United States
Environmental Protection
Agency
Solid Waste
Emergency Response
(5102W)
EPA 542-N-95-005
August 1995
Issue No. 21
xv EPA
The Applied Technologies Journal for Superfund Removals & Remedial Actions & RCRA Corrective Actions
DELAWARE OIL SPILL BIOREMEDIATION
FIELD STUDY
By Albert D. Venosa, EPA National Risk Management Research Laboratory
Last summer, an important field
study was undertaken on the shoreline
at Fowler Beach in the Delaware Bay.
As part of a carefully designed study,
540 gallons of light crude oil were in-
tentionally released onto fifteen repli-
cate plots to evaluate bioremediation.
The research design overcame the flaws
of previous studies reported in the lit-
erature that based conclusions on com-
parisons between one large nutrient-
treated plot and one large control plot.
The problem with those experiments is
that no replicate plots were established
to provide a basis for estimating experi-
mental error.
The objectives of the Delaware study
were to obtain highly credible statistical
evidence (1) to determine if bio-
remediation with inorganic mineral
nutrients and/or microbial inoculation
enhanced the removal of crude oil con-
taminating mixed sand and gravel
beaches; and (2) to compute the rate at
which such enhancement takes place.
Biodegradative loss of oil from the
plots was tracked by GC/MS analysis
of oil analytes normalized to hopane, a
non-biodegrable compound present in
all crude oils. A randomized complete
block design was used to assess treat-
ment effects. Three treatments were
evaluated: a no nutrient control; water
soluble nutrients (biostimulation); and
water soluble nutrients supplemented
with a natural microbial inoculum
from the site (bioaugmentation). An
REMEDIATION
MARKETPLACE
See the special insert in this issue of
TECH TRENDS for information on
the Mid-Atlantic Remediation Mar-
ketplace, sponsored by the EPA's
Technology Innovation Office and
Region 3 and the Pennsylvania De-
partment of Environmental Re-
sources.
unoiled and untreated plot served as a
background control for microbiologi-
cal characterization and baseline bioas-
says. Five replicates (blocks) of each of
four plots (20 plots in all) were set up
in random fashion so that whatever in-
ferences could be ascertained from the
data would be applicable to the entire
beach, not just die test plots.
The experimental design provided
the best chance to detect if significant
treatment effects existed. Statistically
significant differences between treated
and untreated plots were observed at
weeks 2, 4, 8 and 14 for the alkanes
and weeks 8 and 14 for the polycyclic
aromatic hydrocarbons (PAHs). First
order rate constants for treated plots
were significantly different from those
of the untreated plots. At no time
were any significant differences ob-
served between plots treated with nu-
trients alone and plots treated with nu-
trients and the indigenous inoculum.
Notable from the data is the fact that
the hydrocarbon biodegradation oc-
curred to a significant extent in the un-
treated plots. This was attributed to
the high levels of background nitrogen
measured on the beach from Delaware
Bay (upwards of 1 milligram (me) ni-
trate-nitrogen per liter intersritiaTpore
water continuously over the course of
the 14-week study). Had the back-
ground nutrient levels been closer to
those typical of Prince William Sound
beaches (i.e., a thousand-fold less), the
slopes of the control plots would likely
have been much flatter, thus giving rise
to highly significant differences be-
tween treated and control plots for
both alkanes and PAHs, perhaps as
early as two weeks after the release.
The latter conclusion is speculative,
since such low background nutrient
levels were not encountered in Dela-
ware. However, the goals of the
project were met: bioremediation via
nutrient enrichment was demonstrated
Oil Spills
Bio-
remediation
Ocean,
Estuary,
Beaches
unequivocally with statistically credible
data; computed biodegradation rates
were high, with an alkane half-life of
approximately two weeks and a PAH
half-life of approximately four weeks;
and bioaugmentation even with indig-
enous organisms does not stimulate
further declines in hydrocarbons be-
yond simple nutrient addition.
Based on the fact that nutrient levels
in the area were high enough to sup-
port significant biostimulative hydro-
carbon decay, the recommendation
would be not to encourage active
bioremediation activities if an oil spill
were to occur along the Delaware Bay
shoreline. However, for coastlines hav-
ing low natural input levels of nutri-
ents, bioremediation should definitely
be considered a major cleanup option.
The study was conducted under a
Clean Water Act, section 311, permit.
Section 311 allows the Administrator
to approve the intentional discharge of
less than 1,000 gallons of oil for re-
search programs. In October 1994,
the Administrator received an applica-
tion from the State of Delaware to in-
tentionally discharge 540 gallons of
crude oil onto a private beach for a
bioremediation study. The proposal
was strongly supported by EPA s Of-
fice of Research and Development and
Region 3, the U.S. Coast Guard, the
U.S. National Oceanographic and At-
mospheric Agency, the U.S. Depart-
ment of Interior and several Delaware
environmental groups. After an intra-
and interagency review, EPA granted
the permit in June 1994.
For more information on the study, call
Albert D. Venosa at EPA's National Risk
Management Research Laboratory at 513-
569-7668. A report will be available in the
Fall; to get on the mailing send a FAX. re-
quest to Albert Venosa at 513-569-7 105.
Recy clad/Recy clab le
Printed with Soy/Canola Ink on paper
that contains at least 50% recycled fiber
-------
United States
Environmental Protection
Agency
Solid Waste
Emergency Response
(5102W)
EPA 542-N-95-005
August 1995
Issue No. 21
vvERA
The Applied Technologies Journal for Superfund Removals & Remedial Actions & RCRA Corrective Actions
DELAWARE OIL SPILL BIOREMEDIATION
FIELD STUDY
By Albert D. Venosa, EPA National Risk Management Research Laboratory
Last summer, an important field
study was undertaken on the shoreline
at Fowler Beach in the Delaware Bay.
As part of a carefully designed study,
540 gallons of light crude oil were in-
tentionally released onto fifteen repli-
cate plots to evaluate bioremediation.
The research design overcame the flaws
of previous studies reported in the lit-
erature that based conclusions on com-
parisons between one large nutrient-
treated plot and one large control plot.
The problem with those experiments is
that no replicate plots were established
to provide a basis for estimating experi-
mental error.
The objectives of the Delaware study
were to obtain highly credible statistical
evidence (1) to determine if bio-
remediation with inorganic mineral
nutrients and/or microbial inoculation
enhanced the removal of crude oil con-
taminating mixed sand and gravel
beaches; and (2) to compute the rate at
which such enhancement takes place.
Biodegradative loss of oil from the
plots was tracked by GC/MS analysis
of oil analytes normalized to hopane, a
non-biodegrable compound present in
all crude oils. A randomized complete
block design was used to assess treat-
ment effects. Three treatments were
evaluated: a no nutrient control; water
soluble nutrients (biostimulation); and
water soluble nutrients supplemented
with a natural microbial inoculum
from the site (bioaugmentation). An
REMEDIATION
MARKETPLACE
See the special insert in this issue of
TECH TRENDS for information on
the Mid-Atlantic Remediation Mar-
ketplace* sponsored, by the EPA's
Technology Innovation Office and
Region $ and the Pennsylvania De-
partment of Environmental Re-
sources.
unoiled and untreated plot served as a
background control for microbiologi-
cal characterization and baseline bioas-
says. Five replicates (blocks) of each of
four plots (20 plots in all) were set up
in random fashion so that whatever in-
ferences could be ascertained from the
data would be applicable to the entire
beach, not just the test plots.
The experimental design provided
the best chance to detect if significant
treatment effects existed. Statistically
significant differences between treated
and untreated plots were observed at
weeks 2, 4, 8 and 14 for the alkanes
and weeks 8 and 14 for the polycyclic
aromatic hydrocarbons (PAHs). First
order rate constants for treated plots
were significantly different from those
of the untreated plots. At no time
were any significant differences ob-
served between plots treated with nu-
trients alone and plots treated with nu-
trients and the indigenous inoculum.
Notable from the data is the fact that
the hydrocarbon biodegradation oc-
curred to a significant extent in the un-
treated plots. This was attributed to
the high levels of background nitrogen
measured on the beach from Delaware
Bay (upwards of 1 milligram (mg) ni-
trate-nitrogen per liter interstitialpore
water continuously over the course of
the 14-week study). Had the back-
ground nutrient levels been closer to
those typical of Prince William Sound
beaches (i.e., a thousand-fold less), the
slopes of the control plots would likely
have been much flatter, thus giving rise
to highly significant differences be-
tween treated and control plots for
both alkanes and PAHs, perhaps as
early as two weeks after the release.
The latter conclusion is speculative,
since such low background nutrient
levels were not encountered in Dela-
ware. However, the goals of the
project were met: bioremediation via
nutrient enrichment was demonstrated
Oil Spills
Bio-
remediation
Ocean,
Estuary,
Beaches
unequivocally with statistically credible
data; computed biodegradation rates
were high, with an alkane half-life of
approximately two weeks and a PAH
half-life of approximately four weeks;
and bioaugmentation even with indig-
enous organisms does not stimulate
further declines in hydrocarbons be-
yond simple nutrient addition.
Based on the fact that nutrient levels
in the area were high enough to sup-
port significant biostimulative hydro-
carbon decay, the recommendation
would be not to encourage active
bioremediation activities if an oil spill
were to occur along the Delaware Bay
shoreline. However, for coastlines hav-
ing low natural input levels of nutri-
ents, bioremediation should definitely
be considered a major cleanup option.
The study was conducted under a
Clean Water Act, section 311, permit.
Section 311 allows the Administrator
to approve the intentional discharge of
less than 1,000 gallons of oil for re-
search programs. In October 1994,
the Administrator received an applica-
tion from the State of Delaware to in-
tentionally discharge 540 gallons of
crude oil onto a private beach for a
bioremediation study. The proposal
was strongly supported by EPA s Of-
fice of Research and Development and
Region 3, the U.S. Coast Guard, the
U.S. National Oceanographic and At-
mospheric Agency, the U.S. Depart-
ment of Interior and several Delaware
environmental groups. After an intra-
and interagency review, EPA granted
the permit in June 1994.
For more information on the study, call
Albert D. Venosa at EPA's National Risk
Management Research Laboratory at 513-
569-7668, A report mill be available in the
Fall; to get on the mailing, send a FAX re-
quest to Albert Venosa at 513-569-7105.
Racy clad/Recy clab le
Printed with Soy/Canola Ink on paper
that contains at least 50% recycled fiber
-------
SITE Subjects
CONE PENETROMETER MOUNTED SENSORS
By Lary Jack, EPA National Exposure Research Laboratory, Las Vegas
Fiber optics.Laser
In August 1994, the EPA
Superfund Innovative Technology
Evaluation (SITE) Program com-
pleted a field demonstration of cone
penetrometer-mounted chemical'
sensors. Two of the technologies
evaluated were: (1) the Rapid Opti-
cal Screening Tool (ROST®) mar-
keted by Loral Corporation; and (2)
the Site Characterization and Analy-
sis Penetrometer System (SCAPS)
Laser Induced Fluorescence (LIF)
developed by the Army, Navy and
Air Force. These technologies were
designed to detect fluorescing com-
pounds such as petroleum wastes
and coal tars with rapid sampling
and real-time, relatively low cost
analysis of the physical and chemical
characteristics of subsurface soil.
The inherent advantages of such
cone penetrometer sensor systems
are the high rate of penetration
(speed) and the low level of waste
generation. Of particular note, is
that much of the data is acquired
without stoppage and plotted on
continuous logs as the data is ob-
tained.
The SITE demonstration took
place at three sites within EPA Re-
gion 7, which were selected because
of their varying concentrations of
wastes (coal tar and petroleum fuels)
and their ranges in soil textures.
Both technologies are attachments
designed to fit a standard cone pen-
etrometer (CP) rig which simulta-
neously provides a continuous log of
subsurface materials. Two fiber op-
tic cables run from the sensor up
through the center of the penetrom-
eter rods along with the wires com-
ing from the load cells in the CP.
One fiber optic transmits laser in-
duced, monochromatic light from
the truck down to the sensor where
it passes through a sapphire window
on the side of the tool. Since the
cone is pushing its way into the
earth, the window is in direct con-
tact with the soil; and, the light
passes through the window into the
soil. The light energy excites hydro-
carbon compounds and causes them
to fluoresce, with the resultant energy
passing up to the second fiber optic
cable to an analyzer in the truck.
This produces a log of the contami-
nation while the signal from the stan-
dard CP head produces (after transla-
tion by software) a similar continuous
soil classification log.
The SITE demonstration evaluated
the effectiveness of these sensor sys-
tems in sampling and analyzing the
physical and chemical characteristics
of each waste site's subsurface soils,
by comparing each technology's re-
sults to the results obtained using
conventional reference methods. The
demonstration found that the SCAPS
and ROST® technologies produced
screening level data for the contami-
nation while the standard CP pro-
duced screening level soil classifica-
tions. Specifically the qualitative as-
sessment showed that the strati-
graphic and the chemical cross sec-
tions of the SCAPS and ROST8, and
associated CP sensors, were compa-
rable to the reference methods in
their ability to map subsurface con-
taminant plumes at petroleum fuel
and coal tar contamination sites. In
most cases, a consistent relationship
between the fluorescence data and the
reference method data was identified.
Increases in contaminant concentra-
tions were generally followed by in-
creases in measured fluorescence.
Data from both technologies identi-
fied similar zones of subsurface petro-
leum and coal tar contamination at
each of the three demonstration sites
relative to the reference methods.
Both technologies produced continu-
ous profiles, while the reference
methods took only a few selective
samples targeting boundaries and
zones of contamination. For the
SITE demonstration, both technolo-
gies could produce relatively continu-
ous data on petroleum or coal tar
contaminant distribution over a 30-
foot depth in approximately 1.5
hours. (The use of these sensors is
restricted to the maximum push
depth of the cone penetrometer
truck. This depth can be as much as
300 feet, or in the case of the demon-
stration, 30 to 70 feet.)
Based on the SITE demonstration,
these technologies produce screening
level data. A powerful aspect of both
technologies is that they can be ad-
vanced with a standard cone pen-
etrometer to provide continuous de-
scriptions of the subsurface soil con-
currently with the chemical data.
The combination of chemical and
physical sensors allows investigation
and remediation decisions to be
made more efficiently onsite and will
reduce the number of samples that
need to be submitted for costly con-
firmatory analyses. The added ben-
efit of these sensors functioning with-
out requiring physical sampling al-
lows them to produce data in subsur-
face environments that restrict con-
ventional sampling.
The ROST ® technology is cur-
rently available as a service from the
developer. The technology is de-
signed to be operated by trained
technicians. The SCAPS technology
is designed to be operated by trained
technicians from the Army Environ-
mental Center, Army Corps of Engi-
neers, Navy, Air Force and the Wa-
terways Experiment Section. The
SCAPS technology is not currently
available for use by private citizens or
corporations, although it is available
to State and Federal agencies. The
target contaminants are primarily
polycyclic aromatic hydrocarbons;
and, most often, this technology is
applied at petroleum fuel release
sites.
For more information, call Lary Jack
at EPA's National Exposure Research
Laboratory at 702-798-2373. Innova-
tive Technology Evaluation Reports on
ROST6 and SCAPS, as well as the other
cone penetrometer systems evaluated in
the SITE demonstration should be avail-
able by November 1995.
-------
November 29-30,1995
The Warwick Hotel
Philadelphia, Pennsylvania
sponsored by:
U.S. EPA, Technology Innovation Office
co-sponsors:
U.S. EPA, Region III
PA Dept. of Environmental Protection
Southern States Energy Board
The U.S. Environmental Protection Agency Technology Innovation Office
is sponsoring this conference to provide specific information on current
and projected business opportunities in the Mid-Atlantic Region for
innovative environmental technologies at the State and Federal levels and
in the private sector.
The conference will enable attendees to:
/ Identify opportunities in the remediation marketplace
/ Gather information on funding and available assistance
/ Locate points of contact and establish new business relationships
The agenda includes:
State and Federal markets
Public/private partnership opportunities
Prime/subcontractor partnering
Funding of technology development
Overview of international markets
Small business guidance
Case studies of industry perspectives
Will B£>
Our ^
ifu!
-Hoi! -Hours!
Registration information will be sent to you at
the end of September. If you do not receive an
announcement brochure at that time, please fax
your name, address, fax and phone numbers to
SAIC, (215) 628-8916 to receive one.
For Additional Information
Contact:
SAIC
(800) 783-3870
-------
SITE Subjects
GASIFICATION TREATMENT FOR SOILS
By Marta K. Richards, EPA National Risk Management Research Laboratory
Texaco, Inc. has developed waste gas-
ification as an innovative extension of
their conventional fuels gasification
technology. The Texaco Gasification
Process (TCP) was evaluated by EPA's
SITE (Superfund Innovative Technol-
ogy Evaluation) Program to determine
the applicability of the TCP to treat-
ment of a hazardous waste-contami-
nated soil. Since the TGP is operating
af temperatures above the melring
point of the inorganic materials in the
feed stream, the solid residuals form a
glassy slag, while the TGP converts
carbonaceous organic materials into a
gas mixture, called syngas, consisting
primarily of hydrogen and carbon
monoxide. This is accomplished by
reaction with a limited amount of
oxygen (partial oxidation) in a refrac-
tory-lined gasifier at temperatures in
excess of 2,200 F and at pressures
greater than 250 psig. According to
Texaco, these temperatures and pres-
sures are sufficient to destroy hydro-
carbons and organics in the feed and
prevent the formation of undesirable
organic by-products associated with
other coal/fossil fuel conversion pro-
cesses. The TGP produces syngas as a
chemical intermediate product which
can be used in the production of hy-
drocarbons, ammonia, methanol and
PAHs.VOCs
Gasification
Soils
other chemicals. Alternatively, the
syngas can be combusted directly in a
gas turbine to produce electricity.
The process works as follows. Waste
feed, along with coal, oil and/or coke is
ground and mixed with water in a
high-solids' concentration slurry (gen-
erally 55 to 70% by weight). The wa-
ter serves as a reactant, a temperature
moderator and a transport medium.
Liquid and gas fuels, which can be fed
directly to the gasifier, need no pre-
treatment but may require water or
steam moderators. For the SITE dem-
onstration at the Texaco Montebello
Research Laboratory in South El
Monte, California, the slurry feed con-
sisted of coal, water, waste soil from
the Purity Oil Sales Superfund Site in
Fresno, California, and clean soil. The
composite slurry was spiked with
heavy metals (lead ana barium) to en-
sure that the feed contained sufficient
levels of contamination to fail the Tox-
icity Characteristic Leaching Proce-
dure (TCLP) test. The addition of ap-
proximately 6.42 Ibs. per hour of chlo-
robenzene, a volatile organic, was fed
into the gasifier with the slurry to fa-
cilitate the calculation of the Destruc-
tion and Removal Efficiency (DRE).
In the SITE demonstration, slurry
was fed into the gasifier, along with
oxygen, through an injector nozzle.
The chlorinated species in the feed
formed hydrogen chloride in the raw
syngas; the hydrogen chloride dis-
solved into the scrubber water before
being purged from the process. The
sulfur in the feed formed hydrogen sul-
fide or carbonyl sulfide in the product
gas. Commercial acid gas and sulfur
removal processes scrubbed these sul-
fides from the product gas. The re-
sidual ash was recovered as a glassy
slag.
The DRE for chlorobenzene was
greater than 99.99%. The average
composition of the combustible dry
syngas from the TGP was 37% hydro-
gen, 36% carbon monoxide and
21% carbon dioxide. No organic
compound, other than methane
(55 parts per million) was above 0.1
ppm. On the average, the primary
TGP solid product, coarse slag, com-
plied withthe TCLP regulatory re-
quirement for lead [5 milligrams per li-
ter (mg/L)] and the TCLP and Cali-
fornia Waste Extraction Test (WET)
regulatory requirements for barium
(100 mg/L). The coarse slag did not
meet the WET standard for lead (5 mg/
L). Volatile heavy metals, such as lead,
tend to partition and concentrate in
(continued on page 4)
NORTH OF THE BORDER
VACUUM PYROUYSIS FOR ORGANICS IN SOILS
From Environment Canada
Environment Canada's Development
and Demonstration of Site Remedia-
tion (DESRT) Program supported the
evaluation of the Pyrovac vacuum py-
rolysis process developed by Pyrovac
International, Inc. The evaluation
was carried out by the Laval Univer-
sity in collaboration with Pyrovac,
with support from DESRT and the
Ministere de 1'Environnement et de la
Faune du Quebec at the Le Vidangeur
de Montreal Inc. site in Mascouche,
Que'bec. The site, previously a sand-
quarry, was used from 1969 to 1974
ror the dumping of solid residues, as
well as for the incineration of petroleum
sludges. Fissured basins containing liq-
uids and hydrocarbon sludges were
abandoned. Today, hydrocarbon tar
slabs can be found in several places on
the site whose soil and water have been
highly contaminated with hydrocar-
bons, polycyclic aromatic hydrocarbons
and polychlorinated biphenyls (PCBs).
The samples collected at the Mascouche
site were chosen so as to represent a va-
riety of types of soils (sand, clay, tar
slabs) and a variety of contamination
concentrations. The objectives of the
treatment were to reduce the level of or-
Hydrocarbons/^
PAHs, PCBs
Vacuum
pyrolysis
Soils
ganic contamination so as to respect
the Quebec criteria for soil contamina-
tion.
Prior to pilot-scale demonstration,
laboratory scale batch runs were car-
ried out in a vacuum pyrolysis reactor
and confirmed the potential of the
vacuum pyrolysis concept. The evalu-
ation continued with the pilot unit.
Petroleum hydrocarbon reductions in
milligrams per kilogram were: Clay A,
from 290 to 5.0; Clay B, from 150
(continued on page 4)
------- |