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}
------- |