United States
                    Environmental Protection
                    Agency
Hazardous Waste Engineering
Research Laboratory
Cincinnati OH 45268
                                                                                       o\"/
                                                                                       x  -  "'&
                                                                                       *s, -   r -
                    Research and Development
EPA/600/S2 85/129   Dec. 1985
&ER&         Project Summary
                    Treatment of  Contaminated
                    Soils with Aqueous  Surfactants

                    William D. Ellis, James R. Payne, and G. Daniel McNabb
                     The full report presents the results,
                    conclusions, and recommendations of a
                    project performed to develop a technical
                    base for decisions on the use of chemical
                    countermeasures at releases of hazard-
                    ous substances. The project included a
                    brief literature search to determine the
                    nature and quantities of contaminants
                    at Superfund sites and the applicability
                    of existing technology to in situ treat-
                    ment of contaminated soils. Laboratory
                    studies were conducted to develop an
                    improved  methodology applicable to
                    the in situ treatment of organic chemical
                    contaminated soil.
                     Current technology for removing
                    contaminants from large volumes of
                    soils (too large to excavate economical-
                    ly)  has been limited to in situ "water
                    washing." Accordingly, the laboratory
                    studies were designed to determine
                    whether the efficiency of washing could
                    be enhanced significantly (compared to
                    water alone) by adding surfactants to
                    the recharge water and recycling them
                    continuously.
                     The  use of  an  aqueous nonionic
                    surfactant pair for cleaning soil spiked
                    with PCBs, petroleum hydrocarbons,
                    and chlorophenols was developed
                    through bench scale shaker table tests
                    and larger scale soil column tests. The
                    extent of contaminant removal from the
                    soil was 92 percent for the PCBs, using
                    0.75 percent each of Adsee® 799
                    (Witco  Chemical) and Hyonic®  NP-90
                    (Diamond Shamrock) in water. For the
                    petroleum hydrocarbons, the removal
                    with a  2 percent aqueous solution of
                    each surfactant was 93 percent. These
                    removals  are orders of  magnitude
                    greater than obtained with just water
                    washing and represent a significant
                    improvement over existing  in situ
                    cleanup technology.
  Treatability studies of the contami-
nated leachate were also performed to
investigate separating  the surfactant
from the contaminated leachate to allow
reuse of the surfactant. A method for
separating the surfactant plus the con-
taminant from the leachate was devel-
oped; however, all attempts at removing
the surfactant alone proved unsuccess-
ful.
  Based upon the results of the labora-
tory work,  the aqueous surfactant
countermeasure is potentially useful for
in situ cleanup of hydrophobia and
slightly hydrophilic organic contami-
nants in soil, and should be further
developed on a larger scale at a small
contaminated site under carefully con-
trolled conditions. However,  reuse of
the surfactant is essential for cost-
effective application of this technology
in the field.  Accordingly, any  future
work should investigate the use of other
surfactants/surfactant combinations
that may be more amenable to separa-
tion.
  This Pnoject Summary was developed
by EPA's Hazardous Waste Engineering
Research 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
  The Comprehensive  Environmental
Response, Compensation,  and Liability
Act of 1980  (CERCLA  or Superfund)
recognizes the need to develop counter-
measures (mechanical devices, and other
physical, chemical, and biological agents)
to mitigate the effects of hazardous sub-
stances that are released into the envi-

-------
ronment and clean up inactive hazardous
waste disposal sites. One key counter-
measure is  the  use of  chemicals and
other additives that are  intentionally
introduced into the environment for con-
trolling  the  hazardous substance. The
indiscriminate use of such agents could,
however, worsen the contamination
situation.
  The  U.S.  Environmental  Protection
Agency's Hazardous Waste Engineering
Research Laboratory has initiated a
Chemical Countermeasures Program to
define  technical  criteria for the use of
chemicals and other additives at release
situations of hazardous substances such
that  the combination of  the released
substance plus the  chemical or  other
additive, including any resulting reaction
products, results in the least overall harm
to human health and to the environment.
Under  the  Chemical Countermeasures
Program, the efficacy of in situ treatment
of large volumes of subsurface soils, such
as found around uncontrolled hazardous
waste sites, and treatment of large, rela-
tively quiescent  waterbodies contami-
nated with spills of water soluble hazard-
ous substances,  will be evaluated.  For
each situation, the following activities are
planned: a  literature search  to compile
the body of  existing  theory  and  data;
laboratory studies on candidate chemicals
to assess adherence to theory and define
likely candidates for full-scale  testing;
full-scale, controlled-condition, reproduc-
ible tests to assess field operation possi-
bilities; and  full-scale tests  at  a site
requiring cleanup (i.e., a "site of oppor-
tunity").
  This  project, to  develop the use of
aqueous surfactants for in situ washing
of contaminated soils,  was the first
technique to be developed  under  the
Chemical Countermeasures  Program.
The  results and conclusions from an
information  search formed the basis for
the laboratory development work.  Simi-
larly, the results and conclusions from
the  laboratory work are intended to
provide the  basis for another project
involving large-scale testing of a chemical
countermeasure, either in a  large test
tank or under controlled conditions at a
site of opportunity.
Information Search
  The information search was conducted
to determine the nature and quantities of
hazardous  soil contaminants at Super-
fund sites, and to assess the applicability
of existing tech nology for in situ treatment
of contaminated soils. To determine what
                                    2
types of soil  contaminants requiring
cleanup were likely to be found at hazard-
ous waste sites, a survey was made of the
contaminants at 114 high priority Super-
fund sites. The classes of chemical wastes
found at the greatest number of sites, in
order of  decreasing  prevalence, were:
slightly water  soluble organics  (e.g.,
aromatic  and halogenated hydrocarbon
solvents, chlorophenols),  heavy metal
compounds, and  hydrophobic organics
(e.g., PCBs, aliphatic hydrocarbons).
  A variety of chemical treatment meth-
ods were considered  that might prove
effective  in  cleaning up soils contami-
nated  with these  wastes.  However,
methods for in situ chemica I treatment of
soils will  probably be  most effective for
certain cleanup situations, such as those
in which:

• The contamination is spread over a
  relatively large volume of subsurface
  soil, e.g., 100 to 100,000 m3, at a depth
  of 1  to 10 m; or
• The contamination is not highly con-
  centrated, e.g., not over 10,000 ppm,
  or the  highly contaminated portion of
  the site  has  been  removed or sealed
  off; or
• The contaminants can be dissolved or
  suspended in water, degraded to non-
  toxic products, or rendered immobile,
  using chemicals that can be carried in
  water to the zones of contamination.

  For contamination less than 1  m deep,
other methods such as landfarming (sur-
face tilling  to promote aerobic microbial
degradation of organics) would probably
be more  practical. For highly contami-
nated zones of an uncontrolled hazardous
waste landfill or a spill site, methods such
as excavation and removal, or excavation
and onsite  treatment would probably  be
more practical than in situ cleaning of the
soil.
  Findings  under the information search
indicated that  aqueous surfactant solu-
tions might be applicble for  in  situ
washing  of slightly  hydrophilic (water
soluble) and hydrophobic organics from
soils. Texas Research Institute (TRI) used
a combination  of equal parts of Witco
Chemical's Richonate®* YLA, an anionic
surfactant,  and Diamond  Shamrock's
Hyonic® NP-90, a nonionic surfactant, in
several  laboratory  column  and two-
dimensional modeling studies for displac-
ing gasoline from sand packs.
 'Mention of trade names or commercial products
 does not constitute endorsement or recommendation
 for use
  To further verify which organic waste
chemicals should be targeted for counter-
measures development. Field Investiga-
tion Team (FIT) summaries were examined
for the  maximum  concentrations  of
organic contaminants  in  the soil and
groundwater surrounding 50 Superfund
sites. Results of the survey indicated that
many hydrophobics were detected  in the
soils, mainly because hydrophilics tend to
be washed  from soil by infiltrating rain-
water. Hydrophobics  had the  highest
levels of all the organic contaminants,
with 11 compounds averaging in the 1 to
100 ppm range, and  with  chlordane
exceeding 1,000 ppm at one site. The soil
concentrations of slightly hydrophilic
compounds were in the range of 0.001 to
10 ppm.
  Based on these findings, the following
two hydrophobic and one slightly hydro-
philic pollutant groups were  chosen  as
model contaminants for testing  and de-
velopment  of an aqueous   surfactant
countermeasure:

• High boiling point Murban crude  oil
   fraction containing aliphatic and aro-
   matic hydrocarbons (1,000 ppm)
• PCB mixture in chlorobenzenes (Aro-
   clor® 1260 transformer oil) (100 ppm)
• Di, tri-, and pentachlorophenols mix-
   ture (30 ppm)

Laboratory Studies
  The laboratory research was conducted
to determine whether significant improve-
ments to the cleanup  of contaminated
soils with just water, the only in situ soil
cleanup  method demonstrated  to date,
could be obtained using aqueous surfac-
tants. Further laboratory development of
the surfactant countermeasure included
optimizing the concentration of surfactant
used for cleanup, and development  of
contaminated leachate treatment  meth-
ods.
  The aqueous surfactant countermeas-
ure was tested using two basic methods:
shaker table agitation,  to quickly  deter-
mine the soil/aqueous surfactant parti-
tioning of the model contaminants  under
differing conditions; and gravity flow soil
column tests to verify the cleanup be-
havior of the aqueous  surfactant  under
conditions resembling field use. Besides
the optimum surfactant concentration,
the effects  of  leachate treatment and
recycling were also studied.

Soil Characterization
  In choosing a soil for the  surfactant
washing tests, the applicability of the

-------
results to actual field situations was a
primary consideration. Selection included
identifying the native soils  at the ten
Region II Superfund sites for which data
was available, determining the most
commonly occurring soil type series, and
locating a soil of the same soil taxonomic
classification which could  be excavated
and used in the testing experiment. In
addition  to taxonomic classification, a
permeability rating of 10"2to 10'* cm/sec
was desirable since less permeable soils
would take too long to test.
  A Freehold soil series typic hapludult
soil was chosen for the  study. The total
organic carbon content (TOC) of the soil
was 0.12 percent by weight, implying a
relatively low  contribution  by  organic
matter to the adsorption of organic con-
taminants. The cation exchange capacity
(CEC) of the soil was determined to be 8.6
milliequivalentsper 100 gms, an extreme-
ly low value,  indicating an absence of
mineralogic clay in the soil.
  Using a percent moisture content of 11
percent and compacting the soil in the
columns to a density of 1.68 g/cm3 (105
Ib/ft3), an optimum percolation rate of 1.5
x 10~3 cm/sec was obtained  under a
constant 60 cm head.

Surfactant Selection
  The surfactant  combination  used by
TRI  for flushing  gasoline from sand,
Richonate®  YLA  and  Hyonic®  NP-90
(formerly  called Hyonic®  PE-90), was
screened along with several other surfac-
tants and surfactant combinations for the
following  critical  characteristics: ade-
quate water solubility (deionized water),
low clay  particle dispersion, good  oil
dispersion, and adequate biodegradabil-
ity.  The surfactants selected for ultimate
use in the laboratory studies were Adsee®
799 (Witco Chemical) and Hyonic® NP-9O
(Diamond Shamrock).

Soil Contamination Procedures
  Soil was contaminated using an aerosol
spray of the contaminant  mixture dis-
solved in methylene chloride. The meth-
ylene chloride was allowed to evaporate,
and the soil was mixed by stirring in pans.
The soil  was then tested  in shaker or
column studies.

Column Packing
  The soil columns used  in this study
were 7.6 cm (3 in.) inside diameter by 150
cm  (5 ft) long  glass columns. A plug of
glass wool was placed at the bottom of
the  column and  successive plugs of
contaminated soil weighing approximate-
ly 775 g were packed to a height of 10 cm
(4 in.) each. To ensure better cohesion
between layers, the upper 1/4 inch of
each plug was scarified. The soil was
packed to a total height of 90 cm (3 ft) and
compacted to a density of 1.68 to 1.76
g/cm3 (105 to 110 Ib/ft3),  yielding a
percolation rate which was comparable
to its natural permeability.

Shaker Table Tests
  Shaker table partitioning experiments
were conducted to determine the mini-
mum surfactant concentration required
to accomplish acceptable soil  cleanup.
After spiking Freehold soil with PCBs and
hydrocarbons, separately,  surfactants
were used to wash the soil by shaking in
containers on  a  constantly vibrating
shaker table.
  One  hundred grams of contaminated
soil were agitated with  200 ml of the
appropriate surfactant concentration on a
shaker table for one hour, then centri-
f uged, and decanted. Both soil and leach-
ate were analyzed  to determine how
much of the  contaminant  had been
removed.

So/7 Column Experiments
  During the first year of study, the effect
of soil washing with water, followed by
4.0 percent surfactants (2 percent each),
and a final water rinse was investigated
in soil column experiments using Murban
distillate  cut, PCBs and di-, tri-, and
pentachlorophenol contaminants. Free-
hold soil was spiked, separately, with
1,000 ppm Murban distillate cut, 100
ppm PCB, and  30 ppm chlorinated
phenols.
  Results of these column experiments
showed that the initial water wash  had
little effect; however, with  surfactant
washing, 74.5 percent of the pollutant
was removed by the leachate after the
third pore volume (i.e., volume of void
space in  the soil). Additional surfactant
increased the removal to 85.9  percent
after ten pore volumes. The pollutant
concentration in the soil was reduced to 6
percent of the initial spike value after the
tenth pore volume  of surfactant. The final
water rinse also showed only minimal
effects.
  Almost identical  behavior was observed
for the column experiments  using PCB
spiked soil: the initial water wash was
ineffective, but the  soil was cleaned
substantially by the4.0 percent surfactant
solution. After the tenth pore volume, 68
percent of the PCBs were contained in the
leachate, leaving only 2  percent on the
soil.
  Similar soil column experiments were
also conducted using a mixture of di-, tri-,
and pentachlorophenols, and, in contrast
to the PCB and  Murban distillate cut
results, 64.5 percent of the chlorinated
phenols were removed by the first water
wash, and only 0.56 percent remained on
the soil after the tenth  pore volume of
water.

Optimization of Surfactant
Concentration
  To make soil washing techniques cost
effective, it was necessary to determine
the minimum concentration of surfactant
that would yield acceptable soil cleanup.
Surfactant  concentrations were  varied
from 0 to 1.0 percent (2 percent  total
surfactant)  in shaker table experiments
using both PCB and hydrocarbon  con-
taminated  soils.  Column experiments
were  then  undertaken to verify shaker
table data and to further optimize surfac-
tant concentrations.
  Figure 1 shows the effect of surfactant
concentration on PCB partitioning  be-
tween soil  and  leachate. There  was
essentially  no cleanup of the soil  with
surfactant concentrations of 0.25 percent
(0.50 percent total) or below. Similar PCB
partitioning was observed for 0.75 per-
cent and 1.0 percent individual surfactant
concentrations, and the  most effective
cleanup occurred at these levels.
  As Figure 2 shows, similar soil/leach-
ate partitioning behavior was also ob-
served for  Murban hydrocarbons  with
varying surfactant concentrations.  Indi-
vidual surfactant concentrations of 0.25
percent and below were ineffective; in-
creased surfactant concentrations caused
increased soil cleanup from 0.50 to 0.75
percent surfactant; above 0.75 percent
surfactant concentration little enhance-
ment of soil cleanup occurred.

Column Verification
  To ensure that the optimum surfactant
concentration under gravity flow  condi-
tions was not significantly different than
under  equilibrated shaker table  condi-
tions, columns packed with Freehold soil
spiked with 100  ppm PCBs were  also
tested with varying surfactant levels.
  The columns were treated with  one,
two, or three pore volumes of 0.50,0.75,
or 1.0 percent surfactant before sacrifice
and soil analysis. The downward migra-
tion of  PCBs is apparent in  Figure 3,
which presents the PCB concentrations
in the various portions of the columns as
a function of pore volume for each of the
three  surfactant concentrations tested.
PCB mobilization was  not much greater
3

-------
                       702
100

-*. 90
| 80
| 70
1 60
| 50
Relative Contamir,
r? KO to -U
o o o o

ftf* &
t




y
\
%
i
^\\N\\V\\\\\\\\\\\\\\\V\\<
O



0
&
$
i
!
^
\
l
963 S7.2 c>7i





0
E3
%
I
^
I
i





0
x
|
J
^
I
I
1£^* ~ SOU
r~~f = / aa^h



0
'9.'
£
\
\\\\\v\\m\m\\\\mN
t

i
53.4
3;
1.7
\


7 i

~7.
£

6
63.0


              .007
                       .07
                                .70
                             .25
                                                 .50
                                              .75
                                                                   7.0
                             Surfactant Concentration (%J
Figure 1.   PCB Shaker table recoveries vs. surfactant concentration.
100
90
Relative Contaminant Concentration (%l
-*M CO-CkOlO)VJOo
oooooooo
t^ = So/7
<
ffifi

1
2.0
=3 	
77.
P5

3 S
7.6
=3 	
2.\
^

^^
88.9 Q = Leachate
0

5
7.2
=5 	
<•)
I
2
26
4
1.4
I
0
3't
'.8
32.5
26.3 P
fl
                 .007
                           .07
Figure
                         .70       .25      .50

                      Surfactant Concentration (%)

Murban Shaker table recoveries vs. surfactant concentration.

                        4
                                                                 .75
                                                                         7.0
with 0.75 percent surfactant than with
1.0 percent surfactant, and somewhat
less for the 0.50 percent surfactant
concentration. After three pore volumes,
the PCB concentrations at the bottom of
the column were of 244 fjg/g with the
0.50 percent surfactant, compared with
405 A/g/g using 0.75 percent surfactant
and 562  //g/g using  the  1.0 percent
surfactant.
  Results of the column experiments,
coupled with the results of the shaker
table experiments, indicate that the opti-
mum  surfactant concentration for  soil
cleanup is about 0.75 percent of each
surfactant or 1.5 percent total surfactant.

Evaluation of Leachate
Treatment Techniques
  Large amounts of surfactants and wash
water are required for field application of
this countermeasure technology. Surfac-
tants  are  expensive, and for this tech-
nology to be cost effective,  surfactant
recycling is an important consideration.
Accordingly, various leachate treatment
techniques were evaluated for their ability
to remove and concentrate the contami-
nants, while  leaving  the  surfactants
behind  for  further use.  All treatment
methods evaluated were ineffective in
separating the  contaminants  from  the
surfactant.  However,  several  leachate
treatment techniques were able to (1)
concentrate the contaminants to facilitate
disposal, and (2) clean the water enough
that it could be sent to a publicly owned
treatment works (POTW) or reused.
  Four  treatment alternatives  were
tested, and the  conditions for efficient
leachate treatment were optimized in
preparation for large-scale  field testing.
Foam fractionation, sorbent adsorption,
ultrafiltration, and surfactant hydrolysis
were subjected to preliminary laboratory
tests using simulated leachate.
  The results of the foam  fractionation
tests showed that good cleanup of the
leachate was achieved if the concentra-
tion of surfactant  was below about 0.1
percent, while no significant reduction in
surfactant occurred at starting concen-
trations above that.
  Eleven solid sorbents were tested for
their efficiency in removing PCBs and the
surfactants from an aqueous solution.
None of the sorbents was very efficient in
removing PCBs from a surfactant solution.
The most efficient sorbent for PCB re-
moval was the Filtrol XJ-8401, with an
efficiency of 0.00045  g/g; WV-G 12x40
Activated Carbon, and Celkate magne-
sium silicate were most efficient in overall
surfactant and PCB removal (0.195 g/g).

-------
  Hydrolysis treatment of the surfactant
and  contaminant-containing leachate
was also tested. Adsee® 799, a fatty acid
ester, formed a separate organic phase
upon hydrolysis that contained both the
surfactants and 95 percent of the organic
contaminants.
  Further treatment of the aqueous sur-
factant solution with a column of activated
carbon (Westvaco Nuchar WV-B 14x35)
yielded a solution containing only 0.01
ppm of  PCBs. Foam fractionation was
also  used as  a polishing  method for
removing  traces of  surfactants  from
aqueous solutions. A four-column series
of foam  fractionation columns operating
in a continuous countercurrent flow mode
was used. The test results demonstrated
that the residual PCB level (0.0036 ppm)
should be low enough to allow disposal to
a POTW, and low enough to permit reuse
of the leachate water  for soil cleaning.
However,  the  use of hydrolysis was
necessary for the higher surfactant con-
centrations found in the raw leachate.

Evaluation of Leachate
Recycling
  To evaluate the effect of recycling the
untreated aqueous  leachate on  soil
cleanup, column experiments were con-
ducted. The results showed that leachate
recycling—without some sort of treat-
ment—is not an acceptable method, as
contaminants become redistributed back
onto the soil with each successive pass.
However, a column experiment in which
the recycled leachate was  treated be-
tween each  pass showed very effective
cleanup of soil.
  Between passes, fresh surfactant was
added to the treated  leachate prior to
recycling,  and the soil  in the column
received four passes of fresh surfactant;
only the water was recycled. After four
passes,  less than 1.0 percent of the
original  soil contamination remained.


Conclusions and
R ecommendations

Effectiveness of the Surfactants
  Based on bench-scale tests designed to
screen potential surfactants for use as in
situ soil washing enhancers, a 1:1 blend
of Adsee® 799 (Witco Chemical Corp.)
and Hyonic® NP-90 (Diamond Shamrock)
was chosen because of adequate solubil-
ity in water, minimal mobilization of clay-
sized soil fines (to maintain  soil perme-
ability), good oil dispersion, and adequate
biodegradability.
620
580
540

500
460

-^ 400
1 360
I 32°
| 280

§ 240
u
i 200
760
120-
80
40
Pore
Volume:



401





316





33.
$





6
1st

244
210


!P2 77.4
t ft 61-2
$ 16.2 | 5.9\
1 j& Q fA
p
\
\




28.2
&
«s



















233
P
116
1
1

* 	
i




81.
7.06

a = TOP 5*
Q = Middle f

0 = Bottom 5
j;
405 K








3
|

^
^
^
^
^
^

^
332 |



275
n/«4
\
78.9 ^

7.50
'2nd '3rd' . 1st 2nd 3rd . 1st 2
0.5 ^ 	 U. 75








/7<
0
^
X
s
%'*»l
i
^
\
i,
t,
14.9 £
RS ^
i 3rd

                     Surfactant Concentration (%)

Figure 3.    PCB soil column cleanup vs. surfactant concentration.
                         *Samples Lost
  Shaker table and column experiments
show that 4.0 percent of this blend of
surfactants in water removed 93 percent
of the hydrocarbon and 98 percent of the
PCB pollutants from contaminated  soil.
These removals are orders of magnitude
greater  than those  obtained with  just
water washing and represent a significant
improvement to the efficiency of existing
technology.  Chlorinated phenols were
readily  removed from the test soil by
water washing alone.
  Shaker table experiments conducted to
determine the optimum surfactant con-
centration for soil cleanup, with PCB and
petroleum hydrocarbon (Murban) con-
taminated soils, showed the optimum
concentration to be  1.5 percent total
surfactant. Individual surfactant concen-
trations of 0.25 percent or less were
unacceptable for effective soil washing,
and individual surfactant concentrations
above 0.75  percent (1.5  percent total)
were excessive,  since no  significant
enhancement of cleanup resulted. In
addition, similar partitioning between soil
and surfactant solution by the two pollu-
tant types suggests that the resu Its wh ich
would be obtained in further large-scale
experiments with the lowtoxicity hydro-
carbons in a fuel oil like Murban might
reliably model the behavior of other more
toxic hydrophobic pollutant groups, such
as PCBs.
  The experiment which  evaluated the
effect of  leachate recycling, with treat-
ment applied to the PCB leachate between
cycles, showed that:

• Soil cleanup with  1.5 percent total
   surfactant  is good, with  less than 1
   percent of  the PCB remaining on the
   soil.
• The product of hydrolysis represents a
   relatively  small  volume (about  12
   percent of  the total mass of leachate)
   of highly contaminated material, which
   can be further treated by incineration,
   or disposed of for a minimal cost.
• The use of the sa me water for repeated
   cycles precludes  the  generation of
   large volumes of waste leachate.
• The final treated water after four cycles
   contains less than 0.0005 percent of
   the initial contamination encountered
   in the  soil.
  Additional surfactant tests are war-
ranted before this technology  can be
applied in the field. The surfactant com-
bination  used was water soluble, and
effective  in  soil cleanup, and allowed
good soil percolation rates, as the mixture
did not resuspend a significant amount of
the clay-sized particles in the soil, thereby
inhibiting flow. These characteristics are

-------
definitely  important; however, for this
technology to be cost effective, reuse of
the surfactant  is equally important.
Accordingly, it is recommended that other
surfactants/surfactant combinations be
evaluated that have the same "flushing"
characteristics but are also more amen-
able to separation for reuse. The surfac-
tant should  be screened for solubility,
clay dispersion, and oil dispersion, and
should also be screened by mutagenicity
tests to avoid the distinct possibility that
the release situation could be made worse
by the application of a toxic chemical or
other  additive.


Effects of the Test Soil
  The efficiency of cleanup of the hydro-
phobic organic contaminated  Freehold
soil by the aqueous surfactant solution
was directly affected by the low natural
organic carbon content of the  soil. The
low TOC (0.12 percent) represented little
organic matter  in the soil to adsorb the
organic pollutants spiked onto the soil, so
the contaminant removal  could be ex-
pected to be relatively easy compared to a
soil with, for example, a 1  percent TOC.
The removal of hydrophobicorganics from
a 1 percent TOC soil using the Adsee®
799 - Hyonic® NP-90  surfactant pair
would require more surfactant solution.
Also,  the surfactants would become
necessary for  removing  chlorophenols
from  a 1 percent TOC soil; water alone
would not be very effective.
  If additional  laboratory or pilot-scale
testing were undertaken, a second soil
type with greater percentages of organic
carbon should be considered for testing to
expand the overall applicability of the
program results to a broader variety of
soil matrices.
  The hydraulic conductivity of the Free-
hold  soil  packed  in the soil  columns,
which was measured at 1.05  x 10~3
cm/sec,  would  be practical  for field
implementation of the countermeasure.
However, the time required for surfactant
solution tof low through the soil  should be
considered. With this hydraulic conduc-
tivity,  if surface flooding were used to
obtain saturated conditions  from the
surface to a groundwater depth of 10 m
(33 ft), and assuming a porosity of 50
percent, it would take 5.5 days for one
pore volume of solution to flow through
the soil from surface to groundwater. A
flow rate under similar conditions, with a
soil permeability  of 1 x 10~4 cm/sec,
would yield flow rates of about 1.2 m/wk,
which is probably a practical lower limit
for the method.
Potential Target Contaminants
  The types of hazardous chemicals for
which  the surfactant countermeasure
was more effective than water without
surfactant, included hydrophobicorganics
(PCBs and aliphatic hydrocarbons in the
Murban  fraction)  and certain slightly
hydrophilic organics (aromatic hydrocar-
bons in Murban). The chemicals for which
the method is probably not applicable are
heavy metal salts and oxides,  and cya-
nides.  For soils with low TOC values,
chlorophenols and certain other slightly
hydrophilic organics can be removed with
water alone. However, for soils with high
TOC values, the use of aqueous surfac-
tants would  significantly  improve the
removal efficiency of slightly hydrophilic
organics.

Effective Treatment Methods
  A need to  conserve both water and
surfactant prompted the investigation of
leachate reuse or recycling. Recycling of
the untreated leachate is unacceptable
because portions of  the soil that  have
been previously cleaned are recontami-
nated rapidly by the introduction of spent
leachate.  The ideal  treatment method
removes and concentrates contaminants
while leaving the surfactants behind for
further use. However, the same chemical
and physical properties of the surfactant
mixture that help to extract the pollutants
from the soil also inhibit separation of the
contaminants from the surfactants. Due
to the high (percentage) level of surfactant
contained in  the leachate,  most of the
treatment methods evaluated were  inef-
fective. The best treatment that could be
obtained removed  both surfactants and
pollutants, leaving clean water for pos-
sible reuse or easy disposal.
  Additional efforts should be directed
toward optimizing  feasible and  cost-
effective methods of  leachate treatment
and in  particular separation of the sur-
factant for reuse. Ultrafiltration appears
promising and warrants further investi-
gation along with foam fractionation. The
use of already  existing  equipment and
technologies should be examined  in
greater detail to minimize scale-up costs.

Further Countermeasure
Development Before Field Use
  The  testing of a new technique,  in
which hazardous contaminants are rend-
ered more mobile, presents a potentially
greater environmental threat unless the
tests can be readily stopped at  any  point
as  required  to  permit  the immediate
remedy  of any failure  by  established
techniques.  Because the aqueous sur-  fl
factant countermeasure is still develop-  ^
mental, the field tests should be conducted
on a reduced scale until the procedures
are proven workable and the  important
parameters are understood and control-
led.
  The laboratory tests have established
that the technique of in situ washing with
aqueous surfactants is sufficiently effec-
tive for soil cleanup to justify tests on a
larger scale. Pilot-scale testing requires
the use of disturbed soil, and will probably
not further the development of the method
as  much  as controlled-condition .field
testing at a site of opportunity. An appro-
priate site for field testing should have the
following characteristics:

• Moderate to high permeability (coef-
   ficient of permeability of 10"4 cm/sec
   or better)
• Small size (e.g., 30 m x 30 m x 10 m
   deep)
• Minimal immediate threat to drinking
   water supplies
• Hydrophobic and/or slightly  hydro-
   phylic organic contaminants
• Concentrated contamination  source
   removed or controlled
• Low to moderate natural organic mat-
   ter content  in  soil (TOC  0.5  to 2
   percent).

If either small sites, or physically sepa-
rated sections of a large site (e.g., with a
slurry or grout wall) were selected, the
aqueous  surfactant  countermeasure
described  in  this report could be applied,
tested further, and improved to a point of
full  field  countermeasure  applicability.
However,  future work should evaluate
other surfactants  that have  the same
cleanup characteristics as those used in
the  laboratory studies but  are more
amenable to separation for reuse. Also,
prior to any larger scale/site of opportun-
ity studies, the toxicity of the surfactants
should be  ascertained.
                                                                           S. GOVERNMENT PRINTING OFFICE: 1986/646-116/20730

-------
     William D. Ellis, James R. Payne, and G. Daniel McNabb are with Science
      Applications International Corporation, McLean, VA 22102.
     Anthony N. Tafuri is the EPA Project Officer (see below).
     The complete report, entitled "Treatment of Contaminated Soils with Aqueous
      Surfactants,"{Order No. PB 86-122 561/AS; Cost: $11.95. 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:
            Hazardous Waste Engineering Resarch Laboratory—Cincinnati
            U.S. Environmental Protection Agency
            Edison, NJ 08837
United States
Environmental Protection
Agency
                          Center for Environmental Research
                          Information
                          Cincinnati OH 45268
Official Business
Penalty for Private Use $300

EPA/600/S2-85/129
0000129

U  S
                      PS
                        PROTECTION  AGENCY

-------