N11&
                                  Information Is our bustou.
BIOREMEDIATION  OF PETROLEUM HYDROCARBONS
A  FLEXIBLE, VARIABLE SPEED TECHNOLOGY
(U.S.) NATIONAL RISK MANAGEMENT RESEARCH LAB., ADA, OK
1996
 U.S. DEPARTMENT OF COMMERCE
 National Technical Information Service

-------
                                    TECHNICAL REPORT DATA
                             (Please read Imtructions on the reverse before completing]
  1. REPORT NO.
       EPA/600/A-95/140
2.
  4. TITLE AND SUBTITLE
   BIOREMEDIATION  OF PETROLEUM HYDROCARBONS: A
   FLEXIBLE,  VARIABLE,  SPEED  TECHNOLOGY
                              ). RECIPII
                             5. REPORT UA i c
                             6. PERFORMING ORGAmZAT ION CODE
 7. AUTHOR(S)
                              3. PERFORMING ORGANIZATION REPORT NO.
   RICHARD A.'  BROWN (1),  ROBERT HINCHEE  (2), ROBERT
   D. NORRIS  (3), AND JOHN T.  WILSON (4)
 9. PERFORMING ORGANIZATION NAME AND ADDRESS
   GROUNDWATER  TECHNOLOGY,  TRENTON,  NJ (1),  PARSONS
   ENGINEERING  SCIENCE,  S.  JORDAN, UT, (2),  ECKEN-
   FELDER  INC.,  NASHVILLE,  TN  (3), JOHN T.  WILSON,
   US/EPA,  ADA,  OK  (A)
                                                            10. PROGRAM ELEMENT NO.
                             11. CONTRACT/GRANT NO.
                                                              IN-HOUSE  RPJW9
  2. SPONSORING AGENCY NAME AND ADDRESS
   U.S. /  EPA,  NRMRL-ADA
   SUBSURFACE  PROTECTION  &  REMEDIATION DIVISION
   P.O. BOX 1198
   ADA, OK  74820
                             13. TYPE OF REPORT AMD PGRlOOCOVCREO

                               PROCEEDINGS
                             14. SPONSORING AGE'NCY CODE

                               EPA/600/15
  G. SUPPLEMENTARY NOTES
 G. AOSTRACT
       The bioreraediation of petroleum hydrocarbons has evolved into a number of different
       processes. These processes include in situ aquifer bioremediation, bioventing,
       biosparging, passive bioremediation with oxygen release compounds, and intrinsic
       bioremediation. While these processes are often viewed as competing technologies they
       are actually part of a continuum of biodegradation processes governed''primarily by the
       interplay between oxygen and carbon availability.  Generally the more carbon that needs
       to be removed per unit time ;.he more oxygen that needs to be supplied. As the carbon
       availability or desired  removal rate decreases, so does the oxygen requirement.  By
       understanding this continuum approach, bioremediation can be applied  as a  flexible,
       variable-speed technology, where the effort can be increased or decreased through
       oxygen supply. This paper will discuss the carbon-oxygen demands of each process and
       the interplay between  processes.  The paper will provide operating guidelines for
       configuring bioiemediation systems for maximum  flexibility.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
  OISTRIQUTION STATEMENT
  RELEASE TO PUBLIC
                                              b. IDENTIFIERS/OPEN ENDED TERMC
                                              19. SECURITY CLASS iTIni K<-pr>,,i

                                                UNCLASSIFIED
               20. SECURITY CLASS {Tlu'i

                 UNCLASSIFIED
                                             COSATi Picld.CiOup
                                          }i NO OF "AGEs
                                            15
EPAFo
-------
                                                               EPA/600/A-95/140
   Bioremediation of Petroleum Hydrocarbons: A Flexible, Variable Speed Technology

                                 Richard A. Brown
                           Groundwater Technology, Inc.

                                  Robert Hinchee
                            Parsons Engineering Science

                                 Robert D. Morris
                                 Eckenfelder, Inc.

                                   John Wilson
                       U.S. Environmental Protection Agency
ABSTRACT

The bioremediation of petroleum hydrocarbons has evolved into a number of different
processes.  These processes include in situ aquifer bioremediation, bioventing,
biosparging, passive bioremediation with oxygen release compounds, and intrinsic
bioremediation.  While these processes  are often viewed as competing technologies they
are actually part of a continuum of biodegf adation processes governed primarily by the
interplay between oxygen and carbon availability.  Generally the more carbon that needs
to be  removed per unit time the more oxygen that  needs to be supplied.  As the carbon
availability or desired removal rate decreases, so does the oxygen requirement.  By
understanding this continuum approach, bioremediation can be applied as a flexible,
variable-speed technology, where the effort can be  increased or decreased through
oxygen supply. This paper will discuss the carbon-oxygen demands of each process and
the interplay between processes.  The paper will provide operating guidelines for
configuring bioiemediation systems for maximum flexibility.

INTRODUCTION

The bioremediation of petroleum hydrocarbons was established as a key remediation
technology in 1972 when Richard Raymond employed it on a Sun Oil gasoline pipeline
in Ambler, PA (Raymond 1976).  Raymond demonstrated that the controlled application
of oxygen and nutrients could substantially reduce hydrocarbon contamination.
Hydrocarbon bioremediation has evolved considerably since that first application. Its
evolution has taken two pathways. First, the  basic  in situ process (the Raymond Process)
has become more efficient as both oxygenation  systems and nutrient formulations have
substantially improved. Early hydrocarbon bioremediation projects often took five years
and over a million dollars to remediate  a retail  gasoline station. Today  the process can
be accomplished in less than two years at a cost of several  hundred thousand dollars.  A
key factor in this increased efficiency has been the  development of aeration systems, such
as soil vapor extraction (SVE) and air sparging, which can  more easily supply large
quantities of oxygen to the biodegradation process  than the earlier systems.  Second, the

-------
range of application of hydrocarbon bioremediation has evolved from solely the
aggressive treatment of soils and groundwater to include slower and/or more subtle
processes such as bioventing, passive bioremediation, and intrinsic bioremediation. The
focus of early bioremediation development was  on achieving lower clean-up levels in
shorter periods of time.  Today the focus is on processes which accomplish the same
level of remediation at lower costs but over a considerably longer period of time.
Petroleum hydrocarbon bioremediation, today, represents a continuum of processes that
provides a flexibility in effort, time, and cost.  On the one end of the spectrum are in situ
bioremediation which are cost intensive and aggressive processes that can rapidly treat
high levels of soil and/or groundwater contamination. On the other end are intrinsic
bioremediation which is an often inexpensive but lengthy process for removing low levels
of contamination.

Considering the  broad spectrum of hydrocarbon bioremediation processes available,  the
technology can be classified into essentially four levels of application:

Aggressive (Raymond Process). The Raymond  Process is an aggressive source reduction
technology.  Oxygen (air) is aggressively and continuously supplied, and nutrients are
frequently or continuously supplied.  Treatment time  are typically between one and four
years.  This process is applied at sites that have  a high level of contamination and/or
require rapid clean-up. It is effective lor both soil and groundwater contamination.  The
Raymond Process often uses supplemental technologies such as SVE or pump and treat
to accelerate clean-up and to control the potential migration of vapors or dissolved
contaminants.

Moderate Intensity  (Bioventing/Biosparging). These are moderate intensity processes.
Bioventing addresses vadose zone (unsaturated) contamination. With bioventing oxygen
(air) is supplied  at a rate sufficient to support biodegradation while minimizing the
transport and recovery of volatile hydrocarbons. Nutrients (and moisture) are generally
not applied.  Bioventing requires two to ten years to remediate total hydrocarbons, but
will address  BTEX contamination more rapidly  than total hydrocarbons.  The process is
applied to sites that either have relatively low to moderate levels of contamination or
have low levels of volatiles. Bioventing is used to remediate soil contamination.
Biosparging  addresses contamination in  the capillary fringe or below the water table.
With biosparging, oxygen (air) is injected into the aquifer.  This injection volatilizes and
removes VOCs.  With biosparging  the injection  rate is adjusted so that VOCs released
into the vadose zone are biodegraded in the soil column before the sparged air reaches
the ground surface.  Nutrients are generally not applied.   Biosparging may require up to
eight to ten years to remediate total  hydrocarbons. It is applied to sites  that have either
a relatively low to moderate level of contamination, few volatiles,  or a deep vadose zone.

Low Intensity - Passive (Oxygen Releasing Compound - ORC). With passive
bioremediation systems oxygen is supplied by a  slow release oxygen compound - i.e.
magnesium peroxide.  Dispersion of the oxygen  is largely through  advective groundwater
flow.  Nutrients are generally not applied. ORC systems are primarily used for
groundwater control.  They may, however, be used to remediate low level dissolved

-------
  plumes. Treatment time varies with the degree of contamination but may require eight
  to ten years.

  Intrinsic. Intrinsic bioremediation is the least intensive application of bioremediation.  It
  uses the ambient levels of oxygen (or other electron acceptors) and nutrients. It can be
  either an aerobic process (using oxygen) or an anaerobic process.  Intrinsic
  bioremediation is effective in controlling and mitigating dissolved plumes.  Treatment
  times vary considerably depending on the level of contamination.  In some case, intrinsic
  remediation may be as fast or faster than more intensive processes.  Intrinsic
  remediation is most effective with low levels of contamination or when the rate  limiting
  factor is the availability of carbon rather than the supply of oxygen or nutrients.

 A more detailed description of each  level of bioremediation follows.  While each level of
 technology has been developed somewhat independently, there is a common link
 between them.  This link is the interplay between oxygen and carbon availability. Carbon
 availability represents the mobile fraction of contamination  that is available to
 biodegradation; it is the sum of dissolved  and vapor phase carbon.

 Aggressive Bioremediation
 The Raymond Process is based on the principle that naturally occurring biodegradation
 processes can be stimulated by supplying enhanced levels of oxygen and nutrients. The
 early forms of the technology used in-well aeration and the  addition of high
 concentrations of ammonium salts and ortho phosphates. The effectiveness of the early
 technology was limited by the oxygenation system.  These in-well aeration systems were
 able to sustain dissolved oxygen (DO) levels of 3-5 mg/L at 10-20 feet from the well.
 Sites with high degrees of contamination or where oxygen transport was limited were
 slow to respond.

 The second phase of development focused on the use of hydrogen peroxide which was a
 "chemical" carriers for oxygen.  By supplying one to two orders of magnitude more
 available oxygen - 250 to 500 mg/L versus 8 to 10 mg/L - the peroxide based systems
 shortened the time of treatment and extended the range of bioremediation to more
 contaminated sites (Brown et.al, 1984).  In some cases, the peroxide base systems are
 able to sustain DO levels of 5-8 mg/L at distances of 20-30 feet from the injection point.
 The peroxide based system was, however, also limited in effectiveness for highly
 contaminated  sites.

 The third phase of development was the use of aeration technologies.  First, starting  in
 the mid 1980s with soil vapor extraction and later, from 1989 to 1991  with air sparging,
 aeration was recognized as an efficient means of supplying oxygen for bioremediation. By
 1992 most bioremediation projects employed air based  systems rather than chemical
based systems (Brown & Jasiulewicz, 1992) resulting in accelerated clean-ups of highly
contaminated sites.  At some sites, air sparge systems are able to sustain DO levels of 5-
8 mg/L over essentially the entire area of treatment.

-------
The practice of in situ hydrocarbon bioremediation involves the controlled application of
oxygen and nutrients.  There is no upper limit on the amount of available carbon that
can be treated.  At sites with high levels of available carbon, in situ bioremediation is
typically part of a suite of integrated technologies such as NAPL recovery, soil vapor
extraction, air sparging, and groundwater recovery.  In situ  bioremediation is predicated
on ihe assumption that the limiting factors in treatment are the availability of oxygen and
nutrients, particularly oxygen.  Therefore maximum aeration rates are used.  As a result
volatilization rates may be in excess of biodegradation rates and, therefore, vapor
capture and treatment may be required.

The value of aggressive bioremediation diminishes as the carbon availability decreases.
With sites that have less than 20 mg/kg soil - day of carbon availability, less intensive
bioremediation processes may be more  cost  effective. Sites that are contaminated with
insoluble and non volatile hydrocarbon  mixtures - heavy fuel oils, crude oils, weathered
diesel - do not  respond any more to aggressive bioremediation than they do to less
intensive processes such as bioventing.

Moderate Intensity Bioremediation
Bioventing and biosparging are related, moderate intensity  bioremediation technologies.
Bioventing is the introduction of air into unsaturated soils;  biosparging is the
introduction of air into the saturated zone.  Both depend upon inducing air flow in-situ
to provide oxygen.  In most applications this involves the injection of air into the
subsurfar-5.  Bioventing can also utilize  air extraction, allowing air entry from the
atmosphere through the ground surface. The rate  of air injection is controlled to
minimize direct volatilization while enhancing aerobic biodegradation.

Bioventing:  The first documented evidence  of bioventing was reported by the Texas
Research Institute, Inc. in a study of soil vacuum extraction (SVE) for the American
Petroleum Institute (Texas Research Institute 1980). It was observed that a portion of
the  hydrocarbon removed was as the result of aerobic biodegradation and not
volatilization.  Wilson and Ward (1986) suggested that using air as a carrier for oxygen
could be 1,000  times more efficient than water.  In  a review of soil vacuum extraction
processes, (Bennedsen et al 1987) concluded that the SVE  process provided large
quantities of oxygen and in many cases  stimulated biodegradation. The process of
bioventing as currently practiced, however, has evolved from parallel development in the
late  1980s and early 1990s by Chevron (Ely  and  Heffner 1988), the U.S. EPA (Kampbell
et al  1992), the U.S. Air Force  (Dupont et al 199?.,  Miller et al  1991), and the U.S. Navy
(Hoeppel et al  1991).  These efforts have led to wide spread commercial application of
the  technology.  One notable example of this is the U.S. Air Force's Bioventing Institute
in which the technology has been applied at over 120 U.S.  Air Force sites throughout the
United States (Leeson et al 1995).

Bioventing has  been applied at many different sites to a variety of contaminants.  The
technology is applicable to any  contaminant  which  degrades more rapidly under aerobic
than under anaerobic conditions. To date, most applications have been to petroleum

-------
 and petroleum distillate fuels, but success has been reported with PAH compounds and
 other non-petroleum contaminants (Hinchee  and Olfenbuttel 1991).

 Bioventing, although potentially applicable to any aerobically degraded compound, has a
 practical limitation in application to more volatile compounds.  If a compound is too
 volatile it may not biodegrade before  it is released into the soil gas.  The vapor pressure
 above which a compound is difficult to biovent  is somewhat site specific.  On many sites,
 however, it is not safe or feasible to inject air to biovent highly volatile hydrocarbon
 mixtures such as gasoline without air extraction to  control gas release.  In such cases, the
 process begins to look more like conventional soil vapor extraction because of the need
 for vapor capture and treatment.

 One significant limitation to bioventing is the rate  of biodegradation.  After overcoming
 the oxygen limitation, the systems typically become bioavailability limited at rates in the
 range of 2  to 20 mg/kg-day.  Typically the light aromatic fraction, benzene, toluene, and
 xylenes (BTEX) are most rapidly degraded. At most sites, BTEX is remedied in a year
 or less.  The higher molecular weight  hydrocarbons degrade more slowly, frequently
 result!.ig in clean up times on the order of 10 years to reach TPH standards.

 A variety of techniques have been utilized in  attempts to increase bioventing rates.
 Probably the most successful field demonstrated technique for accelerating bioventing
 rates is  soil warming.  Sayles found that by warming soils in Alaska to'20° C bioventing
 rates increased 2 to 3 fold (Sayles et al  1995). Laboratory testing suggests that
 nutrification may increase rates; however, field studies have failed to confirm a benefit
 due to nutrification (Miller et al 1991, Dupont et al 1991, and Leeson et al 1995).

 Biosparging-.  While the use of air sparging to support bioremediation was first suggested
by Raymond's use of in-well aerators  or spargers to introduce transport oxygen into
groundwater, it is likely that in his application the air did not directly enter the aquifer
as a separate phase as is common in the current practice of air sparging.  Dieter Hiller
began applying true air sparging in Germany  in the 1980's (Gudemann and Hiller 1988)
and its application began spreading in the United States in the  late 1980s and early
 1990s.  Early in its development, it was  recognized that air sparging could supply
significant oxygen to support biodegradation (Bianchi-Mosquera et al 1994, Brown and
Jasiulewicz  1992).
                                           *
 In biosparging air is injected into the aquifer  at pressures  in excess of the hydrostatic
pressure. The air moves in channels from the injection point to the vadose zone. As the
 air passes through  the aquifer materials, oxygen dissolves into water and stimulates
biodegradation. If the containment is volatile it may be transported in the air channel to
 the vadose zone.  Once in the vadose  zone the contaminant may biodegrade, or it may
be extracted by an SVE well.  The air flow can  be  adjusted so that the volatilized
 contaminant is completely biodegraded  by the time it reaches the ground surface.

The physics of air movement in an aquifer are not well understood. A key question is
 the amount of oxygen that is transferred from the air phase to the water phase. Some

-------
early workers assumed that air transport was in bubble form, and that the aquifer acted
like a well mixed system.  This lent to overly optimize projections of biosparging's
success (Johnson et aJ  1993). Another problem was that early workers did not
understand the presence of monitoring wells or the impact of the air channeling. In
some cases air  channels intercept monitoring wells resulting  in a "clean" well when the
air enters the well and forms bubbles which rapidly strip volatiles from and add oxygen
to the water in the well.  These effects are not necessarily transferred  to the surrounding
aquifer.  The degree of oxygen transferred to the groundwater and volatile organic
transfer into air channels is now being studied by a variety of groups, including the API,
US EPA, and U.S. Air Force.

Biosparging has the same  basic biodegradation rate limitations as bioventing. It appears
unlikely that biosparging can result in greater biodegradation rates than bioventing, and
in fact, at may sites oxygen transfer may limit biodegradation rates to  less than
bioventing.

Low Intensity (Passive) Bioremediation
For in situ aquifer remediation  using the Raymond method,  air spaiging and soil vapor
extraction are utilized to supply large amounts of oxygen (electron acceptor) to
remediate sites with moderate to heavy contamination.  However, the  conditions at
many, more marginally contaminated sites are such that while the total electron acceotor
demand is limited, it is not high  enough to warrant aggressive effort but is also not low
enough to be met through natural recharge processes.  Thus, for those sites where  there
is sufficient contamination to cause concern but where intrinsic bioremediation is not
sufficient, or is  too slow, or does not provide sufficient control for site owners and
regulators, some form of assistance to the indigenous microbial processes is appropriate.
In general  these efforts focus on providing electron acceptors and possibly nutrients.
Nutrient addition, however, is not Likely to be necessary because marginally
contaminated sites, by definition, have a low level of biodegradable constituents and thus
a low level of demand  of nutrients.  Passive systems that slowly release an electron
acceptor offer the potential to provide nearly maintenance free systems.  For example,
oxygen release compounds represent an approach to providing oxygen as a remedy for
aerobically biodegradable  contaminants.  While relatively small air sparging and
bioventing systems can be generally unobtrusive, they both require installation and
operation and maintenance of mechanical components.  Oxygen release compounds have
been field tested  and are applicable for use a£ migration barriers and  for source
reduction.

Biological Migration Barriers:  The introduction of oxygen above ambient recharge
levels across the plume will result in decreased concentrations of biodegradable
constituents such as BTEX in excess of natural or intrinsic rates.  Oxygen addition  can
be done using a row of air sparging wells, an interceptor trench in which air is release
from a slotted horizontal PVC pipe, or a row of wells containing sleeves of a slow
release oxygen compound  placed perpendicular to the flow  of groundwater. The point
of interception  of the plume is located  to  meet site objectives such as  prevention of
migration across property  lines.  Within a given plume, the closer the  barrier is located

-------
 to the source area, the greater is the demand for electron acceptors.  The location may
 also be influenced by infrastructure restrictions such as buildings, utility trenches,
 driveways, etc.

 Biological migration barriers have been studied on a pilot scale and implemented
 successfully at full scale.  The choice of barrier is typically between air sparging or
 oxygen release compounds. Air sparging barriers offer the advantage of relative wide
 spacing of wells and providing a relative wide oxygenated zone. The disadvantages
 include the need for ongoing maintenance  and the potential for volatile compounds to be
 transferred to the unsaturated  zone and eventually to the atmosphere (in many cases the
 unsaturated soils may serve as an effective biofilter).  The migration of volatile
 compounds is especially bothersome in the vicinity of utility trenches and building? with
 substructures.

 Rows of wells containing oxygen  release compound require closer spacing than air
 sparging wells as distribution of oxygen away from the well is dependent on advection,
 and to a lesser extent, diffusion.  However, there is virtually no potential to transfer
 volatile compounds  to the unsaturated zone.  Further, once the oxygen release compound
 is installed, the only maintenance required is periodic replacement of the oxygen release
 compound. Replacement is required once every four to six months.

 Field studies  using full  scale barriers have been conducted in Alaska (Marlow 1995),
 North Carolina (Kao and Borden 1994), New Mexico (Johnson et.al, 1995), and at the
 Borden Landfill in Ontario (Bianchi-Mosquera 1994) using magnesium peroxide (ORC)
 manufactured by REGENESIS Bioremediation Products.  These workers were able to
 dramatically reduce  BTEX levels down gradient of the release.

The Borden Landfill study consisted cf a controlled release of benzene and toluene up
 gradient from a row of ORC containing wells and included the use of several types of
 controls. Extensive  monitoring of wells and piezometers down gradient of the source
 (ORC) wells showed increases in dissolved oxygen from initial levels of 0.5 mg/1 to levels
 in the range of 8 mg/1  to 11 mg/1 following introduction of ORC.   Benzene and toluene
 levels were reduced  by greater than  97 percent compared to both  initial  levels and levels
 in control wells and  piezometers.

The most extensive  field demonstration is being conducted by GRAM, Inc in close
 coordination with the New Mexico Environmental Department (NMED) in Helen, New
 Mexico. Initially, a  small scale demonstration was conducted to evaluate the feasibility
 of using ORC on this site. Increased DO levels and reduced  BTEX levels  down gradient
of the ORC source wells were  sufficient to justify installation of a  full scale system.
 GRAM designed and installed a full scale system consisting of twenty 6-inch ORC wells.
The vertical distribution of DO and BTEX was measured with probes located 3,  10, and
 17 feet below the water table at each of the 54 down gradient monitoring points.
Oxygen distribution  was estimated by contouring the area! and vertical distribution of
 initial oxygen concentrations and  increased oxygen concentrations  at ten  sampling times
 over a three month  period. BTEX was evaluated using the same methodology.  Within

-------
three to five days oxygen levels increased to levels that could support bioremediation
over at least a twenty foot distance down gradient of the source and remained relatively
constant. Decreased BTEX levels were also quickly evident.

Source Area Migration Control:  In those cases where a source area exists or is under
treatment, it may be possible to reduce overall treatment costs by implementing a limited
level of bioremediation activity near the  source area. This approach has been tried with
air sparging wells and with batch addition of electron acceptors to wells located within or
immediately down gradient of the source area. Air sparging, however,  requires ongoing
maintenance and may  result in concerns  regarding the transfer of volatiles to the
unsaturated zone and,  subsequently, the atmosphere. Batch additions are generally labor
intensive and may be difficult to implement.  Placement of an oxygen release compound
in one or more wells within or slightly down gradient of the source area has the benefit
of requiring little field activity other than monitoring a widely spaced intervals.   In some
cases, sleeves of oxygen release compounds can be placed in existing wells and thus
minimal cost and effort are required beyond .continued monitoring and documentation.

Recently, a sixteen site evaluation of the efficacy of ORC  for reduction in dissolved
phase hydrocarbons has been completed  by several independent contractors in
conjunction with several major oil companies.  In these studies filter socks containing
ORC were placed in a single well.  The increase  in dissolved oxygen and the reduction in
dissolved phase benzene, toluene, ethyl benzene,  and total xylenes (BTEX)  were
monitored.  Dissolved  oxygen  (DO) levels, even  in the presence of dissolved phase
hydrocarbons, generally ranged from 20 to 30  mg/1. The reductions in BTEX levels
typically averaged greater than 80 percent and were greater than 95 perceni in half of
the wells tested. In absolute terms, the reductions in BTEX were as much  as 47.5  mg/1.

Additional studies  are  needed for this and similar application of oxygen release
compounds as well as other methods of introducing electron acceptors  that  are
unobtrusive and relatively low cost.  Bioremediation methods that can address low levels
of contamination will continue to be identified and developed to meet  specific niches.

Intrinsic Bioremediation
Intrinsic remediation is the use of ambient natural biodegradation processes to remove
low levels of dissolved  organics.  Natural bioremediation has been used both
intentionally and inadvertently as a means to control migration. Intrinsic bioremediation
can use a wide range of electron acceptors which are listed in Table 1  (Wilson et al
1994, Borden et al 1994).  The table lists the electron acceptors in order of increasing
reducing conditions.  Of these electron acceptors, oxygen and carbon  dioxide are the
most readily  available being replenished  by natural recharge processes. Sulfate, iron, and
manganese also occur naturally but are dependent on site  mineralogy.  The predominant
sources of nitrate are anthropogenic activities  such as agricultural fertilization. Which of
these electron acceptors is used  is dependent on the redox conditions (aerobic vs.
anaerobic) in the subsurface.  In intrinsic bioremediation oxygen is consumed first, then,
once anaerobic conditions have been established,  nitrate,  sulfate, and the other electron
acceptors are consumed in order of their redox potentials. Oxygen depiction requires the

-------
 presence of a carbon substrate that is readily amenable to aerobic biodegradation (e.g.,
 hydrocarbons).  Carbon sources which are resistant to all but methanogenic processes
 (e.g., PCBs) will not drive the depletion of oxygen and thus will not "activate" other
 pathways.

 Table 1. Common naturally occurring electron acceptors.

       Electron Acceptor         Metabolic Pathway           Natural Sources

          Oxygen (O2)                   Aerobic                    Atmosphere

    Nitrate/Nitrite (NO3'/NO2')        Facultative Anaerobic      Agriculture/Septic//lfttio5/>here*
Sulfatc (SO/5)
Iron (Fe+3)
Manganese (Mo*4)
Carbon Dioxide (CO7)
Anaerobic •
Anaerobic
Anaerobic
Methanogenic
M in e rah /A tm osph ere1
Minerals
Minerals
Atmosphere/A/iAiera/J
urSecon3ary source (Italics)
Intrinsic bioremediation has two uses.  The first is to control the leaching of
contaminants from  untreated or residual sources.  The second is to control and mitigate
low level dissolved plumes.

The first case study illustrates that intrinsic remediation can adequately reduce the risk
of groundwater impact due to residual contamination. Studies of a BTEX plume  at a
gas plant provide evidence that intrinsic bioremediation processes limited the migration
of benzene and other BTEX constituents (Piontek et aJ. 1994).  Groundwater upgradient,
within, and downgradient  of the NAPL zone was analyzed for BTEX, DO, sulfate, iron,
nitrate,  methane and bicarbonate.  It was observed from this data that the electron
acceptors (DO, sulfate, nitrate) decreased in the down gradient direction while the
resulting reduced forms (sulfide, iron (+2), bicarbonate, and methane) increased.  This
paralleled a decrease in the BTEX levels which was in excess of the  decrease that would
be expected based solely on retardation factors and groundwater flow. The changes in
the concentrations of the electron acceptors indicated that sulfate reduction accounted
for the majority of the intrinsic remediation.  The reduction in BTEX demonstrated that
intrinsic remediation could mitigate the impact to groundwater from  source areas.

The second application of intrinsic remediation in mitigating low levels of dissolved
hydrocarbons is illustrated by the following case history. At the site,  groundwater was
contaminated by gasoline  from a leaking underground storage tank with  initial

-------
concentrations exceeding 15 mg/L total petroleum hydrocarbons (TPH).  The geology of
the site consisted of interbedded highly fractured red shales and pebbly sandstones to a
depth of 25 to 30 ft, over a more competent fine-grained sandstone bedrock.  Depth to
water was 30 to 35 ft near the top of the competent bedrock.

The initial remedial system consisted of a centrally located recovery well.  The recovered
groundwater was treated by air stripping, amended with nutrients, and reinjected through
an infiltration gallery located in the original tank pit where the leak had occurred.
Supplemental  oxygen was added to the site through a series of six air sparging wells
equipped with porous stone diffusers connected to air compressors.

During approximately 2.5 years of operation, the  average level of contamination across
the site was reduced by approximately 78%, from 4.5 mg/L to 1.1 mg/L.  Groundwater
concentrations became static after 1.5 to 2 years,  indicating that the rate of
biodegradation was limited.  Dissolved oxygen (DO) levels remained lower than 1 to 2
mg/L in  the monitoring wells. The system was then modified to increase the rate of
oxygenation by adding  hydrogen peroxide at -500 mg/L continuously through the
infiltration gallery and  on a batch basis through individual monitoring wells. DO levels
increased across the site from  1 to 2 mg/L to 3 to 4 mg/L. The modified system was
operated for an additional 1.5  years.  The results were an immediate (3 to 4 months)
drop in contaminant levels from 1.1  to 0.5 mg/L, a 55% decrease.  Over another 6
months of operation, the contaminant level dropped to  -200 to 300 //g/L and stabilized.
DO levels began to rise after ~ 15 months of operation, indicating  that the  rate of
biodegradation was slowing. Because no further benefit was being attained and the cost
of continued operation was unacceptable, the remedial system was shut down and the
site was monitored on a quarterly basis.  As is frequently observed, contaminant levels
rebounded after system shutdown. After the rebound, a steady decline in contaminant
levels from the rebound level of ~900 jig/L was observed to  below the detection limit of
50 /ig/L after  only five years of monitoring. This continued drop in contaminant levels
after the  system was shutdown can be attributed to intrinsic bioremediation.

The proper use of intrinsic remediation requires monitoring and evaluation to determine
that existing environmental  conditions are sufficient for biodegradation to occur and that,
indeed, biodegradation is occurring.  At minimum there should be  documentation that
the compounds of interest are decreasing at a rate that is in excess of what would be
expected from normal dispersion and dilution. Biodegradation is also indicated by the
consumption of electron acceptors (oxygen, sulfate, iron, nitrate) or the presence of by
products  (reduced iron, sulfide, methane, carbon  dioxide, and bicarbonate).
Consumption of electron acceptors or the build up of by-products should  be evaluated by
comparing  their concentrations at points up gradient, within and down gradient of the
contaminated  zone. Additionally general indicators of conditions favorable to
degradation should be  measured.  These include pH, temperature,  redox potential, the
presence of a  viable microbial community, and the presence of low levels of nutrients.
Establishing the presence and efficacy of intrinsic remediation is based on the weight of
evidence.

-------
 SUMMARY AND CONCLUSION

 These four  levels of hydrocarbon bioremediation can be combined to give more cost
 effective treatment.  The combinations can be done either spatially or temporally.
 Spatially, as shown in Figure 1, aggressive and moderate intensity bioremediation are
 best used to treat high levels of contamination, including the source area and high levels
 of dissolved hydrocarbons.  Passive and intrinsic bioremediation are best used to treat
 dissolved plumes.  Low intensity bioremediation is effective for moderate to low levels of
 dissolved hydrocarbons. Intrinsic bioremediation is best used to treat low levels of
 dissolved hydrocarbons.

 Using a combination of these processes can provide cost effective treatment.  Directing
 the application of aggressive or moderate intensity bioremediation processes at the
 source area while focussing the application of low intensity and intrinsic processes on the
 dissolved plume provides a synergistic and cost effective approach. As  the aggressive
 processes reduce the carbon load in the source area, groundwater concentrations will
 drop as there is less  mass to leach into the groundwater.  As the groundwater
 concentrations  decrease at the head of the  plume, the efficacy of the low intensify or
 intrinsic processes will increase resulting in a more rapid attenuation of the groundwater
plume.  This concurrent application of processes gives the fastest remediation.

A second strategy in using the full range of bioremediation systems is to vary their use
temporally.  This is illustrated in Figure 2.  Aggressive remediation is used to address
high levels of available carbon. Available carbon  is the  "sum" of the volatile and
dissolved  fractions to which adsorbed and NAPL phases will eventually contribute.
Aggressive remediation is generally applied in the initial phase of remediation of a highly
contaminated site. Once the available carbon level has decreased or if the site  has
limited available carbon to start, moderate  intensity bioremediation
(bioventing/biosparging) can be employed.  As the source area is removed and  the
residual problem becomes primarily a low-level, dissolved problem, low intensity and
intrinsic bioremediation can be used.  This  sequential, planned decrease in  the level of
activity can  keep the cost of remediation at a minimum while still attaining low  clean up
goals.  The  "hidden  cost" is that a sequential remediation will generally take longer than
a concurrent approach.

The question to be continually asked in the bioremediation of hydrocarbon site  is what is
the comparative value of time and money in the remediation effort. The lower  the
intensity of  the bioremediation process the  lower the cost, but the longer the time it
needs to be  applied.  If time has limited value, i.e. there is no pressure to clean-up the
site within a short time frame, then low intensity and intrinsic processes, or sequential
remediation are the  most cost effective, provided that long-term monitoring and site
management costs do not exceed the difference in treatment costs. If time  is important,
i.e., there is  a need/desire to accomplish clean-up in a defined time period, than
aggressive bioremediation or concurrent remediation is warranted.

-------
In reviewing the application of these various levels of bioremediation to hydrocarbon
sites, there are two important conclusions that can be drawn.  First, all hydrocarbon
plumes come to equilibrium.  As levels of dissolved hydrocarbons decrease through
normal dilution and dispersion, the efficacy of intrinsic remediation will increase.  At
some distance away from the source area the hydrocarbons will be completely mitigated.
Give enough space, intrinsic remediation provides a sufficient and base level of
remediation to control hydrocarbon contamination.  Second, the size and duration of a
hydrocarbon plume can be mitigated by the application of more intensive bioremediation
processes.  The size of the plume is a function of what is leaching into it and what is
being removed from it.  As stated above, the plume attains an equilibrium size based on
the degree of intrinsic remediation. That equilibrium size can be decreased by reducing
the amount of material leaching into the plume, through aggressive or moderate intensity
bioremediation of the source area, or by increasing the amount of removal from ihe
dissolved plume by applying low intensity bioremediation.

REFERENCES

Armstrong, J. and R.D. Morris,  1995  "The Use of Biological Migration Barriers to
Prevent the Spread of Petroleum Hydrocarbons". Presented at Hazmat International
June  14-16, 1995, Chicago, Illinois.

Bennedsen, M.B., J.P. Scott, and J.D. Hartley.  1987. "Use of vapor extraciion systems
for in situ removal  of volatile  organic compounds from soil".  In: Proceedings of National
Conference on  Hazardous Waste and Hazardous Materials.  Washington D.C. pp. 92-95.

Bianchi-Mosquera, G.C, R.M. Allen-King, and D.D. Mackay. 1994. "Enhanced
degradation of dissolved benzene and toluene  using a solid oxygen-releasing compound".
Ground Water Monitor. Rev. pp. 120-128.

Borden, R.C., C.A. Gomez, and M.T. Becker.  1994.  "Natural Bioremediation of a
Gasoline Spill."  In R.E. Hinchee, B.C. Alleman, R.E. Hoeppel and R.N. Miller (Eds.),
Hydrocarbon Bioremediation. pp. 290-295. Lewis Publishers, Boca Raton, FL.

Brown, R.A,, R.D.  Norris, R.L Raymond, 1984.  "Oxygen Transport in Contaminated
Aquifers."  Presented at NWWA and API Conference "Petroleum Hydrocarbon and
Organic Chemicals in Groundwater: Prevention, Detection, and Restorations"  Houston,
TX.

Brown, R.A., and F. Jasiulewicz.  1992. "The Use of Aeration in Environmental Clean-
ups". Presented at Haztech International, Pittsburgh Waste Conference, Pittsburgh, PA.
May 14-16.

Dupont, R.R.,  W.J. Doucette, and R.E. Hinchee. 1991.  "Assessment of In-Situ
Bioremediation Potential and the Application  of Bioventing at a Fuel Contaminated
Site". In R.E.  Hinchee  and R.F. Olfenbuttel (Eds.), In-Situ Bioreclamation. Butterworth-
Heinemann, pp. 262-82, Stoneham, MA.

-------
 Ely, D.L., and D.A. Heffner.  1988. Process for in-situ biodegradation of hydrocarbon
 contaminated soil.  U.S. Patent Number 4,765,902.

 Gudemann, H., and D.Hiller.  1988.  In situ remediation of VOC contaminated soil and
 ground water by vapor extraction and ground water aeration.  In Proceedings of Petroleum
 Hydrocarbons and Organic Chemicals in Ground Water: Prevention, Detection and
 Restoration.   NWWA/API, pp. 147-164, Houston, TX.

 Hinchee, R.E., J.A. Kittel, and H.J. Reisinger (Eds.). 1995.  Applied Bioremediation of
 Petroleum Hydrocarbons.  Battelle Press, pp. 550, Columbus, OH.

 Hinchee, R.E., R.N. Miller, and P.C. Johnson (Eds.). 1995.  In Situ Aeration:  Air
 Sparging, Bioventing, and  Related Remediation Processes. BattelJe Press, pp. 634,
 Columbus, OH.

 Hinchee. R.E., B.C. Alleman, R.E. Hoeppel, and R.N. Miller (Eds.). 1994. Hydrocarbon
 Bioremediation.   Lewis Publishers, 496 pp., Ann Arbor, MI.

 Hinchee, R.E., and  R.F. Olfenbuttel (Eds.).  1991.  In-Situ Bioreclamation. Butterwbrth.
 605 pp., Ann Arbor, MI.

 Hoeppel, R.E., R.E. Hinchee, and M.R. Arthur.  1991.  Bioventing soils contaminated
 with petroleum hydrocarbons.  J. Industrial Microbiology.  8:141-146.

 Johnson, J.,  J. Odencrantz, and S. Koenigsberg, 1995 "Enhanced Intrinsic Bioremediation
 of a dissolved Hydrocarbon Plume Using an Oxygen Barrier" presented at the New
 Mexico Environmental Conference on Bioremediation, June 21-23, 1994, Santa Fe, New
 Mexico.

 Johnson, R.L., P.C.  Johnson, D.B. McWhorter, R.E. Hinchee, and I. Goodman.  1993.
An overview  of in  situ air sparging. Ground Water Monitoring and Remediation
 13(4):127-135.

 Kampbell, D.H., J.T. Wilson, and C.J. Griffin.  1992.  Bioventing of a gasoline spill at
 Traverse City, Michigan.  In:  Bioremediation of Hazardous Wastes. EPA/600/R-
 92/126. Office of Research and Development.

 Kao, C.M., and R.C. Borden.   1994.  Enhanced aerobic bioremediation of a gasoline
 contaminated aquifer by oxygen-releasing barriers.  In: Hydrocarbon bioremediation.
 Boca Raton, FL:  Lewis Publishers.

 Leeson, A.,  P. Kumar, R.E. Hinchee, D. Downey, CM. Vogel, G.D. Sayles, and R.N.
 Miller. 1995. Statistical Analyses of the U.S. Air Force Bioventing Initiative Results.  In  Situ
 Aeration:  Air Sparging, Bioventing, and Related Remediation Processes.  Battelle Press,
 Columbus, OH.  pp. 223-235

-------
Marlow Jr., H.J., H.R. Muniz, and S. Koenigsberg. 1995. 'The Use of Oxygen Releasing
Compound (ORC) for Groundwater Bioremediation in Alaska".  Presented at Emerging
Technologies in Hazardous Waste Management VII, Sept.  17-20,  1995. Atlanta, GA.

Miller, R.N., C.C. Vogel, and R.E. Hinchee.  1991. A Field-Scale Investigation of
Petroleum Hydrocarbon Degradation in the Vadose Zone Enhanced by Soil  Venting at
Tyndall AFB, Florida.  In R. E. Hinchee and R.F. Olfenbuttel (Eds.), In-Situ
Bioreclamation.  Butterworth-Heinemann, Stoneham, MA pp.283-302.

Miller, R.N. 1990.  A field scale investigation of enhanced petroleum hydrocarbon
biodegradation in the vadose zone combining soil venting as  an oxygen source with
moisture and nutrient additions.  Ph.D. Dissertation.  Utah State University. Logan,
Utah.

National Research Council.  1993. In situ bioremediation.  When does it work?
Washington, DC:  National Academy Press.

Piontek, K., T. Sail, S. de Albuquerque, and J. Cruze.  1994. "Demonstrating Intrinsic
Bioremediation of BTEX at National Gas Plant." Symposium on Intrinsk
Bioremediation of Groundwater, Denver, CO. August 1994,  pp. 179-180.

Raymond, R.L. 1976.  Beneficial Stimulation of Bacterial Activity in Groundwater
Containing Petroleum Hydrocarbons.  AICHE Symposium Series 73, pp 390-404.

Salanitro,  J. P.  1994.  "An Industry's Perspective on intrinsic Bioremediation." In Situ
Bioremediation. When Does it  Work?, pp.  104-109 National Academy Press,
Washington, D.C

Sayles, G.D., A. Leeson, R.E. Hinchee, CM. Vogel, R.C Brenner, and R.N. Miller.
1995. Cold Climate Bioventing with Soil Warming in Alaska.  In Situ Aeration: Air
Sparging, Bioventing, and Related Remediation Processes.  Battelle  Press, Columbus,
OH. pp. 297-306.

Texas Research Institute.  1980.  Laboratory Scale Gasoline Spill and Venting Experiment.
American Petroleum Institute.  Interim Report No. 7743-5:JST.

Wilson, J.T., and C.H. Ward.  1986.  Opportunities for bioremediation of aquifers
contaminated with petroleum hydrocarbons. J. Ind, Microbiol.  27:109-116

Wilson, J.T., D.H. Kampbell and J. Armstrong. 1994.  "Natural Bioreclamation  of
AJkylbenzenes  (BTEX) from a  Gasoline Spill in Methanogenk Groundwater."  In R.E.
Hinchee, B.C.  Alleman, R.E. Hoeppel and R.N. Miller (Eds.),  Hydrocarbon
Bioremediation. pp. 201-218. Lewis Publishers, Boca Raton, FL.

-------