United States
Environmental Protection
Agency
Risk Reduction Engineering
Laboratory
Cincinnati, OH 45268
Research and Development
EPA/600/S2-91/023 July 1991
\yEPA Project Summary
Evaluation of Soil Washing
Technology: Results of Bench-
Scale Experiments on
Petroleum-Fuels Contaminated
Soils
Mary E. Loden
The U.S. Environmental Protection
Agency (EPA) through its Risk Reduc-
tion Engineering Laboratory's Release
Control Branch has undertaken re-
search and development efforts to ad-
dress the problem of leaking under-
ground storage tanks (USTs). Under
this effort, EPA is currently evaluating
soil washing technology for cleaning
soil contaminated by the release of pe-
troleum products leaking from under-
ground storage tanks.
The soil washing program evaluated
the effectiveness of soil washing tech-
nology to remove petroleum products
(unleaded gasoline, diesel/home heat-
ing fuel, and waste crankcase oil) from
an EPA-developed synthetic soil ma-
trix (SSM) and from actual site soils.
Operating parameters such as contact
time, washwater volume, rinsewater
volume, washwater temperature, and
effectiveness of additives were investi-
gated. Further work was conducted to
determine what effect, if any, additives
have when added to washwater. The
additives investigated were CitriKleen*
(a biodegradable degreasing agent) and
an organic surfactant. Actual soils from
UST sites in Ohio and New Jersey were
washed using the optimum parameters
derived for the SSM.
The results of the optimization tests
using SSM indicated that greater than
90% of petroleum products could be
removed from the SSM. In experiments
using actual site soils and the same
washing conditions, contaminant re-
moval was lower than it was for the
SSM experiments. Although the SSM
experiments achieved high removals,
only 55% of the washed soil mass was
recovered and a washwater containing
over 20% solids was produced. The
washwaters from the actual site soils
experiments had less suspended sol-
ids, but it also removed fewer contami-
nants from the SSM.
This Project Summary was developed
by EPA's Risk Reduction Engineering
Laboratory, Cincinnati, OH, to announce
key findings of the research project
that is fully documented in a separate
report of the same title (see Project
Report ordering information at back).
Introduction
Based on the Hazardous and Solid
Waste Amendments of 1984 and its Land
Ban Regulations, the EPA discourages
the past practice of excavating contami-
nated soils from around leaking USTs and
disposing of them in landfills. Although
EPA has encouraged the use of on-site
treatment technologies, problems have
plagued the development of on-site tech-
nologies to treat petroleum-contaminated
soils. Technical support is needed to de-
velop effective long-term corrective actions
at leaking UST sites, to design cleanup
program guidance, and to help implement
state programs.
The remedial options available for the
treatment of contaminated soils from UST
sites are broadly segregated into two main
categories: those removing the contami-
* Mention of trade names or commercial products does
not constitute endorsement or recommendation lor
use.
Printed on Recycled Paper
-------
nants without excavation (in situ tech-
niques) and those requiring excavation of
the soil and subsequent cleaning on site.
The first remedial option has not yet dem-
onstrated high efficiency removal of con-
taminants from the subsurface. These
techniques are plagued by the uncertainty
of soil contamination levels in the subsur-
face after treatment. Soil excavation fol-
lowed by extensive cleaning of the soil
will ensure a more complete removal of
contaminants than will in situ techniques.
On-site soil washing of excavated soils
is a viable alternative to in situ techniques
and has been shown to be effective for
cleaning hazardous-waste-contaminated
soils. The goal of this effort is to deter-
mine the feasibility of soil washing for
cleaning petroleum-contaminated soils.
Soil washing is a physical process in
which excavated soils are contacted with
a liquid medium, usually water. The two
principle cleaning mechanisms are dis-
solving the contaminants into the extrac-
tive agent and/or dispersing the contami-
nants into the extraction phase in the form
of particles (suspended or colloidal). Sepa-
rating the highly contaminated fine soil
particles (silts, clay, and colloidal) from
the bulk of the soil matrix can reduce the
volume of the bulk soil. As a result, a
significant fraction of the contaminated soil
is cleaned and can be put back into the
original excavation. Because the con-
taminants are more concentrated in the
fine soil fractions, their removal from the
bulk soil increases the overall effective-
ness. The spent wash waters and the
fine soil fractions need subsequent treat-
ment.
As part of this research program, a syn-
thetic soil matrix (SSM) containing a range
of petroleum products at varying concen-
trations was prepared and subjected to
bench-scale performance evaluations of
soil washing technology. Operating con-
ditions derived from tests using SSM were
used to evaluate s;oil washing technology
on actual samples from leaking LIST sites
in Ohio and New Jersey.
Before preparing the quantities of SSM
needed for the bench-scale tests, several
bench-scale experiments were performed
to develop a dose/response relationship
between the quantity of petroleum prod-
uct added to the SSM and the analysis
quantification. The petroleum products
evaluated during this study included
unleaded gasoline, diesel oil, and waste
crankcase oil. The SSM was then blended
with a specific quantity of petroleum prod-
uct to obtain a predetermined concentra-
tion level. The soils were analyzed for
total petroleum hydrocarbons (TPH) to
verify the concentration levels for diesel
and waste oil, and for benzene, toluene,
ethylbenzene, and xylenes (BTEX) to verify
the concentration levels for gasoline.
The bench-scale washing experiments
were designed to simulate the EPA-devel-
oped pilot-scale mobile soils washing sys-
tem (MSWS). Specifically, the bench-scale
experiments were designed to simulate
the drum-screen washer that separates
the >2-mm soil fraction (coarse material)
from the <2-mm soil fraction (fines) by
use of a rotary drum screen. A high-
pressure water knife operates at the head
of the system to break up soil lumps and
strip the contaminants off the soil par-
ticles.
Synthetic Soil Matrix
Characterization
The basic formula for the SSM was
determined by others from an extensive
review of Superfund sites and a review of
the composition of eastern U.S. soils. The
SSM was a mixture of clay, silt, sand, top
soil, and gravel, prepared by others, in
two 15,000-lb batches.
The existing soil characteristics were
reviewed, and additional tests further de-
lineated the physical and chemical prop-
erties of the SSM. The tests included
particle size distribution, moisture reten-
tion, Atterberg limits, cation exchange ca-
pacity, base saturation, organic matter,
chemical constituents, and mineralogy.
Quantification and assessment of these
specific properties will assist the technical
community to understand the differences
that may be observed between the per-
formance of soil washing technology on
the SSM and on actual UST site soils.
The test results indicated that the SSM
is composed of 60% sand, 19% silt, and
21% clay as determined by particle size
distribution analysis. This composition of
the SSM would be classified (USDA) as a
sandy clay loam texture.
The moisture content of the SSM ranged
from 33.1% at saturation (0 bar) to 8.7%
at the permanent wilting point (15 bars).
The moisture content at field capacity (0.1
bar) was 21.0%. The moisture-retention
curve developed from the moisture con-
tent data was indicative of a finer-textured
soil. The moisture content data can be
used to evaluate moisture and chemical
characteristics of the SSM. For example,
the amount of soil water that can be ex-
tracted from the SSM under typical envi-
ronmental conditions (0 to 15 bars) will be
24.4%. The remaining soil water is con-
sidered "unavailable" and can only be re-
moved by artificially induced vacuums or
pressures.
Bench Scale Soil Washing
Tests
The experiments involved washing the
SSM spiked with gasoline, diesel fuel, or
waste crankcase oil under several operat-
ing conditions to evaluate the sensitivity
of various parameters affecting soil wash-
ing efficiency.
Approximately 1400 g of soil contacted
van/ing amounts of washwater. The con-
tact time was varied as was the rinsewater
volume. The soil and washwater were
shaken in a 2-gal jar in a shaker table
operating with a stroke and frequency of
1.6 in. and 4 Hz, respectively. The soils
were rinsed in a Gilson Wet-Vac Model
WV-1, which both rinsed the soils and
separated the particles into three fractions
with the use of No. 10, No. 60, and No.
140 sieve trays. The process of washing
and rinsing yielded five distinct fractions:
the soils on the three sieve trays, a
washwater, and a rinsewater. All frac-
tions were measured for mass (or vol-
ume) as well as for contaminant concen-
tration. A measure of total BTEX (ben-
zene, toluene, ethylbenzene, and o-, m-,
and p-xylenes) was used on gasoline-
spiked soils, and total petroleum hydro-
carbons (TPH) was used on diesel-spiked
soils.
Preliminary screening tests on soils
spiked with diesel and gasoline determined
the optimum conditions for contact time,
washwater volume, rinsewater volume, and
washwater temperature. The full report
lists the parameters tested and assesses
how the parameter affected removal of
TPH or BTEX from the soil and affected
particle separation during the washing pro-
cess. Most parameters had minimal ef-
fect on the removals and particle separa-
tion for the ranges tested. Increased con-
tact time did somewhat improve the con-
taminant removal and particle separation.
The addition of CitriKleen did not improve
contaminant removal, and, in fact, actu-
ally decreased the removal. Increased
temperature did not improve removals for
soils washed with plain water but did im-
prove removals somewhat for washwaters
containing CitriKleen or surfactant.
The experiments indicated that the op-
timal washing conditions for SSM spiked
with diesel or gasoline are: 20 to 30 min
contact time, 1:1 soil to washwater mass
ratio, 3:1 rinsewater to washwater volume
ratio, and ambient temperature for the
washwater. These conditions resulted in
a 90+% removal of TPH and BTEX in the
No. 10 and No. 60 sieve fractions. Note
that these conditions represent the most
cost-effective operating conditions for
bench-scale treatment of SSM using soil-
-------
washing technology. Operating conditions
for each site soil may vary and should be
determined on a case-by-case basis.
Experiments were then conducted on
site soils from four UST sites in Ohio and
New Jersey. The results indicated that
the removals in the 10- and 60-sieve frac-
tion were significantly lower than the re-
movals achieved for SSM. The site soils
had a significantly different particle size
distribution than did the SSM. The bench-
scale soil washing apparatus recovered
only 55% of the SSM mass after washing,
and 45% was washed off into the
washwater and rinsewater. Since the site
soils contained lower amounts of fines
(7% to 43% by weight for the four sites
tested), greater amounts of soil were re-
covered from the washing process, which
resulted in lower suspended solids in the
washwaters.
Summary and Conclusions
The soil washing experiments involved
washing the SSM spiked with gasoline,
diesel fuel, or waste crankcase oil under
several operating conditions to obtain sen-
sitivity analyses on various parameters af-
fecting soil washing efficiency including:
contact time with washwater, washwater
volume, washwater temperature, and
chemical additives in the washwater such
as a surfactant and a degreasing agent
(CitriKleen). The SSM experiments yielded
highly reproducible results. Since the soils
were prepared in the SSM blending facil-
ity, the soil characteristics and contami-
nant content were homogeneous through-
out the matrix. Experiments were also
conducted using actual site soils, but due
to the heterogeneity of these soils, the
reproducibility of these experiments was
much lower than for the SSM.
The results of the optimization tests us-
ing SSM indicated that greater than 90%
of petroleum products could be removed
from the SSM. As washing conditions
were varied, however, no significant
change in the removals could be detected
for most parameters tested. The contact
time in the wash cycle possibly affected
the percent removal because of improved
mechanical separation of the fines from
the larger particles with more shaking time
in the wash step. Increased volumes of
washwater and rinsewater did not signifi-
cantly improve contaminant removal. The
use of additives such as CitriKleen and
surfactant did not improve contaminant
removal. Increased amounts of CitriKleen
in the washwater decreased contaminant
removal; this may be attributed to en-
hanced adsorption of contaminants to the
soil because of the presence of CitriKleen
adsorbed onto the matrix. Experiments
using surfactant resulted in considerable
foaming; this would lead to operational
problems in pilot- or full-scale operation.
In experiments using actual site soils
and the same washing conditions, con-
taminant removal was lower than it was
for the SSM experiments. The washwaters
for the actual site soils also contained
much lower suspended solids than did the
SSM washwaters. These results indicate
that the primary mechanism for contami-
nant removal for the SSM was particle
separation. Since this mechanism alone
was able to account for a high percentage
of contaminant removal, the experiments
showed no significant improvement when
conditions were varied. The only param-
eters that appeared to improve removals
were those that would improve the par-
ticle separation in the soil matrix.
Although the SSM experiments achieved
high removals, only 55% of the washed
soil mass was recovered and a washwater
containing over 20% solids was produced.
The washwaters from the actual site soils
experiments had less suspended solids,
but it also removed fewer contaminants
from the soil matrix. Therefore, to effec-
tively remove contaminants from soils con-
taining a small fraction of fine materials,
the solubilization mechanism would need
enhancement. The resultant washwaters
would also contain lower suspended sol-
ids. Future work to evaluate enhance-
ment of the solubilization mechanism in
soil washing is recommended.
The full report was submitted in fulfill-
ment of Contract No. 68-03-3409 by Camp
Dresser & McKee Inc., under the sponsor-
ship of the U.S. Environmental Protection
Agency.
•U.S. Government Printing Office: 1993 — 750-071/60253
-------
Mary E. Loden is with Camp Dresser & McKee Inc., Cambridge, MA 02142.
Chl-Yuan Fan is the EPA Project Officer (see below).
The complete report, entitled "Evaluating of Soil Washing Technology," (Order No.
PB91 • 206 599/AS; Cost: $23.00 subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Edison, NJ 08837
United States
Environmental Protection
Agency
Center for Environmental
Research Information
Cincinnati, OH 45268
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Official Business
Penalty for Private Use $300
EPA/600/S2-91/023
------- |