BENCH-SCALE TESTING OF PHOTOLYSIS, CHEMICAL OXIDATION
AND BIODEGRADATION OF PCB CONTAMINATED SOILS AND
PHOTOLYSIS OF TCDD CONTAMINATED SOILS.
by
IT Corporation
Knoxville, Tennessee 37923
Cooperative Agreement No. CR816817-020-0
Project Officer
Mr. Randy Parker
U.S. EPA
Office of Research and Development
Cincinnati, Ohio 45268
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
The information in this document has been funded wholly (or in
part) by the U.S. Environmental Protection Agency under'
Cooperative Agreement No. CR 816817-020-0 to IT Corppration. It
has been subject to the Agency's peer and administrative review,
and it has been approved for publication as an EPA document.
Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
11
-------
FOREWORD
The U.S. Environmental Protection Agency (EPA) is charged by
Congress with protecting the Nation's land, air, and water
resources. As the enforcer of national environmental laws, the
EPA strives to balance human activities and the ability of
natural systems to support and nurture life. A key part of the
EPA's effort is its research into our environmental problems to
find new and innovative solutions.
The Risk Reduction Engineering Laboratory (RREL) is
responsible for planning, implementing, and managing research,
development, and demonstration programs to provide an
authoritative, defensible engineering basis in support of the
policies, programs, and regulations of the EPA with respect to
drinking water, wastewater, pesticides, toxic substances, solid
and hazardous wastes, and superfund-related activities. This
Publication is one of the products of that research and provides
a vital communication link between the researcher and the user
community.
Now in its sixth year, the Superfund Innovative Technology
Evaluation (SITE) Program is part of EPA's research into cleanup
methods for hazardous waste sites around the nation. Through
cooperative agreements with developers, alternate or innovative
technologies are refined at the bench- and pilot-scale level and
then demonstrated at actual sites. EPA collects and evaluates
extensive performance data on each technology to use in
remediation decision-making for hazardous waste sites.
This report documents the results of bench-scale testing of
UV photolysis, chemical oxidation and biological treatment on
soils contaminated with toxic compounds.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
111
-------
ABSTRACT
This report presents the results of bench-scale testing on
degradation of 2,3,7,8-TCDD using W photolysis, and PCB
degradation using UV photolysis, chemical oxidation and
biological treatment. Bench-scale tests were conducted to
investigate the feasibility of a two-phase detoxification process
that would have application to the treatment of soils
contaminated with polychlorinated biphenyls (PCBs) and 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) . The first step in the
process was to degrade the contaminants by using ultraviolet (UV)
radiation facilitated by the addition of a surfactant to mobilize
the contaminants. As an alternative, an advanced oxidation
process using iron (Fe) catalyzed hydrogen peroxide (Fenton's
Reagent) was also tested. Biological degradation, the second
step, was then used to further degrade organic contaminants and
detoxify the soil.
UV photolysis tests were conducted independently using a
medium pressure mercury (Hg) lamp, a 10 hertz (Hz) pulsed Hg lamp
and sunlight. Results from UVtesting on a TCDD soil (200-300
parts per billion) indicated that there was no apparent
destruction of the dioxin on the soil. Surface soil contaminated
with about 10,000 parts per million (ppm) PCBs and a pit soil
containing about 200 ppm PCBs were tested under similar
conditions. The PCB reductions spanned the range up to a maximum
of 69 percent. Batch experiments using the Fenton's Reagent
alternative to degrade PCBs gave similar results with reaction
times of over 100 hours.
Biological treatment on surfactant/UV-treated and untreated
soil was evaluated in two bioslurry treatment experiments. The
bioslurry experiments evaluated PCB degradation on surfactant/UV-
treated and untreated soils using cultures, with and without PCB
degradation inducer chemical addition. The inducers used were
biphenyl and 4-bromobiphenyl. Bioslurry treatment did not
provide significantly different results for the UV-treated
surface soil versus the untreated soil. Percent reductions of
PCBs were highest for an untreated soil containing 350 ppm PCBs
which gave 70, 20 and 30 percent reduction of the di, tri and
tetra-PCBs, respectively. In the enhanced bioslurry experiment
using inducers, the addition of 1,000 ppm biphenyl stimulated
greater reduction in PCB concentrations on the same soil.
KN/9-»*SrrE.ETP(n/SrrE3IUT.REV IV
-------
This report is submitted in fulfillment of cooperative agreement
number CR816817-02-0 by IT Corporation under partial sponsorship
of the USEPA. This report covers the period from September 1990
to July 1993, with the completion of work in July 1993.
-------
CONTENTS
Foreword . . ii i
Abstract iv
Figures viii
Tables , ix
Abreviations and Symbols x
Acknowledgement xii
1. Executive Summary 1
UVPhotolysis Performance 1
Chemical Oxidation Performance 2
Biological Treatment Performance 3
2. Introduction 4
3. UVPhotolysis £
Introduction 6
Experimental Procedures 6
Site Sampling 6
Sample Preparation 7
Bench-Scale Testing 8
Bench-Scale Sunlight Testing 9
Materials and Methods 10
Equipment 10
Chemical Reagents 11
Analysis of 2,3,7,8-TCDD 11
Analysis C-f,££3**,*. * ~~- 11
Quality Assurance/Quality Control 12
Results and Discussion 13
TCDD Photolysis 13
PCB Photolysis Using UVLamps 13
PCB Photolysis Using Sunlight Exposure .. 18
Soil Particle Size Testing 19
Conclusions and Recommendations 24
4. Chemical Oxidation of PCBs 26
Introduction 26
Experimental Procedures 26
Overview 26
Feed Soil Preparation 27
Sampling 27
Materials and Methods 28
Equipment 28
Chemical Reagents 29
pH Measuratnft,v.t 29
Hydrogen Peroxide 29
PCB Analysis 29
Quality Assurance 30
Potassium Permanganate 30
PCB Analysis 30
Results and Discussion . 32
Experiment 1-24 and 92 Hours, No Mixing . 32
Experiment 2-162 Hours,
KN/9-94/srrE.CTF03/srrE3RIT.REV V 1
-------
Continuously Stirred 34
Experiment 3 - 118 Hours, Iron Effect ... 35
Experiment 4-2 Liter Reactor, 850 Hours . 38
Experiment 5-2 Liter Reactor-Surfactant
Addition, 184 Hours 40
Summary of Chemical Oxidation
Testing Results 41
Conclusions and Recommendations 42
5. Biological Treatment 44
Introduction 44
Experimental Design and Test Objectives .'. . . 45
Materials and Method 46
Isolation of PCB-Degraders 46
Rapid PCB Screening Assay 47
Bioslurry Evaluation 48
Enhanced Bioslurry Evaluation 50
E/VPhotolysis 51
Data Handling 51
Results and Discussion 52
Isolation of PCB-Degraders 52
Rapid PCB Screening Assay 54
Bioslurry Evaluation 54
Enhanced Bioslurry Evaluation 59
Conclusions and Recommendations 63
6. References 66
Appendices
A. TCDD Analytical Reports
B. Laboratory Standard Operating Procedures for DCMA PCB
Analysis
C. Summaries of PCB Analytical Results for UV Photolysis
Tests
D. Chemical Oxidation Experiment Data Tables
E. Chemical Oxidation QC Summaries
F. Biological Treatment Analytical Data
G. Bacterial Culture Isolate Data
KN/»-»4/srrE.Eni
-------
FIGURES
Number Pase
1 UV Treatment of Gas Pipeline Soil 22
2 GC Chromatograms of Treated and Untreated Soil 23
3 GC Chromatograms of Fenton's Reagent Treated and Control
Samples 36
4 PCB Concentration vs Time for Fenton's Reagent Exp. #4 . 39
KN/9-94ttrrE.En1<»/srrE>RIT.REV VI11
-------
TABLES
Number Page
1 Summary of TCDD W-Photolysis Testing 13
2 Summary of W Photolysis Testing on PCB Surface Soil 15
3 Summary of UV Photolysis Testing on PCB Pit Soil
Using Mixed Surfactant 16
4 Summary of Surfactant PCB Extraction Screening Tests 17
5 Summary of UV Photolysis Testing on PCB Pit Soil
Using Single Surfactant la
6 Summary of W Photolysis Testing on PCB Pit Soil
Using Solar Irradiation 18
7 Summary of UV Photolysis Testing on Fine Ground
PCB Surface Soil 20
8 UV Photolysis - 20 Hour Test #25 -
PCB Concentration Results 21
9 Fenton's Reagent Experiment 1 33
10 Fenton's Reagent Experiment 2 , 35
11 Fenton's Reagent Experiment 3 37
12 Summary of Chemical Oxidation (Fenton's Reagent)
Testing 41
13 Congener Identification 53
14 Rapid PCB Screening Assay 54
15 Percent Specific Congener PCB Degradation 57
16 Percent Loss of Congener Groups -
DCMA Method Bioslurry Evaluation 58
17 Oxygen Consumption in Treatments 59
18 Percent Specific Congener PCB Degradation 60
19 Comparison of Percent Specific Congener
PCB Degradation 61
20 Percent Loss of Congener Groups -
DCMA Method Enhanced Bioslurry Evaluation .... 62
21 Oxygen Consumption Enhanced Bioslurry Evaluation 63
D-l Initial Conditions of Fenton's Reagent Experiments'
D-2 Fenton's Reagent Experiment 1 PCB Analytical Results
D-3 Fenton's Reagent Experiment 2
D-4 Fenton's Reagent Experiment 3
D-5 Fenton's Reagent Experiment 4
D-6 Fenton's Reaaent Exoeriment 5
IX
-------
ACKNOWLEDGEMENT
This document was prepared under Cooperative Agreement No.
CR816817-02-0 by IT Corporation, Knoxville, Tennessee under the
sponsorship of the USEPA. Randy A. Parker of the Risk Reduction
Engineering Laboratory, Cincinnati, Ohio was the project officer
responsible for the preparation of this document and is deserving
of special thanks for his helpful comments and advice throughout
the project.
Participating in the development of this report for IT
Corporation, were Dr. Duane Root, Arie Groen, Doug Demott, Kandi
Brown, Janet Rightmyer and Dr. John Sanseverino.
IT Corporation would also like to acknowledge the University
of Tennessee Center for Environmental Biotechnology (CEB) and
General Electric Company (GE) for their contribution to this
project. Both institutions lent valuable support in obtaining
and identifying bacterial cultures for the biological treatment
of PCBs.
KN/9-M/SITE.Erra/SrrEJRIT.IlEV X i i
-------
SECTION 1
EXECUTIVE SUMMARY
The tests reported herein were conducted to investigate the
feasibility of a two-phase detoxification process that would have
application to the treatment of soils contaminated with
polychlorinated biphenyls (PCBs) and 2,3,7,8-tetrachlorodibenzo-
p-dioxin (TCDD) . The first step in the conceived process was to
degrade or chemically alter the organic contaminants by using
ultraviolet (UV) radiation. The source of UV radiation may be
either artificial UV light or natural sunlight, but generally
photolytic processes are more rapid with artificial [/Flight.
Alternatively, advanced oxidation processes, such as iron
catalyzed hydrogen peroxide (Fenton's Reagent), may be used to
initiate contaminant degradation. Both photolysis and chemical
oxidation were expected to be facilitated by the application of a
surfactant solution to the soil to mobilize the contaminants and
provide a medium for degradation reactions. These reactions were
expected to convert the contaminants to more easily biodegradable
compounds. Biological degradation, the second step in the
process, would then be used to further oxidize organic
contaminants and detoxify the soil. Biodegradation is typically
enhanced by the addition of microorganisms and nutrients to the
soil and may be further enhanced by the addition of
biodegradation inducers, such as biphenyl or 4-bromobiphenyl.
This report presents the results of bench-scale testing on
degradation of 2,3,7,8-TCDD using UVphotolysis, and PCB
degradation using both t/Fphotolysis and chemical oxidation.
Biological treatment was also performed on both untreated and
post [TVphotolyzed PCB contaminated soils.
UV PHOTOLYSIS PERFORMANCE
UV photolysis testing was performed on three soils; one
containing 2,3,7,8-TCDD contamination and two containing PCB
contamination. The tests were conducted independently using a
medium pressure mercury (Hg) lamp, a 10 hertz (Hz) pulsed Hg lamp
and sunlight, employing different surfactants and surfactant
application procedures.
Testing was performed with a composited TCDD soil from the
Vertac site in Jacksonville, Arkansas using two surfactant
levels, 2.5 percent and 5 percent by weight of the dry soil.
TCDD concentrations on the soil ranged from about 200 to 300
parts per billion. The soil was mixed and sprayed at 1/2 hour
intervals with either surfactant solution or water to a total
irradiation time of 48 hours. Results from these tests indicated
that there was no apparent destruction of the dioxin on the soil
in any of the tests.
-------
Surface soil from a Texas Eastern Gas Pipeline site in
Danville, Kentucky contaminated with about 10,000 parts per
million (ppm) PCBs (Aroclor 1248) and a pit soil from the same
site containing 150 ppm PCBs were tested. Testing conditions
differed from that above by using different surfactants,
application procedures, soil mixing intervals and lamp to soil
distances. The test results showed minimal reduction of PCBs,
ranging from none detected to a maximum of 69 percent. in two
tests, in which soil temperatures were elevated to over 100°C,
loss of 32 to 44 percent of the PCBs due to volatilization was
observed. Typically, in tests in which soil temperatures were
limited to 50 C or less, reductions of soil PCBs were in the
range of 15 to 35 percent. Best results were obtained using a 2-
3 percent surfactant spray loading on soil ground to particle
sizes less than 63 microns with a minimum bed depth (1/4 inch)
and lamp to soil distance (4 inches) . PCB reductions in these
tests ranged from 23 to 69 percent with 6 hours or longer of UV
exposure. Decreases in concentration at temperatures of 50°C or
less occurred for tri- through hepta-PCB homologs while the di-
PCB congener group (homolog) displayed an increase in
concentration because of di-PCB by-product generation.
Generation of some specific Tri- and tetra-PCB by-products was
also detected. These results indicated degradation of higher
chlorinated PCBs to lower chlorinated di-, tri- and tetra-PCBs.
CHEMICAL, OXIDATION PERFORMANCE
Five batch experiments using Fenton's Reagent (H202/Fe) were
performed at ambient temperature. All five used the same surface
soil used in the UV photolysis testing. This soil provided
samples for treatment which ranged from 6,000-10,000 ppm PCBs
(Aroclor 1248). Conditions were established to provide the best
opportunity for observing an effect due to treatment. Each
experimental mixture was pH adjusted to between 2 and 4 and
continuously stirred. Hydrogen peroxide concentration was
monitored throughout the experiments as loss, primarily through
decomposition, was continuous. Additions of hydrogen peroxide
were made as necessary to maintain a concentration (1 to 2
percent). Reagent to soil ratios were high, usually 8:1 to 10:1,
and iron concentrations were varied between experiments, up to
2.5 percent of the soil, to investigate the effect.
Results from these tests showed minimal reduction of PCBs on
the highly-contaminated surface soil tested. The PCB
concentration reductions ranged from none detected to a maximum
of 54 percent in reaction times of well over 100 hours. Highest
reductions were observed with higher iron to soil ratios along
with higher concentrations of hydrogen peroxide. Where
reductions in concentration were noted, the loss of PCBs were
observed more from the lower chlorinated congeners, di and tri-
PCBs, and trended less, progressing through the higher
chlorinated congeners, tetra through Observed
-------
reduction in PCB concentrations are suspected to have been
primarily due to volatilization from solution by gas purging.
Oxygen was continually generated in solution from hydrogen
peroxide decomposition.
BIOLOGICAL TREATMENT PERFORMANCE
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-
biodegradation inducers: biphenyl and 4-bromobiphenyl, on
congener removal.
The bioslurry experiments were conducted 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. In addition, known-PCB
degrading microorganisms were obtained from General Electric
Company (GE). Soils employed were untreated surface soil from
the UVphotolysis testing, surfactant/UV-treated surface soil,
and New England Superfund Site soil. In separate tests, each
soil was treated with bacterial cultures.
Bioslurry treatment did not provide significantly different
results for the UVtreated surface soil versus the untreated
soil. This was not surprising since UVtreatment was not
successful in significantly degrading the higher chlorine level
PCB congeners. Percent reductions of PCBs were highest for an
untreated New England Superfund Site soil which had a
significantly lower concentration of PCB contamination than
either the UVtreated or untreated PCB surface soil from
Danville, Kentucky. The culture isolated from the New England
soil gave 70, 20 and 30 percent reduction of the di, tri and
tetra-PCBs, respectively in the New England soil. PCB reductions
lessened with increasing level of chlorination with no
significant reduction noted for penta, hexa or hepta-PCBs.
Similar results were obtained with inducer additions to the
soils. Biphenyl addition gave even greater reduction in PCB
concentrations for the New England site soil with reductions of
82, 54, 63 and 16 percent for di, tri, tetra and penta-PCBs,
respectively.
-------
LIST ABBREVIATIONS AND SYMBOLS
ABBREVIATIONS
2,3-dhb
BAG
cc
CEB
CFU/mL
CFU/g
cm
DCMA
DOC
DOT
ECD
EPA
g
GC/ECD
GC
GC/FID
GE
hr
HZ
IR
IT
KD
L
/iL
mg 02/kg-hr
mg
mg/kg
mg/L
mL/min
mL
N
ng
nm
PCB
ppb
P Pin1 «
ppm
QA
QC
RCRA
RPD
rpm
RREL
RSD
SARA
2,3 dihydroxybiphenyl
Biotechnology Application Center
cubic centimeter
Center for Environmental Biotechnology
colony-forming units per milliliter
colony-forming units per gram
centimeter
Dry Color Manufacturers' Association
dissolved organic carbon
Department of Transportation
electron capture detector
Environmental Protection Agency
gram
gas chromatography with electron capture
detection
Gas Chromatograph
gas chromatography with flame ionization
detection
General Electric Company
hour
hertz
infrared
IT Corporation
Kuderna Danish
liter
microliter
milligram oxygen per kilogram-hour
milligram
milligram per kilogram
milligram per liter
milliliters per minute
milliliter
Normal
nanogram
nanometer
polychlorinated biphenyl
parts per billion
personal protective equipment
parts per million
quality assurance
quality control
Resource Conservation and Recovery Act
relative percent difference
revolutions per minute
Risk Reduction Engineering Laboratory
relative standard deviation
Superfund Amendments and Reauthorization Act
KN/»-»4/SrrE.En'«H/SrrE3RIT.REV
X
-------
SD
SITE
TCDD
TDL
TSCA
uv
v/v
°C/min
SYMBOLS
Fe
FeS04
H202
H2O2/Fe
H2S04
HC1
Hg
KMn04
NaOH
o2/kg-hr
To
standard deviation
Superfund Innovative Technology Evaluation
2,3,7,8-tetrachlorodibenzo-p-dioxin
Technology Development Laboratory
Toxic Substances Control Act
ultraviolet
volume to volume
degrees Celsius per minute
iron
iron (II) sulfate
hydrogen peroxide
Fenton's Reagent
sulfuric acid
hydrochloric acid
mercury
potassium permanganate
sodium hydroxide
oxygen per kilogram-hour
study initiation
study at 2 weeks
study at 4 weeks
study final
KN/»-94/SITE.ETI'03/SrrE}IUT.REV
XI
-------
SECTION 2
INTRODUCTION
The Superfund Amendments and Reauthorization Act of 1986
(SARA) directed the Environmental Protection Agency (EPA) to
establish an "Alternative or Innovative Treatment Technology
Research and Demonstration Program." In response, the EPA's
Office of Solid Waste and Emergency Response and the Office of
Research and Development established a formal program called the
Superfund Innovative Technology Evaluation (SITE) Program, to
accelerate the development and use of innovative cleanup
technologies at hazardous waste sites across the country.
The SITE program comprises the following five component
programs:
Demonstration Program
Emerging Technologies Program
Measurement and Monitoring Technologies Development
Program
Innovative Technologies Program
Technology Transfer Program
This report is sponsored by the SITE Emerging Technologies
Program. Before a technology can be accepted into the Emerging
Technologies Program, sufficient data must be available to
validate its basic concepts. The technology is then subjected to
a combination of bench- and pilot-scale testing in an attempt to
apply the concept under controlled conditions.
The tests reported herein were conducted to investigate the
feasibility of a two-phase detoxification process that would have
application to the treatment of soils contaminated with
polychlorinated biphenyls (PCBs) and 2,3,7,STetrachlorodibenzo-
p-dioxin (TCDD) . The first step in the process was to partially
degrade or chemically alter the organic contaminants by using
ultraviolet (UV) radiation. Typically the rate of photolytic
degradation is faster with artificial UV light than with natural
sunlight, but both sources of UV radiation were proposed. As an
alternative, an advanced oxidation process, iron catalyzed
hydrogen peroxide (Fenton's Reagent), was investigated as a means
to provide initial contaminant degradation. Both photolysis and
chemical oxidation were expected to be facilitated by the
addition of a surfactant solution to the soil to mobilize the
contaminants and provide a medium for degradation reactions.
Both processes, UV-photolysis and chemical oxidation, were
expected to convert the contaminants to more easily biodegradable
compounds. Biological degradation, the second step in the
overall process, was then envisioned as a final step to further
oxidize organic contaminants and detoxify the soil.
-------
Biodegradation is typically enhanced by the addition of
microorganisms and nutrients to the UV treated soil and can be
further enhanced by the addition of biodegradation inducers.
This two-phase treatment was conceptualized as a potential
in-situ process for shallow contamination on soils. More
probable, however, was the use of the technology for ex-situ, on-
site treatment of excavated soils in a specially constructed
shallow treatment basin, which would meet the reguirements of the
Resource Conservation and Recovery Act (RCRA). The process may
have reguired longer treatment times than other technologies, but
was anticipated to have a trade off in economy. The only residue
generated from this combination of technologies would be soil
contaminated with surfactants and the end metabolites of the
biodegradation processes. The end metabolites depend on the
original contaminants. The surfactants are common materials used
in agricultural formulations.
This report presents the results of bench-scale testing on
degradation of 2,3,7,8-TCDD using UVphotolysis, and PCB
degradation using UV photolysis and chemical oxidation (Fenton's
Reagent). Biochemical treatment testing was also performed on
soil contaminated with PCBs both untreated and after
surfactant/UV photolysis treatment. Soil contaminated with TCDD
was not subjected to chemical oxidation or biodegradation
testing.
Chemical oxidation was proposed as an alternate means to
partially degrade or chemically alter PCB contaminants to more
easily biodegradable products after tests showed little PCB
degradation from UV photolysis treatment. Chemical oxidation
testing using Fenton's Reagent was performed on the same PCB
contaminated soil used in the UVphotolysis tests to compare
these two technologies.
The work presented in this report is divided into three
parts based on the technology employed; UVphotolysis, chemical
oxidation and biological treatment.
Testing in this program involved TCDD soils regulated by the
RCRA and PCB soils regulated by the Toxic Substances Control Act
(TSCA) . The TDL is authorized to perform treatability studies on
RCRA hazardous wastes under the treatability exemptions of
Tennessee Department of Environment and Conservation, Division of
Solid Waste Management (TN Rule Chapter 1200-1-11-.02 (1) (d) 6.).
A TSCA bench scale permit for treatability testing of PCB
contaminated soil was obtained from EPA Region IV Toxic
Substances Control Branch on September 4, 1990.
-------
SECTION 3
UV PHOTOLYSIS
INTRODUCTION
Earlier work performed by IT Corporation (IT) showed a
practical rate of photolytic destruction of PCBs and TCDD (Exner,
et. al., 1984) on soil when the soil surface was treated with a
surfactant solution and irradiated by UV light. The reactions
were aided by the presence of a surfactant, which ideally is
transparent to the UV radiation in the region of activity
(generally 254 nanometers) and which has increased solubility for
the contaminants being destroyed. Conceptually, the irradiation
process can be performed on excavated soils or in situ using
enhanced radiation from lamps or natural sunlight. The process
usually involves the continued application of the solubilizing
aid (surfactant) and continued exposure of fresh surface to the
irradiation source. The solubility aid helps to transfer the
contaminant from the pores of the soil to the soil surface where
the reactions can take place. The surfactant or solubilizing aid
may also act as a medium for the degradation process by providing
labile protons to allow the reaction to proceed more easily.
Because the presence of UV light is usually accompanied by
significant amounts of infrared (IR) radiation or heat, the
solubilizing aid needs to be continually or periodically
refreshed to provide a continued reaction medium.
The testing in this study was performed on three soils, one
contaminated with 2,3,7,8-TCDD and two from a site contaminated
with PCBs. The tests, conducted independently, used a medium
pressure Hg lamp, a pulsed Hg UV lamp, and sunlight as the
sources of W radiation. Different surfactants were tested and
different surfactant application procedures were tried to
establish the procedures that would allow the UVreaction to
proceed. The objective of the tests were to preliminarily
investigate the feasibility of the technology for application to
soils contaminated with TCDD or PCBs. The treatment was
monitored primarily by the disappearance of contaminant with
qualitative notation of any by-product production.
EXPERIMENTAL PROCEDURES
Site Sampling
IT personnel traveled to the Vertac site in Jacksonville,
Arkansas to obtain soil samples contaminated with TCDD. The
soils from several areas within the site were sampled using a
shovel. 5-gallon pails lined with plastic were filled and
sealed (GG3866 and GG3867). The pails were then packaged in
boxes and shipped to the Technology Development Laboratory (TDL)
KN/»-»4/SrrE.ErHn/SITE3MT.REV
-------
located in Knoxville, Tennessee. Workers handling the unpackaged
soils were outfitted with Level C personal protective equipment
(PPE). This level includes a full plastic coated Tyvek suit,
nitrile gloves with PVC under gloves, PVC boots and air purifying
respirators.
To obtain PCB contaminated soil, IT personnel traveled to a
Texas Eastern Gas Pipeline site in Armaugh, Pennsylvania. The
soils were again packaged in 5-gallon pails lined with plastic
bags. The samples were transported back to the TDL in a truck by
IT personnel. These soils were found to be unsuitable for use as
noted below. A second sampling trip to a Texas Eastern site in
Danville, Kentucky by IT personnel was completed on April 1,
1991. The samples were returned to the laboratory on the same
day. All shipping and transportation activities were in
compliance with applicable Department of Transportation (DOT)
regulations.
Sample Preparation
Samples containing TCDD (GG3866 and GG3867) were received at
the TDL on December 21, 1990 and logged into the sample receiving
system. The samples were held in Sample Receiving until January
8, 1991. They were opened and the contents spread into large
aluminum baking pans to air dry. During the air drying, the
soils were crushed and screened to less than 1/8 inch (0.125
inch) particle size. The final weight of the dried and screened
material was approximately 25 kilograms. Testing of these soils
proceeded under IT's Treatability Exemption. At the conclusion
of testing, the TCDD soil was packaged and returned to the site
for disposal.
The first set of PCB contaminated soil samples (Armaugh,
Pennsylvania) were received on February 8, 1991 and logged into
the sample tracking system. The samples were air dried on
February 14, 1991. The soil was a very sticky clay-type material
and dried to a hard cake that dusted badly when it was crushed
and sieved. The soil was also expected to become sticky and form
lumps when the surfactant solution was added during the
experiments. Based on this, IT decided that the soil was not
suitable for testing and another sample site was chosen. This
soil was packaged and returned to the site for disposal.
Samples from a second site (Danville, Kentucky) were
received on April 1, 1991. These samples consisted of soils from
four locations on site and were very different in nature, ranging
from pure gravel to topsoil. The PCB concentration ranged from
100 parts per million (ppm) to greater than 10,000 ppm of Aroclor
1248 in the different samples. Two of the soils which had
moderate PCB concentrations and were available in larger
quantities were selected for testing. These two soils were
identified as surface soil (GG4202) , containing 1700 ppm PCBs
KN/»-94«rrE.EITa3/SrrE3MT.REV
-------
(Aroclor 1248) by analysis, and pit soil (GG4199) , containing 150
ppm PCBs. They were processed by air drying and screening to
less than 1/8 inch (0.125 inch) particle size prior to UV
testing.
Initial testing began using the 1,700 ppm surface soil which
had a high humic content. In subsequent analyses the
concentration of PCBs in the surface soil was found to be greater
than 10,000 ppm instead of the expected 1,700 ppm. This
deviation from the expected value may have been a result of the
initial sampling or the preparation of the soil to exclude debris
and stones.
For the fine ground soil testing (small particle size soil),
the GG4202 surface soil was ground in a standard kitchen blender
in a fume hood until the soil passed a 230 mesh sieve (particle
size less than 63 microns).
Bench-Scale Testing
Testing with the composited TCDD soil began on February 21,
1991.. Analysis of the composite soil gave a 2,3,7,8-TCDD
concentration of 271 parts per billion (ppb) (nanogram/gram).
The initial tests used a 7-inch by 11-inch Pyrex baking dish
filled to a 1 inch depth of soil. Two surfactant levels, 2.5
percent and 5 percent by weight surfactant as a percentage of the
dry soil, were used in testing. The surfactant was applied 'by
spraying approximately one half of the target concentration on
the soil initially and then the remainder was applied during
irradiation by periodic spraying and mixing steps. The
surfactant solution was 8 percent of a 1:1 mixture of nonionic
surfactants: Hyonic NP-90® (Diamond Shamrock Corporation) and
Adsee 799® (Witco Chemical Corporation), in deionized water. A
more dilute solution (less than 2 percent) was used for periodic
spraying during the tests. Hyonic NP-90® is a polyethpxylated
nonyl phenol and Adsee 799* is a polyoxyalkyl fatty acid ester.
It was found that when the surfactant level approached 3.5
percent of the soil, it became very sticky and lumped badly when
it was stirred. In the higher surfactant concentration tests,
the surfactant spraying to reach 5 percent loading was
discontinued and the irradiation continued until the soil dried
sufficiently to be worked. At this point, spraying with water
only was continued to the end of the test.
Tests with a 450 Watt Hanovia medium pressure Hg lamp and a
10 Hz pulsed Hg lamp operating at 450 Watts total power were
carried out with the lamps approximately 10 inches above the soil
and a parabolic reflector above the lamp. The soils were mixed
and sprayed at 1/2 hour intervals with either surfactant solution
or water to a total irradiation time of 48 hours. Most of the
samples were sent to the IT laboratory in St. Louis for TCDD
analysis by Region VII TCDD Rapid Turnaround method. One set of
KNA94mrE.Elm3lslrE3iwr.nEv
-------
duplicate samples was analyzed by the Dioxin Analysis Group at
the TDL using SW-846 Method 8280.
Testing of PCB contaminated soils followed the same general
procedure as described above for the TCDD soil. Lamp to soil
distance, surfactant and surfactant application procedure, as
well as soil mixing/overturning interval and soil particle size
were all variables that were adjusted to optimize degradation of
PCBs.
Bench-Scale Sunlight Testing
UV irradiation by sunlight of TCDD and PCB contaminated soil
was performed during the months of July and August 1991. The
surfactant mix used initially in the TCDD experiments was
discontinued during the test after surfactant extraction tests
showed Hyonic NP-900 to be superior for PCB extraction. The
soils were raked daily and sprayed at the beginning of the day
with the surfactant solution. Subsequent sprayings during the
day used water only to try to maintain a moist surface. The
evaporation rate was very high and it was difficult to keep the
soil moist with only one spraying. If the surface became dry,
the extraction rate of the surfactant became negligible.
Three trays with composited PCB soil and three trays with
composited TCDD soil were weighed out on June 25, 1991. The
trays were 7 x 11 glass Pyrex oven baking dishes. Each tray
contained approximately 1 kilogram of soil. One sample in the
PCB set and one in the TCDD set were preloaded with 2.5 percent
of a surfactant mixture using a 25 percent concentrated
surfactant solution. Another sample in each set was preloaded to
1.2 percent surfactant concentration using 12.5 percent
concentrate. The samples were loaded to a total of 8 to 10
percent moisture content. The third sample in each set was
sprayed with water only. Triplicate aliquots were removed from
each tray for starting analyses for Duplicates were
removed from the TCDD samples for analysis.
The sprayed samples were positioned in a metal tray on the
roof of the TDL building for sunlight exposure. During the
evening and when raining, the tray was covered. For the first
month of exposure the sample was weighed after it was removed
from the sunlight and reweighed after spraying with surfactant
solution. Surfactant was sprayed only for the first month when
the total surfactant had reached 5 percent loading for the high
loading and 2.2 percent for the lower loading test.
During testing the soil became very sticky and produced
balls of material. The top surface caked as the moisture dried
out during the daytime exposure. Samples were removed for
analyses after 0, 40 and 197 hours of exposure. After 1 month
the soil was broken up using a blender to totally remix the soil.
-------
During the second month, testing continued using a solution of
the Hyonic NP-900 surfactant. The surface was sprayed with a
dilute 0.2 percent solution twice per day. The soil was turned
over using a stainless spatula before stirring.
The UV intensity at the surface of each tray was monitored
for the first month along with a position just outside the tray.
The readings ranged from a high of about 360 microwatts/cm* to
below .04 microwatts/cm*. Temperature readings ranged from about
26°C at the start of the testing to a high of 41°C on July 2,
1991. Temperatures usually were in the low 30°C range during the
month of July. The readings from the radiometer were also sent
to a recorder for continuous monitoring. The readout peaks were
about 345 microwatts/cm*. The exposure varied greatly during the
day as the sun rose and moved in and out of clouds. Radiometer
readings were made with a radiometer specific to 254 nanometers
wavelength ultraviolet light.
MATERIALS AND METHODS
Eauinment
The irradiation of PCBs and/or TCDD contaminated soils was
carried out using glass or incoloy metal trays and stainless
implements (spatulas). The application of water or surfactant
solution to the soil was done using a commercially available
plastic spray bottle purchased at a local department store.
These spray bottles are typically used for the application of
agueous solutions in the home (window spray, insecticide or
fungicide solution spray). The spray nozzles were adjusted to
give a fine mist when spraying to provide the best distribution
possible. For the preliminary application, when the greatest
amount of moisture was added, the soil was sprayed incrementally
and mixed with a stainless spatula until the moisture was
uniformly distributed.
The incoloy metal tray was approximately 3 inches by 6
inches by 1/2 inch deep. The glass trays were approximately 7 by
11 by 1 1/2 inch Pyrex baking dishes purchased from a laboratory
supply house and identical to the baking dishes available in
local department stores. The mixing tools were stainless steel
spatulas.
The soils were weighed on a 12-kilogram capacity, digital
top-loading balance (Sartorius Model 1200LC) inside the
laboratory for the determination of moisture weight addition.
For the sunlight experiments on the roof, a lo-kilogram capacity,
Ohaus, top-loading balance was used. The balance was housed
inside a plastic cabinet with a hinged door to allow weighings.
The trays containing contaminated soils for the sunlight
experiments on the roof of the laboratory were positioned inside
a secondary galvanized tray located on several concrete blocks.
10
-------
A lid of plywood with a 2 x 4 inch drip edge was fabricated to
position over the galvanized tray with the drip edge downwards
during the periods when sunlight was not available (rain or
evenings) . The lid was secured to keep it from being blown off.
Chemical Reasents
The commercial surfactants used in the UV photolysis tests
were the following:
Adsee 799® - Witco Chemical Corp., polyoxyalkylated
fatty acid ester
Hyonic NP-90® - Diamond Shamrock Corp., polyethoxylated
(9) nonylphenol
In addition, two other nonionic commercial surfactants were
used in the surfactant extraction tests.
Brij 30* - Supplied by Aldrich Chemical Co.,
polyoxyethylene (4) lauryl ether
Brij 35® - Supplied by Aldrich Chemical Co.,
polyoxyethylene (23) lauryl ether
Analysis of 2.3.7.8-TCDD
Dioxin analyses were performed by two different laboratories
using different analytical techniques. The samples sent to the
IT St. Louis Laboratory were analyzed by USEPA Region VII Rapid
Turnaround Method for TCDD. The dioxin levels contained in the
soil samples being analyzed were much higher than normally
analyzed by this technique and the soil was also somewhat
heterogeneous. The extraction and spiking technique were
modified after consultation with the laboratory to better suit
the sample needs. Copies of the analytical reports are included
in Appendix A.
Samples submitted for analysis at the TDL were extracted and
analyzed using SW-846 Method 8280. The preliminary soil analysis
to establish starting concentration was done at the TDL. One set
of duplicate samples for one of the UV experiments was analyzed
at the TDL for verification of the IT St. Louis Laboratory
method. Agreement between the two laboratories was within
reasonable expectations given the differences in methodology.
Copies of the analytical reports are included in Appendix A.
Analysis of PCBs
The soils were extracted by sonication (SW-846 Method 3550)
or Soxhlet extraction (SW-846 Method 3540) using a mixture of
methylene chloride and acetone with subsequent solvent exchange
to hexane. Samples were then analyzed by gas chromatography with
electron capture detection (GC/ECD).
-------
The analysis and quantification of PCBs was performed by one
of two methods, EPA SW-846 Method 8080 or a PCB homolog
procedure, which is a modified version of the Dry Color
Manufacturers' Association (DCMA) PCB Method, June 1981.
Untreated samples were analyzed and quantified for Aroclor 1248
or Aroclor 1260 using GC/ECD methods consistent with SW-846
Method 8080. Treated samples containing altered PCB patterns
were analyzed by a GC/ECD, semi-specific PCB homolog method
(DCMA). The DCMA method divides the PCB chromatographic elution
window into semi-specific homolog windows. Individual peaks are
quantified versus the appropriate homolog standard based on the
homolog window in which it elutes. Homolog totals are obtained
by summing the individual PCB peak amounts for each homolog
window. The total PCB concentration is then calculated from the
sum of the individual homolog totals. A copy of the laboratory
standard operating procedure for this analysis is included in
Appendix B. Analytical methodology for PCB analysis allows for a
variability of a minimum of plus or minus 15 percent (plus or
minus 25 percent for DCMA method). A statistical determination
of the limit of significance for whether there was a difference
between starting and final PCB concentrations on soils was not
determined because it was beyond the scope of the preliminary
work being performed. In addition to insufficient data, the use
of different methodologies complicates the process of determining
a limit of significance for the percent PCB reduction data based
on the difference of starting and final PCB concentrations. A
PCB reduction of less than 15 percent is clearly not considered
significant based on the minimum variability allowed by the
methodology. This, however, is not intended to signify that 15
percent is the limit of data significance and that any PCB
reduction greater than 15 percent is necessarily statistically
significant.
QUALITY ASSURANCE/QUALITY CONTROL
Because of the nature of the samples under investigation,
many of the samples were taken in duplicate and often the samples
were analyzed in duplicate to compensate for the variability
within the sample matrix. The variability was a result of the
particle size distribution with a significant quantity of small
gravel-like material within the soil. This gravel material tends
to hold a very low quantity of the contaminant under
investigation. If an aliquot is removed which contains no
stones, the analytical result will be disproportionately higher
and the results will be biased. When a large enough aliquot is
taken for the analysis, this bias is either removed or lessened
but the ability to spike the sample at the high levels contained
in the sample becomes impossible.
KN/»-94/SrrE.ETKn/srrE3RFT.RFV 12
-------
RESULTS AND DISCUSSION
TCDD Photolysis
The photolysis of the 271 parts per billion TCDD
contaminated soil using UV lamps under the conditions tested was
not successful in destroying TCDD to a detectable degree. The
lack of destruction may have been a result of many factors, such
as soil depth, surfactant type, lamp distance, soil particle
size, etc. Some of these factors were evaluated using the PCB
contaminated soils. The conditions tested for TCDD destruction
were the two UV lamp types and two surfactant concentrations for
each UV source. The results for the 48 hour tests, shown in
Table 1 shows no significant difference between the final TCDD
concentration in any of the tests and the starting TCDD
concentration (271 ppb).
TABLE 1. SUMMARY OF TCDD UV-PHOTOLYSIS TESTING
Suffactant Final TCDD Percent TCDD
Test. Lamp Type (% of Dry Soil) Cont. (ppb) Reduction'
1
2
3
4
Medium Pressure Hg
Medium Pressure Hg
Pulsed Hg- 10 Hz
Pulsed Hg- 10 Hz
2.5
5
2.5
5
245
356
250
244
10
0
8
10
* Initial soil concentration - 271 ppb TCDD.
Surfactant - Hyonic NP-90«and Adsee799* in 1: 1 ratio.
Soil bed depth -1 inch.
Lamp to soil distance -10 inches.
The soil samples fram the TCDD sunlight tests were not
analyzed for TCDD destruction because of the lack of effect in
the TCDD UVlamp tests and the PCB sunlight tests.
PCB Photolysis Using IJVLamlPs
Following these initial TCDD experiments, the PCB soils were
tested under similar conditions. Since PCB analytical results
were available with a faster turnaround time than TCDD analyses,
the experimental conditions could be adapted to suit the needs of
the experiments.
The initial PCB irradiation experiments used the highly
contaminated (Approximately 10,000 ppm Aroclor 1248) surface soil
13
-------
from Danville, Kentucky. in the first experiment (Test #1), the
pulsed lamp was used with a lower surfactant concentration and
spraying at 1/2 hour intervals to a total irradiation time of 12
hours. The surfactant was the same nonionic mix used in the TCDD
tests. There was no perceptible change in the PCB concentration.
A second experiment to test the effect of stirring the soil
more freguently was carried out with stirring and spraying at 2
minute intervals to a total irradiation time of 12 hours. These
test conditions were also used in testing the Hanovia lamp for 7
hours of irradiation (Test #3) . No change in the PCB
concentration was detected in these tests, again using the high
PCB concentration soil.
A fourth experiment using the Hanovia lamp with air cooling
instead of water cooling in the lamp well appeared to produce a
slight change in the PCB concentration.
A fifth experiment, again using air cooling but with the
lamp at 3.5 inches from the soil surface produced about a 50
percent loss in the PCB concentration after 3.5 hours. However,
the temperature of the soil was significantly higher than in
previous tests (approximately 105°C) because of the lack of well
cooling water and the short distance between the lamp and soil.
In addition, the loss of PCBs was highest for the lighter
chlorinated PCB congeners suggesting loss due to volatilization
at the higher temperature. The same loss was then duplicated in
a separate test (Test #6) by heating the soil in an oven at 140*C
for 4 hours, with spraying and stirring at 1/2 hour intervals.
This temperature was chosen because the bottom of the glass tray
reached temperatures in this region during the fifth irradiation
test. Results from these first six tests are summarized in Table
2.
14
-------
TABLE 2. SUMMARY OF UV PHOTOLYSIS TESTING ON PCB SURFACE SOIL
Test
1
2
3
4
5
6
Lamp
Type
Pulsed
Pulsed
Cont.
Cont.
Cont.
Ovenb
Soil Depth
(in)
0
0
0
0
0
1
.25
.25
.25
.25
.25
Lamp/Soil
Distance (in)
10
10
10
10
3.5
NA
Time
(Hours)
12
12
7
7
3.5
4
Temp.
( 'Cl
25
28
28
40
105
140
Initial PCB
Cone (ppm)
13
7,
7,
8,
,200
240
430
440
6,020*
8,
300
Final
PCB Cone
(ppm)
14,100
7,950
6,960
5,680
4,080
4,690
Percent PCB
Reduction
0
0
6
33
32
44
Initial soil was PCB surface soil.
Pulsed - mercury lamp pulsed at 10 Hz (70 Watts/inch for 6 inch lamp).
Cont. - Hanovia 450 Watt medium pressure continuous mercury lamp.
NA - Not applicable.
Surfactant - Hyonic NP-90* and Adsee 799* in 1 :l ratio at 2 percent of the soil.
* Starting soil was residue from previous treatment experiment.
b Soil was heated in oven at 140' C, no irradiation.
At this point in the testing, a radiometer was used to check
the distribution of light intensity on the soil at various
distances from the lamp. It was found that the intensity was
fairly uniform across the tray at 9-10 inches from the lamp using
the parabolic reflector. The edges fell off rapidly as the tray
was raised closer to the reflector since the edges of the tray
fell outside the reflector.
Based on the results of the UV distribution measurements and
in an effort to observe smaller absolute changes in PCB
concentration, it was decided to test a less contaminated
starting soil, in a smaller tray closer to the lamp. For this
test, the PCB pit soil from Danville, Kentucky with a PCB
concentration of approximately 150 ppm PCBs (Aroclor 1248) was
used. A shallow soil bed with frequent 10 minute raking
intervals was also used. This test used the Hanovia lamp in the
water-cooled light well. Additional air cooling above the soil
reduced the effect of heat generated by the lamp at this close
distance to the soil. A reduction in the PCB concentration of 18
percent was achieved (Test #7) .
A further test (Test #8) used the same soil spiked with
additional Aroclor 1260 to test the hypothesis that spiked
contaminants could be more easily photolyzed than weathered
15
-------
contaminants because they would be easily extracted from the soil
by the surfactants. This suspicion appeared to be confirmed
although the rate of destruction of Aroclor 1260 was not as high
as expected and not that much greater than the destruction of
Aroclor 1248 in the test.
The ninth test repeated the eighth test using the pulsed
lamp instead of the Hanovia lamp. Both lamps performed about the
same in terms of PCB (Aroclor 1260) reduction. Results from
these tests are summarized in Table 3.
TABLE 3. SUMMARY OF UV PHOTOLYSIS TESTING ON PCB PIT SOIL USING
MIXED SURFACTANT
Lamp Soil Depth Lamp/Soil Time Temp. initial Final Percent PCB
Test Type (in.) Distance (Hours) (*C) PCB Cone PCB Cone Reduction
('"•) (ppm)
7
8
9
Cont.
Cont.
Pulsed
0.25
0.25
0.25
3.5
3.5
3.5
10
10
10
50
58
35
194
104'
121'
159
24'
45'
18
77
63
Initial soil was PCB pit soil.
Cont. - Hanovia 450 Watt medium pressure continuous mercury lamp.
Pulsed - mercury lamp pulsed at 10 Hz (70 Watts/inch for 6 inch lamp).
Surfactant - Hyonic NP-90* and Adsee 799* in 1 :l ratio at 2 percent of the soil,
' Concentration of Aroclor 1260 spiked onto soil.
The efficiency of the surfactant solution to extract the
contaminants from the soil was becoming suspect due to the low
destruction rates observed. TO test the ability of the
surfactant to remove the PCBs from the soil, several different
surfactants were evaluated by shaking 2 grams of soil in 20
milliliters of a 3 percent surfactant solution for a total of 60
minutes on a platform shaker and then analyzing the supernatant
solution for PCBs. It was found that the Adsee 799® surfactant
being used in the test program was hindering the extraction
efficiency of the Hyonic NP-90® surfactant. The PCB extraction
screening tests are summarized in Table 4.
KN/9-94/srrE.Eran/srrE3RjT.REv 16
-------
TABLE 4. SUMMARY OF SURFACTANT PCB EXTRACTION SCREENING TESTS
Surfactant.
Brij 30*
Brij 35*
Brij 30* + Brij 35*"
Adsee 799Q
Hyonic NP-90*
Adsee 799* + Hyonic NP-90®"
SDS'
Hyonic NP-90*'
None (Water)
Surfactant
Type
ethoxy alkyl alcohol
ethoxy alkyl alcohol
non ionic mix
oxyalkylated fatty acid ester
ethoxylated nonyl phenol
non ionic mix
anionic
ethoxylated nonyl phenol
NA
PCB Conc.b
(ppm)
11.9
5.4
4.2
2.3
11.1
2.5
8.3
8.3
0.4
Percent
Extracted'
79
36
28
15
74
17
55
55
3
Brij 30®/Brij35*- Aldrich Chemical Co., polyethoxylated alkyl alcohols.
Adsee 799*- Witco Chemical Corp., polyoxyalkylated fatty acid ester.
Hyonic NP-90*- Diamond Shamrock Corp., polyethoxylated nonylphenol.
NA - Not applicable.
* Total surfactant concentration is 3 percent by weight in water.
b PCB concentration in the aqueous surfactant solution.
c Extraction based on 150 ppm PCB in 2 grams of soil in 20 ml_ of extraction solution.
d Mixtures are 1 :l by weight, total is percent.
' Sodium Dodecyl Sulfate.
' Single extraction, all others are averages of duplicate extractions.
Another UVtest (Test#10) was then performed using the
nonionic Hyonic NP-90® surfactant only on the PCB pit soil. This
test used a depth of soil of about 1/2 inch in the large tray at
a distance of 10 inches from the lamp, (water-cooled Hanovia
lamp), with 10 minute raking intervals for a total irradiation
time of 16 hours. A reduction in PCB concentration of
approximately 30 percent was achieved on the weathered,
contaminated soil. Two more tests were then performed using the
same conditions, but using the pulsed ttVlamp instead of the
Hanovia continuous lamp. One had a total irradiation time of 16
hours (Test Xll) and the other was twelve hours (Test X12) .
Results were not guite as good with the pulsed lamp, but were
considered to be within experimental variability to the results
from Test #10 with the continuous Hanovia lamp. Results from
these tests are summarized in Table 5.
17
-------
TABLE 5. SUMMARY OF UV PHOTOLYSIS TESTING ON PCB PIT SOIL
USING SINGLE SURFACTANT
Test
10
11
12
Lamp
Type
Cont.
Pulsed
Pulsed
Soil
Depth
(in.)
0.5
0.5
0.5
Lamp/Soil
Distance
(in.)
9
9
9
Time Temp.
(Hours) CO
16
16
12
30
28
28
Initial
PCB Cone
(ppm)
140
157
170
Final
PCB Cone
(ppm)
98
137
131
Percent PCB
Reduction
30
13
23
Initial soil was PCB pit soil.
Cont. - Hanovia 450 Watt medium pressure continuous mercury lamp.
Pulsed - mercury lamp pulsed at 10 Hz (70 Watts/inch for 6 inch lamp).
Surfactant - Hyonic NP-90* at 2 percent of the soil.
PCB Photolysis Using Sunlight Exnosure
Tests were conducted as described in the Experimental
Section using three different concentrations of surfactant. The
nonionic mix of surfactants was used throughout the first half of
testing and then was changed to the use of Hyonic NP-90® alone
after the results of surfactant PCB extraction tests were
realized. These tests showed no significant change in PCB
concentration after 197 hours (25 days) of sunlight exposure.
Because of the summertime conditions the soil surface dried
rapidly and this is considered partially responsible for the lack
of PCB degradation. Results from these tests are summarized in
Table 6.
TABLE 6. SUMMARY OF UV PHOTOLYSIS TESTING ON PCB PIT SOIL
USING SOLAR IRRADIATION
Test
13
14
15
Lamp
Type
Solar
Solar
Solar
Soil Depth
(in.)
1
1
1
Surfactant
Cone
(%)
4.5
2
0
Time
(Days)
25
25
25
Temp.
CO
26 -41
26 -41
26 -41
Initial
PCB Cone
(ppm)
132
159
171
Final
PCB Cone
(ppm)
156
143
157
Percent PCB
Reduction
0
10
8
Initial soil was PCB pit soil.
Surfactant - Hyonic NP-90* and Adsee 799*.
KN/9-»4/SITE.En»
-------
Samples from the second month of sunlight testing were not
analyzed for PCB degradation because of the lack of effect shown
during the first month of testing.
Soil Particle Size Testing
Following the poor results of the UV testing on the screened
and dried soils (less than 1/8 inch), the effect of particle size
was tested by grinding the clayey PCB contaminated surface soil
to pass a 230 mesh screen (particle size less than 63 microns) .
This ground soil was used as the basis for an additional ten
experiments. The medium pressure Hanovia lamp was again used in
a water cooled quartz light well. Exposure times ranged from 3-
20 hours, and surfactant concentrations were also varied. The
distance of the lamp to the soil and the cooling water rate were
kept constant to maintain a maximum measured soil surface
temperature of 54°C. All of the surfactant (Hyonic NP-90*} was
applied at the beginning of each experiment, the soil moistened
periodically with water only and tilled or raked periodically
during the UV exposure. Results from these tests are summarized
in Table 7.
KN/»-»wrrE.Erpo3/srrE3iuT.REv 19
-------
TABLE 7. SUMMARY OF UV PHOTOLYSIS TESTING ON FINE GROUND PCB
SURFACE SOIL
Test Surfactant Cone. {% Time
of Dry Soil) (Hours)
16
17
18
19
20
21,
22
23
24
25
2.0
2.5
0
2.8
2.1
2.8
2.3
2.0
2.3
2.0
6
3
3
10
3.7
3
3
20
10
20
Initial
PCB Cone
(ppm)
10,970
10,970
10,970
10,970
10,970
10,970
10,970
7,324
6,753
8,572
Final
PCB Cone
(ppm)
3,380
7,100
12,860
8,500
8,930
12,180
9,525
3,537
4,566
5,925
Percent PCB
Reduction
69
35
0
23
19
0
13
52
32
31
Initial soil ground to <230 mesh.
Hanovia 450 watt medium pressure continuous mercury lamp.
Soil depth - 0.25 inch.
Lamp/soil distance - 4 inches.
Surfactant - Hyonic NP-90«.
Temperature - Approximately 50 * C.
The results of Tests #16 through #22 are from analyses of a
single sample of treated soil. Tests #23, 24, and 25 had samples
removed and analyses performed as a function of treatment time.
The PCB concentration of soil moistened with water only (no
surfactant) and irradiated with the UV lamp was unchanged. The
PCB concentration of nearly all soils to which surfactant was
applied and then irradiated showed some decrease. Figure 1 shows
the kinetics for total PCB reaction in the 20 hour UV photolysis
Test #25.
A more detailed look at the effect of UV photolysis on PCB
chlorine level group (homolog) concentration is presented in
Table 8. This data shows the change in each PCB homolog
concentration (di through heptachlorobiphenyl) after UV
treatment. Congeners with three or more chlorine atoms (tri
through hepta-PCBs) showed a relatively consistent reduction in
concentration, whereas there was an increase in dichlorobiphenyl
concentration. The di-PCB fraction of the total PCB
20
-------
concentration increased from 24 percent to 42 percent as a result
of irradiation. The presence of monochlorobiphenyls (single
chlorine atom substituent) was not detected in any of the
samples. Figure 2 shows PCB chromatograms of untreated soil
versus soil irradiated for 20 hours with [/Flight (chromatogram
scales have been adjusted based on sample weights and dilution
volumes used in the analysis to present relative response equal
to relative concentration) . In the I/Vtreated soil, higher
chlorinated PCBs appearing later in the chromatographic analysis
are smaller and some of the peaks in the di and tri-PCB elution
window are larger and a few new peaks are seen in the di-, tri-
and tetra-PCB windows. These data are consistent with
degradation of higher chlorinated PCBs to lower chlorinated (di,
tri and tetra) PCBs.
Summaries of analytical data for Tests #1 through 16 and 23,
24 and 25 are included in Appendix C.
TABLE 8. UV PHOTOLYSIS - 20 HOUR TEST #25
PCB CONCENTRATION RESULTS - SOXHLET EXTRACTION
Starting Soil 20 Hour UV Treated Soil Percent Change
(ppm) (ppm) in PCB Cone.
Dichlorobiphenyls 2,027 2,465 22
Trichlorobiphenyls 1,134 556 -51
Tetrachlorobiphenyls 2,370 1,390 -41
Pentachlorobiphenyls 1,806 826 -54
Hexachlorobiphenyls 1,112 624 -44
Heptachlorobiphenyls 109 53 -52
Total PCB concentration
DCMA 8,570 5,925 -31
KN/9-94/SrrE.ETTO/SfrEJRIT.REV 2 1
-------
Figure 1
UV Treatment of Gas Pipeline Soil
Total Congener Concentration vs Time
3 9-000
Test #25
Irradiation Time (hours)
Figure 1. W Treatment of Gas Pipeline Soil
KN/9-«4/SITE.ETra/SrTE3MT.ItEV
22
-------
shlml'77"
[VO
co
hi
CD
Q
O
O
iu-
hj
O
rt
O
CO
O
hj
CD
rt
CD
(-1-
hj
CD
O>
rt
CD
O
H-
150-
90-
6O
30-
0
Figure 2 - GC of Treated and Untreated Soil
20 Hour UV Treated Soil
Di-PCB Tri Tefra
fTTTTTTrfTTTTTTTTTTTrnTTTTTTTTT
Perita Hexa
rTTTTTTTTTTTTllTTTTTTTTnT
120
100
80-
60-
20^
Untreat
16
•JTTTTJ7T
17
18 19
20 21 22
Retention Time (min.)
23 24
Hepta
TTTTTTTTTTTTTTTTl
;d Soil
TJTTTTTTTTTp-TTT
25 26
-------
CONCLUSIONS AND RECOMMENDATIONS
UV photolysis tests using high intensity UV lamps on TCDD
contaminated soil with surfactant application gave no detectable
change in contaminant concentration for the soil.
In tests using high intensity UV lamps on two different
soils with surfactant applications, PCS'reductions ranged up to a
maximum of 69 percent. In general, changes that were detected in
soil PCB concentrations were obtained using UV lamps and were
less than 50 percent, typically 15 to 35 percent. Best^results
were obtained using a 2-3 percent (Hyonic NP-90*) surfactant
spray loading on fine ground soil (<230 mesh) with a minimum bed
depth of 1/4 inch and a lamp to soil distance of 4 inches. In
these tests, with UVexposure times of six hours or longer, PCB
reductions were consistently in the range of 23 to 69 percent.
Loss of PCBs occurred for the higher chlorine level (tri through
hepta) PCBs. The loss of these PCBs was coupled with generation
of by-products in the di-, tri- and tetra-PCB gas chromatographic
elution windows. It was concluded from this data that
degradation of higher chlorinated PCBs to lower chlorinated PCBs
was occurring to a detectable degree.
Photolysis tests using sunlight exposure on PCB contaminated
soil with surfactant application gave no detectable change in
contaminant concentration for the soil. This was not surprising
as the results from high intensity UVlamp testing did not show
significant effectiveness.
The photolysis of TCDD or PCB contaminated soils using in-
situ or ex-situ configurations appears to be a process with
numerous variables which contribute to its success or failure.
Some of these variables appear to be more significant than others
but the net effect makes the process very difficult to predict.
Because many of the variables are dependent, the scope of testing
reguired becomes enormous. The variability of the analyses
resulting from the heterogeneity of the soil make interpretation
of results difficult. If the variables are to be properly
tested, the individual experiments need multiple replicates that
use the entire sample of each condition tested to remove the
variability introduced by subsampling for analysis. The
criterion of success or failure for each variable tested depends
on the ability of the analyst to rely on the data produced from
the experiment. The method of sample extraction has a
significant effect on the final analytical result. The
experimenter must rely on the analysis of multiple replicates to
interpret data by applying statistical methods.
The process should be tested using the above procedures
after a surfactant or solubilizing aid has been carefully
selected for the soil under consideration. Since this is one of
KN/»-9*SITE.Eni03«rrE3Rl1T.REV
24
-------
the major variables, the selection becomes critical to the
success or failure of the program to follow.
In addition, examination of different soil types should be
performed as results from these tests were much less successful
than results obtained from previous work using similar
conditions, indicating that soil type is a major variable. The
soil used in these tests had a higher humic content than the
sandy soils used in earlier successful testing.
Fine ground surface soil, both untreated and from the 20
hour W photolysis tests were supplied for biological treatment.
The 'soil residue from the 20 hour UV photolysis tests
consistently showed the highest effect from irradiation as given
by the reduction in PCB concentration.
KN/9-M/SITE.Enl(D/SITE3RPr.REV
25
-------
SECTION 4
CHEMICAL OXIDATION OF PCBS
INTRODUCTION
Chemical oxidation by Fenton's Reagent has been used to
destroy organic compounds such as formaldehyde (Murphy, et al.,
1989), azo dyes (Kitao, Kiso and Yahashi, 1982) and chlorinated
phenols (Barbeni, et al., 1987) in groundwater and wastewater.
The reaction is ideally performed at a pH of 2-4 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 decomposition, the reactive hydroxyl radical is
produced and it is capable of oxidizing organic contaminants.
However, if the desired oxidation reaction is slow, significant
amounts of hydrogen peroxide can be consumed in unproductive
decomposition instead of participating in the desired process.
Reaction conditions must be established to provide useful rates
of contaminant oxidation with efficient use of hydrogen peroxide
reagent.
Performing this reaction on soil contamination requires
making a slurry with the soil and the aqueous reagent. Testing
was performed in small batch systems of various sizes under
ambient conditions with concentrations of hydrogen peroxide and
PCBs monitored as a function of time. The tests were performed
on the ground PCB surface soil from Danville, Kentucky (GG4202)
which was used in the UV photolysis testing. The PCB
concentration of this soil was determined to be approximately
10,000 ppm Aroclor 1248.
The objective of the tests was to preliminarily investigate
the feasibility of applying the technology to soils contaminated
with PCBs. This process was investigated as an alternative to W
photolysis to provide initial contaminant degradation to more
easily biodegradable compounds. Conditions were established to
provide the best opportunity for observing an effect due to
treatment; reagent to soil ratios were high, pH maintained in the
range of 2-4 and hydrogen peroxide concentrations were maximized
by periodic replenishment. Iron concentrations were adjusted
from test to test to determine optimum concentration for maximum
PCB degradation.
EXPERIMENTAL PROCEDURES
Overview
Five experiments with Fenton's Reagent were performed at
ambient temperature. All five used the clay/humic, surface soil
GG4202, from the Danville, Kentucky site. This soil had been
26
-------
air-dried in a hood and screened to remove gravel and debris.
Each experiment was conducted in batch mode in covered glass
vessels with vents for gas escape. In each case, soil and
reagents were added to the reaction vessel and conditions were
established at the beginning of the experiment. Periodic
adjustments were made in pH and hydrogen peroxide concentration
as noted for each experiment. In each experiment the
reagent/soil mixture was continuously stirred except Experiment 1
which was stirred only during initial reagent additions and at
the 24-hour sample time. Table D-l in Appendix D is a summary of
soil, water, pH, iron sulfate and hydrogen peroxide initial
conditions for the tests. Further experimental procedural
details are presented elsewhere; however, important points
concerning these tests are the following:
Only Experiment 1 was not stirred continuously.
Experiment 2, Flask 2 was a control: no iron sulfate was
added to this flask.
Experiments 4 and 5 were considerably larger scale and
periodic samples were taken for PCB analysis.
Feed Soil Preparation
The soil used for all experiments was from the air-dried PCB
surface soil sample (GG4202) . The soil was ground in a standard
kitchen blender and sieved. Particles not passing the sieve were
reground in the blender. Fenton's Reagent Experiments 1, 2, and
3 used soil passing 230 mesh standard U.S. Sieve. Fenton's
Reagent Experiments 4 and 5 used soil passing 100 mesh but
retained by 200 mesh.
Sampling
Experiment 1 was sampled for PCB at 24 hours and at the end
of the'experiment at 92 hours. Each jar/flask was stirred for 15
minutes and sampled using a 60 cubic centimeter (cc)
polypropylene syringe. The residual hydrogen peroxide in the
sample was neutralized with sodium bisulfite. After settling,
any aqueous supernatant was removed and the residual wet solids
dried at 48°C for 20 hours. Samples were extracted by Soxhlet
extraction and analyzed by GC/ECD.
Experiment 2 was periodically sampled for hydrogen peroxide
and pH. Typically, pH was measured and stirring was stopped long
enough for a 5- or lo-milliliter (mL) supernatant sample to be
pipetted off for potassium permanganate titration. The solids
were rinsed from the flask into tared wide-mouth jars and allowed
to settle. The reaction flasks were rinsed (3 times) with water,
followed by methylene chloride (3 times). The rinsewater, the
original supernatant, and the supernatant from the transferred
KNAWSIrE.ErPo3ismR.MEv
27
-------
solids were combined and extracted with methylene chloride.
These extracts were combined with the sonication extracts of the
solids and, following solvent exchange, analyzed as a single
sample by GC/ECD.
Experiment 3 was sampled similar to Experiment 2 with the
following changes. The excess hydrogen peroxide was quenched
using sodium bisulfite after titration. The remaining solids
after liquid removal were air dried before extraction. In
addition, the extracts for the liquid and solids were analyzed
separately.
Experiment 4 and 5 samples were withdrawn from below the
surface of the stirred mixture using a 60 cubic centimeter (cc)
polypropylene syringe fitted with a short length of Teflon™
tubing. The sample was then transferred to a clear glass jar
with a Teflon™ lined cap and allowed to settle for at least 45
minutes. Aliquots of the supernatant were then removed and
immediately titrated for residual hydrogen peroxide. The
remaining supernatant was then carefully removed from the jar and
replaced into the reaction flask. The wet solids samples were
then weighed, quenched with sodium bisulfite and then reweighed.
The inside of each sample jar was then rinsed with a small amount
of deionized water and the samples were allowed to air dry in a
fume hood.
MATERIALS AND METHODS
Equipment
Experiment 1 was performed in 250-mL and 500-mL straight-
sided glass jars, Experiment 2 used two 250-mL Phillips' flasks
and Experiment 3 used two 125-mL Erlenmeyer flasks. The mixtures
were stirred with Teflon™-coated magnetic stir bars.
Experiments 4 and 5 were larger scale. Experiment 4 was
done in a straight-sided a-liter (L) Pyrex jar. The soil-water
mixture was agitated with a two-bladed approximately 30° pitch
polypropylene covered steel stirrer driven by a variable speed
lab motor. Stirrer speed requirement was determined by prior
testing of a small amount of sand in water.
The Experiment 5 slurry was reacted in a baffled 4-L
reaction pot/kettle. Stirring was provided by a stainless steel
three-bladed turbine propeller driven by a variable speed
electric motor. As in Experiment 4 a good mixing speed was
determined using a clean sand-in-water mixture.
2 8
-------
Chemical Reasents
Iron (II) Sulfate, FeS04'7H20 - Alfa, ACS reagent
Hydrogen Peroxide - Aldrich, ACS, 30 percent weight,
stabilized
Potassium Permanganate - Mallinkrodt, volumetric
solution; 1.00 ± .005N
Sulfuric Acid - Mallinkrodt, 95.0-98.0 percent, AR
PCB Aroclor 1248 standard - Chem Service, F110
Sodium Bisulfite - Mallinkrodt, AR, granular
Sodium Oxalate - Mallinkrodt, AR
Surfactant -Stepan Co, Bio-Soft S-100, Dodecylbenzene
sulfonic acid
pH Measurement
The pH was measured using a calibrated pH meter and a
combination pH probe. Measurements were made directly in the
reaction vessel contents while they were being mixed.
H vdr og en Peroxide
Hydrogen peroxide concentrations were measured by titration
of a 2.0 - 10 mL sample aliquot diluted with 25 percent sulfuric
acid solution using a potassium permanganate standard solution.
The sample density was taken as 1.0 gm/mL and the peroxide
calculated from:
• ^ a nn KMnOt) (N KMnO.) (1.7)
weight % H202 = - 4 *
mL of sample
PCB Analysis
Sample Preparation--
The air-dried soil samples were crushed to a dust in their
sample jars with a clean stainless steel spatula and mixed
thoroughly. For Experiment 4, the sample jars were not scraped,
but methylene chloride was added after solids removal to extract
any residual PCBs on the walls of the sample container. This
extract was analyzed separately to determine loss from PCBs
adhering to the sample jar surface.
Samples were extracted by one of two methods: sonication or
Soxhlet. Some samples from Experiments 3 and 4 were extracted by
both methods.
Sample Extraction - Sonication—
The sonication extraction procedure was based on EPA Method
3550 (SW-846). Crushed dried soil sample aliquots weighing 2.0 to
2.5 grams were mixed with 2-3 grams of oven-treated sodium
sulfate in a 20 mL glass vial and extracted with lo-12 mLs of 1:1
volume to volume (v/v) acetone to methylene chloride solvent by
KN/9-94/srrE.EtT03/SrrE>MT.REV 29
-------
sonication. After sonication, the extract was gravity filtered
through a bed of sodium sulfate and collected directly into a 50
mL volumetric flask. This process/cycle was repeated three times
for each sample. Sample extracts were solvent exchanged to
hexane by Kuderna Danish (KD) evaporators for analysis by GC/ECD,
or they were diluted with methylene chloride for analysis by gas
chromatography with flame ionization detection. (GC/FID).
Sample Extraction - Soxhlet—
The Soxhlet extraction procedure was based on EPA Method
3540A (SW-846). Soxhlet extractions were done on two scales:
the procedure described in Method 3540A and a micro procedure
essentially identical except that the entire setup is
proportionally smaller (2 gram sample size) . Sample extracts
were solvent exchanged to hexane by KD evaporators if they were
to be analyzed by GC/ECD, or they were diluted with methylene
chloride for GC/FID analysis.
Instrumental Analysis--
The analysis of PCBs was performed by one of two methods:
EPA Method 8080 (SW-846) or a DCMA (PCB homolog) procedure, using
either GC/ECD or GC/FID instrumentation. Samples were initially
analyzed by GC/ECD for its selectivity and sensitivity, but
because of the high concentrations of PCBs in the samples, ease
of sample preparation, and extended linear range, analyses were
switched to GC/FID instrumentation. The analytical methods were
applied in the same manner for either instrument. Samples from
Experiment 3 were analyzed by both GC/ECD and GC/FID. Untreated
samples were analyzed and quantified for Aroclor 1248 using
methods consistent with SW-846 Method 8080. Treated samples were
analyzed and quantified by the DCMA semi-specific PCB homolog
method.
QUALITY ASSURANCE
Potassium Permancfanate
The potassium permanganate (KMn04) titrant solution was
prepared from a 1.00 f.005 Normal (N) standard by diluting an
aliquot 1:20 with deionized water. The titrant concentration was
verified by titration of an accurately weighed sample of sodium
oxalate in 12.5 percent sulfuric acid and found to be 0.052 N.
PCB Analysis
The air-dried soil samples from Experiment 4 were oven, dried
at 106°C after aliquots had been taken for Soxhlet extraction.
The percent moisture ranged from 1.4 to 2.8 percent; the average
was 2.2 percent. These values were low and consequently the
analytical PCB results were not corrected for this amount of
moisture in Experiments 4 and 5.
KN/9-94/SrrE.ETFO
-------
To check for the possibility of PCB adhering to the sample
jar walls, the crushed soil was removed from each Experiment 4
sample jar and the jars themselves filled with methylene
chloride. Insignificant amounts of PCB were found in these jar
soak extracts. PCB loss from adhering to sample jar surface was
found to be less than 2 percent in all cases and was not
considered significant.
Analyses were performed on three samples (feed and flask
samples from Experiment 3) by both GC/FID and GC/ECD
instrumentation to evaluate differences. The relative percent
differences (RPD) in results from analyses by both instruments
were 3, 15 and 31 percent. The difference between instrumental
methods was not considered significant, since data from each
experiment was obtained by one method or the other and results
from GC/ECD analyses were not compared with results from GC/FID
analyses, or vice versa.
Analyses were also performed to determine if PCB recovery
was complete after three sonication extraction cycles of a sample
to assure PCB loss was not occurring from incomplete extraction.
An additional two sonication extraction cycles were performed on
a sample after the extraction procedure using three cycles had
been performed. The fourth and fifth extraction cycle extracts
were analyzed separately and found to contain only two percent of
the PCBs extracted by the first three cycles. The three
extraction cycles were considered sufficient since 98 percent of
the PCBs recoverable by sonication were being extracted. Further
analysis details are supplied in a summary in Appendix E.
A difference was noted during these tests, however, between
sonication and Soxhlet extraction efficiency. In Experiment 4,
samples were analyzed by both procedures and PCB recovery by
sonication ranged from 43 to 74 percent of the PCBs recovered by
Soxhlet extraction. The average ratio of PCBs recovered by
sonication versus Soxhlet extraction was 58 percent with a
relative standard deviation (RSD) of 15 percent. The sonication
extraction results were consistently lower than those obtained by
Soxhlet extraction and although the difference was significant,
the loss of PCBs could be monitored by either method as long as
data from one method of extraction was not used with data from
the other. The results from Experiment 4 showed that the same
conclusions would be reached using data from either PCB
extraction method as long as the data was distinguished by the
extraction method used.
A check on the reproducibility of the micro Soxhlet
extractions for PCB was performed by triplicate extractions of
samples from Experiment 5. This was conducted to evaluate
variability which may have been introduced because of the small
sample sizes (2 grams) used in the micro procedure. Nine sample
sets were extracted in triplicate and one in duplicate. The
KN/9-W/SrTE.ETKQ/SrrE>Wr.REV 3 1
-------
highest RSD or RPD of any set was 4.7 percent; the average
RSD/RPD of all nine sets was 3.0 percent. Detailed individual
and sample set values are presented in a summary in Appendix E.
An unknown PCB quality control (QC) sample was analyzed by
both the TDL and the Biotechnology Application Center (BAG) as an
independent QC check on PCB calibration. The sample was prepared
from an independent source of Aroclor 1248 and provided to both
labs as a QC sample. The percent recoveries reported by both
laboratories engaged in work for this project were well within
the expected +25 percent for demonstration of analytical control.
In addition, the interlab agreement was excellent. There was
less than 3 percent RPD between the two laboratories' results.
Further analysis details are supplied in a summary in Appendix E.
Finally, two micro-Soxhlet sample extracts from Experiment 4
were spiked with an equivalent amount of PCBs from a known
standard to check for interferences and extraneous peaks. Both
spikes were prepared by adding 2.0 milliliters (ml) of an Aroclor
1248 standard (at about the same concentration as the extracts)
to 2.0 mLs of the sample extract. PCB recoveries for the spike
samples were 90 and 102 percent, showing excellent PCB
accountability. Further analysis details are supplied in a
summary in Appendix E.
RESULTS AND DISCUSSION
Experiment 1-24 and 92 Hours, No Mixing
The first experiment consisted of two batch reactions with
the mixtures stirred only during the initial reagent additions
and during the 24 and 92 hour sample times. Table 9 summarizes
the conditions used. The two reactions differed primarily in the
ratio of reagent to soil. Flask 85 (GG4202-1018-85) had a water
to soil ratio of 0.8 and Flask 86 had a ratio of 3.1. All
subsequent tests used higher ratios, in the range of 8 to 10.
After 9'2 hours, dried soils were Soxhlet extracted and aqueous
phases were separatory funnel extracted. Analyses by GC/ECD of
the aqueous and solid phase extracts indicated no significant
change from the starting 'PCB concentration. At least 95 percent
of the PCBs remained at the end of the tests for both reactions.
As shown in Table 9 no significant change in PCB concentration
was found in the samples from this experiment. Further detail on
these analyses is included in Table D-2 in Appendix D.
KN/9-*t/SrrE.ETP03/SrroiUT.REV
32
-------
Fifteen vials were established for each treatment. Three
vials from each treatment set (15 total vials) were sacrificed at
five time points. The time points were study initiation, 24
hours (hr), 48 hr, 94 hr, and 140 hr. Vials were extracted by
sonication with 2 mL of pentane (Aldrich Chemical Co., Milwaukee,
Wisconsin) for one minute in a Bransonic 220 Sonicator Bath.
Pentane extracts were analyzed by a Hewlett Packard 5890A
Gas Chromatograph (GC) with an automatic sampler, ECD, splitless
injector, and Supelco SPB-1 capillary column [75 meter by 0.75
millimeter (internal diameter)].
Nitrogen was used as the carrier and make-up gas. The
carrier gas flow was 2 milliliters per minute (mL/min) at 40°C.
The make-up gas was introduced at 60 mL/min. During sample
analysis, the GC oven initial temperature was 45°C. This was
held for one minute, raised to 150°C at a rate of 10°C/min and
then to 300°C at a rate of 3°C/min.
held for 5 minutes.
Bioslurrv Evaluation
The 300 C temperature was
Three PCB-contaminated soils were evaluated for biological
reduction of PCB congeners. Soils employed were identified as
untreated soil (Sample ID No. GG4202-1018-61), surfactant/UV-
treated soil (Sample ID No. GG4202-1018-96A), and New Englanq
Superfund Site soil.
The following treatments were prepared:
Treatment Bl - surfactant/UV-treated soil, PAS medium,
BAG 17 culture
Treatment B2 - surfactant/UV-treated soil, PAS medium,
H850 culture
Treatment B3 - surfactant/UV-treated soil, PAS medium,
Hydrochloric acid (Killed control)
Treatment B4 - Untreated soil, PAS medium, BAG 17 culture
Treatment B5 - Untreated soil, PAS medium, H850 culture
Treatment 86 - Untreated soil, PAS medium, Hydrochloric
acid (Killed control)
• Treatment B7 - New England soil, PAS medium, BAG 17
culture
Treatment B8 - New England soil, PAS medium, H850 culture
KN/»-»4/srrE.Erw3/srrE3iii>T.REv
-------
TABLE 9. FENTON'S REAGENT EXPERIMENT 1
SUMMARY OF PCB RESULTS
Series: GG4202-1018-85 GG4202-1018-86
Feed Soil PCB, ppm 10,930
24 Hour'
Soil PCB, ppm 11,210
92 Hour'
Soil PCB, ppm 11,360
Decanted aq. (43 ml) PCB, ppb
10,930
10,940
9,710
779
* Duplicate samples, dried at 48°Cfor20 hr,
The flask mixtures were not pH adjusted after reagent
addition, since the pH was less than 4. Additional observations
and comments:
1. The stirring mixtures fizzed slightly during H,0,
addition.
2. After initial stirring stopped, the 85 (1018-85) series
(the thicker mixture) formed a stable foam within 10
minutes which filled the 250 mL jar.
3. Similarly, the 86 (1018-86) series formed a thinner (more
gas) foam layer which was easily reincorporated into the
mixture.
4. The analytical results were corrected for solids added
and for residual moisture remaining after drying at 48°C.
5. Residual moisture was calculated based on known solid
weights and ranged from 2 percent to 14 percent. All
samples appeared dry and crumbled easily.
6. The "single shot" addition of H202 and not stirring the
mixture may have been too restrictive.
7. The ratio of available H202 to total oxidizable components
may have been too low.
To address these possible restrictions a second experiment
was designed.
KN/»-»4/SrrE.En>03/srrE3RIT.REV
33
-------
Experiment 2-162 Hours, Continuously Stirred
Experiment 2 used a higher ratio of reagent to soil (9.7), a
higher concentration of iron (2.5 percent of the soil in Flask
I) , and incorporated a control reaction which had no added iron
(Flask 2). See Table D-l for reagent details.
Table D-3 in Appendix D shows the hydrogen peroxide analysis
and pH data collected during the experiment.
Both flasks were continuously stirred for 162 hours except
for short periodic intervals when the agueous supernatants were
sampled for hydrogen peroxide titrations. At these times,
supplemental hydrogen peroxide was added if low. Ferrous sulfate
solution was also added to Flask 1 to compensate for losses
caused by the removal of aliguots for hydrogen peroxide analysis.
The effects of the iron sulfate and hydrogen peroxide
additions upon pH and temperature were significant. The initial
iron sulfate addition in Flask 1 dropped the pH from 6.3 to 5.0.
The hydrogen peroxide drove the pH further to 2.8 with subsequent
foaming and increase in temperature. In contrast, the Flask 2 pH
went from 6.3 to 6.7 upon addition of the same amount of hydrogen
peroxide with less foaming and no significant temperature change.
Table 10 shows the total congener concentration for each
chlorine level (homolog totals) plus the total PCB results given
by the DCMA analysis for Flask 1 and Flask 2 at the end of the
test. These values include the agueous phase and flask rinse
extracts. The starting soil was also analyzed by sonication
extraction and the concentration was 7,325 ppm PCBs. Percent
reductions in PCB concentrations given by Flask 1 versus the
control, Flask 2, are also presented. The treatment decreased
the total PCB concentration by 45 percent of the control
concentration.
KN/»-»4«rrE.ETW3«rrE3RJ1T.REV
34
-------
TABLE 10. FENTON'S REAGENT EXPERIMENT 2
PCB CONCENTRATION RESULTS - SONICATION EXTRACTS
Flask 1
Starting (H202 and Fe)
Soil (ppml
Flask 2
(Control, no Fe)
(ppm)
Percent PCB
Reduction
(vs. Control)
Dichlorobiphenyls — 125
Trichlorobiphenyls — 169
Tetrach lorobiphenyls — 946
Pentachlorobiphenyls — 1,265
Hexachlorobiphenyls — 1,166
Heptachlorobiphenyls — 86
Total PCB 7,325 3,760
concentration (ppm)
DCMA, GC/ECD
1,272
879
2,125
1,594
896
77
6,840
90
81
55
0
0
0
45
In contrast to the trend of PCB loss observed from UV
irradiation, PCB loss decreased with increasing chlorination
level in the same manner as the oven heated sample described
earlier (I/Vphotolysis testing) . That is, higher percentage
losses were observed for lighter chlorinated congeners, di, tri
and tetrachlorobiphenyls with smaller losses observed for penta,
hexa and heptachlorobiphenyls. Also, as shown in Figure 3, the
GC chromatogram of the treated soil shows no new peaks from.by-
product generation or alteration of the PCB pattern (chromatogram
scales have been adjusted based on sample weights and dilution
volumes used in the analysis to present relative response equal
to relative concentration). What is shown is a decrease in the
pattern trending from late to early elution. These results are
consistent with loss of PCBs through volatilization, although
reaction with hydrogen peroxide cannot be ruled out.
Based on the observations of this experiment a third flask
experiment was designed with the objective of determining the
effect of iron levels in the reaction system.
Experiment 3 - 118 Hours, Iron Effect
Experiment 3 was designed to verify the PCB reduction seen
in Experiment 2 and to investigate the effect of iron
concentration in the reaction mixture. Flask 3A contained the
equivalent of 100 ppm iron while Flask 3B contained the
equivalent of 450 ppm iron. Both iron levels were considerably
35
-------
o>
H-
hi
CD
CO
o
CD O
PL t^
1-1
9'
o
o
CO
CD rt
co O
CO
V
CD
O>
^
CD
3
ri-
300-
240-
180-
120-
60-
0-
GC Chromitogrims of Fenton's Reagent Treated and Control Samples - Exp. #2
Fenton's Reagent Treated Soil - Flask 1
Di-PCB Tri
1
Tetra
Penta
Hexa
: fire 130
140-
120-
100-
80-
60-j
40-;
20;
Hepta
Control - Flask 2
17
l| I I I I [4-N-l | I I I I | I I I I | I I I I | I I I I [ I I I I I I I II [ I I I
18 19 20 21 22
Retention Time (min.)
| I | I | | I
23
I [ | | | I
24
I I | |
25
-------
lower than the Experiment 2 Flask 1 concentration of 2,200 ppm
iron.
Table D-4 in Appendix D shows the hydrogen peroxide analysis
and pH data collected during.the experiment.
In this experiment the pH was adjusted to 2.2-2.5 with
concentrated sulfuric acid after the iron sulfate addition but
before the hydrogen peroxide addition. The temperature rose only
slightly in the high iron flask, 3B, after hydrogen peroxide
addition, but produced a foam for nearly two hours. Flask 3A
foamed only slightly with no perceptible temperature change.
Both flasks were periodically sampled for hydrogen peroxide. The
volume removed was made back up with deionized water or 23
percent hydrogen peroxide solution, as appropriate.
As detailed in the sampling section, the soil extracts and
flask rinses - aqueous phase extracts were analyzed separately to
check for loss of PCB through the intermediate sampling of the
supernatant for hydrogen peroxide determination.
Table 11 summarizes the results of GC/ECD analysis of the
sonicated soil extracts (feed and treated soils) and the flask
solvent rinse - aqueous phase extracts at the completion of 118
hours of reaction.
TABLE 11. FENTON'S REAGENT EXPERIMENT 3
PCB CONCENTRATION RESULTS - SONICATION EXTRACTS
Feed Flask 3A Flask 3B
(118 hours) (1 18 hours)
Percent of total PCB from flask
rinse and 'aqueous extract
Total PCB - soil basis (ppm 1 6,833
Percent reduction of PCBs
4.1
3,171
54
0,2
3,762
45
The low values for the flask rinses and aqueous phase
extracts indicates minimal losses through supernatant sampling or
reaction vessel holdup/wall adhesion, although a clear thin
hydrophobic film was noted on the flask walls during and at the
conclusion of the experiment.
Significant amounts of PCB have either been reacted or lost,
with little significant difference noted for the different
amounts of iron used in the tests for Experiments 2 and 3. The
37
-------
trend of PCB loss as a function of homolog group was also
consistent with results from Experiment 2.
Experiment 4-2 Liter Reactor. 850 Hours
The fourth experiment was designed to allow multiple soil
samples to be taken over time and to duplicate the previous
results on a larger scale. The equipment and sampling procedures
are described in detail in the experimental procedures section.
Experiments 4 and 5 used soil which had been freshly ground to
between 100 and 200 U.S. sieve mesh.
During startup, initial addition of hydrogen peroxide caused
foaming and loss of solution into a containment tray. Addition
of sulfuric acid reduced the foaming and allowed replacement of
the overflow solution. The exterior of the reaction flask was
rinsed and this rinsate was added to the flask. it was estimated
that less than 0.3 percent of the soil was lost in the entire
episode.
During the reaction, small amounts (less than 2 mL) of 50
percent sodium hydroxide or 25 percent sulfuric acid were
periodically added to maintain the reactor pH between 3 to 3.5.
As before, periodic supplemental additions of 30 percent hydrogen
peroxide were also made.
A few sample aliquots were initially extracted by sonication
and soon after fresh aliquots of the same samples were Soxhlet
extracted. All extracts were analyzed by GC/FID.
Table D-5 in Appendix D details the hydrogen peroxide
analysis and pH data collected during the experiment and the
results of PCB analyses that were performed.
The results of the PCB analyses are graphed in Figure 4.
After 845 hours the PCB reduction in the flask was 34 percent.
This reduction is somewhat less, but consistent with the results
from Experiments 2 and 3. The reaction time was much longer than
that for Experiments 2 and 3; however, the decrease in PCB
concentration appears to have occurred in the first 100 hours.
Considerable scatter is evident in the PCB analysis data.
The initial discontinuity in the data from 0-50 hours is most
likely due to the reactor overflow episode; however subsequent
anomalies cannot be fully explained. Some of the PCB variation
in the sample analyses may be a result of particulate size
segregation during sampling.
As noted earlier, the sample jars were tested for residual
PCB after the soils had been sampled and removed, and the
residual PCB in the sample jar was less than 0.7 percent of the
total PCB present in any sample jar.
KN/9-*4«rrE.EriTO«ITE3MT.IlEV 3 8
-------
I
H-
hi
CD
Figure 4 - PCB Concentration vs Time for Fenton's Reagent Exp. #4
10000
U)
VO
o
dd
o
o
3
o
CD
CD
<
w
^
H-
3
CD
Hi
O
hi
^d
CD
^3
rt
O
CD
CU
U3
CD
P
rt
8000
•D
6000
(M
4000
2000
100
200
300
400
Total Hour*
500
600
700
BOO
900
-------
The PCB reduction and trend in the data is similar for the
sonication and Soxhlet extraction data. The consistent
difference between the Soxhlet and sonication values may
represent more tightly adsorbed or shielded PCB in the clay-type
soil which is not recovered during sonication extraction.
Experiment 5, the final Fenton's Reagent experiment, was
planned to address the variation seen in Experiment 4 and to test
a surfactant enhancement.
Experiment 5-2 Liter Reactor-Surfactant Addition, 184 Hours
Experiment 5 differed from Experiment 4 in several aspects.
The all-glass reactor was a four baffled type with a 4-liter
capacity to allow for foaming and had improved mixing to keep
larger particles suspended.
The major differences in the experimental procedure included
the adjustment to pH 2.95 before any hydrogen peroxide addition,
the removal of triplicate sample aliquots at each sampling
interval, and finally, the addition of a surfactant at the
midpoint of the test. These changes resulted in an experimental
startup without incident.
Table D-6 in Appendix D details the hydrogen peroxide
analysis and pH data collected during the experiment and the
results of PCB analyses that were performed. All samples from
Experiment 5 were Soxhlet extracted.
With a starting concentration of 9,400 ppm PCBs and a final
concentration of 8,048 ppm PCBs after 185 hours, the results
indicate destruction of PCBs to be 14 percent. This may have
been due to a lower average hydrogen peroxide concentration 'for
this test, as well as a low iron concentration, .09 percent of
the soil. Hydroxide radicals from hydrogen peroxide reaction
with iron are responsible for advanced oxidation reactions. This
experiment had the lowest combination of iron and hydrogen
peroxide and based on this, should be expected to provide less
effectiveness for PCB reaction. This combination also provides
the lowest amount of hydrogen peroxide decomposition and oxygen
generation which would produce less purging of PCBs from the
reaction mixture.
After 117 hours of treatment, a solution of Bio-Soft S-100
surfactant was added to the reaction flask to bring the solution
concentrationto 100 ppm to see if surfactant addition woul.d aid
PCB degradation. Bio-Soft S-100 is an anionic surfactant,
dodecylbenzene sulfonate, which biodegrades and is relatively
stable in oxidative systems. The PCB concentration was not
detectably affected by the surfactant addition.
40
-------
Summary of Chemical Oxidation Testing Results
Table 12 presents a summary of testing results along with
key experimental parameters. These data indicate that PCB
reductions due to chemical degradation required the presence of
iron, but was not strongly affected by the iron concentration.
Most important was maintaining a high concentration of hydrogen
peroxide in the presence of the iron. This is not easy, however,
because this condition is coincident with a high rate of hydrogen
peroxide decomposition. In order to be effective, efficient use
of the reactive intermediates must be achieved. The other
observation from the data is that relatively long reaction times
(100 hours) under these conditions appear to be necessary in
order to achieve a detectable change.
TABLE 12. SUMMARY OF CHEMICAL OXIDATION (FENTON'S REAGENT)
TESTING
Test/ Soil
Experiment Flask (g)
1b 85 50
86 50
2 1 10
2' 10
3 3A 8.0
3B 8.1
4 1 170
5' 1 196
H202
Water/ Fe Cone. (%}
Soil Ratio pH (% of Soil) Average*
0.8
3.1
9.7
9.4
8.4
9.5
10.1
8.0
3.6
3.3
2.8
6.7-4.5
2.5
2.2
3.1
2.9
0.5
0.5
2.5
0
0.1
0.5
.09
.09
2.5'
0.7'
.07
1.4
1.8
0.87
1.6
0.88
Time
(Hours)
92
92
162
162
118
118
845'
184
Percent
Reduction of
PCB Cone.
0
4
45"
7
54
45
34
14
' Time weighted average .
b Reaction mixtures were not continuously stirred.
c Hydrogen peroxide added at beginning of experiment and was not monitored or adjusted
thereafter.
d As compared to control: Flask 2.
* Control reaction.
' No further decrease in PCB concentration observed after 211 hours.
9 Surfactant addition (100 ppm Bio-Soft S-l 00) was made at 117 hours of experiment,
KN/9-94/SrrE.CTPOJ/SlTE3RllT.REV 4 1
-------
CONCLUSIONS AND RECOMMENDATIONS
Under controlled conditions and using relatively high
reaction medium to soil ratios, PCB concentration reductions of
up to 55 percent were achieved in reaction times on the order of
100 hours. These reactions were conducted at ambient
temperature, but some heat was generated by decomposition of
hydrogen peroxide which was most rapid at the beginning of the
tests and when supplemental additions of hydrogen peroxide.were
made.
PCB loss was not strongly affected by iron concentration in
the range of 0.09 percent to 2.5 percent; however, the presence
of iron (lowest concentration used in the tests was 0.09 percent
of the soil) was required for meaningful effectiveness.
The most important parameter for PCB reduction was
maintaining optimum hydrogen peroxide concentration (2 percent in
the reaction solution) in the presence of iron at concentrations
above 0.1 percent. These attempts are thwarted by high rates of
hydrogen peroxide decomposition under these conditions.
The use of an alkyl benzenesulfonic acid at 100 ppm in the
reaction solution had no detectable impact on the rate of PCB
loss.
The loss of PCBs occurs predominantly through loss of
lighter chlorinated, more volatile PCBs in a smooth trend to
heavier chlorinated PCBs without generation of by-product peaks
or PCB pattern alteration, as seen in the UVphotolysis tests.
This behavior suggests PCB loss is occurring via volatilization.
Volatilization may occur during gas purging (foaming) of the
solution from generation of oxygen by hydrogen peroxide
decomposition. This process could be verified by conducting
experiments in reaction vessels vented through activated carbon
traps. Analysis of the carbon traps for PCBs would quantify loss
through volatilization.
In order for Fenton's Reagent to be of significant use, the
rate of reaction must be increased. The use of a solubilizing
aid or surfactant to increase the solubility of PCBs has
potential. The test described herein used a surfactant, which is
relatively stable to pxidative systems, at low concentration. To
further evaluate the impact of surfactant addition on this
reaction, tests should be performed with higher concentrations of
surfactants, 0.2 percent to 1 percent.
In addition, if the PCB reaction rate is limited by PCB
solubility (mass transfer into solution), the rate of reaction
would be more or less independent of soil PCB content. Moderate
to low PCB concentration soils (100-500 ppm) would be detoxified
at a faster rate than the high PCB content soil used in these
KN/S-M/SrrE.ETKD/SITESIUT.REV
42
-------
tests. Tests should be conducted on a lower PCB concentration
soil, such as the PCB pit soil from Danville, Kentucky, to
evaluate the effect on PCB reaction rate.
4 3
-------
SECTION 5
BIOLOGICAL TREATMENT
INTRODUCTION
The primary objective of this investigation was to evaluate
the effect of surfactant/W-treatment on aerobic, polychlorinated
biphenyl (PCB) biodegradation. Aerobic biodegradation of the
lower chlorinated PCB (1-3 chlorines) has been well-documented
(Ahmed and Focht, 1973; Furukawa and Matsumura, 1976; Furukawa et
al., 1978; Shiaris and Sayler, 1982; Masse et al., 1984; Brunner
et al., 1985; Sylvestre et al., 1985; Barton and Crawford, 1988;
Adrians et al., 1989; Pettigrew et al., 1990). However, the
more highly chlorinated congeners are generally resistant to
microbial attack although, there have been reports of microbial
degradation of the highly chlorinated PCB congeners (greater than
4 chlorines) (Furukawa et al., 1978; Furukawa et al., 1979; Bopp,
1986; Bedard et al., 1987a; Bedard et al., 1987b) . In situ
stimulation of PCB degradation has been shown for Hudson River
sediments (Harkness et al., 1993).
Biological degradation of PCB congeners is highly affected
by chlorination pattern and the number of chlorines per biphenyl.
Congeners chlorinated in the 2,4- and 2,6- positions are
resistant to aerobic metabolism (Furukawa et al., 1978; Bedard
and Haberl, 1990) . Further 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. First, the
bioavailability of the PCB should be increased and second,
decrease the amount of chlorines per biphenyl ring. This study
addresses the bioavailability and microbial attack of PCB after
the combined surfactant/UV treatment of highly contaminated PCB
soil. The theory behind this approach is that surfactants would
render PCB bioavailable and surfactant/UV treatment would affect
dechlorination, making the desorbed PCBs more amenable to
biological treatment. ,
The surfactant/W treated soil used in these tests was
residual soil from the 20 hour UV photolysis tests. The original
source of this material was surface soil from the highly
contaminated Texas Eastern Site in Danville, Kentucky. This
material was fine ground prior to UV photolysis to pass a 230
mesh sieve (particle size less than 63 microns) and had 2 percent
by weight of Hyonic NP-90* surfactant applied during the test.
After trv photolysis, the PCB concentration was reduced to about
half of the starting concentration (starting concentration was
approximately 10,000 ppm PCBs: Aroclor 1248). Fine ground,
untreated soil from this site was also provided for biological
treatment.
KN/9-94/SrrE.ETP03/SrrE3IUT.REV
44
-------
EXPERIMENTAL DESIGN AND TEST OBJECTIVES
Physical dechlorination of weathered PCB-contaminated soil
to produce material which would facilitate biological
transformation of specific congeners was conducted. Materials
produced were subjected to bench-scale biotreatability testing.
The testing objectives included:
Isolating PCB-degrading microbial species from
environmental soil samples
Determining the biological reduction of weathered PCB
congeners in soil samples
Determining impact of PCB-biodegradation inducers and
growth substrates on congener reduction
Determining the effectiveness of the combined physical
and biological PCB treatment.
All test objectives were met during the course of the
investigation.
PCB-contaminated soils treated with Hyonic NP-90® and
exposed to UV light at 254 nm were employed during
biotreatability testing. The investigation examined the
biodegradability of the PCB in the surfactant/UV treated soil,
the untreated soil, and a separate PCB contaminated soil known to
have biological activity against PCB. The biotreatability
laboratory-scale investigation was conducted in four separate
phases to achieve the defined testing objectives. The four
phases of investigation were:
Phase 1 - Isolation of PCB-degrading bacterial cultures
Phase 2 - Rapid PCB Screening Assay
Phase 3 - Bioslurry evaluation
Phase 4 - Enhanced bioslurry evaluation.
During Phase 1 testing, PCB-degrading organisms were
isolated from impacted sail. In addition, known-PCB degrading
microorganisms were obtained from General Electric Company (GE) .
Phase 2 used a Rapid PCB Screening Assay to further characterize
isolates selected during Phase 1. The results of both phases
were evaluated and bacterial cultures were selected for further
testing.
The ability of selected organisms to biotransform PCB
congeners in surfactant/UV-treated and untreated soil was
evaluated during two bioslurry treatment experiments. Phase 3
experimentation evaluated the biological reduction of PCB
congeners in surfactant/UV-treated and untreated soils. A
following bioslurry experiment (Phase 4) evaluated the impact of
KN/»-»4«rre.En«J/SITE3IUT.IlEV
45
-------
PCB-biodegradation inducer and growth substrate addition on
congener removal.
All experiments were conducted under aerobic conditions and
with adequate replication and experimental control to determine
the effect of biological removal. All biological testing was
conducted by IT personnel at IT's BAG and the University of
Tennessee Center for Environmental Biotechnology (CEB). Both
facilities operate under a State of Tennessee exemption for
treatability testing.
MATERIALS AND METHODS
Isolation of PCB-Desraders
Isolation of PCB-degrading bacteria from untreated soils and
from a New England Superfund Site was attempted. Bacteria
demonstrating activity against biphenyl and PCB were found in the
New England Superfund Site soil.
Cultures were isolated by mixing one gram of soil with 25
mLs of phosphate-buffered mineral salts medium, referred to as
PAS medium (Bedard et al., 1987). This medium was augmented with
biphenyl crystals (Mallinckrodt Inc., Paris, Kentucky) until
saturation in the medium was reached. Biphenyl saturation in
water at 25°C is 7 milligrams per liter (mg/L) .
The soil slurries were incubated at 25°C and 200 revolutions
per minute (rpm). Following 2 weeks of incubation, the culture
was transferred, using sterile technique, to fresh PAS medium
containing biphenyl crystals. The culture was incubated for one
week. Following the second incubation period, the enrichment was
plated on R2A agar (Difco Inc., Detroit, Michigan). Once growth
appeared, the plates were sprayed with 2,3_dihydroxybiphenyl
(2,3-dhb) in ether (0.1 percent weight: volume) .
Colonies that turned yellow, indicating cleavage of 2,3-dhb,
were restreaked on R2A medium. Several strains turned yellow and
three were isolated for further characterization. These cultures
were labelled BAG 15, BAG 17, and BAG 19.
Isolates were also characterized by colony hybridization
using the bphC gene probe. This probe codes for the 2,3-dhb
dioxygenase of Pseudomonas pseudoalcaligenes KF707. Bacterial
colonies were transferred to Biotrans" Nylon Membranes (ICN
Biomedical, Costa Mesa, California) and lysed with 0.5 Normal (N)
NaOH for 5 minutes. Filters were allowed to dry and baked for 1
hour at 80°C. Purified probe was labeled with digoxigenin in a
random primed reaction with the Genius DNA Labeling System
(Boeringer Mannheim Biochemicals, Indianapolis, Indiana)
following company protocols. Prehybridization, hybridization and
detection of the digoxigenin probe was according to the Genius
4 6
-------
DNA Labeling System (Boeringer Mannheim Biochemicals,
Indianapolis, Indiana).
Ranid PCB Screening Assay
A Rapid Screening Assay for the determination of bacterial
attack of specific PCB congeners has been developed (Bedard et
al.f 1987). This assay was undertaken to aid in the selection of
cultures for additional bioslurry investigations.
All cultures isolated were evaluated in the screening assay.
In addition, Alcaligenes eutrophus H850 (H850) obtained from GE
was used as the positive control because of its demonstrated
activity against PCB (Bedard et al., 1987). Pseudomonas putida
2440 (2440), a non-PCB degrader obtained from the University of
Tennessee CEB, was used as the negative control for the
experiment.
Five bacterial cultures (i.e., BAG 15, BAG 17, BAG 19, H850,
and 2440) were grown in PAS medium containing biphenyl and 0.005
percent yeast extract. The cultures were grown to an optical
density of 1.0 at 615 nanometer (nm) . Cells were harvested by
centrifugation and washed twice with potassium phosphate buffer
(pH 7.5). Cells were resuspended in potassium phosphate buffer
to an optical density of 1.0 at 615 nm; this solution was
identified as the culture solution.
Each culture solution (5 mL) was aseptically transferred
into 5 sets of fifteen 40-mL glass vials. Each vial was spiked
with 10 microliter (jiL) of a 7-congener mixture. The congener
mixture contained 2, 4, 4 ' -trichlorobiphenyl, 2,3,4-
trichlorobiphenyl, 2,2',5,5'-tetrachlorobiphenyl, 2,3,5,6-
tetrachlorobiphenyl, 2,2',3,3/-tetrachlorobiphenyl, 2,3',4',5-
tetrachlorobiphenyl, and 3, 3', 4, 4 '-tetrachlorobiphenyl; congeners
were obtained from AccuStandard, New Haven, Connecticut. The
final concentration of all congeners in the treatment vial was
approximately 10 mg/L. Biphenyl was also added at a
concentration of approximately 40 mg/L. The following treatments
were prepared.
Treatment Rl - BAG 15, 7-congener substrate, biphenyl
Treatment R2 - BAG 17, 7-congener substrate, biphenyl
Treatment R3 - BAG 19, 7-congener substrate, biphenyl
Treatment R4 - H850, 7-congener substrate, biphenyl
(positive control)
Treatment R5 - 2440, 7-congener substrate, biphenyl
(negative control)
KN/t-M/SrTE.CTKO/SrrEJMT.ItEV
47
-------
Fifteen vials were established for each treatment. Three
vials from each treatment set (15 total vials) were sacrificed at
five time points. The time points were study initiation, 24
hours (hr), 48 hr, 94 hr, and 140 hr. Vials were extracted by
sonication with 2 mL of pentane (Aldrich Chemical Co., Milwaukee,
Wisconsin) for one minute in a Bransonic 220 Sonicator Bath.
Pentane extracts were analyzed by a Hewlett Packard 5890A
Gas Chromatograph (GC) with an automatic sampler, ECD, splitless
injector, and Supelco SPB-1 capillary Column [75 meter by 0.75
millimeter (internal diameter) I.
Nitrogen was used as the carrier and make-up gas. The
carrier gas flow was 2 milliliters per minute (mL/min) at 40°C.
The make-up gas was introduced at 60 mL/min. During sample
analysis, the GC oven initial temperature was 45°C. -This was
held for one minute, raised to 150°C at a rate of 10°C/min and
then to 300°C at a rate of 3°C/min. The 300°C temperature was
held for 5 minutes.
Bioslurrv Evaluation
Three PCB-contaminated soils were evaluated for biological
reduction of PCB congeners. Soils employed were identified as
untreated soil (Sample ID No. GG4202-1018-61) , surfactant/UV-
treated soil (Sample ID No. GG4202-1018-96A), and New England
Superfund Site soil.
The following treatments were prepared:
Treatment Bl - surfactant/UV-treated soil, PAS medium,
BAG 17 culture
Treatment B2 - surfactant/UV-treated soil, PAS medium,
H850 culture
Treatment B3 - surfactant/UV-treated soil, PAS medium,
Hydrochloric acid (Killed control)
Treatment B4 - Untreated soil, PAS medium, BAG 17 culture
Treatment B5 - Untreated soil, PAS medium, H850 culture
Treatment B6 - Untreated soil, PAS medium, Hydrochloric
acid (Killed control)
Treatment B7 - New England soil, PAS medium, BAG 17
culture
Treatment B8 - New England soil, PAS medium, H850 culture
KN/9-M/srre. EnwvsrroRiT. REV
-------
Treatment B9 - New England soil, PAS medium, Hydrochloric
acid (Killed control)
Treatments Bl, B4, and B7 were inoculated with BAG 17.
Treatments B2, B5, and B8 were inoculated with H850. The
cultures were grown in PAS medium as described in the isolation
procedure.
Treatments were prepared using 2 g soil and 8 mL phosphate-
buffered mineral salts medium. All treatments were prepared in
40-mL glass vials with a Teflon™-lined septum screw cap. six
vials per treatment were prepared, with duplicates sacrificed at
3 time points (i.e., study initiation, 2 weeks, and 4 weeks).
Microbial densities of the BAG 17 and H850 culture inoculum
(optical density of 2.0 at 615 nm) were 6.2 x 107 and 9.3 x 108
colony-forming units per mL (CFU/mL), respectively. The cultures
were added to the treatments at an optical density of 1.0 at 615
nm. The estimated cell concentration added to each vial was 3.1
x 107 and 4.7 x 10* CFU/mL for BAG 17 and H850, respectively. The
main carbon source in all treatments was weathered PCB
contamination in the soil.
Treatments B3, B6, and B9 were killed controls established
for each soil evaluated. These treatments were maintained
identically to all biologically-active treatments. Killed
controls were established by the addition of 300 ul of 6 N
hydrochloric acid (HC1) (Mallinkrodt Inc., Paris, Kentucky),
resulting in a pH less then 1. No bacterial cultures were added
to these treatments.
Treatments were shaken at 150 rpm at 25°C in the dark.
Duplicate vials were sacrificed at study initiation (T,), 2 weeks
(T, ) , and 4 weeks (T4). Vials were extracted with 5 mL
dichloromethane (DCM) (Burdick and Jackson, Muskegon, Kentucky)
by sonication (Tekmar 375 watt Ultrasonic Disrupter) and analyzed
for specific PCB congeners and total PCB. DCM was used instead
of pentane due to the increased extraction efficiency achieved
when soil was present. -
After sonication, the solvent layer was separated using an
IECCentra-4B Centrifuge (International Equipment Company). For
improved analysis, the solvent layer was diluted for the
surfactant/UV-treated and untreated soils due to the high PCB
levels present in the soil (approximately 0.4 to 0.8 percent).
The New England soils had PCB levels around 0.03 percent and were
not diluted. Individual PCB congeners were analyzed by GC under
the same conditions previously described.
To assure aerobic conditions in all treatments, oxygen
measurements of vial headspace were made at Day 2, Day 4, Day 7,
4 9
-------
and Day 11. Oxygen measurements were made using a modified
galvanic cell.
Enhanced Bioslurrv Evaluation
An additional study was initiated to look at the effects of
specific inducers and growth substrate on the stimulation of PCB
degradation. It has been shown in previous studies that the
addition of biphenyl, 4-bromobiphenyl (4-BB), 4-chlorobiphenyl,
2-chlorobiphenyl, or other monochlorobiphenyls have induced and
enhanced aerobic PCB biodegradation (Bedard et al., 1987;
Furukawa, et al., 1990; Layton et al., unpublished; Pettigrew et
al., 1990; Phee et al. , 1989).
The objective of this investigation was to determine the
effect of biphenyl and 4-BB (Fluka Ag, Buchf FG) addition on PCB
biodegradation. Two PCB-contaminated soils were analyzed in this
experiment (i.e., New England Superfund Site soil and the
untreated soil).
Inducers (i.e., 4-BB and biphenyl) were dissolved in DCM and
added to the treatment vials. The DCM was allowed to evaporate
before introduction of soil to the treatment vials. Treatments
were established using 2 grams of soil and 8 mL phosphate-
buffered mineral salts medium.
Killed controls were established for each soil evaluated by
the addition of 300 pi, of 6 N HC1. Bacterial culture was also
added to the killed controls to account for any PCB adsorption by
bacterial cell walls.
Based on positive activity against PCB, BAG 17 was the only
culture employed in this investigation. BAG 17 culture inoculum
was added to the treatments at an optical density of 0.9 (615
nm) . BAG 17 was grown following the procedure previously stated
in Section 3.2. The inoculum added to each vial was 9.3 x 10*
CFU/mL, dry weight of 7 milligram (mg).
Treatments were established in triplicate using 40-mL glass
vials. The treatments were:
Treatment El - Untreated soil (Unamended)
Treatment E2 - Untreated soil, BAG 17, and 1,000 mg/L 4-
BB
Treatment E3 - Untreated soil, BAG 17, and 1,000 mg/L
biphenyl
Treatment E4 - Untreated soil, BAG 17, and hydrochloric
acid (Killed)
KN/»-»4«rrE.Erf«wrrE3RfT.R£v 5 0
-------
Treatment E5 - New England soil (Unamended)
Treatment E6 - New England soil, BAG 17, and 1,000 mg/L
4-BB
Treatment E7 - New England soil, BAG 17, and 1,000 mg/L
biphenyl
Treatment E8 - New England soil, BAG 17, and hydrochloric
acid (Killed).
Six vials were established for Treatments El and E5. Three
vials per treatment were sacrificed for initial analyses (T0) .
The remaining 3 vials per treatment were analyzed at TfmU. The
initial analysis of Treatment El vials produced To data for
untreated soil treatments. The initial analysis of Treatment E5
vials produced To data for New England Superfund Site soil
treatments. Treatment vials were incubated at 25°C on a shaker
table at 150 rpm in the dark. All remaining vials were
sacrificed after one week.
Deviation in the extraction procedure described previously
involved the addition of 300 /iL of 6 N HC1 to every treatment
before extraction. This accounted for any differences in the
extraction efficiency due to acid addition.
W-Photolysis
The surfactant/UV-treated soils were prepared in the
photolytic study described previously. In general, the soils
were ground to 200 mesh and treated with the surfactant Hyonic
NP-90® (Henkel Co., Ambler, Pennsylvania) to a concentration of
2.1 percent (wt. NP-90 per wet wt. soil). The experimental set-
up used a 450-watt Hanovia lamp with a parabolic reflector at a
distance of 4 inches. The treated soil temperature did not rise
above 52°C; overheating was prevented by cooling the lamp well.
The soil was periodically raked and moistened throughout the
process. The UV study demonstrated a decrease in the higher
chlorinated PCB with a subsequent increase in the dichlorinated-
PCB. See Section 3 for results.
Data Handling
Soils were evaluated initially and found to resemble an
Aroclor 1248 standard profile. Therefore, soil concentration and
percent degradation of PCB were calculated based on Aroclor 1248
equivalent. Equivalent 1248 is defined as the amount of Aroclor
1248 that it would take to produce a peak of the same size
observed in the soil sample. Total PCB was determined for each
sample by taking the average of the PCB congener 1248 equivalent.
Equivalent concentrations were converted to equivalent mass by
multiplying the equivalent concentration by the mass of the soil
51
-------
used during analysis. Equivalent mass removed was determined by
comparison of final data with To results.
Peak positions for 32 PCB congeners were established (Table
13) based on the pattern of Aroclor 1248 in commercially-
available standards (Ultra Scientific Inc., Kingston, Rhode
Island) and by published congener profile of Aroclor 1248 (Bedard
et al., 1987). Percent degradation was calculated for each
congener by direct comparison of its 1248 equivalent to that
found in the killed controls. Percent degradation was normalized
by subtracting the average percent degradation of the internal
standard peaks. Internal standards were identified as Peaks 32
and -33. These peaks were chosen as internal standards due to
their recalcitrant nature and used to adjust for abiotic loss of
contaminant. Bedard et al., 1987 have shown that A. eutrophus
H850 cannot degrade 2,4,5,3' '4, pentachlorobiphenyl (peak 31),
2,3,4,3',4'-pentachlorobiphenyl/2,3,4,2',3'6'-hexachlorobiphenyl
(peak. 32)' and 2,3,4,2' , 4' ,5'-hexachlorobiphenyl/2, 3 , 5, 6, 3' ,4'-
hexachlorobiphenyl (peak 33) . These peaks are traditionally used
as internal standards to determine extraction efficiency and to
determine biodgradation of other congeners. If the ratio of peak
32 to peak 33 changes then degradation of one of these congeners
has occurred and they cannot be used as internal standards.
Degradation of these peaks did not occur. Degradation of less
than 15 percent was not considered significant based on
analytical and instrument variation. Total percent loss was
determined by comparison of the total average equivalent 1248 to
that of the respective killed control.
Congener groups were also established based on the DCMA
method. Retention time windows were determined for the di-PCB,
tri-PCB, tetra-PCB, penta-PCB, hexa-PCB, and hepta-PCB. Percent
loss of each group was determined by comparison of biologically-
active treatments with the killed controls. Reduction was
normalized by subtracting the average of internal standard Peaks
31 and 32 loss.
Hewlett Packard 5895A GC Chem Station Software system was
used to analyze the data. A complete data package for all
analyses conducted during Phases 3 and 4 is included in Appendix
F.
RESULTS AND DISCUSSION
Isolation of PCB-Dearaders
Colony morphology of isolates BAG 15, BAG 17, and BAG '19
indicated small, off-white colonies with smooth edges. All
isolates grew on biphenyl as the sole carbon and energy source.
BAG 15 and 17 turned yellow after exposure to the compound,
indicating biodegradation of 2,3-dhb, and hybridized with the
52
-------
TABLE 13. CONGENER IDENTIFICATION
Peak No. Congener Identification
1 2,5,2'- trichlorobiphenyl
2 2/4,2'-trichlorobiphenyl/4,4'- dichlorobiphenyl
3 2,3,2'-trichlorobiphenyl/2,6,4'-trichlorobiphenyl
415 2,5,4-trichlorobiphenyl/2,4,4'-trichlorobiphenyl
6 2,3,4-trichlorobiphenyl/2,5,2',6'-tetrachlorobiphenyl
7 2,3,4'-trichlorobiphenyl/2,4,2',6'-tetrachlorobiphenyl
8 2,3,6,2'-tetrachlorobiphenyl
9 2,3,2',6'-tetrachlorobiphenyl
10 2,5,2',5'-tetrachlorobiphenyl
11 2,4,2',5'-tetrachlorobiphenyl
12/13 2,4,3',4'-tetrachlorobiphenyl/2,4,5,2'-tetrachlorobiphenyl
14 2,3,2',5'-tetrachlorobiphenyl
15 3,4,4'-trichlorobiphenyl/2,3,2',4'-tetrachlorobiphenyl
16 2,3,4,2'-tetrachlorobiphenyl/2,3,6,4'-tetrachlorobiphenyl/2,6,3',4'-
tetrachlorobiphenyl
17 2,3,2',3'-tetrachlorobiphenyl
18 2,4,5,4'-tetrachlorobiphenyl
19 2,5,3',4'-tetrachlorobiphenyl
20 2,4,2',4'-tetrachlorobiphenyl/2,3,6,2',5'-pentachlorobiphenyl
21 2,3,6,2',4'-pentachlorobiphenyl
22 2,3,3',4'-tetrachlorobiphenyl/2,3,4,4'-tetrachlorobiphenyl
23 2,3,6,2',3'pentachlorobiphenyl/2,3,5,2',5'-pentachlorobiphenyl
24 2,3,5,2',4'pentachlorobiphenyl/2,4,5,2',5'-pentachlorobiphenyl
25 2,4,5,2',4'-pentachlorobiphenyl
26 2,4,5,2',3'-pentachlorobiphenyl/2,3,5,6,2',6'-hexachlorobiphenyl
27 2,3,4,2',5'-pentachlorobiphenyl
28 2,3,4,2',4'-pentachlorobiphenyl
29 2,3,6,3',4'-pentachlorobiphenyl/3,4,3',4'-tetrachlorobiphenyl
30 2,3,4,2',3'-pentachlorobiphenyl
31 2(3,6,2',4',5'-hexachlorobiphenyl/2,4/5,3',4'-pentachlorobiphenyl
32 2,3,4/3',4'-pentachlorobiphenyl/2/3/4,2'/3'/6'-hexachlorobiphenyl
33 2,3,4,2', 4', 5'-hexachlorobiphenyl/2,3,5,6,3',4'-hexachlorobiphenyl
53
-------
bphC gene of P. pseudoalcaligenes KF707 (Furukawa et al., 1987).
Fatty acid profiles identified the BAG 17 strain as P. cepacia
subgroup B with a similarity index of 0.71 (Appendix G).
BAG 19 did not turn yellow after exposure to 2,3 dhb and was
not tested further.
Rapid PCB Screening Assav
Congener percent reduction was determined for each culture
evaluated (BAG 15, BAG 17, BAG 19, H850 and 2440) using the Rapid
PCB Screening Assay. Data generated during the assay is included
in Table 14. The test results were inconclusive, due to the
substantial congener reduction exhibited by the negative control
(2440). Therefore, growth characterization and hybridization
with the bphC gene probe was used as criteria for selection of
cultures for additional testing. Cultures selected for testing
were BAG 17 and H850.
TABLE 14,
RAPID
Initial
Congener Concentration
(ng///l)
2,4,4'-trichlorobiphenyl
3,4,2-trichlorobiphenyl
2,5,2', 5'-tetrachlorobiphenyl
2,3,5,6- tetrachlorobiphenyl
2,3,2', 3' -tetrachlorobiphenyl
2,5,3,4-tetrachlorobiphenyl
3,4,3',4'-tetrachlorobiphenyl
1.20
1.20
1.20
1.11
1.06
1.20
1.42
PCB SCREENING
ASSAY
Percent Loss per Culture Tested
BAC15
93
96
98
38
99
91
0.7
BAC17
92
95
98
52
97
92
13
BAC19
93
96
98
56
98
91
30
H850
67
93
98
58
98
89
7
2440
75
97
86
67
99
83
56
Bioslurrv Evaluation
Three PCB-contaminated soils were evaluated for biological
reduction of PCB congeners. Soils employed were identified as
untreated soil (Sample ID No. GG4202-1018-61) , surf act ant /UV-
treated soil (Sample ID No. GG4202-1018-96A), and New England
Superfund Site soil.
Untreated, surfactant/UV-treated, and New England Superfund
Site soils used in the bioslurry evaluation were analyzed for
indigenous microbial populations. The microbial density of the
untreated, surfactant/UV-treated, and New England soils were 6.9
x 10s, 1.1 x 105, and less than 3.0 x 10' CFU/g, respectively. It
54
-------
should be noted that, surfactant/UV-treatment reduced the
microbial populations rather than sterilized the soil.
Dissolved organic carbon (DOC) was analyzed in the three
soils evaluated to estimate the distribution of surfactant. The
objective of this testing was to determine if surfactant
distribution may have had an effect on the aerobic respiration
and PCB removal in the varying treatments. The results indicated
that approximately 10 times more DOC was found in the
surfactant/UV-treated soil than in the untreated soil; no.
appreciable DOC was measured in the New England soils.
Surfactant/UV-treated, untreated, and New England Superfund Site
soils had DOC concentrations of 2,500, 220, and 9 mg/kg,
respectively.
Soil pH was also analyzed. Surfactant/UV-treated, untreated
and New England soils demonstrated a slurry pH of 5.5, 6.3, and
7.0, respectively. All pH values were within the range
acceptable for biological activity. Elevated DOC concentration
and low pH in the surfactant/UV-treated soil was a result of
surfactant use during the treatment process.
Initial analysis of the surfactant/UV-treated soil indicated
total PCB concentrations of 4,000 mg/kg. This concentration was
approximately 50 percent less than the untreated soil total PCB
concentration of 8,400 mg/kg. The New England soil had 350 mg/kg
total PCB.
Out of the 32 specific congeners monitored, there was
minimal specific congener loss in the surfactant/UV-treated soil
and untreated soil during this phase of experimentation as shown
in Table 15. Percent removal was determined by comparison of
biologically-active treatments to their respective control
treatments. As expected, BAG 17 preferentially attacked lower
chlorinated compounds consisting mostly of trichlorobiphenyls;
some reduction in tetrachlorobiphenyls and reduction in one
hexachlorobiphenyl congener was also observed (Figure 1) . in the
surfactant/UV-treated soil Treatment Bl, BAG 17 degraded 25, 22,
21, 22, and 20 percent of Peaks 1, 2, 6, 17, and 31,
respectively. In the untreated soil Treatment B4, BAG 17 removed
a greater quantity of the lower chlorinated species,
demonstrating 58, 77, 27, and 46 percent reduction in Peaks 1, 2,
6, and 17, respectively, This isolate demonstrated increased
activity in the New England soil Treatment B7; removal
efficiencies ranging from 17 to 73 percent were measured for
Peaks 1, 2, 4 through 15, 17, and 23 (Table 15) .
H850 demonstrated reduced performance as compared to the BAG
17 culture. Treatment B2 established with surfactant/W-treated
soil demonstrated a 24, 18, 18, 18, 15, 15, and 25 percent
reduction in Peaks 1, 2, 4 through 7, 18, and 31, respectively.
No significant removal was noted in Treatment B5 using untreated
55
-------
soil. Optimum activity of H850 was illustrated in Treatment B8
which evaluated New England soil contaminant reduction. Peaks 1
2, 6, 8 through 15, and 23 were preferentially attacked in this
treatment, resulting in percent removals ranging from 18 to 40
percent (Table 15) .
Both cultures demonstrated unusual degradation of Peak 31 in
surfactant/UV-treated soils. in addition it should be noted
that, BAG 17 exhibited 20, 46, and 50 percent degradation of Peak
17 in untreated, surfactant/UV-treated, and New England soils,
respectively. Reduction of Peak 17, 2,3,2',3'-
tetrachlorobiphenyl, was not demonstrated by H850 (Table 15) .
Based on the DCMA method of classification of PCB congeners,
BAG 17 treatment of New England soil (Treatment B7) illustrated
optimal percent reduction as compared to all other treatments.
Treatment B7 demonstrated di-, tri-, and tetra-chlorobiphenyl
reductions of 70, 20, and 30 percent, respectively (Table 16) .
Treatment B4, which evaluated contaminant reduction in untreated
soils using BAG 17, demonstrated appreciable loss of
dichlorinated species at 67 percent. Reduced performance was
measured in Treatment Bl using BAG 17 and surfactant/W-treated
soils. (Note: Even though specific analysis of congeners
indicated no dichlorobiphenyls present, the DCMA classification
contains overlap between congener groups. Theref9re, some of the
trichlorobiphenyls are grouped with the dichlorobiphenyls, some
of the tetrachlorobiphenyls are grouped with the
trichlorobiphenyls, and so forth.)
KN/9-W/SirE.mHU/SITBRIT RPV
-------
TABLE 15. PERCENT SPECIFIC CONGENER PCB DEGRADATION
BIOSLURRY EVALUATION
Peak
No.
1
2
3
4/5
6
7
8
9
10
11
12/13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
UV-Treated
Soil
Treatment B1 Treatment B2
(BAG 17) (H850)
25'
22
0
0
21
0
0
0
0
0
0
0
0
0
22
0
0
0
0
0
0
0
0
0
0
0
0
0
20
0
0
24
18
0
18
18
15
0
0
0
0
0
0
0
0
0
15
0
0
0
0
0
0
0
0
0
0
0
0
25
0
0
Untreated Soil
Treatment B4
(BAC17)
58
77
0
0
27
0
0
0
0
0
0
0
0
0
46
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Treatment B5
(H850)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
New England
Soil
Treatment B7 Treatment B8
(BAG 17) (H850)
67
73
0
17
37
26
31
34
15
34
33
47
37
0
50
0
0
0
0
0
24
0
0
0
0
0
0
0
0
0
0
39
40
0
0
22
0
23
18
23
22
21
20
18
0
0
0
0
0
0
0
21
0
0
0
0
0
0
0
0
0
0
• Percent degradation less than 15 percent is not considered significant and is reported as zero.
57
-------
TABLE 16. PERCENT LOSS OF CONGENER GROUPS - DCMA METHOD
BIOSLURRY EVALUATION
Congener Group
UV-Treated Soil
Untreated Soil
New England Soil
Treatment
B1
(BAG 17)
Dichlorobiphenyl
Trichlorobiphenyl
Tetrachlorobiphenyl
Pentachlorobiphenyl
Hexachlorobiphenyl
Heptachlorobiphenyl
24
0'
0
0
0
0
Treatment
B2
(H850)
21
16
0
0
0
0
Treatment
B4
(BAG 17)
67
0
0
0
0
0
Treatment
B5
(H850)
0
0
0
0
0
0
Treatment
B7
(BAG 17)
70
20
30
0
0
0
Treatment
B8
(H850)
40
0
0
0
0
0
' Percent degradation less than 15 percent is not considered significant and is reported as zero.
Similar to the results presented in Table 15, PCB removal
based on the DCMA method illustrated reduced performance of H850
as compared to BAG 17 (Table 16). H850 preferentially attacked
dichlorobiphenyls in surfactant/UV-treated and New England soils,
i.e., Treatments B2 and B8. Biological removal of PCB congeners
in the untreated soils by H850 was not evident. Organization of
congener reduction using the DCMA method demonstrated results
similar to those obtained through congener specific analyses.
Respiration is a measurement of oxygen consumption by the
bacteria, indicating microbial activity. Oxygen consumption was
measured by loss of oxygen in the headspace over time. Oxygen
consumption was greater in the treatments containing
biologically-active cultures compared to the killed controls,
where respiration was insignificant. Treatments B2, B5, and B8
containing H850 demonstrated respiration rates of 2.5, 1.5, and
1.0 milligram oxygen/kilogram-hour (mg O2/kg-hr) at 48, 96, and
168 hours, respectively. Oxygen consumption remained at 1.0 mg
O2/kg-hr through 264 hours. In Treatments Bl, B4, and B7
containing BAG 17 culture, oxygen consumption was 2.4, 1.2, 1.0,
and 0.7 mg O2/kg-hr at 48, 94, 168, and 264 hours, respectively.
Oxygen consumption data indicated that the majority of oxygen
demand was satisfied during the first 2 days of incubation (Table
17). Respiration rates were similar in all biologically-active
treatments, although PCB removal rates varied across treatments.
Initial oxygen concentrations in the headspace were considered to
be 300 mg/L (atmospheric concentration).
lCN/».*l/S!TE.En1
-------
TABLE 27, OXYGEN CONSUMPTION IN TREATMENTS
BIOSLURRY EVALUATION
Oxygen Consumed fmg 02/kg-hr}
Time
H850 BAG 17
48 hours 2.5 2.4
96 hours 1.5 1.2
168 hours 1 .0 1.0
264 hours 1.0 0.7
Enhanced Bioslnrrv Evaluation
The objective of this investigation was to determine the
effect of biphenyl and 4-BB (Fluka Ag, Buchf FG) addition on PCB
biodegradation. Two PCB-contaminated soils were analyzed in this
experiment (i.e., New England Superfund Site soil and untreated
soil).
BAG 17 culture was added to all treatments evaluated during
the Phase 4 investigation. Unamended controls for New England
and untreated soil did not receive bacterial culture; congener
removal in these treatments were adjusted for abiotic losses
evident in the killed controls, i.e., Treatments E4 and E8.
Bacterial culture was added to the killed controls to determine
its effect on PCB adsorption.
Treatments established with New England soil were identified
as Treatments E5, E6, and E7. Treatment E5 was the unamended
control for the experimental set. Treatment E6, which received
4-BB, demonstrated substantial removal of Peaks 2, 4 through 7,
10 through 15, 17, 43, and 22 (Table 18). In comparison to
Treatment B7 (Table 15) of the bioslurry investigation,
approximately a two-fold increase in congener removal was
demonstrated in the majority of higher chlorinated congeners,
i.e., Peaks 10 through 15, 17, 19, and 23. Biphenyl addition to
Treatment E7 resulted in substantial increases in congener
removal efficiency. Removal efficiencies ranging from 28 to 100
percent were noted in Peaks 1 through 17, 19, and 23. Once again
this was a notable increase in congener removal as compared to
Treatment B7 which did not receive a growth substrate/metabolic
inducer. The unamended control (Treatment E5) demonstrated a
moderate reduction (i.e., 22 percent) of Peak 17 (Table 18).
Increased congener removal in comparison to Treatment 87 is
illustrated in Table 19.
KN/9-»4/SITE.EIT<»/SrrE)IUT.IlEV 59
-------
TABLE 18. PERCENT SPECIFIC CONGENER PCB DEGRADATION
ENHANCED BIOSLURRY EVALUATION
Peak
No.
1
2
3
415
6
7
8
9
10
11
12/13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Untreated Soil (Percent
Treatment E1
{Unamended)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Treatment E2
(4-BB)
21
38
0
0
0
0
0
0
0
0
0
0
0
0
22
0
0
0
0
0
21
0
0
0
0
0
0
0
0
0
0
Removal)
Treatment E3
(Biphenyl)
28
45
0
0
15
0
0
0
0
0
0
0
0
0
31
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
New England
Soirb(Percent
Treatment E5 Treatment E6
(Unamended} (4-BB)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
22
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
66
0
30
34
75
0
0
38
60
53
94
66
0
100
0
43
0
0
0
43
0
0
0
15
0
0
0
0
0
0
Removal)
Treatment E7
(Biphenyl)
59
100
28
46
87
85
73
38
47
65
71
98
64
27
9.1
0
39
0
0
0
38
0
0
0
0
0
0
15
0
0
0
' Percent degradation less than 15 percent Is not considered significant and Is not reported.
6 All treatments evaluated used SAC 17 culture Inoculum.
60
-------
TABLE 19. COMPARISON OF PERCENT SPECIFIC CONGENER PCB
DEGRADATION WITH AND WITHOUT BIPHENYL AUGMENTATION
New England Soil (Percent Removal)
Peak No.
1
2
3
4/5
6
7
8
9
10
11
12/13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Treatment E5
(Unamendedl
0
0
0
0
0
0
0
0
0
0
0
0
0
0
22
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Treatment B7
(BAG 17)
67
73
0
17
37
26
31
34
15
34
33
47
37
0
50
0
0
0
0
0
24
0
0
0
0
0
0
0
0
0
0
Treatment E7
(Biphenyl)
59
100
28
46
87
85
73
38
47
65
71
98
64
27
91
0
39
0
0
0
38
0
0
0
0
0
0
15
0
0
0
The significant benefit of inducer and growth substrate
addition, specifically biphenyl, in increasing PCB removal was
61
-------
not demonstrated in the untreated soil treatments. Treatment E2
received 4-BB as a metabolic inducer. This treatment
demonstrated reduced performance in comparison with Treatment B4
(untreated soil/BAC 17). Likewise, the addition of biphenyl to
Treatment E3 did not significantly improve PCB congener removal
over that evident in Treatment B4. The unamended (Treatment El)
demonstrated no removal of any PCB congener.
Congener removal was also determined based on the DCMA
method. The New England soil Treatment E7 demonstrated a.82, 54,
63, and 16 percent reduction in the di-, tri-, tetra-, and penta-
PCB, respectively (Table 20) . Treatment E6 showed a 28, 29, and
21 percent reduction in the di-, tetra-, and penta-PCB,
respectively. Unamended Treatment E5 showed
no significant loss of specific congeners.
TABLE 20. PERCENT LOSS OF CONGENER GROUPS - DCMA METHOD
ENHANCED BIOSLURRY EVALUATION
Untreated Soil
Congener group
Dichlorobiphenyl
Triehtorobiphenyl
Tetrachiorobiphenyl
Pentachlorobiphenyl
Hexachlorobiphenyi
Heptachlorobiphenyl
Treatment
E1
(Unamended!
0
0
0
0
0
0
Treatment
E2
(4-BB)
29
0
0
0
0
0
Treatment
E3
(Biphenyl}
37
0
0
0
0
0
New England Soil
Treatment
E5
{Unamended)
0
0
0
0
0
0
Treatment
E6
(4-BB)
28
0
29
21
0
0
Treatment
E7
{Biphenyl)
82
54
63
16
0
. o
DCMA results illustrated no significant congener loss in
untreated soil Treatment El. Treatments E2 and E3 demonstrated
29 and 37 percent reduction of di-PCB, respectively (Table 19).
There was no significant: loss of the higher chlorinated PCB in
Treatments El, E2, and E3.
Oxygen uptake was measured in all treatments at 48 and 96
hours (Table 21) . At 48 hours, BAG 17 demonstrated less than
0.3, 1.4, and 0.6 mg Oj/Jcg-hr in Treatments El, E2, and E3,
respectively. After 96 hours, an increase in oxygen consumption
of 0.5, 2.1, and 1.8 mg 02/kg-hr was noted in Treatments El; E2,
and E3, respectively. At 48 hours, respiration in Treatments E6
and E7 was measured at 2.3 mg and 1.8 mg 02/kg-hr, respectively.
The New England soil treatments demonstrated no appreciable
oxygen consumption at 96 hours.
62
-------
TABLE 21. OXYGEN CONSUMPTION
ENHANCED BIOSLURRY EVALUATION
Oxygen Consumed
Treatment Identification (mg 02/kg-hr)
48 hours 96 hours
Treatment El (Untreated/Unamended)
Treatment E2 (Untreated/4-BB)
Treatment E3 (Untreated/Biphenyll
Treatment E4 (Untreated/Killed)
Treatment E5 (New England/Unamended)
Treatment E6 (New England/4-BB)
Treatment E7 (New England/Biphenyl)
Treatment E8 (New England/Killed)
co. 3
1.4
0.6
co. 3
co. 3
2.3
1.8
<0.3
0.5
2.;
1.8
co. 2
co. 2
co. 2
0.3
<0.2
CONCLUSIONS AND RECOMMENDATIONS
Several obstacles exist to biodegradation of complex
mixtures of PCBs in soil. First, bioavailability is a
significant problem. If bacteria cannot come in contact with the
substrate, the substrate cannot be metabolized. PCBs are very
hydrophobic and sorb readily to surfaces; therefore, in any
biological treatment scenario, desorption of the PCBs is a
primary concern.
Second, the highly chlorinated congeners are resistant,
generally, to biological degradation. Two enzymes are thought to
mediate the initial biotransformations of the lower chlorinated
congeners, biphenyl 2,3-dioxygenase and biphenyl 3,4-dioxygenase.
Highly chlorinated congeners may cause steric hindrance of these
two enzymes inhibiting the initial hydroxylation step
(Abramowicz, 1990; Parsons et a/., 7988J. Although certain
bacteria have demonstrated an ability to cometabolize the highly
chlorinated congeners, extensive aerobic degradation has not been
observed in the environment (Bedard et a/., 1987a; Bopp, 1986)
although this may be due to the lack of bioavailability,
cosubstrates, or thermodynamically unfavorable degradative
pathways.
Third, the inducers of the biphenyl operon must be present
to maintain PCB-degrading activity. Normally, biphenyl and the
lower chlorinated congeners will be degraded first. Biphenyl, 2-
chlorobiphenyl, 4-chlorobiphenyl and 4-bromobiphenyl have been
shown to induce the biphenyl operon (Bedard, 1993; Furukawa, et
a/., 1990; Pettigrew et a/., 1990; Rhee et a/., 1989; Bedard et
a/., 1987; Layton et o/., unpublished data). This group is also
63
-------
responsible for induction of the biphenyl operon. When the
inducers disappear from the environment, no further PCB
degradation would be expected. Addition to a system of non-
toxic, degradable inducers of the biphenyl operon may increase
the extent of degradation of.the highly chlorinated congeners.
Continuous feeding of biphenyl to a PCB-contaminated soil in a
batch slurry bioreactor has given preliminary results of 1,000
mg/kg biphenyl to degrade approximately 10 mg/kg PCB (Layton,
personal communication).
Phase 4 of this investigation studied the effect of the
combined surfactant/UV treatment on biological degradation of
weathered PCBs. In the bioslurry evaluation test strain H850
degraded 21 and 16 percent of the di- and tri-chlorobiphenyls
(Table 16), respectively, in the treated soil while no
significant degradation was observed in the untreated soil. The
opposite situation was observed with strain BAG 17. Sixty-seven
percent degradation of the dichlorobiphenyl was degraded in the
untreated soil versus 24 percent in the treated soil. The
treated soil contained 2,480 mg/kg DOC versus 250 mg/kg in the
untreated soil. This indicates that high amounts of surfactant
were carried through the treatment process and may be inhibitory
to bacterial activity or promote non-PCB degrading activity.
Likewise, the treated soil had a pH of 5.5 which probably was a
result of the surfactant. An additional soil washing step may be
necessary to remove/recycle surfactant from the soil and
neutralize the pH before biological treatment.
Strain BAG 17 removed approximately 30 percent of the
tetrachlorobiphenyls (as defined by the DCMA method) from the New
England soil (Table 16). BAG 17 was originally isolated from
this soil and was expected to perform well.
Augmentation of biphenyl and 4-BB to the New England soil
stimulated biodegradation of the di-, tri-, tetra-, and
pentachlorobiphenyls (Tables 18 and 19) . Biphenyl was the better
growth^substrate and inducer of PCB degradation than the 4-BB
under these conditions. The untreated soil showed stimulation of
only the lower chlorinated congeners. Why degradation was not as
extensive as in the surfactant/UV-treated soil is not understood.
The New England soil with a higher bacterial activity against
PCBs, was composed mainly of sands while the GG4202 soils, with a
low bacterial activity against PCBs, had a strong clay component.
Similar observations were made using a clayey PCB-contaminated
soil from a transformer substation. Correlation of PCB-degrading
activity with soil type, PCB concentration and composition,
biphenyl/PCB concentrations, and bacterial populations needs to
be explored.
In addition, if 1,000 mg/kg biphenyl (or, possibly, any
other inducer) is required to reduce the total PCB concentration
10 mg/kg as suggested earlier (Layton, personal communication),
KN/»WSITE.En
-------
the loss of PCBs in the untreated soil would be masked by the
analytical variability.
Specific conclusions from this study are:
PCB removal in the surfactant/UV-treated soil was
slightly higher when augmented with strain H850.
PCB removal in the untreated and the New England soil was
enhanced by augmentation with strain BAG 17.
Biphenyl was more effective at stimulating PCB
degradation than 4-BB in the untreated and the New
England soil.
Surfactant treatment may have been inhibitory to
microbial activity as evidenced by the high DOC and low
of the treated soil.
lCN/»-94/SrrE.En'(B/SrrE3RJT.REV 6 5
-------
SECTION 6
REFERENCES
Abramowicz, D. A., 1990, "Aerobic and Anaerobic Biodegradation of
PCBs: A Review, " Critical Reviews in Biotechnology, 10( 3) :241-
251.
Adrians, P., H-P. Kohler, D. Kohler-Staub, and D. D. Focht, 1989,
"Bacterial Dehalogenation of Chlorobenzoates and Coculture
Biodegradatiori of 4,4'-dichlorobiphenyl," Applied and
Environmental Microbiology, 55:887-892.
Ahmed, D. and D. D. Focht, 1973, "Degradation of Polychlorinated
Biphenyls by Two Species of Achromobacter, " Canadian Journal of
Microbiology, 19 :47-52.
Barbeni, M., C. Minero, E. Pelizzetti, E. Borgarello, and N.
Serpone, 1987, "Chemical Degradation of Chlorophenols with
Fenton's Reagent," Chemosphere, 16:2225-2237.
Barton, M. R. and R. L. Crawford, 1988, "Novel Biotransformations
of 4-chlorobiphenyl by a Pseudomonas sp," Applied and
Environmental Microbiology, 54 :594-595.
Bedard, D. L., May 1993, "Accelerating the Microbial
Dechlorination of Polychlorinated Biphenyls in Anaerobic Pond
Sediments, " Seminar Presented at the 93rd General Meeting of the
American Society for Microbiology, Atlanta, Georgia.
Bedard, D., R. Wagner, M. J. Brennan, M. L. Haberl, and J. F
Brown, Jr., 1987a, "Extensive Degradation of Aroclors and
Environmentally Transformed Polychlorinated Biphenyls by
Alcaligenes eutrophus H850, " Applied and Environmental
Microbiology, 53:1094-1102.
Bedard, D., M. Haberl, R. J. May, and M. J. Brennan, 1987b,
"Evidence for Novel Mechanisms of Polychlorinated Biphenyl
Metabolism in Alcaligenes', eutrophus H850," Applied and
Environmental Microbiology, 53:1103-1112.
Bedard, D. L. and M. L. Haberl, 1990, "Influence of Chlorine
Substitution Pattern on the Degradation of Polychlorinated
Biphenyls by Eight Bacterial Strains," Microbial Ecology, 20-87-
102.
Bopp, L. H., 1986, "Degradation of highly chlorinated PCBs by
Pseudomonas strain LB400, " Journal of Industrial Microbiology
1:23-29.
66
-------
Brunner, W., F. H. Sutherland, and D. D. Focht, 1985, "Enhanced
biodegradation of polychlorinated biphenyls in soils by analog
enrichment and bacterial inoculation, " Journal of Environmental
Quality, 14:324-328.
Exner, J. H., E. S. Alperin, A. Groen, C. E. Morren, 1984, "In-
Place Detoxication of Dioxin-Contaminated Soil," Hazardous Waste,
1:217-223.
Furukawa, K. and F. Matsumura, 1976, "Microbial metabolism of
polychlorinated biphenyls. Studies on the relative degradability
of polychlorinated biphenyl components by Alcaligenes sp,"
Journal of Agriculture and Food Chemistry, 24:251-256.
Furukawa, K, F. Matsumura, and K. Tonomura, 1978, "Alcaligenes
and Acinetobacter Strains Capable of Degrading Polychlorinated
Biphenyls," Agriculture Biological Chemistry, 42:543-548.
Furukawa, K., K. Tonomura, and A. Kamibayashi, 1979, "Effect of
Chlorine Substitution on the Bacterial Metabolism of Various
Polychlorinated Biphenyls, " Applied and Environmental
Microbiology, 38:301-310.
Harkness M. R., J. B. McDermott, D. A. Abramowicz, J. J. Salvo,
W. P. Flanagan, M. L. Stephens, F. J. Mondello, R. J. May, J. H.
Lobos, K. M. Carroll, M. J. Brennan, A. A. Bracco, K. M. Fish, G.
L. Warner, P. R. Wilson, D. K. Dietrich, D. T. Lin, C. B. Morgan,
W. L. Gately, 1993, "In Situ Stimulation of Aerobic PCB
Biodegradation in Hudson River Sediments,@' Science, 259:503-507.
Layton, A. C., C. A. Lajoie, J. P. Easter, R. Jernigan, M. Beck,
and G. S. Sayler, "Molecular Diagnostics for Polychlorinated
Biphenyl Degradation in Contaminated Soils," Unpublished.
Kitao, T., Y. Kiso, and R. Yahashi, 1982, "Studies on the
Mechanism of Decolorization with Fenton's Reagent," Mizii Shori
Gijutsu, 23:1019-1026.
Masse, R., F. Messier, L. Peloquin, C. Ayotte, and M. Sylvestre,
1984, "Microbial Biodegradation of 4-Chlorobiphenyl, a model
compounds of chlorinated biphenyls, " Applied and Environmental
Microbiology, 47:947-951.
Murphy, P., W. J. Murphy, M. Boegli, K. Price, and C. D. Moody,
1989, "A Fenton-like Reaction to Neutralize Formaldehyde Waste
Solutions," Environmental Science Technology, 23:166-169.
Parsons, J. R., D. T. Sijm, A. Van Laar, and 0. Hutzinger, 1988,
"Biodegradation of Chlorinated Biphenyls and Benzole Acids by a
Pseudomonas strain, " Applied Microbiology and Biotechnology,
29:81-84.
67
-------
Pettigrew, C. A., A. Breen, C. Corcoran, and G. S. Sayler, 1990,
"Chlorinated Biphenyl Mineralization by Individual Populations
and Consortia of Freshwater Bacteria, " Applied and Environmental
Microbiology, 56:2036-2045.
Rhee, G-Y., B. Bush, M. P. Brown, M. Kane and L. Shane, 1989,
"Anaerobic Biodegradation of Polychlorinated Biphenyls in Hudson
River Sediments and Dredged Sediments in Clay Encapsulation,"
Water Research, 23:957-964.
Shiaris, M. P. and G. S. Sayler, 1982, "Biotransformation of PCB
by Natural Assemblages of Freshwater Microorganisms,"
Environmental Science and Technology, 16:367-369.
Sylvestre, M.R. Masse, C. Ayotte, F. Messier, and J. Fautex,
1985, "Total biodegradation of 4-Chlorobiphenyl (4-CB) by a Two
Membered Bacterial Culture," Applied Microbiology and
Biotechnology, 21:192-195.
-------
APPENDIX A
TCDD ANALYTICAL REPORTS
-------
TCDD UV PHOTOLYSIS SAMPLE ANALYSIS CROSS-REFERENCE
FOR KEY SAMPLES
Sample
TCDD Starting Soil
Test 1 Final Soil (48 Hours)
Test 2 Final Soil (48 Hours)
Test 3 Final Soil (48 Hours)
Test 4 Final Soil (48 Hours
Sample No.
GG3866/67 Composite
684-14-2A, 684-14-2B
684-18-2A
684-36-1A
684-39-IA
KN/»-94«rrE.Er?03/SrrE3IUT.REV
-------
CENTRAL FILES
Project No.
INTERNATIONAL c , B ,,/,/u
TECHNOLOGY Sent BY — r>atA V/S/1/
CORPORATION ,
Return Copy to Sender_ A (r*
CERTIFICATE OF ANALYSIS K' —M
ITAS-Knoxville March 28, 1991
304 Directors Drive
Knoxville, TN 37923
Attention: Mr. Ed Aloerin
ITAS-Knoxville Project Number: Site ETP-03/483000
This is the Certificate of Analysis for the following:
Project Number: 217-92
Date Received by Lab: February 26, 1991
Number of Samples: Eighteen (18)
Sample Type; Eiuhteen (18) Soil
Introduction
On March 26, 1991, eighteen (18) samples were received at ITAS -
St. Louis laboratory from ITAS-Knoxville. The list of analytical
tests performed, as well as receipt and analysis, can be found in
the attached report. We were instructed to only analyze samples
below. The samples were labeled as follows:
Soil samples: 1407-001 684-11-1A
1407-002 684-11-4A
1407-003 684-12-1A
1407-004 684-13-1A
1407-005 684-14-1A
1407-006 684-14-2A
1407-007 684-15-1A
1407-008 684-16-1A
1407-009 684-16-2A
1407-010 684-17-1A
1407-011 684-18-1A
1407-012 684-18-2A
1407-013 684-19-1A
1407-014 684-19-4A
1407-015 684-34-1A
1407-016 684-34-3A
1407-017 684-35-1A
1407-018 684-36-1A
Regional Oflice
13715 Rider Trail North . Earth City. Missouri 63045.314-298-8566
If Corporation a a wholly owned subsidiary cl :r,terr.aucr.j. 'ec^r.aogy J.-.-pc.M?...-.'
-------
INTERNATIONAL
TECHNOLOGY
CORPORATION
ITAS-Knoxville
March 28, 1991
Project Number: 217-92
II Methodology Modification
Background
According to Hr. D. Hesse, Organic Technical Director;
samples received from Project 217-92 have been very high
in dioxin, requiring about a 1:25 dilution prior to GC/MS
analysis. In order to maintain the highest accuracy
possible, the following dilution procedure was adopted.
Dilution Procedure
1) The sample extraction weight is reduced from lOgrams
to 5grams. This accounts for a 1:2 dilution.
2) Internal standard and surrogate concentrations are
added at 12.5 times their normal levels. 100-ml of
extract solvent (acetone and hexanes) are added to
the sample then the sample/solvent is shaken as
normal.
The sample is allowed to settle, then 8-mis of the
supernatant extract is pipetted off the sample and
placed onto the cleanup columns, etc. This accounts
for a 12.5 dilution.
This procedure, then, dilutes the sample 25 times, but
results in normal concentrations of the internal standard
and surrogate standard in the final extract.
Calculations
Calculations are performed using the normal DBASE
program, LRGCMS, except that the sample weight is divided
by 12.5 prior to entry into the program to account for
the 12.5 times greater internal standard concentration.
This avoids having to modify the program.
Ill Reoroducibilitv Problems
Ilia Backsround
Duplicate analyses of sample 1407-006, using the 25-fold
dilution procedure, has resulted in poor reproducibility,
giving results of 174 and 268 ng/gm.
Regional Otlice
137 15 Rider Trail North . Earth City. Missouri 63045 . 3 14-2
-------
INTERNATIONAL
TECHNOLOGY
CORPORATION
ITAS-Knoxville
March 28, 1991
Project Number: 217-92
Illb Discussion
Typical reproducibility for the Region 7 methodology is
less than 10%. Therefore, we do not suspect that the
problem lies in the GC/MS analysis, but rather lies
either in the problems associated with obtaining two
equivalent aliguots of the sample or with varying
extraction abilities.
To test the latter, extraction of the same two aliguots
of 1407-006, which had previously given 174 and 268
ng/gm, was continued by sonicating the sample solvent for
10 minutes each. After the sonication (and setting in
solvent overnight), the samples had become fine powders,
as opposed to the half-pea size lumps previously
observed. Concentrations of dioxin were determined at
238 and 356 ng/gm, respectively. The percentage increase
was the same for both. This test shows that extraction,
cleanup and GC/MS analyses are reproducible (based on the
equivalent increases) . However, it also shows that
ultrasonication for these samples is important.
IIIc Conclusion
The irreproducibility lies in the inability to obtain a
homogeneous sample, which we assign to dumpiness of the
sample.
Our best alternative is to analyze each sample in total,
so that no aliquot discrepancies are possible. However,
this would not eliminate sampling errors from occurring
at the sample site.
Reviewed and Approved:
Sally A. (jiane
Project Manager
Regional office
13715 Rider Trad North-Earth City. Missouri 63045'314-2
-------
INTERNATIONAL TECHNOLOGY CORPORATION
ITAS-Knoxville
304 Directors Drive
Knoxville, TN 37923
PROJECT NO.: 217-92
CATEGORY
METHOD
MATRIX
SAMPLE DATE
Client
ID
684-11-1A
684-11-4A
684-12-1A
684-13-1A
684-14-1A
684-14-2A
684-14-2A
NA
NA
NA
684-15-1A
684-16-1A
684-16-2A
684-17-1A
684-18-1A
684-18-2A
DIOXIN
Region VII
SOIL
02/26-03/20/91
LAB
ID
1407-001
1407-002
1407-003
1407-004
1407-005
1407-006
1407-006
DUP
BLK8694A
BLK8694B
SPK8694
1407-007
1407-008
1407-009
1407-010
1407-011
1407-012
PARAMETER
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
REPORT DATE
DATE RECEIVED
DATE EXTRACTED
DATE
ANALYZED
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
DETECTION
LIMIT
NG/GM
3.75
3.75
3.75
3.75
3.75
3.75
3.75
3.75
3.75
NA
3.75
3.75
3.75
3.75
3.75
3.75
: 03/29/91
: 03/26/91
:03/27-28/91
CONC
NG/GM
200.1
208.4
157.6
133.6
191.4
174.4
267.7
ND
ND
101 %
251.9
255.0
258.9
238.4
424.9*
238.9
NOTES: NA=NOT APPLICABLE; ND=NOT DETECTED
-------
.NTERNATIONAL TECHNOLOGY CORPORATION
ITAS-Knoxville
304 Directors Drive
Knoxville, TN 37923
PROJECT NO.:
217-92
CATEGORY
METHOD
MATRIX
SAMPLE DATE
Client
ID
684-11-1A
684-11-4A
684-12-1A
684-13-1A
684-1401A
684014-2A
684-14-2A
NA
NA
NA
684015-1A
684-16-1A
684-16-2A
684-17-119
684-18-1A
68401a-2A
DIOXIN
Region VII
SOIL
: 02/26-03/20/91
LAB
ID
1407-001
1407-002
1407-003
1407-004
1407-005
1407-006
1407-006
DUP
BLK8694A
BLK8694B
SPK8694
1407-007
1407-008
1407-009
1407-010
1407-011
1407-012
PARAMETER
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
REPORT DATE
DATE RECEIVED
DATE EXTRACTED
DATE
ANALYZED
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
DETECTION
LIMIT
NG/GM
3.75
3.75
3.75
3.75
3.75
3.75
.75
3.75
3.75
NA
3.75
3.75
3.75
3.75
3.75
3.75
: 03/29/91
: 03/26/91
:03/27-28/91
CONG
NG/GM
200.1
208.4
157.6
133.6
191.4
174.4
267.7
D A
ND
101 %
251.9
255.0
258.9
238.4
424.9
238.9
NOTES: NA=NOT APPLICABLE; ND=NOT DETECTED
-------
vlTERNAT'.ONAL TECHNOLOGY CORPORATION
ITAS-Knoxville
304 Directors Drive
Knoxville, TN 37923
PROJECT NO.:
217-92
CATEGORY : DIOXIN
METHOD .Region VII
MATRIX : SOIL
SAMPLE DATE: 02/26-03/20/91
REPORT DATE : 03/29/91
DATE RECEIVED : 03/26/91
DATE EXTRACTED:03/27-28/91
Client
ID
684-18-2A
684019-1A
684-19-4A
684-34-1A
684-34-3A
684-35-1A
684-36-1A
684-36-1A
NA
NA
NA
LAB
ID
1407-012
DUP
1407-013
1407-014
1407-015
1407-016
1407-017
1407-018
1407-018
DUP
BLK8727A
BLK8728B
SPK8728
PARAMETER
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
TCDD
DATE
ANALYZED
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
03/28/91
DETECTION
LIMIT
NG/GM
3.75
3.75
3.75
3.75
3.75
3.75
3.75
3.75
3.75
3.75
NA
CONG
NG/GM
471.5*
263.7
249.3
286.8
247.8
218.1
224.3
274.6
ND
ND
102 %
*Concentration
diluted 1:25,
standards.
reported greater than curve level. Sample already
further dilution would be prohibitive based on cost of
NOTES: NA=NOT APPLICABLE; ND=NOT DETECTED
-------
IT Analytical Services • St. Louis
»•»•
frojtct tie.
'S^TL-TI
•3i\-)~ » «— <*7 A^/ fc-
T^o 1 J~ ;
Shlry>«r ^ ^ B^JIB Sr^^ '-' «
Ou. D.I. I ' ^*v ^ - ^ 1
'•f, (:• U!JI !'<••
1 AUIQ ;»:;ifcA-. v j <^1 ^ JL^
| 6* ICA?
|.P8E? UVIS!0»'
0-PIE? ^
CC STCiAli/lCCATICiJ
C1KS
KCJ
mo, „'.- 1 ' ' ' >
x- ' . oinin
*<••
<•'
IWI XO XV XI 1 IX l| II (X CD Cg CR tU H I
IX Ik U V (V II
QfAAl XI II fl II K ,
UXJ UUl
6i-3Civ^/ffpB?/tiJ ^rQ ?" ^'
' 1 A f ! i <_.«'/•'•'
/
MOXHXl
HX2XR5COS
oxie 10 DISPOSE
*
^%'»
teT
,i • i A
i .1 L- • •//
5c/c
X
k* '
^
v
__,A ,
BSSE==S.~ — ™™,-™™™,
M/%1
6Sr-V"
//• H'i
)
_\.
V
X
X
— & —
^ ' /»?3
frH-
/ A • / /i
3 .r/f/
— ~?
y
V
X
V^]
f- % V -
/ ,} • 1 /)
.'i •::&••//
>
V
V
X
/v^
„___
/T~77T
.5 • ' >"/
>
V
.V"
V
^^n
i
/.V -XA
3 • *•! • V 1
-^
X
V
V
"' •>;'•! f
!
/ ^ //i
3 • r •'.'"(
— ">
«'
x:
m_m^-^-m— ,.
r-~"
1
i
^—^
''• 'r *•/ -
n- //i
.! r- v
S^/L
K
*C
X^
fH .,
; 6 ^' / )
;
V
V
V
__^
/, , .,
/ •/ • It]
3 •? r<
i' t
v
V
(
*vmm^^^^~
— rmJ
-------
IT Analytical Services • St. Louis
OKI
LOOIM MO.
o . -\/ _ c.
II id . ^s f r *7 ok
3— K ^_
"g^{o" t a<
It4 Mt. *f Itnolfi 1 ft
tt4 Kiirlx ^\O I C-
^ ' | 1 *UIO CJICIMAl
ciicm ,3— | - \CV\DX ^v 1 1 (° |cx!>
"Oj. Mgr. ^ | L/^/Jc? 0-PUP
i — ^ 1 «^ CC S t C2AO(/LOCAT 1 Cf
Shipper p«.ed , Lr fc • C'.f.s
. . . £-, j./.o
l»yn b«u __ *-} - ^2 T~ ij l'hOJ Mt"
01 "(^
J»«Vlo Morx''
MCX>| JLO At, XJ 1 IX 11 II Cl CO CO Cl CU II <
Sc«-p 10 Wo.
. Cl HO XX MO «A Ml » M || II (| SU t* l(
' 0«icrlp(lon
Tl IK TV U V III 11
ilirpCi Out
•orui M it ri u u
Hiulx
IIKI X1UI
wet . OIIUIFTIOH COXIAIHI* PUIKVXIIVI
^.^.Ki/r^n^/XZ/ /P..Q ? x/o/,!^.^ ^ fr/(^
ITT-K- ' // ' i • ' I
1 1 . JtVj /JL, ' ' /
1 HfO tb14/*{) //1J .^ir-t m^f l>
' / ^
oiiwuii womi
HA2AROCUS
OAlc 10 DISPOSE
PM 5ICWAIUX
Ccnrr
6vt-
l'i-)ri
3-'3-'>l
->
V
X
/\
i-r-'/y
SVM
!• rs.'/-
i'r **»
_ *»
— •)
v
>r
V
A,
'^
Uv-
. •<,-//)
_X-/v-//
-_ -. >
V
\
X
A
/¥'":-'.
xr/t
/:-.-^-
:.?T-//»
.?-;"••.')
^
\
V
V
/•/-•'//•
S&l'l
/.»-..-
V *" /.
."; • ' •; ' '
....->
V
y
X'
/ 1 e •!/
/?>/.
/"
'• :' '/ '
«>•///
"' -..V* ; '
">
-.
\
X"
— i —
-------
CORPORATION'
I'I / IX t I
rr i's 3*^
WHlTF . Trt Armmna/tw &A/nnlA«
-------
INTERNATIONAL
TECHNOLOGY
CORPORATION
PRO.IFCT NAMF S"7
PROJECT NUMBER 'L&.J>P£> Q
PROFIT CENTER NUMBER
PROJECT MANAGER £r- 5
BILL TO
REQUEST FOR ANALYSIS
DAI i" r.AMi'ir r, r,i uppro
LAB DESTINATION
LABORATORY CONTAC F
SEND LAB REPORT TO
'.) 0 0 ^ O "7
R/A Coni/ol No C ~> J •" i /
C/C Cooirol fjo A} '3 ^ ?
^/2 >/-^ /'^ //v'
&+
z. PA
O* i
664
184
3URCHASE ORDER NO.
Sample No.
" // " / n"
'//-4A
-/>.'/^
- / '?-//)•
- i4-yA
-M-3/)-
DATF RFPORT RFOIIIRFn / / ? / /9 1
T6.2-2-" £>Jtr" O3I,. PRQIFHT CONTACT ' H $ <€ £t?C'T"r"
PROJECT CONTACT PHONE NO. .£>.( ^ '&.<>' $1/1 ^
Sample Type
Sb/c^
V
Sample Volume
/o- Jo \
a
y
Preservative . Requested Testing Program Spec .1 m$:-jc: o"
|vj fi""AJ & C l^i" P1 ) ^i? ^^ "H^- p'D A *<•**} I '-if A* C 0
/ ^» jf ^""^ / (5r o * j
[)fctfLC- P? T
; tr^ite^
Kcr*2ii-ci
\/ v7" . . .
* **!
TURNAROUND TIME REQUIRED (Rush must be approved by iheLabOfaioiyPrOjeci Manager ) OC LEVEL iLe-.cis II an= m s-::e:: :o Sv-
\ / SL/om>:!ec lo 1.10 DC:OIC COQ ^
I ^y II ill
Normal
Rush
: •: II>QW -e—i'-'s — ..':
(Subject to rush surcharge ) I ^- » '" - - Pro)ecl Spec.hc
(Please indicate it sample(s) are hazardous mateiiaisand/orsuspeciedtoconta'nnignie.e.so.•: " •*; ••> •
Return to Client Xl Duposal by Lab Archive _. (Indicate number Of months )
1 •£;.'- 5
Sprl •
FOR LAB USE ONLY
Received by
Dale I Time
-------
INTERNATIONAL
CORPORATION
PROJECT NAME/NUMBER
CHAIN-OF-CUSTODY RECORD
67/V^ Jy&.X>6(.\
") "} 7 f C c
R. A Control No ^ > D C r/
r. r. r.,jr,,,,,, ,„, /, 9 *> 2 2 2
SAMPLE TEAM MEMBERS A . (\ Q Q I tV
'Imiifilft
Number
-y5-/#
-/#-//>
- /£ -?^} -/
/r^x -famjt^p
j -y / /
•5 0 ^t>v /t^ A-^?*^^
y .
^f <5 x^A. (s^£lL~ C-A>>-5tv^
1 )llllt IIIK 1 1 tlllll
Colioctvu
3A77/
^ / V
v/^/f /
7/7 /?/
7 AA /
i//ihi
CARRIER/
Sinn|jii*
Typo
Jfi/L
I
/
'WAYBIL
l.«n|
L NO I t- 'S 1 v
tflfl'i
V
/
(N.itt i. ,iM,j o,i'». «*!./•/,. : ».
i
-- I
rs
\
"uctions' C^S tf /t* c/ v? "7j l-)^ t-'c""Z- O/'v" %'/ /"^ '
nple Hazards' rr- ?i o -/"/I^ Receiv<
By^t-C.-.S?-/, ^r—N
led Bv 4. Relmqi
Rv- Receiv
jished B^
3d by. —
jished B^
ed By.
i
9£ 0 Jr V 7 k( t-ki^t -;>,<
/,.,: /¥T,^r , j 7i
l' IH >
-------
INTERNATIONAL
TECHNOLOGY
CORPORATION
REQUEST FOR ANALYSIS
0 1 ~ 0 Q C
R/A Control No - > •
f y I (
C/C Control No * ' ? ? ? ?
PROJECT NUMBER
PROFIT CENTER NUMBER
PROJECT MANAGER
BILL TO
PURCHASE ORDER NO.
V 4 2- 2-
_f 'S> A*L
UAH. :,AMI'I I :, MIIIM'I I)
LAB DESTINATION
LABORATORY CONTACT
SEND LAO REPORT TO
DATE REPORT REQUIRED
PROJECT CONTACT
PROJECT CONTACT PHONE NO
/ '-'.'
Sample No.
- /A- /V?
-if- /A
-18-lflr
Sample Type
Sou-
Sample Volume
/6 '
Preservative
Requesiea Testing
—a.?.
V
r
TURNAROUND TIME REQUIRED: (Rush must be approved by iheLaboraioryPfOieci Manager) OC LEVEL (Levels II .me in suu.c-c: ;o s--cn.vc;o y_.er sat-c • : -v^ •f'-,.--
- - > * SuDmilieO lo lao De'ore Deg on',ngv.or«i
Rush (Subject 10 rush surcharge t I _. x*C^ " '» P-OJCCI Sprc.i.c
Normal
POSSIBLE HAZARD IDENTIFICATION (Please indicateilsampic(s)are hazardous maicnaisaiut/oi • ,u-. p'-cioo to ('.»i...r. n.< ;•> i. ..-••. >.• n ..• n-:
__ *v •«.
Non-hazard Flammable Skin Irnlant . Highly Tone
SAMPLE DISPOSAL (Please indicate disposition ol sample loiiowmganaiyiisL.io .-..n cnarge lor pat' .ng sn.pp." ; .<•:< ..- «'•„•
Rolum lo Client __2^_ Disposal by Lab.. Archive (Indicate number olmonihi )
Olhe,
FOR LAB USE ONLY
Rece
.ved by/?'7i:< Hy / . ": * ^
Dale' Tune
*i >*•)
-------
INTERNATIONAL
I'lHl.JI r;| NAMI/NllMIUII l> ( ^
SAMPLE TEAM MEMBERS ''
CHAIN-OF-CUSTODY RECORD
i AH ui MIMA i ION I I l\
~> •> 4 .it / ' '
R A Control No . 1_±1_ _.'_ '
C C Conlrol No A 93223
'./ ^ c i. / <•
A) ^ X
Sample
Number
-/5"/£
- WH/I
-?4-jA
•34-.? A
-35- M
3&-//J
Special Instr
Possible Sar
Sample
Location «n/^
t
/
Container
Typo
i'l/H.
/
iNamo ano D«te<
x
««or-iV..
--- -
.rtinnc- lj f ff Kf^TH^^ Dfrg-LGfcto ftt ?£P±Cf-t / ^fatO'/lf T7//C
nnlflHa/arrls ^H^ £&*?».?{ CWTtpV *-7-C-^b ^.. _^ [ L£ ill_J>/Wj_. r'Vr
MY- .
SIGNATURES: (Name, Company, Date and Time)
1. Relinquished By: /TTfe^
Received By:
3 Relinquished By.
by _-
2 Relinquished By:
Received By: —
Relinquished By ._ . .
Received «y .
iccompany »impi«j
-------
INTERNATIONAL
TECHNOLOGY
CORPORATION!
I'IK Ml < . I IJAMI
PROJECT NUMBER
PROFIT CENTER NUMBER
PROJECT MANAGER
UILL 1C
REQUEST FOR ANALYSIS
I.//'/
r"/
2
t>AH :.AMTI I :.:.nii'iM I,
LAB DESTINATION
LABORATORY CONTACT
SEND LAD REPORT TO
DATE REPORT REQUIRED
R/A Conirol No 233883
C/C Control No, ""
V'1. ^ '
.JT '?-£ ./I//! t 0.6 * C
1 ? 3
PURCHASEORDER NO.
" #3 1 I "
PROJECT CONTACT
PROJECT CONTACT PHONE NO
J70?
Sample No.
[T-//T
34
- 3 5 - h -
A
Sample Type
Sample Volume j Preservative
4
RoQuesiec Tesfing
' 3
Sso: .1
2.
I*. 3fr"
s /
/H
-Af
TURNAROUND TIME REQUIRED: (Rush must be approved by Ihe Laboratory Proiect Manager) OC LEVEL (Levels II anc in sua.ect to Sorcna»ge p-o;o:' soec'1 c n'i» -e-e-:s •-„»•
Normal 7^-. R u s h (Suojecl lo rush Suicharrje) I ^~ II "I P«OICCI Spec.lie
POSSIBLE HAZARD IDENTIFICATION
(Please indicateil sample(s) are ha^araouSmaicnalsanc/O'SuSroC'OC :o ccf.:a-"' •jfii':.vs"'".i.1.i'CO,.^l...:.l^'.i"'.-.'i
N O n - h a 7 a r ri
SAMPLE DISPOSAL
Return to Client
Flammable
Skin Irritant
Highly Tone'
I
FOR LAB USE ONLY
(Please indicate disposition ot sample loiio.ving analysis La; IMII cna'ge lor pac^iiy sn.pp.ny ,irc''.•/,IMC c ^: ~s.i
Disposal by Lab Archive _ (Indicate number ol months )
3 • .2 6 • -i
Received by
Dace/ Time
-------
INTERNATIONAL
TECHNOLOGY
CORPORATION
I 'I 10.11 ( , I UAMI
PROJECT NUMBER
PROFIT CENTER NUMBER.
PROJECT MANAGER
UILL IO
L.//7
REQUEST FOR ANALYSIS
(-/ r 4 •'•
R/A Conirol No 233889
'2
XH
PURCHASE ORDER NO.
~ 03?/" 6 "3 4
! I:,: ,i iiri Ml;
LAB DESTINATION
LABORATORY CONTACT
SEND LAD REPORT TO
DATE REPORT REQUIRED
PROJECT CONTACT
PROJECT CONTACT PHONE NO
C/C Control Mo
V' '. /
IT
./JJ L 0_i*
U
Sample No.
/T-//V
iC _ /LJL
-54-1 A
-34-5V4
-951-//}-
-3^/ft
Sample Type
^>C> ) (• —
v
^
Sample Volume
/ 0-30*1
\
J
v
Preservative Requested Testing Prog
£tn\ce&b=rt 2(3 7A rc-o-o
!
1
V /
V ! V
t
t
!
I
|
.1 |r>s''uc; o-»
/K
TURNAROUND TIME REQUIRED: (Rush must be approved by the Laboratory Proiect Manager) OC LEVEL (Levels II anc in su-.\eci to s^cnafQe p-o;e:'ssec''c'«.•::-'e-t?-:s •--»•:<
x -/ SiiOmillec! IO I.IO Ce'orw Ct-n "'• "5 >•.(•>•• •
Normal 7^> Rush (SuDjecl lo rush surcharge) I ^~- II '" Proicci Spec-lie
POSSIBLE HAZARD IDENTIFICATION (Please moicateil sample(s) arehaiaroousmatenaisanc/or iusr<;c:oc;:occr-.:a.-.--jn;..•...••.-'n.,.-..-co-.v<...:.•/.!-•.:•*
/
'^ l*~~f
Non-hazard Flammable Skin Irritant .._ Highly Tone
SAMPLE DISPOSAL (Please mdicaie disposition ol sample lolloping analysis La-j \MH cnarge lor pac»in
-------
'llUl51 HAS . St. Louli
Master Sample Login: 1407
Account; 10610
!•! 1 MM 1 »\t ,'ti
Description: Monsanto Thermal
facility:
Project Manager: s. Lane
iMplt NO. Client 10 Key Dter Shlocer
COIL RECV DUE Matrix Analyses
UOrOOl 664-ll-lA
W-11-4A
1407-003 M4-|2-1A
1407-004 M4-13-1A
HOT-MS §H-H*1A
1407-006 6&4-14-2A
1407-00,? M4-15-1A
1407-008 6A4-16-1A
1407-009 684-16-2A
1407-010 6&4-17-1*
02/26 03/26 04/Z'. I CIMX
- .. .02/26 04/J4 f£p-£X
_. . - fli/27 03/2.6. §4/24 ffp-fx.
02/2S 03/26 04/24 fW'«
03/01 03/26. 04/14 ffO-EX
03/04 03/26 04/24 FED-EX
2i££iJJ3/i4JMli_IMi£x__
03/06 03/2604/24 FED-EX
03/06 03^2* 04174 fFn-FX
03707 03/26 Oi/ft ffO-EX
Soil
Soil
Soil
Soil
__Iflil___
Soil
Soil
soil
_JL2il__
Soil
Soil
Soil
_iail__.
Soil
Soil
Sol I
Sqil
Soil
Soil
Soil
Soil
Soil
Soil
soil
__i2J_L_
Soil
Soil
Soil
Soil
Soil
Soil
Soil
Soli
soil
soil
Soil
Soil
Soil
Soil
Soil
OAIA/SIONCP/01
OIXIN/IC007/01
P-OIXIN//02
OATA/STORIP/01
OIXIN/TC007/01
P-OIXIN//02
OATA/SIOREP/01
OIXIM/TC007/01
P-DIXIN//02
DATA/STDREP/01
OIXIN/TC007/Q1
P-OIXIN//02
OATA/STDREP/Q1
. 0IXIN/TC007/01
P-OIX1N//02
DA1A/STOREP/Q1
OIXIH/IC007/01
P-OIXIM//02
OAIA/STOREP/01
OIXIN/1COD7/01
P-OIXIN//02
OAIA/SIOREP/01
OIXIN/TC007/01
P-OIXIN//02
OATA/STDREP/Q1
DIXIN/TC007/01
P-OIXIN//02
DATA/S10REP/01
OIXIM/TC007/01
P-OIXIN//02
482322.38
Class Preservatives
CUMMIN IS;
S
S COLD
S COLD
__ffiMMilL^^
S
S COLD
S COLD
_JMMM£H1S'
S
S COLD
S COLD
__jy2tl£JUJia________
s
S COLD
S COLD
tgf^Hjj^
S
S COLD
S COLD
COMMENTS:
S
S COLD
S COLD
___£g.,M£MJ_|j_
S
S COLD
S COLD
COMMENTS:
S
S COLO
S COLD
COMMENTS:
S
S COLD
S COLD
COMMENTS:
S
S COLD
S COLD
03/26/91
0«AM/ IINA^X NOCO'OeKAKCC
(.OH ': „ _^>
Reviewed By:
Container
40 ML VIAL
co ML VIAL
40 ML vfAL
40 ML VIAL
40 ML VIAL
CO ML VIAL
LO ML VIAL
40 ML VIAL
40 Ml VIAL
40 ML VIAL
LO ML VIAL
CO ML VIAL
CO ML VIAL
40 ML VIAL
CO ML VIAL
40 MLVIAL
CO ML VIAL
40 ML VIAL
LO ML VIAL
LO ML VIAL
^ . (t\C9-i/^
Storage Sire p* ,
e IWE
«109£
™—™ ~™™-^_~™«™__™, „___
t IWE
™~™_~_« — __™™_, — „__ — „ — .„
• 109E
• 109E
• Im
• 109E
• 109€
• 109E
• 109E
• 109E
• 109E
e IOSE
e IWE
• 109E
. . .
e IWE
• 109E
• 109E
• 109C
' -EX
Sol I
------- |