PBDEs in the Environment Time trends, bioaccumulation and the identification

of their successor, decabromodiphenyl ethane

Amelie Kierkegaard

Doctoral thesis

Department of Applied Environmental Science

Stockholm University

Stockholm 2007

ii

© Amelie Kierkegaard, Stockholm 2007 ISBN 91-7155-410-6 Typesetting: Intellecta Docusys Printed in Sweden by Intellecta Docusys, Västra Frölunda 2007 Distributor: Stockholm University Library

iii

Abstract

Polybrominated diphenyl ethers (PBDEs) are important chemical flame re-tardants, but also environmental pollutants. Their behaviour in the environ-ment is a function of their inherent molecular properties, largely governed by the number and character of the bromine atoms substituted, and the micro-environment where they reside. In this thesis different aspects of importance for the understanding of the behaviour of lower brominated and higher bro-minated PBDEs in the environment are addressed.

The contamination of a Swedish freshwater system with lower bromi-nated BDEs was assessed by a retrospective study of pike from Lake Bol-men covering the time period 1967 to 2000. The concentrations of tetra- to hexaBDEs increased exponentially up to the mid-1980s and then leveled off/decreased slowly, possibly reflecting the voluntary reduction in produc-tion and usage of lower brominated BDEs in Europe. Methoxylated PBDEs were found to be present in similar concentrations to the PBDEs. However, there was no correlation between the levels of the two substance groups, and it was therefore concluded that they originated from different sources. To understand the low abundance of higher brominated BDEs in wildlife despite their extensive use and high levels in e.g. sediment, the dietary up-take of the fully brominated BDE, BDE209, was studied in fish. Although it was not expected to be taken up due to its large size and hydrophobicity, it was absorbed to a small extent via the diet. Once absorbed, BDE209 was reductively debrominated to nona- to hexa-brominated BDE congeners.

Reductive debromination in vivo was also demonstrated in dairy cows ex-posed to higher brominated BDEs in their natural diet. The transfer of BDE209 to milk was low (< 0.2 %). In contrast to PCBs and lower bromi-nated BDEs, there was no equilibrium between adipose tissues and milk fat, and for congeners with a log Kow > 7 a progressively smaller fraction of the ingested PBDEs was transferred to the milk. The results indicate that while lower brominated BDEs are excreted in the milk of dairy cows exposed to PBDEs, the higher brominated BDEs are accumulated in the meat.

At the same time that PBDEs were receiving increasing regulatory atten-tion, the next generation of brominated flame retardants was introduced. In this thesis decabromodiphenyl ethane, a replacement for the technical BDE209 formulation, was identified for the first time in the environment.

This thesis identified differences in uptake, metabolism and excretion for brominated compounds compared to the previously thoroughly characterized organochlorines. This knowledge will be useful for future risk assessments given the ongoing use of these brominated aromatic compounds.

iv

“To dare is to lose one's footing momentarily. Not to dare is to lose oneself.”

Sören Kierkegaard

v

List of papers

This thesis is based upon the following papers which are referred to in the text by their Roman numerals.

I Polybrominated diphenyl ethers (PBDEs) and their methoxy-

lated derivatives in fish from Swedish waters with emphasis on temporal trends, 1967-2000. Amelie Kierkegaard, Anders Bignert, Ulla Sellström, Mats Olsson, Lillemor Asplund, Bo Jansson and Cynthia A. de Wit Environ. Pollut., 2004, 130, 187-198.

II Dietary uptake and biological effects of decabromodiphenyl

ether in the rainbow trout (Oncorhynchus mykiss). Amelie Kierkegaard, Lennart Balk, Ulla Tjärnlund, Cynthia de Wit and Bo Jansson Environ. Sci. Technol., 1999, 33, 1613-1617.

III Fate of higher brominated diphenyl ethers in lactating cows.

Amelie Kierkegaard, Lillemor Asplund, Cynthia A. de Wit, Michael S. McLachlan, Gareth O. Thomas, Andrew J. Sweetman, and Kevin C. Jones Environ. Sci. Technol., 2007, 41, 417-423

IV Identification of the flame retardant decabromodiphenyl ethane

in the environment. Amelie Kierkegaard, Jonas Björklund and Ulrika Fridén. Environ. Sci. Technol., 2004, 38, 3247-3253.

The papers are reprinted with the kind permission of the publishers, paper I by Elsevier and papers II-IV by the American Chemical Society. I, Amelie Kierkegaard, made the following contributions: in Papers I and IV, I was responsible for the planning, chemical analysis, data evaluation, and writing the manuscript. In papers II and III, I took part in the planning of the project, was re-sponsible for the chemical analysis, data evaluation and manuscript writing. The exposure, sampling and biological part of paper II was performed by Lennart Balk and his co-workers. The mass balance study that provided the samples in paper III was planned and conducted by Gareth Thomas and his co-workers. The other co-authors each made valuable contributions to the planning, data evaluation, and/or writing of the manuscripts.

vi

vii

Contents

Abstract .......................................................................................................... iii List of papers...................................................................................................v Contents........................................................................................................ vii Abbreviations ............................................................................................... viii Introduction .....................................................................................................1 Objectives .......................................................................................................2

Brominated flame retardants...........................................................................4 PBDEs ........................................................................................................5

Properties and usage.................................................................................................5 Environmental occurrence .........................................................................................8 Toxicity .......................................................................................................................8

Next generation BFRs ................................................................................9 Decabromodiphenyl ethane.......................................................................................9

BFR Look alikes .......................................................................................10 Methoxy-PBDEs.......................................................................................................10

Analytical procedure......................................................................................12 Sample matrices.......................................................................................12 Extraction..................................................................................................13 Clean-up ...................................................................................................15 Instrumental analysis................................................................................16

Gas chromatography ...............................................................................................16 Mass spectrometry...................................................................................................18

Identification & quantification....................................................................20 Quality of the analysis ..............................................................................23

Results and Discussion.................................................................................27 Environmental levels - Temporal trends...................................................27

PBDEs......................................................................................................................27 Methoxy-BDEs .........................................................................................................30

Bioaccumulation .......................................................................................33 Dietary absorption....................................................................................................33 Biotransformation.....................................................................................................36 Aquatic versus terrestrial environment ....................................................................43

Risk assessment / Implications for exposure characterisation ................44 Decabromodiphenyl ethane – a next generation BFR .............................47

Conclusions...................................................................................................50 Acknowledgements .......................................................................................52 References....................................................................................................53

viii

Abbreviations

BAF bioaccumulation factor BCF bioconcentration factor BFR brominated flame retardant BMF biomagnification factor CV coefficient of variation DDT 2,2-bis(4-chlorophenyl)-1,1,1-trichloroethane DeBDethane decabromodiphenyl ethane ECD electron capture detector ECNI electron capture negative ionization ECS effective cross section EI electron ionization GC gas chromatography GPC gel permeation chromatography HRMS high resolution mass spectrometry Kow octanol/water partition coefficient LOD limit of detection LOQ limit of quantification LRM laboratory reference material LRMS low resolution mass spectrometry MeO-PBDE, MeO-BDE methoxylated polybrominated diphenyl ether,

methoxylated brominated diphenyl ether congener MS mass spectrometry PBDE, BDE polybrominated diphenyl ether, brominated di-

phenyl ether congener PCB polychlorinated biphenyl PCDD/F polychlorinated dibenzo-p-dioxin/furan PLE pressurized liquid extraction POP persistent organic pollutant SLE solid/liquid extraction SRM standard reference material STP sewage treatment plant

1

Introduction

Generally, halogenation makes an aromatic hydrocarbon less volatile and less water soluble, and also increases its stability. The chemical inertness of halogenated hydrocarbons also makes them attractive in many industrial processes. Unfortunately, resistance to degradation and lipophilicity are also characteristics of persistent organic pollutants, POPs.

The halogen substituent has a strong influence on the physical-chemical properties of halogenated hydrocarbons. While the small fluorine atom is strongly bound to the carbon, the bond between the large iodine and the car-bon is comparably weak. The resulting differences in the chemical properties are reflected in the uses of halogenated hydrocarbons in technical applica-tions. For instance, the stability of fluorinated hydrocarbons is a useful prop-erty in high temperature applications, whereas the thermal decomposition of brominated hydrocarbons is essential for their function as flame retardants.

According to the Stockholm convention, POPs are defined as “chemicals that remain intact in the environment for long periods, become widely dis-tributed geographically, accumulate in the fatty tissue of living organisms and are toxic to humans and wildlife” (1). The history of POPs includes a number of chlorinated organic compounds, many of which are still present in the environment worldwide, despite regulatory actions having been taken. Examples are pesticides that were deliberately produced to be toxic, such as DDT, industrial chemicals unintentionally released to the environment, such as PCBs, and chemicals formed as by-products in manufacturing and com-bustion processes, such as chlorinated dioxins. The detrimental effects of these chemicals were observed after decades of usage, when the environment was already extensively contaminated. One of the goals of environmental chemistry is to prevent this situation from arising again by providing a knowledge base that allows the environmental behaviour of chemicals to be foreseen or recognized at an early stage.

Although the use of many chlorinated hydrocarbons has decreased, the development and use of hydrocarbons substituted with other halogens con-tinues, with consequences for the environment that are largely unknown. Examples of such compounds are the brominated flame retardants, of which the polybrominated diphenyl ethers (PBDEs) are a prominent representative. Of vital importance for the risk assessment of PBDEs is the manner in which the bromine-carbon bond and changes in the degree of bromine substitution influence their behaviour in the environment.

2

Objectives

The overall objective of this thesis was to improve the understanding of the behaviour of brominated flame retardants in the environment. The results obtained serve to identify potential environmental risks of using brominated flame retardants (BFRs) and to assist environmental authorities in making risk assessments. In this rapidly moving field, the research in this thesis was conducted in the context of the state of the art at the time the work was done.

Paper I

A screening study of lower brominated diphenyl ethers, PBDEs, (up to pentabrominated, denoted pentaBDE) in Swedish biota set the starting point of this thesis (2). The results showed that these compounds were widespread, and the highest concentrations were detected in aquatic ecosystems. In Pa-per I the following questions were addressed: • Are PBDEs an increasing environmental problem in Sweden? Are the

levels of PBDEs in the environment increasing as a result of increasing usage?

Two studies of retrospective temporal trends were initiated, one in pike

from Lake Bolmen and the other in guillemot eggs from Stora Karlsö in the Baltic Proper (3). Both trends were reconstructed using archived material. By the end of the 1990s another family of brominated compounds, methoxy-lated PBDEs (MeO-PBDEs), were identified in fish and seal from the Baltic Sea (4). Their presence was also established in the pike samples from Lake Bolmen and other Swedish freshwater systems. The concentrations of the MeO-PBDEs in these samples were in the same range or higher than the PBDEs. The structural similarities between the two groups were obvious, but the origin of the MeO-PBDEs was unknown. By including the two most abundant MeO-PBDEs in the study, it was possible to investigate if the tem-poral trends of the two groups of compounds were correlated to one another. The question raised was therefore: • Are PBDEs the source for MeO-PBDEs present in Swedish freshwater

systems?

3

Paper II Other studies went on to show that tetra- and pentaBDEs are present in

biota all over the world. However, there were hardly any reports on the most frequently used PBDE product, the Deca-mix formulation, containing mainly the fully brominated congener, decabromodiphenylether or BDE209. A follow-up study of sediment and fish was therefore performed (5) in the same region where PBDEs were first detected in biota (6), the river Viskan, which received wastewater from textile industries known to have used the Deca-mix product. The sediment had high concentrations of BDE209, but only traces were found in fish. These findings led to the questions: • Why is the most frequently used BDE congener, BDE209, not found in

biota? Is it at all absorbed in fish? Could BDE209 metabo-lism/debromination be a source of the more abundant lower brominated BDEs found in fish?

Paper III

The results showed that BDE209 was in fact absorbed from the gut in rainbow trout, although to a small extent. However, further studies of bioac-cumulation in eggs from peregrine falcons revealed a dominance of higher brominated congeners (7), a pattern so far not seen in any organism from the aquatic environment. This led to the hypothesis that the bioavailability of higher brominated BDEs is greater in terrestrial organisms. To investigate the fate of higher brominated BDEs, a mass balance was performed in a ter-restrial organism of large importance for human exposure to many persistent organic pollutants, namely the cow. Since occupational exposure studies had shown that BDE209 was present in humans without known sources of expo-sure, an alternative exposure pathway was suggested to be via the food. The questions addressed were: • How do higher brominated BDEs behave in cows, exposed via their natu-

ral diet, regarding uptake, possible debromination and excretion to the milk? Could dairy products and/or meat be an exposure route for higher brominated BDEs to humans?

Paper IV

While an extensive research effort resulted in close regulatory scrutiny of both the lower brominated and higher brominated BDEs, the search for other persistent brominated chemicals in the environment continued. In a survey of BFRs in Swedish sewage sludge (8) a brominated substance eluting after BDE209 was detected, this led to the question: • Is the next generation of heavy BFR already present in the environment?

4

Brominated flame retardants

The increasing use of flammable polymeric materials in a range of commer-cial and consumer products has increased the demand for fire protection. Flame retardants are chemicals that, when added to materials, reduce their flammability. Different types of chemicals act in different ways to reduce flammability e.g. by cooling via the release of water, or by creating an un-combustible barrier on the surface of the material. Halogenated flame retar-dants inhibit fire mainly by scavenging oxidizing free radicals that are pro-duced during combustion (9). The lack of oxidizing radicals prevents the propagation of the fire. Although all halogens have this function, the bromi-nated compounds are the most effective. They have a high scavenging effi-ciency, a low cost, decompose at lower temperatures compared to chlorin-ated flame retardants (the other major class of halogenated flame retardants), and are compatible with a wide range of polymers (10,11). BFRs are particu-larly suitable for petroleum-based plastic and synthetic materials, and conse-quently they are common in home and office equipment such as casings for electrical appliances, copiers, TVs, computers, and mobile phones. In fact, 90% of all electronic appliances produced contain BFRs (11). Other impor-tant areas of use for BFRs are in furniture upholstered with flame retarded foams and textiles as well as insulating foams and other building materials.

BFRs can either be reactive, i.e. covalently bound to the polymer, or addi-tive, meaning that they are blended in the polymer system. Another way to introduce BFRs into the plastics is via the use of brominated monomers that are mixed with the original monomers before the polymerisation process, resulting in a polymer with built-in flame retardant properties. The most common BFRs are tetrabromobisphenol A (TBBPA), PBDEs and hexabro-mocyclododecane (HBCD). The former is mainly used as a reactive BFR while the latter two are additive products and therefore more prone to leak from the material (12). This thesis focuses on the additive BFR with the largest usage, namely the PBDEs.

5

PBDEs Properties and usage PBDEs are aromatic compounds substituted with up to 10 bromine atoms (Figure 1). Like PCBs, the theoretical number of possible congeners is 209 and the same numbering system as suggested by Ballschmiter et al. (13) is applied. The substitution pattern of the BDE congeners included in this summary is given in Table 1.

Figure 1. Chemical structures of the major substances studied in this thesis, i.e. PBDEs, MeO-BDEs and decabromodiphenyl ethane.

Commercial PBDEs are manufactured as three mixtures named after one of their major components: Penta-mix, Octa-mix and Deca-mix formulations. The industrial production is based on the bromination of diphenyl ether, a process that is terminated at different degrees of bromination (14). Due to

6

the chemical properties of the oxygen directing the bromine to para and ortho positions, and the steric hindrance of the substituted bromines, only a limited number of congeners are present in the PBDE products (10,15). The major components in the Penta-mix are tetra- and pentaBDEs, BDE47 and BDE99 (Figure 1, Table 1) as major congeners, in Octa-mix they are hepta- and octaBDEs, such as BDE183 (Table 1), and in the Deca-mix formulation the main component is decabromodiphenyl ether or BDE209 (Figure 1, Table 1) with small amounts of nonaBDEs.

Table 1. Substitution pattern of BDE congeners included in this summary.

BDE # Bromine substitution BDE # Bromine substitution

28 2,4,4’-triBDE 183 2,2’,3,4,4’,5’,6-heptaBDE 37 3,4,4'-triBDE 188 2,2',3,4',5,6,6'-heptaBDE 47 2,2’,4,4’-tetraBDE 190 2,3,3',4,4',5,6-heptaBDE 49 2,2',4,5'-tetraBDE 194 2,2',3,3',4,4',5,5'-octaBDE 77 3,3',4,4'-tetraBDE 196 2,2',3,3',4,4',5,6'-octaBDE 99 2,2’,4,4’,5-pentaBDE 197 2,2',3,3',4,4',6,6'-octaBDE 100 2,2’,4,4’,6-pentaBDE 198 2,2',3,3',4,5,5',6-octaBDE 104 2,2',4,6,6'-pentaBDE 201 2,2',3,3',4,5',6,6'-octaBDE 126 3,3',4,4',5-pentaBDE 202 2,2',3,3',5,5',6,6'-octaBDE 140 2,2',3,4,4',6'-hexaBDE 203 2,2',3,4,4',5,5',6-octaBDE 153 2,2’,4,4’,5,5’-hexaBDE 204 2,2',3,4,4',5,6,6'-octaBDE 154 2,2',4,4',5,6'-hexaBDE 206 2,2',3,3',4,4',5,5',6-nonaBDE 155 2,2',4,4',6,6'-hexaBDE 207 2,2',3,3',4,4',5,6,6'-nonaBDE 171 2,2',3,3',4,4',6-heptaBDE 208 2,2',3,3',4,5,5',6,6'-nonaBDE 173 2,2',3,3',4,5,6-heptaBDE 209 2,2',3,3',4,4’,5,5',6,6'-decaBDE 182 2,2',3,4,4',5,6'-heptaBDE

PBDEs are applied in a wide range of products, such as thermoplastics used in home or office furnishings, casings for electrical appliances, polyurethane foam, and synthetic fabrics. The typical addition varies between 5-30% by weight of the material to be flame retarded (14). The worldwide consump-tion of PBDE in 2001 (and 1999) was 67 000 tonnes, whereby Deca-mix formulation accounted for 83% (16,17). The usage of the PBDE products differs greatly between the continents. Compared to North America, both Europe and Asia have no or a low consumption of the Penta-mix product and in Europe also of the Octa-mix product (16,17). This is mainly due to differ-ences in regulations and/or voluntary initiatives. The European Union has banned the use of Penta-mix and Octa-mix formulations from 2004 (18) and, at least in Japan, the industry has voluntarily phased out the lower bromi-nated products (17). The Swedish import of technical Deca-mix is decreas-ing (see Figure 2) and a national ban on usage in for example textiles has

7

recently been adopted (2007) (19). However, most of the Deca-mix is im-ported in treated products and semi-finished goods (20).

Figure 2. Annual import of technical Deca-mix formulation to Sweden (tonnes) (21). Imported Deca-mix incorporated in goods and products is not included.

The physical-chemical properties of the PBDEs are in many ways similar to well known environmental contaminants, such as PCBs. The PBDEs are highly hydrophobic compounds with logarithms of the octanol/water parti-tion coefficients (log Kow) ranging from 5.08 to 8.70 for monoBDE to de-caBDE (22). While the log Kow increases with the number of bromine sub-stituents, the water solubility and vapour pressure both decrease. Hence, with increasing degree of bromination there is a diminishing tendency for PBDEs to be found dissolved in water or in the gas phase, they are rather sorbed to particles in the air or in the sediment or soil. The ability for long range trans-port is high for lower brominated BDEs, whereas for higher brominated BDEs the ability is linked to the distance that particles in the air are trans-ported (22,23).

Although the bromine-carbon bond generally is weaker than the corre-sponding chlorine-carbon bond, PBDEs are regarded as persistent in the environment. The physical-chemical properties of the BDE congeners differ greatly, which leads to differences in their environmental behaviour. For example, the reactivity or susceptibility to transformation measured as the rate of the hydrolysis reaction with sodium methoxide has been shown to increase with the degree of bromination (24). Furthermore, higher bromi-nated BDEs are known to be photolytically degradable in the presence of UV light, resulting in mainly lower brominated BDEs and brominated dibenzofurans (25,26). So far, photolysis reactions have only been confirmed in laboratory experiments, while for instance BDE209 in sewage sludge

8

showed no degradation 20 years after application of the contaminated sludge to agricultural soil (27).

Environmental occurrence The first finding of PBDEs in biota was reported in 1981 for pike from the Swedish river Viskan (6). In the years that followed, the presence of PBDEs was confirmed in biota from contaminated sites and from remote regions (e.g. 2,5,28-34), indicating that they are global pollutants. In 1998 it was reported that the level of PBDEs in human milk from Swedish mothers had increased exponentially over the previous two decades (35). In the same year, PBDEs were detected in sperm whales, indicating that PBDEs had reached deep Ocean sea waters (36). After that, the number of investigations on environmental levels of PBDEs worldwide increased exponentially. Sev-eral comprehensive reviews have been published that summarize the pres-ence of PBDEs in biotic as well as abiotic environmental matrices (37-41).

The experience gained from previous studies on other POPs set the initial focus on the aquatic environment. Generally, the dominant PBDEs reported in biota were the tetra- to hexabrominated congeners. The most abundant congeners are BDE47, BDE99 and BDE100, which also are the major con-geners in the technical Penta-mix formulation (42,43), albeit with differing composition. In sediment, dust and sewage sludge the major congener is generally BDE209 followed by the congeners present in the Penta-mix prod-uct. The pattern is similar to that of the commercial mixtures, and thus re-flects the usage and composition of the PBDE formulations (39).

Characterizing the levels and profiles of the PBDEs in the environment was one of the objectives of this thesis, and is further discussed in the Re-sults and Discussion.

Toxicity The knowledge of PBDE toxicity is still insufficient to predict their potential health risks. The interpretation of early studies is furthermore obscured be-cause commercial mixtures were used and the effects seen were at least partly due to more potent halogenated impurities in the products. The current knowledge has been summarized in several reviews such as (14,44-47) and will only briefly be presented below.

Generally, the acute toxicity of PBDEs is low and lower brominated BDEs cause effects at comparably lower doses than higher brominated BDEs (44). The most critical endpoint of PBDEs is the developmental neu-rotoxicity in mice that has been reported for a range of congeners (48-51) including BDE209. Neonatal exposure during a sensitive period of brain development in the fetus affected the spontaneous behaviour of the offspring and caused impaired learning and memory functions after they reached ma-

9

turity. PBDEs also seem to have effects on thyroid hormone transport and metabolism, suggesting them as potential endocrine disrupters (45). Studies in rats and mice have shown reduced levels of thyroid hormone in the serum after exposure to PBDEs (44,45,47). At high dosages, BDE209 has further been shown to induce tumors in rats and mice (44).

Ecotoxicological studies in birds also show that environmentally relevant doses of lower brominated BDEs cause immunosuppression, oxidative stress, reduced thyroid function, and decreased vitamins A and E (52-54). BDE47 and BDE99 were further found to have negative effects on the re-production and development of the invertebrate, Nitocra spinipes (55).

Next generation BFRs The growing concern about the global presence of PBDEs in the environ-ment has led to a search for alternatives. One approach has been to decrease the emissions of BFRs into the environment, beginning with the production and following through to the disposal of the products. Closing the system at the production sites, recycling of BFR treated products, and the use of poly-mers in which the BFR is covalently bound to the monomer are examples of this. Another approach is to find environmental friendly replacements. Ex-amples of BFRs that have been suggested to replace PBDEs are bis(2,4,6-tribromophenoxy)ethane (BTBPE) for the Octa-mix formulation and decab-romodiphenyl ethane (DeBDethane) for the Deca-mix. The latter is included in this study and is described below.

Decabromodiphenyl ethane Decabromodiphenyl ethane, or 1,2-bis(pentabromodiphenyl)ethane (see Figure 1), was introduced to the market already in the mid-1980s (56) but became commercially important as an alternative to the Deca-mix formula-tion in the early 1990s (57). This additive BFR was also designed to meet the strict European regulations on maximum amounts of brominated diox-ins/furans allowed in the product (57,58). The structural difference to BDE209 with carbon linking the aromatic rings also reduces the potential for producing dioxins or furans under pyrolysis conditions (59).

DeBDethane is marketed under the trade names SAYTEX® 8010 (Albe-marle Corp.) and Firemaster® 2100 (Chemtura Corp.) and has the same applications as the Deca-mix PBDE-product, i.e. as an additive to different polymeric materials like high-impact polystyrene (HIPS), and in textiles. Typical applications are in consumer electronics, such as TV cabinets, in building and cable insulation, and in adhesives. The recommended additions are the same as for the Deca-mix formulation. There are currently no figures on the global consumption of DeBDethane, but with increasing concern

10

about the PBDEs, the use of DeBDethane is predicted to increase in the fu-ture. Europe has no production of its own, but the import in 2001 was esti-mated to be a few thousand tonnes, primarily to Germany (57). Furthermore, in Japan there has been a clear shift in consumption away from Deca-mix to DeBDethane (Figure 3) (17).

Figure 3. Annual consumption (tonnes) of Deca-mix and DeBDethane in Japan from Watanabe et al. (17).

Based on the structural resemblance of DeBDethane to BDE209, its physi-cal-chemical properties are assumed to be similar, i.e. a low volatility and low water solubility. The inclusion of the ethane bridge between the aro-matic rings makes it slightly more hydrophobic compared to BDE209, but also gives the molecule conformational flexibility.

The knowledge on the toxicity of DeBDethane is scarce. The oral toxicity in rats was low, possibly due to poor absorption efficiency of the compound (60).

BFR Look alikes Methoxy-PBDEs A new group of brominated compounds structurally related to PBDEs was initially reported in Baltic herring, salmon and grey seals (4). The sub-stances, which were present at similar concentrations to the PBDEs, were identified as methoxylated tetra- and pentaBDEs (4). The presence of MeO-PBDEs was later confirmed in white-tailed sea eagle (61), in salmon (62) and in red algae (63) from the Baltic region. Additionally, early findings of

11

MeO-BDEs in biota from other regions were reported in pilot and beluga whales from Svalbard (64) and in dolphins, whales and dugongs from Aus-tralia (65,66). The common feature of the different MeO-PBDE congeners in Swedish biota was the position of the methoxy group, i.e. ortho to the ether bridge. The major congeners were identified as 6-methoxy-2,2’,4,4’-tetrabromodiphenyl ether (6-MeO-BDE47) and 2’-methoxy-2,3’4,5’-tetrabromodiphenyl ether (2’-MeO-BDE68) (see Figure 1) (62).

There is currently no known anthropogenic source of these compounds. Unlike the PBDEs, both hydroxylated and methoxylated PBDEs are among the wide range of naturally produced organobromine compounds (for a re-view see Gribble, 67). They have been identified in, for example, marine sponges and green algae (68-70). An alternative hypothesis to natural syn-thesis is that they may be metabolites of PBDEs, formed either by metaboliz-ing enzymes via methylation after an initial hydroxylation in vivo or via me-thylating microorganisms after excretion (4). Investigating the relationship between the MeO-PBDEs and the PBDEs was one of the objectives of this study, and this subject is further discussed in the section on environmental levels – temporal trends.

12

Analytical procedure

The analytical methods used in this study were developed from a goal-oriented perspective. As the experience in analysing a class of pollutants increases, the demands placed on the analytical methods typically rise. As the maturity of the analytical methods evolves, it becomes possible to make more detailed and reliable statements about the environmental occurrence and behaviour of the pollutants. An example of this development is the quan-tification in papers I, II and IV, which was accomplished by using the commercial mixtures as external standards, while a large number of single congeners was commercially available in paper III. Nevertheless, pioneer-ing work is a prerequisite for stimulating interest in a new pollutant class and mobilising the resources required for further analytical method development.

Generally, the chemical trace analysis of non-volatile organic environ-mental contaminants involves a number of steps of equal importance. These are sampling, extraction, clean-up, chromatographic separation and detec-tion, and finally identification and quantification. Even though all of the steps are relevant, those parts that distinguish BFRs from other organic pol-lutants will be emphasized in the following discussion.

Sample matrices The type of samples and the way the samples are collected are of importance for the analytical results. The choice of sample matrix depends on the type of question posed and should consider whether the samples are representative of the biological population or the geographical area that is investigated. In environmental monitoring these considerations are especially important, since it is necessary to ensure comparability between samples from different geographical locations and over time. A well-defined procedure for sam-pling/subsampling reduces the variation between parallel samples, thus fa-cilitating the detection of differences due to the variables targeted in the de-sign of the experiment/sampling program.

Most of the work in this thesis was concerned with biota. Sampling biota requires knowledge about the ecology of the organism, e.g. whether it is stationary or migratory, as well as many biological variables, such as age, gender, sexual maturity and nutritional status. The influence of these vari-ables on contaminant levels in biota and the consequences for biota sampling

13

have been thoroughly discussed by Bignert and colleagues at the Department of Contaminant Research, Swedish Museum of Natural History (71,72). The temporal trend in paper I was based on pike, a stationary predatory fish used in the monitoring of PCBs and DDTs in freshwater systems (73,74). In the feeding study (paper II), juvenile rainbow trout were used due to the well-documented background knowledge from their use as experimental fish in biochemical studies (75). Paper III involved samples of grass (silage) and adipose tissues, organs, milk and feces from cows, which together allowed the assessment of the total chemical input and output in the cows over the sampling period. Whereas the tissues were possibly not perfectly representa-tive due to a potential inhomogeneous PBDE distribution in the cows, the milk fat and feces samples were generated from pooled samples that were integrated over a defined time period.

For the purpose of screening “new pollutants” originating from the tech-nosphere, sewage sludge may be a particularly sensitive sample matrix in the sense that it integrates the emissions of chemicals that are used in society, both from domestic sources, traffic and small industries (39). The screening for DeBDethane in paper IV was accomplished with sewage sludge samples from sewage treatment plants (STP) distributed all over Sweden (8,76). Sediment is believed to be the major sink for highly brominated BDEs re-leased into effluents (17,39). The sediment analysed in paper IV originated from an area heavily polluted by Deca-mix (77) and therefore suspected to represent a potential sink for emissions of DeBDethane.

Extraction The physical-chemical properties of the PBDEs (and MeO-PBDEs) make them extractable into organic solvents by methods typically used for tradi-tional lipophilic POPs. Extraction, clean-up and separation/detection meth-ods for PBDEs have been reviewed by Covaci et al. (78). Methods fre-quently used to extract PBDEs from solid matrices are liquid/solid extrac-tions (LSE), either by sequential extractions after vibrating, rotating or soni-cating the sample (5,79,80), or by a continuous solvent extraction as in column (4,81) or Soxhlet extraction (79,82), whereby the latter uses hot sol-vents. The extraction efficiency is dependent on the power of the solvent, the accessibility of the matrix to the solvent, and extraction time. A high temperature will increase the efficiency of the solvent by decreasing its viscosity (thereby increasing solute mobility) and increasing the solubility of the analytes. However, the use of high temperatures and exposure to UV-light may confound the extraction of octa- to decaBDE due to their propensity for degradation.

Other extraction methods used for PBDEs are pressurized liquid extrac-tion (PLE) (83-85), supercritical fluid extraction (SFE) (86-88) microwave

14

assisted extraction (MAE) (89,90) and more recently, solid-phase microex-traction (SPME), with or without prior solvent extraction (91,92). Unfortu-nately, only a few of the “new” extraction methods reported in the literature include higher brominated BDEs. PLE, which lately has become a common method for PBDE extraction, applies temperature and pressure to increase the diffusive power of the solvents. Apart from being automated, it has the advantage of the possibility to use on-line cleanup (84,93).

Generally, lipophilic compounds such as PBDEs (and MeO-PBDEs), are associated with the lipids in the tissues and therefore extraction methods that effectively extract lipids will be efficient in extracting the PBDEs as well. The lipid content is usually determined gravimetrically, either as part of the extraction clean-up (this study) or by applying a total-lipid method such as Bligh & Dyer to an aliquot of the sample. The biota samples in the present thesis (papers I-III) were extracted according to the cold LSE method origi-nally described by Jensen et al. (94). The method comprises repetitive ex-tractions by n-hexane/acetone and n-hexane/diethyl ether. The extraction efficiency of the lipids, mainly triglycerids, in fatty fish such as herring or rainbow trout (paper II), was comparable to the well known method of Bligh & Dyer (95). Later, however, it was found to be less accurate (lipid extraction efficiency ~ 75%) for lean fish such as pike. This was suggested to be due to the increased proportion of phospholipids in the lean fish, and the method was therefore modified to increase the extraction efficiency in matrices with a low fat content (< 1%) (96). Nevertheless, the pike samples in the time trend study (paper I) were all extracted by the original biased method. Moreover, the fat content decreased significantly over the studied period, from the late 1960s to 2000, implying that the lipid content was more biased at the end of the period compared to the beginning. To increase the comparability of the samples, the trend was therefore presented on a fresh weight basis.

A modified version of the LSE original method by Jensen et al., where acetone is replaced by isopropanol was used for the extraction of cow feces in paper III (96). For the extraction of the low-level (with respect to PBDE) samples, i.e. adipose tissue, organs and milk fat, all with a fat content > 1%, the original solvents were used to reduce the contamination from isopropa-nol (97). The silage samples were Soxhlet extracted with dichloromethane.

Sediment, sewage sludge and mineral supplement samples in paper III and IV were cold-solvent extracted with acetone/n-hexane according to Nylund et al. (98), whereas for sediment acetone/toluene was used. The air sample and the insulation tube samples were sonicated with dichloromethane and n-hexane, respectively (paper IV).

15

Clean-up The crude extracts of environmental samples normally contain nonpolar and semipolar compounds that are coextracted in large quantities with lipophilic target analytes, such as the PBDEs. The coextractants may coelute with the analytes (precluding or falsifying identification), distort the chromatography of the analytes, cause unwanted matrix-effects in the ion source, or even change the performance of the GC column and injection system. The general strategy of sample clean-up was to keep the number of steps as few and effi-cient as possible, both to reduce the loss of analyte and to minimize the pro-cedural blanks, e.g. from the solvents used.

The bulk of coextractants present in the sample extract are removed by destructive or non-destructive methods. Liquid partitioning with concen-trated sulfuric acid or column “filtering” by impregnating the acid on silica are destructive methods frequently used for lipid removal in biological sam-ples. In case more labile compounds are to be analysed, a non-destructive method, like gel permeation chromatography (separation mainly based on molecular size) is preferred (78). Other non-destructive methods for lipid reduction are the use of adsorbents, like aluminium oxide, florisil and to some extent silica, applied singly or in combination. The capacities of the latter two are low compared to alumina or treatment with sulfuric acid (78). All substances analysed in this study are resistant to strong acid. Sulfuric acid was used for bulk removal of lipids and other components in the major-ity of the samples in this study (paper I-IV). The lipid content in the sam-ples varied from < 1% in pike up to 100 % in the adipose tissue. In low-level samples where a small final extract volume was necessary, an addi-tional sulfuric acid impregnated silica column was used to further reduce the remaining lipids (paper I, III, IV). For the analysis of milk fat (paper III), the degree of lipid removal needed for an accurate determination of BDE209 could not be achieved with reasonable amounts of sulphuric acid. Therefore, saponification of the lipids with potassium hydroxide (1 M KOH in ethanol) for one hour at 60 °C was used. This method is simple and powerful with respect to lipid removal and keeps the solvent consumption low compared to e.g. GPC, but it can also be hard on labile analytes. Degradation of octa-chlorodibenzofuran/dioxin and o,p’-DDT and p,p’-DDT have previously been reported (99,100,100a). This is the likely explanation for the compara-bly low recovery (53%, coeffient of variation: 12%) recorded for the surro-gate standard, 13C-BDE209 in milk fat (paper III).

Further clean-up usually involves fractionation of different classes of compounds or additional removal of interfering compounds. In papers III and IV an extra column was needed to remove acid-resistant nonpolar satu-rated hydrocarbons, present in varying amounts in the samples. These coex-tractants were removed by fractionation on a silica column, 1 g activated at 450 ºC over night (paper IV). In order to get a more reproducible deactiva-

16

tion of the silica, the activated silica was deactivated with water (2%) in glass ampoules that were immediately sealed until use (paper III).

The presence of elemental sulfur in sediment and sewage sludge, often in high concentrations (mg/g dry weight in typical sewage sludge), distorts the chromatography and interferes in electron capture detection (101). Common methods for sulfur removal are fractionation with GPC or the use of a metal, usually elemental copper, to form metal sulfide (for an overview see 102). The copper powder/granulates can be added to the extract or mounted on the top of a silica column (82). The sulphur present in sewage sludge and sedi-ment (paper IV) was removed by the use of tetrabutylammonium sulphite to oxidize the sulfur to the water soluble thiosulfate (103). This method was, however, not optimal for DeBDethane, since it caused degradation via de-bromination in standard solutions. This problem was probably less pro-nounced in the sewage sludge and sediment samples due to the presence of a protecting matrix. For the objective of this study it was considered tolerable.

Instrumental analysis Gas chromatography The final separation/detection of most BFRs and look alikes such as MeO-PBDEs is generally accomplished by gas chromatography/mass spectrome-try (GC/MS) analysis and, in some cases, also by GC equipped with an elec-tron capture detector (ECD). The instrumental aspects of the analysis of PBDEs were recently reviewed by Stapleton (104). Due to the limited num-ber of congeners present in the commercial products, a 30 m non-polar or semi-polar column is normally sufficient for the separation of lower bromi-nated BDEs in environmental samples. A comparative study of 126 BDE congeners run on 7 capillary columns (~ 30 m) suggested DB-XLB to be the best choice based on coelutions with other BDE congeners or other bromi-nated substances known to occur in environmental samples (105). However, the presence of unidentified congeners produced via abiotic or biotic debro-mination may necessitate further separation. Likewise, other brominated substances of anthropogenic or natural origin such as MeO-PBDEs may require more careful assessment of the chromatographic separation (see later discussion on coelutions). The analysis of higher brominated BDEs, in par-ticular BDE209, requires different instrumental conditions. For example, the DB-XLB column suited for lower brominated BDEs was highly discriminat-ing against BDE209 (106). The difficulty with the analysis of BDE209 is that the temperatures needed for its vaporization in the injector and in the column (if it is to elute within a reasonable time) overlap with the tempera-ture range in which it is thermally degraded. Degradation in the column is seen as a raised baseline before the peak, whereas degradation in the injector

17

results in the elution of lower brominated BDEs (mostly octa- to nonaBDEs) as distinct peaks. To reduce the thermal degradation of BDE209, the resi-dence time in the column as well as in the injector should be minimized (106). A short column of 15 m or less with a thin stationary phase (0.1 µm) is a good choice. The different GC conditions needed for the lower and higher brominated BDEs also imply that a complete PBDE analysis is opti-mally performed on two separate columns.

The injectors usually applied for PBDEs are hot vaporizing injectors such as splitless, pulsed splitless, programmable temperature vaporization (PTV) or cold on-column. The degradation of higher brominated BDEs encountered in splitless injectors is caused partly by the temperature itself, but also by interactions with active surfaces in the injector. The transfer efficiency can be low for splitless injectors, a fact that is beneficial for dirty samples in trapping non-volatile components in the sample that may be detrimental to the column. However, it also discriminates against high boiling analytes. There is no discrimination in on-column injectors, but they require “clean” samples.

Figure 4. Fraction of BDE209 (in %) degraded to nonaBDEs quantified using HRMS with on-column injection or LRMS with splitless injection (3 min. at 280 ºC, paper III). The results are the mean of 41 samples and 11 procedural blanks, re-spectively. Vertical bars represent the 95 % confidence interval for the mean of the sum of the nonaBDEs.

In paper III the cumulative degradation of BDE209 over the whole analyti-cal procedure was quantified using two different injection systems. Quanti-fied on a molar basis, an average of 4 % of the total BDE209 content de-graded to nonaBDEs in samples analysed with a cold on-column injector, probably reflecting the degradation during extraction/cleanup (Figure 4).

18

When a splitless injector was used, the corresponding degradation was 7%, whereby BDE207 and BDE208 accounted for the major difference in the products formed. BDE206 was detected at similar yields in both systems, which suggests that this degradation happened earlier in the analysis and was not governed by thermal reactions in the injector. The formation of BDE207 and BDE208 seemed to be accentuated in the procedural blanks, possibly because other components in the matrix shielded the analytes from exposure to surfaces that could trigger catalytic reactions.

Although degradation of DeBDethane was encountered during extrac-tion/clean-up, it appeared to be less sensitive to thermal degradation than BDE209. Thus the conditions selected for BDE209 were applied to DeBDe-thane as well (paper IV).

Mass spectrometry In a few investigations (e.g. 34) the electron capture detector (ECD), an in-expensive halogen selective detector, has been used in PBDE analysis. The risk for coelutions is, however, large (107). Thus, today, a mass spectrometer is almost always used.

The most frequently applied MS techniques for PBDEs are low resolu-tion MS (LRMS), operated in either electron ionization (EI) mode or in the electron capture negative ionization (ECNI) mode, or high resolution MS (HRMS) operated in EI mode. Similarly to the organochlorines the bromi-nated compounds, with the natural bromine isotope ratio of 50.5% of 79Br and 49.5% 81Br, give rise to fragment ions with isotope distributions charac-teristic for the number of bromine substituents on the fragment ion. ECNI involves the ionization of a reagent gas, like ammonia or methane that via collision with the high energy electrons emitted from the filament generates less energetic thermal electrons that subsequently ionize mainly electronega-tive compounds. The thermal electrons are captured by the analyte either in a dissociative (1) or a non-dissociative (2) process.

Brominated compounds generally form ions from a dissociative electron capturing process, often dominated by the production of bromide ions. Monitoring the bromide ions is one of the most sensitive detection methods for PBDEs (28,108,109) with instrumental limits of detection (LOD, ex-plained later in this section) of 7-400 fg for di- to heptaBDEs (110,111). The detection is selective for brominated substances but gives no structural in-formation and the identification relies solely on differences in the retention time. The fragmentation patterns of brominated substances depend on their

AB + e AB- -therm

A + BAB + e- -therm ( )1

( )2

19

chemical structure, but also on the temperature of the ion source, the electron energy, the system pressure, the choice of reagent gas, and to some extent also on the design of the ion source. The ion source parameters have been optimized in several investigations for lower brominated BDEs (110,111) and for BDE209 (112). The instrumental conditions used in paper I aimed to achieve a robust method. Therefore the ion source was kept at a high tem-perature to favour a complete fragmentation of the analytes and to reduce the condensation of heavy contaminants (in particular BDE209, 84) in the source. The sensitivity drop encountered with methane as the reagent gas (also reported for isobutene, 110) due to the deposition of carbonized mate-rial in the source was avoided by using ammonia, which instead has a clean-sing function. The detection limits achieved with ammonia are similar to those of methane (111). The instrumental conditions used for the analysis of higher brominated BDEs and DeBDethane in paper II-IV were similar al-though different instruments were used.

A common alternative to ECNI is electron ionisation (EI) either in LRMS or HRMS. The ionization is accomplished by the emission of high energy electrons generated from the filament that collide with the analytes, produc-ing characteristic molecular and/or fragment ions. Generally EI gives more structural information about the compounds ionized. The analysis of PBDEs in EI is more selective than ECNI since the ions monitored are [M-2Br]+ or [M]+, which allows distinction between BDE congeners of different homo-logue groups as well as other classes of brominated compounds. The instru-mental LOD for PBDEs analysed in LRMS-EI is however at least one order of magnitude higher than in ECNI, with decreasing sensitivity from mono- to heptaBDEs (110,111), although improvements have been reported for LRMS-EI combined with large volume injection (113) as well as the use of ion storage MS (114). Still, with HRMS-EI both the selectivity and the sen-sitivity are better than with LRMS-EI, and the instrumental LOD for lower brominated BDEs are similar to those in ECNI-LRMS (115,116).

The major advantage of EI compared to other ionization methods is the possibility to apply isotope dilution by the use of 13C-labeled internal stan-dards, which improves the precision of the analysis. The use of isotope dilu-tion is also possible in ECNI for BDE209 due to the abundancy of the phenoxide ion, [C6Br5O]− (112). The formation of phenoxide ions in ECNI was previously reported by Buser (109). Another advantage of phenoxide ions over bromide ion detection is an enhanced signal-to-noise ratio due to reduced chemical noise at higher masses (112). The occurrence of higher mass fragments in ECNI decreases with decreasing number of bromine sub-stituents, but by optimizing the source parameters to enhance the production of phenoxide ions the instrumental LOD has been shown to approach that of bromide detection for 14 of 39 tri- to heptaBDEs investigated (110). The phenoxide fragments can also be used to differentiate between congeners within the same homologue group, such as octaBDE with 3 and 5 bromines

20

in the two aromatic rings versus a congener with 4 bromines in each ring, since different phenoxide ions are produced (43). Some of the unidentified brominated substances detected in paper II were characterized by the iso-tope distribution of the phenoxide ions as discussed in the following section.

In paper III, octa- and nonaBDEs were quantified using larger fragments in ECNI-LRMS (phenoxide ions) as well as in EI-HRMS ([M-2Br]+). A prerequisite for the cow study was the achievement of low detection limits to enable the quantification of a range of congeners in all samples collected. The production of the larger fragments in ECNI varies depending on the pattern and number of bromine substituents on the BDE congener. For ex-ample, the nona-BDE206 and the octa-BDE196 have a lower response com-pared to the nona-BDE207 and the octa-BDE197, respectively, possibly indicating that the number of ortho-substituted bromines favours the re-sponse. Furthermore, La Guardia et al. (43) reported that hepta-BDE171, which is substituted with 3 and 4 bromines in the two rings, produced de-tectable phenoxide ions from both rings, whereas no high mass ions were detected from its homologues BDE190 and BDE173, both of which have 2 and 5 bromines substituted in the rings. A symmetrical octaBDE like BDE197 (and hypothetically BDE194 and BDE202) or BDE209 gives a greater response since only one phenoxide ion is formed, independent of what half of the molecule is ionized. However, even though some octa- and nona-brominated congeners were detected with satisfying sensitivity in ECNI in paper III, others had a lower response. This, together with the less disparate response factors within homologue groups in EI, favoured the choice of HRMS-EI. For BDE209 however, ECNI was preferred due to a higher sensitivity.

Identification & quantification A prerequisite for an accurate quantification is the unambiguous identifica-tion of the peak by minimizing coelutions and the use of good internal and external standards. As mentioned above, bromide ion detection in ECNI makes no distinction between brominated substances. Thus, there is a risk for erroneous results due to coelutions. Coelutions for congeners reported in environmental samples that are present in the technical mixtures, are summa-rized in Table 2.

Among the most frequently occurring PBDEs in environmental samples, some congeners are more subject to coelutions. For example, BDE28 and BDE49 may coelute with other BDE congeners, and BDE99, BDE153 and BDE154 with other BFRs or natural brominated compounds. However, for BDE47, the most abundant congener in biota, no major coelutions have been reported. Moreover, coelutions can be region or matrix specific. For in-

21

stance, the presence of the naturally produced MeO-PBDEs in samples from the Baltic Sea should be considered in the quantification of BDE99 and

Table 2. Coelutions reported for congeners present in the technical mixtures (and reported in environmental samples) detected with ECNI after separation on various nonpolar to semipolar 30 m columns (i.d. 0.25 mm). Coeluting compounds detected in environmental samples are marked in bold, numbers represent BDE congeners. Data is summarized from a (105), b (43), c (117), d (3) e (78), f Paper I, g (62).

Coelutes with BDE# DB-1a DB1-HT b DB-5 a,c,d,e,f CP-Sil-8 g DB5-HT b HT-5 a DB-XLB a

J&W Sci. J&W Sci. J&W Sci.,CAgilent fChrompack J&W Sci. SGE Int. J&W Sci.

28 16, 33 33 16,33 33 16, 33, 38

49 68,80 71, 48 68 degr.prod of HBCD* 71, 48 68 46, 48, 68,

71 47 degr.prod of HBCD 66 42 42 75 51 51 85 6-MeO-BDE99

99

5-Cl-6-MeO-BDE47 6'-Cl-2'-MeO-BDE68 2',6-diMeO-BDE68**degr.prod of HBCD

116

100 4'-MeO-BDE49 unknown br.‡ 109

118 97 97 97 119 120 120 155 126 126 138 166 HBCD 153 TBBPA HBCD

154 BB153† Me-TBBPA††

BB153 Me-TBBPA unknown br.

173 190 190, 171 190 181,190 190 183 BB169 BB169 175 185 192 182 197 204 204 203 198 198

* Degradation products from hexabromocyclododecane, ** named BC-11 by the author (117), † Bromobiphenyl 153, †† Dimethyl tetrabromobisphenol-A. ‡ unknown brominated substance

possibly BDE100 (62). Furthermore, pentabromoethylbenzene may errone-ously be assigned as BDE37 on a DB5 column (L. Asplund, personal comm.). A minor coelution was encountered for BDE154 (paper I), which is why height instead of area was used in the quantification. The possible coelutions among the higher brominated BDEs, mainly hexa-, hepta- and to some extent octaBDEs, are less known since the number of commercially available congeners still are limited compared to lower brominated BDEs.

22

Coelution problems in EI are restricted to congeners within the same homologue groups and to other compounds brominated or non-brominated with interfering mass fragments/molecular ions present in the samples. For the latter type of coelutions HRMS is highly selective. Still, isobaric ions of for instance tetraBDEs [M-2Br]+, pentaPCB [M]+ and heptaPCB [M-2Cl]+ requiring a resolution of 82 000 and 23 000 respectively to separate, have been reported (107,118). For instance, BDE47 coelutes with CB180 (2,2',3,4,4',5,5'-heptachlorobiphenyl) on a 30 m HP-5 column, and on a cor-responding nonpolar column with CB191 (107). Other potential coelutions are tetra- to hexabromodibenzofurans with the corresponding PBDEs. For example [M+4]+ from tetraBDF and [M+6]+ from penta- to hexaBDFs re-spectively, require a resolution of about 30 000 to differentiate from the cor-responding [M+2]+ and [M+4]+ of the tetra- to hexaBDEs (119). However, since the isomer distribution is two amu higher for the BDE congeners, the ratio between the two most abundant ions should reveal major contributions from the corresponding brominated furan. Regarding coelutions with other BDEs however, only one of those presented in Table 2, namely BDE155 and BDE126, represent interference between homologues whereas the re-maining are within homologues and therefore valid also for HRMS (105). Due to the limited number of congeners available, potential coeluting iso-mers of the hepta- and octaBDEs quantified both in EI and ECNI in paper III could therefore not be excluded. BDE203 in the samples may thus poten-tially be assigned to BDE198, BDE197 to BDE204 and BDE173 to BDE190 on a 15 m short column (see Table 2).

A good internal standard (surrogate standard) should have physical-chemical properties similar to the analyte. An ideal choice is therefore 13C-labeled standards because they perfectly mimic losses during the analytical procedure. However, the use of 13C-labeled standards is as previously men-tioned restricted to MS-EI or for BDE209 in ECNI. Another approach is to establish the differences in behaviour between analyte and the surrogate standard of choice in the analytical procedure, by recovery experiments. Examples of surrogate standards used in ECNI are less abundant BDEs, such as BDE77 (115), BDE104, BDE140 (120) or for example decabromobi-phenyl for BDE209 (79)(for additional examples see 78). Monofluorinated derivatives of PBDEs comprise another alternative having similar properties but eluting before the corresponding BDE congener (ortho- and meta-substituted) (121). The internal standard in paper I, 2,2',5,6'-tetrachlorobiphenyl (CB53), was the original internal standard used in the monitoring of PCBs and DDTs (122). CB53 was far from ideal for PBDE analysis, but it was a prerequisite for using the archived extracts. The extrac-tion/clean-up method was the same over the studied period and the relative recoveries between the surrogate standard and the BDE congeners were es-tablished. At the time of the BDE209 feeding study (paper II), no 13C-labeled BDE209 was commercially available. The choice of dechlorane was

23

based on its long term use in previous PBDE investigations (2,5). Dechlorane was also used in the preliminary quantification of DeBDethane (paper IV), which enabled direct comparison to previous results of BDE209 from a larger survey of sewage sludge samples (8). In paper III, the labeled congeners 13C-BDE183 and 13C-BDE209 were used.

For a long time, PBDEs in the environment were quantified against the technical products due to the lack of single congeners. The Penta-mix, Bromkal 70-5DE, the Deca-mix formulation, Dow FR-300BA, and the tech-nical DeBDethane product Saytex 8010®, were used in paper I, II and IV, respectively. To increase the accuracy of quantification in retrospective analysis (paper I), the Bromkal 70-5DE product was characterized and the major components quantified (42). Further characterization of the commer-cial mixtures has recently been published for the purpose of identifying the source of congeners present in the environment (43,105). The two MeO-BDEs present in pike (paper I) were initially quantified using the response factors of BDE47, which was the closest eluting BDE congener with the same degree of bromination. A conversion factor was later calculated for the retrospective correction of the original quantifications.

PBDEs are believed to accumulate in the body lipids like many classic organic pollutants. In this case the concentrations normalized to lipid weight will enable the levels in tissues and organs with different fat content or even between different species to be compared. In some cases it may still be more relevant to normalize to fresh weight, e.g. for food analysis aimed at human exposure. In the present thesis, the intention was to normalize to fat content, but due to several factors, the concentrations in paper I and II were ex-pressed on a fresh weight basis. In paper I the fat content decreased from about 0.8% to 0.4% from the late 1960s to 2000. Apart from the problems encountered with the low extraction efficiency as previously discussed, there was no correlation for pike sampled in a given year at a given location, be-tween lipid content and fresh weight concentration of the PBDEs. This fact by itself is an argument against lipid weight normalization (123). In paper II, the lipid content of the muscle tissue decreased during the experiment whereas the lipid content in liver was about the same. To remove decreasing lipid content as a confoundry factor in the uptake curve of BDE209 a con-servative approach was taken and concentrations were expressed on a wet weight basis.

Quality of the analysis Establishing a good quality control and practice for the analytical procedure is necessary for the comparison of analytical data of different origins. A number of criteria and rules for the analysis of PBDEs in biota have recently been suggested (124) and the results from four interlaboratory studies show

24

satisfactory results for lower brominated BDEs, while the higher brominated BDEs (BDE183, BDE209) still need attention (125,126). The accuracy of the method can be determined by analysing standard reference material (SRM). There are currently seven SRMs with certified and reference values for PBDE congeners (127) and another two candidate SRMs (128). The SRM matrices are cod liver oil, fish tissue, whale blubber, human serum, and flounder (candidate SRM), all of which contain tri- to heptaBDEs. House-dust (SRM) (129) and sediment (candidate SRM), also contain high levels of BDE209. In the following, the special measures taken for the analysis of BDE209 are emphasized.

A Recovery experiment based on spiked samples, e.g. 13C-labeled recov-ery standards, quantified against an injection standard is a measure of how much of the analyte is lost in the analytical procedures (absolute recovery). When internal standards other than 13C-labeled are used, it is vital to investi-gate the recovery of the analytes compared to the recovery of the internal standard (relative recovery) and if there is a large discrepancy to correct for the difference. Relative and absolute recoveries of the PBDEs versus CB53 were determined in spiked samples at two concentration levels (paper I). The relative recoveries were considered acceptable (on average 97-106%), i.e. within the range of the expected variation from parallel samples and were therefore not corrected for. The relative recoveries of dechlorane used in paper II and IV were 110-113 % for tetra- to hexaBDEs in fish (paper II, unpublished) and 114 % for BDE209 (5) and 89% for BDE209 in sedi-ment (unpublished). Despite the somewhat different yield for the BDE con-geners compared to dechlorane, corrections were not made. Dechlorane has on the other hand later been applied in several interlaboratory studies and has shown good results (126,130).

The extraction efficiency in solid matrices can not be evaluated using spiked samples. Adding a surrogate standard to an environmental matrix can as a rule not simulate the form in which the analyte is associated with the matrix. A common way to examine the extraction efficiency is to succes-sively extract the same sample using either the same method or different methods, so called exhaustive extractions. The extraction efficiency for biota and sediment with the methods used in this study was satisfactory for all BDE congeners, with less than 1% remaining for all BDE congeners studied.

The use of a laboratory reference material (LRM) provides a measure of the precision of the method or the within-laboratory reproducibility over time. It also serves as a warning/alarm system for both occasional discrepan-cies and long term deviations. LRMs of herring and pike were included in the temporal trend study from 1996. The coefficients of variation for the low level pike were 10-26 % and for herring it was 6-19 %. These are within the expected range suggested by Horwitz et al. (131). LRM was also prepared and analysed for the different matrices in paper III. However, with the two

25

extraction occasions per matrix performed, the data set was too small for any statistical analysis.

One or several procedural blanks covering the whole or parts of the ana-lytical procedure were analysed in parallel with the samples. In low level samples the concentration in the blanks often dictates if the samples can be quantified. Thus, a protocol for checking solvents and use of separate glass-ware is recommended. In paper III, two blanks for each extraction occasion were used (containing on average 38 pg BDE209/blank), and for the milk fat analysis three additional blanks were employed covering parts of the proce-dure. In all results presented (paper I-IV) the amount in the blanks was sub-tracted from the samples. The addition of a matrix, such as corn oil to the blanks, may present an alternative to avoid losses due to the lack of a “keeper” compared to the samples. This may also prevent a more extensive degradation of BDE209 during extraction/cleanup as well as in the injec-tor/column as previously discussed (Figure 4).

PBDEs can be expected in a laboratory environment equipped with com-puters and other electronic devices. Lower brominated BDEs have been re-ported in laboratory air (132), in offices (133), and high concentrations of BDE209 have been detected in dust (134,135). Subsequently, all handling of extracts exposed to the air should be minimized and the extracts should al-ways be protected from particle deposition. The contamination of the sol-vents (with both higher and lower brominated BDEs) varies between batches and needs to be tested before use. In general, solvent usage should be kept to a minimum. All glassware was heated to 450 ºC overnight and solvent rinsed before use. Other sources of BDE209 contamination are plastic materials such as polyethylene bottles, insulating foams and lubricant oils present in for instance blenders. For the same reason, unnecessary electric appliances and upholstered furniture/chairs were avoided as well as unpackaging of goods in the laboratory where extraction and clean-up took place.

The degradation of BDE209 warrants attention, not only the thermal deg-radation occurring in the GC/MS system but also the degradation induced by UV light or in contact with active surfaces/materials in the analytical proc-ess. The laboratory was equipped with UV-filters on windows and the fluo-rescents lamps. The extracts were kept in brown glassware or covered in aluminium foil. Despite these precautions, BDE209 degraded throughout the whole procedure. The use of isotope dilution compensates for the loss of BDE209, ensuring an accurate quantification, but a problem arises if nonaBDEs and octaBDEs are quantified in the same samples. The major products from the degradation of BDE209 are lower brominated BDEs formed by a successive debromination. In paper III, an approach to trace the degradation is presented. The surrogate standards 13C-BDE183 and 13C-BDE209 were used for quantification of all higher brominated BDEs. By tracing 13C-nona-, 13C-octa- and 13C-heptaBDEs (apart from 13C-BDE183) formed by the degradation of 13C-BDE209 in the samples and subtracting a

26

proportional fraction from the native octa- and nonaBDEs, it was possible to reconstruct and quantify the amount of octa- and nonaBDEs that was present from the start. The correction was made assuming a first order degradation process. A procedural blank containing only 13C-BDE209 was used to trace the potential debromination to 13C-BDE183. Although the input of nonaB-DEs from BDE209 was traced by this method, the concurrent degradation of nonaBDEs to octaBDEs was not accounted for. Additionally, separate calibration standard series for BDE209 and the lower brominated BDEs were used. In general, the calibration curves used in this thesis were prepared from standard solutions at 6-9 concentration levels, analysed 2-3 times in mixed order with the samples.

The lowest level that can be quantified is ultimately defined by the limit of detection (LOD) achieved for the sample matrix with the applied method. The International Union of Pure and Applied Chemistry (IUPAC) has de-fined the limit of detection as “the concentration or quantity derived from the smallest measure that can be detected with reasonable certainty for a given analytical procedure” (136). The smallest measure, XL, is given by the equa-tion:

k is a factor chosen according to the level of confidence desired. A k factor of 3 thus represents the amount where the signal is just distinguishable from the background (with a confidence level of 99.7 % separated from the noise) and is often used to define the LOD (137). The limit of quantification (LOQ), representing the lowest level that can be reliably measured in a sam-ple, requires a higher precision and a factor (k) of 10 is therefore recom-mended (137). A common simplification of the above method is to use a signal to noise ratio (S/N) measured around the analyte retention in the chromatogram. The LOQ concentration is estimated from the signal that corresponds to 2 to 5 times the S/N ratio.

In the analysis of PBDEs, the LOQ for the major congeners in the techni-cal products (BDE47, BDE99, BDE209) are often defined by their presence in the procedural blanks. The LOQ in this thesis was determined for every sample series based on the S/N or the blank levels. In papers I, III and IV it was defined as 5 times the S/N, (corresponding to k > 10 in the IUPAC defi-nition) or, if present in the procedural blanks, as 5 times that amount. In pa-per II the LOQ for BDE209 was defined as S/N > 3 due to a larger peak width compared to later analyses, which was caused by instrumental differ-ences. Despite the reduced height/area ratio, the peak was clearly distin-guished from the noise. Measured values above the LOD but below the LOQ were used in paper I and III in the statistical evaluations and in the figures, respectively, as they represented the best available estimates.

27

Results and Discussion

This thesis made several contributions to the understanding that has been garnered over the last 15 years about the behaviour of BFRs, in particular PBDEs, in the environment. One focus was on the identification and assess-ment of BFRs as an environmental problem. For the lower brominated BDEs, this included determining whether environmental levels were increas-ing and therefore a growing risk (paper I). For DeBDethane, a second gen-eration BFR, the contribution was establishing its presence in the environ-ment (paper IV). A second focus was in the field of bioaccumulation, in particular the absorption, biotransformation, and excretion of higher bromi-nated BDEs (papers II, III). Despite the differences between the systems studied, aquatic versus terrestrial organisms, fish versus mammals, and spiked food versus naturally contaminated food, certain common characteris-tics of bioaccumulation behaviour were identified. These will be discussed in the following section, addressing:

• Levels and temporal trends of BDEs and MeO-PBDEs in the envi-

ronment • Dietary absorption, biotransformation and excretion of higher bro-

minated BDEs in organisms from the aquatic and the terrestrial envi-ronment, and the resulting implications for environmental and hu-man risk assessments

• Occurrence and potential impact of DeBDethane, a second genera-tion BFR

Environmental levels - Temporal trends PBDEs The first papers reporting the presence of PBDEs in environmental samples (in 1979) were from abiotic matrices such as sewage sludge, soil and sedi-ment sampled close to manufacturing facilities in the USA (138,139). In 1981, PBDEs (tetra-hexaBDEs) were for the first time identified in biota (6). The highly sensitive detection method of bromide ions in ECNI (108), en-abled the detection of PBDEs in wildlife (seal, sea eagle and guillemot) from remote regions suggesting that these compounds were global pollutants (28).

28

Other early reports of PBDEs in the environment included avian tissues and eggs, bottlenose dolphins and human adipose tissue from the USA (29,140,141) fish, shellfish, and sediment from Japan (30,142) and fish and seals from Europe (31,32). A more comprehensive screening of Swedish biota from aquatic as well as terrestrial ecosystems was performed in the beginning of the 1990s (2). It revealed that PBDEs (BDE47, BDE99, BDE100), like PCBs, were present in terrestrial organisms but more abun-dant in aquatic species.

Analysis of dated segments of a laminated sediment core indicated that PBDE levels were increasing in Baltic sediment (98). Against this back-ground, two retrospective temporal trend studies were initiated, one in guil-lemot eggs from the Baltic and the other in pike from Lake Bolmen, an area without known sources of PBDEs in the south of Sweden (paper I). Both trends were first presented in 1993 (143), showing increasing levels (sum of BDE47, BDE99 and BDE100) from 1967 to 1991 in pike, whereas the guil-lemot trend indicated a decrease at the end of the period (1989-1992). This was in contradiction to the previously reported PBDE trend in cod liver from the North Sea (31) and in eel from the Rhine and Meuse rivers, but in con-currence with the increasing time trend of eel from the Roer River (144).

Since then, numerous papers have been published documenting the wide-spread distribution of PBDEs in remote regions as well as in “hot spot” areas close to manufacturing plants/user sites. High PBDE concentrations, up to 60 ppm on lipid weight, have for example been reported in fish and earth worms from Sweden (5,27), and in carp, rainbow trout, and Forster’s tern eggs from Virginia, Washington, and California, respectively (USA) (145-147), all collected from contaminated sites. Concentrations of similar magni-tude, but with a larger fraction of higher brominated BDEs, were detected in eggs from Nordic populations of wild peregrine falcons (7,148,149). High levels have also been reported in a number of marine mammals, such as por-poise (150), bottlenose and white-beaked dolphins (151),(150) and pilot whales (152). The presence of PBDEs in deep marine food webs (36,150) as well as in biota from the Arctic (41) and Antarctic (153) regions confirms their global scale distribution and long range transport. Extensive reviews of environmental levels have recently been published (37-41,154).

The initial temporal trend measured in pike from Lake Bolmen, was later extended to include the hexaBDEs, BDE153 and BDE154, and two meth-oxylated tetraBDEs (paper I). All BDE congeners showed increasing con-centrations up to the mid-1980s, which then leveled off or decreased to the end of the period, in 2000. The additional data from 2005 (Figure 5, unpub-lished data, K. Nylund) confirm the decreasing trend.

29

Figure 5. Arithmetic mean concentration (pg/g wet weight) of A) BDE47 and B) 6-MeO-BDE47 in pike from Lake Bolmen, 1967-2005. One outlier with extremely high concentrations in the sample series of 2005 was excluded (n=7).

The variation, both within and between the sampling years, encountered in the pike study was extensive. Between years, the variation may be explained by external factors like temperature or food supply. The within-year varia-tion is to a small extent explained by the analytical uncertainty but is more likely caused by variation in concentrations from food items and from indi-vidual physiological differences, e.g. in the absorption efficiency and in the rate at which the compounds are metabolized and excreted. A way to reduce the within-year and between-year variation is to normalize the data to the concentration of a similar compound that is influenced by climatic or physio-logical differences in a similar way. For example, normalizing the levels of BDE154 to CB153 markedly reduced the variation (Figure 6). CB153, 2,2',4,4',5,5'-hexachlorobiphenyl, is a legacy contaminant that has had stable concentrations in this lake between 1988 and 2000 (data missing for 2001-2004). Even though the decrease of BDE154 was not statistically significant by itself, the BDE154/CB153 ratio was. Thus, relative to CB153, BDE154 was decreasing. This reinforces the interpretation that BDE154, like the other PBDEs, decreased between 1991 and 2005 in pike from Lake Bolmen.

The initial increase and peak of PBDE levels in pike from Lake Bolmen (paper I) was similar to the trend observed in guillemot eggs from the Baltic (3) and in roach from another Swedish fresh water system, Lake Krankesjön (155). The more dramatic decrease in the guillemot trend may be explained by the location of Lake Bolmen, it being closer to local sources of emission. A recently published trend of PBDE levels in mussels from the English Channel (1981-2003) was similar to the Bolmen-trend, with a peak at the beginning of the 1990s followed by a leveling off or slow decrease (156).

30

Figure 6. Concentration ratios of BDE154 and CB153 in pike from Lake Bolmen, 1991-2005 (A). The ratios are based on the years for which both BDE154 and CB153 were analysed in the same specimens (outlier of 2005 included), and show a significant decreasing trend (log-linear regression curve (P = 0.0012). The corre-sponding concentrations (geometric means, 95% confidence intervals) versus time plots for B) BDE154 and C) CB153 are also presented.

While the decrease likely reflects the voluntary reduction in production and usage of PBDEs in Europe, temporal trends from other parts of the world differ. For example, the levels of lower brominated BDEs in a range of North American wildlife have increased exponentially (157-161) up to year 2000. In 1999 and 2001, the American continent accounted for 95% or more of the global demand of the Penta-mix (16,17) and has up to now only vol-untary agreements on reduction in its production and usage. These trends have lately slowed down, while the levels in Arctic biota are expected to continue to increase, partly due to the ongoing processes of long range trans-port to the Arctic regions (38). Thus, temporal trends for seals and beluga whales from the Canadian Arctic showed no signs of leveling off (summa-rized by de Wit et al., 41). Likewise, the PBDE levels (including BDE209) in peregrine falcon eggs from South Greenland increased continuously from 1986 to 2003 (149) although the exposure is partly from their migration to Central and North America during the winter.

Methoxy-BDEs The two tetrabrominated MeO-BDEs, 6-MeO-BDE47 (6-methoxy-2,2’,4,4’-tetrabromodiphenyl ether) and 2’-MeO-BDE68 (2-methoxy-2’,3,4’,5-tetrabromodiphenyl ether) (Figure 1), that were present in the pike samples (paper I), had previously been found in fish, seals and predatory birds from

31

the Baltic region (4,61,162). The levels reported were all in the same range as the PBDE levels, but in contrast to the PBDEs they had no commercial application. The lack of any known direct anthropogenic source, and their structural resemblance to the PBDEs led to several hypotheses, one of which was that these compounds originated from the metabolism of PBDEs (4). Salmon plasma was shown to contain hydroxylated PBDEs (162), which could be interpreted as supporting this hypothesis. Furthermore, the corre-sponding MeO-BDEs were present in the neutral fraction of the plasma, which indicated that they had a common source (162,163). Suggested bio-transformation pathways are enzymatic oxidation followed by methylation, enzymatic or by microorganisms in the intestine, or direct methoxylation by microorganisms in e.g. the sediment (4). Hydroxylated metabolites formed from PBDEs had previously been reported in fish and rodents (164-166). Furthermore, methoxy-hydroxy-BDEs were detected in rats exposed to 14C-BDE209 via the diet (166). However, although 6-OH-BDE47 was tentatively identified in fish dosed with 14C-BDE47, its methylated form was absent (poster of 164). Thus, even if 6-MeO-BDE47 is a plausible metabolite of BDE47 via hydroxylation and subsequent methylation, there is no apparent precursor for 2’-MeO-BDE68 (62,63), suggesting that there are other sources of this compound.

Unlike PBDEs, both hydroxylated and methoxylated PBDEs belong to the wide collection of naturally produced brominated organic compounds (for a review see Gribble, 67). Both MeO-BDEs prevalent in the Baltic biota and in Bolmen pike had previously been found in large quantities in tropical marine sponges (66,69,167,168) and 2’-MeO-BDE68 has also been observed in green algae from Japan (70), further supporting their natural origin. In addition, high concentrations have been recorded in marine mammals from the Southern Hemisphere in which only traces of PBDEs were detected (65).

Although MeO-PBDEs are naturally produced in sponges and algae from the marine environment, it was surprising to also find them commonly pre-sent in fish from freshwater systems (paper I). The purpose of including the MeO-PBDEs in the retrospective temporal trend study of PBDEs was to investigate if there was any covariation between concentrations of the PBDEs and the MeO-PBDEs that could support the theory that the MeO-PBDEs were products of PBDEs. The results were clear, none of the MeO-BDEs showed any correlation to the PBDEs, and the potential metabolite of BDE47, 6-MeO-BDE47, had a decreasing trend over the time period, in contrast to PDE47, which initially showed an exponential increase (Figure 5). A temporal trend in cod liver samples from the Norwegian Arctic coast collected during 1987-1998 showed increased levels (a factor of 10) of MeO-tetraBDEs in 1992 and 1993 compared to the years preceding and fol-lowing (169). The high concentrations found in Bolmen pike in the late 1960s compared to later years suggest a similar variability in concentrations. In conclusion, the results support the hypothesis that the two MeO-BDEs

32

were naturally produced. The natural origin of the same MeO-BDEs was recently confirmed in True’s beaked whale from the North Atlantic by natu-ral abundance radiocarbon measurements, which were consistent with a natural formation as opposed to an industrial synthesis based on petrochemi-cal feedstock (170).

The majority of the naturally produced OH-/MeO-PBDEs reported in the literature so far have the hydroxy/methoxy group at the ortho position (62). While the metabolism of PBDEs at the ortho position could not be excluded, the presence of hydroxylated/methoxylated BDEs at the meta or para posi-tions is more indicative of metabolism of PBDEs (62,171). For example, the 4’-OH-BDE49 identified in Baltic salmon (62), in various fish species from the Detroit river (172), and in glaucus gull from the Norwegian Arctic (173) was suggested to be a metabolite of BDE47 (by 1,2-shift of bromine and hydrogen) and/or BDE49 (62).

The natural source of the MeO-BDEs is still not fully understood. A ten-tative spatial trend of herring collected in 1987 showed increased concentra-tions from the Kattegatt/North Sea to the Bothnian Bay (paper I) and no agreement with the corresponding BDE47 concentrations. This was also observed in 2001 and 2002 (174). A range of OH-/MeO-PBDE pairs have been identified in salmon blood, blue mussels and in the red macroalgae sp. Ceramium from the Baltic region (62,63). Their presence in red algae may represent one of the sources in the Baltic and can possibly explain the lower levels found in herring from the Kattegatt.

Table 3. Characteristics of the sampling location and concentration of 6-MeO-BDE47 and 2’-MeO-BDE68 in fish samples collected in Swedish lakes.

Species, Lake, sampling year n Sampling

season Region Degree of productivity

MeO-BDE68 (ng/g lw)

MeO-BDE47 (ng/g lw)

Pike, Bolmen, 1967-2000

141+17* Spring Woodland Mesotrophic 25-240 50-500

Pike, Storvindeln, LRM, 1993

1* Winter Remote Oligotrophic 21 7

Pike, Storvindeln, 1978-2000

7 Spring Remote Oligotrophic 4-20 2-4

Roach, Krankesjön, 1980-1996**

64 Autumn Agricult. Eutrophied n.d. n.d.

Pike, Roxen, 1972

9 Spring Agricult. Eutrophied 0.3 0.4

Perch, Stensjön Hjärtsjön, Bysjön, 2001***

24 Autumn Woodland Oligotrophic/ mesotrophic ~0.3-0.6 ~0.07-0.15

* pooled samples, ** Results from (155), *** Unpublished results (L. Asplund), n.d. not detected.

33

The natural origin of the MeO-PBDEs in freshwater systems is intriguing. In order to investigate if the source of the MeO-PBDEs was produced by a primary producer and thus related to eutrophication, fish samples from a number of Swedish lakes distributed all over Sweden, from oligotrophic to eutrophic and from populated areas to remote regions, were analysed (Table 3) (paper I). No correlation between MeO-PBDE levels and any of the vari-ables was found. Although the data set in this study is limited, it suggests that the source is neither favoured by eutrophication nor comes from emis-sions connected to human activity.

Bioaccumulation According to the Stockholm Convention, bioaccumulation is one of the key characteristics in the classification of POPs (1). While bioconcentration is the result of the diffusive equilibration of the chemical between the organism and the media in which the organism lives, i.e. water for aqueous and air for terrestrial organisms, bioaccumulation represents the combined uptake from the bioconcentration and from the diet. Some bioaccumulating compounds biomagnify in the food chain. For hydrophobic compounds, biomagnifica-tion can be defined as bioaccumulation leading to higher lipid weight nor-malized concentrations in an organism compared with its food. Here, the first focus is on one of the key processes determining a chemical’s ability to biomagnify, namely the dietary absorption efficiency. This process depends on the physical-chemical properties of the substance, the type of food, and the strength of the chemical’s sorption to the food. The second focus con-cerns other factors that affect the extent of bioaccumulation, namely bio-transformation after absorption as well as excretion, these will be addressed later in this chapter.

Dietary absorption Despite early findings of BDE209 in human adipose tissue in 1991 (141), it was, based on its high molecular weight and extreme hydrophobicity, claimed not to be bioavailable and therefore unable to bioaccumulate in wildlife (175). The initial screening studies performed pointed in the same direction, in contrast to the lower brominated BDEs, no BDE209 was de-tected in fish despite high concentrations in sediment (5,176). The uptake of BDE209 was therefore investigated in rainbow trout fed dried cod chips fortified with the technical Deca-mix (paper II). Both liver and muscle lev-els of BDE209 increased with time of exposure, thus demonstrating that it was absorbed from the gut in fish. The absorption efficiency was difficult to calculate due to the extensive debromination encountered (see later discus-sion), hence only a rough estimate was presented. This comprised the molar

34

sum of the debromination products in the muscle tissue based on estimated response factors interpolated from those of the hexaBDEs and BDE209, which were the only congeners available as pure standards at that time. Comparing the mean molar sum of the lower BDEs and BDE209 with the mean dietary dose of BDE209, the uptake was extremely low, less than 0.13 % after 120 days. This was in contrast to the efficient uptake observed in pike subject to dietary exposure to tetra- to hexaBDE (177). The efficiency in that study was partly attributed to a lipid-mediated facilitation achieved by using living rainbow trout as the matrix for the congener cocktail. Recently, an uptake study of BDE209 in rainbow trout similar to that in paper II was performed (178). The dietary absorption was now determined to 3.5 %. The dose vehicle used was similar to the one in paper II, except that cod oil was used instead of corn oil as the solvent for BDE209 and food pellets were used instead of cod homogenate. The daily dose, however, was 3 orders of magnitude lower (~10 µg/kg body weight). In addition to the differences in the lipid content of the food and the lower daily dose, the lower uptake in paper II may have been due to the fact that the technical product was not fully dissolved in the corn oil before blending it with the cod homoge-nate/gelatine mixture. Hence the BDE209 may have been present in the fish gut as aggregates, which could hinder absorption. The use of a suitable vehi-cle was suggested to at least partly explain the high dietary absorption of BDE209 observed in rats (> 26 %) (166,179) compared to previous studies in the same species (< 0.5 %) (180,181).

Decreasing absorption efficiency with increasing hydrophobicity, (log Kow > 6.5), molecular size or molecular weight has previously been observed for halogenated organic compounds in fish (182,183) and in cows (184,185). These characteristics all coincide with increasing degree of halogenation of the substance. There are currently two major theories on the mechanism responsible for the reduced dietary absorption of large superhydrophobic substances. One is based on the size of the molecule physically restricting its passage through the membrane. A molecular effective cross section (ECS) of 9.5 Å has been proposed as the limit for uptake via the gills and similar lim-its were also suggested for the gut (186,187). According to this theory, not even the pentaBDEs, which have an ECS of 9.6 Å (177), should be taken up. The other theory is based on the combined effect of the compound’s lipid and water solubilities, usually expressed as the log Kow. The first restriction encountered during dietary absorption is here defined as the ability to dis-solve in the lipid/bile salt micelles formed in the gut that act as the transport medium to the intestinal epithelium. The second restriction is connected to the passage through a stagnant boundary layer of water through which the chemicals must diffuse following dissociation of the micelles (188). For superhydrophobic substances such as BDE209 the rate limiting restriction is believed to be the passage through the water layer.

35

The dietary absorption efficiencies of PBDEs have been investigated in pike, zebra fish, lake trout, common carp and in farmed Atlantic salmon (177,189-193). Although there were differences between the species, all except one study (190) showed equal or higher absorption of the lower BDEs in comparison to PCBs (177,191-193). Surprisingly, none of the stud-ies found a clear correlation between the bioaccumulation from food of the BDE congeners and their log Kow (BDE209 was included in only one study (190). There were, however, large differences in the apparent absorption efficiencies between BDE congeners. Because of the inherent difficulties in collecting feces from fish, the uptake was based on the tissue concentrations of the studied substances and thus did not include any biotransformation. Since a range of BDE congeners were added to the food in all studies, the rapid biotransformation of some of the BDEs was suggested to account for a low apparent absorption efficiency of, for example BDE99 and BDE153 in carp (191). Similarly, the unexpectedly high apparent absorption efficiencies may have been confounded by the bioformation of lower brominated BDEs (189-191,193). The extremely high absorption efficiency, close to 100 %, for BDE47 and its very slow approach to steady state were attributed to biofor-mation via debromination of higher brominated congeners (191-193).

In cows (paper III), the dietary absorption could unfortunately not be quantified due to the variability of the PBDE levels in the silage. However, decreasing absorption with increasing log Kow, bromination degree or mo-lecular size was indirectly indicated by a comparison of the ratios of the PBDE concentrations between adipose tissue and silage, focusing on the lowest ratio in each homologue group (Figure 7). The stepwise decreasing ratio from hepta to deca homologues may reflect the influence of bromina-tion degree, molecular size or log Kow on the dietary absorption.

Figure 7. Ratio of adipose tissue and silage concentrations for hepta- to decaBDE in the slaughtered cow (paper III). The lowest ratio within each homologue group is highlighted. * silage levels < LOQ (measured values, not LOQ, are plotted). The congeners are presented in the order of elution from the GC-column.

36

High absorption efficiencies have been reported for BDE209 in rats (>26 %) (166,179) and in grey seals (89 %) (194). This is in accordance to previous findings for other POPs (such as PCBs and PCDD/Fs) showing higher as-similation efficiencies for birds and mammals compared to fish and other poikilothermic organisms (188). Furthermore, in contrast to birds and hu-mans the absorption of substances in both fish and cows has been shown to be dramatically reduced for substances with log Kow > 6.5 (182,184).

Biotransformation The biotransformation of PBDEs is not fully understood and the number of investigations are few (for a review see Hakk & Letcher, 46). Similar to the metabolism of other aromatic xenobiotics, a suggested pathway for PBDEs is oxidative hydroxylation and/or oxidative debromination via an epoxide, mediated by the cytochrome P450 enzyme system. In general, laboratory in vivo-studies identifying polar metabolites of PBDEs have, with the exception of one fish study (164), been limited to rodents. These studies have all iden-tified the presence of hydroxylated metabolites of the BDE congeners stud-ied (164,165,171,195). In pike that were exposed to food spiked with 14C-BDE47, six hydroxy-PBDEs were detected and no lower brominated PBDEs were produced (164). Similar results have been reported in rodents exposed to tetra- to heptaBDEs (e.g. 71,196).

The metabolism of BDE209, however, seems to be more complex. Due to the lack of unsubstituted carbon atoms an oxidation via an arene oxide is most likely preceded by a debromination (46). However, the outcome from the rat studies varied from no hydroxylated or debrominated metabolites detected in the blood (171), to several debrominated-hydroxylated and hy-droxylated-methoxylated metabolites identified (166,179), and recently to lower brominated BDEs indicating debromination of BDE209 (197).

Debromination as metabolic pathway has previously been shown in fish and rats exposed to other brominated substances. Already in 1977, lower brominated biphenyls were detected in Atlantic salmon exposed to a techni-cal octabromobiphenyl mixture administered via the water and via the food (198). Later, reductive debromination of hexabromobenzene was demon-strated in rats (199). It was concluded that intestinal microorganisms were not involved. Similarly, in rainbow trout that were exposed via the diet to the commercial Deca-mix product, a range of hexa- to nonabrominated BDEs were detected (paper II). The levels of all of these congeners increased with the length of exposure. However, assessment of metabolic debromination was confounded by the presence of lower brominated BDEs in the amended food. The technical Deca-mix formulation used in the experiment, DOW FR-300BA, contained more nonaBDEs than later formulations, possibly reflecting changes in manufacturing practices (43). Furthermore, despite precautions, additional degradation of BDE209 to for instance BDE206 may

37

have occurred during the initial separation of the PBDEs from planar impuri-ties and in the preparation/storage of the amended cod chips (see Figure 2 in paper II). Nevertheless, whereas the congener profile in the feed resembled that observed after photolytic degradation of BDE209 (25,26), the profile in fish was different (Figure 8). In fish the first eluting congeners in each homologue group from the hepta- to nonaBDEs were prominent, while the late eluting congeners were more abundant after photodegradation. Despite the apparent accumulation, it was not possible to exclude uptake of minor ingredients in the diet. By estimating the total amount of the different conge-ners in the fish compared to the potential dose via the feed, not even non-detected impurities could be ruled out. Another way to exclude selective uptake was to assess the formation of lower brominated congeners during the depuration period. Although both BDE154 and the first eluting heptaBDE showed an increase during this period in contrast to their potential precur-sors, it was not statistically significant. However, estimated bioaccumulation factors (muscle:feed) after 120 days of exposure differed up to a factor of 400 between isomers within the homologue groups. Since the mechanism of dietary absorption for hydrophobic compounds in general is believed to be a non-selective diffusive process, bioformation was a more likely explanation for the disparate bioaccumulation behavior of the isomers.

Since then, reductive debromination of PBDEs has been confirmed in fish (178,190,192,193,200,201), in starlings (202), and by anaerobic microorgan-isms by single bacteria cultures as well as by inoculum from a sewage sludge digester (203,204). Laboratory experiments with common carp, Atlantic salmon and rainbow trout (178,190,192,193,200) have shown debromination to occur with both lower and higher brominated BDEs as precursors. Fur-thermore, mosquitofish and pumpkinseed captured in the receiving waters of a wastewater treatment plant contained a number of unidentified hexa- to octaBDEs not present in the sediment or effluent waters, which indicated that debromination had occurred also outside the laboratory (201).

Carp was shown to debrominate (possibly BDE153 to) BDE99 to BDE47 and BDE183 to BDE154 (192) as well as BDE209 to octa- down to pentaB-DEs (200). In contrast to rainbow trout (paper II), the parent BDE congener was never detected in the tissues. Carp are stomachless fish with the liver tissue situated along the intestine. This may result in a more efficient absorp-tion and a greater metabolic capacity. A comparative in vitro study of the degradation of BDE209 by liver microsomes from rainbow trout and carp revealed a more efficient debromination (65 % of the total mass of BDE209) by carp than rainbow trout (22 %) over 24 hours (178). The differences in the rate of metabolism were also confirmed in vivo by exposing rainbow trout via the diet. An almost identical congener profile was obtained in this study compared to the previous study (paper II).

The enzymes involved in the debromination have not yet been identified. It has been suggested that the deiodinase enzymes responsible for the re-

38

moval of an iodine atom from thyroxine (T4) to triiodothyronine (T3) are involved (178). These enzymes, which are essential for the regulation of thyroid hormones, are selective for meta-substituted iodine, which is one position reported to be debrominated in fish (192). Another possibility is reductive dehalogenation via the cytochrome P450 enzyme system. Then again, no increase in plausible debromination products was observed in At-lantic tomcod intraperitoneally injected with BDE209 and the cytochrome P450 1A (CYP1A) inducer CB126 (205). Whether other cytochrome P450 enzymes are involved is not known, however, both cytochrome P450 and deiodinase enzyme systems are present in the microsomal fraction used in the in vitro studies described above.

Figure 8. Mass chromatogram (bromide ions in ECNI) for A) rainbow trout muscle after 120 days of dietary exposure to Deca-mix, B) Deca-mix in toluene photode-graded for 120 min (data from U. Sellström) and C) Penta-, Octa-, and Deca-mix technical PBDE products. Coelutions (e.g. BDE197/204 and BDE203/198) cannot be excluded.

39

Figure 9. Combined mass chromatograms ([M]+ for hexaBDEs and [M-2Br]+ for hepta- to decaBDE in HRMS, EI) for A) cow’s milk, B) adipose tissue, C) cow’s feces and D) grass silage. Coelutions cannot be excluded.

Typical signs of debromination were also encountered in the cow (paper III). The congener profile in the cow’s adipose tissues, organs and in the milk differed from that in the major feed, the silage, and the feces (Figure 9). Three major congeners, BDE197, BDE196 and BDE207, seemed to ac-cumulate in the cow. In accordance with the uptake study in fish (paper II),

40

the differences between isomers within the homologue groups pointed to-wards debromination after absorption as the most likely explanation.

The comparison of congener profiles between feces and feed on the one hand and organs/tissues and milk on the other is difficult since other proc-esses apart from debromination influence the outcome. The profile in the tissues also reflects the differences in the absorption efficiency and clearance rate, and, for milk, the transfer efficiency from the mammary gland to the milk. These differences are substantial between homologue groups. By nor-malizing the concentrations to the congener with the lowest level within each homologue group, the similarities and dissimilarities between the silage, the feces, the adipose tissue and the milk are revealed (Figure 10). It seems clear that compared to isomers within the homologue groups there is a pref-erential accumulation after absorption of BDE207 among the nonaBDEs, of BDE197 and BDE196 among the octaBDEs, and possibly by BDE182 among the heptaBDEs. BDE182 was below the quantification limit in the silage and feces but clearly detectable in the adipose tissue and thus only indicative.

Figure 10. Concentrations normalized to the congener with the lowest concentration within each homologue group. The heptaBDE was normalized to BDE183, the oc-taBDEs to BDE203, and the nonaBDEs to BDE208. * < LOQ (measured values, not LOQ, are plotted). The congeners are presented in the order of elution from the GC-column.

Another species from the terrestrial food chain that has been shown to be capable of reductive debromination is the European starling. 76 days after the insertion of BDE209 amended implants under the skin, a range of lower

41

brominated BDEs were detected in the muscle and in the liver (202). The BDE congeners that accumulated most were BDE207, BDE208, BDE197, and BDE206.

Humans exposed to the Deca-mix also had elevated concentrations of octa- and nonaBDEs in the blood (97,206). From the chromatograms the most abundant congeners seem to be BDE207 and an octaBDE (by compar-ing elution order possibly BDE197). Recently, the debromination of BDE209 in rats identified BDE207, BDE197 and BDE201 as major products after 21 days of dietary exposure (197). Thus, at least two of the major ac-cumulating congeners, BDE207 and BDE197, are common to cows, star-lings rats and possibly also to humans.

The major BDE congeners accumulated in the cow were not the same as those with the highest rate of formation in the fish study. The debromination products from BDE209 in rainbow trout were recently identified in a feeding study that was similar to that in paper II (178). The major heptaBDE formed was identified as BDE188, and the two major octaBDEs as BDE202 and BDE201. The debromination pathways are difficult to assess since the accumulation of a congener may represent both a bioformation from higher brominated congeners and/or a resistance to further degradation. Depending on the rate of formation versus the rate of degradation, some congeners ac-cumulate to a larger extent than others. Nevertheless, it is clear that fish are capable of removing meta-substituted bromine atoms, for example in trans-forming BDE99 to BDE47 as reported in carp (192). In addition, fish can also remove bromine atoms in the para-position, as the accumulation of BDE202 implicates. BDE congeners with bromine substituents in the ortho-position appear to be more recalcitrant in fish. The debromination pathway in cows (paper III) seems to be more selective. The major products are re-stricted to the removal of bromines in the meta-position such as BDE207 debrominating to BDE197 and BDE196 debrominating to BDE182, and possibly to bromines being removed from the ortho-position. Thus, BDE196 may originate from an ortho-debromination of BDE207 or a meta-debromination of BDE206, both of which were present in the silage. In con-trast to the debromination pathway demonstrated in fish, no apparent re-moval of bromines in the para-position was observed in cows. Tentative schemes of the debromination pathways of BDE209 encountered in fish (paper II, (178,190,192)) and in cows (paper III) are presented in Figure 11. It is tempting to attribute differences in bioaccumulation profiles to dif-ferences in the capacity of biotransformation between species. Could a less efficient metabolism in terrestrial biota explain the larger proportion of higher brominated BDEs in some of these organisms?

42

Figure 11. Tentative pathways for the metabolic debromination of BDE209. The congeners shown represent a summary of identified lower brominated BDEs de-tected in fish (paper II, (178,190,192), and in cows from paper III. The suggested pathways for the degradation in papers II and III are marked as shadows in the background (no quantitative measures). Solid arrows represent debromination at the meta- and para-positions. The dotted arrows represent a tentative debromination at the ortho-position.

43

Aquatic versus terrestrial environment Similar to most classical organochlorines, the research on PBDEs was for a long time focused on the aquatic environment. Generally, in fish, fish-eating birds and marine mammals, the lower brominated BDEs dominate the BDE pattern typically with BDE47 as the most abundant congener (38,40,77,148,207,208). The few early investigations including samples from terrestrial biota reported low levels of (lower brominated) PBDEs, for instance in rabbits, moose, reindeer and starlings (2). However, the discov-ery of high levels of higher brominated BDEs in the eggs of peregrine fal-cons (7) raised the question of whether organisms in the terrestrial food chain may be more exposed to higher brominated BDE congeners than aquatic organisms. This was further supported by the bioaccumulation of octa- to decaBDE reported in earthworms collected from sewage sludge-amended soil (27). Earthworms represent the base of many terrestrial food chains and thus may constitute a pathway for the bioaccumulation of higher brominated BDEs at higher trophic levels. BDE209 was also the dominant BDE congener in the soil samples from the reference sites, indicating that atmospheric deposition may be an important source of the PBDE contamina-tion in terrestrial ecosystems. A similar profile was also found in air/deposition samples from the Baltic Proper, an area remote from any di-rect emissions (209).

A congener profile indicating a preferential accumulation of higher bro-minated BDEs was also observed in other terrestrial birds of prey such as merlins and golden eagles (148). Furthermore, a survey of birds from the UK showed a clear, positive correlation between the presence of BDE209 and birds feeding from the terrestrial food web (210). Recently, the bioaccumula-tion of higher brominated BDEs was reported in eight terrestrial bird species from the Beijing area in Northern China (211). BDE209 concentrations of up to 12 ppm on a lipid weight basis were found in the liver of the common kestrel, which is the highest BDE209 level reported in wildlife. The kestrels were year-round residents of the urban areas of Beijing and thereby exposed to PBDE sources such as domestic waste. Foxes and grizzly bears are other terrestrial species reported to have a congener profile dominated by BDE209 (212,213). The fox is an omnivore that, when possible, supplements its diet with domestic waste. The difference between terrestrial and aquatic food chains was further reinforced by the PBDE profile in grizzly bears from Brit-ish Columbia, although all specimens contained BDE209, only in those feed-ing entirely on terrestrial food was BDE209 the dominant congener, whereas BDE47 was the major congener in those that ate salmon part of the year (213).

In paper III, BDE209 was the dominant congener in the cow tissues ex-amined, which is consistent with the observations for the other terrestrial feeders listed above. However, the concentrations were low, on average 3.7

44

ng/g lipid weight in adipose tissues, which is at the lower end of the range of levels detected in foxes and grizzly bears. This may reflect the extent and composition of the dietary exposure, as in the example of the grizzly bears, but may also be attributable to physiological differences between species.

There are other variables besides the contamination of the diet that vary between species and that could also play a role in the differences in bioac-cumulation of BDE209 between aquatic and terrestrial organisms as well as among terrestrial organisms. One is the dietary absorption efficiency. Inves-tigations with other chemicals have shown that the dietary absorption of the higher chlorinated PCDD/Fs (log Kow comparable to that of BDE209) is much lower in cows than in terrestrial predators (e.g. humans or birds) (188). This could explain the low levels of BDE209 in cows compared to other terrestrial organisms. The other explanation is related to differences in the biotransformation efficiency. Thus, despite a low absorption of BDE209, the dominance of BDE209 in cows may be explained by a low rate of metabo-lism and excretion. Biotransformation may also be responsible for the differ-ent BDE congener pattern observed in German peregrine falcons compared to sparrowhawks, despite their similar feeding habits (38). In humans, the apparent half life of BDE209 in serum is short (15 days) (214) even though the dietary absorption is high for highly chlorinated POPs, which indicates a rapid metabolism of BDE209. The comparably low levels of BDE209 in aquatic food webs may be attributable to a lower uptake (paper II, 178,188) and/or a more efficient metabolism of this compound by aquatic organisms. Which of these two factors, dietary absorption or metabolism, is more im-portant for the bioaccumulation of BDE209 is, however, far from clear, and may vary from species to species.

Risk assessment / Implications for exposure charac-terisation An important pathway for human exposure to PCBs and PCDD/Fs is via atmospheric deposition to soil and vegetation, transfer to grazing cows, and sequestration in dairy products and meat (215). Since BDE209 is present not just in occupationally exposed workers but also in the general population (206,216), there must be important sources of exposure outside of the work-place. The short half-life of BDE209 in human serum implies that exposure must be semi-continuous in order to explain the blood concentrations ob-served (214). One proposed source of human exposure to higher brominated BDEs is via inhalation/ingestion of household dust. Dust has been reported to contain considerable amounts of BDE209 (134,217,218). Another path-way may be via food.

45

Meat and dairy products from cattle account for a large part of the human exposure to classical chlorinated POPs. Insight into these potential vectors of human exposure to BDE209 was gained in paper III. A clear discrimination against the transfer of higher brominated BDEs to the milk was observed. Whereas the adipose tissues from different lipid compartments in the cow had similar concentrations on a lipid weight base, indicating that the tissues were close to equilibrium, the concentrations in milk fat were 1-2 orders of magnitude lower. This was in contrast to the PCBs and the lower brominated BDEs in the same cows, for which the concentrations in milk fat and adipose tissues were similar (219,220). A comparable behaviour has been reported for the lactational transfer of higher chlorinated PCBs and for tetra- to hex-aBDEs in marine mammals with a high throughput of milk fat (88,221,222). The low transfer encountered for higher brominated BDEs in cows was therefore suggested to be caused by a slow rate of transfer of superhydro-phobic compounds into the milk in combination with the high daily yield of milk fat (about 1 kg/day). Recently, the discrimination in the transfer of higher brominated BDEs was confirmed in human milk versus serum from Japanese mothers (223). BDE209 was the dominant congener in the serum but not in milk, which had 10 times lower concentrations. Compared to hu-mans the transfer of BDE209 in cows was even smaller, a factor of 100 compared to adipose tissue, possibly related to the larger elimination rate of fat per day. The consequence of this discrimination in dairy cows is that, instead of being excreted via the milk like most lipophilic contaminants in lactating cows, the highly brominated BDEs are accumulated in the fat and meat. From the perspective of human exposure, this can be expected to in-fluence the dominant pathways of dietary intake.

Several food basket analyses in Europe have identified fish as the major source of lower brominated BDEs (224-226). However, North American studies have recognized meat as the primary contributor, probably due to the fact that less fish is eaten there (227,228). Only a few studies have included higher brominated BDEs (227,229,230), and in these the BDE209 levels varied widely between food types and items. For example, BDE209 was the dominant congener in cheese but not in butter or milk in one investigation (230), while it was dominant in butter but not in cheese and milk in another study (229). More work is needed to clarify which foods contribute most to dietary exposure to BDE209. The results of paper III suggest that particular attention should be payed to beef.

Debromination of higher brominated BDEs can also have consequences for PBDE risk assessment, since the toxicity of PBDEs seems to be inversely related to the degree of bromination (44). Due to the analytical difficulties associated with the analysis of BDE209, there are fewer investigations in-cluding this congener compared to lower brominated BDEs. However, the number of papers reporting the presence of BDE209 in biota is increasing. In order to assess debromination in environmental samples, there is a need to

46

identify and quantify the debromination products. The limited number of commercially available reference standards for hepta- to octaBDEs is a con-straining factor. Furthermore, the difficulties connected to the quantification of octa- and nonaBDEs due to the degradation of BDE209 during the proce-dures of sample preparation and instrumental analysis pose challenges that were addressed in paper III.

A different area that indirectly is influenced by debromination is the de-termination of bioconcentration factors (BCF), bioaccumulation factors (BAF), or half-lives of BDE congeners from environmental samples or in laboratory experiments where BDE congener mixtures are used. As previ-ously discussed for the absorption efficiencies in fish, the successive debro-mination from higher to lower brominated congeners will give rise to overes-timated absorption efficiencies, BCFs, BAFs or half-lives for the debromina-tion products and underestimated factors for the debromination precursors. Such processes may, for example, explain the unexpectedly high biota-soil BAFs that were reported for BDE197 and BDE207 in earthworms in sewage sludge amended soil (27). Likewise, the differences in apparent half-life between isomers of higher brominated BDEs in human blood may be related to bioformation/debromination (214).

PBDE bioaccumulation has been shown to be highly species and conge-ner specific. Some of these differences may be at least partly caused by dif-ferences in metabolic capacity (146,231,232). If the major metabolic path-way of the species is via debromination, it will influence the BDE congener accumulation in its predator. Biomagnification factors (BMFs) are therefore species dependent and highly specific for the food web they describe. For example, if the debromination capacity increases at the top of the food chain, the BMFs of the debromination products will be elevated and dependent on the levels of precursors present in the prey. As with the BAFs, the conse-quences are an overestimation of the biomagnification potential for the de-bromination products and an underestimation for the debromination precur-sors. Such mechanisms may for instance explain the unexpectedly high bio-magnification potential reported for a heptaBDE in marine food webs from the Baltic Sea and Atlantic Ocean (207).

As of August 2004, the European Union banned the marketing and use of the Penta-mix and Octa-mix formulations. The United States has currently no national legislation concerning PBDEs, but the sole manufacturer of the Penta-mix and Octa-mix voluntarily stopped production of these PBDE products at the end of 2004 (233). Canada has recently (December 2006) proposed strategies to prohibit the use of the Penta-mix and the Octa-mix and to regulate the use of the Deca-mix. In Europe, the Deca-mix is exempt from the prohibition of PBDEs. However, Sweden has from the first of January 2007 introduced a national ban on the use of Deca-mix in other ap-plications than electronic products and vehicles (19). Thus, although the major applications are in electronic goods, it is not allowed in for example

47

textiles. Still, from a global perspective the Deca-mix product is the most frequently used and the least regulated PBDE product. Furthermore, the global presence of BDE209 in the environment may indirectly, via biotic or abiotic debromination, be a long term source of future exposure of biota to lower brominated BDEs.

Decabromodiphenyl ethane – a next generation BFR DeBDethane, a BFR used to replace the Deca-mix formulation, was identi-fied for the first time in the environment (paper IV). DeBDethane was pre-sent in sediment from an area heavily polluted by the Deca-mix, in sewage sludge from 25 (of 50) Swedish STPs and in an air sample from an elec-tronic dismantling facility. In all samples more BDE209 than DeBDethane was present, but the ratio differed (see Figure 12). Its presence in sewage sludge did not show any correlation to STP size or latitude but seemed to be more frequent in highly populated areas. The occurrence was similar to that of BDE209, indicating a diffuse input from households and urban runoff (8,76). DeBDethane has since then been reported in sewage sludge from Spain (234) and Canada (235). The concentrations were in the lower range (up to 15 and 32 ng/g dry weight) of those reported in paper IV.

Figure 12. Concentrations (ng/g dry weight) of DeBDethane and BDE209 in sew-age sludge from 50 Swedish STPs. The DeBDethane concentrations are estimates and the BDE209 concentrations are from Nylund et al. (8,76). Zero-levels represent “non-detects”.

48

The presence of DeBDethane was also confirmed in both air and dust sam-ples (tentatively identified) from a Swedish electronic recycling facility (236,237). Further evidence of its applications in domestic electronic prod-ucts is that detectable DeBDethane levels are reached in normal household dust. For example, DeBDethane was recently reported in 4 of the 5 house-hold dust samples, and in fact exceeded the BDE209 levels in one sample (238). The concentrations did not correlate to the BDE209 concentrations, and DeBDethane was not detected in human plasma from any of the inhabi-tants. Additionally, DeBDethane was detected along with BDEs in tree bark collected in Arkansas, U.S., and their occurrence was traced to emissions from two major BFR manufacturing plants in southern Arkansas (239).

Information on the bioavailability/absorption of DeBDethane is so far scarce. However, recently DeBDethane was included in a study investigating the bioaccumulation of a number of BFRs in a food web from Lake Winni-peg, Canada (240). Although DeBDethane was not detected in zooplankton, mussels or whitefish, it was present in some of the other fish species, such as walleye and burbot. A significant relationship between trophic level and lipid-based concentrations led the authors to suggest that DeBDethane bio-magnified in the food web investigated.

The behaviour of DeBDethane is in many ways similar to that of BDE209, both being large, superhydrophobic and thermolabile compounds. Both compounds also degrade to some extent to lower brominated congeners during sample preparation/analysis. Two nonaBDethane congeners were also present in the technical product, Saytex 8010® (paper IV). BDE209 is known to be photolytically labile undergoing degradation to lower bromi-nated congeners (e.g. 25). Whether DeBDethane is susceptible to degrada-tion via UV light is not known. However, preliminary results from exposing a DeBDethane solution to a daylight fluorescent lamp (Metal Halide Arc) in the laboratory showed that DeBDethane is degraded, producing two nonaB-Dethane congeners as well as a number of peaks tentatively identified as octa-brominated products (Figure 13).

A study of the thermal decomposition (pyrolysis) of flame retarded high-impact polystyrene (HIPS) showed that DeBDethane was degraded mainly to brominated toluenes, while BDE209 produced lower brominated BDEs and brominated furans (241). The difference was suggested to be caused by the weaker C-C-bond in DeBDethane compared to the ether bond in the BDE209 structure. Whether a degradation to pentabromotoluene (PBT) is feasible in the environment is not known. Interestingly, pentabromotoluene has recently been detected in air, river sediment and in blubber from beluga whales (242-244). Although pentabromotoluene is/has been used as a BFR itself (9) and may also be released from the production of BFR oligomers (245), further investigation of its possible formation from DeBDethane is warranted. Future work could investigate the covariation of the concentra-

49

tions as an indication of a common source of DeBDethane and pentabromo-toluene.

Figure 13. Mass chromatogram (bromide ions in ECNI) of technical DeBDethane in n-hexane, A) photodegraded and B) original solution (unpublished data). The raised baseline before the DeBDethane peak is due to thermal degradation in the column.

The present study illustrates the importance of carefully assessing the envi-ronmental behaviour of new chemicals, particularly when they are replace-ments of known problematic substances. It is likely that the use of DeBDe-thane will increase in the future, particularly after the recent regulatory ini-tiatives to reduce the use of the Deca-mix in, for instance, Sweden and Can-ada. On the basis of predicted future widespread use in the UK, an environmental risk evaluation of DeBDethane is currently being performed by the British EPA (draft for peer review 2006).

50

Conclusions

The work presented in this thesis has highlighted some aspects of the behav-iour of BFRs in the environment. The results contribute to the exposure characterization for the environmental risk assessment of the PBDEs in par-ticular, but also a representative of the next generation BFRs, decabromodi-phenyl ethane. From the results the following conclusions were drawn: • The concentrations of lower brominated BDEs in pike from a Swedish

lake reflect the past usage of PBDEs in Sweden/Europe. The levels in-creased up to the mid 1980s and then slowly decreased from the late 1980s up to now.

• The structurally related methoxy-PBDEs, which were present in

amounts similar to lower brominated BDEs in fresh water pike, do not originate from the anthropogenic PBDEs.

• Despite its molecular size and hydrophobicity, the fully brominated

PBDE, BDE209, can to a small extent be absorbed in fish and cows, when ingested via the diet.

• In vivo reductive debromination of BDE209 to lower brominated BDEs

occurs in fish as well as in cows exposed via the diet. • The transfer of PBDEs to the milk in cows decrease dramatically with

increasing molecular size, bromination degree or log Kow of the BDE congener.

• A representative for the next generation of BFRs, decabromodiphenyl

ethane that is marketed as a replacement for the DecaBDE product is already present in the environment.

For PBDEs, the stability and toxicity decrease with increasing degree of bromination. This behaviour is in many ways different from that of the clas-sical chlorinated pollutants. As shown in this thesis, the higher brominated BDEs are subject to degradation, producing lower brominated BDEs both in organisms as well as in the laboratory during analysis. Others have shown their ability to debrominate in the presence of UV-light. While the usage of

51

lower brominated BDEs is subject to regulations, the highly brominated BDEs continue to be used in large volumes. A key question is therefore whether the continued use of the higher brominated BDEs will result in in-creased contamination of biota with the more environmentally detrimental lower brominated BDEs. The lack of persistence of the higher brominated BDEs, which is generally a useful property of an environmental contami-nant, may in this case be detrimental, as BDE209 becomes a source of more persistent lower brominated BDEs in the environment. The differences in the isomer patterns of BDE homologues between the technical products and what is observed in biota suggests that this is happening.

52

Acknowledgements

My supervisors Lillemor Asplund, Cynthia de Wit and Michael McLachlan, are gratefully acknowledged for their support and guidance at different stages of this process.

Others of importance for the outcome are Bo Jansson, a pioneer in the

analysis of brominated substances, Ulla Sellström, my closest colleague with hands-on experience of PBDE-analysis, my coauthors for their exclusive contributions, especially Jonas Björklund, Gareth Thomas, Anders Bignert, Mats Olsson, Lennart Balk and Ulrika Fridén.

Yngve Zebühr and Michael Strandell are acknowledged for their skilful

technical assistance with the HRMS analysis. Sören Jensen for advice in the analytical method development, Eva

Jakobsson and Göran Marsh for gifts in the form of synthesized standards.

Finally, I am grateful for the help from my colleagues at ITM, who made

small and large contributions along the way. Thank you all!

53

References

1. Stockholm convention on persistent organic pollutants (POPs), http://www.pops.int/, Accessed Mars, 2007.

2. Sellström, U., Jansson, B., Kierkegaard, A., de Wit, C., Odsjö, T., Olsson, M., 1993. Polybrominated diphenyl ethers (PBDE) in biological samples from the Swedish environment. Chemosphere, 26, 1703-1718.

3. Sellström, U., Bignert, A., Kierkegaard, A., Häggberg, L., de Wit, C., Olsson, M., Jansson, B., 2004. Temporal Trend Studies on Polybrominated Diphenyl Ethers and Hexabromocyclododecane in Guillemot Egg from the Baltic Sea. Environ. Sci. Technol., 37, 5496-5501.

4. Haglund, P., Zook, D. R., Buser, H.-R., Hu, J., 1997. Identification and quanti-fication of polybrominated diphenylethers and methoxy-polybrominated di-phenylethers in Baltic biota. Environ. Sci. Technol., 31, 3281-3287.

5. Sellström, U., Kierkegaard, A., de Wit, C., Jansson, B., 1998. Polybrominated diphenyl ethers (PBDE) and hexabromocyclododecane (HBCD) in sediment and fish from a Swedish river. Environ. Toxicol. Chem., 17, 1065-1072.

6. Andersson, Ö., Blomkvist, G., 1981. Polybrominated aromatic pollutants found in fish in Sweden. Chemosphere, 10, 1051-1060.

7. Lindberg, P., Sellström, U., Häggberg, L., de Wit, C. A., 2004. Higher bromi-nated diphenyl ethers and hexabromocyclododecane found in eggs of peregrine falcons (Falco peregrinus) breeding in Sweden. Environ. Sci. Technol., 38, 93-96.

8. Nylund, K., Eriksson, U., Haglund, M., Berggren, D., Kierkegaard, A., Allan, A., Asplund, L., de Wit, C., 2002. Analyser av bromerade flamskyddsmedel i avloppsslam från 50 reningsverk i Sverige, Swedish EPA, ISBN 91-620-5188-1-pdf, Stockholm, Sweden.

9. WHO/IPCS, Environmental Health Criteria 192, 1997. Flame Retardants: A General Introduction, ISBN 92 4 157192 6.

10. Alaee, M., Arias, P., Sjödin, A., Bergman, A., 2003. An overview of commer-cially used brominated flame retardants, their applications, their use patterns in different countries/regions and possible modes of release. Environ. Int., 29, 683-689.

11. Bromine Science and Environmental Forum, http://www.bsef.com, Accessed the 23 Mars 2007

12. Hutzinger, O., Sundström, G., Safe, S., 1976. Environmental chemistry of flame retardants part I. Introduction and principles. Chemosphere, 1, 3-10.

13. Ballschmiter, K., Mennel, A., Buyten, J., 1993. Long chain alkyl-polysiloxanes as non-polar stationary phases in capillary gas chromatography. Fresenius J Anal. Chem., 346, 396-402.

14. WHO/IPCS, Environmental Health Criteria 162, 1994. Brominated Diphenyl Ethers, World Health Organization, International Programme on Chemical Safety, Geneva, Switzerland.

15. Marsh, G., 2003. Polybrominated diphenyl ethers and their hydroxy and meth-oxy derivatives. Congener specific synthesis and analysis, Ph.D. the-

54

sis/dissertation, Department of Environmental Chemistry, Stockholm Univer-sity, ISBN 91-7265-616-6.

16. Bromine Science and Environmental Forum, www.bsef.com/bromine/our_industry/; Accessed Sept. 2006, At this date (23 Mars 2007) no longer available.

17. Watanabe, I., Sakai, S., 2003. Environmental release and behavior of bromi-nated flame retardants. Environ. Int., 29, 665-682.

18. Adopted AMENDMENTS to Directive 76/769/EEC. European Commission, http://ec.europa.eu/enterprise/chemicals/legislation/markrestr/amendments_en.htm, Accessed Mars 2007

19. Swedish Chemicals Agency, KEMI, http://www.kemi.se/templates/News.aspx?id=4636, Accessed 22 Mars 2007

20. National Chemicals Inspectorate, 2003. Rapport av ett regeringsuppdrag: Bromerade flamskyddsmedel – Förutsättningar för ett nationellt förbud Stock-holm, Report KEMI 4/03, 101 pp.

21. National Chemicals Inspectorate, 2007. The Products Register Solna, Sweden. 22. Wania, F., Dugani, C. B., 2003. Assessing the long-range transport potential of

polybrominated diphenyl ethers: A comparison of four multimedia models. En-viron. Toxicol. Chem., 22, 1252-1261.

23. Gouin, T., Thomas, G. O., Chaemfa, C., Harner, T., Mackay, D., Jones, K. C., 2006. Concentrations of decabromodiphenyl ether in air from Southern Ontario: Implications for particle-bound transport. Chemosphere, 64, 256-261.

24. Rahm, S., Green, N., Norrgran, J., Bergman, A., 2005. Hydrolysis of environ-mental contaminants as an experimental tool for indication of their persistency. Environ. Sci. Technol., 39, 3128-3133.

25. Söderström, G., Sellström, U., de Wit, C. A., Tysklind, M., 2004. Photolytic debromination of decabromo diphenyl ether (BDE 209). Environ. Sci. Technol., 38, 127-132.

26. Eriksson, J., Green, N., Marsh, G., Bergman, A., 2004. Photochemical decom-position of 15 polybrominated diphenyl ether congeners in methanol/water. En-viron. Sci. Technol., 38, 3119-3125.

27. Sellström, U., de Wit, C. A., Lundgren, N., Tysklind, M., 2005. Effect of sew-age-sludge application on concentrations of higher-brominated diphenyl ethers in soils and earthworms. Environ. Sci. Technol., 39 , 9064-9070.

28. Jansson, B., Asplund, L., Olsson, M., 1987. Brominated flame retardants - ubiquitous environmental pollutants? Chemosphere, 16, 2343-2349.

29. Stafford, C. J., 1983. Halogenated diphenyl ethers identified in avian tissues and eggs by GC/MS. Chemosphere, 12, 1487-1495.

30. Watanabe, I., Kashimoto, T., Tatsukawa, R., 1987. Polybrominated biphenyl ethers in marine fish, shellfish and river and marine sediments in Japan. Chemosphere, 16, 2389-2396.

31. de Boer, J., 1989. Organochlorine compounds and bromodiphenylethers in livers of Atlantic cod (gadus morhua) from the North Sea, 1977-1987. Chemos-phere, 18, 2131-2140.

32. Andersson, Ö. , Wartainen, A., 1992. Levels of polychlorinated camphenes (toxaphene), chlordane compounds and polybrominated diphenyl ethers in seals from Swedish waters. Ambio, 21, 550-552.

33. Loganathan, B. G., Kannan, K., Watanabe, I., Kawano, M., Irvine, K., Kumar, S., Sikka, H. C., 1995. Isomer-specific determination and toxic evaluation of polychlorinated biphenyls, polychlorinated/brominated dibenzo-p-dioxins and dibenzofurans, polybrominated biphenyl ethers, and extractable organic halogen

55

in carp from the Buffalo River, New York. Environ. Sci. Technol., 29, 1832-1838.

34. Allchin, C. R., Law, R. J., Morris, S., 1999. Polybrominated diphenyl ethers in sediments and biota downstream of potential sources in the UK. Environ. Pol-lut., 105, 197-207.

35. Meironyté, D., Norén, K., Bergman, Å., 1999. Analysis of polybrominated diphenyl ethers in Swedish human milk. A time-related trend study, 1972-1997. J. Toxicol. Environ. Health, Part A, 58, 101-113.

36. de Boer, J., Wester, P. G., Lewis, E. W., Boon, J. P., 1998. Do flame retardants threaten ocean life? Nature, 394, 28-29.

37. de Wit, C., 2002. An overview of brominated flame retardants in the environ-ment. Chemosphere, 46, 583-624.

38. Law, R. J., Alaee, M., Allchin, C. R., Boon, J. P., Lebeuf, M., Lepom, P., Stern, G. A., 2003. Levels and trends of polybrominated diphenylethers and other brominated flame retardants in wildlife. Environ. Int., 29, 757-770.

39. Hale, R. C., La Guardia, M. J., Harvey, E., Gaylor, M. O., Mainor, T. M., 2006. Brominated flame retardant concentrations and trends in abiotic media. Chemosphere, 64, 181-186.

40. Law, R. J., Allchin, C. R., de Boer, J., Covaci, A., Herzke, D., Lepom, P., Morris, S., Tronczynski, J., de Wit, C. A., 2006. Levels and trends of bromi-nated flame retardants in the European environment. Chemosphere, 64, 187-208.

41. de Wit, C. A., Alaee, M., Muir, D. C. G., 2006. Levels and trends of bromi-nated flame retardants in the Arctic. Chemosphere, 64, 209-233.

42. Sjödin, A., Jakobsson, E., Kierkegaard, A., Marsh, G., Sellström, U., 1998. Gas chromatographic identification and quantitation of polybrominated di-phenyl ethers in a commercial product, Bromkal 70-5DE. J. Chromatogr. A, 822, 83-89.

43. La Guardia, M. J., Hale, R. C., Harvey, E., 2006. Detailed polybrominated diphenyl ether (PBDE) congener composition of the widely used penta-, octa-, and deca-PBDE technical flame-retardant mixtures. Environ. Sci. Technol., 40, 6247-6254.

44. Darnerud, P. O., 2003. Toxic effects of brominated flame retardants in man and in wildlife. Environ. Int., 29, 841-853.

45. Legler, J. , Brouwer, A., 2003. Are brominated flame retardants endocrine disruptors? Environ. Int., 29, 879-885.

46. Hakk, H., Letcher, R. J., 2003. Metabolism in the toxicokinetics and fate of brominated flame retardants - a review. Environ. Int., 29, 801-828.

47. Birnbaum, L. S., Staskal, D. F., 2004. Brominated flame retardants: Cause for concern? Environ. Health Perspect., 112, 9-17.

48. Viberg, H., Fredriksson, A., Eriksson, P., 2003. Neonatal exposure to poly-brominated diphenyl ether (PBDE 153) disrupts spontaneous behaviour, impairs learning and memory, and decreases hippocampal cholinergic receptors in adult mice. Toxicol. Appl. Pharmacol., 192, 95-106.

49. Eriksson, P., Viberg, H., Jakobsson, E., Örn, U., Fredriksson, A., 2002. A brominated flame retardant, 2,2 ',4,4 ',5-pentabromodiphenyl ether: Uptake, re-tention, and induction of neurobehavioral alterations in mice during a critical phase of neonatal brain development. Toxicol. Sci., 67, 98-103.

50. Viberg, H., Fredriksson, A., Jakobsson, E., Örn, U., Eriksson, P., 2003. Neuro-behavioral derangements in adult mice receiving decabrominated diphenyl ether (PBDE 209) during a defined period of neonatal brain development. Toxicol. Sci., 76, 112-120.

56

51. Viberg, H., Johansson, N., Fredriksson, A., Eriksson, J., Marsh, G., Eriksson, P., 2006. Neonatal exposure to higher brominated diphenyl ethers, hepta-, octa-, or nonabromodiphenyl ether, impairs spontaneous behavior and learning and memory functions of adult mice. Toxicol. Sci., 92, 211-218.

52. Fernie, K. J., Mayne, G., Shutt, J. L., Pekarik, C., Grasman, K. A., Letcher, R. J., Drouillard, K., 2005. Evidence of immunomodulation in nestling American kestrels (Falco sparverius) exposed to environmentally relevant PBDEs. Environ. Pollut., 138, 485-493.

53. Fernie, K. J., Shutt, J. L., Mayne, G., Hoffman, D., Letcher, R. J., Drouillard, K. G., Ritchie, I. J., 2005. Exposure to polybrominated diphenyl ethers (PBDEs): Changes in thyroid, vitamin A, glutathione homeostasis, and oxida-tive stress in American kestrels (Falco sparverius). Toxicol. Sci., 88, 375-383.

54. Murvoll, K. M., Jenssen, B. M., Skaare, J. U., 2005. Effects of pentabromi-nated diphenyl ether (PBDE-99) on vitamin status in domestic duck (Anas platyrhynchos) hatchlings. J. Toxicol. Environ. Health, Part A, 68, 515-533.

55. Breitholtz, M. , Wollenberger, L., 2003. Effects of three PBDEs on develop-ment, reproduction and population growth rate of the harpacticoid copepod Ni-tocra spinipes. Aquat. Toxicol., 64, 85-96.

56. UBA, Umweltbundesamt, 2001. Erarbeitung von Bewertungsgrundlagen zur Substitution umweltrelevanter Flammschutzmittel. An English translation of Part 1 of this report; Substituting Environmentally Relevant Flame Retardants: Assessment Fundamentals (results and summary overview), is available at http://www.umweltdaten.de/publikationen/fpdf-l/1988.pdf, Accessed Mars 2007.

57. Arias, P.A., 2001. Brominated flame retardants - an overview. The Second International Workshop on Brominated Flame Retardants Stockholm, Sweden.

58. Albemarle Corp., Brominated FR for HIPS resin allowing efficient recyclabil-ity for electrical and electronic equipment. http://www.specialchem4polymers.com/tc/brominated-flame-retardants/index.aspx . Accessed Mars 2007.

59. Pettigrew, F. A., Landry, S. D., Reed, J. S., 1992. Recent advances in flame retardants for high-impact polystyrene. Flame Retard. '92, Proc. Flame Retard. '92 Conf., 5th.

60. Hardy, M. L., Margitich, D., Ackerman, L., Smith, R. L., 2002. The sub-chronic oral toxicity of ethane, 1,2-bis(pentabromophenyl) (Saytex 8010) in rats. Int. J. Toxicol., 21, 165-170.

61. Olsson, A., Ceder, K., Bergman, Å., Helander, B., 2000. Nestling Blood of the White-Tailed Sea Eagle (Haliaeetus albicilla) as an Indicator of Territorial Ex-posure to Organohalogen Compounds - An Evaluation. Environ. Sci. Technol., 34, 2733-2740.

62. Marsh, G., Athanasiadou, M., Bergman, Å., Asplund, L., 2004. Identification of hydroxylated and methoxylated polybrominated diphenyl ethers in Baltic Sea salmon (Salmo salar) blood. Environ. Sci. Technol., 38, 10-18.

63. Malmvärn, A., Marsh, G., Kautsky, L., Athanasiadou, M., Bergman, A., As-plund, L., 2005. Hydroxylated and methoxylated brominated diphenyl ethers in the red algae Ceramium tenuicorne and blue mussels from the Baltic Sea. Envi-ron. Sci. Technol., 39, 2990-2997.

64. van Bavel, B., Dam, M., Tysklind, M., Lindström, G., 2001. Levels of poly-brominated diphenyl ethers in marine mammals. Organohalogen Compd., 52, 99-103.

65. Vetter, W., Scholz, E., Gaus, C., Müller, J. F., Haynes, D., 2001. Anthropo-genic and Natural Organohalogen Compounds in Blubber of Dolphins and Du-

57

gongs (Dugong dugon) from Northeastern Australia. Arch. Environ. Contam. Toxicol., 41, 221-231.

66. Vetter, W., Stoll, E., Garson, M. J., Fahey, S. J., Gaus, C., Müller, J. F., 2002. Sponge halogenated natural products found at parts-per-million levels in marine mammals. Environ. Toxicol. Chem., 21, 2014-2019.

67. Gribble, G. W., 2000. The Natural Production of Organobromine Compounds. Environ. Sci. Pollut. Res., 7, 37-49.

68. Handayani, D., Edrada, R. A., Proksch, P., Wray, V., Witte, L., van Soest, R. W. M., Kunzman, A., Soedarsono,O., 1997. Four new bioactive polybrominated diphenyl ethers of the sponge Dysidea herbacea from west Sumatra, Indonesia. J. Nat. Prod., 60, 1313-1316.

69. Cameron, G. M., Stapleton, B. L., Simonsen, S. M., Brecknell, D. J., Garson, M. J., 2000. New Sesquiterpene and Brominated Metabolites from the Tropical Marine Sponge Dysidea sp. Tetrahedron, 56, 5247-5252.

70. Kuniyoshi, M., Yamada, K., Higa, T., 1985. A Biologically active Diphenyl Ether from the Green alga Cladophora fascicularis. Experentia 41(4), 523-524.

71. Bignert, A., Göthberg, A., Jensen, S., Litzén, K., Odsjö, T., Olsson, M., Reutergårdh, L., 1993. The need for adequate biological sampling in ecotoxi-cological investigations: a retrospective study of twenty years pollution moni-toring. Sci. Total Environ., 128, 121-139.

72. Bignert, A., 2003. Biological aspects and statistical methods to improve as-sessments in environmental monitoring. Ph.D. thesis/dissertation, Department of Zoology, Stockholm University, ISBN 91-7265-675-1.

73. Olsson, M., Reutergårdh, L., 1986. DDT and PCB pollution trends in the Swedish aquatic environment. Ambio, 15, 103-109.

74. Bignert, A., 2002. Comments Concerning the National Swedish Monitoring Programme in Fresh Water Biota 2001. Report of the Swedish EPA,

75. Buhler, D. R., Wang-Buhler, J.-L., 1998. Rainbow trout cytochrome P450s: purification, molecular aspects, metabolic activity, induction and role in envi-ronmental monitoring. Comp. Biochem. Physiol., Part C, 121, 107-137.

76. de Wit, C. A., Nylund, K., Eriksson, U., Haglund, M., Kierkegaard, A., Asp-lund, L., 2007. Brominated flame retardants in sludge from 50 sewage treatment plants in Sweden: Evidence of anaerobic degradation of HBCD and TBBPA. Manuscript in prep.

77. de Boer, J., Wester, P. G., van der Horst, A., Leonards, P. E. G., 2003. Poly-brominated diphenyl ethers in influents, suspended particulate matter, sedi-ments, sewage treatment plant and effluents and biota from the Netherlands. Environ. Pollut., 122, 63-74.

78. Covaci, A., Voorspoels, S., de Boer, J., 2003. Determination of brominated flame retardants, with emphasis on polybrominated diphenyl ethers (PBDEs) in environmental and human samples - a review. Environ. Int., 29, 735-756.

79. de Boer, J., Allchin, C. R., Law, R. J., Zegers, B. N., Boon, J. P., 2001. Method for the analysis of polybrominated diphenylethers in sediment and bi-ota. TrAC, Trends Anal. Chem., 20, 591-599.

80. Tollbäck, P., Björklund, J., Östman, C., 2003. Large-volume programmed-temperature vaporiser injection for fast gas chromatography with electron cap-ture and mass spectrometric detection of polybrominated diphenyl ethers. J. Chrom. A, 991, 214-253.

81. Alaee, M., Sergeant, D. B., Ikonomou, M. G., Luross, J. M., 2001. A gas chromatography/high-resolution mass spectrometry (GC/HRMS) method for determination of polybrominated diphenyl ethers in fish. Chemosphere, 44, 1489-1495.

58

82. Covaci, A., Gheorghe, A., Voorspoels, S., Maervoet, J., Steen Redeker, E., Blust, R., Schepens, P., 2005. Polybrominated diphenyl ethers, polychlorinated biphenyls and organochlorine pesticides in sediment cores from the Western Scheldt river (Belgium): analytical aspects and depth profiles. Environ. Int., 31, 367-375.

83. Hale, R. C., La Guardia, M. J., Harvey, E., Mainor, T. M., 2002. Potential role of fire retardant-treated polyurethane foam as a source of brominated diphenyl ethers to the US environment. Chemosphere, 46, 729-735.

84. Eljarrat, E., De la Cal, A., Barcelo, D., 2004. Determination of decabromodi-phenyl ether in sediments using selective pressurized liquid extraction followed by GC-NCI-MS. Anal. Bioanal. Chem., 378, 610-614.

85. Saito, K., Sjodin, A., Sandau, C. D., Davis, M. D., Nakazawa, H., Matsuki, Y., Patterson, D. G., 2004. Development of an accelerated solvent extraction and gel permeation chromatography analytical method for measuring persistent or-ganohalogen compounds in adipose and organ tissue analysis. Chemosphere, 57, 373-381.

86. Hartonen, K., Böwadt, S., Hawthorne, S. B., Riekkola, M.-L., 1997. Super-critical fluid extraction with solid-phase trapping of chlorinated and brominated pollutants from sediment samples. J. Chromatogr., A, 774, 229-242.

87. van Bavel, B., Sundelin, E., Lillbäck, J., Dam, M., Lindström, G., 1999. Supercritical fluid extraction of polybrominated diphenyl ethers, PBDEs, from long-finned pilot whale (Globicephala melas) from the Atlantic. Organohalo-gen Compd., 40, 359-362.

88. Wolkers, H., Hammill, M. O., van Bavel, B., 2006. Tissue-specific accumula-tion and lactational transfer of polychlorinated biphenyls, chlorinated pesticides, and brominated flame retardants in hooded seals (Cistophora cristata) from the Gulf of St. Lawrence: Applications for monitoring. Environ. Pollut., 142, 476-486.

89. Vetter, W., 2001. A GC/ECNI-MS Method for the Identification of Lipophilic Anthropogenic and Natural Brominated Compounds in Marine Samples. Anal. Chem., 73, 4951-4957.

90. Bayen, S., Lee, H. K., Obbard, J. P., 2004. Determination of polybrominated diphenyl ethers in marine biological tissues using microwave-assisted extrac-tion. J. Chromatogr. A, 1035, 291-294.

91. Salgado-Petinal, C., Garcia-Chao, M., Llompart, M., Garcia-Jares, C., Cela, R., 2006. Headspace solid-phase microextraction gas chromatography tandem mass spectrometry for the determination of brominated flame retardants in envi-ronmental solid samples. Anal. Bioanal. Chem., 385, 637-644.

92. Salgado-Petinal, C., Llompart, M., Garcia-Jares, C., Garcia-Chao, M., Cela, R., 2006. Simple approach for the determination of brominated flame retardants in environmental solid samples based on solvent extraction and solid-phase mi-croextraction followed by gas chromatography-tandem mass spectrometry. J. Chromatogr. A, 1124, 139-147.

93. Sporring, S., Björklund, E., 2004. Selective pressurized liquid extraction of polychlorinated biphenyls from fat-containing food and feed samples. Influence of cell dimensions, solvent type, temperature and flush volume. J. Chromatogr. A, 1040, 155-161.

94. Jensen, S., Reutergårdh, L., Jansson, B. FAO/SIDA, 1983. Manual of methods in aquatic environment research. Part 9. Analyses of metals and organochlorines in fish. FAO Fish. Tech. Paper, 212, 21-33.

95. Bligh, E. G. , Dyer, W. J., 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol., 37, 911-917.

59

96. Jensen, S., Häggberg, L., Jörundsdottir, H., Odham, G., 2003. A Quantitative Lipid Extraction Method for Residue Analysis of Aquatic Organisms involving non-halogenated Solvents. J. Agric. Food Chem., 51, 5607-5611.

97. Thuresson, K., 2004. Occupational exposure to brominated flame retardants with emphasis on polybrominated diphenyl ethers. Ph.D. thesis/dissertation, Department of Environmental Chemistry, Stockholm University, ISBN 91-7265-990-4.

98. Nylund, K., Asplund, L., Jansson, B., Jonsson, P., Litzén, K., Sellström, U., 1992. Analysis of some polyhalogenated organic pollutants in sediment and sewage sludge. Chemosphere, 24, 1721-1730.

99. Otaka, H. , Hashimoto, S., 2004. Fast matrix digestion with hot ethanolic alkali plus pyrogallol for polychlorinated dibenzo-p-dioxins, polychlorinated dibenzo-furans and coplanar polychlorinated biphenyls analysis in biological samples. Anal. Chim. Acta, 509, 21-26.

100. Lamparski, L. L., Mahle, N. H., Shadoff, L. A., 1978. Determination of Penta-chlorophenol, Hexachlorodibenzo-Para-Dioxin, and Octachlorodibenzo-Para-Dioxin in Bovine Milk. J. Agric. Food Chem., 26, 1113-1116.

100a Reutergårdh L. 1988. Identification and distribution of chlorinated organic pollutants in the environment. Ph.D. thesis/dissertation, Department of Analyti-cal Chemistry, Stockholm University, National Swedish Environmental Protec-tion Board Report 3465, ISBN 91-620-3465-0.

101. Riis, V., Babel, W., 1999. Removal of sulfur interfering in the analysis of organochlorines by GC-ECD. Analyst, 124, 1771-1773.

102. Smedes, F., deBoer, J., 1997. Determination of chlorobiphenyls in sediments - analytical methods. TrAC, Trends Anal. Chem., 16, 503-517.

103. Jensen, S., Renberg, L., Reutergårdh, L., 1977. Residue analysis of sediment and sewage sludge for organochlorines in the presence of elementar sulfur. Anal. Chem., 49, 316-318.

104. Stapleton, H. M., 2006. Instrumental methods and challenges in quantifying polybrominated diphenyl ethers in environmental extracts: a review. Anal. Bio-anal. Chem., 386, 807-817.

105. Korytar, P., Covaci, A., de Boer, J., Gelbin, A., Brinkman, U. A. T., 2005. Retention-time database of 126 polybrominated diphenyl ether congeners and two Bromkal technical mixtures on seven capillary gas chromatographic col-umns. J. Chromatogr. A 1065, 239-249.

106. Björklund, J., Tollbäck, P., Hiarne, C., Dyremark, E., Ostman, C., 2004. Influ-ence of the injection technique and the column system on gas chromatographic determination of polybrominated diphenyl ethers. J. Chromatogr. A, 1041, 201-210.

107. Alaee, M., Backus, S., Cannon, C., 2001. Potential interference of PBDEs in the determination of PCBs and other organochlorine contaminants using elec-tron capture detection. J. Sep. Sci., 24, 465-469.

108. Greaves, J., Bekesi, G., Roboz, J., 1982. Halogen anion formation by poly-brominated compounds in negative chemical ionisation mass spectrometry. Biomed. Mass Spectrom., 9, 406-410.

109. Buser, H-R., 1986. Selective detection of brominated aromatic compounds using gas chromatography/negative chemical ionization mass spectrometry. Anal. Chem., 58, 2913-2919.

110. Ackerman, L. K., Wilson, G. R., Simonich, S. L., 2005. Quantitative analysis of 39 polybrominated diphenyl ethers by isotope dilution GC/low-resolution MS. Anal. Chem., 77, 1979-1987.

60

111. Eljarrat, E., Lacorte, S., Barceló, D., 2002. Optimization of congener-specific analysis of 40 polybrominated diphenyl ethers by gas chromatography/mass spectrometry. J. Mass Spectrom., 37, 76-84.

112. Björklund, J., Tollbäck, P., Östman, C., 2003. Mass spectrometric characteris-tics of decabromodiphenyl ether and the application of isotopic dilution in the electron capture negative ionization mode for the analysis of polybrominated diphenyl ethers. J. Mass Spectrom., 38, 394-400.

113. Covaci, A., de Boer, J., Ryan, J. J., Voorspoels, S., Schepens, P., 2002. Deter-mination of Polybrominated Diphenyl Ethers and Polychlorinated Biphenyls in Human Adipose Tissue by Large-Volume Injection-Narrow-Bore Capillary Gas Chromatography/Electron Impact Low-Resolution Mass Spectrometry. Anal. Chem., 74, 790-798.

114. Larrazabal, D., Martinez, M. A., Eljarrat, E., Barcelo, D., Fabrellas, B., 2004. Optimization of quadrupole ion storage mass spectrometric conditions for the analysis of selected polybrominated diphenyl ethers. Comparative approach with negative chemical ionization and electron impact mass spectrometry. J. Mass Spectrom., 39, 1168-1175.

115. Thomsen, C., Haug, L. S., Leknes, H., Lundanes, E., Becher, G., Lindström G., 2002. Comparing electron ionization high-resolution and electron capture low-resolution mass spectrometric determination of polybrominated diphenyl ethers in plasma, serum and milk. Chemosphere, 46, 641-648.

116. Pirard, C., De Pauw, E., Focant, J. F., 2006. Suitability of tandem-in-time mass spectrometry for polybrominated diphenylether measurement in fish and shell-fish samples: Comparison with high resolution mass spectrometry. J. Chroma-togr. A, 1115, 125-132.

117. Vetter, W. , Jun, W., 2003. Non-polar halogenated natural products bioaccu-mulated in marine samples. II. Brominated and mixed halogenated compounds. Chemosphere, 52, 423-431.

118. Eljarrat, E., De la Cal, A., Barcelo, D., 2003. Potential chlorinated and bromi-nated interferences on the polybrominated diphenyl ether determinations by gas chromatography-mass spectrometry. J. Chromatogr. A, 1008, 181-192.

119. Hanari, N., Kannan, K., Miyake, Y., Okazawa, T., Kodavanti, P. R. S., Aldous, K. M., Yamashita, N., 2006. Occurrence of polybrominated biphenyls, poly-brominated dibenzo-p-dioxins, and polybrominated dibenzofurans as impurities in commercial polybrominated diphenyl ether mixtures. Environ. Sci. Technol., 40, 4400-4405.

120. Covaci, A., Voorspoels, S., 2005. Optimization of the determination of poly-brominated diphenyl ethers in human serum using solid-phase extraction and gas chromatography-electron capture negative ionization mass spectrometry. J. Chromatogr., B-Analytical Technologies in the Biomedical and Life Sciences, 827, 216-223.

121. Luthe, G., Leonards, P. E. G., Reijerink, G. S., Liu, H. L., Johansen, J. E., Robertson, L. W., 2006. Monofluorinated analogues of polybrominated di-phenyl ethers as analytical standards: Synthesis, NMR, and GC-MS characteri-zation and molecular orbital studies. Environ. Sci. Technol., 40, 3023-3029.

122. Bignert, A., Olsson, M., Persson, W., Jensen, S., Zakrisson, S., Litzén, K., Eriksson, U., Häggberg, L., Alsberg, T., 1998. Temporal trends of organochlo-rines in Northern Europe, 1967-1995. Relation to global fractionation, leakage from sediments and international measures. Environ. Pollut., 99, 177-198.

123. Hebert, C. E. , Keenleyside, K. A., 1995. To normalize or not to normalize? fat is the question. Environ. Toxicol. Chem., 14, 801-807.

61

124. Päpke, O., Furst, P., Herrmann, T., 2004. Determination of polybrominated diphenylethers (PBDEs) in biological tissues with special emphasis on QC/QA measures. Talanta, 63, 1203-1211.

125. de Boer, J., Wells, D. E., 2006. Pitfalls in the analysis of brominated flame retardants in environmental, human and food samples - including results of three international interlaboratory studies. TrAC, Trends Anal. Chem., 25, 364-372.

126. de Boer, J., Cofino, W. P., 2002. First world-wide interlaboratory study on polybrominated diphenylethers (PBDEs). Chemosphere, 46, 625-633.

127. Wise, S. A., Poster, D. L., Kucklick, J. R., Keller, J. M., VanderPol, S. S., Sander, L. C., Schantz, M. M., 2006. Standard reference materials (SRMs) for determination of organic contaminants in environmental samples. Anal. Bio-anal. Chem., 386, 1153-1190.

128. van Leeuwen, S. P. J., Van Cleuvenbergen, R., Abalos, M., Pasini, A. L., Eriksson, U., Cleemann, M., Hajslova, J., de Boer, J., 2006. New certified and candidate certified reference materials for the analysis of PCBs, PCDD/Fs, OCPs and BFRs in the environment and food. Trac-TrAC, Trends Anal. Chem., 25, 397-409.

129. Stapleton, H. M., Harner, T., Shoeib, M., Keller, J. M., Schantz, M. M., Leigh, S. D., Wise, S. A., 2006. Determination of polybrominated diphenyl ethers in indoor dust standard reference materials. Anal. Bioanal. Chem., 384, 791-800.

130. Voorspoels, S., Covaci, A., Schepens, P., 2003. Polybrominated diphenyl ethers in marine species from the Belgian North Sea and the Western Scheidt Estuary: Levels, profiles, and distribution. Environ. Sci. Technol., 37, 4348-4357.

131. Horwitz, W., Kamps, L. R., Boyer, K. W., 1980. Quality Assurance in the Analysis of Foods for Trace Constituents. J. Assoc. Off. Anal. Chem., 63, 1344-1354.

132. Thomsen, C., Leknes, H., Lundanes, E., Becher, G., 2001. Brominated flame retardants in laboratory air. J. Chromatogr. A, 923, 299-304.

133. Hazrati, S., Harrad, S., 2006. Causes of variability in concentrations of poly-chlorinated biphenyls and polybrominated diphenyl ethers in indoor air. Envi-ron. Sci. Technol., 40, 7584-7589.

134. Stapleton, H. M., Dodder, N. G., Offenberg, J. H., Schantz, M. M., Wise, S. A., 2005. Polybrominated diphenyl ethers in house dust and clothes dryer lint. Environ. Sci. Technol., 39, 925-931.

135. Wilford, B. H., Shoeib, M., Harner, T., Zhu, J. P., Jones, K. C., 2005. Poly-brominated diphenyl ethers in indoor dust in Ottawa, Canada: Implications for sources and exposure. Environ. Sci. Technol., 39, 7027-7035.

136. IUPAC Compendium of Chemical Terminology, Electronic version, http://goldbook.iupac.org/L03540.html Accessed Mars 2007.

137. Thomsen, V., Schatzlein, D., Mercuro, D., 2003. Limits of detection in spec-troscopy. Spectroscopy, 18, 112-114.

138. DeCarlo, V. J., 1979. Studies on brominated chemicals in the environment. New York Academy of Sciences, Annals (Ann. N. Y. Acad. Sci.), 320, 678-681.

139. Zweidinger, R.A., Cooper, S.D., and Pellizzari, E.D., 1979. Identification and quantitation of brominated fire retardants. In: Van Hall, C.E. (Ed.), Measure-ment of Organic Pollutants in Water and Wastewater, ASTM STP 686, Ameri-can Society for Testing and Materials, pp. 234-250.

140. Kuehl, D. W., Haebler, R., Potter, C., 1991. Chemical residues in dolphins from the U.S. Atlantic coast including Atlantic bottlenose obtained during the 1987/88 mass mortality. Chemosphere, 22, 1071-1084.

62

141. Stanley, J. S., Cramer, P. H., Thornburg, K. R., Remmers, J. C., Breen, J. J., Schwemberger, J., 1991. Mass spectral confirmation of chlorinated and bromi-nated diphenylethers in human adipose tissues. Chemosphere, 23, 1185-1195.

142. Watanabe, I., Kashimoto, T., Tatsukawa, R., 1986. Confirmation of the pres-ence of the flame retardant decabromobiphenyl ether in river sediment from Osaka, Japan. Bull. Environ. Contam. Toxicol., 36, 839-842.

143. Sellström, U., Kierkegaard, A., de Wit, C., Jansson, B., Olsson, M., 1993. Time trend studies of polybrominated diphenyl ethers (PBDE) in biological ma-terial from the Swedish environment. Subtitle: Pike do it flame retarded! (barred under the statute of limitations) Organohalogen Compd., 14, 211-213.

144. de Boer, J., 1990. Brominated diphenyl ethers in Dutch freshwater and marine fish. Organohalogen Compd., 2, 315-318.

145. Hale, R. C., La Guardia, M. J., Harvey, E. P., Mainor, T. M., Duff, W. H., Gaylor, M. O., 2001. Polybrominated diphenyl ether flame retardants in virginia freshwater fishes (USA). Environ. Sci. Technol., 35, 4585-4591.

146. Johnson, A. , Olson, N., 2001. Analysis and Occurrence of Polybrominated Diphenyl Ethers in Washington State Freshwater Fish. Arch. Environ. Contam. Toxicol., 41, 339-344.

147. She, J., Holden, A., Tanner, M., Sharp, M., Adelsbach, T., Hooper, K., 2004. Highest PBDE levels (max 63 ppm) yet found in biota measured in seabird eggs from San Francisco Bay. Organohalogen Compd., 66, 3939-3944.

148. Herzke, D., Berger, U., Kallenborn, R., Nygard, T., Vetter, W., 2005. Bromi-nated flame retardants and other organobromines in Norwegian predatory bird eggs. Chemosphere, 61, 441-449.

149. Vorkamp, K., Thomsen, M., Falk, K., Leslie, H., Moller, S., Sorensen, P. B., 2005. Temporal development of brominated flame retardants in peregrine falcon (Falco peregrinus) eggs from South Greenland (1986-2003). Environ. Sci. Technol., 39, 8199-8206.

150. Law, R. J., Allchin, C. R., Bennett, M. E., Morris, S., Rogan, E., 2002. Poly-brominated diphenyl ethers in two species of marine top predators from Eng-land and Wales. Chemosphere, 46, 673-681.

151. Kuehl, D. W., Haebler, R., 1995. Organochlorine, organobromine, metal, and selenium residues in Bottlenose Dolphins (Tursiops truncatus) collected during an unusual mortality event in the Gulf of Mexico, 1990. Arch. Environ. Contam. Toxicol., 28, 494-499.

152. Lindström, G., Wingfors, H., Dam, M., van Bavel, B., 1999. Identification of 19 Polybrominated Diphenyl ethers (PBDEs) in Long-Finned Pilot Whale (Glo-bicephala melas) from the Atlantic. Arch. Environ. Contam. Toxicol., 36, 355-363.

153. Corsolini, S., Covaci, A., Ademollo, N., Focardi, S., Schepens, P., 2006. Oc-currence of organochlorine pesticides (OCPs) and their enantiomeric signatures, and concentrations of polybrominated diphenyl ethers (PBDEs) in the Adelie penguin food web, Antarcticae. Environ. Pollut., 140, 371-382.

154. Hale, R. C., Alaee, M., Manchester-Neesvig, J. B., Stapleton, H. M., Ikonomou, M. G., 2003. Polybrominated diphenyl ether flame retardants in the North American environment. Environ. Int., 29, 771-779.

155. Kierkegaard, A., Sellström, U., Bignert, A., Olsson, M., Asplund, L., Jansson, B., de Wit, C., 1999. Temporal trends of a polybrominated diphenyl ether (PBDE), a methoxylated PBDE, and hexabromocyclododecane (HBCD) in Swedish biota. Organohalogen Compd., 40, 367-370.

63

156. Johansson, I., Heas-Moisan, K., Guiot, N., Munschy, C., Tronczynski, J., 2006. Polybrominated diphenyl ethers (PBDEs) in mussels from selected French coastal sites: 1981-2003. Chemosphere, 64, 296-305.

157. Norstrom, R. J., Simon, M., Moisey, J., Wakeford, B., Weseloh, D. V. C., 2002. Geographical distribution (2000) and temporal trends (1981-2000) of brominated diphenyl ethers in Great Lakes herring gull eggs. Environ. Sci. Technol., 36, 4783-4789.

158. Luross, J. M., Alaee, M., Sergeant, D. B., Whittle, D. M., Solomon, K. R., 2000. Spatial and Temporal Distribution of Polybrominated Diphenyl Ethers in Lake Trout from the Great Lakes. Organohalogen Compd., 47, 73-76.

159. Zhu, L. Y., Hites, R. A., 2004. Temporal trends and spatial distributions of brominated flame retardants in archived fishes from the Great Lakes. Environ. Sci. Technol., 38, 2779-2784.

160. She, J., Petreas, M., Winkler, J., Visita, P., McKinney, M., Kopec, D., 2002. PBDEs in the San Francisco Bay Area: measurements in harbor seal blubber and human breast adipose tissue. Chemosphere, 46, 697-707.

161. Lebeuf, M., Gouteux, B., Measures, L., Trottier, S., 2004. Levels and temporal trends (1988-1999) of polybrominated diphenyl ethers in Beluga whales (Del-phinapterus leucas) from the St. Lawrence estuary, Canada. Environ. Sci. Tech-nol., 38, 2971-2977.

162. Asplund, L., Athanasiadou, M., Sjödin, A., Bergman, Å., Börjeson, H., 1999. Organohalogen Substances in Muscle, Egg and Blood from Healthy Baltic Salmon (Salmo salar) and Baltic Salmon that produced Offspring with the M74 Syndrome. Ambio, 28, 67-76.

163. Marsh, G., Athanasiadou, M., Bergman, Å., Jakobsson, E., Asplund, L., 2001. Hydroxylated and Methoxylated Polybrominated Diphenyl Ethers in Salmon Plasma: Synthesis and Identification. Organohalogen Compd., 52, 62-66.

164. Kierkegaard, A., Burreau, S., Marsh, G., Klasson-Wehler, E., de Wit, C., As-plund, L., 2001. Metabolism and distribution of 2,2',4,4'-tetrabromo[14C]diphenyl ether in pike (Esox lucius) after dietary exposure. Or-ganohalogen Compd., 52, 58-61

165. Marsh, G., Athanasiadou, M., Athanassiadis, I., Sandholm, A., 2006. Identifi-cation of hydroxylated metabolites in 2,2 ',4,4 '-tetrabromodiphenyl ether ex-posed rats. Chemosphere, 63, 690-697.

166. Mörck, A., Hakk, H., Örn, U., Wehler, E. K., 2003. Decabromodiphenyl ether in the rat: Absorption, distribution, metabolism, and excretion. Drug Metab. Dispos., 31, 900-907.

167. Anjaneyulu, V., Nageswara, R. K., Radhika, P., Muralikrishna, M., 1996. A New Tetrabromodiphenyl Ether From the Sponge Dysideaherbacea of the In-dian Ocean. Indian J. Chem., 35B, 89-90.

168. Utkina, N. K., Denisenko, V. A., Virovaya, M. V., Scholokova, O. V., Proko-feva, N. G., 2002. Two New Minor Polybrominated Dibenzo-p-dioxins from the Marine Sponge Dysidea dendyi. J. Nat. Prod., 65, 1213-1215.

169. Sinkkonen, S., Rantalainen, A. L., Paasivirta, J., Lahtipera, M., 2004. Poly-brominated methoxy diphenyl ethers (MeO-PBDEs) in fish and guillemot of Baltic, Atlantic and Arctic environments. Chemosphere, 56, 767-775.

170. Teuten, E. L., Xu, L., Reddy, C. M., 2005. Two abundant bioaccumulated halogenated compounds are natural products. Science, 308, 1413.

171. Malmberg, T., Athanasiadou, M., Marsh, G., Brandt, I., Bergman, Å., 2005. Identification of hydroxylated polybrominated diphenyl ether metabolites in blood plasma from polybrominated diphenyl ether exposed rats. Environ. Sci. Technol., 39, 5342-5348.

64

172. Valters, K., Li, H. X., Alaee, M., D'Sa, I., Marsh, G., Bergman, Å., Letcher, R. J., 2005. Polybrominated diphenyl ethers and hydroxylated and methoxylated brominated and chlorinated analogues in the plasma of fish from the Detroit River. Environ. Sci. Technol., 39, 5612-5619.

173. Verreault, J., Gabrielsen, G. V., Chu, S. G., Muir, D. C. G., Andersen, M., Hamaed, A., Letcher, R. J., 2005. Flame retardants and methoxylated and hy-droxylated polybrominated diphenyl ethers in two Norwegian Arctic top preda-tors: Glaucous gulls and polar bears. Environ. Sci. Technol., 39, 6021-6028.

174. Asplund, L., Bignert, A., Nylund, K., 2004. Comparison of spatial and tempo-ral trends of methoxylated PBDEs, PBDEs, and hexabromocyclododecane in herring along the Swedish coast. Organohalogen Compd., 66, 3988-3992.

175. Hardy, M. L., 2000. Properties of the major commercial PBDPO flame retar-dant, DBDPO, in comparison to PBB and PCB. Organohalogen Compd., 47, 233-236.

176. Sellström, U., 1999. Determination of Some Polybrominated Flame Retardants in Biota, Sediment and Sewage Sludge. Ph.D. thesis/dissertation, Department of Environmental Chemistry, Stockholm University, ISBN 91-7265-023-0.

177. Burreau, S., Axelman, J., Broman, D., Jakobsson, E., 1997. Dietary uptake in pike (Esox lucius) of some PCBs, PCNs and PBDEs administered in natural diet. Environ. Toxicol. Chem., 16, 2508-2513.

178. Stapleton, H. M., Brazil, B., Holbrook, R. D., Mitchelmore, C. L., Benedict, R., Konstantinov, A., Potter, D., 2006. In vivo and in vitro debromination of de-cabromodiphenyl ether (BDE 209) by juvenile rainbow trout and common carp. Environ. Sci. Technol., 40, 4653-4658.

179. Sandholm, A., Emanuelsson, B. M., Wehler, E. K., 2003. Bioavailability and half-life of decabromodiphenyl ether (BDE-209) in rat. Xenobiotica, 33, 1149-1158.

180. Norris, J. M., Kociba, R. J., Schwetz, B. A., Rose, J. Q., Humiston, C. G., Jewett, G. L., Gehring, P. J., Mailhes, J. B., 1975. Toxicology of octabromobi-phenyl and decabromodiphenyl oxide. Environ. Health Persp., 11, 153-161.

181. El Dareer, S. M., Kalin, J. R., Tillery, K. F., Hill, D. L., 1987. Disposition of decabromobiphenyl ether in rats dosed intravenously or by feeding. J. Toxicol. Environ. Health, 22, 405-415.

182. Gobas, F. A. P. C., Muir, D. C. G., Mackay, D., 1988. Dynamics of Dietary Bioaccumulation and Fecal Elimination of Hydrophobic Organic-Chemicals in Fish. Chemosphere, 17, 943-962.

183. Muir, D. C. G. , Yarechewski, A. L., 1988. Dietary accumulation of four chlo-rinated dioxin congeners by rainbow trout and fathead minnows. Environ. Toxi-col. Chem., 7, 227-236.

184. McLachlan, M., 993. Mass balance of polychlorinated biphenyls and other organochlorine compounds in a lactating cow. J. Agric. Food Chem., 41, 474-480.

185. Thomas, G. O., Sweetman, A. J., Jones, K. C., 1999. Input-output balance of polychlorinated biphenyls in a long-term study of lactating daily cows. Environ. Sci. Technol., 33, 104-112.

186. Opperhuizen, A., Velde, E. W. v. d., Gobas, F. A. P. C., Liem, D. A. K., Steen, J. M. D. v. d., Hutzinger, O., 1985. Relationship between bioconcentration in fish and steric factors of hydrophobic chemicals. Chemosphere, 14, 1871-1896.

187. Opperhuizen, A., Sijm, D. T. H. M., 1990. Bioaccumulation and Biotransfor-mation of Polychlorinated Dibenzo-Para-Dioxins and Dibenzofurans in Fish. Environ. Toxicol. Chem., 9, 175-186.

65

188. Kelly, B. C., Gobas, F. A. P. C., McLachlan, M. S., 2004. Intestinal absorption and biomagnification of organic contaminants in fish, wildlife, and humans. Environ. Toxicol. Chem., 23, 2324-2336.

189. Andersson, P. L., Wågman, N., Berg, H. A., Olsson, P.-E., Tysklind, M., 1999. Biomagnification of structurally matched polychlorinated and polybrominated diphenylethers (PCDE/PBDE) in Zebrafish (Danio rerio). Organohalogen Compd., 43, 9-12.

190. Tomy, G. T., Palace, V. P., Halldorson, T., Braekevelt, E., Danell, R., Wautier, K., Evans, B., Brinkworth, L., Fisk, A. T., 2004. Bioaccumulation, biotransfor-mation, and biochemical effects of brominated diphenyl ethers in juvenile lake trout (Salvelinus namaycush). Environ. Sci. Technol., 38, 1496-1504.

191. Stapleton, H. M., Letcher, R. J., Li, J., Baker, J. E., 2004. Dietary accumula-tion and metabolism of polybrominated diphenyl ethers by juvenile carp (Cy-prinus carpio). Environ. Toxicol. Chem., 23, 1939-1946.

192. Stapleton, H. M., Letcher, R. J., Baker, J. E., 2004. Debromination of poly-brominated diphenyl ether congeners BDE 99 and BDE 183 in the intestinal tract of the common carp (Cyprinus carpio). Environ. Sci. Technol., 38, 1054-1061.

193. Isosaari, P., Lundebye, A. K., Ritchie, G., Lie, O., Kiviranta, H., Vartiainen, T., 2005. Dietary accumulation efficiencies and biotransformation of polybro-minated diphenyl ethers in farmed Atlantic salmon (Salmo salar). Food Addit. Contam., 22, 829-837.

194. Thomas, G. O., Moss, S. E. W., Asplund, L., Hall, A. J., 2005. Absorption of decabromodiphenyl ether and other organohalogen chemicals by grey seals (Halichoerus grypus). Environ. Pollut., 133, 581-586.

195. Staskal, D. F., Hakk, H., Bauer, D., Diliberto, J. J., Birnbaum, L. S., 2006. Toxicokinetics of polybrominated diphenyl ether congeners 47, 99, 100, and 153 in mice. Toxicol. Sci., 94, 28-37.

196. Örn, U., Klasson-Wehler, E., 1998. Metabolism of 2,2',4,4'-tetrabromodiphenyl ether in rat and mouse. Xenobiotica, 28, 199-211.

197. Huwe, J.K. and Smith, D.J., 2007. Accumulation, whole-body depletion, and debromination of decabromodiphenyl ether in male Sprague-Dawley rats fol-lowing dietary exposure. Environ. Sci. Technol., ASAP available on the web the 22 February.

198. Zitko, V., 1977. Uptake and excretion of chlorinated and brominated hydro-carbons by fish. Fisheries & Marine Service Technical Report No. 737.

199. Yamaguchi, Y., Kawano, M., Tatsukawa, R., 1988. Tissue distribution and excretion of hexabromobenzene and its debrominated metabolites in the rat. Arch. Environ. Contam. Toxicol.,. 17, 807-812.

200. Stapleton, H. M., Alaee, M., Letcher, R. J., Baker, J. E., 2004. Debromination of the flame retardant Decabromodiphenyl ether by juvenile carp (Cyprinus carpio) following dietary exposure. Environ. Sci. Technol., 38, 112-119.

201. La Guardia, M. J., Hale, R. C., Harvey, E., 2004. Environmental debromina-tion of decabrominated diphenyl ether. The Third International Workshop on Brominated Flame Retardants, Toronto, Canada, June 6-9, 231-234.

202. Van den Steen, E., Covaci, A., Jaspers, V. L. B., Dauwe, T., Voorspoels, S., Eens, M., Pinxten, R., 2006. Accumulation, tissue-specific distribution and bio-transformation of BDE 209 in Eorepean starlings ( Sturnus vulgaris) after expo-sure with silastic implants. Organohalogen Compd., 68, 647-650.

203. He, J. Z., Robrock, K. R., Alvarez-Cohen, L., 2006. Microbial reductive de-bromination of polybrominated diphenyl ethers (PBDEs). Environ. Sci. Tech-nol., 40, 4429-4434.

66

204. Gerecke, A. C., Hartmann, P. C., Heeb, N. V., Kohler, H. P. E., Giger, W., Schmid, P., Zennegg, M., Kohler, M., 2005. Anaerobic degradation of decab-romodiphenyl ether. Environ. Sci. Technol., 39, 1078-1083.

205. Lebeuf, M., Couillard, C. M., Legare, B., Trottier, S., 2006. Effects of DeBDE and PCB-126 on hepatic concentrations of PBDEs and methoxy-PBDEs in At-lantic tomcod. Environ. Sci. Technol., 40, 3211-3216.

206. Thuresson, K., Bergman, A., Jakobsson, K., 2005. Occupational exposure to commercial decabromodiphenyl ether in workers manufacturing or handling flame-retarded rubber. Environ. Sci. Technol., 39, 1980-1986.

207. Burreau, S., Zebuhr, Y., Broman, D., Ishaq, R., 2006. Biomagnification of PBDEs and PCBs in food webs from the Baltic Sea and the northern Atlantic Ocean. Sci. Total Environ., 366, 659-672.

208. Law, R. J., Allchin, C. R., Mead, L. K., 2005. Brominated diphenyl ethers in the blubber of twelve species of marine mammals stranded in the UK. Mar. Pol-lut. Bull., 50, 356-359.

209. Ter Schure, A. F. H., Larsson, P., Agrell, C., Boon, J. P., 2004. Atmospheric transport of polybrominated diphenyl ethers and polychlorinated biphenyls to the Baltic sea. Environ. Sci. Technol., 38, 1282-1287.

210. de Boer, J., Leslie, H. A., Leonards, P. E. G., Bersuder, P., Morris, S., Allchin, C. R., 2004. Screening and time trend study of decabromodiphenyl ether and hexabromocyclododecane in birds. The Third International Workshop on Bro-minated Flame Retardants, Toronto, Canada, 125-128.

211. Chen, D., Mai, B., Song, J., Sun, Q., Luo, Y., Luo, X., Zeng, E. Y., Hale, R. C., 2007. Polybrominated Diphenyl Ethers in Birds of Prey from Northern China. Environ. Sci. Technol., ASAP web released the 7 february.

212. Voorspoels, S., Covaci, A., Lepom, P., Escutenaire, S., Schepens, P., 2006. Remarkable Findings Concerning PBDEs in the Terrestrial Top-Predator Red Fox (Vulpes vulpes). Environ. Sci. Technol., 40, 2937-2943.

213. Christensen, J. R., Macduffee, M., Macdonald, R. W., Whiticar, M., Ross, P. S., 2005. Persistent organic pollutants in British Columbia grizzly bears: Con-sequence of divergent diets. Environ. Sci. Technol., 39, 6952-6960.

214. Thuresson, K., Höglund, P., Hagmar, L., Sjödin, A., Bergman, Å., Jakobsson, K., 2006. Apparent half-lives of hepta- to decabrominated diphenyl ethers in human serum as determined in occupationally exposed workers. Environ. Health Persp., 114, 176-181.

215. McLachlan, M. S., 1996. Bioaccumulation of hydrophobic chemicals in agri-cultural feed chains. Environ. Sci. Technol., 30, 252-259.

216. Sjödin, A., Hagmar, L., Klasson-Wehler, E., Kronholm-Diab, K., Jakobsson, E., Bergman, Å., 1999. Flame retardant exposure: Polybrominated diphenyl ethers in blood from Swedish workers. Environ. Health Persp., 107 (8), 643-648.

217. Santillo, D., Johnston, P., Brigden, K., Greenpeace Research Laboratories Technical Note, 2001. The presence of brominated flame retardants and organo-tin compounds in dusts collected from Parliament buildings from eight coun-tries. 03/2001, http://archive.greenpeace.org/toxics/reports/eudust.pdf, 1-22, Ac-cessed Mars 2007

218. Jones-Otazo, H. A., Clarke, J. P., Diamond, M. L., Archbold, J. A., Ferguson, G., Harner, T., Richardson, G. M., Ryan, J. J., Wilford, B., 2005. Is house dust the missing exposure pathway for PBDEs? An analysis of the urban fate and human exposure to PBDEs. Environ. Sci. Technol., 39, 5121-5130.

219. Thomas, G. O., Sweetman, A. J., Jones, K. C., 1999. Metabolism and body-burden of PCBs in lactating dairy cows. Chemosphere, 39, 1533-1544.

67

220. Kierkegaard, A., de Wit, C. A., Asplund, L., McLachlan, M. S., Thomas, G. O., Sweetman, A. J., Jones, K. C., 2007. A mass balance of lower brominated diphenyl ethers (PBDEs) in lactating cows. Manuscript in prep.

221. Sormo, E. G., Skaare, J. U., Lydersen, C., Kovacs, K. M., Hammill, M. O., Jenssen, B. M., 2003. Partitioning of persistent organic pollutants in grey seal (Halichoerus grypus) mother-pup pairs. Sci. Total Environ., 302, 145-155.

222. Debier, C., Pomeroy, P. P., Dupont, C., Joiris, C., Comblin, V., Le Boulenge, E., Larondelle, Y., Thome, J. P., 2003. Dynamics of PCB transfer from mother to pup during lactation in UK grey seals Halichoerus grypus: differences in PCB profile between compartments of transfer and changes during the lactation period. Mar. Ecol.: Prog. Ser., 247, 249-256.

223. Inoue, K., Harada, K., Takenaka, K., Uehara, S., Kono, M., Shimizu, T., Taka-suga, T., Senthilkumar, K., Yamashita, F., Koizumi, A., 2006. Levels and con-centration ratios of polychlorinated biphenyls and polybrominated diphenyl ethers in serum and breast milk in Japanese mothers. Environ. Health Persp., 114, 1179-1185.

224. Darnerud, P. O., Atuma, S., Aune, M., Bjerselius, R., Glynn, A., Grawe, K. P., Becker, W., 2006. Dietary intake estimations of organohalogen contaminants (dioxins, PCB, PBDE and chlorinated pesticides, e.g. DDT) based on Swedish market basket data. Food Chem. Toxicol., 44, 1597-1606.

225. Voorspoels, S., Covaci, A., Neels, H., Schepens, P., 2007. Dietary PBDE in-take: A market-basket study in Belgium. Environ. Int., 33, 93-97.

226. Kiviranta, H., Ovaskainen, M. A. L., Vartiainen, T., 2004. Market basket study on dietary intake of PCDD/Fs, PCBs, and PBDEs in Finland. Environ. Int., 30, 923-932.

227. Schecter, A., Päpke, O., Harris, T. R., Tung, K. C., Musumba, A., Olson, J., Birnbaum, L., 2006. Polybrominated Diphenyl Ether (PBDE) Levels in an Ex-panded Market Basket Survey of U.S. Food and Estimated PBDE Dietary In-take by Age and Sex. Environ. Health Persp., 114, 1515-1520.

228. Ryan, J. J., Patry, B., 2001. Body burdens and food exposure in Canada for polybrominated diphenyl ethers (BDEs). Organohalogen Compd., 51, 226-229.

229. Gomara, B., Herrero, L., Gonzalez, M. J., 2006. Survey of polybrominated diphenyl ether levels in Spanish commercial foodstuffs. Environ. Sci. Technol., 40, 7541-7547.

230. Schecter, A., Päpke, O., Tung, K. C., Staskal, D., Birnbaum, L., 2004. Poly-brominated diphenyl ethers contamination of United States food. Environ. Sci. Technol., 38, 5306-5311.

231. Veltman, K., Hendriks, J., Huijbregts, M., Leonards, P., Heuvel-Greve, M., Vethaak, D., 2005. Accumulation of organochlorines and brominated flame re-tardants in estuarine and marine food chains: Field measurements and model calculations. Mar. Pollut. Bull., 50, 1085-1102.

232. Streets, S. S., Henderson, S. A., Stoner, A. D., Carlson, D. L., Simcik, M. F., Swackhamer, D. L., 2006. Partitioning and bioaccumulation of PBDEs and PCBs in Lake Michigan. Environ. Sci. Technol., 40, 7263-7269.

233. Renner, R., 2004. In US, flame retardants will be voluntarily phased out. Environ. Sci. Technol., 38, 14A.

234. Eljarrat, E., Labandeira, A., Martinez, A., Fabrellas, B., Barceló, D., 2005. Occurrence of the "new" brominated flame retardant, decabromodiphenyl eth-ane, in sewage sludge from Spain. Organohalogen Compd., 67, 459-461.

235. Konstantinov, A., Arsenault, G., Chittim, B., Kolic, T., MacPherson, K., McAlees, A., McCrindle, R., Potter, D., Reiner, E. J., Tashiro, C., Yeo, B., 2006. Characterization of mass-labeled [C-13 (14)]-decabromodiphenylethane and

68

its use as a surrogate standard in the analysis of sewage sludge samples. Chemosphere, 64, 245-249.

236. Pettersson-Julander, A., van Bavel, B., Engwall, M., Westberg, H., 2004. Per-sonal air sampling and analysis of polybrominated diphenyl ethers and other bromine containing compounds at an electronic recycling facility in Sweden. J. Environ. Monit., 6, 874-880.

237. Julander, A., Westberg, H., Engwall, M., van Bavel, B., 2005. Distribution of brominated flame retardants in different dust fractions in air from an electronics recycling facility. Sci. Total Environ., 350, 151-160.

238. Karlsson, M., Julander, A., van Bavel, B., Hardell, L., 2007. Levels of bromi-nated flame retardants in blood in relation to levels in household air and dust. Environ. Int., 33, 62-69.

239. Zhu, L. Y., Hites, R. A., 2006. Brominated flame retardants in tree bark from North America. Environ. Sci. Technol., 40, 3711-3716.

240. Law, K., Halldorson, T., Danell, R., Stern, G., Gewurtz, S., Alaee, M., Marvin, C., Whittle, M., Tomy, G., 2006. Bioaccumulation and trophic transfer of some brominated flame retardants in a Lake Winnipeg (Canada) food web. Environ. Toxicol. Chem., 25, 2177-2186.

241. Jakab, E., Uddin, Md. A., Bhaskar, T., Sakata, Y., 2003. Thermal decomposi-tion of flame-retarded high-impact polystyrene. J. Anal. Appl. Pyrolysis, 68-69, 83-99.

242. Hoh, E., Zhu, L. Y., Hites, R. A., 2005. Novel flame retardants, 1,2-bis(2,4,6-tribromophenoxy)ethane and 2,3,4,5,6-pentabromoethylbenzene, in United States' environmental samples. Environ. Sci. Technol., 39, 2472-2477.

243. Schwarzbauer, J., Ricking, M., Franke, S., Francke, W., 2001. Halogenated organic contaminants in sediments of the Havel and Spree rivers (Germany). Part 5 of organic compounds as contaminants of the Elbe river and its tributar-ies. Environ. Sci. Technol., 35, 4015-4025.

244. Gouteux, B., Alaee, M., Mabury, S., Muir, D., 2006. Emerging brominated aromatic compounds in the environment. P682, Poster at the SETAC North America 27th Annual Meeting in Montréal, Canada, 5-9 November.

245. Gouteux, B., Alaee, M., Muir, D., 2006. Oligomeric BFRs: a potential source for emerging brominated aromatic compounds. Organohalogen Compd., 68, 662.