Roy Harrison OBE is Queen Elizabeth II Birmingham Centenary Professor of Environmental Health at the University of Birmingham. In 2004 he was appointed OBE for services to environmental science. Professor Harrison’s research interests lie in the field of environment and human health. His main specialism is in air pollution, from emissions through atmospheric chemical and physical transformations to exposure and effects on human health. Much of this work is designed to inform the development of policy.

Ron Hester is an emeritus professor of chemistry at the University of York. In addition to his research work on a wide range of applications of vibrational spectroscopy, he has been actively involved in environmental chemistry and was a founder member of the Royal Society of Chemistry’s Environment Group. His current activities are mainly as an editor and as an external examiner and assessor on courses, individual promotions, and departmental/subject area evaluations both in the UK and abroad.

Chlorinated Organic Micropollutants

By R. E. Hester, R. M. Harrison

The Royal Society of Chemistry

Copyright © 1996 The Royal Society of Chemistry
All rights reserved.
ISBN: 978-0-85404-225-8

Contents

Sources and Fates of Polychlorinated Dibenzo-p-dioxins, Dibenzofurans and Biphenyls: The Budget and Source Inventory Approach Stuart J. Harrad, 1,
Human Exposure to Dioxin Curtis C. Travis and April G. Nixon, 17,
Biological Uptake and Transfer of Polychlorinated Dibenzo-p-dioxins and Dibenzofurans Michael S. McLachlan, 31,
Atmospheric Chemistry of PCBs, PCDDs and PCDFs Roger Atkinson, 53,
Human Toxicology of Chlorinated Organic Micropollutants Stephen Safe, 73,
Ecotoxicology of Chlorinated Aromatic Hydrocarbons Pim de Voogt, 89,
Chlorinated Pesticides Alan Turnbull, 113,
Studies of Polychlorinated Biphenyls in the Great Lakes Deborah L. Swackhamer, 137,
Control of PCDD and PCDF Emissions from Waste Combustors Gev H. Eduljee and Peter Cains, 155,
Subject Index, 181,

CHAPTER 1

Sources and Fates of Polychlorinated Dibenzo-p-dioxins, Dibenzofurans and Biphenyls: The Budget and Source Inventory Approach

STUART J. HARRAD

1 Introduction

Polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs) and polychlorinated biphenyls (PCBs) have attracted considerable attention in recent decades, owing to concern over their potential adverse effects in humans and wildlife, which are compounded by their ubiquitous environmental presence and resistance to degradation. Amongst the 75 possible PCDDs, 135 PCDFs and 209 PCBs, there exists wide variation in physicochemical properties, bioaccumulative tendencies and toxicity. Figures 1 and 2 illustrate the basic structures and nomenclature of both PCDDs, PCDFs – collectively referred to as PCDD/Fs – and PCBs.

This chapter reviews our knowledge of several key issues pertaining to the environmental presence of these compounds. Constructing source inventories for a group of chemical pollutants permits the targeting of specific sources in order to reduce environmental emissions and hence human exposure, whilst the establishment of environmental budgets facilitates the identification of major reservoirs, and quantification of the extent to which a given pollutant has been released into the environment and been subsequently ‘lost’ via either biodegradation or environmental transport.

2 Physicochemical Properties and Environmental Levels

PCDD/Fs and PCBs possess low vapour pressures and water solubilities, along with high octan-1-ol/water partition coefficients (KOW values), which are listed for selected congeners in Table 1. When the long biological lifetimes of these chemicals are taken into account (human half-lives of up to 27.5 years have been reported for some PCBs), it is unsurprising that PCDD/Fs and PCBs display significant bioconcentration on ascending food chains, and this is borne out by a summary of their levels in the ambient environment (Table 2).

3 Environmental Budgets

Background and Limitations

In essence, establishing an environmental budget involves quantifying and ranking different environmental compartments as reservoirs of a given pollutant within a defined section of the environment, such as an individual country. The basic principle of an environmental budget is the derivation of a representative concentration for each environmental compartment considered (e.g. 10 µg kg-1 of soil), and its multiplication by an estimate of the volume occupied by that compartment. Whilst obtaining an accurate estimate of compartment volume is not as easy as it would at first appear (requiring answers to such questions as: to what depth are relatively immobile pollutants like PCDD/Fs and PCBs incorporated in soils, and what is the volume occupied by a compartment as loosely defined as ‘biota’?), much of the uncertainty involved in environmental budgets is due to problems in deriving representative pollutant concentrations. To illustrate, several attempts have been made to construct environmental budgets for both PCDD/Fs and PCBs. In each case, the accuracy of such efforts is restricted by the extremely limited database relating to concentrations in different environmental media and spatial variations in such concentrations. With regard to spatial variations, information regarding concentrations of these compounds in rural and remote locations is especially scarce. The significant temporal variations in PCB concentrations reported by some authors also hamper efforts to construct a meaningful budget, and budgets conducted using data recorded over a number of years may be subject to significant inaccuracies.

To illustrate the difficulties in deriving representative mean PCB concentrations in an environment as heterogeneous as the open ocean, whilst Tanabe cited a ΣCPCB concentration of 0.6ng dm-3 in North Atlantic seawater, Harrad et al. employed a value for North Atlantic and North Sea seawater (levels in this latter area significantly exceeded the former) of 0.12ng dm-3. Although Harrad et al. noted the higher estimates of others, they suggested that their own concentration estimate may have been too high, taking into account the concentration decline observed with increased sampling depth, and the fact that the bulk of the samples on which their estimate was based were taken only 6 m below the surface.

Clearly, the derivation of representative concentrations for each of the environmental compartments considered is crucial to the accuracy of any environmental budget. To illustrate, whilst Harrad et al. calculated the ΣPCB burden of a seawater volume of 1.14 x 1017 dm3 (including the North Sea) to be 14t, Lohse used a representative concentration of 3.5 ng ΣCPCB dm-3 to derive a seawater ΣCPCB loading of 150t for the North Sea alone, a volume of 5.25 x 1016 dm3. Such significantly different conclusions concerning the burden of a comparatively well-characterized location illustrates the extent of uncertainty associated with the construction of budgets, and particularly their dependence on accurate concentration data.

Despite these limitations, the construction of environmental budgets plays an important rôle in efforts to understand the environmental fate and behaviour of PCDD/Fs and PCBs, and the following section will examine a selection of the most detailed conducted to date.

PCDD/Fs

Harrad and Jones constructed a budget for the terrestrial UK environment. Unfortunately, although freshwater and freshwater sediments were considered, the absence of sufficiently detailed data relating to PCDD/F contamination of terrestrial biota and the marine environment, as well as the difficulty in deriving a representative concentration for an environmental compartment as diverse in composition as biota, meant that the significance of these potentially important reservoirs of PCDD/Fs was not quantified. The findings of this exercise (summarized in Table 3) were that, within the UK, topsoil represents easily the most important reservoir for tetra- to octachlorinated dioxins and furans, with other compartments such as freshwater sediments, ambient air, freshwater and vegetation making comparatively negligible contributions to the overall burden.

PCBs

The comparative ease of PCB measurement has generated a relatively detailed database relating to the presence of these compounds in the environment. As a result, the distribution of ΣPCB and PCBs number 28, 52, 101, 138,153 and 180 has been considered within an area encompassing the UK surface and the marine environment within a 200 km perimeter zone around the UK shoreline. The findings of this study (summarized in Table 4) are that topsoil contains the majority of the ΣPCB burden within the area considered, with other significant reservoirs identified as seawater and marine sediments.

For comparison, Table 5 lists the findings of Tanabe, who calculated the ΣPCB burden in the global environment. He estimated the global burden to total 374 000t of ΣPCB (31% of total world PCB production), with the overwhelming majority (96%) associated with seawater and coastal sediments. This estimate compares favourably with that of the United States National Academy of Sciences, who calculated that the oceanic water over the North American basin held 66000t ΣPCB. Overall the work of Tanabe and coworkers indicates the open ocean to contain around 61% and the terrestrial and coastal environment 39% of the global environmental burden of PCBs.

There are two apparent discrepancies between the two studies mentioned above. First is the fact that Harrad et al. identified topsoil as the major environmental reservoir of PCBs, whilst Tanabe pinpointed seawater as bearing the most significant fraction of the environmental burden. This difference is mainly attributable to the fact that although the and ocean surface area ratio of the UK environment considered by Harrad et al. is only slightly higher (at 0.44) than that which prevails for the Earth as a whole (i.e. 0.41, assuming global land and ocean surface areas of 1.48 x 1014 and 3.62 x 1014 m2, respectively), the ocean depth (200 m) assumed by Harrad et al. is considerably less than that assumed by Tanabe (3729 m). As a result, the land:ocean volume ratio of the area considered by Harrad et al. is significantly below that of the Earth overall.

The second apparent discrepancy is that whilst Harrad et al. calculated that the overall ΣPCB burden for the area considered amounted to 400t (ca. 1% of total UK sales of PCBs), Tanabe estimated that 31% of world PCB production had been released into the environment. More reasonable agreement between the two studies emerges when UK archived soil concentration data are used to calculate the burden of ΣPCB in UK topsoil at the peak of PCB contamination. The maximum total burden was estimated to be 23 200t (over 50% of total UK sales), a figure that correlates well with that of Tanabe. However, whilst it would appear probable that a significant fraction of PCB production has already ‘escaped’, it is equally apparent that there remains a significant fraction of PCBs with the potential for future release into the environment. Tanabe concluded that the bulk (65%) of world PCB production remained in use, or was ‘locked’ in landfills and hazardous waste dumps, and thus identified an urgent need to develop technologies capable of destroying such land-locked PCBs before they were released into the environment.

Comparison of the peak PCB burden with the present loading for the UK suggests that a significant fraction of UK PCB sales were released into the environment, but have since ‘disappeared’ from the UK. Firm evidence of the fate of these ‘lost’ PCBs is not available, but the most likely loss mechanisms are a combination of anthropogenic and natural degradation, along with atmospheric and pelagic transport away from the UK. Any assessment of the relative importance of these processes must remain largely speculative in the absence of unequivocal supporting data, but comparisons of the relative PCB losses likely from soils due to biodegradation and volatilization indicate that only the latter is likely to be able to account for the dramatic decline in the UK PCB burden over the last two decades. This conclusion is supported by reports that 210Pb corrected PCB fluxes to the Agassiz ice cap in Ellesmere Island, Canada, remained essentially constant over the period 1976/77 to 1985/86, at a level approximately 50% of that observed between 1970/71 and 1975/76. By comparison, levels in soils from several UK sites are reported to have fallen by a significantly greater factor (in 1984, the ΣPCB soil concentration at Woburn was ca. 10% of that in 1972) over a similar time period. Similar rates of decline have been observed in other UK abiotic environmental compartments like grass and air. Such observations lend credence to the hypothesis that PCB volatilization rather than degradation is the principal loss mechanism from the UK, otherwise the rate of decline would have been roughly the same in all locations. This concurs with the finding that by the early 1980s, just 3.6% of US PCB production had been degraded or incinerated.

When coupled with the detection of elevated PCB levels in biota from remote polar locations, the suggestions of large-scale volatilization of PCBs from temperate industrialized nations have spawned the ‘Global Distillation’ hypothesis. The central tenet of this hypothesis is that semi-volatile organic compounds (SVOCs), like PCBs, volatilize from warm and temperate locations, and subsequently undergo long-range atmospheric transport throughout the globe. Following deposition in polar regions, the extreme low temperatures in such regions minimize subsequent volatilization, with the result that over time one would expect a shift of the global PCB loading from temperate industrialized nations to both the Arctic and Antarctic.

4 Source Inventories

Background and Limitations

The production of source inventories is essentially the process of identifying and ranking sources of a given pollutant to a given section of the environment. Such rankings permit the identification of major release pathways, and hence the prioritization of emission control policies. The basic strategy of a source inventory is to derive an emission factor for a specific source activity (e.g. 10 µg per t of waste burnt), and subsequently to multiply this by an activity factor, i.e. the extent to which the activity is practised (e.g. 3 million t waste burnt per year). Generally, activity factors are far more reliable than emission factors (the latter can vary considerably) and uncertainties surrounding the latter are the principal source of inaccuracies in source inventories. In recognition of such problems, it is now becoming commonplace to assess the ‘quality’ of both the emission and activity factors used. Such ‘quality’ assessments (based, for example, on the number and range of concentration values used to derive an emission factor) are unavoidably subjective, but are still to be considered as a welcome development in what remains an extremely important but somewhat ‘inexact’ area of research.

In light of the uncertainties associated with source inventories, it is important that their validity is evaluated as far as possible. One way in which this may be achieved is the comparison of estimated total annual atmospheric emissions with estimated annual atmospheric deposition. For compounds possessing the environmental stability of PCDD/Fs and PCBs, one would in theory expect the two estimates to concur, with any shortfall in atmospheric emissions deemed indicative of a flaw in the source inventory. Of course, even if both emissions and deposition were perfectly characterized, one would not expect total agreement, as both PCDD/Fs and PCBs are subject to environmental degradation. Furthermore, national or regional source inventories fail to consider the influence of environmental transport of emitted material away from the nation/region of concern, as well as the potential ‘import’ of material emitted outside the boundaries considered by the source inventory. Indeed, the latter is the most probable explanation for the presence of PCBs in UK soils prior to the onset of large-scale UK use of these compounds in 1954. Aside of these potential sources of error, there exist possible problems with the derivation of accurate emission factors (discussed above), as well as difficulties in procuring a representative estimate of depositional inputs over a large area, when, for example, what little data are available indicate PCDD/F deposition to be far greater in urban than rural locations. Furthermore, it is by no means certain that current methods of sampling atmospheric deposition ensure 100% collection efficiency, with resultant uncertainty in estimates of depositional input. However, in spite of such problems, it is considered that the comparison of emission and deposition estimates is a worthwhile exercise, and its applications to both PCBs and PCDD/Fs are discussed in the following sections.

PCBs

Little doubt exists as to the principal source of the present environmental burden of PCBs, of which an estimated 1.2 million t were manufactured worldwide between the onset of their production in 1929 and the late 1970s, when their production in most western nations ceased. Despite this, there remains interest in identifying the sources of the continuing input of these compounds. Harrad et al. noted the conclusion of Eduljee that, in the absence of fresh production, volatilization of previously deposited material constituted the principal source of continuing PCB inputs, and using a theoretical treatment of PCB volatilization estimated the extent of such volatilization from topsoil as part of an attempt to construct a PCB source inventory for the UK. Their findings, summarized in Table 6, indicate revolatilization of previously deposited material from topsoil to be the largest current source of PCBs to the UK atmosphere, with other significant sources including leaks from PCB-filled transformers and capacitors remaining in service.

Emissions versus Deposition. Current PCB emissions to the UK atmosphere have been estimated to be around twice the current depositional inputs. Whilst the database on which this conclusion was made was limited, e.g. estimates of PCB volatilization were based on theoretical calculations only, and depositional input was derived from only 6 months data for two urban locations, the inference is that there is presently a net flux of PCBs out of the UK. This conclusion is supported by other workers, who have reported that water bodies such as Lake Superior now constitute a net source of PCBs to the atmosphere as a consequence of outgassing, and lends further support to the ‘Global Distillation’ hypothesis.

(Continues…)Excerpted from Chlorinated Organic Micropollutants by R. E. Hester, R. M. Harrison. Copyright © 1996 The Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.