Thesis Outline

Home , Organobromine compound

Identification of brominated organic compounds in aquatic biota and exploration of bromine isotope analysis for source apportionment

Maria Unger

Doctoral Thesis Department of Materials and Environmental Chemistry (MMK) Stockholm University Stockholm 2010

Printed in Sweden by US-AB, Stockholm 2010 Distributors: Department of Materials and Environmental Chemistry (MMK), Stockholm University Department of Applied Environmental Science (ITM), Stockholm University

Cover photo by Sven-Erik Nilsson. Smedfjärden/Kvädöfjärden. Farleden mellan Huvududden på Stora Askö och Huvudskär.

What will the winter bring if not yet another spring

Kamelot, Serenade, Black Halo 2005

Forward momentum!

Abstract

Brominated organic compounds (BOCs) of both natural and anthropogenic origin are abundant in the environment. Most compounds are either clearly natural or clearly anthropogenic but some are of either mixed or uncertain origin. This thesis aims to identify some naturally produced BOCs and to develop a method for analysis of the bromine isotopic composition in BOCs found in the environment. Polybrominated dibenzo-p-dioxins (PBDDs) in the Baltic Sea are believed to be of natural origin although their source is unknown. Since marine sponges are major producers of brominated natural products in tropical waters, BOCs were quantified in a sponge (Ephydatia fluviatilis) from the Baltic Sea (Paper I). The results showed that the sponge does not seem to be a major producer of PBDDs in the Baltic Sea. In this study, mixed brominated/chlorinated dibenzo-p-dioxins were however discovered for the first time in a background environment without an apparent anthropogenic source. The use of nuclear magnetic resonance spectroscopy (NMR) is unusual in analytical environmental chemistry due to its sample requirements. Preparative capillary gas chromatography was used to isolate a sufficient amount of an unidentified BOC from northern bottlenose whale (Hyperoodon ampullatus) blubber (Paper II) to enable NMR analysis for identification of the compound. The bromine isotopic composition of BOCs may give information on the origin and environmental fate of these compounds. The first steps in this process are the development of a method to determine the bromine isotope ratio in environmentally relevant BOCs (Paper III) and measuring the bromine isotope ratio of several standard substances to establish an anthropogenic endpoint (Paper IV).

Table of contents

Abstract ...... 5 Table of contents ...... 6 Abbreviations ...... 8 List of papers ...... 9 Statement ...... 10 Thesis objectives ...... 11 1. Brominated organic compounds in the marine environment 13 1.1 Naturally produced brominated organic compounds ...... 13 1.2 Natural production of brominated organic compounds ...... 17 2. Bromine isotopes ...... 20 2.1 Kinetic isotope effect ...... 20 2.2 Isotope fractionation processes ...... 20 3. Samples and purification methods ...... 23 3.1 Samples ...... 23 3.2 Extraction and lipid removal methods ...... 24 3.3 Chromatographic separation ...... 27 4. Analysis ...... 29 4.1 GC-MS ...... 29 4.2 NMR ...... 30 4.3 GC-multiple-collector-ICP-MS ...... 31 5. Summary of papers...... 33 5.1 Paper I - Identification and quantification of PBDDs and Br/Cl-DDs ...... 33 5.2 Paper II - NMR-identification of a novel brominated organic compound isolated from whale blubber ...... 34 5.3 Paper III - Development of a GC-MC-ICP-MS method for brominated diaromatic compounds ...... 34 5.4 Paper IV - Measurements of δ81Br in anthropogenic and natural monoaromatic compounds ...... 35

6. Results and Discussion ...... 36 6.1 Identification of natural brominated organic compounds ...... 36 6.2 Development of CSIA-Br ...... 39 7. Conclusions ...... 42 8. Future perspectives ...... 43 9. Sammanfattning på svenska ...... 44 10. Acknowledgements ...... 45 11. References ...... 48

Abbreviations

ACN Acetonitrile BOCs Brominated organic compounds Br/Cl-DDs Mixed brominated/chlorinated dibenzo-p- dioxins CIS Cooled injection system CSIA Compound specific isotope analysis DDT 2,2-bis(4-chlorophenyl)-1,1,1-trichloroethane ECNI Electron capture negative ionization EI Electron ionization GC-MC-ICP-MS Gas chromatography multiple collector inductively coupled ion plasma mass spectrometry GC-MS Gas chromatography – mass spectrometry HRMS High resolution mass spectrometry ICP Inductively coupled plasma KIE Kinetic isotope effect MBB Monobromobenzene MeOMe-BDE42 6-Methoxy-5-methyl-2,2’3,4’-tetrabromo diphenyl ether MeO-PBDEs Methoxylated polybrominated diphenyl ethers NMR Nuclear magnetic resonance spectroscopy OH-PBDEs Hydroxylated polybrominated diphenyl ethers PBB Polybrominated biphenyl PBDDs Polybrominated dibenzo-p-dioxins PBDEs Polybrominated diphenyl ethers PBDFs Polybrominated dibenzofurans PCBs Polychlorinated biphenyls PCDDs Polychlorinated dibenzo-p-dioxins PCGC Preparative capillary gas chromatograph R Isotope ratio RF Radio frequency SIM Selective ion monitoring TCDD 2,3,7,8-Tetrachlorodibenzo-p-dioxin TriBDD s Tribromodibenzo-p-dioxins

List of papers

Paper I Maria Unger, Lillemor Asplund, Peter Haglund, Anna Malmvärn, Kristina Arnoldsson, Örjan Gustafsson. Polybrominated and Mixed Brominated/Chlorinated Dibenzo-p-Dioxins in Sponge (Ephydatia fluviatilis) from the Baltic Sea. Environmental Science & Technology, 2009, 43, 8245-8250

Paper II Maria Unger, Lillemor Asplund, Göran Marsh, Örjan Gustafsson. Characterization of an abundant and novel methyl- and methoxy-substituted brominated diphenyl ether isolated from whale blubber. Chemosphere, 2010, 79, 408-413

Paper III Henry Holmstrand, Maria Unger, Daniel Carrizo, Per Andersson, Örjan Gustafsson. Compound-specific bromine isotope analysis of brominated diphenyl ethers using GC-multiple collector-ICP-MS. Submitted to Rapid Communications in Mass Spectrometry

Paper IV Daniel Carrizo, Maria Unger, Henry Holmstrand, Per Andersson, Örjan Gustafsson, Sean P. Sylva, Christopher M. Reddy. Compound-specific bromine isotope composition of natural and industrially-synthesized organobromine substances. Manuscript

Papers I and II are reproduced with permission from American Chemical Society and Elsevier, respectively.

Statement

I, Maria Unger, have contributed with the following to the papers included in this thesis: In Paper I, I planned and performed the sampling and a major part of the purification. I was responsible for data analysis and interpretation of mass spectra. I had main responsibility for writing the paper. In Paper II, I planned the project. I did a major part of the sample purification and was responsible for the PCGC isolation of the compound, data analysis and interpretation of mass spectra. I took part in the NMR spectra interpretation and had main responsibility for writing the paper. In Paper III, I took part in planning the set up of the instrument and experiments. I took part in the measurements and data evaluation and assisted in writing the paper. In Paper IV, I took part in planning the set up of the instrument and experiments. I took part in the measurements, data evaluation and in writing the paper.

Papers not included in the thesis Emma L. Teuten, Carl G. Johnson, Manolis Mandalakis, Lillemor Asplund, Örjan Gustafsson, Maria Unger, Göran Marsh, Christopher M. Reddy Spectral characterization of two bioaccumulated methoxylated polybrominated diphenyl ethers. Chemosphere, 2006, 62, 197-203

Maria Unger and Örjan Gustafsson PAHs in Stockholm window films: Evaluation of the utility of window film content as indicator of PAHs in urban air. Atmospheric Environment, 2008, 42, 5550-5557

Thesis objectives

The aims of this thesis were to identify natural brominated compounds in the aquatic environment and to develop a method for source apportionment of natural and anthropogenic organobromine compounds based on their stable bromine isotopic composition.

The specific objectives for the papers were: Paper I:  Investigate the role of an abundant sponge in the production of polybrominated dibenzo-p-dioxins (PBDDs) in the Baltic Sea by identification and quantification of halogenated dibenzo-p-dioxins in the sponge.

Paper II:  Isolate a specific unidentified organobromine compound from northern bottlenose whale blubber using preparative capillary gas chromatography to obtain a sufficient amount of pure compound for identification with nuclear magnetic resonance spectroscopy (NMR).

Paper III:  Develop a method for online compound specific isotope analysis of stable bromine isotopes (δ81Br) in polybrominated diphenyl ethers (PBDEs) using a gas chromatograph coupled to an inductively coupled plasma and a multiple collector mass spectrometer (GC-MC-ICP- MS).

Paper IV:  Evaluate the variation in δ81Br in anthropogenic brominated organic monocyclic aromatic compounds to establish an anthropogenic endpoint for bromine isotope analysis of environmentally occurring brominated organic compounds (BOCs).

1. Brominated organic compounds in the marine environment

There are thousands of halogenated compounds of both natural and anthropogenic origin in the environment. The most well- known environmentally occurring organohalogen compounds are DDT, polychlorinated biphenyls (PCBs) and polychlorinated dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs). During the last decades other anthropogenic halogenated substances such as brominated flame retardants (BFRs) (de Wit et al., 2006; Law et al., 2006) and polyfluorinated compounds (Giesy et al., 2001) have received growing attention as environmental contaminants. However, it has also been recognised that a majority of the halogenated organic compounds in the aquatic environment may be of natural origin.

A. B.

C. D.

Figure 1.1. Well-known anthropogenic organohalogen environmental contaminants. Structures of A.) DDT, B) a PCDD, C.) a PCB and D.) a PCDF.

1.1 Naturally produced brominated organic compounds Nearly 5000 naturally produced halogenated organic compounds (Gribble et al., 1999) have been identified, from simple gaseous methyl halides to much larger and sometimes very complex

13 structures (Gribble, 1998; Gribble, 1999; Faulkner, 2000; Faulkner, 2001). Some of the brominated organic compounds (BOCs) that have been identified as natural products are brominated phenols (Whitfield et al., 1999; Fan et al., 2003) methoxylated and hydroxylated polybromodiphenyl ethers (MeO-PBDEs and OH- PBDEs) (Fu et al., 1995; Cameron et al., 2000), brominated bipyrrols (Tittlemier et al., 1999; Teuten et al., 2006c; Teuten et al., 2007), dimethoxylated polybrominated biphenyls (diMeO- PBBs) (Marsh et al., 2005; Teuten et al., 2007; Vetter et al., 2008) and methoxylated and hydroxylated polybrominated dibenzo-p-dioxins (MeO-PBDDs and OH-PBDDs) (Utkina et al., 2001; Utkina et al., 2002). (Figure 1.2). These compounds are mostly detected at low levels but are sometimes present in high amounts in the producing organisms (Unson et al., 1994). While some natural BOCs may, for example, serve as pigments (Saenger et al., 1976), feeding deterrents (Kurata et al., 1997), or have antibacterial effects ( Sharma et al., 1969), the biological role of the majority is still unknown.

A. H B. C. D. H

Figure 1.2. Naturally produced BOCs. A) 2,4,6-tribromophenol, B) a diMeO-PBB C) 1,1’-dimethyl-2,2’-hexabromobipyrrol and D) a OH- PBDD

Brominated phenols Brominated phenols of both natural (Fu et al., 1995; Fielman et al., 1999; Teuten et al., 2006b) and anthropogenic (World Health Organization, 2005) origin are common in the environment. For instance, 2,4,6-tribromophenol is produced by red, brown and green alga (Fielman et al., 1999; Flodin et al., 1999; Whitfield et al., 1999; Flodin et al., 2000; Chung et al.,

2003) and polychaeta (Fielman et al., 1999) but is also anthropogenically produced and used as a flame retardant and for wood preservation (World Health Organization, 2005).

OH-PBDEs and MeO-PBDEs Marine organisms, such as algae, sponges and turnicates, produce OH-PBDEs and MeO-PBDEs (Figure 1.3) (Fu et al., 1995; Gribble, 1998; Gribble, 1999; Gribble et al., 1999; Utkina et al., 2001). These compounds are found worldwide at different trophic levels of the marine food web (Haglund et al., 1997; Marsh et al., 2004a; Melcher et al., 2005; Malmvärn et al., 2008; Pena-Abaurrea et al., 2009).

A. H B.

Figure 1.3. Structure of A) 6-OH-2,2’,4,4’-tetrabromodiphenyl ether (OH-BDE47) and B) 2’-MeO-2,3’,4,5’-tetrabromodiphenyl ether (MeO- BDE68)

OH-PBDEs are produced naturally, but there are also formed through the metabolism of anthropogenic PBDEs (Marsh et al., 2004b; Malmberg et al., 2005). Studies show that metabolised PBDEs tend to have the OH-group in meta- or para-position to the diphenyl ether bond (Marsh et al., 2004b; Malmberg et al., 2005) whereas naturally produced OH-PBDEs tend to be substituted with OH in ortho-position. OH-PBDEs with meta- and para-substitution have been found to dominate in human serum at sites known to be heavily contaminated with PBDEs (Athanasiadou et al., 2008) whereas ortho-substitution dominates for OH-PBDEs and MeO-PBDEs found in the ambient environment (Marsh et al., 2004a; Malmvärn et al., 2008). The

15 biogenic synthesis of MeO-PBDEs is not clear but bacterial O- methylation of phenolic compounds (Allard et al., 1987) presents the possibility that OH-PBDEs are precursors to the MeO-PBDEs. Based on the substitution pattern it is thus likely that the OH- and MeO-PBDEs found in the environment predominantly are of natural origin. This theory has been strengthened by measurements of a contemporary 14C signal of two MeO-PBDEs isolated from whale blubber from the western North Atlantic (Teuten et al., 2005) and by their presence in pre-industrial whale oil (Teuten et al., 2007).

PBDDs and Br/Cl-DDs Polybrominated dibenzo-p-dioxins (PBDDs) and furans (PBDFs) (figure 1.4) are formed during combustion of materials containing PBDEs (Buser, 1986; Thoma et al., 1987; World Health Organization, 1998). If chlorine is present during the combustion process, mixed brominated/chlorinated dibenzo-p- dioxins (Br/Cl-DDs) can also be formed (Weber et al., 2003). These compounds have been found in combustion processes (Sovocool et al., 1988; Sovocool et al., 1989) and in sediments from an industrial area in Japan (Takahashi et al., 2006). The few available studies on the toxicity of PBDDs and Br/Cl- DDs mainly focus on the binding affinity to the aryl hydrocarbon receptor. Their results indicate a similar response as for the more thoroughly studied PCDDs (Birnbaum et al., 2003; Van den Berg et al., 2006; Olsman et al., 2007).

A. B.

Figure 1.4. Examples of A.) a PBDD and B.) a PBDF

High concentrations of mainly tribromodibenzo-p-dioxins (triBDDs) have been reported from blue mussels (Malmvärn et al., 2005; Haglund et al., 2007), alga (Malmvärn et al., 2008) and sponge (Unger et al., 2009) from the Baltic Sea. Lower concentrations have also been reported from several species of fish and shellfish from the Baltic Sea and Swedish west coast (Haglund et al., 2007). Although no natural sources have been identified for these compounds, the congener pattern and lack of clear anthropogenic source have led to the conclusion that these PBDDs probably are of natural origin.

1.2 Natural production of brominated organic compounds Simply described, the natural synthesis of BOCs consists of two steps; first the hydrocarbon skeleton is formed and thereafter bromination occurs (Nielson, 2003) Larger and more complex natural BOCs can also be formed from smaller BOCs.

Formation of the hydrocarbon skeleton There are three main biosynthesis pathways for organic compounds starting with mevalonic, acetic and shikimic acid, respectively (Figure 1.5). Generally, mevalonic acid is the basis for terpenes and steroids, shikimic acid for aromatic compounds and acetic acid for fatty acids and some aromatic compounds (Nielson, 2003). Acetic acid and shikimic acid also combine with simple nitrogen containing compounds to form amino acids.

C COOH A OH O B O

OH OH HO OH OH OH

Figure 1.5. The three starting points for biosynthesis of hydrocarbon compounds. A) Mevalonic acid, B) Acetic acid and C) Shikimic acid.

Halogenation Although chlorine is about 500 times more abundant than bromine in sea water, the naturally produced BOCs are almost as abundant as the natural organochlorine compounds (Gribble et al., 1999). The reason for this is probably that bromine is the more readily oxidized species in the enzymatic halogenation process. Haloperoxidases, enzymes that oxidize halogens through reactions with hydrogen peroxide, are common in marine organisms (Butler et al., 1993; Moore et al., 1996). The general formula for the haloperoxidase reaction is (Nielson, 2003):

+ − 푂푟푔퐻 + 퐻 + 퐻2푂2 + 퐵푟 → 푂푟푔퐵푟 + 2 퐻2푂

The bromination process is a radical reaction with several steps involving the relevant enzyme. The actual bromination of the . organic compound is with the oxidized electrophilic Br+ or Br - formed from reaction of Br with H2O2. The presence of oxidized Br in the halogenation process can be seen in the common bromination pattern in natural products, where Br is found in positions where electron density is highest in a radical reaction such as ortho- and para-positions in phenolic compounds (Nielson, 2003).

Formation of larger structures Larger brominated structures, such as OH-PBDEs and OH-PBBs, can be formed from the brominated phenols (Figure 1.5).

. H .

H H H

Figure 1.5. Formation of OH-PBDE and OH-PBB from brominated phenols by an enzymatically catalysed free radical reaction. The figure is modified from schemes presented by Nielson (2003).

2. Bromine isotopes

Bromine has two naturally occurring stable isotopes, 79Br (50.7%) and 81Br (49.3%). The isotopic composition of Br is usually expressed as the isotope ratio (R) of the heavier-over- lighter isotope, i.e. R= 81Br/79Br. This is used to define a δ-value which is the per mil deviation of the R in a sample compared to R in a reference standard according to equation 1.

푅 − 푅 훿81퐵푟 = 푆푎푚푝푙푒 푅푒푓푒푟푒푛푐푒 × 1000 푅푟푒푓푒푟푒푛푐푒 Equation 1.

The reference standard for stable Br isotopes is the naturally occurring isotope ratio in seawater. This ratio is constant since the residence time for Br in the ocean (Broecker and Peng, 1982) is substantially longer than the time for global ocean mixing (130 million years vs. 1500 years) (Brownlow, 1996).

2.1 Kinetic isotope effect The mass difference between the two bromine isotopes, 79Br and 81Br, induce a slight difference in bond energy for the heavier and the lighter isotope. Thus, the chemical bond strength is isotope specific. The heavier isotope forms a slightly shorter and stronger bond that requires somewhat higher energy to break, causing a slower reaction rate during the scission of C-Br bonds during dehalogenation reactions. This leads to a kinetic isotope effect (KIE), which is the difference in reaction rates for the two isotopes.

2.2 Isotope fractionation processes Due to the KIE, formation and degradation of BOCs might result in bromine isotope fractionation, leading to different bromine isotopic ratios in different compounds. Figure 2.1 shows an overview of organobromine compounds in the marine

20 environment and illustrate processes that may affect their isotopic composition.

Photo degradation Anthropogenic production

Metabolism Natural production Bacterial degradation

Figure 2.1. Overview of the processes that affect BOCs in the environment. Due to the KIE these processes may change the isotopic composition of the BOCs.

A KIE during the synthesis of halogenated compounds may create a specific isotope composition for different synthesis paths. The δ37Cl for anthropogenic abiotically synthesized organochlorine compounds has been shown to range between -5.10‰ and +1.22‰ (Drenzek et al., 2002). In contrast, a study with haloperoxidase showed δ37Cl values of -11‰ and -12‰ for the enzymatic chlorination of organic compounds (Reddy et al., 2002). This difference in δ37Cl potentially allows for differentiation of natural and anthropogenic compounds based on their chlorine isotopic composition. A similar KIE might be observed for BOCs but, due to the smaller relative mass difference, the range in isotopic composition is expected to be smaller for the heavier bromine atoms than for chlorine. One estimate suggests that the difference in KIE for enzymatic and abiotic bromination would yield a δ81Br difference of 4‰ (Sylva et al., 2007). Based on the results of stable chlorine isotope analysis, one of the aims for this thesis was to develop a compound specific isotope analysis (CSIA) method for Br to investigate the possibility of 21 using δ81Br to distinguish between natural (presumably enzymatical bromination) and anthropogenic (presumably abiotic bromination) sources of BOCs in the environment. Since the 79Br-carbon bond is slightly easier to break than the 81Br-carbon bond, the dehalogenation processes that occur in nature may cause isotope fractionation. The isotope fractionation should result in an enrichment of 81Br in the remaining non-degraded fraction compared to the starting material. Determination of the Br isotopic composition of environmental contaminants might thus tell something of the degree of degradation of the organobromine contaminants found in nature. This approach is similar to studies of stable chlorine isotopes, which have been used to assess the extent of degradation of DDT in the Baltic Sea (Holmstrand et al., 2007).

3. Samples and purification methods

In this thesis, BOCs have been studied in four aquatic species An overview of the sample treatment is given in figure 3.1.

Sample extraction (organic solvent)

Lipid removal (Sulfuric acid)

Chromatographic separation

Identification / quantification

GC-MS NMR GC-MC-ICP-MS

Figure 3.1. Schematic presentation of the sample treatment leading to identification, quantification and bromine isotope analysis of BOCs in this thesis.

3.1 Samples A common freshwater sponge (Ephydatia fluviatilis) and blue mussels (Mytilus edulis) were sampled at the same site in the south-western Baltic Sea and analysed for PBDDs (Paper I). The brackish water of the Baltic Sea enables freshwater and marine species to coexist, albeit under some stress due to the too high or too low salinity, respectively. Blubber from a northern bottlenose whale (Hyperoodon ampullatus) was used to isolate several µg of lipophilic BOCs (Papers II and III).This species dwells in the northern Atlantic

23 and the young male in this study was found dead on the Swedish west coast. In Paper IV, dibromophenol isolated from a marine benthic worm (Saccoglossus bromophenolosus) was analysed.

3.2 Extraction and lipid removal methods Extraction and purification of target compounds are prerequisites for accurate quantification and identification. By removing the sample matrix the limits of detection and quantification are lowered.

Small scale sample extraction (Jensen method) The Jensen extraction method (Jensen et al., 1983) was used to extract halogenated dibenzo-p-dioxins from sponge and blue mussel samples (Paper I). This method has been developed to quantitatively extract the lipids, and hence lipophilic compounds, from a complex sample matrix. Briefly, the samples are homogenized in acetone/n-hexane followed by extraction with n-hexane/diethyl ether. Partitioning of the organic solvent phase with a slightly acidic 0.9 % NaCl water phase removes water and water soluble compounds from the extract. The Jensen method is a standard method for extraction of organohalogen compounds, including BOCs, in our laboratory and has previously been used successfully in the extraction of PBDDs. Other possible extraction methods are Soxhlet extraction or accelerated solvent extraction (ASE). However, these methods use heat and/or pressure which could cause debromination of the target BOCs or cyclisation of OH-PBDEs in the sample, forming PBDDs. It is possible that either of these methods could be suitable, though neither have been evaluated for quantitative extraction of PBDDs from biological material.

Large scale lipid extraction (Wallenberg perforator) A sample of 10 kg whale blubber tissue was used to isolate several µg of BOCs (Paper II). Blubber tissue consists of about 95% liquid lipids that can be isolated from the blubber tissue by grinding followed by centrifuging. The lipids are thus separated from the rest of the tissue. The 10 kg whale blubber sample was

24 reduced to approximately 9.5 kg pure lipids by this initial treatment. A custom manufactured Wallenberg perforator as designed by professor Sören Jensen (Jensen et al., 1992) was used to partition the lipophilic compounds from the lipids to an acetonitrile (ACN) phase (Figure 3.2). Aromatic compounds partition into the ACN phase through interactions of their π- electrons with the nitrile group.

Cooling

Collector bottle

Heating

Liquid lipids

Figure 3.2. The Wallenberg perforator (Jensen et al., 1992). Arrows indicate the flow direction for ACN from the cooler through the liquid lipids and back to the heated collector bottle.

The Wallenberg perforator uses refluxing ACN, allowing it to pass repeatedly through the lipids. Briefly, ACN is heated in the collector bottle (Figure 3.2) causing the ACN vapour to rise into the cooler. Liquid ACN flows through a system of glass tubes from the cooler down to the bottom of a flat-bottomed flask

25 filled with liquid lipids. Continuous stirring cause the ACN to spiral through the lipids as it, due to the lower density of ACN compared to the lipids, rise to the top of the flask. The halogenated compounds are extracted with the ACN during this process and are transported back to the collector bottle. The ACN continues to cycle through the system, while the BOCs, having higher boiling points, accumulate in the collector bottle. The extraction is stopped when the amount of ACN having passed through the lipids is the equivalent of 30 times the lipid volume to ensure that more than 95% of the BOCs are extracted. Although some lipids are also dissolved in the ACN and follow the analytes to the collector bottle, the method efficiently removes about 90% of the lipids using a relatively small amount of solvent. A second extraction round remove an additional 70% of the remaining lipids.

Removal of lipid residues Removal of lipid residues was done with sulfuric acid in both Papers I and II. This method was not considered for the entire large blubber lipid sample because of the enormous volume of sulfuric acid that would have been required. Sulfuric acid treatment is a destructive method and many compounds react with the acid and are destroyed or partition into the sulfuric acid phase because of the Lewis acid properties. An example of a non-destructive method for the removal of the lipid residues is gel permeation chromatography which separates the analytes from the lipid molecules by size. This method is time consuming, and was deemed unnecessary since all target compounds were known to survive the sulfuric acid treatment.

3.3 Chromatographic separation Purification by class separation reduces the risk of compounds co-eluting during analysis, which is a prerequisite for both proper identification and quantification.

Column chromatography Before instrumental analysis, the analytes were fractionated according to polarity on a silica gel column. This removed the ubiquitous non-polar PCBs from the slightly polar target MeO- PBDEs and PBDDs, which were also separated into different fractions. The more polar OH-PBDEs are retained on the column and thus removed from the sample. This minimizes the risk of PBDDs forming from OH-PBDEs in the gas chromatography (GC) injector, GC column or in the mass spectrometric (MS) ion source during analysis. In Paper I the PBDD fraction was further purified on a column prepared with active carbon dispersed on Celite (Haglund et al., 2007), separating compounds based on their planarity to isolate the planar dibenzo-p-dioxins and dioxin-like compounds from unplanar substances.

Preparative capillary gas chromatography The NMR method used for identification of an unknown BOC (Paper II) required a much larger amount of pure compound than normally used in analytical work with environmental contaminants. Preparative capillary gas chromatography (PCGC) (Figure 3.3) is a method that can be used to isolate large amounts of single compounds from a sample extract. Different versions of the technique have been used since the 1970s (e.g., (Badings et al., 1975; Roeraade et al., 1986). Environmental applications of the PCGC technique in more recent years include the isolation of single organic compounds for offline CSIA of δ14C (Eglinton et al., 1996; Eglinton et al., 1997; Mandalakis et al., 2004; Teuten et al., 2005; Kumata et al., 2006; Zencak et al., 2007; Sheesley et al., 2009) and δ37Cl (Holmstrand et al., 2006a; Holmstrand et al., 2006b; Holmstrand et al., 2007). The isolation of environmentally occurring BOCs for structural elucidation with NMR (Teuten et al., 2006a, Unger et al., 2010

(Paper II)) and of organic contaminants in ground water for test of genotoxicity (Meinert et al., 2010) have also reported.

Autosampler

Detector Fraction collector CIS 1% 99% Effluent splitter

Cooling

22 24 26 28 Megabore column 22 24 26 28

Figure 3.3. Overview of the PCGC process with schematic picture of the PCGC instrument. A whale blubber sample was injected and a single BOC was isolated in 99% purity. Mass chromatograms were scanned between mass-to-charge ratios (m/z) 50-700.

The PCGC technique was optimized for semi-volatile compounds in environmental samples by Mandalakis et al. (2003). A sample is repeatedly injected into the GC where it is separated and the individual compounds are directed into different cooled quartz traps at the GC outlet. A cooled injection system (CIS) and a megabore column are used to enable larger sample amounts and minimize the number of injections needed. The retention times are carefully monitored and the trap times are adjusted to correlate with changes in retention time over the sampling period of normally a few days. This results in sample purities of >95%.

4. Analysis

Three different instrumental analysis methods have been used for identification (Papers I and II), quantification (Paper I) and bromine isotope analysis (Papers III and IV) of BOCs in this thesis. These analysis techniques are briefly described below.

4.1 GC-MS The most common method for the analysis of environmental contaminants is gas chromatography–mass spectrometry (GC- MS). The method is also applied for the analysis of naturally produced BOCs. Identification of compounds with GC-MS is most reliable when authentic reference standards are used to compare both the GC retention times and mass spectra of the target compounds. Many novel compounds in environmental samples still lack standards, limiting identification to the interpretation of mass spectra fragmentation patterns. In this context, high-resolution MS is particularly useful for determining the accurate mass of the analysed compounds. Based on accurate mass, it is possible to calculate the empirical formula of the analysed compounds. This technique was used for identification of Br/Cl-DDs (Paper I) and as part of the identification of 6-methoxy-5-methyl-2,2’3,4’-tetrabromo diphenyl ether (MeOMe-BDE42) (Paper II). Two MS ionization techniques have been used in the analysis of BOCs in this thesis. Electron ionization (EI) is a reproducible technique that generally provides detailed structural information through characteristic fragmentation patterns for different groups of compounds. Electron capture negative ionization (ECNI) is a softer ionization method where a buffer gas, in this case ammonia, is bombarded with electrons to create low energy thermal electrons. These are captured by the analytes and negative ions are formed. A high temperature in the ion source typically causes BOCs to form mainly Br- ions, resulting in a high concentration of Br- in the detector. Thus, by using selective ion monitoring (SIM) to detect only the Br-, this technique is highly sensitive and permits quantification of small amounts of compounds. This method was used for

29 quantification of MeO-PBDEs (Paper I) and for estimating the concentration of MeOMe-BDE42 (Paper II). The PBDDs and Br/Cl-DDs in Paper I were quantified with high resolution MS using EI and SIM of appropriate ions.

4.2 NMR Nuclear magnetic resonance spectroscopy (NMR) is a routine method for structural elucidations in organic chemistry. However, since NMR requires nearly mg amounts of pure compound the technique is rarely used in analytical environmental chemistry. The 1H NMR spectra provide information on the position of all protons in a molecule, both in terms of what kind of carbon structure they are attached to (aliphatic, aromatic, alcohol etcetera) and where they are located in relation to other substituents in the molecule, as the chemical shift of each proton is affected by its surroundings. Protons attached to adjacent carbons also affect each other’s signal, resulting in signal splitting (Figure 4.1). The amount of sample required for 13C NMR is larger than for 1H NMR because of differences in the intrinsic magnetic properties of the 1H and 13C nuclei, and the low natural abundance of 13C (1.1% of total carbon). When the sample amount is too small for regular 13C NMR, some 13C NMR shifts may be obtained by using heteronuclear single quantum coherence (HSQC). This technique is based on 1H-13C interactions and gives 13C shifts for carbons directly attached to protons by showing the cross-peaks in a two-dimensional spectrum even though the intensity of the 13C spectrum is too low to be seen. Figure 4.1 shows the HSQC NMR spectrum used for identification of MeMeO-BDE42 in Paper II.

- shift

80 (ppm)

160 7.5 5.0 2.5 1 H- shift (ppm)

Figure 4.1. Two-dimensional HSQC NMR spectrum of MeMeO-BDE42. 1H-spectrum is shown on top with enlargement of the area between 6.2 - 7.8 ppm to show signal splitting. 13C-shifts are shown on the y- axis.

4.3 GC-multiple-collector-ICP-MS Inductively coupled plasma (ICP) is an efficient ion source that in connection with a MS equipped with multiple ion beam collectors can be used to determine isotope ratios for a wide range of elements. The plasma is created by Ar gas in an electromagnetic field generated by a radio frequency (RF) current in the RF-coil (Figure 4.2). Inorganic or polar organic compounds in aqueous solution are aspirated into the plasma and ionized. The ions are transferred into the MS, where they are separated based on their mass-to- charge ratio and detected. The continuous flow of sample to the plasma yields a continuous ion signal to the detector. The signal can be scanned repeatedly permitting a large number of measurements of the isotope ratio.

RF-coil MS inlet cones

Sample

Auxillary gas (Ar) Plasma gas (Ar)

Manual injection

Transfer line

Hexapole Mass collision cell analyser ICP torch Multiple collector Faraday cups Ar

Megabore column Ambient pressure High vacuum

Figure 4.2. Schematic drawing of the GC-MC-ICP-MS instrument used for bromine isotope measurements of BOCs in Papers III and IV.

Besides aqueous solutions, it is possible to use other sample introductory systems, including GC. The direct connection of a GC to the plasma and MS (Figure. 4.2) was explored in Papers III and IV. This on-line system simplifies the purification process needed before analysis by separating and directly introducing a specific compound from a complex sample matrix into the ICP ion source. The sample is manually injected into the megabore coloumn of the GC where the compounds are separated, allowing single compounds to elute through the heated transferline into the plasma where they are ionized. In contrast to an aqueous solution, which is continuously aspired into the ICP ion source, each compound generates a transient signal as they elute from the column. The peak width is only a few seconds and probably not isotopically homogenous (Krupp et al., 2005; Holmstrand et al., 2006b). The entire signal thus needs to be integrated in order to obtain the isotope ratio of the compound.

5. Summary of papers

5.1 Paper I - Identification and quantification of PBDDs and Br/Cl-DDs In Paper I the common fresh water sponge Ephydatia fluviatilis was collected from Borgholm harbour in the south-western Baltic Sea and analysed for halogenated dibenzo-p-dioxins. High concentrations of PBDDs, especially 1,3,7- and 1,3,8-TriBDD, were found. As there are no known anthropogenic sources for PBDDs in the Baltic area and the bromine substitution pattern of the detected PBDDs are in agreement with what could be expected for naturally produced compounds, these findings strengthen the idea that PBDDs are produced naturally in the Baltic Sea. Mixed brominated/chlorinated dioxins were also found in the sponge. To our knowledge this was the first time Br/Cl-DDs were reported from a background environment with no apparent anthropogenic sources. This led to the conclusion that Br/Cl- DDs are also probably natural products in this region. It is worth noting that the concentrations of Br/Cl-DDs, though lower than PBDDs, were higher than PCDDs, although PCDD concentrations are considered to be a problem in the Baltic Sea.

A. B. C.

D. E.

Figure 5.1. Examples of the compounds identified in Paper I A.) 1,3,7- triBDD, B.) 1,3,8-triBDD, C.) 1,2,4,7-tetraBDD, D.) a Br2Cl-DD and E.) a Br3Cl-DD.

5.2 Paper II - NMR-identification of a novel brominated organic compound isolated from whale blubber In Paper II an unknown BOC, present in large amounts in northern bottlenose whale (Hyperoodon ampullatus) blubber, was identified. The combination of GC retention times, mass spectra and NMR enabled us to determine the structure of a PBDE with both MeO- and Me-substitution (Figure 5.2). The use of NMR for identification of unknown compounds is unusual in analytical environmental chemistry as it requires several µg of pure compound. In this case it was possible to isolate about 100 µg of pure compound by using a PCGC. To our knowledge this was only the second time PCGC was used to isolate compounds for identification with NMR.

Figure 5.2. MeOMe-BDE42 isolated from whale blubber and identified in Paper II.

5.3 Paper III - Development of a GC-MC-ICP-MS method for brominated diaromatic compounds The bromine isotopic composition of BOCs found in the environment could potentially reveal information of the origin and fate of these compounds. Paper III describes the first attempts to develop a method for CSIA of bromine in PBDEs and MeO-PBDEs with an on-line GC-MC-ICP-MS system. Three different transferlines were tested and evaluated for their efficiency in conducting the analytes from the GC to the ICP. Standard bracketing with monobromobenzene (MBB) with known δ81Br was used to calculate δ81Br for three PBDEs, 2,2’,4,4’-tetrabromodiphenyl ether (BDE-47), 2,2’4,4’,5- pentabromodiphenyl ether (BDE-99) and 2,2’4,4’,6- pentabromodiphenyl ether (BDE-100) (figure 5.3). An attempt was also made to compare the δ81Br of the structurally similar BDE-47 and 6-methoxy-2,2’,4,4’-tetrabromo diphenyl ether (MeO-BDE47).

A. B. C.

Figure 5.3. The three PBDEs analysed in Paper III. A) BDE-47 B) BDE- 99 and C) BDE-100.

5.4 Paper IV - Measurements of δ81Br in anthropogenic and natural monoaromatic compounds In Paper IV δ81Br values were determined for four industrially produced monoaromatic brominated compounds and a brominated naphtalene (Figure 5.4) using a method developed in Paper III. This was done to test the variation in bromine isotopic composition among different industrial products. A naturally produced compound, isolated from the marine worm Saccoglossus bromophenolosus and structurally identical to one of the anthropogenic compounds, was also analysed. No statistical difference could be seen between the anthropogenic and the natural product. More research is needed before any final conclusions can be drawn in this area.

A. B. H C. H D.

E. H F. G.

Figure 5.4. BOCs analysed in Paper IV. A) 2,4-dibromoanisole, B) 2,4- dibromophenol, C) 4-bromophenol, D) 1,4-dibromophenol, E) 2,4,6- tribromophenol, F) 1,3,5-tribromobenzene and G) 1-bromonaphtalene.

6. Results and Discussion

The dual aims of this thesis were the identification of natural BOCs and the development of a method for compound specific bromine isotope analysis. The following section is thus split in two, discussing first the identification of BOCs (Papers I and II) and thereafter bromine isotope analysis (Papers III and IV).

6.1 Identification of natural brominated organic compounds Since the first report of PBDDs in Baltic Sea biota was published in 2005 (Malmvärn et al., 2005) additional studies have been published (Haglund et al., 2007; Malmvärn et al., 2008; Unger et al., 2009 (Paper I); Haglund, 2010) attempting to identify a natural source and the biosynthetic pathways of these compounds. There are several lines of evidence that these compounds are of natural origin. There is no known commercial or other intentional production of PBDDs. There are however unintentional anthropogenic formation of PBDFs and PBDDs via combustion of bromine containing material e.g. PBDEs (Thoma et al., 1987). In this context, PBDFs are present in far higher concentrations than PBDDs. In analysed matrices from the Baltic Sea, the levels of PBDDs greatly exceeded the concentrations of PBDFs (Haglund et al., 2007; Malmvärn et al., 2008; Unger et al., 2009 (Paper I); Haglund, 2010). A general pollution of PBDDs due to combustion ought to also result in similar concentrations in lakes as in the Baltic Sea, which is not reported (Haglund et al., 2007). The congener pattern that is seen in the Baltic Sea samples, where triBDD is clearly dominant, also indicate natural production since combustion of PBDEs yield a larger variety of PBDD congeners. The concentrations of triBDDs found in the sponge (Ephydatia fluviatilis) were similar to the levels reported in blue mussels. Both sponges and blue mussels are filter feeders and their respective PBDD concentrations probably reflect a similar uptake process. This indicates that sponges are not major producers of PBDDs in the Baltic Sea.

Several single congener PBDDs were identified and quantified in ambient biota for the first time as reported in Paper I. Identification and quantification of individual PBDD congeners is a step towards a better understanding of the biosynthesis of these compounds. It has been shown that the most abundant PBDDs found in Baltic Sea biota can be formed through either enzymatic coupling of abundant bromophenols, condensations of common OH-PBDEs or, for some triBDDs, debromination of higher brominated dibenzo-p-dioxins (Haglund, 2010). Identification and quantification of single congeners is also a base for future risk assessment since different congeners most certainly have different toxic effects. A less toxic compound can have a larger effect than a more toxic compound if the concentration is high enough. The triBDDs found in high concentration in Baltic Sea biota were present at 25 000 times higher concentration than the toxic 2,3,7,8-tetrachloro dibenzo- p-dioxin (TCDD) in the sponge (Paper I). Several other PBDD and Br/Cl-DD congeners were also more abundant that the PCDD congeners the sponge (Paper I). The triBDDs appear to be relatively rapidly excreted by fish (Haglund et al., 2010) and are thus less likely to cause toxic effects, but more work is required to confirm this speculation. The findings of Br/Cl-DDs in sponge (Paper I) was the first time these compounds were reported in a background environment. Previous reports have been from sites associated with combustion (Sovocool et al., 1988; Sovocool et al., 1989; Weber et al., 2003; Haglund et al., 2007). Accordingly, the very high levels of these compounds in the sponge, by far exceeding those of PCDDs, are a strong indication a natural production of Br/Cl-DDs in the Baltic Sea. It seems clear that PBDDs and even mixed Br/Cl-DDs are produced naturally in concentrations that exceed the anthropogenic input of these compounds to the Baltic Sea. Since PBDDs in the Baltic Sea are likely formed from OH-PBDEs or brominated phenols, the question of natural or anthropogenic origin should perhaps be expanded to also include the origin of these precursor compounds.

The use of NMR vs. GC-MS for identification of BOCs in the environment The use of NMR for identification of bioaccumulated BOCs is limited by the requirement of near mg amounts of pure compound for analysis. Its main use for environmentally occurring BOCs has thus been for natural products isolated from the producing organisms (Unson et al., 1994; Cameron et al., 2000; Takahashi et al., 2002; Utkina et al., 2002). Although NMR gives more information on the chemical structure than GC-MS, the ability of GC-MS to detect compounds in pg concentrations in sample extracts (as for PBDDs in Paper I) makes it a more useful method for most analytical environmental chemistry purposes. The different requirements for sample amount and purity make it possible to analyse samples with GC-MS that are impossible with NMR, due to the presence of other compounds in the sample or the small amounts of analyte available. Isolating a single compound from a more complex environmental extract using a PCGC, alleviates the problem with pure compounds for NMR analysis. To isolate up to mg amounts of an analyte for NMR, either a sample with very high concentrations of the compound of interest or a very large sample (as in Paper II) is needed. Paper II describes the identification of a novel BOC with a combination of NMR and GC-MS. A diMeO-PBB with the same molecular weight as the unknown compound has been reported (Marsh et al., 2005; Teuten et al., 2007; Vetter et al., 2008). The mass spectra of our compound indicated a MeOMe-BDE structure, as opposed to a diMeO-PBB structure, and the 1H NMR spectra unequivocally showed the presence of one methoxyl group and one methyl group, both placed on an aromatic ring. The sample amount was too low for 13C NMR. Still, with the somewhat limited information given by 1H NMR and HSQC, it was possible to limit the possibilities to three structures. One of these compounds were then synthesised and the final identification of MeOMe-BDE42 was with GC retention times.

6.2 Development of CSIA-Br A range of isotope systems are used for studies of transport and fate of compounds and elements in the environment. This includes the ratios of 15N/14N to study food webs (Broman et al., 1992; Tittlemier et al., 2002), 14C/12C to separate between biogenic and fossil fuel combustion products (Mandalakis et al., 2004; Sheesley et al., 2009) and to determine the natural or anthropogenic origin of BOCs in the environment (Teuten et al., 2005), and 13C/12C for studies of degradation of organic pollutants (Schmidt et al., 2004; Zwank et al., 2005; Holmstrand et al., 2007). Similarly, δ37Cl has been used to study the degradation of DDT in the Baltic Sea (Holmstrand et al., 2007) and for source apportionment of high levels of PCDDs in clay (Holmstrand et al., 2006a). The possibility that accurate measurements of the bromine isotopic composition of BOCs in the environment may be used for determining the sources and fates of these compounds was the reason for the method development described in Papers III and IV.

Standard bracketing Isotope ratio determination using MC-ICP-MS suffers from variations in the absolute isotope ratios due to instrument drift and mass bias. Corrections for these effects can be obtained with different techniques, including the standard-sample- bracketing method. This method relies on a substance with known isotopic composition that is analysed between the samples. Here, MBB with a known isotope ratio was co-injected with all samples and used as the isotopic anchor point (Papers III and IV). A linear drift in the measured isotope ratios was assumed between the two MBB measurements immediately preceding and following the sample peaks. A theoretical 81Br/79Br ratio for MBB was calculated for the time of elution for the PBDEs. This value used to calculate δ81Br values for the PBDEs (Paper III). The difference in δ81Br for the three PBDEs remained constant within all measurement series, although the absolute values varied between series. MBB and PBDEs are two different types of compounds however, and it is likely that the ionization yields in the plasma differ. This could enhance any

39 variation caused by slight differences in the physical setup of the instrument before each measurement campaign and explain the larger difference between data series than within each measurement series This idea was tested with measurements of structurally similar MeO-PBDE and PBDE isolated from whale blubber (Paper III). BDE-47 was used as a standard for MeO-BDE47 and the precision between two measurement series was similar to that previously found within one series of PBDEs. This result supported the idea that a structurally similar reference standard gives a better precision. This was further explored with measurements of four anthropogenic monocyclic aromatic BOCs and a brominated naphthalene using MBB as reference standard (Paper IV). All of these compounds are structurally similar to MBB and have similar GC retention times, which place the measurement of the standard closer in time to the analyte than for PBDEs. The precision for these measurements was 0.8‰. The internal instrument drift and the compound-specific ionization yields apparently call for a reference standard with similar ionization efficiency in the plasma that can be measured close in time to the analyte.

Transfer lines The higher boiling points for PBDEs compared to monocyclic BOCs require a transfer line between the GC and the ICP ion source with enough heating capacity to keep the analytes from condensing. Thus the transfer line needed to be heated all the way from the GC to the plasma with no cold spots on the way. Three different transfer lines were tested and evaluated. A detailed schematic drawing of all three transfer lines is presented in Paper III (Figure 1 in Paper III). The first transfer line (T1 in Paper III) was electrically heated between the GC and the quartz torch of the ICP ion source. It was assumed that the heated gas from the GC and the heat from the plasma would be sufficient to carry the PBDEs through the torch to the plasma. The results showed that the standard MBB passed through but no PBDE signals could be detected. A version of this transfer line was however used during later analysis of monocyclic aromatic BOCs (Paper IV). 40

The second transfer line (T2 in Paper III) consisted of a thin stainless steel capillary that was fed into the torch and electrically heated through its entire length. The steel capillary ended 20 mm from the plasma, any further and the intense heat from the plasma would have melted the steel. Although this system gave good response for PBDE the precision between measurement series was lower than expected and the precision for MBB went from 0.4‰ for T1 to 2‰. We hypothesized that this was an effect of electrostatic interference from the electric heating of the transfer line causing instability in the plasma. The third transfer line (T3 in Paper III) was constructed to keep the sample constantly heated all the way from the GC to the plasma while eliminating the risk of electrostatic effects. Instead of heating the transfer line itself, the Ar gas carrying the sample into the plasma was heated in the GC oven. The sensitivities for both MBB and PBDEs were the same as for T2 but unfortunately the precision did not improve. The variations in isotope ratios might be caused by the heating of the gas affecting the stability of the plasma.

Evaluation of δ81Br for enzymatic and anthropogenic bromination The δ81Br of both industrially synthesised and naturally produced dibromophenol were found to have a difference of 1.4±0.9‰ (Paper IV). The δ81Br for all the anthropogenic compounds in this study ranged from -4.3‰ to -0.45‰. The observed difference between enzymatic and anthropogenic bromination was therefore not statistically significant. The difference in δ81Br for presumably natural MeO-BDE47 and anthropogenic BDE-47, both isolated from whale blubber (Paper III), was even lower at 0.3±0.7‰. With the existing methods and results it is not possible to discern a difference in the bromine isotopic composition for anthropogenic and biogenic BOCs.

7. Conclusions

 Based on their distribution in algae, blue mussels and sponge, it is possible that PBDDs in the Baltic Sea do not derive from a single producing organism, but rather reflect a general production process active in several organisms (Paper I).

 Mixed Br/Cl-DDs are produced in the Baltic Sea (Paper I).

 PCGC is a useful tool to facilitate the identification of unknown compounds in the environment. By allowing a large amount of pure compound to be isolated NMR analysis is enabled (Paper II).

 Reference standards with ionization yields similar to the analytes alleviate the problem of internal instrument drift during isotope measurements with GC-MC-ICP-MS (Papers III and IV).

 With the existing methods and results it is not possible to discern a difference in the bromine isotopic composition for anthropogenic and biogenic BOCs (Papers III and IV).

8. Future perspectives

Although there are strong indications that PBDDs and Br/Cl-DDs are naturally produced in the Baltic Sea the question is not fully resolved. Studies of PBDDs, Br/Cl-DDs, OH-PBDEs and brominated phenols in sediments from the pre-industrial era up to now could bring clarity to this issue. CSIA of 14C, or possibly Br isotopes, could be used for this purpose.

Information on the effects of PBDDs and mixed Br/Cl-DDs is still very limited. More studies are needed in order to evaluate the potential effects these compounds have on the environment.

NMR has a clear advantage over GC-MS for identification of unknown compounds although the requirement of near mg amounts of pure compound is daunting to an analytical environmental chemist. Large scale extraction methods such as the Wallenberg perforator and PCGC for isolation of the compound make NMR a possible choice for the identification of unknown organic compounds in the environment.

Source apportionment of BOCs is increasingly interesting as more are discovered in the environment. The attempts here to use bromine isotopes should be further investigated. It is necessary to repeat the experiment done for organochlorine compounds with chloroperoxidase also for BOCs, using a range of enzymes, to establish a natural production endpoint or range. Until there is a relatively simple, unquestionable way of separating anthropogenic and naturally produced compounds in the environment, the discussion of the relative importance of the different sources to the environment will continue.

9. Sammanfattning på svenska

Bromerade organiska föreningar (här förkortat BOCs, från engelskans brominated organic compounds) är vanliga i miljön. Många har kända källor och är antingen naturliga eller antropogena, dvs. industriellt tillverkade. För en del ämnen finns dock både naturliga och antropogena källor medan andra är av okänt ursprung. Den här avhandlingen syftar till att dels identifiera ett antal naturligt producerade BOCs och dels till att utveckla en metod för att mäta sammansättningen av bromisotoper i BOCs i miljön. Polybromerade dibenzo-p-dioxiner (PBDDs) i Östersjön har bedömts vara naturligt producerade, även om man ännu inte har lyckats identifiera någon specifik naturlig producent. Eftersom många BOCs i tropiska vatten är producerade av marina svampdjur, analyserades förekomsten av BOCs i en vanligt förekommande tvättsvampsart (Ephydatia fluviatilis) från Östersjön (Paper I). Resultaten visade att den här tvättsvampen troligen inte är någon betydande källa till PBDDs i Östersjön. Utöver PBDDs fann vi också dibenzo-p-dioxiner med både klor och brom i tvättsvampen. Detta var första gången sådana föreningar rapporterades från en bakgrundsmiljö, dvs. utan någon antropogen källa i närheten. Kärnmagnetisk resonansspektroskopi (NMR) används sällan i analytisk miljökemi eftersom metoden kräver en stor mängd ren substans. I Paper II användes en preparativ kapillärgaskromatograf (PCGC) för att isolera ca 100 µg av en tidigare oidentifierad BOC från späck från en nordlig näbbval (Hyperoodon ampullatus). Detta var en tillräcklig mängd för att kunna använda NMR för att identifiera föreningen. Bromisotopsammansättningen (δ81Br) i en BOC påverkas av olika processer och skulle därför kunna ge information om dess ursprung och öde i miljön. De första stegen mot en sådan applikation är utvecklingen av en metod för att bestämma bromisotopkvoten i de BOCs man finner i miljön (Paper III) samt att definiera var på δ81Br-skalan de antropogena respektive naturliga ämnena befinner sig. I Paper IV analyserades därför bromisotopkvoten i ett antal antropogena BOCs i ett första försök att definiera det antropogena området på skalan.

10. Acknowledgements

När jag började på ITM var jag 23 år och hade förlorat min mamma ett knappt år tidigare. Att säga att jag var ung och lite vilsen är nog inte helt fel. Under mina år här jag har inte bara lärt mig mer om laborerande, analyser och skrivande utan också blivit många erfarenheter rikare och utvecklats som person. Jag har blivit vuxen. Tack alla ni som varit med mig på den resan!

Först vill jag tacka mina handledare. Tack Örjan för din förmåga att entusiasmera och se möjligheter där jag ser svårigheter. Tack Lillemor för allt ditt stöd och för att du alltid tagit dig tid att lyssna på mig, vare sig det gällt arbete eller privatliv. Tack Per för att du alltid funnits där i bakgrunden, trevlig julsushi och för all hjälp under skrivandet av själva avhandlingen.

Ett stort tack också till Åke för stöd då jag behövt det och för alla kommentarer och tips under avhandlingsskrivandet.

Henry, du förtjänar egentligen ett helt eget kapitel! Din lugna personlighet och speciella humor kan göra vilken dålig dag som helst mycket bättre. Utan dig har jag svårt att se hur jag skulle gått i mål med det här, inte minst med tanke på att pek 3 och 4 hade varit oändligt mycket svårare (jag säger inte omöjligt, för inget är omöjligt!) att nå iland med utan dig.

Tack Karin, för att du tar hand om oss alla och för att du är så bra på att organisera de trevliga små påhitt som skapar sammanhållning.

Bodil, tack för alla tips och råd om extraktioner av jätteprover! Och tack för att du är en underbar människa som både lyssnar och säger rätt saker.

Tack till mina rumskompisar på gamla ITM. Marie och Cissi, som tog hand om mig när jag var ny. Tack Marie för alla långa både vetenskapliga och ovetenskapliga samtal. Nu skiter vi i det här och går hem!  Och tack Cissi för att du ibland kan prata nästan lika mycket som jag. Vi fick inte mycket gjort vissa förmiddagar inte!

Labbsällskapet Kerstin, Hanna, Jorien, Ulla E, Dennis och Anna M, tack för både tips och råd och för mer eller mindre urspårade (freestylade?) diskussioner under många och långa timmar på lab. Tack Yngve för din oerhörda kunskap om allt som rör masspektrometri och ditt intresse och din entusiasm för nya oidentifierade substanser!

Tusen tack Marsha för trevligt sällskap och för att du tagit bort säkert 90 % av alla however, also och thus i den här avhandlingen! 

All current and previous post docs in the MPG group, Manolis, Kuma, Bart, Laura, Rebecca (I seriously miss our morning coffees!), Daniel, Brett, Christoph and August. Thank you Zdenek for taking more time than you really had to help me with everything from handling the PCGC to structuring excel sheets.

De “nya” doktoranderna i MPG gruppen, Axel, Elena, Charline och Emma. Ni är ett härligt gäng! Thank you Elena for introducing me to the wondrous combination of condensed milk, lots of nuts and simple pie!

Tack till doktorandgänget på Miljökemi, Anna S, Linda, Johan F, Hans, Maria S och alla andra både nuvarande och tidigare doktorander för trevliga fester, seminarier, fikan och en och annan lunch. Tack Karin L, min räddare i nöden vid flera tillfällen, för hjälp med kiselgel, standarder, provletande, referenser och allt vad det må vara!

Tack Anita och Birgit (och Ulrika) för att ni alltid har haft tid när jag ringt eller kommit förbi och pratat en stund.

Tack till mina medförfattare för långa och givande diskussioner i samband med peken. Tack Göran för att du visade mig vilken information det faktiskt går att få fram ur ett NMR spektrum och tack Peter för snabba svar om allt som rör dioxiner.

Tack också till alla andra nuvarande och tidigare kollegor på ITM som hjälpt till att göra min tid som doktorand så bra som den varit. Det känns lite sorgligt att lämna er nu men jag tror att jag är redo att möta världen utanför universitetet!

Även livet utanför jobbet påverkas av de fram- och, i mitt fall frekventa, motgångar man drabbas av på jobbet och det är vännerna som får ta smällen. Så, tack hela gänget för att ni, trots att ni egentligen inte fattat vad det är jag pysslar med (hmm, typ kemi va?) ändå snällt lyssnat på mig när jag beklagat mig över trasslande instrument och svårhanterade prover.

Nahal, tack för att du finns. Du gjorde mitt liv lättare när det var svårt.

Tack Lisa för alla luncher, med eller utan promenad. Hur hade det gått för oss om vi inte hade haft vår timme i veckan att ösa ur oss allt för någon som faktiskt förstår vad det är man pratar om?

Ett tack också till mina vänner i Hudik, jag saknar er alla! Cia och Tommy (och barnen) för sköna slöa kvällar, med eller utan bastu(!) och för en fullkomligt underbar skidsemester! Gerald och Katarina för härliga fikastunder och roliga samtal. Jag kommer snart och hälsar på igen! Tack Martin, för att du visade mig ett annat liv.

Tack familjen Nilsson för att ni omedelbart släppte in mig i er gemenskap. Jag ser fram emot många fler besök i Motala och i skärgården!

Jonas, tack för att du hittade mig! Tänk vad lite kyckling kan göra…

Till slut kommer jag till de som nog egentligen är viktigast. Ett stort tack till hela min underbara familj! Jag inser att jag är oerhört lyckligt lottad som har en familj som förstår vad jag håller på med.  Mina systrar Anna och Jenny och deras respektive Joav och Pascal. Världens bästa syskonbarn, Adrian och Elsa! Tack pappa för allt. Och mamma, älskade mamma! Jag önskar att du kunde varit här. Vi saknar dig, nu och för alltid.

Wow, tre sidor…. Nåväl, hur många känner ni som pratar mer än jag? 

This thesis was financially supported by MISTRA Idéstöd, the Swedish Research Council FORMAS and the Stockholm marine research Center (SMF)

11. References

Allard,A.S., Remberger,M., Neilson,A.H., Bacterial O-Methylation of Halogen-Substituted Phenols. Applied and Environmental Microbiology 1987. 53, 839-845.

Athanasiadou,M., Cuadra,S.N., Marsh,G., Bergman,Å., Jakobsson,K., Polybrominated diphenyl ethers (PBDEs) and bioaccumulative hydroxylated PBDE metabolites in young humans from Managua, Nicaragua. Environmental Health Perspectives 2008. 116, 400-408.

Badings,H.T., Vanderpol,J.J.G., Wassink,J.G., Wide-Bore Glass Capillary Columns for Gas-Chromatography - Their Preparation and Application. Chromatographia 1975. 8, 440-448.

Birnbaum,L.S., Staskal,D.F., Diliberto,J.J., Health effects of polybrominated dibenzo-p-dioxins (PBDDs) and dibenzofurans (PBDFs). Environment International 2003. 29, 855-860.

Broecker, W.S. and Peng, T.H., Tracers in the sea. 1982, Eldigio Press Lamont Doherty Geological Observatory

Broman,D., Näf,C., Rolff,C., Zebuhr,Y., Fry,B., Hobbie,J., Using Ratios of Stable Nitrogen Isotopes to Estimate Bioaccumulation and Flux of Polychlorinated Dibenzo-P-Dioxins (Pcdds) and Dibenzofurans (Pcdfs) in 2 Food-Chains from the Northern Baltic. Environmental Toxicology and Chemistry 1992. 11, 331-345.

Brownlow,A.H, Geochemistry. Second Edition ed 1996, New Jersey: Prentice Hall

Buser,H.R., Polybrominated Dibenzofurans and Dibenzo-Para-Dioxins - Thermal-Reaction Products of Polybrominated Diphenyl Ether Flame Retardants. Environmental Science & Technology 1986. 20, 404-408.

Butler,A., Walker,J.V., Marine Haloperoxidases. Chemical Reviews 1993. 93, 1937-1944.

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

Chung,H.Y., Ma,W.C.J., Ang,P.O., Kim,J.S., Chen,F., Seasonal variations of bromophenols in brown algae (Padina arborescens, Sargassum siliquastrum, and Lobophora variegata) collected in Hong Kong. Journal of Agricultural and Food Chemistry 2003. 51, 2619- 2624. de Wit,C.A., Alaee,M., Muir,D.C.G., Levels and trends of brominated flame retardants in the Arctic. Chemosphere 2006. 64, 209-233.

Drenzek,N.J., Tarr,C.H., Eglinton,T.I., Heraty,L.J., Sturchio,N.C., Shiner,V.J., Reddy,C.M., Stable chlorine and carbon isotopic compositions of selected semi-volatile organochlorine compounds. Organic Geochemistry 2002. 33, 437-444.

Eglinton,T.I., Aluwihare,L.I., Bauer,J.E., Druffel,E.R.M., McNichol,A.P., Gas chromatographic isolation of individual compounds from complex matrices for radiocarbon dating. Analytical Chemistry 1996. 68, 904- 912.

Eglinton,T.I., BenitezNelson,B.C., Pearson,A., McNichol,A.P., Bauer,J.E., Druffel,E.R.M., Variability in radiocarbon ages of individual organic compounds from marine sediments. Science 1997. 277, 796- 799.

Fan,X., Xu,N.J., Shi,J.G., Bromophenols from the red alga Rhodomela confervoides. Journal of Natural Products 2003. 66, 455-458.

Faulkner,D.J., Marine Natural Products. Natural Product Reports 2000. 17, 7-55.

Faulkner,D.J., Marine natural products. Natural Product Reports 2001. 18, 1-49.

Fielman,K.T., Woodin,S.A., Walla,M.D., Lincoln,D.E., Widespread occurrence of natural halogenated organics among temperate marine infauna. Marine Ecology-Progress Series 1999. 181, 1-12.

Flodin,C., Helidoniotis,F., Whitfield,F.B., Seasonal variation in bromophenol content and bromoperoxidase activity in Ulva lactuca. Phytochemistry 1999. 51, 135-138.

Flodin,C., Whitfield,F.B., Brominated anisoles and cresols in the red alga Polysiphonia sphaerocarpa. Phytochemistry 2000. 53, 77-80.

Fu,X.O., Schmitz,E.J., Govindan,M., Abbas,S.A., Hanson,K.M., Horton,P.A., Crews,P., Enzyme-Inhibitors - New and Known Polybrominated Phenols and Diphenyl Ethers from 4 Indo-Pacific Dysidea Sponges. Journal of Natural Products-Lloydia 1995. 58, 1384-1391.

Giesy,J.P., Kannan,K., Global distribution of perfluorooctane sulfonate in wildlife. Environmental Science & Technology 2001. 35, 1339- 1342.

Gribble,G.W., Naturally occurring organohalogen compounds. Accounts of Chemical Research 1998. 31, 141-152.

Gribble,G.W., The diversity of naturally occurring organobromine compounds. Chemical Society Reviews 1999. 28, 335-346.

Gribble,G.W., Blank,D.H., Jasinski,J.P., Synthesis and identification of two halogenated bipyrroles present in seabird eggs. Chemical Communications 1999. 2195-2196.

Haglund,P., On the identity and formation routes of environmentally abundant tri- and tetrabromodibenzo-p-dioxins. Chemosphere 2010. 78, 724-730.

Haglund,P., Löfstrand,K., Malmvärn,A., Bignert,A., Asplund,L., Temporal Variations of Polybrominated Dibenzo-p-Dioxin and Methoxylated Diphenyl Ether Concentrations in Fish Revealing Large Differences in Exposure and Metabolic Stability. Environmental Science & Technology 2010. 44, 2466-2473.

Haglund,P., Malmvärn,A., Bergek,S., Bignert,A., Kautsky,L., Nakano,T., Wiberg,K., Asplund,L., Brominated Dibenzo-p-Dioxins: A New Class of Marine Toxins? Environmental Science & Technology 2007. 41, 3069- 3074.

Haglund,P.S., Zook,D.R., Buser,H.R., Hu,J.W., Identification and quantification of polybrominated diphenyl ethers and methoxy- polybrominated diphenyl ethers in Baltic biota. Environmental Science & Technology 1997. 31, 3281-3287.

Holmstrand,H., Gadomski,D., Mandalakis,M., Tysklind,M., Irvine,R., Andersson,P., Gustafsson,Ö., Origin of PCDDs in ball clay assessed with compound-specific chlorine isotope analysis and radiocarbon dating. Environmental Science & Technology 2006a. 40, 3730-3735.

Holmstrand,H., Mandalakis,M., Zencak,Z., Andersson,P., Gustafsson,Ö., First compound-specific chlorine-isotope analysis of environmentally-bioaccumulated organochlorines indicates a degradation-relatable kinetic isotope effect for DDT. Chemosphere 2007. 69, 1533-1539.

Holmstrand,H., Mandalakis,M., Zencak,Z., Gustafsson,Ö., Andersson,P., Chlorine isotope fractionation of a semi-volatile organochlorine compound during preparative megabore-column capillary gas chromatography. Journal of Chromatography A 2006b. 1103, 133-138.

Jensen,S., Athanasiadou,M., Bergman,Å., A Technique for Separation of Xenobiotics From Large Amounts of Lipids. Organohalogen Compounds 1992. 8, 79-80.

Jensen,S., Reutergårdh,L., Jansson,B., Analytical methods for measuring organochlorines and methyl mercury by gas chromatography. FAO Fish Technical Paper 1983. 212, 21-33.

Krupp,E.A., Donard,O.F.X., Isotope ratios on transient signals with GC- MC-ICP-MS. International Journal of Mass Spectrometry 2005. 242, 233-242.

Kumata,H., Uchida,M., Sakuma,E., Uchida,T., Fujiwara,K., Tsuzuki,M., Yoneda,M., Shibata,Y., Compound class specific C-14 analysis of polycyclic aromatic hydrocarbons associated with PM10 and PM1.1 aerosols from residential areas of suburban Tokyo. Environmental Science & Technology 2006. 40, 3474-3480.

Kurata,K., Taniguchii,K., Takashima,K., Hayashi,I., Suzuki,M., Feeding- deterrent bromophenols from Odonthalia corymbifera. Phytochemistry 1997. 45, 485-487.

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

Malmberg,T., Athanasiadou,M., Marsh,G., Brandt,I., Bergmant,A., Identification of hydroxylated polybrominated diphenyl ether metabolites in blood plasma from polybrominated diphenyl ether exposed rats. Environmental Science & Technology 2005. 39, 5342- 5348.

Malmvärn,A., Zebuhr,Y., Jensen,S., Kautsky,L., Greyerz,E., Nakano,T., Asplund,L., Identification of polybrominated dibenzo-p-dioxins in blue mussels (Mytillus edulis) from the Baltic Sea. Environmental Science & Technology 2005. 39, 8235-8242.

Malmvärn,A., Zebuhr,Y., Kautsky,L., Bergman,Å., Asplund,L., Hydroxylated and methoxylated polybrominated diphenyl ethers and polybrominated dibenzo-p-dioxins in red algae and cyanobacteria living in the Baltic Sea. Chemosphere 2008. 72, 910-916.

Mandalakis,M., Gustafsson,Ö., Optimization of a preparative capillary gas chromatography-mass spectrometry system for the isolation and harvesting of individual polycyclic aromatic hydrocarbons. Journal of Chromatography A 2003. 996, 163-172.

Mandalakis,M., Gustafsson,Ö., Reddy,C.M., Li,X., Radiocarbon apportionment of fossil versus biofuel combustion sources of polycyclic aromatic hydrocarbons in the Stockholm metropolitan area. Environmental Science & Technology 2004. 38, 5344-5349.

Marsh,G., Athanasiadou,M., Athanassiadis,I., Bergman,Å., Endo,T., Haraguchi,K., Identification, Quantification and Synthesis of a Novel Dimethoxylated Polybrominated Biphenyl in Marine Mammals Caught Off the Coast of Japan. Environmental Science & Technology 2005. 39, 8684-8690.

Marsh,G., Athanasiadou,M., Bergman,Å., Asplund,L., Identification of hydroxylated and methoxylated polybrominated diphenyl ethers in Baltic Sea salmon (Salmo salar) blood. Environmental Science & Technology 2004a. 38, 10-18.

Meinert,C., Schymanski,E., Kuster,E., Kuhne,R., Schuurmann,G., Brack,W., Application of preparative capillary gas chromatography (pcGC), automated structure generation and mutagenicity prediction to improve effect-directed analysis of genotoxicants in a contaminated groundwater. Environmental Science and Pollution Research 2010. 17, 885-897.

Melcher,J., Olbrich,D., Marsh,G., Nikiforov,V., Gaus,C., Gaul,S., Vetter,W., Tetra- and tribromopbenoxyanisoles in marine samples from Oceania. Environmental Science & Technology 2005. 39, 7784- 7789.

Moore,C.A., Okuda,R.K., Bromoperoxidase activity in 94 species of marine algae. Journal of Natural Toxins 1996. 5, 295-305.

Nielson, A,H., Organic bromine and iodine compounds. In: The handbook of environmental chemistry vol 3 R, 2003, Springer-Verlag Berlin Heidelberg

Olsman,H., Engwall,M., Kammann,U., Klempt,M., Otte,J., van Bavel,B., Hollert,H., Relative differences in aryl hydrocarbon receptor-mediated response for 18 polybrominated and mixed halogenated dibenzo-p- dioxins and -furans in cell lines from four different species. Environmental Toxicology and Chemistry 2007. 26, 2448-2454.

Pena-Abaurrea,M., Weijs,L., Ramos,L., Borghesi,N., Corsolini,S., Neels,H., Blust,R., Covaci,A., Anthropogenic and naturally-produced organobrominated compounds in bluefin tuna from the Mediterranean Sea. Chemosphere 2009. 76, 1477-1482.

Reddy,C.M., Xu,L., Drenzek,N.J., Sturchio,N.C., Heraty,L.J., Kimblin,C., Butler,A., A chlorine isotope effect for enzyme-catalyzed chlorination. Journal of the American Chemical Society 2002. 124, 14526-14527.

Roeraade,J., Blomberg,S., Pietersma,H.D.J., High-Resolution Preparative Gas-Chromatography .1. A Microprocessor-Controlled System for Automated Fraction Collection. Journal of Chromatography 1986. 356, 271-284.

Saenger,P., Pedersen,M., Rowan,K.S., Bromo-Compounds of Red Alga Lenormandia-Prolifera. Phytochemistry 1976. 15, 1957-1958.

Schmidt,T.C., Zwank,L., Elsner,M., Berg,M., Meckenstock,R.U., Haderlein,S.B., Compound-specific stable isotope analysis of organic contaminants in natural environments: a critical review of the state of the art, prospects, and future challenges. Analytical and Bioanalytical Chemistry 2004. 378, 283-300.

Sharma, G.M., Vig,B., Burkholder,P.R., Antimicrobial substances of marine sponges IV. In: Food drugs from the sea. Proceednings 1969. Ed. Youngken,H.W.Jr, Marine Technology Society, Washington, DC, USA.

Sheesley,R.J., Krusa,M., Krecl,P., Johansson,C., Gustafsson,Ö., Source apportionment of elevated wintertime PAHs by compound-specific radiocarbon analysis. Atmospheric Chemistry and Physics 2009. 9, 3347-3356.

Sovocool,G.W., Donnelly,J.R., Munslow,W.D., Vonnahme,T.L., Nunn,N.J., Tondeur,Y., Mitchum,R.K., Analysis of Municipal Incinerator Fly-Ash for Bromo-Dioxins and Bromochloro-Dioxins, Dibenzofurans, and Related-Compounds. Chemosphere 1989. 18, 193-200.

Sovocool,G.W., Mitchum,R.K., Tondeur,Y., Munslow,W.D., Vonnahme,T.L., Donnelly,J.R., Bromo-Polynuclear and Bromochloro- Polynuclear Aromatic-Hydrocarbons, Dioxins and Dibenzofurans in Municipal Incinerator Fly-Ash. Biomedical and Environmental Mass Spectrometry 1988. 15, 669-676.

Sylva,S.P., Ball,L., Nelson,R.K., Reddy,C.M., Compound-specific Br- 81/Br-79 analysis by capillary gas chromatography/multicollector inductively coupled plasma mass spectrometry. Rapid Communications in Mass Spectrometry 2007. 21, 3301-3305.

Takahashi,S., Sakai,S., Watanabe,I., An intercalibration study on organobromine compounds: Results on polybrominated diphenylethers and related dioxin-like compounds. Chemosphere 2006. 64, 234-244.

Takahashi,Y., Daitoh,M., Suzuki,M., Abe,T., Masuda,M., Halogenated metabolites from the new Okinawan red alga Laurencia yonaguniensis. Journal of Natural Products 2002. 65, 395-398.

Teuten,E.L., Johnson,C.G., Mandalakis,M., Asplund,L., Gustafsson,Ö., Unger,M., Marsh,G., Reddy,C.M., Spectral characterization of two bioaccumulated methoxylated polybrominated diphenyl ethers. Chemosphere 2006a. 62, 197-203.

Teuten,E.L., King,G.M., Reddy,C.M., Natural C-14 in Saccoglossus bromophenolosus compared to C-14 in surrounding sediments. Marine Ecology-Progress Series 2006b. 324, 167-172.

Teuten,E.L., Pedler,B.E., Hangsterfer,A.N., Reddy,C.M., Identification of highly brominated analogues of Q1 in marine mammals. Environmental Pollution 2006c. 144, 336-344.

Teuten,E.L., Reddy,C.M., Halogenated organic compounds in archived whale oil: A pre-industrial record. Environmental Pollution 2007. 145, 668-671.

Teuten,E.L., Xu,L., Reddy,C.M., Two abundant bioaccumulated halogenated compounds are natural products. Science 2005. 307, 917-920.

Thoma,H., Hauschulz,G., Knorr,E., Hutzinger,O., Polybrominated Dibenzofurans (Pbdf) and Dibenzodioxins (Pbdd) from the Pyrolysis of Neat Brominated Diphenylethers, Biphenyls and Plastic Mixtures of These Compounds. Chemosphere 1987. 16, 277-285.

Tittlemier,S.A., Fisk,A.T., Hobson,K.A., Norstrom,R.J., Examination of the bioaccumulation of halogenated dimethyl bipyrroles in an Arctic marine food web using stable nitrogen isotope analysis. Environmental Pollution 2002. 116, 85-93.

Tittlemier,S.A., Simon,M., Jarman,W.M., Elliott,J.E., Norstrom,R.J., Identification of a novel C10H6N2Br4Cl2 heterocyclic compound in seabird eggs. A bioaccumulating marine natural product? Environmental Science & Technology 1999. 33, 26-33.

Unger, M., Asplund, L., Marsh, G., Gustafsson, Ö., Characterization of an abundant novel methyl- and methoxy-substituted brominated diphenyl ether isolated from whale blubber. Chemosphere 2010 79, 408-413.

Unger,M., Asplund,L., Haglund,P., Malmvärn,A., Arnoldsson,K., Gustafsson,Ö., Polybrominated and Mixed Brominated/Chlorinated Dibenzo-p-Dioxins in Sponge (Ephydatia fluviatilis) from the Baltic Sea. Environmental Science & Technology 2009. 43, 8245-8250.

Unson,M.D., Holland,N.D., Faulkner,D.J., A brominated secondary metabolite synthesized by the cyanobacterial symbiont of a marine sponge and accumulation of the crystalline metabolite in the sponge tissue. Marine Biology 1994. 119, 1-11.

Utkina,N.K., Denisenko,V.A., Scholokova,O.V., Virovaya,M.V., Gerasimenko,A.V., Popov,D.Y., Krasokhin,V.B., Popov,A.M., Spongiadioxins A and B, two new polybrominated dibenzo-p-dioxins from an Australian marine sponge Dysidea dendyi. Journal of Natural Products 2001. 64, 151-153.

Utkina,N.K., Denisenko,V.A., Virovaya,M.V., Scholokova,O.V., Prokof'eva,N.G., Two new minor polybrominated dibenzo-p-dioxins from the marine sponge Dysidea dendyi. Journal of Natural Products 2002. 65, 1213-1215.

Van den Berg,M., Birnbaum,L.S., Denison,M., De Vito,M., Farland,W., Feeley,M., Fiedler,H., Hakansson,H., Hanberg,A., Haws,L., Rose,M., Safe,S., Schrenk,D., Tohyama,C., Tritscher,A., Tuomisto,J., Tysklind,M., Walker,N., Peterson,R.E., The 2005 World Health Organization reevaluation of human and mammalian toxic equivalency factors for dioxins and dioxin-like compounds. Toxicological Sciences 2006. 93, 223-241.

Vetter,W., Turek,C., Marsh,G., Gaus,C., Identification and quantification of new polybrominated dimethoxybiphenyls (PBDMBs) in marine mammals from Australia. Chemosphere 2008. 73, 580- 586.

Weber,R., Kuch,B., Relevance of BFRs and thermal conditions on the formation pathways of brominated and brominated-chlorinated dibenzodioxins and dibenzofurans. Environment International 2003. 29, 699-710.

Whitfield,F.B., Helidoniotis,F., Shaw,K.J., Svoronos,D., Distribution of bromophenols in species of marine algae from eastern Australia. Journal of Agricultural and Food Chemistry 1999. 47, 2367-2373.

World Health Organization, Environmental Health Criteria 205: Polybrominated dibenzo-p-dioxins and dibenzofurans. 1998.

World Health Organization, Concise international chemical assessment document 66: 2,4,6-tribromophenol and other simple brominated phenols, 2005

Zencak,Z., Klanova,J., Holoubek,I., Gustafsson,Ö., Source apportionment of atmospheric PAHs in the western balkans by natural abundance radiocarbon analysis. Environmental Science & Technology 2007. 41, 3850-3855.

Zwank,L., Berg,M., Elsner,M., Schmidt,T.C., Schwarzenbach,R.P., Haderlein,S.B., New evaluation scheme for two-dimensional isotope analysis to decipher biodegradation processes: Application to groundwater contamination by MTBE. Environmental Science & Technology 2005. 39, 1018-1029.

No regrets.