BIOREMEDIATION  OF HAZARDOUS WASTES
RESEARCH,  DEVELOPMENT, AND FIELD
EVALUATIONS,  1994
(U.S.)  NATIONAL RISK MANAGEMENT RESEARCH LAB., CINCINNATI, OH
SEP 94
 US. DEPABTWtrNT Qf COMMERCE
 National Technical Information Service

-------
                                           EPA/600/R-94/161
                                            September 1994
 BIOREMEDIATION OF HAZARDOUS WASTES
Research, Development, and Field Evaluations
3iosystems Technology Development Program
     Office of Research and Development
    U.S. Environmental Protection Agency
           U.S. Environmental Protection Agency
     Ada, OK; Athens, GA; Cincinnati, OH; Gulf Breeze, FL;
            and Research Triangle Park, NC
                                    t Printed on paper that contains at leas
                                     50 percent recycled fiber.

-------
                                   TECHNICAL REPORT DATA
                                    nstruciiont on the reverie before complennfj
1. REPORT NO.
 EPA/600/R-94/161
4. TITLE AND SUBTITLE
 BIOREMEDIATION  OF HAZARDOUS  WASTES   RESEARCH,
 DEVELOPMENT AND FIELD  EVALUATIONS  - 1994
                                                           6 PERFORMING ORGANIZATION CODE
             3 REC
               REPORT DATE
7 AUTHOR(S)
 Fran Kremer
             a PERFORMING ORGANIZATION HEPORT NO
9 PERFORMING ORGANIZATION NAME AND ADDRESS
                                                            10 PROGRAM ELEMENT NO
 U.S. Environmental  Protection Agency, National Risk
 Management Research Laboratory,  Cincinnati, OH 45268
             11 CONTRACT/GRANT NO
 '2 SPpNSOBING.AGENCY NAME AN.D ADDRESS ,  ,  .    .
 National Risk  Management Research Laboratory
 Office of  Research  and  Development
 U.S. Environmental  Protection Agency
 Cincinnati, OH  45268
             13. TYPE OF REPORT AND PERIOD COVERED
             14 SPONSORING AGENCY CODE
19. SUPPLEMENTARY NOTES
18. ABSTRACT
       The  proceedings of the 1994 Symposium on BioremeuMation of Hazardous Wastes,
 hosted  by  the  Office of Research and Development  (ORD)  of the EPA in San Francisco,
 California.  The symposium was the seventh annual meeting for the presentation  of
 research conducted by EPA's Biosystems Technology Development Program (3TDP)  and by
 affiliated Hazardous Substance Research Centers  (HSRCs).   The document contains
 abstracts  of recent research projects, ranging in scope from laboratory application
 to cleanup evaluations in the field.  4i papers  and  numerous posters presented  at
 the  symposium  are organized into six program  areas:   Bioremediation Field
 Initiative,  Performance Evaluation, Field Research,  Pilot-Scale Research, Process
 Research,  and  Hazardous Substance Research Centers.   The proceedings also contain  a
 brief synopsis of introductory  remarks made  by  opening speakers.
                                KEY WO1DS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS  0. COSATl FitlJ Group
 bioremediation,  biological
 treatment,  hazarous wastes
 ORD,  R&D
18. DlSTRiauTlON STATEMENT


  RELEASE TO  PUBLIC
                                                                Rfptlrt)
70,
               I Thi
                           JJ PO.CF
SPA P~m 3220-1 (».». 4-77)   »««v.ou» to T.ON  • O*«OLITC

-------
                                                          INSTRUCTIONS


                  LP^" . numb., a, ,• appear, on tne cove, of the pub.io.ion.
     2.   LEAVE BLANK

     3.   RECIPIENTS ACCESSION NUMBER
         Reserved for use by ejeh report recipient
         number uid include subtitle for ihe ipecific title


    *'   "h^JtlhaU carry a date md.caunf at lean month and you  Indite lh« "^ on .huh u «... *UMcJ ,. t  Jo,, „, .„„, j,M
         approval, dale of preparation, etc I

    t.   PERFORMING ORGANIZATION COOC
         Leave blank.


    ''   c"e?£mV*l in c -vennonil order Uorir, ft Doc. J  Rotor, Doc  cl, I   LiM julhur\ jllilunun ,1 ..... llerv ir,..n ilu- |HTl,,rmmf ,,,Jni.
         ZJIion.

    8.   PERFORMING ORGANIZATION REPORT NUMBER
         ln«n if performing orpmtation wuhej 10 a»si(tn thi\ number

    9.   PERFORMING ORGANIZATION NAME AND ADORES3
         Give name, street. city, lute, ind ZIP code. Litl no more than two levels ol jn orpjni/jliunjl hircinhy

    10.  PROGRAM ELEMENT NUMBER
         UK the profnm element number under which the repori was prepared.  Subordnulc numlx-r. nu>  lie iiulmU.I m  iviu-nilH^v

    11.  CONTRACT/GRANT NUMBER
         Iniert contract Of (ram number under which report w» prepared.

    12.  SPONSORING AGENCY NAME AND AOORESS
         Include ZIP cod*.

    13.  TYPt OF REPORT AMD PERIOD COVERED
         Indicate interim flaxl, etc.. and if ipplKabte. d*l Mgnitkini mlormjimn <.untjinvil m HK fvi«>ri  II ilu '> i">n .'•MI.IUI>.
        BfniTicani bibliofraphy or literatu/e survey, mention it hen-

    17.  KEY WORDS AND DOCUMENT  ANALYSIS
        (a) DESCRIPTORS • Select from the Thcuurui of hngiiwcnrj jnd SucntiiK ferm> I ic proper ju"inn/ed inrn> Hut ijennly itu- nui"
        concept of the -eieuch  and are mfficicnily tpecilk  index cnirio lor vji^lu^mg

        (b) IDENTIFIERS AND OPEN-ENDED TERMS • Use identifiers for protect n;:mvv code rume "|vn
        ended terms written  in descriptor firm for thoje subjects for which no dcvriutor evuu.

        (c) COSATI  I II LD GROUP • I leld and froup js»i|nments ire to  be taken Irom the IU65 ( OSAll Sul^-it ( ju^'fy I 1-1   Smtc ilw. m
        jonty of documents are multidisciplinary in nature. Ihe Primary I  wld/(.roup j\Mgnmentlvt will be speiiiu JIM iplim .irej nl liunu"
        endeavor, or type of physical ob|ect. The appucalion(s) will be crow-retcrentcd with seiondjry I n  III  (.rum- jsMpnineniv (hji «illl"il'
        the primary  postingts)

    1aX  DISTRIBUTION STATEMENT
        Denote releaiabilily to the public or limitation  for reav>m other th»n \eiuriiy  tor temple  KclejM I  ....... iuil  ' iii  j>\<-  j».ni.ii"i">
        the public, with address and  prue

   It. ft 70. SECURITY CLASSIFICATION
        DO NOT submit classified reports to the Ninonal Technical Informjiion wrvne

   21.  NUMBER OF PAGES
        Insert the total number of pages, including this one and unnumbered pj(cv hut exclude diMnbution IIM ii jnv

   22.  PRICE
        Insert the price set by the National Technical Information Service or the Government I'rmnng Olme  it known
EPA Faro 2720.1  Id... 4.77) (».»....)

-------
                                      Notice
The information in this document has been funded wholly or in part by the U.S.  Environmental
Protection Agency (EPA) and has been reviewed in accordance with the EPA's peer and adminis-
trative review policies and approved for presentation and publication. Mention of trade names or
commerciaJ products does not constitute endorsement or recommendation for use.
                      PROTECTED UNDER INTERNATIONAL COPYRIGHT
                      ALL RIGHTS RESERVED.
                      NATIONAL TECHNICAL INFORMATION SERVICE
                      U.S. DEPARTMENT OF COMMERCE

-------
                                             Contents
Introduction  .  . .  .


Executive Summary,
                                                                                                Page
Section One: Bioremediation Field Initiative '.	5

       Intrinsic Bioremediation of TCE in Ground Water at an NPL Site in St. Joseph, Michigan               7
              John T. Wilson, James W. Weaver, and Don H. Kampbell

       Enhanced Reductive Dechlorination of Chlorinated Ethenes ...                             11
              Zachary C. Hasten, Pramod K. Sharma, James N.P. Black, and Perry L. McCarty

       Bioventing of Jet Fuel Spills I: Bioventing in a Cold Climate
       with Soil Warming at Eielsen AFB, Alaska    	                             14
              Gregory 0. Saytes,  Richard C. Brenner, Robert E. Hinchee,
              Andrea Leeson, Catherine M.  Vogel, and Ross N. Miller

       Bioventing of Jet Fuel Spills II:  Bioventing in a Deep Vadose Zone at Hill AFB, Utah                 18
              Gregory D. Saytes,  Richard C. Brenner, Robert E. Hinchee,
              and Robert Elliott

       In Si*u Bioremediation of a Pipeline Spill Using Nitrate as the Electron Acceptor                     22
              Stephen R. Hutchins, John T.  Wilson, and Don H. Kampbell

       Performance Evaluation of Full-Scale In Situ and Ex Situ Bioremediation of
       Creosote Wastes in Ground Water and Soils	                 25
              Ronald C. Sims, Judy L Sims, Darwin L. Sorensen, David K. Stevens,
              Scott G. Huling, Bert E. Bledsoe, John E. Matthews, and Daniel Pope

       Bioventing Soils Contaminated with Wood Preservatives	29
              Paul T. McCauley, Richard C.  Brenner, Fran V. Kremer, Bruce C.  Alleman. and
              Douglas C. Beckwith

       Field Evaluation of Fungal Treatment Technology  	      .     ...  33
              John A. Glaser, Richard T. Lamar. Diane M. Dietrich, Mark W. Davis,
              Jason A. Chappelle, and Laura M. Main

       The Bioremediation in the Reid Search System (BFSS)	          	    37
              Fran V. Kremer, Linda B. Diamond, Susan P.E. Richmond, Jeff B. Box, and Ivat. 3 Pudn/cki

-------
                                       Contents (continued)
                                                                                                Page

Section Two: Performance Evaluation	39
       Integrating Healtn Risk Assessment Data for Bioremediation .      .          .    .                 40
              Larry D. Claxton and S. Elizabeth George
       Construction of Noncolonizing E. Coli and P. Aeruginosa ...      	       .         42
              Paul S. Cohen
       Toxicant Generation and Removal During Crude Oil Degradation             .  .          .         45
              Linda E.  Rudd, Jerome J. Perry, Larry D. Claxton, Virginia 5. Houk, and Ron 'N.  Williams
       Potentiation of 2,6-Dinitrotoluene Bioactivation by Atrazine in Fischer 344 Rats .  .       .    .     .48
              S. Elizabeth Georqe, Robert W. Chadwick, Michael J. Kohan, Joycelyn C. Allison,
              Larry D. Claxton, Sarah H. Warren, and Ron W. Williams
       Effects of Lactobadllus Reuteri on Intestinal Colonization of Bioremediation Agents                  49
              M'rtra Fiuzat, S.  Elizabeth George, and Walter J. Dobrogosz

Section Three:  Reid Research  	53
       Field-Scale Study of In Situ Bioremediation of TCE-Contaminated
       Ground Water and Planned Bioaugmentation  	       .54
              Perry L. McCarty and Gary Hopkins
       Geochemistry and Microbial  Ecology of Reductive Dechlorination
       of PCE and TCE in Subsurface Material  . .      	        57
              Guy W. Sewell, Candida C. West, Susan A. Gibson, William G. Lyon, and Hugh Russell
       Application of Laser-Induced Fluorescence Implemented Through a Cone
       Penetrometer To  Map trie Distribution of an Oil Spill in the Subsurface              	         62
              Don H. Kampbell, Fred M. Pferfer, John T. Wilson,
              and Bruce J. Nielsen
       Effectiveness and Safety of Strategies for Oil Spill Bioremediation:
       Potential and Limitations  	       .  .            65
              Joe Eugene Lepo, C. Richard Cripe, and PH. Pntchard
       The Use of In Situ Carbon Dioxide Measurement To Determine Bioremediation Success              70
              Richard P.J. Swannell and Francois X. Merlin
       Intrinsic Bioremediation of JP-4 Jet Fuel Contamination at George AFB, California  . .       ....    71
              John T. Wilson, Michael L Cook, and Don H. Kampbell
       Factors Affecting  Delivery of  Nutrients and Moisture for Enhanced In Situ
       Bioremediation in the Unsaturated Zone	72
              James G. Uber, Ronghui Liang, and Paul T. McCauley
                                                 rv

-------
                                        Contents (continued)
                                                                                                Page


                                                                                                   73
 Section Four:  Pilot-Scale Research . ......................................
        Pilot-Scale Evaluation of Alternative Biofilter Attachment Media for Treatment of VOCs        ...      74
               Francis L Smith, George A. Serial, Makram T. Suidan, Pratim Biswas, and Richard C. Brenner

        Biological Treatment of Contaminated Soils and Sediments Using
        Redox Control: Advanced Land Treatment Techniques  ..............         •      79
               Margaret J. Kupferie, In S.  Kim, Guanrong You, Tiehong Huang. Maoxiu Wang,
               Gregory D. Sayles, and Douglas S. Upton                                               ,
        Research Leading to the Bioremediation of Oil-Contaminated Beaches   .....       .             82
               Albert D. Venosa, John R. Haines, Makram T. Suidan, Brian A. Wrenn,
               Kevin L Strohmmer, B. Loye Eberhart, Edith L Holder, and Xiaolan Wang
        Engineering Optimization of Slurry Bioreactors for Treating Hazardous Wastes        .....      .86
               John A. Glaser, Paul T. M^Cauley, Majid A. Dosani, Jennife' S. Plan, and E. Radha Krishnan
        Development and Evaluation of Composting Techniques for Treatment of Soils
        Contaminated with  Hazardous Wastes  ...........................          •   90
               Carl L Potter, John A. Glaser, Majid A. Dosani, Srinivas Krishnan, Timothy Deets,
               and E. Radha Krishnan
        Remediation of Contaminated  Soils from Wood Preserving Sites
        Using Combined Treatment Technologies   ...................             93
               Amid P. Khodadoust,  Gregory J. Wilson, Makram T. Suidan, and Richard C. Brenner
        Pilot-Scale Evaluation of Nutrient Delivery for Oil-Contaminated Beaches .......         .96
               Michael Boufadel, Makram T. Suidan, and Albert D. Venosa
        Field Treatment of BTEX in Vadose Soils Using Vacuum Extraction or Air Stripping and Biofilters       97
               Rakesfi Govind, E. Radha Krishnan, Gerard Henderson,  and Dolloff F. Bishop
        Dechlorination with  a  Biofilm- Electrode Reactor  .......................      99
               John W. Norton, Makram T. Suidan, and Albert D.  Venosa
        Use of Sulfur Oxidizing Bacteria To Remove Nitrate from Ground Water ..............   101
               Michael S. Davidson,  Thomas Cormack, Harry Ridgway,  and Grisel Rodriguez
        Engineering Evaluation and Optimization of Biopiles for Treatment of Soils Contaminated with
        Hazardous Waste   .......................................    103
                 ,1 L Potter and John A. Glaser
Section Five: Process Research   ....................................... 105

       Metabolic and Ecological Factors Affecting the Bioremediation of PAH- and
       Creosote-Contaminated Soil and Water .............................     107
              P.H. Pritchard, Jian-Er Lin. James G. Mueller, and Suzanne Lantz

       Metabolic Pathways Involved in  the Biodegradation of PAHs ..............        114
              Peter J. Chapman,  Sergey A. Se/ifonov, Richard Eaton, and Magda Grifoll

-------
                                Contents (continued)

                                                                                       Page


Environmental Factors Affecting Creosote Degradation by Sphingomonas paucimobi/is
Strain EPA505	           •          117
       James G. Mueller, Suzanne E. Lantz, and P.M. Pritchard
Molecular Genetic Approaches to the Study of the Biodegradation of Polycyclic
Aromatic Chemicals	         	           •  121
       Richard W.  Eaton, Peter J. Chapman, and James D. Nittgrauer
Comparison of Sulfur and Nitrogen Heterocyclic Compound Transport in
Crtosote-Contammated Aquifer Material  ....      	                124
       Ean M. Warren and E. Michael Godsy
Modeling Steady-State Methanogenic Degradation of Ptienots in Ground Water at
Pensacola, Florida   	          •   •  127
       Barbara A.  Batons, E. Michael Godsy, and Donald F. Goertitz
Anaerobic Biodegradation of 5-Chtorovanillate as a Model Substrate for the
Bioremediation of Pacer-Milling Waste     	               130
       B.R. Sharak Genthner, B.O. Blattmann, and P.H. Pritchard
Characterization of  a 4-Bromophenol Dehalogenating Enrichment Culture: Conversion of
Pentachtorcphenol to Phenol by Sediment Augmentation	   133
       XJaoming Zhang, W. Jack Jones, and John E. Rogers
Stimulating the Microbial Dechlorination of PCBs: Overcoming Limiting Factors	136
       John F. Quensen, HI, Stephen A. Boyd, James M. Tiedje, and John E. Rogers
Potential Surfactant Effects on the Microbial Degradation of Organic Contaminants  . .             .  139
       Stephen A.  Boyd, John F. Quensen, III, Mahmoud Mousa, Jae Woo Park,
       Shaobai Sun, and William Inskeep
Enhanced Dechlorination of PCBs in Contaminated Sediments by Addition of Single
Congeners of Chloro-and  BromobiphenyIs           	       142
       W.  Jack Jones, John E. Rogers, and Rebecca L Adams
Effect of Heavy Metal Availability and Toxicrty on Anaerobic Transformations of
Aromatic Hydrocarbons	       .  .    . .  146
       John H. Pardue, Ronald D. DeLaure, and William H. Patrick, Jr.

Biodegradation of Petroleum Hydrocarbons in Wetlands Microcosms	             149
       Rochelle Araujo, Marirosa Molina. Dave Bachoon, and Lawrence D.  LaPlante
Biodegradation of Petrolaum Hydrocarbons in Wetlands: Constraints on Natural and
Engineered Remediation	     153
       John H. Pardue, Andrew Jackson, and Ronald D. DeLaune
Anaerobic Biotransformation of Munitions Wastes	156
       Deborah J.  Roberts, Farrukh Ahmai^, Don L Crawford, and Ronald L. Crawford
Covalent  Binding of Aromatic Amines to Natural Organic Matter  Study of Reaction
Mechanisms and Development of Remediation Schemes	160
       Eric J. Weber, Dalizza Co/dn, and Michael S. Elovitz
                                          VI

-------
                                       Contents (continued)
                                                                                                 Page


        Kinetics of Anaerobic Biodegradation of Munitions Wastes   ....
               Jiayang Cheng, Makram T. Suidan, and Albert D. Venosa
                                                                                                  1 fifi
        Biodegradation of Chlorinated Solvents	
               Sergey A. Selifonov, Lisa N. Newman, Michael E. Shelton, and Lawranca P.  Wackett
        Characterization of Bacteria in a TCE Degrading Biofilter	        17°
               Alec W. Breen, Alex Rooney,  Todd Ward, John C. Loper, Rakesh Govmd, and John R. Haines

        Bioremediation of TCE: Risk Analysis for Inoculation Strategies	                 173
               Richard A. Snyder, Malcolm S. Shields, and PH. Pritchard
        Studies on the Aerobic/Anaerobic Degradation of Recalcitrant Volatile Chlorinated Chemicals
        in a Hydrogel Encapsulated Biomass Biofilter	176
               Rakesh Govind, P.S.R.V, Prasad, and Dolloff F. Bishop
        Metabolites of Oil Biodegradation and Their Toxicity	       ...        ...            178
               Peter J. Chapman, Michael E. Shelton, Simon Akxerman, Steven S. Foss,
               Douglas P. Middaugh, and  William S. Fisher
        TCE Remediation Using a Plasmid  Specifying Constitutive TCE Degradation:
        Alteration of Bacterial Strain Designs Based on Field Evaluations	       ... 179
               Malcolm S. Shields, Allison Blake, Michael Reagirr, Tracy Moody, Kenneth Qverstreet, Robert
               Campbell, Stephen C. Francesconi, and PH. Pritchard
        Degradation of a Mixture of High Molecular-Weight Polycyclic Aromatic Hydrocarbons by a
        Mycobacterium Species	        180
               /. Kelley, A. Selby, and Carl E.  Cemiglia

        Bioavailability Factors Affecting the  Aerobic Biodegradation of Hydrophobic Chemicals    .  .        181
               Pamela J. Morris, Suresh C. Rao, Simon Akkerman, Michael E. Shelton,
              Peter J. Chapman,  and PH. Pritchard

Section Six:  Hazardous Substance Research Centers	183

       In Situ Attenuation of Chlorinated Aliphatics in Glacial Alluvial Deposits	184
              MichaelJ. Barcelona, Mark A.  Henry, and Walter J. Weber, Jr.

       In Situ Bioremediation of Chlorinated Solvent Ground-Water Contamination:
       Scaling up from a Field Experiment to a Full-Scale  Demonstration   	   186
              Perry L. McCarty, Gary D. Hopkins, and Mark N. Goltz

       Bioavailability and Transformation of Highly Chlorinated Dibenzo-/>Dioxins and
       Dibenzofurans in Anaerobic Soils and Sediments   	188
              Peter Adriaens and Quingzhai Fu

       Localization of Tetrachloromethane Transformation Activity in  Shewanella Putrefaciens MR-1  .  .   .189
              Erik A. Petrovskis, Peter Adriaens, and Timothy Vogel
                                                 VII

-------
Formation and Transformation of Pesticide Degradation Products Under Various
Electron Acceptor Conditions	        .                             190
       Paige J. Novak, Gene F. Parkin, and Craig L. Just
Bioremediation of Aromatic Hydrocarbons at Seal Beach, California: Laboratory and
Field Investigations       ....      	    , .                                    19J
       Harold A. Ball, Gary D. Hopkins,  Eva Orwin, and Martin Remhard
Pneumatic Fracturing To Enhance In Situ Bioremediation  .  .                                     196
       John R. Schuring, David S. Kosson, Shankar Venkatraman, and Thomas A. Bo/and
                                         VIII

-------
    Economic ana
 Regulatory Analysis
    Environmental
   Heoitf. Services
     nical \fcnt-ng
      and Editing
     Conference
    Management
        Software
       Educator-
                    Eastern Research Group. Inc
                 DATE:
                 TO:
                 RE:
                                          -MEMORANDUM-
September 7, 1995

Fran Kremer

Jenny Helmick

Bioremediation in the Field Bulletin
Contract 68-C1-0040, Task 40
                 Enclosed is the camera-ready copy for Bioremediation in the Field Bulletin No.  12.  Also
                 enclosed is an electronic PDF (Acrobat) file for this document, as you requested.

                 Please give me a call if you have any questions about this document.
   Grapnic Design
 13 Media Services
        Training
Eaua Opportunity
      Employe'


      Pnntea on
 Recycled ?apet
                   II OHartwell Avenue • Lexington. Massachusetts 02173-3198 •  Phone (617)674-7200  • Fax (617) 674-2851
                   1925 Norm Lynr, Street. Suite 703 « Arlington. Virginia 22W-1707 • Phone (703) 841-0500 • Fax (703) 841-1440

-------
                                         Introduction
Some of the most promising new technologies for solv-
ing hazardous waste problems involve the use of biore-
mediation,  an  engineered  process  that  relies  on
microorganisms such as bactena or fungi to transform
hazardous chemicals into less-toxic or nontoxic com-
pounds.  Until recently, the use of bioremediation was
limited by the lack of a thorough understanding of biode-
gradation  processes,  their appropriate  applications,
their control  and enhancement in environmental matri-
ces, and the engineering techniques required for broad
application of the technology.

Because the U.S.  Environmental  Protection Agency
(EPA) believes that bioremediation offers an  attractive
alternative to conventional methods of cleaning up haz-
ardous waste, it has developed a strategic plan for its
acceptance  and  use by the technical and  regulatory
communities  The Agency's strategic plan is centered
on site-directed bioremediaticn research to expedite the
development and use of relevant technology.

EPA's Office of Research and Development (ORD)  de-
veloped an  integrated  Bioremediation Research Pro-
gram to advance the understanding, development, and
application  of bioremediation  solutions  to hazardous
waste problems threatening human health and the  en-
vironment  The  Bioremediation  Research Program is
made up of  three major research components:  the
Biosystems Technology Development Program, the In
Situ Application Program, and the Bioremediation Field
Initiative.
Related bioremediation studies are being earned out at
five  EPA  Hazardous  Substance  Research  Centers
(HSRCs) under the direction of ORD's Office of Exoiora-
•3ry Research (OER). EPA was authorized to estafcusn
these centers by orovistons in the 1986 amendments to
the Superfund  law calling for research into all aspects
of the "manufacture, use, transportation, disposal, and
management of hazardous substances."

EPA's bioremediation research  efforts have  produced
significant results in the laboratory, at the pilot scale, and
in the field.  The many accomplishments include aquifer
restoration, soil cleanup, process characterization, and
technology  transfer.  This symposti'm was held to pre-
sent and discuss recent developments in bioremediation
research undertaken during 1993 under the Biosystems
Technology Development Program.

In this document, abstracts of paper and poster presen-
tations from the symposium are organized  within five
key research and program areas:

• Bioremediation Feld Initiative

• Performance  Evaluation

• Reid Research

• Pilot-Scale Research

• Process Research

In the last  section of this document are abstracts of
poster presentations on  bioremediation  research per-
formed as part of the HSRC program.

-------
                                    Executive Summary
 The U.S. Environmental Protection Agency's (EPA's)
 Office of Research and Development (ORD) hosted the
 seventh annual Symposium on Bioremediation of Haz-
 ardous Wastes:  Research,  Development, and Field
 Evaluations, in San Francisco, California, June 27-29,
 1994.   The symposium was held in cooperation  with
 EPA's Region 9 offices. More than 500 people attended,
 including leading  bioremediation researchers  and  field
 personnel from federal, :tate, and local agencies as well
 as representatives from industry and academia. Three
 speakers opened the symposium with background infor-
 mation  on btoremediation research.

 Fran Kremer,  Coordinator of the Bioremediation Reid
 Initiative, provided an  introduction  and  overview of the
 Biosystems Technology Development Program (BTDP).
 The BTDP  draws on ORD  scientists who  possess
 unique skills and expertise in biodegradation, toxicology,
 engineering, modeling, biological and analytical chem-
 istry, and molecular biology. These scientists  work out
 of the following laboratories and organizations, all  of
 which are institutional  participants in the program:

 •  Environmental Research Laboratory-Ada, Oklahoma

 •  Environmental    Research   Laboratory-Athens,
  Georgia

 • Center for  Environmental  Research Information-
  Cincinnati, Ohio

 • Risk  Reduction  Engineering Laboratory-Cincinnati,
  Ohio

 • Environmental  Research Laboratory-Gulf  Breeze,
  Florida

 • Health  Effects  Research  Laboratory-Research
  Triangle  Park,  North Carolina

A regional perspective on bioremediation was presented
by Jeffrey Zelikson, Director of EPA's Region 9 Hazard-
ous Waste Management Division in San  Francisco, Cali-
fornia.   According  to   Mr.   Zelikson,  accelerated
development and  use of innovative technologies are
critical to protecting the environment and ensuring the
competitiveness of U.S.  industry  both at home  and
abroad.  Although  research is the key,  from a regional
perspective, two other factors are necessary for s
cess. These factors are diffusion of information or "i
ting the word  out* and community  acceptance.
recent legislative efforts were designed to address tti
issues: the Environmental Technology Initiative, wh
promotes the use of bioremediation, and the Superfi
Reform Act, which expands the role  of communities
decisionmaking at  hazardous waste sites across
country.
Robert Menzer, Director of the EPA's  Environmei
Research Laboratory in Gulf Breeze.  Florida, aiscus:
ORD's Bioremediation Program. ORD has teamed
with the Department of Defense (DOD) and the Oepi
ment of Energy (DOE) to form the Strategic Reseai
and Development Program, which provides funds
bioremediation projects in the field.  Part of this efl
involves setting up the former Wurtsmith Air Force &t
(AFB) near Lake Huron in Michigan as a national cer
for testing btoremediation research and  developme
An industry group, the Bioremediation Technologies
velopment Forum, has already committed itself to a
ducting field test work at the Wurtsmith site.

The 41 papers delivered at  the conference highligfi
recent  program achievements  and  research proje
aimed at bringing bioremediation into more widespre
use.  Taken as a whole, these topic  areas represer
comprehensive approach to bioremediation of haza
ous waste sites. The presentations weie organized ii
five key research and program areas:

1.  Bioremediation Field Initiative.   This  initiative w
   instituted in 1990 to collect and oisseminate perfoi
   ance data on bioremediation techniques from fi
   application experiences.  The Agency assists the
   gions  and states  in conducting  field tests and
   carrying out independent evaluations  of site  de<
   ups using bioremediation.   Through  this initial*
   tests are under way at Superfund sites, Resoui
   Conservation and Recovery Act corrective action
   cilities, and Underground Storage Tank sites. B(
   paper   presentations were  devoted   to this *
   program  area, covering field  evaluations it si
   using   bioventing,  biochemical   techniques, a

-------
   bioremediation under a variety of aerobic and an-
   aerobic conditions.

2.  Performance Evaluation  Performance evaluation
   of various bioremediation  technologies  involves
   assessing ttie extent and rate of cleanup for particu-
   lar bioremediation methods as well as monitoring the
   environmental  fate and effects of compounds and
   •their by-products. Because attempts to remediate a
   contaminated site can result in  the production  of
   additional compounds, an important aspect of per-
   formance evaluation involves assessing the potential
   health effects of processes.  Two papers were pre-
   sented concerning EPA's Health Effects Research
   Laboratory (HERL) and its integrated program devel-
   oped to address the  risk of potential health effects
   and to  identify bioremediation approaches that best
   protect public health.
3.  FiekJ Research.  Once a bioremediation approach
   has  proven  effective in a laboratory or pilot-scale
   treatability study, it must be monitored and evaluated
   at a field site. The objective of this level of research
   is to demonstrate that the particular bioremediation
   process performs as expected in the field.  For most
   bioremediation technologies, certain key factors con-
   cerning applicability (e.g., cost effectiveness) cannot
   be thoroughly evaluated until the approach is scaled
   up and field tested.  Four paper and several poster
   presentations provided information on recent or ongo-
   ing field research.

4.  Pilot-Scale Research.  Pilot-scale research provides
   information on the operation and control of bioreme-
   diation  technologies and the management of proc-
   ess-related residuals  end emissions. As such, it is
   a necessary step in anticipation of full-scale applica-
   tion of a technology.  Grven the expanding base of
   expenence with vanous bioremediation methods, the
   need for pilot-scale research is increasing.  Six pa-
   pers and numerous posters were  presented  con-
   cerning research based on microcosms of field sites.

5.  Process Research.  Process research involves iso-
   lating and identifying microorganism^ that carry cut
   biodegradation  processes  and  the environmental
   factors affecting these  processes. Such -esearch is
   fundamental to the de\ jlopment cf new  biosystems
   for treatment of environmental pollutan's in surface
   waters, sediments, soils, and subsurfp.ee materials.
   Twenty-one papers and numerous poster presenta-
   tions addressed this critical area.
In  addition to presentations on research being carried
out under the STOP, the  symposium's poster session
included presentations from the five EPA  Hazardous
Substance Research Centers (HSRC). The scientists
and engineers involved in the latter program conduct
EPA research sponsored by the following centers:
• Northeast  Hazardous Substance Research Center
  (Regions 1 and 2)
• Great Lakes and Mid-Atlantic Hazardous Substance
  Research Center  (Regions 3 and 5)
• South/Southwest  Hazardous  Substance Research
  Center (Regions 4 and  6)

• Great Plains and Rocky Mountain Hazardous Sub-
  stance Research Center (Regions 7  and 8)
• Western  Region  Hazardous  Substance Research
  Center (Regions 9 and  10)

-------
                           Section One
              Bioremediation Field  Initiative
The Bioremediation  Reid Initiative is one of the major components  of  EPA's
Bioremediation Research Program. The  Initiative was undertaken in 1990 to
expand the nation's field experience in bioremediation techniques. The Initiative's
goals are to more fully assess and document the performance of full-scale biore-
mediation applications, to create a database of current field data on the treatability
of contaminants, and to assist regional and state site managers using or consider-
ing bioremediation. The Initiative is currently tracking bioremediation activities at
more than 150 Superfund sites, RCRAcorrective action facilities, and Underground
Storage Tank sites nationwide, and will soon expand its database to include sites
under private sector jurisdiction  and international sites. Performance evaluations
currently are being conducteo at nine sites, six of which were reported on at  this
symposium.
Investigations at the St. Joseph, Michigan, National Priority List (NPL) site revealed
that natural  anaerobic degradation of trichloroethylene (TCE) contamination was
occurring in ground water at the site. Later sampling was performed to estimate the
contaminant mass flux,  and to  estimate apparent degradation constants. Other
studies  also were performed to determine  whether enhancement of the anaerobic
process might be beneficial,  what microorganisms are responsible for the natural
transformation, and what is  an  effective primary substrate to add to the  ground
water for enhancing the remediation in situ.

Other o.-igoing evaluations include a 3-year field investigation, which began in 1991,
of the use of bioventing to remediate jet fuel spills at two Air Force Base (AFB)
sites. At the Eielsen  AFB near Fairbanks, Alaska,  studies were  performed to
demonstrate bioventing in a cold climate and to evaluate several low-intensity soil
wanning methods. At the Hill AFB site near Salt Lake  City, Utah, studies were
performed to investigate bioventing in deep vadose zone soils and to determine
the influence of air flow rate on the biodegradation and volatilization rates of organic
contaminants.

Research continued on the Reilly Tar and Chemical Corporation site in St. Louis
Park, Minnesota, as part of a 3-year evaluation program that began in November
1992. The research is designed to evaluate the potential of bioventing to remediate
soils contaminated with wood preservatives. In situ bioremediation of a pipeline spill in
Park City, Kansas, using nitrate as an electron acceptor also is being investigated.

At the Libby Ground-Water Site in  Libby,  Montana, performance evaluation  has
been completed of full-scale bioremediation of creosote wastes in  ground water
and soils. This evaluation addressed three separate biological treatment processes:
1) surface soil  bioremediation in  a prepared-bed, lined treatment unit; 2) treatment
of extracted ground water from the upper aquifer in an aboveground  fixed-film
bioreactor and 3) in situ bioremediation of the upper aquifer. These three processes
represent a treatment train approach to site decontamination, where each process

-------
was chosen for remediation of a specific: phase (i.e., soil, oil, and water). Published
results of the study will be available from EPA later this year.
Finally, a demonstration study of bioaugmentation of soil contaminated with pen-
tachlorophenoi (PCP) using  selected strains of lignin-degrading fungi was per-
formed at an abandoned wood treating site in Brookhaven, Mississippi.

The symposium's poster session included information on the Bioremediation in the
Reid Search System (BFSS), a PC-based software application developed by EPA's
Bioremediation Field Initiative. BFSS provides access to a database of information
compiled by the Initiative on haiardous waste sites where bioremediation is being
tested or implemented, or has been completed.

-------
           Intrinsic Bioremediation of TCE in Ground Water at an NPL Site in
                                      St. Joseph, Michigan
                          John T. Wilson, James W. Weaver, and Don H. Kampbell
           Robert S. Kerr Environmental Research Laboratory, U.S. Environmental Protection Agency,
                                               Ada. OK
 The ground water at the St. Joseph, Michigan, National
 Priority List (NPL) site is contaminated with chlorinated
 aliphatic compounds (CACs) at concentrations in the
 range of 10 mg/L  to  100 mg/L.  The chemicals are
 thought to have entered the shallow sandy aquifer either
 through waste lagoons, which were used from 1968 to
 1976, or through disposal of trichloroethylene (TCE) into
 dry wells at the site (1). The contamination was deter-
 mined to be divided into eastern and western plumes,
 as the suspected sources were situated over a ground-
 water divide. Both plumea were found to  contain TCE.
 cis- and trans-1,2-dichloroethylene (C-1.2-DCE and  t-
 1,2-DCE),  1,1-dichtoroethylene (1,1-DCE), and vinyl
 chloride (VC).

 Previous investigation of the site indicated that natural
 anaerobic  degradation  of the TCE was occurring
because of the presence of transformation products and
significant levels of ethene and methane  (2,3).   The
purpose of this presentation is to provide the results of
later samplif :g of the western plume near Lake Michigan,
to estimate the contaminant mass flux, and to estimate
apparent degradation constants. The estimates are
based on  visualization of the data  representing  each
measured concentration by  a zone of influence based
on the sample spacing. The presentation of the data is
free from artifacts  of interpolation,  and extrapolation
of the  data  beyond the measurement locations is
controlled.

Data Summary
In 1991  three transects (1. 2, and 3 on Figure 1)  were
completed near the source  of the western plume (2).
            T|.<«  ShMM Augw Bortng

             *•   a/2i»7
Flgur* 1. SL Joseph, Michigan, NPL sit* plan.

-------
The three transects consisted of 17 borings with a slot-
ted auger. In 1992, two additional transects (4 and 5 on
Figure 1) were  completed consisting of 9 additional
slotted auger borings.  In oach bonng, water samples
were  taken on  roughly 1.5 m  (5 ft) depth intervals.
Onsite gas chromatography was performed to deter-
mine  the width  of the  plume and to find the point of
highest concentration. Three of the transects (2, 4, and
5)  were  roughly  perpendicular  to  the  contaminant
plume. Of the remaining transects, transect 1 crosses
the plume at an angle and transect 3 lies along the
length of the plume. The perpendicular transects form
logical units for study of TCE biotransformation.

The site  data from the  transects are visualized  as sets
of blocks centered around the measurement point. The
blocks are defined so that the influence of a particular
measured concentration extends halfway  to the next
measurement location  both horizontally and vertically.
Thus, the presentation  of the data is simple and direct.
The visualization of the data is performed on a Silicon
Graphics Indigo workstation using a two-dimensional
version of the fully three-dimensional field-data analy-
sis program called SITE-3D, which is under develop-
ment at  the Robert S. Kerr Environmental Researcn
Laboratory.

The mass of each chemical per unit thickness ard the
advective mass flux of each chemical are calculated by
summing over the blocks.  By following this procedure,
the measured chemical concentrations are not extrapo-
lated into the day layer under the site. Neither are they
extrapolated beyond a short distance from the measure-
ment locations (5 ft vertically and  50 to 100 ft horizon-
tally).  Other interpolation schemes, such as  inverse
distance  weighting or  kriging, also  could  be used to
estimate the concentration field and perform the mass
estimates. Figures 2 and 3 show the distributions of VC
and TCE at transect 5  using a logarithmic,  black-and-
white "cotor* scale. Notably, the maximum VC concen-
tration at transect 4 was 1,660 uc/L and at transect 5
was 205 ng/L  Th® maximum  TCE concentration at
transect 4 was 8,720 ug/L and at transect , was
163 u.g/L. As noted previously for other portions of the
site (2,4), the contamination is found near the bottom of
the aquifer. The  highest concentrations of VC and TCE
do not appear to be co-located. In Table ',  mass esti-
mates are presented for the perpendicular transects
ordered from furthest upgradient (transect 2) to furthest
downgradient (transect 5). The data in Table 1 represent
the mass in a volume of aquifer that has an area equal
to the cross-sectional area of the iransect and is 1.0 m
thick in the direction of ground-water flow.

Advective Mass Flux Estimates

Results from the calibrated MODFLOW model of Tiede-
man and  Gorelick (4) were used to estimate the ground-
           St Josaph. Michigan
           VkiytCNonda
           Transact 5
           Man: 0.4811 E-01 kg/m
           tS1   tS2
                           153  ISS
         Ground Surtaca
 10Faa«I
                     100F»e»
        » Appro*. N      I—)

Figure 2.  VC distribution at transact 5.
 Concentration


250.000 -i


 25 000 £


  2.500


  2500


  25008


  2.50C


 0.2500


 0.02501—I
           SL Joaapn. Michigan
           Then toroa* ana
           Transact S
           Masc 0.2821E-01 K
-------
 Tab* 1.  Mass per Unit Thlclcnesa (kg/m) at St Joaapn, Michigan
                                                         Transact
Cnamteal
vc
1.1-DCE
M.2-OCS
C-1.2-DCE
TCE
Methane
Ethan*
Ethane
TOC
Chloride
Sulfate
NOy Nitrogen
NH4- Nitrogen
TKN- Nitrogen
2
1.523
0.23-7
0.566
12.32
10.07
5.855
0.6847
no da*a
no data
129.9
37.05
2.904
1.835
2.987
1
1.8969
0.0816
0.5059
5.1127
5.5804
5.4826
0.8925
no data
no data
148.8
34.376
2.471
25609
3.8357
4
0.4868
0.01451
0.03628
1.890
1.397
4.620
0.1747
0.2085
12.63
213.1
95.78
4.421
0.45S2
0.635.'l
5
004311
0001047
0.007041
0,2832
0.02821
1 373
0.004901
0.001689
8.314
156.2
66 19
8.247
0.2256
0.3646
 using the average hydraulic conductivity result in a total
 flux of 13 kg/y of TCE, c-1,2-OCE, t-, ,2-DCE, 1,1-DCE,
 and VC at transect 5. This value contrasts with the total
 flux of these CACs of 310 kg/y at transect 2, near the
 source of contamination, a 24.4-fold decrease in mass
 flux of CACs  across the  site. Given  the 95 percent
 confidence limits on the  hydraulic  conductivity  deter-
 mined by Tiedeman and Gorelick (4), the total range of
 mass flux of these five chemicals is from 205 kg/y to 420
 kg/y at transect 2 and from 8.4 kg/y to 17 kg/y at transect
 5. The range of fluxes at transect 5 is an upper  bound
 on, and the best estimate of, the flux into Lake Michigan.
Apparent Degradation Constants

The mass per unit thickness of TCE at transects 2, 4,
and 5 was used to estimate apparent first-order degra-
dation constants.  The constants are estimated by ap-
plying the first order rate equation
                  In
0)
to the site data, where q is the average concentration in
the transect j, GJ+, is  the average concentration in the
downgrr-^ent transect j+1, t is the advective travel time
for TCE to move between the  transects, and X is the
apparent degradation  constant. The mass per unit thick-
ness data for TCE and the cross-sectional area were
used to determine the average concentrations q and ci+1
in the up- and downgradient transects. The porosity,
bulk density, fraction  organic carbon, organic carbon
      partition coefficient (5), ground-water gradient, and dis-
      tance between the transects were used to determine the
      advective travel times. The values used in  Equation 1
      are given in Table 3. From these quantities, the apparent
      degradation constant for  TCE  was  determined to be
      -0.0076/week from transect 2 to 4 and -0.024/week from
      transect 4 to 5.
References

1. Engineering Science, Inc. 1990. Remedial investiga-
   tion and feasibility  study,  St. Joseph,  Michigan,
   phase I technical memorandum. Liverpool, NY.

2. Kitanidis, P.K., L. Semprini, D.H. Kampbell, and J.T.
   Wilson.  1993. Natural  anaerobic  bioremediation of
   TCE at  the St.  Joseph, Michigan, Superfund site.
   Symposium   on  Bioremediation  of   Hazardous
   Wastes:  Research,  Development, and Field Evalu-
   ations. EPA/60Q/R-93/054. pp. 57-60.

3. McCarty, P.L, and J.T. Wilson. 1992. Natural anaero-
   bic  treatment of a TCE plume at the St.  loseph,
   Michigan, NPL site. In: U.S. EPA. Bioremediation of
   hazardous wastes (abstracts). EPA/600/R-92/126.
   pp.  47-50.

4. Tiedeman, C., and  S. Gorelick.  1993. Analysis of
   uncertainty in optimal groundwater contaminant cap-
   ture design. Water Resour. Res. 29(7):2139-2153.

5. U.S. EPA. 1990. Subsurface remediation guidance
   table 3. EPA/540/2-90/011b.

-------
Table 2. Mass Flux (kg/y) «t 31 Joseph, Michigan
Chemteal
VC
1.1-OCE'
M.2-OCE
C-1.2-OCE
TCE
Methane
Ethane
Ethane
TOC
CWonoe
SuHata
NOj- Nitrogen
NrV Nitrogen
TXN- Nitrogen
Table 3. CnemfcaJ and

2
18.81
2.934
6.995
152.1
131.7
72.29
8.453
no data.
no data
1.604
457.4
3S.3S
22.66
36.38
Hydraulic Valu«a U*«
Araa with Ma*a Pe-
Non-xare Unit
TCE Thickn««a
Concentration from STTE-30
Trtnaeet (m2) (kgM»)
2 1492

4 2.774

5 1.943
.0.67

1.397

0.0282
















1 In Estimating
Avvraga TCE
Concentra-
tion In tn«
Trana«ct
(kg/nT>)
aand c,., In
Equation 1
6.70a-3

5.040-4

1.440-5
Transact
1 4
36.03 10.69
1 551 0.3185
9.609 0.7963
97.11 41.48
106.0 30.67
104.1 101.4
16.95 3.836
no data 4 577
no data 277.2
2.826 4.678
652.9 2.102
*6.93 97.05
•^.64 10.01
72.6.1 13.95
Apparent Degradation Rataa
Gradient
Estimated
from 'Retarded
Distance Tledeman Seepage
Between and Qorellck Velocity tor
Transects (m) (1993) TCE (m/d)
— — —
260 7.3»-3 0.11
— _ _
160 1.10-2 0.156
_ _ _

5
1 676
0.03648
0.2453
9868
0.9829
4786
0 1708
0.05885
289.7
5.444
2.306
287>
7.861
12.70

Estimated
Travel Time
Between
Transects
(weeks)
At in
Equation 1
—
340
—
145
—
•Constants used in seepage velocity calculation:
 Hydraute conductfvrty: 7.5 m/d
 Retardation factor for TCE:  1.78 • 1  +
 Porosity 9. :  0.30
 Bulk density p»:   1.86 g/crrr1
 KOC:  126 mUg
 foe; 0.001
                                                                10

-------
              Enhanced Reductive Dechlorination of Chlorinated Ethenes
               Zachary C. Hasten, Pramod K. Sharma, Jamos N.P. Black, and Perr/ L McCarty
           Western Region Hazardous Substance Research Center, Stanford University, Stanford, CA
 Reductive dehatogenation of trichloroethylene (TCE) to
 cis-1,2-dichloroethylene  (C-1.2-DCE),  tra',s-1,2-dichlo-
 roethylene (M.2-DCE), vinyl chloride (VC), and ethene
 was found to be occurring at a site in St Joseph, Michi-
 gan, by indigenous microbial populations under anaero-
 bic conditions (1). This has raised two possibilities for
 further study: 1) that the natural anaerobic processes at
 the site may be sufficient to bring about site remediation
 alone; or 2) that the natural process will be incomplete
 without  some enhancement  Further  site charac-
 terization is now under way by the EPA Robert S. Kerr
 Environmental Research Laboratory to  determine the
 extent of natural onsite transformation. This study aims
 to determine whether enhancement of  the  anaerobic
 process might be beneficial, what microorganisms are
 responsible for the natural transformation, and what is
 an effective primary substrate to add to the ground water
 for enhancing  the remediation in situ.  For comparison,
 aquifer  material from a site in Victoria,  Texas, also is
 tsing evaluated. This site is contaminated by tetrachlo-
 roethylene  (perchloroethylene, or PCE)  and is being
 actively bioremediated by the addition of benzoate and
 sulfata (2).

 Methods

 Aquifer material for this study was obtained aseptically
 in the absence of  oxygen from both  Si Joseph  and
 Victoria sites The potential of the  St. Joseph aquifer
 material for TCE transformation and the effect of adding
 different primary substrates were studied using 25 mL
test tubes as small laboratory columns (3).  The fluid
within the test tubes was exchanged after incubation
periods ranging from 1 to 4 months  with  filter-sterilized
site ground  water that was amended with a primary-
substrate and  TCE. Control columns received TCE-
amended, filter-sterilizea ground water without an added
primary substrate.  Between fluid exchanges,  the open-
ings were sealed, and the columns were incubated with-
out fluid exchange in  a  room  temperature anaerobic
glovebox containing 1 to 10 percent hydrogen. Each
primary  substrate was fed to yield  100 mg/L chemical
oxygen  demand (COD) to provide similar reducing
equivalents for each column. Each column was fed only
one substrate from the time the column was prepared.

In addition, microcosms consisting  of 125  mL bottles
containing aquifer material and site ground water were
used to simulate in situ conditions with the Victoria
aquifer material. Only 110 mL of saturated aquifer ma-
terial was used in the bottles to allow for sampling of the
liquid from the remaining 15 mL,  and to provide for bed
fluidization during mixing. These microcosms were incu-
bated without headspace.
Enrichments were developed by the addition of Victoria
aquifer-material to a basal medium (4). This enrichment
was  subsequentty  transferred to aquifer-matorial-free
media. The effect of different metabolic inhibitors was
studied using an inoculum from a benzoate enrichment
culture into 160 mL bottles filled with 120 mL of defined
media amended with PCE, benzoate, yeast extract, and
the respective inhibitor.

Results

The possibility of enhancing biodegradation by the ad-
dition of various primary substrates was studied using
columns of Si Joseph material.  Table 1 shows the re-
sulting  concentrations  of TCE dechlorination products
after a  typical 6-week incubation period. Following this
exchange, the ethanol-fed column was switched to ben-
zoate and  immediately performed similar to the column
that had been fed benzoate from the start.

Of the primary substrates tested, benzoate addition con-
sistently stimulated the most complete dechlorination.
Similar results were obtained with the microcosms con-
taining  Victoria aquifer  material  (data not shown). No
significant  lag time before the onset of dechlorination
was observed with either material.

In the St. Joseph unfed column control, partial dechlori-
nation of TCE to c-1,2-DCE was  observed over several
exchanges spanning several months.  This dechlorina-
tion may have been associated with oxidation of natural
                                                  11

-------
Tabte 1.  Concentration of TCE Oectilorlnetlon Products after 6 Weeks of Incubation In SL Joseph Aquifer Material Columns*

                                       Compounds Remaining after 6 Weeks of Incubation i

Substrate
                       TCE
                                     cOCE
                                                 1,1-DCE
                                                              vc
                                                                          Ethene
                                                                                        Sum
f.ooe
Benzoate
Uctate
Sucrose
Ethand
Metfianol
20.5
0
0.5
2.4
3.6
9.3
4.3
0
4.1
5.4
3.7
6.3
0
0
0
0.7
0.6
0.7
0
11.8
13.5
16.1
14.1
7.9
0
14.4
5.8
5.1
2.4
1.6
25.3
25.0
23.9
29.7
24.4
25.8
 Acetate
                       10.4
                                      3.4
                                                   0.9
                                                               7.1
                                                                            1.9
                                                                                          23.7
 'X-1.2-OCE wu also present in some columns in trace amounts.

 organics within the aquifer material c: of hydrogen that
 diffused  into the column from the glovebox gases. Vic-
 toria microcosms  aiso showed some dechlorination of
 PCE to TCE in the unfed controls.

 For column studies with SL Joseph material, site ground
 water was used that included 0.49 mM  nitrate and 0.50
 mM sutfate. During incubation in the substrate-amended
 columns, nitrate and sutfate were consumed completely,
 and varying amounts of methane were produced. Nitrate
 also disappeared in the unfed control, but no sutfate was
 consumed  or methane  produced. Dechlorination  ac-
 counted for less than 2 percent of the substrate utilized;
 nitrate reduction, sulfate reuuction, and methanogene-
 sis accounted for the rest.

 After several exchanges, the primary substrate-fed col-
 umns became dogged. Small entrapped bubbles were
 visible in the columns as well as a noticeable amount of
 black precipitate.  Considering the amount of primary
 substrate added to the columns, up to about a fifth of the
 pore volume could have been filled by methane forma-
 tion. The extent of the clogging caused by iron sulfide
 precipitate or  biomass  is  unknown, but  after a few
 months,  during which the  columns sat  unfed, the  en-
 trapped bubbles visibly decreased and the  columns be-
 came undogged.  Bubbles also formed  in the Victoria
 microcosms, but they were allowed to come to the sur-
 face during daily  shaking  and were removed during
 analysis.

 PCE was not dechlorinated within 2 months in micro-
 cosms containing a defined mineral media  amended
with only benzoate, while the addition of benzoate and
0.05 percent yeast extract  stimulated dechtorination of
all the PCE completely to ethene (data not shown). The
addition  of benzoate and sulfate stimulated  partial
dechlorination, as did the addition of yeast extract alone.
Studies of the effects of various metabolic inhibitors
were conducted to better understand the role of sul-
fate-reducing and methanogenic bacteria. Table 2 lists
duplicate live bottles from a 3-morth incubation  with
0.416 mM benzoate, 0.01  percent yeast extract, and
various amendments, including 2 mM sulfate, 0.5 mM
bromcethanesulfonic   add  (BESA),  and  0.5  mM
molybdat3, where applicable. No dechlorination was ob-
served in uninoculated or sterile controls, t-1,2-DCE and
1,1-dichloroethylene were not observed  in the enrich-
ment cultures.
Summary and Conclusions

Studies with aquifer material from both contaminated
sites* have shown that all primary substrates tested were
capable of stimulating  dechlorination of some PCE or
TCE to ethene, with benzoate consistently stimulating
the most complete degradation. High sulfate concentra-
tions  appear to  inhibit dechlorination,  although  no
dechlorination was observed in microcosms incubated
without some sulfate or yeast extract. The addition of
molybdate reversed su.fate inhibition, but here dechlori-
nation stopped at c-1,2-DCE. These data show that the
anaercbic dechlorination of PCE or TCE to ethene car
be enhanced by the appropriate addition of a primary
substrate and yeast extract or sulfate.
References

1.  McCarty, PL, and J.T. Wilson. 1992. Natural anaero-
   bic treatment of a TCE plume, St. Joseph, Michigan,
   NPL site. In: U.S. EPA. Bioremediation of hazardous
   wastes (abstracts). EPA/600/R-92/126.  Cincinnati
   OH. pp. 47-5C.
                                                  12

-------
Tibte 2.  Effect* of Inhibitor* on Dechlorlnation*
                                                Remaining In Duplicate Bottles after Incubation
Amendments
B«nzoate and
Yeast Extract
Benzoate, Yeast
Extract and
8ESA
Banzoate, Yeast
Extract and
Molybdate
Benzoate, Yeast
Extract and
Sulfata
Benzoata, Yeast
Extract Sulfate.
and Molytxlate
Benzoate, Yeast
Extract Sulfata,
and BESA
PCE
0.00
0.00
0.00
0.00

0.05
0.01

1.06
1.0Y

0.00
0.00

0.37
0.96

res
0.00
0.00
0.00
0.00

0.06
0.01

0.30
0.31

0.00
0.00

0.33
0.35

cDCE
0.00
0.00
0.00
0.00

1.52
1.65

0.13
0.13

1.78
1.65

0.24
0.24

VC
0.00
0.00
0.00
0.00

0.00
0.00

0.00
0.00

0.00
0.00

0.00
0.00

Htr-ene
1.70
1 76
1 63
1.62

0.00
0.00

0.00
0.00

0.00
0.00

0.00
0.00

Sum
1.70
1 76
1.63
1.62

1.63
1.67

1.51
1.50

1.78
1.65

1.54
1.55

      for PCE ar.* its dechlorinatlon from duplicate cultures .ncubated for 3 months at room temperature.
2. Beeman,  R.E.  1994.  In  situ  biodegradation  of
   ground-water   contaminants.  U.S.   Patent  No.
   5,277,815.
3. Siegrist, H., and P.L. McCarty. 1987. Column meth-
   odologies for determining rorption and biotransfor-
   mation potential of chlorinated aliphatic compounds
   in aquifers. J. Contam. Hydrol. 2:31-50.

4. Tanner,  R.S./and R.S.  Wolfe. 1988. Nutritional re-
   quirements of Methanomicrobium mooile. Appl. En-
   viron. Microbiol. 54:625-628.
                                                   13

-------
                                 Bloventing of Jet Fuel Spills I:
        Bioventing In a Cold Climate with Soil Warming at Eielson AFB, Alaska
                                Gregory D. Sayles and Richard C. Brenner
         U.S. Environmental Protection Agency, Risk Reduction Engineering Laboretory, Cincinnati, OH

                                  Robert E. Hinchee and Andrea Leeson
                        Battelle Memorial Institute, Columbus Division, Columbus, OH

                                           Catherine  M. Vogel
                      U.S. Air Force.-Armstrong Laboratories, Tyndall Air Force Base, FL

                                              Ross N. Miller
                U.S. Air Force, Center for Environmental Excellence, Brooks Air Force Base, TX
 Bioventing is a process that supples oxygen in situ to
 oxygen deprived soil microbes by forcing air through
 unsaturated contaminated  soil  at  low  flow rates (1).
 Unlike soil venting or soil vacuum extraction technolo-
 gies, bioventing attempts  to stimulate  biodegradative
 activity while minimizing stripping of volatile organics,
 thereby destroying the toxic compounds in the ground.
 Previous work (2) has demonstrated that biodegradation
 rates associated with bioventing are temperature de-
 pendent. Briefly, the goal of the current study is to dem-
 onstrate bioventing in  a cold climate and to evaluate
 several low-intensity soil wanning methods ior the ability
 to maintain greater than average soil temperatures and
 rates of biodegradation.

 The EPA Risk Reduction Engineering Laboratory, with
 resources from  EK-"s Bioremediation  Field Initiative,
 began a 3-year  field stucty of in situ bioventing in the
 summer of 1991  in collaboration with the U.S. Air Force
 at Eielson Air Force Base (AFB) near Fairbanks, Alaska.
 The site has JP-4 jet fuel contaminated unsatuiated soil
 where a spill ha'  occurred in association with a fuel
 distribution network. The contractor operating the pro-
ject is Battolle Memorial Institute, Columbus, Ohio. This
 report summarizes tt^e first 21/fc years of operation.


 Methodology

 Site history, characterization, installation, and monitor-
 ing were summarized previously (3,4,5).  Figure 1 shows
a plan view of the project.
Briefly, four 50 ft x 50 ft test plots have been established,
all receiving relatively uniform injection of air. The four
test plots are being used to evaluate three soil warming
methods:

• Passive warming:  Enhanced solar warming in late
  spring, summer,  and early fall using a clear plastic
  covering over the plot; and passive heat retention the
  remainder of the year by applying insulation to the
  surface of the plot.

• Active wanning:  Warming by applying hea;ed water
  from soaker hoses 2 ft below the surface. Water is
  applied at roughly 35°C and at an  overall rate to the
  plot of roughly 1 gal/min. Five parallel hoses 10 ft
  apart deliver the warm water. The surface is covered
  with  insulation year-round.

• Buried heat tape warming:  Warming  by heat tape
  buried at a depth of 3 ft and distributed throughout
  the plot 5 ft apart.  The tape heats at a rate of 6
  W/ft, giving a total heat load to the plot of roughly
  1 W/ft2.

The contaminated control consists of contaminated soil
vented with injected  air with no artificial method of heating.

The passively heated,  actively heated, and control test
plots were installed  in the summer of 1991, and the heat
tape plot was installed in September 1992. Air injec-
tion/withdrawal  wells and soil  gas  and temperature
monitoring points are  distributed throughout the site.
(See Figure 1.) Heating of the actively heated plot was
                                                  14

-------
         \
              N
                                               Taxiway
                                            II
         O    t-l  O
                          11 O   *-•
                                 Active
                                Warming
                                 System
                                                                                     Taxiway
                                                                                0    25'   50'
                                                                                    Scale
                                                                             Qround Water Monitoring Wall
                                                                             Air iniectlon/WtttidrawaJ Wed
                                                                             Three-level Sod Gas Probe
                                                                             Three-level Thermocouple Proo«
                                                                             Air IniecttorvWitharawai WeN
                                                                              (currentty not in use)


 Figure 1.  Plan view of the EPA/U.S, Air Force Moventing system at Eielson AFB near Fairbanks, Alaska. "S' reoresents u,r^-n,,.,
          soil  gas monitoring points, T" represents three-level temperature probes, and -o" and "." represent Inactive and active
          air Injection wells, respectively. Instrumentation In the lower left Is the uncontamlnated background location
 discontinued in July 1993 to compare heated and un-
 heated biodegradation rates at the same location.

 Periodically, in situ respirometry tests (6) are conducted
 to measure in situ oxygen uptake  rates by the microor-
 ganisms. These tests allow estimation of the biodegra-
 dation rate as a function of time  and, therefore, as a
 function of ambient temperature and soil warming tech-
 nique. The rate of oxygen use can be converted into the
 rate of  petroleum use by assuming  a  stoichiomet"' of
 biodegradation (4). Quarterly comprehensive and monthly
abbreviated in situ respiration tests were conducts-1

Final soil hydrocarbon analyses will be conducted in
the summer of 1994 and  compared  with  initial soil
analyses to document actual hydrocarbon loss due to
bioventing.
 Results

 Evaluation of Soil Warming Methods

 Figure 2 displrvs the average temperature of each plot
 and at  an uncontaminated background location  as a
 function of  time during  the  study. By applying warm
 water to the plot the temperature of the actively heated
 plot was maintained in the range of 10°C to 25°C,  com-
 pared  with  the contaminated  (unheated) control plot
 where the minimum winter temperature is roughly 0°C.
 When heating of the actively heated plot was terminated
 in July  1993, its temperature followed the temperature
 of the unheated control plot closely, as expected.

The ability to control temperature in the passive!-  heated
plot was not  as successful. The temperature of the
                                                    15

-------
                                                         10
        1991
                     1992
                                       1993
 Rgwra 2.  Av*raga tampafarum of aach plot and «t an uncoo-
          taminatad background location at  th« EMaon AFB
          btovantlng site aa a function at tlma during tha study.
 passively heated  plot roughly mimicked the contami-
 nated control plot temperature except during the sum-
 mer  of  1992, when the passively heated  plot  was
 roughly 5°C warmer than the control plot The insulation
 applied during the winder has been marginally success-
 ful at best providing  1°C to 2°C temperature elevation
 in the passively heated plot  relative to the control plot

 Heating  by buried heat tape in the surface heated plot
 has been successful at maintaining temperatures be-
 tween  10°C and  22°C  year-round.  Temperatures
 achieved in this plot in the summer were much higher
 than  those maintained in the winter because, although
 the heat input was constant the ambient temperature
 was much higher in the summer.


 Rate of Blodegradation

 The rate of jet fuel biodegradation, estimated by in situ
 respirometry tests, as a function of time for each plot is
 shown in Figure 3. The influence of temperature on the
 rate is clear  the actively warmed and surface warmed
 ptots  maintained rates two to three times greater than
 the unheated control plot year-round. The small differ-
 ence  in temperature between the passively warmed and
 the control plots (see Figure 2) is reflected in the small
 difference in respective rates measured in these plots.

 Researchers commonly believed that bioremediation
 systems  should be shut down for the winter in any cold
 climate because microbial  activity  is thought to ap-
 proach zero at these  low temperatures. The rate was
 nonzero  (roughly 0.5  mg/kg/day), however, in the un-
 heated control plot in the middle of winter in Alaska,
when  the average temperature of the plot was roughly
0°C (see Figure 2).

After July 1993,  when heating of the actively warmed
plot was discontinued, the rate observed in this plot was
not significantly different than the late measured from
                                                       a
                                                                    * Acnv» Aar
                                                                    • Contaminated Control
                                                                    • Surtsc* Warming
                                                             Oa. ii
                                                                           Auouit  *&f.,[ . *«L_
                                                            1991
                                                                          1992
                                                                                             1993
 Flgura 3. Avaraga rata of |«t fuel blodagradauon of rach plot
         at th« Eialaon AF9 bioventing sita. aa m«asur»d by
         In situ rasplromatry, a* i function of tbna during th«
         •tudy.
 the  unheated control plot, consistent with the similar
 temperatures of these two plots.

 Conclusions

 Application of warm water and heat generated by elec-
 trical resistance  has been  successful at maintaining
 summer-like  temperatures in the soil year-round. The
 enhanced temperatures in the plots provided elevated
 rates of biodegradation. The passively warmed plot has
 performed only marginally better than no heating ;the
 contaminated control) with respect to temperature and
 rate.

 At the conclusion of this study, a cost-benefit analysis
 will  be conducted to compare the performance of the
 heating methods in terms of rate enhancement versus
 cost of heating.

 References

 1. Hoeppel, R.E., R.E. Hinchee, and M.F. Arthur. 1991.
   Bioventing soils contaminated wrth petroleum hydro-
   carbons. J. Indust Microbiol. 8:141-146.

 2. Miller, R.N., R.E. Hinchee, and C.M. Vogel.  1991. A
   field-scale investigation  of petroleum hydrocarbon
   biodegradation in the vatfose zone enhanced by soil
   venting at Tyndall AFB, Florida.  In:  Hinchee, R.E.,
   and  R.F.  Olfenbuttel, eds.  In situ bioreclamation.
   Boston, MA:  Butterworth-Heinemann. pp. 283-302.

 3. Sayles, G.D., R.C. Brenner, R.E. Hinchee, C.M. Vo-
   gel, and R.N. Miller. 1992. Optimizing bioventing in
   shallow vadose zones and cold climates:   Eielson
   AFB bioremediation of a  JP-4 spill. In:  U.S. EPA.
   Symposium on bioremediation of hazardous wastes
   (abstracts).  EPA/600/R-92/126.   Washington  DC
   (May), pp. 31-35.

4. Leeson, A., R.E. Hinchee, J. Kittel, G. Sayles C M
   Vogel, and R.N. Miller. 1993. Optimizing bioventing
                                                   16

-------
   in shallow vadose zones and cold climatas.  Hydro-    6. Ong,  S.K.,  R.E.  Hinchee,  R.  Hoeppel,  ano
   logical Sci. 38(4):283-295.                              Schultz.  1991.  In situ  respirometry for determin
e  ^    o ,x  .  ,        r-, r- 11-  ,.     , ix-~  i /•% n       aerobic degradation rates. In:  Hinchee, R.E.,
   ?,ngL r n  *  f50"'  H 'o M  IS"',t^'?', r^'       R-F- Olfenbuttel. eds. //, s/ft; bIOreclamat,on. Bo I
   Vbgal, GD. Sayles and R.N. M, ler. 1W4 Cold d,-       ^  Butterworttl.He,nemann. pp. 541-545.
   mate applications of biovenfing.  In: Hinchee, R.E.,
   etaJ., eds. Hydrocarbon bioremediation. CRC Pr.iss.
   pp. 444-453.
                                                17

-------
                                 Bloventing 'of Jet Fuel Spills II:
                   Bloventing in a Deep  Vadose Zone at Hill AFB, Utah
                                Gregory D. Sayles and Richard C. Brenner
         U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH

                                            Robert E. Hinchee
                        Battelle Memorial Institute, Columbus Division, Columbus, OH

                                              Robert Elliott
                                          Hill Air Force Base, UT
Bioventing is a process that supplies oxygen in situ to
oxygen deprived  soil microbes by forcing air through
unsaturated contaminated  soil  at  low flow rates (1).
Unlike soil venting or soil vacuum extraction technolo-
gies,  bioventing attempts  to stimulate biodegradative
activity while minimizing stripping of volatile organics,
thus destroying the toxic compounds in the ground.
Bioventing technology is especially valuable tor treating
contaminated soils in areas where structures and utili-
ties cannot be disturbed because bioventing equipment
(air injection/withdrawal wells, air blowers, and soil gas
monitoring wells) is relatively noninvasive.

The EPA Risk Reduction Engineering Laboratory, with
resources  from EPA's Bioremediation Field Initiative,
began a 3-year field study of in situ bioventing in the
summer of 1991 in collaboration with the U.S. Air Force
at Hill AFB near Salt Lake City, Utah. The site has JP-4
jet fuel contaminated  unsaturated soil, where  a spill
occurred in association with overfilling of an  under-
ground storage tank. The contractor operating the pro-
ject is  3attelle  Laboratories,  Columbus,  Ohio. This
report summarizes the first 2 Vfc years of the study.

The objectives of this project are to increase our under-
standing of bioventing large volumes of soil and to de-
termine the influence of air flow rate on biodegradation
and volatilization rates of the organic contaminant.

Methodology and Results

See previous reports (2,3) for additional details.

Site Description/Installation

The site is contaminated with JP-4 from depths of ap-
proximately 35 ft to perched water at roughly 95 ft. Here,
bioventing, if successful, will stimulate biodegradation of
the fuel plume under roads, underground utilities, and
buildings without disturbing these structures. A plan view
of the installation  is shown in Figure 1. The single air
injection well installed in December 1990, continuously
screened from 30 ft to 95 ft below grade, is indicated.
      Ground Water Monitoring Wall
    O CW . Soil Vapor Ctusitf W«H
      (1 S) • TPH in Ground Wat»r (mgO.) (Wl)
      A-A' m Crou-S«ctlon Trac*
Flgur» 1.  Plan view of the joint EPA and U.S. Air Force biovent-
         ing activities at Hill AFB, near Salt Lake City, Utah.
         IW la the air Injection well, and CW are cluster soil
         gas monitoring wells.
                                                   18

-------
           479O
           4770-
           4750-
           4730 - S
           4710 - ~
           4«90-
           4070-  U
[*] - Sand ««h OM) «d Ctay      0 - SenMnad Mar«l
S- S«y Sand               -JZ-.  •
Q . Sand
                                                                                         4790
                                                                                        -4770
                                                                                        -4750
                                                                                         4730
                                                                                     <•  -4710
                                                                                        -4890
                                                                                        -4670
<*»«*• SurtK*)
                                                                CW - So* 6m Ouster Wei
                                                                8«0- TPH ki Sol (mg/ho) (9/91)
                                                                        . TPH m Gnaw* Water
Figure 2.  Cross-section view of the bioventing Installation «t Hill AFB. Cross Auction follows the path AA' in Figure 1. Initial sod
         TPH concentrations measured at various depths at the wells ar* Ir jlcatad.
"CW wells  are soil gas 'duster wells," where  inde-
pendent soil gas samples can be taken at 10ft intervals
from  10  ft to 90 ft deep;  CW1  through CW3  were
installed in April 1991, CW4 through CW9 were installed
in September 19?1. A cross section of the site along
path AA' in Figure 1 is shown in Figure 2. The injection
well and the soil  gas monitoring wells are indicated.
Initial soil total petroleum hydrocarbon (TPH) concentra-
tions  measured from the locations indicated are given
also.
Air Injection

To determine the influence of air injection rate on biode-
gradation and volatilization  rates, various air injection
rates have been used during this study:

• August 1991 to October 1992 and December  1992
  to April 1993, 67 ftVmin

• October to December 1992 and April to June 1993,
  40 frVmin

• Jury to November 1993, 117 ftVmin

• November 1993 to present, 20 fr/Vmin
       Soil Gas Composition

       Monthly soil gas measurements during venting are con-
       ducted.  Soil  gas 02, COz, and total hydrocarbons are
       measured at each depth in all wells, providing a three-
       dimensional map of soil gas composition in the vadose
       zone.

       In Situ Respiration Tests

       For each flow rate used, an in situ respirometry  test (4)
       is conducted to evaluate the in situ biodegradation rate.
       Rates are measured at each soil gas monitoring loca-
       tion. Table 1  shows rates at three original well locations
       averaged over depth versus time over a 2-ye^r period.
       These wells  are close enough to the injection well that
       changes in the air injection flow rate did not significantly
       change  oxygen  levels  at these  locations (data not
       shown). Lower rates with time suggest that bioventing
       is rervioving petroleum hydrocarbons from the site at a
       significant rate.

       Operational Paradigm for Bioventing in
       Deep Vadose Zones

       Bioventing of this system appears to degrade jet fuel by
       two mechanisms: 1) providing oxygen for bioremediation
                                                  19

-------
Table 1.  Rate* of Blodegradation, Averaged over the Depth
        «vi Measured by In Situ Reaplratometry, at the
        Three Original Soil On Monitoring Walla

                       Rat* (mg/kg/day)
Wall
                1991
                              1992
                          October
                           1993
CW1

CW2

CW3
1.1

0-26

0.54
0.59

0.13

0.26
0.31

0.16

0.12
of jet fuel  contaminated soils near the injection  well
(Figure 2); and 2) transporting oxygen and volatilized jet
fuel components into the surrounding, relatively uncon-
taminated soils (Figure 2), where the organic vapors are
bkxtegraded. Other studies have demonstrated in situ
hydrocarbon vapor biodegradation (5-8). Evidence  also
exists here to support this operational paradigm. Based
on soil gas measurements averaged from August  and
September 1993 from all depths in all monitoring wells,
Figure 3 shows C02 produced versus Oj consumed as
the air  stream  passes  from  the injection  well  to the
monitoring  point. The approximately linear  relationship
indicates that oxygen is being converted stoichiometri-
calty to carbon dioxide at all locations, contaminated or
not Thus, hydrocarbon vapors are degraded as they are
transported through the  uncontaminated sc.ls.

Based on data  taken in April and September 1991, a
preliminary best-fit linear model for the rate  of oxy-
gen  uptake versus soil gas TPH  and soil TPH was
developed:

Rate(%(Vhr) »
2.5x10-5CSOJ,8MTPM(Ppmv)-i.                    (1)
5.7x1 (T8
where CwuguTPH and CKHTPH are soil gas TPH and soil
TPH concentrations, respectively.  Clearty, the soil gas
hydrocarbon vapors contribute significantly to the total
oxygen  demand. Thus, jet fuel vapor degradation is a
significant mechanism for total jet fuel removal  at Hill
AFB.  The rate function  Rate(CSOi ^ TpH,C,oii TPH)  is
plotted in Figure 4. This  model will be reassessed as
additional soil gas data are reviewed.

Soil Sampling

Final soil hydrocarbon analyses will be conducted in the
summer of 1993 and compared with initial soil analyses v
to document actual hydrocarbon loss due to bioventing.

References

1. Hoeppel, R.E.. R.E. Hinchee, and M.F. Arthur. i991.
   Bioventing soils contaminated with petroleum hydro-
   carbons. J.  Indust Microbiol. 8:141-146.
                                        10
                                                         8 -
6 -
                                                      O
                                                      O
                                                         4 -,
                                         2  •
                                                             .•    .
                                                     .•  •
                                                          10       15

                                                          Oj Consumed (%)
                                                                           20
                                                                                   25
                                      Figure 3.  CO? produced versus Oj consumed as the air stream
                                              peases from the injection well to sach soil gaa moni-
                                              toring point Data Indicate biological activity at all soil
                                              gas monitoring well locations.
                                                             fr»nv)
                                      Figure 4.  Plot of the model (Equation 1 \ the rate of oxygen us*
                                              aa a function of soil gas TPH and soil TPH levels.
                                      2. Saytes, G.D., R.C. Brenner, R.E. Hinchee, C.M. Vo-
                                        gel, and R.N. Miller. 1992. Optimizing bioventing in
                                        deep vadose zones and moderate climates: Hill AFB
                                        bioremediation of a JP-4 spill. In:  U.S. EPA. Sym-
                                        posium on bioremediation of hazardous wastes (ab-
                                        stracts). EPA/600/R-92/126. Washington, DC (May).


                                      3. Sayles, G.D., R.E. Hinchee,  R.C.  Brenner, and  R.
                                        Elliott. 1993. Documenting bioventing of  jet fuel to
                                                  20

-------
   great depths: A field study at Hill Air Force Base,
   Utah. In:  U.S. EPA. Symposium on bioremediation
   of hazardous wastes: Research, development, and
   field  evaluations  (abstracts).  EPA/600/R-93/054.
   Washington, DC (May).
4. Ong, S.K., R.E.  Hinchee,  R.  Hoeppel,  and  R.
   Schultz. 1991. In situ respirometry for determining
   aerobic degradation rates. In:  Hinchee, R.E., and
   R.F. Olfenbuttel,  eds. In situ bioreclamation. Boston,
   MA: Butterworth-Heinemann. pp. 541-545.
5. Ostendorf, D.W., and D.H. Kampbell. 1990. Bioreme-
   diated soil venting of light hydrocarbons. Haz. Waste
   Haz. Mat 7:319-334.
6. Kampbell, D.H., and J.T. Wilson. Bioventing to treat
   fuel spills from underground storage tanks. J. Haz.
   Mat 28:75-80.
7. Miller, R.N., R.E. Hinchee, and C.M. Vogel. 1991. A
   field-scale investigation  of petroleum  hydrocarbon
   biodegradation in the vadose zone enhanced by soil
   venting at Tyndall AFB, Florida. In: Kinchee, R.E,
   and  R.F. Olfenbuttel,  eds. In situ bioreclamation.
   Boston, MA: Butterworth-Heinemann. pp. 283^302.

8. Kampbell, D.H.. J.T. Wilson, and C.J. Griffin. 1992.
   Performance of bioventing at Traverse City, Michi-
   gan. In:  U.S. EPA. Symposium on bioremediation of
   hazardous wastes. EPA/600/R-92/126.  Washington,
   DC (May), pp. 61-64.
                                               21

-------
  In Situ Bloremediatlon of a Pipeline Spill Using Nitrate as the Electron Acceptor
                        Stephen R. Hutchins, John T. Wilson, and Don H. Kampbell
      U.S. Environmental Protection Agency, Robert S. Kerr Environmental Research Laboratory, Ada, OK
In the la» 1970s, leakage of refined petroleum products
from an  underground pipeline contaminated approxi-
mately 24,000 square meters of  a shallow water-table
aquifer in Park City, Kansas. Aerobic in situ bioremedia-
tion was initiated but was unsuccessful due to plugging
of the injection wells or sediments adjacent to (tie well
screen by gas and iron precipitates.

Nitrate WLS selected as an alternative electron acceptor
that might avoid some of the problems with plugging.

Approach

Ground water from the aquifer was amended with so-
dium nitrate and ammonium chloride and returned to the
area of the hydrocarbon spill through a series of infiltra-
tion wells that were installed in a grid. The wells were
spaced 6.1  m apart  The study area contained 157
infiltration wells, spaced over 5,800 m2, which received
3,000 m3 of water in a tracer test  followed by 39,400 m3
of water containing 4,136 kg of sodium nitrate (an aver-
age of 17 mg/L nitrate nitrogen). The circulated water
also contained 50 to 60 mg/L sulfate.

Rgure 1 plots the cumulative flow of ground water to the
infiltration wells against time. Row was unhindered for
the first 150 days of operation, then the system plugged
over the next 100 days.

A total of 7.3  m of recharge was  applied to the spill, of
which 6.8  m contained nitrate.

Procedure to  Distinguish Flushing from
Biodegradation of BTEX

The site was cored, and vertically stacked continuous
cores  from the same borehole were analyzed to deter-
mine the total ma?- of benzene, toluene, ethylbenzene,
and xylene (BTEX) compounds in the aquifer. To esti-
mate the mass of BTEX compounds in ground water in
contact with the hydrocarbon spill, monitoring wells were
installed in the boreholes used to acquire the cores. The
screened interval on the monitoring well was equivalent
to the  depth interval containing NAPL hydrocarbons.
                                                                         Cumulative Flow
                50
                       100      150     JOC

                     Days After Addition of Nitrate
                                             250
Flgur» 1. Cumulative flo*v of ground water am*nd«d with ni-
        trate to ttw study araa (m3).

The following procedure was used to determine the total
mass of BTEX compounds in the aquifer under a unit
surface area. The concentrations of BTEX compounds
in individual core samples (g/kg) were multiplied by the
vertical interval that each core represented (M),  then
multiplied by the  bulk density of sandy aquifer matenal
(1,800 kg/m3). The masses in the depth intervals repre-
sented by the cores then were summed to determine the
total  mass of each BTEX compound under each square
meter (Table 1).

The concentration of BTEX compounds in water under
each  square meter was determined  by multiplying a
square meter by the length of the well  screen to deter-
mine the volume sampled, then by 0.3 to estimate the
volume of ground water, then  by the concentration of
BTEX compounds in ground water sampled from the
well (Table 1). The volume of aquifer sampled by the well
to estimate  mass in ground  water and the volume
summed to estimate total mass w»re equivalent.

The ratio of mass in water to total mass determines the
fraction of total mass that can be flushed away each time
water in the sampled volume is exchanged by the infil-
trating ground water (Table 1).

The volume of water in the sample volume was consid-
ered equivalent to a pore volume in a column experi-
ment; the infiltration of ground water was expressed in
                                                 22

-------
Table 1.  Concentration of BTEX Compounds in Ground
        Water and in the Aquifer at Site 60A, the Mo«t
        Contaminated Site in the Study    '
Compound
Benzena
Toluene
Ethylbenzene
p-Xyten*
m-Xylene
o-Xycentrations In |ug/l_
 Estimate of Treatment Effectiveness

 If  the  concentration of BTEX compounds in ground
 water and in the NAPL are in equilibrium, Raoulf s Law
 can be used to put an upper boundary on the total mass
 of contaminant removed by in situ bioremediation. Con-
 centrations of individual BTEX compounds were com-
 pared  before and after remediation  to determine
 fractional removal in ground water. The fractional remov-
 als in ground  water were multiplied by the initial total
 mass of each  BTEX compound to estimate total mass
 removals.

The amount of BTEX degraded during denitrification is
equivalent  to  the amount of nitrate-nitrogen applied.
Apparentfy, considerably more BTEX was removed than
could be explained  by the quantity  of nitrate supplied
(Table 2). In fact, there was more removal than could be
accounted for by either denitrification or flushing. Sulfate
in well 60A was less than  1.0 mg/L prior to the start of
infiltration; during infiltration concentrations ranged from
57 mg/L to 93 mg/L. During the course of the demon-
stration, concentrations of sulfate in monitoring well 60G
in the study area were  near  10  mg/L, when concentra-
tions of sulfate were in the range of 50 mg/L to 60 mg/L
in the infiltrated water. Removal of 40 mg sulfate per  liter
by sulfate reduction could  have  accounted for as much
as 230 gm/m2 of total BTEX removal. If this is the case,
naturally occurring sulfate in the infiltrated ground water
was  more important as an electron acceptor than  the
nitrate that was intentionally added. Concentrations of
methane in the  infiltrated water ranged from 4.8 mg/L to
6.3 mg/L while  concentrations in well 60A  ranged from
                                                   23

-------
Tabte 2.  UM of Raoulf s Law to Estimate the Total Mam of Contaminants Removed by Nitrate-Based Bloremediatlon at SOA, the
        Moat Contaminated Slta In the Study Area
Compound
Benzene
Toiiiene
o-Xylene
m-Xylene
p-Xy1*n«
Ethylbenzena
Total BTEX removed
Maximum attributed to
Maximum attributed to
Balance, attributed to !
Concentration In Well
Initial
2,010
2,570
776
1,260
958
1,020

nitrate as electron acceptor
flushing
urffnt* orrf
UIIBIV o9 WfJCITWi BUC^n^im
60A(jig/L)
Rnal
174
77.9
209
297
304
26.5



Fraction
Removed from
Water
(percent)
0.913
0.970
0.732
0.7ft*
0.683
0.974



Initial
Concentration In
Core Material
(gm/m2)
17.6
102
78.3
161
68
72



Maaa Removed
(gm/m2)
16.1
98.9
57.2
123
46.4
70.2
411.2
118
131
163
2.8 mg/L to 3.7 mg/L Mettianogenesis cannot explain
the missing mass of BTEX compounds.

The assumption of chemical equilibrium also may be in
error, and  much of the BTEX may not have been in
contact with the ground water. In  this case, the total
BTEX removed would be overestimated, and the nitrate
demand that was exerted would represent that portion
of the  hydrocarbons that exchanged re idily with the
ground water.
                                                  24

-------
     Performance Evaluation of Full-Scale In Situ and Ex Situ Bioremediation of
                        Creosote Wastes in Ground Water and Soils
                  Ronald C. Sims, Judy L. Sims, Darwin L Sorensen, and David K. Stevens
                                    Utah State University, Logan, UT

                         Scott G. Huling, Bert E. Bledsoe, and John E. Matthews
                             U.S. Environmental Protection Agency, Ada, OK

                                             Daniel Pope
                                     Dynamac Corporation, Ada, OK
 The Champion International Superfund Site in Libby,
 Montana, was nominated by the Robert S. Kerr Environ-
 mental  Research Laboratory as a candidate site for
 performance evaluation as part of the FEPA-sponsored
 Bioremediation Reid Initiative. Two forms of wood pre-
 servatives were used at the site: creosote, containing
 polycyclic aromatic hydrocarbons  (PAHs),  and loose
 pentachlorophenol (PCP). PAHs are  currentty the pri-
 mary components of concern at the site. The perform-
 ance evaluation project is directed by Dr. Ronald Sims
 of Utah State University.

 The bioremediation performance evaluation consisted
 of three phases:  1) summarize previous and current
 remediation  activities;  2) identify site characterization
 and treatment parameters critical to the evaluation  of
 bioremediation performance for each of the bioremedia-
 tion treatment units; and 3) evaluate bioremediation per-
 formance based on this information.

 Three biological treatment processes  are addressed in
 the bioremediation performance evaluation:  1) surface
 soil bioremediation in a prepared-bed, lined land treat-
 ment unit (LTU); 2) treatment of extracted ground water
 from the upper aquifer in  an  aboveground fixed-film
 bioreactor and 3) in situ bioremediation of the upper
 aquifer at the site. A description of the site with accom-
 panying figures appears in  the abstract book from the
 1993 EPA-sponsored Symposium on Bioremediation of
 Hazardous Wastes (1).

 Biological Treatment Processes

The LTU has been used for bioremediation of contami-
 nated soil taken from three primary sources, including
tank farm, butt dip, and waste pit areas. Contaminated
soil was excavated and moved to one central location,
the waste pit. Soil pretreated in the waste pit area is
further treated in one of two prepared-bed, lined land
treatment  cells (LTCs). Total estimated  contaminated
soil volume for treatment is 45,000 yd3 (uncompacted).
Contaminated soil cleanup goals (dry-weight basis) are
1) 88  mg/kg total (sum of 10) carcinogenic PAHs; 2) 8
mg/kg naphthalene; 3) 8 mg/kg phenanthrene; 4) 7.3
mg/kg pyrene; 5) 37 mg/kg PCP, and 6) 0.001 mg/kg
2,3,7,8-dioxin  equivalent

The LTU comprises two adjacent 1-acre cells. Compo-
nents  of the soil  bioremediation system for each LTC
include the treatment zone, liner system, and leachate
collection system. Each cell is lined with low-permeabil-
ity materials to minimize leachate infiltration from the
unit Contaminated soil is applied and treated in lifts
(approximately 9-in. thick) in the designated LTC. When
reduction  of  contaminant  concentrations in  all  lifts
placed in the LTU has reached the cleanup goals speci-
fied in the Record of Decision (ROD), a protective cover
will be installed over the total 2-acre unit and maintained
in such a way as to minimize surface infiltration, erosion,
and direct contact

Degradation rates, volume of soil to  be treated, initial
contaminant concentration, degradation  period,  and
LTC size determine the time required for remediation of
a given lift Based on an estimated 45-day ti...e frame
for remediation of each applied lift as determined by
Champion  International,  an estimated 45,000 yd3 of
contaminated  soil, and a 2-acre total LTU surface area,
the projected  time to complete soil remediation is 8 to
10 years.
                                                 25

-------
The upper aquifer aboveground treatment unit provides
biological treatment of extracted ground water for re-
moval  of PAHs  and PCP prior  to reinjection via an
infiltration trench. The biological treatment consists of
two fixed-film reactors operated in series. The first reac-
tor is heated and has been used for roughing purposes,
while the second has been used for polishing and reoxy-
genation of the effluent prior to reinjection. The system
was commissioned in February 1990.

Extracted ground-water treatment system components
include equalization and biotreatment Equalization sys-
tem components include four ground-water extraction
wells  and an  socialization tank, which consists  of a
cylindrical horizontal flow tank with a nominal  hydraulic
residence time of 6 hours at a flow rate of 10 gpm. The
bioreactor treatment system components include nutri-
ent amendment, influent pumping, bioreactor vessels,
aeration, heating, and effluent pumping. The compo-
nents  of the  aboveground treatment  system for ex-
tracted ground water are shown in the 1993 Symposium
abstract book (1).

The pilot upper aquifer area in situ bioremediation sys-
tem involves the addition of oxygen and inorganic nutri-
ents to stimulate the growth of microbes.. The  initial
source of oxygen was a hydrogen peroxioe injection
system that was designed to  maintain  a concentration
of approximately 100 mg/L of hydrogen peroxide.  Injec-
tion flow rate  was approximately 100  gpm into  three
injection clusters. Inorganic nutrients in the form of po-
tassium triporyphosphate and ammonium chloride are
continuously added to achieve concentrations in the injec-
tion water of 2.4 mg/L nitrogen and 1 mg/L phosphorus.

The ROD calls for cleanup levels in the  upper aquifer of
40 parts per trillion  (ppt) total carcinogenic PAHs, 400
opt for total noncarcinogenic PAHs, 1.05 mg/L for PCP,
5 ug/L for benzene, 50 u.o/L for arsenic,  and  a human
hearth  threat no greater than 10"* for ground-water con-
centrations of other organic and inorganic compounds.


Performance Evaluation Activities

Performance of the soil  bioremediation system in the
LTCs involved  evaluating the reduction in concentration
of PAHs and PCP with time and with depth within the
LTD. The primary purpose of the LTD soil sampling
program in this project was to determine the statistical
significance and extent of contaminated soil treatment
at this  site. A quantitative expression of data variability
is necessary to determine an  accurate estimate  of
biodegradation of these contaminants at field  scale.
Such an expression will allow data generated to be used
by others to help estimate the biodegradation  potential
of similar type  wastes under similar conditions at  other
sites.
In most soils and disturbed soil materials, physical and
chemical properties are not distributed homogeneously
throughout the volume of the soil material. The variability
of these properties may range from 1 percent to greater
than 100  percent of the  mean value within  relatively
small areas.  Chemical properties,  including contami-
nants, often have  the highest variability. A first app'oxi-
mation of the total variance m monitoring data can be
defined by the following equation:
V, =
                         + Va/k*n
whore  k  is the number of samples, n the number of
analyses per sample, k'n the total number of analyses,
V, the total variance, Va the analytical variance, and V,
tha sample  vananct  In general,  sampling efforts to
minimize V, result in the most precision. Analytical pro-
cedures frequently achieve precision levels (Vyk*n) of
1 to 10 percent, while soil  sampling variation (V,) may
be greater than 35 percent. Sampling designs that re-
duce the magnituoe of V,  should be employed where
possible. Therefore, the sampling  procedures  used in
this  evaluation  were  designed to  minimize V, and to
provide representative information about the transfor-
mation of PAHs and PCP within the LTCs.

The LTD was sampled in May, June, July, and Septem-
ber 1991, and in September 1992. Field-scale investi-
gations concerning PAH and  PCP concentrations were
supported by laboratory mass-balance investigations of
radiolabeled compounds for determination of minerali-
zation  as  well  as humification  potential for target
contaminants.

Performance evaluation  of the  upper aquifer above-
ground fixed-film treatment system involved evaluating
the bioreactor system. Treatment evaljation focused on
characterizing performance regarding system capability
to remove PAHs and PCP from the ground water, and
on  optimizing  operation within  the bioreactors.  The
aboveground treatment system was sampled  during
1991 and 1992 for chemical, physical, and biological
parameters. In addition, a pilot-scale reactor was con-
structed and operated to evaluate abiotic reactions of
chemicals present  in  the water phaso within the biore-
actors. The  information generated from  the sampling
and  monitoring of  the full-scale reactor and from the
operation of the pilot-scale reactor was combined with
data provided by Champion International  to provide an
in-depth evaluation of performance.

Performance evaluation of the in situ bioremediation
system focused on characterization of the water phase,
the solid phase (aquifer materials), and oil associated
with the aquifer solid material. The aquifer was sampled
during  1 991 and 1 992. An evaluation of the water phase
included  measurements  of  dissolved  oxygen  (DO)
concentrations, the  inorganic  chemicals  iron  and
                                                  26

-------
 manganese to  evaluate potential  abiotic  demand for
 injected hydrogen peroxide, and the concentrations of
 PAHs and PCP. An evaluation of trie aquifer solid phase
 has included PAHs and PCP concentrations in treated
 and Background areas at the site. Laboratory  mass
 balcnce  experiments  using radiolabeled  target cc;.>
 pc jnds were used in conjunction with field-scale meas-
 urements to  provide additipnal information concerning
 biotic reactions (mineralization) and potential abiotic re-
 actions (poisoned controls).

 Summary  of Results

 Analyses of more than 300 soil samples were performed
 from which greater than 5,000 individual chemical con-
 centrations were determined for the 16 priority pollutant
 PAH compounds  using  gas chromatography/mass
 spectrometry  (GC/MS)  and  for  pentachlorophenol
 (PCP) using a gas chromatography/etectron capture de-
 tector (GC/ECD). Results  of  chemical  analyses indi-
 cated that target remediation levels for target chemicals
 were achieved using mean values  at each  depth evalu-
 ated in each LTC, with only two exceptions  where mean
 concentrations were only slightfy higher than the target
 remediation levels. As a result of obtaining vertical sam-
 ples at each sampling event,  downward  migration of
 target chemicals through the LTU was no: observed. Soil
 within the LTU was detoxified 10 control uncontaminated
 soil levels. Toxicity information was based  upon results
 of using both the Microtox assay to measure  water
 extract toxicity and the Ames Salmonella  typhimurium
 mammalian microsome mutagenicrty assay (Ames as-
 say) to measure mutagenicrty  of soil solvent extracts.
 Detoxification to nontoxic levels was evident in all sam-
 ples evaluated for both Microtox and Ames assays.

 Results of the  laboratory evaluation of soil microbial
 metabolic potential demonstrated that PCP and  phen-
 anthrene, the two chemicals evaluated using radiola-
 beled carbon, could be metabolized to carbon dioxide
 by indigenous microorganisms present in the contami-
 nated soil matix present at the site at temperature and
 moisture conditions  representative of the site. In addi-
 tion, significant volatilization of PCP or phenanthrene is
 unlikely based upon the laboratory evaluation. The in-
 formation obtained in the laboratory evaluation corrobo-
 rated the interpretation of apparent decrease in  target
 chemical concentrations in field samples within the LTU
 and in the in situ aquifer samples  at the Libby site as
 due to biological processes rather than pfr'sical/chemi-
 cal processes.

 Results of the aboveground fixed-film bioreactor indi-
cated that removal of PCP and PAHs from extracted
ground water was strongly influenced by hydraulic re-
tention time (HRT). The system removed greater than
80 percent of PCP and 90 percent of PAHs at a flow rate
of 10 gpm, with an HRT of 30 hours. At a  flow rate of
10 gpm, the  effluent concentrations of PCP and total
PAHs .vere 0.3 mg/L to 0.9 mg/L and less than detection
(30 (ig/L), respectively.  When the flow rate was  in-
creased to 15 gpm, with an HRT of 20 hours, removal
of both  PCP  and PAHs decreased significantly. At the
15-gpm flow  rate, effluent concentrations of PCP and
total PAHs were 6 mg/L to-12 mg/L and 0.6  mg/L to 6
mg/L, respectively. Additional  limitations of DO and nu-
trients are addressed in the final report.

Results of the in situ treatment evaluation indicated that,
with respect to the ground-water phase, total PAHs and
PCP were present at lower concentrations in wells con-
sidered to be  under the influence of the treatment injec-
tion  system  consisting  of  nutrients  and   hydrogen
peroxide,  while total  PAHs and PCP were  present at
higher concentrations in wells considered to  be outside
of the influence of the injection system. An evaluation of
the water phase in monitoring wells demonstrated the
presence  of  reduced inorganic compounds, including
iron and manganese, with concentrations inversely re-
lated to DC concentrations. These chemicals may exert
a damand on the oxygen supplied by the  hydrogen
peroxide and  reduce the oxygen available for microbial
utilization.

With respect  to the nonaqueous phase liquid (NAPL)
phase, both  total PAHs and  PCP were found in the
highest concentrations in the NAPL. greater than 10,000
mg/L and 1,000 mg/L, respectively, than in  any other
phase  sampled  at  the  Champion International  Site.
These results indicate that there is potential contamina-
tion of the upper aquifer remaining in the  form  of  a
nonaqueous phase that represents significant potential
contamination of the ground  water by transfer or  con-
taminants from the  NAPL phase to the ground-water
phase.

Total PAH and total petroleum hydrocarbons (TPH) were
present within the aquifer sediment/NAPL samples at
concentrations of 5 mg/kg to 687 mg/kg and  70 mg/kg
to 2,525 mg/kg, respectively. The heterogeneous distri-
bution of total PAH, PCP, and TPH contaminants was
consistent among three  boreholes evaluated from the
water table to the deepest sampling point. Target chemi-
cals associated with sediment/NAPL interfaces may be
more difficult  to bioremediate  in situ than chemicals in
the aqueous phase due to limitations of mass transport
of oxygen and nutrients from the water phase to the
NAPL phase that contain target chemicals.

Chemical  mass balance evaluations  conducted using
radiolabeled target chemicals  in the laboratory demon-
strated that aquifer materials from the site  contained
indigenous microorganisms that had the ability to  min-
eralize phenanthrene.  Up to  70 percent of the radiola-
beled carbon became incorporated into the aquifer matrix
and was nonsolvent extractable. No significant phenan-
threne  mineralization or incorporation  of  radiolabeled
                                                  27

-------
carton was observed in poisoned controls. PCP miner-
alization, however, was  insignificant (less than 2 per-
cent), with results similar for nonpoisoned and poisoned
samples.

The three biological treatment processes evaluated at
the Libby, Montana,  site represent a treatment  train
approach  to site decontamination,  where each of the
treatment  processes  are biological. The soil phase is
treated in  the LTU system, and any leachate produced
can be treated in the aboveground  bioreactor before it
is returned to the LTU as part of soil  moisture content
control and treatment of low levels of PAHs and PCP in
the effluent. The in situ treatment system addresses the
oil and solid phases in the subsurface. At the LJbby site,
therefore,  a different biological process was chosen for
remediation of each contaminated phase (soil, oil, and
water).

Performance Evaluation Reports
While the extended abstract prompted in this report has
been  abridged concerning site  c^iaracterization and
treatment results, separate report* have been prepared
for EPA that  address  each  of the three biological
treatment systems at the site in detail:  1) soil bioreme-
diation in the prepared-bed LTU; 2) aboveground fixed-
film system for extracted ground water; and  3) in s/tu
treatment. Information generated from full-scale charac-
terization and monitoring, pilot-scale studies, and laoo-
ratory treatability studies was combined with information
provided by Champion International to  provide an inte-
grated evaluation of bioremediation performance at the
Ubby, Montana, site. The information obtained can be
used  to evaluate and  select rational  approaches for
characterization, implementation, limitations, and moni-
toring of bioremediation at other sites.

References

1.  Sims, R.C., J.E.  Matthews, S.G. Huling, B.E. Bled-
   soe, M.E. Randolph, and D. Pope. 1993. Evaluation
   of  full-scale in situ  and ex situ  bioremediation of
   creosote wastes  in soils and ground water. In:  U.S.
   EPA. Symposium on bioremediation of hazardous
   wastes: Research,  development, and field evalu-
   ations (abstracts). EPA/600/R-93/054. Washington,
   DC (May).
                                                  28

-------
                 Bloventing Soils Contaminated with Wood Preservatives
                         Paul T. McCauley, Richard C. Brenner, and Fran V. Kremer
           U.S. Environmental Protection Agency, Office of Research and Development, Cincinnati, OH

                                           Bruce C. Alleman
                                 Battelle Memorial Institute, Columbus, OH
                                                              i
                                          Douglas C. Beckwrth
                             Minnesota Pollution Control Agency, Si Paul, MN
 The Reilly Tar and Chamical Corporation operated a
 coal tar distillation and wood preserving plant, known as
 "ie Republic Creosote Company, in St. Louis Park, Min-
 nesota, from 1917 to 1972. During this period, wastewa-
 ter discharges as well as drips, spills, and dumping from
 the wood preserving processes resulted in creosote and
 coal tar contamination of about 80 acres of this site and
 the underlying ground water. In 1972, the City of St.
 Louis Park purchased the site from  the Reilty Tar and
 Chemical Corporation for land use. All onsite  buildings
 were dismantled and removed, and the soil was graded
 and covered with 3 ft of topsoil for  beautification and
 odor control.

 In  1978, the Minnesota Department of Health  began
 analysis of  ground water from municipal wells in  St.
 Louis Park and neighboring communities for carcino-
 genic and noncarcinogenic polycyclic aromatic  hydro-
 carbons  (PAHs).   The  discovery  of  significant
 concentrations of regulated PAHs in  six St. Louis Park
 wells  resulted in their shutdown  during the  period of
 1978 to  1981. St  Louis Park is currentfy maintaining
 gradient control of  the  contaminated ground-water
 plume by pumping and treating. With tne exception of a
 tar plug in one well, little PAH source contamination has
 been removed. Without source control of the  PAHs,
 pumping and  treating of contaminated ground water
 may be required for several hundred  years.

 Background

 Bioventing is a proven technology for in situ remediation
of various types of hydrocarbon contaminants. The tech-
nology  has been used successfully to remediate sites
contaminated with gasoline (1), aviation fuels (JP-4 and
JP-5) (2,3), and  diesel fuel (4). A biological treatment
process, bioventing uses low-rata atmospheric air (or
oxygen enriched air up to 100-percent oxygen) injection
to treat  contaminated unsaturated soil in situ. The air
flow provides  a continuous oxygen  source that en-
hances the growth of aerobic microorganisms naturally
present  in soil, wtth minimal volatilization to the atmos-
phere of any. volatile organic compounds that may be
present  in the soil. The size of the  treatment area is
defined by the  number of wells installed, the size of the
air blower used, and site characteristics such as soil
porosity. The current research evaluates the potential of
bioventing to remediate soil? contaminated with PAHs.

Methods

Site Description

A 50 ft x 50 ft control  and r. iO ft x 50 ft bioventing
treatment plot  were established on the site during the
original soil gas survey (Figure 1). The first 3 ft of soil at
the test plots is uncontaminated topsoil applied after the
cessation of industrial use (Figure 2). A dense, 3-in. to
6-in., hard-packed  layer separates the  topsoil from the
porous sandy layer, which  extends to below the water
table (8 ft to 10 ft below the ground surface). Most of the
PAH contamination was found in the sandy layer.

PAH Sampling

Composite soil samples (120 soil borings per plot) were
taken for PAH analysis  and prepared by homogenizing
the soil obtained from the 4 ft to 8 ft depth of each boring.
The resultant  boreholes were  filled  immediately  with
bentonite. The  PAH soil analyses were recorded as
zero-time PAH concentrations.
                                                  29

-------
                  Louisiana Avenue
                                Treatment Plot

                           ©                  ©
                                               o
                                 T
                                   Thaller
         Control Plot
                                F*oc«
 Figure 1. Placement of Infection and soil gas sampling wells In
         the control and treatment plots.
 Venting Well

A single-vent bioventing system was installed  at  the
center of the treatment area (Figure 2). The vent (injec-
tion) well was screened from 7 ft to 11  ft below  the
surface and packed with sand. The vent well then was
sealed with bentonite from the 5  ft depth to the surface.

Soil Gas Sampling Well
Twelve soil gas probes were installed along diagonals
drawn from the comer of  the square treatment area
(Figure 2), and four were installed in the comers of  the
no-treatment control area.  The  soil gas probes were
constructed so that the soil gas  withdrawal points and
thermocouples were located at 4, 6, and 8 ft below  the
ground surface.

Respirometry
Initial Oj and CO2 measurements were  obtained us.ng
stainless steel gas probes  withdrawing air from  meas-
ured intervals below the ground surface to Gas Teck O2
and  CO2 meters. The gas measurements were  ex-
pressed as percentages of  total soil gas. Gas samples
for the zero-time sampling in November were extracted
using the newty installed soil gas sampling wells. Initial
sampling indicated that due at least in part to the highly
pervious soil at the Reilly site, injected air was migrating
from the test plot 125 ft to 180 ft  into  the  unaerated
                                         Btoventtng Injection and So< Qa< Sampling W*fls
                                                                     Air injection W*
              -11FMI
                       gj  Ha«) Ptdwt Ljyvr                       ^  B*ntont*


                       HjJ  CotrM Sandy Iff*                      ff  Sand and Qrav«l


Figure i  Air lr|*ctlon and soil g*» sampling w*lls Installed In th« treatment plot
                                                    30

-------
 control plot A 10-tt deep bentonite slurry wall was con-
 structed across the near wall of the control plot The
 slurry wall and reduced air injection pressures and flow
 rates effectively prevented further unwanted aeration of
 the control plot

 Shutdown Respiration Tests
 Shutdown respiration tests are being conducted for 2
 weeks at quarterly intervals. Soil gases are brought to
 atmospheric  O2 and C02 levels in the test plots  by
 pumping ambient air into the ground. When ambient O2
 and CO2 levels are achieved and documented, the  air
 flow into the ground is stopped. Soil gases levels are
 taken over measured intervals until an 02 utilization rate
 is defined. The air flow was set st 10 ft3/min, which
 translated at this site to a pressure of 3.5 in. of H2O.

 Results
 In the summer of  1992,  a  field team from the  Risk
 Reduction Engineering Laboratory (RREL), Biosystems
 Branch, conducted  a soil gas survey at the Reilly site
 and determined that soil gases were  below the  esti-
 mated 5-percent oxygen threshold required for aerobic
 metabolism (5). Under a cooperative  project involving
 the Btoremediation Reid Initiative, the Superfund Inno-
 vative  Technology  Evaluation (SITE)  Demonstration
 Program,  and RREL's Biosystems Program, a pilot-
 scale bkjventing field demonstration for PAH bioreme-
 diation was initiated at the Roilty site in November 1992.
 Soil PAH analysis demonstrated significant contamina-
 tion in both plots. The treatment plot demonstrates an
 order-of-magnitude  decrease in  PAH concentration on
 the eastern side of the plot The control plot is contami-
 nated to about the same degree as the western half of
 the treatment plot
 Quarterly shutdown respiration tests have shown respi-
 ration rates ranging from below detection (Figure 3) to
 0.484 percent 02 per hour (Figure 4). The highest res-
 piration rates were found in the western half of the
 treatment  area, where PAH contamination  also  was
 shown to be  the heaviest. Current average measured
 respiration rates are consistent with a 14-percent reduc-
 tion in PAH contamination per year.

 Summary and Conclusion

 A 3-year evaluation program was initiated in November
 1992 with the zero-time sampling. In  situ respiration
 tests are being performed four times each year to deter-
 mine oxygen utilization and CO2 evolution rate?. These
data can  be converted to estimated  bkxJegradation
rates to estimate the disappearance of PAHs (6). Be-
cause of the strong partitioning of PAHs to soil, long-
term bioventing is  expected to be necessary to  fully
                Respiration Curva (MP- K)
                    Shutdown Test
   ,.
 o»

 O
   10
 £
               5^fl	L.
                                     ->—&-J 0
                                             15
                                             12
                                              3
                                            6!
                                              0.

                                            3
    o as-oc-
      0   40    80    120   160   200   240
                      Time, hours

      •  4 Foot         •  8 Foot       *  8 Foot

Flgurv 3.  Solid symbol* represent Oj. Hollow symbols rapi
         sent COj.
                      Shutdown Test
                                             15
                                             12
                                             9
                                          -  6
                                          -  3
     0    40    30   120  160   200  240

                   Tim*, hours

Figure 4. Solid symbol* represent O?. Hollow symbols repr
        MfltCOj.
remediate the site. The target PAH removal rate for tn
3-year project is 30 percent. Successful achievement
this rate would project total cleanup in 10 to 15 years
                                                  31

-------
References

1.  Ostendorf, D.W., and D.H. Kampbell. 1990. Bioreme-
   diated soil venting of light hydrocarbons. Haz. Waste
   Haz. Mat 7:319-334.

2.  Sayles,  G.D., R.C.  Brenner,  R.E.  Hinchee,  A.
   Leeson, C.M. Vogel, R.  Elliot, and R.N. Miller. 1994.
   Bioventing of jet  fuel spills I:  Bioventing in a cold
   climate with soil  warming at Eielson  AFB, Alaska.
   Presented at the  U.S. EPA Symposium on Bioreme-
   diatksn of Hazardous Wastes: Research, Develop-
   ment and  Field Evaluations, San Francisco,  CA
   (June).

3.  Sayles,  G.D., R.C. Brenner,  R.E. Hinchee, and R.
   Elliott 1994. Bioventing of jet spills II:  Bioventing in
   a deep vadose zone at Hill AFB, Utah. Presented at
   the U.S. EPA Symposium on Bioremediation of Haz-
   ardous Wastes: Research, Developmant, and Field
   Evaluations, San Francisco, CA (June).

4.  Kampbell, D.H., and J.T. Wilson. 1991. Bioventing to
   treat fuel spills from underground storage tanks. J.
   Haz. Mat 28:75-80.

5.  Ong,  S.K.,  R.E.  Hinchee,  R.  Hoeppel,  and  R.
   Schultz. 1991. In situ  respirometry  for determining
   aerobic degradation rate. In: Hinchee, R.E., and R.F.
   Olfenbuttel, eds. In situ bioreclamations,  applica-
   tions, and investigations for hydrocarbons and con-
   taminated   site   remediation.    Boston,   MA:
   Butterworth-Heinemann. pp. 541-545.

6.  Hinchee, R.E., and S.K. Ong. 1992. A rapid in situ
   respiration test for measuring aerobic biodegradation
   rates of hydrocarbons  in soils.  J. Air Waste Mgmt.
   Assoc. 42(10): 1,305-1,312.
                                                  32

-------
                     Field Evaluation of Fungal Treatment Technology
                                            John A. Glaser
                U.S. Environmental Protection Agency, Risk Reduction Enginer ring Laboratory,
                                            Cincinnati, OH

          Richard T. Lamar, Diane M. Dietrich, Mark W. Davis, Jason A. Chappelle, and Laura M. Main
                   U.S. Department of Agriculture, Forest Products Laboratory, Madison, Wl
 Bioaugmentation of soil contaminated with pentachlo-
 rophenol (PCP) using selected strains of lignin-degrad-
 ing fungi has been shown to result in extensive and rapid
 decrease in the PCP concentrations for two soils under
 field treatment conditions (1,2). In different soils studied
 under laboratory conditions, the same  behavior was
 observed and extensively evaluated by means of deter-
 mining the pollutant mass balance in the soils (3,4).
 Initial products of fungal biotransformation  were  identi-
 fied. PCP concentrations in excess of 1,000 mg/kg were
 80 to 90 percent biotransformed in soil by selected fungi
 in 56 days (Figure 1).
 A two-phase project consisting of a treatability study in
 1991  and a demonstration study in 1992, was conducted
 at an abandoned wood treating site in Brookhaven,
 Mississippi,  to evaluate fungal treatment effectiveness
 under field conditions. The study site, located 30 miles
 south of Jackson, was identified as a removal action site
 for EPA Region 4. While the wood treating facility was
 in operation, two process liquid  lagoons were drained
 and excavated. The sludge was mounded above the
 ground surface in a Resource Conservation and Recov-
 ery Act (RCRA) hazardous  waste treatment unit The
 excavated material provided the contaminated soil for
 both phases of the  project  The demonstration  phase
 was undertaken as a Superfund  Innovative Technology
 Evaluation (SITE) Program Demonstration  Project

 The fungal  treatment processes reported herein were
 conducted at Brookhaven because the site charac-
 teristics were suitable for conducting field investigations,
 not because the investigators desired to promote  fungal
 treatment as one of the treatment options for the site.

 Methodology

The demonstration study was designed to evaluate the
ability of a single fungal strain (Phanaerochaete sordida)
       % TREATMENT^
              15
          PCP(ng/8)
    67
                                             1017
                                       673
3 tPchrytosponum T.hirsutt^  615

 14 JT. INOCULUM CONTROL UMl 687

 151 N6 TREATMENT CONTROL
                                       737
            IWOOO CHIP CONTROL 1
Figure 1. TrwrtaWltty study p«rform»net.

to degrade PCP in soil. The soil pile was sampled and
analyzed for PCP  and creosote components (i.e., poly-
cyclic aromatic hydrocarbons [PAHs]) prior to develop-
ing  the test site. Analysis of the laboratory results
identified sections of the pile with PCP concentrations
of less than  700 mg/kg. These sections were used to
supply the contaminated  soil for both phases of the
study.

A test location was constructed on an uncontaminated
portion of the wood treating site. The base for the test
plots was formed by using uncontaminated soil to pro-
vide i 1-percent to 2-percent slope to promote better
drainage. Soil beds (Rgure 2) were constructed of gal-
vanized sheet metal. For the demonstration study, the
P. sordida treatment plot measured 30.5 m x 30.5 m and
the  treatment and inoculum  control plots  measured
7.6  m x 15.25 m.  Plot dimensions  were determined in
conjunction with SITE  program personnel. A concrete
                                                 33

-------
pad was constructed to assist tiller entry into the differ-
ent plots and to decontaminate the tiller as it was moved
from plot to plot

Within each plot the base soil was graded  for a V-
shaped indentation  in the central portion of the plot to
permit leachate collection. A leachate collection system
was installed to direct the liquid discharge from all test
plots to a central location for testing and treatment. After
installation of  the leachate system, 25  cm (10  in.) of
dean sand was  layered into each test plot followed by
a 25-cm (10-in.)  lift of contaminated soil.

The treatment plot received 10 percent by weight of an
infested  inoculum containing P.  sordida. The  no-treat-
ment control received no amendments. The inoculum
control plot consisted of  contaminated  soil amended
with noninfested inoculum carrier. All plots were tilled on
the same schedule, weather permitting.  The fungal in-
oculum  was developed jointly  with  the LF.  Lambert
Spawn Co. of Coatesville, PA. The prepared inoculum
and inoculum carrier were shipped to the site by refrig-
erated transport.
The contaminated  soil  was sized  through a 2.5-cm
(1-in.) mesh  screen using a Read  Screen All shaker
screen having a capacity 8.4 nrrVhr (10 yc^/hr). The soil
was deposited in separate piles on a polyethylene tarp.
Further homogenization was accomplished by mixing
different portions of screened soil. The soil then was
mixed with the 10 percent by weight fungal inoculum in
a  Reel Auggie Model 2375 Mixer and  applied  to the
treatment plots using a front end loader.

After inoculation with fungi, each plot was irrigated and
tilled with a garden rototiiler. Soil moisture was moni-
tored on a daily basis throughout the study and main-
tained at a minimum of 20 percent Ambient and soil plot
temperatures were recorded dairy throughout the study.
Rot tilling was scheduled on a weekly basis for the

                                   iMchat* CoHactton
  100ft
                      100 ft
                                     25 n    25 ft
duration of the study. A time series analysis of treatment
performance was accomplished by sampling the plots
before application of the treatments,  immediately after
treatment application, and after 1, 2, 4,  8, 12, and 20
weeks of operation (Figure 3).

Results
The demonstration study was conducted over a 5-month
period between June and November 1992. The greatest
removal  of PCP  (Table 1) was achieved in the  plot
inoculated with P. sordida. Over the course of the study,
this treatment  regime produced 69-percent transforma-
tion of PCP from the contaminated soil initially having a
pH of 3.8. Significant precipitation occurred throughout
the study, leading to unexpected  excursions from the
prescribed treatment protocol specified by the Risk Re-
duction Engineering Laboratory (RREL) Forest Products
Laboratory (FPL) developers. Lack of tilling clearly com-
promised the  ability to evaluate the fungal treatment
technology.
Information collected by both the SITE program and the
RREL/FPL effort demonstrated that fungal activity in the
treatment plot  was significantly lower than expected at
the beginning of the study. Fungal activity in the inocu-
lum  control  increased significantly  during the  study,
which  is most likely attributable to  infestation with  a
wild-type fungal species.
Summary demonstration'removal data for the soil con-
taminants is presented in Table 1 for the treatment using
P. sordida. Concentration decreases  of the three- and
four-ring PAHs were consistently greater following fun-
gal treatment.  Larger  ring PAHs persisted in both the
treatment and  control plots.

Summary and Conclusions

Treatment of PCP by fungal application  had a signifi-
cantly greater ef act when compared with  controls. Loss
of fungal activity was  detected in both the fluorescem
Flgur* 2.  Broofchav*n  damonttritior  treatment  plot
         parapacttva.
Rgure X  Sampling plan layout
                                                   34

-------
Tabto 1.  Summary Results for Demonstration Study (5,6)

                                           Percentage Removal
Analyte
PCP.
(RREL1/FPL data)
2-Rlng PAHs
3-Rlng PAHs
4-fllng PAHs
5-fllng PAHs
TotaJ PAHs
No Treatment Control
13
19
70
83
46
14
65
Inoculum Control
71
30
48
72
67
25
66
Treatment
(P. sordida)
69
69
46
64
58
27
59
diacatate and ergpsterol analyses (Figures 4 through 8).
The specified RREL/FPL treatment protocol could not
be followed in the required time frame because of ex-
cessive precipitation during the testing period. The miss-
ing component of the protocol was the specified tilling
of the treatment beds. The treatment data dearty show
that the inoculum control was infested with a wild-type
fungal species, which contributed to the biotransforma-
tion of the targeted pollutants in that plot

Treatment by the selected fungal species was observed
for PCP concentrations in excess of 1,000 mg/kg, which
is greater than any reported concentrations treated us-
ing bacterial inocula (Figure 9). Despite the remarkable
differences in soil composition and characteristics for
the Wisconsin and Mississippi sites, consistent biotrans-
formattons of 80 to 90 percent were observed for PCP.
One notable soil feature that apparently does not affect
fungal treatment is soil pH, which, for the Wisconsin and
Mississippi sites, was 3.5 and 9.2, respectively.

References

1. Lamar, R.T., and 0. Dietrich. 1990. In situ depletion
   of pentachlorophenol  from  contaminated  soil by
   Phanerochaete   spp.   Appl.   Environ.   Microbiol.
   56:3,093-3,100.
2. Lamar, R.T.,  J.W. Evans, and J.A.  Glaser.  1993.
   Solid-phase treatment of a pentachlorophenol con-
   taminated soil using lignin-degrading fungi. Environ.
   Sci. Techno!.  272,566-2,571.

3. Lamar, R.T., J.A. Glaser, and T.K. Kirk. 1990. Fate of
   pentachlorophenol (PCP) in sterile soils inoculated
   with white-rot basidiomycete Phanerochaete chryso-
   sporiurrr.  mineralization, volatilization, and depletion
   of PCP. Soil Bid. Biochem. 22:433-440.

4. Davis,  M.W.,  JA Glaser, J.W. Evans, and R.T. La-
   mar.  1993. Reid evaluation  of the lignin-degrading
   fungus Phanerochaete sordida to treat creosote-con-
   taminated soil.  Environ.  Sci.  Technol.  27:2,572-
   2,576.

5. U.S.  EPA. 1994. Technology evaluation report:
   Bioremediation of PCP- and creosote-contaminated
   soil using USDA-FPL/USEPA-RREL's fungal  treat-
   ment technology, Vol. 1. Final draft

6. Lamar, R.T,  M.W. Davis, D.M. Dietrich, and J.A.
   Glaser. 1994.  Treatment of a pentachlorophenol- and
   creosote-contaminated  soil using the lignin-degrad-
   ing fungus Phanerochaete sordida. Submitted paper.
                                                  35

-------
       ONutanSoll
FIflur«4.  ToMfun9a4bkMiWM(mgAg)byfluorMMlndl«Mtat*
          •ttinlng.
                                                                Tr««lm«nl Control
                                                                 Inoculum Contra*
                                                                    OUuttonSotl
                                                                                           133
                                                                                           80-1
                                                                                          89.3
                                                                                                        29
                                                                                                        26
                                                                            l« I* '00 « « « 30  9  JO «  90  10  ICO <30 IJO
                                                                                              1  Sw**k 20
                                                                     7.   Actfv*  l»ct^«l  Wom.8. (mgrtig)  by fluor.sc.ln
                                                                         dujc«t«l» staining.
  TrMinwit Control
       OUkmSoH
 Flgura 5.  Aetfva  fungat  btofnan  (mg/kg)  by  fluora«ca4n
          dlac^ata staining.
  Inoculum Control
     OOuDonSol
                            658
                            844

                            656
                                            120
                                            34.8
                                           162
             noo  a  00  «o
                                       X»  «30  (00
                       E
                                1 BWMh 20


Inoculuni
Raw sou
Inoculated soil
Cone (mgrttg)
Found
241
0^
4
ExpwctMl


24
                                                               Flflum S.  Ergooterol evaluation.
                 1248
                            Time (weeks)

          — No Treat      -*• Inoculum   *


Flgura 9.  PCP concentration depletion.
Figure 8.  Total  bacterial  blomaaa  (mg/kg)  by  fluoreseeln
          dlaeetat* staining.
                                                                                                             12     20
                                                           36

-------
                 The Bioremediation in the Field Search System (BFSS)
                                            Fran V. Kremer
          U.S. Environmental Protection Agency, Office of Research and Development, Cincinnati, OH

                 Linda B. Diamond, Susan P.E. Richmond, Jeff B. Box, and Ivan B. Rudnicki
                              Eastern Research Group, Inc., Lexington, MA
The Bioremediation in the Reid Search System (BFSS)
is a PC-based software application developed by EPA's
Bioremediation Held Initiative. BFSS provides access to
a database of information compiled by the Initiative on
hazardous waste sites where biofemediation is being
tested or  implemented, or has been  completed. Sites
include  Comprehensive  Environmental  Response,
Compensation, and Liability Act (CERCLA) sites,  Re-
source Conservation and Recovery Act (RCRA) sites,
Toxic Substances Control Act (TSCA) sites, and Under-
ground Storage Tank  (UST) sites. The database cur-
rently contains information on approximately 160 sites,
primarily those under federal authority. This summer the
Initiative  plans to expand the database by soliciting
information from industry, contractors, and vendors—an
effort that is  expected to double or triple the number of
sites in the database.

BFSS contains both general site information and data
on  the operation of specific biological  technologies.
General site  information includes the location of the site,
site contacts, the predominant site contaminants, and
the legislative authority under which  the site is being
remediated.  Technology-specific information includes
the stage of operation, the type of treatment being used,
the wastes and media being treated,  the cleanup level
goals, and the performance and  cost of the treatment.
Both ex situ and in situ technologies are represented,
including activated sludge, extended  aeration, contact
stabilization,  fixed-film, fluidized bed, sequencing batch,
and slurry reactor treatments; aerated lagoon, pile, and
land treatments;  and bioventing,  air  sparging, in  situ
gror^d-water  treatment,  and  confined  treatment
facilities.
BFSS allows the user to search the system based on
location, regulatory authority for cleanup, media, con-
taminants, status of the project, and treatment utilized.
Based on the search criteria specified by the user, BFSS
generates a list of qualifying sites. BFSS allows the user
to view on-line information about these sites and to print
site  reports  based  on information  contained  in the
database.

The Initiative established the BFSS database to provide
federal and  state project managers,  consulting engi-
neers, industry personnel, and researchers with timely
information regarding new developments in field appli-
cations of bioremediation.  BFSS data and the operation
of the search system have been  reviewed by  repre-
sentatives of the target user community, including per-
sonnet  from   EPA   regional  offices   and   other
professionals in the field of bioremediation. Information
in the database is updated  semiannually and  is
reported in EPA's quarterly Bioremediation in the Field
bulletin, which is published by the Office of Research
and  Development (ORD)  and the  Office  of  Solid
Waste and  Emergency  Response  (OSWER). The
bulletin  provides  a  valuable   information-sharing
resource for site managers using or considering the
use of bioremediation.


Version 1.0 of BFSS will be available by August 1994 on
several EPA electronic bulletin boards—Cleanup Infor-
mation (CLU-IN), Alternative Treatment Technology In-
formation  Clearinghouse  (ATTIC),  and ORD bulletin
board systems—and  on diskette from the  EPA Center
for Environmental Research Information.
                                                 37

-------
                           Section Two
                  Performance  Evaluation
In an effort to evaluate the performance of various bicremediation technologies,
researchers assess the extent and rate of cleanup for particular bioremediation
methods.  They also study the environmental fate and effects of compounds and
their by-products, since remediation efforts at a  contaminated site can produce
intermediate compounds that can themselves be hazardous. Thus, another impor-
tant  aspect  of performance evaluation projects  involves assessing the  risk of
potential health effects and identifying bioremediation approaches that best protect
public health.

To this end,  EPA's Health  Effects Research Laboratory (HERL) has developed an
integrated program to address: 1) the toxicity of known hazardous waste  site
contaminants, their natural breakdown products, and their bioremediation products;
2) the development of methods to screen microorganisms for potential adverse
health effects;  3) the potential for adverse effects when chemical/chemical and
chemical/microorganism interactions occur and 4) the development of methods to
better extrapolate toxicologies I bioassay results to the understanding of potential
human toxicity.

Specific research ongoing within the HERL program includes a study of the con-
struction of noncolonizing  £ colt and P. aeruginosa. Researchers obtained strains
of E. coll that are unable to colonize the lung tissue or the intestines of humans and
animals, thus minimizing the possibility of opportunistic infections that can result in
debilitating disease. These strains could be useful as detoxifiers of chemicals,
agricultural biopesticides,  and  in the prevention of ice nucleation on plants.

The symposium's poster session included presentations on toxicant generation and
removal during crude oil degradation, the effects of Lactobadllus reuteri on intes-
tinal  colonization of bioremediation agents, and potentiation of 2,6-dinitrotoluene
btoactivation by atrazine in Fischer 344 rats.
                                   39

-------
              Integrating Health Risk Assessment Data for Bioremediation
                                 Larry D. Claxton and S. Elizabeth George
                  U.S. Environmental Protection Agency, Health Effects Research Laboratory,
                                       Research Triangle Park, NC
 Scientific literature ciearty indicates that our environ-
 ment contains individual substances, combinations of
 substances, and complex mixtures that are hazardous
 to human health. Additionally, some environmental mi-
 croorganisms historically  considered  nonpathogens
 have been shown to cause disease when humans are
 exposed under "nontypicaT conditions. To protect public
 health, those involved in remediation efforts must under-
 stand the potential for adverse health effects from envi-
 ronmental contaminants and microorganisms before,
 during, and after any type of  remediation. When bioas-
 say information coupled with, chemical characterization
 indicates a measurable loss of toxicity and testing of
 applied microorganisms (if any) shows no adverse ef-
 fects, one can have increased confidence that the reme-
 diation effort will have its intended effect
 Because any human exposure to toxicants in bioreme-
 diation sites is most likely to be of the chronic, low-con-
 centration type, the lexicological endpoint of greatest
 concern typically is carcinogenesis. Some investigators
 report an increased frequency of cancers in counties
 surrounding hazardous waste sites. One study reported
 that age-adjusted gastrointestinal (Gl) cancer mortality
 rates were higher than national rates in 20 of 21 of New
 Jersey's  counties. The environmental variables most
 frequently associated with  Gl cancer mortality rates
 were population density,  degree of  urbfinization, and
 presence of chemical toxic waste disposal sites (1). In
 a study of 339 U.S. counties (containing 593 waste
 sites) where contaminated ground drinkinc water is the
 sole source water supply, the  association between ex-
 cess deaths due to cancers of  the lungs, bladder, stom-
 ach, large intestine, and rectum and the presence of a
 hazardous waste Sud (HWS) was significant when com-
 pared with all non-HWS counties (2). Although studies
such as these do not prove causality between cancer
incidence and release of hazardous substances from
waste sites, they do raise serious questions that should
be examined through more precise research.
There are numerous reasons why large gaps exist in our
ability to assess the health significance of environmental
exposures to chemicals in our environment Exposure
cannot be readily quantified by measuring body burdens
of contaminants, because  rapid metabolism of toxic
agents prevents measurable accumulation. Because of
the complexities  of toxin  uptake,  toxicologists do not
fully understand the relationships between environ-
mental exposure and body burden (i.e., the amount of a
toxin reaching  and interacting with biological targets).
Even  more problematic are the possible  antagonistic
and synergistic interactions that can possibly nullify pre-
dictions  based  on the toxicity of individual compounds.

Bioremediation involves increasing the numbers of pol-
lutant-degrading microorganisms to a level at which they
can have a  significant effect in  a  timely fashion.  This
increase in the microbial population also increases the
likelihood of human exposure to these microorganisms.
Because environmental organisms do have some po-
tential to cause adverse hearth effects,  researchers
must develop methods to  screen bioremediation micro-
organisms for the ability to induce adverse effects.

The Health Effects Research Program

To address the adverse health effects questions associ-
ated wrth bioremediation,  the EPA's Health Effects Re-
search Laboratory (HERL) has developed an integrated
program that addresses key issues. In collaboration with
other  EPA laboratories, HERL examines 1) the toxicity
of known HWS contaminants, their natural breakdown
products, and their bioremediation products; 2) the de-
velopment of methods to screen  microorganisms for
potential adverse health effects;  3) the potential for ad-
verse  effects when chemical/chemical and chemical/mi-
croorganism interactions occur; and 4) the development
of methods to better extrapolate toxicological bioassay
results to the understanding of potential human toxicity.
The program is carried out using known HWS pollutants,
samples from microcosm  studies that model  the biode-
                                                  40

-------
gradation within waste sites, and actual waste site sam-
ples. The HERL program attempts to coordinate its own
efforts with those of the other cooperating EPA labora-
tories and academic researchers funded through coop-
erative agreements.

HERL projects can be grouped into four categories: 1)
the infectivity and pathogenicity of  environmentally re-
leased microorganisms; 2) the toxicrty of metabolites of
environmental toxicants; 3) the toxicity of  products of
bioremediation; and 4) development of microbial con-
structs that decrease  the likelihood of adverse human
hearth effects.
This talk will give  a brief  overview of the specific re-
search ongoing within the HEfiL program,  how the re-
search is interrelated, and how the information coming
from this program could affect developing nsk assess-
ment methods.

References

1. Najem, G., I. Thind, M. Lavenhar, and  D.  Louna.
   1983. Gastrointestinal cancer mortality in New Jer-
   sey counties and the relationship with environmental
   variables. Int. J. Epidemiol. 12:276-289.

2. Griffith, J., R. Duncan, W. Riggan, and A. Pellom.
   1989. Cancer mortality in U.S. counties with hazard-
   ous waste sites  and ground water pollution. Arch.
   Environ. Health 44:69-74.
                                                  41

-------
                Construction of Noncolonizing E. Coli and P. Aeruginosa
                                             Paul S. Cohen
   Department of Biochemistry, Microbiology, and Molecular Genetics, University of Rhode Island, Kingston, Rl
 The wall of the mammalian large inte&ne consists of an
 epithelium containing brush border epithelial cells and
 specialized goblet cells, which secrete a relatively thick
 'up to 400 un), viscous mucus covering (1). The mucus
 layer contains mucin, a 2-MDa gel-forming glycoprotein,
 and a large number of smaller grycoproteins, proteins,
 glycolipids, and lipids (2-4). For many years, we have
 been interested in how Escherichia coli and Salmonella
 typhimurium colonize the large intestines  of mice and
 have come to the conclusion that growth in jhe mucus
 layer is essential (5). Moreover,  when  E. coli and S.
 typhimurium are grown in intestinal mucus  in vitro, they
 synthesize surface proteins that are not  synthesized
 during growth in normal  laboratory media (6). These
 results led us to envision  two approaches for obtaining
 strains of E coli that are perfectly healthy  when grown
 in normal laboratory media but are unable to colonize
 the  large intestines of mice. The first approach is to
 identify and mutate E coli genes that are necessary for
 growth or survival in mucus and determine whether such
 mutants are unable to colonize. The second strategy is
 to identify major nutrients, for growth of £ coli in mucus,
 isolate mutants unable to utilize these  nutrients, and
 determine whether such mutants are unable to colonize.
 This report explains how we have been successful in
 both approaches with £  coli and have now obtained
 strains that are unable to colonize but  that are com-
 pletely healthy in the laboratory. These strains should be
 as effective as their parents for gene cloning yet more
 effectiva for containment of rec^mbinant DMA.

 Technological exploitation of modem genetic techniques
 now holds great promise  for use of members of  the
 genus Pseudomonas for environmental purposes (e.g.,
 as agricultural biopesticides [7], as detoxifiers of chemi-
 cals  [8], and in prevention of ice nucleation  on plants
 [9]). For obvious reasons, the i  ains to be released into
the environment must be strong, competitive organisms.
 Unfortunately, strong, competitive  pseudomonads can
be opportunistically pathogenic (10,11).  Human expo-
sure to these microorganisms may occur in the agricul-
tural   or  industrial   setting   during production   or
application. Because a high concentration of these mi-
croorganisms may be found in the air and water, expo-
sure and subsequent disease  may occur  through
inhalation and ingestion. Clearly, strong, competitive
Pseudomonas strains should be constructed  mat are
unable to colonize  the lung tissue or the intestines of
humans and animals to minimize the possibility of op-
portunistic infections resulting  in debilitating disease.
This report explains our initial attempts at obtaining such
strains using the approaches outlined above for E. coli.

Background

E.Coll

E. coli F-18 was isolated from the feces of a  heslthy
human in 1977 and is an  excellent colonizer of the
streptomycin-treated mouse large intestine. Its serotype
is rough:K1:H5. £  coli F-18CoC, a poor colonizing de-
rivative of £ coli F-18, contains  all the £ coli F-18
plasmids, and its serotype is also rough:K1:H5. These
strains were used in experiments designed to determine
why £ coli F-18CoP is a poor colonizer and to identify
major nutrients required for successful £ coli coloniza-
tion of the mouse large intestine.

Pseudomonas Aeruginosa

P. aenjginosa AC869 is an environmental strain that has
been engineered to utilize 3,5-dichlorobenzoate as the
sole source of carbon and energy (11)  but which has
been found to be pathogenic for mice when adminis-
tered intranasally (11). This strain was used in experi-
ments to determine changes associated with growth in
mouse lung and cecal mucus preparations in vitro.

Results

ECo//

£ coli F-18 DNA was randomly cloned into £ coli F-18
Cor using the plasmid pRLB2. The entire bank was fed
to three streptomycin-treated mice, and all three mice
                                                  42

-------
selected the same clone which contained a 6.5 kb insert.
This insert increased the  colonizing  ability  of E. coli
F-18COI" approximately 1-million-fold.  After subcloning
and sequencing, we identified the gene responsible for
the observed increased  colonizing ability: /euX, which
encodes a leucine tRNA specific for  the rare leucine
codon UUG. An £ co//K-12 derivative,  E. co//XAc supP,
contains a defective leuX gene. This strain was found to
be unable to colonize the large intestines of streptomy-
cin-treated mice; i.e., mice fed 1010 colony forming units
(CPU)  were essentially  free of  the strain by  Day  11
postfeeding. In  contrast,  streptomycin-treated mice fed
10'°CFU of E. coliXAc supP containing the cloned /euX
gene colonized indefinitely at 107  CPU per gram  of
feces. Here, then, is  an  E. coli K-12 strain that is per-
fectty healthy when grown in  normal  laboratory media
but is unable to colonize the mouse intestine.

Glucuronate, a  major carbohydrate in  mouse cecal mu-
cus, i.er, 0.6 percent by dry weight (12), is metabolized
in E. coli via the Ashwell pathway (13). Mutants unable
to grow using glucuronate as the sole  source of carbon
were isolated after mini-Tn10  mutagenesis. One of the
mutants was unable to metabolize glucuronate, glucon-
ate, and galacturonate,  suggesting that it was lacking
2-keto-3-deoxy-6-phosphogluconic    aldolase    (EC
42.1.14), an enzyme encoded  by the eda gene (14).
The mutant eda gene was transduced into wild-type  E.
coli K-12, and the E. coli F-18  eda' strain and the E. coli
K-12 eda' strain were each fed to streptomycin-treated
mice (I0'° CPU per mouse). Both strains were essen-
tially eliminated from the  mouse intestine  by Day 9
postfeeding. When the eda' mutants  were comple-
mented with the previously  cloned eda" gene, both
strains colonized indefinitely at between 10* CPU and
103 CPU per gram of feces. We are presentty construct-
ing E. coli F-18 and  E. coii  K-12  supP~ eda' double
mutants to determine whether such mutants are even
more rapidly eliminated from the mouse large intestine.

P. Aemginosa
Rabbit  antisera were raised against  P. aemginosa
AC869 grown in Luria broth,  mouse lung mucus, and
mouse cecaJ mucus.  P. aemginosa AC869 grown  in
these media were subjected to SDS-PAGE and im-
munoblottjng using the three different rabbit antisera as
probes. Surprisingly, the  major change in P. aemginosa
AC869 observed when grown  in  either mouse lung mu-
cus or cecal mucus was a huge increase in O-side chain
containing lipopolysaccharide  (LPS). In support of this
view, P. aemginosa AC869 grown in  Luria broth was
found to be untypeable with respect to LPS, whereas
the same strain grown in either  mouse lung mucus  or
cecal mucus was typed  as O6. [LPS serotyping was
kindly performed at the Statens Seruminstitut in Copen-
hagen, Denmark.)  This  finding  was of great interest,
since P. aemginosa strains without O-side chain on their
LPS are known to be serum sensitive, i.e., they are killed
by normal human serum (15). We are, therefore, pres-
ently attempting to isolate mutants of P. aeruginosa
AC869 that do not make O-side chains when grown m
either mouse lung mucus or cecal  mucus. It is hoped
that such mutants will be perfectly healthy when grown
in laboratory media, will remain capable of metabolizing
3,5-dichlorobenzoate, yet will  be nonpathogenic when
inoculated intranasally into mice.

Summary and Conclusions

The genes leuX and eda have been  shown to be critical
tor  E  coli  colonization  of  the streptomycin-treated
mouse large intestine. These findings have allowed us
to obtain E.  coli K-12 strains  that grow well  in normal
laboratory media but are unable to colonize the  strepto-
mycin-treated mouse large intestine. Moreover, these
strains are easily transformed with pBR322-based plas-
mids containing chromosomal DNA  inserts. Developing
healthy E. coli K-12 strains for recombinant DNA work
that will not colonize the human  intestine now appears
possible.

We have shown that P.  aemginosa AC869 synthesizes
more O-side chain (O6) when grown in either mouse
lung mucus or cecal mucus than  in Luria broth. Since P.
aemginosa strains that lack O-side chain are serum
sensitive, its seems likely that such mutants of P. aerugi-
nosa AC869 will be less pathogenic in the lungs of mice.
Experiments designed to test this hypothesis are cur-
rently in progress.

References

1. Neutra, M.R.. and J.F. Forstner. 1987. Gastrointesti-
   nal  mucus:   Synthesis, secretion, and function. In:
   Johnson, LR., ed. Physiology of the gastrointestinal
   tract, 2nd ed. New York, NY:  Ravan Press,  p. 975.

2. Kim, Y.S., A. Morita, S. Miura,  and B. Siddiqui. Struc-
   ture of glycoconjugates of  intestinal mucosal mem-
   branes. Role of bacterial adherence. In: Boedecker,
   E.G.. ed. Attachment of organisms to the gut mu-
   cosa. Vol. II. Boca Raton,  PL:  CRC Press, Inc.  p.
   99.

3. Allen, A. 1981. Structure and function of gastrointes-
   tinal mucus. In: Johnson, L.R., ed. Physiology of the
   gastrointestinal tract New York,  NY: Ravan Press.
   p. 617.

4. Slomiany, A., S. Yano, B.L.  Slomiany, and G.B.J.
   Glass. 1978. Lipid composition of the gastric mucus
   barrier in the rat. J. Biol. Chem. 253:3,785.

5. Cohen, P.S.,  B.A. McCormick,  D.P. Franklin,  R.L.
   Burghoff,  and  D.C.  Laux.  1991.  The role of large
   intestine mucus in colonization of the mouse large
   intestine by  Escherichia coli F-18 and Salmonella
                                                  43

-------
    intestine by Eschehchia coli F-18 and Salmonella
    typhimurium. In: Wadstrom, T., A.M. Svennerholm,
    H. Wolf-Watz, and P. Klemm, eds. Molecular patho-
    genesis of gastrointestinal infections.  New York,
    NY: Plenum Press, p. 29.
6.  McCormick, BA, D.C. Laux, and P.S. Cohen.  Un-
    published results.
7.  Obukowicz, M.G., F.J. Pertak, K. Kusano-Kretzmer,
    E.J. Mayer, S.L Bolten, and L.S. Watrud.  1986.
    Tn5-mediated integration of the dctta-endotoxin gene
    from Bacillus thuringiensis into  the chromosome of
    root-colonizing pseudomonads. J. Bacteriol. 168:982.

8.  Leahy, J.G., and R.R. Colwell. 1990. Microbial deg-
    radation of hydrocarbons in the environment Micro-
    btol. Rev. 54:305.

9.  LJndow, S.E. 1985. Ecology of Pseudomonas syrin-
    gae relevant to field use of Ice deletion mutants
    constructed  In  vitro  for plant frost control.  In:
    Halvorson, H.O., D. Pramer, and M. Rogul, eds.
    Engineered  organisms in the environment  Scien-
    tific issues. Washington, DC: American Society tor
    Microbiology, p. 23.

10.  George, S.E., M.J.  Kohan,  D.A. Whitehouse, J.P.
    Creason, and LD. Claxton.  1990.  Influence of  an-
    tibiotics on intestinal tract survival and translocation
    of environmental Pseudomonas species. Appl. En-
    viron. Microbiol. 56:1,559.

11.  George, S.E., M.J. Kohan, D.A. Whitehouse, J.P.
    Creason, C.Y. Kawanishi, R.L Sherwood, and L.D.
    Claxton. 1991. Distribution, clearance, and mortality
    of environmental pseudomonads in mice upon in-
    tranasal  exposure.   Appl.   Environ.   Microbiol.
    57:2,420.

12.  Krivan, H.C., and P.S. Cohen. Unpubl'shed results.

13.  Ashwell,  G. 1962.  Enzymes  of  glucuronic and
    galacturonic acid metabolism in bacteria. Methods
    Enzymol. 5:190.

14.  Falk,  P.,  H.L  Komberg, and E. McEvoy-Bowe.
    1971. Isolation and properties of Escherichia coli
    mutants defective in 2-keto 3-deoxy 6-phosphoglu-
    conate aldolase activity. FEBS Lett 19-225.

15.  Dasgupta, T., T.R. de Kievit, H. Masoud, E. Altman,
    J.C. Richards, I. Sadovskaya, D.P. Speert, and J.S.
    Lam. 1994. Characterization of lipopolysaccharide-
    deficient mutants of Pseudomonas aeruginosa de-
    rived  from  serotypes  03,  O5, and O6.  Infect.
    Immun. 62:809.
                                                44

-------
            Toxicant Generation and Removal During Crude Oil Degradation
                                             Linda E. Rudd
                               North Carolina State University, Raleigh, NC

                            Lairy D. Claxton, Virginia S. Houk, Ron W. Williams
                     U.S. Environmental Protection Agency, Research Triangle Park, NC

                                            Jerome J. Perry
                               North Carolina State University, Raleigh, NC
 As microorganisms are  promoted for  environmental
 bioremediatkxi efforts, the potential risk of adverse ef-
 fects of pollutant exposure to the microbes  must be
 assessed. Although fungi (1,2) and bacteria (3-5) de-
 grade hydrocarbons, the genotoxic consequences of
 deg/Tdation have not been addressed.  Bacterial spe-
 cies use enzyme systems to convert  hydrocarbons to
 metabolites -with increased toxicity (6-8) or to mineralize
 toxic compounds during metabolism (9). This study in-
 volves interactive use  of microbiaJ culture, analytical
 chemistry, and mutagenicity bioassays to investigate the
 genotoxicrty of the oil degradation process. Following
 degradation by two fungi,  Cunninghamella elegans and
 Penidllium zonatum M 0,11), crude oils of low, moderate,
 and high mutagenicity are tested tor their resulting mu-
 tagenic activities.
Methods
Results

Pennsylvania and Cook Inlet Alaska crude oils' myceiial
mat  weights are directly proportional to biologically
linked oil degradation. The fungi consistently form sturdy
mats with Pennsylvania crude; the Cook Inlet mat, how-
ever, is more fragile. Mat weights are not proportional to
West Texas sour crude utilization;  sturdy  mats are not
consistently produced by either organism even though
the oil is used as the sole carbon source. The loss of oil
mass is evidenced by a significant decrease  in C7 to
C20 hydrocarbons as  incubation time increases. Weath-
ered samples of the three oils do not exhibit changes in
mutagenic activity over time. The  mutagenicity of the
most potent oil, Pennsylvania crude, is significantly re-
duced following degradation bv either fungus (Table 1).
The activity of the weakly mutagenic West Texas crude
exhibits little change upon treatment (data not shown).
The nonmutagenic Cook Inlet Alaska crude oil becomes
mutagenic when incubatod with either fungus (Table 2).
C. elegans ATCC 36112 or P. zonatum ATCC 24353 was
inoculated into 500 ml L-Salts medium (12) with 5 mL
of crude oil. Flasks were incubated at 30°C for 4 to 30
days; at  2-day intervals, flasks were sacrificed, and
crude oil was extracted  with methyiene chloride by a
mocfification of the method used by Cemiglia (10,13). Oil
mass determinations were calculated from oil  residue
weights. Extracted oils were analyzed for conversion of
straight chnin  hydrocarbons by gas  chromatography
and for mutagenicity by the spiral  Salmonella assay
(14,15). Controls  included "weathered" (uninoculated)
oil flasks  and fungi grown on 2-percent glucose to test
for mutagenic products from fungal growth atone (fun-
gal mat controls').
Conclusion

The fungal species used in this study may convert crude
oil hydrocarbons to products more mutagenic than the
original compound. Further studies in progress address
effects of oxygenation, nitrogen and phosphorus enrich-
ments, and surfactant addition to the experimental system.
References

1.   Kirk. P.W., and  A.S. Gordon.  1988.  Hydrocarbon
    degradation by  filamentous marine  higher fungi
    Mycotogia 80(6):776-782.
                                                 45

-------
Table 1. Pennsylvania Crude (-i-t-f highly mutaganlc)
Organism           Incubrton (day*)         Mutagente Response
                                          % Biological Loss*
                         Mat Weight (3)
 C. stogans
 2
 4
 6
 8
10
12
14

 2
 4
 6
 8
10
12
14
     7%
     9%
     17%
     23%
     42%
     26%
     32%

     8%
     16%
     21%
     27%
     33%
     29%
     18%
     0
    0.2
    0.6
    0.6
    1.0
    0.4
    O.S

    0
    0
    0.4
    0.5
    0.3
    0.4
    0.3
 •Biological Loss » Amount of oil used by fungus (corrected for procedural nonbiological oil loss)
 Table 2.  Alaska Crwie (- nonmutagenlc)
     Organism           Ineubatlon (day*)
                   Mutagenlc Response
% Biological Loss
Mat Weighty!
    C, ebgans
    P. zonttu/n
    2
    4
    6
    8
   10
   1<
   14

    2
    4
    6
    8
   10
   12
   14
      4%
      4%
      19%
      19%
      18%
      18%
      16%

      5%
      13%
      16%
      28%
      24%
      27%
      24%
     0
     0
     0.1.
     0.1
     0.1
     0.1
     0.1

     0
     0
     0.1
     0.2
     0.2
     0.2
     0.1
2.  Jobson. A., F.D. Cook, and D.W.S. WestJake. 1972.
    Microbial  utilization of crude oil. Appl. Microbiol.
    23(6):1,082-1,089.
3.  Cemiglia, C.E. 1992. Biodegradation o. polycyclic
    aromatic hydrocarbons. Biodegradation 3:351-368.
4.  Perry, J.J. 1968. Substrate specificity in hydrocar-
    bon utilizing  microorganisms. Antonie  van  Leeu-
    wenhoek 34:27-36.
                                5.   WaJker, J.D.,  L Petrakis, and R.R. Colwell. 1
                                    Comparison of the biodegradability of crude
                                    fuel oils. Can. J. Microbiol. 22:598-602.
                                6.   Gibson, D.T.,  V. Mahadevan, D.M. Jerina, H,Y
                                    and H.J.C. Yeh. 1975. Oxidation of the carcinog
                                    benzo[alpyrene and benzo[a]anthracene to JJ
                                    drodiols by a bacterium.  Science 189:295-297
                                7.   Mkjdaugh,  D.P.. S.M. Resnick,  S.E.  Lane,
                                    Heard,  and  J.G.  Mueller.   1993.
                                                      46

-------
    assessment of biodegraded pentachlorophenol: Mi-
    crotox™ and fish embryos. Arch. Environ. Contam.
    Toxicol. 24:165-172.

3.   Liu, D..  R.J. Ma'guire, G.J. Pacepavicius,  and E.
    Nagy. 1992. Microbial degradation of polycydic aro-
    matic hydrocarbons and polycyclic aromatic nitro-
    gen heterocyclics. Environ. Toxicol. Water Qual.
    7(4):355-372.

 9. Burback, B.L, and J.J. Perry. 1993. Biodegradation
    and  biotransformation  of  ground-water pollutant
    mixtures by Mycobacterium vaccae. Appl. Environ.
    Microbiol. 59(4): 1,025-1.029.
10. Cemiglia, C.E., and J.J. Perry. 1973. Crude oil deg-
    radation by microorganisms isolated from the ma-
    rine environment Zeitschrifl fur Allg. Mikrobiotogie
    13(4)^99-306.
11.  Hodges, C.S., and JJ. Perry. 1973. A new species
    of  Eupenicillium from soil.  Mycologia 65(3):697-
    702.
12.  Leadbetter, E.R., and J.W. Foster. 1958. Studies on
    some methane-utilizing bacteria. Arch.  Mikrobiol.
    30:91-118.
13.  Cerniglia, C.E. 1975. Oxidation and assimilation of
    hydrocarbons by microorganisms isolated from the
    marine environment. Dissertation. Raleigh:  North
    Carolina State University.
14.  Maron, D., and B.N. Ames. 1983. Revised methods
    for the Salmonella mutagenicity test. Mutation Res.
    113:173-212.
15.  Houk, V.S., S.  Schalkowsky, and  L.D.  Claxton.
    1989. Development and validation of the spiral Sal-
    monella assay: An automated approach to bacterial
    mutagenicity testing. Mutation Res. 223:49-64.
                                                  47

-------
  Potentiation of 2,6-Dinitrotoluene Bioactivation by Atrazine In Fischer 344 Rats
             S. Elizabeth George, Robert W. Chadwick, Michael J. Kohan, and Joycelyn C. Allison
     U.S. Environmental Protection Agency, Health Effects Research Laboratory, Research Triangle Park, NC

                                  Sarah H. Warren and Ron W. Williams
                        Integrated Laboratory Systems, Research Triangle Park, NC

                                           Larry D. Claxton
     U.S. Environmental Protection Agency, Health Effects Research Laboratory, Research Triangle Park, NC
Because of widespread use, pesticides often are found
as co-pollutants at hazardous waste sites and other
sites  contaminated  by  xenobiotics. The  herbicide
atrazine  is used as a weed control agent during the
cultivation of food crops and is found frequently as a
ground-water contaminant To study atrazine as a co-
pollutant this study explored the effect of atrazine treat-
ment   on  the  bioactivation  of   the  promutagen
2,6-dinitrotoluene (2,6-DNT). For 5 weeks, male Fischer
344 rats (21 d) were administered  p.o. 50 mg/kg of
atrazine. At 1.3, and 5 weeks, both control and atrazine-
pretreated rats were administered 75 mg/kg of 2,6-DNT
by gavage and were placed into metabolism cages for
urine collection. Following urine concentration, a micro-
suspension modification of the Salmonella assay with
and  without metabolic activation  was used  to detect
urinary mutagens. No  significant  change in  mutagen
excretion was observed in  atrazine-pretreated rats. A
significant increase, however, was detected  in direct-
acting urine mutagens from rats receiving atrazine and
2,6-DNT at Week 1 (359 ±68 revertants/mL versus 621
±96 revertants/mL) and Week 5 (278 ±46 revertants/mL
versus 667 ±109 revertants/mL) of treatment. Urinary
mutagericrry was accompanied by an increase in small
intestinal nctroreductase activity. At Week 5, elevations
in  large intestine nitroreductase and B-glucuronidase
were observed. This study suggests that atrazine poten-
tiates the metabolism and excretion of the mutagenic
metabolites of 2,6-DNT by modifying the intestinal en-
zymes responsible for promutagen bioactivation.
                                                 48

-------
              Effects of Lactobacillus reuteri on Intestinal Colonization of
                                     Bioremediatlon Agents
                                              Mitra Ruzat
                   Department of Microbiology, North Carolina State University, Raleigh, NC

                                          S. Elizabeth George
     U.S. Environmental Protection Agency, Health Effects Research Laboratory, Research Triangle Park, NC

                                          Walter J. Dobrogosz
                   Department of Microbiology, North Carolina State University, Raleigh, NC
 Lactobacillus reuteri is the predominant heterofermen-
 tative species of Lactobacillus inhabiting the gastroin-
 testinal (Gl) tract of humans, swine, poultry, rodents, and
 a number of other animals (1). Studies on chicks and
 poults have shown that  oral (probiotic) treatment of
 flocks at hatch with viable, host-specific L reuteri prior
 to challenge at Day 1 posthatch wtth  S. typhimurium
 reduces mortality by 50 percent to 75 percent compared
 with untreated flocks (2). L reuteri  is unique a.nong
 bacteria in its ability to produce and secrete the potent,
 broad-spectrum antimicrobial agent reuterin when incu-
 bated in the presence of gtycerol under physiological
 conditions similar to those which exist in the Gl tract
 (3,4)  Reuterin has been purified, chemically charac-
 terized, and identified  as an equilibrium mixture of
 monomeric,  nydrated monomeric, and cyclic dimeric
 forms of 3-hydroxypropionaldehyde (5,6).

 The environmental release of naturally occurring, mu-
 tant and  recombinant microorganisms has prompted
 questions concerning human health and environmental
 effects (7,8). To date, a variety of microbes have been
 released into the environment for many uses. Currently,
 investigators are engineering microorganisms, primarily
 pseudomonads, for their ability to degrade hazardous
 environmental  contaminants  such as pentachlorophe-
 nol,  2,4,5-trichlorophenoxyacetate,  chtorobenzoates,
 a 1 trichtoroetrtylene. Pseudomonas  spp.,  however,
 have long been recognized as opportunistic pathogens,
 readily occurring in serious  secondary  infections, and
they have been linked to major infections in immunosup-
pressed and leukemia patients as well as those treated
with antibiotics (9-11). Because  of the clinical signifi-
cance of Pseudomonas spp., their potential health ef-
fects have been studied in terms of their ability to com-
pete and  survive in a  CD-1  mouse  model system
(12,13). The effects of antibiotics on their survival and
transtocation to other organs also have been investi-
gated. Results from these studies indicate that environ-
mental pseudomonads can survive in the Gl tract for up
to 14 days, where they can alter the normal microbiota.
Their translocation to the spleen and/or liver also occurs,
indicating the potential for a systemic infection (14,15).
This research was undertaken to determine if L. reuteri
prophylaxis could mitigate the pathogenic effects of
these Pseudomonas spp. in the mouse model system.

Materials and Methods

Bacterial Strains

Three Pseudomonas aeruginosa strains were used in
this  study.  Strain   BC16  degrades  polychlorinated
biphenyl, strain AC869 degrades 3,5-dichlorobenzoate,
and strain  PAO is a clinical isolate. Four mouse-specific
L reuteri strains were used.

Animals

Thirty-day-old CD-1 male mice were used in this study.
These animals were administered /.. reuteri (109 colony-
forming units (CFU)/mL) in sterilized  water daily for 5
days prior to Pseudomonas administration by gavage
(one group 108 CPU and the other group 109 CPU) and
thereafter  during the entire experiment. Control mice
were given only sterilized water. On Day  2 and  Day 7
after the Pseudomonas administrations,  the animals
were sacrificed, and their livers and ceca were analyzed
for presence of L. reuteri and  Pseudomonas spp.
                                                 49

-------
 Detect/on of L reuteri and Pseudomonas spp.

 Mice were sacrificed by C02 asphyxiation. Ceca and
 livers were removed aseptically and homogenized in
 5 mL PBS buffer. Homogenate dilutions were made in
 buffer, and duplicate platings were carried out on Lacto-
 bacillus selection (LBS) agar and Pseudomonas isola-
 tion agar (PIA).  The  LBS  medium  was  used  to
 enumerate the total gut and liver population of lactoba-
 cilli. The subpopulation of L. reuteri colonies on appro-
 priately diluted plates is identified based on the ability of
 L reuteri colonies to convert glycerol to reuterin under
 anaerobic conditions. The  PIA plates were  used for
 Pseudomonas spp. detection in livers and ceca.

 Results and Discussion

 Animals that were treated  with  P.  aemginosa  strains
 BC16 and AC869 and L reuteri were cleared of the
 infectious agent in 7 days. Of animals that were not
 treated with L reuteri,  55 percent and 33 percent re-
 mained infected at that time with P. aemginosa strains
 BC16 and AC869, respectively.  When the mice were
 given  109 cells of P. aemginosa AC869 by Day 7, 83
 percent remained infected compared with a 50-percent
 infection rate in the  L reuteri treated group. Animals
 treated with P. aemginosa PAO (109 cells per mouse) in
 the absence of L reuteri were 75-percent infected by
 Day 7; those treated with L reuteri were only 50-percent
 infected.

 Some indigenous lactobacilli have been shown to inhibit
 colonization of pathogenic bacteria, particularly in the
 small  intestine, by means  of what  has been termed
 colonization resistance (CR) or competitive exclusion
 (CE) (16). Neither the mechanism(s) underlying this
 phenomenon nor the protective effect of L reuteri on the
 Pseudomonas infections described in this report is fully
 understood. Our research has indicated, however, that
 1) L reuteri prophylaxis is beneficial to the host animal's
 health and 2) this treatment  could  have applications
 concerning the protection of animals against Pseudo-
 monas spp. Preliminary studies (17) indicate that
 L. reuteri's efficacy  in this regard could be based on
 its ability to stimulate a protective immune response
 to P. aemginosa infections.

 References

 1.  Kandler, 0., and N. Weiss  1986. Regular gram-
    positive nonsporing rods.  In: Sneath, P.H.A., M.E.
    Sharpe, and J.G. Holt, eds. Sergey's manual  of
    systematic bacteriology, Vol. 2. pp. 1,208-1,234.

2.  Casas, I.A., F.W. Edens, W.J. Dobrogosz, and C.R.
    Parkhurst 1993. Performance of GAIAfeed and
    GAIAspray: A Lactobadllus reuteri based probiotic
    tor poultry.  In: Jensen, J.F., M.H. Hinton,  and
    R.W.A.W.  Mulder, eds.  Prevention and control  of
    potentially pathogenic microorganisms in  poultry
    and poultry meat products. Proceedings 12, FLAIR
    No. 6. Probiotics ana Pathogenicity, DLO Centre for
    Poultry Research and Informational Services. The
    Netherlands: Beekbergen. pp. 63-71.

3.   Axelsson, L.T., T.C. Chung, S.E. Lindgren, and W.J.
    Dobrogosz. 1989. Production of a broad spectrum
    antimicrobial substance by Lactobadllus reuteri. Mi-
    crobial Ecol. Health Dis.  2:131-136.

4.   Chung, T.C., L.T. Axelsson, S.E. Lindgren, and W.J.
    Dobrogosz. 1989. In vitro studies on reuterin syn-
    thesis  by Lactobadllus  reuteri. Microbial  Ecol.
    Health Dis. 2:137-144.

5.   Talarico,  T.L.,  LA.  Casas, T.C.  Chung, and  W.J.
    Dobrogosz. 1989. Production and isolation of reu-
    terin: A growth inhibitor produced by Lactobadllus
    reuteri. Antimicrob. Agents Chemotfier. 32:1,854-
    1,858.

6.   Talarico, T.L, and W.J. Dobrogosz. 1989. Chemical
    characterization of an antimicrobial substance pro-
    duced  by   Lactobadllus  reuteri.  Antimicrobial,
    Agents Chemother. 33:674-679.

7.   Franklin,  C.A.  1988. Modem biotechnology: A re-
    view of current regulatory status and identification
    of research  and regulatory needs. Toxicol.  Ind.
    Health 4:91-105.

8.   Rissler, J.F. 1984. Research needs for biotic envi-
    ronmental effect of genetically engineered microor-
    ganisms. Recomb.  DNA Tech. Bull. 7:20-30.

9.   Guiot, S.F.L., J.W.M. van der  Meer,  and R. van
    Furth. 1981. Selective antimicrobial modulation of
    human microbial flora: Infection prevention in pa-
    tients with decreased host defense mechanisms by
    selective  elimination of potentially pathogenic  bac-
    teria. J. Infec. Dis. 143:644-654.

10. Schimpff, S.C.  1980.  Infection prevention  during
    profound granulocytopenia: New approaches to ali-
    mentary canal  microbial  suppression. Ann.  Intern.
    Med. 93:358-361.

11. Barttett, J.G. 1979. Antibiotic-associated pseudomem-
    branous colitis. Rev.  Infect Dis. 1:530-538.

12. George, S.E.,  M.J. Kohan, D.B.  Walsh, and  LD.
    Claxton. 1989. Acute colonization of polychlorinated
    biphenyl-degrading pseudomonads in  the mouse
    intestinal tract: Comparison of single and multiple
    exposures. Environ. Toxicol. Chem.  8:123-131.

13. George, S.E., M.J. Kohan, D.B. Walsh, A.G. Stead,
    and L.D.  Claxton. 1989.  Polychlorinated biphenyl-
    degrading pseudomonads: Survival in mouse intes-
    tines and competition with normal flora. J. Toxicol.
    Environ. Health. 26:19-37.
                                                  50

-------
14. George, S.E., M.J. Kohan, O.J. Wtiitehouse, J.R   16. Fuller, R. ed. 1992. Probiotics: The scientific basis.

   Creason, and LD. Claxton. 1990. Influence of an-       NY: Chapman and Hall.

   tibiotJcs on intestinal tract survival and transition       Dobrogosz, WJ., H.j. Dunham, F.W. Edens, and


        'nm              naSS6CieS- **'
                                                          ,    .,  ..        ,  ..      ,

    r           H                  - *"*'          '-A- Casas.  1992. LtOobacUus  reuter,  ,rrrruno-
   v,ron. MicrobKX. (In press)                          modulation  of  stressor-associated diseases  in
15 ar ^ D^T' n st^d,tLD-
   Claxton. 1989. Effect of ampiallin-mduced altera-       .    5*28-29
   tions in murine intestinal microbiota on the survival        u^u
   and  competition  of  environmentally   released
   pseudomonads. Fund. Appl. Toxicol. 13:670-680.
                                           51

-------
                          Section Three
                         Field Research
 Field research is essential for evaluating the performance of full-scale bioremedia-
 tion  processes and for conducting accelerated testing on  technologies that are
 appropriate for scaleoXip application. For example, problems associated with the
 use of bacteria used in the laboratory include optimizing the activity of the organism
 under site conditions and defining the risks associated wrth the introduction of a
 non-native microorganism. The objective of this level of research is to demonstrate
 that  the particular bioremediation process performs as expected in  the field.
 Researchers at the symposium provided information on  several  ongoing field
 experiments.
 F«
-------
  Field-Scale Study of In Situ Bioremediatlon of TCE-Contaminated Ground Water
                                and Planned Bloaugmentatlon
                                   Perry L McCarty and Gary Hopkins
           Westom Region Hazardous Substance Research Center, Stanford University, Stanford, CA
 Trichtoroethytene (TCE) and other lessee halogenated
 ethenes are biodegradable through aerjbic co-metabo-
 lism.  Here,  microorganisms that possess oxygenases
 for initiating the oxidation of either aliphatic or aromatic
 hydrocarbons or ammonia fortuitously can oxidize the
 chlorinated alkenes to unstable epoxides. The epoxides
 degrade  further to inorganic  end  products  through a
 combination of chemical and biological transformations.
 To cany  out in situ bfodagradation of such chlorinated
 ethenes  in ground wster,  the appropriate aliphatic or
 aromatic hydrocarbon or ammonia must be added to the
 ground water as a substrate  both to grow a sufficient
 population of the desired organisms and to supply the
 energy required  for maintaining activity of the oxy-
 genase. Paid studies to evaluate the potential of aerobic
 co-mefsooJism  of TCE and other  chlorinated alkenes
 have  been conducted at the Moffett Naval Air Station in
 Mountain view, California,  since 1985 (1-3). Methane,
 phenol, and toluene  now have been added  to ground
 water at this site to determine their effectiveness as
 primary substrates for chlorinated ethene degradation.

 The above  studies have shown  the effectiveness of
 microorganisms indigenous to the subsurface environ-
 ment  at Moffett Field for degrading chlorinated alkenes.
 One potential problem in  attempting to translate the
 results at the Moffett Field site to other field sites is that
 the same primary substrates may  not stimulate the
 growth of microorganisms with  similar effectiveness.
 Many different microorganisms can grow on the primary
 substrates found effective for  TCE co-metabolism, but
 their effectiveness for this purpose can vary  widely- To
 better ensure a high degree of effectiveness, an ability
 to apply bioaugmentation successfully with  organisms
 known to be capable of high .utes of biotransformation
 is highly desirable. In addition,  phenol and toluene, sub-
 strates found to be  highly effective  as primary sub-
strates,  are also  hazardous  chemicals.   Use  of
microorganisms that can use less harardous chemicals
as primary substrates while maintaining  a high degree
of effectiveness is desirable. Efforts to evaluate bioaug-
mentation at the Moffett Held site now are under way.

A summary of the results from the Moffett Reid test site
using indigenous organisms is described below, as are
plans for in situ bioaugmentation.

Moffett Field Test Results

Over the past several years, methane and phenol have
been evaluated for their effectiveness in stimulating
aerobic co-metabolic degra
-------
Table 1  Summary of the Effectiveness of afferent Primary Substrates for In Situ Co-metabollo Blodegradatlon of
        Chlorinated Ethenea at the Moffett Raid Taat Slta

                                                         Primary Subatrataa

Primary Substrate Concentrations (mg/L)
Dissolved Oxygen Concentrations (mg/L)
Methane
6.6
26
Pr-oo.
12.5
30
Toluene
9
28
 Target Compound*
                                      Percent Removal
                                                          Percent Removal
                                                                             Percent Removal
vc
1,1-OCE
M.2-OCE
C-1.2-OCE
TC£ '
95
NE
92
42
19
>98
54
73
92
94
NE
NE
75
>98
93
 NE - Not evaluated
 1 ug/L Here, sufficient oxygen was present for effective
 oxidation. The EPA maximum contaminant level (MCL)
 and maximum contaminant level goal (MCLG) for tolu-
 ene in drinking water is 1,000 ug/L, and the taste and
 odor threshold is in the  range of 20 u.g/L to 40 u.g/L.
 Thus, the low levels achieved after addition to ground
 water in the field suggest that no hazard from toluene
 addition should remain if sufficient oxygen is  present.
 Phenol,  while known to have toxicity similar to that of
 toluene, has no established MCL value, and so its ap-
 propriate safe limits can only be estimated.

 Bioaugmentatlon

 A cooperative study is now under way between the EPA
 Gulf  Breeze  Environmental Research Laboratory, the
 Michigan Center for Vicrobial Ecology at Michigan State
 University, the University of Western Florida,  and the
 Western Region Hazardous Substance Research Cen-
 ter at Stanford University to evaluate the possibility of
 btoaugmentation for enhanced in situ co-metabolic deg-
 radation of TCE. Moffett  Field will be used as the test
 site for this study. The objectives of this study are 1) to
 evaluate at field scale the potential of bioaugmentation
 to enhance and improve in situ bioremediation of ground
 water contaminated with TCE; 2) to determine the move-
 ment fate, and effectiveness of introduced microorgan-
 isms  in  an aquifer;  3)  to determine and  evaluate
 methods for maintaining  dominance of introduced or-
 ganisms over indigenous  organisms; 4) to evaluate en-
 vironmental and ecological factors that affect organism
 dominance in aquifers during in situ bioremediation; and
 5) to evaluate the applicability of molecular tools in the
 monitoring, operation, and control of in situ bioremedia-
tion systems.

A new test leg is being constructed at the Moffett Reid
test site  for this evaluation. Soil samples have been
collected from this test leg for use in laboratory studies
to determine the best approach for carrying out the field
bioaugmentation studies and to maintain dominance by
the introduced  microorganisms. Also,  the laboratory
studies will be used to develop and evaluate molecular
tools for characterizing the phenol and toluene degrad-
ing populations  present, and the fate of the introduced
microorganisms. Possible microorganisms for introduc-
tion also are being evaluated in these laboratory studies.
These organisms include Pseudomonas cepacia G4, an
organism that grows on either toluene or phenol and is
known for its  high effectiveness  in degrading TCE, and
the PR1 mutant of this organism, which  has a constitu-
tive oxygenase  that is induced even when it grows on
nonhazardous substrates such as lactate.

The laboratory studies will be conducted during the first
ongoing  year of  this study.  Field  implementation is
planned for the second year of study. The different insti-
tutions involved in this study will share in the evaluation
of the effectiveness of  bioaugmentation. The Moffett
FiekJ  site offers a good opportunity in  general  for a
comparative evaluation of different approaches to in situ
biodegradation of chlorinated aliphatic compounds, and
offers promise for evaluating bioaugmentation as well.
Acknowledgments

The studies  reported  here  were supported  by EPA
through the Robert S.  Kerr and Gulf Breeze  Environ-
mental  Research  Laboratories, the Bic  ystems Pro-
gram, and  the Western Region Hazardous Substance
Research Center, and  by the U.S. Department of En-
ergy. These agencies have not reviewed this publica-
tion, and no official endorsements by them should be
inferred.
                                                  55

-------
References                                          effects on pilot field-scale in situ ground-water biore-
1  Mrt^tin. r- n  i  c^™*  ,^ DI  Mozart,, 1000       mediation by phenol-oxidizing microorganisms. En-
1. Hopkins, G.D., L Sempnni, and PL McCarty. 1993.            .   Technol 27M2V2 542-2 547
   Microcosm and in situ field studies of enhanced      viron' **'• recnnoL 27(12).2,54,£ ^,547.
   bkjfransfomiation of tricnioroethykine by phenol-util-    3. Semprini, L, P.V. Roberts, G.D.  Hopkins, and PL.
   izing  microorganisms.  Appl.  Environ.  Microbiol.      McCarty. 1990. Afield evaluation of in s/ft/biodegra-
   59(7):2,277-2,285.                                    dation of chlorinated ethenes:  Part 2.  Results of
2. Hopkins, G.D., J.  Munakata. L Semprini,  and PL.      biosMmulatlon  and ^transformation experiments.
   McCarty.   1993.   Trichloroethy.ene  concentration      Ground Water 28:715-727.
                                                 56

-------
    Geochemistry and Microbial Ecology of Reductive Dechlorination of PCE and
                                  TCE in Subsurface Material
                                  Guy W. Sewell and Candida C. West
       U.S. Environmental Protection Agency, Robert S. Kerr Environmental Research Laboratory, Ada, OK

                                 Susan A. Gibson and William G. Lyon
 ManTech Environmental Research Services Corp., Robert S. Kerr Environmental Research Laboratory, Ada, OK

                                            Hugh Russell
       U.S. Environmental Protection Agency, Robert S. Kerr Environmental Research Laboratory, Ada, OK
 Chloroethenes are among the  most common organic
 contaminants of ground water.  In the subsurface and
 other anaerobic environments, they can be transformed
 through a  biologically mediated,  step-wise,  reductive
 removal of chloride ions, known as reductive dechlori-
 nation. Potentially this process can lead to nonchlori-
 nated products that  are  environmentally  acceptable.
 Unfortunately, more mobile and toxic daughter products
 are  intermediates.  If the process 'stalls," as it often
 seems to in the subsurface, before reaching nonchtori-
 nated end  products, the reductive dechlorination proc-
 ess   may  increase  potential  risks to  human  and
 environmental health. Thus, the reductive dechlorination
 process can exacerbate or attenuate the problems cre-
 ated  by the release of chloroethenes such as trichlc-
 roethylene  (TCE) or  tetrachloroethylene (PCE) to the
 subsurface and ground-water environments. In these
 studies, we have attempted to identify the environmental
 parameters that control  the onset  and extent of the
 dechlorination activity.

 Three areas of investigation have been the focus of
 efforts by Robert S. Kerr Environmental Research Labo-
 ratory researchers on the reductive dechlorination of
 chloroethenes. The first is the effects of alternate elec-
 tron acceptors, commonry found in the subsurface, on
 the reductive dechlorination  process. The  second  is to
 develop a conceptual understanding of microbial popu-
 lations and  interactions that carry out the process. The
third  is directed toward identifying organic compounds
that can serve as sources of reducing equivalents for the
dechlorination process under native conditions or  as a
component  of an active biotreatment application.
Results and Discussion

Saturated sandy subsurface sediments from near the
municipal landfill in Norman, Oklahoma, were collected
and used as the  test material in  these studies. Trie
subsurface  environment from  which the material was
collected is  impacted by landfill leachate and classified
as methanogenic. This material has been previously
shown to contain  microbial populations capable of re-
ductively dechlorinating PCE (1). Figure 1 demonstrates
the microorganisms' capacity for complete dechlorina-
tion of PCE in long-term batch enrichments.

Alternate Electron Acceptor Studies

Under anaerobic  conditions,  the oxidation of organic
compounds is linked to the reduction of electron ac-
ceptors other than oxygen. In the  subsurface may be
present many different electron acceptors, such as ni-
trate, ferric iron, sutfate, carbonate, or organic contami-
nants, such as chloroethenes. If multiple acceptors are
present in  physiologically acceptable concentrations,
then the predominant  terminal  oxidation process is
linked to the acceptor £*t will yield  the most energy. As
this acceptor becomes  limiting, the acceptor with the
next highest energy yield is utilized, and so on, until the
acceptor with the  lowest energy yield is utilized, which
is  usually  carbonate (methanogenesis). Previous re-
search  suggests  that  in  the subsurface,  reductive
dechlorination may be only a minor fate (less than 10
percent) for the reducing equivalents generated during
the anaerobic oxidation reactions (2).  Whether this
noncompetitiveness is  because of the physiological
                                                 57

-------
                        PCE
                        Vinyl Chtorid*
                        Ethana
                1390       1400

                    Tim* (day*)
                                    1450
                                               1500
Figure 1.  Production of etnene and vhiyt chloride from re-
         peeled PCE spikes over time In long-term Norman
         Landfill sediment enrichment*. TCE and DCE Inter-
 limitation of the organisms  involved, the low potential
 energy of reactions coupled to reductive dechlorination,
 or as-yet-unrecognized environmental parameters is un-
 ciear.

 Laboratory microcosm studies indicated that nitrate was
 extremely inhibitory to the reductive dechlorination proc-
 ess (Figure 2). In  the presence of  nitrate, oxidizable
 organic carbon is quickly utilized by microorganisms in
 the test material. Whether this was the only mechanism
 of inhibition was unclear. Sulfate appeared to be partially
 inhibitory  under   the   conditions   tested.   Again,
 competition for electron donor appeared to  be the
 mechanism of inhibition. In experiments with different
 initial concentrations  of sulfate,  significant dechlorina-
 tion activity appeared  only after sulfate concentration fell
 below 400 ujirf (Figure 3).

 Mlcroblal Process Studies

 Formation of a conceptual model is the first step in the
 development  of valid mathematical descriptions of in
 situ reductive dechlorination processes. In an effort to
 define the metabolic processes involved in these reac-
 tions and to enhance  our understanding of the ecology
 of the reductive dechlorination process, we have studied
 the effects of metabolic  inhibitors (2-bromoethanesul-
 fonic  acid [BESA], molybdate.  and vancomycin)  on
 butyrate, ethanol, and formate driven reductive dechlori-
 iiation of PCE  in aquifer microcosms. Molybdate  (5 mM)
 and BESA (1  mM and  10  mM) are used as  specific
 inhibitors  of   sulfate-reduction and  methanogenesis,
 respectively. Vancomycin (100 ppm) is  used as a gen-
eral eubacterial  inhibitor. Molybdate appears to be  an
effective inhibitor of reductive  dechlorination under the
                                                                                          PCE (mathanogonte)
                                                                                          PCE(10mM3ulfatB)
                                                                                          PCE OOmM Nltrals)
                                                                                          TCE (malwiogsnic)
                                                                   20
                                                                           40       90
                                                                           Tim* (daya)
                                                                                           ao
                                                                                                   100
                                                       Figure 2. Effects of nltrat* and sulfate on the deehlortnatlon of
                                                               PCE versus time In Norman Landfill microcosms. Vat-
                                                               IM* are an average of five reptlcants. DCE Intermedl-
                                                               atea are not shown.
                                                      conditions tested. BESA completely inhibited dechlori-
                                                      nation in microcosms at 10 mM, but only partially inhib-
                                                      ited  activity  at  1  mM  (Table  1).  The  results  of
                                                      experiments,  such as those shown in Table 1, suggest
                                                      that the dechlorinating organisms access the same pool
                                                      of reducing  equivalents  as  the terminal   oxidizing
                                                      organisms.


                                                      Electron Donor Studies

                                                      We have shown in the laboratory that the availability of
                                                      a suitable electron donor is essential for denalogenation
                                                      of PCE  and TCE to occur  at appreciable rates in oligo-
                                                      trophic  subsurface environments  (3,4). We  and other
                                                      groups have identified a wide variety of organic electron
                                                      donors  that  can  drive biodehalogenation   of  chlo-
                                                      roethenes (2-9). Conceptually, any organic substance
                                                      capable of being catabolized  under  anaerobic condi-
                                                      tions  should  be able to support or "drive"  reductive
                                                      dechlorination.  At some sites, however,  chloroethene
                                                      plumes  are undergoing dechlorination where  significant
                                                      amounts of anthropogenic material  is  not  detected.
                                                      Physical interactions of chtoroethenes with indigenous
                                                      organic  matter in soil, sediment, and aquifer  solids are
                                                      important processes controlling the fate and transport of
                                                      contaminants in the subsurface (10-12). In  many in-
                                                      stances, organic carbon concentrations of aquifer solids
                                                      are assumed to be negligibly low, and in soils they are
                                                      assumed to decrease exponentially with surface depth.
                                                      We have tested a working hypothesis that under certain
                                                      conditions, the  release of chlorinated solvents could
                                                      mobilize soil organic material, which could then serve as
                                                      an anaerobicalry metabolizable carbon source that will
                                                      drive the dechlorination of  chloroethenes.
                                                   58

-------
                          No Added Sulfat*
^     OJ mil Addad Sulfate
                          10  20   30  40  50  60
                             Tim*  (days)
          10  20  30  40   50  60
             TJm«  (days)
                _      1.0 mil Addad Sulfate
 _     5.0 mM Addad Sulfata

 1   «;
co
                                                •fo
                          10  20   30  40   50  60
                             Tim* (days)
                                                               u
 _3
 3
          10   20  30  40  50  60
              Tim* (days)
 Figure X  Effect* erf different Initial sulfate concentrations on the onset of reductive dechtorinatlon activity. -Cl Is earbon-chlorid*
           bonds reduced snd Is equal to [TCE] + 2(DCE). Values are *n aversge of flve repllcsnts.
 TaWa 1.  Effects of Varioci Inrilbttors on Reductive DecNorinstion Activity In Norman Landfill Sedlmenta

                                      Formate                         Ethanol
 Donor
                                      Sutynte
                                ROC
                                                DC
                                                               ROC
                                                                               DC
                                                                                              ROC
                                                                                                              DC
Treatment
BESA (10 mM)
BESA (1 mM)
Mo(5mM)
Mo/SCV (5/10 mM)

SO,- (10 mM)

0
0
0

0
0

+f- 0
+A- 0
0 0

- 0 0
w- - +

0
0
-

0
0

-
-
0

0
•f
Vancomydn HydrochJoride
(100 ppm)
                                                W-
RDC » Raducttve decrilorlnatton activity relative to positive control
DC  • Electron donor catabollam relative to positive control
Mo  - Moryodats (r^MoO<.2H2O)
0    -No activity
W-  • No significant change relative to positive control
-    • Decreased activity relative to positive control
*    « Increased activity relative to positive control
n    » Five each treatment
                                                         59

-------
 Organic carbon was extracted from a spodic soil high in
 humic and tulvic acid concentrations, collected from the
 va;jose  zone of  the Sleeping  Bear site in Michigan.
 Distilled water and distilled water saturated with ICE
 were used  as extractants. The presence of TCE was
 observed to improve the extractability of organic com-
 pounds  (although the specific  identity of these com-
 pounds is unknown at this time, as is the mechanism of
 extraction). Experiments were conducted in which mi-
 crocosms were spiked with the soil carbon extracts in a
 range of concentrations. The extracted organic material
 served as the primary carbon/energy source for subsur-
 face microorganisms in  the microcosms. The micro-
 cosms were monitored over time to determine the ability
 of the extractabie organic carbon to support the dechIon-
 nation of PCE. Figure 4 shows the results of the micro-
 cosm experiments, which indicate the loss of PCE over
 time for both types of extracts when present  in sufficient
 concentrations. The dechlorination of PCE in the active
 experimental treatments  correlated with the production
 of TCE and dtehtoroethylene (DCE) daughter products
 (data not shown), indicating that the extracts provide a
 metabolizable electron donor capable of supporting mi-
 crobial consortia  responsible for reductive dechlorina-
 tton of PCE.

 Summary and Conclusions

 In situ reductive dechlorination holds significant poten-
 tial for use in natural (passive) and active in situ reme-
 diation methods. For  reductive biodehalogenation to
 gain acceptance as a viable alternative to conventional
 physical and biological treatment methods,  however, it
 must be predictable and well understood. Information
 and operational experience are needed concerning the
 environmental  parameters, microbial interactions, and
 metabolic responses that control the initiation, rate, and
 extent of these degradation processes in the subsur-
 face. An understanding of the controlling mechanisms
 and the incorporation of such mechanisms into predic-
 tive  models and operational designs  should allow more
 accurate assessment of the applicability and  implemen-
 tation of anaerobic remediation of chloroethenes at chto-
 roetnene-contaminated sites.

 References

 1. Suflita, J.M., S.A. Gibson, and R.E. Beeman. 1988
   Anaerobic biotransformation of pollutant  chemicals
   in aquifers. J. Indust Microbiol. 3:179-194.

2. Sewell, G.W., and S.A. Gibson. 1991. Stimulation of
   the reductive dechlorination of tetrachloroethane >n
   anaerobic aquifer microcosms by the addition of tolu-
   ene. Environ. Sci. Techno). 25:982-984.

3. Gibson, S.A., and G.W. Sewell. 1992. Stimulation of
   reductive dechlorinatbn of tetrachloroethane (PCE)
   in anaerobic aquifer microcosms by addition of short-
    a
    c
    o
    HI
          0      50     100    150    200    230
                        Tlrrw (days)
                 -•-  NoExtract

                 -&-  Abode

                 -•-  100 rolTCE/wBtar Extract

                 —Ik—  50 ml TCE/watar Extract

                 —a>-  10 rm TC£A»«»r Extract

                 -9-  100 Tdwttar Extract

                 —+—  SO ml watsf Extract

                 —*—  10 ml watar Extract

 Figure 4.  Effaota of watsr and watar/TCE extracts OP r»ducttv«
         oachlorinatlon of PCE In Norman Landfill micro-
         cosm*. TCE and  DCE Intsrmadlatas ir» not shown.
   chain organic acids or alcohols. Appl. Environ. Mi-
   crobiol. 58(4): 1,392-1,393.

4. Gibson, S.A., D.S. Robinson. H.H. Russell, and G.W.
   Sewell. 1994. Effects of addition of different concen-
   trations  of  mixed fatty acids on dechlorination of
   tetrachloroethane in aquifer microcosms.  Environ.
   Toxicol. Chem. 13(3):453-460.

5. Freedman,  D.L.,  and J.M. Gossett. 1989. Biological
   reductive dechlorination  of tetrachtoroethylene and
   trichloroethylene  to  ethylene  under  methancgenic
   conditions. Appl.  Environ. Microbiol. 555,144-2,151.

6. Schdz-Muramatsu, H., R. Szewzyk, U. Szewzyk,
   and S. Gaiser. 1990. Tetrachloroethylene as electron
   acceptor for the anaerobic degradation of benzoate
   FEMS Microbiol.  Lett 66:81-86.

7. DiStefano, T.D.. J.M. Gossett, and S.H. Zinder. 1991.
   Reductive dechlorination of  tetrachloroethane  to
   ethene by an anaerobic enrichment culture in the
   absence of methanogenesis. Appl. Environ Micro-
   biol. 57:2,287-2,292.
                                                  60

-------
8.  BarrioLage, GA, RZ. Parsons, R.S. Nassar, and
   P.A.  Lorenzo.  1987.  Biotransformation of trichlc-
   roethene in a variety of subsurface materials. Envi-
   ron. Toxicol. Chem. 6:571-578.
9.  Fathepure, B2., and SA Boyd. 1988. Dependence
   of tetracnloroetriylene  dechlcrination on  methane-
   genie substrate consumption by Methanosarcina sp.
   strain DCM. Appl. Environ.  Microbiol.  54:2,976-
   2,980.
10. Karickhoff, S.W. 1981. Semi-empirical estimate
    sorption of hydrophobic pollutants on naturals
    merrts and soils. Chemosphere 10:833-846,

11. Schwarzenbach, P.P., and J. Westall. l981.Tu
    port of nonpolar organic compounds from suit
    water to ground water Laboratory sorption stud
    Environ. Sci. Technol. 15:1,360-1,366.

12. Dzombach, D.A., and R.G. Luthy. 1984. Estimai
    adsorption of potycyclic aromatic hydrocarbons
    soils. Soil Sci. 137292-308.
                                                61

-------
     Application of Laser-Induced Fluorescence Implemented Through a Cone
        Penetrometer to Map the Distribution of an OH Spill In the Subsurface
                          Don H. Kampbetl, Fred M. Pfeffer, and John T. Wilson
                             U.S. Environmental Protection Agency, Ada, OK

                                          Bruce J. Nielsen
                            Armstrong Laboratory, Tyndall Air Force 3ase, FL
 Reid monitoring at spill sites usually involves collec-
 tion and analysis of ground water, soil gas, and/or
 core material. Applications for soil gas are limited to
 volatile contaminants in the vadose zone. Ground-
 water assays are useful but detect only contaminants
 associated with the aqueous phase. Total contamina-
 tion of the subsurface, especially by petroleum hydrc-
 arbons, is best  measured by  vertical profile  core
 sampling and analyses. Our field site characterization
 studies of fuel spills involve vertical profile core sam-
 pling for direct analysis of combustible gas and sol-
 vent  extractions  for total petroleum hydrocarbons
 (TPH) by infrared  spectrometry or for aromatic hydro-
 carbons by gas  chromatography and  mass spec-
 trometry.

 Objective

 The objective of the study was to demonstrate the
 usefulness of a  laser-induced fluorescence cone
 penetrometer (LIF-CPT) as an inexpensive and rapid
 alternative  to  taking core samples for  defining the
 three-dimensional boundaries of an immiscible oily
 phase.  Data are for use  in the Bioplume model  to
 determine  the amenability of the site to intrinsic
 bioremediation.

 Operative Components

 Dakota Technologies, Inc., and Applied Research As-
 sociates, Inc.,  under contract with the U.S. Air Force
 (Armstrong  Laboratory's   Environics   Directorate),
 have developed a LIF-CPT tool for mapping the dis-
 tribution of petroleum hydrocarbons as  nonaqueous
 phase liquids (NAPLs). Principal individuals from the
two organizations  involved in development and appli-
cation of the specific LIF-CPT probe used in this study
are Wesley L.  Bratton, Randy St. Germain, Martin  L.
Gildea, Greg D. Gillispie, and James O. Shmn. Basic
operating components are an optical system to deliver
tuneable laser radiation into an optical fiber for transfer
downward through a cone penetrometer to a sensor tip
equipped with a sapphire window. The subsurface ma-
terial next to the window fluoresces upon exposure to
laser radiation. This fluorescence radiation is transmit-
ted back to the surface, where intensity, fluorescent
lifetime, and wavelength are measured.

The LIF-CPT was calibrated for condensed ring aro-
matic hydrocarbons  (specifically,  the naphthalene
class), which are common constituents of petroleum
products. Acquired data were stored on a floppy disk
for later processing. Data plots also were displayed
on  a monitor  screen  for direct interpretation as  the
probe moved downward. The  LIF-CPT  also was used
for continuous profiling of soil  stratigraphy and collec-
tion of soil gas, ground-water, and core samples.

Field Site

The field study site was used extensively as a fire-
fighting training area from 1950 to the mid-1980s. Fire
training  pits were flooded with  water,  and waste jet
fuel mixed with oil and solvents was  floated on  the
water and ignited. The burning oil was extinguished.
Any unbumed oil infiltrated after these exercises. Pits
were constructed in about 70  ft of sand above a con-
fining layer of clay.  The lithology  is unconsolidated
and unifor - glacial outwash sand. The water table is
about 30 ft below the ground surface. The  ground-
water seepage velocity is about 0.4 ft/day.

Less than  3 hours were required to acquire LIF data,
recover the tools, decontaminate, and move to  the next
site. Using the LIF-CPT to collect cores for analyses
took 12  hours Samples could not be collected  more
                                                62

-------
 than 3 ft below the water table. A conventional hollow
 stem auger would have required 24 hours to acquire the
 same samples. The LIF-CPT can detect petroleum hy-
 drocarbons in  material  below the water table where
 rrv..erial cannot be recovered as cores.

 Results
 Vertical  profile  LIF-CPT probe  responses  were
 obtained at nine locations  within  the study  area.
 Figura 1 shows probe  responses  in  a longitudinal
 transect though the fire training area  parallel to the
 direction of ground-water flow. Strip chart displays for
 each location depict relative fluorescence measure-
 ments. Location 840 was within the fire pit. Beginning
 at 15 feet below the land surface, a LIF-CPT response
 positive for NAPL was obtained. This response ex-
 tended another 30 feet downward to a position 5  feet
 below the water table. A core taken at the water table
• contained 125,000 mg  TPH/kg soil.  From combined
 LIF-CPT and TPH information,  an estimated 85  per-
 cent of the oily phase is present above the water table.
 Remediation by vadose zone venting may be able to
remove a majorfraction of the subsurface contaminated
mass.
Test locations 84L and 84F were 1CO feet apart and
7CO  feet downgradient  from  the fire pit (Figure 1),
NAPL was present in the capillary fringe  at both loca-
tions. Core material collected  at the water table depth
at location 84F contained 2,050 ,ng TPH/kg soil. Lo-
cation 84K, located 100 feet downgradient from 84F,
did not have a positive response to LIF-CPT probing.
Therefore, the leading edge of the oily-phase plume
was concluded to be less than 100 feet beyond 34F.

Figure 2  is a display of the  TPH and  LIF-CPT  results
for location 84D and shows a direct relationshio witti
the two  parameters. Other information  will be  pre-
sented to show that results obtained for specific fuel
aromatic hydrocarbons also show a direct relationship
with TPH and LIF-CPT results.

Discussion
The UF-CPT probe used as an onsite rapid assay tool
successfully mapped in three dimensions the oily-phase
Figure 1. UF rMpona* v»r»u» «t«v«t1oo «t sampling location*.
                                                 63

-------
Figure 2.  LIF and TPH versus depth at location 840.
plume studied. Applications of the LIF-CPT technology
will be investigated at other field  spill sites. We  are
continuing system development to apply the LIF-CPT
method to characterization studies at sites with different
classes of hydrocarbons present.
                                                    64

-------
          Effectiveness and Safety of Strategies for Oil Spill Bioremediation:
                                    Potential and Limitations
                                            Joe Eugene Lepo
       Center for Environmental Diagnostics and Bioremediation, University of West Florida, Pensacola, FL

                                    C. Richard Cripe and P.H. Pritchard
          U.S. Environmental Protection Agency, Environmental Research Laboratory. Gulf Breeze, FL
 Background

 A variety of commercial agents are available for use in
 oil spill bioremediatjon. Selection of appropriate biore-
 mediation agents or bioremediation strategies for use in
 the field, however, has been complicated by the lack of
 standard tests for assessing agent effectiveness and
 environmental safety. Acknowledging this problem, EPA
 began an effort to develop protocols for assessing effec-
 tiveness and safety of putative commercial bioremedia-
 tion agents (CBAs) based on a tiered approach (1,2).

 Protocol validation for open-water and beach spill sce-
 narios has progressed using selected CBAs and posi-
 tive control  regimes.  CBAs were characterized  by
 vendors as microtiial, nutrient, enzyme, dispersant, and
 other. Tier I involves the gathering of pertinent informa-
 tion from vendors on potentially hazardous components
 in the agents, putative mechanism(s) of action, and
 methods and rates of application. Tier II monitors oil
 bkxlegradation in a closed, shake-flask test system in
 which the oil  is physically agitated. Tier III  oil spill simu-
 lation  tests  are designed  to model  field  conditions
 thought to significantly affect CBA effectiveness in open
 water or on sandy beaches; effluents can be monitored
 for washed out petroleum hydrocarbons or monitored for
 toxicity. Tier IV testing will be an  actual field evaluation
 of the protocol test systems, conducted on a controlled
 release of oil  or a "spill of opportunity.*

 Because of the nature of bioremediation,  nutrients are
 common components of CBAs; however, most forms of
Inorganic nitrogen exhibit some toxicity to aquatic organ-
isms. The concern for product toxicity is addressed at
the Tier III level with two 7-day chronic estimator tests
associated with effluent toxicity evaluations that use a
crustacean (Mysidopsis bahia, mysids) and a fish (Meni-
 dia beryllina,  inland sih/ersides) (3). The mysid test has
three endpoints—survival, growth, and fecundity—while
the fish test focuses on survival, growth, and develop-
merit In addition to evaluation of toxicity of CBA alone,
CBA toxicity also  is assessed in  the  presence of a
sublethal water soluble fraction of oil to examine poten-
tially detrimental  interactions.

This report focuses on  results of protocol development
for CBA effectiveness and environmental safety using
the Tier III open-water and sandy beach test systems,

Tier III Teat Systems

The Tier III open-water test system provides an intact,
undisturbed oil-on-water slick in a  flow-through design
(Figure 1). A constant influx of seawater below the oil
slick removes CBA microbes and nutrients that do noi
remain  associated with the oil slick, as would be ex-
pected at a field site. Test duration is 7 days.  Effluent is
split  one stream for oil residue analysis and the other
for toxicity testing. The slick is analyzed at the end of the
test If a significant amount of oil is mobilized from the
slick surface to  the water column below (e.g., from
biosurfactant production), a subsequent test assesses
the bkxjegradability of the transported oil.

The Tier III oiled beach test system provides a sandy
beach substratum, colonized for 1  week by  microflora
indigenous to seawater. The system models tidal influx
and egress. (See Figure 2.) The surface is oiled and 2
days later, a CBA or other bioremediation strategy is
applied. Beach test systems run for  28 days, after which
the oil residues can be extracted for analysis. Effluents
are collected for analytical or lexicological examination.

For the purpose of the Tier III  protocol, generic environ-
mental  parameters were selected  for both  the open-
water and the beach test systems.  The oil was applied
to a 0.5-mm thickness, turbulence was standardized,
                                                   65

-------
             H,0
                                     Mcroco*rn A
                                   Synchroooug Motor


 F%gur« 1.  Tl«r H »4mul«t»d op«n »•!•» oil spills t«rt syvtwn.
SttPlat*
                                                                     Vl««)a( Bonom cK
 Flgur* 2. Tl«f tn *Jmul«t»d o
-------
 Two treatments,  in three replicates each, are used:
 1) a control with oil alone; and 2) a treatment with both
 oil and CBA. Criteria for evaluating the effectiveness of
 bioremediafion  in the Tier III  open-water test systems
 are based on statistically  significant (p Z 0.05)  reduc-
 tions in the weight of oil and in tha amount of selected
 gas chromatography/mass spectrometry (GC/MS) ana-
 lytos remaining  in the test vessels and test-system efflu-
 ent relative to the control vessels and effluents.

 Supplemental research  (in progress) will examine  the
 effects of environmental parameters (e.g., salinity, tem-
 perature, water  turbulence, increased treatment time or
 increased CBA application rates) on the effectiveness of
 the CBAs to provide more site-specific information.


 Results
 Validation of Open-Water Test System Using
 Positive Controls and CBAs

 To establish baseline performance for the Tier II! open-
 water test syste:ns, we used positive-control treatments
 that  were surrogates for either  nutrient  CBAs  or
 microb'ai CBAs. Three conditions were tested: 1) Gulf
 of Mexico seawater control, 2) seawater amended with
 nutrients (to test for the ability of nutrients to enhance
 the degradation capability of the natural degraders); and
 3) nutrient-amended  seawater  supplemented with a
 dairy inoculation of hydrocarbon-degrading bacteria as
 a test of competent, high levels of microbial biomass.
 The erfectr/eness of the positive control in the open-
 water test system is presented in Table 1 as a percent
 of the oil remaining relative to controls to which neither
 nutrients nor microbes were supplied. Values represent
 an average of three replicate test chambers. The num-
 ber of the GC/MS endpoints out of a total of 70 analytes
 that were significantly reduced relative to the control for
 each agent also is tabulated. Nutrients alone failed to
 stimulate biodegradation by the microbial population in-
 digenous to Gulf of Mexico saawater. Several analyte
 endpoints, however, were significantly different as trie
 result of action by the hydrocarbon-degrading bacteria
 in the presence of nutrients.


 Table 1  also reports the results of six CBAs selected as
 representativas of each CBA type. Each was applied to
 the oil slick  in the test systems according to the instruc-
 tions supplied by the vendor. Of the six, only the nutrient
 CBA gave a promising  result, effecting a change in 18
 of the GC/MS analytes and  a statistically significant
 reduction (although only 1 percent) in total oil  residue
 weight  In  contrast  the nutrient-amended seawater
 treatment of the positive control experiment effected a
 statistically significant change in only one of the GC/MS
 analytes.
 Only in the positive control experiment in which nutrients
 were  supplied continuously and oil degrading bacteria
 were  applied daily did  we find effects on a relatively
 large  number of endpoints as well as substantial reduc-
 tion in the total weight of the oil recovered.
 Table 1.  Percentage of anaryte remaining relative to control* after 7 days of UeaUiieiit with btoremedlation agents or positive
         control regimes.
 Anaryt*
                  N/M
                                               "CBA or Positive Control Treatment
HA*
                                                                  M/D
Treatment typa:  E » enzyme, N « nutrient, 0 • dlspersant, M » mcrobtaJ, *N » nutrient positive control
 +N/M * nutrient positive control + microbes,
"Number of endpoints snowing a statistically significant change at 0.05 or less
•p S 0.05; "p S, 0.01
                                                                                                     +N/M
Cn
c*
Pnytmne
Pristine
Ruorene
Chrysene
Pnenantrene
N-Alkanes
Aromatics
Total Oil
"Endpoints
97
101
103
103
102
103
102
98
102
99
5
102
100
103
99
106
117
102
105
105
101
1
•92
99
99
103
96
95
99
**92
98
•99
18
92
96
101
104
105
107
102
96
102
103
6
103
102
102
101
102
-90
99
102
102
99
1
105
100
101
99
107
114
103
106
103
102
0
94
99
102
104
99
96
102
96
103
102
1
"34
"57
95
98
95
95
•97
"40
97
•93
30
                                                    67

-------
 Validation of Oiled Beach Test System Using
 CBAs

Table 2 shows ttie percent of oil and oil components
 remaining in the test systems after 28 days of exposure
 to four CBAs in Tier III beach test systems. The control
 treatments, in which seawater flushed the systems in
 the same tidal regimes as in the CBA chambers, lost
 substantial amounts of the lower molecular weight poly-
 cydic aromatic hydrocarbons. Positive  control experi-
 ments and experiments in which we attempted to run
 sterile control treatments have suggested that the dis-
 appearance is biologically mediated,  although whether
 the compounds have been washed intact from the tast
 systems or catabolized is still  being investigated.

 Environmental Safety of CBAs

 An important ecotoxicologicaJ consideration for CBAs is
 the possible production of toxic metabolites.  This con-
 sideration is addressed at the Tier III level with a mysid
 7-day chronic estimator test  on the  effluent from the
 open-water and beach test systems. A key assumption
 is that the test system designs are conservative with
 respect to dilution; thus, if toxicity is not observed under
 these mixing scenarios, it is unlikely to occur in a field
 application. Increased toxicity (compared with the toxic-
 ity of effluent from control systems containing onry oil)
 exceeding that of the product alone (from Tier II testing)
 would suggest the need for further studies that focus on
 potentially toxic metabolites. Table 3  indicates that the
 open-water effluent from most CBAs demonstrated low
 or no toxicity. Safety has not yet been evaluated using
 the beach test system.

 One application of toxicology came as a result of adapt-
 ing a 10-day amphipod (Laptocheirus plumulosus) (5)
 sediment toxicity test to evaluate potential toxic metabo-
 lites associated with the sand of the beach test system
after the 28-day CBA efficacy  test. We observed trial
oiled sediment, whether subjected to biorem jdiation or
not, was toxic. Although  this phenomenon prevented
accurate assessment of potential toxic metabolites in
the sediment,  it led to research to determ-ne  //hether
toxicity testing could be used as an efficacy endpoint,
focusing on  the potential  of a CBA to render  an oiled
sediment suitable for amphipod recolonization. The re-
sults of preliminary studies will be discussed.

Conclusions

We have completed validation of the open-water and
sandy-beach testing systems.  Thus far, the CBAs ex-
amined  during our protocol development work have
shown little toxicity and should pose little environmental
threat to the organisms tested when applied according
to the vendor's suggested regime. Some CBAs effected
significant changes in one or more targeted hydrocar-
bons relative" to the control; however, it should be em-
phasized that the sum of all GC/MS analytes is  less than
6 percent of the total oil.  Moreover, no substantial de-
creases in  oil residue weights were associated with
treatment by CBAs.

By daily addition of microbial biomass and nutrients to
the open-water system, however, we were  able to dem-
onstrate the  greatest biodegradation of oil  components
within the 7-day period, including a significant weight
toss, i.e., significant decreases occurred in 30  of the 70
GC/MS  analytes. Thus, we conclude that  the  test sys-
tem itself was capable of giving a measurable response,
although its  accurate modeling of actual  site-specific
field conditions remains to be evaluated. These resuJts
also may indicate that the rerammended application
rates  of CBAs are insufficient to produce substantial
changes in oil biodegradation.  Daily or more frequent
additions may  be untenable in some open-water field
situations (e.g., large-area spills)' however, spills of a
Tab* 2.  Tlar • EftocttvwwM toauKa of 8«ach Syatam Teeta with C8A»

                                               Percent Remaining Hydrocarbon Analyta
CBATyp*
C18
Phytan*
018/Phyt
FKiorana
Dibenzothioph
Phenamfirene
Chrysene
Gravimetric
Nutrient
-33
-96
"39
32
52
53
106
•*92
Control
90
93
97
23
51
48
106
94
Nutrtontf
Mlcroblal
"20
-53
-37
29
52
51
104
-89
Nutrlentf
Microbial
-25
35
29
43
68
67
100
•91
Control
39
39
100
39
68
68
99
96
DUpersant
39
36
100
50
81
34
100
94
•Mean of 2 replicates; all otters were means of 3 replicates.
'PSO.05; "psO.01.
                                                  68

-------
Table 3.  Tier IR Be«urt» o* 7-O«y Chronic Estimator T»rt» with Myaidopt^ t*hl*
 CBAB      Max Effluent Cone. (%)     7-D«y LCjo
                                                                                 Comparison to Oil Control'
                                                                                   NOEC
                                                                                                  LOEC
£ 63 >«3


N 55 >55


N/M 66 >66


D 10 3.7(3- 4.6)


survival
growth
fecundity
survival
growth
fecundity
survival
jrOWtt)
fecundity
survival
growth
fecundity
63
63
63
55
55
55
66
66
66
3
NE
3
NE
NE
NE
NE
NE
NE
NE
NE
NE
10
NE
C
 •Comparisons were made between the effluent from control systems thai conta ned oil atone and those from systems containing oil and the
  CBA.
 °C8A types as defined In the note to Table 1.
 fecundity data at these effluent concentrations greater than 3% are disregarded because no females were found alive.
 NOEC • no ooserved effect concentration: LOEC * lowest observed effect concentration; NE » no effect
 more confined nature may be reasonably treated with
 higher or more frequent applications.

 There are substantial barriers to effective performance
 of oil-spill CBAs, among them dilution rates, nutrient and
 biomass limitations, and a limited time in which a CBA
 can remain  in contact with the oil spill. Efficacy indices
 from analytical chemistry, coupled with assessments of
 toxicity  for CBAs, should provide useful information to
 an on-scene coordinator. These limitations will be dis-
 cussed in the light of our experience with the Tier III
 effectiveness protocol.
 Acknowledgments

 Validation of the effectiveness protocol for Tier III open-
 water and beach test systems as well as the ecotoxicol-
 ogy for Tier II and Tier III was performed through a
 cooperative agreement (CR-818991-01)  between  the
 University of West Florida Center for Environmental Di-
 agnostics and the EPA Environmental Research Labo-
 ratory at Gulf Breeze. The following people contributed
 ideas and technical assistance during the development
of this project;   Wanda Boyd,  Mike Bundrick, Pater
Chapman. Jim Clark, Carol Daniels, Barbara Frederick,
Tim Gibson, Wallace Gilliam, Jeff Kavanaugh, Joanne
Konstantopolis, Tony Mellone,  Len Mueller, Neve Nor-
ton, Jim Patrick,  Bob Quarles, Mike Shelton, Scott
Spear, Phil Turner, Ling Wan, George Ryan, VicW Whit-
ing, Diane Yates, and Shiying Zhang.
References

1. Lepo, J.E. 1993. Evaluation of Tier III bio'emediation
   agent screening protocol for open water using com-
   mercial  agents:   Preliminary  report  EPA/600/X-
   93/001.    University    of   West    Florida/US.
   Environmental Protection Agency, Gulf Breeze Envi-
   ronmental Research Laboratory, Gulf Breeze, FL.

2. National Environmental Technology Application Cor-
   poration  (NETAC). 1993.  Oil spills bioremediation
   products testing  protocol  methods manual. Pitts-
   burgh,  PA:   University  of Pittsburgh Applied Re-
   search Center (August).


3. U.S. EPA. 1988. Short-term methods for estimating
   the chronic toxicity of effluents and receiving waters
   to  marine  and  estuarine  organisms.  EPA/600/4-
   87/028. Washington, DC.


4. International Organization for Standardization. 1989.
   Crude petroleum oil:   Determination  of distillation
   characteristics using  15 theoretical plates columns.
   Draft international standard ISO/DIS 8708.


5. Schlekat, C.E., B.L. McGee, and E. Reinharz. 1992.
   Testing sediment toxicity in Chesapeake Bay with the
   amphipod Leptochairvs plumulosus: An evaluation.
   Environ. Toxicol. Chem.  11:225-236.
                                                   69

-------
   The Use of In Situ Carbon Dioxide Measurement To Determine Bloremediation
                                              Success
                                         Richard P.J. Swannell
             Biotechnology Services, National Environmental Technology Centre, AEA Technology
                                         Oxon, United Kingdom

                                           Francois X. Merlin
                                    CEDRE, Plouzane, Brest, France
 Monitoring bic.-emediation success involves complex
 analytical chemistry and time-consuming microbiology.
 Potentially, a more valuable tool for the oil spill treatment
 specialist would be one that enabled the efficacy of a
 btoremediation strategy to be determined in real time in
 situ. This poster describes  preliminary researc'. on a
 method for making  in situ measurements of bioremedia-
 tion efficacy based on the estimation of CO2 evolution.
 These studies were conducted in the field near Lande-
 vennec, France. The trial involved the oiling of six plots
 on  a beach consisting largely of shale on a day base.
 Three plots were amended with a slow-release inor-
 ganic nutrient and three plots remained untreated as
 controls. Three  plots also were delimited on the same
 beach to act as unoiled controls.

 Methods

 Two sampling devices were made from stainless steel,
 consisting of a shallow cylinder (0.2 m high and 1.1 m
 in diameter) sealed at one end with a base plate. The
 base plate was  pierced with two steel tubes connected
 to valves  on the outside of the device.  The samplers
 were  pushed gently into the  beach surface, with the
 base plate facing upward and the valves open to the air.
 The CO2 analyzer  then was connected  to the valves,
 and air from the sampler was circulated through it giving
 an  initial C02 reading. The C02 level was then moni-
 tored  periodically over the next 5 to 20 min.  Measure-
 ments were taken at the same coordinates on the oiled
 controls, the unoiled controls, and the plots treated with
oil and fertilizer.  Readings were made 26,116, and 144
days after oiling. Nutrients were applied 11 days after
oiling and monthly thereafter.
Results and Discussion
On each sampling day, the rate of CO2 evolution was
enhanced on oiled plots treated with fertilizer in compari-
son to oiled controls and unoiled controls. The largest
difference was noted  15  days after nutrient  addition
when the rates increased from 3.1  ppm  to 4.0 ppm
COj.min"1 on the oiled controls to 12.6 Dpm to 22.3 ppm
COj.min'1 on the'  fertilized plots. The ijnoiled  controls
gave values between 2.8 ppm and  4.2 ppm CO2.mirv'
These data suggest that nutrient addition stimulated the
CO2 evolution rates when compared with untreated con-
trols. The rates were found to decrease in subsequent
measurements of the fertilized plots but were still 1.5 to
2.0 times greater than the  controls,  suggesting the
stimulation in CO2 production was sustained.

Conclusion

These preliminary  data suggest  that addition of fertilizer
to oiled plots stimulates CO2 evolution. Whether  this
stimulation reflects enhanced oil biodegradation, as we
suspect  remains to  be proven  absolutely using gath-
ered chemical samples. Further, although the measured
values are, by  their nature, relative  rates and not abso-
lute indicators  of C02  production,  the results  suggest
that this technique may provide useful data when exam-
ining the efficacy of bioremediation strategies and prod-
ucts on contaminated  shorelines.  A  second field trial
conducted in  the  United  Kingdom in the summer of
1994, funded by EPA, will  allow a more detailed evalu-
ation of this promising technique.
                                                 70

-------
Intrinsic Bioremediation of JP-4 Jet Fuel Contamination at George AFB, California
                          John T. Wilson, Michael L Cook, and Don H. Kampbell
      U.S. Environmental Protection Agency, Robert S. Kerr Environmental Research Laboratory, Ada, OK
Intrinsic bioremediation is difficult to evaluate from moni-
toring well data. Depending on ttie screened interval and
the pumping rate, a well may produce water from an
uncontaminated part of the aquifer, resulting in a sanple
that is greatly diluted by clean water. In addition, a well
may miss the plume entirely. Both  effects give the false
impression that in situ biological processes are attenu-
ating the contaminants. A rigorous demonstration of in-
trinsic bioremediation should include 1) information on
the use of available electron acceptors; and 2) informa-
tion on the concentration of  a tracer associated with the
plume that can be used to correct for dilution.

Ground water at George Air Force Base (AFB)  was
contaminated by a release of JP-4 jet fuel. Well MW 24
is near the center of the spill. Well MW 25 is 500 feet
from well MW 24 in a direction that is perpendicular to
ground-water flow. Wells MW 27, 29, and 31 are along
a flow path down-gradient of well MW 24. The plume
velocity is near 100 ftryr.
Oxygen and nitrate were depleted downgradient of the
spill. The concentration of benzene was reduced  more
than 300-fold, while the concentration of a more recal-
citrant compound, 1,2,3-trimethylbenzene, was only re-
duced three-fold. After correcting for dilution, benzene
concentrations were reduced at least 100-fold due to
intrinsic bioremediation.
TaM* 1. Intrinsic Blortfiwdlatton of B*nz*fw

IvOcnfon

Oxyg«n
Nitrate

BOTXMM
Toliwn*
1,2^-THnwthylbwwn*
MW24
C«ntw olo» ton*

<0.5
0.8

1,620
1.500
73
and ToliMn*
MW2S
Eclg« of oil lens

8.0
3.7

194
604
30

MW27
700 tt away
(mg/lltaf)
0.6
0.4
frig/liter)
90
<0.5
56

MW29
1,200 ft away

<0.5
0.3

4.8
<0.5
20

MW31
1,800 ft away

1.1
3.1

<0.5
<0.5
<0.5
                                                 71

-------
      Factors Affecting Delivery of Nutrients and Moisture for Enhanced In Situ
                           Bloremediation in the Unsaturated Zone
                                    James G. Uber and Ronghui Liang
          Department of Civil and Environmental Engineering, University of Cincinnati, Cincinnati, OH

                                            Paul V. McCauley
   U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Water and Hazardous Waste
                               Treatment Research Division, Cincinnati, OH
Successful in situ btoremediation in the unsaturated
zone requires that water, oxygen, and trace nutrients be
available in appropriate amounts and correct locations.
To enhance degradation rates, some applications may
require delivery of moisture, oxygen, or trace nutrients
via subsurface or surface application of fluids. Since the
exact locations and geometry of contaminated regions
are unknown, a practical  engineering approach  is to
design fluid delivery systems to uniformly distribute the
fluids to a subsurface region.
This project Investigates limitations of engineered sys-
tems  for delivery of nutrients, either liquid or gas, to
contaminated soils in the unsaturated zone. These limi-
tations are derived from two sources: 1) the basic design
of fluid  delivery  systems (e.g.,  inherent limitations in
using vertical wells or surface irrigation systems to uni-
formly distribute and collect  a fluid in an unsaturated
subsurface region); and 2)  heterogeneity in porous me-
dia properties that affect fluid flow in the unsaturated
zone  (e.g., spatial  variability of saturated hydraulic
conductivity).
Unfortunatery, the design of common fluid delivery sys-
tems and the heterogeneity of hydraulic soil properties
work against achieving the goal of uniform fluid distribu-
tion. Vertical wells and soaker hoses are two means of
fluid delivery, but these are essentially  point or line
sources. Thus, important unanswered  questions exist
about the proper spacing of these devices to achieve a
uniform application rate. A potentially more difficult issue
is the signiflcarrt spatial heterogeneity in  the hydraulic
properties of natural soils. This  heterogeneity  creates
paths of preferred flow on a variety of spatial scales; only
a fraction of the porous media may contribute to fluid
flow, and thus, an engineered system designed to de-
liver moisture, oxygen, or nutrients could fail to achieve
a uniform distribution. Thus, the conventional notion of,
for example, a well's "region of influence" is less clear
and will be critically reexamined through  experimental
and theoretical approaches.
Work in Progress

This poster presents findings from a review of soil sci-
ence and in situ bJoremediation literature, focusing on
the potential effects of preferential flow on in situ biore-
mediation effectiveness. This review was initiated at the
start of the project in January 1994 and is being used to
guide the design of experiments scheduled to begin later
this year. Future plans regarding the experimental
investigations also will be presented.
                                                  72

-------
                          Section Four
                     Pilot-Scale  Research
By studying bioremediation processes under actual srte conditions on a small scale,
researchers can gather critical information on issues such as operation, control,
and management of residuals and emissions before moving to full-scale research.
Thus, pilot-scale research is a critical intermediate step in which the success of
laboratory experiments are further tested in an expanded but controlled setting.

Pilot-scale evaluations were performed on three alternative biofilter attachment
media as part of continuing  research on the development of  biofiltration for treat-
ment of volatile organic compounds (VOCs). A pelletized medium exhibited trie best
and  most consistent performance of the three media tested.  Future work will
concentrate on further optimizing the use of the pelletized medium.
Research  continued on developing methods for operating rand treatment  reactors
using rcdox control. Methods will be tested using  pentachtorophenol (PCP)-con-
taminated soil from the American Wood Products site in Lake City, Florida, in
pilot-scale soil pan reactors. In a related project studies continue on the use of
combined  treatment technologies for remediating  contaminated soils from PCP
manufacturing facilities and wood preserving sites.

A small-scale field study along the Delaware Bay shoreline is planned to evaluate
bioremediation of oil-contaminated beaches. Laboratory and Meld experiments will
be u-»d to test application strategies.

EPA's Testing and Evaluation (T&E) Facility will evaluate the performance of bench-
and pilot-scale slurry bioreactors in treating hazardous waste, as part of a general
research program on engineering assessment and optimal design. Soil contami-
nated with creosote constituents from a site in Si  Louis Park, Minnesota, will be
used to test the reactors. In addition, researchers at EPA's T&E Facility are studying
the ability of compost microorganisms to biodegrade porycyclic aromatic hydrocar-
bons (PAHs) in in-vessel reactors. Soil contaminated with PAHs from  the Reilty Tar
site in St.  Louis Park, Minnesota, will be used to evaluate the performance of this
technology.

The symposium's poster session included presentations on a pilot-scale evaluation
of nutrient delivery for oil-contaminated beaches; field treatment of benzene, tolu-
ene, ethylbenzene,  and xylene  (BTEX)  in vadose soils using extraction or air
stripping and  biofilters; dechtorinatlon with a  biofilm-electrode reactor the use of
sulfur-oxidizing bacteria to remove nitrate from ground water; and an engineering
evaluation and optimization of biopiles for treatment of soils contaminated with
hazardous waste.
                                   73

-------
          Pilot-Scale Evaluation of Alternative Biofliter Attachment Media for
                                       Treatment of VOCs
                  Francis L Smith, George A. Serial, Makram T. Suidan, and Pratim Biswas
          Department of Civil and Environmental Engineering, University of Cincinnati, Cincinnati, OH

                                          Richard C. Brenner
         U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH
Siiice enactment of the 1990 amendments to the Clean
Air Act the control and removal of volatile organic com-
pounds (VOCs) from contaminated air streams has be-
come  a  major public  concern  (1). Consequently,
considerable interest has evolved in developing more
economical technologies for cleaning contaminated air
streams, especially dilute air streams. Biofiltration  has
emerged as a practical air pollution control (APC) tech-
nology for VOC removal In fact, biofiltration can be a
cost-effective alternative to the more traditional tech-
nologies, such as carbon adsorption and incineration,
for removal of low levels of VOCs in large air streams
(2). Such cost effectiveness is the consequence of a
combination of low energy requirements and microbial
oxidation of the VOCs at ambient cjnditions.

Preliminary investigations (3) were performed on three
modia: 1) a proprietary compost mixture; 2) a synthetic,
monolithic, straight-channeled  (channelized) medium;
and 3) a synthetic, randomly packed, pelletized medium.
These media were selected to offer a wide range of
microbial environments and attachment surfaces  and
different air/water contacting geometries. The results of
this preliminary work demonstrated that 95+ percent
VOC removal efficiency could be sustained by all three
media at a toluene loading of 0.725 kg C00/m3-d, but
at different empty bed residence times (EBRTs). For the
pelletized medium, this performance could be achieved
at an  EBRT of 1  min, for the channelized medium at 4
min, and for the compost medium at 8 min. Both syn-
thetic  media  developed  headless  over time, with  the
pelletized medium showing a pressure drop in excess
of several feet of water after sustained, continuous op-
eration. These results left open the question of  which
medium could provide the optimum combination of high
VOC elimination efficiency at high loading with minimum
pressure  drop.
This paper discusses the continuing research being per-
formed for development of biofiltration as an efficient.
reliable, and cost-effective VOC APC technology. The
objectives of the recent research were to conclude the
evaluation of the three media and to develop workable
strategies for the rer.oval and control of excess biomass
from the  (ultimately)  selected pelletized  attachment
medium.


Experimental Apparatus

The biofilter apparatus used in this study consists of
three independent, parallel biofilter trains, each contain-
ing 4 feet of attachment medium: biofilters A, B, and C.
A detailed schematic and equipment description is given
e'sewhere (4). Biofilter A was filled with a proprietary
compost mixture, B with a Coming Celcor channelized
medium, and C with  a Manville Celite pelletized me-
dium.  Biofilters A and B are square and  have an inner
side length of 5.75 in.; biofilter C is round, with an inside
diameter of 5.75 in. The air supplied to each biofilter is
highly purified for conmlete removal of oil, water, CO2,
VOCs, and particulates. After purification, the air flow for
each biofilter is split off, injected with VOCs, humidified.
and fed to the biofilters.  The  air feed  is mass flow
controlled,  and the VOCs  are metered  by syringe
pumps. The flow direction of the air and nutrient inside
each biofilter is downward.  Each biofilter is insulated
and independently temperature controlled.

Buffered nutrient solutions are fed to biofilters B and C.
A detailed  description of  the nutrient composition is
given  elsewhere (4).  Each  of  these biofilters inde-
pendently receives a nutrient solution containing all the
necessary macro- and micronutrients, with a sodium
bicarbonate buffer. The nutrients required in biofilter A
were included as part of the  original compost
                                                 74

-------
 Results

 BlofiltsrA

 This biofiller run on the compost medium was made to
 evaluate the effects of temperature and then loading on
 toluene removal efficiency. Figures 1a and 1b summa-
 rize the biofilter performance. The biofilter was started
 up and, after some operational difficulties, stabilized by
 Day 10 at 52°F, 50 ppmv toluene, 2 min EBRT, and a
 removal efficiency of about 58 percent On Day 17, the
 temperature was raised to 60°F, resulting in a  rise in
 efficiency to about 75  percent,  which decreased after
 Day 24 into the 60s, and after Day 32 into the 50s. On
 Day 41, the temperature was increased to 70°F, result-
 ing in  a gradual increase in efficiency to about 75 per-
 cent by Day  47. On  Day 53,  the  temperature  was
 increased to 80°F, resulting in an incraase in efficiency
 into the low 80s. On  Day 61,  the  temperature  was
 increased to 90°F, resulting in a further increase in effi-
 ciency to the mid-90s (Figure 1 a). After Day 77, the feed
 was increased slowly to about 95 ppmv toluene, result-
 ing in  a drop in efficiency to about 88 percent Further
 increases in the feed concentration to a maximum of 180
 ppmv  toluene on Day 139 resulted in a further decline
 in efficiency to about 58 percent (Rgure ib). The run
 was terminated on Day 215.

 Biofilter B
 This biofilter run was made on the synthetic channelized
 medium to evaluate the effects of temperature and then
 nutrient feed rate on removal efficiency. The biomass in
 the channels of the medium remaining from the previous
 run was removed by hydroblasting the eight 6-in. high
 medium blocks from top  and bottom. The comers  of
 these  square blocks were filled with grout to provide a
 "round" active block. This  last step was taken to match
 a round block cross section with the round pattern of the
 nutrient delivery spray nozzle. Figure 2 shows the biofil-
 ter performance as a function of time. The biofilter was
 started up at 52°F, 50 ppmv toluene, and 2 min EBRT.
 By Day 36, the removal efficiency had  drifted over a
 range  from  about 62 to  80 percent On Day 36, the
 nutrient feed rate was increased from 30 L/day to 60
 L/day,  while keeping the mass loading of the nutrients
 constant  The  increased nutrient flow rate effectively
 doubled the wetting cycle from 20 sec/min to 40 sec/min.
 An immediate increase in  efficiency to 99 percent was
 observed, which then quickly dropped and ranged by
 Day 50 between about 30 and 70 percent. On Day 50,
 the nutrient  feed rate was increased to 90 L/day (in-
creasing the wetting cycle to  60 sec/min), but the effi-
ciency  dropped from 69 percent and ranged by Day 67
from about 22 percent to  65  percent On  Day 67, the
temperature was raised  from 52°F  to 60°F,  and the
efficiency increased to 66  percent. By Day 75, the effi-
ciency  was 87 percent, and this level was maintained to
  100

 "2 90
 8
 f 30
 cc
 <0
 § 70
   50
Toluene Loading
0.45 kg COD/mJ day

EBRT » 2 minutes
             50     80     70     90
                     Temperature, f
                                        90
Figure 1*.  Effect of temperature  on the performance
          compost Moflltar.
  100

   90

 |  80


 I  ?°

 i  60

 t
   50
Temperature « 90°F

EBRT m 2 minutes
     0    0.4  0.6   0.8   1.0   1.2   1.4  1.6  1.8 I
                Toluene Loading, kg COD/day nv>

Figure 16.  Effect of toluene loading on the performance
          eompoat biofllter.
Day 83. After Day 83, the temperature was raised
10°F steps to 90°F, but the efficiency did not improvf
fact, for the rest of the run, at 90°F and 60 L/day
efficiency  ranged between about  58 percent and
percent The run was terminated on Day 152.

Biofilter C

The first biofilter run on the synthetic pelletized medi
was made to evaluate the effects of pressure drop
then temperature on toluene removal efficiency.
biofilter was charged with pellets used in trie previi
run. These pellets were washed by hand in hot w
(150°F) until the accumulated surface biomass
been removed and the pellets were free flowing. Fig
3 presents the biofilter performance as  a function
time. The biofilter was started up at 52°F, 50
toluene, and 2 min EBRT.  By  Day 21, the re*
efficiency was 99 percent, and by Day 27, it had react
100 percent  and  remained at this level  until Day
From Day 51  to Day 57,  the  EBRT was  gra*
reduced to  1 min, causing the efficiency to drop
                                                   75

-------
  100
                          T<*jan« Loading
                          144 kg COO/m» day
             60     80    100    120   140   160
 Flgur* 2.  P«rform«nc« of chann«ilr»d bioflltar with rMp«ct to
         toiuan* removal of an EBRT of 2 min.
            Efftaency
        1
>  80*F 70*F  ,'80"F
      2 _ 2 mm. "gT  .,
                                       '/
                      _1 mm. EBHT
               Toluana Loading kg COD/mJ day! ,'
          20   40    SO    80    100    120    140
                                             30
Figure 1  Performance of pallatized btonitar with reepect to
         toluene removal at 1  min and 2 min EBRT without
84  percent. Subsequently, the  toluene  removal effi-
ciency rapidly increased to the low 90s and remained in
that range until Day 81. On Day 82, the temperature was
raised to  60°F, and the efficiency steadily  rose until
complete biodegradation of the toluene was reached on
Day 89. This essentially 100-percent efficiency in tolu-
ene removal was maintained through Day 97. During the
period between  Day 54 and  Day 97, pressure drop
across the system increased from 0.2 to 5.5 in. water.
From Day 97 to Day 111, the efficiency dropped steadily
from  100  to 86  percent  while  the pressure  drop
increased from 5.5 to  6.0 in. water. On Day 112, fhe
temperature was increased to 70°F, and the efficiency
rebounded by Day 113 to a peak value of 97 percent.
after which it dropped to 85 percent by Day 188. On Day
119, the temperature was raised to 80°F, and  the effi-
ciency rose to about 89 percent by  Day 120. During the
period from Day 112 to  Day  120, the pressure drop
increased from 6 in. water to 18 in. By Day 128, the
efficiency had steadily dropped frorr 89 to 77 percent as
the pressure  drop increased from 18 in. water to 27 in.
This  pattern  of  a steady loss  of efficiency with a
coincident increase in  pressure drop suggests the de-
velopment of short circuiting within  the biofilter medium
because of biomass accumulation, which  results  m a
significant reduction in actual contact time. The run was
terminated on Day 128.

The second biofilter run on this medium was conducted
to evaluate routine biomass control by backwashmg.
The biofilter was charged with a 50:50 mixture  of fresh
pellets and pellets from  the previous run. The used
pellets were thoroughly washed by hand in tepid water
(90°F) until the accumulated surface biomass had been
removed and the pellets were free flowing.  Figure 4
shows the biofilter performance as a  function  of time.
The filter was started up at 90°F, 50 ppmv toluene, and
2 min EBRT.  By Day 4, the removal efficiency was 100
percent (Note: This second run, started up with pellets
washed in tepid water, contrasts with the slower startup
in the first run, where the pellets were washed with hot
water.) On Day 8, the feed was increased to 250 ppmv
toluene; the efficiency dropped to 97 percent and ranged
between 92 and 98 percent until Day 25, when it again
reached  99   percent  Subsequently,  the  efficiency
dropped as tow as 86 percent before regaining 99  De-
cent on Day  81,  after which the efficiency was near!-/
always  99-t-  percent  Initially, backwashing  was  per-
formed once a week by using 100 L of fresh water at a
rate of 6 gallons per minute  (gpm). After Day 28, the
90
x ao
}70
I9"
5 so
I40
J 30
20
10
0
[W-^y «• '-i |"
h Effldeney I ] 1 « ^
Toluene Loading leg COCVn>> ary
«4S

" Pntaure Orep
' ,\|j
-^'.^•JilAri


i,
-»"- 	 -^L_
i
£
10 9
f
3|
0
         23    50   73    100   123    150   17S
                   Sequential Dale, dayi

Figure 4.  Partormanoa of pallatized biofilter with respect to
         •-•	» removal at 2 min EBRT with backwaahlng.
                                                   76

-------
   700
        • VSS Produced (n»rog«n baJ«rv»)
        • VSS LoK (backwash + «fflu«m liquid)
        • VSS n«tan«d
 Rgur* S.  D«v»lopm«nt of p^tattMd bloflttw wttti Urn* (VSS
 frequency was increased to twice por week, and after
 Day 38, the volume was increased  to 200 L These
 changes were made because measurable pressure
 drop was observed between backwashings. On Day 73,
 the backwash rate was increased to 15 gpm to induce
 full fluidization. Although the pressure drop increase was
 minimal, the  efficiency did not improve,  suggesting
 some form of channelizing within the bed. Therefore, on
 Day 80, the length of the backwash period was  in-
 creased to 1 hour by ^circulating the backwash water.
 After this final adjustment the  toluene removal effi-
 ciency,  as mentioned above, achieved and sustained
 99+ percent During this latter period,  the total volume
 of water used per backwash was optimized to 120 L Of
 this  volume, 70 L were  used for the 1-hr backwash
 recycle, while the remaining 50 L were used.to flush the
 released solids from the reactor. Figure 5  shows the
 development of biomass with time.  After Day 38, the
 rate of biomass accumulation declined with the increase
 in the wash volume. After Day 73, the accumulation rate
 became nearly zero with the implementation of full fluidi-
 zation. Since then, no change in the  backwash proce-
 dure has been made, and the accumulation of biomass
 within the biofilter has leveled off at about 180 g with the
 pressure drop between backwashings typically under
 0.2 in. of water.
Conclusions and Future Work

A marked improvement in toluene removal efficiency
with increasing temperature was demonstrated in this
study for the compost mixture, he channelized medium,
and the pelletized medium. The direct consequence of
this finding is that much less medium would be needed
for a biofilter operating at 90°F tnan at 52°F, resulting in
a proportional  reduction in capital cost The economic
tradeoff with the cost of heating the incoming air should
usually favor operation at these warmer conditions.
The modest performance of trie compost mixture with
respect to increased loading complemented our earlier
findings with respect to decreasing EBRT (3). Unfortu-
nately, implicit limitations of the experimental apparatus
may have resulted in reouced performance. Specifically,
the manufacturers recommended using a width-to-depth
ratio of 1:1, rather than 1:8. They also stated that from
their experience the only effective means of controlling
bed moisture content was to weigh the entire biofilter.
Weighing was impossible with the heavy stainless steel
unit used here, which \vas bolted to a support frame.
Several moisture  measurement and control strategies
were attempted, but it was never possible to be certain
that tha bed moisture content was consistency at the
reported optimum range, i.e., between about 50 and 60
percent (5,6). The sometimes erratic performance may
have  been influenced by variations in bed moisture
content The best removal efficiencies achieved by the
compost  mixture,  however, were better than shown by
the channelized media but worse  than shown by  the
pelletized media.
The performance  of the channelized medium also con-
firmed our earlier  findings that this medium is distinctly
inferior to the pelletized medium (3). The best perform-
ance  was achieved during the use of new medium
blocks. After biomass accumulation within the channels
and subsequent removal by hosing, the performance
never  regained the previous, still  modast  levels.  At-
tempts to adjust nutrient flow as a means of testing the
effect of the duration of wetting in the ruthent application
cycle  did not overcome the previously demonstrated
efficiency limitations. The more erratic  performance of
this medium after  removal of the biomass suggests that
this medium may  be unsuitable for sustained efficiency
after periodic cycles of biomass removal.  This erratic
performance, due to suspected random uneven plug-
ging of channels by biomass, coupled with its relatively
tow overall  removal  efficiency,  difficulty  in biomass
removal,  and intrinsically high  medium cost, suggests
that this  medium  may not be  a viable option for this
application.

The  pelletized  medium  exhibited the  best and most
consistent performance of the three media tested. II
rapidly achieved high removal efficiencies at high tolu-
ene loadings. As the first run demonstrated, however, an
excessive accumulation of biomass, shown by a rise in
the pressure drop across the medium, results in a sub-
stantial toss in efficiency, followed by a very rapid rise in
pressure drop. This suggested that efficient sustained
performance might be achieved through early and peri-
odic control of biomass accumulation by backwashing.
In the second  run, the implementation of a suitable
backwashing strategy for biomass control was achievsd
by using full medium fluidization. This strategy permitted
sustained operation of the  biofilter at high loadings with
efficiencies consistent^ at 99+ percent. According to
                                                  77

-------
mass balance calculations, the bkwnass retained within
the biofilter stabilized at a nearly constant level.
Future work will concentrate on further optimizing the
use of the pelletized medium, with the objective of mini-
mizing the medium volume required for a selected ARC
technology application.

References

1.  Lee, B. 1991. Highlights of the Clean Air Act Amend-
   ments of 1990. J. Air Waste Mgmt Assoc. 41(1):16.
2.  Ottengraf, S.P.P. 1986. Exhaust gas purification. In:
   Rehn, H.J., and G. Reed, eds. Biotechnology, Vol. 8.
   Weinham, Germany:  VCH VeriagsgesellschafL
3.  Serial,  G.A.. F.L. Smith, P.J. Smith, M.T. Suidan,  P.
   Biswas, and R.C. Brenner. 1993. Evaluation of biofil-
   ter  media for treatment of  air streams containing
   VOCs. Paper No. AC93-070-002. Proceedings of the
   Water Environment Federation 6€th Annual Confer-
   ence and Exposition, pp. 429-439.

4.  Serial, G.A., F.L. Srrith, P.J. Smith, M.T. Suidan,  P.
   Biswas,  and R.C. Brennor.  1993.  Development  of
   aerobic  bioftter design criteria for treating VOCs.
   Paper No. 93-TP-52A.04. Presented at the 86th An-
   nual  Meeting and Exhibition of the Air and Waste
   Management Association, Denver,  CO (June).

5.  Bohn, H.L. 1993. Btofiltrabon: Design principles and
   pitfalls. Paper No. 93-TP-52A.01. Presented at the
   86th  Annual Meeting and Exhibition of the Air and
   Waste Management Association, Denver, CO (June).

6.  Van Uth, C., S.L David, and R. March. 1990. Design
   criteria   for biofilters.  Trans.  Inst Chem.  Eng.
   688:127-132.
                                                  78

-------
 Biological Treatment of Contaminated Soils and Sediments Using Redox Control:
                           Advanced Land Treatment Techniques
              Margaret J. Kupferte, In S. Kim, Guanrong You, Tiehong Huang, and Maoxiu Wang
          Department of Civil and Environmental Engineering, University of Cincinnati, Cincinnati, OH

                                          Gregory D. Sayles
         U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH

                                          Douglas S. Upton
                            Levine-Fricke Consulting Engineers, Emeryville, CA
 Soils and sediments contaminated with highly chlorin-
 ated  aromatic  compounds such  as polychlorinated
 biphenyls (PCBs), pentachJorophenol (PCP), hexachto-
 robenzene (HC8), and 1,1,1-trichloro-2,2-ois(p-chlo-
 rophenyl)ethane (DDT)  are  found  at many  of  the
 Superfund sites that have been placed on the National
 Priority List for cleanup. Btoremediation has been pro-
 posed as a means for converting these contaminants
 into less toxic or nontoxic substances.

 The biodegradation rates of  many highly chlorinated
 compounds can be accelerated by controlling the redox
 potential  [or oxidation-reduction potential (ORP)] of the
 treatment environment  In general, the  biochemical
 pathway providing the highest rate for the initial steps of
 microbial destruction of highly chlorinated organics is
 anaerobic reductive  dechlorination.  Once partially
 dechlorinated,  the resulting compounds  typically  de-
 grade faster under aerobic, oxidizing conditions. Effi-
 cient  and complete degradation of highly chlorinated
 contaminants is possible when the two redox conditions
 are sequentially applied.

 Sequential treatment techniques have been proposed
 as a means of treating aqueous wastes  and slurries
 containing soils contaminated with highly chlorinated
aromatic  compounds such  as PCBs, PCP, HCB, and
 DDT, among others (1,2). For example, the meta and
para chlorines of highly chlorinated PCBs are removed
by anaerobic reductive dechlorination; however, the or-
 tho chlorines  are  only slowly removed by the same
btoprocess. Aerobic organisms remove the ortho chlo-
rine and complete the mineralization of the compound
relatively  quickly  Thus, sequential anaerobic-aerobic
treatment should provide relatively rapid destruction of
PCBs (3,4). The process •- ;jplied to PCB-contaminated
sediments has been stuu,^ by other research groups
(1,5) and is currently being demonstrater* 'r the field.
Woods et al. (6)  suggested that an anaei  r vaerobic
sequential treatment strategy would be a   ^tractive
treatment alternative for highly chlorinated p..e
-------
land treatment reactors under anaerobic as well as aero-
bic conditions so that a sequential strategy car be read-
ily applied in the  field. Methods of applying multiple
cydes  of  alternating redox  conditions to  achieve
cleanup also are being investigated. During this project
year, these methods will be tested using PCP-contami-
nated soil in pilot-scale soil pan reactors. In subsequent
project years, we plan to investigate soils from several
types of sites, including sites contaminated with DDT.

Methodology

Reactor operating strategies that deliver adequate an-
aerobic and aerobic  microbial  environments are cur-
rently being developed using uncontaminated soil in a
pilot-scale unit with two pans (reactors). Each pan holds
approximately 30 kg of soil. Various methods of main-
taining anaerobic conditions in the soil reactor currently
are being evaluated,  including simply flooding the soil
bed, adding an easily degradable organic compound(s)
to serve as an oxygen scrubber near the surface, and
covering the soil bed with an air-impermeable cover to
inhibit the transport of oxygen. Liquid addition and per-
meate recycle techniques also are being evaluated dur-
ing the  anaerobic phase of operation. Methods for
returning the soil bed to aerobic conditions will be inves-
tigated when the anaerobic phase is complete. The soil
bed will be drained and, if necessary, a vacuum will be
applied befow the bed to assist in drainage and aeration
of the soil. Bulking agent addition  may be  required to
improve aeration of the soil. Hand mixing/tilling methods
and sample collection methods will be investigated dur-
ing both phases.

A source of contaminated soil has  been identified, and
background information about the site and the range of
contaminants and contaminant concentrations has been
obtained. Soil samples (courtesy of Wildemere  Farms,
Inc., Lake City, Florida) from various locations at the
American Wood Products site in Lake City. Florida, rep-
resenting a range of contamination levels  have been
analyzed  for chlorinated  phenolics.  A comparison of
PCP  concenf ations in these samples found  by  our
group and by an  independent  laboratory is  shown in
Table 1.

Trace amounts of less chlorinated intermediates were
noted in some of the samples analyzed in our laboratory,
out the concentrations were under the method detection
limrt (-1 mg/kg). Dioxins, low-level contaminants in tech-
nical grade PCP,  were analyzed by the independent
Iaboratoi7, the congener with the highest concentration
was octochlorinated-dioxin at 18 ppt,  and the highest
risk congener, 2,3,7,8-tetrachlorodioxin, was nondetect-
able. For the pilot-scale work at EPA's Test  and Evalu-
ation (T&E) Facility, soil will be obtained from two of the
sampling points at the site that represent high and low
levels of contamination. Approximately 600 kg of  soil
   Table 1.  Soil Analyala for PCP

              PCP In Analyzed Soil Sample*'
Sample
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Mean
Concentration
(mg PCPrtcg
dry io*l)
12.2
37.8
103
109
8.66
3.54
136
118
209
133
445
69.2
4.21
1.11
Standard
Devt-rton
(mg PCo/kg
dry soil)
0.66
i.a
2
12
4.08
0.19
9
7
15
7
38
4.2
1.00
0.22
Data from
lnd*p*nd«nt
Lab (mg
PCP/Vg
soil**)
16.3
46.4
64.5
59.7
3.29
3.08
115
93.3
178
125
N/A
N/A
N/A
N/A
    * Thro* replicates analyzed per sample
   ~ Dry weight not specifically indicated in report
   from each sampling location will be required. The soil
   will be transported to the T&E Facility, where it will be
   shredded, sieved, mixod, sampled, characterized,  and
   placed in the pilot-scale units.

   Six pilot-scale units with four pans each, a total of 24
   pans, will be employed in this study. The experimental
   design  is shown in Table  2.  Each treatment will be
   duplicated in separate reactors. A 'clean* soil spiked
   with PCP will be tested in addition to the two concentra-
   tions obtained from  the site.  The use of  recycle for
   moving the  liquid through the soil versus the mainte-
   nance ot stagnant liquid in the pan will be one of the
   variables tested. Sterile controls will be run  in parallel
   with each treatment to monitor for abiotic losses. The
   Tabte X  Experimental OeeJgn for Soil Pan Reactor*
wmiiaiimiauufi kevei
Treatment
Biologically
Active

Biologically
Inactivate J

No recycle
Recycle
No recycle
Recycle
Low
2*
2
2
2
High
2
2
2
2
spiked
Clean
Soil
2
2
2
2
   Two reactors per treatment
SO

-------
simplest approach will be tested first The soil will be
flooded wrth site  water,  if it can be obtained, or with
deionized water (dose approximation to rainwater) to
create anaerobic conditions.

Specific treatment assignments to specific pans in the
six four-pan units have been randomly assigned. Ran-
domization is  necessary because this  design  will be
statistically analyzed as  a three-factor analysis of vari-
ance (ANOVA) with replication. The three factors  are
biological activity, soil "type," and recycle. The depend-
ent variable that will be used to compare treatments and
evaluate treatment effectiveness will be  the molar sum
of the chlorinated aromatics (parent compound + meta-
bolites) removed per kilogram of dry soil at a set time
interval (e.g., after 4 months in anaerobic treatment and
after 2 months in aerobic treatment). Molar concentra-
tions will be normalized using the initial concentration in
each treatment so that the treatments can be compared
statistically using ANOVA techniques.
To supplement the statistical comparison, the pans  will
be sampled at 2-week interim time points, and the sam-
ples will be analyzed for the parent contaminant and
chlorinated aromatic metabolites to provide insight into
the pattern of removal.  Other monitoring will include
daily measurement of pH, ORP,  and temperature. Total
and volatile solids will be determined each time a soil
sample is collected so concentration can be calculated
on a dry soil basis and so soil moisture can be monitored
during the aerobic phase.

Serum bottle experiments using soil from the site will be
conducted concurrently with the pilot-scale reactors. In
these  experiments,  alternative  treatment strategies
including co-substrate and nutrient amendments and
inoculation of acclimated organisms will be explored as
means of improving treatment rate and extent. Pilot-
scale evaluation of alternatives found to be optimal is
planned for FY95.

References

1. Zitomer, D.H., and R.E. Speece.  1993. Sequential
   environments  for  enhanced biotransformation of
   aqueous  contaminants.   Environ.   Sci.  Technol.
   27(2)^27-244.

2. Armenante,  P.M.,  D.  Kafkewitz, G.  Lewandowski,
   and C.M. Kung.  1992. Integrated anaerobic-aerobic
   process for bkxJegradation of chlorinated aromatic
   compounds. Environ. Prog. 11(2):113-122.

3. Abramowicz,  D.A.  1990.  Aerobic and  anaerobic
   biodegradation of PCBs: A review. Grit Rev. Micro-
   bio). 10(3):241-251.

4. Bedard, D.L 1990. Bacterial transformation of poly-
   chlorinated biphenyls. In:  Kamely, D., et al., eds.
   Biotechnology and  biodegradation,  Vol. 4. The
   Woodlands, TX:  Portfolio Publishing Co.

5. Avid, P.J., L Nies, and T.M. Vogel. 1991. Sequential
   anaerobic-aerobic  biodegradation of  PCBs  in the
   river model. In: Hinchee, R.E., and R.F. Offenbuttel,
   eds.  Onsite  bioreclamation. Boston,  MA:  Butter-
   worth-Hememan n.

6. Woods,  S.L, J.F.  Ferguson, and M.M. Benjamin.
   1989.  Characterization of chlorophenol and  chic-
   romethoxybenzene biodegradation during anaerobic
   treatment  E.iviron. Sci. Technol. 23:62-68.
                                                 81

-------
        Research Leading to the Bioremediatlon of Oil-Contaminated Beaches
                                  Albert D. Venosa and John R. Haines
         U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH

  Makram T. Sukjan, Brian A. Wrenn, Kevin L Strohmeier, B. Love Eberhart, Edith L. Holder, and Xiaolan Wang
                                  University of Cincinnati, Cincmnar, OH
 During the summer of 1994, EPA, in cooperation with the
 Delaware Department of Natural Resources and Environ-
 mental Control (DNREC), plans to conduct a small-scale
 field study on the shoreline along Delaware Bay involving
 bioremediation of crude oil released in small quantities on
 15 identical plots. The goals of this research project are
 1) to obtain  sufficient statistical evidence to determine  if
 bioremediation with inorganic mineral nutrients and/or rri-
 crooiaJ inoculation enhances the removal of crude oil con-
 taminating mixed sand and gravel beaches; 2) to compute
 the rate at which such enhancement takes place; and 3)
 to establish engineering guidelines on how to bioremeo5ate
 an oil-contaminated shoreline. Prior to conducting such  a
 study,  two important pieces of information need to be
 defined: 1) the minimum nitrogen concentration enabling
 the degrading populations to metabolize the oil compo-
 nents at their maximum rate at all times; and 2) the fre-
 quency at which the nutrients must be added to maintain
 such a concentration. The first question is being addressed
 in the laboratory, the second in the field. This paper dis-
 cusses the design and conduct of laboratory and field
 experiments and presents some of the preliminary data
 answering the two questions posed.

 Two nutrient application strategies were tested, one in-
 volving a sprinkler system spraying water soluble nutri-
 ents on the plot the other incorporating a trench situated
 above  the high tide line but below the underlying water
 table (1). In the latter method, tracer is applied through
 a manifold at the  bottom of  the trench just before high
 tide. The underlying ground water carries  the tracer to
 the treatment zone as tides ebb and flow over time.

 Methodology

 Laboratory Experiment

To  determine the  minimum nitrogen  concentration
needed for  maximum biodegradation over time, s'x
semicontinuous flow respirometric beach reactors able
to mimic tidal flow on a beach (2) were used. A major
advantage of this microcosm is its ability to provide
continuous, real-time monitoring of oxygen uptake and
carbon dioxide evolution without the need for destructive
sampling. Each tidal flow reactor measures 75 mm in
diameter and 260 mm deep and holds approximately 2
kg beach material. The columns are fed from a 20-L
Teflon reservoir containing a flexible inner Teflon bag.
Influent seawatar contained inside  the flexible bag is
continuousty pumped by a "wave" pump into the top of
the reactor through a spray nozzle. The seawater finally
returns to the  20-L carboy outside the Teflon bag to
maintain  separation between influent and effluent The
headspace of the reservoir, the reaeration flasks, and
the reactor column are  all connected to maintain con-
stant pressure  in me system. Oxygen is supplied  auto-
matically to the microcosm system frm a respirometer
whenever a deficit is sensed. The cumulative uptake of
oxygen  is tracked continuously over time, enabling
analysis of reaction kinetics. An experiment was set up
in which six different concentrations of nitrate-N (ranging
from 0 mg/L  to 10 mg/L) were supplied to the reactors,
and biodegradation of heptadecane was followed con-
tinuously. A  mixed culture from the shoreline of Dela-
ware, previously enriched wi+h heptadecane, was used
as the inoculum.

Held Experiment

The field study is located on a sandy and slightly gravelly
beach south of Slaughter Beach,   Delaware. Surface
morphology consists of a loose upper 25-mm thick layer
of smooth gravel ranging  in size  from  4.75 mm to
19.1   mm atop coarse  sand having a  moderately
homogenous particle size distribution. Two plots meas-
uring 5 m x  10 m were set up. Two types of wells were
situated within and outside the vicinity of each plot pie-
zometers and sampling wells. The piezometers consisted
                                                  82

-------
 of black iron rods about 2.5 m long and 3.2 cm inside
 diameter (ID). The bottoms were fitted with a specially
 fritted  brass tip  that allowed  water to enter the well
 filtered of fine sand or peat particles characterizing the
 deeper zone of  the  beach.  The  piezometers  were
 equipped with pressure transducers connected to a data
 logger mounted  to a wooden post  in back of and be-
 tween the plots. The pressure transducers were used to
 measure the water head continuously to provide  accu-
 rate readings of water levels during the tidal cycles.

 The sampling wells were constructed of stainless steel
 and were also about 2.5  m long. Openings of 32 mm ID
 were drilled into the sides of the wells starting at 15 cm
 from the  bottom tip and extending  upward at intervals
 of  15 cm over a total length of 1.8 m. Stainless steel
 tubing of the same diameter was welded to these open-
 ings. The tubing extended inside the wells  from the
 openings to above the tops of the wells, where plastic
 tygon  tubing was attached for collection of water sam-
 ples via syringe. The openings in the sides of the wells
 were covered with a fine-mesh stainless steel screen to
 filter out particulate matter that might  dog the tubing.
 Thus, water samples at each depth interval were totally
 independent from other  water samples, which enabled
 measurement of tracer concentrations at one  depth
 without influence from tracer concentrations at  other
 depths.

 For the sprinkler plot, 20 kg of LJNO3 was dissolved in
 800 I of fresh water. For the trench application, 30 kg
 was dissolved in the 800 L because the trench, being
 5 m wider than the plot  width, required more  tracer for
 an  equivalent amount to reach the desired area of the
 plot Two types of samples were collected at each sam-
 pling event subsurface  sand and water from the sam-
 pling wells. The sand samples were collected with a bulb
 planter at low tide  only, water samples at both low and
 high tides. Water samples  were analyzed for lithium by
 atomic absorption spectrophotometry (3). Sediment
 samples were extracted and filtered, and the pore water
 was measured for lithium by activated alumina (AA).

 Results

 Laboratory Experiment

 Figure 1 summarizes results from two of the six reactors.
 Space limitations preclude presentation of all the data.
 Clearty, the reactor fed 10 mg/L NOr-N exhibited twice
 the O2 uptake and C02 evolution  as the reactor fed 0.5
 mg/L Also, the effluent  nitrate levels measured in the
 reactor fed 10 mg/L were  only slightty lower than the
 influent nitrate levels, whereas effluent nitrate in  the
 reactor fed 0.5 mg/L declined to  virtually undetectabla
 levels.  Thus, 0.5 mg/L  nitrogen appears to limit  the
biodegradative activity. The next higher concentration
used in the experiment was  2.5 mg/L.  which  gave
Figure 1.  Mirwrmllzatton of h«pt»d«can« In continuous flow mi-
         erocoOTM In th» pr»»*nc» of 0.5 mg/L and 10 mg/L
approximately the same results as the 10 mg/L level
Another experiment was designed (results not ready at
the time of this writing) to determine moro closely the
minimum  nitrogen level that still provides maximum
biodegradation.

Field Experiment

The plots were situated in the high intertidal zone corre-
sponding to where the spring high tide would flood the
entire plot The tide experienced, however, was a neap
tide, which means that the high tide did not cover the
plot at all during the first few days of the experiment.
Figure 2 is a three-dimensional mesh graph summariz-
ing the lithium concentrations measured in the upper
12 cm to 13 cm of sand in the sprinkler plot from time C
hr to 37 hr after application of tracer, corresponding to
six tidal cycles. Immediately after appl'  ition, the lithium
concentration in the sediment pore water ranged from
spproximately 120 mg/kg to 200 mg/kg sand. Thus, the
distribution of the tracer by the sprinkler was not as even
as originally hoped. At the next low tide (12 hours later),
the lithium had declined about 50 percent and was more
evenly distributed over the plot surface. At the next low
                                                  83

-------
                               OHoun
                                                                        12Houn
                               2SHoura
                                                                        37 Hours
                            PV*L«ngm,

 Figure 2.  Thr««-dlnwnsional plot showing behavior of IHMum tracer during the first 37 hours •ftsr •ppllcation.
tide (25 hours after application), lithium concentrations
at the bottom of the plot had declined to almost unde-
tectable levels. The previous high tide had covered this
much of the plot, which explains the low levels of tracer
there. Note that the lithium tracer in the upper two-thirds
of the intertidai zone, which had not been wetted by the
high tide,  still persisted at  slightly lower levels than the
previous low tide. At 37 hours, corresponding to the third
full tidal cyde, more of the plot had been covered by the
incoming tide as reflected by the lithium concentrations
shown in  the figure. At the 48-hr  mark, a  storm had
occurred,  causing the tidal waters  to  completely
submerge the plot Lithium levels  were undetectable
(<1 mg/kg) in the surface sediment from about 55 hours
through the  remainder of the experiment which lasted
10 days. Lithium concentrations in the surface sediment of
the trench ptot  were undetectable ur«\ after the storm
event when low levels of lithium finally appeared because
of underlying water carrying the tracer to the surface.

Tracer levels measured in  well water samples from the
ground water below the plot (data not shown) persisted
for the duration of the experiment The tracer moved up
and down with the tides, which is consistent with obser-
vations made by Wise et al. (2) in Alaska.

Conclusions
From the laboratory experiment the minimum nitrogen
concentration needed to stimulate maximum microbial
degradation of hydrocarbons is somewhere between
0.5 mg/L and 2.5 mg/L. From the  field experiment,  it
appears  that application  of fertilizer should be  con-
ducted every day when the tide covers the entire con-
taminated zone. When the tide only covers the lower
intertkJa! zone, nutrient application is not needed, since
the nutrients will likely persist for several days.  During
this period, the microorganisms will  be in constant con-
tact with nitrogen and phosphorus, which will allow time
for biostimulation to proceed. For the trench method to
work, two trenches seem to be  needed, one for the
spring tide and one for the neap tide.

References

1.  American  Public Health Association.  1989. Direct
   air-acetylene flame  method 3111B. In:  Standard
                                                  84

-------
   methods for the examination of water and wastewa-       In Situ and Onsite Bioreclamatio'i Conf&rsr.ce. .Ian
   ter, 17th ed. Washington, DC.                           Diego, CA.

2.  Strohmeier, K.L, M.T. Suidan, A.D. Venosa, and J.R.    3. Wise, W.R., 0. Guven, Fj. Molz, and S C. McC .;--n-
   Haines. 1993. A beach microcosm for the study of       eon. 1993. Nutrient retention time in a i^-r-De'T-e-
   oil biodegradation. Poster presented  at the Battelle       ability oil-fouled beach, j. Erviron. En-j   n or3ss)
                                                85

-------
   Engineering Optimization of Slurry Bioreactors for Treating Hazardous  Wastes
                                    John A. Glaser and Paul T. McCauley
          U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH

                          Majkj A. Dosani, Jennifer S. Platt, and E. Radha Krishnan
                              IT. Environmental Programs, Inc., Cincinnati, OH
 Introduction

 Biological treatment of contaminated soil slurries may
 offer the optimal treatment conditions for soil bioreme-
 diation at  an economically viable cost Despite  this
 promise, slurry  bioreactor treatment of soils has not
 achieved the status of a durable, reliable, and cost-ef-
 fective treatment option. As part of a general program of
 engineering assessment and  optimal design of slurry
 bioreactors. both bench- and pilot-scale reactors have
 been developed to address the pressing needs for miss-
 ing operational  data associated with slurry  bioreactor
 use. These reactors are located at the EPA Testing and
 Evaluation (T&E) Facility located in Cincinnati, Ohio.

 Methodology

 Application of slurry bioreactors to the treatment of con-
 taminated soil has been conducted with a variety of soil
 types (1). Case  studies and cost comparison are avail-
 able, but the information associated with these studies
 is incomplete (2). An EPA best demonstrated available
 technology (BOAT) study has investigated the applica-
 tion of slurry reactors to creosote-contaminated soil (3).

 To systematically evaluate each of the major compo-
 nents of slurry biotreatment,  a research program has
 been organized  along the general divisions of physics,
 otology, and  chemistry.  Each  of  these divisions is a
 major contributor to the slurry biotreatment process. The
 physics of mixing has been the earty focus of the slurry
 research program.  The criteria for optimal  mixing for
 slurries has not received the  required attention.  Five
 different criteria  have been advanced tor the chemical
 processing industry (4-7):  1) maximum  uniformity of
 suspension; 2) complete off-bottom suspension; 3) com-
 plete  on-bottom motion of all particles: 4) filleting but no
progressive fillet  formation; and 5)  height of suspension
 (cumulative particle  s'ze  distribution,  percent  solids,
percent suspension, weight-percent ultimate suspended
solids,  and  percent  ultimate  weight-percent  settled
solids).

For  the  initial  evaluation of the bench-scale reactor
(Figure 1), performance was assessed through the cor-
relation of critical factors contributing to the efficiency of
mixing (Figures 2  through 6).  Solids composition was
       Cleen Air
       
-------
                   30% Sand/Clay
     1O.OOO
        10
           Optfm*

 12


 10


 8
                                                I



                                                I
 Figure 3.  Complete off-bottom  suspension (5  In. between
          Impeller*, bsffle»desfgn 3).
 Figure 3. Cc npiete off-bottom suspension  (S  in. oetween
         Impellers, baffle-design 3).
   1.000
     100
     10
                 OpUnwl A«rtg«
                           '
                                            a. s


                                            •9
,.  !
.   I
                                            0.9
       0      10      13     JO     3O
Figure 4.  Complete off-bottom  suspension  (S In.  between
         Impellers, beffleadeslgn 3).
investigated for its influence on power consumption and
the rotational speed of the impeller (Figures 2 through
4). Clear optimal ranges for air flow are evident in the
recorded  data. The optimal  operating conditions are
found at the point where the lowest power is consumed.

A soil from St. Louis Park, Minnesota, was contaminated
with creosote constituents and  used to evaluate the
performance of bench-scale slum/ reactors. The bench-
scale bioslurry reactor was constructed from a 8-L glass
conventional resin kettle with a four-port cover fitted with
standard taper joints. The reactor vessel was fabricated
to hava three sample ports located 5 cm, 10 cm, and
15 cm vertically from dead center of the reactor bottom.
The ports in the reactor cover permitted introduction of
the stirring shaft, influent and effluent gas lines, and a
thermocouple  temperature probe  into the soil  slurry.
Operational slurry volume was 6 L or 75 percent of the
total reactor volume.

Ten bench-scale reactors were used to assess the effect
of engineering variables on the degradation of porycyclic
aromatic hydrocaroon  (PAH)  constituents over  a
10-week treatment period.

The experimental design of the treatability study is out-
lined in Figure  7.  Experimental  variables  selected for
this  study  were soil loading, rotational speed of the
mixing impeller, and dispersant  Soil solids concentra-
tions of 10 percent and 30 percent (dry weight basis)
were tested. Two mixing speeds were evaluated. A high
mixing rate was selected for  complete off-bottom sus-
pension. A tow mixing rate was arbitrarily set at 200 rpm
tower than the high mixing rate. Effective high mixing
rates were found to  be 650 rpm and 900 rpm  for the
10-percent and 30-percent soil solids, respectively. The
dispersant (Westvaco,  Reax  100M) was added to test
its ability to minimize foam production. Foam formation
is an operational problem associated with  *he aoplica-
tion  of soil bioslurry technology and is thought to be
connected  wrth naturally occurring organics  found in
certain soils.

Two separate reactors were operated under abiotic con-
ditions to serve as bioinactive control reactors. Formal-
dehyde was used as a biocide  in these reactors and
maintained at 2-percent residua!  concentration.

The  following monitoring and  operating conditions held
constant for the reactors:

• Dissolved oxygen greater than 2 mg/L

• pH range of 6 to 9

• Ambient temperature recorded daily

« Treatment duration of 10 weeks

• Nutrient C:N:P ratio = 100:10:1

• Antifoam as needed to control foam
                                                   87

-------
 Results

 For purposes of convenience, the individual PAH con-
 stituents were grouped into two categories: two- and
 three-ring compounds and four- through six-ring con-
 stituents. Initial concentration of total PAHs in the soil
 prior to treatment were 1,750 ppm in the  10-percent
 solids loading slurry and 2,047 ppm  in the 30-percent
 slurry,  indicating a degree of heterogeneity in the soil
 slurry. The total PAH  concentration was reduced to
 408 ppm in the 10-percent slurry (runs 1  through t)
 and 419 ppm in the 30-percent slurry (runs 5 through
 8) after 7 days of treatment. In the  10-percent slurry
 runs, the concentrations of two- and three-ring PAH
 compounds decreased from 709 ppm  to  67.4  ppm,
 and concentrations of  four- through six-ring PAHs de-
 creased from 1,041 ppm to 340 ppm; whereas for the
 30-percent slurry runs, the concentrations of two- and
 three-ring PAH compounds decreased from 798 ppm to
 45.1 ppm, and  concentrations of four- through six-ring
 PAHs  in the 30-percent slurry runs decreased from
 1,249 ppm to 374 ppm.

 Summary and Conclusions

 The totaJ PAH concentration was  reduced by 85  to 90
 percent after  70 days of treatment. The major decrease
 in PAH concentrations occurred in the first 7 days, where
 total PAHs removed ranged from 75 to 32 percent. Soil
 solids concentrations significantty affected removal rate
 and the final treatment  endpoint (PAH concentration). A
 maximum removal for the 30-percent solids loading was
 achieved after 21 days of treatment Continued treat-
 ment after 21 days had little effect on further reduction
 of PAH concentrations. In the 10-percent solids  runs,
 however, PAH concentrations continued  to be reduced
 between Days 21 and 70. The final  concentrations of
 two- and three-ring  and  four-  through six-ring  PAH
 categories, as well as total PAHs, for the  10-percerit
 solids runs were half the levels in the 30-percent soli'te
 conditions.

 These results show that removal efficiencies are appar-
 ently not as sensitive to complete off-bottom suspension
 as we had expected. Similarly, removal rates appear to
 be unaffected by mixing speed ranges. The dispersant
 additive did not effectively suppress foam formation or
 enhance PAH removal.

 This initial study dearty identifies soil solids composition
 as a major factor controlling treatment goals. Lower
 solids compositions and longer treatment duration may
 favor treatment to tower PAH concentrations in the soil.
 Because removal rates observed  in this work may be
 specific to the soil matrix selected for study,  the gener-
alizations arising from this work can be used for guid-
ance for future applications of  soil-slurry bioreactors.
Treatability studies are necessa-y, however, to deter-
mine the most  effective operating variables for each
Figure 5.  Air (low optimization (5 In. b»tw»«n impellers).
Figure 8.  Air flow optimization (6 In. b»fwt»n impellers).
 Run  A
                      Variable
 10
Dtspersot   0 mqn.

Mixing      450/700
Speed       rpm

Sent Solids   0

CHjO       0
50

650/900
  rpm

50 rrgrt.

50 i
Figure 7.  Experimental design (SL Louis Park toil).

waste matrix before embarking on any large-scale treat-
ment. Foaming potential of a contaminated soil should
be evaluated prior to treatment to minimize operational
problems associated with foam fo,Tr.ation at higher sol-
ids concentrations.
                                                  08

-------
References
1.  U.S. EPA. 1990. Engineering bulletin: Slurry biode-
   gradation. EPA/540/2-90/016. Cincinnati, OH.
2.  Ross,  D. 1990. Slurry-phase bioremediation: Case
   studies and cost comparisons. Remediation 1:61-74.
3.  U.S.  EPA. 1991.  Pilot-scale  demonstration  of  a
   slurry-phase biological reactor for creosote-contami-
   nated soil. EPA/540/A5-91/009. Cincinnati, OH.
4. Oldshue, J.Y.  1983. Fluid mixing technology.  In:
   Chemical engineering. New York, NY; McGraw-Hill.
   pp. 94-124.

5. Oldshue, J.Y  1983. Fluid  mixing  technology and
   practice. Chem. Eng. pp. 92-108 (June).

6. Oldshue, J.Y. 1990. A guide to fluid mixing. Roches-
   ter, NY:  LJghtnin.

7. Hic!'s, R.W., J.R. Morton, and J.G. Fenio. 1976. How
   to  design agitators  for desired  process  response.
   Chem. Eng. pp. 102-110 (April).
                                                39

-------
  Development and Evaluation of Composting Techniques for Treatment of Soils
                            Contaminated with Hazardous Wastes
                                    Cart L Potter and John A. Glaser
 U.S. Environmental Protection Agency, Andrew W. Breidenbach Environmental Research Center, Cincinnati, OH

                   Majkj Dosani, Srinivas Krisnnan, Timothy Deets, and E. Radha Krishnan
                             I.T. Environmental Programs, Inc., Cincinnati, OH
 Significant progress in optimizing condrtio'is and apply-
 ing the power of biotechnology to large-scale compost
 systems  requires a working understanding of the proc-
 esses and  mechanisms involved. Prototype  bench-
 scale units have been designed and tested to evaluate
 composting processes using contaminated soils. Identi-
 fication of suitable co-compost and bulking agents, ap-
 propriate  ratios  of  soil  to  organic components, and
 effective  aeration strategies and rates have been se-
 lected as major factors requiring investigation.

 This research program is designed to develop a thor-
 ough engineering analysis and optimization of compost-
 ing  as a  process  to treat soil  contaminated with
 hazardous waste. Bench-scale composters serve  as
 diagnostic tools to estimate the treatment capability of
 larger  systems. Fully enclosed, insulated reactors per-
 mit reliable data collection on mechanisms of metabo-
 lism and  the  fate  of  toxic  chemicals  during  soil
 composting.

 We are currently studying the ability of compost micro-
 organisms to bJodegrade porycydic aromatic hydrocar-
 bons (PAHs) in in-vessel reactors located at the EPA
 Testing and Evaluation Facility in Cincinnati, Ohio. Soil
 contaminated with PAHs was obtained from the Reilry
 Tar Pit Superfund site in Si Louis Park, Minnesota, for
 use in  this study.

 Background

 Composting holds potential to  provide tow-cost treat-
 ment of hazardous waste with  minimal environmental
 controversy. Commercial compost operations are oper-
ated as black-box systems in that optimization is largely
approached through trial and error. Treatment of hazard-
ous waste cannot be conducted with suboptimal con-
trols to meet the specified endpoints.
Some proponents of compost treatment have claimed
significant success in destruction of hazardous wastes
without strong data to support their claims. Disappear-
ance of parent compounds has been used to claim that
microorganisms successfully degraded waste chemi-
cals. Some toxic chemicals,  however, could potentially
adsorb to, or react with, humic substances in the com-
post and become undetectable by chemical  analysis.
Such toxicants might later desorb from humus and mi-
grate  to  the biosphere. This highlights the  need for
well-controlled studies to rigorousry document degrada-
tion rates and to identify metabolic products of hazard-
ous chemicals,  metabolically active microbia) species,
and mechanisms of hazardous chemical transformation
in compost systems.

The  conventional aerobic  compost  process passes
through four major microbiological phases identified by
temperature:  mesophilic (30°C to 45°C), thermophilic
(45°C to  75°C), cooling, and maturation. The greatest
microbial diversity  has been observed in the mesophilic
stage. Microbes found in the thermophilic stage have
been spore forming bacteria  (Bacillus sp.) (1)  and ther-
mophilic fungi (2,2). Microbial recolonization during the
cooling phase brings the appearance of mesophilic fungi
whose spores withstood the high temperatures of the
thermophilic stage. In the final compost stage, the matu-
ration phase, most digestible organic matter has been
consumed by the  microbial  population, and  the com-
posted material  is  considered stable.

Reactor Design

Ten 55-gal, insulated  stainless steel composters have
been constructed to perform closety  monitored treatabil-
ity studies. The units stand  upright, and air  flows up
through the compost mixture. Completely enclosed units
permit periodic analysis of volatile organic compounds
                                                 90

-------
 (VOCs) and online Analysis of oxygen, carton dioxide,
 and methane. Cylindrical reactor design permits mixing
 cf reactor contents by rolling each unit on a drum roller
 at desired intervals.

 Each compcster houses four thermocouples connected
 to a central computer for online temperature measure-
 ments. Thermocouples reside at three equally spaced
 locations within the compost mixture, and a fourth ther-
 mocouple tracks ambient temperature outside the reac-
 tion   vessel.  One  operational   scheme  permits
 temperature control  by introduction of  ambient  air
 through a computer-controlled varving system. If the
 temperature of a unit exceeds a predetermined value,
 the computer switches that unit to high air flow to cool
 the  reaction mixture. After the high-temperature unit
 cools to a specific temperature, the computer switches
 the unit back to low air flow.

 Periodic determination of compost moisture content in
 each reactor unit permits adjustment of total moisture
 content in the compost matrix to 40  to 50 percent.
 Moisture condensers .nside compost units promote re-
 cyding of moisture. Otherwise, each  unit could lose 10
 Ib to 15 Ib of water daily.

 Current Research

 Prototype composter evaluation has proceeded through
 several different designs. The performance of each de-
 sign was evaluated by conducting a treatability experi-
 ment  using the St.  Louis  Park soil. For our design
 criteria, one particular prototype offerad  considerable
 versatility. This design is currently being converted  to
 stainless steel reactor units.

 Current studies focus on defining acceptable operating
 conditions and process charac (eristics to establish suit-
 able parameters for  treatment effectiveness. Parame-
 ters of interest  include aeration,  moisture dynamics,
 heat production, and physical and chemical properties
 of the compost mixture.

 Aeration studies evaluate porosity (air flow) in the com-
 post system and attempt to discover relationships be-
 tween  free air space, forced air flow, and composting
 rate. Aeration studies also investigate roles of anaerobic
 and  aerobic metabolism in chemical  degradation. An-
 aerobic pockets  may benefit  the process by initiating
 degradation of recalcitrant compounds, especially highly
 chlorinated compounds, via reductive  metabolism. After
 ?n initial reductive step, aerobic bkxtegradation of toxi-
 cants may proceed more readily. The research program
 will attempt to identify optimal aeration rates and pile
 mixing frequency for  the most  effective combination  of
 anaerobic/aerobic conditions for biodegradation  of re-
calcitrant substrates.  These  studies will investigate
whether forced anaerobiosis and inoculation with a fac-
 ultative anaerobe prior to development of aerobic com-
post  conditions  enhances  biodegradation  of toxic
wastes.

Studies on moisture dynamics measure rates of change
in moisture content in different regions of the compost
reactor. A compost pile  can lose moisture  through
evaporation and  convection. Moisture  dynamics  are
evaluated  in  terms  of  aeration,  temperature,  and
compost composition (e.g., soil  type  and co-compost
material).

Heat production may be highly variable throughout the
compost reactor. We have devised a method to continu-
ally monitor temperature  changes (heat production) at
various reactor locations.  Bench-top  composters  are
insulated to control heat loss, thereby mimicking a large-
scale compost pile where heat is lost by ventilation and
water evaporation more than by conduction.

Physical properties of the compost mixture include den-
sity (g/cm3), pH changes in various reactor locations,
pressure drop across the pile if it is actively aerated, and
the fraction  of  solkJs,  moisture, and  organics.  These
investigations focus on the  potential to enhance biode-
gradation  by manipulation  of physical and  biological
parameters that influence the process. These studies
also will investigate whether recycling  mature compost
material into fresh compost enhances biodegradation of
contaminants.

Early microbiological studies will foe >s on characterizing
changes in biological activity during the four staqes of
composting. We will also attempt to identify microbial
species responsible for  significant  biodegradation  of
PAHs during each compost stage,  and look for reap-
pearance of fungi and other mesophiles (e.g., Actinamy-
cetes) during the cooling stage.

Future Research

Future  investigations will include technical develop-
ments necessary to imprcve composting applications for
degradation of  hazardrjs wcs*e. This will  involve in-
creased application of pilot-scale compost systems in
additior; to ongoing research in bench-top composters.
Emphasis will be placed on developirg techniques for
trapping VOCs from pilot-scale  systems, determining
mass balance of contaminant degradation in the com-
post and  identifying microtal species responsible for
biodegradaJon of contaminants.

Future studies also will attempt to validate extrapolation
of results from bench-top to pilot-scale and field demon-
stration levels. Maintaining a bench-top system at opti-
mum  conditions is relatively easy  compared with a
large-scale composter, where optimum conditions  will
not prevail at all times. The degree of variance frorr
optimal conditions  requires investigation  and approxi-
mation in small-scale systems.
                                                  91

-------
References                                       2. Fogarty, A.M., and O.H. Tuovinen. 1991. Microbio
1.  Nakasaki. K.. M. Sasaki, M. Shoda, and H. Kubota.       logicaj degradation of Pestic'd<* j" ^ w*ste com-
   1985. Change in microbial numbers during thermo-       P05^' Microb'01- Rev' PP" 225'233 (June)'
   philic composting of sewage sludge wrth reference    3. Strom,  P.P. 1985.  Identification of thermophilic bac-
   to  CO2  evolution   rate. Appl.  Environ.  Microbiol.       teria in solid-waste composting.  Appl. Environ. Mi-
   49(1):37-41.                                          crobiol. 50(4):906-913.
                                                 92

-------
        Remediation of Contaminated Soils from Wood Preserving Sites Using
                             Combined Treatment Technologies
                       Amid P. Khodadoust, Gregory J. Wilson, and Makram T. Suidan
          Department of Civil and Environmental Engineering, University of Cincinnati, Cincinnati. OH

                                          Richard C. Brenner
         U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH
 Pentachtorophenol (PCP), a pesticide used as a wocd-
 preserving compound since the 1930s, has been placed
 on EPA's National Priority List of pollutants (1). The
 cleanup of contaminated soil from PCP manufacturing
 facilities and wood preserving sites has been mandated
 through the Comprehensive Environmental Response,
 Compensation, and Liability Act (CERCLA) (2).
 Among technologies employing physical, chemical, and
 biological processes for the removal of PCP from con-
 taminated soils, solvent washing followed by biological
 treatment of the wash fluid appears to be a viable alter-
 native (3). The selection of the solvent depends on the
 hydrophobic nature of the pesticide and the soil wetting
 capability of the solvent (4,5). Mueller et al. (6) found
 that ethano) effectively removed polycyclic aromatic hy-
 drocarbons (PAHs) from wet contaminated soils. Pre-
 viously, equal proportions of ethanol and water were
 found to have the highest removal efficiencies for above-
 ground batch extractions of PCP from soil at various
 soilrsolvent ratios (7). In addition, 50-percent and
 75-percent ethanol solutions achieved higher removal
 efficiencies at low solvent throughputs in simulated in
 situ soil flushing experiments. Chemically synthesized
 extracts from the  soil washing process  were treated
 using an  anaerobic,  fluidized-bed granular  activated
 carbon (GAC) btoreactor. The PCP was reduced to an
 equimolar concentration of monochlorophenol, which
 caused inhibition of the biological system. Reduction of
 the feed concentration  of PCP to 200 mg/L appeared to
 alleviate reartor inhibition.

 Results and Discussion

 Solvent Extraction Studies
The effectiveness  of the 50-percent ethanol/water mix-
ture was  evaluated for the  removal of PCP from soils
that had been aged for 3 weeks, 3 months, 6 months,
9 months, and 1 year. The aging of soil spiked with
100 ppm PCP occurred in the absence of natural weath-
ering, i.e., the soil was  not exposed to ground and
atmospheric influences. The 50-percent ethanol/water
solution was used for simulated in situ soil flushing of
20 x40 and 100 x 140 U.S. mesh soils and 20 x40 U.S.
mesh soil conditioned at 60°C. The soil washing batch
experiments were conducted on 20 x 40 and 100 x 140
U.S. mesh soils and the clay fraction o! the original soil
and on 20 x 40 U.S. mesh soil conditioned at 60°C. The
in situ solvent washing (flushing) of soil was simulated
by continuously flushing solven* through a packed bed
of soil until the PCP concentration in the effluent did not
decrease. The aboveground soil washing was simulated
by batch extraction tests conducted on PCP-contami-
nated soil.

The 50-percent ethanol solution, applied as the flushing
solvent, consistently produced higher PCP removal effi-
ciencies at various aging periods from the  100 x 140
U.S. mesh soil than  from the 20 x 40 U.S. mesh soil.
The higher PCP recovery from the 100 x 140 U.S. mesh
soil was due to the larger mass transfer area (contact
surface) between the solvent  and the soil that the
smaller soil particle size provided.

The  data in Figure 1 show  the results from the batch
extraction tests performed on the 100 x 140 U.S. mesh
soil.  The results  indicate that the 50-percent ethano;
solution removed more PCP  from the soil than did either
the 100-percent  ethanol  solution or deionized  water.
Similar results were obtained for the other soil fractions.
This higher recovery of PCP by the 50-percent ethanol
solution was consistent throughout the study. The re-
sults also  show that PCP recoveries decreased after
9 months  of aging.  The PCP removal efficiency for
deionized water was  lower than that for the 100-psrcent
                                                 93

-------
ethanol solution after 6 months of aging, indicating that
the solubility of PCP in the hydrophobic  solvent was
contributing more to the removal of PCP from the soil
than was the superior wetting of  soil by water.
In addition to the batch extraction tests with the various
ethanol/water mixtures, sonication and soxhlet extrac-
tions with methanol/methylene chloride were carried out
on the same soil fractions. The results shown in Rgure 2
indicate that the PCP recoveries from the sonication and
soxhlet extractions of 100 x 140 U.S. mesh soil were not
superior to those from the batch extraction tests per-
formed with 50-percent ethanol solution. Similar results
were obtained for the other soil fractions.

Biological Treatment Studies
Anaerobic,  fluidized-bed GAC  anaerobic  bioreactors
were used for the biological  treatment of chemicalry
synthesized extracts (spent solvents)  from the soil sol-
vent washing process. The synthesized spent solvent
solution was fed to GAC bioreactors, where the PCP
content of the wash fluid was  the biodegradable meta-
bolite and ethanol served as the  primary substrate.
The effect of empty bed contact time (EBCT)  on  the
biodegradation of PCP and its  degradation products
was examined using the GAC bioreactor (8). Through-
          3  01 Water                             j
          •  50% Ethanol  Soil: Solvent Ratio of 1g: 100 mL  4
          *  ioo%Bhanol   lOOmgPCP/kgSorf(100ppm)  J
                       6       9
                     So* Age (months)
                                       12
                                                15
Figure 1.  Soil washing batch te*t» for 100 x 140 U.S. meah aoil.
     100 c-
      90 I-
      50 r
      X [•  e Soncatton
      20 !•  ''Sod Washing Batch Test wtth 50% Ethand
      10 i-  100mgPCP/VgSoil(lOOppm)
      0
                                      12
                                              15
Flgura 2.  Sonication and so»hl«t •xtractlona of 100 JT 140
         U.S. trwsh sod.
out the experiments, the influent PCP concentration was
maintained constant at 100 mg/L by doubling the mass
and hydraulic loadings simultaneously. The EBCTs were
based on an effective volume of 7 L (the total volume of
the reactor, 10 L, minus the volume due to a 30-percent
carbon expansion) divided by the  total  hydraulic flow
rate (Table 1).

Effluent concentrations of  PCP and its degradation by-
products are  shown in  Figure 3.  Influent and predicted
effluent (with  no biological activity) PCP concentrations
also are shown. In molar units, a relationship between
influent PCP and the total monochlorophenol concentra-
tion in the effluent indicates nearly complete conversion
of the influent PCP to monochlorophenol. PCP concen-
tration was reduced by at least three orders of magni-
tude  (a  greater  than   99-percent transformation)
throughout the study. No biological inhibition because of
PCP was observed during any phase,  and the EBCT will
be further decreased in future work.

Influent chemical oxygen demand (COO) was contrib-
uted by PCP,  ethanol, and trace salts.  As  seen in Rgure
4, a two-fold  increase in the COD loading rate occurred
each  time the mass end hydraulic loading  rates were
doubled (see Table 1). Only 5 percent  of the influent
COD  persisted in  the e.fluent  COD  throughout all
phases of the study, while 70 percent was accounted for
by the methane produced. The remaining 25 percent of
the influent COD was attributed to biomass production.
                                                       Table 1. Operation Summary of Bloreactor


                                                                            PCP    Ethanol
                                                       Ptiaa*
          Day* of
          Operation
                 Row
                 Rate    EBCT
                 (Ud)     (hr)
                                                       I
          48O-6O6


          607-324


          825-999
                                                                            0.60
                                                                                     4-28
                                                                                              8.0
                                                                                                     28.01
1.?0      8.33      12.0    13.99


2.40      16.66      24.0    7.01
                                                                  Ptl«*§l
                                                                                PTiajel
                                                                    BTfcivn MCP« [•ouaft

                                                                   ' "J'°"^uiuyjiigjtgig0au .njunu B a.!
                                                                                              Pliasa I
                                                                                                   "
                                                                500
                                                                         500
                          700
                          Day.
                                                                                         900
                                                                                                  900
Ftgur* 3.  PCP and PCP Intermediate affluent concentrations.
                                                   94

-------
                      Days
Figure 4. COO talanc*.
Weekly  analysis also was performed on the effluent
chloride ion concentrations, volatile fatty acids, and al-
cohols. The chloride potential is defined as the equimo-
lar amount of chloride from all potential sources (i.e., all
chlorinated phenols in the feed). The delta chloride rep-
resents  the difference between the  measured effluent
chloride concentration and concentration of chloride in
the influent These analyses confirmed that PCP under-
went biological  transformation to monochlorophenols
through the removal of four chlorine atoms per molecule
of the phenol.

References
1. Cirelli, D. 1978. Patterns of pentachlorophenol usage
   in the United States of America. An overview. In:
   Rao,  K.R. Pentachtororhenol. New York,  NY: Mar-
   cel Dekker, Inc. pp. 13-18.
2. U.S. EPA. 1989. Superfund Record of Decision (EPA
   Region   6),  United  Creosoting  Co.,  Conroe,
   Montgomery County, TX (2nd remedial action), re-
   port EPA/ROD/R06-S9/053.

3. U.S. EPA. 1990. Soil washing treatment. Enginee'-
   >rg bulletin. EPA/540/2-90/017. Cincinnati, OH.

4. Voice, T.C., and W.J. Weber,  Jr.  1983. Sorption of
   hydrophobic compounds by  sediments, soils, and
   suspended solids, Vol.  I. Theory and  background.
   Water Res. 17:1,433.
5. Karickhoff, S.W., D.S. B-own,  and T.A.  Scott 1979.
   Sorption of hydrophobic  pollutants on  natural sedi-
   ments. Water Res. 13:241.
          >
6. Mueller,  J.G., M.T.  Suidan, and J.T. Pferfer. 1988.
   Preliminary study  of  treatment  of contaminated
   groundwater from the Taylorville  gasifies site.  RR
   077. Hazardous Waste Research and  Information
   Center.

7. Khodadoust A.P., J.A. Wagner, M.T. Suidan, and S.I.
   Safferman. 1993. Treatment of PCP-contammated
   soils by washing with ethanol/water followed by an-
   aerobic treatment. In:  U.S. EPA.  Symposium on
   bioremediation of hazardous wastes:  Research, de-
   velopment,  and  field   evaluations   (abstracts).
   EPA/600/R-93/054. Washington, DC (May).

8. Wagner, J.A., AP.  Khodadoust M.T. Suidan,  and
   R.C. Brenner. 1993. Treatment of PCP-containing
   wastewater using  anaerobic  fluidized-bed  GAC
   bioreactors. Paper No. AC93-035-003. Proceedings
   of  the Water Environment  Federation 66th Annual
   Conference and Exposition, pp. 189-200.
                                                 95

-------
      Pilot-Scale Evaluation of Nutrient Delivery for Oil-Contaminated Beaches
                                Michael Boufadel and Makram T. Suidan
          Department of Civil and Environmental Engineering, University of Cincinnati, Cincinnati, OH

                                           Albert D. Venosa
         U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH
In situ bioremediation is emerging as an efficient and
economical strategy for the cleanup of oil-contaminated
beaches. The mechanisms and routas of nutrient deliv-
ery in the  presence  of  tides, however, are not well
understood. The main  objective  of  this project is to
investigate  these phenomena to identify the best nutri-
ent application technology.

Results and Discussion

For this  purpose, a pilot-scale beach simulation unit is
being built  This unit will be 8 m long, 0.60 m wide, and
1.8 m tall, and will be equipped with a pneumatic wave
generator. The unit is intended to simulate *aves that
propagate perpendicularly to beaches. The height of the
unit was selected to permit investigation of tidal effects.
Prior to constnjctJon of the pilot-scale  unit a  small
bench-scale unit was constructed and tested to observe
wave generation  and beach erosion. The results  ob-
served from the bench-scale unit were very encourag-
ing. A periodic wave was generated and sustained over
several days.
The initial part of the study will investigate nutrient trans-
port using tracer studies. A distributed computer model
will be developed in parallel. The model parameter will
be estimated  from the results of tracer studies. Sub-
sequently, the mode) will be evaluated at pilot scale and
later on real beaches. The experimental data also will
be evaluated  against the  mathematical model devel-
oped by Wise et al. (1).

References

1.  Wise, W.R., 0. Guvn, F.J. Molz,  and S.C. McCutch-
   eon.  1994. Nutrient retention time in a  high-perme-
   ability oil fouled beach. (In press)
                                                 96

-------
      Field Treatment of BTEX in Vadose Soils Using Vacuum Extraction or Air
                                     Stripping and Biofliters
                                            Rakesh Govind
                Department of Chemical Engineering, University of Cincinnati, Cincinnati, OH

                                E. Radha Krishnan and Gerard Henderson
                           International Technology Corporation, Cincinnati, OH

                                            Doltoff F. Bishop
         U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH
 Spills of fuels and leaking fuel tanks represent a major
 source  of vadose soil contamination. This contamina-
 tion, which includes  the aromatic hydrocarbons ben-
 zene, toluene, ethylbenzene, and the xylenes (BTEX),
 leaches through vadose soil into ground water. Aromatic
 hydrocarbons pose health risks when ground water is
 used as a drinking water supply.

 EPA's Risk Reduction Engineenng Laboratory (RREL),
 in cooperation with the University of Cincinnati, is devel-
 oping engineering  systems to bioremediate fuel-con-
 taminated  vadose soils  or ground  water.  Vacuum
 extraction of soils or air stripping of ground water, which
 transfers the volatile  organic compounds (VOCs) from
 the soils or ground water to air,  is combined with air
 btofiltration to achieve treatment.


 Field Demonstration

 Two types of air biofilters will be studied;   1) packed
 beds witfi  ceramic pellets,  6-mm  average diameter
 (Celite,  Manville Corporation), as the packing material;
 and 2)  straight-passages ceramic monoliths with
 50 square passages per square inch  (as shown in
 Figure  1). A schematic of the experimental system is
 shown in Figure 2. The aerobic mixed cultures, from an
 activated sludge treatment plant are immobilized on the
 surface  of the packing. Nutrient  solutL.i,  needed for
 microbial growth, is trickled  down through  the packed
bed, with the contaminated air flowing countercurrent to
the nutrient ficw. The gas residence time in each biofiiter
is varied between 1 and 3 minutes. Electricity and water
are used to raise the temperature of the extracted air to
approximately 30°C and to prehumidify the air. A syringe
pump is used during startup to contaminate the air with
jet fuel to establish the biofilms in the biofilters.

The  biofilters will be constructed  at EPA's  Test and
Evaluation (T&E) Facility in Cincinnati. The system will
include gas chromatography for analyses of the influent
and effluent gas streams from each biofiiter. The biofilm
on the support media will be preacclimated to jet fuel

              Treated Air
                              Nutrients
                              Biofflm
                              Straight-passages
                              Monolith
               t
           Corrtamwiated Air
Figure 1. Schematic of the straight-passage* monolith media.
                                                 97

-------
(JP-4) hydrocarbons. The skid-mounted biofilters with
acdimated biofilms  will  be transported to the- site for
connection to  the vacuum extraction or air  stripping
system.
                                       The sito for the  field demonstration has not yet been
                                       selected but is likely to be an air force base in Ohio. T're
                                       performance of the integrated  system will be  charac-
                                       terized for approximately 3 months.
                                                                       Blower
                    i
                rN-
Sample Port
ecyde ^


Btoftlter
Media



^J
Sample Port M^g
&

Sample Ports
Valves
^^/N| '".
Rov»
L U^J 1 1 "
Fksw

^••MavJl i llWh^
Blofitter
Media


Nutrient       Biofllter 1     Meter
 Tank
                                                        Me
-------
                      Dechlorination with a Biofilm-Electrode Reactor
                                 John W. Norton and MaKram T. Suidan
           Department of Civil and Environmental Engineering, University of Cincinnati, Cincinnati, OH

                                           Albert D. Venosa
         U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati OH
 Introduction

 Pentachlorophenol (PCP) is a pesticide and bactericide
 that is used widely in tne wood and leather preserving
 industries (1). PCP, however, is a susoecied mute gen
 and carcinogen (2), and, in 1986, EPA set a maximum
 contaminant level of 0.001 mg/L Superfund documents
 have reported PCP levels as high as several hundred
 milligrams per liter in contaminated ground water.

 According to Krumme (3), in systems without a carbon
 or energy source PCP has been snown to be dechlori-
 nated and mineralized to about 40 percent of the influent
 concentration (3).  In systems using a co-substrate, it
 has been demonstrated that PCP can be dechlorinated
 up  to 99.9 percent (4). The addition  of external carbon
 and energy sources, however, could pose difficulties in
 both in situ and ex situ treatment of contaminated sites.
 Cell growth is enhanced by ttia addition of these carbon
 and energy sources, and the disposal of the resultant
 sludge can prove to be costly. In situ treatment  of PCP
 can also pose problems; the addition of a carbon and
 energy source into the ground might cause the formation
 of hazardous scluble compounds. Methods of enhanc-
 ing  microbial activity that could reduce or remove the
 need to provide  external energy  and carbon  sources
 should be examined.

 Results and Discussion

The objective of this project is to examine the dechlori-
nation and mineralization of PCP under anaerobic con-
ditions using the  electrolytic  reduction  of  water  to
provide an external energy source and hydrogen _onor.
Researchers  have demonstrated  that biological proc-
esses can be enhanced when subjected to an  electric
current (5,6).  These studies examined the role of elec-
trolyticalry produced hydrogen in  tha denitrification of
wastewater. Islam et al. (6) found a correlation between
the  applied current and the removal efficiency  of the
reactor system and determined the optimum current to
be 20 mA, for which the removal efficiency was greater
then 98 percent

The reactor is a fixed-film chemostat with trace salts and
nutrients added. PCP dissolved in ethanol is added at
two different feed concentrations (5 mg/L and 50 mg/L),
with a current of 15.0 mA across the junction. The flow
rate is 5 L/day, with a hydraulic detention time of 0.44
days. The reactor was  seeded with biomass from an
anaerobic, expanded-bed,  granular activated carbon
(GAC) reactor that had been successfully dechlormating
PCP. The gas production of about 96 mL/day of methane
and the intermediates in the effluent indicate the pres-
ence of an active growing biofllm.

Good dechlorination of PCP was achieved,  with about
0.24 percent of the  influent PCP remaining as PCP, 0.1
percent as tetrachlorophenol, 0.87 percent as trichlo-
rophenol, 10.28 percent as dichlorophenol, and about
55  percent as moncchlcrophenol on  a mclar basis
(Figure 1). The remaining 33.5 percent was presumed
to be mineralized to HCI, CO2, and H2O. Currently, the
feed alcohol concentration is being reduced stepwise as
the biofilm stabilizes to the operating concentration.

Work on  this project is continuing; new data will be
included in the poster presentation.

References

1. Crosby, D.G. 1981. Environmental chemistry of pen-
   tachlorophenol. Pure Appl. Chem. 53:1,051-1,080.

2. Keith, L.H., and W.A. Telliard.  1979.  Priority pollut-
   ants, I. A  perspective view. Environ. Sci. Technol.
   13:416-423.

3. Krumme, M.L.,  and S.A. Boyd.  1988.  Reductive
   dechlorination of chlorinated phenols in anaerobic
   upfiow bioreactors. Water Res. 22 (2):171-177.
                                                 99

-------
                                             PCP and Intermediates vs Time
                                             Boelectrolytic reactor, 15.0mA
   002
 to
 8
 •3
   0.01  --
                                                a     a     a  o
                                                                       .O. a	QOQ.

                                                                         o
              „
             0°
                                     00
        367
          A  phenol
          o  MCP
          •  OOP
     387


v  tri
x  tetra
O  PCP
                                                 407
                                 427
                                                                                            44;
                                                       Days
influent
Rjur* 1.  PCP «vd IntcrnwdiotM v«r»ua tirn* (b*o«t»ctro»vHe rvactor, 15.0 mA).
4.  Gutfirie, MJV.. EJ. Kirsch, R.F. Wukasch, and C.P.L.
   Grady, Jr. 1984. Pentacrttorophenol bKXJegradation.
   II. Anaerobic. Water Res. 18(4):451-461.
5.  Meltor, R.B..J. Ronnenberg.W. Campbell, and S. Diek-
   mann.  1992. Reductior. of nitrate and nrtnte  in  water
   by immotxlized enzymes. Nature 355(20):717-719.
                               6. Islam, S., J.R.V. Flora, M.T. Suidan, P. Biswas, and
                                  Y. SakaKibara. 1993. Paper No. AC93-039-002. Pro-
                                  ceedings of the Water Environment Federation 66th
                                  Annual Conference and Exposition, pp. 217-225.
                                                  100

-------
        Use of Sulfur Oxidizing Bacteria To Remove Nitrate from Ground Water
                 Michael S. Davidson, Thomas Cormack, Harry flidgway, and Grisel Rodriguez
            Biotechnology Research Department Orange County Water District, Fountain Valley, CA
 The chemoautotrophic bacterium Thiobacillus denitrifi-
 cans is capable  of effective removal of nitrate  from
 ground water under anoxic conditions. This microorgan-
 ism is capable of deriving metabolic energy from oxida-
 tion of inorganic sulfur compounds including elemental
 sulfur,  hydrogen  sulfide,  thiosulfate,  metabisulfite,
 tetrathionate. and sulfite. All carbon required for biosyn-
 thesis is derived from carbon dioxide, carbonate, and
 bicarbonate. The primary products of autotrophic deni-
 trification are nitrogen  gas, sulfate, water, and biomass.
 The potential advantages of using elemental sulfur  (in
 powdered, flaked, or prilled form) are as follows: 1) low
 cost and wide availability of energy source; 2) tow tox-
 icity compared with other energy sources (i.e., methane^
 or ethanol); 3) ease and safety of storage;  4) potential
 for development of water treatment reactors capable of
 operating for long periods (months) at a time -with little
 or no maintenance or operator attention; and 5) potential
 for use  in situ to remediate  nitrate-contaminated
 aquifers.

 A column reactor  (3.6 m tang x  0.051  m ID) has been
 operated continuously for rr.ore than 1 year outdoors.
 The reactor was filled  initially to  a depth of 1.83 m with
 sulfur granules graded -16/+30 Mesh (U.S.  Standard
 Sieve). Well-water nitrate content could be consistently
 reduced to less than 0.3 ppm from an influent level of
 55 ppm with a reactor feed rate of 0.35 L/min. Increasing
 flow to 0.45  Lymin  resulted in  an  effluent containing
 nitrate concentrations ranging from less than 0.3 ppm to
 5 ppm. Maintenance of constant  bed volume for a given
 flow rate required periodic replenishment of the bed with
 fresh  sulfur granules.  As denitrification oroceeds, the
 granules decrease in mass (i.e., are consumed) to the
 point  that their mass is insufficient to remain within the
 reactor. A novel fluidized bed reactor system has been^
 designed that will permit essentially complete utilization
of the smaller particles.

A variety of heterotrophic (organotrophic) bacteria were
found to become established in  reactors fed only inor-
ganic  energy sources  (elemental sulfur or sodium thio-
sulfate). The first survey involved 15 bacterial isolates
recovered from a chemostat reactor operated with pre-
cipitated sulfur slurry as the energy source and nitrate
as the terminal electron  acceptor. The isolates were
recovered by plating dilutions of watsr samples on R2A
(an organic-based  medium) under aerobic conditions.
Isolates were punfied by restreaking on R2A and were
subjected  to a proprietary  identification system.  API-
NFT, designed to identify nonfermentative bacteria. Of
15 isolates, one isolate each was identified as Achromo-
bacter sp., Pseudomonas stutzeri, F!avobacteri-jm sp..
and Pseudomonas putrefaciens.  Seven of the isolates
were Gram-negative "nonidentifiable."  The remaining
four isolates were Gram-positive "nonidentifiable." The
second survey involved 19 isolates recovered from a
chemostat reactor operated with  sodium thiosulfate as
the energy source  and nitrate as the terminal electron
acceptor. Of these, one isolate each was identified as
Achromobacter sp., Pseudomonas pseudoalcaligenes,
and Pseudomonas paucimobilis.  Twelve isolates were
identified as Pseudomonas aeruginosa. Four isolates
were Gram-negative "nonidentifiable." The 'nonidentifi-
able" designation refers to isolates that gave biochemi-
cal reactions  profiles uncharacteristic of the API-NFT
database collection. Work in progress should result in
identification to the jenus level.

Sodium thiosulfate was tested as  an energy source in a
small, prototype fluidized bed reactor. The Pyrex column
(40 cm long x 2.54  cm ID) contained a 16-cm deep bed
of 0.10-mm diameter silica  LTjhsres (settled bed depth
under zero flow conditions).  In this reactor configuration,
the silica spheres serve only as: an inert support matrix.
Sodium thiosulfate is highly  soluole in water and can be
supplied in correct proportion with the aid of a metering
pump. The degree of bed expansion was essily control-
lable between  0 and 100 percent. The reactor demon-
stration involved recirculation  of  14 liters of a defined
mineral salts solution containing 1,227 ppm  nitrate and
2,252 ppm thiosulfate  through the  column. Following
inoculation, flow was  set at 30  ml/mm (equal to 25
percent bed expansion). Approximately 7 percent of the
                                                  101

-------
nitrate was removed by  Day  7.  Nitrate removal had
increased to nearly 35  percent by Day  11. Runs con-
ducted with varying concentrations of nitrate relative to
thiosulfate revealed that acceptable denitnfication effi-
ciency required careful control of the relative proportions
of the two reactants. While technically feasible, the level
of control required to reliably produce denitrified water
on a practical scale might prove difficult. Thiosulfate also
suffers from the disadvantage of higher cost per unit of
nitrate removed in comparison  to elemental sulfur.

Respirometric experiments were conducted using pure
cultures of Thicbacillus denitnficans. Washed cells ob-
tained-from aerobic cultures with  either thiosulfate or
tetrathionate as the energy source were unable to deni-
trify in short-term experiments. This demonstrates that,
as is the case with heterotrophic bacteria, denitnfication
is an inducible rather than a constitutive metabolic ca-
pability. However, anoxicairy grown cells could tolerate
exposure to oxygen without  immediate deterioration or
loss of denitnfication activity. On a practical level, this
suggests that a biological denitnfication reactor could
readily withstand periodic  ingress of oxygen resulting
from penodic  air-scour or  high flow  backwash proce-
dures, as might be required to control formation of ex-
cess biomass  deposits. Rapid recovery of denitnfication
activity following suc^ *rcatments would be  a  decided
advantage.
In conclusion, sulfur-mediated biological denitnficaticn
of ground water appears to  be technically feasible.  A
fluidized bed reactor containing granular sulfur ha,, been
operated for more than  1 year. Autotrophic sulfur bacte-
ria and i^jnautotrophic (organotrophic) bactena appear
to coexist stably. The nature of their relationship (possi-
bly syntrophic or mutualistic) is under  further study. The
use  of readily soluble sulfosalts  as  thiosulfate  or
tetrathionate in reactors containing an inert support ma-
terial is less certain. This approach will require additional
basic research  to determine the relationship between
nitrate concentration and energy-yielding substrate and
their overall effect on denitrification rates and efficiency.
                                                    102

-------
     Engineering Evaluation and Optimization of Biopiles for Treatment of Soils
                            Contaminated with Hazardous Waste
                                    Cart L Potter and John A. Glaser
 U.S. Environmental Protection Agency, Andrew W. Breidsnbach Environmental Research Center, Cincinnati, OH
Biopile systems offer the potential for low-cost treatment
of hazardous waste in  soil. Biopiles  provide favorable
environments for naturally occurring microorganisms to
degrade soil contaminants. The microbial environment
can  be manipulated to promote aerobic or anaerobic
metabolism. Air is supplied to the system by a plumbing
network that forces air through the pile by applying either
pressure or vacuum.
Siopites differ from compost piles in that bulking agents
necessary  for composting are not added to biopiles.
Some nutrients and exogenous microorganisms, how-
aver, may be added to  a biopile in the form of manure
or other nutrient-rich material. Biopiles will normally pro-
duce less heat than compost piles because less organic
substrate is added,  although significant aerobic  micro-
bial activity will produce some heat While heat produc-
tion  is often desired in compost piles, we may wish to
limit heat production in biopiles to avoid killoff of  meso-
philic organisms involved in biodegradation of soil con-
taminants.

The  goal of this project is to evaluate the potential of
biopile systems to remediate soils contaminated  with
hazardous chemicals. Pilot-scale reactors with a volume
of 2  yd3 to 3 yd3 each are being constructed at EPA's
Test and Evaluation (T&E) Facility in Cincinnati. Con-
taminated field soil from selected sites will be brought to
the T&E Facility for this research. Depending on avail-
ability of soil, contaminants may include any or all of the
following: pentachlorophenol, creosote, munitions, and
petroleum hydrocarbons.
Short-term work will focus on designing and construct-
ing pilot-scale biopile reactors and defining suitable op-
erating conditions. Pilot-scale operations may permit
collection of reliable data to develop effective aeration
strategies, document degradation rates ar.d metabolic
products of hazardous chemicals, and identify me'ibo-
lically active microbial species. Physical  and chemical
data to be collected  include heat production; density
(g/cm3); fractions of solids, moisture, and  organics;
pressure drop across sections of aerated biopiles; and
pH changes in various reactor locations. Subsequent
studies will emphasize treatability of contaminated soils.

Future investigations will  fofs on the potential to en-
hance biodegradation by  man'pulation of physical and
biological parameters. For  example, anaerobic treat-
ment may be necessary to initiate degradation of recal-
citrant compounds via reductive  metabolism. Following
reductive metabolism, toxicants may be amenable to
aerobic biodegradation. Research may identify the most
effective combination of  anaerobic/aerobic  conditions
for biodegradation of recalcitrant substrates  in biopile
systems.
                                                 103

-------
                           Section Five
                       Process Research
Process research involves isolating and identifying microorganisms that carry out
biodegradation processes and the environmental factors affecting these processes
In this way. researchers establish the building blocks of new biosystems for treat-
ment of environmental pollutants  in surface waters, sediments, soils,  and subsur-
face materials.Thorough evaluation ia critical at this level of research, since a firm
scientific basis can facilitate the scaling up of a  promising btoremediatior. method
or technology. Process research is being conducted on a number of environmental
pollutants.

Several research projects are focusing on the biodegradation of potycyclic aromatic
hydrocarbons (PAHs) and creosote. Specific areas of study include the metabolic
and ecological factors affecting the bioremediation of PAH- and creosote-contami-
nated soil and water; the  environmental factors  affecting creosote degradation by
catabolicalry  competent  microfiora, such as Spingomonas paudmobilis strain
EPA505; and a comparison of sulfur ar.d nitrogen heterocyclic compound transport
in creosote-contaminated aquifer  material.
Research also is being conducted on phenols, including a study on the modeling
of steady-state methanogenic degradation of phenols  in ground water at an aban-
doned wood treatment facility in Pensacola, Florida; and a study demonstrating the
conversion of pentachlorophenol  (PCP) to phenol in sediment slurries inoculated
with cells from a 4-bromophenol (4-BP) dehatogenating enrichment culture.

Two other  projects focused  on the dechlorination of polychlorinated biphenyls
(PCBs). One study examined limiting factors in order to develop effective methods
for stimulating microbial dechlorination of PCBs. Another study focused en the
addition of single cogeners of chloro- and bromophenyls for enhanced dechlorina-
tion of PCBs in contaminated sediments.

One project investigated the  kinetics of anaerobic biotransformation of munitions
wastes. Two others focused on the degradation of hydrocarbons, specifically the
effect of heavy metal availability and toxicity  on anaerobic transformations of
aromatic hydrocarbons and the biodegradation of petroleum hydrocarbons in wet-
lands microcosms, including constraints on natural and engineered remediation.

Another major focus of process research was the biodegradation of chlorinated
solvents, particularly trichloroethylene (TCE). One study focused on  the charac-
terization of bacteria in a TCE degrading biofilter. Another study  provided a risk
analysis for inoculation strategies in the bioremediation of TC"E. Related research
was conducted on the aerobic/anaerobic degradation of recalcitrant volatile chlo-
rinated chemicals in a hydrogel encapsulated biomass biofilter.

Other process research projects were the use of 5-chtorovanillate  as a  model
substrate for the anaerobic bioremediation of paper-milling waste; the effect of
surfactants  on  microbial degradation of organic contaminants; and reaction
                                   105

-------
mechanisms and  Development of remediation schemes related  to the covalent
bonding of aromatic amines to natural organic matter.
The  symoosium's poster session included  presentations  on metabolites of  oil
biodegradation and their toxicity, the alteration of a plasmid bactenal strain for TCE
degradation; degradation of a mixture of high molecular-weight PAHs by a myco-
bactenum species; and factors affecting the delivery of nutrients and moisture for
enhanced in suu bioremediation in the unsaturated zone.
                                 106

-------
     Metabolic and Ecological Factors Affecting the Bioremediation of PAH- and
                           Creosote-Contaminated Soil and Water
                                             PH. Prtchard
          U.S. Environmental Protection Agency, Environmental Research Laboratory, Gulf Breeze, FL

                                              Jian-Er Lin
                                Technical Resources, Inc., Gulf Breeze, FL

                                  James G. Mueller and Suzanne Lantz
                                 SBP Technologies, Inc., Gulf Breeze, FL
 Polycyclic aromatic hydrocarbons (PAHs) are a class of
 potentially  hazardous chemicals whose natural pres-
 ence in the environment is attributable to a number of
 petrogenic and pr-ytogenic sources (12). Environments
 contaminated with large amounts of these chemicals
 (e.g., creosote  waste,  coal tar processing sites) are
 considered hazardous owing to potential carcinogenic,
 mutagenic, and teratogenic effects of specific PAHs (3).
 Generally, high molecular-weight (HMW) PAHs, contain-
 ing four or more fused nngs, present the greatest poten-
 tial hazard to both the environment and human health
 (4). Consequently, much  interest axists in developing
 remedial methods, such as bioremediation, to selec-
 tively remove these chemicals from contaminated envi-
 ronmental materials.

 When environmental conditions (e.g., -waste load, nutri-
 ents,  oxygen, pH) are suitable, biodegradation of 'ow
 molecular-weight  PAHs by indigenous microorganisms
 readily occurs (5-7). Under the same conditions, how-
 ever,  biotransformation of HMW PAHs is less likely.
 Although bacteria  have been isolated in pure culture that
 grow on HMW PAHs, such as fluoranthene and pyrene
 (7-9), strategies for stimulating this  activity, as well as
 the degradation of other HMW PAHs, in contaminated
 soils are not readily available in part because of a poor
 understanding of the biodegradation ecology of complex
 mixtures of hydrophobic  hemicals in the environment.
 How, for example, do microorganisms interact during a
degradation process  to promote the  degradation of
these complex mixtures?  Can this interaction be en-
hanced  through population management of  microbial
communities or adjustment of specific  environmental
conditions?  And,  have  microbial  communities  in
contaminated soils adapted (genetically and/or physi-
ologically) to utilize hydrophobic PAHs more effectively''
An improvement of our understanding of biodegradation
ecology for PAHs and creosote could, therefore, lead to
new and effective strategies for bioremediation of these
contaminants. This paper provides a summary of our
research efforts in this area, with specific attention given
to cc-metabolic  processes, bioavailafcility,  inoculation,
and microbial community adaptation.

Results and Discussion

Co-metabolism

The process of co-metabolism in bioremediation generally
refers to the transformation (not necessarily mineraliza-
tion) of a  hazardous waste chemical(s) as an indirect or
fortuitous  consequence  of  the metabolism of another
chemical that a bacterium uses as a source of carbon and
energy (growth substrate). Co-metabolism, an intriguing
consequence of broad enzyme specificity, is one of the
important elements in the recent emergence of new biore-
mediation strategies. Unfortunately, however, its occur-
rence in  natural microbial communities is  neither well
documented nor understood, and the process is difficult to
control in the field. In addition, concerns exist regarding the
fate and environmental impact of the partial oxidation prod-
ucts that are thought to be produced. Successful degrada-
tion of HMW PAHs has been argued to involve extensive
co-metabolic reactions (6); that is, enzymes used by spe-
cific bacteria in a microbial community to degrade one type
of PAH fortuitously oxidize other PAHs. Biochemical evi-
dence for this type of reaction is provided in the paper
by Chapman et al.
                                                 107

-------
 The importance of co-metabolism in PAH degradation is
 illustrated by studies in  which  a bacterium  (Sphingo-
 monas paucimoblis strain EPA505) that used fluoran-
 thene (an HMW PAH containing tout fused rings and a
 major constituent of most creosote and ccal tar wastes)
 as a sole source  of carbon and energy was found to
 biotransform many PAHs that  wwe  not growth sub-
 strates (10). This  included fluorene, pyrene,  chrysene,
 and benzo(a)pyrene. If this bacterium and other PAH
 degraders are exposed to the PAH fraction of creosote
 in a standard shake flask assay (8) for 10 days and the
 creosote fraction  is monitored  by extraction and gas
 chromatographic analysis, considerable loss of most of
 the PAHs occurs even though only a  few of  the PAHs
 are used as growth substrates. A comparison of results
 from strain EPA505 and  strain N2P5,  a bacterium also
 isolated from creosote-contaminated  soil,  is  shown in
 Table 1. Strain N2P5 grew only on two- and  three-ring
 PAHs, such as phenanthrene, and had far less capacity
for »his co-metabolic phenotype. A variety of isolates are
currently being studied to more fully characterize this co-
metabolic capability. The resulting partially oxidized degra-
dation products from  this co-metabolism have not been
specifically identified but are likely to be more soluble and
possibly more biodegradable than the parent compound,
perhaps leading to further degradation or metabolism by
other members of a microbial community.

Other bacteria in nature may behave like these PAH
degraders studied in the laboratory, thereby giving mi-
crobial communities  the capability of co-metabolism.
Few experimental results  are available, however, to
show that this is indeed the case. We are conducting
experiments to specifically  relate pure culture studies to
PAH degradation patterns  in natural  microbial commu-
nities. At a  bioremediation site,  where environmental
conditions are established to promote PAH degradation
by  the  indigenous   microflora  (aeration,  inorganic
Tabta 1. Degradation of Craoaota PAH* by Selected Bacterial laolatea
Ompound (mg/L) Unlnoeulated (»d) EPA SOS (ad)
Naphthalene
Thiariaphthene
2-Metfiyl naphthalene
1-M«thylnaphthalene
Blphenyl
2,6-Dtmatftylnapnthalene
2,3-Dimethylnaphthalene
Acenaphthylene
Acenaphthene
Dlbenrofurmn
Ruorene
Dibanzothtophene
Phananthrane
Anthracene
Carbazola
2-Methytarrthracene
Anthraquinone
Ruoranttana
Pyrana
B«nzo(b)fluorana
B«nzo(a)anttiracena
Chrysene
B«nzo
22.46 (1.20)
16.01 (0.84)
19.83(1-22)
6.85 (0.58)
55.22 (3.00)
2.30 (0.16)
2.94 (0.28)
1.02 (0.55)
5.07 (0.76)
26.53 (2.31)
15.92 (1.40)
2.85 (0.20)
5.94 (2.49)
2.42 (1.12)
1.64 (0.20)
0.60 (0.02)
0.04 (0.01)
0.11 (0.03)
0.07 (0.02)
0.04(0.31)
bd
0.10 (0.03)
0.06 (0.03)
0.21 (0.07)
txfl
0.12 (0.02)
0.15 (0.05)
0.28 (0.15)
bcB
0.48 (0.10)
0.35 (0.12)
0.21 (0.11)
1.12 (0.21)
Mi
8.39 (0.75)
0.78 (0.08) .
5.98(1.00)
1.77(0.32)
1.19(0.21)
0.49 (0.06)
% Reduction N2P5 (ad)
100
92
100
99
100
96
91
62
100
99
99
96
100
83
88
79
78
100
47
73
0
27
27
18
0.10 (0.06)
0.79 (0.17)
0.17 (0.03)
0.79 (0.12)
0.62 (0.07)
0.50 (0.05)
0.37 (0.09)
0.48 (0.17)
10.06(1.30)
0.08 (0.01)
0.11 (0.06)
7.42 (0.56)
0.14(0.01)
1.09(0.05)
0.43 (0.11)
1.58 (0.06)
4.43 (0.57)
28.46 (4.09)
16.01 (5.90)
2.68 (0.11)
6.02 (0.09)
2.31 (0.09)
2.34 (0.16)
0.94 (0.11)
% Reduction
100
44
99
87
81
32
45
13
55
100
99
0
100
61
85
0
13
0
0
7
0
0
0
0
TOTAL
                             261.32
                                               21.92
                                                                                88.46
                                                  108

-------
nutrient amendment moisture control, etc.), however,
co-metabolism  may not have its maximum  effect be-
cause the PAHs serving as inducers of the enzymatic
processes responsible tor co-metabolism are not main-
tained at sufficient concentrations. As a consequence, it
may be reasonable to add a specific PAH in low concen-
trations to stimulate microbial communities to cc-meta-
bolicalry  degrade  HMW PAHs, thereby more  easily
bringing  PAH concentrations to stipulated cleanup lev-
els. Clearly, for any long-term bioremedation treatment
involving co-metabolism, more ecological and biochemi-
cal research is  required.

Bloavallabillt/

Because of their strongly  hydrophobic nature,  HMW
PAHs usually occur as contaminants in natural ecosys-
tems and waste treatment systems at mass levels that
exceed  their  water solubility.  In addition, equilibria
strongly favor particle-bound chemicals (e.g., sorbed to
soils). These characteristics large V account for the slow
biodegradation of HMW PAHs  (11). Therefore, under-
standing treatment conditions and environmental factors
that  can  be manipulated to enhance bioavailability and
consequently biodegradation is critical to bioremediation
considerations.

It has been suggested th?t pure cultures of bacteria can
use PAH compounds only in the dissolved state (12-14).
Therefore, the dissolution of PAHs may be a prerequisite
for initial oxidation and mineralization.  Dissolution rates
are usually  determined by  the solid-liquid contact  sur-
face area and the equilibrium concentration of the PAH
compound (11,12,15).  Surfactants can  enhance PAH
solubilization and dissolution, thus increasing the equi-
lijrium concentration of the compound in the aqueous
phase (16)  which should  lead to faster  degradation
rates. The use of surfactants at high concentrations,
however, reduced or inhibited  biodegradation (17,18)
because of surfactant toxicity to the bacteria used in the
study.

On the contrary, Teichm has shown that a variety of
nonionic surfactants are norrtoxic to a Afycotoaeferium
sp. that is able to grow on fluoranthene and pyrene and
consequently increase  rates of  PAH biodegradation
(19).  Likewise, we have studied iha mineralization of
14C-radiolabeted fluoranthene by S. paudmoblis strain
EPAS05, an organism !hat grov.s on this PAH as  a
carbon and energy source,  and initial rates of minerali-
zation were enhanced by the presence of the surfactant
Triton X-100. An example of th'i res,, jnse is shown in
Rgure 1  (top). For this experiment cells were grown in
complex  medium, washed several times in buffer, and
suspended to a final cell density of 8 x 10'° cells/mL in
minimal salts medium containing 20 mg of  unlabeled
fluoranthene, approximately 60,000 dpm of '^-fluoran-
thene, and various concentrations of Triton X-100. The
                       .00   150   200  250

                       Time (hrs)
Figure 1.  Mineralization profile* of  C-fluoranthene by strain
         EP450S In minimal ulta medium with vsriou» con-
         eeftUallooa of Triton X-100 (top) and vtrloua eombf-
         nattona of celle,  soil  partlclea,  *nd surfactant
         (bottom). Inoculum concentration • 8 * 1010 cells/mL;
         fluorantrtene concentration - 0.4  mg/mL; particle
         concentration • 30 mo/l_

surfactant concentrations tested were all abovethecriti-
cal micelle concentration for this surfactant. Initial rates
of mineralization were clearly enhanced by all concen-
trations of the surfactant The reduced extent of miner-
alization at the two highest surfactant concentrations
may have  resulted from the sequestering of fluoran-
thene degradation intermediates (e.g., leaching from the
cells), making them unavailable for mineralization. The
bacterium dearty was  able to  tolerate  high surfactant
concentrations, thus emphasizing the  importance  of
property selecting PAH-degrading microorganisms that
are not inhibited by surfactants or selecting surfactants
that 
-------
 degradation rates, in essence, are counterbalanced by
 greater chemical turnover  as is true in  the  case of
 fluoranthene  degradation. As  shown in Figure  1
 (bottom), an  aqueous  suspension  of soil particles
 (30 mg/mL) and fluoranthene crystals (20 mg)  together
 resulted in greater mineralization rates than  suspen-
 sions  with only fluoranthene crystals, apparently the
 effect of higher solid-liquid contact. Although increases
 ir biomass or in the activity of the biomass as a result
 of exposure to soil particles also may explain the affect,
 this explanation  is unlikely since  the biomass
 (108  cells/mL) and the mineralization rates  were in-
 itially high.  Note that the effect of soil particles was
 equivalent to that of adding surfactant a further indica-
 tion of increased dissolution by either material.

 In  contaminated  soils, fluoranthene  and  other HMW
 PAHs  will likely exist at concentrations far in excess of
 their aqueous solubility. Given that undissolved PAHs
 will not exist as crystals in the environment it is impor-
 tant to know rf they exist in a form in which soil  particles
 provide higher solid-liquid contact or in which surfac-
 tants can promote greater dissolution, or both.
 Research needs to be accelerated in this area  because
 the use of surfactants will almost assuredly play a sig-
 nificant role in future bioremediation procedures. Also,
 engineering strategies for  using surfactants  or other
 means of increasing mass transport in the field must be
 developed. This should include consideration of how to
 remove the bioavailability-enhancing chemical  from the
 field after it has done its job, and how to protect against
 a negative effect on contaminant distribution in the field
 (e.g., seepage into uncontaminated areas).

 Bloaugmentation

 If we define bioaugmentation as the process of introduc-
 ing microorganisms of sufficient biomass into a site in a
 manner in which it can be Documented that the inocu-
 lated  organism(s)  survives to a point of  significantly
 affecting the fate of a target chemical(s), then very few
 scientifically documented examples  exist where  this
 process has been successful on a significant scale. Yet,
 many possible situations can occur in which bioaugmen-
 tation of chemically contaminated sites with microorgan-
 isms  possessing  unique and  specialized metabolic
 capabilities could potentially be a feasible bioremedia-
 tion approach. With more careful attention  to selection
 ar -"• application of the inoculants, it is quite reasonable
 that bioaugmentation could become a major and effec-
tive component of biological cleanup methods.

 Many recognizable limitations to the use of bioaugmen-
tation in bioremediation  exist. Only a few limitations
have been systematically addressed in an experimental
sense (20-23). These include the inability to support the
growth  and/or activity  of  the  introduced organism
because of competition  by the indigenous microflora.
Success, however, can be realized by employing spe-
cialized techniques to reduce competition and to main-
tain  a biomass  high  enough  to  effect  efficient
degradation of the  target chemicals.  In  addition, the
contaminated environment almost certainly will have to
be physically modified, perhaps over an extended pe-
riod, to optimize the bioaugmentation process. This
modification generally means establishing conditions in
which  the  availability of oxygen, inorganic nutrients,
tomperature, degradable substrate, moisture content,
etc., are optimized.

Bioaugmentation using microorganisms with  requisite
metabolic capacities is or' M jgested approach for en-
hancing biodegradation  or inese  HMW  PAHs (6). Al-
though  biodegradation of HMW PAHs  by identified
microorganisms has been reported, suitable strategies
for using these microorganisms as inocula  in the field
need to be further developed. We have  been experi-
menting with the concept of introducing immobilized
cells using  different encapsulation procedures (24). For
example, polyurethane polymer (PU) has been used to
immobilize  S. paucimobilis strain EPA505. The immobi-
lized cells  were tested for their  ability to  mineralize
fluoranthene under these conditions.  As shown in
Figure 2 (top), no significant difference in fluoranthene
mineralization profiles by the PU-immobilized cells of
strain EPA505 occurred when compared  to nonimmo-
bilized  cells. Since the same inoculation size was used
in all flasks during this experiment, the results suggest
that  the immobilization process does  not significantly
affect microbial activity. Cells immobilized  in the PU
polymer remain active for months when stored at 4°C.

Active  immobilized cells  then offer several additional
possibilities for further enhancing biodegradation and
environmental control. For example, inclusion of adsor-
bents in the immobilization matrix can  result in a more
rapid uptake of toxic compounds from the environment,
thereby potentially providing greater accessibility of the
adsorbed chemical  to the immobilized bacteria. Two
issues  need to be  addressed,  however, when using
co-immobilized adsorbents: 1) Is microbial  activity af-
fected  by co-immobilization with adsorbents? and 2) Is
availability of the adsorbed chemical to the immobilized
cells maximal? To study these questions, diatomaceous
earth and powdered activated carbon were co-imrr.cbi-
lized with strain EPA505 in the polyurethane matrix. In
Figure  2 (top), the degrading activity of the cells co-im-
mobilized with the adsorbents was the same as the
nonimmobilized cells,  indicating that the degradation of
the adsorbed fluoranthene was complete.

Another possibility involves in situ bioremediation situ-
ations,  where direct addition of nitrogen and phospho-
rous  into   soil  or  water  may  have  a  negative
consequence because of enhancement of the activity of
undesired indigenous  microflora and/or the leaching of
                                                  110

-------
        60


      0 50

      d
      3«
      i^
      e repli-
cons are quite large, 3,400, 2,300, and 1,200 kilobases
in size. The presence of these mega-plasmids has been
reported for other species of Pseudomonas (25), js well
as other genera of bacteria. The physiological and ge-
netic functions  of these mega-plasmids are unknown,
but they may be related to the large and broad metabolic
capability that these organisms  possess and perhaps
even to the ability  to degrade fluoranthene. By under-
standing more about this  genetic makeup, it may even-
tually be possible to manipulate adaptation  in the field
in a time frame that could accelerate or increase the
extent of bioremediation.

Summary and Conclusions

The  successful  bioremediation  of PAH-contaminated
soils and sediments  requires  a clear understanding  of
the metabolic and ecological factors that can be manipu-
lated to increase the rate and extent of PAH biodegra-
dation. We provide evidence  in this report suggesting
that 1) co-metabolism may bo a potential mechanism for
                                                   111

-------
degradation of HMW PAHs; 2) bioavailotility of PAHs
may be improv-aa through the application of surfactants;
and ?'. f.a success of bioaugmentation "  / be  in-
creased by the use of procedures that immobilize PAH-
degrading    microorganisms,   adsorbents,   and/or
nutrient's.  In addition, the knowledge of how microbial
communities become adapted for enhanced PAH biode-
gradation may play an  important role in developing fu-
ture strategies for bioremediation.

References

1.   Grosser,  R.J.,  D.  Warshawsky, and J.R.  Vestal.
     1991. Indigenous and enhanced mineralization of
     pyrene, benzo[a]pyrene, and  carbazole in soils.
     Appl. Environ. Microbiol. 57:3,462-3,469.

2.   National Academy of Science. 1983. Porycydic aro-
     matic hydrocarbons: Evaluation of sources and ef-
     fects. Washington,  DC:  National Academy Press.

3.   Moore, M.N..  D.R.  Livingstone, and  J. Widdows.
     1989. Hydrocarbons in marine mollusks: Biological
     effects and ecological consequences. In: Varanasi,
     U., ed. Metabolism of PAHs in the  aquatic environ-
     ment Boca Raton,  FL  CRC Press.  Inc. pp. 291-
     328.

4.   U.S.  EPA.  1982.  Wood preservative  pesticides:
     Creosote,  pentachtorophenol,  and the inorganic
     arsenical  (wood uses). Position  Document 213.
     EPA 540/9-82/004. Washington, DC.

5.   Mueller, J.G., S.E  Larrtz, B.O. Blattmann, and PJ.
     Chapman. 1991. Bench-scale evaluation of  alterna-
     tive biological treatment processes for the remediation
     of  pentachlorophenot-  and  creosote-contaminated
     materials: Slurry-phase bioremediation. Environ. Sci.
     Technol. 25:1,055-1,061.

6.   Mueller, J.G., S.E. Lantz. R.J. Colvin, D. Ross, D.P.
     Middaugh, and P.H. Pritchard. 1993. Strategy using
     bioreactors and specially selected microorganisms
     for bioremediation  of ground water contaminated
     with creosote and pentachtorophenol.  Environ. Sci.
    Technol. 27:691-698.

7.  Cemiglia, C.E.  1993. Biodtxjradation  of porycyclic
    aromatic hydrocarbons. Biodegradation 3:351-368.

8.  Mueller, J.G.,  P.J.  Chapman, and P.H. Pritchard,
    1989. Action of a  fluoranthene-utilizing bacterial
    community  on  porycyclic aromatic  hydrocarbon
    components of creosote. Appl. Environ. Microbiol.
    55:3,085-3,090.

9.  Weissenfels, W.D.,  M.  Beyer, and J.  Klein. 1990.
    Degradation of phenanthene, fluorene, and fluoran-
    thene by pure bacterial cultures.  Appl. Microbiol.
    Biotechnd. 34:528-535.
10. Mueller, J.G., P.J. Chapman, E.O. Blattmann, and
    P.H. Pritchard. 1990. Isolation and characterization
    of a fluoranthene-utiiizing  strain of Pseudomonss
    paudmobilis. Appi.  Environ. Microbiol.  56:1,079-
    1,086.

11. Volkerling, F, A.M. Breure, A. Sterkenburg, and J.G.
    van Andel. 1992. Microbial degradation of pclyrvdic
    aromatic hydrocarbons: Effect of substrate avail-
    ability on bacterial growth kinetics. Appl. Microbiol.
    Biotechnol. 36:548-552.

12. Stucki, G., and M. Alexander. 1987. Role of disso-
    lution  rate and solubility in biodegradation of aro-
    matic   comijounds.   Appl.  Environ.   Microbiol.
    53:292-297.

13. Wodzinsto, R.S., and D. Eertolini. 1972. Physical
    state in which naphthalene and bibenzyl are utilized
    by bacteria. Appl. Microbiol. 23:1,077-1,081.

14. WodzinsW, R.S., and J.E.  Coyie. 1974. Physical
    state of phenanthrene for  utilization  by  bacteria.
    Appl. Microbiol. 27:1,081-1,084.

15. Thomas, J.M.,  J.R.  Yordy, J.A.  Amador, and  M.
    Alexander. 1986. Rates of dissolution and biodegra-
    dation of water-insoluble organic compounds. Appl.
    Environ. Microbiol. 52:290-296.

16. Edwards, D.A., R.G.  Luthy, and Z. Liu. 1991. Solu-
    bilization of porycyclic aromatic hydrocarbons  in
    micellar nonionic surfactant solutions.  Environ. Sci.
    Technol. 25:127-133.

17. Aronstein, B.N., Y.M. Calvillo,  and M. Alexander.
    1991.  Effect of surfactants at hw concentrations on
    the desorption and biodegradation of sorted aro-
    matic  compounds  in soil.  Environ.  Sci. Technol.
    25:1,728-1,731.

18. Laha, S., and R.G. Luthy. 1992. Effects of nonionic
    surfactants on the solubilization and mineralization
    of phenanthrene in soil-water systems. Biotechnol.
    Bioeng. 40:1,367-1,380.

19. Tiehm, A. 1994. Degradation of porycyclic aromatic
    hydrocarbons in the  presence of synthetic surfac-
    tants. Appl. Environ.  Microbiol. 60:258-263.

20. Guerin, W.F.,  and S.A. Boyd.  1992. Differential
    bioavailability of soil-sorbed naphthalene to  two
    bacterial  species.   Appl.  Environ.   Microbial.
    53:1,142-1,152.

21. Pritchard, P.H. 1992.  Use of inoculation in bioreme-
    diation. Curr. Opin. Biotechnol.  3:232-243.

22. Goldstein  R.M., LM. Mallory,  and M. Alexander.
    1985.  Reasons for possible failure of inoculation to
    enhance biodegradation. Appl. Environ.  Microbiol.
    50:977-983.
                                                  112

-------
23. Comeau, Y., C.W. Greer, and R. Samson. 1903.       and application  of PAH-degrading microorganisms
    Role of inoculum preparaton and density on the       In: U.S. iJaval  Research Laboratory report. Con-
    bioremediation  of   2,4-D-contaminated  soil  by       tract No. N00014-90-C-2136 thresh Geo-Centers,
    bioaugmentation.  Appl.  Microbiol.  Biotechnol.       Inc., Ne-.vton Upper Falls, MA.
    38:681-687
                                                     25. Hai-Ping, C., and T.G. Lessie. 1994. Multiple repli-
24. Lin, J.E., J.G. Mueller, K.J. Peperstaete, and PH.       cons  composing the genome  cf Pseudomonas
    Pritchard. 1993. Identification of encapsulation and       cepac'a 17616.  J. Bacterio. (In press)
    immobilization techniques for production, storage.
                                                 113

-------
              Metabolic Pathways Involved in the Biodegradation of PAHs
                                           Peter J. Chapman
                          U.S. Environmental Protection Agency, Gulf Breeze, FL

                                          Sergey A. Seltfonov
                                  University of Minnesota, St. Paul, MN

                                            Richard Eaton
                          U.S. Environmental Protection Agency, Gulf Breeze, FL

                                             Magda Grifoll
                                University of Barcelona, Barcelona, Spain
 The principal sources of polycyclic aromatic hydrocar-
 bons (PAHs) in the environment are the us j and spillage
 of fossil fuel-related materials, either petroleum- or coal-
 derived.  Both sources contain  complex  mixtures of
 PAHs but differ in Amount and composition. Coal-based
 materials such as creosote and coal tar are rich in PAHs,
 with relatively litHe aJkyl substitution. Petroleum, on ti",e
 other hand, generally contains a smaller fraction of PAHs
 composed of a wide array of alkyl-substituted homologues.
 Knowledge of the aerobic biodegradation of PAHs derives
 largely from studies of pure bacterial cultures isolated for
 their ability  to utilize for growth single,  unsubstituted aro-
 matic hydrocarbons such as naphthalene, biphenyl, and
 pnenanthrene (1). In all cases studied, catabolism is initi-
 ated by oxygen adding reactions usualry forming c/s-dihy-
 drodiois on arene rings. While biological  methods  for
 removal of  PAH-containing environmental contaminants
 are now seriously considered options for remediation, de-
 tails of the  processes involved are little understood. For
 example, little is known of the extent to which biotransfor-
 matton (co-metabolism) is involved in the remcvaJ of higher
 molecular weight PAHs in complex mixtures and the or-
 ganisms and growth substrates required. Are products of
 biotransfomriation accumulated?  What are their environ-
 mental effects?

Some recent  findings relevant to these  questions are
summarized below.

Naphthalene Degradation: New Insights

Investigation of reactions of naphthalene  degradation
catalyzed by enzymes encoded by the NAH7 plasmkj
was undertaken using a molecular biological ap-roach
involving cloning and subcloning of pathway genes (2).
As a  result, a collection  of  strains  of Ps&udomonas
aeruginosa was obtained containing key genetic se-
quences of the plasmid encoding for the degraaative
pathway extending various distances from naphthalene.
Such  strains were  used to accumulate,  undor physi-
ological conditions,  catabolites of naphthalene other-
wise difficult to isolate and characterize. As a resuli,
tfans-2-hydnxy benzylidene pyruvate was identified as
a metabolite of 2-hva-oxy chromene-2-carboxylic acid
and a new reaction was recognized as responsible for
formation of salicylaldehyde and pyruvate by means of
a novel hydratase-aldolase enzyme.

Degradation of Creosote PAHs

For studies of the  bactenal degradation of  creosote
PAH-*, an aromatic  hydrocarbon fraction free of polars,
resins, and phenols, with little if any N-heterocyclic ma-
terial  was obtained by column chromatography. Enrich-
ments employed this fraction in mineral salts medium to
establish cultures (from creosote-contaminated soils).
These were incubated with shaking at 20°C to 24°C in
the dark, with transfers biweekly. Amounts of remaining
PAHs, determined by gas chromat igraphy/flame ioniza-
tion detector (GC-FID) after methylene chloride extrac-
tion, showed extensive losses of iow molecular weight
PAHs not accounted *or by abiotic losses. Fluoranthene,
pyrane, and PAHs  with higher retention times were
recovered essentially unchanged, being associated with
insolro.'-i  black  resinous  material   accumulated in
cuinjres.  Column   chromatography and  thin-layer

-------
chromatography haj> shown this material contains both
tow molecular weight neutral  products and  complex
polymeric matenal. Among the neutral products identi-
fied were acenaphthenone, fluorenone,  and other ke-
tones formed from naphtheno-aromatics.  Certain of
these products previously have been shown to result
from the action of bacterial reductive dioxygenases (3).

Naphthalene Dioxygenase Action on
Naphtheno-Aromatic Hydrocarbons

With  the cloned genes of naphthalene dioxygenase
available in a strain of P. aervginosa (2), it was possible
to investigate the action of a reductive oxygenase on
simple naphtheno-aromatic hydrocarbons and related
compounds (4). Induced cells were incubated in buffer
with  fluorene,  acenaphthene,  acenaphthylene,  and
other hydrocarbons having benzylic functions; products
were  extracted for  characterization. Ruorenone was
identified as a product of fluorene oxidation, with ace-
naphthenone  formed  from  acenaphthene  and ace-
naphthylene   together  with  a  os-dinydrodiol  and
acenaphthenequinone in the latter case (Figure 1).

Evidentty the first formed secondary alcohols are acted
on by broad-specificity cellular dehydrogenases to give
ketonic er,d products. Apparentty anomalous oxidations
at benzylic positions, such as observed here, may be
expected in situations where btodegradation of mixtures of
aromatic and naphtheno-aromatic hydrocarbons occurs.

Bacterial Utilization  of a
Naphtheno-Aromatic: Fluorene

Given that oxidation of benzylic functional groups may
be unavoidable  when  arene dioxygenases are con-
fronted  by naphtheno-aromatics, it was  of interest to
examine whether such reactions are involved when bac-
teria utilize naphtheno-aromatics as growth substrates.
Accordingly, the reactions employed in the utilization of
fluorene by a Pseudomonas isolate were investigated. An
earlier study with a different strain (5) suggested that the
productive route of catabolism involved initial aromatic-ring
cfioxygenation and cleavage and that fluorenone was a
dead-end metabolite. By contrast  the pathway  estab-
lished for the Pseudotmnas isolate is inrtiated by benzytic
oxidation leading to fluorenooe formation. Subsequent re-
actions include formation of a novel angular did (6) before
opening  the central  five-membered  ring  to  generate a
cfihydroxylated biphenyl carboxylic acid (Rgu^ 2). This
route (7,8) represents a significant difference from earlier
characterized routes  initiated by conversion of arenes to
os-dihydrodiols in that the naphthenic ring is first oxidized
and then opened, thereby  accommodating both fluorene
and fluorenone.

Organisms possessing this biochemistry, therefore,
are equipped to channel  products of  anomalous
Flgunt 1. Tran»formation if n«phtt>«fio-*rom«tlc3 by
        l«f*»
Figure 3. Rout* of fluorvn* degradation In Pafudomonts F274

oxidation by arene dioxygenases into productive cata-
bolicpathways.

References

1. Gibson,  D.T., and V. Subramanian. 1984. Microbial
   degradation of aromatic hydrocarbons. In:  Gibson,
   D.T., ed. Microbial degradation of organic com-
   pounds.  New York, NY. and Basel, Switzerland:  Mar-
   csl-Dekker, Inc. pp.  181-250.

2. Eaton, R.W., and P.J. Chapman. 1992. Bacterial me-
   tabolism of naphthalene:  Construction and use ot
   recombinant bacteria to study nng cleavage of 1,2-
   dihydroxy-naphthalene and subsequent reactions. J.
   Bacteriol. 174:7,542-7,554.

3. Schocken, M.J., and D.T. Gibson. 1984. Bacterial
   oxidation of the polycyclic aromatic hydrocarbons,
   acenaphthene  and  acenaphthylene. Appl.  Environ.
   Microbid. 48:10-16.

4. Selifonov, S., M. Grifoll, R.W. Eaton, and P.J. Chap-
   man. 1993. Oxidation  of the naphtheno-aromatic
   compounds,  acenaphthene, acenaphthylene, and
   fluorene, by  naphthalene oxygenase  cloned  from
   plasmid  NAH7. Abstr.  #Q345.  93rd Annual  ASM
   Meeting, Atlanta, GA.
                                                 115

-------
5.  Grifoll,  M., A.M. Solanas, and J.M. Bayona.  1990.    7.
   Isolation and chaiacterization of a fluorene-degrad-
   ing bacterium:  Identification of ring oxidation and
   ring  fission  products.  Appl.  Environ.  Microbiol.
   58:2,910-2,917.

6.  Seiifonov, SA, M. Grifoll, J.E. Gurst, and P.J. Chap-    8.
   man. 1993. Isolation and characterization of (+)-1,12-
   dihydroxy-1-hydrofluorenone   formed   by  angular
   dioxygenation in the bacterial catabolism of fluorene.
   Biochera Biopnys. Res. Commun. 193:67-76.
Trenz, S.P.,  K.H.  Engesser, P.  Fischer, and H-J.
Knackmuss.  1994. Degradation of fluorene by Bra-
vibacterium sp. strain DPO 1361: A novel c-c bond
deavage  mechanism via  I,l0-dinydro-i.io-dihy-
droxyfluoren-9-one. J. Bactenol.  176:789-795.

Grifoll. M., S.A. Seiifonov, and P.J. Chapman. 1994.
Degradation of fluorene by Pseudomonas sp. F274:
Evidence for a novel degradative pathway. Appl. Er-
viron. Microbiol. (In press)
                                                116

-------
               Environmental Factors Affecting Creosote Degradation by
                        Sphingomonas paucimobilis Strain EPA505
                                 James G. Mueller and Suzanne E. Lantz
                                 SBP Technologies, Inc., Gulf Breeze, FL

                                            P.M. Pritchard
          U.S. Environmental Protection Agency, Environmental Research Laboratory, Gulf Breeze, FL
The presence  of  polycyclic  aromatic  hydrocarbons
(PAHs) in soil and ground water is recognized by EPA
as a priority environmental problem. Because of inade-
quacies intrinsic to the design and operation of wood
treatment facilities of the past coal tar creosote repre-
sents one of the major anthropogenic sources of exces-
sive PAH concentrations in the environment (1). Coal tar
residues  from coal gasification and creosote distillation
processes represent another major source of environ-
mental PAH contamination.

Of  the hundreds of locations potentially affected by
PAHs from industrial operations, most have been thor-
oughly assessed and  characterized. In cases where
remedial  actions to restore soil and ground water have
been prescribed, a variety of treatment alternatives have
been evaluated.  Unfortunately, many of the more con-
ventional  approaches  have  proven ineffective and/or
prohibitively expensive. For example,  ground-water
pump-and-treat approaches have proven ineffective for
PAH-contaminated aquifers (EPA Office of Solid Waste
and Emergency Response  [OSWER]  memorandum,
May 27,1992). For soils, excavation followed by secon-
dary treatment (e.g., soil washing followed  by slurry-
phase biotreatment) is of such a scale that costs and
practicability have become prohibitive. In addition, from
an end-user's perspective, many conventional remedial
technologies are unacceptable  because of regulatory
problems and technical feasibility.

Of the alternative remedial options available for creo-
sote-contaminated soil, bioremediation may represent a
technology of choice. Despite the many potential advan-
tages of bioremediation, the reported effectiveness of
PAH biodegradation  in contaminated media has varied
(2). This variability is due to  a  number of recognized
factors, including the presence of free product as dense
nonaqueous  phase liquid  (DNAPL)  and/or   light
nonaqueous phase liquid (LNAPL), the heterogenous
nature of soil and subsurface matrices, and the use of
ineffective delivery and implementation strategies. From
a biological perspective, effective biodegradation is in-
fluenced, in part, by the presence of catabolically com-
petent microflora at a contaminated site and by certain
environmental factors that enhance the activity of this
microflora,  including availability and concentration  of
electron acceptors, inorganic nutrients, and the target
chemical(s). The ability to control and regulate th"se
factors is the foundation for bioremediation application
to PAH/creosote-contaminated soils.

In an effort to enhance the biodegradation of PAHs in
the environment, we have recently  focused on several
environmental and toxicological factors influencing the
abi'ity  of Sphingomonas  (Pseudomonas)  paucimobilis
strain EPA505 to mineralize PAHs individually and in
complex mixtures (e.g., creosote). We believe that more
effective management of natural microbial community
activities, through control of these factors, may lead to
more efficient bioremediation of soil and water contami-
nated by PAHs. Additionally, these studies should help
inoculant microorganisms be employed more effectively
for site restoration.

Materials and Methods

Evaluation of Temperature and pH Effects

Biometer flasks (3) containing minimal  salts  medium,
radiolabeled fluoranthe..j or phenanthrene, and cells of
strain EPA505 were used to monitor UC02 evolution
over a range of pH and temperature. A mixture of unla-
beled (10 mg PAH) and 14C-labeled PAH (approximately
41,000 dpm) was added to 250-mL biometer flasks from
acetone stock solutions,  and the solvent was evapo-
rated. To each flask was added 50 mL of Bushnell-Haas,
                                                117

-------
 and the contents were sonicated. The pH of the medium
 was adjusted with HCI or NaOH. The buffering capacity
 of Bushnell-Haas was such that the pH was stabilized
 over the course of 2 days at the target pH. For tempera-
 ture studies, the medium was adjusted to pH 7.1,  and
 flasks were equilibrated at various temperatures for aoout
 an hour prior to inoculation. All flasks were maintained at
 a selected temperature over the course of the studies.

 To initiate studies, 1.0 mL of 2N NaOH was  added to
 each sidearm of the biometer flasks to trap 14CO2.  The
 inoculum was prepared  from a cell concentrate (48-hr
 growth on complex medium LB, harvested, washed,  and
 resuspended in 0.05 M phosphate buffer) and added to
 obtain an initial optical density of 0.5 at 600 nm (about
 3 to 5 x 108 cells/mL).  Flasks were run in duplicate,  and
 killed-cell controls also were used. Flasks were shaken
 at 120 rpm at 30°C in darkness for  up to 8 days. NaOH
 samples were collected  intermittently  and analyzed by
 liquid  scintillation the same day.

 Identification of Inhibitory Creosote
 Constituents

 BiometBr flasks again were used to monitor 14CO2 evolution
 from 14C-PAH in the presence of various concentrations of
 creosote  and  its acid-,  neutral-, and base-extractabie
fractions to study the effect of phenols, PAHs, and neu-
trally  extractabte heterocydes  (carbazote,  dibenzothic-
phene, dibenzofuran, arid thianaphtfiene) and other N-,
S-, and  0-containing  heterocydes,  respective^ (4,5).
Synthetic mixtures of each of these fractions were pre-
pared as defined in Table 1 to more accurately evaluate
the effect of  each of these mixtures (6). An "artificially
weathered" (heating the neutral fraction at 65°C ± 5°C for
24 hours), creosote-neutral fraction also was analyzed to
examine  the  effect  of low molecular-weight PAHs (i.e.,
those containing two fused rings). A killed-cell control was
run for each different substrate, and a positive mineraliza-
tion control (no creosote) was run with each set of incuba-
tions.

The incubation  medium  was prepared as  descrioed
above. Bushnell-Haas,  however, was supplemented
with 0.03-percent Triton X-100 to facilitate study of con-
stituents  at concentrations above their natural  water
solubilities. For consistency, Triton X-100 was added to
each  flask.  The appropriate  amount  of  creosote,  or
some fraction thereof, was added via  glass gas-tight
syringe.

Flasks were  shaken 120 rpm at 30°C in darkness for
10 days. NaOH samples  were  collected  daily and
analyzed by  liquid  scintillation the same  day.  At  the
 Table 1.  Composition* of Synthetic Mixture of Creosote Constituents" Used In Mineralization Inhibition Studies

 Neutral Fraction (PAHs)                    Acidic Fraction (Phenollcs)              Basic Fraction (Hetorocycllcs)
Naphthalene
2-Methylnaphthalene
1-Methy
-------
conclusion of these studies, flasks exhibiting inhibition
were cultured for the determinatior. of viable cells. The
remaining  contents of  each flask subsequently were
extracted and analyzed for the concentration of creosote
constituents by a gas chromatography/flame ionization
detector (GC-FID) (4,5).

Results and Discussion

Average (n=2) percent release  of  UCO2 from  14C-
fluoranthene by strain EPA505 was essentially identical
for pH values of 6, 7,8, and 9 (Figure 1). In these flasks,
postincubation pH was lowered by 0.5 to 1 pH unit The
pH-5 flasks quickty reached a plateau, after which min-
eralization  ceased. This plateau was not characteristic
of any of the other pH treatments. The postnoculation
pH of this flask was 4.6. Absence of extensive minerali-
zation in the pH-4 and pH-10 flasks correlated with the
absence of the characteristic color change (colored deg-
radation  intermediates)  normally  associated  with
fluoranthene mineralization by this bacterium (1,7,8).

Strain EPA505 was active  at  all temperature ranges
tested to date (Figure 2), although rates and extents of
mineralization decreased  with decreasing temperature.
At the 2b"G incubation  temperature, mineralization ex-
tent was reduced compared with 30°C and 37°C  but
might eventualry  reach  that seen  with the higher tem-
peratures given incubation times beyond 200 hours. At
18°C, mineralization rates appeared to be leveling off at
values below those seen at  higher temperatures, and it
does not appear that continued incubation beyond 200
hours will increase mineralization much further. We cur-
rently are  evaluating activity of this  strain at a wider
range of temperature and incubation times. The  effects
of pH and temperature on  the mineralization  of  '*C-
phenanthrene by strain  CRE-7, a low molecular-weight
PAH degrader, are currentty under study.

Of the creosote fractions assessed, the acid-extractable
(phenolic) and base-extractable (heterocyclic) fractions
were the most inhibitory to the activity of strain EPA505.
At 50 mg/L, the phenolics fraction slowed the onset of
mineralization; at  70 mg/L,  no mineralization was  ob-
served (Figure 3). The base-extractable fraction (mostty
heterocydes)  was  inhibitory at  35 mg/L (data  not
shown). Whole creosote was inhibitory at 200 mg/L The
neutrally extracted fraction and the weathered neutral
fractions were not inhibitory at any concentration tested
(210 mg/L).

The basis of this inhibition  is not known but could be the
result of direct toxicity  to the cells or isotope dilution
caused by the use of  more readily degradable sub-
strates, or could be an effect of decreased availability of
the radiolabeled substrate. Studies are currently  in pro-
gress using synthetic mixtures of all  fractions to deci-
pher  the inhibitory mechanism and  more accurately
identify   inhibitory  constituents  and  concentrations.
 I30
   10

    0

   -10
 pH 4
' pH 5
 pH 6
 PH 7
 pH 3
. pH 9
. pH 10
              SO      100      150

                Incubation Time (hrs)
                                       200
Figure 1.  Effect of madia pH on I4C-fluoranthen« mineralization
         by strain EPA50S.
   70


   80


   SO


   40 .


   X .


   20 .


   10 .


    0
 37 C
 MC
 25 C
 18C
              SO
                      100      150
                  Incubation Time (hr»)
                                       200
Figure 2.  EPoct of Incubation ta
-------
from studies using natural microbial communities that
have been enriched to degrade creosote.

Summary and Conclusions

If the isolated strains of bacteria under study represent
the potential  activities of bacteria in contaminated site
material, then environmental conditions may have to be
manipulated, in some cases, to provide optimal activity.
Where low temperature and pH extremes are encoun-
tered in the field, substantial effects on PAH mineraliza-
tion can be expected. In addition, if bioaugmentation is
considered as a biotreatment strategy, inoculants may
have to be carefully selected to be effective under these
suboptimai conditions.

These data further support implementation of creosote
bioremediation via a two-stage process (patent pending)
employing co-inoculation  (e.g.,  bacterial strain to de-
grade the toxic" phenolic and  heterocyclic fractions)
and  secondary biotreatment of  more recalcitrant con-
stituents (e.g., strain EPA505 to treat  high molecular-
weight PAHs) (9).

References
1. Mueller, J.G.,  P.J. Chapman, and P.M. Pritchard.
   1989.  Action of  a  fluoranthene-utilizing  bacterial
   community on polycydic aromatic hydrocarbon com-
   ponents  of  creosote.  Appl.  Environ.  Microbiol.
   55:3,085-3,090.

2. Mueller, J.G., S.E. Lantz, R.J. CoMn, D. Ross, D.P.
   Middaugh, and P.M. Pritchard. 1993. Strategy using
   bioreactors and specially selected microorganisms
   for bioremediation of  ground water contaminated
   with creosote and pentachlorophenol. Environ. Sci.
   Technol. 27:691-698.

3.  Mueller, J.G., S.M. Resnick, M.E. Shelton, and P.M.
   Pritchard.  1992. Effect of inoculation on the
   gradation of weathered Prudhoe Bay crude oil. J.
   Indust. Microbiol. 10:95-105.

4. Mueller, J.G., S.E. Lantz, B.O. Blattmann, and P.J.
   Chapman. 1991. Bench-scale evaluation of alterna-
   tive biological treatment processes for the remedia-
   tion of pentacMorophenol- and creosote-contaminated
   materials: Solid-phase bioremediation. Environ. Sci.
   Technol. 25:1,045-1,055.

5. Mueller, J.G., S.E. Lantz, B.O. Blattmann, and P.J.
   Chapman. 1991. Bench-scale evaluation of alterna-
   tive biological treatment processes for the remedia-
   tion of pentachlorophenol- and creosote-contaminated
   materials: Slurry-phase  bioremediation. Environ. Sci.
   Technol. 25:1,055-1,061.

6. Mueller,  J.G.,  P.J. Chapman,  and  P.H.  Pritchard.
   1989. Creosote-contaminated sites: Their potential
   for bioremediation. Environ. Sci. Technol.  23:1,197-
   1,201.

7. Lin,  J.-E., J.G.  Mueller,   S.E.  Lantz,   and  P.H.
   Pritchard. 1994.  Influencing mechanisms  of opera-
   tional factors on the degradation  of fluoranthene by
   Sphingomonas  paucimobilis strain  EPA505.  Bio-
   chem. Eng. Internal review.

8. Mueller, J.G., P.J.  Chapman, B.O. Blattmann, and
   P.H.  Pritchard. 1990. Isolation and characterization
   of a fluoranthene-utilizing  strain of Pseudomonas
   paucimobilis.  Appl. Environ. Microbiol.   56:1,079-
   1,086.

9. Mueller,  J.G., J.-E.  Lin,  S.E.  Lantz,  and  P.H.
   Pritchard. 1993.  Recent developments in cleanup
   technologies:  Implementing innovative bioremedia-
   tion technologies. Remediation  (summer issue), pp.
   369-381.
                                                 120

-------
  Molecular Genetic Approaches to the Study of the Biodegradatlon of Polycyclic
                                      Aromatic Chemicals
                                Richard W. Eaton and Peter J. Chapman
          U.S Environmental Protection Agency, Environmental Research Laboratory, Gulf Breeze, FL

                                         James 0. Nitterauer
   Technical Resources, Inc., Gulf Breeze, FL, and University of Arkansas for Medical Sciences, Little Rock, AK
 Petroleum, coal, and their derivatives are composed of
 a variety  of chemicals,  including  polycydic aromatic
 hydrocarbons (PAHs), heterocyclics, and alkyt-substi-
 tuted aromatics. As these chemicals increase in size
 and complexity, bacteria have more difficulty metabo-
 lizing them. In  addition,  their catabolic pathways  are
 lengthy and often  branched, making it more difficult to
 sftjdy them.
 The approach that we are taking to study the biodegra-
 dation of individual xenobiotic chemicals involves a va-
 riety of strategies;  foremost among these are molecular
 genetic techniques such as 1) cloning genes that  en-
 code enzymes that catalyze reactions of interest and 2)
 isolating transposon-induced mutants that lack enzymes
 of a metabolic  pathway. These approaches allow an
 individual erzyme-catalyzed reaction or set of reactions
 to be  studied in the absence of other reactions that
 complicate analysis. This approach obviously allows the
 simultaneous study of both the enzymes and the genes
 that confer on an organism its metabolic capabilities.
Naphthalene and benzothiophene are simple, fused-
ring compounds that can serve as  models for more
complex polycyclic aromatic chemicals (PACs) in biode-
gradation studies. The pathway  for the bacterial meta-
bolism of  naphthalene (Rgure  1) was characterized
(1,2)  using  recombinant  bacteria  containing genes
cloned from  the naphthalene catabolic plasmid NAH7
(Figure 2). Bacteria  carrying the plasmid, pRE657,
which contains a 10-kb fcofll-C/a) fragment on which
the genes nahA, nahB, and nanCare located, converted
naphthalene  (Figure 1,1) to a mixture of two chemicals,
2-nydroxychromene-2-carboxylate (HCCA, Figure 1, VI)
and frans-o-hydroxybenzylidenepyruvate (tHBPA, Fig-
ure 1, VII). The initial product, HCCA, and tHBPA spon-
taneously  isomerize in aqueous solution to form  an
equilibrium mixture of the two compounds, making their
identification   difficult  Separation   was   possible,
however, using column chromatography on Sephadex
G-25 with  water as solvent; this allowed the rigorous
identification of these compounds using 'H- and
13C-NMR spectroscopy and gas chromatography/mass
Flgur* 1. Pathway for th« bactwW tmt«boll«m of n«phthal«n« to «*Hcyl«t«.
                                                121

-------
llll 1 S£
IrT 1 II 1



01234
V3nii *fnir *r '

ICLJ
Rfl-TWI 	 	 IBamHI
UU/II
KpfMI 	
I I I I I 1
5 « 7 9 a 10
klGbaai palra
IIIIII
MMI
_JHb*j|ll
SU
IB^III
1 -J
11 12
ORE630
pRE657
pRE672
pRE701
pRE714
pRE718
       XI
 Flgura 2.  Ganatlc map of tha raglon of plawnld NAH7 that ancodaa tha matabdlsm of naphthalarta to srtlcylata.

 spectrometry  (GC/MS).   Subclones  pRE701   and
 pRE718 were obtained that encode the enzymes tHBPA
 hydratase-aldolase (Figure 1, E) and HCCA isomerase
 (Figure 1, D), respectively, and act on ihese intermedi-
 ates. These two intermediates, and the enzymes that
 degrade them,  are  characteristic of  pathways for the
 degradation of aromatic compounds  with two or  more
 rings. The genes that encode these enzymes (nahEand
 nahD) thus may have value as specific probes for envi-
 ronmental microorganisms that degrade PACs, servi, ,g
 as part of the justification for the recently completed
 sequencing of these genes (3).

 The sulfur-containing  heterocycle benzothiophene is
 transformed by isopropylbenzene-degrading bacteria to
 a  mixture of products. One of these strains,  Pseudo-
 monas putida RE204, and its Tn5-generated mutant
 derivatives (4) were  used to study these biotransforma-
 tions (5). Three products were formed from benzothio-
 phene  by the  isopropylbenzene-induced  wild-type
 strain RE204: 
-------
   frans-ohydroxybenzylidenepyruvate   hydratase-al-       lism of isopropylbenzene  in Pseudomonas putida
   dolase from the NAH7 plasmid. In preparaton.           RE204. J. Bacteriol. 168:123-131.
4.  Eaton, R.W.,  and K.N.  Timmis.  1986. Charac-    5.  Eaton, R.W., and J.D. Nitterauer. 1994. Siotrarsfcr-
   terization of a plasmid-specified pathway for catabo-       mation  of benzothiophene by isopropylbenzene-
                                                       degrading bacteria. J. Bacteriol. Submitted.
                                                123

-------
      Comparison of Sulfur and Nitrogen Heterocyclic Compound Transport in
                          Creosote-Contaminated Aquifer Material
                                   Ean Warren and E. Michael Godsy
                                U.S. Geological Survey, Menlo Park. CA
Commonly, ground-water solute transport model inputs
are generated from chemical and ground-water proper-
ties that are not comparable with those at the site of
contamination. Care must be taken when assuming that
chemicals with similar molecular structures or charac-
teristics possess equivalent transport properties. In ad-
dition, ground-water characteristics,  such as pH, must
be compared with ionization constants (pKt) to deter-
mine the influence of the sediments ion exchange ca-
pacity.  Simulated transport will not be accurate if the
parameter determined at one pH differs from that of the
ground water.

In this paper, we compare the values of partition coeffi-
cients and retardation factors for the sulfur and nitrogen
heterocyclic compounds benzothtophene, dibenzothio-
pnene, quinoline, 2(1H)-quinolinone, acridine, and car-
bazole on low organic carbon content, lew ion exchange
capacity aquifer material. Column breakthrough curves
(BTCs) were modeled using the local  equilibrium as-
sumption (LEA) for compounds with a log octand-water
partition coefficient (log KoJ of less than 2.5 and the
nonequilibrium assumption (NEA) for compounds with a
log KQW greater than 2.5.

Background

The column material is taken from sediment adjacent to
an abandoned wood-preserving plant within the crty lim-
its of Pensacola, Florida (1). The wood  preserving proc-
ess consisted of steam pressure treatment of pine poles
with creosote and/or pentachlorophenol  (PCP).  For
more than 80 years,  a large but unknown quantity of
waste water, consisting of extracted moisture from the
poles, cellular debris, creosote, PCP,  and diesel  fuel
from the treatment processes, was  discharged to un-
lined surface impoundments that were in direct hydraulic
contact with the  sand-and-gravel aquifer. The ground
water, at a pH of 5.9 and moving at approximately 1 m/d,
is continually dissolving the more soluble compounds
found in creosote, creating an extended contamination
plume. The aquifer material for the laboratory col-
umns consisted of a low organic carbon content
(0.024 percent organic  carbon),  low ion exchange
capacity (2 meo/ 100 g) claylike  sand from the ap-
proximate centroid of the plume (Table 1).

BTCs of the water-soluble heterocyclic compounds in
laboratory columns can be described by the convection-
dispersion equation using the  LEA as proposed by
Hashimoto et al. (2),
where 9 is the porosity (-), P& is the bulk density of the
aquifer material (g/rrr), Ka is the partition coefficient
(m3/g), C is the arueous concentration (g/m3),  t is the
time (d), D is the dispersion coefficient ( n2/d), x is the
distance (m), and v is the linear velocity (m/d).

Transport of hydrophobic chemicals commonly must be
modeled using the NEA as proposed by van Genuchten
and Wierenga (3), which accounts for a readily mobile


Table 1. Aquifer Material and Column Characteristic*

Median PartWe Diameter (m)                0.000375
Percent Organic Carbon (-)                  .024
Cation Exchange Capacity (meo/100 g)        1.6
Column
   Length (m)                          0.354
   Diameter (m)                         .025
   Porosity (-)                           .449
   Bulk Density (g/m3 * 10"*}               1.361
   Row Rate (m3/d x 10*)               140
                                                 124

-------
fraction and a stagnant or immobile fraction of water in
the aquifer matrix (subscripts m and im, respectively),
                    a*2
                         ar
                                                (2)
                                                (3)
where / is the fraction of sorption sites in  the mobile
region (-)  and a defines the transfer rate of the solute
between mobile and immobile water (d"1). As described
by van Genuchten (4), the variables, f and a, from
equations 2 and 3, can be related to two fitted, dimen-
stonless parameters, respectively: (J. the fraction of the
sites in the mobile region where sorptinn is instan-
taneous,  and co, the  ratio of hydrodynamic residence
time to characteristic time of sorption (5). The NEA
model is based on the assumption that convection and
dispersion govern transport in the mobile water, and that
diffusion controls the  transfer of contaminant  between
mobile ard immobile water.

Both models assume a linear isotherm. Retardation fac-
tors, R, wri-ch describe the movement of contaminants
relative to a conservative tracer, can be reiatod to parti-
tion coefficients, bulk densities, and porosity bv
                                                (4)
Parameters were fit to BTCs using nonlinear regression
anatysis  by the computer programs HASHPE  (6),  to
determine flfor LEA, and CFITIM (4), to determine R,
jj, and oo for NEA. The dispersion parameter for all mode)
simulations was determined from CaClj breakthrough.

Brusseau and Rao (7) suggest that, for values of 3 less
than approximately 10, the NEA should be used instead
of the LEA to account for the observed tailing. The
values of 3 for benzothiophene, dioen'.othiophene, car-
bazole, and acridin& (compounds witn log K^,, >2.5) are
well below 10 (Table 2), justifying  trio use of the NEA
model. The NEA model determined that the values of 3
were much greater than 10 for quinoline and 2(1H)-qui-
nolinone (compounds with log KO* <2.5). Thus, the LEA
model was used to determine breakthrough parameters
for these compounds.

Results and Discussion

Fitted  parameters and original coefficients tor ben-
zothiophene, dibenzothiophene, quinoline, 2(lH)-quinc~
linone, carbazole,  and acridine using the models  are
given in Table 2.  The chemical structures are shown in
Figure 1.  The retardation factors for benzothiophens,
quinoline," 2(1 (H)-quinolinone are quite similar to each
other. 2(1H)-Quinolinone, with a pK.of 5.29,  is approxi-
mately 20-percent ionized, and quinoline, with a pKs of
4.9, is approximately 9.1-percent ionized. Zachara et al.
(8) have shown that sorption of quinoline is dominated
by ion exchange  up to 2 pH units above its pKa. 2(1 H)-
Quinolinone, like quinoline, should be retained by both
ion  exchange and organic  sorption.  Benzothiophene,
however,  is nonionic  and  subject to organic sorption
alone.

The values of (3  for the sulfur heterocycles agree with
each other but are greater than those for the nitrogen
heterocydes,  suggesting a larger percentage of sites at
which instantaneous sorption for the sulfur heterocycles
occurs. The value of  co for the sulfur heterocycles is
much less than that for the nitrogen heterocycles, indi-
cating that the characteristic time of sorption contributes
more to the retardation of nitrogen heterocycles, and to
acridine transport in particular.

The retardation of acridine is much greater than that of
dibenzothiophene and carbazole, despite the fact that
all have two benzene rings fused to a sulfur or nitrogen
heterocydic ring (Figure 1 and Table 2) and have similar
tog «<„,; dibenzothiophene and carbazole, however, are
Tabto 3.  pK. (09 K^, Partition Cotffletonts, Retardation Factor*, and Nonaqulllbrium Assumption Parameter Values for
        a*nzotHoph«n«, Dlbanxothloph«fi«, Qulnotlrw, 2(1H>Oulno
-------
                     Quindine
2(1H)-QuinoUnon«
  Dber;oth>opfierie
                     Carbazoto
                                      Acridine
Figure 1.  Chemical structures of banzotMophww, dlb*nzottilo
         phww, quinollna, 2(1 H)-qulnolinon*, cartazofe, and
         •cridln*.
subject to organic sorption alone, whereas acridine is
subject to both organic sorption and ion exchange. The
pK. of acridine is 5.6 and of carbazole is -5.7 (Table 2).
Thus, at pH 5.9, the ionized-fraction of acridine is 0.33,
but carbazole is completely un-ionized. The degree of
affinity (the selectivity) of acridino to charged functional
groups on the aquifer material and the extent of icniza-
tion as well as the sediments cation-exchange  capacity
contributes to the retention capacity.  With an  acridine
concentration of 18 g/m3 (0.10 meq/L), the column ca-
pacity due to ion exchange is 160. The column  capacity
is bfsed on the assumption of total sorption of the
ionized fraction of acridine to the aquifer  material and
complete displacement of calcium ions.

Transport of organic chemicals in ground water must be
modeled using parameters similar to those at the site of
interest Assumptions about solute transport based on
chemical and physical properties of similar but not iden-
tical compounds, aquifer sediments, and ground water
are not always valid.  Field conditions, such as  pH, flow
velocity, and chemical properties  (such as  selectivity
and pK.), must be taken  into consideration  to effectively
model solute transport.
References

1. Godsy. E.M., D.F. Goerfitz, and D. Grbid-Galic. 1992.
   Methanogenic biodegradation  of creosote contami-
   nants in natural and simulated ground-water ecosys-
   tems. Ground Water 30(2):232-242.

2. Hashimoto, I., K.B. Deshpande, and H.C. Thomas.
   1964. Peclet numbers and retardation factors for ion
   exchange columns. Ind. Eng. Chem. Fundam. 3:213-
   218.

3. var  Genuchten, M.T., and P.J. Wierenga.  1976.
   Mass transfer studies in sorting  porous media. I.
   Analytical solution.  Soil  Sci.  Soc.  Amer.  Proc.
   40:473-480.

4. van Genuchten, M.T. 1981. Nonequilibrium transport
   parameters  from  miscible displacement  expen-
   ments. U.S. Department of Agriculture. U.S. Salinity
   Laboratory Research Report 119:88.

5. Brusseau, M.L, and M.E. Reid. 1991. Nonequilibrium
   sorption of organic  chemicals by low organic-carbon
   aquifer materials. Chemosphere 22(3-4):341-350.

6. Oravitz, J.L  1984.  Transport of trace organics with
   one-dimensional saturated flow: Mathematical, mod-
   eling and parameter sensitivity analysis. M.S.C.E.
   thesis. Michigan Technological University,  Depart-
   ment of Civil Engineering.

7. Brusseau, M.L, and P.S.C. Rao. 1989. The influence
   of  sorbate-organic  matter interactions on sorption
   nonequilibrium.  Chemosphere   18(9-10):1,691-
   1,706.

8. Zachara. J M., et al. 1986. Quinoline sorption to sub-
   surface materials:  Role of pH and retention of the
   organic cation. Environ. Sci. Technol. 20:620-627.
                                                  126

-------
 Modeling Steady-State Methanogenic Degradation of Phenols in Ground Water at
                                        Pensacola, Florida
                        Barbara A. Bekins, E. Michael Godsy, and Donald F. Goerlitz
                     Water Resources Division, U.S. Geological Survey, Menlo Park, CA
 The study site is an abandoned wood treatment facility
 in the extreme western end of the Florida Panhandle
 within the city of Pensacola. For about 80 years, creo-
 sote-derived contaminants and pentachlorophenol from
 unlined waste-disposal ponds entered the ground water
 in the underlying sand and gravel aquifer. Concentra-
 tions of phenol and 2-,  3-, and 4-methylphenol have
 been monitored at the study site for more than 12 years.
 The data indicate that a nonaqueous-phase source be-
 low the ponds provides  a constant input of disserved
 phenols that then are degraded within 200 m downgradi-
 ent Figure 1  is a generalized geologic section along a
 flow lire down the axis of the plume together with con-
 tours of total  phenolic compound concentration. The
 degradation process appears to be at steady state be-
 cause the concentration profile has not changed over
 the last 12 years. The aquifer consists of approximately
 90 m of poorty sorted fine to coarse grained deltaic sand
 deposits interrupted by discontinuous  silts and clays.
 Ground-water flow is generally horizontal and south-
 ward toward Pensacola Bay. Row velocities range from
 0.3m/dtol.2m/d(1).

 Model Description

 Godsy et  al.  (2)  determined methanogenic  utilization
 rates for four phenolic compounds in microcosms con-
 taining aquifer sediments. They fit the change in concen-
 tration wrth time and the associated microbial growth to
 the equations for Monod growth and substrate utiliza-
 tion. Their results, given in Table 1, were used in a model
 describing transport and degradation at the field site.

 The modeled profile is 6 m below the surface in the
 methanogenic part of the contaminated zone, below the
 depth at which recharge and floating hydrocarbon at the
 water table affect concentrations and  above the clay
 lenses. A one-dimensional model was used because the
 flow direction is primarily horizontal and perpendicular
to a wide  contaminant source. Acridine orange direct
counts (AODC) indicate that the bacteria population is
             100 m*t*f*
                             Vertical Exaggeration 10x
         0    300 fMt

            ["I Sand
                          Sandy Clay
                             Clay
Rgure 1.  Generalized geologic section along • flow line down
         the center of the plume. Contours of total phenols
         are shown In mg/i..

spatially uniform and low (5 x 103 to 7.6 x 107 AODC/g
dry weight of sediment) relative to subsurface enumera-
tions at other sites (3). The existence of a steady-state
degradation  profile of each substrate, together with a
low, uniform  bacteria density, indicates that the bacterial
population is exhibiting no net growth (4). Thus, the
bacteria concentration in the model is held constant in
time and uniform in space.

We assume  that the substrate  profile at a depth of 6 m
satisfies the  one-dimensional transport equation with a
Monod reaction term:
<*S   n—   3s _
   =      ~V
                              m B
                             Y 9
(1)
where S is the substrate concentration (mg/L); t is time
(d); x is distance downgradient from the first observation
well  (m); D is the dispersion coefficient (m2/d); v is
                                                 127

-------
 Tebta 1.  Kinetic Constant* from Microcosm Studies for Each of the Phenolic Compounds Tested (2)'
Compound
Phenol
2-Methyf phenol
3-Methylphenol
4-Methyt phenol
Growth Rata m, (d'1)
0.111 ± 0.005
.044 ± 0.001
.103 ± 0.078
.099 ± 0.110
Half Saturation K, (mg/L)
1.33 ± 0.07
.25 t 0.82
.55 ± 6.67
3.34 ± 11.1
Yield Vfmg/mg)
0.013
022
.026
.025
 •Yield values were obtained from protein determinations before and after sub'Jtrate utilization.
 average linear velocity (m/d);  u.m is maximum growth
 rate (d~');  Y is yield (mg bacteria per mg S); B is the
 concentration of the active degrading bacteria (mg/L);
 9 is  porosity;  and K, is the  half-saturation  constant
 (mg/L). This equation was solved using a computer code
 described  by Kindred and Celia (5), with boundary and
 initial conditions givon by:
S(0,fl = S0;|?
                       JT.2SO
'0;S(x,0) =
                                                 (2)
 where So is the contaminant concentration 6 m below
 the ground  surface at Site  3, the closest site to  the
 source.

 Model Results

 Two predicted steady-state substrate profiles, along with
 the measured  phenolic-compound concentrations at 6
 m below land surface at each sample site, ara shown in
 Rgure 2. The computed profiles are steady-state solu-
 tions to  a one-dimensional advective-dispersive equa-
 tion with a  biological reaction  term  (Equation 1). The
 upper curve predicts the field  profile that would result
 from the phenol degradation  rate that was measured in
 the lab, whereas the lower curve corresponds to the rate
 measured for  2-methylphend. These two rates wera
 used because they have the  smallest associated errors
 and bracket the rates for the other two  comDOi;..-
-------

0.10
i.
•3 0.08
Giowfli FUla
0 O
2 S
0.02
0
- - No Inhibition
~ ~ ^ - - - Wttfi Hiiaan* 'nhibiOon "•.
\ J
, '" \ 4
' .-'
\
V
t
, s D«cay RIM (div1)
\
               40      80     120
                 Dtatinc* from W*N SIM 3
                                     ISO
200
         Tb«or«tteal  growth rat* computed from th« ph»nol
         concentration, th« Monod growth •rpnMalan, irwi {*»
         growth p»r«m«t»r» m«a»ur*d In th« microcosm ».rm»-
         Ictlans. Th« two curve* are computed with »ru
         out th« «ff»ct of rMdarw Inhibition.
growth. Furthermore, in theory, the functional form of the
positive growth curve cannot be balanced by a constant
decay rate. When the toxicrty of phenol is accounted for
using a Haldane  (7) inhibition mode), the predicted
growth  is about 50 percent lower but still much higher
than the published .decay rate.

Summary and  Conclusions

We have created a model cf methanogenic degradation
of phenolic compounds for a sand and gravel aquifer at
Pensacola, Florida. The mode1 verifies that field disap-
pearance rates of four phenols match those determined
in batch microcosm studies performed by Gocisy et at.
(2). The degradation process appears to be at steady
state because a sustained influx of contaminants over
several decades  has been continuously disappearing
within 150 m downgradient of the source. Goerlitz ot al.
(8) concluded that sorption was insufficient to explain
the observed toss. The existence of a steady-state deg-
radation profile of  each substrate, together with a low
bacteria density in the aquifer, indicates that the bacte-
rial population is exhibiting  no net growth possibly be-
cause  of  the oligotrophic  nature  of the  bacteria
population indicated by  the low value  for K,. A low K,
causes growth and utilization to be approximately inde-
pendent of the phenolic-compound concentration  for
most of the concentration range. Thus, a roughly con-
stant bacteria growth rate should exist over mucn of the
contaminated area.  This growth could be balanced fcy
an unusually high decay or maintenance rate caused by
hostile conditions or predation. Alternatively,  the loss of
bacteria by transport downgradient is being investigated
with column studies.

References

1.  FrsnKS, B.J. 1988.  Hydrogeology and How of water
   in a sand and gravel aquifer contaminated by wood-
   pr^swving  compounds,  Pensacola,  Flonda.  U.S.
   Geological  Survey  Water-Resources  Investigations
   Rftpcrt 87-4260.  p. 72.

2.  Gcxjsy, E.M., D F. Goerlitz. and D. Grbic-Galic. 1992.
   Methanc\jenic degradation kinetics of phenolic com-
   ocundf '.i aquifer-denved microcosms. Biodegrada-
   tion 2'
          3. Godsy, =.M.. O.r. Goerlitz, and D. Grbic-Galic. 1992.
            Meihar""t6. >.c c'oHeqradation  of creosote contami-
            nants in natuici inn .simulated ground-water ecosys-
            »«r.;«. Or wnd \ >ater 30:232-242.

          4. &"Hir:», 6 A., £vi. Godsy, and D.M. Goerlitz. 1993.
            Mcao'iiS'j V.jaay-sti'e methanogenenic degradation
            of  pherK.s  !r  ;,,rcur..-! water. J.  Contam. Hydrol.
            l(ir>;ireci,  ..S., a-iu M.A.  Celia. 1989. Contaminant
            uansv>^^  an- b f..  O.fc, Troutman, E.M. Godsy, and B.J.
            Franks. '>«?,o.  ^igraticn of wood preserving chemi-
            cals in contamir,oted ground water in a sand aquifer
            at Pensacola, Fonda. Environ. Sci. Techno). 19:955-
            961.
                                                  129

-------
      Anaerobic Biodegradation of 5-Chlorovanillate as a Model Substrate for the
                             Bioremediation of Paper-Milling Waste
                                 B.R. Sharak Genttiner and B.O. Blattmann
                                       Avanti, Corp., Gulf Breeze, FL

                                              PH. Pritehard
           U.S. Environmental Protection Agency, Environmental Research Laboratory, Gulf Breeze, FL
 5-Chlorovanillate (5CV; 5-chk>ro-4-hydroxy-3-methoxy-
 benzoic acid) was selected as a model compound for
 studying the biodegradation of paper mill effluents. This
 compound contains the methoxy-, chloro-, and carboxyl
 side groups often present on aromatic chlorinated com-
 pounds released in paper mill effluents. The major path-
 way of 5CV  degradation previously was determined to
 be   stepwise  demethoxylation   to   5-chtoroproto-
 catechuate (5CP;  5-chloro-3,4-dihydroxybenzoic acid),
 decarboxylation to 3-chtorocatechol (3CC; 3-chtoro-12-
 dihydro-xybenzene),  and dechtorination to catechol,
 which was completely degraded (Figure 1). The current
 research further investigates  the  anaerobic bacterial
 species responsible  for  the individual transformation
 steps. Once obtained in pure culture, studies  can be
 performed investigating individual transformation steps
 with reduction in toxicity of paper mill waste.
 Selective  media containing guaiacol  (2-methoxyphe-
 nol).  protocatechuate (dihydroxybenzoic acid)   and
 catechol as the sole energy source were inoculated with
 the original 5CV culture. Transformation of target com-
 pounds in these  enrichment cultures was followed using
 high  performance  liquid chromatography analyses.
 Immediately  upon  completing  the transformation of
 interest the cultures were passed to fresh medium. The
 guaiacol,  protocatechuate, and catechol cultures were
 sequentialry transferred through their respective media
 several times,  followed by several refeedings  of the
 target compound to enrich for the  bacterial species of
 interest These enrichments then  were diluted  in the
 respective media to obtain bacterial cultures respon-
 sible for deriethoxylation (Figure 2), decarboxyiation
 (Figure 3), and catechol (Figure 4) degradation. The
 data indicate that the derrethoxylating and decarboxy-
 lating bacterial species were more numerous by three
 orders of magnitude than the catechol-degrading  bacte-
 rial species. The transforming and degrading activity in
 these cultures has been sustained for several months
 and through several transfers, indicating that the activity
 is stable—a condition necessary for bioremediation ap-
 plications. The demethoxylating and decarboxylating
 cultures continued to transform guaiacol and  proto-
 catechuate in the presence of fairly high concantrations
 of catechol. Demethoxylation  rates begin  to decline
 above 3 mM catechol (Figure 28), while decarboxyiation
 rates did not decline significentfy at 10 mM catechol
 (Figure 3B). Because paper mill waste  contains other
phenolic compounds, applied  bacterial cultures must
tolerate other  toxics while performing  the desired
          COOH
                               COOH
                                          CO,
                                                              cr
                                                    ci
     5-Chtorovanillte        S-Chloroprotocatechuic       3-Chkxocatecho!


Flgur* 1.  Pathway for th« eomptat* degradation of S-chlorovanllllc add.
                                                                                      C0
               Catechol
                                                 130

-------
3.0
2.0 •
      e	«  Gualacol
      O—O  CalectMl
                  *   /° • «
                                     —
0.0
          10    20    30   40    50    60

     •—•  Gualacol
     o—o.  caecnoi                __

               00
0.0
                  w    \  ^

                   \\X..
         0            25            50
                       rime (days)
 Rgure 2.  Enrichment for demethoxylstlng anaerobic bacterial
          specie* (A) and demetnoxylatlng activity In highest
          active (1(T) dilution of a demetnoxyiating (B) anMro-
          bie bacterial consortium.

 transformation. Photomicrographs  of these  cultures
 show apparently pure cultures.  Purity of these cultures
 currently is being confirmed.

 The initial dechlorination of 5CV was investigated using
 a 3-chtorobenzoate-dechlorinating  anaerobic  co-cul-
 ture, which  dechtorinated 5CV to  vanillate and then
 demethoxylated  vanillate to  protocatechuate. Proto-
 catechuate was not further metabolized. A sul'ate-re-
 ducing  bacterium was isolated from this co-culture and
 identified as a new bacterial species, Desulfomicrobium
 escambium (1). Initial investigations with the pure cul-
 ture of  0. escambium showed a decline in the concen-
 tration   of  3-chlorobenzoate   (3CB)   in   defined
 pyruvate/SCB medium, which  depended upon the pres-
 ence of pyruvate. Because reductive dechlorination has
 been shown to be very  specific for  halogen  position
 (2,3), and 3CB and 5CV are both mete-chlorinated, the
 basis for the decline  in  3CB by D.  escambium was
 further investigated.

 Further  studies indicated that  D.  escambium trans-
 formed  not only 3CB but 3-bromobenzoate (3BB) and
 benzoate as well  (Figure 5). Again,  the decline was
 dependent upon the presence of pyruvate. Lactate, for-
 mate, ethanol, and hydrogen,  which are used by D.
 escambium as electron donors for sulfate reduction, did
 not  support  the   transformation   of   these  three
compounds. The  similarity  in transformation  rates


1
1
(Q
C
8

o






f,
I
J
O




20
1.5


1.0


0.5 '

00 f
(
*A
1U

8
6
4 •

2
<
0<
a — * Protocatechuale A
O — O Calechol
^
-
/
-H
* o *• ~3 — °
B • •
A, C D EFG • / *
\A*c«»*»-^o •
y X y v y V "f o x
A A A /• * / /f n- .

3 10 20 30 40 50 60 70
• — 0 Protocatachuata 20"^
O — O Catechol ^o'" s
-?"°"C"
0°
oc?°00
/

>-•$••••• r • . «x • • «^
                                                          0     25    50    75    100    125   150
                                                                        Time (days)

                                                  Figure 1  Enrichment for decarborylating anaerobic bacterial
                                                           species (A) and decarboxylatJng activity In highest
                                                           (10  ) active dilution (B) of decarboxylatlng anaerobic
                                                           consortium.
                                                      500


                                                      400
                                                    |300
                                                      100


                                                        0


                                                      800
                                                    s«*>
                                                      400
                                                      200
25    50     75   100   125   150
                                                          0    25    50    75    100   125   150  175
                                                                         Time (days)

                                                  Figure 4.  Enrichment let catechot-degradlng anaerobic bacte-
                                                           rial species (A) and catechol-degradlng activity In
                                                           highest (10^ active dilution (B) of a catechol-degrad-
                                                           lng anaerobic consortium.
                                              131

-------
   600
                                             600
                                    30
Figure 5.  Reduction at 3C8 (•), 3BB (A), and banzoats (•) to
         3-ehlofO. (T).,3-oromo- (*), end banzyl alcohol (•)
         by d**v/fomVcro6/u/n Mca/nd/u/n strain ESC1. Sym-
         bols:  Opsn, 0.2-parcant pyruv«t»; dosad,  minus
         pyruvats.
between benzoate and the two halogenated benzoates
suggested that the transformation being observed was
not dehalogenation. After derivitization, gas chromatog-
raphy analysis revealed  the presence of two unknown
compounds in each culture. Further investigation using
gas chromatography/mass   spectrometry  (GC/MS)
analysis indicated that 3C3, 3BB,  and benzoate were
being reduced to their respective alcohols without deha-
logenation (Figure 5).
During GC/MS anarysis, the second unknown peak was
identified as  suca'nate.  Under anaerobic conditions,
succinate can result from the carboxylation of pyruvate.
A fdtowup study showed that benzoate was not reduced
in medium containing a gas phase of 100-percent nitro-
gen. The requirement for both  pyruvate  and carbon
dioxide  indicates  that the reduction of the  benzoate
compounds to their respective alcohols by D. escambium
is dependent upon carboxylation of pyruvate to succi-
nate.  If sulfate is added to the pyruvate/benzoate me-
dium, sulfate is reduced, benzoate does not decline, and
pyruvate is degraded to acetate and carbon dioxide.
Apparently, the reducing equivalents in this case are
diverted from the reduction of benzoate to the reduction
of sulfate, energetically a mora favorable reduction. If
reductive dechlorination competes similarly for reducing
equivalents, the presence of sulfate would be unfavor-
able for detoxification of paper mill waste.

Because D. escambium reduces but does not dechlori-
nate 3CB in pure culture, attempts  are currently under
way to isolate the second member of the 3CB-dechlori-
nating co-culture. This bacterial species may be respon-
sible  for dechlorinaticn  of  3CB  and  5CV by the
co-culture or  may provide a factor that  enables  D. es-
cambium to divert reducing equivalents to the dechlori-
nation of 3CB or 5CV.

References
1.  Sharak  Genthner,  B.R.,  G. Mundfrom,  and  R.
   Devereux. 1994. Characterization of  Desulfomicro-
   bium escambium sp.  nov.  and  proposal to assign
   Desulfovibrio desulfuricans strain Norway 4  to the
   genus Desulfomicrobium. Arch. Mic-obiol. (In  press)

2.  Boyd, S.A., and  D.R. Shelton.  1984.  Anaerooic
   biodegradation of chlorophenols  in fresh and accli-
   mated sludge. Appl. Environ. Microbiol. 46:50-54.

3.  Suflita, J.M., A. Horowitz,. D.R.  Shelton, and J.M.
   Tiedje. 1982. Dehalogenation: A novel pathway for
   the anaerobic  degradation of haloaromatic com-
   pounds. Science 218:1,115-1,117.
                                                 132

-------
     Characterization of a 4-Bromophenol Dehalogenating Enrichment Culture:
       Conversion of Pentachlorophenol to Phenol by Sediment Augmentation
                                           Xiaoming Zhang
                 National Research Council, National Academy of Sciences, Washington, DC

                                   W. Jack Jones and John E. Rogers
            U.S. Environmental Protection Agency, Environmental Research Laboratory, Athens, GA
 Pentachlorophenol (PCP), a carcinogen and ionophore
 (energy transfer inhibitor), is included on EPA's list of
 priority pollutants. Reductive dechlorination was found
 to be a significant reaction mechanism for the anaerobic
 degradation of PCP. The sequential removal of chlorines
 from PCP and its intermediate products may lead to less
 toxic products. In this abstract, we present data to demon-
 strate PCP transformation to phenol in sediment slurries
 inoculated with cells from a 4-bromophenol (4-BP) deha-
 logenating enrichment culture. We also describe  partial
 characterization of the 4-BP-dehalogenating enrichment


 Methods

 Sediment samples were collected from a freshwater pond
 in Cha^kee Trailer Part?, near Atfiens, Georgia Sediment
 slurries were adapted to degrade 3,4-dichlorophenol (3,4-
 DCP) by the sequential addition of 3,4-DCP '61 nM) to the
 slurries immediately following the dtsappeaiznce  of  the
 previous addition of 3,4-DCP (every 2 to 3 weeks). After
 12  months, the 3,4-DCP-adapted  sediment slurry was
 transferred (1:1 voWol) to a mineral medium containing 0.1
 percent yeast extract, 02 mM 3,4-DCP, and 50-percent
 (voWof) site water according to Zhang and Wiegel (1). The
 pH of the medium was adjusted to pH 72 to 7.3 with HO.
 Transfers were made when the 3,4-DCP was reductive^
 dechtorinated to at least 3-chtorophenol (3-CP). The 3,4-
 DCP dechlorination activity also could be maintained by
 substituting 4-BP for 3,4-DCP. The  4-BP (0.5 mM  to  0.8
 mM) maintained culture was used in subsequent experi-
 ments tu examine the dechlorination of PCP and its inter-
 mediate products.  These experiments  were performed
 using 1) 4-BP inoculated cultures in yeast extract-contain-
 ing mineral medium; 2) washed cell suspensions prepared
from cells grown with 4-BP in the mineral medium; and 3)
sediment slurries amended  with the 4-BP washed cell
suspension.
Results and Discussion

2,3-DCP, 2,4-DCP, or 3,4-DCP, added to mineral  me-
dium and inoculated (20 percent v/v log phase culture)
with cells grown on 4-BP, were dechlorinated to monc-
chlorophenols (MCPs).  Under the same conditions,
PCP (18.8 uM to 37.5 uM) was not dechlorinated. 4-BP
was dehalogenated to phenol in the control culture (plus
4-BP grown cells) supplemented with, only 4-BP but not
in the culture supplemented with both 4-BP and PCP,
indicating that PCP inhibited growth and/or activity of the
dehalogenating culture.

4-BP grown  cells that were harvested from a late log
culture, washed, and resuspended in phosphate buffer
to concentrate cells 40- to  100-fold,  exhibited dehalo-
genating activity in the presence of pyruvate.  All chlo-
rophenols tested  (19 congeners), except the three
MCPs, were dechlorinated at ortho, meta, or para posi-
tions in the presence of cnloramphenicol, which inhib-
ited any further production of dehalogenating enzymes.
As examples, 2,4-DCP was dechlorinated to 2-CP a. d
4-CP, and 3,4-DCP was dechlorinated to 3-CP, which
was not further transformed. These results are consis-
tent with a previous observation that all six dichlorophe-
nol isomers  were dechlorinated in 3,4-DCP-adapted
sediments (2).

Although PCP (300 uM) was preferentially dechlorinated
at the ortho position by the 4-BP grown cell suspension
(concentrated 40-fold), dechlorination of meta and para
chlorines also was observed.  2,3,4,5-, 2,3,4,6-,  and
2,3,5,6-tetrachlorophenol (TetCP) were identified as in-
termediate products  using a combination of high  per-
formance liquid chromatography,  gas chromatography,
and gas chromatography/mass spectrometry analyses.
Addition of either hydrogen, formate, or ethanol did not
stimulate  the dechlorination  activity. Heat-treated
                                                133

-------
 (10 min at 90°C) or solvent-permeated (toluene-treat-
 ment) cells lost dehalogenating activity. Sulfite, thiosul-
 fate,  and  sulfide  inhibited  the  ortfto  and  para
 dechlorination of 2,4-DCP. The addition  of sulfate or
 sodium chloride had no effect.
 In  a  4-BP grown  cell suspension assay prepared in
 99.9-percent  deuterium   oxide,   2,3,4-trichlorophenol
 (2,3,4-TCP) was transformed to OCRs and MCPs con-
 taining one and two deuterium atoms, respectively. This
 transformation verified the identity of the proton source
 (water) for the dechlorination of 2,3,4-TCP and its inter-
 mediates. This phenomenon also has been observed for
 trie reductive dechlorination of 2,5-dichlorobenzoate
 and 2,3,4,5,6-pentachlorobiphenyl (3,4).

 PCP  (28 |iM) was dechtorinated to phenol  (about 90-
 percent stoichiometric conversion) in 5 days in sterilized
 (autoclaved) and nonsterilized freshwater sediment slur-
 ries inoculated (equivalent to 8-percent inoculation) with
 a washed cell suspension prepared from a 4-BP deha-
 logenating  enrichment culture.  2,3,4,5-TetCP,  3,4,5-
 TCP, 3,5-OCP, and  3-CP were detected as transient
 intermediates (Figure 1).  In addition,  small peaks with
 retention times similar to  those found for 2,3,4,6-TetCP
 and 2,3,5,6-TetCP also were detected. In sterilized and
 in nonsterilized, noninoculated control  slurries, PCP was
 not transformed. The PCP transformation pathway iden-
 tified in this study was  somewhat different than  the
 pathway reported by Bryant et al.  (2)  for 3,4-DCf
 adapted  sediment slurries (or a  combination of 2,4-
 DCP- and 3,4-DCP-adapted sediments) prepared from
 the same site. 2,3,5,6-TetCP and 2,3,4,5-TetCP, either
 atone or together, have been detected as products of
 PCP  transformation in samples from other ecosystems
 (5).
 Specific experimental conditions were modified to iden-
 tify factors affecting PCP transformat!on in nonsterilized
 sediment slurries inoculated with the  4-BP enrichment
 culture. In these studies, the PCP transformation rate
 was dependent on the concentration of  added 4-BP
 grown cells, pH, and temperature. Addition of potential
 electron donors, including pyruvate, formate, and yeast
 extract did  not stimulate the transformation  of PCP,
 suggesting that the concentration of  electron donor in
 the sediment slurry was not a rate-limiting factor for PCP
 transformation. The presence  or absence of 4-BP
 (0.15 mM) in these experiments did not significantty
 affect PCP transformation. The rate of PCP transforma-
 tion in an estuarine sediment slurry amended with 4-BP
 grown cells was 25 percent of the rate observed in the
freshwater sediment slurry.

 In a previous study, Mikesell and Boyd (6) demonstrated
that by inoculating PCP-adapted  sewage sludge into
soil, PCP was dechlorinated to TCPs, DCPs, and MCPs
in 28  to 35 days. In  our study, PCP was converted to
phenol (90-percent recovery) within 5  days when a cell
Rgure 1.
              23       4       5
              Incubation Time (days)
   -0-  2,3,4,5,8-PCP
   -*-  2,3.4.5-TetCP
   -+-  3,4.5-TCP
   _*_  3,5-OCP
   _»_  3-CP
   _a_  Phenol

Dcchlorinatlon of PCP to phanol In • nonstwlle and
unadapMd a*dlm*nt slurry inoculated with calls har-
v*st»d from • 4-BP d«halog«natlng  •nrichmant
culture.
suspension of the 4-BP dehalogenating enrichment cul-
ture was added to freshwater sediment slurries. Taken
togetrwr, these results suggest that bioaugmantation
(and possibly induction) of microbial populations may
provide an alternative method of bioremediating PCP-
contaminated soils and sediments.


References

1. Zhang, X., and J. Wiegel. 1990. Isolation and partial
   characterization of a Clostridium spec, transforming
   para-hydroxybenzoate  and 3,4-dihydroxybenzoate
   and producing phenols as the final transformation
   products. Mterob.  Ecol. 20:103-121.

2. Bryant P.O., D.D.  Hale, and J.E. Rogers. 1991. Re-
   giospecific dechlorination  of pentachlorophenol by
   dichlorophenol-adapted microorganisms in freshwa-
   ter,  anaerobic sediment slurries. Appl. Environ. Mi-
   crobiol. 57:2,293-2,301.

3. Nies, L, and T.M. Vogel. 1991. Identification of the
   proton source for the microbial reductive dechlorina-
   tion of 2,3,4,5,6-oentachlorobiphenyl. Appl. Environ.
   MicrobiOl. 57:2,771-2,774.

4. Griffith, G.D., J.R. Cole, J.F. Quensen, III, qnd J.M.
   Tiedje. 1992. Specific deuteration of dichloroben-
   zoate during reductive dehalogenation by Desulfo-
   monile  tiedjei in  D2O. Appl. Environ. Wicrobiol.
   58:409-411.
                                                  134

-------
5.  Larsen, S., H.V. Hendriksen, and B.K. Ahring. 1991.    6. Mikesell, M.D., and S.A. Boyd. 1986. Complete re-
   Potential for thermophilic (50°C) anaerobic dechlori-       ductive  dechlorination and mineralization  of  pen-
   nation of pentachlorophenol in different ecosystems.       tachlorophenol by anaerobic microorganisms. Appl.
   Appl. Environ. Microbiol. 57:2,085-2,090.                Environ. Microbiol. 52:861-865.
                                                  135

-------
  Stimulating the Mlcrobial Dechlorination of PCBs:  Overcoming Limiting Factors
                        John F. Quensen, III, Stephen A. Boyd, and James M. Tiedje
                               Michigan State University, East Lansing, Ml

                                            John E. Rogers
            U.S. Environmental Protection Agency, Environmental Research Laboratory, Athens, GA
 The discovery that porychlorinated biphenyls (PCBs)
 can be reductivety dechtorinated by microorganisms un-
 der anaerobic conditions has stimulated interest in the
 development of a sequential anaerobic/aerobic biotreat-
 ment process for their destruction. While the aerobic
 degradation of PCBs is generally limited to congeners
 with four or fewer chlorines, the anaerobic process can
 dechlorinate more highly substituted congeners, pro-
 ducing products that  are  aerobically degradable.  In-
 deed, all products from the anaerobic dechlorination of
 Aroclor 1254 (1) have been shown to be aerobically
 degradable by one or more strains of aerobic bacteria
 (2). Also,  the  high  proportion  of monochlorinated
 biphenyls that can accumulate as a result of anaerobic
 PCB dechlorination may serve to induce PCB-degrading
 enzymes in aerobic microorganisms (3). More  highly
 chlorinated congeners can be aerobicaily co-metabo-
 lized but are not inducing substrates (4).

 A greater understanding of the factors controlling the
 anaerobic dechlorination of PCBs is necessary before a
 successful  sequential anaerobic/aerobic biotreatment
 process can be developed for PCBs. In particular, how
 to stimulate more rapid and complete PCB dechlorina-
 tion in areas where the natural rate and/or  extent of
 dechlorination is limited is important to determine. The
 general approach we have taken is to identify site-spe-
 cific factors limiting in situ PCB dechlorination, then to
 apply treatments to alleviate the limitation(s). During the
 first year of this project, our research focused on en-
 hancing the dechlorination of PCBs in soil and in sedi-
 ments from the River Raisin in Michigan.

 Drag Strip Soil Experiment

 Factors most likely limiting PCB dechlorination in soils
are a high redox potential, lack of organic carbon avail-
ability, and absence of PCB-dechlorinating microorgan-
isms. To determine how to alleviate limitations because
of these three factors, we conducted an experiment with
PCB-contaminated drag strip soil from Glens Falls, New
York. Alternate means tested for achieving low redox
conditions were to use a chemical reductant (Na?S) or
to provide carbon so that microbial activity would con-
sume al< oxygen present. The effectiveness of defined
and complex carbon sources were compared. Methanol
was chosen as a defined carbon source because it has
been shown to  enhance microbial dechlorination  of
PCBs (5). Trypticase soy broth (TSB) was used because
it is  a  complex carbon  source used for the general
culture of anaerobic microorganisms. Inocula consisted
of PCB-dechlorinating microorganisms eluted from up-
per Hudson River sediments.


Materials and Methods

The procedure followed was to first weigh 2 g of sieved
soil into each anaerobic culture tube. Depending on the
treatment, sterile liquid medium or inoculum (10 mL)
then was  added  while flushing with  Cyfree N2:C02
(80:20). Sterile (autoclaved) nonreduced media con-
sisted of 1)  minimal salts; 2)  minimal salts plus 0.1
percent methanol; or 3) minimal salts plus 0.1 percent
TSB. Sterile reduced media consisted of these same
three media but purged of oxygen with nitrogen and
amended with Na2S (0.24 g/L). All media were buffered
at pH 7. Inocula were prepared by eluting PCB-dechlori-
nating  microorganisms  from Hudson River  sediments
with each  of these six media. After adding  the proper
inoculum to each tube, the tubes were sealed with Tef-
lon-lined rubber stoppers and a.jminum crimps. Con-
trols  were autoclaved for 1 hour at  121 °C. Triplicate
samples were  analyzed every 4 weeks for  24 weeks.
The entire contents of a culture  tube were extracted for
each observation, and a congener-specific PCB analy-
sis was performed by capillary gas chromatography with
electron capture detection.
                                                 136

-------
To determine if the time required to achieve anaerobic
conditions was related to the lag time before dechlorina-
tion or to the subsequent extent of dechlorination, moni-
toring the redox  of the cultures was  necessary. The
redox indicator indigo disulfonate was added to parallel
treatments for this purpose, reduced to  a colorless form
at an Eh of -125 mV. The concentration of the oxidized
form  was monitored photometrically  during  the first
month of incubation.

Results

Dechlorination occurred only in inoculated treatments
that received a carbon supplement (methanol or TSB)
(Figure  1). The lag time was slightly less (8 weeks) in
the TSB/NazS treatment than in the other dechlorinating
treatments (12 weeks), possibly because reduced con-
ditions were maintained more effectively (Figure 2). By
the end of 24 weeks, about 0.69 and 0.62 meta plus para
chlorines (m & p Cl) per biphenyl had been removed in
the methanol and TSB treatments  without  reductant
(NaaS). The addition of Na^ and methanol gave more
extensive dechlorination (an average loss of 0.87 per-
cent m & p Cl after 24 weeks) than methanol alone, but
NajS did not stimulate further dechlorination with TSB.
Thus, both inoculation and a carbon supplement were
necessary to  initiate PCS dechlorination in this soil.
Apparently, indigenous microorganisms capable of PCS
dechlorination were not abundant enough to express
dechlorination activity within the 24 weeks that the ex-
periment lasted.

The extent of dechlorination achieved in the inoculated
treatments was not simply related to the rate at which
reducing conditions were achieved, as  indicated  by the
reduction of indigo disulfonate (Figure ?.). Whether NajS
was used, the inoculated methanol treatments took sig-
nificantly longer to reduce  all of the indigo disulfonate
than the TSB treatments did. Without NazS, the same
   3.0
   1.5 .
   1.0
Autodavad
Unameodad
Reductant
Mathanol
TSB
Raductant Methanel
Reductant TSB
               5       10      1S
                  Incubation Time (weeks)
                                       20
                                               23
Flgura 1.  Dachlorfnatlon of PC8s In drag strip sod expressed
         a* the decrease In the avaraga numbar of matt and
         par* chlorlna* ovar Umax.
                                               50


                                               40 -
                                               20 -
                                               10 •
     -10


      SO


      40
    r

    I  X


    |  20


    I  10


       0


     -10
          •-• Inoculated Minimal Medium
          •-* Inoculated MeOH Medium
          e-e Inoculated TSB Medium
          •-• Inoculated Autocalved Control
                                                            10
                                                                      20
                                                   •-• Inoculated' Na^S Mkiinai Medium
                                                   «-• Inoculated Na?S MeOH Medium
                                                   •-• InocUated N*jS TS8 Medium
                                                            10         20
                                                             Incubation Time (days)
Flgura 2.  Tha redox Indicator Indigo dlsulfonata was used to
         follow changaa In  radox during tna first month of
         Incubation of tna drag strip soti axparlmant Reduced
         condition* ara Indicated when the concentration of
         tna oxidized form (plotted) raachaa zero. A • without
         chamlcal  raductant  (NaiS). 8 •  with  chemical
         raductant

extent of dechlorination was  achieved with each carbon
source,  but  with Na2S   greater dechlorination  was
achieved with methanol.

River Raisin Sediment Experiment

We are conducting a similar experiment to determine the
minimal amount of manipulation necessary to  dechlori-
nate PCBs  in River Raisin  sediments collected near
Monroe,  Michigan. In a previous  research project, we
found that little in situ dechlorination of PCBs had oc-
curred in these sediments. PCB-dechlorinating microor-
ganisms, however, exist in the sediments, the sediments
support dechlorination in  laboratory assays,  and  the
PCBs are bioavailable because they were dechlorinated
under conditions of our treatability assay. In fact individ-
ual congeners in the contaminated sediment decreased
30 to 70 percent in 24 weeks at rates nearly identical to
rates for the  same congeners  freshly spiked into non-
contaminated sediments.  The  treatability assay  was
conducted using air-dried River Raisin sediments. They
were slurried with an equal weight of air-dried non-PCB-
contaminated sediments    and   reduced  anaerobic
                                                   137

-------
mineral medium (RAMM). The slurry then was inocu-
lated with microorganisms eluted from Hudson River
sediment and ^.SAtrichlorobiphenyl (2-34-CB)  in a
small volume of acetone was added to a concentration
of 500 ng/g sediment The noncontaminated sediment
was added to provide a source of undefined nutrients.
The medium included essential minerals and the chemi-
cal reductant (Na2S) to lower the initial redox potential.
Inoculation ensured that PCB-dechlorinating microor-
ganisms  were  present.  The 2-34-CB was added be-
cause the addition of a single PCS (or potybrominated
biphenyl) can somehow "prime"  the dechtorination of
RGBs already present in a contaminated sediment (6).

The question thus becomes: what aspects of our treat-
ability assay are  necessary to dechlorinate the RGBs
present in the River Raisin sediments? We are conduct-
ing separate experiments with wet and air-dried River
Raisin sediments to answer this question. With the air-
dried sediments, ihe factors being considered are
1) addition of 2-34-CB; 2) addition of the mineral  salts
in RAMM; 3) addition of Naj-S; and 4) addition of the
non-PCB-contaminated  sediments. All treatments with
the air-dried sediments  were inoculated with microor-
ganisms  eluted from Hudson River sediments. These
same four factors also are  being addressed in the ex-
periment  wrth wet (i.e., never air-dried) River Raisin
sediments. In this case, the necessity of inoculating with
Hudson  River  microorganisms also is being tested.
These experiments are still in progress, and data are not
yet available.
References

1. Quensen, J.F., III, S.A. Boyd, and J.M. Tiedje. 1990.
   Dechlorination of four commercial polychlonnated
   biphenyl  mixtures (Aroclors)  by anaerobic microor-
   ganisms  from  sediments. Appl.  Environ. Microbiol.
   56:2,360-2,369.

2. Bedard,  D.L,  R.E. Wagner, M.J. Brennan, M.L.
   Haberi, and J.F.  Brown, Jr. 1987. Extensive degra-
   dation of Aroclors and environmentally transformed
   polychlorinated biphenyls by  Alcaligenes eutrophus
   H850. Appl. Environ. Microbiol. 53:1,094-1,102.

3. Masse, R., F. Messier, L Peloquin, C. Ayotte, and M.
   Sylvestre. 1984.  Microbial biodegradation of 4-chlo-
   robiphenyl,  a model  compound  of  chlorinated
   biphenyls. Appl. Environ. Microbiol. 41:947-951.

4. Furukawa, K., F. Matsumura, and K.  Tonomura.
   1978. Alcaligenes and Acinetobacter strains capable
   of degrading potychlorinated  biphenyls.  Agric. Biol.
   Chem. 42:543-548.

5. Nies, L.  and  T.M.  Vogel. 1990. Effects of  organic
   substrates on dechlorination of Aroclor 1242 in an-
   aerobic   sediments.   Apol.   Environ.   Microbiol.
   565,612-2,617.

6. Bedard,  D.L,  H.M.  Van Dort,   R.J.  May, K.A.
   DeWeerd, J.M. Principe,  and LA. Smullen. 1992.
   Stimulation  of dechiorination of Aroclor  1260 in
   Woods Pond  sediment In: General Electric Com-
   pany research and development program  for the
   destruction of PCBs, 11th progress report Schenec-
   tady, NY:  General Electric Corporate Research and
   Development pp. 269-280.
                                                138

-------
         Potential Surfactant Effects on the Microbial Degradation of Organic
                                           Contaminants
                 Stephen A. Boyd, John F. Quensen, III, Mahmoud Mousa, and Jae Woo Park
                                Michigan State University, East Lansing, Ml

                                    Shaobai Sun and William Inskeep
                                 Montana State University, Bozeman, MT
 The biodegradatJon of poorly water soluble compounds
 in soil or sediment systems is believed to be limited by
 low  bioavailability because of strong  sorption of the
 compounds to natural organic matter (1-4). The use of
 surfactants to increase aqueous concentrations of these
 types of compounds, and therefore their bioavailability,
 often has been suggested as a way of overcoming this
 problem (5,6). Significant solubilization of the target
 compounds, however, usually occurs only above the
 critical micelle concentration (CMC) of the surfactant, a
 concentration often toxic or inhibitory to bacteria (7).

 Petroleum sulfonate oil (PSO) surfactants are different
 from conventional surfactants in that they form stable
 rrtcroemulsions in water rather than micelles, thereby
 enhancing  solubilization at  low concentrations without
 apparent toxic effects to bacteria (5,8). We recentty
 reported a 60-fold decrease in the apparent soil sorption
 coefficient (K*) of 2^',4,4'.5,5'-hexachlorobiphenyl at a
 PSO aqueous concentration of only  30 ppm, and a
 200-fokj decrease in K* at a  170 ppm PSO (4). We,
 therefore, propose to investigate the use of this class of
 surfactants  in  enhancing  the  anaerobic  microbial
 dechlorination of potychlorinated biphenyls (PCBs).

 Although conventional surfactants are ineffective at en-
 hancing  HCH  solubility  at concentrations  below the
 CMC, evidence exists for stimulatory effects on btode-
 gradation of aromatic hydrocarbons in soils even when
 surfactant-induced disassociation from soil was not sig-
 nificant i.e., at concentrations  below the CMC (9). For
 example, mineralization of phenanthrene was substan-
 tially enhanced in a muck soil in the presence of 10 ng
of nonionic surfactant per gram of soil (10 ppm). Similar
effects  on biphenyl mineralization were not observed,
and surfactant concentrations of 100 ppm were either
less stimulatory or inhibited mineralization.
A few reports indicate that sub-CMC concentrations of
surfactants may enhance anaerobic dechlorination of
aromatic  compounds.  Dechlorination  of pentachlo-
robenzene in  sediment  slurries was  stimulated  by
Tween 80 concentrations of 0.06 |ig/mL to 100 ug/mL
and SOS concentrations of 0.3 ug/mL to 40 ug/mL (10),
while Tween 80  at  concentrations below  the CMC
slightiy enhanced the dechlorination of hexachloroben-
zene (11). Triton X-705 at 600 ppm decreased the lag
time before PC'1   xihlorination  took place in Hudson
River sediment slurries but did not affect the subsequent
rate (12). C mcentrations of other  surfactants tested
(sodium dodecyl benzene sulfonate, Triton X-100, and
X-045) were all at or above their CMCs and  inhibited
dechlorination. Because these secondary stimulatory
effects can occur at surfactant concentrations below the
CMC, they do not appear to be related  to contaminant
solubility enhancement. We are attempting to establish
the stimulatory effects on PCB dechlorination of surfac-
tant concentrations below the  CMC for major types of
nonionic, anionic, and PSO surfactants (Table 1) and to
attribute these effects to either solubility enhancement
or secondary mechanisms. The physiological or physi-
cal nature of such secondary mechanisms  is being
investigated.

Results

The surfactants used in this study are listed in Table 1.
These include  several nonionic  surfactants that were
selected  to provide a range of CMC values,  and be-
cause previous  studies have shown that they provide
beneficial effects on  biodegradation  as  described
above. We also have included a twin-head anionic sur-
factant to minimize surfactant sorption to soils.

One of the major objectives of this research is to evalu-
ate the  effectiveness of sub-CMC  concentrations of
                                                 139

-------
 T«bt» 1.  Surfactants Proposed for This Study

 Surfactant                           CMC (mg/L)
 Triton X-100

 Triton X-405

 Triton X-705

 Tween 80

 Alforic 810-60

 C,9DPOS (Dowfax 8390) (14)

 Petroleum sulfonate oil
  130(7)

  620(7)

  625

   13(13)

  275 (9)

4,000

  MA
 NA * not applicable. These products form stable rrtcroemulsfons in
      water and do not exhibit a CMC. They consist of petroleum
      sulfonate  (61 to 63 percent) and mineral oil (33 percent).
 surfactants in increasing the rate and extent of PCS
 dechlorination. To determine what trie exact aqueous
 phase concentration in soil-  or sediment-water slurries
 is and whether'this concentraton is above or below the
 CMC, we need to measure surfactant sorption (i.e.,
 obtain sorption  isotherms) by the soils and sediments.
 To accomplish  this,  we will  use a batch equilibration
 technique, where the amount sorbed is determined from
 the difference between the initial (added) and final (after
 sorption) aqueous phase surfactant concentrations. The
 following three methods for measuring aqueous  phase
 surfactant concentrations have been evaluated:  1) ten-
 siometen 2) UV-absorption; and 3) total organic carbon.
 Sorption isotherms developed using Method 1 indicated
 higher surfactant uptake by  sediment then those ob-
 tained using  Methods 2  and 3.  We  suspect  that the
 presence of disserved or suspended organic matter from
 the sediment may be influencing the surface tension
 measurement, and hence we have elected not to use
 this method.  Methods 2 and 3  resulted in essentially
 identical sorption isotherms for Triton X-100 '->y Hudson
 River sediments. Method 3 is universally applicable to
 all the surfactants listed in Table 1, whereas Method 2
 is only applicable to surfactants with the appropriate UV
 absorption properties. Hence, Method 3 is currentjy be-
 ing used  to obtain sorption isotherms for all the surfac-
 tants listed in Table 1. This information will quantitate the
 aqueous  phase surfactant concentrations in our sedi-
 ment slurries and determine whether these are above or
 below the CMC.

 To separate solubility enhancement effects of surfac-
 tants  (which could increase  bioavailability and hence
 bkxtegradation rates)  from the secondary effects of sur-
 factants on biodegradation rates, we are evaluating the
 sorption of PCBs in sediment-water-surfactant systems
 above and below the CMC. We have now observed the
effect of Triton X-100 on the  sorption of 2,2',4,4',5,5'-
 PCB by soil by measuring the apparent sorption coeffi-
cient K* at different aqueous  surfactant concentrations
 (C*). At Co, values below 200 ppm (approximately the
 CMC of Triton X-100), K* values increased from - 500
 to 1,200 with increasing surfactant concentration. In this
 concentration  range, the added surfactant is strongly
 sorbed by  soil, and the soil-bound surfactant in turn
 enhances  PCS  sorption.  At higher Ct^s (above the
 CMC), K* decreases rapidly and substantially because
 of the formation of surfactant micelles in solution that
 effectively dissolve PCBs and raise the apparent aque-
 ous  phase  PCB concentration. These preliminary re-
 sults strongly suggest that the enhanced contaminant
 biodegradation rates observed previously at low (below
 the CMC) surfactant concentrations are not due to in-
 creased  bioavailability  associated with solubility en-
 hancement effects. Thus,  other  indirect or secondary
 effects may be responsible for the stimulating biodegra-
 dation rates at surfactant levels below the CMC. These
 mechanisms will be investigated in the future.

 References

 1.   Ogram, A.V., R.E. Jessup, L.T. Lou, and P.S.C. Rao.
     1985. Effects of sorption on biological degradation
     rates of 2,4-dichlorophenoxy acetic  acid in soil.
     Appl. Environ. Microbiol. 49:582-587.

 2.   Steen, W.C., D.F. Paris, and G.L Baughman. 1960.
     Effects  of sediment sorption on microbial degrada-
     tion of toxic substances. In:  R.A. Baker, ed. Con-
     taminants  and sediments, Vol. 1. Ann  Arbor, Ml:
     Ann Arbor Science, pp. 447-482.

 3.   Weissenfels, W.D., H.J. Klewer, and J. Langhoff.
     1992. Adsorption of polycyclic aromatic hydrocar-
     bons (PAHs) by soil  particles: Influence on biode-
     gradability and biotoxicity. Appl. Microbiol.  Technol.
     36:689-696.

 4.   Guerin, W.F., and  S.A. Boyd.  1992.  Differential
     bioavailability of soil sorbed naphthalene to two
     bacterial   species.   Appl.   Environ.   Microbiol.
    58:1,142-1,152.

 5.   Sun, S., and SA Boyd. 1993. Sorption of  nonionic
    organic contaminants in soil-water systems contain-
    ing petroleum  sulfonate-oil surfactants.  Environ.
    Sci. Technol. 27:1,340-1,346.

 6.  Laha, S., and R.G. Luthy. 1991. Inhibition of phen-
    anthrene mineralization by nonionic surfactants in
    soil-water systems. Environ. Sci. Technol. 25:1,920-
    1,930.

7.  Kile, D.A.,  and C.T.  Chiou. 1989. Water solubility
    enhancement of DDT  and  trichlorobenzene by
    some surfactants above and  below the critical
    micelle  concentration.   Environ.   Sci.   Technol.
    23:832-838.

8.  Kile, D.T..  C.T. Chiou, and R.S. Helburn. 1990. Ef-
    fects of some petroleum sulfonate oil surfactants on
                                                   140

-------
    the apparent water solubility of organic compounds.
    Environ. Sci. Technol. 24:205-208.
9.   Aronstein, B.N.,  Y.M. Calvillo, and M. Alexander.
    1991.  Effect of surfactants at low concentrations on
    the desorption and bioavailability of sorted aro-
    matic  compounds  in soil. Environ.  Sci. Technol.
    25:1,728-1,731.

10. Mousa, M.A., and J.E. Rogers. 1993.  Enhancement
    of pontachtorobenzene dechlorination by surfactant
    addition.  Abstract Q-155.  Presented at the 93rd
    General Meeting of the American Society for Micro-
    biology, Atlanta, GA.

11. Van Hoff, P.L, and C.T. Jafvert. 1991. Influence of
    nonionic surfactants on hexachlorobenzene degra-
    dation. Abstract 498. Presented at the 12th Annual
    Meeting of the Society of Environmental Chemistr/
    and Toxicology, Seattle, WA.
12.  Ambramowicz, D.A., M.J. Brennan, H.M. Van Dort.
    and E.L Gallagher.  1993. Factors influencing '.he
    rate of  polychlorinated  biphenyl dechlorination  in
    Hudson  River  sediments. Environ. Sci.  Techno!.
    27:1,125-1,131.
13.  Schick, M.J. 1966. Nonionic surfactants. New York,
    NY: Marcel Dekker.
14.  Rouse, J.D., and D.A. Sapatini. 1993.  Minimizing
    surfactant  losses using twin-head  anionic surfac-
    tants in subsurface remediation. Environ. Sci. Tech-
    nol. 27:2.072-2,078.
                                                  141

-------
     Enhanced Dechlorination of PCBs in Contaminated Sediments by Addition of
                     Single Congeners of Chloro- and Bromobiphenyls
                                    W. Jack Jones and John t£. Rogers
             U.S. Environmental Protection Agency, Environmental Research Laboratory, Athens, GA

                                          Rebecca L. Adams          ,
                                 Technology Applications, Inc., Athens, GA
 Bioremediation has been  suggested as  a  technology
 that may be useful in decreasing the level of pollutants
 at contaminated sites. For polychlorinated  biphenyl
 (PCS)  contaminated sediments, reductive dechlorina-
 tion reactions (anaerobic) preferentially transform the
 more highly chlorinated PCS congeners to less chlorin-
 ated derivatives, which are more amenable to aerobic
 degradation. In this instance, the  anaerobic and sub-
 sequent aerobic processes are complementary and re-
 sult in a reduction of toxic (higher chlorinated, coplanar)
 PCB congeners and possibly the biological destruction
 of PCBs through subsequent aerobic oxidations. Before
 using this method at a remediation site,  the ability of
 indigenous microorganisms from the site to transform
 the pollutants must be assessed, to understand factors
 that control the dechlorination reactions, and to develop
 techniques to enhance microbial activities.

 PCB transformation in anaerobic environments, such as
 sediments of lakes and rivers, could be inferred  in the
 mid-1960s from the studies of Brown and coworkers (1).
 These investigators noted that historically contaminated
 sediments from the Hudson River exhibited  an altered
 PCB congener profile compared with the congener pro-
 file of the origin&i contaminating Aroclor. The alterations
 were characterized by a reduction in the concentration
 of the  more highly chlorinated PCB congeners, with
 selective or preferential removal of meta and  para chlo-
 rines, and an increase in the concentration of the more
 lightly chlorinated  and ortfto-substituted congeners.
 Thus, dechlorination of the more highly chlorinated PCB
 congeners was proposed to be catalysed by  anaerobic
 microorganisms residing in the contaminated  sediment.
The  biologically mediated  reductive dechlorination of
PCBs from contaminated sediments was subsequently
demonstrated in several laboratory investigations  (2-4).
In  some studies, the microbial inoculum was obtained
 by "washing*  PCB-contaminated  sediments  with an-
 aerobic medium and collecting  the supernatant (4,5).
 Recently,  reductive dechlorination  of  PCBs was
 suggested to  be enhanced  when  PCB-contaminated
 sediments are amended with PCB mixtures (Aroclors)
 or specific  PCB/polybrominated  biphenyl  (PBB)
 congeners (6).

 To date, only a limited number of studies have attempted
 to  understand  the  factors that affect the reductive
 dechlorination of PCBs in historically contaminated sedi-
 ments. Abramowicz et al. (7) reported that addition of
 inorganic nutrients enhanced reductive dechlorination of
 endogenous PCBs in laboratory  incubations of Hudson
 River sediments. In a recent study using methanogenic
 sediment  slurries  contaminated with Aroclor  1260,
 Bedard and Van Dort  (2) reported  that addition  of
 bromobiphenyl  congeners  simulated  the  reductive
 dechlorination  of endogenous (historical) PCBs. In an
 earlier study, Bedard and coworkers (8) reported that
 amendment of Woods Pond sediment with a high con-
 centration (approximately 1 mM) of either 2,3',4',5-CB
 or 2,3,4,5,6-CB stimulated reductive dechlorination of
 endogenous PCBs and that  transformation of conge-
 ners with para chlorines was especially evident.

The primary objectives of this study were to determine
the reductive dechlorination potential cf PCB-contami-
nated sediments from the Sheboygan and Ashtabula
Rivers and to further test the hypothesis that addition of
PCB  and  PBB  congeners  enhances  the reductive
dechlorination of endogenous (historical) PCBs  by in-
digenous microbial populations.

Materials and Methods

PCB-contaminated sediments were collected from the
Sheboygan River, near  Sheboygan Falls,  Wisconsin,
                                                142

-------
and from the Ashtabula River,  near Ashtabula, Ohio.
Grab samples of sediments were collected from all sites.
Initial gas chromatography data indicated that a signifi-
cant shift in the PCB congener profile had occurred
since the time of PCB deposition, suggesting previous
reductive dechlorination activity.

Biotransformation experiments were prepared by com-
bining one volume of PCB-contaminated sediment with
one volume of anoxic  (N2 sparged) site water and the
mixture was stirred for approximately 5 min. Aliquots of
the sediment slurry (equivalent to 5 g dry sediment)
were dispensed into amber serum vials, and the sedi-
ment slurry was  amended with various chtorobiphenyl
or bromobiphenyl congeners dissolved in acetone. Initial
experiments were  conducted with PCB-contaminated
Sheboygan River sediment (approximately 180 ppm to-
tal PCBs) and Ashtabula River sediment (approximately
100 ppm total PCBs) and were amended with  penta-,
hexa-,  hepta-, or octa-chlorobiphenyl congeners. Addi-
tional experiments were performed with more  heavily
contaminated  Sheboygan River sediment  (approxi-
mately 1,000 ppm total PCBs) and were amended with
either a di- or tetra-cMorobiphenyl congener or the cor-
responding di- or tetra-bromobiphenyl conoener (final
concentration  of 1 mM). Attodaved controls and
nonautodaved, unamended controls also were included
                                             in the study. Triplicate samples were analyzed at 4- to
                                             8-week  intervals for congener-specific  PCBs using
                                             capillary gas  chromatography and electron  capture
                                             detection.'

                                             Results

                                             Enhanced Dechlorination Using Specific
                                             PCB Congeners

                                             Initial dechlorination experiments  were conducted with
                                             PCB-contaminated (-180 ppm) Sheboygan River sedi-
                                             ments    amended  with  20  ppm to  80  ppm  of
                                             2,2',3,3',4,5,S,6'-octachlorobiphertyl  (octa-CB).  The
                                             most prominent PCB homologues detected in the con-
                                             taminated Sheboygan River sediments were  tnchloro-
                                             biphenyls and tetrachlorobiphenyls. The percentages of
                                             octa-CB remaining in the samples after anaerobic incu-
                                             bation for 8 months were 35 percent, 20 percent, and
                                             10 percent, respectively, for sediments amended with
                                             20 ug/g, 40 ug/g, and 80 ^g/g. In all sediment experi-
                                             ments amended with octa-CB, there was a decrease in
                                             the concentration of hepta-, hexa-, penta-. tetra- and
                                             tri-CB congeners occurred as well as an increase in the
                                             concentration of di- and mono-CB congeners. The mole
                                             percentage of mono-CBs was less than 1 percent at the
                                             onset of the  experiment (Figure 1A,  Week  1); after
                                       Sheboygan River Sediment (Control)
         60.00
         40.00
         20.00
          0.00
                                                             g Week 1

                                                                Week 30
                 MONO    DI      TRI   TETRA  PENTA  HEXA  HEPTA   OCTA  NONA   DEC A
                                                Homotog Groups
                                   Sheboygan River Sediment •»• 20 PPM OCTA-CB
I
          60.00


          40.00


          20.00


           0.00
          MONO   DI
                                 TRI    TETRA  PENTA  HEXA  HEPTA  OCTA   NONA  DECA
                                                Homolog Groups
Rgur* 1. Profll« of anwndad «nd •ndootnou* PC3 WotranafoonaMon In (A) unam«id«d control s«dlm«nt» «nd (B) 20 ppm
        2^',W,4,5,6,ff-oct»chlorot>lph«ny1 (oct»-CB) «m«(xl»d ««tllm»nt»,
                                                143

-------
 anaerobic  incubation  for 30  weeks,  this homologue
 group accounted for approximately 8 percent of the total
 PCB congeners in sediments amended with 20 mg/g of
 octa-CB (Figure 1B). The major products of reductive
 dechlorination were di-CB congeners; this homologue
 group increased from  2.5 to 40 mole  percent after 30
 weeks of incubation. The most prominent di-CB peak
 detected in octa-CB amended sediments consisted of
 two ortfto-substitijted congeners (22'-CB and 2,6-CB).
 Two  additional homologue groups,  tri- and tetra-CBs,
 initially  accounted for  approximately 80  percent of  the
 total  PCB homotogues in the contaminated Sheboygan
 sediments but were reduced to less than 50 percent of
 the total  following 30 weeks' incubation in octa-CB
 amended experiments. The  ayerage  number of chlo-
 rines per biphenyl (total of endogenous  plus amended
 PCBs) decreased from 4.2 to 2.8 (±0.1  ), 2.5 (±0.3). and
 22. (±0.3), respectively, in experiments amended with
 20 u,g/g, 40 u.g/3, and  80 u.g/g of octa-CB.
 PCB-contaminated Ashtabula River sediments were
 spiked  with 2,3,3',4,4'-pentachlorobiphenyl (penta-
 CB). 2.3,3',4,4',5-hexachlorobiphenyl  (hexa-CB),  or
 2,2',3,4,5,6,6'-heptachlorobiphenyl (hepta-C3) or com-
 binations thereof and incubated anaerobically. Dechlori-
 nation of the added congeners was observed after  lag
 periods of 5, 4, and 3 months for experiments amended
 with  either the penta-CB, hepta-CB,  or hexa-CB,  re-
 spectively Addition of the chlorobiphenyl congeners sin-
 gly  or  as mixtures resulted in  enhanced  reductive
 dechlorination of endogenous PCB congeners in a man-
 ner similar to that observed for Sheboygan River sedi-
 ment amended with octa-CB. Appreciable decreases in
 the mote percentages  of endogenous  PCB homologue
 groups  (tetra-CB and penta-CB) were coupled with in-
 creases in the mole percentages of  mono-, di-, and
 tri-C8 congeners. The  average number of chlorines per
 biphenyl decreased from approximately 5.2 to  2.7 in
 Ashtabula  River sediments  amended  with any  of the
 three congeners tested. No significant changes in the
 distribution of the PCB  homologue groups were noted in
 control experiments.

 Dechlorination In the Presence of
 PBB/PCB Congeners

 Recently, experiments have been initiated to test the
 hypothesis that  amendment of PBB congeners en-
 hances  the dechlorination of PCSs in  contaminated
 sediments. Highly  contaminated (1,100 ppm  PCBs)
 sediments from the Sheboygan River were amended
 with dibromo-  or dichlorobiphenyl congeners, or with
 tetrachloro- or tetrabromobiphenyl congeners, and de-
 halogenation was followed over the course of 6 months
 incubation.  After 6 months of incubation, no enhance-
 ment  of  dechlorination  of endogenous  PCBc has been
detected  in   sediments  amended  with   2,2',4,5'-
tetrabromobiphenyl   or  2,2',4,5'-tetrachlorobiphenyl
compared with controls. Both meta and para debromi-
nation of the added 2,2',4,5'-PBB congener, however,
was evident after 1 month of incubation, with 2,2'-dibro-
mobiphenyl  observed as th-j major product. Approxi-
mately 25 percent of the parent 2,2',4,5'-PBB remained
after  6 months'  incubation.  Dehalogenation  of the
amended 2,2',4,5'-PCB congener was more rapid than
debromination of the corresponding PBB congener;
more  than 70 percent of the 2,2',4,5'-PCB was trans-
formed to 2,2',4-PCB after 1 month's incubation. As with
the added PBB congener, however, enhanced  dehalo-
genation of the endogenous PCBs was not evident.

In a separate set of experiments, 2,4-, 2,5-, or 2,6-dibro-
mobiphenyl or dichlorobiphenyl congeners were added
to PCB-contaminated  Sheboygan River  sediments.
Greater than 85 percent of the amended 2,4- and 2,5-
dibromobiphenyl were denominated at the para and
meta  positions, respectively, within the initial 3  months
of incubation. No  evidence  of debromination of the
amended 2,6-dibromobiphenyl was noted. Further, ad-
dition  of the dibromobiphenyl congeners has not yet had
an effect on the extent of dechlorination of the endo-
genous PCBs compared with controls. Of the dichloro-
biphenyls examined, significant loss (40 percent) of only
2,5-dichlorobipherryl has been observed. Dechlorination
at the meta chlorine was accompanied by an increase
in 2-chlorobiphenyl. Although results are only  prelimi-
nary,  a moderate reduction  in the average number of
meta plus para chlorines for endogenous PCBs appears
to be  in this data set

The results  from  the  present study demonstrate the
dechlorination capacity of PCB-contaminated Sheboy-
gan River and Ashtabula River sediments. No apprecia-
ble dechlorination of endogenous PCBs  was observed
in unamended sediment slurries. Several explanations
are proposed for the stimulation of reductive dechlorina-
tion of endogenous PCBs in sediments  by addition of
specific PCB congeners: 1) the bioavailability of PCBs
was enhanced, thus providing an available electron ac-
ceptor for oxidation reactions; 2) the growth of indige-
nous   PCB  dechlorinating   microorganisms  was
stimulated;  or 3)  amended  PCB  congeners induced
dechlorinating activity  of indigenous microbiai popula-
tions.  Additional  strategies should  be  considered, for
PCB  bioremediation and may include increasing the
physical-chemical availability of PCBs bound to  sedi-
ments (for example, the addition of surfactants) or cy-
cling between anaerobic and  aerobic conditions.


References

1. Brown, J.F., R.E. Wagner, H. Feng, D.L. Bedard, M.J.
   Brennan, J.C. Carnahan, and R.J. May. 1987.  Envi-
   ronmental dechlorination of PCBs. Environ. Toxicol.
  Chem. 6:579-593.
                                                 144

-------
2. Bedard, D.L. and H.M. Van Dort 1992. Brominated
   biphenyls can stimulate reductive dechlorination of
   endogenous Aroclor 1260 in methanogenic sediment
   slurries. Presented at the  92nd General Meeting of
   trie American Society for Microbiology, p. 339.
3. Quensen, J.F., III, J.M. Tiedje, and S.A. Boyd. 1988.
   Reductive   dechlorination    of    polychlorinated
   biphenyls by anaerobic microorganisms from sedi-
   ments. Science 242:752-754.

4. Quensen. J.F.. Ill, SA Boyd, and J.M. Tiedje. 1990.
   Dechlorination  of four commercial  polychlorinated
   biphenyl mixtures (Arodors) by anaerobic microor-
   ganisms from sediments,  Appl. Environ. Microbioi.
   565,360-2.369.

5. Mies,  L, and T.M.  Vogel.. 1990. Effects of organic
   substrates on dechlorination of Arodor 1242 in an-
   aerobic  sediments.   Appl.   Environ.  Microbioi.
   565,612-2,617.
6.  Van Dort, H.M., and D.L. Bedard. 1991. Reductive
   ortho- and mera-dechlorination of a polychlonnated
   biphenyl congener by anaerobic microorganisms.
   Appl. Environ. Microbioi. 57:1,576-1,578.
7.  Abramowicz, DA, M.J. Brennan, and H.M. Van Dort.
   1990.  Anaerobic and aerobic biodegradation  of en-
   dogenous PCBs. In: General Electric Company re-
   search and development program for the destruction
   of  PCBs, 9th  progress report.  Schenectady, NY:
   General Electric Corporate Research and Develop-
   ment  pp. 55-69.
8.  Bedard. D.L, S.C. Bunnell, and H.M. Van Dort. 1990.
   Anaerobic dechlorination of endogenous PCBs in
   Woods Pond sediment in:   General Electric Com-
   pany research and development program for  the
   destruction of PCBs, 9th progress report Schenec-
   tady, NY: General Electric Corporate Research and
   Development pp. 43-54.
                                                 145

-------
  Effect of Heavy Metal Availability and Toxicity on Anaerobic Transformations of
                                    Aromatic Hydrocarbons
                      John H. Pardue, Ronald D. DeLaune, and William H. Patrick, Jr.
               Wetland Biogeochemistry Institute, Louisiana State University, Baton Rouge, LA
 The existence of co-contaminants (e.g., heavy metals
 and toxic organics) in impacted sediments has created
 concern  over the potential for biodegradation to assist
 in remediating these sites. Heavy metals can be inhibi-
 tory to microorganisms and microbial processes, includ-
 ing  the  decomposition of organic matter and other
 biogeochemical processes (1). The  characteristics of
 this inhibition for biodegradation of toxic organics are
 poorly understood because of the large number of vari-
 ables involved. This study was initiated to determine the
 effect of heavy metals on reductive dechlorination of
 chlorinated aromatic organic*.  Exoeriments are being
 conducted with two metals, cadmium (a single valence
 [+2] transition metal) and chromium (a multivalence [+6
 and +3] transition metal), and two chlorinated aromatics,
 hexachlorobenzene (HCB) and 2,3,4-trichloroaniline
 (2,3,4-TCA). The reductive dechlorination of these com-
 pounds has been demonstrated, and the degradation
 pathways are generally understood (2,3).

 The interactions between metals and organic-degrading
 microbes or cor.sortia  are complex  because the ob-
 served effects are largely a function of the bioavailability
 of botr, the melais and the organic compound. Studies
 have been conducted on  aerobic biodegradation proc-
 esses (4,5), but inhibition of anaerobic biodegradation is
 not understood. At present, the best information indi-
 cates that the soluble fraction of the co-contaminants is
 the 'available" fraction to the microorganisms (6). Under
 anaerobic conditions, metals may be precipitated as sul-
 fides or present as reduced forms of lower toxicity. Solu-
 bility and specJatior of metals is strongly dependent on tK
 redox potential and pH of the sediment An excellent ex-
 ample is  the solubility of chromium, which exists in two
valence states with large Differences in solubility—Cr(Vl)
and Cr(lll)—depending on the redox potential of the sedi-
 ment (7). Other metals with single valence states (e.g.,
Cd2*, Zn2*) adsorb onto redox-sensitive surfaces (e.g., iron
and manganese oxides) and form various complexes un-
der different redox conditions.
Results and Discussion

Experiments are  being conducted to determine the ef-
fect of cadmium  on reductive dechlorination of 2,3,4-
TCA in previously uncontaminated anaerobic freshwater
sediment environments, including a rice paddy soil, a
cypress swamp soil, a bottomland hardwood soil, and a
freshwater marsh soil. These soils differ widely in sedi-
ment properties, including the organic matter concentra-
tion, which ranges from 2.9 percent in the rice paddy soil
to 74  percent in the freshwater marsh. 2,3,4-TCA is a
particularly  useful model compound because chlorine
substituents are present at ortho, meta, and para chlo-
rine positions. Representative results from several sNIs
are discussed here. Microcosms, with continuous moni-
toring of the Eh and pH, were constructed using sedi-
ment  slurries under anaerobic conditions.  Sediments
were  amended with 2,3,4-TCA  (200 mg/kg soil) and
varying concentrations of CO2* (control, 10 mg/kg soil,
100 mg/kg soil, and 1.JOO mg/kg  soil). Periodically, sub-
samples of microcosms were removed for quantification
of metals and 2,3,4-TCA.  Gas  chromatography/mass
spectrometry was used to identify lower chlorinated ani-
line metabolites.

Degradation of 2,3,4-TCA in rice paddy soil is presented
in Figure 1. Data are from  representative replicates.
When no Cd was added, dechlorination proceeded rap-
idly by removal of the ortno chlorine to form 3,4-dichlc-
roaniline (3,4-DCA). 3,4-DCA appeared only fransientfy
and was rapidly dechlorinated to  3-chloroaniline (3-CA).
No further dechlorination was observed. When 10 mg/kg
Cd was added, dechlorination also proceeded rapidly
but by the removal of the para chlorine to form 2,3-DCA.
Two monochloroanilines (2-CA  and 3-CA)  were sub-
sequently formed in nearly equal amounts. When cad-
mium was added at higher concentrations (100 mg/kg
and 1,000  mg/kg), no dechlorination  was observed.
Daily mass balance of chloroanilines for the microcosms
in Figure 1 averaged 103 percent ± 33 percent.
                                                 146

-------
                              O 2.3,4-TricMoroOTilr*
                              • 3,4-Otchtoroar.Um
                              7 3-Chtofoanclin*
                                • 3,4-Otchtoroenttne
                                w 3-Chtoroaniline
                                * 2-Critoioan*ne
                                            21
Figure 1.  Dtchlorlnatlon of 2^,4-trlchloroanlllne In • control
         (no cadmium added) and cadmium amended  (10
         mo/kg aell) mlcrocoem constructed from • rice paddy
         •oil (Crowley silt loam). Soluble cadmium was < 20
         ^^g/\. for tne control and 0.19 mg/L for the cadmium-
         amended microcosm. Soil Eh ranged from -200 to
         •230 mV.
This general trend also has been observed in the cy-
press swamp soil and freshwater marsh soil,  despite
wide differences in the degree of sorptjon of metals and
organics in these soils. Studies are ongoing in the fourth
soil (bottomland hardwood soil). The observed pattern
is ortho dechlorination when no cadmium is added, para
dechlorination  when  a  critical  level  of  cadmium  is
reached, and complete inhibition at another critical level
of cadmium. The trend is poorly predicted by  the total
concentration of cadmium but appears to be well pre-
cficted by 'soluble" cadmium (measured as porewater
cadmium passing through a 0.45-mm filter). Of the three
soils examined, ortho dechlorination occurred when sol-
uble cadmium concentrations ranged from less than 20
mg/L  to 32 mg/L Para dechlorination occurred when
soluble cadmium concentrations ranged from 0.15 mg/L
to 0.2 mg/L Complete inhibition occurred when soluble
cadmium concentrations ranged from  0.2 mg/L to 7.4
mg/L Further experimental replication may refine these
ranges more accurately. These results  are surprising in
light of differences in pore water chemical composition
between these flooded soils. MINTEQ, a  geochemical
speciation model, is being used to estimate concentra-
tions of cadmium complexes,  which may shed further
light on these results.
Preliminary batch studies also have been performed to
determine the effect of Cr(VI) on 2,3,4-TCA dechlorina-
tion in the bottomland hardwood soil. Results indicate
that Cr(VI) additions affect the dechlorination of 2,3,4-
TCA by increasing the lag time necessary for degrada-
tion to occur (Figure 2). Addition of Cr(VI) at 20 M,
50 M, 75 M, and 175 M all  increased the lag time for
dechlorination from  approximately  2  to 10  weeks.
Following the lag time, apparent rates of dechlorination
of 2,3,4-TCA were unaffected by the initial  chromium
addition.

Biogeochemistry of chromium in the bottomland hard-
wood soil has been previously investigated (7). Addition
of Cr(VI) under low Eh conditions is followed by rapid
(1 min) reduction  to Cr(lll), followed  by  precipita-
tion/sorption of Cr(lll) from the soil solution. A critical Eh
for the reduction process has been identified, +300 mV,
below which the reaction proceeds rapidly. In the batch
study (Eh »-200 mV), Cr(VI) was undetectable in solu-
tion (detection limit 5 ppb) immediately following addi-
tion, and only low concentrations of Cr(lll) (50 ppb) were
detected. Methanogenesis, as indicated by the accumu-
lation of CH4 in the vial headspace, was unaffected by
additions of Cr(VI). The mechanism by which chromium
inhibits dechlorination is unclear, although results sug-
gest an initial toxic effect  on the degrading population
that requires time  to overcome (lengthening lag time).
This effect could be direct (mortality of some microbial
population) or indirect (oxidation of some key reductant
crucial to dechlorination).
                      8        12
                         Weeka
                                       16
                                               20
Figure 2.  Effect of Cr(VI)  on dectilorlnatlon of 2,3,4-trl-
         chloroanlllne In flooded bottomland hardwood soil.
         Po4nt» are mean* of triplicate determination*. Coef-
         ficient of vt latJon are < 20 percent with the exception
         of measurements at 2 weeks and 5 weeks for control
         sample* (37 percent and 97 percent, respectively).
                                                   147

-------
References
1. Capone, D.G., D.D. Reese, and R.P. Kiene.  1983.
   Effect of metals on methanogenesis, sulfate reduc-
   tion, carbon dioxide evolution, and microbial biomass
   in anoxic salt marsh sediments. Appl. Environ. Mi-
   crobiol. 45:1,586-1,591.
2. Kuhn, E.P., and J.M. Suflita. 1989. Sequential reduc-
   tive  dehalogenation   of chloroanilines  from  a
   metnanogenic aquifer. Environ. Sci. Technol. 23:848-
   852.
3. Fathepure. BZ., J.M. Tiedje, and S.A. Boyd.  1988.
   Reductive dechlorination of hexachlorobenzene to
   tri- and dichlorobenzenas  in  anaerobic sewage
   sludge. Appl. Environ. Microbiol. 54:327-330.
4. Said, W.A., and D.L Lewis. 1991. Quantitative as-
   sessment of the effects of metais on microbial deg-
   radation   of  organic chemicals.  Appl.  Environ.
   Microbiol. 57:1,498-1,503.

5. Springael, D., L Diels, L Hooyberghs, S. Kreps, and
   M.  Mergeay.   1993. Construction  and  charac-
   terization of heavy metaJ-resistant haloaromatic-de-
   grading Alcaligenes eutrophis strains. Appl. Environ.
   Microbiol. 59:334-339.

6. Duxbury, T. 1985. Ecological aspects of heavy metal
   responses in microorganisms. Adv. Microbiol. Ecol.
   8:185-235.

7. Masscheleyn, P.H., J.H. Pardue, R.D. DeLaune, and
   W.H. Patrick, Jr. 1992. Chromium redox chemistry in
   a lower Mississippi valley bottomland hardwood wet-
   land. Environ. Sci. Technol. 26:1,217-1,226.
                                                 148

-------
         Biodegradatlon of Petroleum Hydrocarbons in Wetlands Microcosms
                                   Rochelle Araujo and Marirosa Molina
            U.S. Environmental Protection Agency, Environmental Research Laboratory, Athens, GA

                                             Dave Bachoon
                        Department of Microbiology, University of Georgia, Athens, GA

                                          Lawrence D. LaPlante
                                 Technology Applications, Inc., Athens, GA
 In the aftermath of several major environmental oil spills,
 it became apparent that spill preparedness did not in-
 dude an up-to-date inventory of bioremediation strate-
 gies or adequate methods for assessing the efficacy of
 bioremecJiatkjn  under field conditicns. Thus, field trials
 of bioremediation (1) preceded rigorous laboratory-  and
 pilot-scale experimentation. A cooperative effort to de-
 velop protocols for evaluating bioremediation strategies
 has led to the adoption of a system of tiered assays for
 determining efficacy and environmental toxicrty of prod-
 ucts that might be applied to spilled oil. Protocols include
 1)  analytical methods for determining the extent of
 biodegradation; 2)  toxicrty  assays for aquatic and sedi-
 ment organisms; 3) flask experiments to determine po-
 tential  for  biodegradation;  4) and laboratory-scale
 microcosms for assessing  the potential for degradation
 in  prototype  environments,  including  open  water,
 beaches, and wetlands.

 Development and Testing of Microcosm
 Protocols

 Oil extraction, refining, and transshipment facilities often
 are located in coastal regions, putting wetlands ecosys-
 tems at risk for exposure to  spilled oil. The inaccessibility
 of sites and the fragile nature  of the ecosystems pre-
 clude mechanical cleanup of oil, making bioremediation
 a preferred option for wetlands. Moreover, the high level
 of indigenous microbial activity suggests a potential for
 biodegradation,  especially if environmental nutrient limi-
 tations can be relieved by fertilizer  additions.

 Results

Sediment microcosms were constructed from homoge-
nized  marsh  sediments from Sapelo Island,  Georgia,
and were flushed on a tidal basis with seawater adjusted
to the salinity of the collection site (20%o). The tidal cycle
was continued until a dear boundary distinguished the
aerobic and anaerobic layers  (3 mm to 5 mm) of the
microcosm. Then.'oil (521 fraction of Alaska North Slope
crude, 0.5 mm depth/3.93 mL) was applied to  the sedi-
ment surface. The numbers of hydrocarbon-degrading
bacteria in the sediment prior to  construction of the
microcosms was in the range 103 cells/g to 104 cells/g,
which  is consistent with nonpristine coastal areas (2).
Products to be tested were applied 1 day after the  appli-
cation of oil. The types of products submitted for testing in
protocol development induded microbial cultures,  nutri-
ents, surfactants, sorbents, and combinations thereof.

Figure 1 shows the composition of the oil, as deter-
mined by gas chromatography/mass spectrometry, af-
ter a 6-week incubation. The alkane  constituents  of
the oil (Figure 1A) were appreciably degraded  in all
treatments relative to the original composition of the
oil. The degradation in the nutrient treatment  (Product
D) was slightiy greater than in the nonfertilized con-
trol. The addition of nutrients  plus microbial inoculum
(Product J) resulted in significant degradation of the
full range of alkanes (C13 to C35); that degradation
was primarily  biological is indicated by  the  reduced
ratios  of C17:pristane and Cl8:phytane. Neverthe-
less, pristane and phytane were reduced in  concen-
tration, indicating that they also are  subject   to
biodegradation,  although at a slower rate than the
normal alkanes. Thus, oil constituents that are  more
resistant to biological degradation than  are  prist  -<*
and phytane are more suitable  for use as inten.al
indices in longer  incubation  experiments;  both ho-
panes (3) and C2-chrysenes have been proposed for
this application.
                                                 149

-------
                                                                         a PRODUCT j
                                                                         D PRODUCT D
                                                                         D CONTROL
                     iHJlll   1   ijMJHJIHhjji'.;
                                      5U5"3533S55953335
s s " 2 *
  G u u « • • ^
      u u u u
          B
                                                    9 PRODUCT J
                                                    CD PRODUCT D
                                                    D
 Rgur* 1.
 Degradation of the aromatic constituents of oil was neg-
 ligible; only the naphthalene series differed in concen-
 tration between treated microcosms and controls after 6
 weeks' incubation. The lack of degradation of aromatics
 n the continued presence of aiVane constituents sug-
 gests that degradation of the two classes of compounds
 may ^ sequential, although Foght et al. (4) concluded
 that degradation of aliphatics and aromatics could occur
concurrently if adapted organisms are present. To test
whether alkane degradation goes to completion before
the onset of degradation  of aromatics, the length of the
microcosm incubation period  in  subsequent experi-
ments was increased from 6 weeks to 3 months.
                            Factors Influencing the Persistence ofPAHs
                            In Sediments

                            In light of the relative degradability of the alkane con-
                            stituents of petroleum and the toxicity and carcinogenic-
                            ity associated with the more recalcitrant  polycyclic
                            aromatic  hydrocarbons (PAHs), the effectiveness of a
                            remediation effort in reducing ecological risk depends
                            largely on the degree to which the latter are degraded.
                            Moreover, PAHs of industrial origin are of environmental
                            concern as soil and sediment contaminants in their own
                            right. Thus, the persistence of PAHs in the microcosms
                            can be considered a shortcoming of bioremediation
                            measures.
                                             150

-------
 Several explanations have been proposed to explain the
 persistence of PAHs in the environment. Intrinsic con-
 trols on the rate of degradation include low solubility,
 toxicity,  and  interactions  between  PAH  compound
 classes; extrinsic controls include environmental factors
 such as salinity,  temperature, nutrient  concentrations,
 and interactions  betwaen PAHs and other classes  of
 compounds, including  natural organic  matter. Interac-
 tions between PAHs and other compounds  may include
 co-metabolism,  the competitive utilization of alterna-
 tive substrates, or the absence of  required inducer
 compounds.

 Bauer and Capone (5) noted that preexposure of marine
 sediments to single PAHs enhanced subsequent degra-
 dation of those compounds and that  cross acclimation
 occurred  between select  PAHs. Similarly,  Kelley and
 Cemiglia  (6) reported  an  interaction  between fluoran-
 thene and pyrene and concluded that the catabolism of
 fluoranthene, pyrene, and  phenanthrene was catalyzed
 by a common enzyme system. Other  researchers (7)
 observed that a  mixed microbiaJ community was re-
 quired for the complete utilization of some  PAHs.

 Results

 We tested the interactions between  PAHs of different
 size classes to determine  if interactions between PAHs
 were responsible for  the persistence  of  those  com-
 pounds in sediments.  The presence of other PAHs,
 either grouped by size classes or as a mixture of 16
 compounds, did not affect the mineralization of pyrene
 by an acclimated microbial culture introduced into sedi-
 ment slurries with inorganic nutrients (Figure 2A). The
 same culture degraded pyrene more slowty when four-,
 five- and six-ring PAHs were present in mineral medium
 enriched with sediment organic extract (Figure 2B), and
 did not degrade pyrene at all when  five-  and six-ring
 PAHs were present in mineral medium (Figure 2C). We
 concluded that large PAHs are inhibitory to the activity
 of organisms capable of degrading pyrene, but that the
 inhibition is removed when the high  molecular-weight
 compounds are sorted to sediments or complexed with
 organic matter. Tbxicrty because of large PAHs, there-
 fore, probably did not explain the persistence of PAHs
 in the microcosm trials.

 Sediments that were inoculated with a culture that had
 not been recently exposed to PAHs adapted to degrade
 pyrene after a lag of 1 day,  unless protein synthesis was
 inhibited with chloramphenicol (Figure 3). When the cul-
 ture was preexposed to pyrene, the addition of chloram-
 phenicol did not appreciably  inhibit degradation upon
 subsequent exposure.  Similarly, the  antibiotic did not
 inhibit  degradation of pyrene by a culture  preexposed
to phenanthrene, although protein synthesis was neces-
sary for pyrene degradation by cultures oreexposed to
naphthalene.  Therefore, we concluded that the  cells
      70
      30

   I  2°
   *  10

      0

      80

   ?  ^
   £  80
   I  50

   1  *
   I  30

   I  m
      10
      0
                          10
                       Time (days)
                                   15
                                            20
Figure 2.  Mineralization of pyrens (8 jig/mL) by an enrichment
         culture In the absence of otfier PAHs (•) «nd In the
         praience of two- 2nd three-ring PAHs (Q), four-ring
         PAHs (»), flv«- end six-ring PAHs («), and a mixture
         of 14 PAHs (A) In sediment slurries amended with
         organic nutrients (A), minimal medium containing or-
         ganic aedlment extract (B), and minimal medium (C).
         The enrichment was previously acdlmated In sedi-
         ment slurries to a mixture of 16 PAHs. No minerali-
         zation occurred In sterile controls.
shared a common enzyme system  for phenanthrene
and pyrene, and another for naphthalene.

Ongoing Research

Current research  includes  the  isolation and  charac-
terization of a Mycobacterium sp. capable of degrading
pyrene as  a  sole carbon source. The  isolate will be
introduced  into  the mixed miurobial  community of the
sediment microcosm to assess survival  and impact on
the degradation of PAHs. The microbial  diversity in im-
pacted and nonimpacted sediments will be assessed by
whole genome hybridization, and specific probes will be
used to compare the activities of oil degraders and
lignocellulose degraders under various nutrient and sur-
factant treatments.

References

1.  Pritchard. P.H., and C.F. Costa. 1991. Environ  Sci
   Technol. 25:372-379.
                                                  151

-------
                        Tlm«(d«y»)
                    Pyr Atttpttd 
-------
  Blodegradatlon of Petroleum Hydrocarbons in Wetlands: Constraints on Natural
                                  and Engineered Remediation
                         John H. Pardue, Andrew Jackson, and Ronald D. DeLaune
                Wetland Biogeochemistry Institute, Louisiana State University, Baton Rouge, LA
 Sensitive wetland ecosystems are susceptible to impact
 from spilled and discharged oils. Major oil recovery and
 processing operations are located  in wetland ecosys-
 tems, including Louisiana, where 40 percent of the U.S.
 coastal wetlands and 15 percent of U.S. crude oil pro-
 duction are located. Understanding the responses of
 these wetland ecosystems to oil-related  impacts is criti-
 cal for the design of remediation strategies. Bioremedia-
 tion' is particularly  attractive because   mechanical
 cleaning or washing operations are usually impossible
 because of the sensitivity of these systems. At present,
 however, little information is available on the constraints
 of bioremediating spilled oils in wetland  ecosystems.

 Coastal marshes are wetland  ecosystems in  the Gulf
 Coast  region where oil production  and transshipment
 are concentrated. Marsh soils  differ from typical bottom
 sediments in fundamental ways that will  affect bioreme-
 diation in these systems: 1) highly organic  marsh soils
 store large amounts of nutrients but very little in readily
 available forms; 2) marshes are heavily  vegetated with
 macrophytes that can serve as conduits for O2 diffusion,
 dramatically increasing aerobic surface  area in the rfii-
 zosphere of marsh soils; and 3)  marshes are charac-
 terized by periods of flooding and drying, which expose
 a larger volume of porous soil to the atmosphere. Be-
 cause  these  features of marshes  and  other wetland
 types are  unique,  this study was recentiy initiated to
 determine the constraints on natural and engineered oil
 biodegradation in wetlands. The project is a cooperative
 agreement with the EPA Environmental Research Labo-
 ratory in Athens, Georgia.

 Background

 Biodegradation of oil components in wetlands has been
demonstrated (1) but rates of degradation  are strongly
dependent  on environmental conditions. These condi-
tions include temperature, salinity, Eh, pH, sorption, and
the oxygen and nutrient status of the environment Stud-
ies have documented changes in  microbial populations
in wetlands in response  to spilled oils  (2,3).  These
responses  were  generally increases in total microbial
populations and  increases in the ratio of oil degraders
to total heterotrophs.

In general,  wetlands are dominated by anaerobic proc-
esses: methanogenesis in freshwater wetlands and sul-
fate reduction in  brackish  and saline wetlands. Several
novel microbial processes have been identified that de-
grade oil components under anaerobic conditions (4).
Aerobic processes, however, are recognized to act on a
broader spectrum of compounds and are more rapid and
complete (e.g.,   mineralization  to  COj  and  H^O).  In
marshes, aerobic heterotrophic activity is concentrated at
the sediment-water interface in a small (several millime-
ters) aerobic layer and around the rhizosphera of rooted
marsh macrophytes. High sediment oxygen demand, cre-
ated by a sequence of events leading from organic matter
diagenesis, prevents further G>2 penetration.

The maintenance of this aerobic layer  is critical to mi-
crobial degradation of petroleum hydrocarbons. In oil-
impacted wetlands, petroleum components provide an
additional overwhelming carbon source and potentially
serve as a physical barrier for O2 diffusion. Some of this
limitation may be overcome by passive diffusion of 02
through marsh plants, although the relative supply and
demand of this process has not been calculated.  Flood-
ing/drying cycles, either tidal or seasonal, also will con-
trol Oj supply to marsh  soils.  In addition to oxygen
limitation, essential nutrients such as nitrogen may be-
come limiting because of  disruption of  natural biogeo-
chemteal cycles and competition from highly productive
macrophytes. Availability of nutrients such as nitrogen
depends on microbial mineralization  processes that
convert  nutrients to usable forms, which are rapidly
assimilated by plants and microorganisms. This "tight"
internal cycling is characteristic of marshes, where ex-
ternally supplied  nutrients are only a fraction of those
required  for observed plant (and  microbial) growth.
                                                 153

-------
 Fertilization may be required to maximize a microbial
 response to oil.

 Preliminary Results
 Study sites that have been selected  in the Barataria
 Basin, Louisiana, include a freshwater marsh and a salt
 marsh located along a salinity gradient extending toward
 the coast. Seasonal samples are being taken from these
 sites, and numerous nutrient, microbial, and geochemi-
 cal analyses are being conducted relating to bioreme-
 diation potential. For example, samples taken during
 January/February  1994 were evaluated  for  aerobic
 btodegradation potential of two oil components, phenan-
 threne and hexadecane, using radiorespirometry. Sur-
 face  marsh  samples were removed from the marsh
 using thin-walled aluminum cores, homogenized, and
 dispensed in center-well respirometry  vials.  Slurries
 were amended with the labeled hydrocarbons in  an oil
 matrix (-1  percent to 2 percent South Louisiana "sweet"
 crude, v/v), and 14CO2 was  quantified using liquid scin-
 tillation.  Treatments included controls, killed controls,
 and fertilization (with nitrogen, phosphorus, and iron).
 Results indicate that fertilization can increase the extent
 of mineralization of hexadecane and phenanthrene. Fer-
 tilization approximately doubled the extent of hexade-
 cane mineralization in both  the salt and  fresh marshes
 (Figure 1).  Fertilization effects on phenanthrene were
 significant in the salt marsh but within the experimental
 error in the fresh marsh (Figure 2). Nutrient availability
 in the winter months are generally highest because of
 the lack of competition from growing plants; therefore,
 fertilization  may have more dramatic effects in  other
 seasons. Most probable numbers of oil-degrading mi-
 croorganisms in the  fresh  marsh (103) were  several
 orders of magnitude higher than in the salt marsh  (10'),
 which may explain observed higher rates of  phenan-
 threne mineralization.  Results will be contrasted with
 seasonal data taken over the next year.

 Current work also is being conducted on other aspects
 of oil degradation in wetlands. The application of stable
 isotope techniques is being investigated as a method of
 measuring oil biodegradation in marshes.  Marsh soils
 have characteristic 513C signatures because of the pres-
 ence of nearty monospecific stands of plants that use
 either the C-3 or C-4 pathway (Table 1). Respired C02
 reflects the carbon signature of the  marsh soil. Crude
 oils also  have stable, characteristic 613C signatures (6)
 that have been used to detect biodegradation in the
 subsurface (5).  Measuring the 813C signature  of 002
 emitted from oiled marshes  is boing  investigated  as an
 indicator of the extent of oil mineralization in these wet-
 lands. This measure may serve as a noninvasive tech-
 nique  for  determining  oil biodegradation  in   spill
 situations. Additional studies are being conducted on oil
degradation using  core  and controlled  Eh-pH micro-
cosms. Variables being investigated include tidal and
         038  9  12  15  18 21  24  27  30  33
         03   6  9  12  IS  18  21  24  27  30  33
Figure 1.  Mineralization of  C-bexadecane (In an oil matrix) In
         fertilized and unfertilized salt marsh and fresh marsh
         •oil* In coaatal Louisiana (aoil samples  takan In
         February 1994).
!  io

r
§  40
I
V  20

I   o
              °  Fertilized
              •  Unfertilized
                                    SM Marstl
   I  80
         0  3   6   9   12  15  18 21  24 27 30 33
                           Days
    60


    40


    20


    0 4*
               0 Fertilized
               * Unfertilized
                                      Fresh Marsh
   0.    03   6   9  12  15  18  21  24  27  30 33
                             Days

Figure 2.  Mineralization of uC-phenanthrene 0" an oil matrix)
         In fertilized  and unfertilized salt marsh and  fresh
         marsh soils In coastal Louisiana (soil samples taken
         In February 1994).
                                                    154

-------
Tabte 1.  S13C («/„) of Marah Soil* of Louisiana Coastal
        Region and of Patrotoum Products (6,7)
Source
                                       513C
Fresh Marsh (Panicum hemitomon)
Intermediate Marsh (Sagittate falcate)
Brackish Marsh (Sparffna paters)
Salt Marsh (Sparing altamrftora)
Crude Oil
-27.9

-26.6

-14.9

-16.5

-30.6
flooding regime, fertilization, vegetation density, and soil
oxygen demand.  Gas  chromatography/mass  spec-
trometry analysis of crudes is being used to quantify 50
to 60 oil components, including alkanes, polycyctic aro-
matic hydrocarbons, naphthenes, and isoprenoids.

References
1. Hambrick, G.A., III, R.D. DeLaune, and W.H. Patrick,
   Jr. 1980. Effect of estuarine pH and oxidation-reduc-
   tion potential on microbial hydrocarbon degradation.
   Appl. Environ. Microbiol. 40:365-369.
2. Hood, MA, W.S. Bishop, Jr., F.W. Bishop, S.P. Mey-
   ers,  and T. Whelam. 1975.  Microbial indicators of
                oil-rich  salt marsh  sediments.
                30:982-987.
                                   Appl.  Microbiol.
3.  Kator, H., and R. Herwig. 1977. Microbial responses
   after two experimental oil spills in an eastern coastal
   plain ecosystem, fn: Proceedings of the  1979 Oil
   Spill Conference. API Publ. No. 4284. Washington,
   DC:  American Petroleum Institute, pp. 517-522.

4.  Milhelcic, J.R., and R.G. Luthy-1988. Microbial deg-
   radation of acenaphthene and naphthalene  under
   denitrification conditions in soil-water systems. Appl.
   Environ. Microbiol. 54:1,188-1,198.
5.  Aggarwal, P.K., and R.E Hinchee. 1991. Monitoring
   ttie in situ biodegradation of hydrocarbons by using
   stable  carbon   isotopes. Environ.  Sci.  Technol.
   25:1,178-1,180.
6.  DeLaune,  R.D. 1986. The use of 13C signature of
   C-3 and C-4 plants in determining past depositional
   environments in rapidly  accreting marshes  of the
   Mississippi River deltaic plain,  Louisiana. Chem.
   Geol. 59:315-320.
7.  Kennicutt, M.C.,  II. 1988. The effect of biodegrada-
   tion on crude oil bulk and molecular composition. Oil
   Chem. Poll. 4:89-112.
                                                   155

-------
                    Anaerobic Biotransformation of Munitions Wastes
                                 Deborah J. Roberts and Farrukh Ahmad
            Department of Civil and Environmental Engineering, University of Houston, Houston, TX

                                Don L Crawford and Ronald L. Crawford
             Center for Hazardous Waste Remediation Research, University of Idaho, Moscow, ID
 An environmental problem associated with U.S. military
 facilities is the presence of soil, sediment surface water,
 and  ground water  contaminated  with toxic explosive
 compounds. With the current emphasis on demilitariza-
 tion and returning land to the private sector, the reme-
 diation of the contaminants from these sites has become
 important Several types of remediation procedures are
 under  investigation for the removal of munitions from
 soils and water. Incineration has been demonstrated to
 be an effective process for the remediation of soils from
 these sites. The physical  process of wet air oxidation of
 munitions  contaminants is under investigation, as well
 as several biological remediation procedures. Kaplan
 (1) reviews the literature concerning the biological deg-
 radation of munitions compounds and shows that under
 aerobic conditions  the compound 2,4,6-trinitrotoluene
 (TNT)  is degraded  by a  reductive process and is not
 mineralized but merely transformed, producing dinrtro-
 toluenes and azoxy compounds as the products of me-
 tabolism. This degradation suggests that a process
 that is reductive in  nature (i.e.,  anaerobic) would be
 the best approach  to the treatment of  soils contami-
 nated with TNT. Under anaerobic conditions,  reduc-
 tive processes would occur at a faster rate, so lower
 amounts of the hydroxylamino intermediates would be
 produced and thus lower amounts of the azoxy dimers
 and polymers.

 Current studies show that an aerobic treatment might be
 a possibility, using Phanerocheate chrysosporiuifi (2-4).
 Boopathy °il al. (5, 6) have published findings concern-
 ing the anaerobic degradation of TNT by a sulfate reduc-
 ing bacteria. Many  investigations are currently under
way concerning the biological degradation of TNT, but
the procedure outlined below  is the first pilot-scale ap-
plication of a biological technology for munitions degra-
dation that has been demonstrated.
Background

A procedure for the anaerobic remediation of munitions
compounds  including TNT,  hexahydrc~1,3,5-trinitro-
1,3,5-triazine   (RDX),  and  1,3,5,7-tetranitro-1,3.5,7-
tetraazocine (HMX) from contaminated soil has been
developed (7-10) and is being demonstrated at Weldon
Springs, Missouri. The procedure, first developed and
demonstrated for the removal of DinoseD from soils,
invoivas flooding the soil with water and adding a carbon
source with a high oxygen requirement (such as starch)
(11-13).  Aerobic heterotrophs deplete the oxygen from
the aqueous phase while utilizing the starch. The aque-
ous/soil  mixture  then will be anaerobic,  allowing  the
degradation of TNT, RDX, and HMX to occur. The pro-
cedure requires that the pH be controlled to between 6.5
and 7 and that the temperature be in the mesophilic
range (8).

The pathway for TNT reduction as seen under anaerobic
conditions is  initially  a reductive one,  where  first 4-
amino-2,6-dinitrotoluene (4A), then 2,4-diaminotoluene
(24DA),  and finally 2,4,6-triaminotoluene (TAT) are pro-
duced. TAT very  rarely accumulates in the cultures but
is rapidly converted to 2,4,6-trihydroxytoluene  (methyl-
phloroglucinol, MPG) by some unknown mechanism.
This conversion  is followed by dehydroxylation reac-
tions, leading ultimately to p-cresol, which can undergo
ring cleavage either anaerobically or aerobically (14,15).
Although the latter compounds do not accumulate in the
soil during regular treatment procedures, they have
been detected in laboratory  cultures degrading TNT
when yeast extract was added as a nutrient supplement
for cultures enriched from soil.

A proposed improvement to the anaerobic remediation
strategy is to  implement an aerobic stage  after  the
reductive  stage  of the procedure is complete. This
                                                156

-------
 improvement would ensure mineralization of the carbon
 to COj rattier than a fermentation to several short chain
 fatty acids. This Improvement requires that the addition
 of starch at the beginning of the procedure be reex-
 ammed, as excess starch always occurs when the treat-
 ment is complete; thus, oxygenating the system is very
 hard (7). To do this, the use of external carbon sources
 that were more defined and thus easier to control than
 starch were investigated. The use of a commercial sol-
 uble starch,  glucose, and acetate was  compared with
 the insoluble starch supplied by J.R. Simplot Co. (Boise,
 Idaho).

 Laboratory experiments were conducted  to determine
 the soil  loading rates  for the treatment of a soil from
 Umatilla, Oregon, contaminated with 12,000 mg TNT/kg
 soil. 3.000 mg RDX/kg soil, and  300 mg  HMX/kg soil.
 These rates led to experiments designed  to determine
 the effect of tite reduced intermediates on the reduction
 of TNT and on the metabolism of the intermediates.

 AJI experiments were performed using a 1 percent (w/v)
 addition of  a  soil  that had  been  contaminated with
 Dinoseb and treated using the anaerobic procedure as
 an inoculum. Experiments to determine the effects of
 carbon source additions were performed  using  4 per-
 cent (w/v) Umatilla soil in phosphate buffer. Experiments
 to determine the effects of 4A on metabolism were per-
 formed in cultures spiked with TNT and 4A at the levels
 indica*9d in Figure 3.  Analyses were performed using
 narrow-bore high performance liquid chromatography,
 as described by Ahmad (16).

 Results

 The results  of the  experiments with various  carbon
 sources led us to glucose as the carbon source of choice
 (Fgure 1). Acetate was not used as a carbon source for
 oxygen depletion in these  cultures. The reason is un-
 known, but the contaminants in the soil possibly either
 inhibited some reaction in the TCA cycle or the glyoxy-
 late shunt the two  main pathway* for the utilization of
 acetate. Commercially available soluble starch did not
 serve as a carbon source  either, probably because of
 the absence of starch-degrading  organisms in the soil
 inoculum. The  insoluble starch was used  as  a carbon
 source for oxygen depletion  in these experiments, as
 had been demonstrated previously (8,11). This starch
 contains its  own microbiaJ component (11),  thus  the
 presence of starch-degrading organisrs in the soil was
 unnecessary. Cultures fed  glucose reduced the redox
 potential to  the lowest values and showed the fastest
 initial degradation of TNT.
 When  the amount of soil used in the treatment proce-
 dure was increased from 1 percent (w/v)  to 4 percent
 (w/v), the first intermediate (4A) accumulated to an ex-
tent that had not been seen before (50 mg/L) (Figure 2).
This accumulation w,is accompanied by a reduction in
     400
          a. F»dOK potential with giucow nioiut>»« starch. ~t
            vXuOK narcn u •xtwnu cartxxi tourco
     300

    e 200
      100 <

   I   o
     -200 .
     -300
                                         ;* Starcl
                                     in»ol' Di« Starch
                   10     15    X    25     30
          b. R*Joi DOtafTttaf wrth ac«tat» or glucDM as
           •xr»maJ carton SOUTCM.
                   tO    15    20
                        Tlm«(day»)
                                    2S
                                          30
          c. TNT concvnmaflon with ^UCOM. m»o*ut*# Katrcfi, ex
           •dub** iiaich a* •*t»>ma* carbon SOUTCM.
                            „ Sdutt* Stare*
              \.             • imMutM Starcfi
              y^x           • oiuaMt

              C^^WltHf
                      •H-niniiiiniM-H-
                        15
                             JO
           . TNT conccntranon wittt acatat* or JIUCOM la
                il carton aourcM.
r;
c
t W
1 70
1 *°
0 40
£ 30
4 20


\




I
a Acalat*
• QtuOOM

-,,,,,-^n.,
                  10
                        15
                             X
                                   15
                                        30
Figure 1.  Th« effect of •xtvmal carbon >ourc«s on r»dox po-
         tential and TNT degradation In cultures containing 5
         percent  Umatflla soll/pho»phat» buffer  and inocu-
         lated wrlth treated aoll.
                                                   157

-------
 the rate and extent of reduction of TNT. To further exam-
 ine this observation, experiments were conducted to
 deteimine the effects of 2A on the reduction of TNT and
 on the degradation of 2A. The results show that when
 2A was spiked into the media containing TNT, a reduc-
 tion in the rate and extent of degradation of TNT and 2A
 occurred (Figure 3).


 Summary and Conclusions

 Glucose 'Mas used successfully  as an external  carbon
 source, allowing an accurate calculation of the oxygen
 demand and a determination of the amount to add that
 would allow consumption of the oxygen present initially
 and maintenance of anaerobic conditions for a specified
 time. Calculations show that 28.8 mg/L of glucose must
 be supplied to remove aJI initial dissolved oxygen (DO)
 and keep the aqueous phase free of DO, assuming an
 initial DO of 9.08 mg/L, a reaeration rate of 0.908 mg/L,
 and an incubation time of 24 days. The  calculation as-
 sumed that all glucose was used for  oxygen consump-
 tion, and no fermentation of  the glucose occurred. To
 correct for this,  a figure of  100 mg/L glucose could be
 used as a  conservative starting point.  Future experi-
 ments  at  the  University of  Houston  will  determine
 whether this figure is sufficient to allow the creation of
 and to sustain  anaerobic  conditions for the required
 period, and whether the  institution of an aerobic stage
 is benaficial to the procedure.

 The process must be engineered towards rapid removal
 of intermediates rather than onty rapid removal of TNT.
 This rapid removal will ensure that buildup of toxic inter-
 mediates will not occur and  that the process may be
 performed reliably in the field.  The development of more
 efficient inocula  that will ensure efficient removal of in-
 termediates produced during TNT degradation currently
 is under investigation at the University of Idaho and the
 University of Houston. The  effects of  the intermediates
 on the growth and metabolic activities of  the organisms
 involved also is  being investigated at the University of
 Houston.


 References

 1.  Kaplan, D.L. 1990. Biotransformation pathways of
    hazardous energetic organo-nitro compounds. In:
    Kamery, D., A. Chakrabarty, and G.S. Omenn, eds.
    Biotechnology and biodegradation.  TX:  Portfolio
    Publishing Company, p. 155.

2.   Fernando, T., and S.D.  Aust, eds. 1991.  Biodegra-
    dation of munition waste, TNT (2,4,6-trinitrotoluene),
    and RDX (hexahydro-1,3,5-trir,itro-1,3,5-tria2ine) by
    Phanerochaete chrysosponum. In: Emerging tech-
    nologies in hazardous waste management Ameri-
    can Chemical Society, p. 214.
   too

   X

   K

   70

   90
A
\

$
/,
/
>

V
A


r
I
*H
9



\,

7


>

x
i
\





\
7



**


\



-



1C


-


4



-


t~

2.4S.THT
4A2.SONT
2.4OMMT
' IJJLi r I n'P'"" • •
'ftrrn"^:::'
n

— X_I_rr • , r ,
15 » 25 X
 Figure 2. Concentration* of TNT and Its metabolic intorrrmdl-
         at»* during the anaerobic remediation erf Umatillas
         oil In culture* Inoculated with treated soil.
                                   • Omgrt. 4A19DWT

                                   • 20 m&. UlKCm

                                   O 40 mglL tA2.«DNT
Figure 3.  Concentration* of TNT In acueoua cultures Inocu-
         lated with treated soil degrading 100 mg/L TNT In trie
         preeence of 4-emino-2,6-dlnitrotoluene.
3.  Fernando, T, and S.D. Aust. 1991. Biological de-
    contamination of water  contaminated with explo-
    sives    by    Phanerochaete    chrysosporium.
    Proceedings of the IGT Symposium on Gas, Oil,
    Coal and Environmental  Biotechnology III. pp. 193-
    206.

4.  Fernando, T., J.A. Bumpus, and S.D. Aust. 1990.
    Biodegradation  of TNT  (2,4,6-trinitrotoluene)  by
    Phanerochaete chrysosporium. Appl. Environ. Mi-
    crobiol. 56:1,666-1,671.

5.  Boopathy, R., and C.F. Kulpa. 1992. Trinitrotoluene
    (TNT) as a sole nitrogen  source for a sulfate reduc-
    ing bacterium Desulfovibrio sp. (B strain) isolated
    from an anaerobic digester. Curr. Microbiol. 25:235-
    241.

6.  Boopathy, R.,  M. Wilson, and  C.F. Kulpa. 1992.
    Biotransformation of 2,4.6-trinitrotoluene (TNT) by
    a sulfate reducing bacterium \B strain) isolated from
    an  anaerobic  reactor treating furfural.  Abstract
                                                  158

-------
    Q143. Presented at the American Society for Micro-
    biology 92nd General Meeting, New Orleans, LA

7.   Funk, S.B., D.L Crawford, D.J. Roberts, and R.L
    Crawford. 1994. Two stage bioremediation of TNT
    contaminated soils. In: Schepart, B.S., ed. Biore-
    mediation of pollutants in soil and water. ASTM STP
    1235. Philadelphia, PA: American Society for Test-
    ing Materials.

8.   Funk, S.B., D.J. Roberts, D.L Crawford, and R.L
    Crawford. 1993. Initial-phase optimization for biore-
    mediaticn  of  munition  compound-contaminated
    soite. Appl. Environ. Microbtol. 59:2,171-2,177.

9.   Funk, S.B., DJ. Roberts, and  R.A.  Korus.  1992.
    Physical parameters affecting  the anaerobic degra-
    dation of TNT in munitions-contaminated soil. Ab-
    stract Q142.  Presented at the  American Society for
    Microbiology 92nd General Meeting, New Orleans,
    LA.

10. Roberts, DJ., S.B. Funk. D.L Crawford, and R.L.
    Crawford.  1993. Anaerobic  biotransformation  of
    munitions wastes. In:  U.S.  EPA. Symposium on
    bioremediation  of  hazardous  wastes:   Research,
    development and Meld  evaluations  (abstracts).
    EPA/600/FI-93/054. Washington. DC (May).

11. K*ake, R.H., DJ. Roberts,  TO.  Stevens,  R.L
    Crawford, and D.L. Crawford. 1992. Bioremediation
    o*  soils contaminated  with  2-seo-butyW,6-dini-
    trophenol  (Dinoseb).  Appl.  Environ.  Microbiol.
    58:1,683-1,689.

12.  Roberts, D.J., R.H. Kaake, S.B. Funk, D.L. Craw-
    ford, and R.L. Crawford. 1992. Anaerobic remedia-
    tion of Dinoseb from contaminated soil: An onsite
    demonstration. Appl. Eiochem. Biotechnol. 39:781-
    789.

13.  Roberts, D.J., R.H. Kaake, S.B. Funk, D.L Craw-
    ford, and R.L Crawford.  1992. Field fcale anaero-
    bic bioremediation of Dinoseb-contaninated soils.
    In: Gealt,  M., and M. Levin, eds. Botreatment of
    industrial and hazardous wastes. New York,  NY:
    McGraw-Hiil.

14.  Roberts, DJ., and D.L Crawford. 1991. Anaerobic
    degradation of TNT. Abstract 0160. Presented at
    the American Society for Microbiology 91st General
    Meeting, Dallas, TX.

15.  Roberts, D.J., S.B.  Funk, and R.A. Ksrus. 1992.
    Intermediary metabolism during anaerobic degra-
    dation of TNT  from munitions-contaminated soil.
    Abstract Q136.  Presented at the American Society
    for Microbiology 92nd General Meeting, New Or-
    leans, LA.

16.  Ahmad,  F, and DJ.  Roberts. 1994. The use of
    narrow bore HPLC-diode array detection to identify
    and quantitate intermediates during the biological
    degradation of 2,4,6-trinitrotoluene. J. Chromatog.
    (In press)
                                                159

-------
       Covalent Binding of Aromatic Amines to Natural Organic Matter: Study of
           Reaction Mechanisms and Development of Remediation Schemes
                                     Eric J. Weber and Dalizza Colon
             U.S. Environmental Protection Agency, Environmental Research Laboratory, Athens. GA

                                           Michael S. Elovitz
                 National Research Council, Environmental Research Laboratory, Athens, GA
 Aromatic amines comprise an important class of envi-
 ronmental contaminants. Concern over their environ-
 mental late arises from the toxic effects that certain
 aromatic amines exnibit toward microbial populations
 and reports that they can be toxic or carcinogenic to
 animals. Aromatic amines can enter the environment
 from the degradation of textile dyes, munitions, and
 numerous herbicides. Because of their importance as
 synthetic building blocks for many industrial chemicals,
 the toss of aromatic amines to the environment also may
 result from production processes or improper treatment
 of industrial waste streams. The high probability of con-
 tamination of soils, sediments, and ground-water aqui-
 fers wifi aromatic amines necessitates the development
 of  innovative, cost-effective in situ remediation  tech-
 niques for their treatment

 Numerous studies have demonstrated  that aromatic
 amines become covalentiy bound to the organic fraction
 of soils and sediments through oxidative coupling or
 nudeophilic addition reactions (1-4).  It is generally ac-
 cepted that once bound,  the bound residue  is less
 btoavailable and less mobile than the parent compound.
 Thus, procedures for enhancing the irreversible binding
 of aromatic amines to soil constituents could potentially
 serve as remediation technologies.

 Model studies suggest that oxidative enzymes derived
 from soil  microorganisms play a significant role in cata-
 lyzing the formation of bound residues (5,6). Stimulation
of these naturally occurring enzymes could provide an
effective in situ method for the treatment of soils, sedi-
ments, and ground-water aquifers contaminated with
aromatic  amines  (7). For example, Berry and Boyd (8)
were able to enhance the covalent binding of the potent
carcinogen 3,3'-dicrsloroben2idine (DCB) in a soil by the
addition of highly reactive substrates (i.e., ferulic acid
and  hydrogen peroxide).  They  concluded that by
 providing the indigenous peroxidase enzymes  with
 highly reactive substrates, the overall level of oxidative
 coupling in the soil was increased, which lead to en-
 hanced incorporation of DCB.

 To gain a more in-depth understanding of the enzyme-
 mediated binding of organic  amines to soils and sedi-
 ments, we   have  studied  the  effects  of enzyme
 amendments to sediments treated with aromatic amines
 such as aniline, reduction products of TNT and atrazine,
 and metabolic reaction products of atrazine.

 Results and Discussion

 Initially, experiments were conducted  to determine trie
 limiting factors controlling the binding of  aniline to
 amended sediments. Figure 1 il'ustrates the effect of the
 addition of various combinations of horseradish peroxi-
 dase, H202,  and ferulic acid to Beaver Dam  sediment-
 water systems treated with aniline at an initial aqueous
 concentration of 5 x 10's M. In each case,  the amend-
 ments were added 24 hours after the addition of aniline.

 The data in Figure 1 show that the binding capacity of
 the sediment for aniline was limited prior to the additon
 of the amendments. Only  10  percent of the initial con-
 centration of aniline was irreversibly bound to the in-
 treated natural sediment All amendments tested gret tly
 enhanced the removal of aniline from the aqueous
 phase of the Beaver Dam sediment-water systems, as
 the concentration of aniline in the aqueous phase was
 bet.« detectable limits in a matter of hours. The obser-
vation  that the addition of H2O2  alone catalyzed the
 removal of aniline suggested that the sediment was not
limited in peroxidase activity or oxidizable substra'.es.

To determine the effect of H2O2 on the binding of aniline
in  a sediment with no peroxidase,  we monitored the
aqueous concentration of aniline in both a nonsteri's and
                                                160

-------
    0.06
Figure 1.  Effect of amendmenti on the aqueoua phaae concen-
         tration of aniline In Beaver Dam sediment-water sys-
         tem:  (•) control, no treatment.  (•) (crude acid,
         peroxtdase, and HjOi; (+) feniUc add and HjOj; and
         (D)rbOi.
    SE-5


    SE-5


    4E-5


    3E-5


    2S-5


    1E-5


    OE*0
                    10
 15    20    25
Time (hour*)
30
Figure 2.  Effact of hydrogen peroxide treatment on the aqua-
         oua concentration of Aniline In a Beaver Dam sedi-
         ment-water system: (•) nonatartla control, no HjOi
         treatment; (») nonsterlle aadlmant traatad with HjOi
         at te24 hr; and (*)  rteat-etartllzed aadlmant traatad
         wrtti HtOt at W4 hf.

a heat-sterilized Beaver Dam sediment with and without
the addition of HjOj (Figure 2). The aqueous concentra-
tion of  aniline was measured for 24  hr prior  to the
addition of H^. As before, the control study (no addi-
tion of H2O2) demonstrated the limited binding capacity
of the sediment for aniline. Surprisingly, the addition of
H3L< 24 hours after the initial addition of aniline had a
significant effect on the aqueous concentration of aniline
in both  the  sterile and nonsterile sediment-water
systems.

Because our initial assumption was that heat steriliza-
tion would  destroy peroxidase activity, the observation
that treatment  of  the  heat-sterilized Beaver  Dam
sediment-water system greatty enhanced the removal of
aniline suggested that a mechanism other than peroxi-
dase activation may exist. The high iron content of the
sediment may have resulted in the iron-mediated reduc-
tion of H202  to form hydroxyl radicals  (Fenton's reac-
tion),  which could subsequently react  with aniline via
hydrogen  abstraction  and ring  addition. Recently, the
chemical oxidation of chlorinated organics by addition of
H2O2 to sand containing iron has been demonstrated by
Ravikumar and Gurol (9).
In an attempt to determine rf the iron-mediated reaction
was occurring, two Beaver Dam sediment-water sys-
tems were treated with H2O2 24 hours prior to the addi-
tion of  aniline. We hypothesized that if  Fenton-type
reactions  were occurring, the  extremely  reactive hy-
droxyl radicals would react quicWy with the organic mat-
ter and subsequently  would  not be available to react
directly  with  aniline upon its addition  24 hours later.
Surprisingly, at both concentrations of H2O2 studied, the
binding capacity of the Beaver Dam sfliment for aniline
was increased by treatment with H2O2 24 hours prior to
the addition of aniline. These findings suggest that hy-
doxyl radicals, like activated peroxidase, may react with
organic matter to produce binding sites for compounds
such as aromatic amines (Figure 3).
In summary, we feel that hydrogen peroxide treatment
of soils and sediments  contaminated with  aromatic
amines  and other classes of reactive chemicals shows
promise as a remediation method.  We  currentty are
extending this remediation technology to other aromatic
amines of interest such as TNT reduction products and
atrazine and its metabolites, whose contamination of
soils and  sediments has  been reported.  Experiments
are also in progress to further our understanding of the
mechanisms by which H2O2 enhances the covaent
binding of aromatic amines.
                                   6E-S
                                  OE»0
                                                      20
                                                              30
                                                                      40
                                                                              50
                               Flgur* a. Effect of HiOj treatment of a Seaver Dam sediment-
                                       water lyatam 24 hours prior to the addition of aniline-
                                       Initial [aniline] - 5 J * 10 5 M.
                                                   161

-------
References
1. Baughrnan, G.L, E.J. Weber, R.L Adams, and M.S.
   Brewer.  1992.  Fate of colored smoke  dyes. Army
   Project No. 88PP3863. U.S. Department of the Army,
   Frederick, MD.
2. Graveel,  J.G.,  LE. Sommers, and  D.W. Nelson.
   1985. Sites of benzidine, -naphthylamine,  and f>
   toluidine retention in soils. Environ. Toxicol. Chem.
   4:607-613.
3. Paris, G.E. 1980. Covalent binding of aromatic
   amines  to  humates. 1. Reactions  with carbonyl
   groups  and  quinones.  Environ.  Sci.  Techno).
   14:1,099-1,105.
4. Scheunert, I.,  M. Mansour, and  F. Andreux. 1992.
   Binding of organic pollutants to soil organic matter.
   Intern. J. Environ. Anal. Chem. 46:189-199.
5. Bollag, J., and W.B. Bollag. 1990. A model for enzy-
   matic binding  of pollutants  in the soil. J. Environ.
   Anal. Chem. 39:147-157.

6. Claus, H., and Z. Filip. 1990. Enzymatic oxidation of
   some substituted phenols and aromatic amines, and
   the behavior of some phenoloxidases in the pres-
   ence of soil related adsorbents. Water Sci. Tech.
   22:69-77.

7. Bollag, J. 1992. Decontaminating soil with enzymes.
   Environ. Sci. Technol. 26:1,876-1,881.

8. Berry, D.F., and SA Boyd. 1985. Decontamination
   of soil through enhanced formation of bound resi-
   dues. Environ. Sci. Technol.  19:1,132-1,133.

9. Ravikumar, J.X..  and M.D. Gurol. 1994. Chemical
   oxidation of chlorinated organics by hydrogen perox-
   ide in the presence of sand. Environ. Sci. Technol.
   28:394-400.
                                                162

-------
               Kinetics of Anaerobic Blodegradatlon of Munitions Wastes
                                   Jiayang Cheng and Makram T. SukJan
          Department of Civil and Environmenta] Engineering, University of Cincinnati, Cincinnati, OH

                                            Albert D. Venosa
         U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH
 2,4-OinitrotDluene (2,4-DNT) is formed during the manufac-
 ture of propeilant and is commonly found in munitions waste-
 water. It has been found to be mutagentc in bacterial and
 mammalian assays and carcinogenic in animal studies (1).
 Because of its toxic nature and large-scale use, 2,4-DNT is
 listed as a priority pollutant by EPA. Earty studes on the
 biodegradation of  2,4-ONT suggested  that 2,4-ONT was
 resistant to biotogical treatment in aerobic processes such
 as activated sludge systems (2). Recently, some investiga-
 tors reported complete degradation of 2.4-DNT by a pure
 aerobic  culture (3,4).  industrial appication of the aerobic
 biodegradation of 2,4-DNT, however, reveals that it is very
 difficult to  achieve compliance with EPA discharge  limits.
 Under anaerobic condrbons, 2,4-DNT can be completely
 transformed to 2.4-diaminotoluene (2,4-DAT) with ethanol
 serving as the primary substrate (5). Subsequently. 2,4-OAT
 can be easily mineralized aerobically (5).

 In this study, the anaerobic botransformatbn of 2,4-DNT
 with ethanol serving as the primary substrate was investi-
 gated. The culture was acciimated in a chemostat with
 2,4-ONT and ethanol as substrates. The pH and the tem-
 perature in the chemostat were kept at 72 and 35°C, re-
 spectively. The hydraulic retention time in the chemostat was
 40 days. Biochemical methane potential (BMP) tests with
 2,4-ONT and ethanol as substrates were conducted jsing
 an anaerobic respirometer with the culture from the chemo-
 stat serving as an inoculum. SooSum surfide and L-cysteine
 hydrochloride were used to maintain a reducing environ-
 ment for the BMP tests. The impact of the reducing  agent
 on the biotransformation of 2,4-DNT and ethanol was stud-
 ed. The effect c4 ^,4-DNT, the biotransformation irrterme*
 ates, and 2,4-DAT on the bkxonversion of ethanol also was
 investigated.

 Results and Discussion

After steady-state operation was established in the che-
mostat (i.e., the effluent composition, the volumetric gas
production rate and composition, and the biomass con-
centration in the chemostat had been constant for over
120 days), mixed culture from the chemostat was used
as  an inoculum for the BMP tests. The culture was
transferred into the BMP  reactors  in  an oxygen-free
anaerobic chamber at 35°C. The pH and the tempera-
ture in the BMP reactors were kept the same as those
in the chemostat Different initial 2,4-DNT concentra-
tions were used in the BMP tests, while the initial con-
centration of ethanol was the same in all of the reactors.
Fgure 1 illustrates the biotransformation process of 2,4-
DNT,  in the  presence of ethanol and 50 mg/L sodium
surfide hydrate and 100 mg/L  L-cysteine hydrochloride
as the reducing agents. 2,4-DNT was completely trans-
formed to 2,4-DAT, with  2-amino-4-nitrotoluene (2-A-4-
MT) and 4-amino-2-nitrotoluene (4-A-2-NT) appearing
as  intermediates. The initial  transformation rate  de-
creased with increasing initial 2,4-DNT concentrations
(Figure 1a).  Note that at a low initial concentration of
2,4-DNT, a greater buildup of 4-A-2-NT occurred com-
pared with 2-A-4-NT (Figures  1c and 1d). As the initial
2,4-DNT concentration increased, more 2,4-DNT was
transformed  via 2-A-4-NT (Figures 1c  to 1f). A higher
concentration of 2-A-4-NT than 4-A-2-NT was formed at
the high initial 2,4-DNT concentration (Figure 1f). The
results suggest two pathways leading to the complete
biotransformation of 2,4-DNT to 2,4-DAT (Figure 2), with
pathway (a) occurring faster at high initial 2,4-DNT con-
centrations and pathway (b) occurring faster at low initial
2,4-DNT concentrations.

Another BMP test was conducted under similar condi-
tions except,  in this instance, the reducing agent was
200 mg/L NajS 9H2O. The rate of biotransformation of
2,4-DNT was much higher, and 2,4-DNT exhibited much
less inhibition to its biotransformation as a result of the
presence of a higher concentration of surfide. The pres-
ence of the higher concentration of sulfide provided a
more reducing environment, which was favorable to the
                                                  163

-------
0.20
             20       40       60    '790   800
                                                          0.020
                                                                           20           40
                                                                                 Tlrr« (hours)
                                                                                                  780   800
                                                          0.00 »
                                                               0    20    40    60    80  100     780   800
                                                                                  Time (hours)
         O  .  •  2.4-ONT
         O  ,  •  4A2NT
         A  .  A  2A4NT
         V  .  ^  2,4-OAT
                                        	   Averag* 2,4-ONT
                                        	   Avarag* 4A2NT
                                        	   Avarag«2A4NT
                                        	   Av«rag« 2,4-OAT

Flgur* 1.  Ana*roble blotr«n*(ormatton o« 2,4-ONT wltt) «thano< •* primary substrct*.
                                                  164

-------
                                        NH,
                                      2.4-OAT
Rgur* 2.  PMhwvy of •na*rot*j Hotrwwtorrwtlon o( 2,4-ONT.

biotransformation of 2,4-ONT. An abiotic test was con-
ducted to evaluate the potential for chemical reduction
of 2,4-DNT. Results suggest that 2,4-DNT is chemically
reduced  to 2,4-OAT via 2-A^-NT or 4-A-2-NT in the
presence of high concentrations of sulflde and minerals.

The bioconversJon  of ethanol was also affected by the
reducing agent used in the BMP test L-cysteine hydrc-
chlonde Is widely used as a reducing agent in anaerobic
experiments. When L-cysteine (100 mg/L) and Na2S
(50 mg/L) were used as reducing agents in the co-me-
tabolic biodegradation   of 2,4-DNT,  propionate  was
formed during the byconversion of the primary  sub-
strate ethanol when the initial concentration of 2,4-DNT
was tower than 6 mg/L (6). No such propionate produc-
tion,  however, was observed when sulfide (200 mg/L)
was the sole reducing agent L-cysteine hydrochloride
may contribute to the formation  of proptonate during the
fermentation of ethanol in the presence of  2,4-DNT.
References

1.  Ellis, H.V., C.B. Hong, C.C. Lee, J.C. Dacre, and J.P.
   Glennon.  1985.  Subchronic and chronic  toxicity
   study of 2,4-
-------
                          Blodegradation of Chlorinated Solvents
             Sergey A. Selifonov, Lisa N. Newman, Michael E. Shelton, and Lawrence P. Wackett
        Department of Biochemistry and Institute for Advanced Studies in Biological Process Technology
                                  University of Minnesota, St Paul, MN
 Hakxxganics comprise the largest single group of chemi-
 cals on the EPA list of priority pollutants (1) because many
 of these industrially important compounds have been dem-
 onstrated to be mutagenic and carcinogenic in mammals.
 Successful application of chlorinated solvent bioremedia-
 tion requires extensive knowledge of underlying molecular
 mechanisms of btodegradation. Such knowledge will allow
 a rationale for selection of organisms  and treatment
 schemes, and prevent slow, costly empiricaJ approaches
 to btoremedate every different site.

 Microbial action on chlorinated solvents often involves
 co-metabolism or cases of fortuitous metabolism, which
 provide no net benefit to the organism involved. An
 example of this is the bacterial degradation of  trichlc-
 roethylene (TCE), a widespread ground-water pollutant.
 Gratuitous metabolism of ICE has been observed to be
 catalyzed by a number of different oxygenases: toluene
 dtoxygenase (2,3), toluene-4-monooxygenase (4), am-
 monia monooxygenase (5), soluble methane monooxy-
 genase (sMMO) (6),  propane  monooxygenase  (7),
 toluene-2-monooxygenase (8), phenol hydroxylase (9),
 and isoprene oxygenase (10). Currently methanotrophs
 expressing  sMMO oxidize TCE most rapidly in small-
 scale  laboratory  studies.  In practice,   the  use  of
 methanotrophs suffers from 1) inactivation  of  sMMO
 resulting from . Jkylation by  acyl chlorides derived from
 TCE oxidation; 2) formation of toxic chloral hydrate as a
 TCE byproduct 3) cooxidation of co-contaminants to
 more toxic materials (i.e., chlorobenzene to chlorophe-
 nols); 4) inhibition with methane; and 5) inability to main-
tain sMMO under field conditions.

 In light of the above, other TCE-degrading organisms
might outperform  methanotrophs, or  toluene  dioxy-
genase-expressing strains, over sustained periods and
under field conditions. One of our experimental models
is the  strain  of Pseudomonas cepacia G4  (8,11),
whose TCE-degrading ability is based on co-metabolic
action of the toluene-2-monooxygenase system.  The
performance and safe application of TCE-biodepraders
necessitates a greater understanding of the mecha-
nisms of oxygen addition to TCE and rigorous determi-
nation of the final recoverable products. Purification of
TMO activity from P. cepacia G4 will facilitate determi-
nation of the complete product stoichiometry of TCE
oxidation. These questions are important in the context
of understanding the physiological basis  by which P.
cepacia (toluene-2-monooxygenase, TMO) is less influ-
enced by toxic effects resulting from TCE oxidation than
are Pseudomonas  putida Ft (toluene dioxygenase,
TOO) and  other organisms.

Understanding the biochemical basis of advantages of
TMO over other chtoroethene-degraders may open new,
direct approaches for searcn of more effective strains
and enzymes.

Physiology and Biochemistry of TCE
Oxidation by P.  cepacia G4

In Vivo Studies with P. cepacia G4

Generally,  in vivo studies have focused on measuring
the disappearance of chlorinated compounds. Supple-
menting this information, however, with a deeper knowl-
edge of the products obtained from chlorinated solvent
oxidation is crucial.  TCE oxidation has been investi-
gated most extensively, but only substoichiometric  ac-
counting  of products  has been accomplished.  The
present study addresses possible formation of epoxides
from chloroethenes and of products arising  from chlo-
ride migration during oxygen addition.

Identification of TCE Blodegradation Products

In experiments with TCE, 200 uM was essentially quan-
titatively degraded by P. cepacia G4. At that time, culture
filtrates were extracted and analyzed by gas chromatog-
raphy (GC) for the  presence of the possible chloride
rearrangement  products  2,2,2-trichloroacetaidehyde
and  2,2,2-trichloroethanol.  Neither  compound  was
detected above the level of 0.25 percent of the total TCE
                                                166

-------
transformed (less than 0.5 uJvl). Analysis of culture fil-
trates  obtained  in  experiments  with  (14C]-TCE  and
washed cell suspensions of P.  capada G4 was  per-
formed by high performance  liquid  chromatography
(Bio-Rad Aminex organic acid column). The major de-
tectable  metabolite, in  all  cases,  comigrated  with
authentic gtyoxylate and  accounted for 2.5 percent, 29
percent and 19 percent of the added TCE at 0 min, 30
min, and 60 min of incubation, respectively. (Zero time
control contained live induced cells  centrfuged  with
TCE, so several  minutes elapsed beforj the cells were
actually removed from the culture supernatant fluid.) In
subsequent experiments with 10 mM gtyoxylate  added
as cold trap, more than 60 percent of the products were
accounted  for as gtyoxylate. The data  indicate  that
gtyoxylate is a likely major product and Is further meta-
bolized by P. cspacia G4. Two minor products also were
observed transiently; one of them may be formate, the
identity of other is unknown. These analyses provided
no evidence for the formation of trichloroacetate, dichlc-
roacetate, oxalate, and glycolate by P. capada G4 from
[t4CJ-TCE.

Evidence of Epoxide  Formation from
Chloroethenes by P. cepacia G4

Production of  glyoxylate infers the formation  of TCE-
epoxide as precursor. While TCE-epoxide is unstable in
water  (t^ < 1 min),  frans-1,2-dichloroethylene epoxide
undergoes   hydrolysis  and  isomerization  relatively
stowty. trans-1,2-Dichtoroethylene (trans-1,2-DCE) was
used as a model compound to obtain evidence for epox-
ide formation, frans-1,2-DCE was readily oxidized by P.
cepacia G4 induced with toluene vapor at a starting
concentration of 200 uM 85 percent of frans-1,2-DCE
was transformed after 60 min. Only 3 percent  of the
transformed frans-1,2-DCE was recovered, however, as
its colored  epoxide  adduct  with 4-(p-nitrobenzyl)-pyri-
dine (12). Noninduced P.  cepacia G4 showed no signifi-
cant  production  of material  forming  the  colored
4-(p-nrtrobenzyl)-pyridine adduct

GC/mass spectrometry (MS) and GC/Fourier transfer in-
frared (FT1R) was used to analyze pentane extracts of cell
supematants after incubation of P. cepacia G4 with trans-
1,2-DCE. A compound was found with the same R,, mass
and infrared spectra as synthetic frans-1,2-DCE epoxide.
Synthetic 2^-
-------
 Figure 1.
               HO  OH
               a   CM,
                                        MO
                                        a*
 Flgur* 2.

 Compared with TMO of P. cepada G4, enzymes such
 as TOO or methane monooxygenase are inactivated in
 vivo by reactive  intermediates generated during ICE
 oxidation; cells expressing these activities experience
 cytotoxicity from  oxidizing TCE  (13). Generally,  most
 known TCE oxidation reactions are characterized by low
 reaction  rates and formation of  harmful metabolites.
 With respect to TCE (or PCE) co-metabolism, the bac-
 teria cannot help themselves to  select against or for
 such fortuitous reactions. These  reactions provide no
 net benefit to cells as energy  and carbon sources.
 Counter argument would point out that TCE and PCE
 are not natural products, and are found only recently in
 soil and water, so natural selection has not had time to
 select against this deleterious co-metabolism.

 Using surrogate carbon and energy sources may offer
 a practical solution to finding microorganisms that 1) are
 capable of  not forming toxic substrates; and 2)  have
 higher reaction rates of TCE  and PCE oxidation compa-
 rable with the conversion rates for growth  (catabolic)
 substrates.  Either direct dihydroxylation or a monooxy-
 genation/hydration sequence would produce intermedi-
 ates (Rgure 2) capable of serving as carbon and energy
 sources.  Therefore, the enrichment  culture approach
 may provide a selection tool for finding new biological
' mechanisms capable of attacking the hindered double
 bond of PCE and TCE in an appropriate electrophilic
 environment.

 Neither TMO or TOO can oxidize such hindered com-
 pounds  as  1,1-dichloro-2-methyl-1-propene  or  1,1,2-
 trichloro-1-propene.  The  less  hindered  compound,
 1,1-difluoro-2,2-dichloroethylene, however, is oxidized
 by TOO and sMMO. This fact indicates that strong steric
hindrance rather than the electrophilic environment of
the double bond appears to be a lim.ting factor deter-
mining the success of oxidative reactions on PCE and
TCE.

This work is supported  by Cooperative Agreement
EPA/CR820771-01-0 between the U.S. EPA Environ-
mental Research Laboratory, Gtlf Breeze, and the Uni-
versity of Minnesota.

References

1.   Leisinger, T. 1983. Microorganisms and xenobiotic
    compounds. Experientia 39:1,183-1,191.

2.   Nelson,   M.J.K.,   S.O.   Montgomery,  and   PH.
    Prrtchard. 1988. Trichloroethylene metabolism by
    microorganisms that degrade aromatic compounds.
    Appl. Environ. Microbid. 54:604-606.

3.   Wackett, LP., and D.T. Gibson. 1988. Degradation
    of trichloroethylene by  toluene  dioxygenase  in
    whole cell studies with Pseudomonas putida F1.
    Appl. Environ. Microbiol. 54: 1,703-1,708.

4.   Winter, R.B., K.-M. Yen, and B.D.  Ensley. 1989.
    Efficient  degradation of trichtoroethylene by  a re-
    combinant Escherichia coli. Biotechnology 7:282-
    285.

5.   Arciero, D.,  T. Vanneili, M. Logan, and A.B. Hooper.
    1989. Degradation  of  trichloroethylene  by the
    ammonia-oxidizing   bacterium    Nitrosomonas
    europaea.  Btochem.  Biophys.   Res.  Commun.
    159:640-643.

6.   Oldenius, R., R.L Vink, J.M. Vlnk, D.B. Janssen,
    and B. Witholt 1989. Degradation  of chlo.inated
    aliphatic     hydrocarbons    by    ."tethylosinus
    trichosporium OB3b expressing  soluble methane
    monooxygenase.    Appl.   Environ.    Microbiol.
    55:2.819-2,826.

7.   Wackett  LP., G.  Brusseau, S.  Householder, and
    R.S.   Hanson. 1989.  Survey of  microbial  oxy-
    genases:  Trichloroethylene  degradation by  pro-
    pane  oxidizing bacteria. Appl. Environ.  Microbiol.
    55:2.960-2,964.

8.   Folsom,  R.R, P.J. Chapman, and PH.  Pritchard.
    1990. Phenol and trichloroethylene degradation by
    Pseudomonas cepada G4:  Kinetics and interac-
    tion between substrates. Appl. Environ.  Microbio'
    56:1,279-1,285.

9.   Montgomery,  S.O., M.S. Shields,  P.J. Chapman,
    and PH. Pritchard. 1989. Identification  and charac-
    terization of trichloroethylene degrading bacteria.
    Abstract K-68:256. Presented at  the Annual Meet-
    ing of the American Society for Microbiology.
                                                  169

-------
10. Ewers, J., D. Frier-Shroder, and H.J. Knackmuss.
    1990. Selection of trichtoroethylene (TCE) degrad-
    ing bacteria  that  resist inactrvation by TCE. Arch.
    Mtorobiol. 154:410-413.

11. Nelson, M.J.K., S.O. Montgomery, E.J. O'Neill, and
    P.H. Pritchard. 1986. Aerobic metabolism of trichlo-
    roethylene by a bacterial isolate. Appl. Environ. Mi-
    crobiol. 52:383-384.
12.  Fox,  B.G., J.B. Bomeman,  L.P. Wackett, and J.D.
    Upsccmb. 1990. Haloalkene oxidation by the  sol-
    uble  methane monooxygenase from Methylosmus
    tridicsporium OB3b:  Mechanistic  and  environ-
    mental applications. Biochemistry 29:6,419-6,427.
13.  Wackett L.P., and S.R. Householder. 1989. Toxicity
    of trichloroethylene to Pseudomonas putida F1 is
    mediated by toluene  dioxygenase. Appl.  Environ.
    Microbiol. 55:2,723-2,725.
                                                 169

-------
                Characterization of Bacteria In a TCE Degrading Biot'liter
                        Alec W. Breen, Alex Rooney, Todd Ward, and John C. Loper
                 Department of Molecular Genetics, University of Cincinnati,  Cincinnati, OH

                                           Rakesh Govind
                Department of Chemical Engineering, University of Cincinnati, Cincinnati, OH

                                           John R. Haines
         U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH
A trichloroethylene- (TCE-) degrading vapor  phase
biofilter was investigated  to determine the microbial
population(s)  mediating degradation.  Initial observa-
tions suggested that ammonia-oxidizing bacteria could
be responsible for TCE degradation. The biofilter being
studied had been ma;ntained in the presence of a gas
stream containing methylene chloride,  benzene, ethyl-
benzene, toluene, and TCE. During operation, a micro-
bial community was established that could oxidize TCE
when all other substrates were removed from the gas
stream. Twenty to thirty percent removal of TCE at an
inlet concentration of 21 ppmv  (0.113 mg/L) and a gas
residence  time of  1  minute was  experimentally ob-
served. TCE degradation capability remained intact for
more than  12 months. The standard OECD mineral salts
solution wrth excess ammonia was trickled over the
biofilter. The fact that ammonia was present in the nutri-
ent solution provided  circumstantial evidence  that it
could serve as a co-metabolite for nitrifying bacteria
mediating  TCE degradation. The  ammonia monooxy-
genase (AMO) system, responsible for the conversion
of ammonia to hydroxylamine, has been snown to carry
out a co-metabolic  oxidation  of TCE (1,2). Charac-
terization of the biofilter community was undertaken to
establish if ammonia-oxidizing bacteria were responsi-
ble for TCE oxidation.

Background

Studies on the aerobic metabolism of TCE have  shown
that a diverse group or organisms can oxidize this com-
pound in a co-metabolic fashion (3). The initial observa-
tion  by   Wilson   and  Wilson  (4)  demonstrating
co-metabolism  of TCE by methanotrophs was followed
by reports  of TCE degradation by toluene oxidizers (5),
propane oxidizers  (6),  and ammonia oxidizers  (1).
Strategies for the treatment of TCE containing wastes
often focus on the optimization of degradation using the
addition of a cc-metebolite to the appropriate group of
organisms.

Experimental System and Results

The presence of nitrifying bacteria was monitored by
most probable number (MPN) methodology and by gene
probing with an AMO gene probe. The data generated
showed that levels of ammonia oxidizers were low, gen-
erally below the level of detection of the AMO probe and
102 to 10* per gram of biofilter biomass. Following gene
probing and MPN  analysis,  TCE  degradation experi-
ments were begun.

A series  of TCE degradation experiments were con-
ducted with  biofilter biomass in a batch degradation
assay using '*C-TCE as a tracer and trapped 14CO2 as
the ultimate product of oxidation. The radiolabel experi-
ments were conducted  in 40.0-mL screw cap vials. The
vials were capped with Teflon-lined septa, allowing in-
jection into the vial. An inner vial containing 0.4 N NaOH
was placed inside the larger to serve as a CO2 trap. The
trap was assayed by scintillation  counting. The vials,
inoculated with biomass, contained 2.0 mL of media and
38.0 mL of head space. After the appropriate incubation
period, vials were acidified with 0.2 mL of 2 N H2S04 to
drive off COj. The sterile control values were subtracted
from experimental values when determining conversion
to CO2. All data reported represent the mean value of
three vials. Mass balance calculation on sterile controls
were conducted by assaying the discharge per minute
(dpm)  in the NaOH trap, the aqueous phase, and a
2.0 mL hexane extract. Greater  than  85 percent of
added  TCE could  be accounted for at  the end of the
                                                170

-------
experimental incubation time. Counts in the sterile control
were always less than 2 percent of the totaJ dpm added.

The initial phase of this study was designed to test the
hypothesis that autotrophic ammonia-oxidizing bacteria
were responsible for TCE degradation. Figure 1 shows
the effect of nitrapyrin, an inhibitor of autotrophic ammo-
nia oxidation, on TCE degradation (7,8). A number of
batch treatments on the biomass were carried ou» as
part of this study. The effects of ammonia, nitrate, phe-
nol, and glucose, in both the presence and the absence
of nitrapyrin, were examined. None of the  treatments
tosted, including those to which nitrapyrin was added.
greaty affect TCE mineralization.  These results sug-
gested that ammonia oxidizers were not responsible for
TCE mineralization.

A time course experiment was conducted over a range
of TCE concentrations in  both  the presence and the
absence of ammonia. In this experiment the oxidation
of ammonia was assayed  by a  colorimetric method to
detect both nitrite and  nitrate. For this experiment, three
TCE concentrations (0.021, 0.149, and 0.372 mg/L) and
three time points (0, 20, and 44 hr) were chosen. Am-
monia supplemented (+ ammonia) and nitrate (- ammo-
nia)  batch tests were inoculated  with  0.003 mg of
biofilter biomass. Data from this experiment are shown
in Table 1. After 1 hour, no conversion of TCE to C02
was observed  at any TCE  concentration, either with or
without ammonia. After 20 hours,  TCE  mineralization
Rgura 1.  Nitrapyrin Inhibition axparimant Bloflltaf Womasa
         (0.01 mg blomaaa/vlal) waa uaad to test tha affact of
         an Inhibitor on TCE oxidation In tha presanca of vari-
         oua Inducar compound*. Culturaa wara incubatad In
         tha prasanca of TCE (0.4 ugAvlal) for $ days prior to
         acidification. Raaults ara  raportad aa parcant of
         addad radlolabal racovarad aa CO?:  1) haat-killad
         control; 2) tima 0; 3) ammonia traatad; 4) ammonia
         plua nitrapyrin; 5) nltrata; 8) nttrata plua nitrwpyrin;
         7) phano* traatad; 8) phand plua nitrapyrin; 9) glu-
         coaa traatad; and 10) glucoaa plua nitrapyrin.
occurred at lower TCE concentrations. No miren-za-
tion occurred at the highest TCE concentration a: ;:
hours or at 44 hours. Conversion to CO,, m the va:s at
the lowest TCE concentration appeared to level ctf n 20
hours, showing little increase after 44 hours. The 0.*49
mg/L  TCE concentration continued to demonstrate in-
creased TCE conversion  at 44 hours.  The effect  or
ammonia does not appear to be great at any concentra-
tion. A slight enhancement of mineralization m the am-
monia-treated sample occurred after 20 hours, and a
slight decrease in  the  ammonia-treated  sample oc-
curred after 44 hours, vials from the 44-hours time point
were  assayed  for  nitrite  and nitrate by  colonmetnc
assay. No nitnte or nitrate  was detected in any vials.
suggesting that little ammonia oxidation was occurring.
The nitrogen  source had no effect on TCE mineraliza-
tion. At this point, the biomass was examined to deter-
mine  which organisms were mineralizing TCE without
co-metabolite addition.

The persistence of  aromatic hydrocarbon  oxidizers  in
the biofilter suggests that they may be responsible for
TCE  oxidation.  Enrichment  cultures  using   biofilter
biomass were incubated in 50.0 rnL flasks in 10 ml of a
mineral salts medium. These flasks were placed in 5-gal
desiccators and exposed to 0.5 mL of either toluene or
benzene. These flasks grew to turbidity and produced a
yellow metabolite indicative of  aromatic ring cleavars.
The yellow metabolite was observed at the  greatest
dilutions tested (10"*). These enrichment cultures were
tested for mineralization in mineral salts in the absence
of toluene or  benzene, and showed high levels of TCE
mineralization. The  predominant culture appearing on
vapor phase  plates appears to be unique relative to
previousfy described organisms and  is being charac-
terized. In contrast, TCE mineralization assays of posi-
tive MPN cultures did not mineralize TCE.
Conclusions

• Ammonia oxidizers are present in the biofilter, but at
  low levels.

• Removal of ammonia from the medium did not effect
  TCE mineralization by the biomass.

• Addition of the inhibitor nitrapyrin did not effect TCE
  mineralization by the biomass.

• Nitrifier enrichment cultures from the biofilter did not
  mineralize TCE.

• A high level  of toluene/benzene oxidizers is present
  in the biofilter, and enrichment cultures can  mineral-
  ize TCE without addition of an organic co-metabolite.
  These cultures are robust in the biofilter environment
  and have persisted in the bicfirter for more than 1 year.
                                                  171

-------
Table 1.  TCE Mineralization by Blofilter Blomaaa wtth and without Ammonia Addition

                                                          TCE Mineralization

  NK.
              TCE
                                     1 hr
                                                                20 Hr
                                                                                            44 hr
* 0.4
+ 2.9
+ 725
0.0
0.0
0.0
0.0
0.0
0.0
15.0
4.3
0.0
0.06
0.12
0.0
16.2
12.3
0.0
0.06
0.36
0.0
-•
-
-
0.4
2.9
725
0.0
0.0
0.0
0.0
0.0
0.0
11.6
3.0
0.0
0.05
0.09
0.0
15.3
13.1
0.0
0.06
0.38
0.0
•NKrata substituted tor ammonia
% • Percent of added radtolabel recovered aa COj
References
1. Arciero, D., T. Vannelli, M. Logan, and A.B. Hooper.
   1989. Degradation of trichloroethylene by the ammo-
   nia oxidizing bactarium Nitrosomonas europea. Bk>
   chem. Biophys. Res. Commun. 159:640-643.
2. Hyman, M.R., R. By, S. Russell, K. Williamson, and
   D. Arp. 1993.  Co-metaboiism of TCE by nitrifying
   bacteria. In:  U.S. EPA. Symposium on bioremedia-
   tion or hazardous wastes: Research, development
   and field evaluations (abstracts). EPA/60Q/R-93/054.
   Washington. DC (May).
3. Enstey,  B.D. 1991. Biochemical  dversity of trichto-
   roethylene metabolism. Ana Rev. Microbio). 45283-300.
4. Wilson, J.T., and B.H. Wilson. 1994. Biotransforma-
   tton of trichloroethylene in soil. Appl. Environ. Micro-
   bkDl. 49:242-243.
                                                     5. Nelson, M.KJ., S.O. Montgomery,  E.J. O'Neil, and
                                                        P.H. Pritchard. 1986. Aerobic metabolism of trichlo-
                                                        roethylene by a bacterial isolate. Appl. Environ. Mi-
                                                        crobiol. 52:383-384.

                                                     6. Wackett, L.P., G-A. Brusseau, S.R. Householder, and
                                                        R.S. Hanson. 1989. Survey of microbial oxygenases:
                                                        Trichloroethylene degradation by propane oxidizing
                                                        bacteria. Appl. Environ. Microbiol. 55: 2.960-2.964.

                                                     7. Oremland, R.S., and  D.G. Capone.  1988.  Use of
                                                        "specific* inhibitors  in  biochemistry and microbial
                                                        ecology. Adv. Microb. Ecol. 10:285-383.

                                                     8. Powell,  S.J., and J.I. Prosser.  1984. Inhibition of
                                                        biofilm populations  of Nitrosomonas europea. Mi-
                                                        crob. Ecol. 24:43-50.
                                                  172

-------
            Bloremedlatlon of TCE:  Risk Analysis for Inoculation Strategies
                                 Richard A. Snyder and Malcolm S. Shields
       Center tor Environmental Diagnostics and Bioremadiation, University of West Florida, Pensacola, FL

                                              P.M. °ritchard
           U.S. Environmental Protection Agency, Environmental Research Laboratory, Gulf Breeze, FL
 The introduction of non-native species has a colorful past
 tor metazoan organisms. Controlled introductions of nort-
 native or genetically engineered bacteria to data have not
 been documented to cause undesirable effects. The ubiq-
 uity of microorganisms has been largely assumed, provid-
 ing a rationale for the safe release of "non-native" bactena.
 The ubiquitous dsTibution argument assumes that all mi-
 croorganisms have equal opportunity to occur in all envi-
 ronments,  and  that  selective  pressures  determining
 Distribution and abundance will eliminate introduced micro-
 organisms that do not already occur in the target environ-
 ment The most successful  introductions have  resulted
 from isolating an organism from the targeted environment
 modifying it and returning it to its previous  niche, e.g.,
 Rhizotoum spp. (1). An alternate strategy that has proved
 effective is to modify the environment to provide a niche
 for the phenotype of interest and to allow natural selective
 processes to occur (2). The  success of these strategies
 supports the ubiquity  argument The history of virulent
 pathogen distribution, however, provides a model to warn
 us that the microbial world is not entirely homogeneous,
 and that some environments may be subject to invasion
 by non-native microorganisms. With the development of
 bactena with potentially novel genetic combinations, we
 have a responsibility to determine  if released organisms
 wiR be constrained by the selective pressures of the target
 environment

 Bacterial populations  in nature are under constant se-
 lective pressures from physical and chemical conditions,
 substrate availability for growth, competition between
 species, and predatory/viral interactions. The balance of
 these forces determines both bacterial species  compo-
 sition and individual species' abundance. The  relative
significance of the biological factors (growth, competi-
tion, and predation)  is determined  by physical  and
chemical factors, as the limits of individual species' tol-
erance are reached within trophic or contaminant gradi-
ents. The addition of bactena to environmental microbial
 communities may locally and temporarily change
 the balance of selective pressures, but these cells
 would ultimately face the selective forces of the target
 environment.

 We have begun to address the abiotic and biological
 parameters for survival of Pseudomonas ceoacia G4
 PR-1  in  laboratory microcosms utilizing ground water
 and sediment from the aquifer beneath the Borden Ca-
 nadian Armed Forces Base in Ontario, Canada. This site
 is proposed for a bioremediation test using a funnel-and-
 gate  technique (3) to control ground-water flow  and
 force  a trichloroethytene-  (TCE-) contaminated  plume
 through biocassettes colonized with PR-1. This  bacte-
 rium constitutivety expresses a toluene orthomonooxy-
 genase that mineralizes TCE (4). The Borden aquifer is
 oligotrophic (3.5 to 6 mg DOC L"'), with a ground-water
 flow of approximately 10 cm/day"1 through a well-sorted
 fine sand sediment  (5). Determining  the transport of
 bacterial cells from a treatment zone as  well as their
 survival necessitates the development of field tracking
 methods for the organism and the plasmid that confers
 the ability to mineralize TCE.

 Approach and Preliminary Results

 Results obtained from analysis of the behavior of PR-1
 in aquifer material in laboratory tetts will be compared
 with the response at field scale c!'jnpg the release. This
 combination is hoped to highlight basic biological char-
 acteristics of bacteria that can be assessed in the labo-
 ratory; in this manner,  future genetically engineered
 microorganism releases can be evaluated without ex-
 pensive testing of the organism in mesoscale or semi-
 contained systems prior to release.

 Characterization of Native Organisms

This initial phase is targeted toward  identifying potential
competitors,  predators, and  viruses  in  the  target
                                                  173

-------
environment Selective plating and gene probing are
employed to identify G4-like  organisms that may be
displaced by the addition of PR-1 or that may contribute
to the loss of PR-1. Phenol-utilizing bacteria in the rela-
tively pristine Border* aquifer represent about 3 percent
of the colony-forming units (CFUs)  obtained on the
ground-water medium R2-A. In contrast,  aquifer mate-
rial from a TCE-contaminated site in Wichita, Kansas,
had 62.6 percent of the R2-A CFUs appearing on phenol
plates. Whether these differences will affect PR-1 sur-
vival remains to be determined.

We have enumerated protozoan  predators of PR-1  in
most probable number (MPN) growth assays using PR-
1  cells  as  the growth substrate.  Both  flagellates
(391  gdw"'), naked  amoebae (298 gdw'), and tes-
taceans  (52 gdw'1) have been  recovered that respond
quickly and grow very well on PR-1 cells. The species
diversity and numbers of protozoans recovered by this
method  are  higher  when  sterile-filtered site  ground
water is used as a diluent rather than a phosphate buffer
(6) or sodium pyrophosphate as a mild surfactant

Both viruses and competitive interactions between PR-1
and native bacteria isolated on plates will be assayed
using overlay plates with  PR-1 cells and  scoring for
clearing  zones. Native viruses have not been reported
from aquifer environments as yet but their widespread
distribution ir, terrestrial  and aquatic environments al-
most ensures their occurrence. Whether active viruses
against PR-1 cells exist in  the target environment re-
mains to be determined.

PR-1 Tracking

A monoclonal antibody has been  prepared against the
o-side chain  of PR-1 LPS (7). We have  tested this
monoclonal against a wide variety of bacteria, including
other P.  cepacia strains  and isolates  from the Borden
aquifer, without evidence of cross reactivity. We have
also tested the use of the monoclonal by tracking sur-
vival of PR-1  in laboratory microcosms  by direct im-
munofluorescence  and immunobtots of colonies from
plates.

We are developing a  polymerase  chain reaction (PCR)
detection assay for PR-1 utilizing the unique junction
sites of Tn-5 from the insertion mutagenesis in both the
plasmid and the genome.  A set of three primers has
been used to target an IS50 on the plasmid: two flanking
primers and one asymmetrically situated  in the interior
sequence. This primer set yields a two band "fingerprint"
when the PCR product is run out on gels.

PR-1 Survival

Tests  for survival  of  PR-1  in  ground  water, sediment
slurries  in shake  flasks, and flow through sediment
columns  are  being conducted with the  site material.
Preliminary results suggest that the abiotic conditions of
the aquifer are not limiting to PR-1  survival. When we
introduced 1  x 107 PR-1  ml"' into sterilized  ground
water,  no loss of PR-1  calls was observed  by im-
munofluorescent counts over 30 days, and plate counis
dropped approximately an order of magnitude and then
stabilized  for 25 days. Seven months later, both direct
counts and plate counts had dropped an additional order
of magnitude each. In nonsterile ground water, however,
PR-1 was eliminated within 10 days, despite a stable
population of total bacteria determined by direct counts
with the fluorochrome DAPI. In shaken sediment slur-
ries, 2 x  107 PR-1 was eliminated  within 4 days, and
numbers  of protozoa  increased concomitant with the
decrease in PR-1, suggesting that predation may be an
important mechanism for loss of the bacterium from the
system. Shifts in the bacterial community structure were
apparent in the slurries based on colony morphologies
on the  heterotrophic friedium R2A.

Presterilized and nonsterile sediment columns were
set up  using 50 cm long x 2 cm diameter tubes with
10 sampling ports sealed with silicone stoppers. A con-
tinuous culture of PR-1 set to a generation time of
approximately 100 hours and a cell yield of 6 x 107 cells
mL' was  used as a source to feed to the top of the
columns, with excess flow shunted off to a waste con-
tainer.  Flow through  the column  was controlled by a
pump at the column outflow and set to 10 cm/day"' as
found in the aquifer. PR-1 cells were detected in the
effluents with fluorescent antibodies after one void vol-
ume passed through the column  (4.5 days). After two
void volume  replacements,  the  inflow  of  cells  was
stopped and switched to basal salts in an attempt to
elute PR-1 from the columns. As in the ground water and
sediment slurries, PR-. persisted at higher levels in the
sterile versus the nonsterile column, and we detected
high numbers of bacterivorous flagellates in the nonster-
ile system. Unlike the ground water and sediment slur-
ries, PR-1 persisted through 22 days of elution in the
presence of predators. Extraction of the sediments with
0.1  percent sodium pyrophosphate at the termination of
the experiment  indicated that more of the PR-1 cells in
the nonsterile system were partide associated than free
in the pore water compared with the presterile system.
Conclusions

The preliminary results from our laboratory tests indicate
that the abiotic conditions of the aquifer will not affect
the persistence of PR-1, but losses to biological vectors
will be a major factor. Cells free in the pore water will be
quickly eliminated, but PR-1 may find refuge from pre-
dation in association with sediment particles that will
allow long-term persistence of the organism in the target
environment.
                                                 174

-------
Acknowledgments
This work was supported by EPA Cooperative Research
Agreement CR822568-01-0. Steven Francisconi (NRC
Post-Doc at U.S. EPA GBERL) contributed sequencing
data and probe design. Thanks  also to technicians
Wendy S. Steffensen and Shiree Enfinger, and to under-
graduate assistants Margo Posten, John  Millward, and
Angela Andrews.

References

1.  Pritchard. P.H.  1992.  Use of inoculation in bioreme-
   diation. Curr. Opin. Biotechnol. 3232-243.

2.  Hopkins, G.D., J. Munakata. L Semprini, and P.L
   McCarty. 1993. Trichtoroethylene concentration ef-
   fects on pilot field-scale in situ ground-water biore-
   mediafJon  by  phenol  oxidizing  microorganisms.
   Environ. Sci. Technol. 272,542-2,547.

3.  Starr, R.C., J.A. Cherry, and E.S. Vates. 1992. A new
   type of steel  sheet  piling with  sealed  joints for
   ground-water pollution control. Proceedings of  the
   45th Canadian Geotechnical Conference, Toronto.
   pp.  75-1 - 75-9.

4.  Shields, M.S., and  M.J. Reagin. 1992. Selection of
   a Pseudomonas cepacia  strain constitutive for  the
   degradation of trichloroethylene. Appl.  Environ.  Mi-
   crobiol. 58:3,977-3,983.

5.  Sudicky, E.A. 1986. A natural gradient experiment on
   solute transport in a sand aquifer Spatial vanability
   of hydraulic conductivity and its role in the dispersion
   process. Water Resour. Res. 222,069-2.082.

6.  Sinclair, J.L, and W.C. Ghiorse. 1987. Distribution of
   protozoa  in  subsurface  sediment  of a  pristine
   ground-water study site in  Oklahoma. Apd. Environ.
   Microbiol. 53:1,157-1,163.

7.  WinWer, J., K.N. Timmis, and R.A. Snyder. Tracking
   survival of Pseudomonas  cepacia  introduced into
   aquifer sediment and ground-water  microcosms. In
   preparation.
                                                 175

-------
        Studies on the Aerobic/Anaerobic Degradation of Recalcitrant Volatile
        Chlorinated Chemicals In a Hydrogel Encapsulated Biomass Biofilter
                                  Rakesh Govind and P.S.R.V. Prasad
                Department of Chemical Engineering, University of Cincinnati, Cincinnati, OH

                                           Dolloff F. Bishop
         U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH
Trichtoroethylene (TCE) and tetrachlorethylene (PCE)
are  organic  solvents most  frequently  detected  as
ground-water contaminant-  Both TCE and PCE un-
dergo  reductive dechlorinatJon  in anaerobic environ-
ments. PCE is aerobically recalcitrant

In an ongoing biofilter study, experimental work is being
conducted to evaluate the potential of gel-entrapped
biomass for treating volatile chlorinated solvents, such
as TCE and PCE, in the gas phase. Entrapped biomass
offers the possibility of aerobic/anaerobic environments
in the gel bead interior while aerobic conditions are
maintained outside the bead. The reduced environment
allows contaminants such  as TCE and PCE to be de-
graded in a biofilter column packed with gel beads con-
taining entrapped biomass.

Background

TCE degrades under anaerobic conditions, forming in-
termediates  such as  vinyl chloride, dichloroethylenes,
and  ethylene (1).  TCE also degrades under aerobic
conditions usually as a co-metabolite in the presence of
a primary substrate. A number of compounds serve as
primary substrates for TCE degradation, including aro-
matics, such as toluene and phenol (2,3); alkanes, such
as methane and propane (4.5); and 2,4 dichloroprie-
noxyacetic  acid (1).  These  microorganisms degrade
TCE because of the enzymes expressed in response to
the   primary  substrate;   for   example,   toluene
monooxgenase, which enables microorganisms to de-
grade toluene and other aromatics, allows degradation
of TCE. The primary metabolite-to-TCE ratio has been
found to be 2 g/g to 40 g/g in a recent study (6). Studies
of TCE degradation (6) were conducted in a gas-lift loop
reactor. TCE concentrations  of between  300 u,g/L (60
ppmv) and 3,000 u,g/L (600 ppmv) were degraded with
95 percent or better efficiency. Results of another TCE
study indicate that certain bacteria may be  able to
express the above enzyme even  in the absence of
toluene or phenol (7). Recently, biofiltration studies with
a 25 ppmv  gas-phase inlet concentration of TCE in a
ceiite-pellet packed bed have shown that TCE can be
successfully degraded with phenol present in the trick-
ling nutrients (8).

Materials and Methods

Activated sludge biomass in an aqueous bioreactor was
acclimated to toluene and TCE by exposing the sludge
to air contaminated with toluene and TCE tor a period
of 30 days. The reactor was supplied with mineral nutri-
ents, and the inlet and exit gas phase concentrations
were  periodically analyzed.  After acclimation  was
achieved, complete toluene conversion and abou'. 30
percent TCE conversion were observed in the reactor.
The biomass was then removed from the reactor, mixed
with k-Carragenan at 50°C, and extruded into 0.5 cm x
1.5 cm cylindrical beads. The beads,  once extruded,
were quenched in a mineral medium and then packed
in a biofilter. The experimental biofilter consists of a 1 -in.
diameter, 5-in. height bed packed  with  k-Carragenan
beads, with biomass encapsulated in each bead.

Contaminated air stream was obtained by injecting the
substrate into  the air stream by means of a syringe
pump (Harvard Apparatus, Model 11). The flow rate of
air was controlled by an MKS thermal mass flow control-
ler (Controller 1259, Control Module 247). Because both
air flow rate and substrate injection rate were precisely
controlled, uniformity of the substrate composition in the
air stream was ensured. The contaminated air stream
was introduced at the bottom of the biofilter to ensure
uniform distribution. OECD nutrient solution was intro-
duced at the top of the biofilter bed at a flow rate of 300
rnL/day.  TCE  concentrations were  analyzed  on  a
                                                176

-------
Hewlett-Packard 5710A gas chromatograph with a 20-ft
long, 1/8-in. diameter column having the packing (PT
10-percent  Alltech  AT-100 on Chromosorb  W-AW
80/100). Gamer gas was nitrogen, and the detector was
flame ionization (FID). Chloride ion concentrations in the
nutrient solution were measured by an Orion solid-state
chloride ion combination electrode  (#9617BN) on  an
Accumet 1003 pH/mV/ISE  meter. The pH  of nutrient
solutions was measured by a combination pH electrode
connected to the above meter. Ammonia-nitrogen con-
centration in nutrient solution was measured by an Orion
gas sensing ammonia electrode (#9512BN). Nitrite ions
in nutrient solutions were detected using a Hach NI-7
nitrite detection kit

Results and Discussion

Separate studies were conducted with toluene at 300
ppmv inlet concentration at various gas phase residence
times. Figure 1 shows the removal efficiency as a  func-
tion of gas phase  residence time for toluene. Toluene
degrades aerobically in the biofilter, achieving 100-per-
cent removal efficiency at less than 1 min residence
time.
Studies also were conducted with 25 ppmv inlet concen-
tration of TCE at various gas phase residence times. No
toluene was present in the inlet gas stream. Complete
mineralization of TCE was observed at a gas residence
time exceeding 4 min,  suggesting a nonaerobic  path-
way. Corresponding increases in chloride ion were ob-
served in the liquid nutrient phase, which demonstrated
that TCE was mineralized to carbon dioxide and chloride
ion. No  partially chlorinated by-products were observed
in the exit gas phase.
  100
                              To)u»fl« (300 ppmv)

                              TCE (25 ppmv)
Studies are currently being conducted to 1) measure the
disserved oxygen concentration as a function of depth
in the hydrogel bead using a microsenson 2) investigate
the effect of bead-size on reactor removal efficiency for
TCE (as the  bead size decreases,  the  extent of  the
anaerobic zone is expected to decrease); 3) develop a
mathematical model for the hydrogel bead biofilter and
validate the model using the experimental data; and 4)
extend the TCE study to other chlonnated solvents, such
asPCE.

References

1. Marker,  A.R., and Y. Kim.  1990. Trichioroethylene
   degradation by two independent aromatic degrading
   pathways  in Alcaligenes eutrophus JMP134. Appl.
   Environ. Microbiol. 56:1,179-1,181.

2. Folsom, B.R., P.J. Chapman, and PH.  Pritchard.
   1990. Phenol and trichloroethylene degradation by
   Pseudomonas cspacia G4: Kinetics and interactions
   between  substrates.   Appl.  Environ.   Microbiol.
   56:1,279-1,285.

3. Wackett. L.P., and S.R. Householder. 1989. Toxicity
   of trichloroethylene to Pseudomonas putida P1 is
   mediated by toluene dioxygenase. Appl. Environ. Mi-
   crobiol. 55:2.723-2,725.

4. Wilson, J.T., and B.H. Wilson. 1985. Biotransforma-
   tion of trichloroethylena in soil. Appl. Environ. Micro-
   biol. 49:242-243.

5. Kampbell,  D.H., J.T. Wilson, H.W.  Read, and  T.T.
   Stocksdale. 1987. Removal of volatile aliphatic hy-
   drocarbons in  a  soil bioreactor. JAPCA  37:1,236-
   1,240.

6. Ensley, B.D. 1993. Biodegradation of chlorinated hy-
   drocarbons in a vapor phase  reactor. Final report
   under contract no. 02112407. Springfield, VA:  Na-
   tional Technical Information Service.

7. Shields, M.S., R. Schaubhut, R. Gerger, M. Reagin,
   C. Soinerville, R. Campbell, and  J.  Hu-Primmer.
   1993. Bioreactor and in situ applications of a consti-
   tutive trichloroethylene degrading bacterium. Paper
   97c. Presented at the AlChE Spring National Meet-
   ing, Houston, TX (April).

8. Bishop,  D.F., and  R. Govind. 1993. Environmental
   remediation using  biofilters. Presented at Frontiers
   in Bioprocessing III, Boulder, CO (S*>otember 19-23).
                      20       30
                     R«SK*»nc« Tim* (min)
Flour* 1.  Wot of p«rc«nt removal •fflctoncy fof taluut* and
         TCE In  th«  g*l-bMd  Woflltar  «rtth ancaoaulatad
         btomas*. ToliMn* and TCE studto w«r« conducted
         ••oarataty.
                                                  177

-------
                   Metabolites of Oil Biodegradatlon and Their Toxiclty
                                          Peter J. Chapman
          U.S. Environmental Protection Agency, Environmental Research Laboratory. Gulf Breeze, FL

                                          Michael £. Shetton
                     University of Minnesota, Department of Biochemistry, SL Paul, MN

                                           Simon Akkerman
      University of West Florida, Center for Environmental Diagnostics and Bioremediatkjn, Pensacola, FL

                        Steven S. Foss, Douglas P. MkJdaugh, and William S. Fisher
          U.S. Environmental Protection Agency, Environmental Research Laboratory, Gulf Breeze, FL
Development of strategies  for the bioremediation of
crude oil and refinery processed petroleum must build
on a basic understanding of microbial degradation of oil
and its many chemical constituents, as well as the limi-
tations imposed on these processes by environmental
factors. Numerous studies document microbial activities
on bulk oil and its components (1,2), yet little is Known
of the formation,  accumulation, and toxicity  of com-
pounds during oil biodegradation. Recent reports of pe-
troleum-derived oxidation  products in ground water (3)
and in the tissues of mollusks (4) indicate the need to
characterize products formed during crude oil biodegra-
dation and to assess their environmental effects.  This
work addresses some of these questions.

Amounts of neutral and acidic materials recovered from
different oil-degrading cultures (from both marine  and
terrestrial sources) were significantly greater than from
sterile controls. Biologically generated neutral materials
were toxic (100-percent moitality) to larvae of Mysidop-
sis baftia (S). to grass shrimp embryos (6), and to em-
bryos  of  Menidia  beryllina  (7)  at  concentrations
matching those at which they were formed in cultures.
Menidia  embryos  exhibited developmental  defects.
Work is continuing  to define the nature of  the toxic-
 ^mponents of these neutral fractions, their precursors
in oil, and the microorganisms and processes that lead
to their formation.

References

1. Atlas, R.M. 1984. Petroleum microbiology. New York,
  NY: Macmillan.
2.  Leahy, J.G., and R.R. Colwell. 1990. Microbial deg-
   radation of hydrocarbons in the environment. Micro-
   btol. Rev. 54:305-315.

3.  Cozzarelli, I.M., M.J.  Baedecker,  R.P. Eganhouse,
   and D.F. Goeriitz. 1994. The geochemical evolution
   of low-molecular-weight-organic acids  derived from
   the degradation of petroleum contaminants in ground
   water. Geochim. Cosmochim. Acta 58:863-877.


4.  Bums, K.A. 1993. Evidence  for the importance of
   including hydrocarbon oxidation products in environ-
   mental studtes. Mar. PolluL Buil. 26:77-85.


5.  U.S. EPA. 1987. Short-term methods for estimating
   the  chronic toxicity  of effluents  and  receiving
   waters   to  marine   and  estuarine   organisms.
   EPA/600/4-87/028. Cincinnati, OH. pp.  171-238.


6.  Fisher,  W.,  and S. Foss. 1993. A simple test for
   toxicity  of number 2 fuel oil and oil dispersants to
   embryos of grass shrimp, Palaemonetes pugio. Mar.
   Pollut Bull. 26:385-391.


7.  MkJdaugh,  D.P..  R.L Thomas, S.E.  Lantz,  C.S.
   Heard, and J.G. Mueller 1994. Reid-scale testing of
   a hyperfiltration unit  for removal of creosote  and
   pentachkxophenol from ground water Chemical and
   biological assessment Arch. Environ. Contam. Toxi-
   col.  26:309-319.
                                                 178

-------
    TCE Remediation Using a Plasmid Specifying Constitutive TCE Degradation:
           Alteration of Bacterial Strain Designs Based on Field Evaluations
   Malcolm S. Shields, Allison Blake, Michael Reagin, Tracy Moody, Kenneth Overstreet, and Robert Campbell
    Center for Environmental Diagnostics and Bioremediation, Department of Cellular and Molecular Biology,
                                University of Wes^ Florida, Pensacola, FL

                                        Stephen C. Francesconi
    National Research Council, U.S. Environmental Protection Agency, Environmental Research Laboratory,
                                           Gulf Breeze, FL

                                             P.M. Pritchard
          U.S. Environmental Protection Agency,  Environmental Research Laboratory, C-ilf Breeze, R_
An integrated study was undertaken to determine the
potential for field application of altered  strains  of
Pseudomonas cepatia G4 (PRla and PR131) devel-
oped by us  for the btoremediation of trichkxoethylene
(TCE). The  investigation demonstrated the ability of
PR 1 a to degrade TCE without inducer substrates via
the constitutive expression of toluene ortno-monooxy-
genase (TOM). Two fundamental areas of research are
detailed: 1) the effectiveness of the PR la phenotype in
a field bkxeactor and 2) laboratory transfer of the con-
stitutive degradative phenotype to two new bacterial
strains selected for their capacity to colonize bkxeactor
matrices. PR123 was field tested in a 100-L plugged flow
reactor receiving contaminated water at 2 L/min and a
daily batch input of cells (6 L) for a period of 2 weeks.
Under these conditions, PR1& was able  to effectively
degrade TCE and os-DCE  in  contaminated aquifer
water at concentrations up to 700 ug/L (70- to 95-per-
cent removal). The PR1a constitutive TOM phanotype,
therefore, was desirable and effective. PR123. however,
gave no indication of successful colonization of the re-
actor matrix, and biodegradation activity quickfy fell fol-
lowing cessation of cell input The TCE degradative
genes and the genetic alteration responsible for  their
constitutive expression are present on a self-transmis-
sible plasmid (pTOM). PR13, was used to allow trans-
mission of the degradative plasmid (pTOM31c) contain-
ing the constitutive TOM phenotype to two alternate
Pseudomonas strains selected for superior colonization
potential.

Both strains acted as competent recipients for pTOM3,c,
consfttutively expressing the  encoded TOM and forming
active biofilms in laboratory columns containing a diato-
maceous earth matrix. This nonrecombinant transfer of
constitutively expressed TCE degradative genes to bac-
teria prescreened for  their stability in a particular envi-
ronment represents a significant advantage over past
strategies, which require that conditions be tailored to
PRla  or PR13i  survival. Such an approach may be
extended to in situ treatment scenarios by transferring
the constitutive phenotype to strains actually isolated
from TCE-contaminated sites. The resulting organisms
would have the advantage of being intrinsic to the par-
ticular site and of possessing an affective, nonrecombi-
nant degradative activity.

Portions of this research were performed under a  grant
from the U.S. Air Force, Armstrong Laboratories,  Envi-
ronfcs Directorate, Tyndall Air Force Base, Florida.
                                                179

-------
       Degradation of a Mixture of High Molecular-Weight Poly cyclic Aromatic
                       Hydrocarbons by a Mycobacterium Species
                                I. Kelley, A. Selby, and Carl E. Camiglia
    U.S. Food and Drug Administration, National Canter for lexicological Research, Division of Microbiology,
                                           Jefferson, AR
A Mycobacterium so., which was previously tested for
its ability to mineralize several individual porycydic aro-
matic hydrocarbons (PAHs), simultaneously degraded
phenanthrene, anthracene, fluoranthene, pyrene, and
benzo[a]pyrene in a six-component synthetic mixture.
Chrysene, however, was not degraded to any significant
extent When provided with a primary carbon source, the
Mycobacterium sp. degraded more than 74 percent of
the total PAH mixture during 6 days of incubation. The
Mycobacterium sp. appeared to degrade pheranthrene
preferentially. No significant difference in  degradation
rates was observed between fluoranthene and pyrene.
Anthracene degradation was slightly delayed, but, once
initiated, degradation proceeded at approximately the
same rate. Benzo(a]pyrene was degraded to a  lesser
extent Additionally, degradation of a crude mixture of
benzene-soluble PAH components from sediments re-
sulted in a 47-percent reduction of the material in 6 days
compared with autoclaved controls. Initial experiments
using environmental  microcosm test systems  indicated
that mineralization rates of individual [14C] labeled com-
pounds were significantly lower in the mixtures than in
equivalent doses of thase compounds alone.  Minerali-
zation of the complete mixture was estimated conserva-
tively to be between 49.7 percent and 53.6 percent in 12
weeks. Mineralization was nearly 50 percent  within 30
days of incubation when all compounds were radiola-
beled. These  result:, strengthen the argument for the
potential  application of  this  Mycobacterium  sp.  in
bioaugmentation of PAH-contaminated wastes.
                                                180

-------
             Bioavailability Factors Affecting the Aerobic Biodegradation of
                                     Hydrophobic Chemicals
                                            Pamela J. Morris
        Soil and Water Science Department, University of Florida, U.S. Environmental Protection Agency,
                            EnvironmentaJ Research Laboratory, Gulf Breeze, FL

                                             Suresh C. Rao
                   Soil and Water Science Department, University of Florida, Gainesville, FL

                                            Simon Akkerman
             Center for Environmental Diagnostics and Btoremediation, University of West Florida,
           U.S. Environmental Protection Agency. Environmental Research Laboratory, Gulf Breeze, FL

                                           Michael E. Shelton
          Department of Biochemistry, University of Minnesota, U.S. Environmental Protection Agency,
                            Environmental Research Laboratory, Gulf Breeze, FL

                                   Peter J.  Chapman and PH. Pritchard
           U.S. Environmental Protection Agency, Environmental Research Laboratory, Gulf Breeze, FL
 We are currerrtty studying interactions between complex
 waste mixtures and microorganisms that are capable of
 transforming organic  components of  these  mixtures.
 Our goal is to integrate methodologies used to study the
 abiotic behavior of hydrophobic organics in soil with the
 biological degradation of the organics. Sorption of hy-
 drophobic   compounds,  such   as  polychtorinated
 biphenyls (PCBs), to soil represents a potential barrier
 to their degradation and detoxification in the environ-
 ment, and influences the relative accessibility of these
 compounds to a number of  physical, chemical, and
 biological processes. We find the concept of bioavail-
 abilrty  a unique opportunity to couple interesting basic
 research to applied bioremediation problems. Our long-
 term objectives include 1) the study of the desorption of
 PCBs  from historically  contaminated  soils and sedi-
 ments; 2) the determination of the influence of co-con-
 taminants,   cosolvents,   and  surfactants  on  PCB
 desorption enhancement; and 3) the coupling of PCB
 desorption and biodegradation kinetics. The soil that we
are studying is from a former racing drag strip  in Glen's
 Falls, New York, contaminated with Aroclor 1242. Pre-
vious studies have shown that approximate^ half of the
 PCBs  present in the soil are unavailable for aerobic
biodegradation. This surface soil, classified as a sand
 (95 percent sand, 4.2 percent silt, and 0.8 percent clay),
 contains 1.9 percent organic carbon and 1.43 percent
 oil and grease.  Mineralogical analyses show that the
 soil minerals consist of 40 percent quartz, 45 percent
 chlorite, and 15 percent Ca-albite (all low internal sur-
 face-area minerals). Heavy metal analysis suggests that
 only lead levels are somewhat high, averaging 190 ppm.
 Specific surface-area analysis indicates a low value of
 0.1444 m2/g. The total pore volume is 0.0016 cnrvVg, and
 the average pore diameter  is 443.78 A. We  also are
 characterizing the following drag strip soil fractions indi-
 vidually:  medium sand (2.00 mm to 0.425 mm), fine
 sand (0.425 mm to 0.08 mm), and silt/clay (< 0.08 mm).
 Studies on the biodegradation of PCBs found in each of
 the three fractions suggest that biodegradation of PCBs
 from the silt/day fraction is  less than biodegradation
 from the fine and  medium  sand fractions. Since the
 silt/clay fraction represents the major reservoir for or-
 ganic carbon, oil and grease, heavy metals, and PCBs
 due to its high  surface area, the release of PCBs from
this fraction may be essential to enhancing  PCB biode-
gradation. The biodegradation of the PCBs found in this
fraction is currency  the focus of our  studies. We are
using the traditional batch method to examine congener-
specific desorption from the drag strip soil and the three
                                                 181

-------
fractions. In addition, we will compare trie miscible dis-
placement technique with results from batch  studies.
The miscible displacement technique uses  preparative
high performance liquid chromatography (HPLC) glass
columns packed with drag strip soil and high-precision
HPLC pumps to provide a steady flow rate.  Column
effluent fractions are collected after passage through a
flow-through variable-wavelength UV detector. Both the
batch method and miscible displacement technique al-
low us to examine the influence of cosolvents and sur-
factants (biological and synthetic) on PCS desorption
and mobility. Enhanced  desorption and mobility  may
contribute to increased availability to biodegradation
processes. In addition, we are examining the biodegra-
dation of the oil and grease in the drag strip soil. Analysis
of the oil and grease by column chromatography shows
the distribution of organics to be 81.9 percent hydrocar-
bons, 16.9 percent polars, and 1.2 percent asphaltenes.
This oil is  very weathered  and contains few readily
biodegradable components.  We  are in the process of
enriching for microorganisms capable of  transforming
this oil matrix and will test whether biodegradation of the
oil results in nhanced availability and biodegradation of
the PCBs present
                                                 182

-------
                           Section  Six
                                        Research Centers
The Hazardous Substance Research Centers (HSRO) conduct EPA research on
bioremediation under the direction of ORD's Office of Exploratory Research (OER).
Research is sponsored by the following centers: Northeast Hazardous Substance
Research Center (Regions 1 and 2), Great Lakes and K^id-Atlantic Hazardous
Substance Research Center (Regions 3 and 5), South/Southwest Hazardous Sub-
stance Research Center (Regions  4 and 6),  Great Plains and  Rocky Mountain
Hazardous Substance Research Center (Regions 7 and  8), and the Western
Region Hazardous Substance Research Center (Regions 9 and 10).
The symposium's poster session included presentations on in situ attenuation of
chlorinated aliphatJcs in glacial alluvial deposits; scaling up from a field experiment
to a full-scale demonstration of in situ bioremediation of chlorinated solvent ground-
water contamination; the bioavailability and transformation of highly chlorinated
dibenzo-p-dioxins and dibenzofurans in anaerobic soils and sediments; localization
of tetrachloromethane transformation activity in Shewanella putrfaciens MR-1; the
formation and transformation of pesticide degradation products  under various
electron acceptor conditions; laboratory and field investigations of bioremediation
of aromatic hydrocarbons at  Seal Beach, California; and pneumatic  fracturing to
enhance in situ bioremediation.
                                 183

-------
       In Situ Attenuation of Chlorinated Aliphatics in Glacial Alluvial Deposits
                       Michael J. Barcelona, Mark A. Henry, and Walter J. Weber, Jr.
   Great Lakes and Mid-Atlantic Hazardous Substance Research Center, University of Michigan, Ann Arbor, Ml
The National Center for Integrated Bioremediation Re-
search and Development (NCIBRD) has located opera-
tions  at the recently decommissioned Wurtsmith Air
Force Base (WAFB) in Oscoda, Michigan. NCIBRD is
dedicated to the evaluation of decontamination tech-
nologies for hazattous wastes and remediation of spii!
and disposal sites. These activities are administered by
the University of Michigan and oversight is provided by
a srience advisory board comprised of the directors of
»**» Hazardous Substance Resource Centers, repre-
sentatives of the EPA Biosystems Group, and nationally
recognized enginears and scientists  from government
and private sectors.

WAFB is  ideally suited for  in situ bioremediation re-
search activities. The 7-square-mile base is bordered by
the Au Sable River to (tie south  and  west and by Van
Etten Lake to the east The property sits en a 20-m bed
of highly transmissiva glacial sand underlain by a thick
silty-clay aquitard. The ground water is found at about
6 m throughout the study area. The U.S. Air Force has
been working with the U.S. Geological Survey (USGS)
to characterize the extent of contamination at WAFB for
the past 12 years, resulting in a large database and an
array of approximately 600 permanent monitoring wells.
An excess of 70 sites are tainted by a variety of sorbed,
dissolved, and nonaqueous-phase petroleum hydrocar-
bon mixtures, chlorinated solvents, and he?vy metals.
Air Force remediation activities have been limited to the
installation of three  conventional air strippers for the
containment of the largest plumes. These systems will
provide the capture zone needed for  the eventual con-
trolled release of tracer chemicals, allowing an in-depth
field study of the fate and transport of contaminants.

The USGS  database provided  information indicating
that natural bioattenuation of aromatic and chlorinated
aliphatic compounds was occurring at WAFB. A sam-
pling program is currently being  implemented to study
the process at two of these sites:  FT-02 (a heavily used
fire training area) and OT-16 (a  former jet engine test
cell).
Fire training was conducted at FT-02 from 1952 to 1993.
Typically, 3,000 L of jet fuel (and some incidental chlo-
rinated solvents) wis pumped over a simulated aircraft
structure, ignited, £°d extinguished. Unfortunately,  un-
bumed fuel and solvents infiltrated into the aquifer. The
USGS and Air Force install^ 49 monitoring wells in 17
clusters to ifack the movement of the plume originating
from this site. Preliminary well monitoring and solid bor-
ings  have shown evidence of a large  plume, with total
volatile organic compounds exceeding 1,000 mg/L, that
is undergoing natural biotrarsformation. Concentrations
of these compounds in the aquifer solids reflect co-me-
tabolic transformations;  in other words,  upgradient
vadose zone levels of trichloroethylene (5 mg/kg), BTEX
(600 mg/kg), and dissolved oxygen decrease  and con-
centrations  of cis-1,2-dichloroethylene  increase to 5
mg/kQ downgradient from the site. This site is located
approximately 300  m from OT-16 and is hydraulically
connected;  plumes from these  sites are believed to
merge downgradient

The  jet engine test cell was used for a variety of ttst
activities. Cleanup of this structure typically  involved
washing solvents off the floor into an oil-water separator,
which eventualry failed, allowing the solvents to enter
the aquifer. The plume contains high concentrations of
BTEX (4 mg/L) and moderate amounts of  chlorinated
solvents  (70  mg/L). The Air Force installed  19 wells
downgradient of this site, but littlo sampling has beer.
done. NCIBRD has just begun ute characterization ef-
forts at this site.

Future work at ^ese two sites will supplement existing
physical-chemical information with  location and geo-
physical surveys, meteorological monitoring, additional
borings  and monitoring  well emplacements,  soil  gas
surveys,  permanently installed water level recorders,
grain-size and hydraulic conductivity determinations, as
well as chemical property measurements (e g., mineral-
ogy,  carbonate, organic carbon, metal and mefal oxide
content cation-exchange capac-ty, etc.).  In  addition,
routine well sampling will document not only  contami-
                                                 184

-------
nant concentrations but also changes in metabolic lev-    conducted by consulting, private industry, and academic
els  in the aquifer. Tnis effect will support experimental    professionals.
applications of  in situ remediation technologies to be
                                                 185

-------
    In Situ Bioremediation of Chlorinated Solvent Ground-Water Contamination:
          Scaling up from a Field Experiment to a Full-Scale Demonstration
                          Perry L McCarty, Gary D. Hopkins, and Mark N. Goftz
          Western Region Hazardous Substance Research Center, Stanford University, Stanford, CA
Studies conducted at an experimental field site at Mof-
fett Naval Air Station have demonstrated  that trichto-
roethylene (TCE) can be  effectively bkxtegraded
co-metabolicalry through the introduction into the sub-
surface of a primary substrate (such as phenol or tolu-
ene) and oxygen to support the growth ano energy
requirements of a native population of microorganisms.
Additional preliminary experimental work at Moffett Field
now has been conducted in preparation for a full-scale
demonstration.

A full-scale demonstration at a real hazardous waste site
is likely to encounter a plume  with  multiple contami-
nants.  It was desirable, therefore, to determine how
other contaminants which could potentially be present
might affect the rate and extent  of TCE degradation. In
particular, previous laboratory studies at Stanford Uni-
versity have indicated that the degradation products of
1,1-dichloroethylene (DCE) are toxic to methane-oxidiz-
ing bacteria. Follow-on  field work conducted at Moffett
Field demonstrated  that the  presence  of  1,1-DCE
inhibited TCE degradation by phenol-oxidizing microor-
ganisms. Thus, 1,1 DCE should not be present at the
site  selected for a full-scale  demonstration of  this
technology.

An effective method »o provide the indigenous microor-
ganisms with sufficient  oxygen  to oxidize  the primary
substrate is needed for  the field demonstration. In past
studies  at Moffett Field, molecular oxygen has been
used as an oxygen source. Molecular  oxygen, however,
is difficult to transfer to  solution. Hydrogen peroxide is
an alternative oxygen source that has been used in
bioremediation of petroleum hydrocarbons and is much
easier to apply to the subsurface  than molecular oxygen.
Preliminary work at Moffett Field showed that hydrogen
peroxide worked as effectively as -nolecular oxygen in
degrading TCE.

Another question that needs to be answered prior to
full-scale implementation of this  technology is how best
to mix a primary substrate, an oxygen source, and TCE
and to deliver the  mixture to  the microorganisms. At
Moffett Field, mixing of these three components was
accomplished aboveground, with the mixture then intro-
duced into the subsurface through an injection well. In
a full-scale demonstration,  the TCE will, of course, al-
ready be in the ground water. A major objective of this
demonstration will be to investigate how a primary sub-
strate and an oxygen source can be efficiently mixed
and transported to indigenous microorganisms, to pro-
mote co-metabolic degradation of TCE. For the demon-
stration, a subsurface recirculation system similar to that
described by Herriing (1) and McCarty and Semprini (2)
is expected to be used. The  remediation system will
consist of a single well, screened  at two depths. In
operation, a submersible pump installed between the
two screens would  draw TCE contaminated water into
the well at one screened interval. The primary substrate
and oxygen will then be  introduced into  the  water
through feed lines,  and the water, which now contains
TCE, primary substrate, and oxygen, will be discharged
into the aquifer from the second screened interval. In
essence, an in situ treatment zone will be created in the
aquifer around the discharge screen. Based on the Mof-
fett  Field results, this  treatment zone is expected to
cover an area within  approximately 1 day's ground-
water travel distance out from the well.

Ultimately, these studies, in which the laboratory and the
Moffett Reid site are being used to make predictions
regarding processes and to help design systems at a
"real-world" site, hopefully will help lead to a  better
understanding of how laboratory and field investigations
can best be scaled up to  make better real-world
predictions.

References

1. Herriing, B. 1991. Hydraulic circulation system for in
  situ bioreclamation and/or in situ remediation of strip-
  pable contamination. In:  Hinchee,  R.E., and  R.F.
                                                 186

-------
Olfenbuttel, eds. Onsite bioreclamation. Boston, MA:    2. McCarty, P.L, and L Semprini. 1993. Ground-water
Butterworth-Heinemann. pp.  173-175.                   treatment for chlorinated solvents. In:  Norris, R.D.,
                                                    et al., eds. Handbook of bioremediation. Boca Raton,
                                                    FL: Lewis Publishers, pp. 87-116.
                                             187

-------
  Bioavailability and Transformation of Highly Chlorinated Dibenzo-p-dioxins and
                   Dibenzofurans in Anaerobic Soils and Sediments
                                   Peter Adriaens and Quingzhai Fu
                   Great Lakes and Mid-Atlantic Hazardous Substance Research Center,
          University of Michigan, Department of Civil and Environmental Engineering, Ann Arbor, Ml
Polychlorinated dibenzc-p-dioxins (PCDDs) and poly-
chtorinatad dibenzofurans (PCDFs) are introduced via
several industrial and municipal channels into both aero-
bic   and  anaerobic  environmental  compartments.
Because of their high toxicity and uncertain genotoxic
potential, their determination and fate in environmental
samples is of great interest. The fate of highly chlorin-
ated PCDD/PCDF congeners was studied in both high
and low organic carbon anaerobic microcosm incuba-
tions. The inocula were derived  from historically con-
taminated    anaerobic   environments   such   as
polychlorinated biphenyl-contaminated  sediments  and
creosote-contaminated  aquifer   samples, and  were
amended with a mixture of aromatic and aliphatic acids
for methanogenic growth. The samples were analyzed
and quantified using high resolution  9-33 chromatogra-
phy coupled with an electron capture detector and a low
resolution mass selective detector operated in  selected
ion  monitoring  (SIM) mode ([M*], [M*+2], and [M%4]
ions). Recovery efficiencies after soxhlet extraction and
sample  cleanup  were 40 to 70 percent basad  on
1,2,3,4-tetrachlorodibenzo-p-dioxin   as  an  internal
standard. The long-term ( 2 yesrs) removal patterns of
sediment-sorbed  PCDDs/PCDFs in both sediments
could be explained by labile and resistant PCDD/PCDF
desorption components, presumably because of intra-
particle diffusion-controlled mass transfer limitations.
Mass transfer limitations were based on incubation time-
dependent   decreased  extraction   efficiencies   of
PCDDs/PCDFs from inactive controls. The net first-or-
der initial rate constants of disappearance ranged from
0.30 to 0.75 (x 10~3) d"' for aquifer sediments and from
0.46 to 1.87 (x 10"3) d"1 for high organic carbon Hudson
River sediments. Moreover, the overall  decrease in
PCDDs/PCDFs from the sediment paricles in active
microcosms sacrificed after 30 months was as much as
20 percent greater compared with the autoclaved con-
trols. Lesser chlorinated  congeners were found in all
active microcosms analyzed. Isomer-specific analysis of
the lesser chlorinated congeners  indicated that  the
1,4,6,9-chlorines were removed preferentially, thus en-
riching the medii-m in 2,3,7,3-substituted conger.ers  and
increasing the overall relative toxicity. These observa-
tions contribute to our knowledge regarding the fate of
PCDDs/PCDFs in anaerobic soils and sediments,  and
indicate the importance of co; .genar "fingerprinting" dur-
ing environmental source!?. •*» analysis.
                                                188

-------
     Localization of Tetrachloromethane Transformation Activity In Shewanella
                                        Putrefaclens MR-1
                          Erik A. Petrovskis, Peter Adriaens, and Timothy M. Vogel
         Great Lakes and Mid-Atlantic Hazardous Substanca Research Center, University of Michigan,
                     Department of Civil and Environmental Engineering, Ann Arbor, Ml
Investigations of pollutant transformation by pure cul-
tures may enhance our understanding of in situ natural
attenuation processes in these environments. Stiewan-
eila putrefadens MR-1, an Fe(lll)- and Mn(IV)-reducing
facultative anaerobe, has been shown to dechtorinatr
tetrachtoromethane (CT) to chloroform (24 percent), af-
ter growth under nitrate- or Fe(lll)- respiring conditions.
Mass balance for carbon included 56-percent incorpo-
ration in biomass, 4.1-percent formation of nonvolatile
products, and 5.5-percent mineralization. Product distri-
bution was independent of growth conditions. Amend-
ment of MR-1 cell suspensions with lactate, formate, or
hydrogen  increased  CT transformation activity, while
methanol did not The rate and extent of CT transforma-
tion increased  for MR-1 cells grown with electron ac-
ceptors having  more positive half-reduction potentials
(Etf). Nitrate did not inhibit CT transformation. In tne
presence  of Fe(lll), reductive dechlorination was en-
hanced  and  resulted  in  the  production  of  dichto-
romethane (DCM), presumably by abiotic mechanisms
involving Fe{ll).
In MR-1 cell extracts.  NACH was  the most effective
electron donor for CT transformation. Addition of  FMN
increased  the activity 3- to  10-fokJ. Furthermore, CT
transformation  activity has been localized primarily to
membrane fractions (89 percent).
The effects of respiratory inhibitors on CT iransformation
activity have been examined. Rotenone, an inhibitor of
NADH dehydrogenase, reduced CT transformation ac-
tivity in MR-1 whole-cell  suspensions using lactate or
NADH as an electron donor. Quinacrine, an inhibitor of
flavins, enhanced this activity. No significant effect was
seen  in the  presence of pCMPS,  sodium azide, and
sodium cyanide or in the presence of the cytochrome
inhibitors HQNO and Antimycin A.  These results sug-
gest that transformation of  CT may be mediated by a
rtonheme electron  transfer agent
Respiratory mutants of MR-1 have been screened for
CT transformation  activity. Rates of CT transformation
for MR-1 mutants in Fe(lll) reductase, Mn(IV) reductase,
or fumarate reductase were equivalent or greater than
those  for the MR-1 wild-type strain. MR-1 mutants that
did not synthesize  menaquinones (MK) and so lost the
ability to couple nitrate, Fe(lll), or fum^rate reduction for
growth also lost 90 percent of CT transformation activity.
When cell suspensions of  MK-deficient mutants were
complamented with an MK precursor, CT transformation
rates  returned to MR-1 wild-type levels. These results
indicate that MK or another electron transfer  mediator
reduced by MK but  not a terminal reductase may be
responsible for CT transformation by MR-1.
                                                 189

-------
          Formation and Transformation of Pesticide Degradation Products
                       Under Various Electron Acceptor Conditions
                             Paige J. Novak, Gene F. Parkin, and Craig L Just
   Great Plains and Rocky Mountain Hazardous Substance Research Center, University of Iowa, Iowa City, IA
Pesticide contamination of ground-water supplies is a
serious and growing problem in the United States. More
than 600 active chemicals exist that are used to protect
crops from target pests (1). Pesticides can remain in the
environment for a long time, entering the air or ground-
water supply by partitioning to or diffusing through the
soil column. Transformation of these chemicals  to one
or more principal metabolites often occurs with unknown
and unmonrtored results. To develop systems to destroy
these contaminants and formulate intelligent policies to
regulate or restart their use, an understanding of the
reactions that these compounds undergo in the environ-
ment is essential.

The herbicides alachlor and atrazine, the two most com-
monly used pesticides in  the nation, together account
for 25 percent by weight of total pesticide use (2). These
herbicides are also the two most frequently detected
pesticide contaminants in ground-water supplies in the
Midwest (2). Many xenobiotics can undergo mineraliza-
tion to carbon  dioxide and water by biological means;
alachlor and atrazine, however, undergo very little min-
eralization under typical environmental conditions. Min-
eralization   has  been  observed  by  onty  a  few
researchers, generally at quantities of less than 5 per-
cent of the initial  herbicide concentration. As a single
exception, a recently  completed study revealed that
atrazine, when serving as the sole nitrogen source for a
microbial  population,  was mineralized  at  levels of
greater than 80 percent of the initial concentration, with
a half-life of 0.5 to 2.0 days using a microbial consortium
that had undergone more than  5 months of subculturing
and enrichment in the laboratory (3). With little natural
mineralization  occurring under typical  environmental
conditions, transformation intermediates of alachlor and
atrazine may be formed and may be accumulating in the
soil and ground water.

The specific objectives of this research project were to
identify the  transformation products  of alachlor and
atrazine  under   four  common  electron   acceptor
conditions (aerobic, denitrifying, sulfate-reducing, and
methanogenic) and, to the extent possible,  • letermine
kinetic coefficients that describe the rate of formation
and disappearance of these metabolites.


Experimental Design

Four  &-L, fill-and-draw reactors  were  established  to
maintain  specific environmental conditions. Each reac-
tor was fed a mineral nutrient solution typical of ground
water under the redox condition o.' interest. Temperature
was maintained at 20°C in the dark to  mimic environ-
mental conditions. Each of the reactors was fed acetate
as the carbon and energy source, with some of the batch
denitrifying experiments carried out with citrate as an
electron donor as well. Alachlor and atrazine were fed
at approximatety  100 |ig/L each, along with a phosphate
buffer to maintain a neutral pH. In addition, the specific
electron acceptor for each system was added in excess:
O2 for the aerobic reactor, KNO3  for the denitrifying
reactor, and MgSO« 7H2O (at a high sulfate-to-orgamc
ratio)  for the sulfate-reducing reactor. The bacteria in
each system were acclimated to alachlor and atrazine
prior to the start of the experiments.

Control experiments  were set up to  determine which
physical and chemical means of alachlor and atrazine
transformation  were important. The potential role of the
phosphate buffer in catalyzing chemical hydrolysis of
alachlor and atrazine was  studied with a phosphate
control. Reactions with resazurin, a color  indicator of
redox potential used in the denitrifying reactors, also
were studied using several control reactors with varying
res^urin concentrations. A mercuric-chloride-killed bio-
logical control was  used to investigate  sorption  to
biomass, and  to further assess the role of resazurin.
Finally, a deionized water control was employed to iden-
tify  mixing problems, the significance of alachlor and
atrazine sorption to the reactor itself, and potential vola-
tilization, chemical  hydrotysis, or photolysis reactions.
                                                  190

-------
AJI experiments were carried out in a batch format. An
initial dose of alachlor and atrazine was added to the
reactor and allowed to mix for approximately 45  mm,
then samples for pesticide analysis were taken from the
reactor at various time intervals.  The denitrifying and
control experiments were carried out in 2-L Pyrex bottles
built like the larger 9-L reactors, so several different
conditions could be tested without affecting the  stock
enrichment  culture. The experiments  involving  the
methanogenic and sulfate-reducing systems were car-
ried out in the 9-L reactors.

Results

Initially, alachlor and atrazine disappeared  in batch
reactors maintained under all terminal electron  ac-
ceptor conditions  except aerobic conditions.  Further
experiments involving the aerobic reactor were aban-
doned because of the absence of noticeable degrada-
tion of parent compounds. Resazurin was added only
to the denitrifying reactors to indicate whether  the
proper conditions were  maintained. This  compound
was found to be involved in the abiotic transformation
of alachlor  and atrazine. Second-order degradation
constants for alachlor and atrazine transformation are
given in Table 1; these constants are averaged values
for four experiments for each  of the different terminal
electron acceptor conditions.  Each of these rate con-
stants has been corrected for the abiotic  transforma-
tion of atrazine and alachlor in the denitrifying reactors
due  to  resazurin, and the abiotic transformation of
alachlor in the methanogenic and sulfate-reducing re-
actors due to the  bisulfide ion. Therefore, the values
given in Table 1  represent only the biological transfor-
mation of alachlor and atrazine.

The  standard  deviation of these rate  constants is rela-
tively high,  for two reasons.  First,  in  the  denitrifying
experiments  duplicate  reactors were used that con-
tained different quantities of biomass and  most likely
slightiy varying microbial populations as well. A slight
change in the relative numbers of the different microor-
ganisms present could result in the differences that were
observed in alachlor and atrazine transformation rates
among  the  different  reactors. For  the  experiments
                            involving the methanogenic and sulfate-reducing envi-
                            ronments, one reactor was  used 'or the four  experi-
                            ments. Upon complete degradation of alachlor, 1 to 2
                            weeks were allowed to pass with  no pesticides added
                            to the reactors while electron dono' and acceptor levels
                            were maintained. At this point,  alachlor again was
                            dosed to the reactors, and  the next experiment was
                            started. Over the course of  the four experiments, the
                            rate of alachlor transformation decreased considerably
                            under both methanogenic and sulfate-reducing condi-
                            tions. At the  end of the fourth expenment, no acetate
                            utilization was observed in either reactor, and no meth-
                            ane production occurred in the methanogenic reactor. At
                            this point, 2 I of fresh ground-water media was added
                            to each  of the reactors  and the  normal fill-and-draw
                            feeding was resumed, but no pesticides were added to
                            either reactor. After  2  months, no recovery of either
                            population was observed. This effect on  the microbes
                            was thought to have been a result of the builduo of
                            nonmetabolizable and toxic  alachlor or  atrazine
                            metabolites.

                            Several metabolites of alachlor were positively identi-
                            fied in these systems.  Under denitrifying conditions
                            with  resazurin   and  organisms present,  aniline,
                            m-xylene, acetyl alachlor, and diethyl aniline were
                            positively identified as products of alachlor degrada-
                            tion. Aniline, identified and quantified by gas chroma-
                            tography/mass  spectrometry  (GC/MS),  appeared
                            between Days 12 and  17 of the 45-day expenment
                            and had degraded below detection limits by the  last
                            day. At the maximum aniline concentration, 35 percent
                            of the initial  alachlor added had degraded to aniline.
                            Aniline formation and degradation constants are listed
                            in Table 2:  these rate  constants are  based  on  the
                            assumption that aniline is formed as a  direct result of
                            alachlor degradation and that biomass remains con-
                            stant throughout the experiment. Aniline formation
                            was assumed to have occurred to some maxima, at
                            which  point degradation  began.  Experiments  are
                            presently under way to study the degradation of ani-
                            line in reactors fed only this compound.  The presence
                            of  aniline in ground water as a result of alachlor
                            degradation  is possible, but the  high  rate of  aniline
Table 1.  Second-Order Degradation Constants for Alaehlor and Atrmxlne under Three Terminal Electron Acceptor Condition!

                                                 Second-Order Degradation Constant
Condi tJona
                                         Alaehlor
                                                                                  Atrazine
Denitrifying Reactor

Memanogene Reactor

Sulfate-reductng Reactor

Rasazunn

Bisulfide Ion (4)
7.9 x KT* (± 4.1 x 10"*) Umg VSS day

2.9 x 10° (± 1.6 x 1IT3) Umg VSS day

1.5 x 1(T2(t 1.4 x 10'*) Umg VSS day

5.0 x 10"* (± 5.4 x 10 *) Umg res-day

1.5 x 10° Umg VSS day
87 x 10'* ( 5.3 x 10-*) Umg VSS day

8.4 x 1Q-*L/mg VSS day

6.5 x 10"* Umg VSS day

4.2 x 10'2 ( 4.2 x 10-*) Umg res day
                                                  191

-------
Tatta 2.  Sacond-Ordaf Formation and Degradation
        Constants for Aniline in the Reactor Containing
        Both Resazurin and Denitrifying Organisms
Second-Ordar
Formation Constant
Second-Order
Degradation Constant
8.4 x lO* Umg VSSday
4.8 x 1CT3 LVmg VSS day
removal by aerobic microorganisms makes the persist-
ence of this substance for a period of longer than a few
days unlikely. Under reducing conditions in an aquifer,
however, aniline may persist for a few weeks or conju-
gate to form compounds such as diphenylamine.

In the denitrifying reactors containing resazurin  and
acetate-utilizing organisms, nvxylene, a  suspected
human carcinogen, appeared between Days 17  and
22 of the experiment and had disappeared by Day 31.
On Day 22, the highest nvxylene concentration was
present in the  reactor sample and corresponded to
approximately 9 percent of the initially fed alachlor.

m-Xylene also was detected  in an abiotic reactor con-
taining only resazurin,  atrczine, and alachlor under de-
nitrifying  conditions. On Day  45, the highest observed
m-xylene concentration was present in this reactor and
accounted for 17 percent of the initial alachlor concen-
tration. Because this compound also is readily biode-
gradable,  it is unlikely tfiat m-xylene would persist in
ground water as a result of alachlor contamination and
subsequent transformation. The role of resazurin  was
not clearly defined. Biomass growth was observed in the
reactor containing only resazunn, alachlor, and atrazine,
indicating that resazunn most likely served as an elec-
tron  donrr for organism growth. Therefore, it is unclear
whether resazurin itself or the organisms that were ca-
pable of growth on only resazurin were responsible for
the formation of m-xylene in this reactor.

One of the denitrifying reactors contained only biomass;
in this reactor,  neitfier aniline  nor m-xylene  was de-
tected. Resazurin, or perhaps some compound that fa-
cilitates electron transfer, such as vitamin B12,  may be
required for at least one step in the degradation pathway
that leads to aniline and m-xylene production.

In the methanogenic  and sulfate-reducing reactors,
diethyl aniline and acetyl alachlor were detected. Be-
cause  these  conditions are highly reducing, acetyl
alachlor is an expected product and  is likery formed
as a  result of reductive dechlorination. Acetyl alachlor
could not be quantified because the sample received
from Monsanto had evaporated to  a residue. Diethyl
aniline is a product of further microbial attack of the
ether and carbonyl groups of alachlor. At the highest
observed concentration, diethyl aniline represented
9 percent and 20 percent of  the initial alachlor added
to the system in the methanogenic and sulfate-reducing
reactors, respectively.  Two unidentified metabolites,
SM1 and SM2, accumulated in both reactors, perhaps
causing the toxicity that eventually caused the organ-
isms to stop their degradation of acetate, alachlor. and
atrazine.

Using the gas chromatograph with  both an electron
capture detector (GC/ECD) and a nitrogen-phospho-
rous detector (GC/NPD), along with the GC/MS, many
transformation products were observed in all of the
reactors yet could not be positively identified. By pre-
liminarily identifying these compounds using a spectra
library from the National Bureau of Standards on the
GC/MS, an idea of the identity of some of these prod-
ucts was gained. Some of the compounds were long,
branched,  saturated, and  unsaturated hydrocarbon
chains and were probably caused by  the breakdown
and microbial metabolism of acetate and citrate. Other
compounds appeared to be caused by the conjugation
or substitution of two or more substances. Transfor-
mation products  appeared to be formed by many dif-
ferent mechanisms, such as dealkylation or reductive
dechlorina*!on, and had widely varying concentration
profiles. Compounds  like acetyl alachlor in the denitri-
fying reactor  appeared and  disappeared  in a  few
days. Other compounds,  such as diethyl aniline and
the unknown metaoolites SM1 and  SM2 detected  in
the methanogenic and sulfate-reducing reactors, were
long-lived,  persisting in the reactor  over a period  of
weexs.

No transformation products of atrazine were identified
under  any  of the  conditions investigated.  Since
atrazine disappearance was measured in the denitri-
fying, methanogenic, and sulfate-reducing systems,
and  complete mineralization  to carbon dioxide and
water was very unlikely, metabolites should  have
been formed in these reactors. The C-18 solid-phase
extraction column used is reportedly  not very effective
at trapping  polar substances. It  is  likery that polar
transformation  products   such as  hydroxyatrazine
were produced; the polar  products probably were lost
during  sample extraction  because  only  those com-
pounds  that were extractable by the use of the C-18
column were analyzed. Their loss  is  a possible expla-
nation for the lack of detected transformation products
of atrazine. As new  solid-phase  extraction columns
are developed for effective extraction of pesticides
and their polar metabolites, more transformation prod-
ucts will be identified in these systems.

Summary and Conclusions

The speed and specific degradation  steps followed in
the transformation of alachlor and atrazine, and the
various  degradation  products that are formed  as a
result of this transformation,  are  strong functions  of
                                                 192

-------
environmental conditions, namely, the terminal electron
acceptor conditions present. In alachlor degradation,
aniline and m-xylene were products detected only in the
denitrifying reactors. On the other hand, acetyl alachlor
was identified under denitrifying,  methanogenic, and
sulfate-reducing conditions. The product formation and
transformation  patterns juring alachlor degradation
were very different in each of these systems. Analytical
limitations prevented the identification of likely pola.
products of atrazine degradation.  Further study is re-
quired to identify more of the metabolites that are formed
and to  try to formulate  a degradation pathway for
alachlor and atrazine. The electron acceptors present,
and consequently the microbial population developed in
these systems, affect the rate of herbicide transforma-
tion, the pathway that this degradation takes, and the
products that are formed that may accumulate  in the
systems. The conditions under which herbicide degra-
dation takes  place also can  result in the formation of
compounds that are human health hazards and could
be a threat to ground-water supplies.
References

1.  Somasundaram,  L, J.R.  Coats,  K.D. Racke,  and
   V.M. Shanbhag. 1391. Mobility of pesticides and their
   hydrolysis metabolites in soil.  Environ. Toxicol.
   Chem. 10:185-194.

2.  Lynch, N.L 1990. Transformation of pesticides and
   halogenated hydrocarbons in the subsurface envi-
   ronment. Ph.D. dissertation. University of  Iowa, De-
   partment of  Civil and Environmental Engineering
   (May).
3.  Mandelbaum, R.T., L.P.  Wackett,  and D.L.  Allan.
   1993. Mineralization of the s-triazine ring of atrazine
   by stable bactenal mixed  cultures. Appl. Environ.
   Microbiol. 59(6): 1,695-1,701.
4.  Wilber, G.G.  1991. Kinetics of alachlor, atrazine, and
   chloroform transformation under various electron ac-
   ceptor conditions. Ph.D.  dissertation. University of
   Iowa, Department of Civil and Environmental Engi-
   neering (August).
                                                  193

-------
         Bloremedlation of Aromatic Hydrocarbons at Seal Beach, California:
                             Laboratory and Field Investigations
                      Harold A. Ball, Gary D. Hopkins, Eva Orwin, and Martin Reinhard
                   Western Region Hazardous Substance Research Center, Stanford, CA
The objective of this study was to develop our under-
standing of processes that are important in (tie anaero-
bic  biodegradation  of  aromatic  hydrocarbons  in
contaminated ground-water aquifers. The focus of the
investigation was a site at the Seal Beach Naval Weap-
ons Station in  Southern California, where a significant
gasoline spill resulted in contamination of  the ground-
water aquifer. The project was divided into laboratory
and field components,  which  were interrelated. The
goals of the laboratory experiments were to determine
the  capability  of  the aqurfer  microbial  community to
transform aromatic hydrocarbon compounds under vari-
ous anaerobic conditions  and to understand the effect
of environmental  factors  on  the  transformation proc-
esses. Feld experiments were carried out on site at Seal
Beach.  The objectives of the  field experiments were to
evaluate potential in situ application of anaerobic biore-
mediation processes and to attempt to apply laboratory
results to the field. The results from the field experiment
will be  used to design a  remediation  proposal for the
aquifer at the Seal Beach  site.

Approach and Results

Labo story Study

In a laboratory microcosm experiment we evaluated
several factors that were rrypothesi-?ed to influence field-
scale bioremediation. Individual  monoaromatic com-
pounds (e.g., benzene, toluene, ethylbenzene, and nv,
p-, and o-xylene) were the primary substrates. To test
the influence of liquid-phase composition on the hydro-
carbon degradation potential of Seal Beach aquifer sedi-
ment, the sediment was placed in native ground water,
native ground water with nutrient amendments, and vari-
ous other laboratory media formulations including deni-
trifying,  sulfate-reciucing, and methanogenic media. In
replicate bottles during  the first 52 days of the study,
toluene  and m+p-xylene (here, m-xylene and p-xylene
were measured as a summed parameter) were biotrans-
formed in the unamended ground-water samples under
presumed sulfate-reducing conditions. Addition of ni-
trate to  the  ground water increased  rates of toluene
bfotransformation coupled to nitrate reduction, stimu-
lated biotransformatjon  of ethylbenzene,  and inhibited
the complete toss  of m-t-p-xylene that was observed
when nitrate was not added and sulfate-reducing condi-
tions prevailed. Addition of the nutrients ammonia and
phosphate had no effect on either the rate of aromatics
transformation  or the distribution  of aromatics trans-
formed.  In contrast to nitrate-amended ground water,
ethylbenzene was always transformed first followed by
toluene  in the microcosms prepared  with  denitrifying
media. In  sulfate-reducing media, lag times were in-
creased, but toluene  and  m-xylene  were  ultimately
transformed just as in the microcosms with ground water
alone. Although methane had been detected in the field,
there appeared  to be no  transformation activity in the
methanogenic microcosms during the period of the
experiment.


Blorcactor Study

A pilot-scale facility  consisting of 90-L reactors was con-
structed at the Seal Beach site. The facility was de-
signed for  the  operation of three anaerobic in situ
bioreactors. The reactors consisted of aquifer sediment
filled stainless steel cylindrical vessels with the capabil-
ity to control and monitor both hydrodynamic flow and
supplements to  the composition of the native ground-
water influent. Initial operation of the  three anoxic/an-
aerobic  reactors  focused  on   evaluating  anaerobic
bioremediation strategies for aromatic hydrocarbons un-
der existing (presumed sulfafe-reducing) and enhanced
denitrifying conditions. Biorbuctor results  were consis-
tent with the laboratory microcosm experiments. Tolu-
ene and  m+p-xylene  were degraded  in  both the
unamended and nitrate-amended bioreactors. Degrada-
tion of ethylbenzene was stimulated by nitrate addition.
Evidence indicated that benzene or o-xylene was not
                                                 194

-------
transformed  in either reactor.  The final percentage
removal efficiency appeared to be higher in the una-
mended bioreactor, where flow was slower.

Field Study

Reid experiments have been conducted to assess an-
aerobic bioremediation of a test zone within the contami-
nated aquifer at the Seal Beach  site. A network of eight
observation wells and one extraction well was installed
at the Seal Beach site. Hydrodynamic evaluation of the
well field indicated that two of the welis were satisfactory
for further experimentation. Experiments have  been
conducted using a slug test  experiment design in which
a single well  was used for the injection of the "slug* or
test pulse and the same well was used to extract the test
pulse. The results of the experiments were inferred by
differences measured in the samples  collected during
extraction. Since the native ground water contained a
variety of  electron acceptors and the water used for the
injected pulses was water that had previously  been
extracted  from the  test zone, the ground water was
treated  to control the concentration of all electron ac-
ceptors during the injection of  the test pulse.  Before
injection, the  desired salts were added back to the de-
oxygenated injection stream and  the stream metered
into the injection well. Sodium bromide was added as a
conservative  tracer. Under  this  scenario, fre different
electron acceptors investigated (e.g., nitrate and sulfate)
could be added as desired.  During initial tracer studies,
Die injection  water  was  organics  free, and thus the
source of the organics was desorption from the in situ
aquifer  solids. In subsequent and ongoing bioremedia-
tion studies, benzene, toluene, ethylbenzene, m-xylene,
and o-xylene were added with the injection  pulse at a
concentration of approximately 200 ng/L each.
The inrtial  bromide tracer data showed stable  tracer
concentrations and indicated no substantial encroach-
ment of native ground water detected in the first 0.4 pore
volumes. A very small hydraulic gradient existed at the
site, hence recovery of the bromide mass from the test
wells ranged from  93 to 99 percent with the extraction
of three pore volumes over a 103-day period. During the
tracer test, the equilibrium desorption concentrations for
the aromatic hydrocarbons when the electron acceptors
nitrate  and sulfate  were  absent from the ground water
were evaluated. Benzene, ethylbenzene, and o-xylene
concentrations remained relatively stable and 'hus ap-
peared to be at an equilibrium. The toluene and m+p-
xylene concentrations had a downward trend relative to
benzene once the native ground water encroached after
approximately 0.4 pore volumes,  suggesting that the
nitrate and sulfite concentrations available in the native
ground water supported some biological activity in the
latter part of the experiment for toluene and m+p-xylene
removal.

In a nitrate augmentation experiment, nitrate and aro-
matics were added to the injection pulse, resulting in
complete consumption of toluene and ethylbenzene fol-
lowed by m-xylene within the first 2 weeks. o-Xylene
was degraded slowly, and its concentration approached
zero by  Day 60. No apparent loss of berzene occurred
when compared with the inert tracer. The addition of
nitrate to.the test region appeared to enhance the natu-
ral  anaerobic  denitrifying  population,  confirming  the
presence of an already active nitrate-reduc'ng popula-
tion in the aquifer whose activity was enhanced by the
addition of nitrate. With the exception of o-xylene trans-
fo*mation, these results were comparable with those
from the nitrate-amended microcosm and bioreactor ex-
periments,  wherein  toluene,  ethylbenzene,  and  m-
xylene were transformed under denitrifying conditions.

During the tracer study, metfiane was detected in  the
test wells. With the encroachment of the native ground
water and associated increase in nitrate  and sulfate
concentrations,  the methane concentration decreased
to values dose  to zero,  suggesting  that nitrate  an-i
sulfate inhibit methanogenesis at this site.

Additional experiments are under way to determine
more precisely  some of the kinetic constants in  the
aquifer under denitrifying conditions and  to evaluate
rates and removal of aromatics under  sulfate-reducing
and methanogenic conditions.

Acknowledgment

Funding for this study was provided by the EPA Office
of Research and Development  under agreement
R-815738-01  through the Western Region Hazardous
Substance Posearch Center. The content of this study
does not necessarily represent the views of the agency.
Additional funding was obtiined from the Chevron Re-
search and Technology Company, Richmond, California
                                                  195

-------
               Pneumatic Fracturing To Enhance In Situ Bioremedlatlon
                                           John R. Schuring
       Northeast Hazardous Substance Research Center, New Jersey Institute of Technology, Newark, NJ

                               David S. Kosson and Shankar Venkatraman
          Department of Chemical and Biochemical Engineering,  Rutgers University, Piscataway, NJ

                                          Thomas A. Boland
       Northeast Hazardous Substance Research Center, New Jersey Institute of Technology, Newark, NJ
In situ bioremediation often is limited by the transport
rate of nutrients and electron -jcceptors (e.g., oxygen,
nitrate) to microorganisms, particularly in soil formations
with moderate to low permeability. An  investigation is
under way to integrate the process of pneumatic fractur-
ing with bioremediation to overcome these  rate limita-
tions.  Pneumatic fracturing is an innovative  technology
that uses high pressure air to create artificial fractures
in contaminated  geologic formations,  resulting  in en-
hanced air flow and transport rates in the subsurface.
The pneumatic fracturing system also can be used to
inject nutrients and other biological supplements directly
into the formation.
A project to investigate the coupling of these two tech-
nologies has been sponsored by EPA under the Super-
fund Innovative Technology Evaluation (SITE) Emerging
Technologies  Program and is scheduled for  completion
in the  summer of 1994. Laboratory and field studies are
being  carried  out simultaneously to degrade benzene,
toluene, and xylenes (BTX) in gasoline. The laboratory
studies are examining the physical and biological proc-
esses at and near the fracture interfaces, including dif-
fusion, adsorption, and  biodegradation.  Both column
and batch studies are being used to observe and quan-
tify the individual and combined effects of these proc-
esses. For  the  field portion of the studies,  a  pilot
demonstration is under way at an industrial site contami-
nated with gasoline that  is underlain by fill and  natural
claylike soils. First, a full-size prototype of the integrated
pneumatic fracturing/bioremediation system was devel-
oped. The site then was pneumatically fractured, and
periodic injections of nutrients are continuing  over a
period of  10 months.  Off-gases from the monitoring
wells  are being analyzed for BIX, oxygen, methane,
and carbon dioxide to evaluate process effectiveness.
Preliminary results from the laboratory studies and field
demonstration available  at the time of the conference
will be presented.
                                                 196

-------

-------
        •d
         CL)
 *H   O  -JH
 0   b   8  v
•tt   *>   %  >
     0*
     0>
    •d
 a

 a!
    &  .?
VH  0>
a>  u
o
a
    
 a   >  •£  a
 v?   t~\  **s  n}

-------