49 (40), endosulfan ND-260 (116), heptachlor ND-0.8 (0.25) and PCBs ND-224 (87), respectively. The occurrence and levels of OCs in 9 rivers in Ondo State, a major cocoa growing area of Nigeria were also studied [28]. The author detected (in ng/L): lindane ND-6.4 (2.4), heptachlor ND-5.0 (2.1), endrin ND-21 (5.1), aldrin ND-3.5 (1.0) and dieldrin ND-2,150 (1,062). PCBs, DDT and metabolites were not detected (Table 4). The detection of 10 OCs residues including PCBs in surface waters in Ibadan, the largest city in the sub-region were also reported [29]. The concentration ranges in ng L–1 of some of the compounds quantified are, a-HCH 1–302, lindane 7–297, aldrin ND-40, dieldrin 17.8–657, endrin ND-19, heptachlor 4–202, endosulfan ND-430, HCB ND-92 and total DDT ND-1,266; PCBs were detected but not quantified (Table 4). These results show higher loads of OCs in the water bodies compared to concentrations elsewhere. This study confirms that organochlorine pesticide residues are widely distributed in the surface waters studied, even at sites remote from point sources The presence of pesticide residues in Lagos Lagoon water has been confirmed with concentrations in ng L–1 as: lindane 85.3; aldrin 1 9.3; DDE 12 ; HCB 1.9; endrin 1 2.5 and dieldrin 28.0 [30]. Considering the OCs levels in Nigeria with the rest of the continent, the OCs with the lowest concentrations are generally heptachlor (ND-11) and HCB if we leave out single high values. Lake Mariut in Egypt appears the most polluted wa-

Dieldrin

East Africa – Kenya Lake Nakuru Southern Africa Hartbeespoort Dam, RSA Voelvlei Dam, RSA L. Mcllwaine, Zimb. L. Mcllwaine, Zimb.

North Africa – Egypt Lake Mariut Lake Mariut Nozha Hydrodrome

<100 <100 <100 200 (<10–530)

<100

West and Central Africa – Nigeria Ibadan, Streams 250 (17.8–657) River Ogun River Imo Cross River Awba Dam, Ibadan Kainji Lake R. Ero Dam, Ondo 560 River Ero, Ondo 740 River Osse, Ondo 2150 R. Owesse, Ondo 1120 River Apomu, Ondo 1380

INLAND WATERS

Location/water type

HCB

100 (26–27)

<100

120

1.0 (ND-5) 0.8 (ND-2.5)

150 (1–302) 17 (ND-92)

a-HCH

2091 1310 1100

100 (7–297) 13.3 (1.4–41.9) 0.2 (ND-0.6) 0.3 (ND-1.1) 61 (9–167) 0.12 (ND-0.24) 2.0 2.0 2.0 6.4 4.8 0.47 (ND-3.84) 3.3 ND 5.0 1.6 4.6

72 (4–202) 0.25 (ND-0.8) 4 (ND-11.4) 2 (ND-8.6)

g-HCH (Lindane) Heptachlor

Table 4. Concentration of chlorinated hydrocarbons in African and coastal waters (ng l–1)

100 (<10–120)

20 (ND-40) 40 (5.1–49) 13 (ND-40) 36 (ND-143) 18 (12–29) 0.55 (ND-3.05) 3.1 ND ND ND 3.5

Aldrin

ND ND ND ND ND

98 (ND-430) 116 (ND-260) 13 (ND-41) 20 (ND-80) 20 (ND-30)

Endosulfan

330 O. Osibanjo

Dieldrin

Lake Mariut Lake Mariut Nozha Hydrodrome Nozha Hydrodrome

North Africa – Egypt

Ibadan, Streams River Ogun River Imo Cross River Awba Dam, Ibadan Kainji Lake R. Ero Dam, Ondo River Ero, Ondo River Osse, Ondo R. Owesse, Ondo River Apomu, Ondo

West and Central Africa – Nigeria

INLAND WATERS

Location/water type

North Africa – Egypt Abu-Kir Bay

4493 6630 1540

ND ND ND ND ND

p,p¢-DDE

West and Central Africa – Nigeria Lagos Lagoon 8 (ND-24)

COASTAL WATERS

Location/water type

Table 4 (continued)

4.88

a-HCH

ND ND ND ND ND

p,p¢-DDD

2 (0.8–4.1)

HCB

134 9820 600

ND ND ND ND ND

p,p¢-DDT

18.05

182 (16–634)

13610

5100 21440

3.88 (ND-10.05) ND ND ND ND ND

310 (ND-1266)

Total DDT

ND

g-HCH (Lindane) Heptachlor

29 (ND-86)

Endosulfan

ND ND ND ND ND

87 (ND-244) 121 (ND-241) 120 (ND-470) 330 (ND-1000)

PCB

53 (3–190)

Aldrin

Organochlorines in Nigeria and Africa

331

ND = Not detected.

North Africa – Egypt Abu-Kir Bay

West and Central Africa – Nigeria Lagos Lagoon

COASTAL WATERS

83 (ND-344)

<100

100

3 (ND-15)

100 <100

<100

p,p¢-DDD

100 <100

<100

East Africa – Kenya Lake Nakuru

Southern Africa Hartbeespoort Dam, RSA Voelvlei Dam, RSA Lake Kariba, Zimbabwe Lake Kariba, Zimbabwe L. Mcllwaine, Zimbabwe L. Mcllwaine, Zimbabwe

p,p¢-DDE

Location/water type

Table 4 (continued)

ND

<100

100 <100 (<20–300)

<100

p,p¢-DDT

24.3

2,500 (ND-9,000)

400 (30–700)

350 (<10–700)

300 <200

Total DDT

<1,000

2,000 <1,000

<1,000

PCB

332 O. Osibanjo

333

Organochlorines in Nigeria and Africa

ter body, based on exceedingly high values for) g-HCH (lindane) of 1,310 ng L–1, for p,p′-DDE of 6,630 ng L–1 and for total DDT of 21,440 ng L–1 [31]. Lake Nakuru in Kenya, Lake Mcilwaine in Zimbabwe and the South African reservoirs are contaminated by PCBs indicating pollution from industries as source of contamination.According to data from Nigeria [27, 30] and Egypt [32] coastal waters are just as contaminated as freshwaters except for the Egyptian lakes which have very high values. Furthermore concentrations above 100 were found for γ-HCH, DDE and total DDT in Egyptian lakes, lbadan streams, Nigeria (DDT only) and PCBs in Hartbeespoort Dam, South Africa, respectively, indicating pollution by these chemicals. These results are similar to level of DDT reported in Colombia by GEMS/WATER and the g-HCH values reported for some major rivers in the United Kingdom [33], Japan [34] and India [35]. However, the majority of the African inland waters have g-HCH concentrations below 10 ng L–1, which is the recommended guideline for the protection of freshwater aquatic life in Canada [36] similar to the results reported for Niagara River, Canada [37]. The problem of gross contamination of ground water by organochlorine pesticides has also been identified in parts of Nigeria [38]. Table 5 shows that the mean concentrations of total DDT and heptachlor exceed the WHO limits for these chemicals in drinking water. The disinfection of water by chlorination which produces trihalomethanes (THMs), which are potential carcinogens, is a common practice in Nigeria and most of the other African countries. Table 6 shows the concentration of individual and total THMs in tap water from some Nigerian towns [39]. Of the THMs, only bromoform was not detected at all and only Ibadan water samples contained the three remaining THMs. Tap water samples from Ibadan, Lagos, Oyo town, Ilorin and Kaduna contained detectable levels of THM. The highest THM value of 220 mg L–1 detected in Ibadan is higher than the permissible limits in Germany (25 mg L–1), Holland (75 mg L–1) and USA (100 mg L–1) but lower than permissible limits in Canada (340 mg L–1) and Britain, respectively. The concentrations of THMs in other African countries are not readily available for comparison. Table 5. Chlorinated hydrocarbon concentration (mg 1–1) in ground water, Ibadan, Oyo State,

and Nigeria Chlorinated hydrocarbon

Concentration range; mean

% occurrence in samples

WHO limit (mg 1–1)

Aldrin BHC Lindane Total DDT Heptachlor Endosulphan

ND-0.367; 0.0190 ND-0.1250; 0.0211 ND-0.1866; 0.0184 0.2791–11.89; 2.200 0.2890–17.50; 2.256 ND-1/4; 0.318

15 95 25 100 100

0.03

N.B First figure is range, followed by mean.

3.0 1.3 0.10

334

O. Osibanjo

Table 6. Concentration (mg 1–1) of individual and total trihalomethanes in tap water from some Nigerian towns

Sample location

Chloroform CHCl3

BromodiDibromoBromo- Total chloromethane, chloromethane form TrihaloCHBr2Cl CHBr3 methanes CHBrCl2

Ibadan (Oyo State) Lagos (Lagos State) Oyo Town (Oyo) Ilorin (Kwara State) Offa (Kwara State) Kagara (Niger State) Kaduna City (Kaduna State) % Positive occurrence

ND-12.6; 62.8 ND ND ND ND ND ND

5.3–55.7; 22.8 ND-8.8 ND-8.4 2.5 ND ND ND

ND-43.1; 11.71 ND-2.8 ND-2.8 ND ND ND ND

ND ND ND ND ND ND ND

220 11.6 11.2 2.5 ND ND ND

31.6%

62.2%

47.4%

0%

68.47%

Permissible levels in: W. Germany Holland USA Canada Britain

25 75 100 340 341

5.2 Concentrations of Organochlorines in Sediments

Apart from a few hot-spots due to localised pollution problems, African inland water sediments are relatively unpolluted by OCs and the values obtained are in most cases lower than concentrations reported for developed countries [40, 41]. Table 7 indicates the contamination level of sediments in Nigeria compared to other African countries. The concentration range (mean) in ng g–1 dry weight of 20 sediments samples analysed from streams and rivers in Ibadan city, Oyo State were: dieldrin ND-6 (1.4), a,b-HCH ND-1.6 (0.2), g-HCH ND-2 (0.3), aldrin ND-0.04 (0.002), DDE ND-50 (6.8) and PCBs ND-14 (1.8); heptachlor, endosulfan and endrin were not detected. The occurrence and levels of OCs in 23 bottom sediment samples from Lekki Lagoon in Lagos State have also been investigated. Eleven organochlorine pesticides and HCB were detected. PCBs were not detected (Table 7). The ranges of concentration with means in parenthesis, reported in ng g–1 dry weight were: lindane 0.1 1–4.9 (1.1), aldrin ND-347 (56), p,p-DDE 11–555 (263), o,p′DDD ND-348 (88), endosulfan 7–1,1 55 (30), heptachlor ND-1 845 (64),6-HCH, ND-260 (66), A-HCH ND-1 1 6 (18.6), HCB ND-3.3 (0.4), endrin ND-1 29 (1 6.5), dieldrin 1 90–8,460 (4,560). Compared to other parts of the world, the sediments of Lekki Lagoon are to be considered fairly contaminated with organochlorine pesticides.

5.0 (ND-1.6)* 4 2 <1

0.09

COASTAL WATERS North Africa – Egypt Abu-Kir Bay

p,p¢-DDE

13 (ND-32) 7 (ND-14) 9 (ND-50) 263 (11–555)

West and Central Africa River Ogunpa, Ibadan, Nigeria River Ona, Ibadan, Nigeria River Oniyere, Nigeria Lekki Lagoon, Nigeria

16 (2–42)*

Location/water type

10 (2–39)

HCB

ND ND ND ND

p,p¢-DDD

0.7 (ND-2) 0.5 (ND-0.9) 0.1 (ND-0.4) 18.6 (ND-116)

a-HCH

0.1 (ND-5.3) 1 (ND-2) ND 88 (ND-438)

p,p¢-DDT

0.05

15.7 (0.07–62.1)

0.9 (ND-6.8)

Total DDT

ND ND ND 30 (7–11155)

Endosulfan

ND ND 4 (ND-14

PCB

1.0 (ND-12)*

131 (ND-251)

ND ND ND 56 (ND-347)

Aldrin

ND ND ND 64 (ND-1845)

g-HCH (Lindane) Heptachlor

0.5 (ND-1.2) ND 0.9 (ND-2.0) 0.4 (ND-3.3) 1.1 (0.11–4.9) 2.3 (0.5–19) 17.8 (ND-125.8) 3.2 (0.01 -13.4) 3.0 (0.07–19.8) <1.0 4 (3–6) 1 (ND-4)

0.9 (ND-1.8) 0.3 (ND-0.5) 2.0 (ND-6) 4560 (190–8460)

Dieldrin

INLAND WATERS West and Central Africa River Ogunpa, Ibadan, Nigeria River Ona, Ibadan, Nigeria River Oniyere, Nigeria Lekki Lagoon, Nigeria Ebrié Lagoon, Ivory Coast Ebrié Lagoon, Ivory Coast Lake Nakuru, Kenya Lake Nyumba ya Mungu, Tanzania (Tanzardin) Lakes, Uganda Southern Africa Lake Mcllwaine, Zimbabwe Lake Mcllwaine, Zimbabwe Hartbeespoort Dam, RSA Voelvlei Dam, RSA

Location/water type

Table 7. Concentration of chlorinated hydrocarbons in African inland and coastal waters sediments (ng g–1 dry weight) [6]

Organochlorines in Nigeria and Africa

335

10 1 (ND-1)

15 10 5 (5–15)

East Africa Lake Nakuru, Kenya Lake Nyumba ya Mungu Tanzania

Southern Africa Lake Mcllwaine, Zimbabwe Lake Kariba, Zimbabwe Lake Mcllwaine, Zimbabwe Hartbeespoort Dam, RSA Voelvlei Dam, RSA

* ng g–1 wet weight. ND=Not detected.

North Africa – Egypt Abu-Kir Bay

COASTAL WATERS

840.5 456.0

7.4 (0.1–149)

p,p¢-DDE

West and Central Africa Ebrié Lagoon, Cote d’Ivoire Ebrié Lagoon, Cote d’Ivoire Nozha Hydrodrome Nozha Hydrodrome

Location/water type

Table 7 (continued)

40 18 2 (2–40)

10 1 (ND-3)

28.1 (0.2–803)

p,p¢-DDD

2 13 6 (2–13)

<10 3 (2–7)

91

15.7 (0.2–354)

p,p¢-DDT

1.73

76 (32–146)* (40– 740) 57 45 13 (13–740)

17.1 (1.1–997) 46.2 (2.5–242.8)

Total DDT

120 320 70 (70–320)

<20

46.7 (2–213) 355.5 (8.5–1014)

PCB

336 O. Osibanjo

Organochlorines in Nigeria and Africa

337

5.3 Concentrations of Organochlorines in Fish

Fish samples from fresh water were found to contain significantly higher concentration of these chemicals than sediments and water . In Nigeria the detection of 10 organochlorine pesticides, HCB and PCBs in 40 freshwater fish samples collected from various locations in Oyo and Ogun States had been reported [42]. The relative occurrence of some of the OCs identified were lindane 100%, PCB and endosulfan 97%, DDT and metabolites 75%. The concentration ranges with mean in parenthesis in ng g–1 fresh weight were (Table 8): lindane 7–106.0 (25.6), p,p-DDE 2.0–30.0 (3.4), p,p′-DDD 2.0–60.0 (7.8), p,p′-DDT 3.0–18.0 (2.9), total DDT 3.3–161 (20.6), PCBs (Aroclor A1250) 8.0–130 (28.7), heptachlor 1.0–300 (50.0), endosulfan 3–904 (173), HCB 9.0–130.0 (12.7) and a-HCH 0.2–5.0 (1.3) (Table 8). In an earlier study in 1980 [43], DDT concentrations of 0.08 to 4.4 mg/g, g-HCH 0.08 to 4.4 mg g–1 and HCB 0.06 to 0.60 mg g–1 in fish from Oyo State Nigeria had been found. The very high values in this study was attributed to suspected cases of deliberating killing fish with lindane and DDT insecticides for human consumption. The detection and quantitation of 9 OCs in southeastern Nigeria has also been reported [44]. The relative occurrence of some of these compounds were PCBs, aldrin, lindane, and a,b-HCH 100%, endosulfan, p,p′DDD, p,p′-DDE and heptachlor 44.4%, 33.3%, 61.1% and 72.2%, respectively. The concentration ranges with means in parenthesis in ng g–1 fresh weight were: a-HCH 0.2–7.4 (1.8), lindane 0.6–13 (4.4), heptachlor ND-1.0 (0.3), aldrin ND-14.9 (5.5), endosulfan ND-89.6 (14), p,p′-DDE ND-4.2 (1.8), DDD ND-8 (0.7), and PCBs 0.7–14 (3.8). Contamination of marine fishes to a lesser extent than fresh water fishes had been reported in Nigeria [45] based on the analyses of 94 samples of 25 marine fish species over 1983–1985 and 14 samples of 7 shellfish species in 1987.The concentration ranges in fresh weight were found to be for HCB 0.04–9.48, lindane ND-5.30, endosulfan ND-4.95, DDT 0. 1 5–18.6, aldrin ND-54.60 and for PCB 1 1.0–225 (Table 9). Fish contained higher concentrations of aldrin, heptachlor, HCB and lindane than shellfish, while the reverse was observed for DDT and PCBs. The concentrations of residues obtained were found to be lower than those reported in literature for industrialised countries. Furthermore, predator fish species were found to concentrate more residues in muscle tissue. Table 9 further indicates that inland water fish are highly contaminated by DDT and metabolites and lindane which reflect the heavy usage of these chemicals. In particular, fish from Kenyan rivers [46] and even more from the South African reservoirs are heavily contaminated with levels of 370 ng g–1 for DDT (Vodiviei Dam) and 920 ng g–1 fresh weight for PCBs (Hartbeespoort Dam). Data for OCs levels in marine fish (Table 8) are available only for West and Central Africa. In general, marine fish are less contaminated than freshwater species. Data for marine fish from Cameroon, however, show high values for DDTs and PCBs. in plankton feeders. The DDT/PCB ratios were less than 1, indicating a predominance of industrial activities over agricultural activities as the source of contamination of the marine environment. The fish Galcoides decadactylus was proposed as a potential bio-indicator organism for chlorinated hydrocarbon pol-

Dieldrin

5 (2–27)

3*

East Africa – Uganda Lakes

East Africa – Tanzania Lake Nyumba ya Mungu

East Africa – Kenya Lake Nakuru 7* Lake Nakuru 1.5–2.4 Tana River (Masinga Dam) Tana River (Hola Irrigation Scheme) Lake Victoria 10 Lake Victoria (7–70) Lake Victoria

North Africa – Egypt Lake Mariut Lake Mariut

West and Central Africa – Nigeria South West (Ogun and Oyo States) Southeast (Cross River and Akwa Ibom States) Ibadan 68 (ND-173) Oyo, Lagos and Cross River States

Country/Location

1*

(1–7)

(1–40) 0.83 (0.11–3.08)*

91.4 (4–295)

34.98 80.06

14 (9–21)

2.03

248 (0.2–598)

0.15 (0.04–0.24) 1.8 (0.2–5.0)

25.6 (7–106)

g-HCH (Lindane)

4.4 (0.6–13)

12.7 (9–130)

HCB

1.8 (0.2–7.4)

a-HCH

0.5 (0.1–1.3)

0.3 (ND-1.0)

50 (1–300)

Heptachlor

Table 8. Concentration of chlorinated hydrocarbon residue in fish from African inland waters (ng g–1 fresh weight) [6]

20

1.4 (0.1–3.8)

5.5 (ND-14.9)

Aldrin

20 (ND-110)

ND

14 (ND-89.6)

173 (3–9904)

Endosulfan

338 O. Osibanjo

(10–146)

HCB

<13* <11

90.4 (85–1185)

171.9 (52–1125)

58 (6–184) 60.76 84.5

8 (5–14) 17.36 31.8 <3* <2

2950 (270–16,000)

1470 (180–8200)

780 (50–4200)

3* <7

15 (0.5–36)

7 (ND-23)

ND ND

Aldrin

3 (0.2–6.6)

20.6 (3–161)

Total DDT

Heptachlor

2.5

2.9 (3–18)

p,p¢-DDT

g-HCH (Lindane)

0.7 (ND-8)

7.8 (2–60)

p,p¢-DDD

64.1 (ND-240)

a-HCH

West and Central Africa – Nigeria South West (Ogun 3.4 (2–30) and Oyo States) Southeast (Cross River and 1.8 (ND-4.2) Akwa Ibom States) Ibadan North Africa Gezira Research Farm, 670 (20–3600) Sudan Lake Nubia, Sudan 48 (1–153) Lake Mariut 38.96 Lake Mariut 10.1 East Africa – Kenya Lake Nakuru 7* Lake Nakuru 1.3–2 Lake Nakuru 25* Lake Naivasha 46 Lake Naivasha 2–5* Lake Baringo 14–48* Tana River 121.4 (15–220) (Masinga Dam)

p,p¢-DDE

(0.3–9) 1.3 (ND-24) 10–40* 80* 27*

Southern Africa Khwai River, Botswana L. Mcllwaine, Zimbabwe L. Mcllwaine, Zimbabwe Hartbeespoort Dam, RSA Voelvlei Dam, RSA

Location/water type

Dieldrin

Country/Location

Table 8 (continued)

<150*

3.8 (0.7–14)

28.7 (8–130)

PCB

30 (ND-190)

Endosulfan

Organochlorines in Nigeria and Africa

339

100* 20*

77* 160*

67* 190*

<3–47*

48 (5–125)*

20 3.0 (0.17–7.4)*

3 (ND-10)

p,p¢-DDT

* Original data on dry weight basis, converted to fresh weight values by dividing by 3. ND = Not detected.

33–60*

27–80*

Southern Africa Khwai River, Botswana L. Mcllwaine, Zimbabwe L. Mcllwaine, Zimbabwe Lake Kariba, Zimbabwe Hartbeespoort Dam, RSA Voelvlei Dam, RSA

2*

(10–20) 7.8 (2.67–16.03)*

10 (3–360) 4.4 (0.43–10.9)*

5*

8 (ND-40)

p,p¢-DDD

247 (ND-1300)

p,p¢-DDE

East Africa – Tanzania Lake Nyumba ya Mungu Lake Tanganyika

East Africa – Kenya Tana River (Hola Irrigation Scheme) Athi River Lake Victoria Lake Victoria Lake Victoria Lake Victoria

Location/water type

Table 8 (continued)

(19–128) 66.6 60–150* 170 244* 370*

8* 165 (50– 330)*

36 25 (90–460) 15.2 (9–29)* 3.0

258 (ND-1350)

Total DDT

920* 600*

400–770*

90.4 (20–332)*

PCB

340 O. Osibanjo

3.72 (0.13–14.70) 15 (2–36) 0.23 1.79

0.12 (ND-1.05) 11 (2–30) 1.86

p,p¢-DDT

0.37 (0.17–0.9)* 1.08

0.31 (<0.03–0.73)* 0.74 0.18 0.98 (0.28–1.76) 1.44 (ND-5.3)

3.07 1.74

p,p¢-DDD

1.35 (ND-4.16)

2.41 (ND-21.0)

0.16 (ND-4.95)

Endosulfan

196 (ND-983) 94.5 (37–287) 342 (ND-705) 209 (ND-716)

37.0 (4.73–152) 1.0 (0.17–1.9)* 244 (76–540) 113 (ND-181)

40.9 (11.0–225) 90 (3–825)

PCB

ND 1.71 (ND-12.0)

0.65 (0.17–1.3)*

0.52 (ND-1.94)

2.4 (ND-13.3)

<0.006 0.18 (0.03–0.7)*

2.85 (ND-54.60)

Aldrin

4.37 (0.15–18.60) 46 (7–116) 3.88 1.92 (0.13–4.3)* 89.5 (ND-393)

Total DDT

0.02 0.1 (<0.6–0.23)* 0.15

0.80 (ND-1.69)

0.58

<0.016

0.22 (ND-0.80)

1.29 (ND-21.40)

Heptachlor

0.83 (ND-5.30) 12.4 (2–55) 0.10 0.41 (<0.03–0.8)* 0.029 1.60 (ND-7.31)

g-HCH (Lindane)

0.92 (0.04–9.48)

HCB

* Original data on dry weight basis, converted to fresh weight values by dividing by 3. ND=Not detected.

West and Central Africa Finfish Nigeria Sierra Leone Benin Cote d’Ivoire Cameroon Shellfish Nigeria: shrimp, crab, oyster, snail Cote d’Ivoire: shrimp Cameroon: shrimp Cameroon: shrimp

Country/Location/ Species

p,p¢-DDE

Dieldrin

West and Central Africa Finfish Nigeria Sierra Leone Benin 0.29 Cote d’Ivoire 0.46* Gambia Cameroon Shellfish Nigeria: shrimp, crab, oyster, snail Cote d’Ivoire: shrimp Gambia: shrimp Gambia: oyster Cameroon: shrimp Cameroon: shrimp

Country/Location/ Species

Table 9. Concentration of chlorinated hydrocarbons residue in fish from coastal and marine African waters (ng/g fresh weight)

Organochlorines in Nigeria and Africa

341

342

O. Osibanjo

lution monitoring in the study area in view of the high correlation between OCs concentrations in tissues, fish weight, length and fat [47]. Table 9 further provides an overview of OCs in Nigerian fish compared to fishes from other African countries. Data for OCs levels in marine fish (Table 9) are generally available only for West and Central Africa. In general, marine fish are less contaminated than freshwater species. Data for marine fish from Cameroon, however, show high values for DDTs and PCBs. Not the least the levels of OCs in fish from African waters are relatively low compared to values from other parts of the world and are similar to results obtained by GEMS/FOOD for 1984–1985 [48] except in a few cases around pollution hot-spots. The levels found are also much lower than the permissible limits for OCs in fish for human consumption. 5.4 Concentrations of Organochlorines in Foods

Ingestion of contaminated foods is a major source of human exposure to OCs. Hence in Nigeria exposure to OCs through dietary source had been established by the collection of foodstuffs: 217 fruit and vegetable samples; four major cereals (rice, maize, sorghum and soyabean) as well as foodstuffs of animal origin from different locations in the country and analysing them for the presence and levels of the OCs [49, 50–52]. Table 10 shows the summary of mean concentrations of the OCs residues in Nigerian foodstuffs. Most samples had maximum residue levels below the FAO’s maximum residue limits (MRL). However some samples of meat, cereals and pulses had DDTs and aldrin+dieldrin levels above the MRL. Thus meat and pulses form the greatest sources of human exposure. The high levels obtained for DDT and aldrin+dieldrin could be correlated with house treatment with DDT for malaria control and aldrin +dieldrin with house treatment for termite control, respectively. The study further established that the dietary intakes of HCHs, DDTs, aldrin and dieldrin came predominantly from tubers, pulses and cereals. The dietary intakes (ADI) of aldrin and dieldrin in Nigeria were estimated for the first time and found similar to ADI values for India [53] but higher than FAO/WHO [54], Japan and European countries ADI values. But ADI values for DDTs and HCHs were below the acceptable daily intake (ADI) of the FAO and those of the developed countries (Table 11). Additional dietary exposure of humans to OCs was also established through the determination of these compounds in hospital total diets (Table 12) and infant foods marketed (Table 13) in the country. Although the food samples were contaminated with OCs, the maximum residue levels were below FAO’s MRL and the standards set by some European countries. Children are not exempt from risk due to exposure to OCs as infant foods marketed in the country are contaminated by these chemicals (Table 11). The OCs concentrations are however much lower than values reported for European countries.

1000

100 e

*

5.8 (1.3–16.4) 6.9 (2.2–22.6) ND

1.6 (1.2–6.2)** 1.5 (1.2–120) (1.2–120) 17.4 (1.9–13.6) 1.9 (1.6 —18.3) *

Fruits

*

4.7 (1.1–13.1) 28.3 (4.3–49.8) ND

1.4 (1.2–1.8) 2.9 (1.2–120) (1.2–120) 5.7 (2.2–13.2) 2.1 (1.6–4.6) *

Vegetables

ND

10.1 (2.0–32.0) 14.0 (10–46) 8.0 (4.0–18) 32.0 ) (20–67) 12.0 (4.0–21) 30.4 (8.0–44.0) ND

ND

Tubers

20 f

100

20 e

500

MRL

8.0 (1.0–51) 17.0 (1.0–123) 8.0 (2.0–30) 45 (6.0–410) 29 (2.0–126) 81 (5.0 0 410) 9.5 (2.0–18) 18 (2.0–33)

ND

Cereals b

108 (11–402) 203 (30–428) 36 (25–51) 208 (58–290) 113 (28–189) 189 (103–334) 156 (38–320) 156 (85–330)

5.0

Pulses

ND

35 (30–70) 2,000 50 (30–110) 200 e 28 (15–50) 312 (20–2160) 106 (15–233) 5000 164 (30–302) 200 f ND

ND

Meat MRL Cow

* destroyed by acid clean-up process. a Maximum Residue Limits (FAO/WHO, 1986). b Osibanjo & Adeyeye (1995). c Highest of the means in the liver, kidney, heart and muscle of each animal (Osibanjo & Adeyeye, 1997). d Osibanjo & Bamgbose, 1990 (mean figures from Table 2, 1985 results). e aldrin + dieldrin. f heptachlor + heptachlor epoxide. ** Range in bracket. ND = below detection limit.

Heptachlor expoxide

Heptachlor

Total DDT

DDE

Dieldrin

Aldrin

Total HCH 2000

Lindane

HCB

MRL a

Table 10. Overall mean concentration (µg kg–1) of the organochlorine residues in Nigerian foodstuffs

ND

226 (40–450) 244 (62–625) 70 (20–190) 337 (44–1420) 374 (44–890) 510 (140–960) ND

ND

Pig

ND

54 (32–82) 61 (4–112) 14 (4.0–30) 145 (62–640) 90 (10–224) 141 (10–556) ND

ND

Goat

*

3.60

4.80

–

*

–

–

1.77

2.31

Fish d

Organochlorines in Nigeria and Africa

343

344

O. Osibanjo

Table 11. Estimated daily intake of HCHs, aldrin+dieldrin and DDTs by Nigerians in compar-

ison with those of some other countries and the ADI of the FAO/WHO Pesticide

Nigeria

India

USA a

Japan

FAO/WHO a

HCHs Aldrin & Dieldrin DDTs

13.1± 8.2 18.0±10.8 28.8±15.3

155 19.0 48.0

0.17 0.29 1.6

0.47 0.05 3.1

600 b 6.0 1200

a

Values converted from mg kg–1 body weight/day to mg/person/day, using an average body weight of 60 kg/person, for ease of comparison. b Value for gamma-HCH alone. Sources: India (Kannan et al, 1992), Japan (Matsumoto et al, 1988), USA (FDA, 1989), FAO/WHO (1986).

Table 12. Concentration of chlorinated hydrocarbons in hospital diets

Pollutant

Concentration (mg kg–1 fat weight)

% occurrence

Lindane Aldrin Heptachlor Endosulphan Hexachlorobenzene p,p-DDE Total DDT PCBs

0.70 (0.04–1.79) 0.33 (ND-1.64) 0.06 (ND-0.94) 0.11 (ND-1.25) 0.14 (0.02–0.20) 0.17 (ND-0.70) 0.29 (ND-1.16) 0.02 (ND-0.18)

100 76 19 0 100 71 81 19

Table 13. Chlorinated pesticides, range and mean concentrations in infant food marketed in

Nigeria Concentrations ng g–1 Nigeria a,b-BHC Lindane Aldrin Dieldrin p,p-DDE Total DDT Heptachlor

0.10 (ND-0.89) 0.43 (ND-2.30) 0.06 (ND-0.13) 2.25 (0.08–8.60) 0.14 (ND-2.60) 0.59 (ND-5.50) 0.09 (ND-0.87)

Malawi

Italy 5.0 (ND-24.30) 4.30 (ND-21.30)

55.00 (40–70) 120 (100–140)

66.60 (ND-737) 110.25 (ND-782) 9.80 (ND-72.40)

Germany 1.55 (1.3–3.50)

Organochlorines in Nigeria and Africa

345

5.5 Concentrations of Organochlorines in Nigerian Wildlife

Wildlife is a major source of foreign exchange earning for some African countries where tourism has been well developed especially in East and Southern African countries and to some extent in Nigeria. Some classes and species are, however, important sources of animal protein for the rural population and a delicacy for the urban dwellers in some countries like Nigeria. Their socio-cultural importance is exemplified with wild life being important ingredients of traditional medicinal preparations and in witchcraft in the African traditional society. There is a paucity of data on OCs levels in wildlife in Africa. In view of the proven ecotoxicological hazards of OCs usage in the developed countries and concerns that Nigerian wildlife resources have been dwindling with many animals and birds becoming extinct, the only baseline study on OCs levels in Nigeria thus far [55] indicated the relative occurrence of the OCs detected in tissues of Nigerian wildlife to be DDEpp (88.7%), BHC (73.6%), endosulfan (64.1%), heptachlor (52.8%), HCB (34.0%) and aldrin (17.0%), respectively. The concept that the OCs body burden depends on the feeding habit is supported by the fact that carnivores and birds concentrate more OCs in their tissues than herbivorous wildlife in the study. The highest concentrations (µg g–1 fresh weight) of total DDT of 1.98 mg g–1, 1.27 mg g–1 and 1.25 mg g–1, respectively were found in the heart; muscle; and liver of owl. The heart (0.49 mg g–1), muscle (0.41 mg g–1), and kidney (0.09 mg g–1) of Saddle-bill stork, respectively, also contained the highest concentrations of g-HCH. The heart accumulated the highest concentration of organochlorine pesticides in most cases. (Table 14). The values are generally lower than residue values in wildlife from areas following intense pesticide spraying [56, 57] but higher than levels of OCs (DDE 3.8–46.5; 21.7; DDT 0.6–93.2; 48.6) in duck liver samples from South Africa [58]. However PCBs were detectable in 53% of the tissues analyzed. The highest concentrations of PCBs were found in the brain of vulture (0.52 mg g–1); owl (0.51 mg g–1) and cattle egret (0.38 mg g–1), respectively, whereas the organochlorine pesticides concentrations were highest mainly in the hearts and livers of the samples studied. Generally the DDT/PCB ratio is greater than 1 thereby implicating agricultural activities rather than industrial activities as the source of OCs contamination of the wildlife in Nigeria. The high levels of OCs obtained for some of the species might have adverse toxicological effects on the wildlife. The deleterious effects of OCs usage on wildlife are yet to be investigated in the country. 5.6 Organochlorine Concentrations in Human Breast Milk from Nigeria

The concentrations of organochlorine contaminants in human breast milk provide reliable indicator on the exposure of different populations to chlorinated hydrocarbon chemicals.Although data abound for developed countries, studies on organochlorines contaminant in human breast milk are few in Africa. A survey of OCs in human breast milk in 1986 [59] indicated that organochlorine pesti-

1.16 ND-1.9 100

ND – 0

91 – 67

2.6 ND-4.6 67

A B C

A B C

Lindane

Aldrin

Endosulphan A B C

Heptachlor

3

3

ND – 0

ND-19.7 – 67

70 55 120 635 118 ND-135 0.3–104 49–139 265–1340 67 100 100 100

ND – –

487 – 33

174 ND-524 33

63 ND ND-188 – 50 –

PCB

A B C

130 79 164 870 60 106–154 32–130 73–302 383–1784 32–96 100 100 100 100 100

TOTAL DDT A B C

A B C

2 ND-6.0 33

12 ND-13 67

8 ND-14 67

p,p-DDE

0.1 0.2 1.4 ND-0.30 ND-0.7 ND-4.2 33 33 33

11.3 6.7 ND-21 ND-20 67 67

ND ND 0

2

34 – 100

ND – 0

– – 0

90 ND 50

ND – 0

ND ND 0

2

40 Aug-72 100

3.5 ND-7 100

ND – 0

ND ND 0

2

Heart

51 27–71 100

ND – 0

58 13–44 100

ND – 0

30 ND-6 50

ND – 0

11.3 22.7 50

2

Muscle

7–530 13–89 100

ND – 0

29; 17 ND-34 50

190 ND-380 50

ND-44 50

7–513 100 22

– 100 260

– 0 269

52 19–39 ND

2

ND – 0

– 0 42

433 41–45 ND

ND-3 50

– 100 1.5

ND-1 50

8–92.0 100 0.5

8–75.0 7–596 50 100 33 50

– 0 119

ND

–

ND – 0

249 ND-487 50

2

Heart

ND – 0

460 ND-920 50

2

Brain

ND-315 50

11–741 100 158

– 100 376

– 0 301

ND – ND

– –

– – –

– –

– 0 –

– – ND

716 492 – 248–1184 257–727 – 100 100 –

ND – 0

38 95 ND-76 ND-189 50 50

2

Kidney Liver

Saddle bill stork

596 355 346 ND-1191 273–436 – 50 100 50

380 – 100

5.7 ND-11.4 50

2

Brain

195 694 56 148–242 322–1067 14–98 100 100 100

49 20–48 100

ND – 0

ND 29–87 100

273 11 ND-181 ND-22 50 50

ND – 0

ND ND 0

2

Kidney Liver

163 33 114–215 – 100 100

30 72–163 100

ND – 0

9 ND-18 50

11 ND-22 50

ND-22.2 ND – – 67 0

19.4 8.8–26 100

3

ND – 0

9.1 ND -25 67

8.4 17 108 ND-19.6 15–18 ND-242 67 100 67

ND – 0

1.4 3.9 4.9 0.4–2.5 ND-10.73.0–6.7 100 67 100

3

Brain

Hexachloro- A benzene B C

A B C

3

Number of Samples

Heart

Muscle

Liver

Muscle

Kidney

Cattle egret

Hooded Vulture

Table 14. Chlorinated hydrocarbon residue in tissues of some Nigerian wildlife (µg g—1) [55]

346 O. Osibanjo

A B C

B C

PCB

A

Hexachlorobenzene

A B C

A B C

Heptachlor

TOTAL DDT

100 187

A B C

Endosulphan

A B C

–

A B C

Aldrin

p,p-DDE

32 – 100

A B C

Lindane

– 100

– 100 100

– 100 211

– 100 23

22 – 100

ND – 0

1

– 100

– – 8

– 100 93

100 66

–

ND 0 8

ND – 0

ND – 0

ND – 0

1

– 100

– 100 23

– 100 216

100 134

–

ND 0 23

ND – 0

ND – 0

ND – 0

1

– 100

– 100 4

– 100 263

100 180

–

ND 0 17

ND – 0

74.8 – 100

143.8 – 100

1

Heart

Brain

– 100

– 100 16

– 100 30

100 27

–

ND 0 16

ND – 0

ND – 0

ND – 0

1

ND – 0

1270 – 100

1052 – 100

– 0

ND

ND – 0

81 – 100

ND – 0

0.7 – 100

1

42 – 100

664 – 100

640 – 100

– 0

ND

ND – 0

139 – 100

ND – 0

ND – 0

1

Kidney

Muscle

Liver

Muscle

Kidney

Owl Otus ireneae

African Hawk Eagle Hieratus bellicosus

Number of Samples

Table 14 (continued)

ND – 100

1254 – 100

892 – 100

– 0

ND

14 – 100

46 – 100

ND – 0

49.2 – 100

1

Liver

425 – 100

1985 – 100

1445 – 100

– 0

ND

31 – 100

ND – 0

ND – 0

ND – 0

1

Heart

512 – 100

133 – 100

1210 – 100

– 0

ND

23 – 100

ND – 0

ND – 0

3.9 – 100

1

Brain

ND – 0

2 – 100

2 – 100

– 100

9

ND – 0

23 – 100

2.7 – 100

2.3 – 100

1

33 – 100

2.3 – 100

2.4 – 100

– 0

ND

ND – 0

25 – 100

4.6 – 100

2.7 – 100

1

20 – 100

4.2 – 100

14 – 100

– 100

0.3

ND – 0

22 – 100

8.9 – 100

1 – 100

1

Muscle Kidney Liver

Monkey, Colubus polykmos

91 – 100

3.4 – 100

3 – 100

– 100

1

ND – 0

54 – 100

7.8 – 100

9.7 – 100

1

ND – 0

3 – 100

– 0

ND

ND – 0

18 – 100

2.8 – 100

2.7 – 100

1

Heart Brain

Organochlorines in Nigeria and Africa

347

348

O. Osibanjo

Table 14 (continued)

Squirrel Xerus erythropus

Number of Samples

Muscle

Kidney

Liver

Heart

Brain

1

1

1

1

1

Lindane

A B C

0.8 100

0.7 100

ND 100

1.5 100

5.3 100

Aldrin

A B C

ND – 0

ND – 0

ND – 0

ND – 0

ND – 0

Endosulphan

A B C

ND – 0

ND – 0

ND – 0

ND – 0

ND – 0

Heptachlor

A B C

14 – 100

7 – 100

ND – 0

3 – 100

ND – 0

Hexachlorobenzene

A B C

0.7 – 100

0.7 – 100

ND – 0

1 – 0

ND – 0

p,p–DDE

A B C

ND – 100

ND – 100

ND – 100

ND – 100

ND – 100

TOTAL DDT

A B C

ND – 100

ND – 100

ND – 100

ND – 100

ND – 100

PCB

A B C

ND – 0

ND – 0

ND – 0

ND – 0

ND – 0

cides and PCBs grossly contaminate breast milk in Nigeria. The levels (µg g–1 fat weight) of OCs in breast milk varied widely as follows (Table 14) indicating mean values and range in parenthesis : p,p-DDE 1.95 (0.18–9.01), p,pDDT 1.27 (0.01 –6.69), total DDT 3.60 (0.18–13.82), aldrin 0.04 (<0.01–0.40), HCB 0.33 (<0.01 –4.87), (0.18–13.82), a,b-HCH 0.03 (<0.01–0.30), lindane 0.46 (ND-6.55), heptachlor 0.06 (<0.01–0.38), endosulphan 0.64 (<0.01–10.03) and PCBs 0.02 (ND-0.29). This shows that Nigerian mothers are exposed to and accumulate considerable amounts of DDE, DDT, lindane, HCB and endosulphan and with low exposure to a,b-HCH and PCBs, respectively. The organochlorine pesticides concentrations are relatively higher than values reported in literature for European countries and South Africa (non-occupationally exposed population) but much lower than values for Hong Kong [60] except for HCB (see Table 15). The occurrence of relatively high levels of DDE, DDT, HCB, lindane and endosulfan in human breast milk is of concern in view of Nigerian government

Nigeria

1.95 (0.18–9.01)

1.27 (<0.01–11.90)

0.04 (<0.01–0.40) – 0.33 (<0.01–4.87) 0.03 (<0.01–0.30) 0.46 (<0.01–6.55) 0.06 (<0.01–0.38) 0.64 (<0.01–10.03) 0.02 (<0.01–0.30)

Country

p,p-DDE

p,p-DDT

Aldrin Dieldrin HCB a,b-HCH Lindane Heptachlor Endosulfan PCBs

and Asia

– – 1.44 (1.22–2.00)

– 0.02 (0.01–0.03) 0.14 (0.11–0.20) 0.17 (0.14–0.22)

0.26 (0.21–0.33)

1.64 (1.15–2.83)

Sweden

– 0.5 (<0.1–2.1)

– 0.08 (<0.01–0.55) 0.14 (<0.01–1.0) 0.22 (<0.01–4.4) –

0.11 (<0.01–1.2)

1.6 (<0.01–7.3)

Great Britain

– – 0.72 (<0.01–5.38) 0.44 (<0.01–9.10) – – – 2.04 (<0.01–12.0)

–

1.51 (<0.01–12.8)

West Germany

South Africa*

11.67 (4.07–22.96) 8.65(0.5–46.9)exposed people; 0.65(ND-4.73)nonexposed people 2.17 (0.67–4.04) 6.77(0.42–28.8)exposed people 0.04(ND-0.36) nonexposed people – 0.24 (0.04–0.80) 0.05 (<0.01–0.29) 15.96 (2.91–27.24) – – – 0.64 (0.25–1.43)

Hong Kong

Table 15. Concentrations of chlorinated hydrocarbons (mg g–1 fat weight) in human breast milk from Nigeria and from other countries in Europe

Organochlorines in Nigeria and Africa

349

350

O. Osibanjo

vigorous campaign that mothers breast milk is best for children.. It has been established by studies in South Africa that OCs can be transferred to infants via breast milk [60, 61]. Thus infants are being exposed to these xenobiotics while the toxicological hazards and risks have not been studied in Nigeria and many African countries. 5.7 Organochlorine Concentrations in Human Blood from Nigeria

Blood level of OCs is indicative of the degree of exposure to these chemicals and the potential chemical intoxication and other deleterious health effects in the population exposed. Table 16 provides data on OCs levels in the blood of Nigerian among different groups namely: occupationally exposed farm workers with over 20 years experience in spraying organochlorine pesticides especially lindane and DDT in agriculture and vector control; industrial workers in a factory formulating lindane and blood of mothers in an urban area [61]. The fourth group is residents of Koko port where about 4000 tonnes of hazardous wastes from a European country were illegally dumped in 1988 [62]. Some containers of PCBs wastes were included in the dangerous cargo. g-HCH (lindane), DDT and its metabolite DDE are the OCs, which occurred most frequently in the blood of this group of workers. This is not surprising since the workers have often sprayed these two insecticides. High levels (mg g–1 fat weight) of DDT (ND-101; 17.70); DDEpp (ND-24.05; 7.20), HCB (ND-13.53; 2.64) and lindane (0.19–5.97; L 1.09) were obtained. Lindane was detected in all the blood samples from the workers of the lindane formulation plant. The OCs level in the blood of this group of workers was much higher than those of the farm workers. For example, most of the plant and individuals in the town workers had blood levels of lindane 4.0–9.0 mg g–1 as compared with 0.2–0.6 mg g–1 in the farm workers group. The two groups of workers have much higher burden of OCs compared to the levels in the general population(e.g., lindane ND-0.50; 0.14; total DDT ND-12.78, 1.34) and mothers blood (lindane ND-31;3.03; total DDT ND-12.78;7.33). The blood levels in the Nigerian farm workers compared with levels reported in occupationally exposed Sudanese farm workers [63]. Blood samples from the port town, Koko, scene of illegal dumping of toxic wastes from Italy in 1987/88 were collected from five categories of individuals namely: residents close to dumpsite, dockworkers, neighbors, and hospital workers [61]. The concentrations of % fat, PCBs and lindane recorded in the blood of residents near the Koko dump site namely: 0.40%, 708.5 ng g–1 and 0.80 ng g–1 were the highest in the study area. The PCBs result was comparable to the range of 11–720 ng g–1 in the blood of PCB intoxicated patients in Taiwan [64].

ND-31 (3.03)

ND-0.50 (0.14)

ND-6.4 (1.33)

ND

ND-2.28 (0.26)

ND-1.2 (0.05)

Mother’s blood on fat weight

General population human blood on fat weight Koko Toxic waste dump residents (ng g–1)

NB: * fresh weight basis. ND=not detected.

0.22–9.63 (3.98)

Blood of industrial 0.13–1.05 workers at in(0.72) secticide formulation plant

γ-HCH

0.19–5.97 (1.09)

α1β-HCH

Blood of farm 0.10–1.70 workers at Ibadan (1.19) University

Sample type

ND-5.0 (3.2)

ND-1.62 (0.42)

ND-8.2 (1.16)

ND-0.93 (0.09)

ND-1.08 (1.08)

Aldrin

–

–

ND-6.24 (0.63)

ND-33.6 (5.18)

ND-13.53 (2.64)

HCB

ND-68.1 (22)

ND-12.32 (1.34)

ND-12.78 (7.33)

ND-35.65 (10.97)

ND-101 (17.70)

Total DDT

Table 16. Summary of OCs levels in human blood in Nigeria (mg/g fat weight) *

ND

1.18–8.31 (3.63)

ND-2.5 (0.9)

ND-5.30 (0.50) ND-53.9 (16.4)

ND-4.1 (1.8)

ND

ND-18.33 (1.67)

–

–

ND

ND-0.18 (0.21)

Heptachlor Endosulphan Dieldrin

0.16–11.07 ND-5.3 (2.81) (1.60)

ND-24.05 (7.20)

ND-92.30 (14.20)

p,p-DDE

ND-730 (199)

0.30–11.50 (2.76)

ND-23.0 (4.6)

Detected but not quantified

Detected but not quantified

PCBs

Organochlorines in Nigeria and Africa

351

352

O. Osibanjo

6 Issues of Concern There is lack of data on OCs production and use in Nigeria and in most of the African countries. Pesticides use and associated ecotoxicological problems will increase as population, demand for food and vector disease control skyrocket Anthropogenic activities provide the primary point of source of OCs into the environment either through deliberate application or accidental release into the environment. But there is a low level of public awareness on the health hazard and risks associated with OCs usage. Hence there is gross misuse and abuse of OCs including unorthodox pathways of human exposure. Environmental contamination by organochlorines of the aquatic environment, land, wildlife, foodstuffs, human diets, human blood and mothers’ breast milk in Nigeria and different sub-regions in the African continent has been established in the foregoing. Levels of aldrin+dieldrin in some meat and pulses were above FAO maximum residue limits. Contamination of human blood and mothers’ breast milk place children especially and the human population at risk from exposure to OCs. In other words the persistence, mobility and contamination of the environmental by organochlorines of the aquatic environment, land, wildlife, foodstuffs, human diets, human blood and mothers’ breast milk in Nigeria and different sub-regions in the African continent has been established in the foregoing. Levels of aldrin+dieldrin in some meat and pulses were above FAO maximum residue limits. Contamination of human blood and mothers’ breast milk place children especially and the human population at risk from exposure to OCs. In other words the persistence, mobility and contamination of environmental Degradation studies on aldrin, DDT and lindane show that these OCs degrade much faster with shorter half-lives of a few weeks in Nigerian soil compared to half-lives of several weeks in cold temperate soils. This raises a fundamental issue on the need for research in developing countries as some of the OCs which have been severely restricted or banned in developed countries due to their persistence and toxicity might still be tolerated in the tropical weather regimes of Africa because of huge debt burdens of the countries. This also underscores the need for national monitoring of OCs residues, as well as the problems of transposing environmental data from developed countries to developing countries. Furthermore, lack of national data on OCs use make regional assessment of OCs sources, uses and emissions difficult in the Africa region. Since urgent control of OCs is required, the reliance on estimates using models becomes imperative. The question arises which model best approximates the real environmental scenario in a region taking cognisance of abiotic and biotic factors with understanding and identification of the properties that make a chemical susceptible to global transport and polar deposition. Another important issue of great concern in African countries is the need for capacity building and the strengthening of national institutions that deal with the sustainable management of chemicals and chemical wastes. The foregoing issues need to be addressed urgently if Nigeria and other African countries are to implement in a sustainable manner any global instruments/protocols that may emerge eventually to tackle the problems of persistent toxic substances in the environment.

Organochlorines in Nigeria and Africa

353

7 References 1. 2. 3. 4. 5. 6. 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. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43.

Ondieki JJ (1996) Sc Tot Env 188:30 Ei-Zorgani GA (1976) Pestic Sci 7:150 Denloye S (1985) MSc thesis University of Ibadan, Nigeria Dejoux C (1988) La pollution des eaux continentals africaines. Experience acquise, situation actuelle et perspectives. Trav Doc Inst Fr Rech Sci Dev Coop 213:513 FAO (1988) Report of the twenty-ninth Session of the FAO desert locust control committee, Rome, 13–17 June 1988. Meeting Report No. AGP/1988/M/2, 38 p Osibanjo O, Biney C, Calamari N, Kaba N, Mbome I L, Naeve H, Ochumba PBO, Saad MAH (1994) Chlorinated Hydrocarbon substances, FAO Fish Report 502:7 Koh D, Jeyaratnam J Sc Tot Env 188:78 Choudhury AW (1989) In: Lehtinen S, Kurpa K, Korhonew E, Saarinen L (ed.), Proceedings of East Africa Regional Symposium on Chemical Accidents and Occupational Health. pp 70. Atuma SS Okor DI (1985) Ambio 14:34 Arebun O H (1990) MSc thesis University of Ibadan, Nigeria UNEP/FAO/WHO/IAEA (1990) MAP Tech Re Ser 39:1 Croll BT (1991) J Int Water Environ Mgmt 5:389 Eisler R (1977) Mar Pollut Bull 8(11):260 FAO (1999) Inventory of obsolete pesticides stock in Nigeria. A Country Report. Osibanjo O (1999) UNEP Mission Report on Preliminary Inventory of Hazardous Waste in the Gambia Perfect J (1980) Ambio 9(1):16 Spencer WF Farmer W J Cliath MM (1973):Res Rev 49:1 Osibanjo O, Ajewole K (1998) J Chem Soc Nigeria Edwards CA (1976) Persistent Pesticides in the Environment, 2nd edn. CRC Press Khan S (1987) Pesticide in the soil environment. Academic Press, New York Sleicher CA Hoperaft J (1984) Env Sci Tech 18:514 Samuel T, Agarwal HC, Pillar MKK (1988) Pest Sci 22:1 Talekar N S, Lian-Tien S, Her-min L, Jiag-Sing C (1977) J Agric Fd Chem 225(2):348 UNEP/WHO (1988) GEMS/WATER, Nairobi, 80p IOC-UNESCO (1985) IOC Workshop Report, Paris Agunloye TO (1984) MSc Thesis University of Ibadan, Nigeria Tongo AA (1985) MSc Thesis University of Ibadan, Nigeria Nwankwoala. AU, Osibanjo 0 (1992) Sc of Tot. Env. 119:179 Ogunlowo SO (1991) MSc Thesis University of Ibadan, Nigeria Okonna SI (1985) MSc Thesis University of Ibadan, Nigeria Saad MAH (1981) Research Report, Vienna, IAEA, 126p Tayel F (1981) MSc Thesis University of Alexandria, Egypt Croll BT (1991) Water Treat Exam 18:389 Suzuki M, Yamato Y, Watanabe T (1977) Env Sci Tech 11:1109 Ramesh A (1990) Env Pollut 67:289 Merriman JC, Metcalf JL (1986) Tech Bull Environ Can 160:1 Oliver BG, Nicol KD (1984) Sc Total Env 39:57 Osibanjo O, Aiyejuyo O (1994) Nig J Sc 5:14 Osibanjo O (1996) Present Water Quality Status in Nigeria. In: Aina EOA,Adedipe NO (eds), Water Quality Monitoring and Environmental Status in Nigeria., FEPA Monograph Lagos Eisenreich SJ, Hollad GJ, Johnson TC (1979) Env Sci Tech 13:569 Oliver BG, Charlton MN (1984) Env Sci Tech 18:903 Amakwe C (1984) BSc Thesis University of Ibadan, Nigeria Osibanjo O, Jensen S (1980) Ecological and Environmental Perspectives of Pesticides Pollution. In: Akinyele O, Omueti JA (eds), Proceedings Conference on Water Pollution and Pesticides residue in foods.

354 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64.

O. Osibanjo: Organochlorines in Nigeria and Africa Fayomi SF (1987) BSc Thesis University of Ibadan, Nigeria. Osibanjo O, Bamgbose O (1991) Mar Poll Bull 21:581. Munga D (1985) MSc Thesis University of Nairobi, Kenya. Bamgbose O, Jinadu K, Osibanjo J Env Sci Health 28(2):321 UNEP/FAO/WHO (1988) Assessment of Chemical Contaminants in Food. Report 110p, A. (1995): Osibanjo O, Adeyeye A (1995) Bull Environ Contam Toxicol 54:460 Osibanjo O, Adeyeye, A. (1997) Bull Environ Contam Toxicol 58:206 Adeyeye A, Osibanjo O (1999) Sci Total Env 231:227 Adeyeye A, Osibanjo O (1999) Sci. Total. Env Kannan K, Tanaabe S, Ramesh A, Subramanian AN, Tatsukawa R(1992) J Agric Food Chem 40:518 FAO/WHO (1986) Codex Alimentarius, vol. XIII, 2nd edn Osibanjo O, Jinadu KA (1999) Nig. J Sc Koerman JH, Genderen HV (1966) J Appl Ecol 3:99 Kendall RJ, Environ Sci Technol 16:448 Evans SW, Bouwman H (1993) Ostrich 64:47 Osibanjo O (1986) Tech Report, Fed Min Works & Env, Nigeria Ip HMH, Phillips DJH (1989) Arch Env Contam Toxicol 18:490 Odaro O (1987) MSc Thesis University of Ibadan, Nigeria Osibanjo O (1988) Tech Report, Fed Min Works & Env, Nigeria El Zorgani GA, Musa A (1976) Bull Environ Contam Toxicol 16(1):16 Chen PH Gaw JM, Wong CK, Chen CJ (1980) Bull Environ Contam Toxicol 25:325

CHAPTER 13

Sources, Fates and Effects of Persistent Organic Pollutants in China, with Emphasis on the Pearl River Delta M. H. Wong · B. H. T. Poon Institute for Natural Resources and Environmental Management and Department of Biology, Hong Kong Baptist University, Kowloon Tong, Hong Kong SAR, PR China E-mail: [email protected]

Contamination from persistent man-made chemicals is a pervasive global problem that urgently demands a global solution. This paper reviews the current situation of several persistent organic pollutants (POPs), namely polycyclic aromatic hydrocarbon (PAHs), aldrin, dieldrin, endrin, DDT, chlordane, heptachlor, mirex, toxaphene, hexachlorobenzene, polychlorinated biphenyls (PCBs), polychlorinated dioxins (PCDDs) and furans (PCDFs) in China with emphasis on South China including Hong Kong. The general information on the production and use of POPs in the region, their emission and environmental contamination levels, and their toxic effects on human health are reviewed. More investigations and support are urged in order to have a more complete assessment of POPs in the region. Keywords: Hong Kong, Polycyclic aromatic hydrocarbons, Pesticides, DDTs, Polychlorinated biphenyls, Dioxins, Furans, Bioaccumulation, Toxicity

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356

1

Introduction

2

General Information of POPs in the Region

3

Environmental Contamination of POPs

3.1 3.2 3.3

POP Pesticides . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358 PAHs and PCBs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359 PCDDs and PCDFs . . . . . . . . . . . . . . . . . . . . . . . . . . 360

4

Biological Effects of POPs . . . . . . . . . . . . . . . . . . . . . . 362

4.1 4.2

Bioaccumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . 362 Toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364

5

Conclusion

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365

6

References

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366

. . . . . . . . . . . . 356

. . . . . . . . . . . . . . 358

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

356

M.H. Wong · B.H.T. Poon

1 Introduction The Stockholm Convention on Persistent Organic Pollutants list twelve priority persistent organic pollutants (POPs) including eight pesticides (aldrin, dieldrin, endrin, DDT, chlordane, heptachlor, mirex and toxaphene), two types of industrial chemicals (PCBs and hexachlorobenzene) and two families of unintended manufacturing by-products, use and/or combustion of chlorine and chlorinecontaining materials (dioxins and furans) for special investigation [1]. All the twelve priority POPs are chlorinated and carbon-based compounds. Although these chemicals are now being banned for manufacture and/or restricted for use in most developed countries, living organisms including human beings can still come in contact with them by breathing contaminated air, eating contaminated food, and drinking or washing in contaminated water. Polycyclic aromatic hydrocarbons (PAHs) are also classified as persistent organic compounds, and they are formed by combustion and burning of organic compounds. Their occurrences are related to anthropogenic processes, and contamination of PAHs in river sediment is especially serious in high-density industrial areas [2]. There has been increasing evidence that the use, production and even unintentional generation of POPs is fundamentally unmanageable and unsustainable. This highly toxic class of man-made chemicals has become an ubiquitous element in our environment over the past decade. These chemicals last a long time in the environment, they are subject to global transport, moving from warm areas to colder areas in many cases, and build up in the food chain due to biomagnification, and finally impose human health hazards, even at very small quantities [3]. The main objective of this article is to review the current situation on production, emission levels, environmental contamination levels and biological effects of these POPs in China, with emphasis on the Pearl River Delta.

2 General Information of POPs in the Region The economy in China, especially the Pearl River Delta, has grown vigorously because of the economic reform and the open-door policy since 1978. The environment is under serious stress due to heavy domestic and industrial pollution. Land use in this region has been transformed from agricultural to industrial purposes. By moving a number of factories from Hong Kong across the border, Hong Kong economy has been changed from small-medium industrial enterprises to financial and serving centres. In spite of this, a substantial amount of industrial and manufactured chemicals has been imported to Hong Kong. Table 1 lists the recent Hong Kong import and export statistics of relevant POPs. Most of the food products consumed in Hong Kong come from Guangdong Province due to the decline of agricultural activities in Hong Kong, e.g., about 65% (355,212 t) of the total fresh and chilled vegetables imported (552,782 t)

357

Sources, Fates and Effects of Persistent Organic Pollutants in China Table 1. Import [4] and export [5] of persistent organic compounds in Hong Kong

Commodities Import statistics Insecticides Halogenated aromatic hydrocarbons Halogenated, sulphonated or nitrated phenols Phthalates Phthalate inducers Liquid Dielectric Transformers Export statistics Commodities Insectides Phthalate inducers Liquid Dielectric Transformers

unit

1996

1997

1998

1999

2000

tonnes tonnes

12,544 1,973

10,047 531

11,415 619

8,374 376

6,059 376

tonnes

592

611

884

507

851

tonnes tonnes no.

114,736 181,542 10,209

120,160 195,820 48,468

108,690 269,465 843,144

unit tonnes kg no.

1996 3934 0 0

1997 3479 0 0

1998 1635 0 0

125,274 80,630 271,649 299,839 37,510 478,784 1999 2653 0 0

2000 579 150 1

was supplied from China [4]. Therefore, most of the imported pesticides are for non-agricultural usages, such as domestic and outdoor (e.g., golf courses) pest control. According to the survey conducted by Friends of The Earth, most of the above listed POP pesticides are restricted for use in Hong Kong, except mirex which is under control by Agriculture, Fisheries and Conservation Department [6]. DDT was also banned from use in Hong Kong since 1988. Organophosphates (dichlorvos, dimethoate and disulfoton), carbamates (carbaryl and carbofuran) are common pesticides used in Hong Kong [7]. In general, over ten thousand tonnes of pesticides have been imported to Hong Kong in the past few years, with the import amount well in excess of the export amount. Moreover, the number of liquid dielectric transformers imported has drastically increased from 47 in 1993 [4] to almost half a million in 2000. They usually contained PCBs and PCB substitutes, which are used as dielectric heat stabilizers. The percentage of PCB-containing transformers is unknown/ needs to be evaluated. More than 10,000 t of DDT had been produced annually in China before it was banned in 1983 [8]. Chlordane and toxaphene are still manufactured, but no statistical information related to the production rates is available [9]. Trading statistics of pesticides during 1996 to 1999 are listed in Table 2. There were two types of Chinese manufactured PCBs – PCB3 and PCB5, similar to Aroclor 1242 and Aroclor 1254, respectively. Production of these PCBs started from 1965 and ceased in 1974, with more than 10,000 t produced during that period. They had multiple uses such as coolant/dielectric for transformers and capacitors, plasticizers and in manufacture of carbonless paper. Pentachlorobiphenyl is still manufactured and applied as one of the additives in the production of paint today [11]. Due to their heat and electrical resistant properties, industrial PCBs had been widely used in transformers as stabilizers during the past three decades. Starting from 1985, more than 2000 large-size trans-

358

M.H. Wong · B.H.T. Poon

Table 2. Trading amount of pesticides and insecticides during 1996 to 1999 in China [10]

Year

1996

1997

1998

1999

Import (US$ 1000) Export (US$1000)

421,458 428,482

424,563 446,986

450,153 477,361

505,029 610,095

formers have been demolished in China. There has been concern related to the demolition of transformers [12]. In addition, the annual production rate of pentachlorophenol (PCP), a potential PCDD/PCDF inducer, is about 5000 t in China [13].

3 Environmental Contamination of POPs 3.1 POP Pesticides

DDT was banned from use in China and Hong Kong in 1983 and 1988, respectively [7], but due to the highly persistence and illegal usage, DDT and its metabolites contaminations have been frequently reported in various studies (Table 3). Our previous study also tested five POPs pesticides, namely aldrin, dieldrin, DDTs, endrin and heptachlor, in agricultural soils and vegetables collected from South China. Only DDT and dieldrin were detected. Table 4 summaries their conTable 3. Environmental concentrations of DDT in South China

Sector

Locations

Total DDTs

Reference

Marine sediment Marine sediment River sediment River sediment River water River water River water

Victoria Habour Pear River Estuary Pearl River Hong Kong Guangzhou Shenzhen Hong Kong

24.0 mg/kg (d.w.) 2.84 mg/kg 11.1 mg/kg (d.w.) <0.5–38 mg/kg (d.w.) 0.01–0.50 mg/L <0.01–0.06 mg/L <0.01–0.01 mg/L

[14] [15] [16] [17] [18] [17] [17]

Table 4. Persistent organic pesticides in agricultural soils at the South China region

Conc (mg/kg d.w.)

Location Donggun

Huizhou

Guangming

Hong Kong

Dieldrin DDE DDT Total DDT

ND 14.3– 65.3 21.4– 60 59.6–138.4

ND 9.5–21.7 18.2–27.6 34 –62.6

ND 5.2– 11.9 14.6–22.6 23 –37.5

<2 –10.9 7.4– 21.7 27.5–45.9 39.2–67.7

ND=under detection limit.

359

Sources, Fates and Effects of Persistent Organic Pollutants in China

centrations in agricultural soils.All the sites are intensively farmed for vegetable production [19]. The ratio of DDE to DDT of two farmlands (Donggun and Huizhou) in China was 0.8, which was greater than that in Hong Kong (0.5). This reflected the situation that although China had banned DDT before Hong Kong, illegal use was still going on. The ratio of DDE to DDT in Guangming is similar to Hong Kong because good agricultural practice (GAP) is adopted. The results also indicated that Hong Kong has used dieldrin on agricultural land. 3.2 PAHs and PCBs

PAHs are also emitted through discharges from industrial and wastewater treatment plants, attached on solid particles and settled to the bottoms of lakes, river and sea [20]. The primary source of anthropogenic PAHs in the environment is atmospheric deposition based on pyrolysis of fossil fuels [21].With a heavy traffic and no raw petroleum industry in Hong Kong, PAHs emissions are mainly non-point sources which are difficult to control. Hong Kong had no PCBs production industry and China ceased to produce PCBs after 1974. It is expected that PCBs entered the air, water and soil in the region during their manufacture and use before 1974. PCBs can be released into the environment from hazardous waste sites containing PCBs, illegal or improper dumping of PCB wastes, and leaks from electric transformers containing PCBs [22]. Hong Kong initiated the monitoring program of PAHs and PCBs in marine sediment since 1994 [24]. Table 5 shows the levels of PCBs and PAHs contamination levels in different monitoring stations from 1993 to 1997 [23]. The results indicated that PAHs and PCBs contamination was more serious in Victoria Harbour and most of the typhoon shelters compared with less than 50 and 10 mg/kg for PAHs and PCBs, respectively, obtained from the open sea monitoring stations (southern Water and Mirs Bay). Being an industrial centre, Kwun Tong sedTable 5. Monitoring results of PAHs and PCBs in some locations of Hong Kong [23]

Locations

Victoria Habour (central) Tolo Habour Southern Water Mirs Bay Deep Bay Port Shelter Lamma Channel Tuen Mun Shelter Rambler Channel Shelter Kwun Tong Shelter Causeway Bay Shelter

Average concentration in sediment (1993–1997) Total PAHs (mg/kg d.w.)

Total PCBs (mg/kg d.w.)

400 (51–2787) 50 (39–129) 25 (18–37) 32 (24–49) 172 (39–828) 51 (39–78) 154 (40–421) 352 (48–1742) 568 (263–1480) 1044 (352–3001) 303 (93–1011)

11 (5–40) 7 (5–30) 2 (1–5) 6 (1–8) 7 (5–13) 9 (5–29) 8 (5–20) 13 (5–29) 28 (5–110) 133 (40–336) 17 (5–31)

360

M.H. Wong · B.H.T. Poon

Table 6. Contamination levels of PAHs and PCBs in river sediments of three major cities located

in the Pearl River Delta region River sediment (mg/kg)

Guangzhou Pearl River

Shenzhen Shenzhen River

Hong Kong North West New Territories

Total PAHs Total PCBs

1137 (504–1623) 3.4 (1.9–5.1)

477 (178–822) 3.4 (0.9–6.1)

702 (170–2145) 8.6 (0.9–19.8)

iments contained the highest levels of both total PAHs (1044 mg/kg) and PCBs (133 mg/kg), among all sites. The same as other regions, the elevated contamination of these compounds in marine sediment is originated from trace amounts presented in domestic sewage, stormwater runoff and industrial discharges [15, 25]. Our recent survey comparing the levels of PAHs and PCBs in river sediments collected from Guangzhou, Shenzhen and Hong Kong with populations of 6.8, 3.8 and 6.9 million, respectively, revealed that the pollution levels of PAHs in river sediments are substantially higher than that in marine sediments, whereas the levels of PCBs are lower than 10 mg/kg obtained in open sea sediment monitoring stations [17]. The contamination levels seemed to be related to the size and industrial activities, with Hong Kong having a medium level of PAHs and the highest level of PCBs among the three cities. Similar data have also been obtained in other studies. At Xiamen Bay, the levels of 158–337 mg/kg PAHs [26] and 0.19 to 72.7 mg/kg PCBs [27] have been noted. Another study indicated that the concentration of PCBs in river sediments at the Pearl River Delta ranged from 0.18 to 1.82 mg/kg [28]. 3.3 PCDDs and PCDFs

It is commonly recognized that man-made sources and activities are far greater contributors to the environmental burden of polychlorinated dioxins (PCDDs) and furans (PCDFs) than natural processes, especially since the 1930s, from which time there had been a steady increase in environmental levels coinciding with the large-scale production and use of chlorinated chemicals [29, 30]. Man-made sources of PCDDs and PCDFs can be divided into three main categories: chemical processes, combustion processes and secondary sources [31]. Municipal waste incinerator, coal combustion, chemical waste incineration plant, clinical waste incinerator, landfill gas combustion, crematoria, animal carcass incinerator and cement manufacture are the main sources of PCDDs and PCDFs in Hong Kong [32]. Table 7 shows the major contributors of PCDDs and PCDFs and their estimated amounts in 1997. Owing to the lack of chemical manufacture, emissions of PCDDs and PCDFs by chemical processes are comparatively lower than those by combustion processes in Hong Kong.

361

Sources, Fates and Effects of Persistent Organic Pollutants in China Table 7. Estimated PCDD/PCDF emissions to the atmosphere from Hong Kong [32]

Sources

Activity (1997)

Inventory (1997) (g I-TEQ)

Industrial sources Coal combustion (power) Landfill gas combustion Cement manufacture MSW combustion Chemical waste combustion Clinical waste combustion

6.1 MT 0.26 MT CH4 1.5 MT clinker 116,508 t (old) 10,198 t 3,650 t (old)

0.4–2.0 0.2–0.3 0.32 21–27 0.004 0.4–1.8

16,250 bodies

0.024

2,049 M km 2,237 M km 2,515 M km 2,000 M km 2,288 M km 612 M km 287 M km

0.002–0.45 0.001–0.03 0.002–0.03 0.001–0.02 0.06–0.09 0.016–0.023 0.0001– 0.006 23–33 g I-TEQ

Non-industrial sources Crematoria Cars – leaded – unleaded – diesel Light Goods Vans (diesel) Heavy Goods Vans (diesel) Buses (diesel) Motorcycles Total

Municipal solid waste incineration was the major source of PCDDs and PCDFs, which contributed more than half of the amount emitted in Hong Kong in 1997. Since the old incinerators were closed in 1998, it is expected that PCDD and PCDF emissions would be reduced to about 5 g I-TEQ annually [32]. Table 8 summarizes the PCDD/PCDF levels of ambient air in Hong Kong [32]. The level of dioxin in the industrial area was higher than that in the commercial area throughout the three years. This was possibly due to the fact that the monitoring station of the industrial area is located closed to the large-scale chemical waste incineration plant. The urban air PCDD/PCDF data obtained in Hong Kong are similar to those of other countries: 0.16 (0.08–0.28), 0.25 (0.09–0.45) and 0.25 (0.07–0.53) pg I-TEQ m–3 of PCDD/PCDFs measured in Japan, USA and Spain, respectively [33, 34, 35]. The generation of PCDDs and PCDFs in the pulp and paper industry has been investigated in China, where PCDDs ranged from 140 to 390 ng/kg dry pulp and PCDFs ranged from 7 to 80 ng/kg dry pulp [36]. This study showed that chemical processing is another major route for the emission of PCDDs and PCDFs to the environment. The environmental concentrations of PCDD/PCDFs in marine sediment and soil in different locations of China (Table 9) indicated that contaminations of PCDD/PCDFs are widespread, although not serious [13, 37].

362

M.H. Wong · B.H.T. Poon

Table 8. Summary of ambient air PCDD/PCDF level in Hong

Kong [32] Year

1997 1998 1999

Average Dioxin Concentration (pg I-TEQ m–3) Commercial Area

Industrial Area

0.122 (0.063–0.173) 0.078 (0.021–0.210) 0.138 (0.031–0.469)

0.155 (0.098–0.246) 0.098 (0.022–0.350) 0.215 (0.036–1.149)

Table 9. Environmental concentrations of PCDD/PCDFs in China

Location Matrix

Yellow Sea Sediment

East China Sea Sediment

Pacific Ocean Sediment

Ya-Er Lake Sediment

Ya-Er Lake Soil

[PCDD/PCDFs] (I-TEQ)

0.14 mg/kg (0.13–0.14)

0.13 mg/kg (0.11–0.14)

0.22 mg/kg (0.09–0.43)

0.65 mg/kg (0.07–1.42)

0.03 mg/kg (0.02–0.03)

Reference

[37]

[37]

[37]

[13]

[13]

4 Biological Effects of POPs The POPs contamination on natural environment and food is a worldwide phenomenon.A number of incidents of acute toxic effects in human, including death, has occurred. Exposure to contaminated environment and food can also pose chronic health risks, including cancer, but the long-term implications of low level dosage are not fully understood. Food contamination is simply caused by their bioaccumulative properties. 4.1 Bioaccumulation

Contamination of POPs in food may occur through environmental pollution of the air, water and soil. Due to their hydrophobic properties, accumulation of POPs in the lipid content of animals is a common phenomenon [38]. In China and Hong Kong, studies concerning bioconcentrations of POPs are mainly related to pesticides and PCBs. It has been demonstrated that ingestion of sediment and/or contact with contaminated particles are the predominant pathways for sediment-sorbed PCBs accumulation in Tilapia (fish) rather than water [39]. It was also indicated that feeding habits of different fish species intervene in the PCBs accumulation process. Higher contents of PCBs and chlorine number are found in black bass which is located in the highest tropic level [40]. Table 10 summarizes concentrations of DDTs, chlordane and PCBs in animals collected from Hong Kong and South China. It have been observed that SPCBs concentrations in Tilapia and shrimp collected from Mai Po Marshes, a relatively remote area designated as a nature re-

363

Sources, Fates and Effects of Persistent Organic Pollutants in China Table 10. Bioconcentration of several POPs in some Hong Kong and China animals

Animal Part

Mai Po Freshwater HK River

Hong Kong Coast

South China Sea

Tilapia shrimp whole body

Tilapia liver

dolphin porpoise blubber

whale blubber

0.071 0.006 0.043

46 / 24

33.01 0.28 1.79

[42]

[43]

SDDTs (mg/kg, ww) / Chlordane SPCBs (mg/kg, ww) 0.009 Reference

/ 0.006

[41]

39 / 12

[44]

serve, increased with the increase of lipid content rather than body size [41]. Concentrations of PCBs and DDTs in high lipid content organs, such as liver of fish and blubber of dolphin and porpoise, were about 10 times higher than in the whole body [42, 43]. It has been noted that PCBs IUPAC #138, 153 and 180 were the major congeners found in animal tissues. DDT concentrations were much higher than chlordane in animal tissues at this region, and most of the DDT had been converted into DDE. By comparing the data obtained in the whale blubber [42], and dolphin and porpoise blubber [44], it seemed that PCBs pollution was more serious along coastal areas of Hong Kong than the South China Sea. When moving up the food chain, the concentrations of PCBs and pesticides in biota tended to increase (shrimp<Tilapia<dolphin), indicating they are biomagnified in the ecosystem and the magnitude for persistence of these chemicals in animals is greater than that for biodegradability. POPs can be eliminated from human body through maternal milk in breast feeding. Levels of DDE, the main metabolites of DDT, have been identified in many studies in maternal milk samples from different countries in the world [45]. The bioaccumulative property of POPs can also be further illustrated by determining POP concentrations in human breast milk (Table 11). When comparing the results obtained from a recent study in South China [46] with those conducted in Europe, the level of contamination by DDTs is much more

Table 11. POPs (mg/kg fat) in human breast milk

Country

Year

Sample #

SDDTs

SPCBs

Reference

Guangzhou Hong Kong Hong Kong Sweden Sweden Sweden Canada Russia Brazil

2000 1999 1985 1997 1984 1972 1992 1993 1992

54 132 25 40 102 227 58 30 40

3854 3146 13840 143 561 3130 247 1474 1700

33.6 42 640 324 600 1090 238 429 150

[46] [46] [47] [48] [49] [49] [50] [51] [52]

364

M.H. Wong · B.H.T. Poon

serious whereas PCBs is less serious in South China region. There seem to be a general trend that both DDTs and PCBs in human breast milk decreased throughout the years in both Hong Kong and Sweden due to regulatory measures. 4.2 Toxicity

It has been estimated that global cumulative production of DDT and PCBs amounted to 1.36 and 1.17 million t, respectively [9]. In addition, more than 2 million tonnes of PAHs reached the marine environment and distributed worldwide [53]. In general, POPs have a high carcinogenic and mutagenic activity. These chemicals can act as endocrine disrupters and interfere with the body’s own hormones [54]. Such hormone-disrupting persistent contaminants can be hazardous at extremely low doses and pose a particular danger to those exposed in the womb. During prenatal life, POPs can alter development [55] and undermine the ability to learn, to fight off disease and to reproduce [56]. Pesticide concentrations in cooking smoke and aerosols have been investigated recently in South China and Hong Kong [57]. Both 4,4¢-DDT, 2,4¢-DDT and their metabolites (0.20–60.78 pg/m3) were detected in the smoke from cooking of several kinds of meat. This indicated that inhalation is another pathway for POPs entering human body, in addition to oral intake. Several epidemiological studies showed that there was a linkage between exposure of pesticides and human diseases. For example, diagnosis for 276 incident primary brain tumor cases among women in Shanghai indicated that risks of brain cancer were elevated with likely exposure to pesticides [58]. Data collected from 29 hospitals in Shanghai showed that women exposed to pesticides during pregnancy had an increased risk of birth defects and miscarriage [59]. A recent study conducted in Hong Kong revealed that duration of the exposure of pesticides was also correlated to the risk of Parkinson’s diseases [60]. Our recent study showed that the cytochrome P450 1A1 (CYP1A1) gene of fish can act as a sensitive, “early warning” method to reflect the levels of PCBs and PAHs contained in marine sediment [61, 62].An animal toxicological model was used to test harmful effects of PAHs by exposing clean green-lipped mussels to PAH contaminated sites in Hong Kong for 30 days. The results showed that DNA adducts in the gill tissue were related to tissue concentrations of benzo[a]pyrene as well as total PAHs of individual animals [63].A similar study on human beings was also conducted in which DNA adducts in human white blood cells in Chinese populations exposed to respirable particles in urban, rural and occupational settings where the particles originated from coal and petroleum fuel combustion, coke production, and other coal-tar aerosols were monitored. These particles contain carcinogenic PAHs that are known to form DNA adducts through covalent binding.At low to moderate environmental exposures to PAHs, DNA adduct levels in the white blood cells were significantly correlated with exposure [64]. Two epidemiological studies also indicated that high lung cancer mortality was associated with PAH exposure in China. The results of the first study showed that residents of Xuan Wei City were exposed to PAHs at occupational levels be-

Sources, Fates and Effects of Persistent Organic Pollutants in China

365

cause of non-vented coal or wool smoke during cooking resulted in a high lung cancer mortality rate. The potent carcinogen, 4.9±1.3 mg/m3 of benzo[a]pyrene, was found in the indoor air. In addition, their hydroxy-PAH levels in urine were higher than their counterparts living in other cities. This indicated that PAHs are the major mutagens and may be etiologically important in causing lung cancer [65]. The second study conducted in Liaoning showed that was a positive correlation between aromatic DNA adducts and urinary 1-hydroxypyrene levels of workers exposed to coke oven. There was also a significant correlation between serum p53 protein levels and the cumulated benzo[a]pyrene exposure dose [66]. PCBs showed a similar adverse effect as PAHs. Two cases of accidental PCBs contamination of rice cooking oil in Japan and Taiwan occurred in the late 1960s and mid-1970s, which exposed thousands of inhabitants to high concentrations of PCBs. Miscarriages and birth defects erupted within the populations [67]. It has been shown that cytochrome P450-dependent monooxygenases of Tilapia were induced by a variety of environmental pollutants including PCBs. Induction of fish monooxygenases may serve as a biological monitor for PCB-types of environmental contaminations [68]. Pentachlorophenol salts were sprayed in central China since the 1960s for the control of snail-borne schistosomiasis. It was found that the International Toxic Equivalents (I-TEQ) of PCDD/PCDFs ranged from 9.0 to 16.3 ng/kg lipid in blood of the residents of the sprayed area while in the general population it ranged from 4.8 to 6.4 ng/kg. The relatively high level of PCDD/PCDFs was also found in human breast milk with 5.4 ng/kg lipid (I-TEQ) PCDD/PCDFs, which was about double that of mothers from the general population (2.6 ng/kg) [69]. Most of congeners of the PCDD/PCDFs are strong carcinogens. Among them, 2,3,7,8 tetrachlorinated dioxin has the strongest hepatocarcinogenicity [70]. An attempt has been made to assess the contents of PCDD/PCDFs in tea leaves and it was revealed that brick tea (made of old leaves, and even branches, roots and fruits) had an I-TEQ of 2.58 ng/kg which is 16 times higher than the level contained in green tea (made of very young tea leaves). The results indicated that individuals of certain populations in China who consume a large quantity of brick tea liquor (up to 3 L/day) would be at risk [71].

5 Conclusion In order to assess the sources, fates and effects of POPs, a detailed inventory for their existing trade, use and potential sources, releases, and discharge should be developed within our region. It is recognized that the food chain is responsible for the majority of POP uptake into humans due to bioaccumulation and biomagnification. Direct contact with ambient air, water and soil by the general public should be taken into account for the health assessment of POPs. However, in the absence of or lacking in data of POPs in different environmental media (air, water, soil and food), it is difficult to obtain a complete picture to assess this issue in the region. Therefore, it is essential to enlarge the database of POPs in different ecological compartments, which could help to establish the local standards

366

M.H. Wong · B.H.T. Poon

and guidelines. More surveys and studies related to POPs should be encouraged with the aim of eliminating them at sources, and/or minimizing their environmental and human effects. Acknowledgement. The authors thank the Research Grants Council of Research Grants Committee (Central Allocation – Group Research, No. HKBU-2/00 C) of Hong Kong and the S. H. Ho Foundation Ltd. for financial support.

6 References 1. World Health Organization (1999) Persistent Organic Pollutants. Programme for Promotion of Chemical Safety Division of Control of Tropical Diseases, and Food Safety Unit, World Health Organization 2. Aboul-Kassim TA, Simoneit BRT (1995) Aliphatic and aromatic hydrocarbons in particulate fallout of Alexandria, Egypt: sources and implications. Environmental Science and Technology 29:2473–2480 3. Weinberg J (1998) Overview of POPs and Need for a POPs Treaty. Public Forum on Persistent Organic Pollutants – The International POPs Elimination Network 4. Census and Statistics Department (1993, 1996–2000) Hong Kong Trade Statistics – Import. Census and Statistics Department, Hong Kong 5. Census and Statistics Department (1996–2000) Hong Kong Trade Statistics – Export and Re-exports. Census and Statistics Department, Hong Kong 6. Hong Kong Special Administration Region Government (1993) Pesticides Ordinance and Regulations, Chapter 133 7. Monks J (1994) Pesticidal Madness. Friends of the Earth, Hong Kong 8. Hua X, Shan Z (1996) The production and application of pesticides and factor analysis of their pollution in environment in China. Advances in Environmental Science, 4(2):33–45 9. International Programme on Chemical Safety (1999) Assessment Report on the 12 prioritized POPs 10. Food and Agriculture Organization of the United Nations, FAO, (2000) Statistical database for agriculture – pesticides and insecticides 11. Jiang K, Li LJ, Chen YD, Jin J (1997) Determination of PCDD/Fs and dioxin-like PCBs in Chinese commercial PCBs and emissions from a testing PCB incineration. Chemosphere 34(5–7):941–950 12. Beijing Municipal Environmental Protection Bureau (1995) Handbook of Regulations on Environmental Protection in China. Resources for the Future, Washington, DC, USA 13. Wu W.Z., Schramn K.W, Henkelmann B,Yediler A, Kettrup A (1997) PCDD/Fs, PCBs, HCHs, and HCB in sediments and soil of Ya-er lake in China: results on residual levels and correlation to the organic carbon and the particle size. Chemosphere 34(1):191–202 14. Connell DW, Wu RSS, Richardson BJ, Leung K, Lam PKS, Connell PA (1998) Fate and risk evaluation of persistent organic contaminants and related compounds in Victoria Harbour, Hong Kong. Chemosphere 36(9):2019–2030 15. Hong H, Xu L, Zhang L, Chen JC,Wong YS,Wan TSM (1995) Environmental fate and chemistry of organic pollutants in the sediment of Xiamen and Victoria Harbours. Marine Pollution Bulletin, 31:229–236 16. Wu Y, Zhang J, Zhou Q (1999) Persistent organochlorine residues in sediments from Chinese river/estuary systems, Environmental Pollution 105:143–150 17. Poon BHT, Cheung KC,Wong MH Environmental Assessment of water quality in the Pearl River Delta, South China: Part II – Persistent Organic Pollutants (manuscript in preparation) 18. Yang Y, Sheng G, Fu J, Min Y (1997) Organochlorinated Compounds in Waters of Pearl River Delta Region. Environmental Monitoring and Assessment 44:569–575 19. Wong AWM, Cheung, KC, Wong, MH Pesticides in vegetables and agricultural lands in South China (manuscript in preparation)

Sources, Fates and Effects of Persistent Organic Pollutants in China

367

20. Agency for Toxic Substances and Disease Registry (ATSDR) (1995) Toxicology profile for polycyclic aromatic hydrocarbons. US Department of Health and Human Services, Public Health Service, Atlanta, GA 21. Witt G (1995), Polycyclic aromatic hydrocarbons in water and sediment of the Baltic Sea, Marine Pollution Bulletin, 31 (4–12):237–248 22. LaGrega DM, Buckingham LP, Evans CJ, The Environmental Resources Management Group (1994) Hazardous Waste Management: 4–6, McGraw-Hill, Singapore 23. Environmental Protection Department (1998) Marine water quality in Hong Kong – 1997. Water Policy and Planning Group, Environmental Protection Department, Hong Kong Government 24. Environmental Protection Department (1994) Marine water quality in Hong Kong – 1994. Water Policy and Planning Group, Environmental Protection Department, Hong Kong Government 25. Makepeace DK, Simth DW, Stanley SJ (1995) Urban stormwater quality: summary of contaminant data. Critical Reviews in Environmental Science and Technology 25: 93–119 26. Wang X, Xu L, Chen W, Zhang L, Hong H (1999) The vertical distributions and sources of PAHs in sediment of Xiamen Bay. Chinese Journal of Oceanology and Limnology 17(3):247–251 27. Chen W, Zhang L, Xu L, Wang X, Hong H (1996) Vertical distribution characteristics of organochlorinated pesticides and polychlorinated biphenyls in sediments of Xiamen Bay. Marine Science 2:56–60 28. Hong H, Chen W, Xu L, Wang X, Zhang L (1999) Distribution and fate of organochlorine pollutants in the Pearl River Estuary. Marine Pollution Bulletin 39:376–382 29. Fortin C, Caldbick D (1997) Are dioxins and furans predominantly anthropogenic? Organohalogen Compounds 32:417–429 30. Alcock RE, Gemmill R, Jones KC (1998) Improvement to UK PCDD/F and PCB atmospheric emission inventory following an emissions measurement programme. Chemosphere 38:759–770 31. Fiedler H (1993) Formation and sources of PCDD/PCDF Organohalogen Compounds 11:221–228 32. Environmental Protection Department (2000) An assessment of dioxin emissions in Hong Kong: Final Report.Air Policy and Planning Group, Environmental Protection Department, Hong Kong SAR Government 33. Abad E, Caiaxach J, Rivera J (1997) PCDD/PCDF from emissions sources and ambient air in northeast Spain. Chemosphere 35:453–463 34. Hunt GT, Maisel BE, Zielinsha B (1997) A source of PCDD/PCDFs in the atmosphere of Phoenix. Organohalogen Compounds 33:145–150 35. Seike N, Yoshida M, Mastuda M, Kawano M, Wakimoto T (1997) Seasonal concentrations and compositions of PCDD/DFs in atmospheric environment. Organohalogen Compounds 33:168–174 36. Zheng MH, Bao Z.C, Wang KO, Xu XB (1997) Level of PCDDs and PCDFs in the bleached pulp from Chinese pulp and paper industry. Bulletin of Environmental Contamination and Toxicology 59:90–93 37. Hashimoto S Wakimoto T, Tatsukawa R (1995) Possible natural formation of polychlorinated dibenzo-p-dioxins as evidenced by sediment analysis from the Yellow Sea, the East China Sea and the Pacific Ocean. Marine Pollution Bulletin 30(5):341–346 38. Sijm DTHM, Linde AVD (1995) Size-dependent bioconcentration kinetics of hydrophobic organic chemicals in fish based on diffusive mass transfer and allometric relationships. Environmental Science and Technology 29:2769–2775 39. Zhou HY, Wong MH (2000) Accumulation of sediment – sorbed PCBs in tilapia. Water Research 34(110):2905–2914 40. Zhou HY, Cheung YH, Wong MH (1999) Bioaccumulation of organochlorines in freshwater fish with different feeding modes cultured in treated wastewater. Water Research 33(12):2747–2756

368

M.H. Wong · B.H.T. Poon

41. Liang Y, Wong MH, Shutes RBE, Revitt DM (1999) Ecological risk assessment of polychlorinated biphenyl contamination in the Mai Po marshes nature reserve, Hong Kong. Water Research 33(6):1337–1346 42. Chan HM, Chan KM, Dickman M (1999) Organochlorines in Hong Kong Fish. Marine Pollution Bulletin 39:346–351 43. Minh TB, Watanabe M, Nakata H, Tanabe S, Jefferson TA (1999) Contamination by persistent organochlorines in small crustaceans from Hong Kong coastal waters. Marine Pollution Bulletin 39:383–392 44. Parsons ECM, Chan HM, Kinoshita R (1999) Trace metal and organochlorine concentrations in a Pygmy Bryde’s Whale (Balaenoptera edeni) from South China Sea. Marine Pollution Bulletin 38(1):51–55 45. Brouwer A, Ahlborg UG, Leeuwen FXRV, Feeley MM (1998) Report of the WHO working group on the assessment of health risks for human infants from exposure to PCDDs, PCDFs and PCBs. Chemosphere 37(9–12):1627–1643 46. Wong CKC, Leung KM, Poon BHT, Lan CY,Wong MH (2001) Organochlorine hydrocarbons in human breast milk collected in Hong Kong and Guangzhou. Archives Environmental Contamination and Toxicology (in press) 47. Ip HMH, Phillips DJH (1989) Organochlorine chemicals in human breast milk in Hong Kong. Archive of Environmental Contamination and Toxicology 18:490–494 48. Noren K, Meironyte D (2000) Certain organochlorine and organobromine contaminants in Swedish human milk in perspective of past 20 – 30 years. Chemosphere 40:1111–1123 49. Noren K (1988) Changes in the levels of organochlorine pesticides, polychlorinated biphenyls, dibenzo-p-dioxins and dibenzofurans in human milk from Stockholm, 1972–1985. Chemosphere 7:39–49 50. Newsome WH, Davies D, Doucet J (1995) PCB and organochlorine pesticides in Canadian Human milk – 1992. Chemosphere 30:2143–2153 51. Polder A, Becher G, Savinova TN, Skaare JU (1998) Dioxins, PCBs and some chlorinated pesticides in human milk from Kola Peninsula, Russia. Chemosphere 27(9–12):1795–1806 52. Paumgartten FJR, Cruz CM Chahoud I, Palavinskas R, Mathar W (2000) PCDDs, PCDFs, PCBs, and other organochlorine compounds in Human Milk from Rio de Janeiro, Brazil. Environmental Research 83:293–297 53. Law RJ (1986) Polycyclic aromatic hydrocarbons in the marine environment: an overview. ICES Coop Res Rep 142:88–100 54. Brucker-Davis F (1998) Effects of environmental synthetic chemicals on thyroid function. Thyroid 8(9):827–56 55. Colborn T, Clement C (1992) Chemically-induced alterations in sexual and functional development: the wildlife/human connections. Princeton Scientific Publishing, Princeton, New Jersey 56. Garry VF, Schreinemachers D, Harkins ME, Griffith J (1996) Pesticides appliers, biocides, and birth defects in rural Minnesota. Environmental Health Perspectives 104:394–399 57. Cheng Y, Sheng GY, Shao B, Lin Z, Min YS, Fu JM (2000) Characteristic and sources of organochlorine pesticides from cooking smoke and aerosols. China Environmental Science 20(1):18–22 58. Heineman EF, Gao YT Dosemeci M, Mclaughlin JK (1995) Occupational risk factors for brain tumors among women in Shanghai, China. Journal of Occupational and Environmental Medicine 37(3):288–293 59. Zhang J, Cai WW, Lee DJ (1992) Occupational hazards and pregnancy outcomes.American Journal of Industrial Medicine 21(3):397–408 60. Chan D K, Woo J, Ho SC, Pang CP, Law LK, Ng PW, Hung WT, Kwok T, Hui E, Orr K, Leung MF, Kay R (1998) Genetic and environmental risk for Parkinson’s disease in Chinese population. Journal of Neurology and Neurosurgery Psychiatry 65(5):781–784 61. Wong CKC, Yeung HY, Cheung RYH, Yung KK L, Wong MH (2000) Exotoxicological assessment of persistent organic and heavy metal contamination in Hong Kong coastal sediment. Archives of Environmental Contamination and Toxicology 38:386–493

Sources, Fates and Effects of Persistent Organic Pollutants in China

369

62. Wong CKC, Yeung HY, Woo PS, Wong MH (2001) Specific expression of cytochrome P4501A1 gene in gill, intestine and liver of tilapia exposed to coastal sediments. Aquatic Toxicology 54:69–80 63. Xu L, Zheng GJ, Lam PKS, Richardson BJ (1999) Relationship between tissue concentrations of polycyclic aromatic hydrocarbons and DNA adducts in green-lipped mussels (Perna virdis) Ecotoxicology 8(2):73–82 64. Lewtas J Walsh D Williams R, Dobias L (1997) Air pollution exposure-DNA adduct dosimetry in human and rodents: evidence for non-linearity at high doses. Mutat Research 378(1–2):51–63 65. Mumford JL, Li X, Hu F, Lu XB, Chuang JC (1995) Human exposure and dosimetry of polycyclic aromatic hydrocarbons in urine from Xuan Wei, China with high lung cancer mortality associated with exposure to unvented coal smoke. Carcinogenesis 16(2):3031–3036 66. Pan G, Hanaoka T, Yamano Y, Hara K, Ichiba M, Wang Y, Zhang J, Feng., Shujuan Z, Guan D, Gao G, Liu N, Takahashi K (1998) A study of multiple biomarkers in coke oven workers – a cross sectional study in China. Carcinogenesis 19(11):1963–1968 67. Iida T, Hirakawa H, Matsueda T, Takenaka S,Yu M L, Guo YLL (1999) Recent trend of polychlorinated dibenzo-p-dioxins and their related compounds in the blood and sedum of Yusho and Yu-Cheng Patients. Chemosphere 38(5):981–993 68. Ueng YF, Liu TY, Ueng TH (1995) Induction of cytochrome P450 1A1 and monooxygenase activity in tilapia by sediment extract. Bulletin of Environmental Contamination and Toxicology 54(1):60–67 69. Schecter A, Jiang K, Paepke O, Fuerst P, Fuerst C (1995) Comparison of dibenzodioxin levels in blood and milk in agricultural workers and others following pentachlorophenol exposure in China. Chemosphere 29 (9–11):2371–2380 70. Chang HY, Lee CC, Kuei CH, Guo TL (1999) The exposure and toxicity of polychlorinated dibenzo-para-dioxin-like chemicals. Chinese Journal of Public Health 18(1):13–27 71. Fiedler H, Cheung KC, Wong MH (2001) PCDD/PCDF, chlorinated pesticides and PAHs in Chinese teas. Chemosphere 46:1429–1433

CHAPTER 14

DDT in Mexico Fernando Díaz-Barriga 1 · Víctor Borja-Aburto 2 · Stefan Waliszewski 3 · Leticia Yáñez 1 1

2 3

Autonomous University of San Luis Potosi, School of Medicine, Laboratory of Environmental Toxicology, Avenida Venustiano Carranza No. 2405, Col. Lomas los Filtros, CP 78210, San Luis Potosi, SLP, Mexico E-mail: [email protected] National Center of Environmental Health, Mexico Institute of Forensic Medicine, University of Veracruz, Veracruz, Mexico

DDT was heavily used in Mexico in agriculture and for the Malaria Control Program. Therefore, in numerous communities, human exposure to DDT has been reported as a result of the presence of this insecticide in different environmental media. In this work, data needed for a health assessment of DDT in Mexico are summarized. The information includes environmental data (soil, household dust, water, sediment, aquatic biota and food); exposure data (human milk, adipose tissue and blood DDT concentrations in women, men and children); and some studies in relation to the biological effects that have been observed in individuals exposed to DDT. With all these data, we can conclude that DDT is a contaminant of concern that deserves further studies in Mexico, as the exposure will continue due to its persistence. Furthermore, now that DDT has been replaced with pyrethroids in the Malaria Control Program, a new scenario will need investigation, that is: the possible interaction between DDT and pyrethroids. Keywords: DDT, DDE, DDD, Malaria, Mexico

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372

1

Introduction

2 2.1 2.2 2.3 2.4 2.5

Environmental Pathways of Exposure to DDT Soil . . . . . . . . . . . . . . . . . . . . . . . Water . . . . . . . . . . . . . . . . . . . . . . Sediments . . . . . . . . . . . . . . . . . . . . Food and Biota . . . . . . . . . . . . . . . . . Air . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

373 374 375 375 376 377

3 3.1 3.2 3.3

Human Exposure to DDT . . Breast Milk . . . . . . . . . . Serum . . . . . . . . . . . . . Adipose Tissue . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

378 378 379 380

4 4.1 4.2 4.3

Health Effects . . . . . . . Cancer . . . . . . . . . . . Endocrine Disruption . . . Genotoxicity . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

382 382 383 384

5

Final Comments

6

References

. . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . 386

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386 The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

372

F. Díaz-Barriga et al.

1 Introduction DDT (dichlorodiphenyltrichloroethane) was first synthesized in 1874, and its insecticidal properties were discovered in 1939 by Paul Hermann Müller [1]. The U.S. military began using DDT extensively for mosquito control in 1944, particularly in the Pacific, where much of the action of World War II took place in highly malarious areas [1]. In the next year, and for the first time, houses were sprayed with DDT in Temixco, Morelos, Mexico [1]. The spray was applied to the walls and ceilings of residences. Studies done two months after the spraying, showed that there was a 99% reduction in the incidence of Anopheles [1]. In 1947–48, the spraying of DDT began in other Mexican regions, such as Veracruz, Mexico City and Baja California [1]. By 1948, the first clear evidence of malaria control appeared in the areas first sprayed with DDT; the overall parasite rate in the state of Morelos was found to be 10%, and the rate in the sprayed towns was found to be 1% [1]. In 1936 it was estimated that half of the Mexican population lived in endemic regions and was subject to a malaria mortality rate of 0.5%, or about 36,000 deaths per year [1]. During the 1930s and 1940s, malaria became the third cause of death in the country. However, the antimalaria campaign was not generalized until 1956 [2]. The success of DDT was outstanding, malaria cases decreased from 41,000 in 1955 to 4,000 in 1960 [3]; in 1970 the campaign was relaxed and the cases increased to 57,000 [3]. However, this was also the time in which DDT production peaked in Mexico, with more than 80 thousand tonnes produced annually [2]. In recent years, the incidences of malaria have declined significantly, to less than 5,000 cases. Since 1982 there have been no deaths from this disease. Malaria is a long-standing public health problem that has inhibited development in large areas of the country. Sixty percent of Mexico’s territory, from sea level to 1,800 meters above sea level, presents favorable conditions for malaria transmission. This includes the Pacific coast, the Gulf of Mexico slopes, the Yucatan peninsula and interior basins of the high plateau. Some 45 million people live in these areas [2]. Currently, Mexico operates a malaria control program that has substantially reduced the incidence of this disease, while gradually decreasing reliance on DDT. In 1995, Mexico initiated an integrated pest management approach for malaria to reduce the heavy dependence on pesticides. Improved sanitation, surveillance and minimum use of pesticides to control mosquitoes and larvae are considered key elements in this new approach [4]. Since 1998, DDT was substituted with pyrethroids in the malaria control program. In the area of agriculture, as much as 1,000 tonnes per year were used [2]. Application rates in the north of Mexico, were among the highest in the world [2]. However, the growing concern about DDT persistence has had a significant impact on agricultural practices in Mexico. During the early 1970s the US Food and Drug Administration (USFDA) began rejecting the importation of commodities due to high residue levels, especially of DDT [2]. Therefore, some agricultural areas changed to newer pesticides in order to comply with the USFDA regulations. By 1990, DDT was limited to campaigns addressing public sanitation [2].

DDT in Mexico

373

2 Environmental Pathways of Exposure to DDT The exposure pathways are the processes by which DDT may be transported from the pollution source to living organisms. In the malaria areas, the source of DDT was the household-spraying of the insecticide. Since the beginning of the control program of malaria, DDT was sprayed on the ceilings and walls, both indoors and outdoors. Therefore, after spraying, indoor dust (or indoor soil in some cases), and the external surface soil in those areas next to the dwellings, were the media first to become contaminated with DDT. From these points, the insecticide could be transported from one medium to another by different processes. For example: (A) due to its semi-volatility, DDT may volatilize from hot regions but will condense and tend to remain in colder regions [5], this property confers on DDT the capacity to be transported over long distances [5]; (B) DDT may be washed off the soil by rainwater, into nearby bodies of surface water; (C) DDT in excess of water solubility limits is adsorbed onto sediments which act as the primary reser-

Fig. 1. DDT (Tons) used in Mexico (1959–1999) in the Malaria Control Program. * Estimated for 1999. Source: Secretaria de Salud, Mexico

374

F. Díaz-Barriga et al.

voir for excess quantities of DDT; and (D) the lipophilic properties of DDT, combined with the extremely long half-life, may result in bioaccumulation and biomagnification. The only difference between this malaria scenario with agricultural practices is that, in the latter, DDT was sprayed directly to the soil. The very high persistence of DDT and related compounds in the environment (half lives of 10–15 years [5]), coupled with the amount of DDT that was used in Mexico in the last 40 years (Fig. 1) [6], provides the necessary conditions for DDT to become a contaminant of concern for Mexico. However, the transport and fate of DDT in the environment are issues that have not been sufficiently studied in Mexico. Thus, data are scarce, but still give a good perspective of the environmental distribution of DDT. In the next sections, we shall concentrate the analysis on the information obtained from the malaria areas, but some results from agricultural zones will also be presented. 2.1 Soil

Malaria areas are located in the tropics; therefore, it is important to remember that DDT degrades faster in warm and moist tropical soils compared to temperate ones [7]. Faster disappearance is due to more volatilization and biodegradation processes [7]. However, in these malarious areas, DDT was also applied indoors; thus, the concentration of this insecticide could be higher in household soil/dust, due to a slower degradation rate. Under aerobic conditions, slow conversion to DDE normally occurs; whereas, under anaerobic conditions, conversion to DDD results and is much more rapid than the aerobic conversion to DDE [8]. In a Mexican malaria community and in a control area, DDT levels have been quantified in household soil (indoor soil) and external surface soil (Table 1). The levels in the community were higher than those found in the control area. Furthermore, DDT (and its metabolites) concentrations in the malaria community, especially in household soil, were as high as those reported in a superfund site, where contamination was due to 60 years of mixing and batching of insecticides [8]. It can also be observed that the ratio DDT/DDE was higher in the malaria area.

Table 1. DDT, DDD and DDE levels in surface soil (mg/kg)

Sample

Location

DDT

DDD

DDE

DDT/DDE

Household Soil

Control Area Malaria Area Control Area Malaria Area Superfund Site *

0.37 82.7 0.6 49.5 61.0

0.02 41.02 0.62 13.24 70.0

0.2 13.8 0.22 5.68 10.0

1 6 3 9 6

External Soil

Our own unpublished results. * Baird and McGuire, Superfund Site in Holbrook, Massachusetts, USA [8].

375

DDT in Mexico

Considering these concentrations and taking into account the persistence of DDT, the risk of human exposure to polluted soil in malaria areas will remain a public health issue for the next years, especially for indoor soil. In contrast, the levels of DDT and its metabolites in external soil may depend on the loss of these compounds during runoffs. 2.2 Water

DDT, DDD and DDE (DDTs) are only slightly soluble in water, with solubilities of 3.4 ppb, 160 ppb and 120 ppb, respectively [8]. Therefore, sedimentation is the most important factor for the disappearance of DDT from water. However, it has also been suggested that contaminated sediments are a main source of DDT inputs to the water column [9]. In order to study the degree of pollution in water bodies located in malaria areas, DDTs were quantified in a relatively small stream. The levels of total DDT found in the malaria area were 280 ppb, while in the control area, no detectable levels were found. Considering that the DDT/DDE ratio had a value of 24, it can be assumed that the insecticide was recently sprayed in the area. This result may be an example of transport from the spraying area to water bodies. Taking into account 40 years of spraying DDT, considering the persistence of DDT in river water (higher than eight weeks [10]), and sediments (half life>3 years for p,p′-DDT [11]), and recognizing that the metabolites can be as toxic as the parent compound; an environmental surveillance program in the bodies of water and their sediments is urgently needed in Mexico. 2.3 Sediments

As stated above, sediments act as the primary reservoir for excess quantities of DDT. Therefore, it is very important to analyze the concentrations in this medium. In Table 2 it is shown that DDT concentrations in Mexican samples (malarious areas) are lower than those detected in other countries, where DDT was used either for the control of malaria or for agricultural practices. Whether this difference can be explained by an increased degradation or by a DDT mass reduction caused by water currents carrying suspended DDTs out of the contaTable 2. DDT levels in sediments (mg/kg)

Location

Total DDT

DDT

Ref.

Mexico (bay) Mexico (river) Mexico (lagoon) Viet Nam (canals) Egypt (river) Egypt (lake) USA (creek)

1.8 27.3 4.9 79.9 126.0 557.0 415.0

– – 2.2 23.0 9.0 18.0 110.0

[12] [13] [14] [15] [16] [16] [17]

376

F. Díaz-Barriga et al.

minated area, are issues that deserve further research. However, we cannot exclude another explanation. The Mexican studies, results of which are shown in Table 2, were not designed to assess the amount of DDT in sediments due to spraying. Therefore, the Mexican sites, although located in malarious areas, may not be as high risk as other sites closer to the sprayed communities. In fact, a sediment sample collected in a lagoon near an area where the insecticide was used intensively for vector control, had DDT concentrations of up to 16.6 mg/kg [18]. 2.4 Food and Biota

Due to their lipophilic attributes and high persistence, the DDTs may bioaccumulate significantly in animal species [19]. Furthermore, biomagnification has been observed; for example, DDT concentration increased with each successive trophic level in a food chain [19]. Taking into account these properties, food ingestion can be considered a pathway of exposure in malaria areas. In Mexico, studies have been done in different food items, such as fish, hen’s egg, butter and cow’s milk and muscle. In Table 3, total DDT levels in different aquatic organisms are presented. It can be observed that concentrations in agricultural areas are similar to those found in malarious areas. Furthermore, in freshwater shrimp collected in a river from a very malarious area, the ratio DDT/DDE was 7.0, with a DDT concentration of 1383.0 mg/g fat basis [18]. Considering fish, the concentrations of DDT, in organisms collected in Mexico, are above normal values. As is shown in Table 4, where DDTs levels in Tilapia are depicted for different countries. Table 3. Total DDT concentration in aquatic species

Location

Organism

Total DDT (ng/g)

Type of Area

Ref.

Palizada, River Palizada, River Carmen, Lagoon Palizada, River Baja Cal, Estuary

Shrimps Oysters Oysters Mussels Mussels

0.25 1.49 6.20 1.44 9.16

Malarious Malarious Malarious Malarious Agricultural

[20] [20] [14] [20] [21]

Table 4. DDTs concentration in fish (ng/g)

Country

p,p′-DDT

p,p′-DDD

p,p′-DDE

Ref.

Mexico Brazil Egypt Hong Kong

38.3 13.5 0.07 6.2

– – 2.0 16.3

3,687.0 – 6.4 17.6

[18] [22] [23] [24]

Tilapia sp. Concentrations were quantified in muscle. Brazil data are total DDT.

377

DDT in Mexico Table 5. DDTs concentration in cows’ sources (mg/kg fat)

Item

DDT

DDD

DDE

Ref.

Milk Muscle Kidney Liver Abdominal fat Kidney fat

0.085 1.64 1.96 0.96 0.46 0.21

0.014 ND ND ND ND –

0.05 0.078 0.084 0.088 0.035 0.026

[25] [26] [26] [26] [26] [27]

(ND) no detectable levels.

The analysis of the DDT/DDE ratio in hen’s egg revealed evidence of recent exposure to DDT. This ratio was higher in an endemic area of malaria than in a cotton area where DDT was used in the past. Concentration of DDT in cow’s milk was higher than the FAO/WHO Maximum Tolerance Level of 0.05 mg/kg (Table 5).Also noteworthy is that p,p′-DDT levels in cow’s milk were significantly higher in the months when the insecticide was sprayed due to anti-malaria actions [25]. Although, DDT concentrations in cow’s milk were higher than the guidelines, the levels of the insecticide in butter samples were lower than the FAO/WHO Maximum Tolerance Level of 1.25 mg/kg [28]. Also in Table 5 is depicted a finding that may need further research, DDT and DDE concentrations were higher in cows’ soft tissues (muscle and kidney) than in cows’ adipose tissue. Because liver and kidney are part of the Mexican diet, the social relevance of this finding is obvious. The information obtained through the analysis of food items is proving that DDTs are being transported among environmental media, and that they are being incorporated into the human food chain. Therefore, surveillance of DDTs in food articles deserves special attention. 2.5 Air

Because DDTs have a Henry’s Law constant value of 10–4/10–5 atm m3 mol, they are considered moderate volatile compounds [5]. Therefore, these compounds can be transported by air, either in the gaseous phase or adsorbed to atmospheric particles [5]. Photodegradation of DDT occurs slowly; thus, residues of these pesticides are ubiquitous in the atmosphere, although at lower concentrations. In a malaria area two issues are of concern: indoor air pollution and aerial transport and deposition. The use of DDT in the anti-malaria program causes indoor air contamination. It has been reported that, in Germany, indoor air pollution is a pathway of exposure to DDT [29]. However, the exposure scenario in malarious areas needs further research as, in these areas, the amount of DDTs in indoor air would depend on the indoor degradation rate, but it will also depend on household ventilation, which in tropical areas, is greater than in non-tropical countries like Germany.

378

F. Díaz-Barriga et al.

Aerial transport of compounds from sources and subsequent deposition is the major pathway for introduction of many contaminants, including DDTs. Airborne chemicals in the vapor phase are accumulated on plant surfaces in cuticle waxes, which are present in all terrestrial plants. This property of plants has been used to monitor spatial and temporal variations in ambient concentrations of DDT around the world [30–32]. Another important topic in relation to semivolatile compounds is the contamination of pastures. For example, residues of DDE in milk of cows grazing in pastures with contaminated soil were greatest when soil moisture and temperature conditions would tend to favor DDE volatilization (these are the normal conditions in the malarious areas) [33]. Concentrations of DDT may be higher in soil, but normally DDE levels would be higher in grass and milk, a finding that can be attributed to the eight-fold higher vapor pressure of DDE [33]. Therefore, considering the ambient conditions of tropical areas, and recognizing that there are no Mexican data in relation to DDT concentrations in air, it would be important to start a monitoring program of DDTs in indoor air, plant surfaces and pastures. The air pathway may be relevant, not only for the transportation of DDTs through long distances, but also for human exposure.

3 Human Exposure to DDT A biomarker of exposure is a xenobiotic substance or its metabolite (s), that is measured within a compartment of an organism. The preferred biomarkers of exposure are generally the substance itself or substance-specific metabolites in readily obtainable body fluid(s) or excreta. DDT and its metabolites DDD, DDE, DDA, and MeSO2-DDE (3-methylsulphonyl-DDE), can be measured in adipose tissue, blood serum, urine, feces, semen, or breast milk. In this document, we shall present data for breast milk, blood serum and adipose tisssue. Results have been obtained in workers (sprayers) and in non-occupational exposed individuals. Furthermore, studies have been done in women and men, in adults and children. Finally, considering that in Mexico DDT was first banned for agricultural purposes, and just recently has been eliminated from the anti-malaria program, most of the studies were done in malarious areas. 3.1 Breast Milk

Due to their lipophilic properties, DDT and its metabolites are primarily stored in fat-rich tissues and subsequently translocated and excreted through milk fat. A major concern is that milk is the first (and in some areas the only) food for the newborn child. Furthermore, it is important to take into account that in some malarious areas, the length of the lactation period may be more than two years. Concentrations of DDTs (DDT, DDD and DDE) in human milk have been shown to be higher in communities exposed to this insecticide, than in non-exposed populations (Table 6). For example, levels from a cotton area where DDT

379

DDT in Mexico Table 6. DDTs concentration in human milk (mg/kg fat) exposed vs. non-exposed areas

Compound

Cotton area *

Malarious suburban

Malarious urban

Mexico City

N p,p¢-DDE p,p¢-DDT

15 10.3 1.9

320 5.5 2.5

290 1.8 0.4

50 0.6 0.2

TOTAL DDT

13.2

8.4

2.0

0.9

* The numbers are mean values. Data in the cotton area were obtained during 1976, when DDT was been used in agricultural areas [34]. Data from the malarious area was collected between May 1994 and December 1995 [35]. The samples from Mexico City were obtained between December 1994 and June 1995 [36].

was used for agricultural purposes were similar to those obtained in samples collected in a malarious area where DDT was extensively used (malarious suburban in Table 6).And both were higher than the concentrations quantified in urban areas, where DDT has never been used (malarious urban and Mexico City). Also noteworthy, the total DDT levels were two times higher in the urban area in the malarious zone than in those from Mexico City (located in a non-malarious area). Although there has been a reduction in the use of DDT in recent years, the concentration of this insecticide in human breast milk did not decline in an exposed malarious area between 1994 and 1997 [37, 38]. This result agrees with the estimated population half-life for DDT in human milk of 4.2–5.6 years [39]. It can be expected that after the ban is completed, the concentration of DDT will begin to decrease. This fact will be important, considering that the current daily intake of DDTs through human milk consumption is higher than the World Health Organization’s Acceptable Daily Intake (ADI) of 20 mg/kg/day [40]. For example, considering a body mass of 5 kg, a milk intake of 0.85 kg/day, a proportion of fat in milk of 0.035, and a DDT concentration in milk of 10.4 mg/kg (total DDT concentration in samples collected during 1996–1997 in a suburban malarious area [38]), the estimated daily intake is three times higher than the ADI. However, if the maximum range concentration found in some studies (36.5 mg/kg [35]) is taken into account, the ADI is surpassed 11 times . 3.2 Serum

Studies done in workers showed that blood levels of DDTs are higher in persons exposed to greater amounts of DDT [8]. In Mexico, similar results were obtained in studies done in non-occupationally exposed populations. Levels in women living in Veracruz (which is located in a malarious area exposed to DDT), or in a rural malarious zone, were higher than to those registered in women living in a control zone not-exposed to DDT (Table 7). It has also been reported that, in Mexican women, DDE serum levels in mother’s blood correlated with those in umbilical cord serum [42] whereas, DDD and DDT concentrations did not have

380

F. Díaz-Barriga et al.

Table 7. DDTs concentration in human serum (mg/L) reproductive age women

Compound

Veracruz

Malarious rural

Control

p,p¢-DDE p,p¢-DDD p,p¢-DDT

14.5 0.3 1.2

7.1 1.7 5.7

1.8 0.4 1.8

The numbers are mean values. Data from Veracruz were collected between October 1997 and June 1998, they includes urban and suburban areas [41]. The samples from the malarious area in a rural zone, and from the control area (non-exposed to DDT) were obtained during 1998 in San Luis Potosi (unpublished data). Table 8. DDTs concentration in human serum (mg/L)

Compound

Children

Adults

Workers

p,p¢-DDE p,p¢-DDD p,p¢-DDT

79.3 1.7 61.1

60.8 1.3 27.1

69.7 1.5 43.3

The numbers are mean values. The samples were collected in the State of Chiapas during 1999 (unpublished data).Adults are older than 15 years old.Workers from the anti-malaria program in Chiapas were randomly selected for this study.

a significant correlation. These results demonstrate that at least DDE readily passes through the placental barrier. Veracruz and San Luis Potosi (Table 7), are malarious zones located on the coast of the Gulf of Mexico (east of Mexico). In Table 8 are depicted data from samples collected in the State of Chiapas, which is located on the Pacific coast (southwest of Mexico). Serum DDTs concentrations were higher in Chiapas. In addition, it can be observed in Table 8 that children had higher DDTs levels than adults, and even their concentrations were somewhat higher than those found in workers occupationally exposed to DDT due to their participation in the spraying of this insecticide. Our results clearly identify three high risk groups: children, pregnant women and workers. The concentration of DDE and DDT in these groups is enough higher to justify a health assessment program. Moreover, preliminary data, from samples collected in San Luis Potosi and Chiapas, show high concentrations of MeSO2-DDE, which is a metabolite that has been related with toxic effects in the adrenal glands [43, 44]. 3.3 Adipose Tissue

Adipose tissue biopsy has been used in epidemiological studies to assess chronic exposure to DDT. This is a logical choice because the DDTs are accumulated in adipose tissue due to its lipid solubility. The half-life of DDT in human adipose tissue is approximately seven years [45].

381

DDT in Mexico Table 9. DDTs concentration in human adipose tissue (mg/g lipid basis)

Compound p,p¢-DDE p,p¢-DDD p,p¢-DDT Total DDTs

Workers

Malarious

Agricola

Urban

60.9 0.9 31.0

28.9 – 8.1

18.4 0.2 0.8

6.0 0.4 1.1

104.5

38.5

21.5

8.3

The numbers are mean values. The samples from the agricultural area and urban zone (Mexico City) were collected during 1975, when DDT was being used in the former [46]. The malarious area in Veracruz was studied during 1992 [47]. Workers: the results are expressed as a geometric mean, the individuals were from the anti-malaria program in Veracruz, the samples were collected during 1995 [50].

As in serum, DDTs in adipose tissue are a good biomarker of exposure for communities exposed to DDT, both in agricultural [46] and malarious areas [47]. When compared to an urban non-exposed community [46], the levels of DDTs (especially those of DDE), were higher in the exposed population (Table 9). In the same table it can be observed that the concentrations of DDT in adipose tissue from workers of the malaria program were higher than the levels found in people living in an agricultural area or in malarious areas. In the workers, a linear model that included an index of chronic exposure, the use of protective gear, and recent weight loss explained 55% of the variation of p,p′-DDE conncentrations in adipose tissue. The index of chronic exposure was constructed according to worker position and based on the historical duration and intensity of DDT application [48]. When the concentrations of DDTs in adipose tissue were expressed by age group, two groups were identified as the most exposed. Those groups were children and elderly people [49]. The levels in elderly people can be explained by the accumulation of DDT in a chronic exposure scenario, whereas the concentration in children may be the result of an exposure to multiple pathways (soil, household dust, air, water, food, etc.). It is interesting that the group less exposed to DDT was the 0–2 years, a group that may be exposed to DDT through lactation [49]. A monitoring program of DDTs in adipose tissue is needed in order to assess the body burden, now that in Mexico this insecticide has been eliminated from the malaria program. However, due to ethical constraints, it is not always possible to obtain adipose tissue samples form healthy individuals. Therefore, alternative matrices are needed; for example, a good correlation between adipose tissue concentration and levels in human milk [50] or human serum [51] has been reported . When the geometric DDE levels in lipid bases are used for the estimation of the adipose tissue/serum DDE ratio, a value near unity is obtained [51].

382

F. Díaz-Barriga et al.

4 Health Effects Besides the well-recognized neurotoxicity of DDT, the health effects on human populations are unclear. Symptoms of acute poisoning include paraesthesia, tremors, confusion, headache, fatigue and vomiting [8]. The evidence for public health implications of the use of DDT in Mexico is scarce. Here we present a review of published and preliminary results of ongoing studies. Two different kind of studies have analyzed possible health impacts: 1) those studying declared clinical illness; and 2) the evaluation of potential health effects looking at early pre-clinical indicators of abnormalities before adverse health effects appear. Clinical and epidemiological studies provide clear evidence of effects in human populations; however, they are expensive, and difficult to conduct because of the large sample size usually required. The evaluation of potential health effects using early pre-clinical indicators is a good alternative that uses more sensitive indicators without the need to wait until the disease is manifest. 4.1 Cancer

Although it has been suggested that the estrogenic activity of DDE may be a contributing factor for development of breast cancer in women, levels of these compounds are not consistently elevated in breast cancer patients. It was initially reported that levels of p,p′-DDE were elevated in breast cancer patients (serum or tissue) versus controls [52]. More recent studies and analysis of organochlorine levels in breast cancer patients versus controls show that these contaminants are not elevated in the latter group [53–56]. The study of occupationally exposed workers has not found clear increased risks for other cancers [57]. Two case-control studies of breast cancer have been carried out in Mexico City, with conflicting results. The first study, conducted by Lopez-Carrillo et al. [58] in Mexico City, compared 141 cases of breast cancer with 141 age-matched controls. All subjects were identified at three referral hospitals between March 1994 and April 1996. The arithmetic mean of serum DDE in lipid basis was 562 ppb±676 for the cases and 505 ppb±567 for the controls. The age-adjusted odds ratios for breast cancer regarding the serum level of DDE were 0.69 (95% confidence interval, 0.38–1.24) and 0.97 (CI, 0.55–1.70) for the contrasts between tertile 1 (lowest level) and tertiles 2 and 3, respectively. These estimates were unaffected by adjustment for body mass, accumulated time of breast-feeding and menopause, and other breast cancer risk factors. These results do not lend support to the hypothesis that DDT is causally related to breast cancer. The second study conducted by Romieu et al. [59] compared 120 cases and 126 controls, selected from six hospitals in Mexico City, from 1989 to 1995. Serum DDE levels in lipid basis were higher among cases (mean=3840 ppb±5980) than among controls (mean=2510 ppb±1970 ).After adjusting for age, age at menarche, duration of lactation, Quetelet index, and serum DDT levels, serum DDE levels were positively related to the risk for breast cancer (adjusted ORQ1 – Q2 =1.24),

DDT in Mexico

383

(CI, 0.50–3.06; ORQ1 – Q3 =2.31, 95 percent, CI, 0.92–5.86; ORQ1 – Q4 =3.81, CI, 1.14–12.80). The increased risk associated with higher serum DDE levels was more apparent among postmenopausal women (ORQ1 – Q4 = 5.26, 95%, CI, 0.80–34.30). Serum DDT level was not related to the risk for breast cancer. In addition to the differences in the comparison of cases and controls, the difference in the serum DDE levels among the women studied is remarkable. Participants from both studies came from similar hospitals, and there were no apparent differences between case and control selection that could explain this divergence. Differences in laboratory procedures is the most feasible explanation. 4.2 Endocrine Disruption

DDT is known to have adverse effects on wildlife via endocrine disruption. Clear effects include thinning of the eggshell, feminization, reproduction impairment and development effects [60]. In Mexico two studies in humans have reported findings in this area. Gladen and Rogan [61] found that DDE might affect women’s ability to lactate in a study conducted in an agricultural town in northern Mexico. Two hundred and twenty-nine women were followed from childbirth until weaning or until the child reached 18 months of age. DDE was measured in breast milk samples taken at birth, and women were followed to see how long they lactated. Median duration was 7.5 months in the lowest DDE group and 3 months in the highest. The effect was confined to those who had lactated previously – but not for first pregnancies – and it persisted after statistical adjustment for other factors. Rodriguez et al. [62] conducted a study aimed at determining the capability of long-term exposure to DDT of altering the normal endocrine function of the hypothalamus-hypophysis-gonads axis in humans. This included 70 workers dedicated to control malaria in the State of Guerrero, Mexico. The main activities of these workers were the application of pesticides, detection of malaria cases and promotion of preventive measures for control vectors. The average time of exposure to technical grade DDT was 25 years (range: 4–35), their last exposure being 5 months before sampling. An interview gathered information on the occupational history, reproductive performance, life styles and other relevant factors. Blood and urine samples were collected to measure serum levels of DDT and metabolites as well as levels of LH, FSH, prolactine, and testosterone. Participants ranged in age from 22 to 69 years, and had been employed in the sanitation campaign from 4 to 37 years. Ninety-seven percent of the participants were sprayers of DDT at some time in their occupational history, and 15% are current sprayers. Average levels of DDT and metabolites expressed as µg/g of extractable lipids were: total DDT, 60.1; p,p′-DDE, 37.41; p,p′-DDT, 21.52; p,p′-DDD, 1.07 and o′p-DDT 0.11. Results show a positive association of LH and FSH with DDT metabolites. An increase of 10 mg/g of p,p′-DDE was associated with an increase of 1.95 UI/L in LH (p=0.01); and 1.10 UI/L per each 10 mg/g of p,p′-DDT (p=0.02). FSH increased 1.09 UI/L per each 10 mg/g of p,p′DDT (p=0.03). There was a negative association of DDE with testosterone,

384

F. Díaz-Barriga et al.

especially for those participants under 55 years of age. These associations suggest direct toxicity to the testicles, especially the Leydig cells, as observed with antineoplasic drugs. 4.3 Genotoxicity

Some studies have reported genotoxic effects in humans heavily exposed to DDT [8]. Therefore, this area has been studied in Mexico. Studies were done in workers from the control program of malaria, and in women living in malarious areas. Herrera et al. [63] evaluated chromosomal translocations in a sample of the above-mentioned workers. Nineteen male sprayers (median age 46 years; range: 22–64), working in campaigns to control malaria vectors in the state of Guerrero, Mexico were included in this study. DDT data obtained in a previous study from eleven individuals, 5 women and 6 men, living in Mexico City, and occupationally unexposed to DDT were used as reference group. Chromosomal aberrations in lymphocytes were analyzed with a chromosome painting technique, a highsensitivity technique for detecting complex chromosomal translocations. o,p′DDT was the only isomer significantly associated with the frequency of chromosomal translocations (p=0.003). Individuals presenting the higher levels of o,p′-DDT in serum (>0.79 mg/g fat; n=4) had a mean frequency of chromosomal translocations (5.1¥1000 metaphases), two times higher than that observed in workers occupationally exposed to 0.5 Gy of radiation.A positive relationship between the duration of exposure to DDT, measured as years working for the vector control program, and chromosomal translocations was observed (Fig.2). These results suggest an increased risk for diseases with a genetic component, such as cancer. Yañez et al. [64] evaluated the association of blood DDT levels and DNA damage using the single cell gel electrophoresis assay. A group of 53 postpartum women were selected from two different areas in San Luis Potosí to assure different exposure levels, one with antecedents of malaria and DDT spraying and the other without malaria. Mean and range levels of DDT, DDE and DDD in whole blood were 5.57 ppm (0.02–20.69), 6.24 ppm (0.04–39.16) and 1.16 ppm (0.01–5.63), respectively. The significant correlation of DNA damage, measured as DNA migration with the logarithm of DDT, DDE and DDD was 0.60, 0.62 and 0.43, respectively. Fig. 3 shows the shape of the association of DNA migration with DDE concentration in whole blood, as obtained by the regression: DNA migration =71.58+7.62(logDDE), this association was not modified by age, smoking habits, nutrition or occupation. This observational finding in the epidemiological study with postpartum women was reevaluated in an in vitro study. Human blood cells were exposed to three doses of DDT, DDD and DDE. DNA damage was assessed by two different techniques: single-cell electrophoresis and flow cytometry. Results obtained by either technique showed that DNA damage was induced by the three organochlorides and a dose-response was observed with DDT. The data suggest a DDT-induced DNA fragmentation, and this outcome was also observed with DDE and DDD.

DDT in Mexico

385

Fig. 2. Frequency of chromosomal translocations in peripheral lymphocytes from 19 individuals occupationally exposed to technical grade DDT for a period of 4 to 35 years. Source: Herrera et al. (2000)

Fig. 3. DNA migration as evaluated in the single-cell gel electrophoresis assay of a group of 53 postpartum women in San Luis Potosí vs. DDE concentrations in whole blood. ** Fitted values obtained by: DNA migration=71.58+7.62(logDDE)

386

F. Díaz-Barriga et al.

5 Final Comments During the last decade Mexico has achieved a great reduction in DDT use. Current practices to control malaria in Mexico with alternative methods will reduce human and environmental exposures to DDT and its metabolites. New practices are being evaluated and community participation has been improved, which will make malaria control more sustainable. However, as a result of its persistence, the exposure to DDT will continue even now that it has been replaced with pyrethroids. Therefore, more studies are needed, not only to assess the biological effects associated with the chronic exposure to DDT, but now, to investigate the interactions of DDT with pyrethroids. For example, it has been reported in animals, that a neonatal exposure to DDT induces increased susceptibility to pyrethroid (bioallethrin) exposure at adult age [65].

6 References 1. Stapleton DH (1998) Parassitologia 40:149 2. North American Commission for Environmental Cooperation (1998) North American Regional Action Plan on DDT. In: The Sound Management of Chemicals Initiative. Regional Commitments and Action Plans. Montreal, Canada, p 67 3. Fernández de Castro J (1988) Panorama histórico y epidemiológico del paludismo en México. Secretaría de Salud, México 4. Implementation Plan to Control Malaria with Alternatives Methods to DDT in Mexico 1999–2001. Government of Mexico, January 1999 5. Ritter L, Solomon KR, Forget J, Stemeroff M, Leary CO (1995) Persistent organic pollutants. An assessment report on: DDT, aldrin, dieldrin, endrin, chlordane, heptachlor, hexachlorobenzene, mirex, toxaphene, polychlorinated biphenyls, dioxins and furans. International Program on Chemical Safety 6. INE-Semarnap (1997) Programa de gestión ambiental de sustancias tóxicas de atención prioritaria. Instituto Nacional de Ecología, Secretaría de Medio Ambiente, Recursos Naturales y Pesca, p 48 7. Mitra J, Raghu K (1998) Bull Environ Contam Toxicol 60:585 8. ATSDR (1994) Toxicological Profile for DDT, DDE, and DDD. Agency for Toxic Substances and Disease Registry, Atlanta, Georgia 9. Zeng EY, Yu CC, Tran K (1999) Environ Sci Technol 33:392 10. Eichelberger JW, Lichtenberg JJ (1971) Environ Sci Technol 5:541 11. Susarla S, El Hefnawy MM, Masunaga S,Yamashita N,Yonezawa Y, Rizk MMS, Urushigawa Y (1997) Toxicol Environ Chem 62:149 12. Noreña-Barroso E, Zapata-Pérez O, Ceja-Moreno V, Gold-Bouchot G (1998) Bull Environ Contam Toxicol 61:80 13. Vázquez-Botello A, Ponce-Vélez G, Toledo A, Díaz-González G, Villanueva S (1992) Ciencia y Desarrollo XVII (102):28 14. Botello AV, Díaz G, Rueda L, Villanueva SF (1994) Bull Environ Contam Toxicol 53:238 15. Phuong PK, Son CPN, Sauvain JJ, Tarradellas J (1998) Bull Environ Contam Toxicol 60:347 16. Yamashita N, Masunaga S, Rizk MS, Urushigawa Y (1997) Toxicol Environ Chem 58:151 17. Brown LR (1997) Arch Environ Contam Toxicol 33:357 18. Albert LA (1996) Rev Environ Contam Toxicol 147:1 19. Fisher SW (1995) Rev Environ Contam Toxicol 142:87 20. Gold-Bouchot G, Silva-Herrera T, Zapata-Pérez O (1995) Bull Environ Contam Toxicol 54:554

DDT in Mexico

21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57.

387

Gutiérrez-Galindo EA, Flores-Muñoz G, Villaescusa-Celaya JA (1988) Cienc Mar 14:91 Caldas ED, Coelho R, Souza LC, Silva SC (1999) Bull Environ Contam Toxicol 62:199 Abd-Allah AM, Ali HA (1994) Toxicol Environ Chem 42:107 Zhou HY, Cheung RY, Wong MH (1999) Arch Environ Contam Toxicol 36:424 Waliszewski SM, Pardío VT,Waliszewski KN, Chantiri JN,Aguirre AA, Infanzón RM, Rivera J (1998) Fresenius Envir Bull 7:238 Waliszewski SM, Pardío-Sedas VT,Waliszewski KN, Chantiri-Pérez JN, Infanzón-Ruiz RM, Rivera J (1996) Rev Int Contam Amb 12:53 Waliszewski SM, Infanzón RM, Rivera J (1995) Fresenius Envir Bull 4:342 Waliszewski SM, Pardío VT, Waliszewski KN, Chantiri JN, Infanzón RM (1996) J AOAC Int 79:784 Neuber K, Merkel G, Randow FFE (1999) Toxicol Lett 107:189 Eriksson G, Jensen S, Kylin H, Strachan W (1989) Nature 341:42 Jensen S, Eriksson G, Kylin H (1992) Chemosphere 24:229 Strachan W, Eriksson G, Kylin H, Jensen S (1994) Environ Toxicol Chem 13:443 Fries GF (1995) Rev Environ Contam Toxicol 141:71 Albert L, Vega P, Portales A (1981) Pest Monitor J 15:135 Waliszewski SM, Aguirre AA, Infanzon RM, Rivera J, Infanzon RM (1998) Fresenius Envir Bull 7:709 Torres-Arreola L, Lopez-Carrillo L, Torres-Sanchez L, Cebrian M, Rueda C, Reyes R, LopezCervantes M (1999) Arch Environ Health 54:124 Waliszewski SM, Pardio Sedas VT, Chantiri JN, Infanzon RM, Rivera J (1996) Bull Environ Contam Toxicol 57:22 Pardio VT, Waliszewski SM, Aguirre AA, Coronel H, Burelo GV, Infanzon RM, Rivera J (1998) Bull Environ Contam Toxicol 60:852 Smith D (1999) Int J Epidemiol 28:179 Lu FC (1995) Reg Toxicol Pharmacol 21:352 Waliszewski SM,Aguirre AA, Benitez A, Infanzon RM, Infanzon R, Rivera J (1999) Bull Environ Contam Toxicol 62:397 Waliszewski SM, Aguirre AA, Infanzon RM (1999) Fresenius Envir Bull 8:171 Jönsson C, Lund BO, Bergman A, Brandt I (1992) Reprod Toxicol 6:233 Jönsson CJ, Rodriguez-Martinez H, Lund BO, Bergman A, Brandt I (1991) Fund Appl Toxicol 16:365 Woodruff T, Wolff MS, Davis DL, Hayward D (1994) Environ Res 65:132 Albert L, Méndez F, Cebrian ME, Portales A (1980) Arch Environ Health 35:262 Waliszewski SM, Pardio-Sedas VT, Infanzon RM, Rivera J (1995) Bull Environ Contam Toxicol 55:43 Rivero-Rodríguez L, Borja-Aburto VH, Santos-Burgoa C, Waliszewski SM, Ríos C, Cruz V (1997) Environ Health Perspect 105:98 Waliszewski SM, Pardio V, Chantiri J, Aguirre A (1996) Fresenius Envir Bull 5:357 Waliszewski SM, Aguirre AA, Infanzon RM, Benitez A, Rivera J (1999) Bull Environ Contam Toxicol 62:685 López-Carrillo L, Torres-Sánchez L, López-Cervantes M, Blair A, Cebrián ME, Uribe M (1999) Environ Res 81:142 Falck F, Ricci A, Wolff M, Godbold J, Deckers P (1991) Arch Environ Health 1991:47:143 Zheng T, Holford TR, Mayne ST, Ward B, Carter D, Owens PH, Dubrow R, Zahm SH, Boyle P, Archibeque S, Tessari J (1999) Am J Epidemiol 150:453 Helzlsouer KJ,Alberg AJ, Huang HY, Hoffman SC, Strickland PT, Brock JW, Burse VW, Needham LL, Bell DA, Lavigne JA, Yager JD, Comstock GW (1999) Cancer Epidemiol Biomark Prev 8:525 Liljegren G, Hardell L, Lindstrom G, Dahl P, Magnuson A (1998) Eur J Cancer Prev 7:135 Hunter DJ, Hankinson SE, Laden F, Colditz GA, Manson JE, Willett WC, Speizer FE, Wolff MS (1997) New Engl J Med 337:1253 Cocco PL, Blair A, Congia P, Saba G, Flore C, Ecca MR, Palmas C (1997) Arch Environ Health 52:299

388

F. Díaz-Barriga et al.: DDT in Mexico

58. Lopez-Carrillo L, Blair A, Lopez-Cervantes M, Cebrian M, Rueda C, Reyes R, Mohar A, Bravo J (1997) Cancer Res 57:3728 59. Romieu I, Hernandez-Avila M, Lazcano E, Weber JP, Dewailly E. Am J Epidemiol in press 60. Olsson PE, Borg BB, Brunström B, Hakansson H, Klasson-Wehler E (1998) Endocrine disrupting substances. Swedish Environmental Protection Agency, Report 4859 61. Gladen BC, Rogan WJ (1995) Am J Publ Health 85:504 62. Rodriguez TR (1999) Master’s thesis. CINVESTAV-IPN, Mexico 63. Herrera LA, Borja-Aburto VH, Cebrián M, Rodríguez T, Verdorfer I, Gebhart E, OstroskyWegman P (2000) Submitted 64. Yañez L (2000) PhD thesis. Autonomous University of San Luis Potosi, Mexico 65. Eriksson P, Johansson U, Ahlbom J, Fredriksson A (1993) Toxicology 77:21

CHAPTER 15

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities Hans-Ulrich Hartenstein Energy and Environmental Consultants (E&EC) GmbH, Postfach 3507, 51535 Waldbröl, Germany E-mail: [email protected]

Polychlorinated dibenzo-p-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF) are formed unintentionally as byproducts of many combustion and industrial thermal processes. Today PCDD/PCDF is known as one of the most toxic group of anthropogenic organic substances. Furthermore, it is perceived to be carcinogenic. Although many questions about the toxicological effects of PCDD/PCDF are still unanswered, all possible reduction measures should be taken to minimize their release into the environment. PCDD/PCDF are also included in the United Nations’ developing international treaty on the ban of several Persistent Organic Pollutants (POPs), and thus are of particular global concern. Consequently, reduction technologies for PCDD/PCDF are needed not only for densely populated urban areas but also in rural regions where the world population’s food is produced. The uptake of PCDD/PCDF via the food chain must be considered to be the major pathway of exposure for humans as well as for animals. After a brief presentation of the regulatory and technical background, this chapter will give an overview of the formation routes of PCDD/PCDF in combustion and other industrial thermal processes. Some of the most important sources of PCDD/PCDDF emissions are also described. Second, the most commonly applied PCDD/PCDF control technologies will be presented for various types of waste incinerators and other industrial thermal processes to demonstrate today’s state-of-the-art flue gas cleaning technology. Modern municipal solid waste (MSW) combustors, hazardous waste incinerators, and sewage sludge incinerators as well as an iron ore sintering facility provide actual examples of full-scale systems in operation. The final part of the chapter will give an outlook towards newly developed PCDD/PCDF abatement technologies for more economical PCDD/PCDF reduction. Keywords: PCDD/PCDF, De-novo-synthesis, PCDD/PCDF emission control equipment, PCDD/PCDF removal efficiency

1

Legislative Activities . . . . . . . . . . . . . . . . . . . . . . . . . 390

2

Dioxin and Furan Formation Routes

3

Sources of Dioxins and Furans

4

Dioxin Removal Technologies . . . . . . . . . . . . . . . . . . . . 400

4.1 4.2

Adsorbent Injection . . . . . . . . . . . . . . . . . . . . . . . . . 401 Entrained Flow Reactor . . . . . . . . . . . . . . . . . . . . . . . 402

. . . . . . . . . . . . . . . . 392

. . . . . . . . . . . . . . . . . . . 398

The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

390

H.-U. Hartenstein

4.3 4.4

Activated Carbon Reactor . . . . . . . . . . . . . . . . . . . . . . 403 Tail-End Selective Catalytic Reaction . . . . . . . . . . . . . . . . 405

5

Examples of Full Scale Plants

5.1 5.2 5.3 5.4 5.5 5.6 5.7

MVB Hamburg . . . . MHKW Leverkusen . MVA Stapelfeld . . . . AVI Gevudo Dordrecht RVA Böhlen . . . . . . VERA Hamburg . . . Sinter Belt 2 Thyssen .

6

New Dioxin Control Processes

7

References

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . . . . . . . . . . . . . . . 407 . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

407 409 411 413 414 416 417

. . . . . . . . . . . . . . . . . . . 421

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422

1 Legislative Activities Due to a highly developed environmental sensitivity of the public in Germany and other Central European countries, more and more regulations regarding air and water quality are being implemented. Especially air emission regulations for power stations, waste incinerators, crematoria, and other thermal industrial processes created a new market for air pollution control equipment specifically for the effective control of PCDD/PCDF. Around 1990 the governments of Austria, Germany, the Netherlands, Sweden, Switzerland, and other European countries drastically tightened the emission guidelines for many pollutants through implementation of respective legislation. The German 17. BImSchV (17th Implementation Directive to the Federal German Immission Protection Act), which formed the basis for the current legislation within the European Union (EU), is used as an example to outline the implications of such legislation for all kinds of waste incineration plants. Among other things, the minimum acceptable combustion operating conditions in terms of flue gas residence time above a minimum temperature as well as maximum permissible emission concentrations for many air pollutants were fixed. One of the most significant achievements of this new directive was the first time introduction of an emission limit for PCDD/PCDF of 0.1 ng I-TEQ/Nm3. Since the group of PCDD/PCDF consists of 210 individual compounds (75 PCDD congeners and 135 PCDF congeners) with different levels of toxicity, the group is commonly referred to in concentration numbers of toxic equivalents. Several different methods for determining toxic equivalents were defined. Table 1 compares the three most commonly used methods for the calculation of toxic equivalents, namely the International Toxicity Equivalency Factors – I-TEF, the World Health Organization’s Toxicity Equivalent Factors – WHO-TEF, and the German Umweltbundesamt/Bundesgesundheitsamt (Federal Environment Agency/Federal Health Agency) Toxicity Equivalency Factors – UBA/BGA-TEF.

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

391

Table 1. Commonly used toxicity equivalent factors

Congener

I-TEF

WHO-TEF

UBA/BGA

2,3,7,8-Cl4DD 1,2,3,7,8-Cl5DD 1,2,3,4,7,8-Cl6DD 1,2,3,6,7,8-Cl6DD 1,2,3,7,8,9-Cl6DD 1,2,3,4,6,7,8-Cl7DD Cl8DD 2,3,7,8-Cl4DF 1,2,3,7,8-Cl5DF 2,3,4,7,8-Cl5DF 1,2,3,4,7,8-Cl6DF 1,2,3,6,7,8-Cl6DF 1,2,3,7,8,9-Cl6DF 2,3,4,6,7,8-Cl6DF 1,2,3,4,6,7,8-Cl7DF 1,2,3,4,7,8,9-Cl7DF Cl8DF

1 0.5 0.1 0.1 0.1 0.01 0.001 0.1 0.05 0.5 0.1 0.1 0.1 0.1 0.01 0.01 0.001

1 1 0.1 0.1 0.1 0.01 0.001 0.1 0.05 0.5 0.1 0.1 0.1 0.1 0.01 0.01 0.0001

1 0.1 0.1 0.01 0.1 0.01 0.001 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.01 0.01 0.001

It is interesting to note the slight differences, even though the use of the I-TEF seems to prevail in most countries (see Table 1). Consequently, the most commonly used nomenclature based on I-TEF is expressed in units of ng/m3 at STP. However, there are numerous different standards for the term STP (standard temperature and pressure), and other parameters such as the O2-content and the moisture content for the definition of a m3. Often seen are the terms Nm3 (Normal m3 – most commonly used in Europe), Sm3 (Standard m3 – most commonly used in the United States and some Asian countries), and Rm3 (Reference m3 – used in Canada). All standards are on a dry basis; thus, the moisture content of the flue gas and the resulting dilution is eliminated, which leads to an increase in the reported over the measured I-TEF concentration. An Nm3 is based on 11 vol.% O2 , 1013 mbar, and 0 °C; a Sm3 is based on 7 vol.% O2 , 1 atm, and 77 F (25 °C); and a Rm3 is based on 11 vol.% O2 , 1 atm, and 20 °C. Consequently, the results reported can only be compared in any meaningful way as long as the standard used to define a m3 is also reported.All the concentration figures to be compared must be converted to a uniform standard before they are compared to make such a comparison meaningful. The following example illustrates this statement. The seemingly identical numerical values of 0.10 ng I-TEF/Nm3 and 0.10 ng I-TEF/Sm3 are in reality two greatly different concentrations. In order to allow a comparison of both values, both need to be brought to the same basis, i.e., ng I-TEF/Nm3. Thus, the following steps are necessary: [(21 vol.% O2 – 7 vol.% O2)/ (21 vol.% O2 – 11 vol.% O2)]=1.4 2. Temperature Correction: (298 K/273 K)=1.1 3. Pressure Correction: not necessary, since 1 atm=1013 mbar

1. Oxygen Correction:

392

H.-U. Hartenstein

0.10 ng I-TEF/Sm3 ¥(1.1/1.4 Sm3/Nm3) =0.78 ng I-TEF/Nm3 The example shows that the real difference of the two numerically identical values amounts to more than 20%! Thus, it is of crucial importance for data quality and reporting to ensure the same basis before comparing any data. It is equally important to understand fully and appreciate which basis was chosen as a standard for a respective piece of legislation. Without this information, no valid understanding of the technological challenge to achieve compliance of a particular facility can be developed. Consequently, no correct evaluation of and selection from the numerous choices of potentially applicable process technology solutions is possible. The 17th BImSchV also requires certain combustion conditions to be maintained, namely a minimum flue gas temperature of 850 °C after the last combustion air injection combined with a flue gas residence time at or above said temperature of at least 2 s. Also an oxygen concentration of at least 6 vol.% must be maintained at all times to avoid starved combustion conditions and potentially incomplete combustion at all times. These requirements to insure good combustion practice are mandatory for the incineration of MSW, sewage sludge, and other such wastes, which do not contain significant quantities of halogenated hydrocarbons. In case of other waste containing mostly chlorinated hydrocarbons, the minimum combustion temperature must be raised to at least 1200 °C with the other two requirements regarding residence time and oxygen concentration remaining unchanged. The reason for these requirements is the significant thermal stability of many halogenated hydrocarbons such as PCDD/PCDF. In order to effectively destroy these compounds, the flue gas must be exposed to sufficiently high temperatures over a long enough period of time while the availability of oxygen for the oxidation is ensured at all times. However, the 17th BImSchV also allows for combustion conditions different from the ones required by the directive as long as individual measurements at the individual facility claiming an exception provide clear proof that the emission concentrations, especially of PCDD/PCDF, polycyclic aromatic hydrocarbons (PAHs), and polychlorinated biphenyls (PCBs) are not higher than at the combustion conditions specified in the directive. This provision allows the operator to optimize the actual operation of the individual facility based on economical and other considerations without compromising on environmental compliance. Most hazardous waste incineration facilities which burn large quantities of halogenated hydrocarbons and other POPs in Germany take full advantage of this flexibility given by the 17th BImSchV in order to lower their operating cost as well as the wear of the refractory lining in the case of a rotary kiln furnace. 4. Overall Correction:

2 Dioxin and Furan Formation Routes After the first measurements at MSW incinerators revealed rather high PCDD/ PCDF emissions at the stack, it was believed that these toxic substances were formed in the furnace. Today, much more about the formation mechanisms of

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

393

PCDD/PCDF is known. Consequently, a well designed and operated combustion furnace has been recognized to almost completely destroy the incoming dioxins and furans. Figure 1 details the PCDD/PCDF destruction efficiency of a combustion process which ensures adequate temperature and sufficient residence time. Nevertheless, the measurable emissions from waste incineration plants were still not considered acceptable to protect public health and the environment. Since it was unclear where and how the PCDD/PCDF formation occurred, a lot of measurements and intensive research was performed.After nearly two decades the answer is clear, although several details still have to be investigated. The formation of relevant PCDD/PCDF concentrations in waste incineration plants takes place downstream of the furnace/combustion chamber in the back end of the heat recovery steam boiler and also during dust removal in case of an electrostatic precipitator (ESP) being installed. Figure 2 reveals the fact that two basic mechanisms of reformation of PCDD/PCDF after the combustion process occur during cooling of the flue gas. This phenomenon is virtually independent of the actual destruction efficiency of the combustion process and is responsible for the bulk of the PCDD/PCDF concentrations in flue gases downstream of the furnace. The first mechanism occurs between 300–800 °C and is a homogenous gas phase reaction. Dioxins and furans are formed from so-called “precursors” or

Fig. 1. Thermal destruction of organic substances in air

394

Fig. 2. Dioxin/furan destruction and formation in an incineration plant

Fig. 3. Possible formation routes for dioxins/furans

H.-U. Hartenstein

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

395

“pre-dioxins” (Fig. 3). Such precursors are, for example, polychlorinated benzenes, phenols, and biphenyls. However, this mechanism does not contribute significantly to the overall PCDD/PCDF concentration found after the furnace. The second mechanism of reformation is the so called de-novo-Synthesis of PCDD/PCDF. It is reasonable to assume that de-novo-Synthesis contributes the predominant portion to the total PCDD/PCDF emissions from modern MSW incineration plants. For older MSW incineration plants or special incinerators (i.e., hospital waste incinerators or crematoria) the individual share of the two formation mechanisms can yet widely differ. Due to very poor combustion conditions, extremely high emission values of PCDD/PCDF have been observed. Two main theories are commonly accepted concerning the de-novo-Synthesis reaction process. Both theories assume PCDD/PCDF formation as a heterogeneous gas-solid phase reaction on the surface of fly ash particles. Inorganic chlorides such as NaCl or HCl will form elemental chlorine (Cl2) in the presence of oxygen in conjunction with catalytic active metallic chlorides like CuCl2 or FeCl3, according to the well known Deacon reaction shown in Fig. 4. Subsequently, Cl2 reacts with aromatic components in the flue gas or fractions from the carbon in the fly ash to form chlorinated organic compounds and fragments, which combine to dioxins or furans in a next reaction step. The first theory [1] postulates a dualistic principle of catalytic PCDD/PCDF destruction depending on temperature and oxygen concentration (dechlorination/hydrogenation) and catalytic PCDD/PCDF formation by means of chlorine. PCDD/PCDF destruction by dechlorination increases exponentially with temperature whereas PCDD/PCDF formation is limited with increasing temperature by the reaction velocity of chlorine formation. Due to the mentioned influencing factors such as chlorine concentration and carbon catalytic surface activities, a temperature range results where the PCDD/PCDF destruction is actually substantially higher than the PCDD/PCDF

Fig. 4. Deacon reaction and chlorination of aromatics

396

H.-U. Hartenstein

Fig. 5. Dualistic principle of dioxin formation and destruction

formation (Fig. 5). Thus, a well designed, operated, and maintained waste incinerator always acts as an ideal overall sink for PCDD/PCDF due to an overall destruction efficiency of over 99.99%. The second theory [2] formulates a limiting control-mechanism for the chlorination reaction of organic compounds. The in-situ formation of Cl2-gas, according to the copper catalyzed Deacon reaction, increases with decreasing temperature, increasing oxygen concentration, and decreasing water vapor concentration. The kinetics of both reactions, i.e., formation of chlorine (Cl2) and the chlorination of aromatics, are enhanced with an increase in temperature. These reactions indicate that aromatic ring structures and chlorine present in the flue gas are the ingredients potentially needed for the formation of PCDD/PCDF. However, the chlorination of aromatics is limited when sulfur is present in the flue gas. If an excess of sulfur dioxide relative to Cl2 exists in the flue gas, the competitive oxidation reaction of SO2 to SO3 dominates.As detailed in Fig. 6, chlorine is effectively intercepted by SO2 and consequently not present in sufficient quantities any more for the formation of chlorinated aromatics. Following this theory, a chlorine/sulfur ratio in the flue gas of less than approximately 0.1 would be sufficient to largely prevent PCDD/PCDF formation, because the chlorine interception reaction should predominate (Fig. 7). Both theories of the de-novo-Synthesis have been supported by measurements and experiments. Therefore, it cannot be definitely decided if one of them is incorrect. The de-novo-Synthesis is most active in a temperature range of 200–500 °C with a maximum at approximately 350 °C. From the theoretical knowledge about PCDD/PCDF destruction during waste combustion and PCDD/PCDF reformation in the waste heat recovery steam boiler, several equipment design and combustion operating principles have been

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

397

Fig. 6. Deacon reaction and alternative reaction routes of chlorine

Fig. 7. Cl/S Ratio, dioxin emission levels, and deacon equilibrium constant

derived as primary measures to minimize PCDD/PCDF emissions from the incinerator before entering the flue gas cleaning plant. Today, such primary measures are consequently applied to the design and construction of new plants. It has also been suggested that these measures should be combined with the addition of inhibitor substances into the boiler to suppress the Deacon reaction, leading to the unwanted liberation of chlorine. Nevertheless, an emission limit of 0.1 ng I-TEQ/Nm3 cannot be ensured without additional flue gas cleaning equipment for the removal of remaining PCDD/PCDF.

398

H.-U. Hartenstein

3 Sources of Dioxins and Furans According to the above mentioned facts it can be concluded that not only waste combustion plants, but virtually all combustion and thermal industrial process categories in which chlorine occurs in combination with a carbon source at a temperature above 180 °C, are potential sources of PCDD/PCDF formation and emissions. Therefore, some countries in Europe such as Germany, the Netherlands and Great Britain as well as in North America (Canada and the U.S.) decided to examine most or all processes with such conditions precedent. National PCDD/PCDF inventories were established based on very extensive emission testing campaigns of all potential industrial sources. Their individual contribution to the total national emissions of PCDD/PCDF were determined and legislative control measures were proposed. Among these processes were such categories as: – – – – – – – –

waste incinerators of all kinds; coal, oil, and wood combustors; vehicle traffic; most metallurgical industries, especially iron ore sintering processes; high-temperature processes; accidental fires; chemical production processes; and numerous others.

(see Table 2 [4]). Some results of intensive measurements as well as information from the literature are presented in Tables 2 and 3. Some ten years ago, the combustion of municipal solid waste was among the most important source of PCDD/PCDF. The introduction of stringent emission limits in the early 1990s led to a dramatic change in this situation and, today, waste incineration has become a rather insignificant source. Up to then, measured PCDD/PCDF concentrations in the flue gas at the stack of MSWI plants varied between 1 and 92 ng I-TEQ/Nm3 [3]. Since then most of the existing facilities were retrofitted with extensive flue gas clean-

Table 2. Emission sources of PCDD/PCDF in Germany [4]

Waste incineration (all types) Metal industry Power stations (all fossil fuels) Industrial combustion Other thermal industrial processes Domestic coal and wood combustion Crematoria Vehicle traffic

1989/90 g I-TEQ/a

1994/95 g I-TEQ/a

Estimate 1999/2000 g I-TEQ/a

400 740 5 20 1 20 4 10

30 240 3 15 <1 15 2 4

< 1 <40 < 3 <15 < 1 10 < 1 < 1

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

399

Table 3. Emission sources of PCDD/PCDF in Great Britain [5]

MSW combustion Chemical waste combustion Hospital waste combustion Iron ore sintering plants Iron and steel production Non ferrous metal industry Industrial coal combustion Domestic wood combustion Domestic coal combustion Crematoria Vehicle traffic a b c

Present estimate (1995)

Future projected estimate

Estimate quality

g I-TEQ/a

g I-TEQ/a

(H=high, M= medium, L=low)

460–580 1.5–8.7 18–88 29–54 3–41 5–35 5–67 2–18 20–34 1–35 1–45

15 a –1.5 b 0.3 b 5 29–47 14 10 5–67 2–18 c 20–34 c 1–35 c 1–45 c

H/M M/M H/M M/L M/L M/L H/M L/L L/L H/L H/L

Since December 1996, IRP 5/3 with 1 ng I-TEQ/Nm3 has been valid. Assumed to comply with EU Draft Directive COM (92), Final-SYN 406 (0.1 ng I-TEQ Nm3). No reduction has been assumed. This is not necessarily valid (for example with Crematoria and Traffic).

ing systems. The remaining facilities were either shut down and demolished or rebuilt completely. Of course, all new plants fully comply with all the new standards (see Table 3). It is worth noting that the German 17th BImSchV does not differentiate between the various types of wastes regarding their emissions from incineration. Thus, the emission limits for municipal waste, hospital waste, sewage sludge, hazardous waste, wood waste, and other types of waste are identical. Consequently, the PCDD/PCDF emissions from all types of waste combustion were lowered significantly, too. In the European Union an emission regulation similar to the 17th BImSchV has been introduced for hazardous waste combustors in 1994, and for municipal waste combustors in 1999. Consequently, the PCDD/PCDF emission limit within the 15 countries of the European Union is now 0.1 ng I-TEQ/Nm3 (at 11% O2 , 1013 mbar, 0 °C, dry basis). Today, the most important PCDD/PCDF source in Europe is the metallurgical industry, especially the iron ore sintering processes in the primary steel industry and similar processes in the secondary non-ferrous industry.Within the iron ore sintering process, a mixture of various substances is combined and baked at high temperature (1000–1200 °C). Therefore, the crushed iron ore material is mixed with coke and transported on a moving chain grate. Burners above the grate heat the top of the material to the required temperature and cause it to ignite eventually. By drawing air through this heated mixture of coke and iron ore, the flame front is moved through the sinter bed. The heat resulting from the coke combustion causes the iron ore to agglomerate and sinter to larger particles suitable for the use in the blast furnace process. The air drawn through the sinter bed results in flue gas containing large amounts of particulate matter. Since the par-

400

H.-U. Hartenstein

ticulate matter consists mostly of iron ore it is being recovered from the flue gas and put back onto the sintering grate. Iron ore sintering plants are typically only equipped with ESPs as flue gas cleaning equipment for the recovery of the iron ore dust drawn from the bed with the air flow. Stack testing for PCDD/PCDF at such plants revealed PCDD/PCDF emissions up to 47 ng I-TEQ/Nm3. Combined with volumetric dry flow rates typically ranging from 250,000 to over 1,000,000 Nm3/h, the mass flow rates of emitted PCDD/PCDF are rather significant [4]. Although the current regulations for these processes do not yet require PDD/PCDF control equipment, such emission limits can be expected to be implemented with in the near future in Austria, Germany, and the Netherlands, and subsequently in the European Union. As shown already in Table 3, in Great Britain the emission situation is slightly different.Various municipal solid waste combustion plants were responsible for the highest PCDD/PCDF emissions until the mid-1990s. For MSW combustors the regulation IRP 5/3 has been valid since December 1996 lowering PCDD/PCDF emissions of each plant to 1.0 ng I-TEQ/Nm3. In the meantime, new plants are also designed and older ones are being retrofitted with adequate flue gas treatment technology to meet the new European Union emission limit of 0.1 ng I-TEQ/Nm3.

4 Dioxin Removal Technologies German engineering companies were among the first in the world to develop suitable PCDD/PCDF reduction techniques. Depending on the client’s individual requirements, the installed PCDD/PCDF emission control equipment is capable of lowering the emissions far below the legal emission limit. There is a great lack of space for landfilling of any kind of waste in Europe. Additionally, landfilling is considered to be a source of future pollution of air, water, and soil, which requires some kind of remediation in the future. Consequently, Germany was among the first European countries to introduce legislation requiring a carbon content of less than 5% for any material to be landfilled after 2005. This leads to the need to incinerate most of the wastes produced in the country.Also, the reutilization of the residues and the byproducts from flue gas treatment became worth to consider, although most of the commonly known byproducts are considered hazardous waste by law. Therefore, German companies developed and built recovery processes for HCl (production of marketable HCl, NaCl, or CaCl2), SO2 (production of marketable gypsum) as well as bottom ash and fly ash treatment processes. Today the German environmental engineering industry is fully capable of supplying a wide variety of PCDD/PCDF reduction processes individually tailored to the clients needs as well as the legal requirements. These technologies are mostly based either on physical adsorption of PCDD/PCDF onto, e.g., activated carbon or on catalytic destruction of PCDD/PCDF. Table 4 provides an overview over these most commonly applied control technologies for removing PCDD/PCDF from flue gases of waste incinerators and other thermal industrial processes. It also lists the individual operating temperature as well as the main equipment features of these technologies. However, these PCDD/PCDF

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

401

Table 4. PCDD/PCDF control technologies

Process a

Adsorbent

Conventional Main equipment Operation Temperature (°C)

Adsorbent injection upstream dedusting devices

Activated carbon, hearth oven coke, special minerals

135–200

Entrained flow reactor

Activated carbon, 110–150 hearth oven coke, special minerals, mixtures of these substances with Ca(OH)2 or inert material

Fresh adsorbent supply, FF, recirculation system, spent adsorbent system

Activated carbon reactor (ACR)

Hearth oven coke, activated carbon

Fresh adsorbent supply, Fixed bed reactor, spent adsorbent system

Tail-end catalytic reaction

None (ammonia only 130 (260)–350 for NOx removal)

a

110–150

Fresh adsorbent supply, Injection system, FF or ESP for new installations

Reactor with catalyst, (ammonia supply)

Combinations of these processes are possible (e.g., entrained flow reactor with adsorbent injection or adsorbent injection with tail end catalytic reaction).

control technologies differ significantly with respect to their costs and their removal efficiencies for PCDD/PCDF and other pollutants (e.g., acid gases and heavy metals). The objective of the process design engineers is to choose carefully among these various processes in order to supply the client with the most cost-effective solution for each individual problem and application (see Table 4). 4.1 Adsorbent Injection

The lowest cost approach to achieve compliance for PCDD/PCDF reduction is usually the injection of an activated carbon based adsorbent upstream of a particulate matter collection device.As shown in Fig. 8, a baghouse or fabric filter respectively is typically used. In many cases this simple technology is sufficient to achieve compliance with an emission limit of 0.1 ng I-TEQ/Nm3 (dry basis, 11% O2). The adsorbent adsorbs the PCDD/PCDF on its way to the fabric filter and on the filter bags itself while it is in contact with the flue gas. The fabric filter separates the adsorbent from the flue gas together with other particulate matter such as fly ash or reaction products from a preceding dry or semi-dry acid gas scrubbing system. The adsorbent injection technology is most commonly used in conjunction with such dry or semi-dry systems. The PCDD/PCDF removal efficiency depends on the quality of the adsorbent injection, the effectiveness of the adsorbent-flue gas mixing system, the type of particulate filter, and the operation of the system.Also a critical parameter is the

402

H.-U. Hartenstein

Fig. 8. Principle of adsorbent injection upstream a particular filter

mass flow rate of the injected adsorbent. For applying this technology most effectively a baghouse should be used because of the extended contact time of the adsorbent with the flue gas in the gas stream and in the filter cake on the filter bags. Lower adsorbent consumption will also result from applying a baghouse rather than an ESP. A critically limiting condition, especially for retrofits, is the boiler outlet temperature, which often exceeds 200 °C. This, however, must be considered the maximum permissible temperature for this technology due to the adsorption capabilities and also safety reasons associated with the use of activated carbon.Adsorbent injection is particularly advantageous if subsequent flue gas cleaning components are installed for high quality byproduct recovery (e.g., wall board quality gypsum from SO2-absorption). The most common adsorbents are powdered hearth oven coke made from lignite or powdered activated carbon. The use of special minerals such as aluminum oxides providing a very high specific surface area has been developed more recently. 4.2 Entrained Flow Reactor

The entrained flow reaction technology is a typical tail-end process for flue gas polishing following a conventional flue gas cleaning system (see Fig. 9).With this technology, most of the remaining regulated air pollutants such as SOx, HCl, HF, Hg, and other heavy metals, as well PCDD/PCDF (except NOx) can be removed to values below most legal emission limits. PCDD/PCDF testing at facilities equipped with entrained flow reactors as the last stage of a modern multi stage air pollution control train revealed values as low as 0.01 ng I-TEF/Nm3 (dry basis, 11% O2). For this technology the same adsorbents are applied as used for the adsorbent injection process. However, the adsorbent is usually applied in a mixture with hydrated lime and/or other inert materials such as limestone, quick lime, or sodium bicarbonate. Upstream the entrained flow reactor for flue gas polishing, a conventional flue gas cleaning system is required for removing the bulk of the fly ash and the acid gases. Such a pre-cleaning system may be a dry, semi-dry, or wet scrubbing system.

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

403

Fig. 9. Principle of an entrained flow absorber

The process consists of three major components, namely (1) a fresh adsorbent supply and injection system, (2) a fabric filter for the capture of the injected adsorbent, and (3) an adsorbent recirculation and storage system for spent adsorbent. Depending on the overall configuration of the air pollution control system, a flue gas reheating system prior to the entrained flow reactor, a bypass, and a start-up heating system might be required. The spent adsorbent is commonly injected into the combustion zone of the furnace and incinerated, thus destroying the adsorbed PCDD/PCDF thermally. The adsorbent recirculation system enhances the utilization of the active surface area of the adsorbent to a maximum possible extent. It also minimizes the consumption of fresh adsorbent while still ensuring the required high concentration of particulate matter in the entrained flow reactor.A minimum particulate concentration in the reactor is important to ensure the required adsorbent distribution in the flue gas duct and for the cleaning intervals of the filter bags. The pressure drop across a fabric filter operated as an entrained flow reactor is usually slightly higher compared to a conventional baghouse for fly ash collection only. An additional sink for mercury collected on the adsorbent might also be needed, such as the injection of Na2S4 [6]. This technology has found widespread use not only in the waste-to-energy industry, but also in the steel and metal industry, cement kilns, glass industry, and many other industrial thermal processes. 4.3 Activated Carbon Reactor

Another highly effective tail-end solution for the control of PCDD/PCDF and other POPs as well as all heavy metals and acid gases is the activated carbon reactor (ACR). It commonly uses granular hearth oven coke made from lignite as an adsorbent. The particle size of the adsorbent is approximately 20 times larger than the powdered adsorbent used in the entrained flow technology.An ACR can be arranged similarly to the above described entrained flow reactor at the tail end of an conventional modern multi stage flue gas treatment system. With an ACR most pollutants can be reduced to an extremely low level, often even to concen-

404

H.-U. Hartenstein

raw gas chamber

Fig. 10. Schematic of a two-bed activated carbon reactor

trations below the detection limits. Among these pollutants are all POPs, especially PCDD/PCDF, PCBs, and PAHs.An ACR is also capable of completely buffering even extreme pollutant peaks in the inlet concentration. Even NOx can be reduced using special activated carbon derivatives [7, 8]. Very extensive PCDD/PCDF testing at numerous full scale waste incinerators showed achieved emission values consistently below 0.001 ng I-TEQ/Nm3 (dry basis, 11% O2). Two basic design principles of the ACR technology have been developed in Germany: the counter flow principle, where the flue gas enters the ACR at the bottom and flows upwards through the downwards moving coke bed and the cross flow principle which is used as an example here. Both are modular systems and thus extendable to 2, 4, 6, or 8 beds depending on the flue gas volume flow rate to be treated. Figure 10 shows the principle of a two bed cross flow ACR. The raw gas enters the reactor at the bottom through the raw gas distribution chamber. Due to an equal pressure drop across each coke bed, the flue gas distributes itself evenly among the beds and flows horizontally through each bed. During this time the adsorption process takes place. Hearth oven coke, activated char, or activated carbon enters the reactor from the top and migrates down vertically through the reactor to the bottom. Figure 11 details the design of one bed. Each bed consists of an inlet gas distribution system, three independent layers of adsorbent, which are separated by perforated shrouds, and an adsorbent retention system at the flue gas outlet. Each bed is also equipped with several cylindrical discharge units and screw conveyors at the bottom to collect the discharged adsorbent. The discharge rate depends on the guiding parameters time, pressure drop, and pollutant load. The polished flue gas leaves the ACR through the clean gas chambers. Subsequently, the clean gas ducts are recombined into the main flue gas duct. Dampers are required in the clean gas ducts in order to take an individual bed briefly off line while dis-

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

405

Fig. 11. Cross flow reactor design concept

charging some spent coke from the last layer. The first layer serves for the filtration of particulate matter and the adsorption of PCDD/PCDF and heavy metals as well as SO2 . In the second layer the remaining SO2 and HCl are adsorbed. The third layer is a spare layer and doesn’t contribute to the adsorption process during normal operation. The discharge of spent adsorbent from each individual layer is performed according to a special adsorbent discharge program. Due to the high residence time of the adsorbent within the ACR, a very low overall adsorbent consumption rate is achieved. The spent adsorbent is commonly fed into the combustion zone of the furnace or combusted externally in order to destroy adsorbed POPs thermally.As mentioned before, a reheating system and a bypass might be required. However, a separate heating system for start-up and an inert gas supply station are mandatory. The ACR technology has been in full scale operation for 10 years treating volumetric flow rates up to 250,000 Nm3/h.ACR systems are in operation in coal fired power plants as well as in municipal and hazardous waste combustion plants, medical waste incinerators, sludge incinerators, crematoria, cement kilns, and other industrial installations. 4.4 Tail-End Selective Catalytic Reaction

Besides adsorption processes the catalytic destruction of PCDD/PCDF provides another viable control option, which has been widely applied. Figure 12 shows the oxidation reactions for two tetra-chlorinated PCDD/PCDF species, which occur on the surface of a catalyst in the presence of oxygen. Mainly CO2 and HCl are the resulting oxidation products. Honeycomb catalysts similar to those known from the SCR (selective catalytic reaction) DeNOx technology are used for this residue-free PCDD/PCDF destruction process. The achievable

406

H.-U. Hartenstein

Fig. 12. Principle reaction mechanism of the catalytic PCDD/PCDF destruction

PCDD/PCDF reduction rate depends on the installed catalyst volume, the reaction temperature, and the space velocity of the flue gas through the catalyst. Therefore, the process can be designed to meet specific requirements. PCDD/ PCDF testing showed that emission values lower than 0.01 ng I-TEQ/Nm3 (dry basis, 11% O2) can be achieved. The main advantages of this process are an easy operation and no residues apart from very little spent catalyst. Catalyst lifetime is usually in excess of five years and the spent catalyst can be taken back and recycled by the suppliers. At low operating temperatures (<200 °C) safety precautions are not required. For higher operating temperatures a security bypass is recommended in order to protect the catalyst against possible destruction caused by excess temperature. Such temperature excursions may result from very high VOC and CO emissions during extreme incinerator upset conditions. These combustible gases are also oxidized by the catalyst and the exothermic oxidation can liberate sufficient energy to cause temperature excursions up to 1000 °C and more. Typically, in MSWI catalytic PCDD/PCDF removal is combined with an SCR DeNOx system within one reactor. Due to an undesired reaction of the ammonia needed for the DeNOx reaction and the remaining SO2 and SO3 concentration downstream of the preceding flue gas cleaning components, the minimum operation temperature of the catalytic system is determined by the residual SO2/SO3 concentration upstream of the SCR reactor. As shown in Fig. 13, such a system typically consists of a regenerative heat exchanger, a natural gas or oil fired burner or a steam reheater, the ammonia supply and injection system (in case of combined DeNOx SCR), and the reactor with the catalyst. A bypass and start-up system is necessary for most of the applications. The higher the operating temperature the lower is the required catalyst volume. However, the overall

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

407

Fig. 13. High temperature SCR reactor for simultaneous NOx and PCDD/PCDF Removal

cost depends on the sum of the operating cost from reheating and the investment cost for the catalyst and the reactor. PCDD/PCDF control by means of an oxidation reactor is usually not applied downstream a MSWI unless combined with a DeNOx SCR. Normally either NOx or heavy metals like mercury have to be removed in the flue gas cleaning plant as well. Therefore, combined processes are preferred. Nevertheless, the first full scale system for the a PCDD/PCDF oxidation reactor after an iron ore sintering plant went into operation in Germany in October of 1998 [9]. The system operates at a design temperature of 130–140 °C and is preceded only by an ESP and an I.D. fan. This tail-end solution for the PCDD/PCDF emission control from iron ore sintering was chosen due to its low overall pressure drop and the lack of additional residue treatment.

5 Examples of Full Scale Plants 5.1 MVB Hamburg

The municipal waste incineration plant Müllverbrennungsanlage Borsigstrasse (MVB) in Hamburg consists of 2 identical 520 ton per day combustion lines and was designed for a total capacity of 320,000 tons per year. Each line includes a combustion system including the heat recovery boiler as well as an entire flue gas cleaning plant. The recovery systems for hydrochloric acid and gypsum are common to both lines. The plant has been in commercial operation since 1994. Unconditional continuous compliance, cost-effectiveness,

408

H.-U. Hartenstein

Fig. 14. MVB Borsigstrasse process flow diagram

and high availability have been demonstrated during year-round operation up to now. Figure 14 shows the generalized process flow scheme for the flue gas cleaning plant. In order to reduce NOx by means of the SNCR selective non-catalytic reduction process, aqueous ammonia is injected into the furnace. This primary NOx control technology allows the federal emission limit of 200 mg/Nm3 (dry basis, 11% O2) to be met. The secondary air pollution control equipment starts after the waste heat boiler. The flue gas exiting the boiler with a temperature between 200 and 230 °C is quenched down to approximately 170 °C in an evaporative cooling tower with process water. Upstream of the quench cooler, powdered HOC is injected as adsorbent for the removal of the bulk of the heavy metals as well as PCDD/PCDF. The next stage comprises of a fabric filter, which is not only used for the collection of particulate matter but also for an almost complete removal of PCDD/ PCDF and heavy metals including mercury, all adsorbed on the HOC.A two-stage co-current acid gas scrubber, which uses water at a pH-value of 0 or less as a scrubbing liquor, serves for the removal of HCl and HF. No neutralizing agent is added to the scrubbing liquor. Make-up water is continuously added in order to replace the evaporative losses from cooling the flue gas down to near saturation. Raw hydrochloric acid of approximately 10 wt% concentration is extracted from the scrubber to the HCl recovery plant. The next stage consists of a counter current scrubber for SO2-removal. In order to enhance absorption and provide a reaction partner for the absorbed SO2 , quicklime (CaO) is constantly added to the scrubbing liquor. This allows the scrubber to be operated at a constant pH-value of approximately 6. Due to the SO2 absorption and sulfite oxidation process in the scrubber, gypsum is formed. The scrubbing liquor containing the suspended gypsum crystals is fed to the gypsum dewatering system for the recovery of wall board quality gypsum. The last stage of the APC train is a wet ESP for the addi-

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

409

tional removal of fine particulate matter and also to avoid the discharge of aerosols to the atmosphere. The excellent PCDD/PCDF control efficiency has been demonstrated over the last five years of operation. The actual PCDD/PCDF emissions downstream the fabric filter are far below the plant’s permit limit of 0.05 ng I-TEQ/Nm3 (dry basis, 11% O2). Originally, the HOC-injection system for PCDD/PCDF and heavy metal control upstream of the wet scrubbers was primarily considered for an enhanced quality and purity requirements of the end products, the 30 wt% hydrochloric acid and the wall board quality gypsum. The gypsum quality is similar to gypsum from FGDs in power stations and used for the production of wall board and building materials. 5.2 MHKW Leverkusen

The MHKW Leverkusen was originally designed for an annual incineration capacity of 256,000 tons per year of municipal waste. Between 1994 and 1996, the facility was retrofitted with a new flue gas cleaning system to meet the highest standards. Before the extensive retrofit, each of the three units (furnace and waste heat boiler) was only equipped with a spray dryer absorber and an ESP for flue gas cleaning. This APC plant was not acceptably suited for insuring continuous compliance with the newly introduced German legislation, the 17th BImSchV. Especially the lack of effective control step for PCDD/PCDF, mercury, and NOx required substantial upgrading of the existing APC system. However, the existing system remained in place and was incorporated into the new flue gas cleaning plant as detailed in the chosen process flow scheme in Fig. 15. The retrofitted flue gas cleaning system starts after the existing ESP and consists of a cross-flow tubular heat exchanger, a dual stage co-current HCl-scrubber, a single stage counter-current SO2-scrubber, an entrained flow reactor, and a low temperature SCR for NOx removal. The old spray absorber was converted to operate as a quench cooler. However, in case of upset conditions in the sodium chloride salt or the gypsum recovery systems it can also be used as a spray dryer. The existing ESP remained for the removal of the bulk of the particulate matter. The HCl-scrubber is operated at a pH-value of approximately 1. This level is maintained by the continuous addition of NaOH for partial neutralization of the absorbed HCl. The resulting NaCl containing raw acid of approximately 60 g Cl–/l is extracted from the scrubber and fed to a multi-stage NaCl recovery plant for the production of marketable NaCl salt. Wall board quality gypsum is also produced by the same process as described for the MVB facility. The entrained flow reactor serves as a sink for PCDD/PCDF and also reduces mercury and the all other heavy metals as well as the remaining acid gases and particulate matter.A mixture of HOC, hydrated lime, and/or limestone is used as an adsorbent. The spent adsorbent is discharged back into the furnace in order to avoid another residue for disposal. PCDD/PCDF emission tests confirmed PCDD/PCDF concentrations emitted to be significantly lower than the permitted limit. As a result of this elaborate six-stage APC process, all actual emission

Fig. 15. Process flow scheme of MHKW Leverkusen retrofit

410 H.-U. Hartenstein

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

411

concentrations of all regulated air pollutants are remarkably below those required by the 17th BImSchV and also substantially lower than those required by the plant’s permit. 5.3 MVA Stapelfeld

The MVA Stapelfeld consists of two identical 550 tons per day units and went into full operation in 1981. Each furnace and boiler was originally designed for a capacity of only 19 metric tons of waste per hour and was recently upgraded to its current capacity. The design of the furnace ensures that the flue gases from the combustion furnace remain at all time at a temperature of above 850 °C for more than 2 s after the last injection of the secondary combustion air. In the heat recovery boiler they are cooled down to approximately 190 °C before entering the flue gas cleaning plant. Upstream the second economizer of the boiler, an ESP, is installed for the removal of particulate matter. After the second economizer, the flue gas is led through a multi-stage wet scrubbing system removing most of the SO2 , HCl, and HF. Prior to that, a cross-flow tubular glass heat exchanger uses the enthalpy of the raw gas downstream the second economizer to reheat the saturated flue gas discharged from the scrubber system to approximately 130 °C. Subsequently, an I.D. fan is installed, which originally discharged the flue gas through the stack. Even though this APC system was absolutely state-of-the-art at its time, the newly introduced standards for PCDD/PCDF, heavy metals, and NOx could not be met without a retrofit. In 1996, the plant’s APC retrofit and upgrade was completed to meet the new requirements of the German 17th BImSchV. The additional gas cleaning equipment for flue gas polishing is shown in Fig. 16. The new design volumetric flow rates are now 140,000 Nm3/h (wet basis, 11% O2) for each line. This additional flue gas polishing system represents a typical example of an upgrade chosen for numerous facilities in Europe in order to meet the new legislation, especially with respect to the extremely effective control of PCDD/PCDF. An activated carbon reactor (ACR) and a low temperature SCR plant were added. The ACR removes heavy metals, PCDD/PCDF, particulate matter, and other pollutants such as remaining HCl and SOx to levels around or even below the detection limits. The ACR of each APC train consists of two independent reactors with four beds each to ensure maximum flexibility and a certain redundancy.Additionally installed booster fans overcome the pressure drop of the ACR and SCR and discharge the flue gas through the existing stack. During start-up and shut-down these fans are also used to preheat and cool the ACR and SCR, respectively. The one SCR reactor per APC train is operated at a relatively low temperature of 200 °C. Such a low temperature could be chosen due to the virtually complete removal of SO2 and SO3 from the flue gas by the ACR. Thus, the risk of the formation and deposits of ammonia sulfate ((NH4)2SO4) and ammonia bisulfate (NH4HSO4) is avoided. Otherwise such formation and deposit could shorten the lifetime of the catalyst dramatically. For heat recovery and reheating purposes a regenerative heat exchanger and a steam reheater are installed respectively.Ammonia, which is needed as a reducing agent for NOx , is mixed into an extracted

Fig. 16. Process flow scheme of the MVA Stapelfeld retrofit

ammonia injection grid

412 H.-U. Hartenstein

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

413

flue gas slip stream to ensure complete evaporation of the injected aqueous ammonia solution and an optimal NH3/NOx distribution after being reintroduced into and mixed with the main flue gas stream. To avoid the loss of municipal waste incineration capacity through the replacement by the high calorific value of the spent activated carbon in the main furnace, the spent HOC is incinerated separately in a specially developed external combustion chamber. The adsorbed organic pollutants are completely decomposed at a temperature of approximately 800 °C. The flue gas of the external combustion chamber is mixed with the flue gas of the main furnaces and treated in the existing gas cleaning system. As in other plants, where the spent activated carbon is fed back onto the grate, a sink for highly volatile heavy metals such as mercury is needed upstream of the activated carbon filter. In the MVA Stapelfeld the sink for mercury is the wet scrubbing system. It absorbs more than 70% of the total incoming mercury. In cases of dry or semi-dry processes a separate sink for mercury must be installed. This problem can be solved very cost effectively using the newly developed sodium tetra sulfide (Na2S4) injection technology. This proprietary technology utilizes Na2S4 to react with elemental as well as ionic mercury (mostly present as HgCl2) to form HgS (cinnabar) [6, 10]. The results obtained during extensive PCDD/PCDF testing demonstrated very impressively the enormous potential of the ACR technology for the virtually complete removal of PCDD/PCDF, PAHs, PCBs, and other POPs as well as heavy metals, acid gases, and particulate matter.As such, this technology represents the most effective way to control simultaneously all pollutants to levels around or below their detection limit. Numerous European incineration facilities for hazardous waste, municipal waste, medical waste, sewage sludge, and even coal fired utility boilers have been retrofitted with the ACR technology. 5.4 AVI Gevudo Dordrecht

The MSW incineration plant Gevudo in Dordrecht, Holland consists of four 310 ton per day incineration lines connected to two identical flue gas cleaning trains. The retrofit of the existing system focused on the adsorbent injection systems for powdered HOC and sodium tetra sulfide (Na2S4) downstream the evaporative cooling towers in which the flue gas is quenched down to approximately 170 °C. The subsequent flue gas cleaning plant consists of a fabric filter, a twostage wet scrubbing system, and a wet ESP. The tail-end of the retrofitted system presents a high temperature SCR plant for combined NOx reduction and integrated oxidation catalyst for the simultaneous removal of PCDD/PCDF (Fig. 17). This technology is commonly referred to as SCR-DeNOx/DeDiox technology, which is also widely applied in Europe. As a peculiarity, the APC train is equipped with two mercury reduction systems (HOC and Na2S4) and two dioxin reduction systems (HOC and oxidation catalyst). This results from the fact that very high emissions of elemental mercury and PCDD/PCDF were expected, because the flue gases treated by each APC train originate from one incinerators with a waste heat recovery boiler and one without a heat recovery system. The new flue gas treatment plant was commissioned

414

H.-U. Hartenstein

Fig. 17. Process flow scheme of flue gas cleaning plant Gevudo

in 1997. The results of PCDD/PCDF tests proved that the European emission limit of 0.1 ng I-TEQ/Nm3 (dry basis, 11% O2) can be met at all times. At the chosen operating temperature of 320 °C the described SCR-DeNOx/DeDiox processes as well as the ACR technology drastically reduce polychlorinated biphenyls (PCB), chlorobenzenes, and polycyclic aromatic hydrocarbons (PAH) as well. Therefore, the emissions of these persistent toxic organic pollutants will also be minimized significantly if PCDD/PCDF reduction processes are applied. Reduction efficiencies of more than 99% have been achieved depending on the control process and raw gas concentrations. 5.5 RVA Böhlen

The RVA Böhlen was originally built in 1976 as a residue incinerator for the chemical olefin production in East Germany. After reunification, the facility was taken over by a hazardous waste company who decided to reconstruct the entire facility except the office building, the pit building, the machine and maintenance shop, and the stack. This decision led to the construction of a virtually new 80 tons per day hazardous waste incineration facility, which today is the most modern and advanced state-of-the-art rotary kiln high temperature hazardous waste incinerator in Europe. Due to very high design inlet pollutant concentrations for the flue gas treatment system combined with local permitting requirements of on the average one order of magnitude below the German 17th BImSchV, an extremely elaborate seven stage arrangement had to be chosen. As a first step after leaving the heat recovery boiler, the flue gas passes through a two-field ESP in order to reduce the concentration of particulate matter to well below 30 mg/m3. Subsequently, Na2S4 is injected in order to react with mercury in the flue gas as described for the Gevudo plant earlier. The following spray dryer is used to cool the flue gas further to about 170–190 °C by evaporating the effluent from the first scrubber. The dried salts, mainly calcium chloride and calcium fluoride as well as the mercury sulfide formed are collected together with residual particulate matter in the downstream baghouse. The next two stages consist of a conventional wet scrubbing system with a two-stage HCl scrubber

Fig. 18. Process flow scheme of the RVA Böhlen APC system

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

415

416

H.-U. Hartenstein

operated at a pH below 2 and a single loop SO2 scrubber operated at a pH of around 6. Both scrubbers receive CaO solution as a neutralizing agent in order to maintain the set pH value. The scrubbing liquor extracted from the second scrubber is liberated from the formed gypsum by means of a centrifuge. The effluent from the first scrubber is combined with the excess water removed from the second scrubber, neutralized, and treated for flocculation and precipitation of heavy metals. Finally, the solution is injected into the spray dryer for flue gas cooling and drying of the reaction salts. The next step in the APC train is the main stage for flue gas polishing including PCDD/PCDF removal (Fig. 18). Even though high temperature rotary kiln type incinerators such as the RVA Böhlen already ensure very low concentrations of PCDD/PCDF as well as other persistent organic pollutants (PTS) at the outlet of the secondary combustion chamber, the extraordinarily low permitted values imposed by the local permitting authority still require an ACR system as an integral pert of the APC train as described earlier. The result of the ACR-technology applied within this APC train are emission concentrations for PCDD/PCDF and other PTS at or below the detection limit. Additionally the ACR system ensures emission values for heavy metals, especially mercury as well as acid gases and particulate matter of well below the permitted values. Reduction efficiencies of more than 99.99% are easily achieved with such a system. The low temperature SCR system operated at the tail end is only designed to remove NOx but also has an additional effect in reducing PCDD/PCDF even further. The APC system installed at the RVA Böhlen represents the most elaborate state-of-the-art technology for controlling PCDD/PCDF emissions. However, it also induces very significant costs for investment as well as operational requirements and burdens the operator and ultimately society with a substantial amount to be paid for achieving close to zero emissions. 5.6 VERA Hamburg

In 1993, the City of Hamburg decided to build a central sewage sludge incinerator which consist of three identical fluidized bed furnaces, heat recovery boilers, and APC trains. Each train has a capacity of 3.5 tons per hour of pre-dried sludge. The plant was erected adjacent to the existing sludge drying facility and went into operation in 1995. The flue gas treatment system (Fig. 19) consists of four stages including an entrained flow reactor at the back end of the system for PCDD/PCDF removal. The first stage employs an ESP for the removal of the fly ash. As a second and third stage the commonly used two-stage wet scrubbing system is used. The effluent from the first scrubber goes back to a waste water treatment system to remove the heavy metals and is then reintroduced to the sewage treatment plant. The gypsum formed in the second scrubber is dewatered in a centrifuge and used as construction material. Following the wet scrubbing system, the saturated flue gas is super-cooled by means of an indirect water cooler. The condensate removed from the flue gas is used as make-up water within the system when needed. However, depending of the residual moisture content of the incoming sludge, the sys-

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

417

Fig. 19. Process flow scheme of the VERA sludge incinerator

tem produces an excess amount of water to be discharged back to the sewage treatment plant. The super-cooled flue gas is then reheated via a cross flow heat exchanger to reach the temperature required for the entrained flow adsorber for PCDD/PCDF control. A mixture of Ca(OH)2 and activated carbon is used as an adsorbent. The adsorbent is recirculated at a high rate in order to make better use of its adsorption capacity and minimize the consumption of adsorbent. During acceptance testing the PCDD/PCDF concentrations measured were well below the legal limit of 0.1 ng I-TEQ/Nm3 in the stack. Thus, this type of PCDD/PCDF control also represents a viable and economically interesting alternative for effectively controlling PCDD/PCDF emissions. 5.7 Sinter Belt 2 Thyssen

As mentioned in Sect. 4.4, catalytic oxidation reactors without simultaneous NOx removal are also an option for effective PCDD/PCDF control after a particulate removal stage. The advantages of such an arrangement compared to a tail-end solution in combination with NOx removal are considerable cost savings resulting mainly from the avoidance of flue gas reheating (gas-gas heat exchanger, burner, etc.) and a lower pressure drop. Figure 20 shows the principle of such a reactor, which is equipped with additional soot blowers to remove possible residual dust deposits. This type of reactor arrangement is known from high dust SCR systems widely used in fossil fuel fired power plants.

418

Fig. 20. Thyssen SB2 DeDiox catalyst reactor

Fig. 21. Iron ore sintering plant process flow scheme

H.-U. Hartenstein

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

419

Within the iron ore sintering processes as shown in Fig. 21, a mixture of various substances is coalesced at high temperature (1000–1200 °C). The material is mixed with coke, which provides the source for the carbon as well as the chlorine, and transported on a chain grate. Burners above this grate heat the top of the material to the required temperature, causing the coke to ignite. By sucking air through the mixture, the flame front is moved through the sintering bed. Consequently, ideal conditions for the formation of dioxins/furans, and other PTS which can not be avoided, exist locally regarding temperature and material composition. Sintering plants are normally equipped with electrostatic precipitators (ESPs) for the removal of particulate matter as the only flue gas cleaning devices. Stack measurements at such plants in Europe revealed average PCDD/PCDF concentrations in a range from 1 to 5 ng I-TEQ/Nm3 at volumetric dry flow rates ranging from 250,000 to 1,000,000 Nm3/h. Thus, due to the very high gas volume flow rates the actual mass flow rates of PCDD/PCDF emitted from iron ore sintering were found to be substantially higher than from waste incineration. However, current regulations for these processes do not require emission control equipment for PCDD/PCDF and some countries such as the US have not even investigated the PCDD/PCDF emissions from steel production as of today. After very extensive pilot testing using a small slip stream reactor yielded very promising results indicating that the target value of PCDD/PCDF emissions of less than 0.1 ng I-TEQ/Nm3 could be achieved, a consortium of European steel producers awarded the contract for the installation of the first full-scale catalytic oxidation reactor on October 14th, 1997. The project was intended to serve as the first application of this kind and a demonstration plant in order to determine the viability of this known PCDD/PCDF control technology with respect to iron ore sintering. Figure 22 shows the process flow arrangement chosen for the demonstration plant which was set up at Thyssen’s Sinter Belt 2 in Duisburg, Germany, which is one of the production facilities of the German Thyssen-Krupp AG steel company. The DeDiox reactor is arranged downstream the existing ESP and I.D. fan. The flue gas enters the reactor from the top and leaves at the bottom to the existing stack. In this case a tail-end solution was selected due to its high availability, low pressure drop, and the lack of additional residue treatment. Besides that, any other arrangement would have required major downtime of the sinter belt due to significant space restrictions. The maximum permissible pressure drop for the total retrofit was limited to 10 mbar. Other PCDD/PCDF control technologies such as ACR, adsorbent injection, or entrained flow reactors would have led to a much higher pressure drop (15 to 20 mbar) resulting not only in the corresponding increased electricity cost and additive cost but also in the requirement of an additional I.D. fan .The reactor is designed for an operating temperature of 130–140 °C and a flue gas volumetric flow rate of 400,000 m3/h (wet at STP). The catalyst was designed for a PCDD/PCDF reduction efficiency of at least 75%. Due to the residual dust content of approximately 50 mg/Nm3 and the lack of experience concerning the dust properties, a large pitch of 6.5 mm was chosen and additional soot blowers were installed. As shown in Fig. 22, the reactor can accommodate six catalyst layers, four of which are equipped with catalyst. Two of them are spare layers, which can be

420

H.-U. Hartenstein

Fig. 22. DeDiox reactor for PCDD/PCDF control downstream a sintering plant

equipped later to enhance the overall catalyst lifetime. In addition, soot blowers are arranged upstream of each catalyst layer in order to remove any possible residual dust deposits on the catalyst. Supplied with compressed air, they are operated on a discontinuous basis. Upstream the catalyst, a natural gas burner is installed fulfilling the following two purposes: 1. Preheating of the catalyst to temperatures above the water dew point during start-up of the sinter plant 2. Adjusting the SCR temperature to investigate the temperature influence on the PCDD/PCDF control efficiency Due to the longer preheating process of the sintering belt itself, the start-up time of the catalyst is completely integrated in the overall start-up process. Consequently, no loss of time and production of sinter occurs. The reactor itself has been in continuous operation without problems. It fully complies with the operational requirements of the sintering plant without im-

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

421

posing any restrictions on sinter production. The actual pressure drop is well below the limit of 10 mbar.After six months of good operating experience, the soot blowers of the third and fourth layers had been switched off without observing an increase in pressure drop. Since the plant operates automatically, no additional operating staff is required. A very comprehensive dioxin/furan sampling and measuring program has been started by the consortium, which will continue, to the middle of 2000. The technology of catalytic oxidation of PCDD/PCDF for the control of these PTS has been proven to be a reliable and economical solution for a wide range of applications. In this case the emissions from an iron ore sintering plant could be reduced substantially as well as economically without compromising the product requirements and the operational demands of the sinter production.

6 New Dioxin Control Processes Another low cost process which was recently introduced is the so-called AKTINERT technology. It is based on a new adsorbent consisting of an inert limestone center coated with activated carbon dust (Fig. 23).AKTINERT refers to the sorbent’s active surface on an inert center. The activated carbon fraction is only 0.5–3 wt%. The first measurements with the new process were carried out at iron

Fig. 23. The AKTINERT adsorbent

422

H.-U. Hartenstein

Fig. 24. PCDD/PCDF control using the AKTINERT process

ore sintering plants. The PCDD/PCDF adsorption efficiency was about 99.98% (from 20 to 0.005 ng I-TEQ/Nm3 (dry basis, actual O2)). Most of the advantages of the ACR process remain while the investment and operating costs as well as safety requirements will decrease significantly. The principle design of the AKTINERT process is shown in Fig. 24. It consists of a fresh adsorbent supply system, a fixed bed reactor similar to the ACR, and an adsorbent preparation system. Here a very small amount of spent adsorbent is removed and the inert center is re-coated with fresh adsorbent. Due to the lower carbon content in the fixed bed the operating temperature range is widened to 110–200 °C. Since the main application of this technology is the dioxin/furan and heavy metal control only, the bed thickness can be significantly decreased. These pollutants will be caught in the first layer of a bed. Estimates quantify a total cost savings potential of approximately 25% compared to an entrained flow reactor and about 30% compared to a conventional ACR.

7 References 1. Hagenmaier H, Beising R (1989) Untersuchung von Kraftwerksrauchgasen auf polychlorierte Dibenzodioxine und Dibenzofurane. VGB Kraftwerkstechnik 69, Heft 10 2. Griffin, R (1986) A new theory of dioxin formation in municipal solid waste combustion. Chemosphere 15:1987–1990 3. De Koning J (1996) Dioxin emissions into the air in the Netherlands. VDI-Berichte 1298, VDI, Düsseldorf 4. Bruckmann P, Bröker G, Gliwa H (1996) Meßprogramm zur systematischen Erfassung relevanter Dioxinemittenten in den deutschen Bundesländern.VDI-Berichte 1298,VDI, Düsseldorf

Dioxin and Furan Reduction Technologies for Combustion and Industrial Thermal Process Facilities

423

5. Dyke P, Coleman P (1900) Dioxin measurement programmes in Great Britain.VDI-Berichte 1298, VDI, Düsseldorf 1996 6. Schüttenhelm W, Hartenstein H-U, Licata T (1998) An optimized concept of flue gas cleaning downstream of MWCs using sodium tetra sulfide for mercury removal. Sixth North American Waste-To-Energy Conference, Miami Fl, USA 7. Hartenstein H-U (1993) A fixed bed activated coke/carbon filter as a final gas cleaning stage retrofitted for a hazardous waste incineration plant – the first 6 months of operation experience. 86th Annual AWMA Meeting and Exhibition, Denver CO, USA 8. Hartenstein H-U (1993) Activated carbon filters for flue gas polishing of MWIs. Third International Conference on Municipal Waste Combustion, Williamsburg VA, USA 9. Hartentein H-U, Schüttenhelm W, Wemhöner R, Werner K (1999) Reduction of PCDD/F emissions from iron ore sintering plants – the first full-scale SCR installation. Dioxin 99 – 19th International Symposium on Halogenated Environmental Organic Pollutants and POPs, Venice, Italy 1999 10. Kubisa R, Schüttenhelm W (1996) Vereinfachte Konzepte für Abgasreinigungen hinter Müllverbrennungsanlagen unter Berücksichtigung wirtschaftlicher Aspekte und Emissionsanforderungen VDI-Seminar Vereinfachte Hightech – Verbesserte Additivtech. Dioxin- und Gesamtemissionsminimierungstechniken mit Betriebserfahrungen, September 1996, München

CHAPTER 16

Alternative Technologies for Destruction of PCB and Other POPs Ian D. Rae Department of History and Philosophy of Science, The University of Melbourne, Parkville, Vic 3052, Australia E-mail: [email protected]

Alternatives to incineration for the destruction of organochlorine materials involve the use of non-oxidative conditions and thereby avoid the formation of toxic products such as polychlorodibenzodioxins and -furans. These alternatives include the use of plasma arc facilities in which molecules are broken down to atoms and allowed only to recombine to form small molecules, through chemical reductions using gaseous hydrogen or other sources of this element, to nucleophilic substitution reactions which displace chloride ion from the molecules of concern and also bring about other changes. A small group of other thermal technologies includes vitrification of contaminated soil, and combustion of wastes under alkaline conditions in a soda furnace. The chemistry, to the extent that it is known, of each process is discussed, together with description of its advantages and disadvantages and approximate costs of establishment and operation. Keywords: Plasma, Reduction, Paraffin, Nucleophilic, Vitrification

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 426

1

Introduction

2

Why Not Incineration? . . . . . . . . . . . . . . . . . . . . . . . . 427

3

Alternative Technologies

4

Plasma Arc Destruction

5

Reduction Processes . . . . . . . . . . . . . . . . . . . . . . . . . 429

5.1 5.2 5.3

Base-Catalysed Decomposition . . . . . . . . . . . . . . . . . . . 429 EcoLogic Hydrogenation . . . . . . . . . . . . . . . . . . . . . . . 431 Catalytic Hydrogenation . . . . . . . . . . . . . . . . . . . . . . . 432

6

Nucleophilic Substitution

6.1 6.2

Fluidex (PCB Gone) Alkoxide Process . . . . . . . . . . . . . . . . 433 Ball Milling with Lime . . . . . . . . . . . . . . . . . . . . . . . . 434

7

Other Processes

. . . . . . . . . . . . . . . . . . . . . . 428 . . . . . . . . . . . . . . . . . . . . . . . 428

. . . . . . . . . . . . . . . . . . . . . . 433

. . . . . . . . . . . . . . . . . . . . . . . . . . . 434 The Handbook of Environmental Chemistry Vol. 3, Part O Persistent Organic Pollutants (ed. by H. Fiedler) © Springer-Verlag Berlin Heidelberg 2003

426

I.D. Rae

7.1 7.2

Soil Vitrification . . . . . . . . . . . . . . . . . . . . . . . . . . . 434 Soda Furnaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435

8

Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . 436

9

References

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437

1 Introduction Any archeologist will tell you that disposing of unwanted material is a practice as old as humankind. The excavation of thousands of rubbish dumps, some as old as thousands, and in some case tens of thousands of years, gives us information about life in the long distant past [1]. Discarded material is different from the carefully selected and arranged material of a burial site, and in fact often complementary in what we can learn from it.Among the most durable material of the rubbish dump are the stone chips and discarded tools of Neolithic times, followed by ceramics and then food detritus – shells and bones. Putrescible material fares less well over the ages, although vegetable material such as seeds and pollen, and animal matter such as faeces and hair are recognisable even after millennia. Anything that will decay within a reasonable period of time may be regarded as “disposed of ” or even “destroyed”, but the distinction between the two is an important one that we shall apply critically to the wastes of modern society. If we discount slags and ashes from early smelting, then the first industrial wastes becoming available for disposal were those of the nineteenth century chemical industry. Landfilling – to use a modern name for it – was the popular method of disposal, sometimes in natural depressions or old quarry holes and occasionally (at first) in specially dug pits. In other cases waste materials remained above ground, an outstanding example being the calcium sulfide byproduct of Le Blanc alkali production, which is stored in large mounds along the south bank of the Tyne, near Newcastle, UK. An interesting case of disposal by entombment of an organic waste was revealed at ICI’s Blackley works, near Manchester, in the 1980s when removal of a road embankment uncovered a red substance which turned out to be an out-of-specification batch of the dye Turkey Red. Burial of organic wastes must have been quite common, if we are to judge by the contamination uncovered when sites of old works are rehabilitated. It was not until the middle of the twentieth century, however, that the industrial incinerator became a preferred method for the destruction of organics, as environmental laws were passed in many countries to ban the discharging of liquids to rivers and the open burning of waste solids [2]. Burning waste material represented destruction, rather than mere disposal. In the related field of domestic waste disposal, it was specially suitable for reducing the burden placed on landfills, where paper, wood and other combustible materials were frequently set alight but the major impetus for the adoption of incineration of these materials was the prevention of the spread of disease. The eventual predominance of the germ theory

Alternative Technologies for Destruction of PCB and Other POPs

427

of disease over the earlier miasma theory came about in the late nineteenth century, at about the time that the municipal waste incinerators began to appear in Britain, to be closely followed by their erection in the United States and continental Europe. Today, in Europe, landfill space is at a premium and much domestic and industrial waste is burned in specially constructed and operated incinerators, while chemical wastes such as PCBs are burned in even more sophisticated high-temperature incinerators [3].

2 Why Not Incineration? Incineration of waste materials has a number of advantages, including volume reduction (only ash remains to be landfilled) and energy recovery (generation of steam and/or electricity).Apart from the capital costs of such facilities, then, one might wonder why there should be opposition to the construction and operation of high-temperature incinerators. The answer lies in the release to atmosphere of certain by-products of combustion, notably the highly toxic polychlordibenzodioxins and -furans by industrial incinerators of the 1980s and early 1990s [4, 5]. These emissions are the subject of an on-going campaign by the non-governmental environmental organisation Greenpeace, which has extended its opposition to include the use of all chlorinated materials because of the potential production of dioxins and furans during incineration of wastes. Polyvinyl chloride, a plastic material used in packaging and construction, has been a particular target, but release of dioxins following chlorine-bleaching of paper pulp has also come in for critical attention. The regulatory response to concerns about dioxins and furans has been to set stringent limits for discharges to air. In the United States, for example, 30 ng Nm–3 for post-1990 facilities, and up to 125 ng Nm–3 for existing facilities; in Europe, at first 3 ng Nm–3 and later 0.1 ng Nm–3 (dioxin/furan quantities are expressed as Toxic Equivalent – TEQ). Such low emission levels are achieved by careful control of operating conditions and sophisticated treatment of combustion gases. Before these standards were introduced, however, environmentalists led by Greenpeace, expressed strong opposition to high temperature incineration of chlorine-containing materials. Such opposition was a major influence on the Australian government’s decision to abandon plans to build such a facility in the early 1990s, and to encourage instead the development of alternative technologies for destruction of the country’s stockpiles of PCBs and other organochlorines. Petts [6] offers the following criteria for the adoption of alternative technologies to incineration: – the ability to handle large volumes of heterogeneous waste, including municipal solid waste, – versatility in terms of handling both liquid and solid wastes, – energy consumption requirements and costs, – the ability to consistently equal, and preferable better, any emissions values associated with existing technology, and – investment requirements and commercial availability.

428

I.D. Rae

The extent to which the alternative technologies meet these criteria will be discussed in subsequent sections, but before leaving the Petts list it should be noted that it includes only technical factors and overlooks the citizen distrust of incineration [7] which regulatory authorities and potential technology holders, aware of technical and financial considerations, need to take it into account in their decision making.

3 Alternative Technologies Destruction of PCBs by non-incineration methods has been practiced in a number of countries but has never posed serious competition to incineration. One reason for this has been simply economic, while another is that most alternative methods are somewhat limited in the kinds of feedstocks they can accept, whereas the incinerators may be used for a broad range of materials of varying chemical nature and physical disposition. In Australia, where no high-temperature incinerator has been constructed, and government has enforced a ban on exports, the alternative technologies have flourished. Their operation provides the basis for most comment in this essay although the limited experience with alternative technologies in other countries will also be mentioned. We distinguish four classes of destruction technology: 1. 2. 3. 4.

Plasma arc destruction, Reductive methods, Nucleophilic substitution, Others.

Dioxins and furans are known to form from other materials under the oxidative conditions which exist in incinerators and in combustion gas streams. The plasma arc method ensures that constituent molecules of the feedstock are broken down with formation of atoms and from the plasma chlorine and carbon atoms may be trapped before there is opportunity for larger molecules to form. The reductive methods are the complete antitheses of incineration, in that chlorine atoms in the molecules of waste materials are replaced by hydrogen, and not only is there no opportunity for formation of dioxins and furans, but any present in the waste are reacted along with other organochlorines. Nucleophilic reactions offer further ways of replacing chlorine and thus obviating dioxin/ furan formation. The final category, ‘others’ does not allow of such simple description, but will be discussed in detail in the text that follows.

4 Plasma Arc Destruction Conceptually the simplest of the alternative technologies is destruction of the chlorinated material in-flight in a plasma where temperatures of several thousand degrees Celsius are reached. This is usually struck in argon gas which, although experiments have been conducted with the cheaper nitrogen, remains the vehicle of choice. The organic molecules are broken down in the plasma to their

Alternative Technologies for Destruction of PCB and Other POPs

429

constituent atoms (carbon, hydrogen and chlorine, in the case of PCBs and most POPs) and the post-plasma gas stream is quenched with steam or oxygen to prevent recombination of atoms into molecules larger than di- or triatomic.As a result, very high destruction efficiencies (six to eight nines) are achieved and Plascon systems easily comply with emission limits of 0.1 ng Nm–3, including cases where dioxins and furans are known to be present in the feedstocks. Apart from electric power, the only other system requirements are cooling water (a closed loop) and alkali for the rapid quenching of the gas stream. Liquids from the quenching process contain materials such as sodium chloride and bicarbonate which can be disposed of by conventional means. Operating conditions need to be optimised so as to avoid build-up of elemental carbon in the system, with the object being to produce gaseous products which can be released or absorbed. Plasma arc units for destruction of organic wastes were developed through the joint work of Australia’s government research body, the Commonwealth Scientific and Research Organisation, and the private engineering firm Siddons Ramset, and are now marketed by the specialist company SRL Plasma Limited. The inventions are protected by a number of patents [8]. The first “Plascon” units were installed at a chemical company which produces 2,4-dichlorophenoxyacetic acid, starting with chlorination of phenol. In 1990 the company was the subject of a Greenpeace “raid” on account of its discharges of dioxin-containing material to sewer. The resulting inquiry recommended cessation of this practice, and the company moved to purification of crude products of chlorination followed by destruction in plasma arc systems of a waste stream containing organochlorines. In early 1998, a Plascon unit was installed at BCD Technologies in Brisbane, where it is used to destroy concentrated PCB liquids. The plasma arc requires liquid or gaseous feed and is unsuitable for solids, although slurries have been proposed. The presence of fluorinated material in the feedstock leads to production of the highly corrosive hydrogen fluoride which is converted to sodium fluoride in the wash liquor, but extensive use of the Plascon to destroy CFCs and Halons has shown that problems which might thus arise can be overcome by suitable choice of construction materials. POPs derived from agricultural sources occasionally present as mixtures with arsenicals, and it is believed that such mixtures can be processed in the Plascon, with destruction of the organochlorine material and recovery of arsenic by conventional methods from the wash liquor. No commercial venture of this type is known to have taken place, however.

5 Reduction Processes 5.1 Base-Catalysed Decomposition

The Brisbane company, BCD Technologies, has operated for some years the basecatalysed decomposition (BCD) process which was invented some years ago [9] by the US EPA following experiments at their Cincinnati Risk Reduction Research Laboratory into dechlorination including an early process known as KPEG, so

430

I.D. Rae

called because it involved potassium hydroxide and polyethylene glycol. Three companies operate the process under licence in Australia: BCD Technologies,ADI (discussed below) and a smaller company in Melbourne, Haz-Waste. The US EPA licenses users in the United States, of which there have been a number although incineration has always been the major destruction route for PCBs, but licensing outside the country is the responsibility of the BCD Group, consisting of three former employees of US EPA. Under their auspices, the process has been licensed in Spain (for destruction of HCH) and in Japan (for PCB contaminated oils), as well as in Australia. Base-catalysed decomposition is a batch process involving as a hydrogen source hot paraffin (approximately 300 °C) in the presence of an inorganic base (caustic soda or sodium carbonate) and a proprietary catalyst. It destroys POPs under reductive conditions, where no dioxins and furans can form (and existing quantities are destroyed). In one version of the process, the catalyst is carbon derived from in situ decomposition of sucrose. Chemically, the major reactions involved in the process involve hydride transfer from the a hydrogen-rich donor substance (often paraffin, as described below) to the chlorinated molecules, with expulsion of chloride ion which ends up as sodium chloride (salt), disposable by conventional means.Water is also formed and is distilled from the reaction mixture, its volume providing a convenient monitor of the extent of reaction. The BCD process is most commonly applied to dilute solutions of PCB in paraffin, of the kind which have arisen through draining of PCB-containing equipment in the electricity industry in the 1980s and back-filling with paraffin. Concentrations up to 100,000 mg/kg (10% by weight) are said to be suitable for treatment but in practice the PCB concentrations are generally much lower. Alkali needs to be added in proportion. Representing the organochlorine as Ar–Cl, we can write the following [Eq. (1)] to represent the reactions taking place, although their exact mechanism is unknown: Ar–Cl+NaOH +–CH2–CH2–CH2– =Ar–H +NaCl+H2O+–CH2–CH=CH–CH2– There is some evidence for stepwise removal of chlorines from PCB molecules, and this has raised concerns that part-way through a run there may be produced lightly-chlorinated PCBs which have the highest dioxin-like toxicity. In practice this has not led to releases of such materials. Recovery of biphenyl from reaction mixtures might be expected if the reactions were simply those of hydro-dechlorination, but aromatic substances are also destroyed and carbon is among the reaction products, so the overall suite of reactions taking place is likely to be very complex. The paraffin is degraded, as double bonds are introduced into the molecules bringing about an increase in dielectric constant of the bulk material which makes it unsuitable for further use in the electricity industry. The BCD process typically can reduce PCB levels in paraffin to below 2 mg kg–1, at which level it is acceptable for energy recovery in appropriate furnaces. Materials presenting for destruction with higher organochlorine concentrations may be diluted with paraffin or blended with lower concentration materials to achieve optimum concentrations for the chemical reaction. A case in point is the neat PCB removed from small capacitors, which is removed by solvent

Alternative Technologies for Destruction of PCB and Other POPs

431

washing of such equipment. Unfortunately, shredded capacitors cannot be added directly to the BCD process because the alkali used therein would attack the aluminium cases, consuming alkali and producing hydrogen gas, and so removal of PCB from the carcasses is performed in a preliminary step. In some cases the organochlorine has been obtained by indirect thermal treatment of contaminated soil or other material, and is then added to paraffin for hydro-dechlorination. An improved, faster version of the BCD process has been developed by another Australian company, ADI, in conjunction with a New Zealand Crown Research Institute. The process has been used, in conjunction with thermal desorption, to destroy POPs and rehabilitate soil at the Sydney Olympics site. Designated ADOX, the ADI process uses a proprietary catalyst, said to be among those types listed in the original patents, which makes it much faster than the original BCD process. Batch reactions are complete within two hours, as opposed to eight or more hours for the prototype. There is a hidden advantage in this speed, since failure of the apparatus is less likely to occur when the dechlorination reaction is partly completed and the most toxic, lightly chlorinated, biphenyls are present. The thermal desorption process developed by ADI results in substantial destruction or organochlorines in an indirectly heated kiln at 250–480 °C. This achieved by charging the contaminated material together with additional reactants including lime, with the result that only a small proportion of organic material is desorbed. Volatilised material and fines may be recycled into the desorber or taken on for treatment in the ADOX base-catalysed dechlorination unit. The material initially containing up to 100,000 mg kg–1 organochlorines, when discharged from the kiln contains typically less than 0.02 mg kg–1, and less than 1 mg kg–1 dioxins (TEQ). No explanation for the chemistry which might be taking place has been advanced by the proponents, but the process obviously has similarities with some of those described under ‘nucleophilic substitution’ in Sect. 6. 5.2 EcoLogic Hydrogenation

In Kwinana,Western Australia, ELI EcoLogic operated a hydrogenation plant using technology invented in Canada by Eco Logic International Inc. The first commercial plant came on stream in Australia in 1995, and a second one at St Catherines, Ontario, in the following year although this has since ceased operation. In the process, hydrogen gas at approximately 900 °C brings about non-discriminant reduction of organic substrates, in which their carbon content appears as (predominantly) the hydrocarbon methane, and other elements are accordingly converted to their hydrides. Thus, PCB and other organochlorines produce of hydrochloric acid and methane and, where fluorine is present in the wastes being destroyed it is converted to hydrogen fluoride which may be a corrosion hazard. The EcoLogic process is only superficially similar to hydrogen gas reductions carried out under laboratory conditions – actually selective hydrogenations, although the adjective is seldom used – or in other sections of the chemical industry in that (a) no catalyst is used, the rate at high temperatures being suffi-

432

I.D. Rae

ciently high that catalyst is unnecessary, (b) no selectivity is evident, all substances being reduced, and (c) the reducing power of the system is very high, transforming paper into methane and water, for instance. The reductive conditions ensure that no dioxins and furans are produced and that any present in the feedstock are destroyed, the general destruction efficiency for organochlorines reaching eight nines. Arsenic passes through the system probably being converted to arsine in the reactor and back to arsenic trioxide during quenching, which emerges as a contaminant of the wash water. Initially, the hydrogen for the process was generated by the water gas shift reaction with recycling of the methane so that little extra hydrogen had to be added to the system, and this arrangement proved unsatisfactory and was eventually replaced by commercial hydrogen gas sources. Alternative plans for energy recovery from the methane were abandoned in favour of simple combustion. Similarly, early marketing of the hydrochloric acid produced by the EcoLogic system gave way to neutralisation and disposal of salt by landfilling. While material may be introduced directly into the reactor, the EcoLogic plant also included an batch vaporiser, an oven or thermal desorber in which containers such as drums or large capacitors may be heated to approximately 400 °C, while hydrogen gas circulates through and flushes volatilised material into the higher-temperature reactor. The combination of vaporiser and reactor has been used to destroy large quantities of PCBs and OCPs, especially DDT, and is especially suitable for equipment such as capacitors. Material after treatment meets stringent requirements for residual PCBs or OCPs, and stack gases contain less than 2 mg Nm–3 organochlorines and less than 0.1 ng Nm–3 dioxins (TEQ). In late 1999, trials were conducted with hexachlorobenzene (HCB) waste derived from past operations of an ethylene dichloride plant near Sydney. If this technology is chosen for the destruction of the HCB then a larger scale hydrogenation plant will need to be constructed. 5.3 Catalytic Hydrogenation

A catalytic hydrogenation process, using hydrogen gas and a sulfide catalyst, has been developed by the Coal and Energy Division of Australia’s government research organisation (CSIRO) in conjunction with an electricity distributor, and has recently been applied successfully to destruction of PCBs derived from the electricity industry. Few details are available of costs and scale of operation but it is clearly a useful addition to the suite of technologies available to holders of PCBs and possibly other POPs. The organochlorines are treated in solution at modest temperatures in a closed system. A major virtue of the process is that, whilst achieving six nines destruction efficiency for PCBs treated in paraffin solutions, the dielectric properties of the paraffin are largely restored and so it may be reused in the electricity industry. The value of such recycled oil, approximately $AUS 1500 ($US 900)/tonne, may be offset against the treatment cost with consequent improvement in the overall economics of the process. The chlorine from the PCB molecules is re-

Alternative Technologies for Destruction of PCB and Other POPs

433

leased as hydrogen chloride and ultimately removed as sodium chloride, while the aromatic residue may be retained in the oil as biphenyl or hydrogenated biphenyls. The catalytic hydrogenation process would seem to be applicable to other organochlorines as well, although these would need to be dissolved in a solvent such as paraffin which is inert to the hydrogenation conditions and arrangements would need to be made to remove hydrodechlorination products – benzene or cyclohexane, for example, from hexachlorobenzene – from the reaction mixture.

6 Nucleophilic Substitution 6.1 Fluidex (PCB Gone) Alkoxide Process

The process developed by S. D. Myers, and known originally as PCB Gone, is specific for treatment of PCB-contaminated paraffin in electrical equipment and is designed to be operated without removing the equipment from service, resembling in this aspect a kidney dialysis machine. The reagent is a mixture of potassium hydroxide (some descriptions of the process speak of ‘sodium’ and it is possible that metallic sodium is used as the source of the alkali) and a monomethyl ether of a polyethylene glycol having a molecular weight in the vicinity of 350 Dalton. The process bears some similarity to the base-catalysed dechlorination, and one major reaction path would be hydrodechlorination mediated by hydrogen transfer from the polyethylene glycol. As before, the chlorine would appear as chloride ion, in this case part of potassium chloride, but the dehydrogenated polyethylene glycol (being at the aldehyde oxidation level) is likely to resinify under the influence of the strong alkali and be removed in the clean-up step described below. However, it is also likely that chloride ion would be displaced from the PCB molecule by nucleophilic reaction with an alkoxide ion formed in equilibrium from the polymer alcohol, as shown in Eqs. (2) and (3): CH3O–(CH2CH2O)n–CH2CH2OH +KOH =CH3O–(CH2CH2O)n–CH2CH2O–K+ +H2O CH3O–(CH2CH2O)n–CH2CH2O–K+ +Cl–Ar =CH3O–(CH2CH2O)n–CH2CH2O–Ar+KCl For this reason we have classified the technology as one involving predominantly nucleophilic substitution, even though hydrodechlorination must be acknowledged as a co-reaction. In the PCB Gone process the organic by-products of reaction were absorbed by Fullers Earth, and regenerated paraffin was returned to the equipment. The PCB Gone process has been used in the United States, Canada and other countries for some years now, and over 70 million litres of PCB-contaminated oil has been treated to <2 ppm, a level widely adopted as constituting minimal hazard from PCBs, at least to the extent that any such hazard would be outweighed by hazards associated with the oil itself [10, 11].

434

I.D. Rae

S. D. Myers’ acquisition of the South African company Fluidex has seen the technology pass into their hands, and be enhanced with superior clean-up procedures involving regeneration of the Fullers Earth beds by proprietary processes. Energy Services International, a company established just north of Brisbane by the electricity distribution company Powerlink, uses a version of this process involving a sodium alkoxide to destroy PCBs which are found dissolved in paraffin oil following earlier efforts to retrofill drained electrical equipment. The facility is relocatable, having been transferred to New Zealand on one occasion but mainly operating in the Brisbane area where it is capable of treating up to 3000 litres of PCB contaminated oil in an eight-hour shift. No information is available about destruction of other than PCB materials and it seems unlikely that it would be used to destroy a broader range of POPs. 6.2 Ball Milling with Lime

Just as the addition of lime to soil contaminated with organochlorines, and treatment in an indirectly heated kiln can destroy a range of organochlorines (presumably through nucleophilic attack – see Sect. 4.1), efforts have been made to bring about reaction of organochlorines with lime, CaO, in ball mills. Research into this possibility was conducted jointly by the University of Western Australia and the Advanced Technology Department of the international minerals company Conzinc Riotinto Australia (CRA). Although the ball mill operated at comparatively low temperature, there was considerable heating at the point of impact of the balls and under these conditions rapid reactions were stimulated. The products of the reaction included graphite, calcium chloride and calcium hydroxide, and were of low toxicity, but simple chemical reactions cannot be written for the process without further information, which is not available. On a laboratory scale, the process was capable of handling a wide range of physical conditions of material, and destruction of PCB and DDT was achieved with high efficiency and no gaseous emissions. The process has not been commercialised, possibly because of the availability of competing technologies in Australia, but has attracted interest from scientists in Russia.

7 Other Processes 7.1 Soil Vitrification

In situ vitrification technology sometimes known as Geomelt was developed by the Pacific Northwest Laboratories Division of the Battelle Memorial Institute in the 1980s under a contract with the US Department of Energy. Geosafe Corporation was opened in 1989 to commercialise the technology and it has been used since 1993 in the remediation of a number of contaminated sites. The Geomelt

Alternative Technologies for Destruction of PCB and Other POPs

435

process involves the use of electric currents – relatively low voltage but high amperage – passed between large carbon electrodes to melt the contaminated soil. The obvious application is to inorganic contamination because, after the melting of the target volume is complete the mass is allowed to cool and contaminants remain trapped in the resulting glass. For example, this in situ vitrification has been applied successfully to immobilisation of radioactive wastes remaining after British nuclear tests were conducted in the 1950s in central Australia. Less obvious applications of the technology have involved the treatment of wastes in the United States and Japan containing organochlorines such as PCP, dioxins, PCBs, benzene hexachloride and other chlorinated pesticides. In partnership with Amec Engineering, Geosafe Australia has recently conducted trials in which hexachlorobenzene (HCB) wastes have been successfully destroyed. In these applications the waste is mixed with soil, sometimes in the presence of alkalis such as sodium carbonate or alumina, and melted at temperatures of 1100–1200 °C in a crucible, while steam is injected around the melt zone. The degradation of organochlorines appears to take place on the surface of soil particles with production of carbon monoxide, hydrogen chloride, and possibly hydrogen. The off-gases pass through a porous refractory, together with additional steam and oxygen, thus converting the carbon monoxide to dioxide and completing the reaction of organic products. Finally, these gases pass to a thermal oxidiser where they experience long residence times at high temperatures, ensuring that destruction efficiencies reach five to six nines, before the emergent gases are quenched and neutralised. Overall, the destruction reaction may be represented by Eq. (4): C6Cl6 +3H2O+4.5O2 =6CO2 +6HCl With the HCB waste, trials were conducted with 16.5 and 33% waste in the mixture, in crucibles holding 300 or 600 kg. If the process is adopted for the approximately 8000 tonne of HCB which is stored near Sydney awaiting treatment, a much larger facility would need to be constructed. 7.2 Soda Furnaces

As well as the highly engineered high temperature incinerators which dominate northern hemisphere waste destruction technology, there have been a number of attempts to use existing furnaces for destruction of PCBs and other POPs. Cement kilns, for example, are widely used for this purpose in the United States, although it is sometimes remarked that, in a reversal of normal priorities, some such kilns produce cement as a by-product of hazardous waste destruction. Cement kilns have the twin advantages of long hold-up times at high temperatures, and a strongly alkaline environment in which organochlorines might be expected to be chemically attacked as well as merely oxidised. These virtues also attach to the soda furnaces which can form part of the recovery system in pulp mills, where alkaline solutions of degraded lignins are burned and alkali is recovered for use in the wood-pulping process [12]. Temperatures of 1100–1400 °C are reached in the soda furnace and hold-up times as

436

I.D. Rae

long as 20–30 seconds, and trials at two Russian pulp mills (Archangelsk and Solombola) have demonstrated efficient destruction of PCBs with emission levels of dioxins typically 1–5 pg/Nm3, well below conventional emission targets. Other sludges and municipal wastes have also been treated successfully in the soda furnace trials, but no commercial-scale work with PCBs or other POPs has been reported. The experiments are interesting, however, in that existing furnaces with little modification except perhaps enhanced health and safety conditions for operators, may be used for destruction of hazardous wastes in ways that will be of great interest to developing countries and emerging economies.

8 Concluding Remarks When compared to other developed countries, the situation regarding destruction of PCBs and other POPs in Australia is an unusual one. Export of POPs materials is difficult and expensive, but the size of the local stockpiles are probably too small to justify the cost of establishing a high temperature incinerator similar to those used in Europe and North America. Some idea of the capacities and costs of establishing and operating these alternative facilities in Australia is given by the data in Table 1. On a unit cost basis, the alternatives are uncompetitive with high temperature incineration, but cost of transport (including insurance) to such a facility, when one is not established in a country which is distant from North America or Western Europe, must be kept in mind when weighing alternatives. Another factor influencing Australia’s decision to prefer the local development of alternative technologies is the strength of the environment movement in Australia and the opposition by its members to incinerator technology. This view has received bureaucratic and political support, overcoming whatever support existed among holders of industrial waste for the establishment of an incinerator facility. However, these industries have worked within the no-incinerator/no-export framework and have been prepared, with adequate notice, to pay for the destruction of the wastes they hold. In a few cases, shareholder and auditor pres-

Table 1. Australian treatment facilities for PCBs (1999)

Technology

Establishment cost $US¥106

Capacity Charge tonne/year $US/tonne

Typical feed

Incinerator

50

50,000

~300 Europe 200–3000 US <2000 ~4000

Broad range

4000–6000 1150–1650

PCBs, OCPs PCB solutions reaction (in-line)

Plascon (plasma arc) Base-catalysed dechlorination Ecologic hydrogenation Sodium alkoxide

1 0.2 10 ?

450 2200 1000 ?

Liquids, gases Solution in paraffin

Alternative Technologies for Destruction of PCB and Other POPs

437

sure has begun to be exerted on companies to remove POPs liabilities from their balance sheets. In order for Australia’s experience to be reproduced in other countries, all or most of these features would need to be reproduced. A number of other alternative technologies may be found in the literature, including those in wide-ranging surveys by the Canadian [13] and Australian [14] governments and by the United Nations Environment Programme [15], as well as in commercial information packages and the open literature. Most are variants of the reductive or nucleophilic substitution types discussed above, but none has yet found on-going commercial application. In continuation of their long crusade, the international Greenpeace organisation has recently endorsed the use of non-incinerator technologies for the destruction of persistent organic pollutants [16]. The most important “alternative” afforded by the alternative technologies is the avoidance of dioxin and furan formation which inevitably results when an oxidative process is applied to waste destruction. It is true that formation of these highly toxic materials may be minimised by choice of appropriate operating conditions, and reduced to extremely low levels in the gases emanating from dioxinremoval systems applied to flue gases. However, as with the old medical adage, “prevention is better than cure”, and only the reductive methods offer this advantage. Because the alternative technologies are selective, they often lack applicability to a broad physical range of material – solids, liquids, sludges – such as one associates with a hazardous waste incinerator or even a cement kiln. However, there is advantage in this selectivity because materials such as aluminium which are associated with the organochlorine wastes may be recovered for separate recycling. The trend to destruction rather than mere “disposal” of wastes means that archaeologists of the future will find nothing to remind them of the industrial chemical culture of the late twentieth century. The future’s loss is the present’s gain, however, since many toxic substances are thereby prevented from entering environmental compartments where they are unwelcome. To a present day historian the toxic substances released to air seem to represent a threat to human health like that posed by the miasmas of old and concern over these modern miasmas forces us to look back beyond the germ theory of disease to “something in the air”.

9 References 1. Lambert JB (1997) Traces of the Past. Unraveling the Secrets of Archaeology through Chemistry. Helix Books/Addison Wesley, Reading, Mass. 2. Theodore L, Reynolds J (1987) Introduction to Hazardous Waste Incineration. Wiley, New York, p 3 3. Chandler AJ, Eighmy TT, Hartlén J, Hjelmar O, Kosson DS, Sawell SE, van der Sloot HA, Vehlow J (1997) Municipal Solid Waste Incinerator Residues. Elsevier, Amsterdam, p 1 4. Eduljee GH (1994) Organic Micropollutant Emissions from Waste Incineration. In: Hester RE, Harrison RM (eds), Waste Incineration and the Environment. Royal Society of Chemistry, Cambridge, p 71

438

I.D. Rae: Alternative Technologies for Destruction of PCB and Other POPs

5. Williams PT (1994) Pollutants from Incineration. In: Hester RE, Harrison RM (eds) Waste Incineration and the Environment. Royal Society of Chemistry, Cambridge, p 37 6. Petts J (1994) Incineration as a Waste management Option. In: Hester RE, Harrison RM (eds),Waste Incineration and the Environment. Royal Society of Chemistry, Cambridge, p 1 7. Oppelt ET (1999) Overview. In: Rickman WS (ed), CRC Handbook of Incineration of Hazardous Wastes. CRC Press Inc, Boca Raton, p 3 8. Electric Arc Reactor. International Patent Application PCT/AU89/00216. Electric Arc Generating Device. International Patent Application PCT/AU89/00396. Material Processing, International Patent Application PCT/AU93/00089. 9. Method for the Destruction of Halogenated Organic Compounds in a Contaminated Medium. US Patent 5 019 175, 28 May 1991. Method for the Destruction of Halogenated Organic Compounds. US Patent 5 039 350, 13 August 1991. Method for the Base-Catalysed Decomposition of Halogenated and Non-Halogenated Organic Compounds in a Contaminated Medium. US Patent 5 064 526, 12 November 1991. 10. Brunelle DJ, Singleton DA (1983) Destruction/removal of polychlorinated biphenyls from nonpolar media. Reaction of PCB with poly (ethylene glycol)/potassium hydroxide. Chemosphere 12(2):183–196 11. DeFilippis D, Chianese A, Pochetti F (1997) Removal of PCBs from mineral oils. Chemosphere 35(8):1659–1667 12. Bukhteyev BM, Grudinin VP,Yufit SS (1999) Dioxin-free technique to dispose harmful and dangerous waste of any type and a commercial reactor for disposing polychlorinated biphenyls (PCBs) and other hazardous substances, personal communication 13. Piersol P (1989) The Evaluation of Mobile and Stationary Facilities for the Destruction of PCBs. Environment Canada, Ottawa 14. Environment Australia (November 1997) Appropriate Technologies for Treatment of Scheduled Wastes Review Report Number 4. Commonwealth of Australia, Canberra. Earlier reports were published in November 1994, November 1995, and August 1996 15. UNEP Chemicals and the Secretariat of the Basel Convention (1998) Inventory of Worldwide PCB Destruction Capacity. Inter-Organization Programme for the Sound Management of Chemicals, Geneva 16. Costner P, Luscombe D, Simpson M (1998) Technical Criteria for the Destruction of Stockpiled Persistent Organic Pollutants. Greenpeace International Science Unit