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

-------