United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/600/9-91/036
February 1992
&EPA
Bioremediation of
Hazardous Wastes
-------
-------
EPA/600/9-91/036
February 1992
BIOREMEDIATION OF HAZARDOUS WASTES
by
Biosystems 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
Printed on Recycled Paper
-------
I
Biorernediatiori of Hazardous Wast^
DISCLAIMER
Theinformationinthisdocumenthasbeen funded wholly or in part by the U.S. Environmental Protection
Agency It has been subjected to the Agency's peer and administrative review by the respective laboratories
responsible for the research presented in the authors' abstracts, and approved for publication as an EPA
document. Mention of trade names or commercial products does not constitute endorsement or recommenda-
tion for use.
-------
Bioremediatibn of Hazardous Waste
CONTENTS
EXECUTIVE SUMMARY 1
INTRODUCTION '. 5
SECTION ONE: GROUND-WATER TREATMENT
Bioventing of an Aviation Gasoline Spill: Design
and Operation of a Field Demonstration 7
Bioventing of an Aviation Gasoline Spill: Performance
Evaluation of a Field Demonstration 9
Laboratory and Field Studies of the Kinetics of Bioventing 10
SECTION TWO: TREATMENT IN A REACTOR
Treatment of Wastewater with the White Rot Fungus
Phanerochaete Chrysosporium < 13
Treatment of CERCLA Leachates by Carbon-Assisted
Anaerobic Fluidized Beds 15
Improved Prediction of GAC Capacity in a Biologically Active
Fluidized Bed 17
Treatment of TCE and Degradation Products Using
Pseudomonas Cepacia 19
Aerobic Biodegradation of Volatile Organic Compounds
inaBiofilter 21
SECTION THREE: SOIL/SEDIMENT TREATMENT
Use of Lignin-Degrading Fungi in the Remediation of
Pentachlorophenol-Contaminated Soils 23
Anaerobic Degradation of Chlorinated Aromatic Compounds 26
Influence of Nonionic Surfactants on the Anaerobic
Dechlorination of Hexachlorobenze 28
Anaerobic Degradation of Chloroaromatic Compounds under
Different Reducing Conditions 30
Para-Hydroxybenzoate as an Intermediate in the Anaerobic
Transformation of Phenol to Benzoate 31
Aerobic Degradation of Polycyclic Aromatic Hydrocarbons 33
III
-------
Bioremediatibh of Hazardous Waste
SECTION FOUR: COMBINED TREATMENT ,
Bacterial Degradation of KPEG-Modified PCBs in Anaerobic
and Aerobic Enrichment Cultures 36
Onsite Biological Pretreatment Followed by POTW Treatment
of CERCLA Leachates , .. '- 40
Aerobic Biodegradation of Creosote :........ 42
SECTION FIVE: SEQUENTIAL TREATMENT
Anaerobic/Aerobic Treatment of CERCLA Leachates in POTWs:
An Innovative Treatment Approach 45
Methanogenic Degradation Kinetics of Phenolic Compounds 47
Degradation of Naphthalene, PAHs, and Heterocyclics :. 50
Anaerobic Degradation of Highly Chlorinated Dioxins
and Dibenzofurans 52
SECTION SIX: METABOLIC PROCESS CHARACTERIZATION
Degradation of Halogenated Aliphatic Compounds by the
Ammonia-Oxidizing Bacterium Nitrosomonas Europaea 55
Degradation of Chlorinated Aromatic Compounds under
Sulfate-Reducing Conditions 57
Ring-Fission of Polycydic Aromatic Hydrocarbons by
White Rot Fungi 59
Aerobic Biodegradation of Polychlorinated Biphenyls: Genetic
and Soil Studies 60
Manipulation of TCE-Degradative Genes of Pseudomonas Cepacia 62
SECTION SEVEN: RISK ASSESSMENT
Genotoxicity Assays of Metabolites from Biological Treatment Process 65
SECTION EIGHT: BIOREMEDIATION FIELD INITIATIVE
Results of the Bioremediation Field Initiative 67
SECTION NINE: OIL SPILL BIOREMEDIATION PROJECT
Oil Spill Bioremediation Project 69
IV
-------
Bioremediation of Hazardous VVaste
EXECUTIVE SUMMARY
EPA's Office of Research and Development (ORD) initiated the Biosystems Technology Development
Program to anticipate research needs that can be applied to our nation's waste management problems. In April
1991, ORD hosted the fourth annual Symposium on Bioremediation of Hazardous Wastes: U.S. EPA's
Biosystems Technology Development Program in Falls Church, Virginia/ to discuss recent achievements of the
program and research projects aimed at bringing bioremediation into more widespread use.
At this year's conference, papers were presented in six key media-based or process-oriented research
areas and on three current programs.
1. Ground-Water Treatment. Effective bioremediation of ground water is constrained by the geology,
hydrology, and geochemistry of the subsurface environment. Design of any effective remedial action
for contaminated ground water requires a greatdeal of site-specific information. Current ground-water
research is focused on enhancement of degradation using bioventing to treat gasoline-contaminated
vadose zones.
2. Treatment in a Reactor. In reactors, hazardous pollutants are brought into contact with microorganism
to accelerate the degradation process. Landfill leachates are a good example of a liquid waste that is
amenable to reactor treatment. In addition, toxic waste from Superfund sites can be treated in reactors.
EPA researchers are exploring use of reactors at publicly owned treatment works (POTWs) as a model
for such reactors.
3. Soil/Sediment Treatment. Decontamination of soils and sediments is one of the most difficult
problems found at hazardous waste sites. This is due to the cost involved, heterogeneity of the media,
and adequate transfer of amendments (e.g., oxygen nitrates).
4. Combined Treatment. Most hazardous waste sites contain complex mixtures of biologically persistent
organic and inorganic contaminants that can be remediated only by a combination of treatment
techniques. EPA researchers are developing methods to combine various physical, chemical, and
biological treatment technologies, and comparing the effectiveness of the various combinations.
5. Sequential Treatment. Sequential treatment is generally applied to two waste types: compounds that
degrade into stable intermediates that can be further degraded under different conditions than those
used for the parent compound; and complex mixtures of wastes, which are generally degraded in order
of their thermodynamic behavior. EPA researchers are investigating the most effective coupling and
sequence of treatments for such wastes.
6. Metabolic Process Characterization. EPA's metabolic process research generates a better understand-
ing of the processes by which microorganisms degrade chemicals, expanding the range of organisms
that can be used in biosystems technologies. Based on the insights gained from this research, scientists
can then choose indigenous organisms or enhanced organisms to meet needs in pollution cleanup and
control.
7. Risk Assessment. A number of the high-priority compounds that require disposal are known
carcinogens or precarcinogens. Since biodegradation does not necessarily result in total degradation
to carbon dioxide and water, researchers need to assess whether ultimate or procarcinogens are created
by a given biological treatment. Public health evaluations must be conducted to determine the toxicity
of substances at Superfund sites, assess the safety of nonindigenous organisms, and compare
bioremediation with other potential technologies.
-------
Bioremediation of Hazardous Waste
8. Bioremediation Field Initiative. As part of its overall strategy to increase the use of bioremediation
to treat hazardous wastes, EPA implemented the Bioremediation Field Initiative. This program assists
the regions and the states in conducting field tests and evaluations of this technology. At more than 140
sites in the United States, treatability studies are being conducted and bioremediation is being planned,
is in full-scale operation, or has been completed.
9. Oil Spill Bioremediation Project. EPA's Oil Spill Bioremediation Project in Prince William Sound,
Alaska, examined whether the addition of nutrients to oil-contaminated beaches would sufficiently
enhance oil degradation rates to enhance biodegradation. The success of this project demonstrated that
bioremediation should be considered as a key component in any cleanup strategy for future oil spills
impacting the shoreline.
In pursuing research in these areas, the Biosystems Technology Development Program has identified
a number of avenues through which to enhance the use of this technology. These are:
• Process characterization. Isolateandidentifymicroorganismsthatcarryoutbiodegradationprocesses.
Search out and characterize biodegradation processes in surface waters, sediments, soils, and subsur-
face materials in order to identify those that may be used in biological treatment systems and develop
process-based mathematical models to evaluate potential treatment scenarios.
• Process development. Develop newbiosystems for treatment of environmental pollutants. Biosystems
would include naturally selected microorganisms, consortia, bioproducts, and genetically engineered
microorganisms. i
• Process engineering. Determine, evaluate, optimize, and demonstrate the engineering factors neces-
sary for applying biological agents to detoxify or destroy pollutants in situ or at a centralized treatment
facility.
• Environmental risk. Determine environmental fate and effects of, as well as risks involved in the use
or release of, degrading microorganisms or their products to detoxify or destroy pollutants.
• Mitigation of adverse consequences. Develop means to mitigate adverse consequences resulting from
the accidental or deliberate release of microorganisms for pollution control.
• Technology transfer. Transfer information on advances in the technology to the user community.
Provide evaluation of full-scale projects and a central repository of field projects.
The Biosystems Technology Development Programs draws on ORD scientists who possess unique skills
and expertise in biodegradation, toxicology, engineering, modeling, biological and analytical chemistry, and
molecular biology. Participating laboratories and organizations are:
Environmental Research Laboratory—Ada, Oklahoma
Environmental Research Laboratory—Athens, Georgia
Environmental Research Laboratory—Gulf Breeze, Florida
Health Effects Research Laboratory—Research Triangle Park, North Carolina
Risk Reduction Engineering Laboratory—Cincinnati, Ohio
Center for Environmental Research Information—Cincinnati, Ohio
As hazardous wastes become increasingly more diverse with respect to contaminants and contaminant
mixtures/ our nation will need to rely more arid more on innovative technologies for improved treatment
-------
Bioremediation of Hazardous Waste
efficiency at lower costs. In February 1990, EPA Administrator William Reilly held a meeting with over 70
representatives of biotreatment companies, contractors, environmental organizations, academia, and other
federal agencies. The purpose of the meeting was to develop an agenda outlining strategies for increasing the
use of bioremediation for cleaning up hazardous waste sites and petroleum products. As a result, ORD together
with the Office of Solid Waste and Emergency Response (OSWER) launched a new Bioremediation Field
Initiative with three primary goals to be carried out over the following two years.
The first goal is to more fully document the performance of full-scale bioremediation field applications.
OSWER and ORD will evaluate treatment effectiveness, operational reliability, and costs at both in situ and ex
situ bioremediation projects, focusing on in situ biological treatment for surface and subsurface contamination.
The second area covered by the initiative is to provide technical assistance to EPA and states overseeing
bioremediation projects or considering the use of bioremediation. ORD Technical Support Centers in Ada,
Oklahoma, and Cincinnati, Ohio, will provide assistance with site characterization, treatability study design, or
the interpretation of results. The third part involves the creation of a data base, which will be a central repository
of current data on progress in the field in determining the treatability of various contaminants.
To date,,the Bioremediation Field Initiative has identified over 140 sites across the country where
bioremediation projects are being considered, planned, or are currently underway. These sites include
Comprehensive Emergency Response, Compensation, and Liability Act (CERCLA), Resource Conservation
and Recovery Act (RCRA), and Underground Storage Tank (UST) sites. Approximately one-third of these
projects are in the planning stages, one-third are undergoing or have completed treatability studies, and one-
third are being designed or implemented. Data gathered from the initiative indicates that bioremediation is
being undertaken for three major waste categories: petroleum, creosote, and solvents. These three types of
waste comprise about two-thirds of wastes being biologically remediated.
According to data gathered by the Bioremediation Field Initiative, soil alone and soil and ground water
together are the media most often treated with bioremediation. Bioremediation is less often used to treat ground
water only; sediments,, and surface water. The most frequently applied bioremediation technique'is in situ
treatment (over 80 sites), followed by land treatment and liquid treatment in a reactor (over 60 sites each).
The Field Initiative is also considering four sites for performance evaluations: 1) a creosote site, 2) a site
contaminated with trichloroethylene, dichloroethylene, and vinyl chloride, 3) and ethylene glycol site, and 4)
an underground storage tank.
-------
-------
Bioremediation of Hazardous Waste
INTRODUCTION
The US. Environmental Protection Agency (EPA) is responsible for protecting public health and the
environment from the adverse effects of pollutants. EPA's authority to develop regulations and to conduct
environmental health research is derived from major federal laws passed over the last 20 years that mandate
broad programs to protect public health and the environment. Each law—including the Clean Air Act, the Safe
Drinking Water Act, the Clean Water Act, the Toxic Substances Control Act, the Federal Insecticide, Fungicide,
and Rodenticide Act, the Resource Conservation and Recovery Act, and the Comprehensive Environmental
Response, Compensation, and Liability Act (CERCLA, known as Superfund)—requires that EPA develop
regulatory programs to protect public health and the environment.
For the control and cleanup of hazardous wastes, the Superfund law gives EPA broad authority to
respond directly to releases of hazardous materials that endanger public health or the environment. Also, the
Superfund Amendments and Reauthorization Act of 1980 (SARA) expands EPA's authority in research and
development, training, health asssessments, community right-to-know, and public participation. EPA's Office
of Research and Development (ORD) conducts basic and applied research in health and ecological effects,
hazardous wastes, and remediation development and demonstration of control technologies. Technologies are
designed to provide efficient, cost-effective alternatives for cleaning up the complex mixtures of pollutants
found at Superfund sites or at other locations, such as oil spills. As the technologies advance, ORD transfers
information on their use and enhancement to groups that apply technologies at specific sites.
Some of the most promising new technologies for solving hazardous waste problems involve the use of
biological treatment systems. Biological treatment uses microorganisms, such as bacteria or fungi, to transform
harmful chemicals into less toxic or nontoxic compounds. These microorganisms can break down pollutants to
obtain energy to live and reproduce. They have a wide range of abilities to metaboloize different chemicals;
scientists can tailor technology to the pollutants at specific sites and in specific media (e.g., contaminated
aquifers, waste lagoons, contaminated soils) by using an organism in the treatment system that breaks down a
particular pollutant. Where possible, technologies are developed to utilize native microorganisms that have
been demonstrated to" metabolize the pollutants on the site. In other cases, organisms known to metabolize the
pollutants can be introduced and supplemented if necessary to accelerate biodegradation.
Biodegradation is an attractive option because it is "natural," and the residues from the biological
processes (such as carbon dioxide and water) are usually geochemically cycled in the environment as harmless
products. These processes are also carefully monitored to reduce the possibility of a product of a process being
more toxic than the original pollutant. Another advantage of biological treatments—particularly in situ
treatment of soils, sludges, and ground water—is that they can be less expensive and less disruptive than options
frequently used to remediate hazardous wastes, such as excavation followed by incineration or landfilling.
Other methods of applying biological treatments, such as spreading contaminated soils on controlled land plots
or mixing sludges with water for treatment in contained vessels, are currently used with varying degrees of
success/depending on the chemical contaminants and environmental conditions. Additional research in such
technologies will broaden their applicability and effectiveness. Finally, bioremediation holds another clear
advantage over many technologies relying on physical or chemical processes: instead of merely transferring
contaminants from one medium to another, biological treatment can degrade the target chemical.
Perhaps one of the most well-known recent application of bioremediation technology was in cleaning up
portions of the shoreline of Prince William Sound, Alaska, in the wake of the March 1989 Exxon Valdez tanker
accident. This project demonstrated EPA's ability to work with private industry and academia to develop and
transfer expertise in handling hazardous wastes. The Federal Technology Transfer Act of1986 (FTTA) provided
a mechanism whereby industry and EPA could share research costs and results.
-------
Bioremediatioh of Hazardous Waste
The FTTA encourages the development of commercial pollution control technologies by making possible
cooperative research and development among federal laboratories, industry, and academic institutions. Under
the FTTA, EPA and industry can cooperate in developing and marketing biological treatment technologies in
any of the six key areas of biosystems research identified by ORD's Biosystems Technology Development
Program: ground-water treatment, liquid reactors, soil/sediment treatment, combined treatment, sequential
treatment, and metabolic processes research.
This document is divided into six sections based on research performed in these media-based or process-
oriented areas. Each section contains an introduction to the specific technology and summary of recent research,
followed by abstracts of EPA projects presented at the April 1991 Symposium on Bioremediation of Hazardous
Wastes.
-------
SECTION ONE
GROUND-WATER TREATMENT
Effective bioremediation of ground water is limited by the geology, hydrology, and geochemistry of the subsurface
environment. Engineers must control the hydrology of the system to direct the biological treatment to the area of
contamination. They also must understand the geochemistry of the site to avoid mineral precipitation and the subsequent
plugging of aquifers. Thus, the design of any effective remedial action for contaminated ground water requires a great deal
of site-specific information.
Because oxygen has limited solubility in water, contaminated ground water is frequently anoxic. This problem can
be offset by artificially supplying oxygen to the pollutant plume so that aerobic degradation is not limited by a poor oxygen
supply. Nutrients can also be added to make environmental conditions suitable for microbial activity. Oxygen can be
delivered to ground water by withdrawing the ground water, adding oxygen, and reinjecting the water using injection
wells and trenches.
.In a series of research projects managed by the U.S. EPA R.S. Ken Environmental Research Laboratory, the U.S.
Coast Guard in Traverse City, Michigan, has been evaluating the biodegradation of hydrocarbon-contaminated vapors
within the unsaturated zone using bioventing. Bioventing is the engineered advection of air blown through contaminated
soil to promote aerobic degradation. The Bioventing Reclamation Pilot Study is being performed on subsurface contami-
nation resulting from an aviation gasoline spill of about 35,000 gallons that occurred in 1969. One study area is using
injected aeration only; the other employs injection as well as extraction/reinjection. Fuel hydrocarbons were reduced by
about40percentinbothtreatmentareasdunngtheftrst3monthsofoperation.Laboratoryandfieldstudiesofkineticsun^
natural and engineered advective conditions are helping EPA researchers determine the optimal conditions for bioventing.
BIOVENTING OF AN AVIATION
GASOLINE SPILL:
DESIGN AND OPERATION OF A
FIELD DEMONSTRATION
John M. Armstrong, Ph.D. and Christopher
]. Griffin, P.E., The Traverse Group, Inc.,
Ann Arbor, MI.
The Bioventing Reclamation Pilot Study is de-
signed to evaluate the biodegradation of
hydrocarbon-contaminated vapors within the un-
saturated zone during induced volatilization. This
study is being conducted at the U.S. Coast Guard
Air Station in Traverse City, Michigan, which is the
site of a spill of about 35,000 gallons of aviation
gasoline which occurred in 1969. After 20 years, a
major portion of the spill still persists in the subsur-
face as a plume that is about 1,100 feet long and 250
feet wide. This study is being conducted as a coop-
erative effort between the U.S. Coast Guard and the
U.S. Environmental Protection Agency's Robert S.
Kerr, Environmental Research Laboratory.
The subsurface conditions at the site consist
of a uniform beach sand extending to depths of
about 50 feet, underlain by a gray glacial silty clay.
The water table is located at a nominal depth of
about 15 feet below the ground surface, but over the
past 6 years the water table elevation has fluctuated
6 to 8 feet.
The 90- by 75-foot study area has been di-
vided into two equal areas of 45 by 75 feet to evaluate
the effects of different flows and extraction patterns.
The northern area has an injection system, while the
southern area has an injection and extraction/
reinjection system. The pneumatic properties of the
-------
Ground-Water Treatment
unsaturated zone were evaluated by the perfor-
mance of a pneumatic pump test, resulting in a
design radius of influence of 10 feet. The work plan
calls for ambient air to be injected into both areas at
an initial rate that would replace the calculated
volume of air-filled pore space over 24 hours. The
flow rate would be increased to a vapor recharge
rate of 8 hours or less as the system becomes accli-
mated.
The blower package, therefore, has to be ca-
pable of extracting vapors in the south study area, at
depths of 15 to 18 feet (depth of the water table) and
flow rates ranging from 5 to 63 cubic feet per minute
(cf m), then reinjecting the vapors at the same rate, at
a depth of 10 feet. Additionally, the system has to be
able to inject ambient air at the same flow rate
within both the extraction/reinjection plot (south
area) and the air injection plot (north area). Accord-
ingly/because theambientairinjected will be placed
in twice the area (two test plots), the blower has to
be able to inject air at flow rates ranging from 10 to
128 cfm.
The construction of the Bioventing Project con-
sisted of installing, in the north area, 15 aeration
injection points placed on 10-foot centers in a 3- by
5-foot grid and screened just above the water table.
In the south area, eight sets of injection points
coupled with seven extraction points, 10 feet on
center, were installed, with screens placed justabove
the water table. Eight reinjection wells were in-
stalled, with the screens placed at a depth of 10 feet.
The blower package used is a 45 URAI Roots
vacuum pump with a maximum flow rate of 130
cfm. This system extracts vapors at a vacuum of 4 to
6 inches of mercury and reinjects the vapors at a
pressure of 6 pounds per square inch (psi). The
vacuum pump is driven by a 10 HP 3-phase electric
motor. Similarly, an additional 45 URAI Roots
pump, also with a maximum flow rate of 130 cfm, is
used for the ambient air injection. This blower is
driven by a 7.5 HP 3-phase electric motor at an
operating pressure of 6 psi. All the equipment is
explosion proof.
The moni toring requirements of the EPA work
plan called for the installation of several different
types and depths of monitoring equipment and/or
sample points. To monitor vapor hydrocarbon and
oxygen concentrations, six5-point cluster wells were
installed with three cluster wells per plot. The
cluster wells consisted of 1/4-inch diameter tubing
with a wire mesh screen covering the tip. The five
points of each cluster well were installed at 3.28-foot
(1-meter) depth increments throughout the unsat-
urated zone. Additionally, we installed three
14-point cluster monitoring wells (well screens at
1.5-foot intervals from ground surface to 21 feet—
one per plot and one at an upgradient location) and
one set of moisture/temperature probes per plot.
The moisture/temperature probes are Soil Test Se-
ries 300 moisture/temperature cells consisting of
thermistor soil cells buried at depths of 5,10, and 15
feet below grade.
The development of a sufficient microbial
population to degrade the hydrocarbon vapors re-
quires adequate quantities of nitrogen, phosphorous,
and potassium. The EPA Bioventing Work Plan
called for an initial application of 64 pounds of
nitrogen, 13 pounds of phosphorus, and 5 pounds of
potassium to be applied to each area prior to startup.
Additionally, during the growing season, 10 pounds
of nitrogen, 2 pounds of phosphorous, and 1 pound
of potassium were to be applied to each area monthly.
These nutrients were applied in an aqueous solu-
tion by sprinklers until they were detected in the
ground water, indicating that they had moved com-
pletely through the treatment zone.
The Bioventing Project is sampled and/or
monitored daily, biweekly, and monthly. Daily
monitoring consists of measuring the blower's op-
erating parameters, such as flow rate, pressure, and
vapor temperature. Combustible gas concentration
within the vapor reinjection flow line is determined
daily withaBacharachThresholdLimit Value (TLV)
combustible gas meter.
Biweekly monitoring includes determining the
combustible gas and oxygen concentration within
the three 5-point cluster wells located in each plot.
The combustible gas concentration is determined
using the TLV gas meter, and the oxygen concentra-
tion is determined using a Bacharach Oxygen
indicator. Additionally, the soil moisture content
and soil temperature are measured biweekly in the
moisture/temperature probes.
8
-------
Gro'uird-Water Treatment
The surface emissions are sampled at two loca-
tions within each of the study areas biweekly and at
two upgradient locations weekly. The samples,
taken over a 4-hour period, are pulled using an
Ismatec peristalic pump set for a flow of approxi-
mately one liter per hour. A 19-inch diameter
stainless steel bowl having a volume of 4.3 gallons
(16 liters) is inverted and placed flush on the ground.
The sample is pulled from the bowl through flexible
vinyl tubing that is attached to the bowl by a 174-
inch diameter steel ball valve tapped into the bottom
of the bowl. Any water that collects within the
emission chamber is removed by a water trap, lo-
cated upstream from the sample trap, consisting of
a flask containing a drying agent (Dried te).
Water quality data are obtained by sampling at
two depths in each of the 14-point monitoring wells.
The water samples are analyzed for nutrients and
BTEX.
BIOVENTING OF AN AVIATION
GASOLINE SPILL:
PERFORMANCE EVALUATION OF A
FIELD DEMONSTRATION
; Don Kampbell, U.S. Environmental
Protection Agency, Ada, OK.
A spill of about 35,000 gallons of aviation
gasoline occurred in 1969 at a U.S. Coast Guard air
station. Much of the spill persists after 22 years as an
oily phase residue at the water table near a depth of
5 meters. The subsurface matrix is a fairly uniform
beach sand to 15 meters.
Aerobic soil microcosms were used in the labo-
ratory to simulate the ability of the spill-site soil to
biodegrade aviation gasoline vapors. Reaction rates
with acclimated microcosms were rapid, with dis-
appearance curves showing typical first-order
kinetics. Degradation rates within a temperature
range of 12 to 23° C were high. A nutrient addition
of ammonia, nitrate, phosphorus, and potassium
increased by several fold the bacteria count, degra-
dation rate, and active biomass. A suppressive
effect was shown on degradation of dimethyl and
trimethyl pentanes mixtures when compared to
singular components. A Lineweaver-Burk recipro-
cal plot was used to calculate a biochemical reaction
kinetic maximum velocity value of 5.7 mg fuel/kg
soil-hour and half saturation constant of 7.0 mg
fuel/kg soil for aviation gasoline vapors. Extrapo-
lation of the rates to field conditions suggests
consumption of the gasoline vapors during
bioventing within 8 hours in the unsaturated zone.
The pilot demonstration systems have been
operational slightly more than 3 months. One area
has injected aeration only; the second has injection,
extraction, and reinjection. Turf was established to
cover both treatment areas. A nutrient solution of
nitrogen and phosphorus was dispersed through-
out the unsaturated zone. Subsurface flow
characteristics will be defined with a sulphur
hexafluoride tracer test. Core material, soil gas, and
underground water are being analyzed to deter-
mine the extent of remediation. Core material fuel
hydrocarbons have been reduced about 40 percent
in both treatment areas during the first 3 months of
operation. Surface emissions have been less than 1
percent of total volatile hydrocarbons detected in
the soil gas aeration stream at a 1-meter depth.
Objectives of the project are to demonstrate that:
• Remediation will be completed in a reason-
able time.
• Surface emissions of gasoline do not occur.
• Remediated core material will be <10 mg
fuel carbon/kg.
• Final benzene levels in the ground water
will not exceed 5 |4./g liter.
• Performance and economical advantages
will be applicable to full-scale remediation.
The project described in this presentation is
jointly funded by the U.S. Environmental Protection
Agency and the U.S. Coast Guard. The work de-
scribed is in the initial phase of data compilation.
Therefore, the contents have not been subjected to
the Agency's review policy and no official endorse-
ment should be inferred.
-------
Ground-Water Treatment
LABORATORY AND FIELD STUDIES OF
THE KINETICS OF BIOVENTING
David W. Ostendorf,AssociateProfessor, Civil
Engineering Department, University of Mas-
sachusetts, Amherst, MA; Don H. Kampbell,
Research Chemist, R.S. Kerr Environmental
ResearchLaboratory, U.S. EPA, Ada, OK; and
Ellen E. Moyer, Research Assistant, Civil En-
gineering Department, University of
Massachusetts, Amherst, MA.
The interaction of laboratory and field investi-
gations of biodegraded hydrocarbon vapors are
discussed under naturally diffusive and engineered
advective conditions.
The coupled transport of aviation gasoline
and oxygen vapors has been measured and mod-
eled at the U.S. Coast Guard Air Station in Traverse
City, Michigan, as part of a series of research projects
managed by the R.S. Kerr Environmental Research
Laboratory of the U.S. EPA. The data consist of a
group of stainless steel tubing clusters set at 1-meter
depth increments over the 5-meter thick unsatur-
ated zone in the uniform sand at the site. The
clusters have been sampled over a 13-month period
with hydrocarbon and oxygen meters, calibrated
againstknownheadspacestandard gases. Themoclel
is a time averaged steady balance of diffusion and
Michaelis-Menton kinetics, coupled stoichiometri-
cally under the assumption of abundant oxygen.
An implicit solution is put forth by Ostendorf and
Kampbell:
Z_{DK,
** "" V **T » '
1/2 (H}
! K
(diffusion)
(1)
with elevation z above the contaminated cap-
illary fringe, soil moisture diffusivity D, half
saturation constant K, maximum reaction rate V,
and hydrocarbon concentration H (1). The integral
function I(H/K) is evaluated by Ostendorf and
Kampbell (1), and the corresponding oxygen con-
centration follows from the stoichiometry of the
reaction. This natural diffusion model is calibrated
with the field soil gas data at four clusters, yielding
the following kinetics:
Unacclimated sandy soil
V = 8.6 x 10'9 kg/m3-s
K = 0.10mg/L
Bioventing is the engineered advection of air
blown through contaminated soil, subject to aerobic
biodegradation. Ostendorf and Kampbell model
this process as a balance of advection and Michaelis-
Menton kinetics, yielding a simple prediction for
the effluent hydrocarbon vapor concentration. HE:
K
H
HfHE
K
(advection) (2)
with pneumatic residence time t and influent con-
centration Hj (2). Thus the reaction kinetics play a
major role in determining the removal efficiency of
the bioventing reactor. The foregoing theory de-
scribes field data from an acclimated clay soil
bioreactor treating a propane/butane waste gas
mixture at Racine, Wisconsin (3).
Ostendorf and Kampbell apply laboratory
microcosm data to the diffusion and advection trans-
port models cited above (2). Aseptic soil samples
from the Wisconsin and Michigan sites were dosed
with appropriate gases in headspace vials, and the
subsequent decay of concentration was measured
in a gas chromatograph (subject to abiotic control).
The temporal decay of the initial concentration H0 is
described by a balance of storage, headspace sorp-
tion, and Michaelis-Menton kinetics:
t=
V
K
H-H
K
(microcosm) (3)
with retardation factor RD. The microcosm-based
kinetics for Traverse City agree fairly well with the
field calibrated values listed above. The Wisconsin
soil kinetics, obtained from the microcosms, are:
Acclimated clay soil
V = 4.1xl(T5kg/m3-s
K = 0.088 mg/L
10
-------
Grpuhd-Water Treatment
These values, when substituted into the
bioventing advection model (Equation 2) yield a 92
percent predicted removal rate that compares fa-
vorably with the observed range of 90 to 99 percent
(3). We conclude that laboratory microcosms yield
kinetics that are consistent with field values, imply-
ing no transport limitations on the field or microcosm
scale in unsaturated soil.
We note the close correspondence of half
saturation constants for two dramatically different
soil types. The wide range of maximum reaction
rates may be due to biomass variation in clays and
sands, and also suggests that biostimulation may be
possible in natural soils. Ostendorf and Kampbell
explored the latter possibility by running a series of
microcosm studies for stimulated soil samples at
Traverse City with the results (2):
Acclimated sand with nutrients
V = 2.8xl(r6kg/m3-s
K = 0.25mg/L
These kinetics are input to a series of simula-
tions of bioventing effectiveness at the site. The
optimal conditions, consisting of a high concentra-
tion influent and a biostimulated soil, result in very
effective bioventing, with complete removal of prod-
uct in about 1 month of system operation. Natural
kinetics are too slow for effective bioventing at
Traverse City, and an auxiliary exhaust' air treat-
ment process would have to be considered at the
site.
References
1. Ostendorf, D.W. and D.H. Kampbell. 1991.
Biodegradation of hydrocarbons in the un-
saturated zone. Water Resources Research.
(In press)
2. Ostendorf, D.W. and D.H: Kampbell. 1990.
Bioremediated soil venting of light hydro-
carbons. Hazardous Waste and Hazardous
Materials 7:319.
3. Kampbell, D.H., J.T. Wilson, H.W. Read,
and T.T. Stocksdale. 1987. Removal of
volatile aliphatic hydrocarbons in a soil
bioreactor. Journal Air Pollution Control
Association 37:1236.
11
-------
-------
SECTION TWO
TREATMENT IN A REACTOR
Work is ongoing for treatment of liquid and gaseous waste streams in a reactor. In reactors, pollutants are brought
mtocontactwithmicrowganismstoacceleratethedegradationprocess.Landfillleachateisagoodexampleofatype
waste that is amenable to reactor treatment. Over 130 million tons of solid waste are disposed ofeachyear in landfills across
the nation, and various organic chemicals and ions often leach from the waste.
A current laboratory-scale project is exploring the use of white rot fungus (Phanerochaete chrusosporium) in a
reactor to degrade organic pollutants in wastewater. This fungus has been found to decolorize the bleachplant effluent that
results from the manufacture of white paper. It can also dechlorinate the organic chlorine compounds in the effluent, and
can transform high molecular weight compounds into lower molecular weight compounds. It may be possible to use
immobilized P. chrysosporium as a pretreatment for a variety ofwastewaters prior to conventional secondary treatment.
In recent years, granular activated carbon (GAC) has received increased attention for removing synthetic organic
compounds (SOCs)from water. One study tested two anaerobic CAC expanded-bed bioreactors as pretreatment units for
decontaminating hazardous leachates. One leachate was representative of those from old, stabilized waste landfill sites; the
second was typical of younger, more active sites. In both reactors, the majority of the SOCs were removed by biological
activity. Another study investigated the effects ofmolecular oxygen on theadsorptive capacity of GAC for different organic
compounds found in wastewater.
Other research is exploring the treatment of ground water containing trichloroethylene (TCE) and degradation
• products in bioreactors using Pseudomonas cepacia Strain G4. Another recent EPA study analyzed'the biodegradation
of three volatile organic compounds (VOCs) in an aerobic biofilter containing acclimated biomass. Nearly 100 percent
removal of the three compounds was achieved.
TREATMENT OF WASTEWATER
WITH THE WHITE ROT
FUNGUS PHANEROCHAETE
CHRYSOSPORIUM
Thomas W. Joyce and Hou-min Chang,
Department of Wood and Paper Science, North
Carolina State University, Raleigh, NC., and
John A. Closer, U.S. EPA, Risk Reduction
Engineering Laboratory, Cincinnati, OH.
The purpose of this summary is to provide an
overview of the use of the white rot fungus
Phanerochaete chrysosporium in the treatment of aque-
ous wastewaters, with emphasis on wastewaters
from paper manufacture bleach plants.
White rot fungi, and in particular P.
chrysosporium, are able to degrade a wide variety of
organic pollutants (I, 2, 3). Those studies were
extensions of research that found that the fungus
was able to decolorize the bleach plant effluent
resulting from the manufacture of white paper (4).
As the chromophoric material in the effluent was
destroyed, the organic chlorine-containing com-
pounds were also degraded ,(5). The degradation is
believed to occur as the result of secondary meta-
bolic activity, i.e., the fungus cannot utilize the
substrate' for carbon or energy purposes (6). The
degradation of pollutants is now known to be
mediated by a family of enzymes excreted by the
fungus (7).
13
-------
Treatment in a Reactor
The initial treatability studies on bleach plant
effluent were done using 125-mL shake flasks. After
evaluation of numerous reactor designs, it was de-
termined that a rotating biological contactor (RBC)
was best suited to the physiological needs of the
fungus. Our standard laboratory-scale reactor has a
capacity of about 2.1 L. It contains eight partially
wetted (40 percent) disks rotating at 1 rpm (8). It has
also beenfoundthatahigher-than-atmospheric oxy-
gen tension promotes organic pollutant destruction.
Typically, the fungus is grown at 37°C; after sub-
stantial mycelial growth, the temperature is lowered
to 30°C for enzyme production (9). The pH is main-
tained in the range of 3 to 5. After 2 to 4 days of fungal
growth/ the nitrogen is removed from the waste-
water and the fungus soon enters secondary
metabolism, when the enzymes actually responsible
for organic pollutantdestruction are excreted. Other
researchers have found that the fungus can be im-
mobilized on porous plastic (10) or in alginate (11),
or that the enzymes themselves can be immobilized
and remain viable (12).
With respect to bleach plant effluents, the use
of P. chrysosporium may be particularly advanta-
geous. Not only does the fungus reduce the color of
the effluent but it also can dechlorinate the organic
chlorine compounds as measured by AOX. At
present, typical biological treatment systems, aer-
ated lagoons and activated sludge, do not remove
substantial amounts of color or AOX from the efflu-
ent. The immobilized fungi can remove some 70
percentofthecolorandaboutSOpercentoftheAOX.
We have found that the high molecular weight or-
ganic compounds in bleach plant effluent can be
transformed into lower molecular weight com-
pounds that exhibit more toxicity as measured by
the MICROTOX bioassay (13).
We believe the most logical use for the immo-
bilized fungusmay beina two-stage system whereby
the fungus is used as a pretreatment prior to conven-
tional secondary treatment (14). The fungus would
thus be used to dechlorinate the AOX compounds
and reduce the average molecular weight of the
remaining lignin compounds. The conventional
secondary treatment could then remove a signifi-
cant portion of the remaining organic compounds.
In some cases, it may be economically advan-
tageous to concentrate by ultrafiltration the effluent
to be treated by the fungus. We have found that the
rate of color removal is generally proportional to
color concentration. In one example case, the cost of
fungal treatment for a bleach plant effluent was
$12.37/ton pulp; when ultrafiltration was added to
the treatment flowsheet, the cost decreased to $4.46/
ton pulp despite the added cost of the UF system
(15).
To date, the immobilized fungus has not been
tested at even pilot plant scale. We believe, how-
ever, that it can be effectively integrated into an
existing treatment scheme to pretreata wide variety
of wastewaters to make them more amenable to
conventional treatment.
References
1. Bumpus, J., et al. 1985. Science 228:1434.
2. Messner,K.,etal.l988.ForumMikrobiologie
11:492.
3. Pellinen, J., et al. 1988. J. Biotech. 8:67.
4. Eaton, D., et al. 1982. TAPPI J. 65:89.
5. Huynh, B., et al. 1985. TAPPI J. 68:98.
6. Kirk, T. and R. Farrell. 1987. Ann^ Rev.
Microbiolol. 41:465.
7. Kirk,T.,etal. 1986.EnzymeMicrob.Technoi.
8:27.
8. Chang, H-m., et al. 1985. U.S. Patent
4,554,075. ;
9. Asther, M., et al. 1988. Appl. Environ.
Microbiol. 54:3194.
10. Messner, K., et al. 1988. S. Eur. Patent
Al-0 286 630.
11. Livernoche, D., et al. 1983. Biotech. Bioeng.
25:2055.
14
-------
Treatment in a Reactor
12. Presnell, T., et al. 1990. Proc. Cellucon '90,
Brastislava, Czechoslovakia, 28-31 August.
13. Fukui, H., et al. 1991. Proc. Symp. Cellulose
and Lignocellulosic Chem., Guangzhou,
China, 13-15 May.
14. Yin, C-f., et al. 1989. J. Biotech. 10:77.
15. Yin, C-f., et al. 1990. Proc. 24th EUCEPA
Conf., Stockholm, Sweden, 8-11 May.
TREATMENT OF CERCLA LEACHATES
BY CARBON-ASSISTED
ANAEROBIC FLUID1ZED BEDS
M.T. Suidan, R. Nath, and A.T. Schweder,
Department of Civil and Environmental
Engineering, University of Cincinnati, Cincin-
nati, OH; E.R. Krishnan, PEI Associates, Inc.,
Cincinnati, OH; and R.C. Brenner, U.S. Envi-
ronmental Protection Agency, Cincinnati, OH.
Rainfall arid surface runoff percolating through
landfill are contaminated with a number of organic
and inorganic compounds, and direct discharge of
exiting leachate to municipal wastewater treatment
plants can result in inadequate removal of many
hazardous substances. Many volatile and
semivolatile synthetic organic chemicals (SOCs) used
as solvents, degreasers, and components in indus-
trial products are present in leachates originating
from hazardous waste sites. These SOCs are often
inadequately treated in aerobic wastewater treat-
ment processes; as volatiles are subject to air
stripping, many semivolatiles simply pass through
untreated, and highly chlorinated compounds are
difficult to degrade aerobically.
In this study, two anaerobic granular activated
carbon (GAC) expanded-bed bioreactors were tested
as pretreatment units for the decontamination of
hazardous leachates. Two municipal leachates, ren-
dered hazardous with the addition of 10 volatile and
four semivolatile organic compounds commonly
found in leachates, were fed to two identical bench-
scale (10.2cm diameter x 96.5 cm high) expanded-bed
reactors. One leachate, with a chemical oxygen
demand (COD) of approximately 1,100 mg/L, was
representative of those that emanate from old, sta-
bilized waste landfill sites where only a smallfraction
of the leachate COD is made up of volatile fatty
acids. Sulfate concentrations in this leachate aver-
aged 89 mg SO4/L. The second leachate, with a
COD of approximately 3,800 mg/L, was typical of
younger, more active landfill sites, where the ma-
jority of the COD is attributable to volatile fatty
acids. This leachate had a very low sulfate content.
The different characteristics of the two leachate feed
streams resulted in one reactor operating in a sulf ate-
reducing mode and the second in a strictly
methanogenic environment. Both reactors were
operated with a 6-hour unexpended empty-bed
contact time and achieved SOC removals acceptable
for pretreatment units. In both reactors, the majority
of the SOCs were removed by biological activity,
with GAC adsorption providing stability to each
system during startup and buffering against load
fluctuations during long-term operation.
The SOCs and their respective concentra-
tions added to the two leachates are summarized in
Table 1. During the first phase of the study, chloro-
form was deleted from the SOC supplement because
of toxicity problems associated with this volatile
organic compound. The effect of chloroform on the
performance of the treatment systems was evalu-
ated during the second phase of the study.
Phase One (No Chloroform Addition)
The first phase of the study extended over a
period of 400 days, during which both reactor sys-
tems exhibited excellent removal efficiencies for all
the SOCs listed in Table 1. A summary of the
performance of the two systems averaged over the
last 260 days of operation is given in Table 2. A
comparison of the two systems indicates that the
sulfate-reducing environment may produce equal
or better performance than a methanogenic envi-
ronment in removing a consortium of hazardous
chemicals from waste streams. All three volatile
aromatic compounds (toluene, chlorobenzene, and
ethylbenzene) in the SOC consortium were removed
at higher rates in the sulfate-reducing environment.
Also, the persistence of the intermediate biodegra-
dation product of nitrobenzene, aniline, was only
significant in the methanogenic reactor.
15
-------
Treatment ih a Reactor
Phase Two (Chloroform Addition)
During the second phase of the study, chloro-
form was added to the SOC supplement at an initial
feed concentration of 2 mg/L. This feed concentra-
tion was maintained for a period of 105 days. During
this period, chloroform was not observed to exert an
adverse effect on the performance of either the sul-
fate-reducing or the methanogenic reactors.
Encouraged by the ability of both reactors to handle
this feed level of chloroform, the influent chloro-
form concentration was raised to 3.5 mg/L. The
methanogenic reactor responded with a rapid de-
cline in methane production and COD reduction,
and corresponding increases in effluent SOC con-
centrations. The sulfate-reducing reactor, on the
other hand, was able to withstand this increased
feed concentration of chloroform, and continued to
maintain excellent removal of all the SOCs for a
period of 156 days. The current plan is to continue
increasing the feed concentration,of chloroform to
the sulfate-reducing reactor. Recovery of the
methanogenic reactor was rapidly achieved after
chloroform addition was terminated.
Table 1. Composition of SOC supplement added to the leachates.
Compound
Concentration
(pg/D
VOLATILE ORGANIC COMPOUNDS
Acetone
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Trichloroethylene
1,1-Dichloroethane
Methylene Chloride
Chloroform ;
Chlorobenzene
Ethylbenzene
Toluene
SEMIVOLATILE ORGANIC COMPOUNDS
Phenol
Nitrobenzene
1,2,4-Trichlorobenzene
DibutylPhthalate
' 10,000
5,000
1,000
400
100
1,200
0 to 3,500
1,100
600
8,000
2,600
500
200
200
16
-------
Treatment in a Reactor
Table 2.
Summary of SOC concentrations in reactor effluents.
Compound
Sulfate
Reducing
Reactor
Acetone
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Trichloroethylene
Methylene Chloride
1,1-Dichloroethane
Chlorobenzene
Ethylbenzene
Toluene
Phenol
Nitrobenzene
1 ,2,4-Trichlorobenzene
Dibutyl Phthalate
Effluent
189* (216)#
70 (68)
35 (16)
8 (10)
65 (50)
20 (17)
67 (42)
34 (19)
436 (303)
22 (23)
6 (16)
10 (15)
26 (29)
Effluent
Methanogenic
Reactor
410* (577)#
220 (150)
58 (25)
5 (3)
46 (44)
14 (8)
165 (86)
85 (41)
1,102 (744)
93 (63)
8 (17)
14 (16)
36 (30)
*A11 concentrations in ng/L.
^Standard deviation.
IMPROVED PREDICTION OF
GAC CAPACITY IN A BIOLOGICALLY
ACTIVE FLUIDIZED BED
M.T. Suidan and R.D. Vidic, Department of
Civil and Environmental Engineering, Uni-
versity of Cincinnati, Cincinnati, OH; andKC.
Brenner, U.S. Environmental Protection
Agency, Cincinnati, OH.
The first applications of activated carbon for
water quality control involved the addition of pow-
dered activated carbon (PAC) to drinking water
treatment plants for taste and odor control. In recent
years, granular activated carbon (GAC) has received
increasing attention for removing synthetic organic
compounds from water. Moreover, the possibility
of GAC reuse have accentuated the applicability of
granular activated carbon to drinking water
treatment.
Interest in removing biologically resistant or-
ganic contaminants led to the application of activated
carbon in wastewater .treatment. PAC has been
used in activated sludge systems. (1), while GAC
found application in anaerobic fixed film processes
(2) and in aerobic wastewater treatment (3). In
addition to its adsorptive properties, activated car-
bon has been reported to provide an excellent surface
for microbial attachment (4,5).
Nakhla (6) found the adsorptive capacity of
GAC for o-cresol in an expanded-bed anaerobic
GAC bioreactor treating a synthetic mixture of phe-
nol, acetic acid, and o-cresol to be much lower than
that determined from an adsorption isotherm con-
ducted on virgin carbon using the standard
bottle-point technique. Further experimentation
revealed that the adsorptive capacity obtained from
the biological anaerobic reactor agreed very well
with capacities of virgin GAC determined from
isotherm experiments conducted in the absence of
molecular oxygen. The adsorptive capacity of vir-
17
-------
Treatment in a Reactor
gin GAC exhibited for o-cresol in the presence of
molecular oxygen (oxic conditions) was almost 200
percent above the adsorptive capacity of virgin
GAC attainable in the absence of oxygen (anoxic
conditions). Since both aerobic and anaerobic bio-
logical GAC reactors are in use, this study was
designed to further investigate the effects of mo-
lecularoxygenontheadsorptivecapacity of activated
carbon for different organic compounds.
To evaluate the effect of the same functional
group substituted at different positions on the par-
entphenolmolecule/bothoxicand anoxic adsorption
isotherm experiments were conducted using 16 x 20
U.S. Mesh virgin F-400 GAC (Calgon Carbon, Pitts-
burgh, PA) as an adsorbent and 2-, 3-, and
4-methylphenol as adsorbates. Experimentally de-
termined data on adsorption capacity for these three
compounds on GAC are presented in Figure 1. All
the adsorption isotherms were found to be well
described by the Freundlich isotherm equation
qe = K * Cc1/n. The straight lines on Figure 1 were
obtained by nonlinear least square fit of experimen-
tal data to the Freundlich isotherm equation^ As is
apparent from this figure, the absence of molecular
oxygen from the test environment resulted in al-
most identical GAC adsorptive capacities for all
three compounds. On the other hand, the presence
of molecular oxygen had a diverse impact on the
adsorptivecapacityofGACfordifferentcompounds.
This is well illus trated in Figure 2, where the ratio of
oxic and anoxic adsorptive capacities for all three
compounds is plotted as function of the respective
equilibrium adsorbate liquid phase concentrations.
The most significant increase in oxic adsorptive
capacity was observed in the case of 2-methylphenol
followed by 4-methylphenol, while the adsorptive
capacity of GAC for 3-methylphenol was the least
affected by the presence of molecular oxygen. An-
other important conclusion from this figure is that
the two adsorption isotherms (oxic and anoxic)
have very different values of the coefficient 1 /n in
the Freundlich isotherm equation, since the ratio of
the two adsorptive capacities depends on the liquid
phase concentration of the adsorbate.
Further investigation of the observed phe-
nomenon was continued by extracting GAC that
was preloaded with each of the adsorbates during
the adsorption isotherm experiments. Extraction
was performed in a soxhlet extraction apparatus
using methanol for 1 day followed by further extrac-
tion with methylene chloride for an additional period
of 3 days. The results of these experiments are
presented in Figure 3. Extraction efficiency was
calculated as the ratio of the mass of adsorbate in the
extract to the mass of adsorbate loaded on the car-
bon during the adsorption isotherm experiment.
The extraction efficiencies obtained from the
carbons used in anoxic isotherm experimentsshowed
very little dependence on the carbon loading. On
the average, 90 percent of the adsorbed compound
was extracted from these carbons. On the other
hand, the extraction efficiencies for the carbons used
in the oxic isotherm experiments were significantly
lower and exhibited strong dependency on the equi-
librium carbon loading. Furthermore, the least
amount of adsorbate in the oxic isotherm experi-
ments was extracted from the GAC in the case of
2-methylphenol, followed by 4-methylphenol and
3-methylphenol. This arrangement is in agreement
with the degree of influence of molecular oxygen on
the adsorptive capacity of GAC for these three ad-
sorbates.
Gas chromatographic/mass spectroscopic
analyses of the extracts from several samples of
carbon that were preloaded with o-cresol under oxic
conditions revealed the presence of significant
amounts of dimers, trimers, and even tetramers of o-
cresol. The discovery of polymers of o-cresol sug-
gested that some polymerization reactions are tak-
ing place on the surface of GAC in the presence of
molecular oxygen. These polymerization reactions
offer a possible explanation for the increased re-
moval of adsorbate from the liquid phase under oxic
conditions.
References
1. Weber, W.J., Jr. and B.E.Jones. 1986. Toxic
substances removal in activated sludge and
PAC systems. Water Engineering Research
Laboratory, Office of Research and Devel-
opment, U.S. EPA, Cincinnati.
2. Lowry, T.D. and C.E. Burkhead. 1980. The
role of adsorption in biologically extended
activated carbon columns. /. Water Pollut.
Control Fed. 52:389.
18
-------
Treatment in a Reactor
10'
3. Suidan, M.T., C.E. Strubbler, S.W. Kao, and
J.T. Pfeffer. 1983. Treatment of coal gasifi-
cation wastewater with anaerobic filter tech-
nology. /. Water Pollut. Control Fed. 55:1263.
4. Characklis,W.G. 1973. Attached microbial
growth, attachment and growth. Water
Research 2:113.
5. Fox, P., M.T. Suidan, and J.T. Bandy. 1990.
A comparison of media types in acetate fed
expanded-bed anaerobic reactors. Water
Research 24:827.
6. Nakhla, G.F., M.T. Suidan, and U.K.
Traegner. 1989. Steady-state model for the
expanded-bed anaerobic G AC/Reactor with
GAG replacement and treating inhibitory
wastewaters. Proceedings,IndustrialWaste
Symposium, 62nd WPCF Annual Meeting,
San Francisco, CA.
o 2-Methylphenol
° 3-Methylphenol
•> 4-Methylphenol
10* 10' 10' 10*
C. mg/L
Figure 1. Oxic and anoxic adsorption isotherms of
methylphenol on GAG.
o
g-8.0
1,
2-Methylphenol
3-Uothylphnnol
4-Uothytphenol
10'
C, mg/L
Figure 2. Ratio of oxic to anoxic adsorptive capaci-
ties of GAG for methylphenol.
100
90
ao
70
80
ISO
40
30
20
anoxic
o 2-Hethylphenoi
o 3-Methylphenol
« 4-Methylphenol
0 100 200 300 400 900
q. mg/g
Figure 3. Extraction efficiency of methylphenol
from GAG.
TREATMENT OF TCE AND
DEGRADATION PRODUCTS
USING PSEUDOMONAS CEPACIA
Malcolm Shields, TRI/U.S. Environmental
Protection Agency, Gulf Breeze, PL.
Summary
The constitutive trichloroethylene (TCE)-de-
grading Pseudomonas cepacia strain G4 Phel was
found to be very stable, with no detectable loss of
activity (or marker) after 100 generations of
nonselective growth. Toluene monooxygenase ac-
tivity towards trifluoromethyl phenol was constant
throughout the growth curve as measured by the
rate of production of 2-hydroxy-7,7,7-
trifluoroheptadienoic acid. TCE degradation rates
measured for G4 Phel (0.5 - 1 nmole TCE muv1
mg-1 protein) are approximately 15 to 30 percent of
the maximal wild type G4 phenol induced activity.
G4 Phel was found to degrade TCE over a fairly
wide range of physical conditions: 4 to 30 °C, pH 4 to
9,3 to 30 mg I-1 oxygen, and 0 to 20 0/00 salinity. A
slight lowering of growth rate in G4 Phel was
observed in the presence of TCE at an aqueous
concentration of 530 jo.M. Chlorobenzene and 2-
chlorophenol were found to be inhibitory to TCE
degradation at 1 to 10 joM. Following introduction
of pROlOl (a transposon derivative of pJP4) 2-
chlorophenol and chlorobenzene levels inhibitory
to TCE degradation were increased to 100 |J.M.
19
-------
Treatmeht in a Reactor
Results and Discussion
Ground-water contamination by organic pol-
lutantsisasubjectofoverwhelmingconcern through-
out the industrialized world. Chief among these
pollutants are those categorized as volatile organics.
These include the chloroethylenes: TCE,
tetrachloroethylene, fnms-l,2-dichloroethylene
(DCE), 1,1-DCE, and vinyl chloride (ranked first,
second, third, fifth, and more than tenth, respec-
tively, of all volatiles detected as ground-water con-
taminants in the United States) (3). The constitutive
TCE degrader, Pseudomonas cepada G4 Phel, that
employs a unique toluene monooxygenase (4) for
the degradation of TCE (5), was investigated for its
bioremediation potential.
Rates of TCE degradation were determined
using a glass syringe with a Teflon plunger and no
air headspace in the bioreaction chamber as previ-
ously described (1). This assay was adopted for rate
u
1.1.
1J.
u.
0)
2
Q.
I
I
o
(A
T
LT
CMUflMUMtf *
10 II 20 «
•c
U 7.7 at J1J
Oxygm (mg/L)
analysis because it more closely resembles a con-
taminated aquifer, and multiple samples can be
taken without the introduction of an air headspace.
The range of certain physical variables (tempera-
ture, oxygen concentration, pH, and salinity) likely
to affect the rate of TCE degradation by P. cepada G4
Phel under environmental conditions was investi-
gated using this technique.
The most profound effect on the rate of TCE
degradation was found at the lowest temperatures
used (Figure la) where despite cooling to 4°C, G4
Phel maintained approximately 30 percent of the
rate of TCE degradation measured at 30°C. TCE
degradation rates of approximately 20 and 45 per-
cent (relative to the maximal pH 7 value) were
evident at pH extremes of 4 and 9, respectively
(Figure Ib). Little effect on TCE degradation rates
was observed over the ranges of oxygen concentra-
tion (2.8 to 31.3 mg I/1) (Figure 1 c) or salinity (0 to 20
0/00) tested (Figure Id).
b
T
'' /*
'i «
' ^
t ' /
'* /
" ' » '* '
//:
;, '*
>
•^ ^
f5
< •.,
s'
s vf
'/',
^%
T
t"" i
? * -^
;;/
T
<
r1!
pH
1.1
1.0
OJ
OJ
0.7
OJ
0.1
OJ
OJ
0.2
0.1
I
0 • JO
8»llnlly (ODO)
Figure 1. Effects of varying temperature, pH, dissolved oxygen, and salinity on the rate of TCE degradation
by Pseudomonas cepada G4 Phel.
20
-------
Treatment in a Reactor
Possible toxicity of a metabolite of TCE to-
wards G4 Phel was investigated. G4 does not affect
TCE in the absence of an aromatic inducer (2).
Therefore, any toxicity in the presence of TCE would
be due to direct toxicity. G4 Phel differs only in one
significant respect from G4 (i.e., the constitutive
expression of the toluene and TCE degradative en-
zymes). Any toxic effects beyond those seen with
G4 would therefore be attributable to the active
metabolism of TCE. A slight toxic effect was mea-
sured as a depression of growth rate for G4 Phel on
glucose and yeast extract as a result of the metabo-
lism -of TCE at levels c.a. 2 mM as if all were in
aqueous solution (measured at 530 |iM).
References
1. Folsom, B.R., P.J. Chapman, and P.H.
Pritchard. 1990. Appl. Environ. Microbiol.
56:1279-1285.
2. Nelson, M.J.K., S.O. Montgomery, W.R.
Mahaffey, and P.H. Pritchard. 1987. Appl.
Environ. Microbiol. 53:949-954.
3. Rajagopal,R.I986.Environ.Prof. 8:244-264.
4. Shields, M.S., S.O. Montgomery, P.J.
Chapman,S.M. Cuskey,andP.H. Pritchard.
1989. Appl. Environ. Microbiol. 55:1624-1629.
5. Shields, M.S., S.O. Montgomery, S.M.
Cuskey, P.J.Chapman, and P.H. Pritchard.
1991. Appl. Environ. Microbiol. Submitted.
AEROBIC BIODEGRADATION OF
VOLATILE ORGANIC COMPOUNDS
IN A BIOFILTER
Vivek Utgikarand Rakesh Govind, Department
ofChemicalEngineering, University of Cincin-
nati Cincinnati, OH; and Fred Bishop, Chief,
Biosystems Branch Risk Reduction Engineer-
inglaboratory, U.S. Environmental Protection
Agency, Cincinnati, OH.
In recent years the emission of volatile organic
compounds (VOCs) has received increased atten-
tion from EP A, OSH A, and other government agen-
cies due to the serious human health hazards these
compounds present as pollutants. The origins of
these VOCs can be from manufacturing processes
or wastewater treatment plants, where the waste
stream is stripped of the VOCs during aeration.
Another significant source of these pollutants is
landfill leachate.
The conventional physical/chemical treat-
ment methods for these gaseous pollutants are
adsorption on a solid, absorption in a solvent, in-
cineration, or catalytic conversion. An alternative
to these conventional treatment methods is the
biological destruction of the VOCs. This method
has the advantages of pollution destruction (as
compared to transfer to another medium) and lower
operation and maintenance costs.
The biodegradation can be carried out in a
biofilter. A biofilter consists of a packed column
containing biologically active mass. The biologi-
cally active matter (biomass) can exist either as a
biofilm on the support medium or as biomass
particles trapped in the void spaces between the
support material.
Biodegradation of three volatile organic com-
pounds in an aerobic biofilter was studied. The
three chemicals (substrates) were studied at the
following concentrations: toluene: 520 ppm; meth-
ylene chloride: 180 ppm; trichloroethylene: 25 ppm.
The substrates were fed upflow to the biofilter
through the gas phase. The requisite composition of
the substrates in the gas phase was achieved by
making the synthetic gas mixtures in a cylinder and
subsequently blending them with air. The biofilter
was packed with pelletized activated carbon sup-
port material Nutrient solution was circulated
counter to the gas through the bed. The inlet and
outlet gas streams were analyzed for the above
three chemicals.
The biofilters contained active acclimated bio-
mass. The results showed that nearly 100 percent
removal of the three compounds was achieved in
the biofilter.
21
-------
-------
SECTION THREE
SOIL/SEDIMENT TREATMENT
Soil and sediment contamination is one of the most difficult problems encountered at hazardous waste sites. Soils
at industrial sites are often contaminated with a complex mixture of pollutants. Biologically treating these soils in situ is
much more effective and inexpensive than excavating the soils—in itself a major task — and then performing the treatment.
EPA's biosystems research on soils and sediments currently focuses on pentachlorophenol (PCP), hexachlorobenzene, and
other chlorinated aromatic compounds.
PCP is a common soil contaminant at wood-preserving facilities. Recent EPA studies have shown that a lignin-
degrading fungus, Phanerochaete chrysosporium. can effect a rapid and extensive depletion of PCP from soil and other
contaminated materials. Current research is underway to further investigate the feasibility of using P. chrysosporium and
other lignin-degrading fungi to remediate soils contaminated with wood-preserving chemicals.
Another recent project focused on the ability of unacdimated microorganisms and sediment microorganisms
acclimatedtodechlorinate2,4-and3,4-dichlorophenoltodegrade2,4-D,2,4,5-T,DDT,hexachlorobenzene,andchloroanisoles.
EPA researchers also recently examined the effects of various nonionic surfactants on the anaerobic dechlorination of
hexachlorobenzene. The use of surfactants has been suggested as a way to enhance soil treatment, since surfactants can
mobilize pollutants into micellar solution in the presence of soil or sediment solids, desorbing them from these media.
AnotherrecentEPAstudyexaminedtheanaerobicdegradationofchlorophenolundermethanogenicandsulfidogenk
conditions. The same study also examined the degradation of chlorinated phenols and benzoic acids under methanogenic,
sulfidogenic, and denitrifying conditions. The results indicate that degradation of these chlorinated aromatic compounds
can take place under various reducing conditions.
Researchers have also examined the anaerobic transformation of phenol to benzoate using a bacterial consortium.
Phenol was transformed to benzoate without complete mineralization of benzoate; the results showed that transformation
was via para-carboxylation.
USE OF LIGNIN-DEGRADING
FUNGI IN THE REMEDIATION OF
PENTACHLOROPHENOL-
CONTAMINATED SOILS
R.T. Lamar, DM. Dietrich, and T.K. Kirk,
Institute for Microbial and Biochemical
Technology, Forest Products Laboratory,
Madison, WI, and J.A. Glaser, U.S. EPA,
Risk Reduction Engineering Laboratory,
Cincinnati, OH.
Recently, we reported that a lignin-degrad-
ing fungus, Phanerochaete chrysosporium, was capable
of depleting pentachlorophenol (PGP) from three
soils in the laboratory. This depletion was found to
result primarily from the transformation of PCP to
nonvolatile products. The nature of these prod-
ucts—whether soil-bound or extractable—was
greatly influenced by soil type. We believe that the
soil-bound products are humic acid-PCP hybrid
polymers that are formed as a result of the activity of
extracellular enzymes (lignin peroxidases, Mn per-
oxidases, laccases) with phenol-oxidizing activity,
which are produced by the lignin-degrading fungi.
Our objective is to determine the feasibility of using
lignin-degrading fungi to remediate soils that are
contaminated with wood-preserving chemicals. The
results of some of our studies are summarized
below.
23
-------
Soil/Sediment Treatment
In investigations of white-rot fungal degrada-
Uonofhazardouscompounds/P.cfoT/sosporiMmstrain
BKM-F-1767 has been used almost exclusively as
the experimental organism. However, there are an
estimated 1,400 species of lignin-degrading fungi,
and there is great diversity among these organisms
in their ability to degrade lignin. Thus, there is
reason to believe that this same diversity will be
seen in xenobiotic degradation.
Several studies were conducted to determine
the degree of interspecific and intraspecific varia-
tion among selected Phanerochaete spp. in growth
rate and in their ability to tolerate and degrade the
wood preservative PCP. Mycelial extension rates of
selected strains of Phanerochaete chrysorhiza,
Phanerochaete laevis, Phanerochaete sanguinea,
Phanerochaete filamentosa, Phanerochaete sordida,
Inonotus circinatus, and Phanerochaete chrysosporium,
and the ability of these organisms to tolerate and
degrade PCP in an aqueous medium and in soil,
were measured. Most of the tested species had
maximum mycelial extension rates in the range of
<0.5 to 1.5 cm d4, but there were large interspecific
differences. A notable exception, P. sordida, grew
very rapidly, with an average mycelial extension
rate of 2.68 cm d*1 at 28°C. There were also signifi-
cant intraspecific differences in mycelial extension
rates. For example, mycelial extension rates among
strains of P. sordida ranged from 1.78 to 4.81 cm d'1.
Phanerochaete spp. were very sensitive to PCP
in 2 percent malt agar. Growth of several species
was prevented by the presence of 5 ppm PCP. How-
ever, P. chrysosporium and P. sordida grew at 25 ppm
PCP, albeit at greatly decreased mycelial extension
rates. We have observed lignin-degrading fungi in
PCP-contaminated soils with PCP concentrations
exceeding 500 ppm. Therefore;, growth of lignin-
degrading fungi on malt agar in the presence of PCP
should be used as a screening tool for determining
relative sensitivities and not as an indication of
growth performance of these organisms in PCP-
contaminated soil.
In an aqueous medium, mineralization of PCP
by P. sordida strain 13 was significantly greater than
that by all other tested P. sordida strains and P.
chrysosporium (Table 1). This strain was tested in a
field investigation with P. chrysosporium to deter-
mine the ability of these fungi to deplete PCP in
contaminated soils.
Inoculation of a field soil contaminated with a
commercial wood preservative product and con-
taining 250 to 400 n,g g'1 PCP with either Phanerochaete
chrysosporium or P. sordida resulted in an overall
decrease of 88 to 91 percent of PCP in the soil in 6.5
weeks. This decrease was achieved under suboptimal
temperatures for the growth and activity of these
fungi, and without the addition of inorganic nutrie-
Table 1. Percentage of total [MC] PCP mineralized and volatilized in liquid cultures of P. chrysosporium
and several strains of P. sordida after 30 days3.
Strain
P. chrysosporium
P. sordida 1
P. sordida 8
P. sordida 9
P. sordida 13
Control
Mineralization
1.97 (0.22)b
2.67 (1.09)b
1.92(0.44)b
1.22(0.54)b
11.64(2.54)c
0.17 (0.03)d
Volatilization
^ /O )••
• 13.82 (1.20)b
12.91 (2.79)b
8.88 (2.21 )b
11.92 (2.35)b
8.48 (0.52)b
0.06 (0.03)c
Totiil 14C Evolved
15.79
15.58
10.80b
13.14
20.12
' 0.23
a Figures in parentheses represent the standard deviation of three observations. Means followed by
the same letter are not significantly different.
24
-------
Soil/Sediment Treatment
Table 2. Percentage of mineralization and volatilization of pentachlorophenol and pentachloroanisole
in liquid cultures of P. chrysosporium or P. sordida 13 and from control (noninoculated) cultures.
Culture
Compound
Mineralization
Volatilization
(*)
P. chrysosporium
P. sordida
Control
PCP
PCA
PCP
PCA
PCP
PCA
8.91
7.95
16.13
13.07
1.05
0.40
7.95-
9.54
8.25
10.02
0.08
15.92
ents. A small percentage (8 to 13 percent) of the
decrease in the amount of PCP was a result of
fungal methylation to pentachloroanisole (PCA).
However, both of these organisms can also trans-
form PCA (Table 2), and thus it would also be
expected to be depleted from soil over time.
For soil studies, inocula have consisted of
aspen chips (0.65 to 1.3 cm) preinfested with pure
cultures of a single fungus. We have found in
laboratory studies and in a field study that aspen
chips absorb PCP. Therefore, we have also investi-
gated the metabolism of PCP by lignin-degrading
fungi in wood chips.
Inoculation of PCP-contaminated softwood
or hardwood chips (whether sterilized or not) with
either P. chrysosporium or P. sordida resulted in a
decrease in the PCP concentration of the chips. No
decrease in the PCP concentration was observed in
noninoculated chips, indicating that the observed
PCP decreases were due to the activities of P.
chrysosporium or P. sordida. Depletion in hardwood
and softwood chips inoculated withP. chrysosporium
was rapid and extensive (63 to 72 percent decrease
after 6 weeks), except in nonsterile softwood chips.
In nonsterile softwood chips, depletion of PCP was
very slow and resulted in only a 30 percent decrease
after 6 weeks. This lower rate of PCP depletion may
have been the result of a lower rate of colonization
of these chips by P. chrysosporium due to competi-
tion from indigenous microbes.
Depletion of PCP by P. sordida was also af-
fected by sterilization. Inoculation of nonsterile
softwood and hardwood chips resulted in only a 50
percent and a 45 percent decrease in the PCP con-
centration, respectively, after 42 days. However,
the PCP concentration in both hardwood and soft-
wood chips that had been sterilized was decreased
by ca. 66 percent by P. sordida after 42 days. Again,
this lower rate of decrease was probably due to
competition from indigenous microbes.
Depletion of PCP was always accompanied by
an increase in the concentration of PCA. Accumula-
tion of PCA in sterile cultures was much greater
than in nonsterile cultures of both fungi. This was
particularly true for cultures inoculated with P.
sordida. Only 7 percent and 19 percent of the PCP
decrease in nonsterile softwood and hardwood cul-
tures, respectively, was due to conversion of PCP to
PCA. However, this low rate of conversion was
associated with relatively low amounts of PCP deple-
tion.
In nonsterile hardwood and softwood chips
inoculated with P. chrysosporium, 65 percent and 72
percent, respectively, of the PCP decrease was due
to conversion of PCP to PCA. In sterile chips inocu-
lated with either fungus, virtually all of the PCP
decrease was due to conversion to PCA. The differ-
ence between sterile and nonsterile chips in the
amount of PCP loss due to methylation to PCA
suggests that the indigenous microbes inhabiting
the nonsterile chips, which were not able to metabo-
lize PCP, were able to metabolize PCA and thus
prevent its accumulation.
25
-------
Soil/Sediment Treatment
Inoculation of PCP-contaminated softwood
chips with the lignin-degrading fungus Trametes
Mrsuta also resulted in a rapid and extensive re-
movalofPCP(Figurel). Approximately62percent
of the PCP was removed after 28 days of incubation.
ThisissimilartoandgreaterthantheamountofPCP
removed by P. chrysosporium and P. sordida, respec-
tively, after thesame incubation time. However, the
removal was not due to conversion of PCP to PCA.
Since no PCA accumulated in cultures of this fun-
gus, we are in the process of determining the fate of
PCP in soils supporting growth of Trametes hirsuta to
determine the ability of this organism to transform
PCP in soils to innocuous products.
The results of these and other studies have
shown that lignin-degrading fungi can effect rapid
and extensive depletion of PCP from soils and other
contaminated media (i.ev wood chips), and that
there is a great diversity among fungal species in
their ability to effect decreases of PCP and in their
metabolism of PCP. Further studies are needed to
confirm the incorporation of PCP into humic mate-
rials hi soils, to assess the stability of these hybrid
polymers, and to continue the screening process to
identify fungi with superior PCP-degrading capa-
bilities.
Control
10 20
Time (days)
30
Figure 1. Effect of inoculation with lignin-degrad-
ing fungus Trametes hirsuta on the PCP
concentration of PCP-contaminated soft-
wood chips.
ANAEROBIC DEGRADATION OF
CHLORINATED AROMATIC
COMPOUNDS
John E. Rogers, U.S. Environmental
Protection Agency, Athens, GA; Dorothy Hale,
Wren Howard, and Frank Bryant, TAI, Athens,
GA; and Mahmoud Mousa and Shiu-Mei Hsu,
UGA, Athens, GA.
Over the last few years, our laboratory has
been investigating the anaerobic degradation of a
variety of nitrogen heterocyclic and chlorinated aro-
matic compounds. These compounds have included
pentachlorophenol, 2,4-D; 2,4,5-T, DDT and chlori-
nated anisoles, anilines, benzenes, benzoic acids,
biphenyls, and phenols. In 1991, we reported on the
reductive dechlorination of pentachlorophenol by
sediment microbial communities acclimated to de-
chlorinate 2,4- and 3,4-dichlorophenol (DCP). We
report this year on the degradation of 2,4-D, 2,4,5-T,
DDT, hexachlorobenzene, and chloroanisoles.
The ability of unacclimated sediment microor-
ganisms and sediment microorganisms acclimated
to dechlorinate 2,4- and 3,4- (DCP) to degrade 2:,4-D
and 2,4,5-T was investigated. Acclimated sediment
microorganisms were prepared from sediment col-
lected in November 1987 and March 1990. When the
experiments were conducted using sediments from
November 1987, the 2,4-DCP acclimated microor-
ganisms dechlorinated 2,4-D without a lag to
4-chlorophenoxyaceticacid,the3,4-DCP acclimated
microorganisms did not dechlorinate 2,4-D over
several months of exposure, a mixture of the two
acclimated microbial populationsparalleled the 2,4-
DCP acclimated microorganisms, and the
unacclimated sediment microorganisms paralleled
the 3,4-DCP acclimated microorganisms. The 4-
chlorophenoxyacetic acid was produced in
stoichiometric quantities and was stable for the
duration of the experiment. Similar results were
observed with sediments collected in March 1990;
however, 4-chlorophenoxyacetic acid was readily
degraded in the 2,4-DCP acclimated sediment. 2,4,5-
T was degraded at the same rate with acclimated
sediment microorganisms and unacclimated sedi-
mentmicroorganisms;nodegradation intermediates
. were detected and dechlorination followed a lengthy
lag period (>14 days).
26
-------
Soil/Sediment Treatment
Previous studies in our laboratory (1) have
shown that sediment microorganisms acclimated to
dechlorinate 2,4- or 3,4-DCP could dechlorinate the
respective chlorinated anilines but not the respec-
tive chlorinated benzoic acids. A partial explanation
for the dechlorinating specificity was that the phe-
nol and aniline are ortho/para directing, whereas
benzoic acid is meta directing. To further test this
hypothesis we have examined the dechlorination of
chlorinated anisoles. Because of the similarity of the
sigma values for the methoxy substituent of anisole
and the hydroxy and amine substituents of phenol
and aniline, the methoxy substituent should also be
ortho/para directing for dechlorination. Therefore,
we have examined the degradation of 2,4-
dichloroanisole (DCAn) by the 2,4- and 3,4-DCP
acclimated sediment microorganisms (March 1990).
When added to unacclimated sediment microor-
ganisms, 2,4-DCAn was slowly demethylated (50
percent in 24 days) to 2,4-DCP following a 5-day lag
period. The 2,4-DCP was subsequently degraded
over the next 2 weeks. Following a second addition
of 2,4-DCAn, a 50 percent loss of 2,4-DCAn was
observed in 2 days; however, the product observed
was 4-chloroanisole. When 2,4-DCAn was added to
an equal mixture of 2,4- and 3,4-DCP acclimated
sediment microorganisms or to separate 2,4-and
3,4-DCP acclimated sediment microbial communi-
ties, greater than 50 percent of the 2,4-DCAn was
lost in 2 days in all cases. The only products iden-
tified were 4-chlorophenol and phenol.
Unfortunately, we were unable to determine whether
dechlorination and demethylation occurred con-
comitantly or sequentially. The finding that
4-chloroanisole was the primary product from the
second addition of 2,4-DCAn, however, is indica-
tive of rapid reductive dechlorination of 2,4-DCAn
by a microbial community acclimated to dechlori-
nate 2,4-DCP.
To date, we have investigated the reductive
dechlorination of several compounds having some
structural resemblance to the acclimating substrates.
Recently we have investigated the ability of accli-
mated sediment microorganisms to degrade
compounds with little structural resemblance. DDT
added to unacclimated sediment microorganisms,
both autoclaved and nonautoclaved, was rapidly
converted to DDE within 4 days. Similar results
were observed for DDT added to 2,4- and 3,4-DCP
acclimated sediment microorganisms, both auto-
claved and nonautoclaved. However,
4-chlorophenol was observed as a significant prod-
uct with acclimated microorganisms (autoclaved
and nonautoclaved) but not with unacclimated mi-
croorganisms. The mechanism of 4-chlorophenol
formation is currently under investigation.
Hexachloro- benzene (HCB) was investigated in
sediments collected from two different ponds. In
both cases, a loss of HCB (70 ppm) was observed
after a substantial lag phase (90 to 100 days). Less
than 20 percent of the HB remained after 170 days of
incubation. Mixtures of lesser chlorinated conge-
ners were identified as intermediate degradation
products. A different mixture was observed in each
of the two sediments, and no one product domi-
nated in either case. Hexachlorobenzene was stable
in autoclaved sediment.
The reductive dechlorination of chlorinated
aromatic compounds requires a source of reducing
equivalents. Dolfing and Tiedje (2) have identified
molecular hydrogen as a source of reducing equiva-
lents for the dechlorination of 3-chlorobenzoate by
an anaerobic consortium. Gibson and Suflita (3) and
Nies and Vogel (4) have shown that organic sub-
strates can also be the source of reducing equivalents
for the reductive dechlorination of 2,4,5-T and PCBs.
We are currently testing molecular hydrogen and a
number of organic substrates as sources of reducing
equivalents for the reductive dechlorination of 2,4-
and 3,4-DCP by microorganisms in sediments slur-
ries. In some cases, our results are different from
previous reports. We found, for example, that
changes in headspace gas composition resulted in
the following order of increasing reductive dechlo-
rination: COj/N^N^HVN2. We also observed
that propionate, formate, and butyrate, which have
been shown to stimulate the degradation of 2,4,5-T,
inhibited or did not enhance the reductive dechlori-
nation of 2,4-DCP. In a parallel study, we have
tested a number of complex organic substrates as
possible sources of reducing equivalents. Addition
of sterile sediment or a sediment extract supported
reductive dechlorination of 2,4-DCP. Lake water
collected from just above the sediment source only
marginally supported 2,4-DCP degradation. Or-
ganic mixtures such as landfill leachate and rumen
27
-------
Soil/Sedirnent Treatment
fluid did not stimulate dechlorination of 2,4-DCP
when added to sediments at low concentrations
(1 to 2 percent) and inhibited activity at higher (15 to
20 percent) concentrations.
References
1. StruijsJ. and J.E.Rogers. 1989. Reductive
dechlorination of dichloroanilines by
anaerobic microorganisms in fresh and
dichlorophenol-acclimated pond sediment.
Appl. Environ. Microbiol. 55:2527-2531,
2. DolfingJ.andJ.M.Tiedje. 1991. Influenceof
substituents on reductive dehalogenation
of 3-chlorobenzoate analogs. Appl. Environ.
Microbiol. 57:820-824.
3. Gibson, S.A. and J.M. Suflita. 1990.
Anaerobic biodegration of 2,4,5-
trichlorophenoxyacetic acid in samples from
a methanogenic aquifer: Stimulation by
short-chain acids and alcohols.
4. Nies,L.andT.M.Vogel. 1990. Effects of
organic substrates on dechlorination of
Aroclor 1242 in anaerobic sediments. Appl.
Environ. Microbiol. 56:2612-2617.
INFLUENCE OF NONIONIC
SURFACTANTS ON THE
ANAEROBIC DECHLORINATION OF
HEXACHLOROBENZENE
Patricia VanHoof, U. ofGeorgia/U.S. Environ-
mental Protection Agency, College StationRoad,
Athens, GA;and Chad T. Jafoert, Environmen-
tal Research Laboratory, U.S. Environmental
Protection Agency, Athens, GA.
Background
Surfactants can solubilize pollutants into mi-
cellar solution in the presence of soil or sediment
solids, effectively desorbing them from these natu-
ral media (1,2). This phenomenonhas been suggested
as a possible tool to enhance the treatment of con-
taminated sediments or soils (3,4,5). The rationale
is that surfactant micelles (or monomers or emul-
sions) will solubilize precipitated or sorbed
compounds, making them more readily available
for biological remediation, pump-and-treat opera-
tions, or soil washing operations.
The effectiveness of surfactants in removing
contaminants from soils or sediments is largely a
function of 1) the sorption reactions of pollutants to
the sedimentary materials, 2) the solubilization of
pollutants by the surfactant micelles (and / or mono-
mers), and 3) the interactions of surfactant monomers
and micelles with sediment or soil components.
These processes have recently been examined in
several freshwater sediments spiked with various
polycyclic aromatic hydrocarbons (PAHs) and the
anionic surfactant sodium dodecylsulfate (SDS) (1,
2). Also, we have recently examined the solubiliza-
tion of PAHs and other compounds in solutions of
various nonionic surfactants, including Brij 35,
Tween 80, and Tween 20, as well as observing some
of the interactions of these surfactants with sedi-
ment solids.
The structures of Tween 80 and Brij 35 are
shown in Figure 1. Unlike SDS, which can be
purchased in pure form (>99 percent), these surfac-
tants are homolog mixtures of differing ethoxy chain
lengths. Also, unlike SDS, whose primary interac-
tion with sediment components under conditions
of interest is the precipitation of its calcium salt,
nonionic surfactants sorb to sediment solids—pos-
sibly through both hydrophobic and hydrophilic
mechanisms. At surfactant concentrations around
their critical micelle concentration (cmc), and at
sediment concentrations of 5 to 10 percent, a large
portion (>95 percent) of the surfactant is sorbed.
However, because of their lower toxicity to microor-
ganisms (relative to anionic surfactants), we chose
to examine the effects of enhanced solubilization by
these nonionic surfactants on microbially mediated
transformation reactions of trydrophobic organic
pollutants. Hexachlorobenzene (HCB) was chosen
as a test compound, primarily because of its low
water solubility, 5.0 |0.g/L (6) (i.e., high sediment-
water partition coefficient), and because it is known
to degrade relatively slowly in anaerobic sediments
(7,8).
28
-------
ISoii/Sbdiment Treatment
Anaerobic Dechlorination of HCB
The dechlorination of low levels of
hexachlorobenzene (2.8 x 10"3 mM) incubated in
anaerobic pond sediments (7 percent solids) oc-
curred after an initial lag period of 10 days at a rate
of 5.2 x 10'5 mM/day. Fresh sediments inoculated
with acclimated sediments dechlorinated HCB (2.2
x 10-3 mM) at a slightly faster rate of 6.5 x 10s mM/
day with no initial lag phase. Initially, only one
dechlorination pathway was observed with
pentachlorobenzene (PCB) and 1,2,3,5-tetra chloro-
benzene (1,2,3,5-TTCB) as intermediates, and
1,3,5-trichlorobenzene (1,3,5-TCB) as a possible end
product. After 55 days, a second dechlorination
pathway was observed with 1,2,4,5-TTCB and 1,2,4-
TCB as intermediates, and all three dichlorobenzenes
(DCBs) as possible end products. Both pathways
have been observed during anaerobic incubation of
fresh digester sludge (7). After 75 days, degradation
of HGB is complete, leaving 1,3,5-TCB, 1,2,4,-TCB,
and 1,3-DCB as the major products.. In acclimated
sediments, a shorter delay of 10 days occurs before
onset of the second pathway.
With additions of Tween 80, the
polyoxyethylene sorbitan monooleate surfactant,
the aqueous phase concentration of HCB was in-
creased by one to two orders of magnitude over that
of controls, which were generally at half aqueous
saturation (2 to 3 |xg/L) in the sediment slurry. After
1 week, however, the surfactant was degraded, and
the concentration of aqueous HCB decreased to
control levels. Initially, the rates of HCB dechlorina-
tion and product formation were similar to that of
controls for low levels of Tween 80 (1,500 mg/L).
After 40 days, however, dechlorination ceased. As
Tween 80 concentrations were increased to 5,000
mg/L, initial rates of dechlorination decreased to
almost imperceptible levels. In addition, there was
no evidence of the second pathway.
In acclimated sediments exposed to low levels
of Tween 80 (900 and 1,200 mg/L), rates of HCB
dechlorination are similar to those of controls. At
even lower levels of Tween 80 (300 and 600 mg/L),
dechlorination rates are slightly faster. Both path-
ways are evident in these acclimated sediments
Contrasting the influence of Tween 80 on HCB
dechlorination is the effect of Brij 35, a
polyoxyethylene alcohol surfactant. HCB dechlori-
nation is completely suppressed in acclimated sedi-
ment slurries in the presence of low levels of Brij 35
(1,000 mg/L). In fresh sediment slurries, the surfac-
tant addition results in longer lag periods. Again,
the surfactant appears to be readily degraded, thus
preventing aqueous solubility enhancement effects
from being observed. Currently, another alcohol
surfactant, Brij 30 [C12H2SO(CH2CH2O)4H], is being
tested and shows no signs of being degraded, itself,
to date (after 4 weeks).
References
1. Jafvert, C.T. and J.K. Heath. 1991. Environ.
Sci.Technol. 25:1031-1038.
2. Jafvert, C.T. 1991.
25:1039-1045.
Environ. Sci. Technol.
3. Roy, W.R. and R.A. Griffen.1988. Sur-
' factant-and Chelate-Induced Decontami-
nation of Soil, Report 21. Environmental
Institute for Waste Management Studies,
The University of Alabama.
4. Nash, J.H. 1987. In: Field Studies of In
Situ Soil Washing. EPA /600 /2-87 /100.
U.S. Environmental Protection Agency,
Cincinnati, Ohio.
5. Vigon, B.W. and AJ. Rubin.1989. Journal
W.P.C.F. 61:1233-1244.
6. Chiou, C.T. and D.W. Schmedding. 1982.
Environ. Sci. Technol. 16:4-10.
7. Fathepure, B.Z., J.M. Tiedje, and S.A. Boyd.
1988. Appl. Environ. Microbiol. 54:327-330.
8. Mousa, M.A. and J .E. Rogers. Personal
communication.
TWEEN 80
R,--(CH2CHp))H
R,--(CH,CH,O)yH
R'° OR>
BRIJ 35
Figure 1. Tween 80 and Brij 35 structures.
29
-------
Soil/Sediment Treatment
ANAEROBIC DEGRADATION OF
CHLOROAROMATIC COMPOUNDS
UNDER DIFFERENT REDUCING
CONDITIONS
MM. HSggblom, MJD. Rivera, andL.Y. Young,
Depts.oJMicrobiologyand Environmental Medi-
cine, NYU Medical Center, New York, NY; andJ.
Rogers, U.S. Environmental Protection Agency,
Athens, GA.
Chlorophenol Degradation under Methanogenic
Conditions
If only dechlorination of a chloroaromatic sub-
strate is taking place and the aromatic ring remains
intact, no carbon is provided for microbial growth.
In the environment, carbon sources in addition to a
chloroaromatic are likely to be available, which may
affect the metabolism of the chloroaromatic. In or-
der to examine this, sediment enrichment cultures
were set up with 2,4-dichlorophenol and 4-
chlorophenol under methanogenic condition with
and without the addition of a supplementary car-
bon source. Propionate was chosen as a readily
utilizable carbon source, and para-cresol was used
as a structurally similar nonchlorinated substrate.
2,4-Dichlorophenol was dechlorinated to 4-
chlorophenol without a lag in 25 to 60 days in both
the presence and absence of an auxiliary carbon
source. It took approximately 50 days before the
onset of 4-chlorophenol degradation. Degradation
of 4-chlorophenol was stimulated by either p-cresol
or propionate. In cultures without an auxiliary
carbon source,4-chlorophenol persisted. By repeated
feedings of the chlorophenols and auxiliary sub-
strates, the degradation rates were significantly
enhanced. After 6 feedings of p-cresol and
chlorophenol over a period of 450 days, the rate of
2,4-dichlorophenol and 4-chlorophenol degradation
was enhanced over 10-fold to 40 and lOnmolliter^day1,
respectively.
Dechlorination at the ortfzo-position could be
sustained, and by repeated dilution into fresh me-
dium and refeeding, a stable microbial enrichment
culture free of sediment, which degraded 2,6-
dichlorophenol, was established. 2,6-Dichlorophenol
was sequentially dechlorinated to 2-chlorophenol
and phenol and ultimately mineralized to CH4 and
CO2. On the other hand, the ability to degrade 4-
chlorophenol was lost by transferring the culture
into fresh medium, but, by refeeding, the 4-
chlorophenol degrading culture could be
maintained.
Cultures adapted to 2,4- and 2,6-
dichlorophenol readily dechlorinated other
dichlorophenols containing an ortfzo-chldrine. De-
chlorination of 2,3- and 2,5-dichlorophenol yielded
3-chlorophenol. Dichlorophenols with no ortho-
chlorines persisted. 2,3,6-Trichlorophenol was
dechlorinated at the ori /zo-positian yielding first 2,3-
and 2,5-dichlorophenol, and then 3-chlorophenol.
Similarly, 2,4,6-trichlorophenol was sequentially
dechlorinated to 2,4-dichlorophenol and 4-
chlorophenol. This preferential removal of
orf/zo-chlorines,withmeffl-orpflnj-chlorines removed
at slower rates, appears to be characteristic for
methanogenic cultures.
Chlorophenol Degradation under Sulfidogenic
Conditions
Degradation of chlorophenols under sulfate-
reducing conditions was studied with an estuarine
sediment inoculum (East River). After an initial lag
period of approximately 50 to 100 days, 2-, 3- and 4-
chlorophenol and 2,4-dichlorophenol (0.1 mM) were
completely removed in 120 to 220 days. 4-
Chlorophenol was detected as a transient metabolite
of 2,4-dichlorophenol, but no metabolites of the
monochlorophenols were detected. The rate of
chlorophenol degradation was greatly enhanced
after repeated refeeding of the substrate to the sedi-
ment cultures. In acclimated cultures the
monochlorophenols (0.16 mM) were degraded in 6
to 20 days, corresponding to rates of 8 to 40 nmol-
liter'day1, which are similar to the degradeition
rates in methanogenic cultures. The relative rates of
degradation were 4-chloropheriol > 3-chlorophenol
> 2-chlorophenol, 2,4-dichlorophenol. No degrada-
tion of chlorophenols was observed in sterile
controls.
30
-------
iSoil/Sediment Treatmieht
During degradation of all three
monochlorophenol isomers in the sediment cul-
tures, there was a concomitant loss of sulfate,
corresponding to the stoichiometric values expected
for complete oxidation of the chlorophenol to CO2,
according to the following equation:
C6H5C1O + 3.25 SO42' + 4 H2O —> 6 HCO3' + 3.25 H2S + Ck + 0.5 H*
Formation of sulfide was confirmed with 4-
chlorophenols using a radiotracer technique. No
methane was produced in the cultures, verifying
that sulfate reduction was the main electron sink.
Addition of molybdate, a specific inhibitor of sulfate
reduction, inhibited chlorophenol degradation com-
pletely. These results indicate that chlorophenols
can be mineralized under sulfidogenic conditions
and that oxidation of the chlorophenol is coupled to
sulfate reduction.
The sulfidogenic cultures were propagated by
repeated refeeding of chlorophenols and dilution
into fresh media, and are able to utilize the
chlorophenol as a source of carbon and energy. This
is the opposite of what was observed under
methanogenic conditions, where 4-chlorophenol
degrading cultures could not be subcultured with-
out loss of activity. The sulfidogenic cultures were
very specific and only degraded the
monochlorophenol isomer to which they were accli-
mated. However, all the cultures rapidly degraded
phenol, suggesting that it may be an intermediate in
chlorophenol degradation.
Degradation of Chlorinated Phenols and Benzoic
Acids under Three Reducing Conditions
Anaerobic enrichment cultures, under
methanogenic, sulfidogenic, and denitrifying con-
ditions were established on each of the three
monochlorophenol and monochlorobenzoate iso-
mers with Hudson River sediment from two different
sites (HR1, HR2). In addition, denitrifying cultures
were also established on the same compounds with
East River sediment. Initial results monitoring sub-
strateloss indicated thatallthreemonochlorophenols
and 3-chlorobenzoate were degraded under
methanogenic conditions in HR1 cultures.
Sulfidogenic HR1 cultures were active against all
three monochlorophenols and 3- and 4-
chlorobenzoate. Transient accumulation of phenol
was detected in some of the chlorophenol-amended
cultures under both methanogenic and sulfidogenic
conditions, indicating that reductive dechlorination
is taking place. HR2 cultures, in general, showed the
same activity but at a much slower rate. The rate of
degradation was enhanced with refeeding of the
substrates. Under denitrifying conditions, degrada-
tion of 3- and 4-chlorobenzoate and 2-chlorophenol
was observed in HR1 sediment cultures. These re-
sults indicate that degradation of these chlorinated
aromatic compounds can take place under more
than one reducing condition.
PA/M-HYDROXYBENZOATE AS AN
INTERMEDIATE IN THE ANAEROBIC
TRANSFORMATION OF PHENOL TO
BENZOATE
Barbara R. Sharak Genthner, Technical
Resources, Inc., Environmental Research
Laboratory, U.S. Environmental Protec-
tion Agency, Gulf Breeze, FL.
Summary
Anaerobic transformation of phenol was stud-
ied using a bacterial consortium which transformed
phenol to benzoate without complete mineraliza-
tion of benzoate. Products of monofluorophenol
transformation indicated para-carboxylation. Phe-
nol and benzoate were detected during
para-hydroxybenzoate (p-OHB) degradation. p-OHB
was detected in phenol-transforming cultures con-
taining 6-hydroxynicotinic acid (6-OHNA), a
structural analogue of p-OHB, or at elevated initial
concentrations of phenol (>5 mM), or benzoate (>10
mM).
Results and Discussion
The original phenol-degrading consortium
(2) first transformed phenol to benzoate followed by
complete mineralization of benzoate. Subculture
B-l transformed all of the phenol to benzoate, but
failed to completely mineralize the resulting
benzoate.
31
-------
Soil/Sediment Treatment
Subculture B-l stoichiometricallytransformed
2-fluorophenol (2FP) to 3-fluorobenzoate (3FB) in
the presence or absence of phenol. The rate of 3FB
formation (12.9 moles-lM"1) was similar to the rate of
2FP decline (13.5 molesl'M'1). In the presence of
phenol, a small amount (3 percent) of 3-fluorophenol
was transformed to 2-fluorobenzoate. 4-
Fluorophenol was not transformed in the presence
or absence of phenol. Neither 2-, 3-, nor 4-
fluorobenzoate was used as an energy source. These
results are the same as those obtained with the
original phenol consortium (3). Thus, transforma-
tion was via para-carboxylation as previously
concluded for the original phenol consortium.
parfl-Carboxylation of phenol implies the
formation of p-OHB as an intermediate of transfor-
mation. Therefore, we examined the degradation of
p-OHB by both the original phenol consortium and
subcultureB-1. After3days,p-OHB was completely
degraded in both cultures. Phenol was the Only
compound detected during p-OHB degradation by
the original phenol consortium (Figure la). By
contrast, both phenol and benzoate were detected in
subculture B-l (Figure Ib). After 2 weeks, phenol
was no longer detected in either consortium, but 900
_M benzoate was present in subculture B-l. Thus,
the original phenol consortium had completely de-
graded the phenol formed via decarboxylation of
p-OHB, whereas subculture B-l had subsequently
transformed phenol benzoate. Failure to detect
benzoate during p-OHB degradation by the original
phenol consortium may be a result of rapid benzo-
ate turnover. The detection of benzoate during
p-OHB degradation by subculture B-l may be the
result of diluting out a bacterial species essential for
complete benzoate degradation.
p-OHB was not detected in the original phe-
nol consortium, but small amounts (2 uM) of a
compound with the retention time of p-OHB were
detected in subculture B-l during transformation of
phenol. This suggested thatbenzoate accumulation
inhibited p-OHB breakdown. Consequently, stud-
ies were devised to enhance p-OHB formation using
analogue-, product-, and substrate-inhibition. The
highest concentrations of p-OHB (14 to 43>M) were
detected in the original phenol consortium (data not
shown) and in subculture B-l at initial phenol con-
10 20 30 40
TIME (HOURS)
8 O.OI
10 : 20
TIME (HOURS)
Figure 1. Degradation of para-hydroxybenzoate by
the original phenol consortium (a) and
subculture B-l (b). Symbols: (•) phenol;
(A) para-hydroxybenzoate; (•) benzoate.
centrations >5 mM (Figure 2a). The intermediate
was also detected (6 to 8 >M) in the presence of >100
avi 6-OHNA or >10 mM benzoate (Figures 2b and
2c, respectively).
The intermediate in subculture B-l grown with
10.5 mM phenol cochromatographed with authen-
tic p-OHB under the two isets of separation
parameters. Its UV spectrum matched that of au-
thentic p-OHB. Its identity was confirmed by GC/
MS analysis. The mass spectrum matched that of
the TMS-derivative of authentic p-OHB with peaks
at m/e 282 and m/e 267 (-CH3). Neither 2- nor 3-
hydroxybenzoate was detected during this analysis.
Identifying p-OHB in cultures that were transform-
ing phenol provided a direct indication of
para-carboxylation. Since p-OHB was not detected
in the original phenol consortium with initial phe-
nol concentrations <5 mM, it is possible that p-OHB
was not detected in other studies (1,4,5) because of
the use of lower phenol concentrations (>2 mM).
32
-------
iSoil/Sediment Treatment
10
(1—0.
Ld
s
M
LJ
CD
B
0 O.
25 50 75
TIME (DAYS)
100
30
20
10
l
0
30
20
10
0
30
20
10
0
NI
2
.•ft--
Figure 2. Detection of para-hydroxybenzoate (A) as
an intermediate in the transformation of
phenol(•) to benzoate (•) in the presence
of (a) 10 m M phenol, (b) 6-hydroxynicotinic
acid (1 mM), or (c) 10 mM benzoate.
References
1. Knoll,G.andJ.Winter. 1989. Appl.Environ.
Biotechnol. 30:318-324.
2. Sharak Genthner, B.R., G.T. Townsend, and
P.J. Chapman. 1989. Biochem. Biophys.
Res.Gomm. 162:945-951.
3. Sharak Genthner, B.R., G.T. Townsend, and
P.J. Chapman. 1990. Biodegradation 1:65-74.
4. Zhang X., T.V. Morgan, and J. Wiegel. 1990.
FEMS Microbiol. Lett. 67:63-66.
5. Zhang, X. and J. Wiegel. 1990. Appl.Environ.
Microbiol. 56:1119-1127.
AEROBIC DEGRADATION OF
POLYCYCLIC AROMATIC
HYDROCARBONS
Peter J. Chapman, U.S. Environmental
Protection Agency, U.S. EPA Environmen-
tal Research Laboratory Gulf Breeze, FL.
Polycyclic aromatic hydrocarbons (PAHs) are
chemicals possessing three or more aromatic rings
found in complex mixtures in fossil fuels and their
combustion products. Because a number of these
chemicals are of concern to human health, their
removal from contaminated sites is an important
priority for remediation.
Many studies (reviewed in references 1, 2
and 3) have documented the abilities and mecha-
nisms by which different axenic bacterial cultures
degrade individual chemicals of this class, particu-
larly those of lower molecular weight. However,
the widespread occurrence of these chemicals as
constituents of complex PAH mixtures, often to-
gether with chemicals of other types, requires an
assessment of the abilities and conditions under
which microorganisms can best effect degradation
of mixtures of these chemicals, many of which have
very limited aqueous solubility. Laboratory studies
were undertaken to study the biodegradation of
PAHs, as they occur in crude fossil fuels, and their
products and wastes. In developing appropriate
approaches and methods for studying biodegrada-
tion of mixed chemicals, it was anticipated that
research would also lead to the isolation of PAH-
degrading bacteria of value in other areas of
biodegradation research. Initially, the neutral frac-
tion of coal tar creosote, which contains all of the
PAHs (Table 1) found in this lumber preservative
(approximately 85 percent by weight of original
creosote) together with a number of neutral oxygen-
and sulfur-containing heterocycles, was chosen for
study of its microbial degradation.
The plan of this study was to determine how
readily various PAH constituents were degraded
by aerobic bacterial cultures, which had been en-
riched by growth in a mineral salts medium
containing the neutral creosote fraction as a carbon
and energy source. It was anticipated that this
33
-------
Soil/Sedi inent Treatment
Table 1, Predominant polycydic aromatic
hydrocarbons in coal tar creosote*.
Compound
Coal Tar Creosote
(% of Total PAHs (range)
Naphthalene
2-MeNaphthalene
1-McNaphthalene
Blphcnyl
2,6-DiMoNaphthalene
2,3-DiMeNaphthalene
Acenaphthene
Fluorcnc
Phenanthrcne
Anthracene
2-Me Anthracene
Anthraquinone
Fluoranthene
Pyrene
2,3-Benzotb]Fluorene
Chrysene
BenzolaJPyrene
3.0 - 15.8
2.1 - 14.2
2.1 - 14.2
2.3-2.8
2.0-2.3
2.0-2.4
4.1-9.0
9.6-10.0
4.6-21.0
1.5-2.0
0.5-2.6
0.1 - 1.0
6.8 - 10.4
2.2-8.5
2.0-4.6
2.8-3.0
0.1 - 1.0
"For sources see reference 6.
approach would facilitate establishment of differ-
ent cultures that could be isolated for their ability
to utilize individual aromatic hydrocarbons and
that could be used in defined mixed cultures to
simulate the degradative performance of unde-
fined enrichment cultures. In this way,
contributions to the overall process of PAH degrar
dationby individual axenic cultures, singly and in
different combinations, could be assessed.
Phenols and basic N-heterocycles were re-
moved from crudecreosotebysuccessiveextraction
of its methylene chloride solution with sodium
hydroxide and sulfuric acid. After washing, dry-
ing, and solvent removal, the resulting creosote
neutral fraction (CNF) was added (0.1%) to a min-
eral salts medium (4) also containing dimethyl
sulfoxide (DMSO) at 0.02% to enhance the
bioavailabilityof chemicals withlow aqueous solu-
bility. Soils from two contaminated lumber
treatment sites (Live Oaks, Florida, and American
Creosote Works, Florida) were used as sources of
microorganisms. Mixed soil samples were used to
inoculate shake-flask cultures (25°C, 200 rpm) for
enrichment of CNF-degrading bacteria. Methyl-
ene chloride extraction of the entire flask contents
after 14 days of incubation followed by capillary
GC-FID analysis (5) of the concentrated extracts
showed that, once established, sequential enrich-
ment cultures effected significant and reproducible
breakdown of a large proportion (70 percent) of
CNF constituents, principally those having three
rings or less. Compounds with four or more rings,
e.g., pyrene or fluoranthene, were not significantly
affected (Figure 1). By comparison, 14-day incuba-
tions of uninoculated flasks showed losses only of
the more volatile constituents such as naphthalene.
A number of axenic cultures were isolated from
these enrichment cultures, and also directly from
the site soils, by using individual aromatic hydro-
carbons as selective substrates. Microorganisms
able to utilize biphenyl, naphthalene, and phenan-
threne were obtained from enricliment cultures while
acenaphthene-, fluorene-, pyrene-, and fluoranthene-
' utilizing isolates were isolated directly from site
soils. Organisms isolated from the enrichment cul-
tures were shown to degrade their respective growth
substrates when cultured in CNF-mineral salts me-
dia. By contrast, organisms obtained directly from
soil were generally devoid of action towards their
growth substrates when added to CNF-medium
either singly or in combination with others. By
systematically assembling defined mixtures of iso-
lates, it was shown that the performance of the
undefined enrichment cultures could be simulated
by a five-member mixed culture (Figure 2). At-
tempts to construct defined cultures with fewer
microbial members resulted in more limited degra-
dation of aromatic hydrocarbons, such as
acenaphthene, whether the organisms omitted were
biphenyl-, naphthalene-, or phenanthrene-utilizers.
These and other observations s uggest that in mixed
cultures, cooxidation may be an important mecha-
nism of degradation of certain constituents of CNF
and that consequently the rates of degradation of
such chemicals will depend upon the numbers and
activities of cells whose growth is supported by
other chemicals.
Evaluating the extent to which enrichment
and defined cultures employ cooxidation processes
and accumulate cooxidation products is continu-
ing, as is work to establish and define new biological
systems that degrade the more recalcitrant four-
and five-ring PAHs and higher molecular weight
members of this series.
34
-------
Soil/Sediment Treatmdrtt
Acknowledgements
Technical assistance was provided by Walter
Randall and Sol Resnick, Technical Resources, Inc.
(TRI). Maureen Downey and Beat Blattman (TRI)
assisted with chromatographic analyses, as did
Suzanne Lantz who together with Jim Mueller (both
of SBP Technologies, Inc.), provided valuable dis-
cussion and insights.
References
1. Gibson, D.T. and V. Subramanian. 1.984.
Microbialdegradationofaromatichydrocar-
bons. In: D.T. Gibson, ed.,Microbial Degra-
dation of Organic Compounds. Marcel
Dekker, Inc. pp. 181-252.
2. Gibson, D.T. 1987. Microbial catabolism of
aromatic hydrocarbons and the carbon cycle.
In: S.R. Hagedorn, R.S. Hanson, and
D.A. Kunz, Microbial Metabolism and
the Carbon Cycle. Harwood Academic
Pub. pp. 33-58.
3. Cerniglia,C.E. 1984. Microbial transforma-
tion of aromatic hydrocarbons. In: R.M.
Atlas, ed., Petroleum Microbiology.
MacMillanPub. Co. pp. 99-128.
4. Hareland, W.A., R.L. Crawford, P.J.
Chapman, and S. Dagley. 1975. Metabolic
function and properties of 4-
hydroxyphenylacetic acid 1-hydroxylase
from Pseudomonas acidovomns. J. Bacteriol.
121:272-285.
5. Mueller, J.G., S.E. Lantz, B.O. Blattman, and
P.J. Chapman. 1991. Bench-scale evalua-
tion of alternative biological treatment pro-
cesses for the remediation of pentachloro-
phenol- and creosote-contaminated materi
als: solid phase bioremediation. Environ.
Sci. Technol. 25:1045-1055.
6. Mueller, J.G., P.J. Chapman and P.H.
Pritchard. 1989. Creosote contaminated
sites: their potential for bioremediation.
Environ. Sci. Technol. 23:1197-1201.
Capillary GC / FID Analysis of Creosote Neutrals
Action of Undefined Mixed Culture
T = 0
T = 14D (BIOL.)
TIME (mln)
Figure 1. Action of undefined enrichment culture on
creosote neutrals after 14 days of incuba-
tion (lower trace) compared to starting
material (upper trace) and 14 days incuba-
tion without incubation (middle trace).
Capillary GC / FID Analysis of Creosote Neutrals
Action of Defined Mixed Culture (5 Isolates)
T = 14D (NON BIOL.)
Undefined Mixed Culture
= 14D
Figure 2. Action of defined mixed culture (5 organ-
isms) on creosote neutrals after 14 days
(lower trace) compared with 14 day incuba-
tionof undefined enrichmentculture (middle
trace) and starting material (upper trace).
35
-------
SECTION FOUR
COMBINED TREATMENT
Most hazardous waste sites contain complex mixtures of persistent organic and inorganic contaminants that can be
deaneduponly by a combination of treatment techniques. Researchers are developingmethods to combine various physical,
chemical and biological treatment technologies, and comparing the effectiveness of the various combinations. For example,
a chemical treatment— adding potassium polyethylene glycol (KPEG) to the soil at a site — may be used to dechlorinate
PCBs. Then, biological treatment, which is more effective after dechlorination, can be used to complete the soil restoration.
In one project, researchers are studying the degradation of KPEG-treated PCBs in both aerobic and anaerobic primary
enrichment cultures. Current data support the possibility of biodegradation ofPCB-KPEG products as part of a combined
diemicalfbiological treatment of PCBs, at least under aerobic conditions.
Combined treatment is also being examined as an alternative to conventional treatment methods for the destruction
of volatile organic compounds (VOCs). Researchers are studying the effectiveness of fixed-film anaerobic biological
processes for treating and decontaminating leachates containing synthetic organic chemicals (SOCs). This process can
result in less air stripping of VOCs than with aerobic processes. Two types of anaerobic pretreatment processes are being
evaluated: an upflaw anaerobic filter reactor and a GAC anaerobic fluidized bed reactor. :
Researdiers also conducted bench-scale tests of solid-phase (land treatment) and slurry-phase bioremediation of
pentachlorophenol-and creosote-contaminated sediment and soil. The data indicate that slurry-phase bioremediation can
beusedeffectivelytotreatmaterialscontaminatedwithcreosoteandpossiblyPCP.Slurry-phasebiotreatmenttechnologie^
have been integrated into a multi-phasic treatment process, currently under evaluation at the pilot-scale level.
BACTERIAL DEGRADATION OF KPEG-
MODIFIED PCBS IN ANAEROBIC
AND AEROBIC ENRICHMENT
CULTURES
Wale Adewumni and Joseph A. Krzycki, De-
partment of Microbiology, Ohio State
University, Columbus, OH.
In the 1960s and 70s it became apparent that
polychlorinated biphenyls (PCBs) were recalcitrant
to biological degradation, and were accumulating
in organisms throughout the food chain. Subse-
quently, PCBs were phased out of use; but now it is
estimated that at least a half million tons of PCB
currently exists in landfills and closed systems await-
ing detoxification (1). Destruction is accomplished
by transport of contaminated materials to incinera-
tors. An alternative method is treatment of
contaminated materials with a formulation of KOH
and polyethylene glycol,400 (KPEG). KPEG attacks
chlorine-bearing carbons, resulting in the formation
of chloroarylpolyglycols (2,3,7). These compounds
resemble PEG-based nonionic surfactants, arid it
appeared probable that PCB reacting with KPEG
would be rendered soluble in water. Considerations
such as these led to the suggestion that KPEG treat-
ment coupled with biological degradation could be
an alternative method of PCB disposal. We have
been testing the feasibility of this idea by studying
metabolism of a KPEG-treated PCB congener in
both aerobic and anaerobic primary enrichment
cultures.
Reaction of KPEG with a polychlorinated
biphenyl may yield numerous products, since nu-
cleophilic attack of KPEG can occur at several
different chlorine-bearing carbons and result in a
mixture of chloroarylpolyglycols with different de-
grees of PEG substitution, as well as small amounts
36
-------
Combined Treatment:
of chlorobiphenylols (2,3). Therefore, we elected to
carry out our studies using uniformly 14C-labeled
2/2'/4/4'/5/51 hexachlorobiphenyl (14C-HCBP) in or-
der to facilitate identification of the products of the
initial chemical reaction, as well as biodegradation
products. This congener is 8.2 percent of Arochlor
1260 and 3.3 percent of Arochlor 1254 (4), and has
been identified as a major congener found in both
human adipose tissue (5) and milk (6).
KPEG was formulated as described by
Kornell and Rogers (7) except that 1.3 mol of KOH
was dissolved with heating into 1.0 mol of PEG 400.
The mixture added to 14C-HCBP absorbed to glass
at a ratio of 100 mg congener to 1 mL KPEG. The
reaction was incubated at 84°C and quenched by
addition of water. Figure 1 details a timecourse of
the reaction which demonstrates that KPEG effects
a rapid phase transfer of 14C-HCBP.
After the reaction had proceeded for 60 min-
utes, 95 percent of recovered radioactivity had been
converted to a form that remained in the aqueous
phase following repeated.extractions with hexane.
The products formed from 14C-HCBP after a 48-hr
reaction with KPEG remained soluble in water. In
control experiments where 14C-HCBP was incu-
bated with PEG, rather than KPEG, phase transfer
was not observed and the congener remained hex-
ane soluble. This experiment indicated that KPEG
reaction did convert HCBP to a water- soluble form,
presumably the chloroarylpolyglycols first de-
scribed by Brunelle and Singleton (3). Reverse-
phase HPLC analysis (Figure 2) using an instru-
ment equipped with an on-line radioactivity
detector demonstrated increasing product com-
plexity with time. Two major water -soluble
products were formed within minutes; at later times,
these were converted to forms which eluted rela-
tively quickly from the reverse- phase column.
These products accumulated and, with a 12-hr re-
action time, six products from the aqueous phase
could be resolved. These ma)^ represent
chloroarylpolyglycols, with higher degrees of PEG
substitution. On the basis of these experiments, a
reaction time'of 1 hour was chosen in order to
minimize product homogeneity in the subsequent
biodegradation studies.
4000
g
3000-
2000-
1000 -
to—D -
""" ... Hexane
******-..
0 10 20 30 40 50 60
Time "(min)
Figure 1. Phase transfer of HCBP during reaction
with KPEG. Vials containing 100 mg of
14C-HCBP(20mCi/mg)andl mL of KPEG
reagentwere incubated at84°C before the
reaction was quenched with water at the
indicated times.
0.04
0.03-
^230
0.02-
0,01-
1500
I- 1000
CPM
-500
20
40 60
Time (min)
80
100
0.04
0.03-
0.02-
0.01-
600
- 400
CPM
- 200
20 40 60 80
Time (min)
100
Figure 2.14C-HCBP/KPEG products after 1 and 12
hours of reaction time analyzed on a
silica-based C-18 HPLC column eluted at
0.35 mL/min with a 30 to 50 percent
isopropanol gradient.
37
-------
Com hi iled Treatment
Biodegradation experiments were initiated
with anaerobic sewage sludge taken from the Jack-
son Pike municipal waste treatment plant
(Columbus, OH). MC-arylpolyglycols were incu-
bated with either undiluted sludge or primary
enrichmentcultures. Enrichment cultures were made
using a mineral medium and incubated under a
nitrogen headspace. The KPEG/14C-HCBP reaction
mixture was neutralized with HCL and added to
cultures to final concentrations of 0.2 mM total
arylpolyglycols and 25 mM PEG. Enrichments were
set up in duplicate under methanogenic (no addi-
tions), sulfidogenic (15 mM Na2SO4), iron-reducing
(0.1 percent amorphous iron oxide), or denitrifying
(20 mM KNO3) conditions. Controls were estab-
lished for each condition investigated by inoculating
cultures with autoclaved sludge. The fate of the
labeled arylpolyglycols was followed by periodic
ethyl acetate extraction of aliquots from each cul-
ture. Both arylpolyglycols and unreacted 14C-HCBP
could be extracted using this solvent. The ethyl
acetate extract was then evaporated, and partition-
ing of radioactivity in the residue between hexane
and water determined. Figure 3 shows the average
results obtained under methanogenic conditions for
active cultures and controls. At the start of incuba-
tion, 80 percent of expected counts could be
recovered in the ethyl acetate fraction, 19 percent
was left in the culture supernatant. Of radioactivity
in the ethyl acetate fraction, only 4 percent was
s oluble in hexane, the remainder was in the aqueous
fraction. After incubation for 28 days at 37°C, 71
percent of expected counts could be recovered by
ethyl acetate extraction of the cultures. However, 60
percent of radioactivity in the ethyl acetate fraction
was now soluble in hexane; the remaining radioac-
tivity was still soluble in water. In contrast, killed
controls showed no change in the phase distribution
of label. Similar results were obtained with enrich-
ment cultures established under sulfate, nitrate, or
iron-reducing conditions, as well as with 14C-
chlorobiphenylpolyglycols incubated in undiluted
sludge. These results indicated that in this anaerobic
sludge, bacteria were capable of converting
chlorobiphenylpolyglycols into a water-insoluble
form. A possible mechanism is hydrolysis of the
PEG moiety from the aromatic ring, leaving a rela-
tivelyinsolublepolychlorobiphenylol.Ananalogous
reaction has been documented with nonoxynol (Tri-
ton N), a PEG-based nonionic surfactant (8).
Biodegradation in sewage sludge resulted in cleav-
age of the PEG moiety from nonoxynol and liberation
of the recalcitrant, hydrophobic compound,
nonylphenol.
Biodegradation of the labelled arylpolyglycols
was also tested in aerobic primary enrichment cul-
tures. Metabolism of the compounds did occur;
however, unlike anaerobic cultures, the products
remained water soluble. Aerobic primary enrich-
ments were established in triplicate sealed 160 mL
vials which were periodically replenished with oxy-
gen. Each culture received 0.1 mM labeled
arylpolygycol and 12 mM PEG. Duplicate enrich-
mentcultures were also established with autoclaved
inocula as controls. Results are illustrated in Figure
200
aqueous
hexane
not extracted
69
Figure 3. Phase transfer of 14C-HCBP/KPEG reac-
tion products during incubation in
methanogenic enrichment cultures. The
partitioning of extracted radioactivity
from 0.5 mL of culture between hexane
and water is shown. The amount of radio-
activity not extracted from the culture by
ethyl acetate is also indicated.
38
-------
Combined Treatment I
4. At the start of incubation, an average of 66 percent
of total radioactivity added to active cultures could
be recovered by ethyl acetate extraction, while 16
percent was not extractable and remained in the
culture supernatant. In the ethyl acetate fraction, 90
percent of radioactivity was water soluble. After 41
days of incubation, only 19 percent of radioactivity
added to the culture could be recovered by ethyl
acetate extraction, while 57 percent was not re-
moved from the cultures after repeated extraction
with ethyl acetate. The amount of radioactivity that
was soluble in both ethyl acetate and hexane did not
increase. This indicated that aerobic biodegradation
of the labeled arylpolyglycols had resulted in prod-
ucts that were soluble in water but no longer soluble
in ethyl acetate.
DAY
80
60-
40H
20-
Active
culture
E§ja aqueous
HU organic
Ifll not extracted
43
DAY
Figure 4. Phase partitioning between hexane and
water of 14C-HCBP/KPEG reaction prod-
ucts extractable by ethyl acetate after
incubation for 43 days in aerobic enrich-
ment cultures and killed controls. "Not
extracted" refers to radioactivity leftin the
culture sample after four sequential 1 mL
extractions with ethyl acetate.
0.08
1200
0.06 - i
A230
0.04 -
0.02 -
0 20 40 60 SO 100
0.08
0.06 -
A230
0.04 -
0.02-
Autoclaved
control
800
CPM
•^400
20 40 60 80
Time (min)
100
Figure 5. Reverse phase HPLC of aerobic culture
supernatants after 41 days of incubation.
The column was eluted with a 30 to 50
percent isopropanolgradientat 0.25mL/min.
Reverse-phase HPLC analysis of the aqueous
supernatant of cultures (Figure 5) confirmed that
metabolism of the products occurred. Killed con-
trols had radioactive HPLC elution profiles similar
to those of the starting material; however, the su-
pernatant of active cultures no longer contained
any of the KPEG/14C-HCBP reaction products. In-
stead, nearly all the radioactivity present in the
sample eluted immediately from the column. Using
a different HPLC methodology, we have identified
two major radioactive peaks with absorbance at 230
nm, indicating the breakdown products still con-
tain an aromatic ring. Currently, we are determining
if these compounds are end products or transient
intermediates. They may represent products in
which a PEG-substituted aromatic ring has been
cleaved following hydrolysis of the PEG moiety
from the ring, resulting in a chlorinated aromatic
carboxylic acid.
39
-------
Combined Treatment
In summary, we have examined both aerobic
and anaerobic degradation of the arylpolyglycols
produced by KPEG reaction with 2,2',4,4?,5,5'
hexachlorobiphenyl in enrichment cultures estab-
lished with inocula froma municipal waste treatment
plant Under anaerobic conditions, the PEG moiety
was apparently cleaved from the chlorq-
arylpolyglycols, resulting in hydrophobic products.
In contrast, aerobic enrichments carried out degra-
dation of the ethyl acetate and water- soluble
chloroarylpolyglycols into two water- soluble, but
ethylacetateinsoluble,aromatic compounds; possi-
bly chlorinated arylcarboxylic acids. Our data
support the possibility of biodegradation of PCB/
KPEG products as part of a combined chemical/
biological treatment of PCBs. However, the process
may be feasible under only aerobic, and not
anaerobic, conditions.
References
1. Reineke,W.andH.-J.Knackmuss. 1988.
Ann. Rev. Microbiol. 42:263.
2. Brunelle, D.J. and D.A. Singleton. 1983.
Chemosphere 12:183.
3. Brunelle, D.J. and D.A. Singleton. 1985.
Chemosphere 14:173.
4. Albro, P.W.,J.T. Corbett and J.L. Schroeder.
1981. J. Chromatog. 205:103.
5. Jensen, S. and G. Sundstrum. 1974. Ambio
3:70.
6. Safe, S., L. Safe, and M. Mullin. 1985. J.
Agric. Food Chem. 33:24.
7. Kornel, A. and C. Rogers. 1985. J. Hazard.
Mater. 12:161.
8. Stephanou,E.andW.Giger. 1982. Environ.
ScLTechnol. 16:800.
ONSITE BIOLOGICAL PRETREATMENT
FOLLOWED BY POTW TREATMENT OF
CERCLA LEACHATES
E.R. Krishnan,R.C. Haught,andM.L. Taylor,
PEI Associates, Inc., Cincinnati, OH; ALT.
Suidan and M. Islam, Dept. of Civil and Envi-
ronmental Engineering, University of Cincin-
nati, Cincinnati, OH; and R.C. Brenner, U.S.
Environmental Protection Agency, Cincinnati,
OH.
The objective of this research is to assess the
effectiveness of fixed-film anaerobic biological pro-
cesses in treating and decontaminating leachates
containing synthetic organic chemicals (SOCs) that
may be regulated under the Comprehensive Envi-
ronmental Response, Compensation, and Liability
Act (CERCLA). This is an attractive proposition
because anaerobic processes result in less air strip-
ping of volatile organic compounds compared to
aerobic processes due to their lower gas production
rates. Anaerobic pretreatment processes are also
expected to reduce problems associated with in-
complete degradation of chlorinated compounds as
well as pass-through of semivolatile organic com-
pounds, which can occur when CERCLA leachates
are discharged without any pretreatment to pub-
licly owned treatment works (POTWs). Two types
of anaerobic pretreatment processes are being evalu-
ated in this study: an upflow anaerobic filter reactor
and a granular activated carbon (GAC) anaerobic
fluidized-bed reactor. The tests are being conducted
at U.S. EPA's Test and Evaluation (T&E) Facility in
Cincinnati, Ohio.
Leachate Characteristics
The leachate for the experiments is obtained
from a large commercial municipal landfill in
Georgetown, Ohio. The leachate is highly variable
in composition (chemical oxygen demand
[COD ] levels ranging from 300 to 2,500 mg/L) with
low to moderate levels of biodegradability. Sulfate
concentrations in the leachate range from 3 to
300 mg/L. Due to its relatively low biodegradable
content during the first 6 months of the project, the
leachate was supplemented with a mixture of acetic,
40
-------
Combined Treatment
propionic, and butyric acids to increase its total
COD to approximately 1,600 mg/L. Later, the total
COD of the leachate increased significantly to a
maximum of 2,500 mg/L, at which time, volatile
acids addition was discontinued. The leachate was
fed without any addition of volatile acids during the
remainder of the project. During most of the latter
part of the project, the total COD of the raw leachate
remained relatively stable, ranging from 400 to
1,000 mg/L. The leachate is rendered hazardous by
supplementing it with a mixture of 10 volatile and 4
semivolatile organic compounds, shown with their
corresponding target concentrations in Table 1.
Chloroform was not added to the leachate until the
later stages of the project because of its potential
toxicity.
Treatment Systems
The treatability of the leachate is being evalu-
ated in three parallel trains. One train consists of
leachate pretreatment in a bench-scale upflow
anaerobic filter reactor (6-in. diameter x 48-in. high)
packed with 1-in. Pall rings, followed by mixing
with raw municipal wastewater in a ratio of 95
percent wastewater to 5 percent leachate and treat-
ment in a bench-scale activated sludge POTW unit.
The second train is similar in scale to the first, with
the exception that a fluidized-bed reactor (4-in. di-
ameter x 42-in. high) filled with 16 x 20 U.S. mesh
G AC is used for leachate pretreatment instead of the
upflow filter reactor. The third treatment train
consists of a pilot-scale anaerobic filter reactor (4.25-
ft diameter x 7.5-ft high) followed by a pilot-scale
activated sludge POTW unit. The objective of this
process train is to evaluate the scale-up of anaerobic
filters and to observe potential problem areas such
as bed plugging and wall effects. The anaerobic
pretreatment systems are operated at 35°C. The
empty bed contact time in the GAC fluidized-bed
reactor is maintained between 6 and 8 hours, while
a longer detention time of 48 to 96 hours is employed
in the anaerobic filter reactors.
Results
The COD removal efficiency in the bench-
scale anaerobic filter and GAC fluidized-bed system
averaged 42 and 48 percent, respectively. The aver-
age influent COD was about 1,100 mg/L. The COD
removal efficiency for the systems was found to
increase with increasing influent COD. During the
period of the volatile acids addition,.the primary
COD removal mechanism was methanogenic. Af-
ter the volatile acids addition was stopped and the
COD of the leachate decreased, the COD removal
mechanism was due to a combination of
methanogenesis and sulfate reduction. The average
sulfate reductions in the GAC fluidized-bed and
anaerobic filter reactors were 71 and 65 percent,
respectively, corresponding to .an average influent
sulfate concentration of 116 mg/L. In the bench-
scale anaerobic filter reactor, most of the organic
compounds were removed in excess of 90 percent.
Chlorobenzene, ethylbenzene, 1,1-dichloroethane,
and trichlorobenzene exhibited gradual
degradation, indicating the need for longer acclima-
tion periods. In the bench-scale GAC fluidized-bed
reactor, all organic compounds exhibited removal
efficiencies exceeding 95 percent with the exception
of 1,1-dichloroethane, which required a longer ac-
climation period. Despite some differences in the
design and operation of the pilot-scale anaerobic
filter-reactor compared to the bench-scale anaerobic
filter reactor, the performance of both systems was
similar prior to chloroform addition to the leachate.
Within 3 weeks after the addition of chloroform,
however, thepilot-scale system showed a decline in
the removal of some of the SOCs (including chloro-
form). SOC removals continued to decline over a
period of 4 months, at which time chloroform addi-
tion was discontinued. The pilot-scale system is
now being monitored to determine whether perfor-
mance will return to prechlorofbrm addition levels
while the SOC consortium (excluding chloroform)
continues to be added to the leachate feed.
41
-------
Combihed Treatment
Table 1. Composition of SOC supplement to
the leachates.
Compound Concentration ( g/L)
Volatile Organic Compounds
Acetone 10,000
Methyl Ethyl Ketone 5,000
Methyl Isobutyl Ketone 1,000
Trichloroethylene 400
1,1-Dichloroethane 100
Methylene Chloride 1,200
Chloroform 0 to 2,000
Chlorobenzene 1,100
Ethylbenzene 600
Toluene 8,000
Scmivolatilc Organic Compounds
Phenol 2,600
Nitrobenzene 500
1,2,4-Trichlorobenzene 200
DibutylPhthalate 200
AEROBIC BIODEGRADATION OF
CREOSOTE
James G. Mueller, Suzanne E. Lantz, Ron L.
T}iomas,and Ellis L. Kline, Southern BioProd-
nciSflnc., Sabinelsland, Gulf Breeze, FL; Peter
J. Chapman, Douglas P. Middaugh, and P. Hap
Pritchard, U.S. Environmental Protection
Agency, Environmental Research Laboratory,
Saline Island, Gulf Breeze, FL; and Richard J.
Colvin, Allan P. Rozich, and Derek Ross, En-
vironmental Resources Management, Inc.,
Exton,PA.
Summary
Performance data on solid-phase (land farm-
ing) and slurry-phase bioremediation of
pentachlorophenol- (PCP-) and creosote-contami-
nated subsurface soil (sediment) and surface soil
were generated at the bench-scale level (1,2). Soil
samples from slurry reactors and from specially
designed land farming chambers were extracted
and analyzed by gas chromatography for PCP and
42 monitored creosote constituents to delineate the
activity of indigenous microorganisms. Changes in
microbial biomass were also recorded. In general,
slurry-phase bioremediation resulted in rapid and
extensive removal (3 to 5 days to biodegrade >50
percent of targeted compounds) of monitored con-
stituents. However, removal rates from surface soil
slurries were slower than those observed with sub-
surface soil slurries, and removal was generally
confined to the more readily biodegradable, lower-
molecular-weight compounds. In all cases,
solid-phase bioremediation was much less effec-
tive. The general order of biodegradation was
phenol ics> low-molecular-weight
PAHs>heterocycles>high-molecular-weight
PAHs=PCP. These data suggest that slurry-phase
bioremediation strategies can be effectively em-
ployed to treat creosote-contaminated, and possibly
PCP-contaminated, materials. The efficiency of an
integrated, multi-phasic slurry treatment process is
currently being evaluated at trie pilot-scale level.
Results and Discussion
Surface Soil (SS) Bioremediation. To sim-
plify data presentation, PAHs were arbitrarily
divided into three groups: groups 1,2, and 3 consist
of PAHs containing two,three, and four or more
fused rings, respectively. In the absence of inor-
ganic nutrient supplements (SS- treatment),
biodegradation of phenolic, heterocyclic, and lower-
molecular-weight PAH constituents of creosote was
most readily apparent (for example, see Figure 1).
Biodegradation of the more persistent chemicals
was less extensive. Biodegradation of PCP began
' after a 2-week lag period, ultimately resulting in the
removal of approximately 70 percent of this chemi-
cal over the course of the study (90 days).
When surface soils were .amended on a weekly
basis with inorganic nutrients (SS+ treatment), the
rate of biodegradation of monitored chemicals was
accelerated. As observed in the unamended soils,
biodegradation of phenolics, heterocyclics,, and
low-molecular-weight PAHs was mostrapid. More-
over, the extent of biodegradation of the more
persistent chemicals (i.e., group 3 PAHs, PCP) was
increased.
Biodegradation of all monitored creosote
constituents was most rapid and extensive under
slurry-phase conditions (2). Within 7 days of slurry
incubation, 91, 90, 45, 47, and 30 percent of the
42
-------
Combined Treatment!
phenolics, group 1 PAHs, heterocyclics, group 2
PAHs, and group 3 PAHs, respectively, were biode-
graded. However, PCP was not biodegraded by
indigenous microflora established in the reactor.
Little change in the total amount of chemical biode-
graded was apparent with continued incubation (30
days). This suggests that conditions for biodegrada-
tion became limiting (e.g., nutrient limitation,
accumulation of bacteriotoxic metabolites, etc.).
Alternatively, depletion of the readily biodegrad-
able carbon sources (i.e., group 1 PAHs, phenolics)
prevented further catabolism of monitored
chemicals.
Subsurface Soil (SBS) Bioremediation. Sub-
surface soils recovered from a depth of 5 m beneath
the highly contaminated, capped solidified sludge
material present at the American Creosote Works
Superfund site, Pensacola, Florida, contained ap-
proximately 7 percent (by weight) unweathered
creosote plus PCP (3). Initial microbial population
estimates showed that this material was essentially
sterile (data not shown). This was presumably due
to the high organic loading rate and the presence of
fly-ash (added to stabilize the above-lying sludge),
which resulted in a soil pH of 10 to 11. Hence,
biodegradation of these chemicals during aerobic,
solid-phase bioremediation was slow to initiate.
After a 1- to 2-week lag phase, however, readily
biodegradable chemicals (i.e., phenolics, heterocy-
clics) were removed, but a majority of the other
compounds resisted biological attack. The addition
of inorganic, soluble nutrients had little effect on the
rate and extent of biodegradation.
As observed with surface soil, biodegradation
occurred more rapidly, and was more extensive,
duringslurry-phasetreatmentthanwithsolid-phase
treatment (2). Within 7days of slurry incubation, 95,
90, 85, 65, and 50 percent of the group 1 PAHs,
phenolic, group 2 PAHs, heterocyclics and group 3
PAHs, respectively/were biodegraded. As before,
PCP was not degraded by the microbial community
established in the reactor.
Based on these and other data, slurry-phase
biotreatment technologies have been integrated into
a multi-phasic remediation strategy to ameliorate
soil and water contamination by creosote, PCP, and
related wastes (4). The ability of this system to
remove >90 percent of monitored chemicals from
contaminated wastes has been demonstrated at the
bench-scale level (3, 5). In association with the
Superfund Innovative TechnologyEvaluation (SITE)
Program, pilot-scale performance data are currently
being generated at the American Creosote Works
Superfund Site, Pensacola, Florida.
Acknowledgments
Technical assistance was provided by Beat
Blattmann, Maureen Downey, Mike Shelton, and
Miriam Woods (Technical Resources, Inc.). Susan
Franson (U.S. EPA, EMSL,,Las Vegas, NV) gra-
ciously offered a QA/QC review of these'studies.
Dan Thoman (U.S. EPA, BSD, Athens, G A) obtained
subsurface soil samples and performed indepen-
dent chemical analyses. Assistance from Natalie
Ellington and Beverly Houston (U.S. EPA, Region
IV) is also gratefully acknowledged.
Financial support for these studies was pro-
vided by the U.S." EPA Superfund Program (Region
IV). The ongoing Pilot-Scale Technology Demon-
stration Project is supported by the U.S. EPA SITE
Program, Cincinnati, Ohio.
This work was performed as part of a Coop-
erative Research and Development Agreement
between the Gulf Breeze Environmental Research
Laboratory and Southern Bio Products, Inc. (At-
lanta, G A) as defined under the Federal Technology
Transfer Act, 1986 (contract no. FTTA-003).
References
1. Mueller, J.G.,S.E. Lantz, B.O. Blattmann,
and P.J. Chapman. 1991. Bench-scale
evaluation of alternative biological treat
ment processes for the remediation of creo-
sote-contaminated materials: Solid-phase
bioremediation. Environ. Sci. Technol. In
press.
2. Mueller, J.G., S.E. Lantz, B.O. Blattmann,
and P.J. Chapman. 1991. Bench-scale
evaluation of alternative biological treat-
43
-------
Combined Treatment
ment processes for the remediation of
creosote-contaminated materials: Slurry-
phasebioremediation. Environ. Sci.Technol.
In press.
3. Mueller, J.G., S.E. Lantz, B.O. Blattmann,
andP.J. Chapman. 1990. Alternative bio-
logical treatment processes for remediation
of creosote-contaminated materials: Bench-
scale treatability studies. EPA/600/9-90/
049. 103 p.
4. MuellerJ.G./P.J.Chapman/R.Thomas,E.L.
Kline, S.E. Lantz, and P.H. Pritchard. 1990.
Development of a sequential treatment sys-
tem for creosote-contaminated soil and
water: Bench studies. Proceedings U.S.
Environmental Protection Agency's sym-
posiumon Bioremediation of Hazardous
Wastes:U.S. EPA's Biosystems Technology
DeveopmentProgram. EPA/600/9-90/041,
pp. 42-45.
5. Middaugh, D.P., J.G. Mueller, R.L. Thomas,
S.E. Lantz, M.J. Hemmer, G.T. Brooks, and
P.J. Chapman. 1991. Detoxification of creo-
sote-and PCP-contaminated ground water:
Chemical and biological assessment. Arch.
Environ. Contam. Toxicol. In press.
it
024 6 8 10
INCUBATION TIME (weeks)
I— group 1PAHs *— groupSPAHs O— PCP
t— group 2 PAHs —H
12
phonoiics
heterocycEes
Kgure 1. Solid-phase bioremediation, SS.
44
-------
SECTION FIVE
SEQUENTIAL TREATMENT
Sequential treatment is generally applied to two types of waste: compounds that degrade into stable intermediates,
which can be further degraded under different conditions than those used for the parent compound; and complex mixtures
of wastes, which are generally degraded in order of their thermodynamic behavior. The classic example of the first type
of waste is DDT, which is effectively degraded under alternating anaerobic and aerobic conditions. For waste mixtures,
an understanding is needed of the degradation pathways for the waste components and intermediate compounds, the
sequence by which complex mixtures are degraded under field conditions, the physical and chemical factors that can
influence the degradation pathways, and the availability of organisms or adapted bacterial communities that are able to
degrade or transform components of the mixture or intermediate degradation products. .
Sequential technologies can vary from the simple coupling of sequential aerobic and anaerobic processes for the
degradation of a single compound to the use of sequential reactors containing bacterial cultures adapted for degrading
specific compounds or compound classes that are components of a hazardous organic mixture.
A number of recent studies are providing information that may be useful in developing sequential treatment
processes. Current research examines an innovative approach to anaerobic/aerobic sequential treatment of hazardous
waste leachates in publicly owned treatment works (POTWs). This approach is proposed to address some of the problems
in conventional aerobic treatment of these wastes (e.g., pass-through of toxics, air stripping of.volatiles, and insufficient
anaerobic contact for dechlorination of toxics.) In this process, a GAC-expanded bed is placed between the primary clarifier
and the aeration basin. This contact/sorption stage is designed to reduce the pass-through of toxics, retaining them for
treatment in an off-line reactor.
Another study examined the melhanogenic degradation kinetics of phenolic compounds. To estimate substrate
utilization and concomitant bacterial growth, laboratory microcosms containing aquifer material were used to simulate
the biotic and abiotic interactions at an abandoned-wood preserving plant.
EPA researchers also investigated the bacterial -degradation of naphthalene, more complex polychlorinated
.aromatic hydrocarbons, and structurally related heterocydic aromatic hydrocarbons. Another study examined anaerobic
degradation of highly chlorinated dioxins and dibenzofurans. The results of this study indicated that chlorinated dioxins
and dibenzofurans could serve as alternative electron sinks, as is the case for polychlorinated biphenyls (PCBs).
ANAEROBIC/AEROBIC SEQUENTIAL
TREATMENT OF CERCLA LEACHATES
IN POTWS: AN INNOVATIVE
TREATMENT APPROACH
Margaret J. Kupferle and Paul L. Bishop,
University of Cincinnati, Cincinnati, OH;
and Dolloff F. Bishop and Steven I.
Safferman, U.S. Environmental Protec-
tion Agency,RiskReductionEngineering
Laboratory, Cincinnati, OH.
The presented innovative approach to anaero-
bic/aerobic sequential treatment of hazardous waste
leachates in POTWs is proposed to mitigate some of
the problems with conventional aerobic treatment
of these wastes, i.e., pass-through of toxics, air strip-
ping of volatiles, and lack of sufficient anaerobic
contact for dechlorination of toxics. Figure 1 is a
conceptual schematic of the process. A contact/
sorption state consisting of a granular activated
carbon (GAG) expanded bed, which would be placed
between the primary clarifier and the aeration basin
of the POTW, is designed to reduce pass-through of
45
-------
Sequential Treatment
Primary
Effluent/
Leachate
+ Toxics
T3
O>
CD
U.
i
CD
"o
=
o
0)
To Aeration
Basin „
CONTACT/SORPTION
STAGE
* On-line expanded bed
* Ambient temperature
T
o
<
CD
•a
CO
a»
CD
0}
E
Make-up
Carbon
Pollutant-Laden GAC
Gas
Pollutant-
Laden GAC
•4—
Regenerated
CD
"o
O
(U
ANAEROBIC
STABILIZATION
STAGE
* Off-line expanded bed
*-Temperature ~35°C
Gas [KEVl
Liquids
GAC/Liquid
Slurries
Semi-Continuous
Flow
Separate
GAC from
supernatant
Mix pollutant-
laden GAC
with 35 °C
supernatant
Separate
GAC from
supernatant
SEMI-CONTINUOUS
CARBON EXCHANGE
Figure 1. Conceptual schematic of proposed system.
46
-------
Sequential Treatmenti
Two 87-L/day bench-scale systems are in
operation for proof-of-concept studies. Each sys-
tem has two stages. The first stage is operated as a
contact/sorption unit and the second as a stabiliza-
tion unit. The carbon retention time (CRT) in the
first stage is 2 days, with a hydraulic retention time
(HRT) of 30 minutes. The target CRT in the second
stage is 15 days. One system (control) treats pri-
mary effluent only and one system (test) treats
primary effluent spiked with 5 percent landfill
leachate and a mixture of nine volatile and five
semivolatile hazardous organic compounds. Con-
centrations of the spiked organics, as well as
conventional wastewater treatment parameters
such as chemical oxygen demand (COD), are rou-
tinely monitored. Results to date indicate little or
no pass-through to the aeration basin of most of the
spiked organics, and average COD removals in the
40 to 50 percent range.
METHANOGENIC DEGRADATION
KINETICS OF PHENOLIC COMPOUNDS
E. MichaelGodseyand DonaldF. Goerlitz,
U.S. Geological Survey, Menlo Park, CA;
and Dunja Grbic-Galic, Stanford Uni-
versity, Stanford, CA.
Introduction
Microbiologists have come to appreciate the
power of the quantitative approach in their re-
search. It is no longer enough to simply describe the
organisms that occupy a given habitat; the rates at
which they carry out metabolic functions of ecologi-
cal importance must be estimated. Only when
quantitative information about metabolic activities
is coupled with knowledge of organismal types can
our understanding of the concerted actions of the
members of a community be considered complete.
The quantitative approach in microbial ecol-
ogy involves the estimation of parameters in
equations chosen to represent the process under
study, such as substrate depletion and concomitant
growth. A major factor affecting activity and growth
of microorganisms in many environments (e.g.,aqui-
fer sediment) is related to the presence of solid
surfaces in those environments. Surfaces may alter
the availability of organic chemicals, change the
levels of various organic and inorganic nutrients,
and/or retain microorganisms. Bacterial cells that
are attached to subsurface materials may have physi-
ological activities quite different from cells that are
in suspension. Most of these phenomena have not
been investigated for subsurface environments.
The generally accepted equations describing
substrate utilization and concomitant bacterial
growth without decay are the ones proposed by
Monod (1):
ds = Um X S
dt Y(K+S)
dt K+S
47
-------
Sequential Treatment
where:
jlm = maximum specific growth rate, 1/day
Kf = half-saturation constant, mg/L— numeri-
cally equal to that substrate concentration
that yields a growth rate equal to one-half
Y = yield coefficient, mg cells/mg substrate
utilized
S = substrate concentration at time t, mg/L
X = biomass at time t, mg/L
The above relationships were developed from
experiments using pure cultures of bacteria utiliz-
ing single organic compounds. It remains to be
determined if these expressions describe the degra-
dation in the subsurface environment by a complex
mixed microbial population of single compounds in
a complex mixture of compounds.
In this study, we present evidence that the
Monod equations adequately describe both the uti-
lization of phenolic compounds at very low
environmental concentrations (oligotrophic) and the
concomitant bacterial growth.
Materials and Methods
The study site is located in Escambia County
within the city of Pensacola, Florida, adjacent to an
abandoned wood preserving plant (2). The wood
preserving process consisted of steam pressure tr eat-
ment of pine poles with creosote and/or
pentachlorophenol (POP). Largebutunknownquan-
titiesof waste waters,consistingofextractedmoisture
from the poles, cellular debris, creosote, PCP, and
diesel fuel were discharged to surface impound-
ments. The impoundments were unlined and
hydraulically in direct contact with the sand-and-
gravel aquifer. The aquifer consists of deltaic
deposits of fine-to-coarse graveled quartz
interbedded with discontinuous silts and clay
intervals.
Microcosms used for the study were prepared
in4-L sample bottles and contained approximately
3 kg of aquifer material collected from a depth of 5
to 6 m at a site 30 m do wngradient from the contami-
nation source. Phenolic compounds were added to
2.5 L of mineral salts solution (3) at concentrations
similar to the aquifer concentrations, 20 to 40 mg/L.
Amorphous FeS was used as a reducing agent (4) to
ensure methanogenic conditions. The microcosms
were prepared, stored, and sampled in an anaerobic
glove box containing an O2-free argon atmosp>here
at22°C.
Samples for substrate utili zation were remo ved
from the microcosms at approximately 3-day inter-
vals after gentle mixing. Aralyses were done by
reverse-phase gradient-elution HPLC using a UV
detector set at a wavelength of 280 nm.
Nonlinear Parameter Estimation
Substrate depletion and bacterial growth curves
were fitted to the Monod equations using nonlinear
regression analysis. The method of Marquardt (5)
was used for the estimation of parameter values that
minimized the sum of the squared residuals. Be-
cause the Monod equations do not have explicit
analytical solutions for substrate and biomass con-
centrations as a function of time, a simultaneous
solution of both equations was accomplished using
a fourth-order Runge-Kutta numerical procedure.
The statistical basis for these analyses is presented
by Robinson (6), and requires that the sensitivity of
the independent variable to changes in each of the
parameters be calculable. The partial derivatives of
the substrate (S) with respect to \im, Ks, and Y satisfy
this requirement. These expressions are derived
from the integrated Monod equation by implicit
differentiation. Unique estimates of the parameters
can be obtained when the initial substrate concen-
tration (So) is in the mixed-order region and proceeds
through the first-order region during the course of
the experiment.
Parameter estimates with the 95 percent confi-
dence intervals are given in Table 1. The time
interval before the onset of rapid methanogenesis
varied from 28 days for 3-methylphenol to 119 days
for 2-methylphenol, even though the inoculum his-
tory suggests that all of the microorganisms in the
microcosms had been exposed to all of the phenolic
compounds for a considerable length of time (~ 80
years).
48
-------
Sequential Treatment
Discussion and Conclusions
Laboratory microcosms containing aquifer
material simulate the same biotic and abiotic
interactions that occur at the Pensacola site. An
important consideration in determining the
ultimate environmental fate of contaminants is the
adsorption of both the substrate and the biomass to
the aquifer sediment. Studies in the laboratory and
at the research site have shown that substrate ad-
sorption to aquifer sediments of the four phenolic
compounds tested was insignificant (Retardation
Factors ranged from 1.01 for phenol to 1.10 for 4-
methylphenol), and that greater than 99 percent of
the biomass was associated with the aquifer sedi-
ment. However, for modeling purposes, the biomass
on the sediment may be treated as if it were uni-
formly distributed throughout the liquid volume.
These considerations justify the modeling of sub-
strate utilization in themicrocosms as batch reactions.
The bacterial substrate utilization and growth
data for all of the compounds tested could be mod-
eled successfully using the Monod equations. The
long apparent lag times for the phenolic compounds
coulcfbe attributed to extremely low initial biomass
concentration in the microcosms. The kinetic con-
stants for all of the compounds are very similar, and
given that the inocula were acclimated to all of the
phenolic compounds, it is unclear why the range of
onset times was so great. There also appears to be a
lack of correlation between the onset times and the
model parameters.
Although we know of no other kinetic studies
conducted under similar conditions, it appears that
the values of the parameters obtained are reasonable
and consistent with values expected of organisms
from oligotrophic environments. The extremely
low Y values for the phenolic compounds suggest
that these organisms have adapted to this environ-
ment by utilizing 99+ percent of the available energy
for maintaining cellular integrity, or that they are,
very inefficient at capturing the free energy avail-
able. This phenomenon is currently under
investigation. !
References
1. Monod, J. 1949. The growth of
bacterial cultures. Annual Review of Mi-
crobiology 3:371-394.
2. Godsy, E.M., D.F. Goerlitz, and D.
Grbi-Gali. 1991. Methanogenic biodegra-
dation of creosote contaminants in
natural and simulated ground water
ecosystems. Ground Water (in journal
review).
3. Zeikus, J.G. 1977. The biology of
methanogenic bacteria. Bacteriological
Reviews 41:514-541.
4. Brock,T.D. andK. CKDea. 1977. Amorphous
ferrous sulfide as a reducing agent for
culture of anaerobes. Applied and
Environmental Microbiology 33:254-256.
5. Bard, Y. 1974. Nonlinear parameter estima
tion. Academic Press, Inc., New York.
6. Robinson, J.A. 1985. Nonlinear regression
analysis in microbial ecology. Advances in
Microbial Ecology 8:61-114.
Table 1. Results.
Compound j^, I/day
Phenol
2-Methylphenol
3-Methylphenol
4-Methylphenol
0.1 04 ±0.022
0.040 + 0.010
0.122 ±0.068
0.095 ±0.045
K^mg/L
2.00 + 6.10
0.27 ±0.35
0.40 + 1.11
1.90 ±10.1
Y, mg/mg
0.003 ±0.003
0.003 ±0.004
0.002 ±0.004
0.052 + 0.139
49
-------
Sequential Treatment
DEGRADATION OF NAPHTHALENE,
PAHS, AND HETEROCYCLICS
Richard W. Eaton and Peter J. Chapman,
EnvironmentalResearch Laboratory, U.S.
Environmental Protection Agency, Gulf
Breeze, FL.
Naphthalene is the simplest fused polycyclic
aromatic hydrocarbon. Information obtained from
studies of its bacterial degradation may be used in
understanding and predicting the pathways used in
the metabolism of more complex polycyclic aro-
matic hydrocarbons and structurally related
heterocyclic aromatic compounds.
In spite of the relative simplicity of naphtha-
lene, much about its bacterial metabolism has,
remained unclear, particularly the steps in the meta-
bolic pathway by which 1,2-dihydroxynaphthalene
(DHN/Figurel/Dismetabolizedtosalicylaldehyde.
This is primarily because of the chemical instability
of various chemical intermediates implicated or
identified in this pathway. Thus DHN is rapidly
and spontaneously oxidized in water to 1,2-
naphthoquinone, and the potential ring-cleavage
products c/s-o-hydroxybenzalpyruvate(Figurel,rV),
frans-o-hydroxybenzalpyruvate (Figure 1, III), and
2-hydroxychromene-2-carboxylate (Figure 1, V) all
Figure 1.1,2-Dihydroxynaphthalenediroxygenase
(nah C product).
undergo isomerizations in water.
Davies and Evans (2) identified a product of
the oxidation of DHN by cell extracts of a naphtha-
lene-grown Pseudomonas strain as
o-hydroxybenzalpyruvate (HBP A), initially isolated
as its perchlorate, and suggested that it was prob-
ably the cis- isomer based on its chemical properties.
Cell extracts metabolized both cis- and trans- iso-
mers to salicylaldehyde. The cis- isomer of HBPA
was spontaneously converted at neutral pH to its
hemiketal, 2-hydroxychromene-2-carboxylate
(HCCA), which was not metabolized by cells or cell
extracts and was considered to be an artifact.
Barnsley (1) subsequently demonstrated that,
instead of being an artifact, HCCA was the initial
product of the enzymatic ring cleavage of DHN. He
did this by incubating cell extracts, and he also
purified dihydroxynaphthalene dioxygenase (4)
with 1,2-dihydroxynaphthalene at pH 5.5. (At this
pH, the autoxidation of DHN to 1,2-naphthoquinone
is somewhat reduced.) After 2 minutes, the incuba-
tion mixture was applied to and eluted from a
column of Sephadex G-25 in order to separate the
large protein components of the cell extracts from
lower -molecular- weight reaction products which
were collected and freeze-dried. In this way, a small
amount of a single chemical, 2-hydroxychromene-
2-carboxylate, was obtained, and it was proposed
that this was the initial ring-cleavage product. An
HCCA-metabolizing enzyme (isomerase), which
catalyzed the conversion of tha.t compound to Irans-
HBPA at pH 10 (Km = 0.2 mM), was also evident.
Both of these studies were hampered by their
use of 1,2-dihydroxynaphthalene as substrate. In-
cubations required large amounts of enzyme, and
yielded product only in quantities that were insuf-
ficient to identify rigorously.
Accordingly, a way to identify the DHN ring
cleavage product was adopted that involved clon-
ing the genes encoding the first: three enzymes of the
pathway away from genes encoding enzymes cata-
lyzing subsequent steps. Bacteria carrying these
genes would be able to transform the stable sub-
strate, naphthalene, to the ring-cleavage products,
which could then be prepared on a large scale and
identified. Such clones would also be useful for the
preparation of analogous pathway intermediates
from other hydrocarbons and heterocyclics.
50
-------
i - -1 I i' ; ' j
Sequential Treatment
The NAH7 plasmid from Pseudomonas putida
G7 carries genes encoding the complete degrada-
tion of naphthalene. These genes are grouped in
two operons (5, 6). The operon encoding the me-
tabolism of naphthalene to salicylate has been cloned
in the plasmid vector pMMB277 on an 11 kb EcoRI-
Hindlll fragment (Figure2). Severaldeletionsand
subclones have been obtained that eliminate DNA
between nahC (the 1,2-dihydroxynaphthalene
dioxygenase gene located 2.15 to 3.05 kb from the
Hzndlll site [3]) and the HmdIII site and which
inactivate the gene encoding the enzyme that de-
grades the product of 1,2-dihydroxynaphthalene
cleavage. One of these subclones is a 10 kb EcoRI-
C/fll fragment inserted in pMMB277. Pseudomonas
aeruginosa PAO1 carrying this plasmid transforms
naphthalene to the ring-cleavage products which
accumulate. These products have been separated
by chromatography on Sephadex G-25 and identi-
fied as frans-o-hydroxybenzalpyruvate and
2-hydroxychromene-2-carboxylate (the hemiketal
of cz's-o-hydroxybenzalpyruvate) by NMR and mass
spectrometry.
Botha's- and t rans-isomers are probably formed
by the spontaneous rearomatizafa'on of an unstable
ring-cleavage product (Figure I, II). They are both
rapidly degraded to salicylaldehyde by cell extracts
of strain PAO1 carrying the cloned EcoRI-HmdHI
fragment. On addition of NAD+, salicylaldehyde is
oxidized to salicylate.
2-Hydroxychromene-2-carboxylate could have
been rapidly formed from the accumulating ds-
isomer by nucleophilic attack of the orf/zo-hydroxyl
oxygen on the carbonyl carbon. This hemiketal is
not metabolized at neutral pH by cell extracts of
nah genes A
PAO1 carrying the cloned EcoRI-HmdIII fragment.
In wild-type naphthalene-degrading bacteria, the
ds- isomer is probably metabolized before the for-
mation of the hemiketal can occur.
References ,
1. Barnsley, E.A. 1980. Naphthalene
metabolism by pseudomonads: The
oxidation of 1, 2-dihydroxynaphthal-
ene to 2-hydroxychromene -2- carboxylic
acid and the formation of 2'
hydroxybenzalpyruvate. Biochem.
Biophys. Res. Commun. 72:1116-1121.
2. DaviesJ.L.and W.C. Evans. 1964. Oxida-
tive metabolism of naphthalene by
soil pseudomonads. The ring-fission
mechanism. Biochem. J. 91:251-261.
3. Harayama,S.andM.Rekik. 1989. Bacterial
aromatic ring-cleavage enzymes are
classified into two different gene families.
J.Biol. Chem. 264:15328-15333.
4. Patel, T.R. and E.A. Barnsley. 1980. Naph
thalene metabolism by pseudomonads:
Purification and properties of 1,2-dihydroxy-
naphthaleneoxygenase. J.Bacteriol. 143:668-673.
5. Yen, K.-M. and I.C. Gunsalus. 1982. Plas-
mid gene organization: Naphthalene/
salicylate oxidation. Proc. Natl. Acad.
Sci. USA 79:874-878.
6. Yen, K.-M. and C.M.Serdar. 1988. Genetics
of naphthalene catabolism in pseudo-
monads. CRC Grit. Revs. Microbiol.
15:247-268.
F C D E
EC _ X — X
p^c B o c r; c = £ 5 o o o> o =r c o = o E
•S£ 5- S. 5$ • £$ £ Sl-Sgi £ S$Z£ $&
\{( \ III ^^If^
EcoRI 1
EcoRI 1
EcoRI 1
EcoRI 1
EcoRI 1
75 «• Q. I
OT O M
-------
Sequential Treatment
ANAEROBIC DEGRADATION OF
HIGHLY CHLORINATED
DIOXINS AND DIBENZOFURANS
Peter Adriaens and Dunja Grbic-Galic,
Department of Civil Engineering,
Stanford University, Stanford, CA.
Summary
Araclor 1242-contaminated Hudson River
sediments (HR), creosote-contaminated Pensacola
soil (PS), and chlorophenol adapted Cherokee Pond
sediments (ACP) wereanaerobically incubated with
50 p-g/L of the following dioxin (dibenzo-p-dioxin)
and dibenzofuran congeners: 1,2,3,4,6,9-hexa
(HexaCDD), 1,2,4,6,8,9/1,2,4,6,7,9-hexa isomer
(HexaCDDi), and 1,2,3,4,6,7,9-hepta (HeptaCDD)
chlorinated dioxins,and 1,2,4,6,8-penta (PentaCDF)
and 1,2,3,4,6,7,8-hepta (HeptaCDF) chlorinated
dibenzofurans. The initial analytical data from the
first 8 weeks of incubation suggest that HR sedi-
ments exhibit a more extensive activity towards the
congeners than the PS inoculum. No data are avail-
ableyetfor ACP sediments. Substrate disappearance
was observed for PentaCDF (35 percent), HexaCDDi
(<10 percent), and HeptaCDD (10 percent) in one or
more replicates of HR incubations. No detectable
levels of intermediates have been observed, except
for a small amount (3 to 4 jig/L) of HexaCDD
obtained from the incubation with HeptaCDD. This
intermediate was identified by GC/MS; however,
isomer assignment was impossible due to the ex-
tremely low concentration levels. Although the
initial results do not show extensive activity against
the congeners tested, the presence of a lower chlori-
nated metabolite may be the first indication that
highly chlorinated dioxin and dibenzofuran iso-
mers could serve as alternative electron sinks, as is
observed forpolychlorinated biphenyls (PCBs).
Experimental Procedures
Experimental Setup and Sample Preparation.
Three replicates of microcosms (50 mL total liquid
volume, 50 ± 2 g solids) were spiked with 100 |J.L of
the respective dioxin and dibenzofuran from 50
mg/L (exceptfor HexaCDDi,5 mg/L) nonane stock
solutions to give a final concentration of 50 M-g/L (5
H,g/L for HexaCDDi). In addition, duplicate killed
(autoclaved) biological controls, live biological con-
trols without PCDD or PCDF, and chemical controls
without inocula, have been established and were
monitored along with the cultures.
All bottles were manually shaken, decapped,
and sampled (5 mL) with a 10-mL glass syringe to
contain both sediment or soil and aqueous phase,
under a continuous stream of nitrogen in the
headspace. Two volumes of hexane/acetone (9:1)
and 0.5 |Ag of octachloronaphthalene (as an internal
standard) were added to each sample, which was
then shaken overnight on a wrist-action shaker. The
extraction solvent was decanted from the soil, sedi-
ment, or aqueous phase, and extracted with 2 mL of
concentrated H2SO4. The extract was then back-
extracted with 2 mL of a 2 percent CaCL. solution (in
distilled water), and dried over Na2SO4. The result-
ing extract was eluted over a Pasteur pipette packed
with Florisil (60 mesh)/Cu-powder (40 mesh) (1:4
ratio), to remove excess sulfate. The sample was
then concentrated to 1 mL under a constant stream
of N2. To this fraction, 100 ]iL of dodecane was
added. The sample was then further concentrated
to 100|0,L under a gentle stream of N2 and used for
GC/MS and GC-ECD analyses.
Analytical Procedures. The samples were
analyzed both on a Triple Stage Quadrupole TSQ 70
Finnigan MAT GC/MS, and on a 5890A Hewlett-
Packard Gas Chromatbgraph equipped with an
electron cap ture detector (BCD).
GC/MS operating conditions: Column: DB-5,
60 m, 0.32 urn I.D., 0.25|xm film thickness; column
head pressure: 25 kPa; injection: on-column; injec-
tor temperature: 60°C; initial temp.: 90°C, hold 5
min; rate 1: 25°C mur1; temp. 2: 200°C, hold 15 min;
rate 2: 4°C mur1; temp. 3: 250°C, hold 15 min.
Gas chromatographic conditions: Column: DB-
5,30 m, 0.32 jim I.D., 0.25 Jim film thickness; carrier
gas: helium (linear flow velocity: 25 cm s'1); make-
up gas: argon/methane; column head pressure: 14
kPa; injection: split/splitless (10:1 ratio); injector
temperature: 250°C; detector temperature: 275°C;
initial temperature: 250°C, hold 5 min; rate 1: 2°C
min-1; final temperature: 300°C, hold 5 min. The
high initial temperature caused the Araclor in
samples from bottles containing Hudson River
52
-------
Sequential Treatment
sediments to elute in the first 10 min, without inter-
fering with the dioxin (25.8 min) and dibenzofuran
isomers (penta: 20min,hepta: 24 min). The internal
standard (octachloronaphthalene) eluted after 19
minutes.
Results and Discussion
Table 1 represents the time zero samples, and
Table 2 shows the results from samples taken after
2 months. The recovery efficiencies of all isomers
from the chemical controls always exceeded 92 per-
cent, while they varied between 20 and 45 percent
from the samples containing inoculum. The low
value obtained for the killed control of penta CDF
incubations (Table 1) cannot be ascribed to low
recovery efficiencies, as the data are corrected for
sorption with the internal standard. Hence, they
have to be explained either by incorrect spiking with
the furan, or a volatilization loss when these con-
trols were autoclaved twice. To prevent these kinds
of losses, killed controls of the Cherokee Pond incu-
bations were spiked with 0.5 mL of concentrated
H2SO4, instead of twice autoclaved.
The recoveries of penta CDF hi the live HR
incubations and killed controls, after normaliza-
tion, wereonly60and87percent,respectively (Table
2). Further analysis of the 8-month incubation
samples might yield more information with respect
to this phenomenon. In the case of HR incubations
with HeptaCDD, however, the third replicate yielded
a much lower concentration than both other repli-
cates. Although no discernible peak could be
observed during full scan GC/MS analysis of this
sample between masses 100 and 480, single ion
monitoring (SIM) between masses 380 and 410 indi-
cated the presence of a chlorinated compound
(Figure 1). Scans for the base peak M+ (m/z 390),
M+2+ (m/z 392) and M-2* (m/z 388) ion abundances
in the molecular ion cluster agreed well with those
of a HexaCDD standard (1,2,3,4,6,9-HexaCDD) and
published values (1).
Table 1. Time zero analysis of sediment and soil incubations with five dioxin and dibenzofuran isomers.
All concentrations [(x + s) ppb] were normalized with respect to octachloronaphthalene.
Isomer
HeptaCDF
PentaCDF
HeptaCDD
HexaCDD
HexaCDDi
Inoculum1
HR
PS
HR
PS
HR
PS
HR
PS
HR
PS
Live2
40.4 ±0.1
44.9 ±2.9
41.3 + 1.4
49.9 ±4.1
47.3 ±3.0
48.9±il.l
50.9 ±0.9
55.7 ±6.7
4.9 ±0.4
4.7 ±0.6
Killed2
42.2 ±3.5
45.5 + 0.5
35.5 + 2.9
50.7 ±2.9
47.6 ±2.3
47.4+1.6
52.8 ±2.1
51.8 ±8.1
4.6 ±0.6
4.3 ±0.8
Controls3
Biological
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Chemical
51.8 ±0.9
51.1 + 0.5
51.0 + 6.7
50.2 ±0.9
44.3 ± 1.2
46.6 + 0.1
48.6 ±1.2
56.2 ±0.8
4.8 ±0.1
4.6 ±0.4
1 Abbreviations: HR, Hudson River; PS, Pensacola Soil.
^Average of three samples.
•'All data from the controls are average values from two samples.
53
-------
Sequential Treatment
Table 2. Analysis of sediment and soil incubations after 2 months
were normalized with respect to octachloronaphthalene.
as in Table 1.
All concentrations [(x ± s) ppb]
The number of replicates is the same
Isomer
HeptaCDF
PentaCDF
HeptaCDD
HexaCDD
HexaCDDi
Inoculum
HR
PS
HR
PS
HR
PS
HR
PS
HR
PS
Live
42.5 ±5.0
45.6 ±2.9
27.4 ±1.0
50.7 + 0.2
, 41.4±5.8l
48.6 ±1.2
49.9 ±7.1
48.1 ±7.7
4.4 + 1.1
4.5 ±0.3
Killed
43.4 + 1.5
44.1 ±0.9
38.5 + 1.9
48.8 + 0.8
48.0 + 3.3
48.3 ±2.4
52.2 ±2.3
48.8 ±3.2
4.3 ±0.4
4.3 ±0.8
Controls
Biological
0.0
0.0
0.0
0.0
0.0
0.0
0.0 ,
0.0
0.0
0.0
Chemical
43.1 + 1.2
43.8 + 4.9
44.0 + 1.3
48.2 + 2.4
47.9 + 2.6
48.2 ±0.8
49.2 ±0.7
52.5 + 0.8
4.6 ±0.3
4.3 ±0.6
e concentration of HeptaCDD in one replicate decreased by ± 10 percent.
Except for the presence of the m/e 327 ion,
which was out of scan range, the relative abundance
percentages of the three main ions (Figure 2) were
well within the U.S. EPA acceptance criteria for
isomer identification (2). Since background ion
abundance levels interfered strongly with the peak
identification, a total extract of this replicate will be
analyzed to conclusively identify this ion cluster as
a HexaCDD.
References
1. Kleopfer, R.D., R.L. Greenall, T.S.
Viswanathan, C.J. Kirchmer, A. Gier, and
J.Muse. 1989. Determination of polychlori-
nated dibenzo-dioxiris and dibenzof urans
in environmental samples using high
resolution mass spectrometry. Chemo-
sphere 18:109-118.
2. Alford-Stevens, A.L., J.W. Eichelberger,
T.A. Bellar, and W. L. Budde. 1986.
Determination of chlorinated dibenzo-p-
dioxins and dibenzofurans in soils and
sediments by gas chromatography/ mass
spectrometry. EPA Report: Physical and
Chemical Methods Branch.
54
-------
SECTION SIX
METABOLIC PROCESS CHARACTERIZATION
EPA's metabolic processes research generates a better understanding of the processes by which microorganisms
degrade chemicals, expanding the range of organisms that can be used in biotreatment technologies. Based on the insights
gained from this research, scientists can choose indigenous organisms or enhanced organisms to meet needs in pollution
cleanup and control.
Data from one research project suggest that actively nitrifying Nitrosomonas europaea have a potential role in
biotreatment of halogenated aliphatic compounds. In another project, researchers are developing sulfate-reducing bacterial
cultures that can degrade chlorinated aromatic compounds, and are evaluating sewage sludge and freshwater sediments
for sulfidogenic activity that results in microbial transformation of certain of these compounds. Another study is
investigating ring fission of poly cyclic aromatic hydrocarbons by white rot fungi (Phanaerochaete chrusosporium). a
phenomenon previously thought to be unique to certain bacteria. Other researchers are conducting genetic and soil studies
to develop strains of the bacteria Pseudomonas with sufficiently high soil survivability for use in biotreatment of.
polychlorinated biphenyls (PCBs). Researchers are also studying the degradation of trichloroethylene (TCE) by
Pseudomonas cepacia. A major limitation of these organisms for TCE bioremediation is their requirement for exogenous
aromatic inducers. Current research is focusing on a mutant of G4 that does not require induction.
DEGRADATION OF HALOGENATED
ALIPHATIC COMPOUNDS BY
THE AMMONIA-OXIDIZING BACTERIUM
NITROSOMONAS EUROPAEA
Todd Vanelli and Alan B. Hooper, Univer-
sity of Minnesota, St. Paul, MN; and Peter
Chapman, Environmental Research Labora-
tory, U.S. Environmental Protection Agency,
Gulf Breeze, FL.
The ubiquitous soil-, marine-, and freshwa-
ter-dwelling ammonia-oxidizing nitrifying bacteria
are obligate chemolithoautotrophic aerobes. They
depend for growth on the activity of the enzyme
ammonia monoxygenase (AMO): 2 H+ + 2 e~ + O2 +
NH3 -> NH2OH + H2O. The two electrons for the
AMO reaction originate in the subsequent reaction
catalyzed by hydroxylamineoxidoreductase (HAO):
H2O + NH2OH -> 4 e- + 5H+
NO2-(1).
Nitrosomonas, like the methylotrophs (2), is
capable of the oxidation of many organic com-
pounds. In addition, our laboratory at Minnesota
(3,4) and the laboratory of Arp at Corvallis (5) have
observed degradation of many halogenated aliphatic
compounds as summarized in Table 1.
In no case is there evidence that oxidation of
an organic substrate will support growth. Response
to inhibitors of AMO (2-chloro-6-trichloromethyl
pyridine["nitrapyrin"],acetyleneor-dipyridyl)and
the fact that the organic co-oxidized substrate inhib-
its ammonia oxidation indicates that degradation is
catalyzed by AMO. For all compounds, the con-
comitant oxidation of ammonia is required for deg-
radation of halogenated hydrocarbons. Depending
on the substrate, hydroxylamine and/or hydrazine
(a nonbiological substrate for HAO) can also serve
as electron donors for degradation in cells. Low
levels of degradation are observed for a short time in
the absence of added electron-donating co-substrate;
it is presumed but not known that an endogenous
electron donor is involved.
Clearly Nitrosomonas is able to degrade a wide
spectrum of halogenated aliphatics. The exceptions
are tetrachloroethylene and carbon tetrachloride.
The observation that the soluble methane-oxidizing
enzyme from methylotrophs will oxidize
fluorotrichlorethylene (6) suggests that the absence
of a C-H bond need not be limiting.
55
-------
Metabolic Process Characterization
Table 1. Degradation of halogenated aliphatic compounds.
Substrates"
Products
Haloalkanes:
Bromomethaneb
Chloromethanec
Dibromomethane
Dichloromethane
Trichloromethane
Tetrachloromethane4
Bromoethane
Chloroethanec
Huoroethane0
Iodoethanec
1,2-Dibromoethane
1,1,2-Trichloroethane
Chloropropane
1,2,3-Trichloropropane
Chlorobutane0
Haloalkenes:
Chloroethylene
cis 1,2-Dibromoethylene
trans l/2-Dibromoethylened
1,1 -Dichloroethylene
cis 1,2-Dichloroethylene
trans l,2-Dichloroethylened
Trichloroethylene
Tetrachloroethylened
1,3-Dibromopropene
2,3-Dichloropropene
1,1,3-Trichloropropene
Nitrapyrin
Formaldehyde
Formaldehyde
Acetaldehyde, 2-Chloroethanol
Propionaldehyde, 3-Chloro-l -propanol
1 -Chloro-2-propanol
Butyraldehyde, 4-Chloro- -butanol
a. All compounds reported in reference 4 except as indicated.
b. See ref. 1 and 2.
c. See ref 5.
d. Tested but not degraded.
56
-------
- ' ! l !
Metabolic m
oeess Characterization
The product of oxidation of some substrates by
Nitrosomonas, possibly an epoxide intermediate,
forms an irreversible derivative of AMO and thus
inactivates the enzyme. Reaction with acetylene,
the best example of this kind of compound, results
in derivatization of a specific membrane polypep-
tide (7). Nevertheless, the oxidation of most haloge-
nated compounds can continue for days provided
that the concentration of ammonia remains high
enough.
We have recently demonstrated the ammonia-
dependent degradation of 1,3-dibromopropene,
1,1,3-trichloropropene, 2-chloro-6-trichloromethyl
pyridine nitrapyrin, and 2,3-dichloropropene by
Nitrosomonas. The rates were 4.0, 1.6, 5.8, and 29
Hmoles hr1 wet weight1. Halogenated substrates
were measured by electron capture detector after
gas chromatography. At a concentration of 30 to 40
^iM, the first three compounds completely and
irreversibly inhibit oxidation of ammonia. Ring
14C-labeled nitrapyrin derivatizes all membrane
proteins equally. Thus the reactive product of oxi-
dation of nitrapyrin appears to be membrane soluble
and long lived.
In nature or in pollution treatment, actively
nitrifying Nitrosomonas would appear to have a
potential role in the degradation of halogenated
aliphatic compounds.
References
1. Hooper, A.B. 1989. Biochemistry of the
nitrifyinglithoautotrophic bacteria. In: Au-
totrophic Bacteria (H.G. Schlegel and B.
Bowien, eds.) Sci. Tech. Publishers, Madi-
son, WI, pp. 239-265.
2. Bedard, C. and R. Knowles. 1989. Physiol-
ogy, biochemistry, and specific inhibitors of
CH4, NH4+, and CO oxidation by
methylotrophs and nitrifiers. Microbiol. Rev.
53:68-64.
3. Arciero, D., T. Vannelli, M. Logan, and A'.B..
Hooper. 1989. Degradation of trichloroeth-
ylene by the ammonia-oxidizing bacterium
Nitrosomonas europaea. Biochem. Biophys.
Res. Commun. 159:640-643.
4. Vanelli, T., M. Logan, D.M. Arciero, and
A.B. Hooper, 1990. Degradation of
halogenated aliphatic compounds by
the ammonia-oxidizing bacterium
Nitrosomonas, europaea. Appl. Envt.
Microbiol. 56:1169-1171.
5. Rasche, M.E., R.E. Hicks, M.R. Hyman, and
D J. Arp. 1990. Oxidation of monohal genated
ethanes and n-chlorinated alkanes by whole
cells of Nitrosomonas europaea. ]. Bacteriol.
172:5368-5373.
6. Fox, B.C., J.G. Borneman, L.P. Wackett, and
J.D. Lipscomb. 1990. Haloalkene oxidation
by the soluble methane monoxygnease from
Methylosinustrichosporium OB3b: Mechanistic
and environmental implications. Biochem.
29:6419-6427. ,
t
7. Hyman, M.R. and P.M. Wood. 1985. Sui-
cidal inactivation and labeling of ammonia
monooxygenase by acetylene. Biochem. J.
227:719-725.
DEGRADATION OF CHLORINATED
AROMATIC COMPOUNDS
UNDER SULFATE-REDUCING
CONDITIONS
Patricia J.S. Colberg, Department of Molecu-
lar Biology, University of Wyoming, Laramie,
WY; and John E. Rogers, U.S. EPA Environ-
mental Research Laboratory, Athens, GA.
Despite the recent progress made in describ-
ing microbial transformations that occur under
anaerobic conditions, our understanding of the role
sulfate-reducing bacteria may play in the
remediation of environmental contaminants is still
very much in its infancy. The aromatic nucleus is
the second most common naturally occurring or-
ganic residue in the biosphere and is the basic struc-
tural unit of innumerable anthropogenic materials
that are produced in enormous quantities all over
the world. Yet the first description of an aromatic
57
-------
Metabolic Process Characterization
compound being used by a sulfate reducer was not
published until 1980 (1).
The number of aromatic substrates known to
be amenable to microbial transformations under
sulfidogenic conditions is still small, but includes a
rapidly growinglistofnonhalogenated compounds.
Some of these transformations are catalyzed by
pure cultures of sulfate-reducing bacteria, though
none of the metabolic pathways have been eluci-
dated.
The number of halogenated aromatic com-
pounds reportedly susceptible to microbial trans-
formations under sulfate-reducing conditions is as
yet limited to the five chlorinated phenols whose
structures are shown below. Results from several
laboratory studies have suggested that sulfate may
inhibit the anaerobic degradation of chloroaromatic
compounds by preventing dehalogenation
(2,3,4/5,6). The mechanism of this apparent sulfate
inhibition of dehalogenation is unclear. Some in-
vestigators have suggested that the inhibitory effect
of sulfate may be caused by competition for an
electron donor between the sulfate-reducing bacte-
ria and the organism(s) responsible for
dehalogenation; that is, the sulfidogens are able to
outcompete dehalogenators for available hydrogen
(2,6,7). Itis interesting to note, however, that sulfate
does not inhibit dehalogenation by Desulfomonik
tiedjei (8,9), the only known obligately anaerobic
dechlorinating bacterium which also happens to be
a sulfidogen.
Based perhaps, in part, on the rather consis-
tently observed disparity in transformation poten-
tial between thesulfate-reducingandmethanogenic
regions within their shallow aquifer study site, Kuhn
and his co-workers (6) have speculated that the
anaerobic dehalogenation potential might be lower
in environments that maintain a higher redox
potential (i.e., denitrifying and sulfidogenic), espe-
cially for less highly halogenated aromatic com-
pounds. Vogel et al. (10) have proposed a similar
explanation for the anaerobic dehalogenation of
aliphatic compounds.
Despite the evidence linking sulfate with in-
hibition of reductive dehalogenation, it is worth
noting that several of the reports of this phenom-
enon are from the same laboratory and used material
from the same shallow aquifer (2,3,6), so it is perhaps
premature to make any generalizations. Based on -
results of a recent study in which chlorophenol deg-
radation occurred during sulfidogenesis (4) and an-
other in which haloaromatic degradation was shown
to be coupled to sulfate reduction (11), it is possible
that sulfate inhibition of dehalogenation may be a
site-specific characteristic.
The objectives of this research are to:
1. Develop sulfidogenic cultures or consortia
that are able to dehalogenate chlorinated
benzenes, phenols, benzoates, anilines, and
biphenyls. ,
2. Determine the activity of the cultures ob-
tained over a range of environmental cond-
itions:
3. Evaluate the substrate specificity of the cul-
tures towards all of the compounds classes,
4. Evaluate in microcosm experiments the
ability of these cultures to enhance the
degradation of hazardous chlorinated aro-
matic compounds in contaminated soils and
sediments. ,
This report will summarize some of the work in
progress aimed at evaluating sewage sludge and
several freshwater sediments for sulfidogenic activ-
ity that results in microbial transformations of a se-
lected number of chlorinated aromatic compounds.
References
1. Widdel, F. 1980. Anaerober Abbau von
Fettsauren und Benzoesauren durch neu
isolier te Arten sulfat-reduzierender Bakterien.
Doctoral Dissertation, University of
Gottingen. ,
2. Gibson, S.A. and J.M.Suflita. 1986. Extrapo-
lation of biodegradation results to ground-
water aquifers: Reductive dehalogenation.
of aromatic compounds. Appl. Environ.
Microbiol. 52:681-688.
58
-------
; 's. ::; •- ;'! : :••;:. i
Metabolic Process Characterization
3. Gibson, S.A. and J.M. Suflita. 1990.
Anaerobic biodegradation of 2,4,5-
trichlorophenoxyacetic acid in samples
from a methanogenic aquifer: Stimula-
tion of short-chain organic acids and alcohols.
AppLEnviron. Microbiol. 56:1825-1832.
4. Kohring, G-W., X. Zhang, and J. Wiegel.
1989. Anaerobic dechlorination of 2,4-
dichlorophenol in freshwater sediments in
the presence of sulfate. Appl. Environ.
Microbiol. 55:2735-2737.
5. Genthner,B.R. Sharak, W.A. Price II, and
P.H.Pritchard. 1989. Anaerobic degrada-
tion of chloroaromatic compounds under
a variety of enrichment conditions.
Appl. Environ. Microbiol. 55:1466-1471.
6. Kuhn, E.P., G.T. Townsend, and J.M.
Suflita. 1990. Effect of sulfate and organic
carbon supplements on reductive
dehalogenation of chloroanilines in anaero-
bic aquifer slurries. Appl. Environ.
Microbiol. 56:2630-2637.
7. Suflita, J.M., S.A. Gibson, and R. E. Beeman.
1988. Anaerobic biotransformations of pol-
lutant chemicals in aquifers. J. Ind.
Microbiol. 3:179-194.
8. DeWeerd, K.A., L. Mandelco, R.S. Tanner,
C.R. Woese, and J.M. Suflita. 1990.
Desulfomonile tiedjei gen. nov. and sp nov.,
a novel anaerobic dehalogenating,
sulfate-reducing bacterium. Arch.
Microbiol. 154:23-30.
9. Linkfield,T.G.,andJ.M.Tiedje. 1990. Char-
acterization of the requirements and sub-
strates for reductive dehalogenation by
strain DCB-1. J. Ind. Microbiol. 5:9-15.
10. Vogel,T.M.,C.S.Criddle,andP.L.McCarty.
1987. Transformations of halogenated ali-
phatic compounds. Environ. Sci. Technol.
21:722-736.
11. Haggblom, M.M. and L.Y. Young. 1990.
Chlorophenol degradation coupled to sul-
fate reduction. Appl. Environ. Microbiol.
56:3255-3260.
Acknowledgement
This project is supported by the EPA Envi-
ronmental Research Laboratory, Athens, GA,
through Cooperative Agreement CR-816398-01-0.
RING-FISSION OF POLYCYCLIC
AROMATIC HYDROCARBONS
BY WHITE ROT FUNGI
Kenneth E. Hammel, State University of New
York, College of Environmental Science and
Forestry, Syracuse, NY., and John A. Glaser,
U.S. EPA Risk Reduction Engineering Labo-
ratory, Cincinnati, OH.
Polycyclicaromatichydrocarbons(PAHs)are
major pollutants of both anthropogenic and natural
pyrolytic origin, occurring in soils, sediments, and
airborne particulates. The crucial step in their bio-
degradation is oxidative fission of the fused aro-
matic ring system, an event previously thought
unique to certain bacteria. Recent evidence necessi-
tates a revision of this view: the lignin-degrading
fungi that cause white rot of wood have also been
shown to mineralize a wide variety of aromatic
pollutants, including certain PAHs, under culture
conditions thatpromote the expression of ligninolytic
metabolism. A key component of the fungal
lignonoly tic system is thought to consistof extracel-
lular lignin peroxidases (LiPs), which have been
shown to catalyze the one-electron oxidation of
various lignin-related substrates. LiPs have also
been shown to oxidize certain PAHs and other
aromatic pollutants in vitro, and it has been pro-
posed that these enzymes play an important role in
fungal xenobiotic metabolism. However, it has
never been demonstrated that any PAH is oxidized
by LiPs in vivo, or that the products of such a reaction
are subsequently cleaved to smaller, monocyclic
compounds. To address these questions, we have
examined the fate of one PAH, anthracene, in
cultures of the ligninolytic basidiomycete
Phanaerochaete chrysosporium.
Anthracene (AC) is the simplest PAH to be a
LiP substrate. We found' that it was oxidized to a
59
-------
Metabolic Process Characterization
single end product by both crude and purified
preparations of the enzyme, and that this product
was indistinguishable from 9,10-anthraquinone
(AQ) when subjected to thin-layer chromatography
(TLC) on silica or gas chromatography/electron
impact mass spectrometry. Since other PAHs that
have been examined in detail give mixtures of
products when oxidized by LiP, we concluded that
AC was the PAH most likely to yield diagnostic
metabolites in fungal cultures, and selected it for
furtherstudies. In fungal cultures, ["C^^JACand
[«C hen ,]AQ were mineralized to the same extent,
with" i1U±3.9 percentof AQ, and 12.9±1.3 percent
of AC, oxidized to CO2 in 14 days. Moreover, the
cultures rapidly oxidized AC to AQ. The quinone
was the predominant neutral AC metabolite found
when the culture medium was analyzed by reverse-
phase high performance liquid chromatography
(HPLC) or TLC on silica, and an isotope dilution
experiment done on the extracellular medium and
myceliumshowed that AQ accounted for38 percent
of the AC originally added after 48 hours in culture.
The abiotic oxidation of AC in uninoculated cultures
gave only 1 percent conversion to AQ in this time.
These results support a role for LiP in AC oxidation
by Phanerochaete, and show that the pathway
AC-*AQ ->CO2 is quantitatively important in AC
metabolism by the fungus.
Analysis of the acidic metabolites formed
from AC and AQ by P. chrysosporium showed that
both compounds were cleaved to phthalic acid. The
identification of the ring-fission metabolite as
phthalate was based on three findings: 1) it was
indistinguishable from authentic phthalic acid by
ion exclusion HPLC, 2) it recrystallized with au-
thentic phthalic acid to constant 14C specific activity
in an isotope dilution experiment, and 3) after treat-
ment with diazomethane, it was indistinguishable
fromauthenticdimethylphthalatebyTLConsilica.
The isotope dilution experiment showed that both
AC and AQ were accumulated as phthalic acid in 12
percent yield after 12 days in culture. Phthalic acid
was not a dead-end metabolite in fungal cultures: it
also was mineralized, but at only about one-third
the rate that AC and AQ were. The relative persis-
tence of phthalate in the cultures probably explains
our success in identifying it as an intermediary
metabolite, and the bulk of AC/AQ mineralization
is presumably due to further degradation of the
moiety that is cleaved from AQ to give phthalate.
Our results show that the pathway AC-> AQ-*
phthalate is a major one in AC ring-fission by
Phanerochaete. This fungal pathway clearly differs
from the classical bacterial one, which proceeds AC
•^AC-cz's-l,2-dihydrodiol-»>saIicylate. Itisnotewor-
thy that the principal oxidized products to accumu-
late from AC, namely AQ and phthalic acid, can
both be degraded by Phanaerochaete if they are given
to freshly ligninolytic cultures. This result shows
that the cessation of organopollutant mineralization
activity after 2 to 4 weeks that is generally observed
in P. chrysosporium cultures is not due in every case
to the accumulation of recalcitrant products that the
fungus cannot further metabolize. This result is
rather an artifact of laboratory culture conditions. A
principal direction for future work, accordingly,
should be the development of methods to prolong
biodegradative capability in fungal cultures.
AEROBIC BIODEGRADATION OF
POLYCHLORINATED BIPHENYLS:
GENETIC AND SOIL STUDIES
Frank J. Mondello and Bruce D. Erickson,
Biologkal Sciences Laboratory, General Elec-
tric Corporate Research and Development,
New York, NY.
A practical process for aerobic bioremediation
of PCB-containing soil is dependent upon isolating
or developing suitable organisms. Two of the most
important characteristics for such organisms are 1)
high levels of degradative activity against many
PCB congeners, and 2) the ability to survive on the
soil long enough for significant PCB degradation to
occur. Recombinant DNA technology is currently
being used to develop bacterial strains with these
and other desirable properties and to study the
genes involved in PCB degradation from a strain of
Pseudomonas designated LB400.
Pseu'domonas sp. strain LB400 is able to de-
grade a wide variety of PCBs but has many charac-
teristics which make it unattractive for use in a
bioremediation process. One serious shortcoming
is that the organism has been reported to lose viabil-
ity very rapidly on soil. This would make it neces-
60
-------
Metabolic Process Characterization
sary to add organisms frequently to a site or reactor,
thus making a process less feasible. Previous studies
have shown that Escherichia coli strain FM4560 (a
genetically modified organism containing the LB400
bph A, B, and C genes) degrades PCBs nearly as well
as LB400 without exhibiting many of its undesirable
traits. For instance, FM4560 had bettersurvivability
than LB400 in laboratory media containing PCBs
and did not require growth on biphenyl for high
levels of degradative activity.
The abilities of FM4560 to survive and to
degrade PCBs on soil were compared with those of
LB400 to determine if the recombinant strain is
potentially more useful than the naturally occurring
organism. Survivability on soil was examined
using PCB-contaminated material from a site in
Glens Falls, New York (dragstrip soil). This mate-
rial contained approximately 550 parts per million
of highly evaporated Aroclor 1242, and therefore
appeared similar in composition to Aroclor 1248. A
series of 2-dram vials containing dragstrip soil were
inoculated with either FM4560 or LB400, sealed, and
incubated at 23? C without shaking. Colony for-
mation on selective media was used to measure cell
survival. The presence of active bph genes in LB400
was determined by growth on biphenyl, while bph
gene activity in FM4560 colonies was demonstrated
by their ability to produce yellow meta-cleavage
product when sprayed with an ether solution of
biphenyl or 2,3-dihydroxybiphenyl.
Similar survival curves were obtained for
both FM4560 and LB400. Early time points showed
significant increases in cell number, presumably
resulting from growth on stored intracellular nutri-
ents. After 72 hours, cell numbers returned to their
original level and continued to decrease such thatby
8 days approximately 20 percent of the cells re-
mained culturable. After 28 days of incubation, the
number of culturable cells was 2 percent of the
initial value.
The presence of active bph genes in the sur-
viving cells was examined at each time point. For
LB400 the number of viable cells unable to grow
using biphenyl remained relatively stable at ap-
proximately 2.5 percent for the first 8 days and then
increased to 4.2 percent for the remainder of the
experiment. All of the FM4560 colonies contained
active bph genes.
The ability of E. coli strain FM4560 to degrade
PCBs on soil from the Glens Falls site was compared
with that ofPseudomonas sp. strain LB400. All experi-
ments were conducted in sealed, sterile, 2-dram glass
vials containing 0.1 gram of nonsterile soil and 1.0 mL
of bacterialculture. Under most of the experimental
conditions, PCB degradation by strain FM4560 was
significantly greater than that by LB400.
Analysis of the LB400 bphA Gene
Biphenyl/PCB dioxygenase (encoded by the
bphA gene) is the enzyme primarily responsible for
PCB degradation. Knowledge of the structure, func-
tion, and regulation of bphA is therefore crucial for
the development of genetically modified bacteria
with superior PCB-degrading abilities. The nucle-
otide sequence of the gene(s) for biphenyl/PCB
dioxygenase from strain LB400 has been obtained.
This sequence was compared with that of toluene
dioxygenase from Pseudomonas putida strain Fl, a
multi-component enzyme made up of 4 distinct sub-
units. The genes for these subunits are co-transcribed
and arranged in the order todCl (ISPTOL large sub-
unit), todC2 (ISPTOL small subunit), todB
(FerrodoxinTOL), and todA (Reductase^). Computer
analysis of the bphA nucleotide sequence identified 4
open reading frames whose DNA and protein se-
quences were similar to those encoding the 4 sub-
units of toluene dioxygenase. For example, 67.5
percent identity in the nucleotide sequence and 65.5
percent identity in the protein sequence between
todCl and the putative ISP large subunit of biphenyl/
PCB dioxygenase were obtained. Sequences up-
stream of the identified homologous coding region
showed no significant homology. These data suggest
a relationship between the dioxygenase responsible
for biphenyl/PCB degradation and those used to
degrade a variety of other aromatic hydrocarbons,
including toluene and benzene.
Transcription of the bphA region is being ex-
amined using the technique of SI nuclease mapping.
Two mRNA 5' ends are visible that map to the area
immediately before the first bphA coding region. An
additional transcript has been detected that maps at
least 1,500 base pairs before this coding region. The
start site of this transcript has not yet been
determined. RNA 5' ends can arise from either
transcription initiation c»r an RNA-processing event.
61
-------
Metabolic Proieess Characterization
To differentiate between these two possibilities,
DNA fragments corresponding to the putative pro-
moter regions were cloned into a promoter detec-
tion vector containing the promoter-less
galactokinase gene. Several small fragments (< 300
base pairs) that span the start of the bphA coding
region demonstrated promoter activity, strongly
suggesting that transcription of biphenyl dioxy-
genase originates from within this region. Studies
are currently under way to examine the effect of
biphenyl on promoter activity, since it is known that
PCB degradation by LB 400 is enhanced when the
organism is grown using biphenyl.
MANIPULATION OF TCE-DEGRADATIVE
GENES OF PSEUDOMONAS CEPACIA
Malcolm Shields, TRI/U.S. Environmental Pro-
tection Agency, Gulf Breeze, FL.
Summary
Pseudomonas cepacia G4, like several other
toluene-utilizing bacteria, can degrade trichloro-
ethylene CTCE) only when the requisite oxygenase is
expressed in response to aromatic inducers. A major
limitation to the application of these organisms for
TCE bioremediation is their requirement for exog-
enous inducers. A mutant of G4 (G4 Phel) was
selected for its ability to degrade TCE, 1,1
dichloroethylene, and czs-and trans-1,2-
dichloroethylene without requiring induction.
Experiments to determine the genetic basis for this
altered regulation indicated that enzymes involved
in the conversions of cresols to ring cleavage prod-
ucts were constitutively expressed. Genetic stability
of the strain was assessed followinglOO generations
of nonselective growth. The constitutive phenotype
was completely stable under these conditions. G4
Phel was anticipated to be highly sensitive to chlo-
rinated aromatics as a result of the constitutive
aromatic meffl-fission pathway. This anticipated
problem was partially ameliorated through the
introduction of a derivative of the pJP4 2 ,4D-
degradativeplasmid,pRO101(Tnl721:pJP4), which
encodes an orffto-fission chlorocatechol pathway.
Results and Discussion
Pseudomonas cepacia strain G4 possesses a
novel pathway of toluene catabolism (1). Mutants
unable to hydroxylate toluene, o-cresol, m-cresol,
and phenol were also shown to be incapable of the
degradation of TCE, suggesting that an ortfzo-acting
toluene monooxygenase may be responsible for its
degradation (2). Pursuant to a more complete ge-
netic description several mutant classes were ana-
lyzed. A variant was produced from one class
lacking detectable toluene monooxygenase activity
that constitutively degraded TCE, 1,1-
dichloroethylene, and cis- and trans-1,2-
dichloroethylene (Table 1). G4 and its constitutive
derivative G4 Phel were examined for their ability
to convert 3-trifluoromethyl phenol (TFMP) to 2-
hydroxy-7,7,7-tri- fluoro-heptadienoicacid (TFHA)
during growth on lactate. This chromogenic reac-
tion requires two enzymes: a toluene/cresol
monooxygenaseand a catechol-2,3-dioxygenase (2)
(Figure 1). G4 clearly exhibited an inducible re-
sponse in its rate of production of TFHA (measured
by absorbance at 385 nm). G4Phel, however, lacked
any detectable inductive response, demonstrating
instead a fairly consistent level of activity through-
out growth. The high specific activity of the second
enzyme of TFMP metabolism, catechol-2,3-
dioxygenase (C23O), was determined under both
induced and noninduced conditions. Table 2 clearly
indicates that C23O was constitutive inG4 Phel. In
addition, the evidence suggests that C23O in G4 is
under the control of more than one promoter/
operator. ;
Stability of the constitutive property of G4
Phel was assessed after growth under non-selective
conditions (i.e., basal salts medium with sodium
lactate at 20 mM as the sole carbon source) and serial
dilution (allowing ca. 10 generations per transfer)
for 100 generations. All independent isolates exam-
ined (>1,000) were found to have maintained
constitutivity as determined by their ability to im-
mediately metabolize TFMP to TFHA.
Since chloroaromatics are frequent co-con-
taminants with TCE at waste sites, the constitutive
aromatic monooxygenase of G4 Phel might cause
serious problems since it can produce toxic
chlorocatechols from these chloroaromatics.
62
-------
Metabolic Process Characterization
G4 Phel metabolized chlorobenzene to 2-
chlorophenol and then to 3-chlorocatechol (not fur-
ther metabolized by the C23O of G4). Plasmid
pROlOl (pJP4::Tn!721), which encodes a
chlorocatechol-l,2-dioxygenase (that accepts both
mono- and dichlorinated catechols), was introduced
into this strain. G4 Phel (pROlOl) utilized 2,4-
dichlorophenoxyacetic acid as a sole carbon source,
was resistant to phenylmercuric acetate, and did not
accumulate 3-chlorocatechol. Concentrations of 2-
chlorophenoland chlorobenzene thatinterfered with
the degradation of TCE by G4 Phel (pROlOl)
(lOOuM) were 10-fold higher than those with an
inhibitory effect on the plasmid-free strain G4 Phel
(1-1
References
1. Shields, M. S., S. O. Montgomery, P. J.
Chapman,S. M. Cuskey, and P. H. Pritchard.
1989. AppL Environ. Microbiol.55:1624-1629.
2. Shields, M. S., S. O. Montgomery, S. M.
Cuskey, P. J. Chapman, and P. H. Pritchard.
1991. Appl. Environ. Microbiol. Submitted.
a,
M
8 6
'
S
H „
4-
I 2
TFHA Production by
G4
200 400 600
1.2
1.0
o.e
0.6
-0.4
0.2
X
7
1 pj
"3 8~
o
e.
bfl A
B 6~
T
•5 4—
g
M
V
"5 2-
E
' C
TFHA Production by
G4 5223 Phel ^^^^-*^^
.s**
^^_ _^/
^•"'~
/
i
i
•
X
^-'"^
-1.4
-1.2
-0.8
-0.6
-0.4
-0.2
Minutes
Figure 1. Rates of conversion of TFMP to TFHA during growth.
100
200 300
Minutes
400
500
63
-------
Metabolic Process Characterization
Table 1.
Action of P. cepacia G4 Phe(l) on chlorinated ethylenes.
% Chloroethylene Remaining3
Strain.
Uninoculated
G4 Uninduced
Phe(l)
1,1-DCE
100
±2
104
±3
50
±3
cis-
1,2-DCE
100
±4
69
±19
1212
±9
trans-
1,2-DCE
100
±9
107
±5
0M
TCE
100
±3
133
±5
2
±2
PCE
100
. ±7
' 103
±7
104
±3
a Substrate remaining expressed as percentage of that determined in uninoculated controls. Abbreviations:
1,1-DCE, 1,1-Dichloroethylene; ds-l,2-DCE, czs-l,2-Dichloroethylene, trans-l,2-T)CE, £«ms-l,2-Dichloro-
ethylene; TCE, Trichloroethylene; PCE, perchloroethylene;M, Metabolite detected by GC.
Table 2. Differences in specific activity of catechol-2,3-dioxygenase in wild type and mutants of G4.
Catechol-2,3-dioxygenase
nmoles min"* mg protein"^ ;
Substrate
Strain
G4
G4
G45223
' G45223
Phe(l)
Phe(l)
Inducer
none
phenol
none
phenol
none
phenol
Cata
2.1
53.6
0.07
13.1
156.0
48.0
SmCat
3.7
62.5
2.2 ;
31.4
50.4 ..
34.9
aCat, catecol; SmCat, 3-methylcatechol.
64
-------
SECTION SEVEN
RISK ASSESSMENT
A number of the high-priority compounds that require disposal are known carcinogens or precarcinogens. Since
biodegradation does not necessarily result in total degradation to carbon dioxide and water, researchers need to assess whether
ultimateorprocarcinogensarecreatedbyagivenbiologicaltreatment.Publichealthevaluationsmustbeconductedtodetermine
the toxicity of substances at Superfund sites, assess the safety of nonindigenous organisms, and compare bioremediation with
other potential technologies. ' .
Work in 'this research area is designed to develop comparative risk assessment methods to evaluate and contrast
mutagenic/carcinogenic products potentially generated by different microbial treatment processes in different environmental
settings. Current EPA projects are addressing the use of in vitro bioassays to evaluate the dynamic processes that occur within
Superfund sites. Results suggest that using these tests on metabolites from bioremediation, as well as products of other
remediation technologies, can be a valuable asset in evaluating treatment alternatives.
GENOTOXICITY ASSAYS OF
METABOLITES FROM BIOLOGICAL.
TREATMENT PROCESS
Larry D. Claxton, Health Effects Research
Laboratory, Research Triangle Park, North
Carolina.
The present framework for public'health
evaluations at Superfund sites and for the develop-
ment of health-based performance goals is pro-
vided in the U.S. EPA Superfund Public Health
Evaluation Manual (SPHEM) (1). Public health
evaluations are one of the driving forces for investi-
gations that determine the need to undertake reme-
dial action, for feasibility studies during the cleanup
phase, and for determining the effectiveness of the
cleanup procedures.
The health evaluation process as outlined in
the SPHEM typically involves five steps. In the first
step, indicator compounds are selected from among
the list of compounds known to be present at the
site. The selection of these indicator compounds is
based on known toxicity, physical/chemical fac-
tors, and concentration at the site. All evaluations
after this step are based upon knowledge about
these selected indicator compounds. The second
step is using knowledge about the fate and transport
of the indicator compounds to estimate exposure
concentrations. For the third step, human intake is
estimated using "standard assumptions" for daily
water and air intake. The, fourth step involves a
review of the toxicity of the indicator chemicals. The
final step is calculating human health risks from the
exposure and toxicity information.
These health evaluations are applicable in
deciding when a site requires remedial action, to aid
in feasibility studies for cleanup alternatives, and to
evaluate the effectiveness of any cleanup proce-
dures used for a site. Each, of these decision
processes, therefore, is dependent upon the identi-
fication and quantification of indicator chemicals
and the toxicity information for these compounds.
Several weaknesses associated with this pro-
cess are identified in the SPHEM. The following
statements, for example, are made:
"...important chemical data are frequently un-
available."
"...toxicity testing has not kept pace with the
need for information on many chemicals...."
"...exposure assessment often requires many
assumptions."
".. .it would be unrealistic to expect that all data
necessary to determine precisely the health risks
associated with every site will be available."
65
-------
Risk! Assessment
Although many of these weaknesses would re-
sult in an underestimation of human risks associated
with a site, a number of the data and knowledge gaps
may cause an overestimation of the human health
risks.
For most remediation technologies (not just
bioremediation), the products from the remediation
efforts are not predicted or monitored. Instead, the
process is moni tored to understand to what extent the
pollutant or indicator chemical is depleted. There are
a few exceptions to this statement. For example,
incinerators may be monitored for the production of
dioxins. The interaction of the known toxic pollutants
and other ancillary components and the degrading
microorganisms will not always produce nontoxic
substances; however, there is currently no provision
for examining this potential.
Although present knowledge and procedures
provide a usable framework for human risk assess-
ment associated with the remediation of Superfund
sites, those risk assessments (applied to all technqlo-
gies) are highly imprecise because of major data and
knowledge gaps.
Public health evaluations for Superfund sites
span a continuum of complexity, detail, level of effort,
accuracy, and precision. As the complexity of the site
increases and/or the detail of information decreases,
risk assessments typically become less .precise and
accurate. There is a need to fill important data and
knowledge gaps concerning the toxicology of sub-
stances found within Superfund sites, the safety of any
nonindigenous microorganisms, the toxicology of
bioremediation products, and the making of multime-
dia assessments for comparing bioremediation with
other potential technologies. In addition, substances
within Superfund sites have the potential for under-
going natural phototransformation and ecological
transformation thus creating additional products that
are also potentially toxic. There is a primary need,
therefore, to have bioassay monitoring and research
capabilities that can address these complex issues in a
more direct and reliable manner.
This presentation will illustrate how in vitro
bioassays that detect mutagens and many genotoxic
carcinogens can be used to evaluate the dynamic pro-
cesses that can occur within Superfund sites. For
example, sites containing high levels of the wood
preservative pentachlorophenol (PCP) could be ex-
pected to have other chlorinated phenols. Testing of
other chlorinated phenols demonstrated that some are
less mutagenic and others are more mutagenic (2).
Next, the presence of one toxicant may alter the toxic-
ity of another pollutant. For example, PCP potentiates
the genotoxicity of 2,6-dinitrotoluene (3). During
bioremediation, it is also possible for the products of
bioremediation to be genotoxic (3). Because of its
persistent nature, the herbicide 2,4,5-
trichlorophenoxyacetic acid (2,4,5-T) remains an envi-
ronmental hazard. 2,4,5-trichlorophenol (2,4,5-TCP),
one of the three reported metabolites of 2,4,5-T, is 100-
fold more genotoxic than 2,4,5-T (4). However, when
one tests the metabolites of 2,4,5-T produced by a
bioremediation organism (Pseudomonas cepacia
AC1100), the genotoxicity of 2,4,5-T is reduced when
2,4,5-T is the sole carbon source (4). Because 2,4,5-TCP
did not accummulate to a high enough degree to
increase mutagenicity, results would support the use
of bioremediation as a viable alternative for the treat-
ment of 2,4,5-T sites (4). .1
These and other results demonstrate that the
use of short-term mutagenicity assays can be a valu-
able asset in evaluating treatment alternatives includ-
ing bioremediation.
References
1. Office of Emergency arid Remedial Response.
1986. Superfund Public Health Evaluation
Manual. EPA/540/1-86/060. U.S. Environ-
mental Protection Agency, Washington, D.C.
175 pp. ;
2. DeMarini, D.M., H.G. Brooks, and D.G.
Parkes, Jr. 1990. Induction of prophage
lambda by chlorophenols. Environmental
and Molecular Mutagenesis 15:1-9.
3. Chadwick, R.W., S.E. George, J. Chang, M.J.
Kohan, J.P. Dekker, J.E. Long, M.C.Duffy,
and R.W. Williams. 1991. Potentiation of
2,6-dinitrotoluene genotoxicity in Fischer
344 rats by pretreatment with pentachloro-
phenol. Pesticide Biochemistry and Physi-
ology 39:168-181. ;
4. George, S.E., D.A. Whitehouse, and L.D.
Claxton. 1991. Genotoxicity of 2,4,5-
trichlorophenoxyacetic acid biodegradation
products in the Salmonella reversion and
lambda prophage-induction bioassays.
Environmental Toxicology and Chemistry.
In press.
66
-------
SECTION EIGHT
BIOREMEDIATION FIELD INITIATIVE
As part of Us overall strategy to increase the use ofbioremediation to treat hazardous wastes, EPA implemented the
Bioremediation Field Initiative. This program assists the regions and the states in conducting field tests and evaluations
of this technology. At more than 140 sites in the United States, treatability studies are being conducted and bioremediation
is being planned, is in full-scale operation, or has been completed. Petroleum, creosote, and solvent wastes make up almost
three quarters of the waste types undergoing bioremediation.
RESULTS OF THE BIOREMEDIATION
FIELD INITIATIVE
Fran Kremer, U.S. Environmental Protec-
tion Agency, Cincinnati, OH; John Wilson,
U.S. Environmental Protection Agency;and
Walter Kovalick and Nancy Dean,-U.S.
Environmental Protection Agency,
Washington, DC.
Background
As we approach the treatment of hazardous
wastes, which are increasingly more diverse with
respect to the contaminants and the contaminated
matrices, we will be more reliant on innovative
technologies for improved treatment efficiencies
and lower costs. To meet these objectives, the U.S.
Environmental Protection Agency's (EPA) Admin-
istrator, William Reilly, has sought to develop an
agenda for the 1990s to identify strategies for in-
creasing the use ofbioremediation for the treatment
of hazardous wastes. To develop this agenda, assis-
tance has been received from biotreatment compa-
nies, site cleanup contractors, industry, academia,
environmental organizations, and other federal
agencies, in addition to the various offices within
EPA.
One of the initial recommendations from this
consortium was the need to expand our field expe-
rience using this technology. Even though
bioremediation is a viable technology to treat some
hazardous wastes, it has not been fully utilized for
the many different types of wastes and site condi-
tions requiring remediation. It was recommended
that EPA serve as a focal point in fostering field tests,
demonstrations, and evaluations ofbioremediation,
using good test protocols and documentation of
results.
Based on this recommendation, the Office of
Solid Waste and Emergency Response (OSWER)
and the Office of Research and Development, (ORD)
have instituted a Bioremedialion Field Program.
This program provides assistance to the regions and
the states in conducting field tests and carrying out
evaluations of site cleanups using bioremediation.
Sites considered in this program include Superf und
sites, RCRA corrective actionsites, and Underground
Storage Tank (UST) sites. The program is designed
to:
1. More fully assess and document perfor-
mance of full-scale field applications of
bioremediation.
2. Provide technical assistance at various stages
of site remediation, from .site characteriza-
tion to full-scale implementation.
3. Regularly provide information on
bioremediation projects being-under taken
nationally.
As solid, full-scale performance data are
needed to assess the capabilities of this technology,
evaluations of field operations are being under-
taken. Sites considered for evaluation have field
biological units for treatment of wastes in situ or ex
situ, i.e., treatment of solids or ground water in place
or treatment in a reactor or land treatment facility.'
67
-------
Bioremediation Field Initiative
Technical assistance is available to the regions
and the states on treatability and field pilot studies.
This is to ensure adequate site characterizations,
proper design of treatability studies, and interpreta-
tion of results. In some cases, EPA may conduct the
treatability work. This assistance is available through
EPA's Technical Support Centers. Assistance is cur-
rently being provided on a number of creosote sites
and chemical facilities.
Data are being compiled on laboratory-, pilot-,
and full-scale projects in order that EPA will have a
central repository of treatment information.
Treatability data are being collected from the re-
gions, states, other federal agencies, and the private
sector. These data will be available through the Risk
Reduction Engineering Laboratory's Treatability
Database Program (513-569-7503) and also through
the Alternative Treatment Technology Information
Center (ATTIC) (301-816-9153). The Treatability
Database Program provides specific information on
the treatability of specific chemicals for a variety of
technologies,includingbioremediation. ATTIC is an
on-line information retrieval network that provides
current information on innovative treatment meth-
ods for hazardous wastes. The Treatability Database
Program and ATTIC are currently available.
To date, over 130 sites have been identified
across the country where treatability studies are
being conducted and bioremediation is being
planned, is under full-scale operation, or has been
completed. These sites include CERCLA, RCRA,
and UST sites. Petroleum, creosote, and solvent
wastes make up almost three quarters of the waste
types being biologically treated.
68
-------
SECTION NINE
OIL SPILL BIOREMEDIATION PROJECT
EPA's Oil Spill Bioremediation Project in Prince William Sound, Alaska, examined whether the addition of
nutrients to oil-contaminated beaches would sufficiently enhance oil degradation rates to enhance biodegradation. The
success of this project demonstrated that bioremediation should be considered as a key component in any cleanup strategy
for future oil spills impacting the shoreline.
OIL SPILL BIOREMEDIATION PROJECT
Parmely H. Pritchard, Environmental
Research Laboratory, U.S. Environmental
Protection Agency, Gulf Breeze, FL.
Introduction
In the several weeks following the Exxon
Valdez oil spill in Prince William Sound, Alaska,
several million gallons of Prudhoe Bay crude oil, a
well-studied oil with respect to previous cold water
biodegradation studies, had contaminated almost
300 miles of rocky coastline in Prince William Sound.
This confronted Exxon, the state of Alaska, and the
U.S. Coast Guard with the largest cleanup effort in
U.S. history. As a variety of cleanup options were
assessed and implemented, it became clear to EPA's
Office of Research and Development and its scien-
tists that bioremediation was also a reasonable
cleanup option despite the complexity of the envi-
ronmental setting. We reasoned that the oil would
become quickly colonized with oil-degrading bac-
teria but that their ability to degrade oil would be
limited by the availability of nitrogen and phospho-
rus nutrients. Artificially adding these nutrients
would therefore enhance biodegradation rates,
something that has been observed many times
in laboratory studies. Thus, the Alaskan
Bioremediation Project was initiated. An approach
was developed to determine whether the addition
of nitrogen- and phosphorus-containing fertilizers
to oil-contaminated beaches would sufficiently en-
hance oil biodegradation rates to permit consider-
ation of bioremediation as a secondary cleanup tool.
A plan was conceived to conduct an initial field
demonstration of this approach; if it were success-
ful, recommendations for wider scale application
would be made to Exxon. EPA would then provide
a followup field study as a definitive indication of
the success of the large-scale application.
The Field Demonstration
Field operations were begun in early May
1989. Two sites were selected: Snug Harbor and
Passage Cove. These beaches were mainly com-
posed of large cobblestone overlying a mixed sand
and gravel base. Both beaches had a thin layer of oil
covering the surface of the cobblestone, as well as oil
mixed into the sand and gravel under the cobble to
varying depths.
Selection of fertilizers was based on applica-
tion strategies, logistical problems for large-scale
application, commercial availability (particularly if
large-scale applicationbecamereasonable), and their
ability to provide nitrogen and phosphorus nutri-
ents to the microbial communities on the surface
and the subsurface beach material over sustained
periods. Three application strategies were adopted
for testing: commercially available slow release
formulations, an oleophilic fertilizer, and water-
soluble fertilizer applied as a solution.
Commercial slow release fertilizer formula-
tions were screened for the best nutrient release rate
characteristics. The strategy was to apply the best
product to the beach surface and then allow tidal
action to'disperse the released nutrients over the
contaminated area of the beach. The product had to
69
-------
Oil Spill Bioremediation Project
remain on the beach for several weeks while still
delivering sufficient quantities of nutrients.
Oleophilic fertilizers are thought to essentially dis-
solve the nutrients into the oil when applied directly
to the oiled beach material. Nutrients sequestered
in the oil phase would presumably facilitate bacte-
rial growth on the surface over sustained periods.
The oleophilic fertilizer Inipol EAP 22, produced by
Elf Aquataine Company (Artix, France), was se-
lected. Fertilizer granules (about 2 to 3 mm in
diameter), produced by Sierra Chemicals (Milpitas,
California)/ were selected as the main slow release
fertilizer formulation. These granules (Customblen)
have a N:P:K ratio of 28:8:0 and slowly release
ammonia, nitrate, and phosphate from inorganic
ammonium nitrate and ammonium phosphate en-
capsulated within a diene-treated vegetable oil
coating. The fertilizer granules were broadcast onto
the beach surface at a concentration of 90 gm/m2
using a mechanical seed spreader. Their high spe-
cific gravity, propensity to adhere to the oil, and
tendency to entrain under rocks and in interstitial
spaces ensured that they would remain on most low
and moderate energy beaches in Prince William
Sound for 2 to 3 weeks.
The third type of fertilizer application in-
volved spray irrigation with an aqueous fertilizer
solution. This approach produced the most de-
fined, controlled, and reproducible introduction of
nutrients into the oiled beach material, particularly
for oilbelow the beach surface. Itwas accomplished
by dissolving commercially available sources of
ammonium nitrate (34-0-0) and triple phosphate (0-
45-0) into seawater pumped from below the beach.
The resulting fertilizer solution was then sprayed
over the beach surface at low tide using a pump and
lawn sprinkler heads.
The first application of the oleophilic fertil-
izer occurred July 8,1989, at the Snug Harbor site.
Approximately 2 to 3 weeks following application,
the treated beach showed a visually pronounced
reduction in the amount of oil on the surface of tiie
cobblestone. This produced a striking "window"
against the oiled beach background. Differences
between treated and untreated portions of thebeach
were dramatic. Close examination of the beach,
however, revealed that significant quantities of oil
remained under the cobblestone as well as within
the beach subsurface. Over the next few weeks,
however, even this oil slowly disappeared. This
contrasted with the untreated control areas, in which
there was little visual change. Subsequent studies in
the laboratory verified that Inipol was not a chemi-
cal rock washer.
Definitive information on the role of biodeg-
radation in this event was established by extracting
oil from surface samples of cobble in the oleophilic
fertilizer-treated beach and analyzing the extracts
by gas chromatography. The sampling and analysis
showed that this visual disappearance of the oil was
accompanied by significant decreases in total oil
residues (i.e., weight of extractable material) and
changes in hydrocarbon composition. This change
in hydrocarbon composition was largely due to
biodegradation. Thus, it is reasonable to assume
that the decreases in oil residue weight on the cobble-
stone were caused by biodegradation. We suspect
that after a certain extent of oil biodegradation was
achieved, the physical nature of the oil changed into
a less sticky, flaky consistency,, and that this innocu-
ous degraded material was then easily scoured from
the rock surfaces by tidal action.
Summary and Conclusions
Results from our initial field studies were
sufficient for Exxon to consider the use of
bioremediation on a large scale as a finishing step
for their cleanup effort. We recommended that the
Oleophilic fertilizer, Inipol, be applied to beaches
with only surface oil, and that a combination of
Inipol and the fertilizer granules (Customblen) be
applied to beaches with both surface and subsurface
oil. The granules provided a simple means of releas-
ing nutrients into the beach subsurface by tidal
action and thereby potentially enhancing biodegra-
dation of subsurface oil.
Exxon began fertilizer application in early
August 1989 to approximately 50 miles of beach in
Prince William Sound that had been physically
washed. Increasing the biodegradation rate of oil at
this point was very important because meiximal
degradation could be achieved before winter condi-
tions slowed biodegradation processes. In many
cases, the results of large-scale fertilizer application
were as dramatic as our initial observations at Snug
70
-------
Oil Spill Bioremediation Project
Harbor; that is, where the oil was spread thinly over
the cobble surface (as was the case on many beaches
that had been physically washed), the oil disap-
peared over a 20-day period. Unexpectedly, it also
appeared that even oil underneath the cobble had
disappeared in a shorter time period than that ob-
served at Snug Harbor. Although it is difficult to
prove experimentally, we believe that the physical
cleaning process used by Exxon dispersed the oil
throughout the beach material to such an extent that
the exposed oil surface area was greatly increased,
allowing greater bacterial colonization and subse-
quent biodegradation.
Our Passage Cove study was initiated in late
July 1989 as the definitive technical support site for
the large-scale application of Inipol and Customblen
fertilizers. These fertilizers were applied in combi-
nation to a large test beach and samples of beach
material were analyzed for changes in oil residue
weight and aliphatic hydrocarbon composition.
These changes were compared to those observed in
an untreated control beach. In addition, a beach
treated with a seawater solution of inorganic fertil-
izer (applied via a sprinkler system) was examined
in the same way.
Approximately 2 to 3 weeks following initia-
tion of this study, oil on the cobble surfaces in the
Customblen/Inipol- and the fertilizer solution-
treated beaches had been degraded to the point of
producing visibly cleaner surfaces, much as we had
seen in Snug Harbor. Surface oil on the control
beach, however, was still very apparent, showing
no visual reduction in the amount of oil. Disappear-
ance of oil from the rock surfaces on the beach
treated with the fertilizer solution provided defini-
tive proof that biodegradation (and not chemical
washing) was responsible for the oil removal, as
there was no other reasonable mechanism to ex-
plain the effect of nutrient addition.
Despite sampling and interpretation compli-
cations resulting from the high variability in oil
distribution on the beaches, we have been able to
show, statistically, that oil biodegradation (as mea-
sured by changes in oil chemistry) was significantly
greater on the beach treated with the fertilizer solu-
tion than it was on the control beach. Based on this
information, we projected that after 45 days ap-
proximately 4 to 5 times more oil remained on the
control test beach than on the fertilizer solution-
treated test beach. This corresponded to an
enhanced biodegradation rate of about two to three
fold. On the Inipol/Customblen-treated beach, it
appeared that accelerated biodegradation
(approximately a two- to threefold increase in bio-
degradation rates) occurred early in the test when
nutrient concentrations were highest.
The long-term benefit of fertilizer application
was realized during examination of the beaches in
Passage Cove in November 1989 and early June
1990. At the later sampling, virtually no oil was
observed on either of the treated beaches, at both the
surface and the subsurface (8-in. depth). However,
the untreated control still showed areas of heavy oil
contamination in the subsurface beach material.
These observations provided the final definitive
demonstration of the long-term success that can be
expected from bioremediation of oil-contaminated
beaches.
As a result, in spring of 1990, bioremediation
became an integral part of a cleanup plan for the
remaining oil-contaminated shorelines in Prince
William Sound. To follow the success of this treat-
ment, a joint bioremediation monitoring program
was conceived and implemented by scientists from
Exxon, EPA, ADEC and the University of Alaska
(usinglogistical and resources support from Exxon).
The central success of this monitoring program
paved the way for multiple reapplication of the
fertilizers during summer 1990, a necessary step in
many cases because of the large quantities of oil
remaining in some areas.
The success of our field demonstration pro-
gram has now set the stage for the consideration of
bioremediation as a key component in any cleanup
strategy developed for future oil spills. Its use and
effectiveness will depend on the amount of oil present
in the contaminated environmental matrix, i.e., a
longer time will be required for degradation of high
concentrations of oil, and consequently a longer
period of fertilizer application will also be required.
In addition, location of the oil (in the absence of
physical cleanup, subsurface oil may only be treat-
able by bioremediation) and the acceptability of
other cleanup options must be considered. In most
71
-------
Oil Spill Bioremediation Project
aquatic environments, enrichments of oil-degrad-
ing microbial communities occur relatively soon
after oil contaminates shorelines. It is unlikely that
natural sources of nitrogen and phosphorus will be
sufficient to give maximal degradation rates in light
of the available degradable organic carbon from the
oil. Thus, the application of fertilizers should en-
hance degradation and eventually remove the oil.
Although oxygen may become limiting in certain
situations (e.g., fine-grain sandy beaches), the high
porosity and large tidal fluxes characteristic of Prince
William Sound beaches precluded this as a
limitation.
Acknowledgements
The success of this project would not have
been possible without the able assistance and col-
laboration of the following EPA scientists and their
relevant expertise: Dr. John Rogers (microbial analy-
ses), Drs. Al Venosa and John Glaser (fertilizer
evaluation), Dr. Fran Kremer (field site coordina-
tion), Mr. DanHeggem (quality assurance), Drs. Jim
Clark and Rick Coffin (ecological assessment), Dr.
Larry Claxton (genotoxicity), and Dr. John Haines
(fertilizer application). We wish also to acknowl-
edge the special contributions from John Baker,
SteveMcCutcheon,DickValentinetti,Dennis Millar,
and Steve Safferman. Without the resources and
support of Exxon this project would not have been
possible; we acknowledge the valuable scientific
interaction with Drs. Steve Hinton, Roger Prince,
and Russ Chianelli at Exxon. We gratefully ac-
knowledge the guidance, encouragement, and
wisdom offered by Dr. Eric Bretthauer and John
Skinner, director and deputy director, respectively,
of the EPA Office of Research and Development.
72
•&U.S. GOVERNMENT PRINTING OFFICE: 1992 - 648-003/41848:
-------
-------
"D
CD
"05"
g. o>
18
a.
s
a
»'
-
3 ®
nt ^
m
ro
m
•
8
SL
•= 5"!
™ a c
r.3-
S<
g 8
8g-
?3 nr ^*
; a.
33
o
•o
y.
8
•Q
:D
o
O
------- |