Identification of Hydroxylated and Methoxylated Polybrominated Diphenyl Ethers in Baltic Sea Salmon (Salmo salar) Blood

Article abstract of DOI:10.1021/es034671j

Methoxylated and hydroxylated polybrominated diphenyl ethers (MeO-PBOEs and OH-PBOEs) have recently been reported to be present in wildlife from Northern Europe. The structures of a majority of these compounds have however been unknown. In the present study, nine OH-PBOEs and six MeO-PBOEs were identified in Baltic Sea salmon (Salmo salar) blood. All OH- and MeO-PBOEs identified were substituted with four or five bromines, and five of these had one chlorine substituent. Fourteen of the OH- and MeO-PBOEs have the methoxy or hydroxy group substituted in the ortho position to the diphenyl ether bond. Identification was done by comparison of relative retention times of authentic reference standards with compounds present in salmon plasma on two gas chromatographic columns of different polarities. The identification was supported by comparisons of full-scan mass spectrometric data: electron ionization (EI) and electron capture negative ionization (ECNI). Nine of the 15 OH- and MeO-PBDEs identified have not previously been reported to occur in the environment. The structures of several identified OH- and MeO-PBDEs support natural origin. However, at least one of the OH-PBDEs may be a hydroxylated metabolite of anthropogenic polybrominated diphenyl ether (PBDE).

Full text of DOI:10.1021/es034671j

Research
ether (2′-OH-BDE-66) (2). Other environmentally relevant
PBDE congeners, 2,2′,4,4′,5-pentabromodiphenyl ether (BDE-
Identification of Hydroxylated and
Methoxylated Polybrominated
Diphenyl Ethers in Baltic Sea
Salmon (Salmo salar) Blood
9
9) (4) and decaBDE (BDE-209) (5), have been shown to form
OH-PBDE metabolites in rats. On the other hand, OH-PBDEs
have been isolated, structurally identified, and determined
as natural products in marine sponges (6, 7) and ascidians
(tunicates) (8, 9). OH-PBDEs have also been detected in red
alga (10). Methoxylated PBDEs (MeO-PBDEs) have been
reported as natural products present in marine sponge (11,
12) and green alga (13). The occurrence of these compounds
in environmental samples can therefore originate either from
natural or anthropogenic sources.
G O¨ R A N M A R S H , * , †
†
M A R I A A T H A N A S I A D O U ,
†
A° K E B E R G M A N , A N D
‡
L I L L E M O R A S P L U N D
OH-PBDEs and MeO-PBDEs have been shown to be
present in species at high trophic levels. Two MeO-tetraBDEs
were reported from green turtle and marine mammals
Department of Environmental Chemistry and Institute of
Applied Environmental Research, Stockholm University,
SE-106 91 Stockholm, Sweden
(
whales, dolphins, dugong, and seals) from the Southern
Hemisphere and at concentrations higher than the levels of
PBDEs (14, 15). The concentrations of PBDEs in Australian
marine mammals were at least 2 orders of magnitude lower
as compared to the MeO-PBDEs (15). One of these MeO-
PBDEs, 2′-methoxy-2,3′,4,5′-tetrabromodiphenyl ether (2′-
MeO-BDE-68), and a diMeO-PBDE, 2′,6-dimethoxy-2,3′,4,5′-
tetrabromodiphenyl ether (2′,6-diMeO-BDE-68), were ident-
ified (14). The reference standards used for these identifica-
tions were based on compounds isolated from a marine
sponge collected within the same geographic area as the
majority of the investigated animals. Hence these results
support a natural origin of these identified compounds. Three
MeO-tetraBDEs with unknown structure were reported to
be present in whales collected in the Atlantic (16), and a
human milk sample from the Faeroe Islands was shown to
contain 2′-MeO-BDE-68 and a MeO-tetraBDE with unknown
structure (17). From Northern Europe, in the Baltic Sea area,
MeO-PBDEs have been reported in marine fish (18, 19),
freshwater fish (20), birds (21), and seals (18). OH-PBDEs
were reported in fish (19) and humans (22). 6-Methoxy-
Methoxylated and hydroxylated polybrominated diphenyl
ethers (MeO-PBDEs and OH-PBDEs) have recently been
reported to be present in wildlife from Northern Europe.
The structures of a majority of these compounds have however
been unknown. In the present study, nine OH-PBDEs
and six MeO-PBDEs were identified in Baltic Sea salmon
(Salmo salar) blood. All OH- and MeO-PBDEs identified
were substituted with four or five bromines, and five of these
had one chlorine substituent. Fourteen of the OH- and MeO-
PBDEs have the methoxy or hydroxy group substituted
in the ortho position to the diphenyl ether bond. Identification
was done by comparison of relative retention times of
authentic reference standards with compounds present in
salmon plasma on two gas chromatographic columns of
different polarities. The identification was supported
by comparisons of full-scan mass spectrometric data:
electron ionization (EI) and electron capture negative ionization
2
,2′,4,4′-tetrabromodiphenyl ether (6-MeO-BDE-47) has been
(ECNI). Nine of the 15 OH- and MeO-PBDEs identified
detected in fish (19, 20) and birds (21). The corresponding
phenolic compound, 6-OH-BDE-47, has been identified in
salmon (Salmo salar) plasma (19) and in human plasma (22).
Nevertheless, the majority of the OH- and MeO-PBDEs, all
substituted with 4-5 bromine atoms, indicated in previous
studies have not yet been identified (18, 19). In the latter
studies, it was not clarified whether the OH- and MeO-PBDEs
originated from natural or anthropogenic sources.
have not previously been reported to occur in the environment.
The structures of several identified OH- and MeO-PBDEs
support natural origin. However, at least one of the OH-
PBDEs may be a hydroxylated metabolite of anthropogenic
polybrominated diphenyl ether (PBDE).
The objective of this work was to identify the structures
of OH-PBDEs and MeO-PBDEs present in Baltic Sea salmon
plasma. This was done by comparison of synthesized MeO-
PBDE standards with compounds extracted from the salmon
plasma using relative retention times on a polar and nonpolar
GC column and by comparing MS data. Coelution between
PBDE and MeO-PBDE congeners and between MeO-PBDE
congeners are reported, and the potential sources of these
compounds are discussed.
Introduction
During the past decade, an increased interest has been
focused on brominated flame retardants (BFRs) and par-
ticularly on polybrominated diphenyl ethers (PBDEs), which
are known as ubiquitous environmental contaminants (1).
PBDEs are synthetic anthropogenic compounds and have
not to our knowledge been shown to occur as natural
products. Hydroxylated PBDEs (OH-PBDEs) have been
detected as metabolites in fish (2), mice, and rats (3) after
exposure to 2,2′,4,4′-tetrabromodiphenyl ether (BDE-47), the
most abundant PBDE congener in wildlife (1). Two of the
OH-PBDE metabolites found in fish have tentatively been
identified as 6-hydroxy-2,2′,4,4′-tetrabromodiphenyl ether
Materials and Methods
Chem icals. The chemical name and abbreviation of the
individual MeO-PBDE congeners used as authentic reference
standards are given in Table 1 and were synthesized as
described elsewhere (23, 24). Thirty-three PBDE congeners
(
6-OH-BDE-47) and 2′-hydroxy-2,3′,4,4′-tetrabromodiphenyl
(
-
abbreviated according to Ballschmiter et al.) (25): BDE-30,
32, -17, -25, -28, -33, -35, -37, -75, -51, -49, -71, -47, -66, -77,
*
Corresponding author phone: +46-8-163677; fax: +46-8-163979;
e-mail: goran.marsh@mk.su.se.
†
‡
-100, -119, -99, -116, -85, -155, -105, -154, -153, -139, -140,
-138, -166, -128, -183, -181, -190, and BDE-203 (presented in
Department of Environmental Chemistry.
Institute of Applied Environmental Research.
1
0
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 1, 2004
10.1021/es034671j CCC: $27.50

2004 Am erican Chem ical Society
Published on Web 11/18/2003

TABLE 1. Relative Retention Times (RRTs) Given in Elution Order of All MeO-PBDEs Used as Reference Standards on CP-SIL 8 and
a
SP-2331 GC Columns
CP-SIL 8 CB (nonpolar)
SP-2331 (polar)
MeO-PBDE
abbreviation
RRT
ID no.
abbreviation
RRT
ID no
6
4
2
3
4
6
2
6
4
3
2
5
4
3
6
5
4
4
5
6
6
3
4
2
6
2
6
′-m ethoxy-2,2′,4-triBDE
′-m ethoxy-2,4,6-triBDE
′-m ethoxy-2,4,4′-triBDE
′-m ethoxy-2,4,4′-triBDE
6′-MeO-BDE-17
4′-MeO-BDE-30
2′-MeO-BDE-28
3′-MeO-BDE-28
4′-MeO-BDE-17
6′-MeO-BDE-49
2′-MeO-BDE-68
6-MeO-BDE-47
4′-MeO-BDE-69
3-MeO-BDE-47
2′-MeO-BDE-66
5-MeO-BDE-47
4′-MeO-BDE-49
3′-Cl-6′-MeO-BDE-49
6′-Cl-2′-MeO-BDE-68
5-Cl-6-MeO-BDE-47
4′-MeO-BDE-121
4-MeO-BDE-42
5′-Cl-2′-MeO-BDE-66
6-MeO-BDE-90
6-MeO-BDE-99
3-Cl-6-MeO-BDE-47
4-MeO-BDE-90
2-MeO-BDE-123
6-MeO-BDE-85
BDE-138 (IS)
0.754
0.755
0.768
0.783
0.783
0.813
0.824
0.836
0.839
0.850
0.851
0.854
0.856
0.864
0.869
0.870
0.886
0.888
0.898
0.900
0.903
0.905
0.929
0.935
0.947
1.000
1.044
4′-MeO-BDE-30
6′-MeO-BDE-17
2′-MeO-BDE-28
6′-MeO-BDE-49
4′-MeO-BDE-17
2′-MeO-BDE-68
3′-MeO-BDE-28
3′-Cl-6′-MeO-BDE-49
6′-Cl-2′-MeO-BDE-68
5-Cl-6-MeO-BDE-47
6-MeO-BDE-47
4′-MeO-BDE-69
2′-MeO-BDE-66
4′-MeO-BDE-121
3-MeO-BDE-47
6-MeO-BDE-90
4′-MeO-BDE-49
6-MeO-BDE-99
5-MeO-BDE-47
5′-Cl-2′-MeO-BDE-66
3-Cl-6-MeO-BDE-47
2-MeO-BDE-123
4-MeO-BDE-90
4-MeO-BDE-42
6-MeO-BDE-85
BDE-138 (IS)
0.495
0.544
0.554
0.571
0.582
0.587
0.609
0.647
0.658
0.662
0.669
0.671
0.683
0.689
0.695
0.724
0.731
0.731
0.743
0.760
0.832
0.836
0.836
0.853
0.946
1.000
b
1
2
′-m ethoxy-2,2′,4-triBDE
′-m ethoxy-2,2′,4,5′-tetraBDE
′-m ethoxy-2,3′,4,5′-tetraBDE
-m ethoxy-2,2′,4,4′-tetraBDE
′-m ethoxy-2,3′,4,6-tetraBDE
-m ethoxy-2,2′,4,4′-tetraBDE
′-m ethoxy-2,3′,4,4′-tetraBDE
-m ethoxy-2,2′,4,4′-tetraBDE
′-m ethoxy-2,2′,4,5′-tetraBDE
′-chloro-6′-m ethoxy-2,2′,4,5′-tetraBDE
′-chloro-2′-m ethoxy-2,3′,4,5′-tetraBDE
-chloro-6-m ethoxy-2,2′,4,4′-tetraBDE
′-m ethoxy-2,3′,4,5′,6-pentaBDE
-m ethoxy-2,2′,3,4′-tetraBDE
′-chloro-2′-m ethoxy-2,3′,4,4′-tetraBDE
-m ethoxy-2,2′,3,4′,5-pentaBDE
-m ethoxy-2,2′,4,4′,5-pentaBDE
-chloro-6-m ethoxy-2,2′,4,4′-tetraBDE
-m ethoxy-2,2′,3,4′,5-pentaBDE
-m ethoxy-2′,3,4,4′,5-pentaBDE
-m ethoxy-2,2′,3,4,4′-pentaBDE
,2′,3,4,4′,5′-hexaBDE
1
2
3
5
6
7
3
4
5
6
7
8
4
9
8
9
-m ethoxy-2,2′,3,4,4′,5-hexaBDE
6-MeO-BDE-137
6-MeO-BDE-137
a
Recorded by ECNI m /z 79 and 81. Matching com pounds from the salm on blood sam ples are m arked in bold with identification num bers as
b
in Figures 1 and 2 (chrom atogram s) and in Figure 3 (structures). Did not elute within the GC program used.
elution order on a CP-Sil 8 CB GC column; BDE-17/ BDE-25
and BDE-28/ BDE-33 are coeluting pairs) were used to
investigate any potential chromatographic coelution with
MeO-PBDEs. The PBDE congeners have been synthesized
as previously reported (26, 27), but the preparation of BDE-
sodium hydroxide (8.4 mg, 0.21 mmol) and tetrabutylam-
monium hydroxide (34 mg, 0.14 mmol) in water (2 mL) after
which iodomethane (20 µL, 0.35 mmol) in dichloromethane
(2 mL) was added and the reaction mixture was stirred at
room temperature for 4 h. The layers were separated, and
the water phase was extracted with dichloromethane (2 mL).
The combined organic phases were dried with sodium sulfate,
filtered, and concentrated. The crude product mixture was
purified on a silica gel column (2 × 10 cm) from polar
substances using dichloromethane/ n-hexane (1:1) as the
mobile phase.
The partial chlorination and the subsequent methylation
of 6-OH-BDE-47, 6′-OH-BDE-49, and 2′-OH-BDE-66 resulted
in two isomeric monochlorinated 6-MeO-BDE-47; in one
monochlorinated isomer product from each of 6′-MeO-BDE-
49 and 2′-MeO-BDE-66, respectively; and in a mixture with
the starting compound. The structures were tentatively
assigned as follows: the monochlorinated MeO-PBDEs had
1
05, -155, -139, and -183 are unpublished. The abbreviation
system used for the OH-PBDE and MeO-PBDE congeners is
the same as that used for polyhalogenated biphenylols and
their methyl ether derivatives (28). The nomenclature used
for OH- and MeO-PBDEs deviate from IUPAC rules (i.e.,
phenoxyphenols and methoxyphenoxybenzene), and these
compounds are instead referred to as diphenyl ethers in the
present work to simplify comparison with PBDE congeners.
All solvents, acids, and other chemicals used in the synthesis
and analysis were of pro analysis quality.
Synthesis of Reference Standards. Three partially chlo-
rinated MeO-PBDE mixtures were prepared from 6-OH-BDE-
4
7, 6′-hydroxy-2,2′,4,5′-tetrabromodiphenyl ether (6′-OH-
BDE-49), and 2′-OH-BDE-66 and used as monochlorinated
MeO-tetraBDE reference compounds. An attempt was also
made to chlorinate 2′-hydroxy-2,3′,4,5′-tetrabromodiphenyl
ether (2′-OH-BDE-68) in the same manner as described
below, but no monochlorinated OH-tetraBDE product was
formed. The mixtures were prepared as follows: Aqueous
sodium chlorate (50 µL, 50 µmol) was added to a solution of
the OH-tetraBDE congener (23) (35 mg, 70 µmol), mentioned
above, in acetic acid (4 mL) and concentrated hydrochloric
acid (0.25 mL) at room temperature. The reaction mixture
was stirred at room temperature for 0.5 h, and water (5 mL)
was added. The aqueous phase was removed from the
precipitated product mixture. This mixture was washed with
water (4 mL) and dissolved in dichloromethane (3 mL). The
organic phase was washed with saturated aqueous sodium
hydrogen carbonate (3 mL) followed by water (2 × 3 mL),
and the products were concentrated in a rotary evaporator.
The crude chlorinated OH-PBDE mixture was treated with
3
(i) a loss of BrCH as determined by MS in electron ionization
(EI) mode, indicating the methoxy group in the ortho position
of the diphenyl ether bond (23); (ii) the two fragment ions
-
-
[M - C
6
HBr
and/ or [M - C
2
ClOCH
Br
3
]
2
m/ z 249 and [M + H - OC
3
6
H
3
Br
2
]
-
6
H
3
CH
] m/ z 298 that were shown by MS
in electron capture negative ionization (ECNI) mode, indi-
cating the numbers of halogen atoms in each ring (18, 29);
(iii) a molecular ion at m/ z 546 determined by EI; and (iv)
an ion isotopic halogen pattern that was in agreement with
the expected pattern for all ions assigned. One of the
methylated monochlorinated products from 6-OH-BDE-47
had identical MS spectra and RRT as the authentic 5-Cl-6-
MeO-BDE-47 previously synthesized (23). Accordingly, the
other isomer formed is most likely 3-Cl-6-MeO-BDE-47. Since
there was only one monochlorinated product formed from
both 6′-OH-BDE-49 and 2′-OH-BDE-66, the most likely
substitution position is para to the hydroxy group (the ortho
positions are occupied), which support the structures of 3′-
VOL. 38, NO. 1, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1 1

Cl-6′-MeO-BDE-49 and 5′-Cl-2′-MeO-BDE-66, respectively,
after methylation. Furthermore, the authentic reference 6′-
Cl-2′-MeO-BDE-68 (23) gives distinct loss of BrCl by MS in
EI mode, while the chlorinated and methylated product of
2, 20% dichloromethane in n-hexane (4 mL); and fraction 3,
50% dichloromethane in n-hexane (4 mL). Fractions 2 and
3 were finally pooled, and the solvent volume was reduced
to 50 µL prior to GC-MS analysis.
The neutral fraction of muscle and egg samples from the
same salmons, which were used for supporting ECNI
information in the present work (of four MeO-PBDEs), were
cleaned up as described earlier (19). Furthermore, they were
similarly pooled and fractionated on silica as the salmon
blood samples.
2
′-OH-BDE-66 only gave traces of this loss, indicating that
the chlorine is substituted in the 5′-position and not in the
′-position. As mentioned, 2′-OH-BDE-68 did not give any
6
monochlorinated OH-tetraBDEs products at all under the
preparation conditions used. This further supports the ortho/
para directing properties of the aromatic hydroxy group as
the conclusive factor during the chlorination, despite the
fact that the aromatic ring is substituted with four functional
groups. On the basis of the data and the reasoning above,
the structures of the chlorinated MeO-PBDEs are suggested
to be 3-Cl-6-MeO-BDE-47 and 5-Cl-6-MeO-BDE-47 from
Results
The identification of the OH- and MeO-PBDEs in the salmon
blood plasma was supported by similar RRTs versus BDE-
1
38 on two columns compared to the synthetic references
6
-OH-BDE-47, 3′-Cl-6′-MeO-BDE-49 from 6′-OH-BDE-49,
as shown in Table 1. The RRTs were determined from full-
scan ECNI chromatograms by recording the ions m/ z 79
and 81. In all cases, the difference in RRT between each
reference standard and the corresponding MeO-PBDE in the
salmon blood sample was less than (0.001 on both columns
used. Mass chromatograms of the phenolic and the neutral
fractions on both nonpolar and polar GC columns are shown
in Figures 1 and 2, respectively. The identified compounds,
nine OH-PBDEs and six MeO-PBDEs, are labeled on the mass
chromatograms, and their structures are shown in Figure 3.
MS of Major Com pounds. 6′-MeO-BDE-49, 2′-MeO-BDE-
and 5′-Cl-2′-MeO-BDE-66 from 2′-OH-BDE-66.
Instrum ents. Gas chromatography-mass spectrometry
(
7
GC-MS) analyses were performed on a Finnigan MAT SSQ
10 with a Varian 3400 gas chromatograph and a split/ splitless
injector operated in the splitless mode. Two fused-silica
capillary columns with different polarities were used. A
nonpolar column with a 5% phenyl, dimethylpolysiloxane
phase, CP-Sil 8 CB (30 m × 0.25 mm i.d. and 0.25 µm film
thickness) was purchased from Chrompack (EA Middelburg,
The Netherlands). The column was programmed as follows:
-
1
8
0 °C (2 min), 10 °C min to 300 °C (10 min). The injector
6
8, 6-MeO-BDE-47, and 6-MeO-BDE-90 (1-3 and 8) from
and transfer line temperatures were 260 and 270 °C,
respectively. The other GC column was a polar column with
an 80% biscyanopropyl, 20% cyanopropylphenyl, siloxane
phase, SP-2331 (30 m × 0.25 mm i.d. and 0.2 µm film
thickness) obtained from Supelco (Bellefonte, USA). The
the neutral fraction and the methylated corresponding OH-
PBDEs as well as 6′-Cl-2′-OH-BDE-68 from the phenolic
fraction (1-3, 6, and 8) all had matching EI mass spectra
with comparison to the authentic references. A full-scan EI
spectrum is shown in Figure 4a for 6′-Cl-2′-MeO-BDE-68 (6)
-
1
column temperature program was 80 °C (1 min), 20 °C min
from the methylated phenolic fraction. In EI mode, the
-
1
to 200 °C (1 min), 3 °C min to 270 °C (12 min). The injector
and transfer line temperature were both 260 °C. Helium was
used as carrier gas.
+
presence of a [M - CH
3
] fragment ion is characteristic for
MeO-PBDEs with the methoxy group in the para position to
the diphenyl ether bond, the [M - BrCH
+
3
] fragment ion is
Two different ionization techniques were used: electron
ionization (EI) and electron-capture negative ionization
characteristic for the methoxy group in the ortho position,
and the absence of these two fragment ions indicates that
the methoxy group is substituted in the meta position, as
recently reported (23). This fragmentation is also in agreement
with previous observations of chlorinated analogues (i.e.,
ortho-, meta-, and para-substituted methoxylated polychlo-
rinated diphenyl ethers (MeO-PCDEs)) (30, 31). A specific
(
2
ECNI). For ECNI, methane (quality 4.5, <5 ppm O , AGA,
Stockholm, Sweden) was used as the reagent gas. The electron
energy was 70 eV, and the ion source temperature was 150
°
2
C for both MS techniques. The spectra were scanned from
00 to 750 m/ z in EI and from 32 to 750 m/ z, in ECNI. The
chromatographic data were recorded and processed with
the ICIS data system (Finnigan MAT, USA).
Sam ples. Thirty female sea-run Baltic salmon from the
Dal a¨ lven River population sampled in 1995 were used for
the preparation of the pooled samples (for a more detailed
description, see Asplund et al. (19)).
3
loss of BrCH was observed for all compounds mentioned
above, indicating that the methoxy group is substituted in
the ortho position to the diphenyl ether bond, which is in
agreement with the authentic references. 6′-Cl-2′-MeO-BDE-
68 (6) had a distinct loss of BrCl giving the fragment ion [M
+
- BrCl] (Figure 4a), as observed for the authentic reference
Cleanup and Analysis. The salmon blood plasma (5 g)
were extracted and cleaned up as described elsewhere (19).
In brief: the plasma samples (5 g) were denaturated by
hydrochloric acid (6 M) and 2-propanol, followed by extrac-
tion of the organohalogens with n-hexane/ methyl tert-butyl
ether. Neutral and phenolic compounds in the extract were
separated by partitioning between n-hexane and potassium
hydroxide (0.5 M). After acidification of the aqueous phase,
phenolic compounds were extracted into an organic phase
and derivatized by diazomethane. Lipids present in the
phenolic and neutral fractions were removed by HR-GPC
and further cleaned up on a silica gel column (0.5 g) eluted
with dichloromethane. The neutral phase was fractionated
by HPLC on a nitrophenyl silica column, eluted with
n-hexane. Two fractions were collected: fraction 1, where
the PCBs eluted, and fraction 2 (back flush), where PBDE
and MeO-PBDE eluted. The phenolic and neutral fractions
were pooled separately. The solvent volume was reduced to
approximately 0.5 mL and transferred to an activated (300
standard, but this fragment ion was absent or only present
in trace amounts for all other investigated monochlorinated
MeO-tetraBDEs used as reference standards. The identities
of these compounds were also supported by ECNI full-scan
spectra. An ECNI full-scan spectrum of the methylated 6′-
Cl-2′-OH-BDE-68 (6) from the phenolic fraction is shown in
-
Figure 4b. Besides the bromine ions m/ z 79 and 81 [Br] and
-
159, 161, and 163 [HBr
2
]
and also the mixed bromine/
-
chlorine ions 115, 117, and 119 [HBrCl] for the chlorine-
containing compounds, fragment ions above m/ z 163 were
also detected but with low abundance as compared with
-
[Br] . For the compounds 6′-MeO-BDE-49, 2′-MeO-BDE-68,
6-MeO-BDE-47, and 6-MeO-BDE-90 (1-3 and 8), the higher
m/ z ions were more clearly detected in muscle and egg
samples from the same salmon as compared to the phenolic
and neutral fractions in the salmon blood sample. Also
in ECNI mode, the major compounds identified formed
specific fragment ions, which give important structural
information. Due to cleavage at the diphenyl ether oxygen
°
C overnight) silica column (1 g in a Pasteur pipet). Three
bond, the nonmethoxylated phenyl ring forms the fragment
-
fractions were collected: fraction 1, n-hexane (6 mL); fraction
ion [M - C
6
H
1-2Br2-3Cl0-1OCH
3
]
and the methoxylated
1
2
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 1, 2004

FIGURE 1. GC-MS (ECNI) chromatograms of bromide ions (m/z: 79, 81) of phenol-type substances (upper chromatogram) and neutral
substances (lower chromatogram) in salmon blood analyzed on a CP-Sil 8 GC column. The phenolic compounds were methylated prior
to GC-MS analysis. The peak identity numbers in the chromatograms refer to the chemical name and abbreviations given in Table 1,
and the structures of these compounds are shown in Figure 3. A schematic chemical structure of the unidentified compound UC is also
shown in Figure 3.
phenyl ring forms [M - C
6
H
3
Br
2
CH
3
]- and/ or [M + H -
were the only identified compounds with an abundant clear
[M] in ECNI mode.
-
-
OC
6
H
3
Br
2
] as shown for 6′-Cl-2′-MeO-BDE-68 (6) in Figure
4
b. This fragmentation has previously been reported for MeO-
One OH-PBDE and its corresponding MeO-PBDE were
verified in the salmon blood and marked with UC in Figures
1-3. The methylated derivative of this OH-PBDE was thus
identical to the MeO-PBDE and had a molecule ion at
m/ z 468 with a chlorotribromo isotope pattern, as detected
both with EI and ECNI. These compounds were substituated
with a hydroxy/ methoxy group in the ortho position to
the diphenyl ether bond and with two bromine atoms in
the nonhydroxylated/ nonmethoxylated ring as well as one
bromine and one chlorine atom in the hydroxylated/
methoxylated ring, according to EI and ECNI. Unfortunately,
no methoxylated chlorotriBDEs standards were available to
confirm the identity of these compounds.
MS of Minor Com pounds. EI full-scan spectra, present
in relatively low abundance as compared with the major
MeO-PBDEs, were recorded for three of the methylated OH-
PBDEs from the phenolic fraction (i.e., 4′-MeO-BDE-49, 3′-
Cl-6′-MeO-BDE-49, and 6-MeO-BDE-99 (4, 5, and 9)) but
could nevertheless give some structural information. 3′-Cl-
6′-MeO-BDE-49 and 6-MeO-BDE-99 (5 and 9) also had a
PCDEs with the methoxy group in the ortho position to the
diphenyl ether bond (29). The fragment ion [M + H -
OC
tetraBDE with a unknown bromine substitution pattern but
with the methoxy group determined to be in the ortho
position, according to EI (18). In each case the fragment ion
-
6
H
3
Br
2
]
has previously been discussed for a MeO-
-
[M -C
6
H
1-2Br2-3Cl0-1OCH
3
] obtained from the MeO-PBDEs
in the salmon samples was a dibromophenolate ion since
this 2,4-dibromo-substituted (ortho and para to the diphenyl
ether bond) nonmethoxylated diphenyl ether ring was the
same for all compounds. Thus, these two specific ECNI
fragment ions provide information about the number of
halogen atoms in each phenyl ring, and this was also the
case for the major MeO-PBDEs in the salmon samples, which
was similar for the synthesized reference compounds.
However, for 6-MeO-BDE-47 (3), this fragmentation was poor
and shown in some MS runs but not in others, and this was
the case both for the salmon samples and the reference
standard. 6′-Cl-2′-MeO-BDE-68 (6) and 2′-MeO-BDE-68 (2)
VOL. 38, NO. 1, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1 3

FIGURE 2. GC-MS (ECNI) chromatogram of bromide ions (m/z: 79, 81) of phenol-type substances (upper chromatogram) and neutral
substances (lower chromatogram) in salmon blood analyzed on a polar SP 2331 column. The phenolic compounds were methylated prior
to GC-MS analysis. The peak identity numbers in the chromatograms refer to the chemical name and abbreviations given in Table 1,
and the structures of these compounds are shown in Figure 3. A schematic chemical structure of the unidentified compound UC is also
shown in Figure 3.
loss of BrCH
3
whereas 4′-MeO-BDE-49 (4) did not. However,
and indicates the
PBDEs. An additional number of both MeO-PBDEs and OH-
PBDEs were indicated in the salmon plasma but not possible
to confirm since their concentrations were too low. All
coelutions found from all reference compounds (PBDEs and
MeO-PBDEs) on both polar and nonpolar GC columns are
summarized in Table 2.
4
′-MeO-BDE-49 (4) had a loss of CH
3
methoxy group in the para position to the diphenyl ether
bond. This fragment ion was absent or only present in trace
amounts for all other MeO-PBDEs detected, including the
major once. All other compounds (i.e., the methylated 5-Cl-
6
-OH-BDE-47 (7) from the phenolic fraction and 5-Cl-6-MeO-
Discussion
BDE-47 (7) and 6′-Cl-2′-MeO-BDE-68 (6) from the neutral
fraction) did not give any supporting EI information. No clear
ECNI full-scan mass spectra could be obtained from these
compounds that were present in the salmon blood plasma
in low concentrations. However, these compounds all gave
PBDEs and Coelutions. The six PBDE congeners detected
in the plasma correspond to the most dominating ones
reported in biota (1, 32, 33). Of the identified compounds in
the present study, BDE-99, 6′-Cl-2′-MeO-BDE-68 (6), and
5-Cl-6-MeO-BDE-47 (7) coeluted on the nonpolar CP-Sil-8
column whereas 4′-MeO-BDE-49 (4) and 6-MeO-BDE-99 (9)
coeluted on the polar SP 2331 column. The coelution between
BDE-99 and 6′-Cl-2′-MeO-BDE-68 (6) as well as 5-Cl-6-MeO-
BDE-47 (7) on the common nonpolar DB-5 type CP-Sil-8
column must be considered to avoid errors in the quanti-
fication of BDE-99. BDE-99 has previously been reported to
coelute with 2′,6-diMeO-BDE-68 in environmental samples
of marine mammals from the Southern Hemisphere (14, 15)
and in a human milk from the Faeroe Islands (17).
-
the bromine ions m/ z 79 and 81 [Br] and 159, 161, and 163
-
[HBr
2
] , and the chlorine-containing compounds gave the
-
mixed bromine/ chlorine ions 115, 117, and 119 [HBrCl] . It
should be emphasized that the 4′-MeO-BDE-49 (4) gave very
poor fragmentation above m/ z 163 for the authentic standard;
consequently, it was difficult to find fragmental ions matching
with the corresponding sample compound.
Both the neutral fraction and the phenolic fraction from
the salmon blood contained a number of other compounds
that interfered with the MeO-PBDEs and methylated OH-
1
4
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 1, 2004

FIGURE 3. Structures of the compounds identified in the salmon blood. The identification numbers of corresponding OH-/MeO-PBDEs are
the same (Figures 1 and 2) since the OH-PBDEs are methylated prior to GC-MS analysis.
Structural Inform ation of the Identified OH- and MeO-
PBDEs. MS data that gave characteristic structural informa-
tion of the MeO-PBDEs was valuable for two purposes. As
pointed out in the results, the major MeO-PBDEs detected
in the salmon samples and partly also three of the minor
compounds gave MS structural information, which strengthen
the identity besides the matching RRTs. MS structural
information was also used for the selection of potential MeO-
PBDEs candidates, which were synthesized with respect to
the present work. The MS information could together with
an assumption (see below) reduce the relative large number
of theoretical MeO-PBDE isomers to only a few, as illustrated
in Table 3. The majority of the MeO-PBDEs used as reference
compounds (23) were prepared according to this guidance.
Hence, the major MeO-PBDEs in the salmon samples gave
to 6 MeO-tetraBDEs, 12 MeO-chlorotetraBDEs, and 4 MeO-
pentaBDEs for the major compounds detected (Table 3). Of
these compounds, 4 of the 6 MeO-tetraBDEs, 5 of the 12
MeO-chlorotetraBDEs, and all 4 MeO-pentaBDEs were made
available as reference standards in the present study.
RRTs of certain synthesized MeO-PBDEs, with the meth-
oxy group in the ortho position, gave also structural
information since 2,3,4- and 3,4,5-trihalogenation in the
methoxylated phenyl ring of these compounds seems to have
longer RRTs as compared to equivalent isomers with 2,3,5-
and 2,4,5-trihalogen pattern on the nonpolar column (Table
1). This latter information was especially useful for the
preparation of MeO-chlorotetraBDEs (5-Cl-6-MeO-BDE-47
(7) and 6′-Cl-2′-MeO-BDE-68 (6)), which were the most
recently synthesized compounds. Additional structural in-
formation about the major MeO-chlorotetraBDE investigated,
originating from the phenolic fraction (identified as 6′-Cl-
2′-OH-BDE-68 (6)), was obtained from the loss of BrCl in EI
mode that indicated the chlorine substituent to be in the
ortho position to the diphenyl ether bond.
Environm ental Occurrence of the Identified OH- and
MeO-PBDEs. Nine compounds (5-Cl-6-OH-BDE-47, 5-Cl-
6-MeO-BDE-47, 6′-OH-BDE-49, 6′-MeO-BDE-49, 4′-OH-
BDE-49, 3′-Cl-6′-OH-BDE-49, 6′-Cl-2′-OH-BDE-68, 6′-Cl-2′-
MeO-BDE-68, and 6-MeO-BDE-90) are here reported for the
first time to occur in the environment. The following OH-
and MeO-PBDEs, reported herein, have previously been
isolated, structurally identified, and confirmed as natural
products: 6-OH-BDE-47 has been detected in marine
sponges (7, 35-37) and in ascidians (tunicates) (8, 9); 2′-
MeO-BDE-68 has been identified in marine sponge (38) as
a loss of BrCH
3
indicating the methoxy group in the ortho
position to the diphenyl ether bond, and the fragment ions
-
-
[
M - C
6
H
1-2Br2-3Cl0-1OCH
3
] and [M - C H Br CH ] and/
6 3 2 3
-
or [M + H - OC
6
H
3
Br
2
] in ECNI mode gave the number of
halogens in each phenyl ring, which in each case were two
bromine atoms in the nonmethoxylated phenyl ring and,
consequently, the resulting halogens in the methoxylated
phenyl ring. Furthermore, we assumed that the non-
methoxylated ring is 2,4-bromo-substituted since all OH-
and MeO-PBDEs with natural origin reported in the literature,
with two bromine atoms in the nonmethoxylated ring, are
all substituted in this way. The over all dominating substitu-
tion of technical PBDEs (34) as well as PBDEs found in biota
(
1, 32, 33) with two bromine atoms in one of the phenyl rings
is similarly, 2,4-bromo-substituted. Having this in mind, the
potential candidates of MeO-PBDE isomers can be reduced
VOL. 38, NO. 1, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1 5

FIGURE 4. EI (a) and ECNI (b) mass spectra of 6′-chloro-2′-methoxy-2,3′,4,5′-tetraBDE (6′-Cl-2′-MeO-BDE-68), originating from the phenolic
fraction in the salmon blood, as 6′-chloro-2′-hydroxy-2,3′,4,5′-tetraBDE (6′-Cl-2′-OH-BDE-68).
well as in green alga (13); 6-MeO-BDE-47 (11), 2′-OH-BDE-
8 (7, 12, 36, 37), and 6-OH-BDE-99 (6) have been identified
in a marine sponge. In addition, 6-OH-BDE-47, 2′-OH-BDE-
8, 6-OH-BDE-90, and 6-OH-BDE-99 have been detected in
red alga by comparison of RRTs and ECNI full-scan mass
spectra with authentic reference compounds (10). As pointed
out in the Introduction, 6-OH-BDE-47, 6-MeO-BDE-47, and
and fish), and no MeO-PBDEs have hitherto been reported
with that origin (2-4). However, it cannot be excluded that
MeO-PBDEs are formed metabolically from PBDEs but then
only as a minor pathway. Methoxyhydroxy-PBDE metabolites
containing 6-7 bromine atoms have been shown in rats
exposed to BDE-209 (5). OH-PCDEs have been reported to
be microbially methylated (46), and this seems to be one
possible pathway also for the methylation of OH-PBDEs.
The structure of a OH-PBDE (i.e., the position of the
hydroxy group and the bromine atoms) may indicate its
origin. Environmental samples and technical pentaBDE
products contain the dominating tetraBDE congener BDE-
47 as well as the minor tetraBDEs (BDE-49 and BDE-66), and
the dominating pentaBDEs are BDE-85, BDE-99, and BDE-
100 (32-34). According to what is known about metabolism
of halogenated aromatic compounds via cytochrome P450-
mediated transformations (47), via direct hydroxylation, or
via hydroxylation with possible 1,2-shift resulting in a
bromine or hydrogen migration, it is not possible to obtain
more than four of the OH-PBDE discussed in the present
paper (i.e., 6-OH-BDE-47, 6′-OH-BDE-49, 6-OH-BDE-99, and
4′-OH-BDE-49) from environmental relevant PBDEs. The
presence of a chlorine atom in five of the identified
compounds in the salmon blood excludes PBDEs as their
precursors. On the other hand, one hydroxylated chlorotet-
rabromodiphenyl ether has been structurally identified and
6
6
2
′-MeO-BDE-68 have previously been reported in biota of a
high trophic level.
Sources of the Identified OH- and MeO-PBDEs. To our
knowledge, OH- and MeO-PBDEs are not industrially pro-
duced and have not been reported as impurities in any
brominated technical products. However, various technical
chlorophenol products have been reported to contain
byproducts of which OH-PCDEs were found as the major
polychlorophenol dimer byproduct (39-42). MeO-PCDEs
have also been determined in commercial polychlorophenol
products (43, 44). Contradictory, two technical polybro-
mophenol mixtures (i.e., 2,4,6-tribromophenol and pent-
abromophenol) were found to contain no OH-PBDEs or other
dimeric byproducts, such as MeO-PBDEs (45). According to
the present knowledge about OH- and MeO-PBDEs, their
appearance in environmental samples is most likely either
as natural products and/ or metabolites of anthropogenic
PBDEs. OH-PBDEs that originate from PBDEs have so far
only been observed in laboratory experiments (in mice, rat,
1
6
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 1, 2004

Cl-6-MeO-BDE-47)) (35), and additionally one hydroxylated
chlorotetrabromodiphenyl ether has been detected in red
alga present in the Baltic Sea (10). 6-OH-BDE-47, 6′-OH-
BDE-49, and 6-OH-BDE-99 may be metabolites of BDE-47,
BDE-49, and BDE-99, respectively, and 6-MeO-BDE-47 and
6′-MeO-BDE-49 could possibly be formed through a meta-
bolic methylation process. On the other hand, both 6-OH-
BDE-47 and 6′-OH-BDE-49 have their corresponding MeO-
PBDEs present in the salmon blood, similar to five other
OH-PBDEs that also form MeO-PBDEs, but none of these
are likely metabolites of PBDEs. Forming such a OH-/ MeO-
PBDE pair at the actual concentrations is thus suggested as
evidence of natural formation of at least the major part of
TABLE 2. Coelutions Encountered for PBDE and MeO-PBDE
Congeners on CP-Sil 8 and SP 2331 GC Columns, Presented
in Elution Order
CP-Sil 8 (nonpolar)
vs
PBDEs
MeO-PBDEs
BDE-71
3′-MeO-BDE-28
4
′-MeO-BDE-17
BDE-77
BDE-100
2′-MeO-BDE-68
5-MeO-BDE-47
4
′-MeO-BDE-49
BDE-85
BDE-99
6-MeO-BDE-99
5-Cl-6-MeO-BDE-47
6
′-Cl-2′-MeO-BDE-68
6
-OH/ MeO-BDE-47 and 6′-OH/ MeO-BDE-49. In the salmon
MeO-PBDEs
vs
MeO-PBDEs
blood, the concentrations of these MeO-PBDEs are ap-
proximately in the same range as the PBDEs, and the
corresponding OH-PBDEs were estimated to be 20-30% of
the MeO-PBDEs (19). In addition, 6-OH-BDE-47 and 6-OH-
BDE-99 have been reported in red alga from the Baltic Sea
6
3
3
6
4
6
′-MeO-BDE-17
′-MeO-BDE-28
-MeO-BDE-47
′-Cl-2′-MeO-BDE-68
′-MeO-BDE-121
-MeO-BDE-90
4′-MeO-BDE-30
4′-MeO-BDE-17
2′-MeO-BDE-66
5-Cl-6-MeO-BDE-47
4′-MeO-BDE-42
(
10), which supports natural origin.
5′-Cl-2′-MeO-BDE-66
4
′-OH-BDE-49, however, may be a metabolite of BDE-47
and possibly also of BDE-49 because the hydroxy group is
attached to the para position to the diphenyl ether bond. All
natural occurring OH-PBDEs so far reported in the literature
have the hydroxy group in the ortho position whereas mice
and rats exposed to BDE-47 have the hydroxy group also in
the meta and para positions (3). 4′-OH-BDE-49 is most
probably one of the two metabolites with the hydroxy group
in the para position as determined with EIMS formed in rat
and mice exposed to BDE-47 (3), and most likely also in fish
(exposed to BDE-47) (2) when comparing the characteristic
chromatographic peak patterns in these two works. Findings
of OH-PBDEs with the hydroxy group in a meta or para
positions may indicate PBDE metabolism.
SP 2331 (polar)
vs
PBDEs
MeO-PBDEs
BDE-66
BDE-155
2′-MeO-BDE-68
6-MeO-BDE-47
4
’-MeO-BDE-69
BDE-99
BDE-154
BDE-153
4′-MeO-BDE-121
5-MeO-BDE-47
2-MeO-BDE-123
4
-MeO-BDE-90
MeO-PBDEs
vs
MeO-PBDEs
6
6
2
-MeO-BDE-99
-MeO-BDE-47
-MeO-BDE-123
4′-MeO-BDE-49
4′-MeO-BDE-69
4-MeO-BDE-90
To summarize, the present knowledge about OH- and
MeO-PBDEs as well as the structures of the identified OH-
and MeO-PBDEs in the salmon blood indicate that all
determined as natural product from a marine sponge (i.e.,
-chloro-6-methoxy-2,2,′,4,4′-tetrabromodiphenyl ether (3-
3
TABLE 3. Step-by-Step Reduction of Possible MeO-PBDE and Methylated OH-PBDE Isomers Detected as Major Components of
These Classes of Compounds in Baltic Sea Salmon Blooda
a
The num ber of MeO-PBDE isom ers are given in parentheses.
VOL. 38, NO. 1, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1 7

compounds except one (4′-OH-BDE-49) originate from
natural sources. However, parts of 6-OH/ MeO-BDE-47, 6′-
OH/ MeO-BDE-49, and 6-OH-BDE-99 present in the salmon
blood might be metabolites of the corresponding PBDE
congeners. 4′-OH-BDE-49 is most likely a metabolite of BDE-
(19) Asplund, L.; Athanasiadou, M.; Sj o¨ din, A.; Bergman, A° .; B o¨ rjeson,
H. Ambio 1999, 28, 67-76.
(
20) Kierkegaard, A.; Sellstr o¨ m, U.; Bignert, A.; Olsson, M.; Asplund,
L.; Jansson, B.; de Wit, C. Organohalogen Compd. 1999, 40, 367-
3
70.
(
21) Olsson, A.; Ceder, K.; Bergman, A° .; Helander, B. Environ. Sci.
4
7 and possibly also one of BDE-49. Recently, Reddy at al.
Technol. 2000, 34, 2733-2740.
1
4
(
48) reported that C analysis is a useful tool to determine
(22) Hovander, L.; Malmberg, T.; Athanasiadou, M.; Athanassiadis,
I.; Rahm, S.; Bergman, A° .; Klasson Wehler, E. Arch. Environ.
Contam. Toxicol. 2002, 42, 105-117.
whether a compound is of natural or synthetic origin or both.
PBDEs that are manufactured from petrochemicals are “free”
(
23) Marsh, G.; Stenutz, R.; Bergman, A° . Eur. J. Org. Chem. 2003,
566-2576.
1
4
of C whereas natural products are not. Thus, proof of the
origin of the OH- and MeO-PBDEs can be obtained if
sufficient amount of the actual compounds can be isolated
2
(24) Marsh, G.; Bergman, A° .; Bladh, L.-G.; Gillner, M.; Jakobsson, E.
Organohalogen Compd. 1998, 37, 305-308.
1
4
(25) Ballschmiter, K.; Mennel, A.; Buyten, J. Fresenius J. Anal. Chem.
for a C measurement.
1
993, 346, 396-402.
(
(
(
26) Marsh, G.; Hu, J.; Jakobsson, E.; Rahm, S.; Bergman, A° . Environ.
Acknowledgments
Sci. Technol. 1999, 33, 3033-3037.
27) O¨ rn, U.; Eriksson, L.; Jakobsson, E.; Bergman, A° . Acta Chem.
Scand. 1996, 50, 802-807.
We are grateful to Ioannis Athanasiadis for MS assistance
and also to Hans B o¨ rjeson and Anna Malmv a¨ rn for their
valuable contributions. This project was made possible
through financial support from The Foundation for Strategic
Environmental Research, MISTRA, of the NewS program,
FORMAS, and from the EU R&D 5th framework program for
support of Compare.
28) Letcher, R. J.; Klasson Wehler, E.; Bergman, A° . Methyl Sulfone
and Hydroxylated Metabolites of Polychlorinated Biphenyls. In
New Types of Persistent Halogenated Compounds; Paasivirta, J.,
Ed.; Springer-Verlag: Berlin, 2000; Vol. 3, Chapter 11.
29) Campbell, J.-A. B.; Griffin, D. A.; Deinzer, M. L. Org. Mass
Spectrom. 1985, 20, 123-133.
(
(
30) Deinzer, M.; Lamberton, J.; Griffin, D.; Miller, T. Biomed. Mass
Spectrom. 1978, 5, 566-571.
Literature Cited
2
(
(
1) de Wit, C. Chemosphere 2002, 46, 583-624.
2) Kierkegaard, A.; Burreau, S.; Marsh, G., Klasson Wehler, E.; de
Wit, C.; Asplund, L. Organohalogen Compd. 2001, 52, 58-61.
3) O¨ rn, U.; Klasson Wehler, E. Xenobiotica 1998, 28, 199-211.
4) Hakk, H.; Larsen, G.; Klasson Wehler, E. Xenobiotica 2002, 32,
(
(
3
69-382.
(
(
(
(
(
5) M o¨ rck, A.; Hakk, H.; O¨ rn, U.; Klasson Wehler, E. Drug Metab.
Dispos. 2003, 31, 900-907.
6) Bowden, B. F.; Towerzey, L.; Junk, P. C. Aust. J. Chem. 2000, 53,
2
99-301.
7) Fu, X.; Schmitz, F. J.; Govindan, M.; Ackerman, R. A. J. Nat.
Prod. 1995, 58, 1384-1391.
8) Fu, X.; Hossain, M. B.; Schmitz, F. J.; van der Helm, D. J. Org.
Chem. 1997, 62, 3810-3819.
9) Schumacher, R. W.; Davidson, B. S. Tetrahedron 1995, 51, 10125-
1
0130.
(
10) Asplund, L.; Malmv a¨ rn, A.; Marsh, G.; Athanasiadou, M.;
Bergman, A° .; Kautsky, L. Organohalogen Compd. 2001, 52, 67-
7
0.
(
(
11) Anjaneyulu, V.; Nageswara Rao, K.; Radhika, P.; Muralikrishna,
M. Ind. J. Chem. 1996, 35B, 89-90.
12) Handayani, D.; Edrada, R. A.; Proksch, P.; Wray, V.; Witte, L.;
Van Soest, R. W. M.; Kunzmann, A.; Soedarsono J. Nat. Prod.
1
997, 60, 1313-1316.
13) Kuniyoshi, M.; Yamada, K.; Higa, T. Experientia 1985, 41, 523-
24.
14) Vetter, W.; Stoll, E.; Garson, M. J.; Fahey, S. J.; Gaus, C.; M u¨ ller,
J. F. Environ. Toxicol. Chem. 2002, 21, 2014-2019.
15) Vetter, W. Anal. Chem. 2001, 73, 4951-4957.
16) van Bavel, B.; Dam, M.; Tysklind, M.; Lindstr o¨ m, G. Organo-
halogen Compd. 2001, 52, 99-103.
(
(
5
(
(
(
(
17) Vetter, W.; Jun, W. Chemosphere 2003, 52, 423-431.
18) Haglund, P. S.; Zook, D. R.; Buser, H. R.; Hu, J. Environ. Sci.
Technol. 1997, 31, 3281-3287.
1
8
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 1, 2004