PDF Archive

Easily share your PDF documents with your contacts, on the Web and Social Networks.

Share a file Manage my documents Convert Recover PDF Search Help Contact

Environ Sci Technol 36, 2002, 1995 2002 .pdf

Original filename: Environ Sci Technol 36, 2002, 1995-2002.pdf
Title: No Job Name

This PDF 1.2 document has been generated by Parlance Publisher 5.0/(Xyvision Postscript Formatter) 3.0 3 / Acrobat Distiller Command 3.01 for Solaris 2.3 and later (SPARC), and has been sent on pdf-archive.com on 01/11/2015 at 17:59, from IP address 71.17.x.x. The current document download page has been viewed 620 times.
File size: 125 KB (8 pages).
Privacy: public file

Download original PDF file

Document preview

Environ. Sci. Technol. 2002, 36, 1995-2002

Photochemical Mass Balance of
2,3,7,8-TeCDD in Aqueous Solution
under UV Light Shows Formation of
Chlorinated Dihydroxybiphenyls,
Phenoxyphenols, and Dechlorination
Department of Chemistry, P.O. Box 3065, University of
Victoria, Victoria, British Columbia, Canada V8W 3V6
Contaminants Science Section, Institute of Ocean Sciences,
Department of Fisheries and Oceans Canada,
Sidney, British Columbia, Canada V8L 4B2
Department of Chemistry and Biochemistry, University of
Guelph, Guelph, Ontario, Canada N1G 2W1

Polychlorinated dibenzo-p-dioxins (PCDDs) are a class of
halogenated diaryl compounds that are environmentally
important because of their high toxicity and bioaccumulatory
properties. There is an incomplete understanding of their
photochemistry because the majority of photoproducts, as
indicated by incomplete mass balances, have not been
identified. We studied the photochemical transformation of
2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TeCDD) in
aqueous solution using 302 nm light. Our results allow for
the first photochemical mass balance (92-99%) for this
compound and confirm the operation of a novel photochemical
pathway, which gives rise to 2,2′-dihydroxy-4,4′,5,5′tetrachlorobiphenyl (4,4′,5,5′-TeCDHBP) at >50% conversion
from the starting material. Rearrangement from 2,3,7,8TeCDD to 4,4′,5,5′-TeCDHBP takes place following preferential
homolytic C-O bond cleavage via a proposed mechanism
analogous to the parent dibenzo-p-dioxin system.
Photochemical conversion from the starting material to
dechlorinated PCDDs or chlorinated phenoxyphenols are
minor pathways, although the exact contribution of the strictly
dechlorination pathways remains uncertain because of
the complexity of the system. The results suggest that the
photochemical conversion of PCDDs to chlorinated
dihydroxybiphenyls, which are also polychlorinated biphenyl
(PCB) metabolites, may be an important part of their
environmental fate.

The environmental fate of polychlorinated dibenzo-p-dioxins
(PCDDs) is of interest, owing to their high acute toxicity and
potential as immune disrupters (1). Evidence suggests PCDD
emissions into the environment increased after 1940, reaching
* Corresponding author phone: (250) 721-8976; fax: (250) 7217147; e-mail: pwan@uvic.ca.
10.1021/es011311s CCC: $22.00
Published on Web 04/02/2002

 2002 American Chemical Society

a peak in the 1960s and 1970s, and then declined up to the
present date (2). However, their persistence and ubiquity in
both biota and sediments at low ng‚kg-1 levels (3, 4) with
higher levels observed in waterways near industrial or
populated regions (5, 6) is of concern. Hence, comprehensive
investigations into the environmental fate of PCDDs are
Along with biologically mediated and various abiotic
reactions, photolysis of organic contaminants is an important
environmental transformation. In general, the majority of
UV light in natural fresh and coastal marine waters is absorbed
within the top 2 m of the water column (7). Rates of
photodegradation decrease quickly with increasing depth,
although photolysis still occurs in deeper waters provided
light of sufficient energy is present (8). Only small dependences on temperature have been noted for the photolysis of
organic compounds, with a 10 °C increase in temperature
enhancing the reaction rate by a factor of 1.15-1.50 (9, 10).
Hydroxyl radical concentrations in natural waters are also
typically too low to play a major role in photodegradation
(11). Bond cleavage following absorption of a photon may
proceed by homolytic, heterolytic, or mesolytic pathways, of
which homolysis (where the electrons forming the covalent
bond are distributed equally between previously bonded
atoms) is thought to be the major photochemical path for
Overall, photochemical decomposition quantum yields
for PCDDs decrease with increasing chlorination (12-14).
PCDDs substituted at the 2,3,7,8 positions are more photochemically stable than non-2,3,7,8-substituted congeners
(15-17). Relationships between the electronic and photolytic
properties of PCDDs have been observed. For example,
increasing PCDD quantum yields have been inversely related
to the largest positive charge on a Cl atom (qCl) and the dipole
moment and directly related to the ELUMO, EHOMO, and (ELUMO
- EHOMO) values (18, 19). Experiments with the octachlorodibenzo-p-dioxin (OcCDD) suggest either a short-lived
triplet or both singlet and triplet states may be involved in
homolytic dechlorination (20). Among the various PCDD
congeners, 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8TeCDD) is the most toxic, with LD50 values <2 µg‚kg-1 (21).
This congener has been assigned a toxic equivalence factor
of 1.0 relative to other polychlorinated diaryl systems such
as polychlorinated biphenyls (PCBs) and polychlorinated
dibenzofurans (PCDFs) (22). While numerous studies have
sought to achieve a quantitative photochemical mass balance
of 2,3,7,8-TeCDD on photolysis in either organic (23-27) or
aqueous solvents (17, 26, 28), identifying all of the photochemically converted starting material has remained elusive.
Early studies suggested reductive dechlorination was only a
minor photolytic pathway for 2,3,7,8-TeCDD, even in organic
solvents (29).
In comparison to other tetrachlorinated congeners,
2,3,7,8-TeCDD has the most rapid photodegradation rate in
solution but the slowest in the solid state (30). One study
found a linear relationship between the photolysis rates and
toxicity of various 2,3,7,8-substituted PCDDs. The author
suggested that the photolytic mechanism may have a related
intermediate to the biological end point, such that a common
molecular electronic requirement must be met (21). Relationships such as these in environmental science are
intriguing and deserve further attention.
Previous work has shown that the parent dibenzo-p-dioxin
system (DBD) photochemically rearranges in aqueous solution to 2,2-dihydroxybiphenyl (DHBP) as the major primary
pathway via a novel mechanism involving initial C-O bond



homolysis (31). We wanted to extend this work to a relevant
chlorinated dioxin system to investigate if a similar rearrangement could take place and form a major environmental
degradation pathway for 2,3,7,8-TeCDD.

Experimental Section
Reagents. DHBP, 2-phenoxyphenol (PP), 3-chlorophenol,
4-chlorophenol, and 3,4-dichlorophenol were purchased
from Aldrich and recrystallized prior to use to obtain >99%
purity. DBD and 2,2′-dihydroxy-4,4′,5,5′-tetrachlorobiphenyl
(4,4′,5,5′-TeCDHBP) had been previously synthesized and
characterized (25, 31). 2,3,7,8-TeCDD was obtained from
Cambridge Isotope Laboratories (CIL; Andover, MA) and
certified at a concentration of 67.8 ( 2.3 ng‚L-1 (23 °C) in
2,2,4-trimethylpentane (isooctane). Internal and recovery
standards (13C-labeled 2,3,7,8-TeCDD, 1,2,3,4-TeCDD, and
1,2,3,7,8,9-HxCDD; >99.8% labeled) were also obtained from
CIL. Trichlorinated dioxin impurities were certified at 1
ng‚mL-1 (23 °C) and were accounted for by use of appropriate
blanks (discussed next). All solvents (CH3CN, acetone,
toluene, hexane, and CH2Cl2) were of distilled-in-glass quality.
NaCl and NaSO4 used in sample extraction and cleanup were
washed with toluene, hexane, acetone, and CH2Cl2 and then
baked at 350 °C for a minimum of 12 h before use. A similar
treatment was performed on all glassware, syringes, and
Teflon and stainless steel utensils; no other material types
were used, to minimize contamination. Filter papers were
rinsed with toluene, hexane, acetone, and CH2Cl2 and dried
at 105 °C. N2 was ultrahigh purity grade, and deionized water
was obtained from a Millipore Milli-Q water system.
Procedure. The reaction vessel was a 3-mL quartz cuvette
with a Teflon cap. A 1.75 × 10-12 M solution of 2,3,7,8-TeCDD
in H2O/CH3CN (95:5, v/v) was placed in the cuvette (1695 pg
of 2,3,7,8-TeCDD in 3 mL of solvent) and sparged with N2 for
30 min. This concentration is below the reported solubility
limit of 6.0 × 10-11 M (32), and the 5% organic cosolvent
ensured complete dissolution. Dilutions of 2,3,7,8-TeCDD
were prepared by injecting 1695 pg of 2,3,7,8-TeCDD in 25
µL of isooctane cosolvent into the photochemical cuvette.
The isooctane solvent was then evaporated by gentle heating
under a steady stream of purified N2. To ensure the 2,3,7,8TeCDD was not volatilized, recovery of the 2,3,7,8-TeCDD
following isooctane evaporation was examined by gas
chromatography/high-resolution mass spectrometry (GC/
HRMS). The GC/HRMS analyses showed 95-104% recovery
of the original quantity of 2,3,7,8-TeCDD. Thus, negligible
quantities of isooctane (the amount remaining after evaporation would be far less than the 5% acetonitrile cosolvent)
were present during photolysis. This is important, as even
small quantities of isooctane present during photolysis could
act as a good hydrogen-donor (unlike acetonitrile) and
thereby skew the results in favor of strict dechlorination
products. Following evaporation of the isooctane cosolvent,
the 2,3,7,8-TeCDD solution was made up to 3 mL volume
with H2O/CH3CN (95:5, v/v). Duplicate photolyses were
performed at each irradiation time.
Several blanks were performed to ensure that the losses
due to volatilization during sparging of the H2O/CH3CN
solution and from thermal reactions were negligible. Levels
of all analytes in the blanks ranged from nondetectable to
2.1% of that found in the corresponding irradiated samples.
Three solvent and three thermal blanks (to ensure no “dark
reactions” were taking place) were performed. The levels of
analytes in the blanks were not subtracted from the reported
The cuvette was then sealed with the Teflon cap and
irradiated with a Ultra-Violet Products UVM-57 302 nm lamp
located 5 cm from the cuvette. This wavelength is near the
reported absorption maxima of 304-307 nm in various
organic solvents (26, 29, 33), although an absorption maxima



in water has yet to be determined, likely because of solubility
issues. Immediately following irradiation, 1000 pg of 13Clabeled 2,3,7,8-TeCDD was added to the cuvette as an internal
standard. Solutions were then transferred to a separatory
funnel, 3 mL of a saturated aqueous NaCl solution was added,
and the solution acidified to pH 2 with 6 N HCl. The resulting
aqueous fractions were extracted with 3 × 25 mL of CH2Cl2.
Sample extractions were evaporated to 3-5 mL on a rotary
evaporator (25-30 °C) and transferred to a glass centrifuge
tube with toluene and evaporated under N2 and gentle heating
to ∼100 µL. Samples were then transferred into amber
microvials with toluene and evaporated under nitrogen to
∼20 µL. Recovery standards (1000 pg each of 13C-labeled
1,2,3,4-TeCDD and 1,2,3,7,8,9-HxCDD) were added prior to
capping the microvial for GC/HRMS analysis to monitor
recovery of the internal standards and the analytes during
the sample workup procedure.
Sample extracts were analyzed by GC/HRMS using a VGAutospec high-resolution mass spectrometer (Micromass,
Manchester, U.K.) equipped with a Hewlett-Packard model
5890 series II gas chromatograph and a CTCA200S autosampler (CTC Analytics, Zurich, Switzerland). The GC was
operated in the splitless injection mode, and the splitless
injector purge valve was activated 2 min after sample
injection. The volume injected was 1 µL of sample plus 0.5
µL of air. The analyses were conducted using a 60 m DB-5
fused silica capillary column (0.25 mm i.d. with 0.1 µm film
thickness) from J&W Scientific (Folsom, CA), with UHP He
as the carrier gas at a constant head pressure of 25 psi to
maintain a linear velocity of 35 cm‚s-1. The temperature
program for PCDD analytes was as follows: the initial column
temperature was held at 100 °C for 2 min after injection and
increased at 20 °C‚min-1 to 200 °C, then at 1 °C‚min-1 to 215
°C, held for 7 min followed by a ramp of 4 °C‚min-1 to 300
°C, and held for 3 min. All sample injections were performed
using the CTC A200S autosampler. The MS was the only online detector attached to the GC system. The splitless injector
port, direct GC-MS interface, and the MS ion source were
maintained at 280, 290, and 290 °C, respectively.
The high-resolution MS was a sector instrument of EBE
geometry coupled to the GC via a standard Micromass GCMS interface. For PCDD analyses, the MS was operated under
positive EI conditions with the filament in the trap stabilization mode at 600 µA and an electron energy of 28-35 eV. The
instrument operates at 10 000 resolution, and data were
acquired in the selected ion monitoring (SIM) mode for
achieving maximum possible sensitivity. Two or more ions,
M+ and (M + 2)+, of known relative abundance were
monitored for each molecular ion cluster representing a
group of isomers, as were two for each of the 13C-labeled
surrogate standards. Compounds were identified only when
the GC/HRMS data satisfied all of the following criteria: (1)
two isotopes of the specific congeners were detected by their
exact masses with the mass spectrometer operating at 10 000
resolving power or higher during the entire chromatographic
run; (2) the retention time of the specific peaks was within
3 s to the predicted time obtained from analysis of authentic
compounds in the calibration standards; (3) the peak maxima
for both characteristic isotopic ions of a specific congener
coincided within 2 s; (4) the observed isotope ratio of the two
ions monitored per congener were within 15% of the
theoretical isotopic ratio; (5) the signal-to-noise ratio resulting
from the peak response of the two corresponding ions was
g3 for proper quantification of the congener.
To ensure that no water soluble phenolic photoproducts
avoided extraction and analysis, the aqueous fractions
remaining after extraction with CH2Cl2 were nonselectively
derivatized using established CH3I methods for the phenolic
metabolites of PCDD/Fs and PCBs (34-36). Solutions were
then processed and analyzed by GC/HRMS using the

FIGURE 1. Time-resolved photoproduct profiles for the irradiation
of 2,3,7,8-TeCDD in H2O/CH3CN (95:5, v/v) at 302 nm. Individual plots
show quantities of individual analytes from the major product classes
of (a) chlorinated DHBPs, (b) chlorinated PPs, and (c) PCDDs,
respectively. Abbreviations for individual compounds are provided
in the text. Error bars are ranges of analyte quantities from duplicate
instrument conditions as given previously except with the
following temperature program for SIM analyses: the initial
column temperature was held at 80 °C for 1 min after injection
and increased at 10 °C‚min-1 to 180 °C, then at 1 °C‚min-1
to 200 °C, followed by a ramp of 20 °C‚min-1 to 300 °C, and
held for 4 min. The splitless injector port, direct GC-MS
interface, and the MS ion source were maintained at 220,
220, and 280 °C, respectively. Full scans at 5000 resolution
were also performed on the derivatized samples using the
following temperature program: the initial column temperature was held at 100 °C for 1 min after injection and
increased at 10 °C‚min-1 to 180 °C, then at 1 °C‚min-1 to 200
°C, followed by a ramp of 20 °C‚min-1 to 280 °C, and held
for 6 min. The splitless injector port, direct GC-MS interface,

and the MS ion source were maintained at 220, 220, and 280
°C, respectively. Both SIM and full scans analyses showed no
methylated photoproducts.
Tri- through monochloro-2,2′-dihydroxybiphenyl compounds (TrCDHBPs, DiCDHBPs, and MoCDHBP, respectively) were prepared by irradiating a 500 ng‚mL-1 solution
of 4,4′,5,5′-TeCDHBP in toluene for 1 min at 302 nm. As the
percent conversion to dechlorinated products was not known,
the compounds formed were used only for identification
and not quantification. To identify and quantitate non-PCDD
photoproducts for which analytical standards were not
available (e.g., all chlorinated phenoxyphenols and MoCDHBP through TrCDHBPs), sample extracts were first analyzed
by GC/HRMS with the MS in full scan mode at 5000 resolution
using the following temperature program: the initial column
temperature was held at 100 °C for 1 min after injection and
increased at 10 °C‚min-1 to 180 °C, then at 1 °C‚min-1 to 200
°C, followed by a ramp of 20 °C‚min-1 to 280 °C, and held
for 3 min. The splitless injector port, direct GC-MS interface,
and the MS ion source were maintained at 220, 220, and 280
°C, respectively. The approximate GC retention times of these
photoproducts were estimated using the identification criteria
specified previously. Where library spectra were available,
these were used to aid in assigning tentative retention times.
Using these retention time windows, SIM analyses of the M+
and (M + 2)+ peaks were performed at 10 000 resolution on
all extracts using the temperature program described previously for methylated photoproducts with SIM. Compounds
were identified only if they met the criteria noted previously
and if their retention times were consistent with other
members of the same class (e.g., dichloro analytes must elute
prior to trichloro analytes of the same class) and between
classes (e.g., chlorinated phenoxyphenols will elute before
chlorinated DHBPs provided they have the same number of
chlorine substituents). Because of the low MS response for
analytes with quantities <0.30 pmol using SIM, sample
extracts from the duplicate photolyses performed at 5, 15,
30, 45, and 60 min were subsequently combined and
evaporated under nitrogen to ∼20 µL. The combined extracts
were then analyzed by GC/HRMS in SIM mode at 10 000
resolution using the temperature program described previously for methylated photoproducts with SIM.
For quantitation, relative response factors (RRFs) (the
instrument response per molecule of analyte, normalized to
a common analyte, which in this case was 4,4′,5,5′-TeCDHBP)
for the chlorinated phenoxyphenol series were assumed equal
to their chlorinated DHBP homologues. All compounds will
have a distinct RRF, and these are linked together by the
internal standard (IS); RRFs used in the present work have
been corrected to the IS to reduce variance between samples.
To estimate the RRFs of the MoCDHBP through TrCDHBP
products, we examined the effects of substitution on the
RRFs of 3-chlorophenol, 4-chlorophenol, and 3,4-dichlorophenol by monitoring the intensity of the M+ and (M + 2)+
peaks for each compound under SIM at 10 000 resolution.
The number and location of chlorine substituents on the
chlorophenol models was found to have only a minor effect
on the RRF, resulting in estimated IS-corrected RRFs of 1.10,
1.15, and 1.20 for TrCDHBP, DiCDHBP, and MoCDHBP,
respectively. While the authors recognize that there may be
some uncertainty in assigning RRFs to these compounds via
chlorophenol models, it is important to note that only
MoCDHBP is a significant contribution (>25%) to the mass
balance. The continuity of the mass balance (92-99%)
suggests that such errors were minimal.

Results and Discussion
Nature and Distribution of Photoproducts. In H2O/CH3CN
(95:5, v/v), the major primary photoproduct of 2,3,7,8-TeCDD
is 4,4′,5,5′-TeCDHBP. A time-resolved photoproduct profile



TABLE 1. Quantities of Photoproducts and Starting Material from the Photolysis of 2,3,7,8-TeCDD in H2O/CH3CN (95:5, v/v) and
Deionized H2O at 302 nma
H2O/CH3CN (95:5, v/v)
0 min


5 min

15 min

30 min

45 min

60 min

60 min


0.21 ( 0.03(4)
1.52 ( 0.12(29)
0.08 ( 0.02(2)
0.58 ( 0.07(11)

0.53 ( 0.01(10)
1.55 ( 0.10(29)
0.13 ( 0.03(3)
2.36 ( 0.11(45)

1.19 ( 0.04(23)
1.88 ( 0.08(36)
0.03 ( 0.02(1)
0.58 ( 0.07(11)
0.29 ( 0.06(6)

1.18 ( 0.08(22)
1.99 ( 0.05(38)
0.18 ( 0.03(3)

2.08 ( 0.07(40)
1.02 ( 0.12(19)
0.02 ( 0.01(0.4)
0.13 ( 0.03(2)

3.71 ( 0.19(71)
0.44 ( 0.09(8)
0.01 ( 0.01(0.2)
0.02 ( 0.01(0.4)
0.04 ( 0.02(0.6)

5.26 ( 0.26(100)
5.26 ( 0.26(100)

0.11 ( 0.02(2)
2.04 ( 0.06(39)
0.29 ( 0.05(5)
4.84 ( 0.37(92)

0.01 ( 0.01(<1)
0.05 ( 0.01(1)
0.52 ( 0.06(10)
0.03 ( 0.02(1)
0.02 ( 0.01(0.4)
5.21 ( 0.36(99)

0.16 ( 0.03(3)
0.30 ( 0.03(6)
0.48 ( 0.05(9)
0.07 ( 0.03(1)
4.99 ( 0.41(95)

0.27 ( 0.06(5)
0.73 ( 0.13(14)
0.57 ( 0.04(11)
0.04 ( 0.02(0.8)
4.96 ( 0.43(94)

0.19 ( 0.04(4)
0.01 ( 0.01(0.2)
0.01 ( 0.01(0.2)
0.73 ( 0.12(14)
0.82 ( 0.04(16)
0.05 ( 0.02(1)
5.05 ( 0.47(96)

0.07 ( 0.03(1)
0.21 ( 0.03(4)
0.39 ( 0.02(7)
0.02 ( 0.02(0.4)
0.05 ( 0.02(1)
4.95 ( 0.44(94)

a Quantities are averages of triplicate photolyses for 0 min, and duplicate photolyses for 5, 15, 30, 45, and 60 min and are presented in picomoles.
Error bars are range of replicate quantities. Percent of mass balance is in parentheses. Abbreviations: prefixes indicate degree of chlorination
(Mo ) mono, Di ) di, Tr ) tri, Te ) tetra); C ) chloro; DHBP ) dihydroxybiphenyl; DBD ) dibenzo-p-dioxin; CDD ) chlorodibenzo-p-dioxin; PP
) phenoxyphenol. b Authentic standard available for identification and quantification. c Relative response factors approximated using chlorophenol
models. d Two isomers were differentiated on the GC column but could not be identified. e Assumed to have same detector response as chlorinated
dihydroxybiphenyl analogues. f nd ) not detected.

for the H2O/CH3CN system is shown in Figure 1, and
individual analyte data is provided in Table 1 for clarity.
Compound abbreviations are explained in the text and below
Table 1. Also shown in Table 1 is the photoproduct distribution in 100% H2O after 60 min of irradiation. While nonextractable photoproducts as measured by 14C activity have
been reported for 2,3,7,8-TeCDD in distilled water (16, 37),
we were able to account for 92-99% of the starting material
and photoproducts extracted by CH2Cl2. The improvement
in mass balance over previous studies may be partly a result
of acidifying the photolyses after irradiation to pH 2 and
then extracting. The presence of several electron-withdrawing
Cl substituents on such phenolic moieties may reduce their
pKa to the point where they are substantially dissociated in
aqueous solution and hence nonextractable. Working at low
starting material concentrations (1.75 × 10-12 M) may also
assist in preventing photochemically induced polymerization
of 2,3,7,8-TeCDD.
Figure 1 shows that the majority of starting material
(>50%) is converted to 4,4′,5,5′-TeCDHBP, which may
subsequently dechlorinate by photolysis to TrCDHBP,
DiCDHBPs, and MoCDHBPs, and ultimately to the parent
DHBP. 4,4′,5,5′-TeCDHBP (in methylated form) from the
irradiation of 2,3,7,8-TeCDD in isooctane has been tentatively
reported (38); however, an authentic standard was not
available for definitive identification, nor was there a rational
mechanism proposed for its formation. These authors found
that the proposed 4,4′,5,5′-TeCDHBP product made up <10%
of converted starting material in the organic solvent, far less
than our value for a primarily aqueous system. Lower
conversion to 4,4′,5,5′-TeCDHBP in organic solvents is
expected, as a better H-donating solvent (e.g., isooctane)
will favor formation of 4,4′,5,5′-tetrachlorophenoxyphenol
(4,4′,5,5′-TeCPP) following homolytic C-O bond cleavage
(see Scheme 2 and the following discussion for mechanistic
details), rather than rearrangement to 4,4′,5,5′-TeCDHBP. In
comparison to other organic solvents, CH3CN is a poor H
donor and is not expected to significantly alter the aqueous
nature of the system. No TrCDHP was observed following
the photolyses, while two DiCDHP isomers were observed



after 5 min, reaching peak levels at 30 min. One isomer is
formed in clear preference to the other (11% vs 1% mass
balance after 30 min), but the lack of authentic standards
precluded identification. Significant quantities of a MoCDHBP isomer were present during all photolyses and accumulated to a peak of 38% mass balance after 45 min.
The large mass balance of MoCDHBP after 5 min (29%)
precedes the 4,4′,5,5′-TeCDHBP peak and suggests conversion of dechlorinated dioxins (e.g., those with three Cl
substituents) to the corresponding DHBP. These possible
pathways are illustrated in Scheme 1. The specifics of this
process are unclear, as several pathways are available to arrive
at MoCDHBP. Conversion of 2,7-DiCDD to 2,2′-dihydroxy4,4′-dichlorobiphenyl has been previously reported (38),
supporting our observations. In any case, this suggests that
the primary dechlorination pathway to 2,3,7-TrCDD may be
more significant than first thought on the basis of observed
dechlorination products, possibly accounting for ∼30%
conversion of starting material. Hence, several dechlorination
steps (taking place either from chlorinated DHBPs or PCDDs
with two Cl) and one rearrangement step (from a PCDD to
a chlorinated DHBP) would need to proceed more rapidly
by at least a factor of 3 than the rearrangement step for 2,3,7,8TeCDD to 4,4′,5,5′-TeCDHBP in order for part of the
MoCDHBP peak to precede that for 4,4′,5,5′-TeCDHBP. Both
PCDDs and PCBs have been shown to have increasing
quantum yields with decreasing chlorination (26), and this
increased efficiency may help explain our findings. Such
results are also consistent with our observations which show
that the majority of converted starting material resides either
as tetrachlori-nated DHBP or PP or as the unsubstituted
parent compounds (DHBP, PP, and DBD). Thus, build-up of
dechlorinated photoproducts in any of the three major
pathways shown in Scheme 1 is prevented by their more
rapid photolysis than the corresponding primary photoproducts.
Minor primary photoproducts include 4,4′,5,5′-tetrachlorophenoxyphenol (4,4′,5,5′-TeCPP) and 2,3,7-trichlorodibenzo-p-dioxin (2,3,7-TrCDD), which at no point contribute
greater than 5% and 2% of the mass balance, respectively.

SCHEME 1. Possible Photochemical Pathways of 2,3,7,8-TeCDD in Aqueous Solutiona


Each transformation requires the absorption of a photon.

As mentioned previously, the dechlorination pathway may
contribute more than the presence of observable products
indicates, possibly up to 30% conversion. The lack of
observable dechlorinated PCDDs is consistent with previous
studies in aqueous and organic solvents (<20% mass balance)
(27, 28). Quantum yields for dechlorinated PCDDs resulting
from 2,3,7,8-TeCDD are ∼4 times larger than the starting
material. Combined with the comparative molar absorbances,
the rate of photolysis of 2,3,7-TrCDD is ∼3 times more rapid
than 2,3,7,8-TeCDD (26). Hence, we did not observe or expect
an accumulation of dechlorination products. Preferential
C-Cl cleavage of PCDDs has been examined by both product
studies and computational methods (39, 40). Our results are
consistent with these reports in that we observe only 2,7and 2,8-DiCDD as dechlorination products, not 2,3-DiCDD.
Because of the low yields of these two isomers (see Table 1),
we were not able to distinguish which was the major product,
although 2,7-DiCDD is predicted to dominate (39, 40). No
2-MoCDD is observed at any time while the parent DBD

system appears after 30 min (3%), and its yield does not
substantially change for the 45 and 60 min irradiations (5%
and 4%, respectively).
The relative low yield of 4,4′,5,5′-TeCPP (5%) from 2,3,7,8TeCDD is analogous to the parent DBD system, which
displays <10% conversion to PP (31), and is a smaller yield
than similar photolyses (∼10% conversion) performed in
hexanes (25). Such a finding is consistent with the H2O/
CH3CN system (95:5, v/v) being a poorer H donor than organic
solvents, which would promote rapid H-abstraction following
homolytic C-O bond cleavage rather than rearrangement to
4,4′,5,5′-TeCDHBP. Conversion to 4,4′,5,5′-TeCPP peaks after
5 min (5%) and then declines with no mono- and trichlorophenoxyphenols (MoCPP and TrCPP, respectively) observed. One dichlorophenoxyphenol (DiCPP) isomer is seen
at low yield (1%) after 15 min, but conversion of dechlorinated
phenoxyphenols appears more rapid than production of
4,4′,5,5′-TeCPP; hence, we do not observe a build-up of monothrough trichlorophenoxyphenols. As with the DHBP and



FIGURE 2. Time-resolved contributions of the different compound
classes to the photoproduct distribution of 2,3,7,8-TeCDD in H2O/
CH3CN (95:5, v/v) at 302 nm. Error bars are ranges of analyte quantities
from duplicate photolyses.
DBD systems, the yield of parent PP increases after first
appearing at 30 min (9%) to a maximum after 60 min (16%).
These findings are summarized in Figure 2, which shows
the relative contributions of each photoproduct class (DHBPs
with 0-4 Cl, PPs with 0-4 Cl, PCDDs with 1-3 Cl plus the
parent DBD, and the quantity of starting material) to the
mass balance at each irradiation time. DHBPs dominate the
photoproduct distribution and account for up to 87% mass
balance after 15 min, whereas the summation of PPs and
PCDDs + DBD each account for <20% mass balance at all
times. On the basis of the observed photoproduct profiles,
the relative importance of each of the three major photochemical pathways for 2,3,7,8-TeCDD were assigned as ∼55%
through 4,4′,5,5′-TeCDHBP, ∼30% by 2,3,7-TrCDD, and ∼15%
via 4,4′,5,5′-TeCPP. Using the 2,3,7,8-TeCDD quantities
reported in Figure 2 at 0, 5, 15, and 30 min (converted to
molar concentrations), the rate constant for the disappearance of starting material was calculated (data at 45 and 60
min not used). As the rate-limiting step in the photochemical
conversion of 2,3,7,8-TeCDD to products is assumed to be
either C-O or C-Cl bond cleavage, a first-order decay process
was fitted to the 2,3,7,8-TeCDD data

) -k[2,3,7,8-TeCDD]
The resulting rate constant, k, was calculated to be 0.0931
min-1 (R2 ) 0.90), which provides a 2,3,7,8-TeCDD half-life
of 7.4 min under our experimental conditions. An estimate
of the 2,3,7,8-TeCDD sunlight photolysis half-life in water
has been reported elsewhere (33); the much longer t1/2 ) 118
h compared to our value for t1/2 ) 0.12 h suggests that our
light intensity was 2-3 orders of magnitude more intense at
302 nm than the solar spectrum. However, the qualitative
distribution of 2,3,7,8-TeCDD photoproducts in aqueous
solution should remain constant, regardless of light intensity.
From these findings, we propose a scheme illustrating
the photochemical pathways of 2,3,7,8-TeCDD in aqueous
systems (Scheme 1). As shown in this figure, pathways exist
for interconversion between secondary photoproducts. Previous studies have shown that 2-phenoxyphenol cannot
convert to 2,2-dihydroxybiphenyl (31); hence, the DHBP and
PP pathways are not connected. Although photochemical
ring closure of chlorinated phenoxyphenols to PCDDs has
been observed and is thought to occur from the triplet state
(41), they generally require ortho-Cl substituents on one of
the phenyl systems. While such a geometry is not expected
in this case, and thus we assume that the PCDD f chlorinated
phenoxyphenol pathways are not reversible, the increase in
2,3,7,8-TeCDD quantity from 30 to 45 min may contradict



FIGURE 3. Comparison of photoproduct class distribution after 60
min of irradiation of 2,3,7,8-TeCDD in H2O/CH3CN (95:5, v/v) and
100% H2O at 302 nm. Error bars are ranges of analyte quantities from
duplicate photolyses.
this claim. Reversibility in any of the photochemical pathways
shown in Scheme 1 would be quite unusual, although the
decrease in DHBP contribution to mass balance from 15 to
60 min is otherwise difficult to explain.
Photochemical C-O bond cleavage of polychlorinated
phenoxyphenols has been reported, leading to chlorophenols,
chlorocatechols, and chlorobenzenes. Both C-O and C-Cl
bond cleavage for these compounds are thought to take place
from the singlet state (41). In this case, we did not look for
C-O bond cleavage products from the phenoxyphenol
pathway, as their GC retention time under our instrumental
conditions (and with toluene as the solvent) would have made
them difficult to identify and quantitate. PCDD/Fs may be
formed from polychlorinated phenols (42), and this contribution cannot be ruled out. However, the nearly complete
mass balance for all extracts suggests these pathways are not
overly significant but may help explain the remaining 1-9%
of unaccounted for mass. Conversion of DHB to 2,2′,3trihydroxybiphenyl (THB) was not observed, although it had
been suggested in irradiations of lake water (17).
Effect of Photolysis on the System’s Toxicity. As 2,3,7,8TeCDD has the highest TEF (1.0) of all PCDD congeners for
which this rating scheme has been developed, the conversion
of 2,3,7,8-TeCDD to other PCDDs via photolysis can only
decrease the PCDD-TEF equivalent of a solution. However,
the aqueous photoproducts of 2,3,7,8-TeCDD, such as
chlorinated DHBPs and PPs, may also have serious health
effects which manifest themselves in a chronic, rather than
acute (as measured by TEFs), manner. Chlorinated DHBPs,
which may also arise through PCB metabolism, have been
identified in humans and wildlife (43, 44). While some
chlorinated DHBPs are readily excreted, bioaccumulation of
some congeners may be related to their high binding affinity
for the serum thyroid hormone binding protein transthyretin
(TTR) (45, 46). In addition, chlorinated DHBPs may exert
endocrine disrupting abilities (47, 48), with some congeners
having up to 10 times the binding affinity for TTR relative
to the endogenous and major blood thyroid hormone,
thyroxine (T4) (49). However, the structure of chlorinated
DHBPs is an important determinant of their biological effects;
those congeners with ortho-OH groups are significantly less
disruptive to biological systems than congeners with metaand para-OH functions (50). Because only ortho-OH biphenyls can be produced photolytically from PCDDs, it appears
as though the photodegradation of PCDDs will not produce
the most potent endocrine disrupting chlorinated DHBPs.
Chlorinated PPs have also been shown to have deleterious
biological effects such as mutagenicity (51) and toxicity (albeit
at levels much lower than that of 2,3,7,8-TeCDD) (52). Overall,

SCHEME 2. Proposed Mechanism for the Photolysis of 2,3,7,8-TeCDD in Aqueous Solution

it appears as though the aqueous photolysis of 2,3,7,8-TeCDD,
resulting in the production of chlorinated DHBPs and PPs,
will produce a solution of lower acute toxicity but one with
potential chronic endocrine disrupting and mutagenic effects.
Influence of Water on the Rate of Reaction. Figure 3
shows that the photolysis rate of 2,3,7,8-TeCDD is more rapid
in pure water than when an organic cosolvent is present
even in small amounts (e.g., 95:5 H2O/CH3CN, v/v). This
result is consistent with previous reports involving PCDFs
which showed reduced dechlorination rates in polar, hydroxylic solvents such as water (15, 28), as expected because
H-abstraction from H2O is not possible. Studies on the parent
DBD system also show photolytic rate enhancement by water
(31). Some authors have suggested the conversion of dibenzofuran (DF) to DHBP at 300 nm in natural waters (17). We
could not reproduce this result in H2O/CH3CN (1:1, v/v).
Our preliminary investigation by preparatory photolysis
showed no photochemical conversion of DF after up to 2 h
irradiation at 254 nm in a Rayonet photochemical reactor
with 16 lamps as followed by 1H NMR. In contrast to our
observations and those in the literature for PCDFs, PCDD
photolysis rates have generally been reported to be higher
in organic solvents (28, 53) with the presence of hydrides
enhancing the rate to an even greater extent (54). Our
observed rate increase with an increase in water content is
consistent with our proposed photolysis mechanism discussed next.
Mechanism of Rearrangement. A proposed mechanism
for 2,3,7,8-TeCDD photochemistry in aqueous solution is
shown in Scheme 2. Such knowledge of the mechanisms by
which environmental transformation of contaminants takes
place is critical in developing remediation processes and
understanding biological effects. This mechanism is extended
from work with the parent DBD system (31) and is based on
observations of 4,4′,5,5′-TeCDHBP, 4,4′,5,5′-TeCPP, and 2,3,7TrCDD as primary photoproducts. Two major photochemical
pathways exist for 2,3,7,8-TeCDD following absorption of a
photon. These are shown in Scheme 2 as homolytic cleavage
of a C-O bond (path a) or homolytic cleavage of a C-Cl
bond (path b). Photochemically generated carbocation

intermediates via either C-Cl bond homolysis followed by
electron transfer to the chlorine atom or via heterolytic C-Cl
cleavage have been previously proposed for PCDDs. It was
thought that such a carbocation intermediate would be
stabilized in the 2,3,7,8 (lateral) positions and destabilized
in the 1,4,6,9 positions, thereby explaining the more rapid
photolysis rates of 2,3,7,8 substituted PCDDs versus other
congeners (55). On the basis of work with OcCDD (20), we
believe that bond homolysis as shown in Scheme 2 (followed
either by rearrangement or H-abstraction from the solvent)
is the operative mechanism, rather than either heterolytic
cleavage or homolytic cleavage involving electron transfer.
Following cleavage by path a, a phenyl radical resides at
the 2-position, which may subsequently abstract an H atom
from the CH3CN solvent. This pathways results in the
production of 2,3,7-TrCDD. Homolysis of the C-O bond by
path b was shown to be the dominant pathway and results
in a biradical. Abstraction of two H atoms from the solvent
leads to 4,4′,5,5′-TeCPP. Conversely, the system may rearrange via a chlorinated spiro intermediate (CSP) to the
chlorinated biphenylquinone (BPQ), after which H-abstraction from CH3CN affords 4,4′,5,5′-TeCDHBP. Assuming this
mechanism operates much as with the parent DBD system,
conversion of 4,4′,5,5′-TeCPP to 4,4′,5,5′-TeCDHBP is thought
to be negligible, as already demonstrated for the analogous
non-chlorinated pair (31).
Our investigation did not attempt to show whether singlet
or triplet states were involved in the conversion of 2,3,7,8TeCDD, although work with OcCDD suggests that both states
are reactive (20). Because of the increase in spin-orbit
coupling, all PCDDs are expected to have shorter singlet and
triplet state lifetimes than the parent DBD system. The
increased photolysis rate in the pure aqueous system can be
qualitatively rationalized by examining the structures of
intermediates and products in Scheme 2. Homolytic cleavage
of the C-O bond (path b) results in the subsequent formation
of the chlorinated biradical, CSP, and BPQ intermediates,
which are all more polar species than the starting material
(or the radical intermediate from path a). Thus, a more polar
solvent enhances the rate of reaction by possibly lowering



the activation energies needed to generate these intermediates and the energies of the intermediates themselves. In
addition, pure water increases the overall exothermicity of
reaction through stabilization of the highly polar 4,4′,5,5′TeCDHBP product.

S.R. and P.W. acknowledge the Natural Sciences and Engineering Research Council of Canada (NSERC) for financial
support. S.R. thanks Maike Fischer at the Institute of Ocean
Sciences for the GC/HRMS analyses.

Literature Cited
(1) Staples, J. E.; Murante, F. G.; Fiore, N. C.; Gasiewicz, T. A.;
Silverstone, A. E. J. Immunol. 1998, 160, 3844-3854.
(2) Alcock, R. E.; Jones, K. C. Environ. Sci. Technol. 1996, 30, 31333143.
(3) Devault, D.; Dunn, W.; Bergqvist, P. A.; Wiberg, K.; Rappe, C.
Environ. Toxicol. Chem. 1989, 8, 1013-1022.
(4) Rappe, C.; Andersson, R.; Bergqvist, P. A.; Brohede, C.; Hansson,
M.; Kjeller, L. O.; Lindstrom, G.; Marklund, S.; Nygren, M.;
Swanson, S. E.; Tysklind, M.; Wiberg, K. Chemosphere 1987, 16,
(5) Evers, E. H. G.; Vanberghem, J. W.; Olie, K. Chemosphere 1989,
19, 459-466.
(6) Norwood, C. B.; Hackett, M.; Pruell, R. J.; Butterworth, B. C.;
Williamson, K. J.; Naumann, S. M. Chemosphere 1989, 18, 553560.
(7) Sinkkonen, S.; Paasivirta, J. Chemosphere 2000, 40, 943-949.
(8) Frank, R.; Klopffer, W. Chemosphere 1988, 17, 985-994.
(9) Larson, R. A.; Weber, E. J. Reaction mechanisms in environmental
organic chemistry; Lewis Publishers: Boca Raton, FL, 1994.
(10) Schwarzenbach, R. P.; Gschwend, P. M.; Imboden, D. M.
Environmental organic chemistry; John Wiley and Sons:
New York, 1993.
(11) Haag, W. R.; Yao, C. C. D. Environ. Sci. Technol. 1992, 26, 10051013.
(12) Choudhry, G. G.; Webster, G. R. B. Chemosphere 1985, 14, 893896.
(13) Choudhry, G. G.; Webster, G. R. B. Chemosphere 1985, 14, 9-26.
(14) Choudhry, G. G.; Webster, G. R. B. J. Agric. Food Chem. 1989,
37, 254-261.
(15) Dung, M. H.; Okeefe, P. W. Environ. Sci. Technol. 1994, 28, 549554.
(16) Friesen, K. J.; Muir, D. C. G.; Webster, G. R. B. Environ. Sci.
Technol. 1990, 24, 1739-1744.
(17) Friesen, K. J.; Foga, M. M.; Loewen, M. D. Environ. Sci. Technol.
1996, 30, 2504-2510.
(18) Chen, J. W.; Quan, X.; Peijnenburg, W. J. G. M.; Yang, F. L.
Chemosphere 2001, 43, 235-241.
(19) Chen, J. W.; Quan, X.; Schramm, K. W.; Kettrup, A.; Yang, F. L.
Chemosphere 2001, 45, 151-159.
(20) Konstantinov, A.; Bunce, N. J. J. Photochem. Photobiol., A 1996,
94, 27-35.
(21) Mamantov, A. Environ. Sci. Technol. 1984, 18, 808-810.
(22) Kutz, F. W.; Barnes, D. G.; Bottimore, D. P.; Greim, H.; Bretthauer,
E. W. Chemosphere 1990, 20, 751-757.
(23) Koshioka, M.; Yamada, T.; Kanazawa, J.; Murai, T. Chemosphere
1989, 19, 681-684.
(24) Buser, H.-R. J. Chromatogr. 1976, 129, 303-307.
(25) Konstantinov, A. D.; Johnston, A. M.; Cox, B. J.; Petrulis, J. R.;
Orzechowski, M. T.; Bunce, N. J.; Tashiro, C. H. M.; Chittim, B.
G. Environ. Sci. Technol. 2000, 34, 143-148.
(26) Dulin, D.; Drossman, H.; Mill, T. Environ. Sci. Technol. 1986,
20, 72-77.




(27) Colombini, M. P.; DiFrancesco, F.; Fuoco, R. Microchem. J. 1996,
54, 331-337.
(28) Kim, M. K.; O’Keefe, P. W. Chemosphere 2000, 41, 793-800.
(29) Crosby, D. G.; Wong, A. S.; Plimmer, J. R.; Woolson, E. A. Science
1971, 173, 748.
(30) Nestrick, T. J.; Lamparski, L. L.; Townsend, D. I. Anal. Chem.
1980, 52, 1865-1874.
(31) Guan, B.; Wan, P. J. Photochem. Photobiol., A 1994, 80, 199210.
(32) Marple, L.; Brunck, R.; Throop, L. Environ. Sci. Technol. 1986,
20, 180-182.
(33) Podoll, R. T.; Jaber, H. M.; Mill, T. Environ. Sci. Technol. 1986,
20, 490-492.
(34) Tulp, M. T. M.; Olie, K.; Hutzinger, O. Biomed. Mass Spectrom.
1977, 4, 310.
(35) Tulp, T. H. M.; Olie, K.; Hutzinger, O. Biomed. Mass Spectrom.
1978, 5, 224.
(36) Rozemeijer, M. J. C.; Olie, K.; deVoogt, P. J. Chromatogr., A 1997,
761, 219-230.
(37) Dougherty, E. J.; Mcpeters, A. L.; Overcash, M. R. Chemosphere
1991, 23, 589-600.
(38) Kieatiwong, S.; Nguyen, L. V.; Hebert, V. R.; Hackett, M.; Miller,
G. C.; Miille, M. J.; Mitzel, R. Environ. Sci. Technol. 1990, 24,
(39) Makino, M.; Kamiya, M.; Matsushita, H. Environ. Int. 1993, 19,
(40) Makino, M.; Kamiya, M.; Matsushita, H. Chemosphere 1992, 24,
(41) Freeman, P. K.; Srinivasa, R. J. Agric. Food Chem. 1983, 31, 775780.
(42) Skurlatov, Y. I.; Ernestova, L. S.; Vichutinskaya, E. V.; Samsonov,
D. P.; Pervunina, R. I.; Semenova, I. V.; Shvydky, V. O. Acta
Hydrochim. Hydrobiol. 1998, 26, 31-35.
(43) Bergman, A.; Klassonwehler, E.; Kuroki, H. Environ. Health
Perspect. 1994, 102, 464-469.
(44) Hovander, L.; Malmberg, T.; Athanasiadou, M.; Athanassiadis,
L.; Rahm, S.; Bergman, A.; Wehler, E. K. Arch. Environ. Contam.
Toxicol. 2002, 42, 105-117.
(45) Brouwer, A.; Vandenberg, K. J. Toxicol. Appl. Pharmacol. 1986,
85, 301-312.
(46) Brouwer, A.; Klassonwehler, E.; Bokdam, M.; Morse, D. C.; Traag,
W. A. Chemosphere 1990, 20, 1257-1262.
(47) Toppari, J.; Larsen, J. C.; Christiansen, P.; Giwercman, A.;
Grandjean, P.; Guillette, L. J.; Jegou, B.; Jensen, T. K.; Jouannet,
P.; Keiding, N.; Leffers, H.; McLachlan, J. A.; Meyer, O.; Muller,
J.; Rajpert De Meyts, E.; Scheike, T.; Sharpe, R.; Sumpter, J.;
Skakkebaek, N. E. Environ. Health Perspect. 1996, 104, 741803.
(48) Korach, K. S.; Sarver, P.; Chae, K.; Mclachlan, J. A.; Mckinney,
J. D. Mol. Pharmacol. 1988, 33, 120-126.
(49) Brouwer, A.; Morse, D. C.; Lans, M. C.; Schuur, A. G.; Murk, A.
J.; Klasson-Wehler, E.; Bergman, A.; Visser, T. J. Toxicol. Ind.
Health 1998, 14, 59-84.
(50) Schuur, A. G.; Legger, F. F.; van Meeteren, M. E.; Moonen, M.
J. H.; van Leeuwen-Bol, I.; Bergman, A.; Visser, T. J.; Brouwer,
A. Chem. Res. Toxicol. 1998, 11, 1075-1081.
(51) Onodera, S.; Takahashi, M.; Ogawa, M.; Suzuki, S. Jpn. J. Toxicol.
Environ. Health 1995, 41, 212-219.
(52) Hamilton, S. J.; Cleveland, L.; Smith, L. M.; Lebo, J. A.; Mayer,
F. L. Environ. Toxicol. Chem. 1986, 5, 543-552.
(53) Ritterbusch, J.; Vogt, R.; Lorenz, W.; Bahadir, M.; Hopf, H.
Chemosphere 1994, 29, 457-464.
(54) Epling, G. A.; Qiu, Q. W.; Kumar, A. Chemosphere 1989, 18, 329332.
(55) Mamantov, A. Chemosphere 1985, 14, 897-900.

Received for review September 25, 2001. Revised manuscript
received January 9, 2002. Accepted February 13, 2002.

Related documents

environ sci technol 36 2002 1995 2002
photochem photobiol sci 4 2005 876 886
environ int 35 2009 425 437
environ int 32 2006 575 585
wat res 37 2003 551 560
sierra rayne phd thesis

Related keywords