Phototransformations of polychlorobiphenyls in brij 58 micellar solutions

Article abstract of DOI:10.1021/es970960u

Our purpose in conducting these studies was to examine photolysis as a destructive process for polychlorobiphenyls (PCBs) extracted from soils with surfactant solutions. Surfactants have shown promise as agents for removing free phase and sorbed contaminants from soils, yet information on ultimate disposal options and recycle/recovery strategies for the surfactants is generally lacking. For arylhalides, photodechlorination may result in decontamination, eliminating the need to physically separate these contaminants from the washing solution. Photochemical reactions of the PCB congener mixture, Aroclor 1254, and the specific congener, 2,3,4,5- tetrachlorobiphenyl (2,3,4,5-TeCB), were investigated in aqueous solutions containing surfactant micelles with UV light at 253.7 nm. Photoreduction through photodechlorination was shown to be the main decay pathway in which lesser chlorinated congeners were formed as intermediates. In experiments with 2,3,4,5-TeCB, final noncarbon-containing products included Cl^- and H⁺, both produced nearly stoichiometrically from the starting materials. The quantum yield for decay of 0.1 μM 2,3,4,5-TeCB in 0.5 mM Brij 58 micellar solutions was over six times greater than in water alone. Sequential extraction from a soil and photoreduction of 2,3,4,5-tetrachlorobiphenyl by Brij 58 solutions proved to be limited by surfactant loss to the soil. Our purpose in conducting these studies was to examine photolysis as a destructive process for polychlorobiphenyls (PCBs) extracted from soils with surfactant solutions. Surfactants have shown promise as agents for removing free phase and sorbed contaminants from soils, yet information on ultimate disposal options and recycle/recovery strategies for the surfactants is generally lacking. For arylhalides, photodechlorination may result in decontamination, eliminating the need to physically separate these contaminants from the washing solution. Photochemical reactions of the PCB congener mixture, Aroclor 1254, and the specific congener, 2,3,4,5-tetrachlorobiphenyl (2,3,4,5-TeCB), were investigated in aqueous solutions containing surfactant micelles with UV light at 253.7 nm. Photoreduction through photodechlorination was shown to be the main decay pathway in which lesser chlorinated congeners were formed as intermediates. In experiments with 2,3,4,5-TeCB, final noncarbon-containing products included Cl- and H⁺, both produced nearly stoichiometrically from the starting materials. The quantum yield for decay of 0.1 μM 2,3,4,5-TeCB in 0.5 mM Brij 58 micellar solutions was over six times greater than in water alone. Sequential extraction from a soil and photoreduction of 2,3,4,5-tetrachlorobiphenyl by Brij 58 solutions proved to be limited by surfactant loss to the soil.

Full text of DOI:10.1021/es970960u

Environ. Sci. Technol. 1998, 32, 1989-1993
(2, 3). However, little information exists on ultimate disposal
Phototransformations of
Polychlorobiphenyls in Brij 58
Micellar Solutions
options and recycle/ recovery strategies for the surfactant(s).
Our hypothesis is that if surfactants are used in soil-washing
and pump-and-treat technologies to assist the solubilization
of arylhalide contaminants (1, 4, 5), photodechlorination may
result in arylhalide decontamination, eliminating the need
to separate these contaminants physically from the washing
solution.
W E I C H U , † C H A D T . J A F V E R T , * A N D
C L A U D E A . D I E H L
It is well established that many aryl halides undergo
photochemical homolysis to generate aryl and halogen
radicals (6-8). Bunce et al. (8) deduced that halogen
substituents on the aromatic rings increase intersystem
crossing, resulting in higher observed decay rates for the
more substituted aryl (Ar) halides. Accordingly, the following
mechanism was proposed, in which, upon absorption of light,
ArCl produces the singlet excited state, 1ArCl, which through
intersystem crossing (ISC) gives the triplet, 3ArCl, which upon
dissociation produces the aryl radical, Ar•, and a chlorine
atom, Cl•.
School of Civil Engineering, 1284 Civil Engineering Building,
Purdue University, West Lafayette, Indiana 47907-1284
K A R E N M A R L E Y A N D
R I C H A R D A . L A R S O N
Department of Natural Resources and Environmental
Sciences, The University of Illinois,
Urbana-Champaign, Illinois 61801-4723
Our purpose in conducting these studies was to examine
photolysis as a destructive process for
hν
ArCl 98 1ArCl ISC8 3ArCl f Ar• + Cl•
(1)
polychlorobiphenyls (PCBs) extracted from soils with
surfactant solutions. Surfactants have shown promise as
agents for removing free phase and sorbed contaminants
from soils, yet information on ultimate disposal options and
recycle/recovery strategies for the surfactants is generally
lacking. For arylhalides, photodechlorination may result in
decontamination, eliminating the need to physically separate
these contaminants from the washing solution. Pho-
tochemical reactions of the PCB congener mixture, Aroclor
1254, and the specific congener, 2,3,4,5-tetrachlorobiphenyl
(2,3,4,5-TeCB), were investigated in aqueous solutions
containing surfactant micelles with UV light at 253.7 nm.
Photoreduction through photodechlorination was shown
to be the main decay pathway in which lesser chlorinated
congeners were formed as intermediates. In experiments
with 2,3,4,5-TeCB, final noncarbon-containing products
included Cl- and H+, both produced nearly stoichiometrically
from the starting materials. The quantum yield for decay
of 0.1 µM 2,3,4,5-TeCB in 0.5 mM Brij 58 micellar
Bunce et al. (8) found that, in isooctane at 254 nm, ortho-
chlorinated biphenyls were more photolabile than non-ortho-
chlorinated biphenyls, presumably because of increased
efficiency in intersystem crossing leading to the triplet state,
and a reduced half-life of the triplet state due to a steric
effect. From a remediation standpoint, this is significant as
ortho-chlorines tend to be more recalcitrant to anaerobic
dehalogenation.
In the presence of hydrogen sources or other electron
donors, the reaction also may proceed through an electron-
transfer process. In the electron-transfer process, the donor
molecule furnishes an electron to the excited aryl chloride
(ArCl*), forming an unstable aryl radical anion which cleaves
to form the aryl radical and chloride.
D + ArCl* f D•+ + ArCl•- f Cl- + Ar• R-H8 ArH (2)
As in homolysis, cleavage of the C-Cl bond forms an aryl
radical, which may abstract hydrogen from the hydrogen
donor, forming the product ArH.
Freeman and coauthors (9-11) further defined three
subpathways within this general mechanism: singlet frag-
mentation, fission of the triplet, and fragmentation via the
triplet excimer. In the case of pentachlorobenzene, pho-
tolysis in CH3CN at 254 nm gave 1,2,3,5-tetrachlorobenzene
as a major product along with 1,2,4,5-tetrachlorobenzene
and 1,2,3,4-tetrachlorobenzene as minor products. This was
proposed to occur mainly by way of the triplet and triplet
excimer pathways with a minor contribution (2-6%) from
the direct fission of the singlet. In the presence of the electron
donor, triethylamine, a reversal in the regiochemistry occurs,
such that 1,2,4,5-tetrachlorobenzene is formed as the major
product due to the generation of pentachlorobenzene radical
anions through the transfer of an electron from triethylamine
to pentachlorobenzene. A similar reversal in regiochemistry
occurred when the reaction was carried out in 200 mM hexa-
decyltrimethylammonium bromide (CTAB) with 165 mM
triethylamine, presumably with a portion of the triethylamine
residing near the surface of the CTAB micelles, with pen-
tachlorobenzene concentrated within the micelles.
solutions was over six times greater than in water alone.
Sequential extraction from a soil and photoreduction of
2,3,4,5-tetrachlorobiphenyl by Brij 58 solutions proved to be
limited by surfactant loss to the soil.
Introduction
In a previous paper, we examined the facile nature of
photodechlorination reactions of chlorobenzenes exposed
to 254 nm ultraviolet light at room temperature and pressure
in aqueous solutions containing surfactant micelles (1). In
this paper, we extend this work to examine similar reactions
of Aroclor 1254 and the specific congener 2,3,4,5-tetrachlo-
robiphenyl (2,3,4,5-TeCB). Our purpose in conducting these
studies is to examine photolysis as a destructive process for
arylhalides extracted from soils with surfactant solutions.
Many studies have examined solubilization and/ or mobi-
lization of free phase or sorbed contaminants by surfactants
in ex situ soil-washing or in situ pump-and-treat operations
Few photochemical studies with chlorinated biphenyls
have been conducted in micellar solutions. Epling et al. (12)
reported quantum yields of about 0.1 for 4-chloro- and 4,4′-
dichloro-biphenyl in highly concentrated solutions (130 mM)
* Corresponding author phone: (765)494-2196.
† Current address: Civil and Structural Engineering, Hong Kong
Polytechnic University.
9
S0013-936X(97)00960-7 CCC: $15.00
Published on Web 05/20/1998
1998 Am erican Chem ical Society
VOL. 32, NO. 13, 1998 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 9 8 9

of Brij 58, both approximately 100-fold greater than in 9:1
acetonitrile-water mixtures. This suggests that similar
enhancements may occur for other congeners.
transferred to a 25 mL test tube, diluted with approximately
10 mL of distilled water, and carefully overlaid with 2 mL of
isooctane. Tubes were gently agitated just slightly off vertical
with a twist-action shaker in the dark at 20 °C for 24 h. With
this method, the extraction efficiency of each PCB congener
was determined to be approximately 100%.
However, for photolysis to be a feasible destructive process
for arylhalides extracted with surfactants from soils, infor-
mation regarding surfactant reuse or recovery and quantum
yields for decay in these complex systems must be known.
Previously, we and others have shown that precipitation (13)
and sorption (14) are processes that control recovery of
anionic and nonionic surfactants, respectively, from aqueous
slurries of some typical North American soils. In the latter
case, it was shown that sorption of polyethoxyated alcohol
surfactants increased with increasing polyethoxy chain length
within a homologous series of linear ethoxylated alcohols.
Sorption isotherms display plateau maxima at aqueous
concentrations exceeding the critical micelle concentration
(cmc) due to micelle formation. At surfactant doses sufficient
to form micelles, the aqueous phase of soil-water slurries
often is quite colored due to solubilization of humic material.
The effect of these materials on quantum yields in micellar
solutions has not previously been investigated.
PCBs were quantified with a Hewlett-Packard 5890 Series
II gas chromatograph (GC) equipped with a splitless injector,
an electron capture detector (ECD), an integrator, and a
28 m × 0.323 mm DB-5 column (film thickness of 0.25). In
experiments with 2,3,4,5-TeCB, the molecular masses of
products were determined by GC-MS analysis with a Finnigan
MAT 800 mass spectrometer and HP 5890A GC equipped
with the same length and diameter DB-5 column as that
used for product quantification. Relative retention times of
ECD-sensitive species were compared to those reported by
Mullin et al. (17) to further confirm identifications, with the
measured retention times of 2,3,4,5-TeCB and 4-CB used as
calibration points. For this purpose, the GC temperature
was ramped from 100 to 220 °C at a rate of 2.2 °C/ min.
Approximate yields of products were estimated from the
relative response factors (17). For analysis of Aroclor 1254,
the GC temperature was ramped from 65 °C (for 2 min) to
200 °C at a rate of 30 °C/ min, to 250 °C at a rate of 2 °C/ min,
and to 300 °C (for 2 min) at a rate of 30 °C/ min. In some
experiments, chloride was determined with a Dionex 2000i/
SP ion chromatograph, equipped with an Ion Pac AS4A
4 mm chromatographic column and conductivity detector.
The mobile phase was a carbonate-bicarbonate buffer (2.8
mM NaHCO3 and 2.3 mM Na2CO3). pH was measured with
a Fisher Scientific Accumet 925 pH/ ion meter and a general
purpose Orion glass Ross combination electrode.
The purpose of this study was to survey the reactivity of
Aroclor 1254 in various mixtures containing micelles of the
nonionic surfactant Brij 58, to investigate the photodechlo-
rination patterns of 2,3,4,5-TeCB in similar micellar solutions,
and to test the reactivity of 2,3,4,5-TeCB after its extraction
from contaminated soils with Brij 58.
Materials and Methods
Chem icals. All solvents were HPLC grade, purchased from
Fisher, Mallinckrodt, or Aldrich. 2,3,4,5-Tetrachlorobiphenyl
(2,3,4,5-TeCB) and 4-chlorobiphenyl (4-CB) were purchased
from Aldrich and used without further purification. Sodium
borohydride (SBH) (99%), purchased from Aldrich, was often
added as an additional electron donor. All solutions were
prepared with distilled water, made by distilling deionized
water which contained 0.01 M KMnO4 and 0.01 M KOH added
to the still to oxidize organic impurities. Brij 58 (C16H33(OCH2-
CH2)20(ave)OH) was purchased from Aldrich Chemical Co. and
used without further purification. As with other formulations
of nonionic ethoxylated surfactants, Brij 58 is a mixture of
homologues containing various ethoxy chain lengths.
Methods. All photochemical experiments were con-
ducted within a RPR-100 Rayonet photochemical reactor
equipped with a merry-go-round apparatus (Southern New
England Ultraviolet) placed in a dark room. All solutions
were illuminated in cylindrical quartz cuvettes placed within
the merry-go-round. Light sources were two 253.7 nm
phosphor-coated low-pressure mercury lamps (Rayonet). The
light intensity within each tube was determined from separate
Initial decay rates of 2,3,4,5-TeCB were determined from
the integrated first-order decay expression
Pt ) P0e-k t
(3)
p
where P0 and Pt are the concentrations of the 2,3,4,5-TeCP
at time 0 and t, respectively, and kp is the first order decay
rate constant (s-1). The quantum yields with the mono-
chromatic light source were calculated from kp and the light
intensity by (16)
kp
ΦP )
(4)
2.303Iλ,0ꢀp,λ
l
where ΦP is the quantum yield, Iλ,0 is the light intensity term
(einstein L-1 s-1) (16), ꢀp,λ is the molar absorptivity at 254 nm
(L mol-1 cm-1), and l is the cell path length term (cm).
In a final experiment, two samples of a soil, designated
EPA-11 (18) were contaminated with 0.58 and 1.4 mmol of
2,3,4,5-TeCB/ kg of soil by plating the compound on the walls
of glass containers (upon volatilization of the solvent carrier)
and mixing dry soil samples with water (1:10 soil mass to
volume) in the containers for 48 h. Subsamples (5.0 and
0.5 g) of the 0.58 and 1.4 mmol/ kg contaminated samples,
respectively, were extracted with 100 mL of water containing
1 mM Brij 58 for 4 h and centrifuged at 10 000 rpm for
10 min. The aqueous solutions were illuminated in the
photoreactor for three minutes, measuring the concentration
of 2,3,4,5-TeCB before and after illumination by extracting
2 mL of the solution with isooctane. This procedure was
repeated several times, each time using the same illuminated
solution to extract the solute from a new 5.0 or 0.5 g soil
subsample.
tubes containing the chemical actinometer, potassium
2-
ferrioxalate (15). The zero-order decay product, Fe(C2O4)2
,
was determined spectrophotometrically upon Fe(II) com-
plexation with 1,10-phenanthroline. The number of photons
entering each tube per unit volume per unit time, Iλ)254.7
(16), was generally about 5.6 × 10-6 einstein L-1 s-1
.
In each experiment, 5 mL aqueous solutions each
containing identical concentrations of 2,3,4,5-TeCB or Aroclor
1254, Brij 58, and often SBH and/ or NaOH were irradiated
in the photoreactor. Tubes were removed as a function of
time with the concentration at time 0 determined from
analysis of an unexposed sample to which SBH was not added.
The initial experimental stock solutions of 2,3,4,5-TeCB were
saturated solutions in water or the surfactant solution,
centrifuged to remove excess solid phase. Stock solutions
of the Aroclor were centrifuged and water-saturated. No
additional solvents were introduced.
Results and Discussion
Aroclor 1254 Photodecay. Aqueous saturated solutions of
Aroclor 1254 containing as additives 0.5 mM Brij 58, 50 mM
Compounds and products were extracted into isooctane
prior to analysis. To avoid the formation of emulsions in the
extraction step, 4 mL of each illuminated sample was
9
1 9 9 0 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 32, NO. 13, 1998

5 (SBH and NaOH), however, approximately 6% of the total
area remained on the chromatograph, despite the addition
of the strong base-stabilized reducing agent. Qualitatively,
Brij 58 at 0.5 mM was more effective than 50 mM SBH in
facilitating the overall reactions.
Sample GC traces of the Aroclor before and after il-
lumination are shown in Figure 1. This particular sample
contained 0.5 mM Brij 58, and the sampling time was 5 min.
Evident on the illuminated trace is the disappearance of
higher molecular weight homologues, the appearance of
peaks with shorter retention times, and the increased peak
areas of some lower molecular weight homologues.
Kinetic experiments were performed with Aroclor 1254
with these same additives. Relative peak areas for the 17
most prominent congener or coeluting congener peak areas
are shown in Figure 2 for solutions containing (1) water only,
(2) 0.5 mM Brij 58, and (3) 0.5 mM Brij 58, 50 mM SBH, and
10 mM NaOH, after 0, 1, 3, and 24 min illuminations. The
addition of Brij 58 with or without SBH enhanced the total
rate of decay. The formation and decay of some of the lower
chlorinated biphenyls (shorter retention times) is a direct
indication that the higher chlorinated biphenyls at longer
retention times are undergoing photodechlorination.
2,3,4,5-TeCB Photodecay. 2,3,4,5-TeCB was chosen as
a representative congener for more quantitative studies. Upon
illumination of 6.1 µM 2,3,4,5-TeCB containing 2.0 mM Brij
58, the chloride ion concentration increased from 0.74 to
27 µM within 60 min. During the same time period, the pH
of the reaction mixture decreased from 7.03 to 4.56. Assuming
unity activity coefficients to calculate proton ion concentra-
tions and no buffer capacity of the solution, the formation
of chloride and proton ions matches nearly stoichiometrically.
FIGURE 1. Photodecay of Arochlor 1254 in a Brij 58 micellar solution.
Arochlor 1254 GC trace before (top) and after (bottom) 5 min of
illumination at 253.7 nm. An aqueous saturated Arochlor 1254 was
-1
used. The light intensity was 5.7 × 10-6 einstein s-1 L .
SBH, and/ or 10 mM NaOH were illuminated in the photo-
chemical reactor for 24 min. The combination of additives
was (1) water only, (2) Brij 58 only, (3) Brij 58 and SBH, (4)
Brij 58 and NaOH, (5) SBH and NaOH, and (6) Brij 58, SBH,
and NaOH. Small volumes of each additive were combined
with the aqueous saturated solution of Aroclor 1254. After
24 min of illumination of sample 1 (water), about 15% of the
PCB peaks (by ECD), by number and area, remained, whereas
in the presence of Brij 58 (sample 2), only one peak was
evident. Samples with Brij 58 and SBH (sample 3), Brij 58
and NaOH (sample 4), and Brij 58, SBH, and NaOH (sample
6) resulted in no quantifiable peaks by the ECD. For sample
FIGURE 2. Photodecay of Arochlor 1254 with various additives. Brij 58 (5.0 × 10-4 M), SBH (0.05 M), and NaOH (0.01 M) were used as
additives to a saturated Aroclor 1254 solution. The cell path was 1.0 cm. Peak number indicates order of retention time from short to long.
9
VOL. 32, NO. 13, 1998 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 9 9 1

of the orginal 2,3,4,5-TeCB is decayed at this point, with nearly
stoichiometric recovery of the remaining as chlorinated
products. The photoreactivity of the ortho-chlorine in
organic solvents is well documented (see refs 6-8), so it is
not surprising that the majority of the parent compound
decays to 3,4,5-trichlorobiphenyl (3,4,5-TriCB). The data
indicate that this intermediate is reached through two
pathways. The first may occur either by homolysis, forming
a biphenyl radical, which may abstract a hydrogen from the
surfactant hydrocarbon, or possibly by the triplet excited
state abstracting a hydrogen from the surfactant hydrocarbon.
Both mechanisms result in photodechlorination and forma-
tion of the chlorine radical, which likely abstracts an electron
or H-atom from the surfactant, forming the observed chloride
ion product.
The second pathway leads to an overall 1,4-shift isomer-
ization reaction, forming 2′,3,4,5-TeCB in high yields; whether
radical separation occurs is unknown. To our knowledge,
this specific isomerization reaction has not previously been
reported. Isomerization of chlorinated benzenes has been
observed previously in aqueous micellar (1), methanolic (19),
and aqueous acetonitrile (20) solutions.
FIGURE 3. Photodecay of 2,3,4,5-TeCB in Brij 58 micellar solution.
Initial concentrations were 2.0 × 10-3 M Brij 58 and 6.1 × 10-6
M
2,3,4,5-TeCB. The cell diameter was 1.3 cm.
Many studies in homogeneous nucleophilic solvents (i.e.,
CH3OH) indicate that dimerization reactions may occur, but
indication of these products was not evident. This was
anticipated on the basis of the low occupancies of the PCB
congeners in the micelles that occur at the PCB-surfactant
ratios employed in this study, and because of the greater
hydrogen-donating ability of the surfactant compared to
methanol. With on average less than one congener per
micelle, the reactive aryl free radicals are likely isolated from
each other in restricted space.
The decay quantum yield of 0.1 µM 2,3,4,5-TeCB was
determined in solutions of (i) distilled water, (ii) 0.5 mM Brij
58, and (iii) 0.5 mM Brij 58, 50 mM SBH, and 10 mM NaOH
and found as 0.034, 0.21, and 0.24, respectively. These values
are based on the measured 2,3,4,5-TeCB molar absorptivity
of 8630 L mol-1 cm-1 at 254 nm in isooctane. The inability
of SBH to significantly enhance the rate above that of Brij 58
alone is not entirely surprising as the affinity of BH4- for the
hydrocarbon core of the micelle, where the excited state
2,3,4,5-TeCB is most likely located, is very low.
Soil Washing. The results of a series of extractions and
illuminations, using the same Brij 58 solution each time as
the extracting agent on a new subsample of the contaminated
soil, are displaced graphically in Figure 5. When a 0.5 g/ 100
mL soil solution contaminated with 1.4 mmol/ kg 2,3,4,5-
TeCB was used, over 80% of the 2,3,4,5-TeCB was recovered
from each subsample (≈5.8 µM) for the first three serial
extractions. A previous study with a historically contami-
nated sediment showed that increased mass recovery of PCB
congeners is possible by performing serial extractions on
the same soil (4). Once extracted, the 2,3,4,5-TeCB was
dechlorinated to less than 10 nM in less than 3 min of
irradiation. This time was not sufficient to dechlorinate
reaction products, although longer irradiation times could
have been used for this purpose. After three cycles, no loss
in extraction or photolysis efficiency was observed; however,
the fourth cycle showed loss of extraction efficiency, most
likely due to surfactant adsorption to the soil. Absorbance
of the solution at 254 nm increased after each cycle, indicating
accumulation of humic material in the reused micellar
solution.
FIGURE 4. Decay pathway of 20 µM 2,3,4,5-TeCB in 0.5 mM Brij 58
solutions. Numbers in parentheses indicate percent yield at 30 s
-1
with a light intensity of approximately 5.7 × 10-6 einstein s-1 L .
The monochlorobiphenyls are observed at later times.
The photodecay of 2,3,4,5-TeCB and the formation of Cl-
and H+ with time are shown graphically in Figure 3. Over
this time period, 10 peaks appear and most disappear on the
chromatrograms. All peaks were identified by GC-MS
analysis and relative retention times (RRTs) to be congeners
with the following substitution patterns: 2′,3,4,5-, 2,3,4,5-,
3,4,5-, 2,3,4-, 2,4,5-, 2,3,5- or 2′,3,5-, 3,4-, 3,5-, 4-, and 3-.
From this information, the decay pathway shown in Figure
4 is proposed. The percent yield values indicated by each
congener on this figure are those relative to the total molar
concentration of chlorinated products observed at 30 s for
a 20 µM 2,3,4,5-TeCB, 0.5 mM Brij 58 solution. About half
In a separate experiment, we irradiated 2,3,4,5-TeCB in
solutions containing Brij 58 and material extracted from soil
with this surfactant, having light absorbances at 254 nm of
about 0.5, and measured quantum yields equal to or slightly
exceeding those in similar solutions containing Brij 58 alone.
It is likely that light attenuation due to these materials at this
9
1 9 9 2 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 32, NO. 13, 1998

Literature Cited
(1) Chu, W.; Jafvert, C. T. Environ. Sci. Technol. 1994, 28, 2415-
2422.
(2) West, C. C.; Harwell, J. H. Environ. Sci. Technol. 1992, 26, 2324-
2330.
(3) Sabatini, D. A.; Knox R. C.; Harwell, J. H. Surfactant-Enhanced
Subsurface Remediation: Emerging Technologies; A. C. S.
Symposium Series 594; American Chemistry Society: Wash-
ington D. C., 1995; 300 pp.
(4) Jafvert, C. T.; Van Hoof, P. L.; Chu, W. Water Res. 1995, 29,
2387-2397.
(5) Jafvert, C. T.; Van Hoof, P. L.; Heath, J. K. Water Res. 1994, 28,
1009-1017.
(6) Ruzo L. O.; Zabik, M. J.; Schuetz, R. D. J. Am. Chem. Soc. 1974,
96, 3809-3813.
(7) Bunce, N. J.; Safe, S.; Ruzo, O. J. Chem. Soc., Perkin Trans. 1
1975, 1607-1610.
(8) Bunce, N. J.; Kumar, Y.; Ravanal, L.; Safe, S. J. Chem. Soc., Perkin
Trans. 2 1978, 880-884.
(9) Freeman, P. K.; Ramnath, N. J. Org. Chem. 1991, 56, 3646-3651.
(10) Freeman, P. K.; Hatlevig, S. A. Extended Abstracts of the Div. of
Environ. Chem.; ACS Annual meeting, April 5-10, San Francisco,
CA, American Chemical Society: 1992; pp 272-275.
(11) Freeman, P. K.; Lee, Y. S. J. Org. Chem. 1992, 57, 2846-2850.
(12) Epling G. A.; Florio E. M. J. Chem. Soc., Perkin Trans. 1 1988,
703-706.
(13) Jafvert, C. T.; Heath, J. K. Environ. Sci. Technol. 1991, 25, 1031-
1038.
FIGURE 5. Photolysis of 2,3,4,5-TeCB (left scale) extracted from
EPA-11 soil with 1 mM Brij 58, with reuse of the surfactant solution.
The absorbance at 254 nm (right scale) increases with each reuse.
(14) Yuan, C.; Jafvert, C. T. J. Contam. Hydrol. 1997, 28, 311-325.
(15) Hatchard, C. G.; Parker, C. A. Proc. R. Soc. London 1956, 235A,
518-536.
(16) Zepp, R. G. in The Handbook of Environmental Chemistry;
Hutzinger, O., Ed.; Springer-Verlag: Berlin, Germany, 1982; Part
B, Vol. 2, pp 19-41.
absorbance is offset by increased indirect or sensitized
photolysis.
Contaminating the soil with 0.58 mmol of 2,3,4,5-TeCB/
kg of soil, and extracting 5 g of this soil with 100 mL of water
containing 1 mM Brij 58, results in a concentration of 2,3,4,5-
TeCB similar to that in the above case. However, as shown
in Figure 5a, the light absorbance of this solution at 254 nm
exceeded 2.0, and illumination for 3 min was not sufficient
to dechlorinate the congener due to optical inhibition. The
second serial extraction resulted in readsorption of the
congener rather than extraction and a decrease in light
absorption, both indicating the absence of micelles. At this
soil concentration, loss of Brij 58 homologues to the soil
matrix inhibits effective contaminant extraction, yet excessive
extraction of humic materials inhibits photodechlorination.
(17) Mullin, M. D.; Pochini, C. M.; McCrindle, S.; Romkes M.; Safe,
S. H.; Safe, L. M. Environ. Sci. Technol. 1984, 18, 468-476.
(18) Hassett, J. J.; Means, J. C.; Barnwart, W. K.; Woods, S. G. Sorption
properties of sediments and energy-related pollutants; U. S.
EPA: Washington, D. C., 1980; 600/ 3-80-041.
(19) Choudhry, G. G.; Roof, A. A. M.; Hutzinger, O. Tetrahedron Lett.
1979, 22, 2059-2062.
(20) Choudhry, G. G.; Hutzinger, O. Environ. Sci. Technol. 1984, 18,
235-241.
Received for review November 3, 1997. Revised manuscript
received April 3, 1998. Accepted April 21, 1998.
ES970960U
9
VOL. 32, NO. 13, 1998 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 9 9 3