xvEPA
                            United States
                            Environmental Protection
                            Agency
                         National Risk Management
                         Research Laboratory
                         Ada, OK 74820
                            Research and Development
                         EPA/600/S-97/005
September 1997
ENVIRONMENTAL
RESEARCH   BRIEF
              Innovative Measures for Subsurface Chromium Remediation:
                   Source Zone, Concentrated Plume, and Dilute Plume

                 David A. Sabatini1, Robert C. Knox1, Edwin E. Tucker2 and Robert W. Puls3
Introduction

  Many sites  in the United States are contaminated with
toxic metals such as lead, cadmium, and chromium. Under
normal conditions, subsurface chromium contamination
exists in two stable oxidation states:  hexavalent, [Cr(VI)]
and trivalent [Cr(lll)].  Cr(VI) is both toxic and mutagenic;
Cr(lll) is of less health concern and, because of its lower
water solubility, its aqueous concentrations are generally
below water quality standards.  In the subsurface, Cr(VI)
generally exists in theanionicchromate(CrO42-)or bichromate
(Cr2O72-) forms which are relatively soluble and mobile. Thus,
the risk associated with ground-water contamination is high
and remediation of ground water contaminated with Cr(VI) is
of critical importance.  The conventional  approach for
remediating contaminated ground water sites has been
water-based pump-and-treat. In recent years it has been
recognized that this approach can require protracted periods
of time to approach treatment goals (Keely, 1989; Palmer and
Wittbrodt, 1991; Palmer and Fish, 1992).

  The behavior of chromium in soils depends upon various
factors, including: the form of chromium present, soil pH and
mineralogical properties, and presence of organic matter.
'Civil Engineering and Environmental Science and Institute for Applied
  Surfactant Research, University of Oklahoma, Norman, OK 73019
2Chemistry and Biochemistry, University of Oklahoma, Norman, OK
  73019
3 Subsurface Protection and Remediation Division, National Risk
  Management Research Laboratory, ORD,  U.S. EPA, Ada, OK
  74820
                     Several authors have studied the behavior of chromium in
                     soils as a function of these factors (Bartlett and Kimble,
                     1976; Zachara et al., 1989; Ainsworth et al., 1989; Eary and
                     Rai, 1991; Anderson et al., 1994). The presence of organic
                     matter and Fe(ll) is responsible for reduction of Cr(VI) to
                     Cr(lll) (Eary and Rai, 1991; Anderson et al., 1994). Addition
                     of multivalent anions (e.g., phosphate and sulfate) was
                     found to decrease chromate adsorption due to competition
                     for the same adsorption sites (i.e., ion exchange — Bartlett
                     and Kimble, 1976; Zachara et al., 1989; Puls  et al., 1994a).
                     Tucker et  al. (1992) showed the ability of a  cationic
                     polyelectrolyte to partially solubilize a solid BaCrO4 phase.
                     Puls etal.(1994a) showed the ability of an anionic surfactant,
                     sodium dodecyl sulfate (SDS), to significantly enhance the
                     elution of chromate in column studies.

                       Subsurface chromium contamination at the U.S.  Coast
                     Guard Support Center, near Elizabeth City, North Carolina, is
                     the focus of this research. A brief description of the site is
                     given here to provide a context for this research. For further
                     details on the site, the interested reader should consult Puls
                     et al. (1994a, b). A chrome plating shop was operated at the
                     Coast Guard facility for more than thirty years. Activities in
                     this shop resulted in the release of chromic acid into the
                     soils below the shop. While some Cr(VI) was reduced to
                     Cr(lll) in the vadose zone soils, the reducing capacity of
                     these soils was eventually overwhelmed and ground-water
                     contamination resulted.  Although the most concentrated
                     portion of the dissolved plume is down gradient from the
                     source  zone, the plume has not yet separated from the
                     source zone. This indicates that the remaining source zone
                     soils continue to leach chromium into the ground water. The

-------
dilute portion of the plume has migrated  down gradient
towards an adjacent river.

   As such, the contamination at this site can be divided into
three regions: (1) the source zone soils, (2) the concentrated
portion  of the ground-water plume, and (3) the dilute portion
of the ground-water plume. Applying conventional pump-
and-treat methods to all three regions will be highly inefficient.
Pump-and-treat remediation of  the  concentrated  plume
without regard forthe other regions will also be inefficient, as
the dilute portion of the plume will continue to migrate to the
river and the source zone  materials will continue to leach
into the plume. Therefore,  using an integrated approach to
simultaneously address these three  regions will be most
effective.  The technologies chosen  for each of the three
regions must be tailored to the unique characteristics of that
region.

   This  environmental research brief  reports on innovative
measures for addressing each  of these three regions.  For
the source zone, surfactant-enhanced chromium extraction
is evaluated for expediting the removal of chromium from
thesourcezonesoils, thereby mitigating thecontinual feeding
of the ground-water plume.  For the concentrated plume,
polyelectrolyte-enhanced ultrafiltration (PEUF) is evaluated
as an innovative treatment process with desirable operating
characteristics (less sludge production, higher quality final
water, etc.). Relative to the dilute plume, the hydrogeological
effectiveness of hydraulically passive, chemically reactive
barrier systems is evaluated (i.e., in situ reduction of Cr(VI)
to Cr(lll)). It is proposed that the collective use of these three
innovative technologies  will  significantly improve  the
remediation of subsurface chromium contamination.
Research evaluating each of these technologies is the focus
of subsequent sections.


Surfactant-Enhanced Chromium Extraction: Source
Zone Soils

Background:

   This section focuses on an innovative technology forexpediting
elution of chromium from the source  zone soils.  If the source
zone soils are not addressed, the plume will continually be
replenished and the efficacy of the plume treatment methods will
be compromised. As mentioned above, introduction of multivalent
anions and surfactants has been observed to enhance chromium
elution from contaminated soils. This research further evaluates
the use  of surfactant systems to  expedite chromium extraction
from the source zone soils.

   Surfactants (surface-active-agents) are amphiphilic molecules
consisting of lipophilic and hydrophilic groups.  This amphiphilic
structure results in the surface active  nature of surfactants  and
causes  them to concentrate in  interfacial regions.   Above  a
specific concentration, surfactant molecules form dynamic
aggregates or micelles. The concentration above which micelles
form is  known as  the critical micelle  concentration (CMC).
Aqueous-phase micelles  have a hydrophobic interior and  a
hydrophilic exterior, causing them to  behave like dispersed oil
drops.   Solubilization is the  phenomenon by which non-polar
species  partition into the organic interior of the micelles.  Micelles
and polar/ionic species interact mainly through hydrogen bonding
and electrostatic forces (Shimamoto and  Mima, 1979). Recently,
surfactant-enhanced pump-and-treat  remediation has been of
great interest (Nash, 1987; Abdul and Gibson, 1991;  Harwell,
1992; West, 1992;  Edwards et al., 1992; Palmer et al.,  1992;
Fountain, 1992; West and Harwell, 1992; Rouse et al.,  1993;
Shiau et al., 1994; and Sabatini et al., 1995).

   The main factor considered in evaluating surfactant systems
for chromium remediation is the efficiency of chromate extraction
(ratio of chromate removal by surfactant to chromate removal by
water).  It is hypothesized that surfactants can  displace the
adsorbed  chromate by  either ion  exchange,  precipitation-
dissolution  and/or counterion  binding mechanisms,  and that
further enhancement in extraction may be achieved if surfactants
with solubilized complexing agents are used.  Laboratory batch
and column studies were conducted to evaluate these hypotheses
using contaminated soil from the  U.S. Coast Guard site.

   The surfactants used in this research along with some of their
relevant properties are summarized in Table I. These surfactants
were selected based on  their type (anionic, zwitterionic,  etc.),
their susceptibility to losses (e.g., precipitation and sorption),
their ease of regulatory acceptance (having USFDA direct food
additive status), and experience with them in prior research
(Rouse etal., 1993; Shiau etal., 1994). Chromium contaminated
soil samples were obtained from the U.S. Coast Guard Support
Center, Elizabeth City, North Carolina, at depths from 1.5 ft to
6.5 ft. The fraction organic carbon content in Elizabeth City soils
varied from 0.0006 to 0.0027 (Puls et al., 1994a). Batch and
column studies,  as well as chemical analyses,  were conducted
according to standard  procedures, as documented elsewhere
(Nivasetal., 1996).

Batch Extraction Studies:

   Batch studies were conducted  using 3.0g of soil and 15 ml of
solution. Figure I compares the Cr(VI) extraction from Elizabeth
City soil using  deionized (D.I.)  water  and  surfactants.   The
amount of Cr(VI) removed is observed to increase with anionic
surfactant  concentration.  At concentrations greater than the
CMC, Cr(VI) removal was relatively constant (see Figure  I and
Table II).  Upon  equilibration D.I.  water solubilized 2.64 ppm of
Cr(VI).  The ratio of maximum  Cr(VI) removal by surfactants to
that of  D.I.  water ranged from 2.1 for Dowfax 8390 to 2.8 for
Deriphat-160 (see Table II).

   Anionic surfactants  could enhance Cr(VI) extraction by ion
exchange, precipitation-dissolution, and counterion binding. The
experimental data are not consistent with counterion binding as
the enhancement would  begin and increase above the  CMC.
Surfactant precipitation-Cr(VI) dissolution also does not appear
to be the mechanism since surfactants with high precipitation
resistance are expected to be much less efficient (e.g., Dowfax
8390).  However, the data are  consistent with ion exchange as
the primary extraction  mechanism because the enhancement
occurs  below the CMC and is  independent of  the precipitation
potential of the  surfactant.  Additional analysis of the sorption
data further corroborates this conclusion (as presented in  Nivas
etal., 1996).
   It was hypothesized that the extraction of chromate  from the
contaminated soil could be enhanced if a chromium complexing
agent is solubilized into the surfactant micelles.  Solubilization
assays were conducted for sodium dodecyl  sulfate (SDS) and
aerosol OT (AOT) using diphenyl  carbazide (DPC) as the
solubilizate. In both casesthe aqueous solubility of DPC increased
with  concentration  beyond  the CMC (maximum surfactant

-------
Table 1 Relevant Properties of Surfactants and Complexing Agent Used in this Study
Surfactant
SDS*
AOTd
Dowfax-8390
Deriphat-160
T-Maz 20
DPCo
Type
Anionicf
Anionicf
Anionic
Zwitter ionic
A/on ionic?
—
Chemical
Formula
C12H25OS03Na
(C9H1702)2CH2CHS03Na
C16H33C12H70(S03Na)2
R-NH(CH2COONa)2
C12H2402H3(CH2CHO)2002CH2CH03
C6H5(NH)2CO(NH)2C6H5
Average
Molecular
Weight
288.4
444.6
642
373
1228
242
CMC
(mM)
8.20F
1. 124s
3. Off
-
0.039"
—
Obtained from
Fisher Scientific
American Cyanamid Co.
Dow Chemical Co.
Henkel Corp.
PPG/Mazer
Fisher Scientific
  * SDS - sodium dodecyl sulfate
  b USFDA Direct Food Additive
  0 Mukerjee and Mysels, 1971
  d AOT - aerosol OT
  eShiau, etal., 1994
  'Obtained from Dow Chemical Company
  9 Complexing agent - Diphenyl Carbazide
    10
I
o
o
9
     8-
     6-
     2-
      A.   _|_  A    A    A
       •    T JJk   A    A

           A




     CMC Dowfax  CMC SDS


CMC AOT
Water

AOT

SDS
 A
Dowfax 8390
 X
Deriphat-160
                                           I             T
                   5           10          15           20
                  Initial Surfactant Concentration, mM
Figure I   Extraction of Cr(VI) by surfactants from Elizabeth City soil in
          batch systems.
concentration evaluated was 20 mM). While the aqueous DPC
concentration is 1.56 mM, at 20 mM surfactant concentrations
SDS and AOT solubilized 8.67  mM and  5.33 mM of DPC,
respectively.

   Figure II  shows the extraction of Cr(VI)  using surfactant-
solubilized DPC.  Use of DPC in the batch studies required
shifting to a solids based  concentration (mg Cr / Kg soil). The
results of these  studies are summarized in Table III.  Upon
equilibration, water solubilized  11.3  mg of Cr/Kg of soil. The
chromium removal increased with surfactant concentration, with
maximum removal being  evidenced above the CMC.  The
maximum ratio of chromium removal, with and without surfactants,
ranges from 9.3 for SDS to 12.0 for Dowfax 8390 (see Table III);
thus, the addition of a chromium complexing agent enhanced the
extraction of Cr(VI) by an order of magnitude greater than that
obtained  with D.I. water. The ratio of surfactant concentration at
which these removals were achieved to the CMC  ranged from 2.5
for SDS to 8.3 for AOT.  In all  cases the surfactant-solubilized
DPC outperformed the surfactant only results (see Table II). This
demonstratesthatthe micellar-DPC extraction can further improve
chromium removal relative to  surfactant  only  systems.  The
results of these studies show an enhancement of 3.7 to 5.7 times
over those of surfactants alone.
Table II    Results of Cr(VI) Extraction Studies by Surfactants and Hydrotropes from Elizabeth City Soil
Extracting
Agent
D.I. water
AOT
SDS
Dowfax 8390
Deriphat-160
Extracting agent
cone, at max.
Cr(VI) removal,
mM
3
10
1
5
Ratio
to
CMC
2.7
1.2
0.3
Max.
Cr(VI)
Cone.,
ppm
2.6
5.2
6.4
5.6
7.4
Ratio ofCrfVI)
removal by extracting
agent to that by water
1
2.0
2.5
2.1
2.8

-------
Total Cr removed, mg/kg of soil
8 Ł g g 8 i i
A A
+
+
Z
CMC Dowfax
CMC AOT
A
A
X
CMC SDS
x 2
Water
AOT
X
SDS
A
Dowfax 8390
                  5            10          15

                 Initial Surfactant Concentration, mM
20
Figure II   Extraction ofCr(VI) by surfactants with DPC from Elizabeth
          City soil in batch systems.
Column Extraction Studies:

   Column studies were conducted in 2.5 cm diameter by 15 cm
long glass chromatography columns; the average porosity and
pore water velocities were  0.39 and 9.1  cm/h, respectively.
Figure III shows the results of column studies evaluating Cr(VI)
removal using water, 10 mM AOT and 10 mM  Dowfax 8390.
Deriphat-160 was not included in these experiments due to its
extensive sorption in the batch studies. The results are presented
as the Cr(VI) concentration versus the number of pore volumes
of solution injected. Table IV summarizes the results from these
column studies (for further details on column studies see Nivas
et al., 1996).   Dowfax 8390  showed maximum Cr(VI) removal
followed by AOT and D.I. water.  In all three cases the maximum
Cr(VI) effluent concentration was obtained between 1.34  and
1.76 pore volumes.

   The trends in the column results are similar to those observed
for batch tests.   For the column  studies, when  the  Cr(VI)
concentrations in the effluent went below 0.5 ppm, injection  was
stopped.  After a week of no flow, the pumping of 10 mM Dowfax
8390 was continued in order to determine if the Cr(VI) removal
mechanism was rate limited,  as shown in Figure III.  The effluent
Cr(VI) concentration increased from 0.34 ppm to 3.34 ppm after
the flow interruption. This demonstrates that the Cr(VI) removal
mechanism may have been rate limited.  Injection of 10 mM
Dowfax 8390 to the originally waterflushed column also increased
the effluent Cr(VI) concentrations from 0.2 ppm to 2.38 ppm (see
Figure III).

   Figures IV and V show Cr(VI)  removal  using  surfactant-
solubilizedDPC. The results ofthese studies are also summarized
in Table IV. The maximum Cr(VI) removal in the effluent is higher
than that observed with surfactants alone (10.3 ppm and 19.7 ppm
for AOT-DPC and  Dowfax 8390-DPC columns, respectively).
This is about 1.7 to 1.9 times  greater  than that obtained  in
columns flushed by AOT and Dowfax 8390 alone.  In  batch
systems,  addition of DPC  enhanced the extraction by 3.7  to
5.7 times over that of surfactants alone.  Thus, the removal
efficiency enhancement  in column studies  is less than that
observed in the batch systems.  The number of pore volumes
required to reach maximum Cr(VI) removal concentrations was
longer than anticipated as evidenced in Figure V (11.22 and
12.06 pore volumes for AOT  and Dowfax  8390  with  DPC,
respectively, versus 1.54 and 1.76, respectively, for surfactants
alone).  The slope of the percentage removed relative to water
alone curves is very low until about 10 pore volumes.  The lack
of peaks at 1.54 to 1.76 pore volumes, which were observed in
surfactant only - no DPC columns, indicates that monomer/Cr(VI)
ion exchange  is not occurring.

   It is postulated that two competing mechanisms are taking
place in the surfactant-DPC systems.  First,  DPC exits the
micelles and  forms a Cr(VI)-DPC complex with the solids-
associated chromate (this process is apparently rapid enough to
prevent monomer-Cr(VI) ion exchange). Second, the extraction
of the Cr(VI)-DPC complex into the micellar core occurs, due to
complexation with the micellar-phase DPC.  It appears that prior
to  10 pore volumes the former  mechanism dominates; hence,
there is no rapid increase in the removal of Cr(VI) in the early
stages of injection.  During this phase the formation of Cr(VI)-
DPC complex was clearly seen by the  pink coloration of the
medium (the pink color being indicative of the Cr-DPC complex).
Once the solids-associated chromium has been complexed, the
extraction  of  chromate anions  and solubilization of Cr-DPC
complex takes place. This is indicated by the increase in the
slope of both curves after 10 pore volumes in Figure V. However,
at  about  12 pore volumes the  removal  efficiency of  both the
systems again starts to decrease, possibly due to the diminishing
solubilization potential of micelles already saturated with  DPC.
At this point injection of pure micellar solutions of surfactants
would be expected to enhance the solubilization  of the solid
associated Cr(VI)-DPC complex.

   At about 14 pore volumes the AOT-DPC column was switched
to  flushing  with  20 mM AOT  only (no DPC).  This initially
increased the Cr(VI) removal concentration in the effluent from
Table III   Results ofCr(VI) Batch Extraction Studies by Surfactants with Solubilized DPC from Elizabeth City Soil
Extracting
Agent
D.I. water
SDS
AOT
Dowfax 8390
Extracting
agent cone, at
max. Cr
removal, mM
20
10
10
Ratio to
CMC
2.5
8.3
3.3
Max. Aqueous
Cr Cone.,
mg/Kg of soil
11.3
106
119
135
Ratio of Cr removal
by extracting agent to
that by water
1
9.3
10.4
12.0

-------
                                                                                            120
                                             Flow interruptions followed
                                          by 10mM Dowfax 8390 injection
      O
                            10
      20            30            40
   Pore Volumes Injected
50
         Water

         % Removal, Water   "
      10mM AOT                  -*- 10 mM Dowfax 8390

—    % Removal, 10mM AOT   	% Removal Dowfax
Figure III  Removal ofCr(VI) by D.I. water and surfactants in columns.


0.86ppmto4.22ppm(seeFigure IV); however, the enhancement
was only temporary. Subsequent introduction of Dowfax 8390
alone, at a concentration of 10 mM, again increased the Cr(VI)
concentrations in the effluent (from 1.83 ppm to 7.11  ppm, as
shown in Figure IV). The injection of 10 mM Dowfax 8390 was
continued until 25 pore volumes, by which time around 175% of
Cr(VI) was removed relative to D.I. water.

   Similarly, the column flushed with solubilized  DPC in 10 mM
Dowfax  8390 was switched to surfactant only at about 16 pore
volumes. This switch increased the  Cr(VI) concentration in the
effluent  from 10.11  ppm to 15 ppm. The injection of 10 mM
Dowfax  8390 was continued until Cr(VI) removal in the effluent
was negligible.  The total Cr(VI)  removed was 213% relative to
                    D.I. water in less than 20 pore volumes, as opposed to the
                    surfactant only case which achieved only 125% after 25 pore
                    volumes. An interesting point to note is that in this column run,
                    the tailing of the concentration with time was not observed (see
                    Figure IV). The lack of an elution tail is obviously advantageous
                    for chemically-enhanced pump-and-treat processes.

                      It is suggested that the removal of Cr(VI) from the soil can be
                    further enhanced by optimizing the time of switching from injection
                    of surfactant with DPC to surfactant alone and/or by increasing
                    the surfactant concentration.  The time of switching should be
                    immediately after the complete formation of Cr-DPC complex of
                    the solids-associated chromate.  For example, at about 10 pore
                    volumes the formation of Cr-DPC complex  appeared to be
Table IV  Cr(VI) Extraction from Columns by Water, Surfactants Alone, and Surfactant- Solubilized DPC
Extracting Agent
D.I. Water
AOT
Dowfax 8390
AOT with DPC
Dowfax with DPC
Max. Cr(VI)
removed in
effluent, ppm
5.3
7.0
11.8
10.3
19.7
Ratio of Cr(VI)
removed by
extracting agent to
that by D.I. water
1
1.3
2.2
1.9
3.7
Total Pore
volumes
flushed
35.9
26.5
24.4
24.3
18.9

-------
                           i
                          10        15
                       Pore Volumes Injected
20
25
Figure IV  Removal ofCr(VI) by surfactants with DPC in columns.
                                                                 250
               I  200

               2
               Ł
               «  150-
                                                                       20mMAOTw/DPC


                                                                       10mM Dowfaxw/DPC
                                                                                         10mMDowfax8390.
                                                             (D
                                                             EC
                                                             o
                                                                 100-
          I

10       15        20

Pore Volumes Injected
                                                                      25
               Figure V
                Percentage removal of Cr(VI) by surfactants with DPC in
                columns.
complete. Switching at this time to pure micellar solutions would
have resulted in extraction of Cr(VI) in fewer pore volumes. By
increasing the concentration of the surfactant-DPC mixture, the
surface capacity for DPC could be satisfied in fewer pore volumes;
higher surfactant concentrations for complex extraction would
likewise expedite this process.   It should be noted that the
concentrations for the column runs were  selected based on
results from batch studies which indicated that extraction efficiency
was leveling off at surfactant concentrations of 10-20 mM; this
illustrates the importance of conducting column studies and not
relying solely on batch results.

   In summary, batch resultsdemonstratethatsurfactant systems
have the potential to enhance chromium elution by a factor of 2
to 3 (surfactants alone) to an order of magnitude  (surfactant-
complexing agent systems versus water alone). While column
studies corroborated results of batch tests, they also illuminated
operational considerations that require additional research. Thus,
whilethistechnology shows great promise forenhancing chromium
extraction from soils, further laboratory and field scale studies are
necessary prior to full scale implementation.

Polyelectrolyte-EnhancedUltrafiltration:
Concentrated Plume

Background:

   The research presented in this section focuses on an innovative
treatment  process for ground water extracted from the
concentrated portion  of a  contaminant plume. This treatment
process has improved operational  characteristics  relative to
existing technologies and will thus  further optimize  the pump-
and-treat methodology for the concentrated plume. The  next
section will discuss a passive approach for dealing with the dilute
portion of the plume (where pump-and-treat would become less
efficient).
                  For several years, a particulararea of research atthe University
               of Oklahoma has  involved the combination of  colloids (e.g.,
               surfactant  micelles  and polyelectrolytes)  with ultrafiltration
               membranes to  remove  contaminant materials from aqueous
               streams (Christian et al., 1988; Sasaki et al., 1989; Christian
               et al., 1990; Christian et al., 1992; Tucker et al., 1992;  Krehbiel
               etal.,  1992).  These  colloid-enhanced ultrafiltration  (UF)
               techniques have the potential to be used in low-energy,  efficient,
               and selective processes for removal of target ions or molecules.
               One focal point has been the development of an efficient process
               for removal of toxic metals and metalloids in anionic form (e.g.,
               chromate, arsenate  and selenate) from  aqueous streams or
               ground water. In addition to removal of these ions to potable water
               levels, the ultrafiltration  method does not, unlike many current
               techniques,  further degrade the overall  water  quality in the
               process of removing one or more toxic components. A specific
               goal has been to selectively remove the toxic ion in the form of a
               compact solid waste and produce effluent water with essentially
               the same ionic strength as that which enters the process.

                  In contrast with reverse osmosis, ultrafiltration is not generally
               a direct method for removal of ions from aqueous streams. The
               pore size  of UF membranes is characterized by a molecular
               weight cutoff (MWCO) value which is a rough estimate of the size
               of molecules which are retained by the membrane. For instance,
               a  MWCO of 5000  Daltons  indicates that a molecule with a
               molecular weight larger than 5000 will be largely retained by the
               membrane.  In a hypothetical example,  inorganic ions  such as
               chloride, sodium, and chromate in aqueous solution are much
               smaller than this and will freely pass through the UF membrane.
               However, if a soluble polymer of molecular weight greater than
               the MWCO is added to the stream to be filtered and, if the polymer
               binds chromate, then both the polymer and the chromate will be
               retained bythemembrane. FiguresVI andVII presentqualitatively
               a  polyelectrolyte/chromate mixture  and  subsequent binding of

-------
   cr
                                              r
              Na+
                  CrOf
Na+
                                                               Ci;
                                                                                                      cr
                                                             Cl-     CrOf
                                                                          Na+
Cl-   Na+
Figure VI
Homogeneous aqueous mixture of polyelectolyte and
chromate.
                  Figure VII  Chromate ion replaces chloride ion on the cationic polymer.
the chromate ion to the polymer. The solution passing through the
membrane (permeate) will be depleted in chromate while the
solution retained by the membrane (retentate) will be enriched in
chromate  (and  polymer). The  concentration of  chloride  and
sodium ions will be essentially  the same on both sides of the
membrane since these ions are not bound to the polymer.

   Conventional removal of chromate is commonly done through
lowering aqueous pH by addition of acid, adding a reducing agent
such as sodium sulfite, and finally, adding lime or other base to
raise the solution pH and precipitate the reduced chromium as
the Cr(lll) hydroxide. The chromate reduction process uses a
substantial quantity of acid, base and reducing agent with the
result that the effluent water from this conventional process has
increased ionic strength due to salt loading (Kosarek, 1981). For
example,  the product water from one recently  implemented
chromate  reduction  process has a sulfate ion concentration of
1700  ppm. This may be compared with an input level of only
34 ppm sulfate in the feed ground water (Buehler, 1993). Further
use or disposition  of the product water containing such high
sulfate ion concentrations might be restricted.  Additionally, a
large  sludge volume is produced from the reduction process
because other metal hydroxide compounds precipitate along
with the chromium hydroxide. Chromium is a minor component in
this precipitate due to both indiscriminate hydroxide precipitation
and the fact that the hydrous oxides thus produced have a very
low solids content, typically on the order of 10%-20%.

   The polyelectrolyte-enhanced ultrafiltration (PEUF) process
for chromate (CrO42~) removal  involves addition  of a cationic
polyelectrolyte[poly(dimethyldiallylammonium chloride)] to a feed
stream containing Cr(VI); the solution is then filtered with a UF
membrane cartridge, so that the chromate-polymer mixture will
be retained in a small volume (the  retentate), while a large
fraction of the feed solution is produced as highly purified water.
Either chromate  or bichromate  {HCrCy} will  bind to  the
polyelectrolyte.  Model calculations show that at pH 6.5 there is
essentially a 50:50 mixture of the chromate and bichromate
anions in aqueous  solution (with a trace amount of dichromate
anion).  At pH 7.0,  ca. 76% of the total  chromium exists  as the
chromate  anion.  As long  as the solution pH is  above  6.0, a
significant fraction of the Cr(VI) in solution will exist as a divalent
anion and will replace monovalent chloride  counterions  on the
polymer.  Other anions  in ground water such as sulfate  and
phosphate will also  bind to the polymer but sulfate is expected to
be the main competing ion in the pH range 6-8 due to mineral
                                                  solubility considerations.

                                                     The binding of chromate with polymer creates membrane
                                                  selectivity since the polymer-chromate complex is too large to
                                                  pass through pores in the membrane. Ions which are not bound
                                                  to the polymer pass freely through the membrane. Due to the
                                                  membrane transparency for unbound ions there is no salt or brine
                                                  retention by the membrane and the process can be carried out at
                                                  low applied  pressures  (10-100  psi).  Figure  VIII depicts  the
                                                  membrane separation.

                                                     Permeate water from the membrane is of lower ionic strength
                                                  than the feed water and is substantially reduced in chromate
                                                  concentration. In laboratory experiments PEUF has been used to
                                                  remove 99.9% of chromate from feed water containing as much
                                                  as 5 ppm Cr(VI). The retentate solution from the membrane is a
                                                  concentrated solution of polymer and chromate (and sulfate, if
                                                  present). This mixture is treated to separate the polyelectrolyte
                                                  for reuse in the process.  Although several mechanisms might be
                                                  used for polymer-chromate separation in the retentate liquid,  one
                                                  method is theaddition of barium chloride at less than stoichiometric
                                                  ratios to precipitate sulfate and chromate as  a compact solid
                                                  waste.

                                                  Field Test:

                                                     A small scale field test of the PEUF process forthe removal of
                                                  chromate from ground water at the U. S. Coast Guard Support
                                                  Base in Elizabeth City, North Carolina, was conducted in March
                                                  1993. Figure IX gives a schematic of the overall process which
                                                  was employed. The field test consisted of a complete process
                                                  implementation to remove Cr(VI). The ground water from three
                                                  different monitoring wells contained Cr(VI) levels of 2.1 to 3.8 ppm,
                                                  as well as several hundred ppm of dissolved solids including ca.
                                                  Feed Solution
                                                        Permeate
                                                         Na+   Cl-
                                                        Retentate  t
                                                      poly + chromate
                                                  Figure VIII Ultrafiltration of the aqueous mixture.

-------
              POLYELECTROLYTE-ENHANCEDULTRAFILTRATION
                                      for Chromate Removal
                                        Membrane System
        Ground
        Water _
        Feed
                    Polymer
                    Feed
           Polymer
           Makeup
                                                                       Permeate      Disposal
                      H2O,Na+,CI_-    °r.  .   ..
                                        Remjection
Retentate
 Polymer      1[
                                                                 Feed
 Figure IX   Qualitative schematic of Elizabeth City UF test apparatus.
90 ppm sulfate, as shown below.  The feed water pH was above
6.0 and no pH modification was used in this test. The Cr(VI) and
sulfate concentrations detected in several wells were as follows:
Well #2, 3.1 ppm, 85 ppm; Well #12, 1.5 ppm, 82 ppm; Well #13,
3.8 ppm, 86 ppm.

   The single stage pilot UF apparatus was capable of producing
150 gal/day of purified water.  The pilot test was conducted over
a five day period.  The maximum operating time per day was
12 hours and the  longest continuous period of  operation was
8 hours.  An initial charge of 7 gal. of polymer concentrate was
used and this was not augmented during the test period. The UF
system was operated at an applied pressure of 50 psig with a feed
water recovery (as permeate) fraction of 0.75 to  0.80.  Ground
water was pumped from a monitoring well at a nominal rate of 450
mL/min, mixed with polyelectrolyte solution (150 mL/min), and
then  ultrafiltered to produce a permeate (purified water) and a
retentate  solution  containing polymer, chromate,  and sulfate
ions.

   The  retentate solution was treated by addition  of a  small
quantity (less than stoichiometric) of barium chloride to precipitate
barium sulfate and barium chromate and to partially regenerate
the polymer. The renewed polymerwasthen filtered and recycled
to mix with fresh ground water.  A very limited quantity of solid
precipitate was  produced in  this test.   It was not possible to
accurately measure the total barium chromate and barium sulfate
solids because these collected in both the polymer retentate tank
and on the cartridge filters  used in the test.  An estimate of the
maximum quantity of precipitate which could be  produced can
         easily be made.  If it is assumed that the well water contains
         sulfate at 90 ppm and chromate at 3 ppm (as hexavalent chromium)
         then complete  removal of both these ions as the barium salt
         would produce 0.83 grams of BaSO4 and 0.055 grams of BaCrO4
         for each gallon  of ground water treated.

           The polyelectrolyte-enhanced ultrafiltration process performed
         flawlessly during the several day test period. Some permeate
         analyses for Cr(VI) were done on site by ion chromatography with
         UV  detection  of the diphenylcarbazide complex.   Ion
         chromatography  analyses  were also  conducted for sulfate.
         Chromate levels in the permeate over a several day period were
         at virtually the detection limit ofthe ion chromatography apparatus
         [<50 ppb as Cr(VI)]. Sulfate levels were in the range of 1 to 5 ppm
         in the permeate.  Additional on-site  colorimetric analysis with
         diphenylcarbazide, and analysis of permeate samples brought
         back to the laboratory, indicated  a  possible range of  Cr(VI)
         concentration of  30 to 70  ppb, as  measured  in various UF
         permeate samples over the course ofthe test.

           The PEUF process has several advantages in ground-water
         treatment for chromate removal. The purified water produced by
         the PEUF process has an ionic strength similarto or less than the
         original ground  water. In the Elizabeth City test a few ppm of
         Cr(VI) and ca. 80  ppm of sulfate were removed from the ground
         water and  replaced with chloride ion. The product water was
         substantially reduced in ionic strength relative to the feed ground
         water.  This may  be contrasted with the great increase in ionic
         strength of product water (relative  to feed water) from the
         conventional reduction-precipitationtreatmentforCr(VI) removal.

-------
Thus, the PEUF product water is suitable for reuse, or reinjection
to aid in the  remediation process, or direct disposal.   The
chromate waste is concentrated into a small volume which makes
possible either further treatment or easy recovery of Cr(VI).  The
PEUF process is continuous and does not use large volumes of
acid or base for pH modification in contrast with conventional
reduction treatment.  The polyelectrolyte can be reused many
times in the process. Additional direct applications of the PEUF
process are found in the removal of arsenic and selenium from
aqueous streams. Toxic anion removal from aqueous streams by
the PEUF process  is the subject of  a patent  issued to  the
University of Oklahoma (U.S. #5,302,290).

Hydraulically-Passive Redox Barriers:  Dilute Plume
Background:

   A  variety of field,  laboratory and  modeling  studies were
conducted to evaluate the viability of using hydraulically-passive,
chemically-reactive  barriers  for remediation  of the dilute
subsurface chromium contaminant plume at Elizabeth City. Ideally,
these barrier systems would operate by installing the reactive
barrier in the path of migration of the contaminant plume.  The
plume would move through the reactive barrier due to the existing
regional gradient; hence,  the  barrier would be hydraulically
passive. Material in the reactive barrier would cause an in situ
reaction that results in immobilization of the dissolved chromium.

   Various in situ reactions or contaminant removal mechanisms
have  been  proposed  for reactive barriers. Starr and Cherry
(1994) discuss five groups of ;ns/fu reactors based on contaminant
removal mechanisms  (e.g.,  sorption,  precipitation),  chemical
delivery (e.g.,  dissolution of reactants, nutrients), or  physical
process (e.g., air sparging) proposed for the reactive material or
reactive portion of the barrier. The focus of their study was on the
hydraulic efficiency of a dual barrier (funnel-and-gate) system
that uses low permeability sections (funnels) to direct contaminated
water through  higher  permeability materials (gates) in which
reactions would effect contaminant removal.
     A redox barrier has been proposed for the dilute chromium
   plume at Elizabeth City, North Carolina (Powell etal., 1995; Puls
   et al., 1995). The reactive material in the redox barrier would  be
   native aquifer material amended with iron filings. The proposed
   redox  reaction would involve reduction of Cr(VI) to Cr(lll) and
   oxidation of ferrous (Fe+2) iron to ferric (Fe+3) iron. The reacted
   species would then precipitate as insoluble hydroxides (Powell et
   al.,  1995). This research focuses on  the  hydrogeological
   considerations of implementing such barriers.

   Field Studies:

     A series of site characterization activities were completed at
   the Elizabeth  City site. These activities focused on quantifying
   hydraulic characteristics of the shallow ground-water formation.
   Hydraulic conductivity values  for  the various  monitoring wells
   were developed from slug tests (Bouwer, 1978). Variations in the
   regional gradient  were assessed from  synoptic water level
   observations of the monitoring well network. Finally, the potential
   impacts on the regional gradient of the river (e.g., wind waves,
   tidal fluctuations) immediately adjacent to the field site were
   assessed through continuous water level recorders installed in
   selected monitoring wells. Ongoing drilling operations provided
   aquifer material for  subsequent laboratory studies. Analyses
   from periodic sampling episodes were used to generate estimates
   of total mass  and migration of the center of mass of the dilute
   chromium  plume.

     In general, the monitoring wells at  the Elizabeth City site
   (Figure X)  are completed at two different depths (15 ft or 20 ft)
   within  the shallow aquifer. The shallow wells are screened over
   the bottom 5 feet; the deeperwells are screened over the bottom
   10 feet. Hydraulic conductivity values derived from the slug tests
   (Table V)showtheaquiferto be a fairly uniform, highly conductive,
   sandy material.  However, there  does appear to  be a slightly
   lower permeability lens running east-west (parallel to the river)
   about  60  to 80  feet  out from the source zone. The regional
   gradient (Figure XI) is slightly southeast to northwest and does
   not deviate seasonally more than  a few degrees.
                                                  Hangar Bldg. 79
                                  MW3
                                                MW21 •
                                                      MW26
                                                MW17   -
                               MW18

                                   MWs 30, 29, 28
                                                                      MW15
                                                             MWs 24,23,22
                                                                   MW16
                                                                                  Nprth
                                                                                     MW27
                                                              MW33

                                                 MW19  RX5             MW20
                                  20ft
MW34_         +
    MW32          sparge wells
                                                   Pasquotank River
 Figure X   Monitoring well locations at Elizabeth City.

-------
Table V   Hydraulic Conductivity Values from Slug Tests on Selected Monitoring Wells (Calculated Based on Both Partially and Fully
          Penetrating Assumptions)
Well
Ml/1/3
MW11
MW16
MW16a
MW17
MW18
MW19
MW20
MW20b
MW22
MW27
MW27b
MW28
MW28b
MW29
MW29b
MW32
MW32b
Casing
Diameter
(in)
2
4
2

2
2
2
2

2
2

2

2

2

Total
Depth
(ft)
15
25
15

15
15
15
15

50
20

20

20

15

Elevation
Top of
Casing
(ft)
107. 16
107. 15
106.74

106. 74
106.7
105.86
105.88

106.8
106.29

106.45

106. 73

106.4

Screened
Interval below
T.O.C.
(ft)
92. 16 ®
82. 15 ®
91.74®

91.74®
91.7®
90.86®
90.88®

56.8®
86.29 ®

86.45 ®

86. 73 ®

91.4®

97.16
87.15
96.74

96.74
96.7
95.86
95.88

66.8
96.29

96.45

96.73

96.4

Hydraulic Conductivity
(ft/day)
partially fully
penetrating penetrating
5.1
15.9
2.4
2.6
2.9
1.3
2.6
8.2
5.3
4.7
0.9
0.3
8.9
8.8
5.3
5.4
6.1
2.5
4.9
15.2
2.3
2.5
2.8
1.2
2.5
8
5.2
4.3
0.8
0.3
8.6
8.6
5.1
5.3
5.9
2.4
   The observed movement of the contaminant plume is slower
than would  be predicted  based  on the measured hydraulic
gradient and conductivity values and projected adsorption effects.
It was hypothesized that weather-related or tidal effects in the
adjacent river could have reversed the gradient of the shallow
ground-water formation for extended periods  in  the past. The
supposed gradient  reversal  was  proposed as one factor
contributing  to the slow migration  of the  contaminant plume. A
three-month  period of continuous water level observations showed
infrequent disturbances in the regional gradient. In addition, the
magnitudes  of these disturbances were small and  readily
attenuated by the highly conductive aquifer.

Laboratory Studies:

   Laboratory studies were conducted to assess the hydraulic
characteristics of aquifer materials amended  with iron filings.
Aquifer materials  from four  different  depth  ranges  in the
unsaturated  zone were retrieved from RX5 (Table VI). Constant
head column tests, using three different flow rates, were completed
on the materials from each depth range.  Hydraulic conductivity
values reported are the geometric mean  of the three measured
values. The  same tests were conducted  on samples of aquifer
materials amended with iron filings. The results of the hydraulic
conductivity determinations showed no definite trend with respect
to the addition  of iron filings.  Although  more decreases  than
increases in hydraulic conductivity were observed for the iron
amended samples, the magnitude of the changes between the
different depth  ranges far exceeds the  variations  due to the
addition of iron filings.
   Changes in hydraulic conductivity associated with redox
barriers will result in conflicting effects. The hydraulic conductivity
of the reactive barrier must be higherthan the surrounding aquifer
in order to induce flow of the contaminant plume through the
reactive materials. However, increased  hydraulic conductivity
values will result in decreased residence times within the reactive
Figure XI  Regional gradient at Elizabeth City.
                                                            10

-------
Table VI   Laboratory Hydraulic Conductivity Values and Ratios of
	Aquifer Materials With and Without Iron Filings	
    Source (depth)
%Fe     K (ft/day)    K Ratio
RX5(10'-11')
RX5(13'-14')
RX5(15'-16')
RX5(17'-18')
0
5
20
0
20
0
5
0
20
0.56
0.15
0.28
0.71
0.19
73.6
34.8
6.36
7.41
1.00
0.27
0.49
1.00
0.27
1.00
0.47
1.00
1.16
barrier, which could reduce the effectiveness of the redox reaction.
Moreover, to achieve the higher hydraulic conductivity values, a
medium with  large grain sizes would have to be added to the
native materials. The larger grain sizes will have reduced specific
surface areas which could also reduce the  effectiveness of the
redox reaction. However,  Starr and Cherry (1994)  note that
changes in hydraulic conductivity of more than 1 orderof magnitude
above the native material result in relatively little increase in the
amount of flow through a reactive barrier.

   Breakthrough curves for chloride were developed  for the
aquifer materials with and without iron filings (Figure XI I) at the 13
to 14 foot depth range. The iron amended  materials appear to
show less dispersion than the native materials.

Modeling Studies:

   Estimates  of total  chromium mass in the subsurface and
location of the center of mass were developed for four sampling
episodes. The mass estimates (Table VII) and  center of mass
locations were developed using spatial moments and triangular
elements  (Freyberg, 1986; Knox, 1993). Refinements  in the
monitoring  well network over time  resulted in an increased
number of wells. However, not all chromium sampling episodes
included  all available wells. If the  spatial network includes the
monitoring wells with low (background) levels of chromium that
are located outside  the actual dissolved  chromium plume, the
areal extent of the chromium plume appears  to increase.  Hence,
total mass estimates of dissolved chromium are artificially inflated.
The latter sampling episodes focused only on those monitoring
wells that  had  previously shown  elevated concentrations of
dissolved chromium.

   Spatial moment calculations show that the center of mass of
the chromium plume has migrated out approximately 80 feet from
the source area toward the river (Figure XIII). However, samples
from monitoring wells near the river taken during the most recent
sampling episode indicate that the chromium plume could  be
discharging into the  river. Total mass of dissolved chromium in
the plume is probably less  than 15 kg (Table VII).

   Numerical simulations of alternate barrier configurations were
completed  using the USGS Method of Characteristics (MOC)
model  and input  data derived from the field  and laboratory
studies. All simulations utilized the configuration of the plume
fromthe most recent sampling episode as initial conditions. Initial
simulations showed that the permeable barrier had to be specified
as having a transmissivity value 1000 times higher than the
surrounding native  materials in order to  cause discernable
changes in the flowlines of the formation. For each configuration,
the solid contour lines indicate the impact of the barrier on the flow
path and shape of the plume after eight years of transport. The
solid contour lines indicate what portions of the plume would
actually pass through the barrier. Those portions of the plume
passing through the barrier are assumed to be totally reduced
(other research evaluates Cr(VI) removal; Powell et al., 1995).
Observation wells were placed on the periphery of the barriers to
assess how much of the  plume was captured.

   Four basic barrier configurations (Figure XIV) were considered;
the transverse barrier, the longitudinal barrier, the Y-shaped
barrier, and the funnel-and-gate system. The funnel-and-gate
system was  similar to the  Y-shaped barrier,  but utilized
impermeable  funnels to direct flow through  the permeable gate
similarto the systems analyzed by Starr and Cherry (1994). Each
barrier configuration occupied the same number of cells in the
transport grid.

   Figure  XV depicts  simulated chromium concentrations in
peripheral observation wells overtime for each of the four barrier
configurations; hence, the figure depicts how much chromium is
not removed  by each configuration.  In general, the transverse
barrier performed best in terms of capturing the plume. Starr and
Cherry (1994) also found that barriers were  most effective when
they were  oriented  predominantly orthogonal to the  regional
gradient. The Y-shaped barrier performed slightly betterthan the
funnel-and-gate system. Given the  increased complexity of
construction of the funnel-and-gate systems, questions arise as
to the viability of these systems. However, these simulations also
showed that the hydraulic  conductivity  of the barrier  must  be
much higher  (two orders of magnitude) than the surrounding
media  in  order to significantly alter the  ground-water  flow,
regardless of the configuration. Forthe highly conductive formation
at Elizabeth City, these results indicate the most viable alternative
to be a transverse  redox  barrier wider than the contaminant
plume be placed down gradient of the plume.
                        BTCRX5-513-141
Figure XII  Breakthrough curves for aquifer materials with and without
          iron filings.
                                                           11

-------
Table VII Mass Estimates of Dissolved Chromium
Sampling Episode

June 1991
April 1992
July 1992
March 1993
June 1993
September 1993
Total Mass
(kg)
9.4
8.0
4.12
4.63
5.04
14.3
Monitoring
Samples in
12
10
12
7
11
8
Well
Grid






Summary / Recommendations

   Advanced remediation  efforts  should  consider  innovative
processes for each of three regions:  (1) the source zone soils,
(2) the concentrated portion of the contaminant plume, and (3)
the dilute portion of the plume. This environmental research brief
reports on innovative measures for addressing each of these
three regions.   For the source  zone,  surfactant-enhanced
chromium extraction was evaluated for expediting the removal of
chromium from the source zone  soils, thereby  mitigating the
continual feeding ofthe ground-water plume. Forthe concentrated
plume,  polyelectrolyte-enhanced ultrafiltration (PEUF) was
evaluated as an innovative treatment process with desirable
operating characteristics (less sludge production, higher quality
final water, etc.). Relative to the dilute plume, the hydrogeological
effectiveness of hydraulically-passive, chemically-reactive barrier
systems were evaluated (in situ reduction of Cr(VI) to Cr(lll)).

   Batch studies  demonstrated that  surfactant systems can
enhance chromium elution by a factor of 2 to 3 (surfactants alone)
to an order of magnitude (surfactant-complexing agent systems)
versus water alone. While column studies corroborated results
of batch tests,  they also illuminated operational considerations
that require  additional research.   Thus, while this technology
shows great promise for enhancing chromium extraction from
soils,  further laboratory and field-scale studies are  necessary
prior to full-scale implementation.
               The polyelectrolyte-enhanced ultrafiltration (PEUF) system
            performed very well during the pilot-scale field demonstration,
            reducing  permeate chromium concentrations to near or  below
            the detection limit (<50 ppb as Cr(VI)).  Permeate sulfate levels
            were reduced to the 1 to 5 ppm range as well. These results were
            achieved without significant increases in the ionic strength ofthe
            treated water,  as experienced  using conventional  reduction-
            precipitation processes.  This along with the reduced chemical
            demands ofthe PEUF process, make it a very attractive process.
            The successful field demonstration reported herein will help to
            expedite  its utilization in full scale systems.

               The hydraulic characterization  activities demonstrated that
            the aquifer is relatively homogeneous with respect to hydraulic
            conductivity  and that the natural  gradient, which is from the
            source area toward  the  Posquotank River,  did not deviate
            significantly with time.  Numerical modeling studies focused on
            identifying optimal  orientation  and configuration for  the
            hydraulically-passive, chemically-reactive barrier. The difference
            in  hydraulic  conductivity between the barrier and  the aquifer
            necessary to alter the flow lines of the contaminant plume was
            found to be three orders of magnitude. These differences are not
            practical given  the fairly high  conductivity ofthe natural aquifer.
            The optimal orientation forthe barrierwas found to be orthogonal
            to  the flow path of the plume.  Hence,  a  simple linear,  down
            gradient barrier, wider than the plume and orthogonal to the flow
            path, is recommended. As constructed, the barrier is a continuous,
            hanging (i.e., not keyed into an underlying impermeable formation)
            wall 150 feet long. The barrier is only slightly more permeable
            than the  natural aquifer, but  is longer than the plume is wide;
            hence, the entire contaminant plume should eventually traverse
            the barrier. Headless  across the  barrier will be monitored to
            determine residence times ofthe flowing fluids.

               While  each  of these  processes is  at  different  stages of
            development, it is proposed that the collective use of these three
            innovative technologies will significantly improve the remediation
            of  subsurface chromium contamination. Future research should
            continue to evaluate the synergism of these and other innovative
            technologies as we strive to optimize remediation of subsurface
            contamination.
                          June 93
                           June 91
                                     April 92
    MW17
                                         July 92
               Sept 93
                                                 North
March 93
           5ft
                              RX5
                               +
                       Pasquotank River
Figure XIII  Center of mass  locations based  on  spatial moment
           calculations.
Disclaimer

   The U.S. Environmental Protection Agency through its Office
of Research and Development partially funded and collaborated
in the research described  here under Cooperative Agreement
No. CR-820736 to Rice University. It has been subjected to the
Agency's peer and administrative review and has been approved
for publication as an EPA document.  Mention of trade names or
commercial  products  does not constitute  endorsement or
recommendation for use.

Quality Assurance Statement

   All research projects making conclusions or recommendations
based on environmentally related measurements and funded by
the Environmental Protection Agency are required to participate
in the Agency Quality Assurance Program.   This project was
conducted under an approved Quality Assurance Program Plan.
The procedures specified in the plan were used without exception.
Information on the plan and documentation ofthe quality assurance
activities and results are available from the Principal Investigator.
                                                           12

-------
                                                           Transverse Barrier
                                         Cr Plume
ouu
700
600 •
500 •
400
300
200
100 •

,*. Observation
*** Well

§r « years
^T^

Initial Plume
i i i i i i i i i
                                 100
200     300
    X(ft)
400   500
                                                                     ~
                                              Heads and Velocities
                                                                        800
                                                                        700 - -t 11111111111111111111111111111111111
                                                                             iMtittttttio.o1111111111111o.o
                                                                        600
                                                                             tttttttttttttttttttttttttttttttttttt
                                                                            . .11111111111.200 1111111111111.2001
                                                                        500 - -t 1111111111111111111111111 '„•„• 1111111
                                                                             M 111111 M t.eoo-t t M M H H H11? TTTTtTT
                                                                            . .t 11111111111111111111111111111111111
                                                                        200
                                                                        100-
                                                                                        K.200 t I M I t t t I ) I I-.200 I I I I I I
                                                                                       M-.400 I I I I I I I I I I I I [-400
                                                                             HttttBarrier
                                                                                          'MfHHMMVLi^ ttttttt
                                                                                          04111111111 rnrr 1111111
                                                                        300 - -t 111111111111111111111111 L; 20.
                                                                             tt ttttt****-1 .Zi"V^V>ja.ffj'"*] < 11,t 11
                                                                                   10°    20°
                                                          300    400    500
                                                          Longitudinal Barrier
                                           Cr Plume
                    fc
700
600
500
400

300
200
100
n


& ^\ 8 years
• ynr
Barrier — ~fj -1\
y
Initial Plume

i i i i i i i i
                                                                       800
                                                                       700
                                              Heads and Velocities
                                                                       600  -
                                                                       500  -11111111111
                                                                       200
                                                                        100
                                                                             tttttttttttttttttttttttttttttttttttt
                                                                             .t 11111111111111111111111111111111111
                                                                             Httttttttt. 200I I I I M I « I I I -.200
                                                                                        0.0 I I I II I I
                                                                                                     H-0.0
                                                                             t t t t t t t t t 11 t t t t t t t t t t 11 t t t t t t t t t t t t t
                                                                             tttttttttttttttttttttttttttttttttttt
                                                                             I I II I I I I I t t .200 I I I I I I I I I I I I .200 | | | | | | I
                                                                                               Httttttttt"
                                                                                               tl!!tttT:«0'
                                                                             1111111111 ri'yVTT
                                                                             tttttttttttttttttttttttttttttttttttt
                                                                                        -61

                                                                                                          ttttttt
                                                 iiitttt'mttriu,t
                                   . 1111111111111 \ \ \ \	
                                              t'' *-»r<«[nF>»J-Lt •'•ttttttt
                             fc-.n- i	f•I80(! 11 \l|l IW/ M 11 i"" ttttttt
                             -E- 4°° " 11111111111111111 fit tl 111111111111111
                                    111111111111111 »//|Ml\t 111" yji-LLU
                                              LI oo-uirnvvM-n1.-0.0; ttttttt
                                              • - tttTrmttttt
                               300
                                    t m 11 M m *. V,, I I 11 11 I 111 r,1i"t ttttttt
                                    ttttttttttt I ITtt t t t t t t t t t t t t t t t t t t t t
                                    't 11111111111111111111 tLU.H.AO'H I I 11 IT
                                                                             tHttttttili
                                                                             Ittttttttttt
                                                                                        1 1 1 1 1 1 1 1 1 ttiu.H.
                                                                                        ; ioj-i HII rntt 1
                                                                                                      '" «J t * * * *'
                                                                                        ttttttt tttj^t-t-1.601 ttttttt
                                                                            -ttttttttttt* •jjJ^HTTt 11111111111111
                                                                             I I I I II I I I nli i 111111111 H 1JJW-M i I I I I
                                    ,t 111111111111111111 IUAS^ . 1111111111
                                    1111111111111 \ \\^Aff\\ \ 1 \\UJJ4-M-H-t
                                    ttttttHJJA\.^T\\\\\\\>fp-TTT\1111111
                                    tttttttttll 11 \ \ \ \\\\v< «\ \ \1U 1111 M
                                    11111111111 11 \ \ \ \<.^V<^\i;.°T^''' 111111
                                 100
200     300
   X(ft)
400    500
                                                                                  100
200     300
   X(ft)
                                                                                                        400    500
Figure XIV  Redox barrier configurations used in numerical simulations (a) transverse barrier; (b) longitudinal barrier; (c) Y-shaped barrier; (d) funnel
            and gate.
                                                                   13

-------
                                   Cr Plume
              800
              700  -



              600  -



              500  -



           §400  -
           I*


              300



              200  -



              100  -
                           Y-Shaped Barrier


                                     800
            Heads and Velocities
8 years
            "" Initial Plume
                                                                       tt
                                                                      ..tt
  700



  600



  500



§400
                                      200
                                      100 -
                                                                       4-M-
                                                                                 4-..200) 11 m 11 n 11 -.200
                                           t I I tt I I I tt I 0.0 I I I I I I I I I I I I 0.0
                                           ttt
                                                      ttt
                                                      T.200 t t t t t
                                           tt
                                           11111111111'''11 ******* t tl*••1111111
                                           I I I I I I I I I IT;4.°9 111111111111 -.4PP
 tt
- -t 11111tttttt1 i-M-rm~Lt 1111 •••ttttttt
 11111111117-600. 11 tTlt 1111111 T T 11111111
- -t 111111111111111 titIt 11111111111111111
                      f' <' 'lann-t ******
 t 111 t M t I t t.800MjJo3JUH-TTTTf90, 1111111
 11111111111111 turniyt 111111111111111
 1111111111111111111111111 VooU-U-t-44-t
              OO'-ti 111 tt-rrrn 1111111111
                              ttttttttt
                                      300 - -t 111111111111111 ttttttttt*'.jut tttt tt.
                                           *ttt"*"*-*-i"r*	" r  • •* 1111111
                                                  ttttti
                                                        111111111 tt_U.i.40-f-l I I 111T
                                                        .4oJ-w-t-nTfTttt 1111111111
                                          - -t 1 1 1 1 1 1 1 1 1 1 •
                                                                       1 1 t_U-U-*-U
                                                                       .60t 1 1 1 1 1 1 1
 T T T T T tttt
             ttt tt ttrmj-t-t-,1
             • i-iu-H-nT111111   	
             V?i 111111 M \u w-rrrrrnTt
 11111111 M m u n»t)>rtsP. u m u 111
 111111111 \\ \\ \JJJ«T\ \ \ \ \ \\xuui i ttt
                        »^rtT\ \nuttt
                                                                       ttttttttttttt\»\\v
                         100     200     300    400    500
                                    X(ft)
                                                  1	1	h

                                          0      100    200     300    400    500
                                                           X(ft)
                                                      Funnel and Gate
                                    Cr Plume
\J\J\J
700 -
600
500 -
/•— s
Ł400 -
*
300


200
100 -
o -



8 years / g
llli
-jjjjaL-*.

\XA/
r<Ł^M
{^)j
"~ Initial Plume

i i i i i i i i i
I I I I I I I I I
0 100 200 300 400 50
X(ft)
                                                                 800
                                                     Heads and Velocities
                                                                      . .t 11111111111111111111111111111111111
                                                                        t t ) t t t t t t t .2001 I t I I M I t M I -.2001 I I I I I I
                                                                 700  +i



                                                                 600



                                                                 500
                                            II III III It o0 I I I I I I I I I I I I 0.0
                                          - -t 111111111111
                                          . .1 I I I I I I I I I I .200 I I I I  I I I I I I I I .200 ttttttt
                                           tttttttttttttttttttttttttttttttttttt
                                           t t t t t t t t t t t• " t ttttttttttt» * * t t t t t t t
                                           1 I I I I I I I I I I -;4P9 T t t t  t t t t t t t t -40Q'
                                                               §400
                                                                 300



                                                                 200



                                                                 100




                                                                    0
                                                                 n
                                           ttttttttttttttttttttttttttttttt
                                          - -t 111111111111111 U4Jt t tttt ™'_ttttttt
                                                                             ttt
                                          - -t 111111111111111 tf

                                           t t t t t t t t t t t.800l
                                                                                 i 1.60-11 i i i i i i i  i n tttttttttt
. .t t t t t t 11 t 11 t t t t t t t t t t tttt t,'onJ-*-'-t-u-u
 tttttt*""i zo+ttrrTTTTTT, i11 ttttttt
 11111111111 t't 11111111111111111111111
 t t t t t t 11 t 11 t t t t t t t t t ttt_U. 1.404-1 I I I I If
 111 ttttttui 40-1-1111 rTTTtt t tttttttttt
 TTTTTTTTTtt t't 11111111111111 'jit'""
 1111111111111111 111 tLU-H-,1.6bt ttttttt
- -t 1111111111 •' u-m-rTtt 11111
 111 H 1111 rrVi 11111111 M •»* •
 1111111111 \ \ \\\ \\\ \ \ UA
 1111111111\\\\\
 tttttttf t«-i. \ sST
•tttttttttlti1\\\\\\v
 11111111111.11 \ N \ \ NNV'^S iorfVt \\ \\ 111
                                                                     ,»u»u 1111
                                                                            100
              H	h
              200    300
                 X(ft)
                                                                       400    500
Figure XIV Redox barrier configurations used in numerical simulations (a) transverse barrier; (b) longitudinal barrier; (c) Y-shaped barrier; (d) funnel
           and gate.
                                                              14

-------
                                      Cr Concentrations at Observation Wells
                  0.5
                  0.4 --
               to.3
               o
               o
               O
               6 °-2
                                                             'V  ,*w*
                                                                 ''
                                                        / Longitudinal Barrier
/••v
                                                         Time (years)
Figure XV Chromium concentrations for monitoring wells peripheral to redox barriers.
References

Abdul,  A.S., and Gibson,  T.L. (1991) "Laboratory Studies of
    Surfactant-Enhanced Washing of Polychlorinated Biphenyl
    from  Sandy  Material", Environmental Science and
    Technology, 25, 665-671.

Ainsworth, C.C., Girvin, D.C., Zachara, J.M.,  and Smith, S.C.
    (1989) "Chromate Adsorption  on Goethite:  Effects of
    Aluminum Substitution", Soil Science Society of America
    Journal, 53, 411-418.

Anderson,  L.D., Kent, D.B., and Davis,  J.A. (1994) "Batch
    Experiments Characterizing the  Reduction of Cr(VI) using
    Suboxic Material from a Mildly Reducing Sand and Gravel
    Aquifer", Environmental Science and Technology, 28, 178-
    185.

Bartlett, R.J., and Kimble, J.M. (1976) "Behavior of Chromium in
    Soils: II.   Hexavalent  Forms", Journal of Environmental
    Quality, 55, 383-386.

Bouwer, H. (1978).  Groundwater Hydrology. McGraw-Hill, New
    York,  pp. 114-117.

Buehler,  K.  Private  communication from  Directorate of
    Environmental Management, Tinker Air Force  Base,
    Oklahoma, September 21, 1993.

Christian, S. D., Bhat, S. N., Tucker, E. E., Scamehorn, J. F.,  and
    El-Sayed, D. A. (1988)  "Micellar-Enhanced Ultrafiltration of
    Chromate Anion from Aqueous Streams."  AlChE Journal,
    34(2), 189.
         Christian, S.D., Scamehorn, J.F., and Tucker, E.E. (1992).
            "Innovative Methods for Chromium Remediation Project,
            Final Report",  Prepared for the U.S. EPA,  Robert S. Kerr
            Environmental Research Lab, Ada, Oklahoma.
         Christian, S. D., Tucker, E. E., Scamehorn, J. F. (1990). "Colloid-
            Enhanced Ultrafiltration Processes for Purifying  Aqueous
            Streams  and Reservoirs," American Environmental
            Laboratory, 2(1),  13.

         Eary, L.E.andRai, D. (1991) "Chromate Reduction by Subsurface
            Soils under Acidic Conditions",  Soil Science Society of
            America Journal,  55, 676-683.

         Edwards, D.A., Laha,  S., Liu,  Z., and Luthy, R.G. (1992)
            "Solubilization and Biodegradation of Hydrophobic Organic
            Compounds  in  Soil-Aqueous  Systems  with  Nonionic
            Surfactants", In Transport and Remediation of Subsurface
            Contaminants, D.A. Sabatini and R.C. Knox Eds., ACS
            Symposium Series 491, American  Chemical  Society,
            Washington, D.C., 159-168.

         Fountain, J.C. (1992)"Field Tests of Surfactant Flooding: Mobility
            Control of Dense  Nonaqueous-Phase Liquids", In Transport
            and Remediation  of Subsurface Contaminants, D.A. Sabatini
            and R.C. KnoxEds., ACS Symposium Series 491, American
            Chemical  Society, Washington, D.C., 182-191.

         Freyberg, D.L.  (1986) "A Natural Gradient Experiment on Solute
            Transport in a Sand Aquifer  2. Spatial Moments and the
            Advection and Dispersion of  Nonreactive Tracers", Water
            Resources Research, 22(13), pp. 2031-2046.
                                                         15

-------
Harwell, J.H. (1992) "Factors Affecting Surfactant Performance
    in Groundwater Remediation Applications", In Transport and
    Remediation of Subsurface Contaminants,  D.A.  Sabatini
    andR.C. Knox, Eds., ACS Symposium Series 491, American
    Chemical Society, Washington, D.C., 124-132.

Keely, J. (1989). "Performance Evaluations of Pump-and-Treat
    Remediations." U.S. EPA, EPA/540/4-89/005, 19 pp.

Knox, R.C. (1993) "Spatial Moment Analysis for Mass balance
    Calculations  and Tracking of  a Subsurface Hydrocarbon
    Mound",  Ground  Water Monitoring and  Remediation,
    Summer, pp. 139-147.

Kosarek, L. J. (1981). "Removal of Various Toxic Heavy Metals
    and Cyanide from  Water by Membrane Processes",  in
    Chemistry in Water Reuse, W. J. Cooper, Ed.,  Ann Arbor
    Science, Ann Arbor, Michigan, 261 pp.

Krehbiel, D. K., Scamehorn, J. F., Ritter, R., Christian, S. D. and
    Tucker, E. E.  (1992) "Ion-Expulsion Ultrafiltration to Remove
    Chromate from Wastewater."  Separation  Science &
    Technology, 27(13), 1775.

Mukerjee, P.  and Mysels,  K.  J.  (1971)  Critical Micelle
    Concentrations of Aqueous  Surfactant  Systems. NSRDS-
    NBS 36, U. S. Dept of Commerce, Washington,  D.C.

Nash, J.H. (1987) "Field Studies  of In-Situ Soil Washing", EPA/
    600/2-87/110,  U.S.  Environmental Protection Agency,
    Cincinnati, Ohio.

Nivas, B.T., Shiau, B. J., Sabatini, D. A. and Harwell, J. H. (1996)
    "Surfactant Enhanced Remediation of Subsurface Chromium
    Contamination."  Water Research, 30(3), 511-520.

Palmer, C.D. and Fish,  W. (1992) "Chemical Enhancements to
    Pump-and-Treat  Remediation,"  U.S. Environmental
    Protection Agency, U.S. Government Printing  Office,
    Washington,  DC; EPA/540/S-92/001.

Palmer, C., Sabatini, D.A., and Harwell, J.H.  (1992)"Sorption of
    Hydrophobic Organic Compounds and Nonionic Surfactants
    with Subsurface Materials", In Transport and Remediation of
    Subsurface Contaminants, D.A. Sabatini and  R.C. Knox
    Eds., ACS Symposium Series 491, American Chemical
    Society,  Washington, D.C., 169-181.

Palmer, C.D. and Wittbrodt,  P.R. (1991) "Process Affecting the
    Remediation  of   Chromium-Contaminated  Sites",
    Environmental Health Perspectives, 92, 25-40.

Powell, R. M., Puls,  R. W., Hightower, S. K. and Sabatini, D. A.
    (1995)  "Coupled Iron Corrosion and Chromate  Reduction:
    Mechanisms for Subsurface Remediation." Environmental
    Science and Technology, 29(8), 1913-1922.

Puls, R.W., Clark, D.A., Paul, C.J., and Vardy, J. (1994a)"Transport
    and Transformation of Hexavalent Chromium Through Soils
    and into Ground Water", Journal of Soil Contamination, 3,
    203-224.

Puls,  R.W.,   Clark, D.A., Carlson,  C.,  and  Vardy,  J.  (1994b)
    "Characterization of Chromium-Contaminated Soils Using
    Field-Portable X-Ray  Fluorescence", Ground Water
    Monitoring and Remediation, Summer, 111-115.
Puls, R.W., Powell, R.M. and Paul, C.J. (1995)"lnSitu Remediation
    of Ground Water  Contaminated  with  Chromate and
    Chlorinated Solvents Using Zero-Valent Iron: A Field Study."
    In Preprints of Papers Presented at the 209th ACS National
    Meeting, Anaheim,  California, April 2-7,  1995, American
    Chemical Society, Division of Environmental Chemistry, Vol.
    35, No. 1, pp. 788-791.

Rouse, J.D., Sabatini, D.A., and Harwell, J.H. (1993)"Minimizing
    Surfactant Losses Using Twin-Head Anionic Surfactants in
    Subsurface Remediation",  Environmental Science and
    Technology, 27, 2072-2078.

Sabatini, D.A., Knox, R. C. and Harwell, J. H. (1995) Surfactant
    Enhanced Subsurface Remediation: Emerging Technologies,
    ACS Symposium Series, American Chemical Society, in
    press.

Sasaki, K. J., Burnett,  S. L, Christian, S. D., Tucker, E. E. and
    Scamehorn, J.  F. (1989). "Polyelectrolyte Ultrafiltration of
    Multivalent Ions. Removal of Cu(ll) by Sodium Polystyrenesul-
    fonate."  Langmuir, 5, 363.

Shiau, B.J., Sabatini, D.A.,  and Harwell, J.H. (1994) "Surfactant
    Properties Affecting  the Performance of Subsurface
    Chlorinated Solvents Remediation",  In Preparation.

Shimamoto, T., and Mima, H. (1979) "Effect of Polyols on the
    Interaction of p-Hydroxybenzoic  Acid Esters  with
    Polyoxyethylene Dodecyl Ether", Chem. Pharm. Bull., 27,
    2602-2617.

Starr, R.C. and Cherry, J.A., 1994, "In  Situ Remediation of
    Contaminated Ground Water: The Funnel-and-Gate System",
    Ground Water,  32(3), 465-476.

Tucker, E.E., Christian, S.D., Scamehorn, J.F., Uchiyama, H.,
    and Guo, W. (1992) "Removal of Chromate from Aqueous
    Streams by Ultrafiltration and Precipitation", In Transport
    and  Remediation  of Subsurface  Contaminants,  ACS
    Symposium Series 491, D.A. Sabatini and R.C. Knox,  Eds.,
    Washington, D.C.

West, C.C. and Harwell, J.H. (1992)"Surfactants and Subsurface
    Remediation", Environmental Science and Technology, 26,
    2324-2330.

West, C.C.  (1992) "Surfactant-Enhanced  Solubilization  of
    Tetrachloroethylene and  Degradation Products in  Pump
    and Treat Remediation", In Transport and Remediation of
    Subsurface Contaminants, D.A. Sabatini and R.C.  Knox
    Eds.,  ACS Symposium  Series 491, American Chemical
    Society, Washington, D.C., 149-158.

Zachara, J.M., Ainsworth, C.C., Cowan, C.E., and Resch, C.T.
    (1989)" Adsorption of Chromate by Subsurface Soil Horizons",
    Soil Science Society of America Journal, 53, 418-428.
                                                         16

-------