United States
Environmental Protection
Agency
EPA/540/SR-94/531
March 1995
SUPERFUND INNOVATIVE
TECHNOLOGY EVALUATION
Emerging Technology
Summary
Bench-Scale Testing of
Photolysis, Chemical Oxidation,
and Biodegradation of PCB
Contaminated Soils, and
Photolysis of TCDD
Contaminated Soils
The tests reported herein were con-
ducted by IT Corporation, Knoxville, TN,
to investigate the feasibility of a two-
phase detoxification process that would
potentially have application to the treat-
ment of soils contaminated with PCBs
and 2,3,7,8-tetrachlorodibenzo-p-dioxin
(TCDD). The first step in the process
was to degrade the organic contami-
nants by using ultraviolet (UV) radia-
tion. The source of UV radiation was
either artificial UV light or natural sun-
light, but generally photolytic processes
are more rapid with artificial UV light.
Alternatively, advanced oxidation pro-
cesses, such as iron catalyzed hydro-
gen peroxide (Fenton's Reagent), were
used to provide primary contaminant
degradation. Both photolysis and
chemical oxidation were expected to
convert contaminants to more easily
biodegradable compounds. Biological
degradation, the second step, was then
used to further destroy organic con-
taminants and detoxify the soil. Bio-
degradation is enhanced by the addition
of microorganisms and nutrients to the
UV-treated soil.
The results of bench-scale testing on
degradation of TCDD by using UV pho-
tolysis, and PCB degradation by using
both UV photolysis and chemical oxi-
dation, indicate that there was no ap-
parent destruction of the dioxin on the
soil, 23% to 69% destruction of PCBs
in UV Tests, and 0% to 53% reduction
in chemical oxidation tests.
Bioslurry experiments evaluated the
biological reduction of PCB congeners
in surfactant/UV-treated and untreated
soils. Experiments were also conducted
to evaluate the impact of PCB-blodeg-
radation inducers, biphenyl and 4-
bromobiphenyl, on congener removal.
Bioslurry treatment did not provide sig-
nificantly different results for the UV-
treated surface soil versus the
untreated soil.
This Project Summary was developed
by EPA's Risk Reduction Engineering
Laboratory, Cincinnati, OH, to announce
key findings of the SITE emerging tech-
nology that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Polychlorinated dioxins, furans, and bi-
phenyls (PCBs) are the most recalcitrant
Printed on Recycled Paper
-------
vvEPA
United States
Environmental Protection
Agency
EPA/540/SR-94/531
March 1995
SUPERFUND INNOVATIVE
TECHNOLOGY EVALUATION
Emerging Technology
Summary
Bench-Scale Testing of
Photolysis, Chemical Oxidation,
and Biodegradation of PCB
Contaminated Soils, and
Photolysis of TCDD
Contaminated Soils
The tests reported herein were con-
ducted by IT Corporation, Knoxville, TN,
to investigate the feasibility of a two-
phase detoxification process that would
potentially have application to the treat-
ment of soils contaminated with PCBs
and 2,3,7,8-tetrachlorodibenzo-p-dioxin
(TCDD). The first step in the process
was to degrade the organic contami-
nants by using ultraviolet (UV) radia-
tion. The source of UV radiation was
either artificial UV light or natural sun-
light, but generally photolytic processes
are more rapid with artificial UV light.
Alternatively, advanced oxidation pro-
cesses, such as iron catalyzed hydro-
gen peroxide (Fenton's Reagent), were
used to provide primary contaminant
degradation. Both photolysis and
chemical oxidation were expected to
convert contaminants to more easily
biodegradable compounds. Biological
degradation, the second step, was then
used to further destroy organic con-
taminants and detoxify the soil. Bio-
degradation is enhanced by the addition
of microorganisms and nutrients to the
UV-treated soil.
The results of bench-scale testing on
degradation of TCDD by using UV pho-
tolysis, and PCB degradation by using
both UV photolysis and chemical oxi-
dation, indicate that there was no ap-
parent destruction of the dioxin on the
soil, 23% to 69% destruction of PCBs
in UV Tests, and 0% to 53% reduction
in chemical oxidation tests.
Bioslurry experiments evaluated the
biological reduction of PCB congeners
in surfactant/UV-treated and untreated
soils. Experiments were also conducted
to evaluate the impact of PCB-blodeg-
radation inducers, biphenyl and 4-
bromobiphenyl, on congener removal.
Bioslurry treatment did not provide sig-
nificantly different results for the UV-
treated surface soil versus the
untreated soil.
This Project Summary was developed
by EPA's Risk Reduction Engineering
Laboratory, Cincinnati, OH, to announce
key findings of the SITE emerging tech-
nology that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Polychlorinated dioxins, furans, and bi-
phenyls (PCBs) are the most recalcitrant
Printed on Recycled Paper
-------
environmental contaminants. To remove
or destroy contamination on soils, rela-
tively high-temperature thermal treatment
or harsh chemical treatment is required.
These treatments are expensive and can
significantly alter the composition of the
soil.
The use of UV light to destroy dioxins
or RGBs in solutions has long been known.
More recently, the use of UV light to de-
stroy these compounds on soils was in-
vestigated. Typically, the reaction is aided
by the presence of a solvent or solubiliz-
ing aid, such as a surfactant, which is
transparent to the UV radiation in the re-
gion of activity (generally 254 nanometers)
and which has increased solubility for the
contaminants being destroyed. The irra-
diation process can be performed on ex-
cavated soils or in situ with the use of
enhanced radiation from lamps or natural
sunlight. The process usually involves the
continued application of the solubilizing
aid and continued exposure of fresh sur-
face to the irradiation source. The solubil-
ity aid helps to transfer the contaminant
from the pores of the soil to the surface of
soil where the reactions can take place.
The surfactant or solubilizing aid may also
act as a medium for the degradation pro-
cess by providing labile protons to allow
the reaction to proceed more easily.
Chemical oxidation by Fenton's Reagent
has been used to destroy organic com-
pounds such as formaldehyde, azo dyes,
and chlorinated phenols in groundwaters
and wastewaters. The reaction is ideally
performed at a pH of 2 to 4 by using
hydrogen peroxide as the oxidant in the
presence of a ferrous salt. Ferrous ions
catalyze the decomposition of hydrogen
peroxide. In the process of decomposi-
tion, the reactive hydroxyl radical is pro-
duced, and it is capable of oxidizing
organic contaminants. If, however, the de-
sired oxidation reaction is slow, significant
amounts of hydrogen peroxide can be con-
sumed in unproductive decomposition in-
stead of participating in the desired
process. Reaction conditions must be es-
tablished to provide useful rates of con-
taminant oxidation with efficient use of
hydrogen peroxide reagent.
Aerobic biodegradation of the lower chlo-
rinated PCB congeners (1 to 3 chlorines)
has been well documented. The more
highly chlorinated congeners are, however,
generally resistant to microbial attack, al-
though there have been reports of micro-
bial degradation of the higher-chlorinated
PCB congeners (>4 chlorines). Biological
degradation of PCB congeners is highly
affected by the chlorination pattern and
the number of chlorines per biphenyl. Fur-
ther hindering microbial biodegradation of
PCB is their hydrophobicity, which inhibits
their bioavailability. To increase the rate
and extent of PCB biodegradation, two
conditions are necessary: the bioavailability
of the PCB should be increased, and the
chlorination level per biphenyl ring should
be decreased. This study addressed the
bioavailability and microbial attack of PCB
after the combined surfactant/UV treat-
ment of highly contaminated PCB soil.
Process Description
This two-phase treatment is envisioned
as a potential in-situ process for shallow
contamination on soils. More probable,
however, is the use of the technology for
ex-situ on-site treatment of excavated soils
in a specially constructed shallow treat-
ment basin—one meeting the requirements
of the Resource Conservation and Recov-
ery Act (RCRA). The entire process may
require longer treatment times than other
technologies, but have a trade off in
economy. In addition, the only residue gen-
erated from this combination of technolo-
gies is soil contaminated with surfactants
and the end metabolites of the biodegra-
dation processes. The end metabolites will
depend on the original contaminants, and
the surfactants are common materials used
in agricultural formulations.
UV Photolysis Performance
UV photolysis testing was performed on
three soils, one containing TCDD con-
tamination and two containing PCB con-
tamination. The tests were conducted
independently by using a medium-pres-
sure Hg lamp, a 10-Hz pulsed lamp and
sunlight, and by employing different sur-
factants and surfactant application proce-
dures. Tests were carried out with either
the 450-watt medium pressure Hg lamp
or the 70 watts/in, pulsed lamp operating
approximately 10 in. above the soil with a
parabolic reflector above the lamp.
A composited TCDD soil from a Vertac
site in Jacksonville, AR, using two surfac-
tant levels, 2.5% and 5% by weight of the
dry soil, was tested. TCDD concentrations
on the soil ranged from about 200 to 300
ppb. The soil was mixed and sprayed at
1/2-hr intervals with either surfactant solu-
tion or water for a total irradiation time of
48 hr. Results from these tests indicated
no apparent destruction of the dioxin on
the soil in any of the tests.
Surface soil from a Texas Eastern Gas
Pipeline site in Armaugh, PA, contami-
nated with about 10,000 ppm PCBs
(Aroclor 1248) and a pit soil from the
same site containing about 200 ppm PCBs
were tested under the same conditions
given above using different surfactants,
application procedures, soil mixing, etc.
The results of all of the experiments are
summarized in Table 1.
Chemical Oxidation
Performance
Five batch experiments using the H20,/
Fe (Fenton's) reagent system were done
at ambient temperature. All five used the
same clay/humic surface soil that was used
in the UV photolysis testing. This soil pro-
vided samples for treatment that ranged
from 5000 to 10,000 ppm PCBs (Aroclor
1248). Conditions were established to pro-
vide the best opportunity for observing an
effect due to treatment. Each experimen-
tal mixture was pH adjusted to a pH be-
tween 2 and 4 and continuously stirred.
Hydrogen peroxide concentration was
monitored throughout each experiment and
additions were made as necessary to maxi-
mize concentration. Reagent-to-soil ratios
were high, and Fe concentrations were
varied between experiments to investigate
the effect. The treated soil was analyzed
either at the end of the treatment time or
sampled at selected times throughout the
run if sufficient soil was used in the test.
Results from testing are summarized in
Table 2.
Biological Treatment
Performance
The ability of selected organisms to
biotransform PCB congeners in surfac-
tant/UV-treated and untreated soil was
evaluated during two bioslurry treatment
experiments. The first bioslurry experiment
evaluated the biological reduction of PCB
congeners in surfactant/UV-treated and
untreated soils. A subsequent enhanced
bioslurry experiment evaluated the impact
of PCB-biodegradation inducers on con-
gener removal. Previous studies show that
the addition of biphenyl, 4-bromobiphenyl
(4-BB), 4-chlorobiphenyl, 2-chlorobiphenyl,
or other monochlorobiphenyls have in-
duced and enhanced aerobic PCB bio-
degradation.
The bioslurry experiments were con-
ducted under aerobic conditions at 25°C
using PCB-degrading organisms from two
sources. PCB-degrading organisms were
isolated from an impacted New England
Superfund Site soil (BAG 17). In addition,
known-PCB degrading microorganisms
were obtained from General Electric Cor-
poration (GE; H850). Three PCB-contami-
nated soils were evaluated for biological
reduction of PCB congeners. Soils em-
ployed were identified as untreated sur-
face soil from the UV photolysis testing,
surfactant/UV-treated surface soil, and
New England Superfund Site soil. In sepa-
rate tests, each soil was treated with the
bacteria cultures, BAG 17 and H850. In
-------
Table 1. Summary ofUV Photolysis Results on PCB Contaminated Soil
Test
1
2
3
4
5
6
7
8
9
10
11
Soil Type
surface
surface
surface
pit
pit
pit
pit
pit
pit
surface
surface
Soil Depth in.
0.25
0.25
0.25
0.5
0.5
0.5
1.0
1.0
1.0
0.25
0.25
Surfactant, %
2
2
2
2
2
2
4.5
2
0
2.5
2.5
Lamp Type
pulsed
medium pressure
pulsed
medium
pressure Hg
medium
pressure Hg
pulsed
pulsed
solar irradiation
solar irradiation
solar irradiation
medium Hg
medium Hg
Temperature
°C
25
28
40
30
28
28
30-40
30-40
30-40
Time (Hr)
12
7
7
16
16
12
25 days
25 days
25 days
20
20
Initial PCB
Cone., ppm
7240
7430
8440
140
157
170
132
159
171
10000
10000
Final PCB
%
Reduction
<15
<15
33
30
13
23
<15
<15
<15
52*
32
'Increase in concentration noted for di-PCBs, decrease in concentration for tetra through hepta-PCBs.
Table 2. Summary of Chemical Oxidation Testing
Test
1
2
3
4
5
Amount of Soil, g
10
8.0
8.1
170
196
w/surfactant
Water/Soil Ratio
9.7
8.4
9.5
10.1
8.0
pH
2.8
2.5
2.2
3.1
2.9
Fe, % of Soil
2.5
0.1
0.5
.09
.09
H2O2Conc. (%)
Average
.07
1.8
0.87
1.6
0.88
Time (Hr)
162
118
118
845'
184
Percent
Reduction of
Starting PCB
Concentration
44
55
45
35
<,5
'No further decrease in PCB concentration observed after 211 hr.
-------
addition a killed control sample was pre-
pared for each soil. Table 3 summarizes
the reductions measured for the PCB ho-
mologs after four weeks.
In the enhanced bioslurry experiment,
the effect of adding 1,000 mg/L biphenyl
inducer and 1,000 mg/L 4-BB inducer on
PCB biodegradation was investigated. The
BAG 17 culture was used with the New
England Superfund Site soil and the un-
treated soil because these combinations
demonstrated the highest PCB reductions
in the bioslurry experiment. The soils were
tested with both inducers. Table 4 sum-
marizes the reductions measured for the
PCB homologs after one week.
Conclusions
UV photolysis treatment gave no de-
tectable decrease in soil TCDD concen-
tration in the tests conducted and minimal
reduction of PCBs. The PCB reductions
ranged from less than 15% to a maximum
of 52% The decreases in concentration
were highest for higher level chlorinated
PCBs (tetra through hepta-PCBs). In some
cases greater decreases in concentration
of lower chlorine level PCBs were noted.
These were suspected to have been lost
due to volatilization in tests at increased
temperatures caused by the heat gener-
ated by the UV lamp, along with the high
soil surface area exposed.
Chemical oxidation with Fe catalyzed
hydrogen peroxide also provided minimal
reduction of PCBs on the highly contami-
nated surface soil tested. The PCB con-
centration reductions ranged from less than
15% to a maximum of 55% in reaction
times of well over 100 hours. Highest re-
ductions were observed with higher Fe-to-
soil ratios and maximized concentrations
of hydrogen peroxide, up to 2%, by peri-
odic additions. Where concentrations were
reduced, the losses of PCBs were ob-
served more from the lower chlorinated
homologs (di and tri-PCBs) and less from
the higher chlorinated homologs (tetra
through hepta-PCBs). .
Bioslurry treatment did not provide sig-
nificantly different results for the UV-treated
surface soil versus the untreated soil. This
was not surprising since UV treatment was
not successful in significantly degrading
the higher chlorine level PCB homologs.
Percent reductions of PCBs were highest
for a New England Superfund Site soil
that had a significantly lower concentra-
tion of PCB contamination. The culture
isolated from the New England soil gave
70% 20% and 30% reduction of the di,
tri and tetra-PCBs, respectively, in the
New England soil. PCB reductions less-
ened with increasing level of chlormation
with no significant reduction noted for
penta, hexa, or hepta-PCBs. Similar re-
sults were obtained with inducer additions
to the soils. Biphenyl addition gave even
greater reduction in PCB concentrations
for the New England site soil with reduc-
tions of 82%, 54%, 63% and 16% for di,
tri, tetra, and penta-PCBs, respectively.
Recommendations
Although the percent of PCB degrada-
tion was low, meaningful destruction rates
may have been masked by the high con-
centration of PCBs in the surface soil that
was used in many of these tests. Although
the lower PCB concentration pit soil was
used in some of the photolysis tests with
little difference in results, a more detailed
analysis of the processes would be al-
lowed if soils with lower, but still practical
PCB concentrations (<1000 ppm), were
used in this early phase of testing.
High amounts of surfactant were car-
ried through the treatment process and
may have been inhibitory to bacterial ac-
tivity or promoted non-PCB degrading ac-
tivity. Likewise, the treated soil had a pH
of 5.5, which may have been inhibitory to
the bacteria and probably was a result of
surfactant addition. An additional soil wash-
ing step may be necessary to remove/
recycle surfactant from the soil and neu-
tralize the pH before biological treatment.
In addition, correlations of PCB-degrad-
ing activity with soil type, PCB concentra-
tion and composition, biphenyl/PCB
concentrations, and bacterial populations
need to be explored.
Both of these processes may be en-
hanced by adding surfactants to the soil
to solubilize the contaminants and to pro-
vide a medium for reaction mass transfer
processes. .
The full report was submitted in fulfill-
ment of Cooperative Agreement No.
CR816817-02-0 by IT Corporation under
the sponsorship of the U.S. Environmen-
tal Protection Agency.
Table 3. Percent Loss of Homolog Groups at Four Weeks Bioslurry Evaluation
Dichlorobiphenyl
Trichlorobiphenyl
Tetrachlorobiphenyl
Pentachlorobiphenyl
Hexachlorobiphenyl
Heptachlorobiphenyl
24
0
0
0
0
0
21
16
0
0
0
0
Untreated Soif
67
0
0
0
0
0
0'
0
0
0
0
0
New England So//0
70
20
30
0
0
0
40
0
0
0
0
0
*Surfactant/UV-treated surface soil - 4000 mg/Kg total PCB.
"Untreated surface soil - 8400 mg/Kg total PCB.
'New England Superfund Site soil - 350 mg/Kg total PCB.
" Segrad^on /ess tnan 15% is not considered significant and is reported as zero..
------- |