United States
Environmental Protection
Agency
Solid Waste and
Emergency Response
(5102G)
EPA 542-N-99-004
June 1999
Issue No. 32
DNAPL Flushing
with Alcohol
Cost-Effective
Sampling of Ground
Water Monitoring
Wells
In Situ Chemical
Oxidation through
Recirculation
New Dry-Cleaner
Site Remediation
Group Convened
Pg. 1
Pg.2
Pg-3
Pg.4
This issue features new
techniques for ground water
monitoring and active
remedies for ground water
contaminated with organics.
by Michael Annable, Ph.D..,
University of Florida, and
Randall Sillan, Ph.D., LFR
Levine-Fricke
University of Florida researchers are ;
evaluating new in situ co-solvent flushing
methods to remediate ground water i
contaminated with dense non-aqueous
phase liquids (DNAPL) commonly •
resulting from dry cleaning activities. At
an abandoned site known as Sage's Dry
Cleaners in Jacksonville, FL, a pilot-scale
demonstration involving the injection of
water and alcohol was conducted in 1998
to remove perchloroethylene (PCE) from
ground water. More than 60% of the JPCE
existing in the subsurface was extracted.
Researchers estimate that a pump and treat
system would need to operate for 10 years
to extract an equivalent volume of PGE.
This project was conducted through the
State of Florida's Dry Cleaning Program
and Center for Solid and Hazardous Waste
Management, with assistance from the
U.S.EPA.
The source area appeared to be oblong-
shaped, approximately 7.3 meters by 2.7
meters. PCE had been released into the
subsurface through a sump, and extended
7.9-9.8 meters below ground surface.
Tests indicated a NAPL saturation range
of 8-35%, and PCE concentrations in soil
as high as 45,000 mg/Kg. Based on the
results of ground water flow modeling, a
total of three injection wells and six
recovery wells were installed in a pattern
designed to contain the injected fluids
hydraulically and provide adequate ;
coverage of the source area. Various tracer
tests indicated that approximately 69 liters
of PCE existed within the aquifer.
The co-solvent solution was delivered to
specified depths using packers. Approxi-
mately 34,000 liters of a 95% ethanol
solution was delivered to the subsurface.
(Large quantities of flushing agents were
required due to the uneven spatial
distribution of the contaminants.) The
combined flow rate of water and co-
solvent initially was held constant at 4.2
liters per minute for two of the injection
wells and at 6.8 liters per minute for the
third injection well. Flow rates were
adjusted after two days to improve co-
solvent contact with the NAPL. To focus
more of the co-solvent in the lower portion
of the DNAPL-affected area, packers were
used in the injection wells to separate a
lower co-solvent zone from an upper water
flood zone. During the experiment, the
packers were raised to increase the size of
the co-solvent flood. After 3.5 days, the
co-solvent flood zone was reduced by
lowering the packers and displacing the
co-solvent from the formation with water.
Following eight days of flooding, 92% of
the injected co-solvent was recovered. In
order to minimize cost associated with the
disposal of waste containing high levels of
PCE, a macro-porous polymer extraction
(MPPE) system was used to separate PCE
from the extracted co-solvent solution.
During treatment, the waste stream was
passed through a column containing
MPPE material into which the PCE
preferentially partitioned. After loading
with PCE, the columns were regenerated
with low-pressure steam stripping, and the
vaporized steam was condensed into
free-phase PCE for disposal. Approx-
imately 42 of the estimated 69 liters of
PCE were separated from the waste stream.
Recycled/Recyclable
i Printed with Soy/Canola Ink on paper that
-------
The estimated cost of this demonstration
was $440,000, including research, design,
construction, and operation. Researchers
predict that re-use of ethanol could reduce
the initial amount of ethanol needed in full-
scale implementation of this technology by
as much as 50%. It is estimated that full-
scale use of this technique at the Sage's Dry
Cleaning site could be completed in one
month at a cost of $110,000. In contrast,
pump and treat methods could require up to
40 years and $2.2 million. Based on the
success of this project, the Florida Dry
Cleaning Program is considering this
technology for other PCE-contarninated
sites. For more information, contact Dr.
Michael Annable (University of Florida) at
352-392-3294 or E-mail
manna@eng.ufl.edu.
by Maureen Ridley and Don
MacQueen, Lawrence Livermore
National Laboratory
Lawrence Livermore National Laboratory
(LLNL) has begun implementing a new
methodology for estimating the lowest-
frequency ground water sampling schedule
that will provide sufficient information for
regulatory and remedial decision-making.
Known as Cost-Effective Sampling (CES),
this approach has resulted in a 40 percent
reduction in the annual number of routine
ground water samples taken at LLNL's •
Main Site and Site 300 in Livermore, CA.
This reduction has saved LLNL $390,000
annually in sampling, data management,
and analysis costs.
The original method for determining
sampling frequencies at LLNL was based
on the proximity of a well to a contaminant
plume. This approach caused the majority
of the wells to be sampled quarterly,
including those that had shown no change
over an eight-year period, and did not
account for the slow rate of contaminant
migration occurring. In contrast, the CES
approach bases a frequency sampling on
concentration changes observed at a given
well. CES recommends that sampling
frequencies be based on quantitative
analysis of the trends, variability, and'
magnitude of contaminant statistics, which
then are interpreted through decision trees.
The underlying principle of CES is that
sampling should be determined primarily
by the rate of change in concentrations at a
given location. Accordingly, a higher :rate
of concentration change (whether upward
or downward) indicates the need for
frequent sampling, while minimal change
suggests a more relaxed sampling sched-
ule. A second rationale for more frequent
sampling is the degree of uncertainty
displayed hi the measured concentrations.
Low overall rates of change can be offset
by a higher degree of variability, requiring
that a more frequent schedule be main-
tained tq define
the likely degree
of contamination.
Conversely, a high
rate of change that
is highly predict-
able warrants a
lower sampling
frequency. Lastly,
the magnitude of
the measured
concentrations
affects the
interpretation that
is placed on rates
of change. For
example, a yearly
change of 50 parts
per billion is
interpreted
differently when
the median
concentration is 10
versus 1,000. The
significance of the
absolute concen-
trations also varies
by compound.
To be eligible for
CES application, a
sampling location
must have been
sampled previ-
ously on at least
six occasions (or through approximately 18
months of quarterly sampling), have good
site characterization/understanding, and
clean downgradient wells (guard wells).
CES implementation requires that newly
installed wells be sampled frequently to
build a history for analytical purposes. For
each sampling location, decision trees must
be used independently for each contaminant
in the target list, and the sampling schedule
should be equivalent to the most frequent
schedule estimated for any individual
contaminant. The overall flow of CES is
shown in Figure 1.
Use of CES at more than 700 LLNL ground
water monitoring wells has resulted in a
reduction of quarterly sampling frequency
from 418 to 216 events.. For more informa-
tion, contact Maureen Ridley (LLNL) at
925-422-3593 or E-mail ridleyl@llnl.gov.
Set
Adjuj
F
Nc
Evalus
: Figure 1. Overview of CES Steps
Sufficie
Yes
! ' 1
NO
r
^ .Consider First/Next
' Compound
' ^
r
Stepl
Frequency Based on Recent Trends
; i
r
Step 2
st Frequency Based on Overall Trends
i
r
1 Step 3
.educe1 Frequency Based on MCL
^
>
Yes
! 1
T
n pound?
r
ition Sheet & Recommended Schedule
1
Engineering/Scientific ^ If Guard Well, then
Review Quarterly
-------
The U.S. EPA will sponsor a conference at the Renaissance Dallas Hotel in Dallas,
TX, August 31-September 2, 1999, to address in situ abiotic options for
remediating contaminated ground water, The conference will communicate ;
information regarding the current design, performance, technical feasibility, "•
implementation impediments, and costs for abiotic applications. EPA's Office or
Research and Development/National Risk Management Research Laboratory and
Office of Solid Waste and Emergency Response/Technology Innovation Office are
hosting this conference jointly in an effort t§ promote less costly and more
effective site clean-up processes. :
Remedial project managers, on-scene coordinators, government decision-makers,
vendors, researchers, academia, and consultants involved with contaminated
ground water sites are invited to attend. Registration is available on the Internet
at www.epa.gov/ttbnrmrl or by calling 412-741-5462. No fees are required to :
attend the conference, but space is limited.
by Olivia West, Ph.D., Oak Ridge
National Laboratory
Researchers from Oak Ridge National
Laboratory (ORNL) recently completed
field testing of a new technology for
degrading pure-phase trichloroethylene
(TCE) in saturated aquifer sediments using
recirculation as a reagent delivery
technique. In Situ Chemical Oxidation
through Recirculation (ISCOR) involves the
flushing of oxidant solutions through a
contaminated aquifer by injection and
extraction in multiple horizontal and
vertical wells. The ISCOR approach offers
an advantage over conventional techniques
because there is more control over oxidant
and contaminant migration within the
treatment zone and between the injection
and extraction wells. This demonstration
also constituted the first full-scale
application of in situ chemical oxidation
with potassium permanganate (KMNO4)
using a horizontal well system to treat dense
non-aqueous phase liquids (DNAPL) in a
saturated, relatively permeable subsurface
medium.
The demonstration was conducted at the
U.S. Department of Energy's Portsmouth
Gaseous Diffusion Plant (PORTS) in
Piketon, OH. The test area was located
within the PORTS X-701B site, a region
surrounding an unlined holding pond that
was used from 1954 to 1988 for neutraliza-
tion and settling of metal-bearing waste
water, solvent contaminated solutions, • and
acidic waste water. The holding pond had
been drained, and the contaminated sludge
and underlying silt and clay had been
removed. High levels of TCE (200-1,000
mg/L) continued to
be detected in
monitoring wells
screened within the
underlying aquifer,
indicating that
DNAPL are serving
as continuous
sources of ground
water contamina-
tion in the area.
The confined
aquifer consists of
approximately five
feet of silty gravel
overlain by 30 feet
of silty clay and
underlain by shale.
A pair of 200-foot-long horizontal wells
and a network of piezometers were installed
within the plume area at the X-701B site.
The wells were oriented perpendicular to
the overall ground water flow direction, and
located approximately 35 feet below
ground surface and 90 feet apart from one
another. Ground water was pumped from
the upgradient horizontal well at a rate of
six gallons per minute. Crystalline KMNO4
was then added to the extracted groundwa-
ter to achieve oxidant concentrations
ranging from 2 to 4 percent. The oxidant
solution was injected into the
downgradient horizontal well, while
ground water was extracted continuously
from the upgradient well. The system
operated in a recirculation mode for three
weeks with periodic assessment of treat-
ment performance based on ground water
samples collected from the vertical piezom-
eters within the horizontal well flow-field.
Figure 2 provides an overview of the
ISCORprocess.
To ensure the ISCOR process was working
properly, the injection and extraction wells
were sampled daily during active injection
and extraction of the oxidant. Upon
completion of the oxidant recirculation,
post-treatment soil and ground water
sampling was conducted to compare pre-
and post-treatment TCE levels. Analytical
results indicated that wherever MnO4" was
detected, TCE concentrations were reduced
to non-detect levels. Soil and water
Figure 2. In Situ Chen
(
Upgradient
Horizontal Well
4J
o
S3
Extraction Well
ical Oxidation Through Recirculation
KMnO4 Feed System
" Pump
0+ O+ O+ /
^_^ »W>'
O+ O+ 0+ /"'
o+ o+ o+ f
o+ 0+ o+/
o + o •*• o +L *
O+ O+ O+ \ j.
o+ o* o+ \
O+1 o+ o+ V
90ft
PLAN VIEW
> Downgradient
"Horizontal Well
I
C
o
S3
O
f
O + Pre- and Post
Treatment Sampling
Locations
-------
Ground Water Currents is on the NET!
View or download it from:
http://www.epa.gov/tio
http://clu-in.org
Ground Water Currents
welcomes readers' comments and contribu-
tions, and new subscriptions. Address
correspondence to:
Ground Water Currents
8601 Georgia Avenue, Suite 500
Silver Spring, Maryland 20910
Fax: 301-589-8487
chemistry parameters, including residual
oxidant levels; also were measured to
determine the effects of potassium perman-
ganate on the geochemistry of the treated
aquifer, and negligible effects were found.
For more information on the demonstration,
contact Dr. Olivia West (ORNL) at 423-576-
0505 or E-mail qm5@ornl.gov.
1999, in Washington, DC, to plan goals and
future activities for the newly-formed State
Dry Cleaners Remediation Project. EPA's
Technology Innovation Office (TIO) is
sponsoring the project in an effort to help
promote the use of cost-effective, innova-
tive remediation technologies .and provide
resources for small dry-cleaner sites where
limited funding is available. During this
kick-off meeting, project subgroups were
formed to address related state program-
matic issues, technical concerns, and;
outreach methods.
HO will facilitate she project by continuing
to enhance communications among |
technology vendors, responsible parties,
regulators, and consulting engineers and to
act as an information broker through tools
such as CLU-IN, an on-line information
resource located on the Web at http://
www.clu-in.org. In addition, the National
Ground Water Association will provide
regional, training sessions for state govern-
ment representatives and potentially for
state project manager, hydrogeologists,
consultants, dry cleaners, and members of
the fabric-care industry. For more informa-
tion, contact Richard Steimle (EPA) at
703-603-7195 or E-mail
steimle.richard@epa.gov.
Representatives from 11 states with
existing or pending cleanup programs for
dry-cleaner sites convened April 13-14,
The U.S. EPA will sponsor the
conference innovative Clean-Up
Approaches: Investments in
Technology Development, Results, &
Outlook for the Future at the Indian
Lakes Resort in Bloomingdale, IL,
November 2-4, 1999. Stakeholders
in hazardous waste site remediation
projects, including EPA's partners
from other government agencies,
academia, and the private sector, will
have an opportunity to share the
latest information on technology
development, demonstration, and
commercialization. Participants also
will evaluate the success of past
efforts and discuss future research
and information needs. Workshops
will be available to provide
information on the SITE Program,
Brownfields Program, funding
sources, and electronic information
resources. Additional information
can be obtained on the Internet at
www.epa.gov/ttbnrmrl or by calling
412-741-5462.
United States
Environmental Protection
Agency
Solid Waste and
Emergency Response
(5102G)
EPA 542-N-99-004
June 1999
Issue No. 32
------- |