NC0000396_App E GW Remediation EPRI 2006_20160219

ELECTRIC POWER
■ -- , f=01 I RESEARCH INSTITUTE
Effective December 6, 2006, this report has been made publicly
available in accordance with Section 734.3(b)(3) and published in
accordance with Section 734.7 of the U.S. Export Administration
Regulations. As a result of this publication, this report is subject
to only copyright protection and does not require any license
agreement from EPRI. This notice supersedes the export control
restrictions and any proprietary licensed material notices
embedded in the document prior to publication.
Groundwater Remediation of Inorganic
Constituents at Coal Combustion Product
Management Sites
Overview of Technologies, Focusing on Permeable Reactive
Barriers
Technical Report

Groundwater Remediation
of Inorganic Constituents
at Coal Combustion
Product Management Sites
Overview of Technologies, Focusing
on Permeable Reactive Barriers
1012584
Final Report, October 2006
EPRI Project Manager
K. Ladwig
ELECTRIC POWER RESEARCH INSTITUTE
3420 Hillview Avenue, Palo Alto, California 94304-1338 • PO Box 10412, Palo Alto, California 94303-0813 • USA
800.313 3774 • 650 855 2121 • askepri®epri.com • www.epri corn

DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN
ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH
INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE
ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM:
(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (1)
WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR
SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS
FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR
INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL
PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USER'S
CIRCUMSTANCE; OR
(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER
(INCLUDING ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE
HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR
SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD,
PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT.
ORGANIZATION(S) THAT PREPARED THIS DOCUMENT
Natural Resource Technology, Inc.
Southern Company Generation
Powell and Associates
NOTE
For further information about EPRI, call the EPRI Customer Assistance Center at 800.313.3774 or
e-mail [e-mail].
Electric Power Research Institute and EPRI are registered service marks of the Electric Power
Research Institute, Inc.
Copyright @ 2006 Electric Power Research Institute, Inc. All rights reserved.

CITATIONS
This report was prepared by
Natural Resource Technology, Inc.
23713 W Paul Road, Suite D
Pewaukee, WI 53072
Principal Investigator
B. Hensel
Southern Company Generation
Earth Science and Environmental Engineering
42 Inverness Center Parkway, Bin B426
Birmingham, AL 35242
Principal Investigator
J. Pugh
Powell and Associates
140 Pleasant Lake Drive
Waterford, MI 48327
Principal Investigator
R. Powell
This report describes research sponsored by the Electric Power Research Institute (EPRI),
The report is a corporate document that should be cited in the literature in the following manner:
Groundwater Remediation of Inorganic Constituents at Coal Combustion Product Management
Sites: Overview of Technologies, Focusing on Permeable Reactive Barriers. EPRI, Palo Alto,
CA: 2006. 1012584.
iii

REPORT SUMMARY
This report reviews constituents that potentially may trigger groundwater remediation at coal
combustion product (CCP) management sites and briefly summarizes various in situ and ex situ
remediation technologies and their applicability to treat these constituents. The report provides a
more detailed discussion for one potentially promising in situ remediation technology, permeable
reactive barriers (PRBs).
Background
Relatively few instances of significant groundwater contamination at CCP disposal sites have
required active remediation, and new designs make it unlikely that sites developed since about
the mid-1990s will encounter problems. However, many older and closed sites pre -date current
designs, and it is important that the industry develop remediation technologies to address
groundwater issues if they arise.
Objectives
• To identify constituents most likely to trigger a groundwater remedial action at CCP
management sites.
• To present an overview of remediation technologies for treating these constituents,
focusing on in situ remediation in general and permeable reactive barriers in particular.
Approach
The project team identified potential constituents based on their concentrations in CCP leachate
as listed in EPRI field and laboratory databases; their mobility as compiled by USEPA, EPRI,
and other sources; their historical record in triggering remedial actions at CCP management sites;
and interviews with utility environmental managers. The team then conducted a literature review
to summarize technologies that may be applicable to remediating these constituents.
Results
Five primary constituents were identified: arsenic, boron, chromium, selenium, and sulfate.
These constituents have 90th percentile concentrations greater than state or federal groundwater
quality standards, can be mobile in groundwater under certain conditions, and have each
triggered at least one known remedial action at a CCP management site. While these constituents
are the focus of this technology review, it is important to note that they will not necessarily be
present in leachate or groundwater at all CCP sites and that other inorganic constituents may
require remediation at some sites.
Since all the constituents are inorganic and not subject to decay, remediation must focus on
processes that will remove them from solution or cause them to become immobile in the
subsurface. Ex situ and in situ processes summarized in the report include precipitation,
v

adsorption, membrane filtration, ion exchange, electrolysis, and phytoremediation. The
effectiveness of these processes vary by constituent and environment. For example, arsenic can
be treated using several different processes while boron, a key indicator of coal ash leachate, is
only effectively removed from water by ion exchange and membrane filtration, with the latter
only effective at high pH.
The report focuses on permeable reactive barriers as an in situ remediation technology for
groundwater at CCP management sites. A PRB is a subsurface wall, gate, or container filled with
a reactive media. As groundwater passes through the PRB under natural gradients, dissolved
constituents in the groundwater react with the media and are immobilized. A variety of media
have been used or proposed for use in PRBs. The most commonly cited medium is zero-valent
iron (ZVI), which has been used to treat a wide variety of contaminants, including four of the
five constituents identified for CCP sites —arsenic, chromium, selenium, and sulfate. Alternative
media, or a combination of media, are needed to treat boron. The only documented PRB project
at a CCP site, a pilot -scale test at a site in Canada, used a combination media consisting of ZVI,
organic matter, and a boron -selective ion exchange resin to treat the groundwater. Results
showed that the combination of media had potential for treating groundwater at CCP sites.
Permeable reactive barriers require further development for application at sites where the
remedial objectives call for cleanup of the mixture of constituents typically present at CCP sites.
A considerable amount of work has been done on arsenic, chromium, and cationic metals; but
more research is needed on difficult to treat constituents such as boron, sulfate, molybdenum,
and antimony and on the unique interactions that may occur in the CCP leachate matrix. Future
research is recommended on removal effectiveness, long-term performance, and the cost of
reactive media; innovative physical PRB configurations; and continued geochemical
characterization of potential constituents of concern at CCP sites.
EPRI Perspective
Regulatory guidelines expected within the next few years will result in increased monitoring
at both new and existing CCP sites. EPRI's groundwater remediation research, along with risk
assessment studies currently underway, provides a framework for addressing groundwater issues
when they arise. Additional bench- and pilot -scale research is planned to identify the most
appropriate media to treat the constituents most commonly found at CCP sites. Other in situ
approaches, including in situ fixation, will also be evaluated.
Keywords
Remediation
Groundwater
Arsenic
Boron
Chromium
Selenium
Sulfate
Permeable Reactive Barrier
vi

CONTENTS
1 INTRODUCTION ......................................... ................. .......................................................... 1-1
2IDENTIFICATION OF POTENTIAL CONSTITUENTS OF CONCERN..................................2-1
Coal Combustion Product Production and Reuse.................................................................2-1
Potential Constituents of Concern at CCP Management Sites ............................. ........... .....2-1
LeachateCharacteristics.................................................................................................. 2-2
Mobility.............................................................................................................................2-5
UtilityInterviews................................................................................................................2-7
Remediation Case Studies at CCP Sites.........................................................................2-7
Compilationof PCOCs.....................................................................................................2-9
3 OVERVIEW OF GROUNDWATER REMEDIATION TECHNOLOGIES.................................3-1
Treatment/Removal Processes for Dissolved Inorganic Constituents..................................3-1
Precipitation/Co-Precipitation...........................................................................................3-1
Adsorption........................................................................................................................3-1
MembraneFiltration..........................................................................................................3-2
IonExchange....................................................................................................................3-2
Electrolysis...................... ......................... .......... ............................................................. 3-2
Uptake by Plants/Phytoremediation.............................................................. ...............3-3
Constituent Properties Affecting Remediation............................................. .......................... 3-3
Arsenic............................................................................................... ........................... 3-4
Boron......................... .................................... ................................................................... 3-4
Chromium.........................................................................................................................3-4
Selenium.......................................................................................................................... 3-4
Sulfate.... .................................... ........ ......... ................ ............ ................. ........................ 3-5
Remediation Technology Descriptions..................................................................................3-5
SourceControl..................................................................................................................3-5
Capping......................................................................................................................3-6
vii

Removal/Excavation....................................................................................................3-6
Barriers...................................................... ............. ..................................................... 3-6
Chemical in Situ Source Treatments .................................. ......................................... 3-7
Stabilization/Solidification................................................ ............................................ 3-8
Ex Situ Groundwater Remediation...................................................................................3-8
GroundwaterExtraction...............................................................................................3-9
Interception/Drainage Trenches.................................................................................3-11
In Situ Groundwater Remediation..................................................................................3-11
Bioremediation.......................................................................................................3-11
Phytoremediation...................................................................................................3-12
Electrokinetics............................................................................................................3-13
Chemical Injection to Promote in Situ Immobilization................................................3-14
Permeable Reactive Barriers. .................................................................................... 3-14
Monitored Natural Attenuation ............................ ....................................................... 3-16
Groundwater Remediation Alternatives at CCP Sites.........................................................3-17
4 PERMEABLE REACTIVE BARRIERS...................................................................................4-1
Introduction...........................................................................................................................4-1
IntellectualProperty..........................................................................................................4-2
Designand Construction.......................................................................................................4-2
DesignCriteria..................................................................................................................4-2
Site Characterization Considerations for the Installation of a PRB..................................4-2
PRBConfigurations..........................................................................................................4-3
Continuous................................................................................................................... 4-3
Funnel-and-Gate.......................................................................................................... 4-4
InSitu Reactive Vessels.......................................................................................... 4-4
ReactiveMedia............................................................... .................................................. 4-5
Chemical Mobility Controls ........................... ............ ............................................ ....... 4-6
ReactiveMedia Summary............................................................................................4-8
TreatabilityTesting...................................................... ...... ............................................. 4-13
Construction Methods............................................................ ............................... .......... 4-14
ExcavatedPRBs.................................................... .................................................... 4-14
Direct Placement PRBs ...........................................................................................4-15
PRBCosts...........................................................................................................................4-15
CapitalCosts..................................................................................................................4-15

Operationand Maintenance........................................,..................................................4-17
Long -Term Performance..................................................................................................4-18
Monitoringof PRBs................... ............ .............................................................. ............ 4-18
PRB Longevity and Maintenance...................................................................................4-19
Importance of Iron Metal Corrosion...............................................,......,....................4-19
MicrobialActivity., .... ................................................................................................. 4-20
Combining PRBs with Monitored Natural Attenuation— ...................... ............................... 4-21
Applicability of PRBs for Remediation of Groundwater at Coal Combustion Product
ManagementSites..............................................................................................................4-22
5 RESEARCH ISSUES AND RECOMMENDATIONS ................. ............................................. 5-1
PRB Research Needs and Recommendations for Application at CCP Sites ........................5-1
6 REFERENCES.......................................................................................................................6-1
A PRB REACTIVE MEDIA DESCRIPTIONS............................................................................
A-1
Field Tested Reactive Media................................................................................................
A-1
Zero-Valent Iron................................................................................................,............
A-1
OrganicMatter.................................................................................................................
A-1
Phosphate -Based Precipitation.......................................................................................
A-2
Limestoneand Hydrated Lime.........................................................................................
A-2
Zeolites and Surfactant -Modified Zeolites.......................................................................
A-3
BasicOxygen Furnace Slag............................................................................................
A-3
Sodium Dithionite (NaS2Oj and Polysulfide Compounds .................................................
A-3
BaUXSOITM, ViromineTM, Acid-B Extra TM ...........................................................................
A-4
Other Proposed Reactive Media for Inorganic Constituents ................................................
A-4
HumasorbTM....................................................................................................................
A-4
AmorphousFerric Oxide................................................................................. .............
A-4
DissolvedOxygen Barriers..............................................................................................
A-4
DiatomaceousEarth........................................................................................................
A-5
ActivatedAlumina............................................................................................................
A-5
IonExchange Resins.......................................................................................................
A-5
ADSORBIATM (Titanium Oxide -Based Adsorbent)..........................................................
A-5
SORBPLUSTm Adsorbent (Mg -AI oxide)..........................................................................
A-5
FerrousSulfate................................................................................................................
A-6
FORAGERTM Sponge.....................................................................................................
A-6
ix

RareEarth Elements....................................................................................................... A-6
KanchanTmArsenic Filter................................................................................................. A-7
GranularFerric Hydroxide®............................................................................................. A-7
BPRB CASE STUDIES............................................................................................................ B-1
CCP Landfill, Ontario, Canada..............................................................................................
B-1
Former DOE Mill Site, Monticello, UT...................................................................................
B-2
Savannah River Site TNX Area, Aiken, SC..........................................................................
B-2
Haardkrom Site, Kolding, Denmark......................................................................................
B-3
Y-12 Site, Oak Ridge National Laboratory, Oak Ridge, TN .................................................
B-3
U.S. Coast Guard Support Center, Elizabeth City, NC ........................................................
B-4
FryCanyon Site, Fry Canyon, UT........................................................................................
B-4
BodoCanyon, Durango, CO................................................................................................
B-5
Rocky Flats Environmental Technology Site (Solar Ponds Plume), Golden, CO .................
B-5
Nickel Rim Mine Site, Sudbury, Ontario, Canada................................................................
B-6
Tonolli Superfund Site, Nesquehoning, PA..........................................................................
B-6
Public School, Langton, Ontario, Canada............................................................................ B-7
Chalk River Laboratories, Ontario, Canada......................................................................... B-7
Large Experimental Aquifer Program (LEAP) Demonstration Facility, Portland, OR........... B-8
DuPontSite, East Chicago, IN............................................................................................. B-8
GiltEdge Mine, SD............................................................................................................... B-8
100 D Area, Hanford Site, Benton County, WA.................................................................... B-9
Success Mine and Mill, Wallace, ID...................................................................................
B-10
Cyprus AMAX Minerals Company/AMAX Realty Development, Inc., Carteret, NJ ............
B-10
E.I. DuPont, Newport Superfund Site, DE..........................................................................
B-11
Universal Forest Products, Inc., Granger, IN.....................................................................
B-11
Cotter Corporation Uranium Mill, Canon City, CO..............................................................
B-12
Columbia Nitrogen Site, Charleston, SC............................................................................
B-12
x

LIST OF FIGURES
Figure 2-1 Constituent Velocity Relative to Groundwater Velocity as a Function of the
Distribution Coefficient.......................................................................................................2-7
Figure 3-1 Concentration Plot Showing Decreasing Effectiveness of Groundwater
Extraction at CCP Case Study Site 8 (from EPRI, 2001 b)................. ............ .................. 3-10
Figure 4-1 Continuous Reactive Barrier (from USEPA, 1998)....................... ............................ 4-3
Figure 4-2 Funnel -and -Gate System (from USEPA, 1998)........................................................4-4
Figure 4-3 In Situ Reactive Cell Design (from ITRC, 2005).......................................................4-5
Xi

LIST OF TABLES
Table 2-1 Coal Combustion Product Production, Management, and Use for 2004 ...................2-1
Table 2-2 Summary of Representative Laboratory and Field CCP Leachate Data ...................2-3
Table2-3 Mobility Assessment..................................................................................................2-6
Table 2-4 CCP Management Sites Requiring Remedial Action .................................... .............
2-8
Table 2-5 Potential Remediation Constituents in CCP Leachate............................................2-10
Table 3-1 Potential Constituents of Concern and Removal Processes.....................................3-3
Table 4-1 Major Chemical Attenuation Mechanisms for Constituents in CCPs .........................4-6
Table 4-2 Immobilization Mechanisms of Potential Reactive Media..........................................4-7
Table 4-3 Probable Aqueous Species in Pure Water and with Common Complexes................4-8
Table 4-4 Potentially Important Solubility and Sorption Controls...............................................4-9
Table 4-5 Field -Tested PRB Reactive Media for Inorganic Constituents.................................4-10
Table 4-6 Other Potential PRB Reactive Media for Inorganic Constituents.............................4-12
Table 4-7 Potential Reactive Media by Constituent.................................................................4-13
Table 4-8 Major Capital Costs Associated with PRBs (After USEPA, 2002b) .........................4-16
Table 4-9 Mineral Precipitates in Zero-Valent Iron PRBs (from USEPA, 2003).......................4-20

1
INTRODUCTION
There have been relatively few instances of significant groundwater contamination at coal ash
disposal sites, and new designs make it unlikely that sites developed since about the mid-1990s
will encounter problems. However, because of the large number of older and closed sites pre-
dating current designs, it is important that the industry has technologies available to address such
needs should they arise.
This report addresses remediation technologies for the suite of inorganic constituents typically
associated with coal combustion products (CCPs). Information is presented on these constituents
related to their presence in CCPs, the likelihood they will be released to groundwater, and
their fates in a subsurface system. The chemical properties of these constituents affecting
their remediation are addressed, followed by a discussion of remediation systems that can use
these properties to enhance removal. Physical removal techniques are briefly discussed, with
regard to source elimination. Both ex situ and in situ groundwater remediation technologies are
considered but the emphasis is on in situ approaches due to the known difficulties of efficiently
and completely extracting groundwater contaminants for surface treatment. In particular, one in
situ remediation approach that holds promise for a wide range of potential constituents is the use
of permeable reactive barriers (PRBs).
Section 2 of this report provides a description of typical constituents found at CCP sites, and
identifies those that are most likely to lead to a groundwater remediation action. Section 3
provides an overview of remediation technologies and their applicability to inorganic
constituents of interest. In Section 4, PRBs are discussed in more detail; descriptions of the
media used in PRBs are included in Appendix A, and summaries of several case studies of the
application of PRBs to inorganic contaminants are included in Appendix B. Section 5 provides
recommendations for research specific to the constituents identified as most important at CCPs.
1-1

2
IDENTIFICATION OF POTENTIAL CONSTITUENTS OF
CONCERN
Coal Combustion Product Production and Reuse
The term coal combustion products refers to fly ash, bottom ash, boiler slag, and flue gas
desulfurization (FGD) solids (Table 2-1). These materials are managed using a variety of
methods including beneficial reuse, dry -management in landfills, and wet -management in
impoundments. Bottom ash and boiler slag have a relatively low leachability and are widely used
in a variety of use applications, including roadbeds, structural fills, blasting grit, and roofing
granules, that are unlikely to result in groundwater contamination. Similarly, beneficial use of fly
ash and FGD gypsum typically involves incorporation of the CCP into low -leachability materials
such as concrete, wallboard, or stabilized soils with little potential to affect groundwater.
The focus of this report is therefore on fly ash and FGD solids management in landfills
and impoundments.
Table 2-1
Coal Combustion Product Production, Management, and Use for 2004
Fly Ash
Bottom Ash
Boiler Slag
FGD Products
Total Production
70.8
17.2
2.2
32.3
Percent Reused 40%
47%
90%
34%
Source: American Coal Ash Association, 2004
z Production values in million short tons
Potential Constituents of Concern at CCP Management Sites
Many remediation technologies are designed to treat specific constituents. As a result, evaluation
of remediation technologies requires knowledge of the potential constituents of concern (PCOC)
in groundwater that may require treatment.
Most naturally occurring elements can be found in CCPs, but only a small subset of these
elements and their compounds are found in leachate at concentrations high enough to be
environmentally significant, and only a subgroup of these constituents has sufficient mobility
in groundwater to potentially trigger a remedial action. Therefore, a multi -tiered approach was
used to identify PCOCs at CCP management sites:
2-1

Identification. of Potential Constituents of Concern
• Leachate data from a broad range of power plants and CCP management sites were compared
to water quality standards to quantitatively determine constituents with the potential to leach
to groundwater in significant concentrations.
• Generalized mobility data from USEPA and EPRI studies were reviewed to determine
constituents that may be mobile in groundwater.
• Case studies of CCP remediation sites were examined to identify constituents that have
triggered remedial actions.
This information was supplemented with qualitative interviews in which utility environmental
managers were queried about constituents of concern to their operations.
Leachate Characteristics
In the event of a Leachate release to groundwater, the potential for an exceedance of a
groundwater quality standard is a function of the constituent's concentration in leachate relative
to the relevant state groundwater quality standard, its mobility in groundwater (discussed later),
and the hydrogeologic and chemical environment, which is not discussed because it is too
site -specific for this broad -based assessment. Leachate quality was summarized based on two
comprehensive studies covering a range of power plant and CCP management site conditions
(Table 2-2):
• EPRI (1987b) lists results of hot-water leachate extracts on 94 unweathered CCP samples
collected at 39 power plants.
• EPRI (2006a) reports on 81 field leachate samples collected from 29 CCP management sites.
Each study used consistent sample collection methods, and each study used a single laboratory
for analysis, thereby eliminating these two possible causes of external variability.
The data were then compared to the federal maximum contaminant levels (MCLs) or secondary
maximum contaminant levels (SMCLs), which are often used as the basis for state groundwater
standards. For constituents with no MCL or SMCL, a representative state groundwater standard
was used. A ratio less than 1.0 suggests that there is little likelihood of an MCL exceedance in
groundwater for a given constituent in the event of a leachate release. A ratio greater than 1.0
suggests that an MCL exceedance is possible, but other factors, such as the hydrogeologic
system and mobility of the constituent, must be considered.
The leachate data indicate that concentrations of antimony, arsenic, cadmium, chromium,
selenium, and thallium were higher than health -based MCLs in at least 10 percent of the
samples.' In addition, the 90" percentile concentrations of boron, lithium, manganese,
molybdenum, sodium, sulfate, and vanadium were higher than alternative drinking water criteria.
These constituents are more likely to trigger a remedial action in the event of a leachate release
than constituents that typically have leachate concentrations lower than drinking water standards.
' i.e_, those samples with concentrations at or greater than the 90"' percentile.
2-2

Identification of Potential Constituents of Concern
Table 2-2
Summary of Representative Laboratory and Field CCP Leachate Data
a. Constituents with Health -Based Maximum Contaminant Levels (MCLs)
Parameter'
Source 2 !Count
Field
%BDL
Median
901h Percentile'
Maximum
Standard
Ratio'
Antimony
81
7%
0.002
0.020
0.059
0.752
0.006
3.3
Lab
94
67%
BDL
0.114
0.178
0.006
19.0
Arsenic
Field
81
0%
0.026
1.380
0.010
17.8
Lab
94
62%
BDL
0.340
14.040
0.010
34.0
Barium
Field
81
5%
0.089
0.250
0.657
2.000
0.1
Lab
94
0%
0.160
0.677
2.990
2.000
0.3
Beryllium
Field
81
94%
BDL
BDL
0.009
0.004
NC
Lab
0
Cadmium
Field
81
10%
0.002
0.013
0.065
0.005
2.6
Lab
94
73%
BDL
0.012
0.792
0.005
2.3
Chromium
Field
81
48%
0.001
0.025
5.100
0.100
0.2
Lab
94
81
46%
19%
0.014
0.212
5.320
0.100
2.1
Copper
Field
Lab
0.003
0.021
0.494
1.300
<0.1
94
64%
BDL
0.037
61.600
1.300
<0.1
Fluoride
Field
0
Lab
Field
Lab
94
10%
0.163
1.312
8.850
4.000
0.3
Lead
81
94
72%
BDL
0.0004
0.0080
0.0150
<0.1
96%
BDL
BDL
3.7600
0.0150
NC
Mercury
Field
30
0%
0.000004
0.000029
0.000079
0.002000
<0.1
Lab
0
Nitrate
Field
0
Lab
90
9%
0.100
0.682
20.000
10.000
<0.1
Nitrite
Field
0
Lab
94
27%
0.015
0.077
5.800
2.360
1.000
<0.1
Selenium
Field
81
0%
0.018
0.181
0.050
3.6
Lab
0
Thallium
Field
81
53%
BDL
0.005
0.018
0.002
2.6
Lab
0
1. All concentrations in mg/L.
2. Field = EPRI (2006a) field leachate samples; Lab = EPRI (1987) hot water extracts.
3. Ratio is the 901h percentile divided by the standard.
4. NC indicates that the ratio was not calculated because the 90"' percentile was below laboratory detection limits
5. BDL indicates that the median or W percentile was below laboratory detection limits.
2-3

Identification of Potential Constituents of Concern
Table 2-2
Summary of Representative Laboratory and Field CCP Leachate Data (Continued)
b. Constituents with Non -Health Based SMCLs or State Standards
Parameter'
Boron
Source 2
Count
81
%BDL
Median
90'h Percentile
14.0
Maximum
112.0
Alt. Standard3
Ratio'
28.0
Field
0%
2.6
0.5-2.0
Lab
94
5%
1.4
7.8
82.4
0.5-2.0
15.6
Chloride
Field
80
0%
28
74
2,330
200-250
0.4
Lab
94
0%
1
3
517
200-250
<0.1
Iron
Field
81
56%
BDL
0.05
25.60
0.3-5.0
0.2
Lab
94
17%
0.01
0.03
39.40
0.3-5.0
I 0.1
Lithium
Field
81
14%
0.15
0.43
23.60
0.17
2.5
Lab
94
80
0%
0.20
0.51
8.68
0.17
3.0
<0.1
Magnesium
Field
9%
13
34
5,810
400
Lab
94
13%
1
3
143
400
<0.1
Manganese
Field
81
20%
0.060
0.202
4.170
0.05-0.15
4.0
Lab
94
10%
0.004
0.043
3.080
0.05-0.15
0.9
Molybdenum
Field
81
2%
0.36
1.39
60.80
0.035-0.073
39.7
Lab
94
27%
0.06
0.22
1.86
0.035-0.073
6.3
Nickel
Field
81
17%
0.005
0.014
0.011
0.597
0.1
0.1
Lab
94
73%
BDL
8.520
0.1
0.1
Sodium
Field
80
0%
58
312
4,630
120
2.6
Lab
94
0%
6
19
2,008
120
0.2
Strontium
Field
81
1 %
1.2
3.9
16.9
4.6
0.8
Lab
94
0%
0.8
2.5
23.7
4.6
0.6
Sulfate
Field
80
0%
485
1613
30,500
250-400
6.5
Lab
94
0%
198
518
4,600
250-400
2.1
Vanadium
Field
81
6%
0.03
0.01
0.16
5.02
0.0045
35.6
Zinc
Lab
0
48%
73%
0.29
121.20
Field
81
94
0.02
0.01
5.0
<0.1
Lab
BDL
5.0
<0.1
1. All concentrations in mg/L.
2. Field = EPRI (2006a) field leachate samples; Lab = EPRI (1987) hot water extracts.
3. The range of alternative standards is based on non -health based Federal SMCLs (italicized) and representative state drinking
water standards and groundwater clean-up criteria (based on California, Illinois, Michigan, and Wisconsin).
4. Ratio is the 901h percentile divided by the lowest alternative standard.
2-4

Identification of Potential Constituents of Concern
Mobility
Chloride and sulfate are highly mobile in most natural groundwater environments. However,
the mobility of most other inorganic constituents is dependent on site -specific conditions such
as soil type (clay versus sand), redox environment (reducing versus oxidizing), pH, and the
concentration of the constituent in groundwater. As a result, a constituent that may be mobile
in groundwater at one site may be immobile at another site.
Mobility often is quantified by the linear distribution coefficient (K), which is a measure of the
mass of a constituent dissolved in solution to the mass of constituent attached to the soil or rock
matrix. When Kd is zero, the constituent migrates at the rate of groundwater flow; when K, is
greater than zero, the constituent interacts with the soil/rock matrix and migrates at a rate slower
than the rate of groundwater flow. Therefore, the rate of migration, or mobility, of a constituent
decreases as the Kd value increases. While many constituents have non -linear adsorption
isotherms, linear K, provides a useful comparison of relative mobility. For this assessment,
ranges of linear K, were compiled (Table 2-3) from the following sources:
• USEPA (1992): published ranges of K, values for three pH conditions (pH 4.9, 6.8, and 9.0).
For each constituent, the Kd range for the pH values likely to be encountered in groundwater
at ash sites (6.8 and 9.0) was selected.
• USEPA (1996): used a geochemical model to calculate soil K, values as a function of pH.
The range of K, values selected was for the pH range of 6.5 to 8.0 (the maximum pH value)
expected in groundwater at CCP sites.
• EPRI has published reports listing mobility data for arsenic (EPRI, 2004), boron (EPRI,
2005a), and selenium (EPRI, 2006b). For boron, K, is a function of concentration, and the
range selected is for concentrations lower than 10 mg/L, which was the first reported
concentration range below the 90`h percentile boron concentration in CCP leachate. The
arsenic and selenium ranges were based on field conditions reproduced for three CCP sites.
For purposes of evaluating inorganic constituents of interest in this study, a qualitative mobility
ranking system was developed based on the K, ranges in Table 2-3. A mobility ranking of high
was assigned to any constituent where the lowest K, was less than 2.0 L/kg. This value was
selected because it is the approximate K, where the constituent velocity is equal to one -tenth the
groundwater velocity, and is also the approximate inflection point (shown in Figure 2-1) beyond
which constituent velocity is very slow relative to groundwater velocity. A mobility ranking of
moderate was assigned to any constituent where the lowest Kd was greater than 2.0 L/kg and less
than 5.0 L/kg, and the overall range was less than 500 L/kg. A mobility ranking of low was
assigned to all remaining constituents, which often had K, ranges extending beyond 1,000 L/kg.
It must be stressed that the K, ranges are relatively large, and mobility can vary greatly for a
given constituent. For example, in an EPRI study at three ash management facilities, the range of
Kd values reported for arsenic is 5 to 50 L/kg for As(III), but 30 to 350 L/kg for As(V) (EPRI,
2004). This suggests that only As(III) is categorized as moderately mobile in this evaluation, and
that even this species would not be categorized as moderate much of the time. There has been a
significant amount of research on the adsorption of arsenic on pure mineral surfaces and soil
particles that suggest a much larger range of arsenic K, values (EPRI, 2000a).
2-5

Identification. of Potential Constituents of Concern
Table 2-3
Mobility Assessment
Kd (Ukg)
USEPA (1992) USEPA (1996) Other*
Potential
Mobility
Antimony
4.2 — 126
45
Moderate
Arsenic
3.5 — 275
28 — 31
5 — 350
Moderate
Barium
13 — 1,442
37 — 52
Low
Beryllium
29 — 12,848
280 — 100,000
Low
Boron
0.4 — 3.0
High
Cadmium
8.6 — 2,818
52 — 4,300
Low
Chloride
0.0
High
Chromium
0.5 — 383
14 — 20
High
Copper
27 — 18,232
Low
Lead
48 — 432
Low
Lithium
0.03 — 0.25
High
Magnesium
57 — 1025
Low
Mercury
22 — 200
Low
Molybdenum
0.6 — 500
High
Nickel
8.3 — 1,675
50 — 1,900
Low
Selenium
0.1 — 66
2.2 — 6.1
2 — 500
High
Strontium
0.33 — 0.59
High
Sulfate
0.0
High
Thallium
20 — 430
66 — 96
Low
Vanadium
1000
13 — 500
Low
Zinc
14 — 3,989
51 — 530
Low
Sources for other K, values:
Arsenic: EPRI, 2004
Boron: EPRI, 2005a
Molybdenum: USEPA, 2005a (literature search, pH range of 4 to 10 was greater than used here)
Selenium: EPRI, 2006b
Vanadium: USEPA, 2005a (literature search, pH range of 4 to 10 was greater than used here)
Chloride: Considered conservative in groundwater
Sulfate: Considered conservative in groundwater under most conditions, may not be conservative in strongly reducing
environments
Lithium and Strontium: confidential CCP site data for a sand aquifer at neutral pH (CCP Case Study Site 1, Table 2-4).
The potential mobility ranking shown in Table 2-3 indicates that boron, chloride, chromium,
lithium, molybdenum, selenium, strontium, and sulfate are likely to be mobile in groundwater
under the widest range of conditions, while arsenic and antimony may be mobile under certain
conditions. The other listed constituents are less likely to be mobile in groundwater.
2-6

V�/V9W = 1.0 when Kd = 0
0.01
Identification of Potential Constituents of Concern
0.1
V jV9W = 1/Rd = 1 + Kd * Bulk Density / Porosity
rn
1
J
Where:
Y 10 V�/V9W = Ratio of constituent velocity to groundwater velocity
Rd = Retardation coefficient
100 Kd = Distribution coefficient
Plot assumes a bulk density of 1.85 kg/L and porosity of 0.3
1000 —
1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0
Vc/Vgw
Decreasing Constituent Velocity =_>
Figure 2-1
Constituent Velocity Relative to Groundwater Velocity as a Function of the Distribution
Coefficient
Utility Interviews
Eight utility environmental managers were interviewed to determine constituents of significance
to their CCP management programs. The respondents were asked to rank the constituents in
order of significance to their sites. Twenty-two constituents were mentioned at least once, and
13 different constituents were ranked among the top three in importance. Five constituents were
ranked among the top three by at least half of the respondents: arsenic, boron, chromium,
selenium, and sulfate; although no constituent was ranked among the top three by all
respondents.
Remediation Case Studies at CCP Sites
Table 2-4 summarizes CCP management sites where groundwater remedial actions have been
performed. This list is not all-inclusive, but is representative of documented remedial actions
performed by the industry. The list includes primary parameters, meaning the parameters
that triggered remedial actions, other parameters that were prominently mentioned in site
documentation, remedial actions performed to date, and references, if publicly available.
2-7

Identification of Potential Cojistituents of Concern
Table 2-4
CCP Management Sites Requiring Remedial Action
Site
Code
Site
Type'
Primary
Parameters
B, Se, Li
Other
Parameters
Remedial Actions
Published
Reference
Confidential
1
LF
SO,, Mn, Sr
Closed; fully-encirculating barrier
wall with gradient control; soil cap
2
1
Se, B
Sb, Li
Groundwater extraction
Confidential
3
1
As, B, Li
SO,, Fe
Natural attenuation (for As); barrier
wall with gradient control
Confidential
4
1
B
SO,, Mn
Closed; soil cap
EPRI 1005165
(2001 a)
5
1
B, SO,
Closed
EPRI 1005165
(2001 a)
6
1
B, SO,
TDS, Mn,
Fe
Closed
EPRI 1005165
(2001 a)
7
1
B
SO„ Fe,
Mn, pH
Closed; portions excavated
Confidential
8
LF
B, SO,
Se
Groundwater extraction; PVC cap
EPRI 1005214
(2001 b)
9
LF
SO,, B, Mo
Se, Cr
Closed; HDPE cap
EPRI 1005262
(2002)
10
LF
B, SO,
Mo
PVC cap; alternative water supply;
excavation of small area with
saturated ash
Confidential
11
1
B, SO,
Mn, TDS
Closed; will be capped
Confidential
12
LF
B, SO,
As, Se
Interceptor trench; cap TBD
Confidential
13
LF
Se, SO,
V
Alternative water supply, soil/clay
caps, groundwater
extraction/gradient control
http://www.epa.go
v epaQswer/other/f
ossil/ nd ray. df
14
LF
B, Mo
Alternative water supply; cap TBD
hV.L)://www.epa.go
vftealai0sites/pin_
esl
15
LF
As, B, Cr,
Mo, Se, V
SO,
Permeable reactive barrier
(demonstration project)
McGregor et al.
(2002)
16
1
B, SO,
Fe, Mn,
TDS, pH
Capped with synthetic liner for new
impoundment (with leachate
collection)
Confidential
1. 1 = impoundment; LF = landfill

Identification. of Potential Constituents of Concern
All of these landfills and impoundments sites were unlined. In some cases, the leachate
chemistry may have been influenced by comanagement of pyrite with coal ash, a practice that
has since been addressed by the industry (EPRI, 1999a). Most of the sites either closed as part of
the remedial action, or were already closed or inactive. Remedial actions included the following:
• Closure of seven sites;
• Caps at nine sites (three synthetic, two soil or clay, and five yet to be determined);
• One site was "capped" by constructing a new ash impoundment with a synthetic liner and
leachate collection over the top;
• Hydraulic controls (groundwater extraction or interceptor trenches) at four sites;
• Excavation of saturated ash at one site;
• Partial excavation (for a new lined impoundment) at one site;
• Two barrier walls with hydraulic gradient control to assure inward flow of groundwater;
• Three remediation programs included provision of alternative water supplies;
• A permeable reactive barrier was constructed at one site (see PRB case study 1,
Appendix B).
The primary constituents driving remediation were boron (15 of 16 sites), sulfate (10 sites),
selenium (4 sites), arsenic (2 sites), lithium (2 sites), molybdenum (2 sites), and chromium
(1 site).
Compilation of PCOCs
A listing of leachate constituents, and an evaluation of whether or not they are PCOCs from a
remediation perspective, based on the leachate data, mobility data, and case studies, is presented
in Table 2-5. A constituent was listed as a potential remediation constituent (4) based on leachate
data if the ratio of 90`h percentile leachate concentration to MCL was greater than 1.0. Mobility
was categorized as potentially high if a K, value of 2.0 L/kg or lower was listed in Table 2-3,
or moderate if a Kd value between 2.0 and 5.0 L/kg was listed. The potential remediation
constituents identified for the case studies were those that were listed in Table 2-4 as primary
drivers for remediation.
Selenium and chromium have three unqualified check marks in Table 2-5 and are therefore
included in the list of potential constituents of concern for CCP management sites. Several other
constituents had two unqualified check marks and one qualified check mark: arsenic, boron,
lithium, molybdenum, and sulfate. Arsenic, boron, and sulfate were each listed as one of the
three most important constituents by at least half of the utility interview respondents, while
lithium and molybdenum were not prominently mentioned; therefore, arsenic, boron, and sulfate
are grouped with selenium and chromium as the primary PCOCs for CCP management sites.
Lithium, molybdenum, and antimony (the only constituent with two check marks) are considered
secondary PCOCs.
2-9

Identification. of Potential Constituents of Concern
Table 2-5
Potential Remediation Constituents in CCP Leachate
Leachate Data
Mobility
Case Studies
Antimony
Arsenic
J
(J)
J
Barium
-
-
-
Beryllium
-
-
-
Boron
(J)
J
J
Cadmium
J
-
-
Chloride
-
J
Chromium
J
J
J
Copper
-
-
-
Fluoride
-
Iron
-
**
Lead
-
-
-
Lithium
(J)
J
J
Magnesium
-
J
-
Manganese
-
Mercury
-
-
-
Molybdenum
(J)
J
J
Nickel
-
-
-
Nitrate-
Nitrite-
Selenium
J
J
J
Sodium
-
Strontium
-
J
-
Sulfate
(J)
J
J
Thallium
J
-
-
Vanadium
Zinc
-
-
-
4 denotes potential remediation constituents identified within each category.
(4) indicates that leachate concentration for this constituent exceeds a SMCL or state standard, rather than a
MCL, or that mobility is moderate.
- indicates constituents with low potential to trigger a remedial action.
— denotes constituents not rated because mobility data were not available.
2-10

3
OVERVIEW OF GROUNDWATER REMEDIATION
TECHNOLOGIES
Treatment/Removal Processes for Dissolved Inorganic Constituents
Inorganic constituents vary widely in their removal from aqueous systems by treatment
processes. In some cases, the properties that make a constituent mobile in groundwater also make
it difficult to remove using a cost-effective treatment approach. There are a variety of approaches
that are used to remove inorganic constituents from water, but only a few basic processes that are
implemented by these approaches. These include:
• Precipitation/Co-precipitation
• Adsorption
• Membrane Filtration
• Ion Exchange
• Electrolysis
• Phytoremediation
Precipitation/Co-Precipitation
Precipitation occurs when a constituent exceeds its solubility limit in water. This can occur
for a variety of reasons including a simple increase in the concentration of the constituent,
or a change in pH, Eh (oxidation/reduction potential), temperature, or ionic strength. Co -
precipitation is the removal of a contaminant by precipitation of another constituent that
is typically present in higher concentrations, usually Fe(III) or Al(III) salts (often added as
coagulants in water treatment systems). The constituent of concern is removed by trapping
within or adsorption to the precipitates as they form. Ex situ precipitation/co-precipitation via
pump and treat technologies allows separation of the aqueous phase from the precipitate with
subsequent disposal/recovery of the material. Precipitation/co-precipitation in situ can be a viable
groundwater treatment strategy provided that conditions are sufficiently stable that re -dissolution
and remobilization do not occur.
Adsorption
Adsorption is a surface -chemical phenomenon wherein accumulation of a constituent occurs at
the interface between the aqueous phase and solid materials in contact with the aqueous phase.
This process can also be used in treatment systems implementing a variety of substrates as the
solid phase (zeolites, granular activated carbon, etc.), whether in situ or ex situ.
3-1

Overview of Groundwater Remediation Technologies
Adsorption of inorganic contaminants and metals involves either complexation of the
contaminants on the surfaces by bond formation or electrical (coulombic) interactions between
the contaminant and the surfaces. Adsorption can be either relatively weak (reversible) or
relatively strong (often irreversible) depending on whether outer -sphere or inner -sphere
complexation has occurred, respectively. Outer -sphere complexes involve electrostatic attraction
whereas inner sphere complexes have covalent bonding and ionic bonding characteristics.
Therefore, outer -sphere bonds are less stable than inner -sphere bonds. The chemistry of
adsorption is very complex and beyond the scope of this document except to say that pH,
ionic strength, and other chemical aspects of the system can affect adsorption.
Membrane Filtration
Membrane filtration is a physical process that uses a semi -permeable membrane to separate
the contaminants. The water is passed through the membrane via pressure, which allows low
molecular weight chemicals to pass and blocks contaminants with high molecular weight.
Various membrane filtration techniques can be used to meet very distinct liquid separations.
Examples of membrane filtration processes include microfiltration, ultrafiltration, nanofiltration,
and reverse osmosis. Reverse osmosis is the most effective membrane filtration process and
results in water with very low total dissolved solids concentrations. The typical membranes
used for reverse osmosis are synthetic (TFC - thin film composite) or organic (CTA - cellulose
triacetate) and have pores that are less than 0.001 micron in size. The reverse osmosis process is
known to remove dissolved inorganic compounds such as sulfate, arsenic, boron (at high pH),
chromium, and selenium. Given current technology, membrane filtration is not amenable to in
situ remediation.
Ion Exchange
Ion exchange is a common water treatment and is widely used in homes to eliminate problems
with hard water and other trace constituents. It is the replacement of one adsorbed, readily
exchangeable ion on the surface of a supporting substrate with another ion. This involves the
concept of outer -sphere complexes discussed previously for adsorption. When the exchange
medium becomes saturated with the ions that are being removed from the system, the medium
can be replenished and reactivated by soaking it in a concentrated solution of K+ or Na', thus
desorbing the Ca``, Mg", or other constituents to a waste stream. The relatively low cost and
simplicity of ion exchange processes has made this approach a popular choice for ex situ
treatment of water. In situ usage in groundwater is less well developed, partially due to the
fact that the predominant concentrations of major dissolved ions, such as Ca", can out -compete
the contaminants for adsorption sites because the contaminants have much lower concentrations.
The adsorption of major ions can deplete the ion -exchange capacity of some media; however
recently developed adsorbents are more constituent -specific.
Electrolysis
Electrolysis is the application of an electric current through a liquid using at least two electrodes,
one of which is positively charged, and the other of which is negatively charged. The current
results in the migration of ions in the liquid to the electrode of opposite charge. Metallic ions and
metalloids will precipitate or "plate out" onto the surfaces, or in the vicinity of the surface, of the
3-2

Overview of Groundwater Remediation Technologies
electrode. Electrolysis has been a classical technique in chemistry laboratories but such
separation of trace metals from the aqueous phase tends to function most efficiently in
relatively pure, low ionic strength solutions. In groundwater systems, the presence of fairly high
concentrations of dissolved ionic species, which will themselves migrate to and plate out on the
electrodes, can cause passivation of the electrodes and rapid decreases in the removal efficiency
of the contaminants of interest. Additionally, these approaches have been very difficult and
expensive to implement and maintain in situ.
Uptake by Plants/Phytoremediation
Phytoremediation is a process that uses vegetation to remove, transfer, stabilize, or destroy
contaminants in groundwater (or soil and sediment). Various plants and trees are used for this
process depending on the type of phytoremediation application, the contaminants of concern, the
depth of the contaminants to be addressed, and the media to be addressed. Dissolved inorganics
can be treated by adsorption or precipitation onto plant roots or absorption into plant roots. The
constituent may remain on the root, within the root, or be taken up and translocated into other
portions of the plant, depending on the constituent, its concentration, and the plant species, for
example, Cr(III) and As(V) can sorb to plant roots (GWRTAC, 1997a). Phytoremediation of
dissolved inorganics is generally applicable at relatively low concentrations.
Constituent Properties Affecting Remediation
Properties of the most significant PCOCs for remediation of CCP sites (arsenic, chromium,
selenium, boron, and sulfate) are described in this section, and the chemical treatment/removal
processes effective on each are summarized in Table 3-1.
Table 3-1
Potential Constituents of Concern and Removal Processes
Constituent
Removal Processes
Arsenic
Adsorption, best with As(V)
Reductive Precipitation
Co -Precipitation
Ion Exchange
Membrane Filtration
Chromium
Adsorption, best with Cr(III)
Reductive Precipitation
Co -Precipitation
Ion Exchange, primarily Cr(VI)
Membrane Filtration
Selenium
Adsorption
Reductive Precipitation
Ion Exchange
Membrane Filtration
Boron
Ion Exchange
Membrane Filtration (at high pH)
Sulfate
Adsorption
Reductive Precipitation
Ion Exchange
Membrane Filtration
3-3

Overview of Groundwater Renediation Technologies
Arsenic
Arsenic is a metalloid that is very redox-labile and, as such, can exist in several different
valence states (-3, 0, +3, +5) and a wide variety of species in the environment. In natural aqueous
systems, such as surface and ground waters, arsenic occurs primarily as an oxyanion, usually
as arsenate [As(V), AsO.,-"] and arsenite [As(III), AsO, °]. Arsenic concentrations in CCP field
leachate range from 1 to more than 1,000 µg/L (EPRI, 2006a) with a median of 26 µg/L and
a 901h percentile concentration of 178 µg/L (Table 2-2). The dominant species in CCP leachate
is usually arsenate (EPRI, 2006a). Arsenite is typically more toxic, more soluble in water, and
more mobile in groundwater than arsenate. Arsenic can be removed from water by a variety of
processes, including adsorption, precipitation, and co -precipitation, often with conversion of
any arsenite to arsenate prior to treatment.
Boron
Boron compounds tend to be soluble, mobile in the subsurface, and difficult to remediate.
Boron has the highest concentration of minor and trace elements in coal ash leachate, ranging
from 0.2 to more than 100 mg/L (EPRI, 2006a), with a median concentration of 2.6 mg/L and
a 90`h percentile concentration of 14 mg/L (Table 2-2). Boron typically occurs as a neutral boric
acid species and is not readily removed by common water treatment procedures. EPRI (2005a)
reports that boron can be removed from water using boron -selective ion exchange and boron -
selective solvent extraction. Reverse osmosis is not effective at acidic or near -neutral pH, but
is effective at high pH (>9.24). Other removal mechanisms have been proposed, but none have
been fully developed.
Chromium
Chromium is less commonly found in CCP field leachate than the other PCOCs discussed here,
as 48 percent of the field leachate samples were below the detection limit of 0.2 µg/L. The
median field leachate concentration was 1 µg/L, and the 90`h percentile was 25 µg/L, although
concentrations as high as 5,000 µg/L were observed at one site (Table 2-2). When detected,
Cr(VI) typically predominates in CCP leachate (EPRI, 2006a). In groundwater, Cr(VI) is mobile
as an oxyanion, chromate (Cro,2'), which is also its more toxic form. Like arsenic, chromium can
be removed from water by a variety of processes (Table 3-1).
Selenium
Selenium is considered by some to be non-metallic and by others to have metallic properties
(Irwin, 1997), with behavior somewhat similar to sulfate (EPRI, 1994a). Like arsenic, selenium
is generally present in predominantly two oxyanion forms in natural waters. Se(IV) as selenite
ion SeO, `, and Se(VI) as selenate ion SeO4 Z. Selenium concentrations in CCP field leachate
range from less than 1 to more than 2,000 µg/L (EPRI, 2006a), with a median concentration of
18 µg/L and a 90`h percentile concentration of 181 µg/L (Table 2-2). Selenite tends to dominate
in impoundment settings when the source coal is bituminous or a mixture of bituminous and
subbituminous, while selenate tends to predominate in landfill settings and when the source coal
is subbituminous/lignite (EPRI, 2006a). Selenate is generally soluble and mobile, and is readily
ME

Overview of Groundwater Remediation Technologies
taken up by organisms and plants. Selenite is less soluble and mobile than selenate, therefore
reductive precipitation/co-precipitation of selenium could serve as a viable remediation
approach. However, re -oxidation is a potential problem. Phytoremediation has also been reported
and adsorption has been used.
Sulfate
Sulfate is naturally present in many minerals and its occurrence in groundwater is ubiquitous.
It also has the highest concentration of any constituent in CCP leachate (EPRI, 2006a), with a
median concentration of 485 mg/L in field leachate samples and a 90"' percentile concentration
of 1,613 mg/L (Table 2-2). Sulfate is mobile in groundwater except under strongly reducing
conditions, where it is reduced to sulfide. Sulfate is typically removed from water by
precipitation or by reductive processes. For precipitation to occur, concentrations of precipitating
cations (e.g., Ca2+, Ba21, etc.) must be present in sufficient quantity and under the right pH
conditions to lower the sulfate concentration via precipitation as BaSO4 or CaSO4. The reduction
of sulfate to elemental sulfur (S) and hydrogen sulfide (HzS) is microbially mediated and the
presence of organic matter is required. Sulfate reducing conditions do not typically occur in
surficial sand and gravel alluvial aquifers typical of those present beneath most CCP
management sites because these aquifers tend to be mildly oxic.
Remediation Technology Descriptions
There are many technologies that can be used individually or in combination to minimize the
release of contaminants into groundwater or attempt to remediate groundwater once a release has
occurred. Minimizing the potential release of contaminants to groundwater is clearly preferable
to remediating contaminated groundwater, because remediation often requires allocation of
resources for prolonged periods of time and is expensive. Approaches for minimizing the
potential for releases from CCP sites are dependent on site -specific attributes such as underlying
geology, climate, and the management method utilized (wet or dry); and can vary from dry -
stacking with an eventual evapotranspirative cap to liners with leachate collection systems.
The focus of this report, however, is on remedial technologies applicable after a release to
groundwater has occurred from a CCP management unit. Remediation technologies can be
broadly grouped into source control, ex situ remediation, and in situ remediation. In source
control, the CCP source is physically or chemically removed or constrained such that the release
cannot extend beyond the applicable boundary of the CCP management unit. Ex situ remediation
consists of removal of groundwater for treatment at the land surface, while in situ remediation
consists of in -place treatment of a groundwater plume as it migrates downgradient from the
source.
Source Control
Controlling an active source of groundwater contamination can reduce remedial costs by
shortening the time that remedial processes have to be maintained. In some cases, source control
coupled with monitored natural attenuation is all that is required at a site. Source control options
can be particularly beneficial for CCP sites die to the long leaching time of these materials;
however, the range of feasible options at CCP sites is often limited due to their large size.
3-5

Overview of Groundwater Remediation Technologies
Capping
Capping is usually performed to prevent or reduce infiltration of water into CCPs, which
subsequently reduces the volume of leachate generated. Caps can be installed on both legacy
and recently filled CCP sites. Depending on climatic conditions, designs can range from barrier
caps utilizing low permeability materials such as PVC, to evapotranspirative caps that utilize soil
sequencing and vegetation to promote runoff and evaporation of water. Caps are not effective
when CCP is filled below the water table, because groundwater flowing through the CCP will
generate leachate even in the absence of vertical infiltration through the CCP.
Caps can be effective for relatively long-term mitigation of source contaminant intrusion into the
subsurface when properly installed and maintained, and when the CCP is above the water table,
but their installation may not be sufficient as a complete remedial solution once a groundwater
plume has formed. More than half of the remediation case studies listed in Table 2-4 included
capping as a component, and in at least two cases (sites 4 and 9) the cap and closure approaches
have resulted in documented significant decreases of PCOC concentrations.
Removal/Excavation
Source removal is an effective, but often impractical, method of contaminant source control.
Due to the large areal and vertical extent of most CCP disposal sites, excavation, removal, and
alternative disposal of the materials can be either impossible or at least cost prohibitive when
risk and alternative approaches are considered. However, there are scenarios where excavation
of some material can be both technically feasible and cost effective. For example, in CCP
case study 10 (Table 2-4), review of historical aerial photos revealed that the area of greatest
groundwater impact was downgradient of a low area (pond) that existed at the site prior to filling
with CCP. Groundwater fate and transport modeling showed that excavating the saturated ash
filled in this area would enable remediation without implementation of a groundwater extraction
system, and economic analysis showed the cost of excavating this 2-acre (0.8-ha) area, even
though it was below 35 feet (11 m) of unsaturated ash, would be lower than in situ stabilization
or implementation of a long-term groundwater extraction system. The final remedial approach
therefore included: 1) removal and stockpiling of overlying unsaturated CCP; 2) excavation of
saturated CCP and refilling of the excavation with clean sand below the water table;
3) replacement of the unsaturated CCP; 4) placement of a PVC cap; and 5) provision of
alternative groundwater supplies to users within the plume area.
Barriers
When excavation or removal is impractical and the source materials are in the groundwater
or infiltration from these materials cannot be mitigated, it is sometimes possible to contain the
source by the implementation of physical or hydraulic barriers. Barriers are installed to either
surround the source or to divert groundwater around and away from the source.
Physical barriers are constructed from a variety of materials (cement, bentonite, sheet piling) and
installed using widely differing techniques (jetting, trench and fill, pile driving). In order to be
effective, there must be a low -permeability lower confining layer into which the barrier can be
3-6

Overview of Groundwater Rernediation Technologies
keyed, and it must be at a technically feasible depth. Furthermore, if the barrier fully surrounds
the site, some form of gradient control is typically needed to avoid a build-up of head within the
barrier. However, since an inward gradient can be maintained within the wall at relatively low
pumpage rates, it may be possible to close such a site with a simple soil cap, which results in
savings in short term construction costs, relative to an engineered cap, and potentially enables
more options for future site development. A remediation approach utilizing this technology was
used at CCP case study site 1 (Table 2-4).
Hydraulic barriers are used to contain leachate discharging from a site or to change hydraulic
gradients such that the plume can be managed on site. Containment is accomplished by
groundwater extraction wells or trenches that intercept all groundwater flowing from a CCP unit.
Depending on the concentration of PCOCs in the extracted groundwater, it may be possible to
route the extracted groundwater to an existing treatment system, either for plant production
water, as was done at CCP case study site 2 (Table 2-4), or to a sanitary sewer.
Chemical in Situ Source Treatments
The in situ treatment of contaminant sources is a relatively recent approach to source control.
Most in situ source treatment has been focused on organic chemical contaminants such as
chlorinated hydrocarbons and petroleum hydrocarbons. These systems have included:
• Cosolvent/surfactant flushing
• Thermal stripping
• Air sparging
• Steam enhanced extraction
• Surfactant flushing with enhanced bioremediation
• Chemical fixation/in situ oxidation
A few in situ source treatments have been attempted for metals and inorganics. At the U.S.
Coast Guard Air Support Center (Elizabeth City, NC), sodium dithionite was injected into a
high concentration Cr(VI) source zone, directly below an old chrome plating shop, to reduce
the chromate to Cr(III) and immobilize it. The process of in situ chemical fixation using ferrous
sulfate has been evaluated in the laboratory for arsenic -contaminated soils at utility substation
sites (Donahoe, 2006a). As a result of successful laboratory studies, ferrous sulfate was injected
into soil contaminated by arsenic -containing herbicides to mitigate release of arsenic to
groundwater. The arsenic is immobilized by co -precipitation with, or adsorption onto, iron
precipitates (Redwine, 2001).
Whereas most treatments for organic contaminants depend upon actual destruction or removal,
either oxidizing the materials to CO, and water or increasing the vapor pressure to enhance
volatilization, inorganic contaminants and metals are typically immobilized by reduction or
oxidation followed by precipitation or adsorption; or, less commonly, the contaminants may
be intentionally mobilized to facilitate capture and removal.
3-7

Overview of Groundwater Remediation Technologies
Some key issues for in situ source remediation are:
• Locating the source
• Source dimensions
• Source contaminant mass
• The ability to comingle the contaminants and reactants in the subsurface
• Competing subsurface reactions (that consume added reactants)
• Hydrologic characteristics of the source and subsurface vicinity
• Delivery options for the cleanup procedure(s)
• Capture of any contaminants mobilized by the procedures
• Long term stability of any immobilized contaminants
Stabilization/Solidification
Stabilization/solidification (S/S) reduces the mobility of organic and inorganic substances
through both physical and chemical means by addition of binder materials to the soil and
groundwater. In stabilization, chemical reactions are used to reduce the leachability of a waste
product, while solidification encapsulates the waste, decreasing the available surface area from
which leaching can occur.
The typical binder materials are cement -based (i.e., cement or blast furnace slag) and can be
added ex situ or in situ. Cement -based S/S reduces the mobility of inorganic compounds by
formation of insoluble hydroxides, carbonates, or silicates; substitution of the metal into a
mineral structure; sorption; and physical encapsulation in a stabilized monolith. The process also
reduces the hydraulic conductivity of the material, which limits the contact between the transport
fluid and contaminants. Because it reduces hydraulic conductivity, there may be cases where it
will be advantageous to use S/S technology to construct a barrier wall. For in situ applications,
the binder material is mixed using vertical augers, conventional equipment such as backhoes,
or injection grouting (Redwine, 2001).
Cement -based S/S involves a complex series of reactions and there are many potential
interferences that can prevent attainment of S/S treatment objectives for physical strength and
leachability. Treatment of arsenic and chromium with cement -based S/S is not always effective
because the high pH of the cement may inhibit formation of insoluble precipitates (USEPA,
1997).
Ex Situ Groundwater Remediation
Historically, ex situ remediation, such as pumping groundwater followed by treatment at the
surface (pump and treat), was the standard approach to managing groundwater plumes. In recent
years, the emphasis has shifted to in situ technologies as limitations of the pump and treat
approach, as described below, became apparent. However, there will continue to be applications
where pump and treat and other ex situ technologies are the optimal remedial approach.

Overview of Groundwater Remediation Technologies
A dedicated treatment system is not always necessary for ex situ remediation at CCP sites,
depending on the PCOCs present, their concentration, and the availability of alternative water
management options. For example, it may be possible to discharge extracted groundwater that
contains relatively low concentrations of inorganic constituents to a storm sewer, sanitary sewer,
or water treatment system used by the power plant.
Groundwater Extraction
Pumping groundwater to the surface and treating it to remove contaminants has a long history of
use as a plume control and remedial technology. Numerous surface treatments have been devised
to remove virtually any contaminant from the water, including those found in CCP leachate.
Surface treatment of inorganics results in materials that must be subsequently disposed or
recycled. Unlike organic compounds, which can be completely destroyed by certain treatments,
inorganics persist and are concentrated by the treatment processes and must be managed within
the treatment train. Examples of these residues are sludges resulting from reductive precipitation,
concentrated "reject" water from reverse osmosis, depleted resins or backflush water from ion
exchange, and plated electrodes from electrolysis.
Treated water must also be managed. Management options include but are not limited to:
• Reinjection to groundwater if the treated water meets groundwater clean-up criteria and if
allowed by state regulation;
Discharge to surface waters or storm sewers if available and if the treated water meets
surface water effluent limits;
• Discharge to sanitary sewers if the only available option or if the treated water does not meet
surface water effluent limits. Depending on the concentration of the extracted groundwater,
it may be possible to discharge groundwater from a CCP site directly to sanitary sewers,
thereby avoiding treatment costs. This approach was used at CCP case study site 1
(Table 2-4).
Pump and treat approaches can be effective at lowering contaminant concentrations, but they
require long-term operation and maintenance for every aspect, from pumping the groundwater
to operating the treatment train and ultimately disposing of the residuals. The expenditures
of energy and money for operating pump and treat systems may, in and of themselves, be
acceptable if cleanup to remedial goals is attainable within a reasonable time frame. However,
pump and treat systems do not always perform as designed. It has been repeatedly observed that
rebound of contaminant concentrations sometimes occurs when pump and treat systems are
turned off (Keeley, 1989). This is due to processes such as slow diffusion of sorbed/adsorbed
contaminants from the mineral surfaces and high contaminant concentrations contained in
restricted pore spaces of the aquifer materials that gradually diffuse/bleed/pinch off into the
surrounding groundwater. Another potential problem is deterioration of extraction wells, as
occurred at CCP case study site 8 (Table 2-4), where the groundwater extraction system was
eventually abandoned due to decreasing well efficiency (Figure 3-1), and an alternative remedial
approach (a cap upgrade) was subsequently implemented.
3-9

Overview of Groundwater Renzediation Technologies
150
III
J
m
E
C
O
C° 50
0
1/1/1978
Pumping W-4 Pumping
Started ; _ Ended
Rr,rr,n
1500
1000
E
O
500 cn
. . 0
1/1/1982 1/1/1986 1/1/1990 1/1/1994 1/1/1998 1/1/2002
Time
Figure 3-1
Concentration Plot Showing Decreasing Effectiveness of Groundwater Extraction at CCP
Case Study Site 8 (from EPRI, 2001 b)
Remediation times for pump and treat systems can be very long (some sites have reported
periods of 50-100 years). The National Research Council (1994) has listed the processes
responsible for affecting pump and treat cleanup times. These are:
• Mixing of clean groundwater with contaminated groundwater (inevitable during pump
and treat remediation), resulting in much larger quantities of water requiring treatment.
• Geologic heterogeneities (difficulties flushing low permeability zones, contaminant diffusion
limited cleanup).
• Non -aqueous phases (the presence of NAPLs and their slow dissolution —not an issue for
CCP sites).
• Sorbed contaminants (desorption rate limited cleanup processes —a potentially significant
issue for CCP sites).
• Leachate from remaining contaminant sources (remaining source terms).
The NRC has also stated the situations in which it thinks pump and treat can be used as a viable
treatment technology. These are:
• May be able to restore the water to health -based standards at relatively simple sites;
• May be able to clean up the dissolved portion of the contaminant plume at more complex
sites while either additional pumping or other methods contain the remaining contamination.
3-10

Overview of Groundwater Reinediation Technologies
Most sites are not geologically homogeneous and isotropic, but instead vary widely with
regard to complexity over short distances. Containment of the remaining contaminants by
active technologies (e.g., pumping) is also not generally a cost-effective long-term management
strategy. These issues have driven research into and adoption of in situ remediation strategies
that are less impacted by pumping limitations.
Interception/Drainage Trenches
For shallow plumes, interception or drainage trenches might be a viable alternative to
groundwater extraction wells. The trenches are similar to a French drain except that the
captured water originates in the subsurface rather than on the surface. Generally, the trench is
excavated to intersect groundwater perpendicular to the flow path and to a depth within the zone
of contaminant migration. Contaminated groundwater then flows into this trench. A variety of
methods may be used to remove the water from the trench and into a treatment system (gravity
flow, pumps, etc.). These systems are generally limited to relatively shallow depths, and issues
such as trench stability, etc., are important to an effective design.
In Situ Groundwater Remediation
The conceptualization and initial development of in situ remediation approaches for groundwater
contamination began about 15 to 20 years ago, and have gained impetus during the last decade.
Most early efforts at in situ remediation were via bioremediation, i.e., using microorganisms
to alter or eliminate the groundwater contaminants, initially focusing on plumes of dissolved
hydrocarbons. More recently, these technologies have been applied to plumes containing
inorganic constituents; an area of continuing research.
Bioremediation
Most efforts at bioremediation have been for the remediation of organic contaminants,
notably petroleum hydrocarbons and chlorinated ethenes. Some success has been attained using
microbial processes for inorganic compounds as well, especially using the ability of certain
microorganisms to create reducing environments under the proper conditions. One example is
the reduction of sulfate concentrations by sulfate reducing bacteria with the concomitant removal
of metals as sulfide precipitates (e.g., PbS, CdS). Potential bioremediation processes for
inorganics include (Bolton and Gorby, 1995):
• Oxidation
• Reduction
• Methylation
• Demethylation
• Metal -organic complexation
• Ligand degradation
• Precipitation
• Bioaccumulation
• Biosorption
3-11

Overview of Groundwater Reinediation Technologies
Biological processes are always present in soils and aquifer materials, sometimes including
those processes that serve to remove contaminants from the water migrating through these
materials. Occasionally these processes are sufficiently active, and receptors of concern are
sufficiently distant that contaminant removal to below regulatory requirements will occur
by natural processes. This is the premise of monitored natural attenuation. When the natural
processes are not sufficient, conditions are not in the right ranges (pH, Eh, nutrients, etc.),
the native microorganisms are not appropriate, or downgradient receptors are too close, then
enhanced bioremediation might be required.
Enhanced bioremediation processes have been extensively investigated since the 1980s and have
included the evaluation of numerous modifications to subsurface environments in an attempt to
improve contaminant removal/transformation by microorganisms. Typically these have involved
the additions of electron acceptors (e.g., oxygen, peroxide, sulfate) and/or electron donors (e.g.,
readily oxidizable organic matter), carbon sources, and nutrients by injection and infiltration.
They have occasionally included the addition of non-native bacterial species or microbial
consortia as well, when the native microbes did not have the needed capabilities. Certain
oxidized metals and inorganic species can serve as electron acceptors in the subsurface,
themselves becoming reduced in the process. If the reduced species are more readily adsorbed,
decreased in toxicity, or precipitated because of this reduction in oxidation state, then this
approach can be used for in situ remediation of these constituents. Oxidized species of PCOCs
sometimes found in CCP leachate that may be amenable to bioreduction and immobilization
include chromate, sulfate, selenate, and molybdate. Other constituents, however, may become
more mobile if converted to a more reduced state, arsenic being a notable example. For the near -
neutral pH range of 6.0 to 8.5 typical of groundwater (Hem, 1989), As(III) is more toxic, has
higher solubility, and is more mobile than As(V).
Another important consideration when evaluating potential for bioremediation at CCP sites is the
effect of iron and sulfate reactions. Fe(III) is a potential electron acceptor but tends to precipitate
as hydrous iron oxides or oxyhydroxides. When subsurface conditions are such that Fe(III)
becomes the favored electron acceptor (anoxic environment) and appropriate microbial
populations or consortia are present, the Fe(III) forms can be reduced to Fe(II), which tends to
remain in solution as an aquo complex. If groundwater containing Fe(II) migrates to an oxic
zone, then the Fe(II) can reoxidize and precipitate, which can cause coprecipitation of other
metallic constituents present in the water. Microbial reduction of sulfate can remove divalent
metal aquo complexes by precipitation of the metal as a sulfide, e.g., CdS, FeS, PbS.
Phytoremediation
Plants are able to remove, transfer, stabilize, or destroy contaminants in groundwater (or soil and
sediment). One of the mechanisms used to remove dissolved inorganic contaminants from water
is rhizofiltration, which involves the adsorption or precipitation of contaminants onto plant roots
or the absorption into the roots of contaminants that are in solution surrounding the root zone.
This mechanism is primarily used on extracted groundwater or surface water, although it can
also be used as an in situ treatment. When used ex situ, the extracted groundwater is pumped
to an ex situ engineered tank system with plants grown under hydroponic conditions whereby
the water must come into contact with the roots. The plants are harvested and disposed as the
roots become saturated with contaminants. When used in situ, the tanks would be placed in a
downstream or downgradient location where the water can enter and exit by gravity drainage.
3-12

Overview of Groundwater Remediation Technologies
Phytovolatilization is a phytoremediation mechanism that can be used for in situ remediation.
It involves contaminant uptake by the plant, transpiration, and eventual release to the atmosphere
of a volatile contaminant, a volatile degradation product of the contaminant, or a volatile form
of an initially non-volatile contaminant. For effective phytoremediation, the degradation product
or modified volatile form should be less toxic than the initial contaminant. An example includes
transformation of selenate to the less toxic dimethyl selenide gas (USEPA, 2001).
The primary limitation for in situ phytoremediation of groundwater is the depth of the plume. It
is limited to shallow depths in unconfined aquifers in which the plume is accessible to the plant
roots.
Phytoremediation can also be used as a hydraulic control measure to influence the movement of
groundwater through the uptake and consumption of large volumes of water (typically associated
with plantations of poplar or willow trees) (EPRI, 1999b). This mechanism may be used as a
supplement to another groundwater remediation technology as a containment measure or as a
means to draw water from one location to another.
A considerable volume of research has been performed on phytoremediation including:
• The U.S. Army Corps of Engineers: littl):/lel.erdc.usticc,.tirll]y.niii/phi
• USEPA:llttp.-Hw\.�,w.epa.gov/ORD/NRMRL/irl2ccl/1-r/iiliytores.litni
• ITRC: htt ://www,itrcweb.orNad Ph to.as
• EPRI:
- The Tennessee Valley Authority Constructed Wetland at Widows Creek: Role of
Vegetation in the Removal of Toxic Trace Elements, TR-114220, 1999
- Improvement of Plants for Selenium and Heavy Metal Phytoremediation Through
Genetic Engineering, TR-114219, 11/09/1999
- Phytoremediation of Trace Elements by Wetland Plants, 1005185, 08/23/2001
- The Allegheny Power Service Constructed Wetland at Springdale: The Role of Plants
in the Removal of Trace Elements, 1006504, 11/05/2001
- Phytoremediation at an Arsenic Contaminated Site, 1011760, 2005
Electrokinetics
Electrokinetic techniques use low voltage direct current to separate and extract metals and other
contaminants from groundwater. The goal of electrokinetic remediation is to effect the migration
of subsurface contaminants by developing an electrical potential between electrode pairs that
have been implanted in the ground on each side of the contaminated area. The charged particles
are mobilized toward the electrodes by the current. Upon their migration to the electrodes, the
contaminants may be removed by electroplating, precipitation/co-precipitation, pumping near
the electrode, or complexing with exchange resins. In situ electrokinetic remediation has been
developed largely to enhance contaminant removal in low permeability soils (GWRTAC,
1997b). This technology has been used on arsenic in soils at a utility substation (EPRI, 2000b)
and at an MGP site (EPRI, 1994b). Results from the utility substation pilot test suggested that
electrokinetic remediation of arsenic was not cost -competitive with competing technologies
(EPRI, 2000b).
3-13

Overview of Groundwater Renzediation Technologies
Chemical Injection to Promote in Situ Immobilization
This approach covers a broad range of technologies where liquid or gaseous reactants are
injected into the substance to cause immobilization of dissolved inorganic constituents.
Examples of reactants that have been used or are in development include:
Ferrous sulfate (FeSO4.7H20): Ferrous sulfate has been used for in situ remediation of
chromium -contaminated sites. The oxidation of ferrous iron (Fe21) to ferric iron (Fe") is
proposed to reduce Cr61 to Cr". A field application of ferrous sulfate to treat chromium at the
Townsend Saw Chain site in Pontiac, South Carolina, showed that, despite ultimate reduction
of chromium concentrations in the aquifer, initial concentrations of chromium increased due
to displacement of adsorbed chromium by excess sulfate ions (USEPA, 2000). Due to the
strong adsorption affinity of arsenic for iron oxyhydroxides, it is likely that ferrous sulfate
can also be utilized to remediate arsenic -contaminated groundwater.
Nanoscale zero-valent iron: This ultrafine "nanoscale" iron powder is injected into a plume
as a slurry and flows with the groundwater to remediate contaminants in situ. The nanoscale
iron particles are 10 to 1000 times more reactive than conventional iron powders because
their smaller size gives them a much larger surface area. An oxidation/reduction reaction
occurs with the iron that reduces the dissolved metals to an insoluble form that remains in
place (Zhang, 2003).
Organo-phosporus nutrient mixture (PrecipiPHOSTM ): The PrecipiPHOSTM method
involves injection of a carbon feed source and a gaseous organo-phosphorus nutrient mixture
to promote the growth of microbes that in turn produce inorganic phosphates as a by-
product! The inorganic phosphates immobilize the metals through precipitation, co -
precipitation, or adsorption.
Sodium dithionite: Sodium dithionite is a strong reductant that is injected into the
subsurface to reduce dissolved metal species, such as Cr(VI) to Cr(III), which then
precipitate. Sodium dithionite solutions can also be injected into a plume to reduce natural
ferric iron -bearing minerals to ferrous iron -bearing minerals. The ferrous minerals then
reduce oxidized species dissolved in passing groundwater. This technology was applied
at a site in Washington state as described in PRB case study 17 (Appendix B).
Permeable Reactive Barriers
Among the most promising of the innovative in situ remediation technologies are permeable
reactive barriers (PRBs). Permeable reactive barriers have been defined as:
An emplacement of reactive materials in the subsurface designed to intercept a contaminant
plume, provide a flow path through the reactive media, and transform the contaminant(s)
into environmentally acceptable forms to attain remediation concentration goals
downgradient of the barrier (Powell and Powell, 1998; Powell et al., 1998).
liiii):1/ovww;etlz► t?lov/AIL: lLuti�lltaru�,zrxti3tif;zi��Ill�tlilu��_ la�i ci �i .l�t�tir
3-14

Overview of Groundwater Remediation Technologies
The concept of PRBs resulted from the observation of chlorinated hydrocarbon concentrations
during investigations of well casing material effects on groundwater sampling (Reynolds et al.,
1990). It was realized that halocarbon concentrations decreased when exposed to iron and steel
casing materials. From this, researchers at the University of Waterloo derived the concept of
using reactive media emplaced in the subsurface to intercept plumes of contaminants and
transform them to harmless by-products.
Initial experiments to test the hypotheses of contaminant removal were done using scrap iron
(e.g., iron filings and granules) in batch and column tests in the laboratory, followed by pilot -
scale field demonstrations and full-scale remedial implementations. Initially the focus was
on the remediation of chlorinated hydrocarbons via successive dechlorination reactions, e.g.:
PCE —> TCE —> DCE —> VC -> Ethane -> Ethane
Realization that a reductive process was occurring resulted in the application of the process
to inorganic contaminants as well.
The remediation of reducible contaminants by iron metal is based on the corrosion of the iron
metal itself. Corrosion occurs because zero -valence iron, or Fe(0), such as the metallic iron
chips, filings, and granules used in most PRBs, are thermodynamically unstable and can serve
as electron donors for the reduction of oxidized species, themselves becoming oxidized in the
process while the contaminants are reduced. Zero valence iron (ZVI) is unstable in the natural
environment and has to be created using high temperature metal refining processes (Evans,
1960; Sculley, 1975; Snoeyink and Jenkins, 1980). It tends to revert to a form that is more
thermodynamically stable; for example, iron metal oxidizes to Fe,O, in the earth's oxygen -rich
atmosphere. At low temperatures, the rate of atmospheric oxidation of iron and steel is negligible
due to the oxide surface films that form and inhibit further surface exposure. However, when the
iron is immersed in an aqueous salt solution, as would be the case for a reactive barrier of iron
chips or filings in groundwater, an electrochemical corrosion mechanism will occur.
Electrons are given up by the metal in one area (the anodic region) forming soluble cations of
the metal, and taken up by oxidized species that become reduced, at another part of the metal
surface (the cathodic region). The instability of the iron itself can provide the necessary energy
for oxidation-reduction reactions without external energy input, provided suitable coupled
electron -accepting reactions can occur with reducible species at the cathode. Typically, dissolved
oxygen is the preferred oxidant, or electron acceptor, during aerobic corrosion processes.
These systems can, however, become anoxic or anaerobic if oxygen is depleted by the reactions.
When present, inorganic contaminants such as chromate (CrO,'-), selenate (SeO,'-), or highly
halogenated organic compounds such as PCE and TCE can serve as the oxidants, accept
electrons, and become reduced. Protons (H+) can also be reduced and paired to form hydrogen
(H). As long as electron acceptors are present, corrosion processes and electron transfer within
the metal can continue.
In addition to metallic iron, numerous other reactive materials have been investigated for use in
PRBs for remediation of a variety of metals, inorganics, and organic contaminants. These have
included bimetallic media (such as platinum, palladium, nickel, or copper -coated iron), organic
matter to promote biological reactions (leaf litter, sphagnum peat, etc.), crushed limestone,
phosphate rock (and fishbone), scrap iron, tin, aluminum, zeolite, pyrite, iron oxides, zinc,
3-15

Overview of Groundwater Remediation. Technologies
stainless steel, copper, brass, biotite, vermiculite, and others. The mechanisms of contaminant
remediation by these materials vary widely and include bioreduction, adsorption, precipitation,
and molecular sieves. A number of inorganic contaminants are now known to be amenable
to remediation using PRB technology and the appropriate selection of reactive media. These
include chromium, sulfate, selenium, nickel, lead, uranium, technetium, iron, manganese, copper,
cobalt, cadmium, zinc, molybdenum, nitrate, phosphate, and arsenic. Notably, four of the five
primary PCOCs for CCP sites are known to be treatable by PRB technology, and research has
been performed for the other constituent, boron (McGregor et al., 2002).
PRBs are discussed in more detail in Section 4.
Monitored Natural Attenuation
Monitored natural attenuation (MNA) is a relatively new approach to subsurface/groundwater
management that is not an active remediation but is also not a "no action" approach. MNA relies
on natural physical/biological processes being sufficient to protect potential receptors and
regulated boundaries downgradient of the source/plume. To ascertain that the processes are
sufficiently protective requires:
• thorough site characterization,
• a conceptual model of the biogeochemical behavior of the contaminants over time in this
milieu,
• careful calculation of distances, travel time, source mass/concentration,
• determination of removal or degradation rates, and
• ongoing monitoring of dissolved contaminant concentrations and geochemical indicators.
The concept of MNA initially developed from observations that certain contaminant plume
dimensions did not increase in size (or did not increase at the expected rate) even when an
ongoing source term was present. Plumes that decreased in volume without advancing were also
noted. The process was referred to as natural attenuation or intrinsic bioremediation until the
importance of the monitoring aspects of the approach were fully realized, at which time it
became generally redesignated as monitored natural attenuation. As with PRBs, the concept was
originally applied to plumes of organic contaminants, in this case petroleum hydrocarbons, and
subsequently considered for chlorinated hydrocarbons and inorganic contaminants. MNA differs
from active bioremediation because, in general, no additions are made to the subsurface to alter
the geochemical or microbiological environment, with the possible exception of the removal of
the source term whenever practical.
The mechanisms that result in the natural attenuation of metals and inorganics along a plume
flow path are generally the same used for active remediation processes, as listed previously.
These include sorption, adsorption, precipitation, ion exchange, biodegradation, surface
complexation, oxidation/reduction reactions, etc. In addition to these, the physical processes of
dispersion and dilution can also contribute to natural attenuation. MNA of inorganic constituents
differs from MNA of organic constituents in that the constituents are not degraded —rather they
disperse, or remain in an immobilized form as a precipitate or adsorbed to the solid matrix.
3-16

Overview of Groundwater Remediation Technologies
Monitoring for MNA may be more extensive than for other remedial approaches, possibly
including a series of wells along the length, and often breadth, of the plume. This allows
observation of whether or not the plume size is increasing, is at a pseudo -steady state, or is
diminishing in volume. In addition to simple observation, it is possible to use these data to
get at least approximate rate values for this increase or decrease (usually calculated as pseudo -
first order rate constants) and a travel time estimate should the plume be advancing in spite of
naturally attenuating processes. Additionally, understanding the geochemistry and microbiology
along the flow path can yield information on whether or not the attenuating processes will
continue, become depleted, increase, or decrease in rate downgradient of the current plume
boundaries.
USEPA is currently developing a framework document for the assessment of MNA for inorganic
constituents. Key elements of the framework are that 1) immobilization is the primary viable
process for MNA of non -radioactive inorganic constituents, and that two "plumes" must be
managed —the liquid plume (immobilization) and solid phases (so they do not remobilize)
(Puls, 2006). The document will provide a framework for evaluating whether or not MNA is
viable, and guidance for determining rates of attenuation, long-term stability, and establishment
of a monitoring and contingency plan. Although MNA is now often proposed as a stand-alone
approach to management of contaminant plumes, it can also be evaluated adjunctive to other
remediation technologies.
Groundwater Remediation Alternatives at CCP Sites
The optimal groundwater remediation alternative for a given CCP site will depend on a variety
of factors such as:
• The location of the site and PCOCs present in groundwater. A site leaching selenium
to an aquifer with nearby downgradient receptors may require a higher level of effort (such
as active downgradient remediation and/or provision of alternative water supplies) due to
health -based water quality issues, than a site leaching sulfate with no downgradient receptors.
• Whether the CCP was managed wet or dry. If managed wet, the readily leachable fraction
of some constituents may have been largely removed during sluicing and the remaining mass
may only leach low concentrations. Case studies have shown that in some hydrogeologic
environments, it may be possible to remediate groundwater at a CCP impoundment by
closing and dewatering the impoundment (CCP case study sites 4 and 5, Table 2-4).
• Whether the CCP lies above the water table. In many cases, an engineered or
evapotranspirative cap will effectively remediate a CCP source if the CCP is above the water
table (CCP case study sites 8 and 9); while a cap alone may not be effective at sites where
CCP was filled below the water table.
• The remaining capacity of the site. It may be more advantageous to install an active
downgradient remediation at sites with a large amount of remaining capacity than to
prematurely close the site.
• Hydrogeologic conditions. A permeable reactive barrier is potentially feasible in a non-
lithified aquifer when the plume is within the limits of PRB trenching technology (currently
40 to 50 feet, 12 to 15 m). A different alternative, such as groundwater extraction, may be
necessary or more cost effective in lithified (rock) aquifers, and when the plume extends
deeper than the limits of PRB trenching technology.
3-17

Overview of Groundwater Renzediation Technologies
• Sensitive environmental areas. It may not be feasible to excavate trenches for barrier walls
or permeable reactive barriers through environmentally sensitive areas downgradient of some
CCP sites. This restriction factored into the selection of the groundwater extraction system
installed at CCP case study site 2.
The case studies listed in Section 2 list the following general approaches to remediation:
• Closure
• Capping
• Excavation
• Barrier walls with hydraulic gradient control
• Provision of alternative water supplies
• Groundwater extraction
• Permeable reactive barrier
• Monitored natural attenuation
The first five approaches are fairly standard applications. The sixth approach, groundwater
extraction (with or without ex situ treatment), is a relatively mature technology, and has
limitations, but may be the only viable alternative in some cases. The last two alternatives are
forms of in situ remediation, a topic of developing interest for CCP sites.
In situ remediation approaches, which allow the plume to follow its natural flow path to
treatment, are very attractive, relative to groundwater extraction with ex situ treatment, because
they:
• require no energy/financial expenditure to control the groundwater flow;
• minimize the treatment of clean groundwater along with the contaminated;
• reduce or eliminate the need to dispose of treatment by-products;
• eliminate the need to manage extracted groundwater; and
• reduce potential for cross -media contamination (groundwater to air, surface water, or land).
The next section in this report describes the potential for application of one particular in situ
remediation technology at CCP management sites: the permeable reactive barrier. The PRB
shows promise because it can be used to treat a variety of inorganic constituents. Furthermore,
this technology is still being actively researched, and there are multiple pilot scale and field
applications; therefore, there is a sizeable body of published research on the topic.
3-18

4
PERMEABLE REACTIVE BARRIERS
Introduction
Permeable reactive barriers (PRBs) have potential to provide cost-effective, passive, in situ
treatment of groundwater contamination. Remediation is achieved as contaminated groundwater
passes through a reactive subsurface zone that either removes the constituent of concern from
groundwater or facilitates its transformation into a less toxic form. The implementation of PRBs
has resulted primarily from the recognized ability of reactive materials (e.g., zero-valent iron)
to remediate organic compounds. The favorable oxidation kinetics of zero-valent iron result in
effective dechlorination of organic solvents such as TCE by reductive processes. The success of
zero-valent iron in passively remediating organic contaminants has led to more recent advances
in every aspect of PRB technology. These advances include design and construction techniques,
reactive media development, application to inorganic constituents, and increased understanding
of PRB economics and long-term performance.
Comprehensive descriptions of PRB design, installation, and monitoring have been published by:
• The Interstate Technology & Regulatory Council: fittL:/Iwww.itrcweb.orag/gd PRB.asg
• USEPA Office of Research & Development: I t_tp://ww% el)a.Gov/ztci ilptibs/t-e])ot-ts.him]
Additional PRB references, including case studies, studies of specific reactive media, and links
to other organizations can be found at:
• USEPA CLU-IN web site: 17tt1}:/1�ltt-%n.orglconlllta'cll7rh/1'eso�lrce.iltl7t
• USDOE: littp-//,Alww.gio.doe.gov/12criii-bai--i-/
• The United Kingdom's Engineering and Physical Sciences Research Council (EPSRC):
till I}://ww-,v.prb-i t.ort.or6,
• The University of Waterloo has been a leader in PRB research and holds several patents on
the technology:
17tt17://www.sc.icnce.Liwaterloo.c,,i/rescZtrclil,,,� r/PermeableRc4ictiveBLirriers/.Perineabl.CR��activ
eBarriers.html
This section summarizes the application of PRB technology for remediation of inorganic
constituents typically associated with CCPs in groundwater. It does not attempt to reproduce all
of the information that can be obtained from the above -listed sources, as well as other sources,
but instead serves as a primer to familiarize the reader with the general concepts of PRBs as they
may be applied to CCP sites, and to identify areas of potential CCP-focused research.
M1

Permeable Reactive Barriers
Intellectual Property
Many of the technologies discussed in this summary report are protected intellectual property,
including design (e.g., U.S. patents 5,362,394 and 5,514,279; Canadian patent 2,062,204),
installation techniques (e.g., U.S. patent 6,357,968 for a jet -grouting approach to injecting PRB
material), and reactive media (e.g., U.S. patent 5,876,606 for a media composed of basic oxygen
furnace oxides or slag). Patented material (for example, many reactive media are patented) can
be researched online at 11 lip://www.tisl)tci-gov/patft/iiidcx.litiiil (for United States patents) and
hltl :Ill}=�trpits .ic.%.cdii7tm-e.litntl (for Canadian patents).
Design and Construction
Design Criteria
Permeable reactive barriers must be engineered with appropriate reactive media, effective
residence time of contaminated water in the reactive media, and strategic location to passively
capture the entire extent of the plume. Various design configurations have been constructed,
including continuous barriers, funnel -and -gate systems, in situ deep slurry injections, and in situ
reactive vessels (ITRC, 2005). Important design considerations include:
• Characterization of site hydrogeology, including delineation of the plume —which provides
necessary data for feasibility analysis and the physical design of the PRB.
• Site constraints such as buildings and major overhead or underground utilities that will affect
PRB positioning and construction methods.
• Characterization of the chemical composition and redox environment of the plume and
background groundwater —which is needed for feasibility analysis and evaluation of reactive
media.
• Regulatory acceptance: will the design feature an unusual physical configuration or new
reactive media, such that a high degree of performance monitoring will be required?
Site Characterization Considerations for the Installation of a PRB
Although a good site characterization is needed for any type of groundwater remediation
scenario, it is especially important for the installation of a PRB. This is because no active
pumping of the water (i.e., flow control) will occur for a typical installation. The site
characterization must be sufficient to allow a thorough understanding of the contaminant plume
in three dimensions, groundwater flow directions and flow rates, and, if possible, the mass of the
contaminant source term. The PRB must lie directly in the flow path of the plume and its design
and construction must be such that the plume cannot bypass the treatment system by flowing
over, under, or around the distal ends of the PRB. This means that the PRB must have a cross -
sectional area (including funnels if so designed) sufficient to more than encompass the cross -
sectional area of the advancing plume. The PRB must also contain an adequate volume of
reactive media and the constituents of interest must have a residence time within the media
(based on flow rate and media thickness) that is long enough to accomplish the remediation
concentration goals. These issues are addressed in greater detail in publications such as Powell
et al. (1998).
WJ

Permeable Reactive Barriers
The surface characteristics of a site (e.g., buildings, utility lines, property boundaries, and a
variety of other considerations) may also impact the design and construction approach for PRB
installation. In most cases, it has not been possible to actually locate the PRB in front of the
advancing plume. Typically, the PRB is installed in a manner that transects the plume, as near its
leading edge as possible, ahead of the zone with highest concentrations. Of necessity, this leaves
the leading edge of the plume downgradient of the PRB; however, it may be possible to treat this
portion of the plume using MNA.
PRB Configurations
Continuous
In a continuous PRB, the reactive media extends across the entire path of the contaminant plume
(Figure 4-1). For shallow applications, the PRB is typically trenched. When plumes are too deep
to feasibly excavate a trench, it may be possible to use injection methods such as jet -grouting or
hydrofracturing to place the reactive media.
f`rtin#in��rl��c DQB
ninant Plume
Figure 4-1
Continuous Reactive Barrier (from USEPA, 1998)
Keyed PRBs are installed such that the bottom of the PRB is keyed into a layer of relatively
low hydraulic conductivity, such as clay, to reduce the potential for contaminant migration under
the barrier. Hanging PRBs are not keyed into a zone of low hydraulic conductivity —these must
be carefully designed to assure that groundwater will preferentially flow through, rather than
below, the PRB. Very long PRBs might have hanging and keyed sections, dependent upon the
stratigraphy. At least one installation has had an additional PRB installed downgradient of the
primary PRB, a multiple installation (PRB case study 15, Appendix B).
4-3

Permeable Reactive Barriers
Funnel -and -Gate
In funnel -and -gate systems, barrier walls are installed to control groundwater flow through
a permeable gate containing reactive media (Figure 4-2). The walls are usually constructed
using sealed -joint sheet piling, or by excavating a trench and installing bentonite or other low -
permeability media (see PRB case studies 2, 12, and 22 in Appendix B). Walls must be keyed
into a layer with low hydraulic conductivity to assure that the plume is directed toward the gate
rather than flowing beneath the wall. Several gate configurations have been developed, including
a baffled design that forces water to flow up and down through reactive media to increase
residence time.
Reactive Gate
1
Figure 4-2
Funnel -and -Gate System (from USEPA, 1998)
In Situ Reactive Vessels
mfvqwvw� �r
M
anninant Plume
These installations use funnels and/or collection trenches to capture a plume and pass the
water through one or more reactive vessels (Figure 4-3). The vessels can be located within the
contained area, within the funnel, or at a distance downgradient from the plume (ITRC 2005).
These systems use gravity or hydraulic head to pass groundwater through the treatment vessels.
This configuration is designed to facilitate reactive media replacement. A patent -pending in situ
reactive vessel design (GeoSiphonTM) is currently in use at the Savannah River site (PRB case
study 3, Appendix B; Phifer et al., 2005). The Y-12 site at Oak Ridge also utilized reactive
vessels (PRB case study site 5, Appendix B).

Co11cction "I reach vo,
flmtlxa w;I'tL i,wraci
RC Ill Ldi'I[Cd r
Groat dWa;t:t'� __W
FiC"•--..._ n 1 _.
'F
f,;1'!1la�; Ks rhT Frrt!'<,lPrt:Rtrf 1'rrrNr4P _ 1 i-irrr 'fry EM, Inc .
Figure 4-3
In Situ Reactive Cell Design (from ITRC, 2005)
Reactive Media
Permeable Reactive Barriers
Reactor CC%
u�f rcactrvc
A
Flow
Dttc don
The main considerations in selecting reactive media are as follows (Gavaskar et al., 1998):
• Reactivity: The media should be of adequate reactivity to immobilize a contaminant within
the residence time of the design.
• Hydraulic performance: The media should provide adequate flow through the barrier,
meaning a greater particle size than the surrounding aquifer materials. Alternatively, gravel
beds have been emplaced in front of barriers to direct flow through the barrier.
• Stability: The media should remain reactive for an amount of time that makes its use
economically advantageous over other technologies.
• Environmentally compatible by-products: Any by-products of media reaction should be
environmentally acceptable. For example, iron released by zero-valent iron corrosion should
not occur at levels exceeding regulatory acceptance levels.
• Availability and price: The media should be easy to obtain in large quantities at a price
that does not negate the economic feasibility of using a PRB.
Of these considerations, the most important criterion for media selection is its ability to
immobilize the PCOCs or convert them to non -toxic forms. Immobilization is the most important
control for the PCOCs encountered at CCP sites, while conversion is of interest for remediation
of plumes containing organic compounds. Some constituents are immobilized by pH control,
while others rely on redox transformation, precipitation, sorption, or biological transformation.
4-5

Penneable Reactive Barriers
Chemical Mobility Controls
EPRI (1984) identified major factors influencing chemical attenuation of constituents found in
CCPs (Table 4-1). The major processes are the combinations of adsorption/desorption and
precipitation/dissolution.
Table 4-1
Major Chemical Attenuation Mechanisms for Constituents in CCPs
Mechanisms
Adsorption/Desorption
Precipitation/Dissolution
Important Factors
Only uncomplexed rather than complexed ions are effectively adsorbed
(e.g., CdCl2 is not adsorbed, but Cd2' is adsorbed).
Hydrous oxides of Al, Fe, and Mn, amorphous aluminosilicates, and
organic carbon are important sorbents.
Oxyanions (e.g., As, Cr, Se, Mo, V, SOJ adsorb most strongly at low pH
and cations (e.g., Pb, Cd, Ni) adsorb most strongly at high pH.
Competing ions and complexing ligands generally reduce adsorption
(e.g., phosphate effectively competes with arsenate).
Specific adsorption (strong, inner sphere adsorption) predominates at
lower concentrations for most elements.
Precipitation is the primary attenuation mechanism for Fe, Al, and Mn.
Solubility -controlling carbonate and hydroxide phases of Cd, Pb, Cr, and
Cu have been observed in alkaline conditions.
Formation of Fe -containing solids may be an important attenuation
mechanism for both cationic and anionic elements.
Precipitation of solid solutions [e.g., (Fe,Cr)(OH)3, (Ca,Cd)CO3] is
expected to be very important.
Some potential reactive media, such as zero-valent iron, promote both adsorption and
precipitation of a broad range of constituents, while others, such as hybrid ion exchange resins,
specifically promote one reaction to target a single or narrow range of constituents (Table 4-2).
Table 4-3 summarizes aqueous species that may exist in either pure water or in a hypothetical
groundwater (EPRI, 1984; USEPA, 2004). The species that exist are primarily a function of pH,
Eh, and overall ion composition of groundwater. Aqueous speciation changes as conditions of
Eh, pH, and ion composition change. Potential solubility controls for various elements are
summarized in Table 4-4.
4-6

Permeable Reactive Barriers
Table 4-2
Immobilization Mechanisms of Potential Reactive Media
Constituents*
Mechanism
Media
Comments
Oxyanions
Adsorption
Zero-valent iron (ZVI)
Neutral -to -acidic pH is
(e.g., As, Se,
.
Surfactant modified zeolites
optimal
V, Cr, Sb, Mo,
Basic oxygen furnace slag
SOa)
Rare-earth or Fe -doping
Amorphous ferric hydroxide
improves adsorption
Neutralized red mud
capacity
Diatomaceous earth
Ferrous sulfate (HFO)
High levels of sulfate may
depress adsorption
Activated alumina
Hybrid ion exchange resin
g
Rare earth elements
KanchanTM arsenic filter
Granular Ferric Hydroxide TM
Clays
Precipitation
ZVI
Obtained by chemical
Ferrous sulfate (Cr)
reduction or as solid
solutions with Fe
Sodium dithionite (Cr)
Organic carbon
Neutral -to -alkaline pH is
Cations
Adsorption
ZVI
(e.g., Fe, Mn,
Humasorb rM
optimal
Cd, Pb, Ni,
Ferrous sulfate (HFO)
Be, Ba, TI)
Zeolites
Clays
Precipitation
Phosphates
May include sulfides,
Limestone
sulfates, carbonates,
oxides, and hydroxides
ZVI
Organic carbon
Neutralized red mud
Oxygen sparging
treat any one constituent in
' The listed reactive media
will not treat all constituents
listed, nor will a reactive media necessarily
all groundwater environments.
4-7

Permeable Reactive Barriers
Table 4-3
Probable Aqueous Species in Pure Water and with Common Complexes
Element
Valence States
Oxidized
(Pure Water)
Reduced
(Pure Water)
Other*
As
5+, 3+, 0
HAsO4z-
H,As03
Thio(S)-species
B
3+
B(OH),,B(OH),
B(0H),,B(0H),
`*BF;
Ba
2+
Bat+
Bat'
BaSOq , BaCO3
Be
2+
Be2+
Bee'
BeF+, BeSO4
Cd
2+
Cd2+
Cd2'
CdSO4 , CdCO3
Cr
6+, 3+
Cr042-
Cr3+, Cr(OH),(s)
CrF2+
Fe
2+, 3+
Fe3+
Fe 2+
FeF', FeSO4
Mn
4+, 3+, 2+
Mn4+,Mn2+
Mn3+, Mn2+
MnSO4
Mo
6+, 5.33+, 5+, 4+
HMo04
M002+1 Mo,Oa(s), MoS2(s)
NA
Ni
2+, 3+
Ni3+ Ni2+
Ni2+
NiHCO,+, NiSO4
Pb
2+
Pb 2+
Pb 2+
PbCO3 , Pb(CO3)22-
CaSO4
S
4+, 6+, 2-
S042-, SO,2-
HS-, H2S
Sb
5+, 3+, 0
Sb(OH)s
Sb(OH)„ Sb(s)
NA
Se -
6+, 4+, 0, 2-
Se042-
Se0,2-, Se(s), HSe
NA
TI
4+, 3+, 1+
TI4+, TI3+
T120(S), TI+
NA
V
5+, 4+, 3+
H2VO4 , HVO,2-
V022+ , V(OH)3
NA
NA = Not applicable
Data from USEPA (2004) and EPRI (1984)
Species listed under "Other' only form under specific circumstances, for example under a specific redox condition when a
certain ion is present
"EPRI (2005a) — BF; included as boron species
Reactive Media Summary
A summary of reactive media that have been used in PRBs for inorganic constituents is provided
in Table 4-5. Table 4-6 contains a listing of other media, primarily sorbents, that have been
tested for water treatment and may have potential application as a component of a reactive media
used in a PRB. A listing of different reactive media options for the PCOCs identified in Section 2
is presented in Table 4-7. Media descriptions are provided in Appendix A.
By far, the most commonly used reactive media for inorganic constituents has been zero-valent
iron. This media can immobilize four of the five most likely PCOCs identified for CCP sites
(arsenic, chromium, selenium, and sulfate). It is also proven effective for immobilization of
molybdenum; however, it has not been proven for boron, antimony, and lithium.

Table 4-4
Potentially Important Solubility and Sorption Controls
Element
Solids _
Sorbents
Clays, Fe -/AI -oxides
As
FeAs04, AsS,
As2S3, ettringite*
B
Hydroxyborates,
Fe -oxides, clays, AI-/
Ca -borate,
Mg -hydroxides
ettringite*
Cr
Cr(OH),, FeCr,O,,
Fe -oxides, Mn-oxides,
(Fe,Cr)(OH),,
clays, organic matter
ettringite* _
Clays, Fe- and AI-
Se
Se(s), metal-
selenides,
oxides
ettringite*
S
CaSO4,
Fe -oxides, amorphous
AI,(OH),,SO,.5H2O,
aluminosilicates
KAI,(SO,),(OH),
ettringite*
Mn
MnCO,, Mn-oxides
NA
Mo
PbMo04, FeMoO„
Fe -oxides, amorphous
Fe,(MoO,)„ MoSZ,
aluminosilicates
ettringite*
Sb(OH),
Amorphous Mg- or AI -
Sb
hydroxides
Ba
BaSO,
Clays
Be I Be(OH), I AI -oxides
Cd CdCO„ Cd,(PO,),, Fe -oxides, Mn-oxides,
CdS clays
Fe Fe(OH)„ Fe,(OH)a, NA
FeCO„ FeS,
Ni NiS, NiFe,O, Fe -oxides, Mn-oxides
Pb Pb(OH),, PbCO,, Fe -oxides, Mn-oxides,
Pb,(PO,),, clays
Pb,O(PO,),,
Pb,(PO,),OH, PbS
TI TI,O„ TI(OH)„ TIS MnO, clays
(only above pH 12)
Permeable Reactive Barriers
Key Factors
As(V) adsorbs best at acidic pH under
oxidizing conditions; can form a sulfide
under reducing conditions.
Sorption maximums on Fe -oxide observed
at pH 8-10.
Adsorbs at pH < 8 under oxidizing
conditions or precipitates as Cr(OH),under
reducing conditions.
Adsorbs at acidic pH under oxidizing
conditions; insoluble as Se(s) when
reduced. _
Can be sequestered in sulfides under
reducing conditions; adsorbed under
oxidizing conditions at acidic pH; forms
sulfate solids.
Precipitation is key to immobilization.
Highly mobile in oxidized systems at pH >7;
adsorbed at pH <7 under oxidizing
conditions; forms sulfide at low Eh.
Only precipitates at or above 26 pg Sb L-';
adsorbed at acidic pH.
Barite solubility -- 32 pg/L; precipitation is
likely solubility control.
Sparingly soluble above pH 6; pH control is
major factor.
Immobilized at alkaline pH by adsorption
and precipitation; forms a sulfide under
reducing conditions.
Immobile under oxidizing or very reducing
conditions; wide pH stability range as
Fe(OH),.
Adsorbs at alkaline pH; will form a sulfide
under reducing conditions.
Immobilized at alkaline pH by adsorption
and precipitation; forms a sulfide under
reducing conditions.
Immobile as TI-oxide under all but highly
reducing conditions; only forms sulfide
under highly reducing conditions and pH
>12.
V Fe,(V03)21 Fe -oxides Easily reduced and mobilized by organics;
VO(OH)2-H101 adsorbed at acidic pH under oxidizing
ettringite* conditions.
After USEPA (2004) and EPRI (1984)
Formation of ettringite occurs at pH >11, which is greater than the pH of most groundwaters. However, if formed, ettringite may
sequester of variety of oxyanions (As, B, Cr, Se, SO„ Mo and V) (Hasset et al., 2003)
4-9

Permeable Reactive Barriers
Table 4-5
Field -Tested PRB Reactive Media for Inorganic Constituents
Media
Zero-valent
iron (Fe' or
ZVI)
Treated Inorcianics'.2
As, Cr, Se, SO,, Mn, Mo, Ba,
Ni, Pb, U, Tc, Fe, Cu, Co, Cd,
Zn, NO,,, PO,, Hg, V
Organic As, Se, SO,, NO,, Cd, Pb, Co,
matter I Cu, Ni, Zn, Fe, PO,
Cost'
$350-
400/ton
(USEPA,
1998)
Field Applications'
1. CCP Landfill, Ontario, Canada
(McGregor et al., 2002; Blowes et
al., 2006)
2. Former DOE Mill Site, Monticello,
UT (Morrison et al., 2002)
3. Savannah River Site TNX Area,
Aiken, SC (Phifer et al., 2005)
4. Haardkrom Site, Kolding,
Denmark (Kjeldson and
Fuglsang, 2000)
5. Y-12 Site, Oak Ridge National
Laboratory, Oak Ridge, TN
(Watson et al., 1998)
6. U.S. Coast Guard Support
Center, Elizabeth City, NC (Puts
et al., 1998)
7. Fry Canyon Site, UT (Feltcorn
and Breeden, 1997)
8. Bodo Canyon, Durango, CO
(httpwww_rtdf.org)
9. Rocky Flats Environmental
Technology Site, Golden, CO
(htip:www.rtdf.org)
20.Newport Superfund Site, DE
(Bronstein, 2005)
22.Uranium Mill, Canon City, CO
(USDOE, 2005)
23.Columbia Nitrogen Site,
Charleston, SC (Bronstein, 2005)
1. CCP Landfill, Ontario, Canada
(McGregor et al., 2002; Blowes et
al., 2006)
9. Rocky Flats Environmental
Technology Site, Golden, CO
(hltp;www.11(,.-Um)
10.Nickel Rim Mine Site, Sudbury,
Ontario, Canada (Benner et al.,
2000)
12.Public School, Langdon, Ontario,
Canada (Baker et al., 1998)
23.Columbia Nitrogen Site,
Charleston, SC (Bronstein, 2005)
4-10

Permeable Reactive Barriers
Table 4-5
Field -Tested PRB Reactive Media for Inorganic Constituents (Continued)
Media Treated Inorganics" Cost' Field Applications'
Phosphates
As, SO,, Mo, Mn, Pb, Cd, Ni, Ba,
$350/ton
7. Fry Canyon Site, UT (Feltcorn
(Bone
U, Zn, NO,
(Conca
and Breeden, 1997)
char/Apatite
and
18.Success Mine and Mill Site, Idaho
IITM)
Wright,
(Conca and Wright, 2006)
2006)
Ion
B
NI
1. CCP Landfill, Ontario, Canada
exchange
(McGregor et al., 2002; Blowes et
resin
As, Se, SO,, Mo, Cd, Pb, U, Cu,
al., 2006)
Limestone,
$95/ton
11.Tonolli Superfund Site,
hydrated
Ni, Zn
(USGS,
Nesquehoning, PA (USEPA,
lime,
2006)
2005b)
dolomitic
12.Public School, Langdon, Ontario,
limestone
Canada (Baker et al., 1998)
19.Cyprus AMAX Minerals
Company, Carteret, NJ
(Bronstein, 2005)
1. CCP Landfill, Ontario, Canada
Zeolites and
Cr, Ba, Sr
$20/ton
surfactant-
(SMZ =
(McGregor et al., 2002; Blowes et
modified
$425/ton)
al., 2006)
zeolites
(Ott, 2000)
13.Chalk River Laboratories,
Ontario, Canada
(fitt :www rtdf.or )
141EAP Permeable Barrier
Demonstration (http:www.rtdf.gfg)
Basic
As, SO,, Pb, Zn
$4 to
15.DuPont Site, East Chicago,
oxygen
$7/ton
Indiana (Wilkens et al., 2003)
furnace slag
(USGS,
(BOF)
Cr
2006)
Sodium
NI
17.100D Area, Hanford Site, WA
dithionite/
(Naftz et al., 2002; Bronstein,
calcium
2005)
polysulfide
21.Universal Forest Products, Inc.,
Granger, IN (Ott, 2000)
Neutralized As, Cr, SO,, Sb, Cd, Co, Cu, Pb, NI 16.Gilt Edge Mine, SD (McConchie
red mud Fe, Mn, Ni, Zn, NO, et al., 1999)
(BauxsolTM
Viromine TM,
Acid-B
Extra TM)
1. The list of constituents for each media may be incomplete; furthermore, a reactive media that successfully treated a
constituent in one application may not necessarily be successful in another application due to differences in environmental
conditions.
2. The most likely PCOCs for CCP Leachate (As, B, Cr, Se, and SO,) are indicated by bold italics; additional PCOCs
determined in Section 2 (Li, Mo, Sb) are indicated by italics.
3. NI = information not available
4. Numbers indicate case studies listed in Appendix B.
4-11

Permeable Reactive Barriers
Table 4-6
Other Potential PRB Reactive Media for Inorganic Constituents
Media'
HumasorbTM
Surfactant -modified
zeolites
Amorphous ferric
hydroxide (AFO)
Dissolved oxygen
Diatomaceous earth/
HFO impregnated
diatomite
Activated alumina
Hybrid ion exchange
resins (ArsenXn,)
Titanium oxide
(ADSORBSIATM)
Mg -AI oxide
(SORBPLUSTM)
Ferrous sulfate
Treated Inorganics"'
As, Hg, Pb, Al, Cu, Cd,
Fe, Ni, Zn,
As, B, Se(VI), Mo, V
As, Se, U
As
As
As
As, Cr, Mo, V, P, Ra,
U, F, possibly Sb
As, Pb
Cr, Cu, Co, Ni
As, B, Cr, Mo, Ni, and
V
Mechanism IReference
Adsorption htip:/Iwwkjv.arctc c h.com1
Adsorption I Donahoe (2005, 2006b)
Adsorption I Feltcorn and Breeden (1997)
Precipitation US Patent 6,254,786
Adsorption Jang et al. (2005)
Adsorption McRae et al. (1999)
Adsorption http.://www.,puroiite.coml
Adsorption I http://www.dow.c�oM/IiqUdsep
I a,/prodlpt as.htm
Adsorption Evanoff et al. (1992)
Reductive USEPA (2000), Donahoe
precipitation; (2006b)
adsorption
ForagerTM Sponge As, Pb, Cu, Cd Adsorption )Itt,p://www.dynaphore.comf
Rare earth elements As, Se Adsorption Tokunaga and Hakuta (2002),
Harck et al. (2004), Zhang et
al. (2005)
KanchanTM Arsenic Filter As, Fe Adsorption http://www.irc.nf—/p—a-qe/2517
Granular Ferric As, Se, Cr, Sb, Cu, Adsorption 7ttp:/Iwww.usfilter.comlen/Pro
Hydroxide TM PO4, duct+Lines/General Filter Pr
oduc s General Filter Produc
is/general filter gfh.h#m
1. The authors were unable to locate field application data for the reactive media listed in this table; however, that does not
necessarily mean that these media have not been used in field applications. Check with the listed vendors or references (see
also Appendix A) for information on bench, pilot, or field -scale applications.
2. The list of constituents for each media may be incomplete; furthermore, a reactive media that appears successful in one
application may not necessarily be successful in another application due to differences in environmental conditions.
3. The most likely PCOCs for CCP Leachate (As, B, Cr, Se, and SO,) are indicated by bold italics; additional PCOCs
determined in Section 2 (Li, Mo, Sb) are indicated by italics.
4-12

Table 4-7
Potential Reactive Media by Constituent
Constituent' Reactive Media with PRB Case
Studies
As ZVI, organic matter, phosphates,
limestone, BOF, red muds
B I Boron ion exchange resin
Cr ZVI, zeolites, sodium dithionite, red
muds
Se I ZVI, organic matter, limestone
SO4 ZVI, organic matter, phosphates,
limestone, BOF, red muds
Li none
Mo ZVI, phosphates, limestone,
Sb I Red muds
Permeable Reactive Barriers
Other Potential Reactive Media
HumasorbTM, surfactant -modified zeolites,
AFO, diatomaceous earth, activated
alumina, hybrid ion exchange resin, Ti
oxide, ferrous sulfate, Forager TM sponge,
rare earth elements, KanchanTM filter,
Granular Ferric Hydroxide TM, clay minerals
Surfactant -modified zeolites, ferrous
sulfate
Hybrid ion exchange resin, Mg -AI oxide,
ferrous sulfate, Granular Ferric
Hydroxide TM
Surfactant -modified zeolites (Se(VI) only),
AFO, rare earth elements, Granular Ferric
Hydroxide TM
Surfactant -modified zeolites, hybrid ion
exchange resin, ferrous sulfate
Hybrid ion exchange resin, Granular Ferric
Hydroxide TM
1. The lists of reactive media for each constituent may be incomplete; furthermore, a reactive media that successfully treated a
constituent in one application may not necessarily be successful in another application due to differences in environmental
conditions.
The only proven reactive media for boron is a boron -specific ion exchange resin, although
ferrous sulfate and, to a lesser extent, surfactant modified zeolites have shown potential in
laboratory experiments. Neutralized red muds have been demonstrated to immobilize antimony,
and granular ferric hydroxide and hybrid ion exchange resins have potential for immobilization
of antimony. No reactive media for lithium were identified during this review; it is unknown
whether this is due to a lack of need to remediate for lithium, or due to difficulties in
precipitating or adsorbing this constituent.
Treatability Testing
Treatability testing is performed during the design of a PRB to evaluate the potential
effectiveness of a reactive media in a bench -scale hydrogeological environment comparable to
that expected in the field. This is a critical portion of the design process because environmental
factors such as pH, Eh, and competing ions in the groundwater will affect the ability of various
media mixtures to immobilize the PCOCs present in site groundwater, as well as the rate at
which a media may degrade due to clogging and due to reactions with constituents that are not
targeted PCOCs.
4-13

Permeable Reactive Barriers
Laboratory studies using samples of site groundwater are needed to determine the effectiveness
of the selected reactive media at the site. Two types of studies are usually carried out: batch
studies and column studies. Batch studies are usually a "first cut" type of study in which samples
of groundwater from the site and different mixtures of reactive media are concurrently shaken or
mixed for a specified period, and then analyzed to see if contaminant removal has occurred and
to what extent. Column studies are more sophisticated one-dimensional flow studies that are used
to determine rates of removal for the best performing mixture(s) determined during the batch
study. A column is filled with the selected reactive media and site groundwater is pumped
through the saturated media at a flow rate scaled to that of the groundwater flow rate at the site.
Samples are taken down the length of the column periodically to assess the degradation/removal
rate. The zone wherein most of the reactions are occurring can also be observed as it moves
gradually down the length of the column. The benefits of column testing following batch testing
are as follows (Gavaskar et al., 2000):
• Design parameters are determined under dynamic flow conditions. Installation of
intermediate sampling points along the length of the column allows measurement of changes
in contaminant concentrations through the barrier media.
• Non -linear sorption to non -reactive sorption sites is better simulated in columns.
• Reaction products may accumulate in batch reactors, but could be washed away in columns,
depending on porosity of the geologic material.
Construction Methods
A variety of construction methods are available for PRB emplacement. Selection of the most -
appropriate method will vary depending on the configuration and depth of the PRB, as well as on
the soil type and on the availability of materials. Construction methods are described in detail in
Gavaskar et al. (2000), and summarized below.
Excavated PRBs
Excavated PRBs are built using traditional soil excavating techniques and are subject to the same
limitations, e.g. water intrusion, collapse of the excavation walls, etc. Some PRBs have been
constructed where only excavation dewatering was required but most excavated PRBs have
required supports. Some trenches have been constructed using sheet piling as supports. More
recently, dense biodegradable polymer solutions, such as guar gum, have been used to support
the trench while the reactive media is emplaced. At least three excavation methods have
been used, with low relative cost (compared to injected PRBs) that increases with depth:
• Backhoes and clamshell excavators remove the soil in one operation and then fill once
the excavation has reached design depth. Backhoe excavations are limited to relatively
shallow depths (30 feet, 9.1 m), while clamshell excavators are capable of depths greater
than 100 feet (30 m); although maximum depth in both cases is dependent on the ability
to keep the trench from collapsing before the reactive media is placed.
• Trenching machines have been built that are able to excavate the aquifer materials and soils
with immediate replacement of the solids by reactive media within the cut (see PRB case
example 6). Trenching machines used for PRB installations have been limited to depths of
about 40 feet (12 m) (ITRC, 2005); although there are trenching machines used for barrier
wall applications capable of excavating to depths of 100 feet (30 m) or more.
4-14

Permeable Reactive Barriers
• Caissons can be driven, while excavating material within, to construct PRBs to depths of
50 feet (15 m) or more. A potential disadvantage of this approach is that removal of the
caisson can be difficult due to friction with native soil on the outside of the caisson and
friction with the reactive media on the inside (Gavaskar et al., 2000).
Direct Placement PRBs
With injected PRBs, reactive media is placed in the subsurface without prior excavation.
Injection methods allow placement of reactive media at depths greater than is feasible by
excavation, although cost is higher. Depth and relative cost data are from Gavaskar et al. (2000).
• Mandrels/Tremie Tubes are essentially hollow tubes (typically rectangular) with drive shoes
on the bottom. They are driven to the desired depth, filled with reactive media, and then the
tubes/mandrels are withdrawn. Maximum reported depth using this method is 50 feet (15 m),
and cost is moderate —a factor of about 2 higher than excavation.
• Augered/deep soil mixing techniques use augers to drill vertical boreholes and mix in
reactive media. Overlapping boreholes are used to create the PRB. This method is capable
of achieving depths up to 150 feet (46 m), but cost is high.
• High pressure jetting approaches were developed from the construction industry where
grouts are injected to increase soil stability. Under extremely high pressure, the reactive
media is jetted into the subsurface, displacing the native soils/aquifer materials. A linear
configuration of the media results if the jets are not rotated. Rotation of the jets produces
thicker cylindrical media zones. This method can install PRBs to depths of 200 feet (61 m).
Cost is relatively high —a factor of 5 to 10 greater than excavation.
• Hydraulic and sand fracturing are rarely -used high-pressure PRB installation techniques that
induce fractures in sand or force reactive media into existing bedrock fractures. Achievable
depths up to 120 feet (37 m) have been reported and cost is high.
PRB Costs
The most reliable and thorough cost information has been obtained from sites using zero-valent
iron as the reactive media. Where information is available, costs are usually reported in terms
of two categories: 1) capital costs and 2) operation and maintenance. The cost -benefit of PRB
technology over other remediation technologies (e.g., pump -and -treat) is realized in reduced
operation and maintenance. The extent of operation and maintenance that is required will depend
largely on the longevity of the reactive media.
Capital Costs
Capital costs for PRB implementation at CCP sites include:
• Site characterization: hydrogeologic, geotechnical, and geochemical assessments;
• Design: treatability studies, modeling, data collection, licensing fees, cost evaluation,
cost comparisons, and work plan development;
• Construction: media, mobilization, emplacement, waste disposal, health and safety,
and site restoration.
4-15

Permeable Reactive Barriers
Documented capital costs for PRB applications ranged from $0.09 million to $1.45 million
(Table 4-8). The largest portion of capital expenses associated with installing PRBs is related to
construction. Emplacement costs will be determined by the following (Gavaskar et al., 1998):
• Plume and aquifer depth: Greater depths increase both emplacement and media costs.
• Plume width: Wider barriers increase both emplacement and media costs.
• Geotechnical considerations: Boulders, rocks, or highly consolidated material increase
the difficulty of emplacement.
Table 4-8
Major Capital Costs Associated with PRBs (After USEPA, 2002b)
PRB/ Primary
Site" Media Depth Site Design* Construction* Total* Stated Cost
Length (Feet) Characterization
Source
(Feet)'
USCG Support 152/152
Center
Moffet Federal 50/10
Airfield
Dover AFB, DE
68/8
Kansas City
130/130
Plant, MO
Aircraft
Maintenance,
650/100
OR
Nickel Rim,
50/50
Ontario
Pease AFB, NH f 150/150
24 1$150,000 1$145,000 $500,000 J$795,000 1 RTDF (2000)
25 $100,000 $175,000 $332,375 $607,375 RTDF (2000)
39 $165,000 $200,000 $296,000 $661,000 RTDF (2000)
30 �$150,000 $100,000 $1,200,000 $1,450,000 RTDF (2000)
29 $350,000
14 $25,000
33 $400,000
$35,000 1$700,000
$30,000 J$35,000
$200,000 1$500,000
$1,085,0001 RTDF (2000)
J$90,000 1 RTDF (2000)
$1,100,000 Gavaskar et
al. (2000)
"Nickel Rim reactive media consists of organic matter, all others are ZVI.
First number is total PRB length, second is reactive media length, which may be less than the total PRB length for funnel and
gate systems.
*Costs adjusted from primary stated cost sources by USEPA (2002b) to reflect cost by category. Costs are not adjusted to
present-day dollar amounts and should be used only as indications of relative categorical expenses.
Although highly site -specific, the installation costs will be determined mostly by the length and
depth of installation. However, the costs of other factors, such as media unit costs, should not be
overlooked when selecting a design. The unit cost of the media will depend on the type of media
selected. Total reactive media cost, however, is also driven by the amount of reactive material
that is required. The following considerations will determine how much media should be used
in a specific PRB (Gavaskar et al., 1998):
4-16

Permeable Reactive Barriers
• Type and concentrations of contaminants: Those with slower reactivities with selected
media will require a PRB design that includes a greater volume of media.
• Regulatory criteria: More stringent regulatory criteria may require a greater treatment
volume, depending on the reactivity of the contaminant with selected media.
• Groundwater velocity: Higher groundwater velocity will require a greater barrier thickness,
requiring a greater volume of media.
• Groundwater flow and contaminant distribution: Heterogeneous flow and contaminant
distribution may lead to inefficiencies in barrier design, resulting in "wasted" media.
• Geochemical and biogeochemical conditions: Biofouling and mineral precipitation can
substantially reduce the effective treatment surface area of barrier media. These processes
can reduce efficiencies or even clog the media over time.
Operation and Maintenance
The most important operation and maintenance (O&M) costs include the following (Gavaskar et
al., 1998):
• Compliance monitoring: This will vary from site to site, depending on local regulatory
requirements.
• Long-term performance monitoring: Monitoring objectives will determine the value
and frequency of long-term performance monitoring.
• Replacement/rejuvenation of the reactive media: This will vary, depending on site -
specific geochemical, biological, and hydrogeological factors.
Operation and maintenance costs for PRB systems are not well-defined due to the fact that
PRBs are a relatively new technology with little history of recorded cost data. However, O&M
costs of PRBs have been compared to other remedial technologies. PRB construction costs
can be relatively high, particularly when costly reactive media, such as granular iron, are used.
Conversely, groundwater extraction and treatment systems tend to have lower construction costs
than PRBs, with higher O&M costs. The only annual O&M cost for a typical PRB system is
routine monitoring to ascertain that it is functioning properly ($30,000 to $90,000 per year for
the sites listed in Table 4-8; EPA, 2002b). Groundwater extraction systems, with or without on -
site treatment, have O&M costs for the wells, pumps, treatment system or water management,
and monitoring.
More difficult to quantify is the replacement cost of reactive media, if necessary. This is because
the technology is relatively new and the industry has little real experience with replacement of
reactive media. However, should the reactive media become depleted or an impediment to flow
due to precipitate blockage, causing a need for replacement, then a large maintenance cost will
result. Research is ongoing into methods for treating the upgradient surfaces of PRBs to disrupt
any precipitate blockages that might impede the groundwater flow.
4-17

Permeable Reactive Barriers
According to Battelle (2002), a PRB system will be more cost-effective than a pump -and -
treat system if the reactive media functions for greater than 10 years without the need for
rejuvenation or replacement. If the media functions for only 5 years or less before rejuvenation
or replacement, then pump -and -treat remediation may be more cost-effective. ITRC (2005)
suggests that a general rule for iron -based reactive media is to expect some form of maintenance
every 10 years at a cost of approximately 25 to 30 percent of initial construction costs. However,
very recent data are beginning to suggest that this 10-year rule of thumb may be overly
conservative, and that some PRBs may be effective for more than 10 years, and possibly
as long as 30 years (Blowes et al., 2006; Puls, 2006; ITRC, 2005).
Long -Term Performance
Monitoring of PRBs
Two types of monitoring may be used for PRB remediations:
• Compliance Monitoring
• Performance Monitoring
Compliance monitoring is used to determine whether or not regulatory concentration goals are
being met at the compliance boundary, e.g., at agreed upon downgradient monitoring wells,
property boundaries, etc. These monitoring approaches are usually regulated as part of the
facility's operating/closure permit or as part of the remedial action plan.
Performance monitoring is used to ascertain that the PRB itself is meeting its design goals.
PRBs may be performance monitored in several locations relative to the PRB itself, e.g.:
• Immediately upgradient of the PRB
• Immediately downgradient of the PRB
• At the distal ends of the PRB
• Occasionally within the reactive media of the PRB itself
PRB performance monitoring is typically done using small diameter wells (as small as 3/a-
inch i.d. or even bundled tubes with short screens less than 1 foot in length), using low flow and
semi -passive purging and sampling techniques (Powell and Puls, 1997). The concept of
performance monitoring is to provide a performance baseline for the PRB for later comparison;
therefore, it should be implemented shortly after PRB completion. It should accomplish a
number of objectives, including the determination of:
• Short-circuiting of contamination through or around the PRB
• Changes in reactivity
• Decreases in PRB permeability
• Changes in upgradient contaminant concentration, etc.
Mr.

Permeable Reactive Barriers
As the number of FRB installations has increased to over 100, there has been decreasing
emphasis on highly detailed performance monitoring other than at research sites or other sites
incorporating new configurations, reactive media, etc. This is due to the increased confidence
in the performance of the PRBs.
PRB Longevity and Maintenance
PRB longevity has been a much debated and researched topic. Early lifetime estimates for PRBs
with ZVI as the reactive media were for 5 to 10 years. Many of these early PRBs have now been
successfully operating, with little sign of deterioration or plugging by precipitates, for more than
5 years, and recent estimates have been revised upwards to somewhere between 10 to 20 years, if
not longer (ITRC, 2005). Less information is available for the longevity of other reactive media,
such as organic matter (which may still be from 5 to 10 years).
Geochemical, biogeochemical, and biological processes can have significant effects on the long-
term hydraulic and reactive performance of PRBs. Permeable reactive barriers remove not only
contaminants from groundwater, but also aqueous species that are not necessarily targeted for
removal, such as inorganic carbon, calcium, magnesium, nitrate, and silica. These species are
removed by mineral precipitation, adsorption, and biochemical transformations. In zero-valent
iron barriers, the major geochemical, biological, and biogeochemical processes that occur are
iron metal corrosion, microbial sulfate and nitrate reduction, adsorption, gas production, and
mineral precipitation (USEPA, 2003).
Importance of Iron Metal Corrosion
The major consequences of iron corrosion are (USEPA, 2003):
• the production of OH- ion (pH increases to pH 9-11 in the wall),
• decreases in oxidation-reduction potential (Eh sometimes <-500 mV),
• increases in hydrogen concentration (under anaerobic conditions),
• release of ferrous iron (under anaerobic conditions), and
• precipitation of iron -bearing minerals.
The rate of iron corrosion is significantly affected by the anion composition of groundwater. It is
expected that corrosion rates will be faster in chloride -rich water than in bicarbonate- or sulfate -
rich water. This is because the tendency of ferrous iron to form complexes with anions follows
the order Cl>HCO3>SO4>OH (USEPA, 2003). CCP leachate is usually dominated by sulfate,
suggesting lower iron corrosion rates than in alkaline or chloride dominated waters.
The anion composition will also dictate the types of mineral precipitates that will form in iron
PRBs (Table 4-9). If minerals form as surface coatings on reactive iron surfaces, reactivity
may be reduced or enhanced, depending on the contaminant. For example, the uptake rate of
chromium is increased over time, likely due to an increase in available sorption sites. However,
excess precipitation may eventually result in long-term porosity and permeability loss. Li et al.
(2005) performed reactive transport simulations to assess the impact of mineral fouling on the
4-19

Permeable Reactive Barriers
hydraulic behavior of continuous -wall PRBs that employ zero-valent iron in a carbonate -rich
alluvial aquifer. They found that only subtle changes in hydraulic behavior occur during the
first ten years, and that significant changes do not occur until at least thirty years of operation.
However, they noted that water with high TDS can increase the rate of fouling, and decrease
the time necessary for changes to occur. The Elizabeth City, NC PRB (PRB case study 6,
Appendix B) has experienced minor mineral accumulation because of the low-TDS groundwater
at the site. Studies have shown that volume loss in iron barriers results from the formation of
precipitates containing carbon, sulfur, and iron (USEPA, 2003).
Table 4-9
Mineral Precipitates in Zero-Valent Iron PRBs (from USEPA, 2003)
Precipitate Type
Oxides and Hydroxides
Carbonates
Sulfides
Microbial Activity
Mineral
Ferrihydrite — Fe(OH)3
Lepidocrocite — Fe00H
Goethite — Fe00H
Hematite — Fe203
Maghemite — Fe203
Green rust 1 — FeJOH)12CO3xH2O
Magnetite — Fe30,
Calcite — CaCO3
Aragonite — CaCO3
Iron carbonate hydroxide — Fe2(OH)2CO3
Siderite — FeCO3
Mackinawite — Fe,,,S
Greigite — Fe3S,
Pvrite — FeS,
Microbial activity will have significant effects on iron -based PRB performance. Metallic
iron is a major energy source for microorganisms, and will be utilized as such in natural
environments. Gu et al. (2002) investigated the microbial population at the Oak Ridge Y-12
Plant site (PRB case study 5, Appendix B) and found that concentrations of biomass within the
ZVI media were one to three orders of magnitude greater than that found in adjacent aquifer
material. Similar results have been found at the Elizabeth City (PRB case study 6) and Denver
Federal Center; sites (USEPA, 2003).
Negative effects of microbial activity include changes in PRB hydraulic conductivity, masking
of active sites, removal of active chemical species, mineral precipitation, production of gas
bubbles, and competition for reducing equivalents (USEPA, 2003).
It has been reported that the release of ferrous iron may not be detrimental to PRB performance.
For example, if the targeted contaminant can be reduced by either Fe' or Fee,, then the zone of
reactivity is essentially increased beyond the surfaces of Fe". Additionally, if sulfate -reducing
bacteria begin to proliferate, then constituents such as arsenic, antimony, molybdenum,
cadmium, nickel, and lead may be immobilized by metal sulfide precipitation.
' Not listed in Appendix B because this PRB was designed to treat organic compounds.
4-20

Permeable Reactive Barriers
The presence and utilization of dissolved hydrogen can result in bacterial growth and biofilm
formation. Biofilm growth in a porous medium such as an aquifer or reactive wall may cause
reductions in total pore volume, but quantification of pore volume reductions is difficult (Taylor
et al., 1990; Thullner et al., 2002). USEPA (2002c) suggests that while the overall growth of
biomass in a barrier may seem insignificant, the localization of these growths can lead to
significant heterogeneities in barrier hydraulic performance.
Combining PRBs with Monitored Natural Attenuation
Because it is not always possible to locate a PRB downgradient of the advancing front of a
plume, PRBs are sometimes constructed to transect the plume. This approach manages the
upgradient plume as it impinges the PRB but does little other than eliminate the ongoing source
for the aquifer downgradient of the PRB. Given this situation, there are several possible options
for managing the downgradient plume:
• In some instances, regulators have been content with simple monitoring to ascertain that
the downgradient contamination is gradually decreasing.
• If necessary, the rates of natural attenuation can be evaluated in the downgradient portion
and an assessment of the travel times and concentrations that will reach downgradient
receptors can be made. Should natural processes be sufficient, nothing more is needed.
This is monitored natural attenuation in combination with the PRB.
• If neither simple monitoring nor monitored natural attenuation is sufficient to satisfy
stakeholders and regulators, then it might be necessary to actively treat the downgradient
water.
If treating the downgradient water is necessary then one approach is to pump the downgradient
water and reinject it upgradient of the PRB. This must be done at a withdrawal/injection rate
that is not disruptive to the PRB goals (i.e., avoid significantly enhancing the flow rate through
the PRB; contaminant residence time is important). If it is known that remediation of the
downgradient waters will be required, this should be factored into the PRB design. At some point
the downgradient waters, having no source term, should be sufficiently low in concentration to
allow the pump(s) to be turned off.
Another possibility for combining PRBs with MNA is to rely on MNA to complete the cleanup
that was mostly accomplished by the plume passing through the PRB (ITRC, 2005). In this case
the PRB could be deliberately designed in such a manner that:
• The PRB is not expected to remove 100 percent of the contaminant,
• Contaminant breakthrough could occur because multiple contaminants are present in the
plume and the PRB has been designed primarily to remove those of greatest concern, or
• Some unexpected breakthrough has occurred.
If it can be shown that MNA is sufficient to avoid impacts on sensitive downgradient receptors
and protect human health and the environment, then a strong argument can be made that the
remediation is sufficient.
4-21

Permeable Reactive Barriers
Applicability of PRBs for Remediation of Groundwater at Coal Combustion
Product Management Sites
Only one of the 23 PRB case studies compiled for this technology review was performed at a
CCP site (PRB case site 1, Appendix B). Eight of the sites were mining or metal refining sites,
three were metal plating sites, and 11 were of miscellaneous use. PRB case study 1 demonstrated
a promising approach for remediating boron and other PCOCs in groundwater using a mixed
media approach that included ZVI, organic matter, and a boron -selective ion exchange resin.
However, this was a pilot scale application —the trench was only 42 feet (13 m) long —and cost
data for the mixed media were not available, so cost effectiveness relative to other remedial
technologies could not be evaluated.
Despite the limited application of this remediation technology at CCP sites to date, permeable
reactive barriers appear to be a viable alternative if remedial objectives require groundwater
restoration, either with or without source control. There are a variety of PRB configurations
and reactive media currently available that can be employed for targeted remediation of PCOCs
that may exceed health -based standards, such as arsenic, selenium, and chromium. One of these
media is zero-valent iron, for which there is a considerable amount of case -study application
data. However, other media also show potential for application in mixtures at CCP sites; cost and
compatibility in mixtures with other media to immobilize boron will determine media selection.
This technology requires more development for application at sites where the remedial objectives
call for clean-up of the mixture of constituents typically present at CCP sites. A considerable
amount of work has been done on arsenic, chromium, and cationic metals, but more research
is needed on difficult to treat elements such as boron, sulfate, molybdenum, and antimony, and
unique interactions that may occur in the CCP leachate matrix.
4-22

5
RESEARCH ISSUES AND RECOMMENDATIONS
PRB Research Needs and Recommendations for Application at CCP Sites
A review of relevant published literature regarding permeable reactive barrier technology
has identified a variety of short-term and long-term research needs. In the context of CCP
management sites, the most important future research needs relate to reactive media selection
and long-term media performance. With respect to remediation designs and remedial
alternatives, the least known variables relate to media life expectancy.
Short-term research needs include:
• Media selection: Evaluation of reactive media mixtures that target utility PCOCs
(specifically oxyanions and boron). Media such as ZVI, organic matter, ion -exchange resins,
surfactant -modified zeolites, ferrous sulfate, red muds, and Granular Ferric HydroxideTM
show promise for remediation of utility constituents. Overall, performance and cost data
are needed for a wide range of constituents, and for varying mixtures of these constituents.
• PRB configurations: Related to media selection, research on innovative PRB
configurations, such as sequential PRBs designed to treat different PCOCs, may yield an
alternative that will be effective for all PCOCs likely to be encountered in groundwater at
CCP sites.
• Geochemical characterization: Methods of geochemical characterization of a site should
be targeted to a wide range of utility constituents. For example, selenium and arsenic are both
redox-sensitive, and valuable information can be derived from knowledge of elemental redox
states. Advanced field and laboratory methods as they pertain to utility constituents should be
compiled and used as site characterization guidance.
Long-term research needs include:
• Long-term performance: Laboratory experiments designed to simulate long-term media
life expectancy should be performed on a variety of media, including predictive modeling of
geochemical and biogeochemical reactions that occur in field settings. Experiments should
address the potential for biofouling and the resulting permeability and reactivity loss.
• Media rejuvenation: Improved methods of in situ and ex situ media replacement and
rejuvenation may result in procedures to reduce future operation and maintenance costs.
• Innovative applications: Additional cost -benefit may result from innovative methods of
incorporating reactive media into construction of CCP storage sites; for example, reactive
liners.
5-1

6
REFERENCES
American Coal Ash Association (ACAA), 2004. 2004 Coal Combustion Product (CCP)
Production and Use Survey, http=//www.,icaa-usa.org/PDF/2004 CCP Surve 9y-9-05Lpdt'.
Amonette, J. E., J. E., Szecsody, H. T. Schaef, J. C. Templeton, Y. A. Gorby and J. S. Fruchter,
1994. Abiotic reduction of aquifer materials by dithionite: a promising in situ remediation
technology. In Proceedings of the 33rd Hanford Symposium on Health and the Environment —
In Situ Remediation: Scientific Bases for Current and Future Technologies. Battelle Press,
Richland, WA. pp. 851-882.
Bain, J., L. Spink, D. Blowes, and D. Smyth, 2002. The removal of arsenic from groundwater
using permeable reactive materials. In Proceedings: Tailings and Mine Waste '02, Jan. 27-30,
Ft. Collins, CO. Netherlands: A.A. Balkema Publishers, pp. 213-216.
Baker, M.J., D.W. Blowes, and C.J. Ptacek, 1998. Laboratory development of permeable
reactive mixtures for the removal of phosphorus from onsite wastewater disposal systems.
Environmental Science and Technology, Vol. 32, pp. 2308-2316.
Battelle, 2002. Final cost and performance report, evaluating the longevity and hydraulic
performance of permeable reactive barriers at Department of Defense sites.
Benner, S. G., D.W. Blowes, and C.J. Ptacek, 2000. Long-term performance of the nickel rim
reactive barrier: a summary. In Proceedings, ICARD 2000, Fifth International Conference on
Acid Rock Drainage, Society for Mining, Metallurgy, and Exploration, Inc. pp. 1221-1225.
Benner, S.G., D.W. Blowes, C.J. Ptacek, 1997. A full-scale porous reactive wall for prevention
of acid mine drainage. Ground Water Monitoring and Remediation, Vol. 17, Issue 4. pp. 99-107.
Blowes, D.W., D. Smyth, and C. Ptacek, 2006. Permeable reactive barriers for treating
groundwater contaminated by dissolved metals. Presentation to the EPRI Workshop on
Remediation of Inorganic Contaminants, July 11-12, 2006, Baltimore, MD.
Bolton, H. Jr., and Y.A. Gorby, 1995. An overview of the bioremediation of inorganic
contaminants. Bioremediation of Inorganics. Hinchee, R.E., Means, J.L., and D.R. Burris, eds.
Bioremediation 3(10), 1-16, Battelle Press, Columbus, OH.
Bostick, W. D., R. J. Jarabek, D. A. Bostick, and J. Conca., 1999. Phosphate -induced metal
stabilization: use of apatite and bone char for the removal of soluble radionuclides in authentic
and simulated DOE groundwaters. Advances in Environmental Research, Vol. 3, pp. 488-498.
6-1

References
Bronstein, K., 2005. Permeable reactive barriers for inorganic and radionuclide contamination.
Preparedfor USEPA. Available online at,httpalwww.clLl-ill.crrg.
Clark, M.W., 2000. Geochemical cycling of heavy metals in Brisbane River estuary dredge
sediments during sub -aerial disposal, and the consolidation at the Port of Brisbane Authority's
Fisherman Islands reclamation paddocks. PhD Thesis, Southern Cross University, NSW.
Craig, M., 2004. Biological PRB used for perchlorate degradation in ground water. Technology
News and Trends, Issue 10 (February).
Conca, J.L., and J.B. Wright, 2006. An Apatite II permeable reactive barrier to remediate
groundwater containing Zn, Pb and Cd. Applied Geochemistry, V. 21, No. 8, August 2006,
pp. 1288-1300.
Cummings, M. and S. Booth, 1997. Cost effectiveness of in situ redox manipulation for
remediation of chromium -contaminated groundwater. LA-UR-97-165. March 1997.
Donahoe, R.J., 2006a. In situ chemical fixation of arsenic -contaminated soils. Presentation to the
EPRI Workshop on Remediation of Inorganic Contaminants, July 11-12, 2006, Baltimore, MD.
Donahoe, R.J., 2006b. Remediation of trace elements associated with coal combustion by-
product disposal facilities. Presentation to the EPRI Workshop on Remediation of Inorganic
Contaminants, July 11-12, 2006, Baltimore, MD.
Donahoe, R.J., 2005. Personal communication.
Chemical Attenuation Rates, Coefficients, and Constants in Leachate Migration. Volume L-
A Critical Review. EPRI, February 1984. Final report EA-3356.
Chemical Characterization of Fossil Fuel Combustion Wastes. EPRI, September 1987.
Final Report EA-5321.
Chemical Attenuation Reactions of Selenium. EPRI, February 1994a. Final Report TR-103535.
Electrokinetic Removal of Coal Tar Constituents from Contaminated Soils. EPRI, March 1994b.
Final Report TR-103320.
Guidance for Comanagement of Mill Rejects at Coal -Fired Power Plants. EPRI, 1999a. Final
Report TR-108994.
Evaluation of EcolotreeTM Cap for Closure of Coal Ash Disposal Sites. EPRI, January 1999b.
Final Report TR-112442.
Improvement of Plants for Selenium and Heavy Metal Phytoremediation through Genetic
Engineering. EPRI, November 1999c. Final Report TR-114219.
6-2

References
The Tennessee Valley Authority Constructed Wetland at Widows Creek: Role of Vegetation
in the Removal of Toxic Trace Elements. EPRI, November 1999d. Final Report TR-114220.
Environmental Chemistry of Arsenic: a Literature Review. EPRI, November 2000a. Final Report
1000585.
Electrokinetic Remediation of Arsenic in Soil and Groundwater. EPRI, 2000b. 1000587.
Evaluation and Modeling of Cap Alternatives at Three Unlined Coal Ash Impoundments.
EPRI, September 2001a. Final Report 1005165.
Evaluation of Remedial Actions at an Unlined Coal Ash Landfill: CS Site. EPRI, December
2001b. Final Report 1005214.
Phytoremediation of Trace Elements by Wetland Plants. EPRI, August 2001c. Final Report
1005185.
The Allegheny Power Service Constructed Wetland at Springdale: the role of plants in the
removal of trace elements. EPRI, November 2001d. Final Report 1006504.
Evaluation of a Remedial Action at an Unlined Coal Ash Landfill: PI Site. EPRI, December
2002. Final Report 1005262.
Chemical Attenuation Coefficients for Arsenic Species Using Soil Samples Collected From
Selected Power Plant Sites. EPRI, December 2004. Interim Report 1005505.
Chemical Constituents in Coal Combustion Product Leachate: boron. EPRI, March 2005a.
Final Report 1005258.
Phytoremediation at an Arsenic Contaminated Site. EPRI, August 2005b. Technical Update
1011760.
Characterization of Field Leachates at Coal Combustion Product Management Sites: Arsenic,
Selenium, Chromium, and Mercury Speciation. EPRI, December 2006a. Technical Report
1012578.
Chemical Attenuation Coefficients for Selenium Species Using Soil Samples Collected from
Selected Power Plant Sites. EPRI, December 2006b. Technical Report 1012585.
Evanoff, S.P., M. Kunka, and P.M. Stearns, 1992. A pilot study of novel hexavalent chromium
removal from chemical film rinsewater at General Dynamics — Fort Worth division. SAE,
Annual Aerospace/Airline Plating and Metal Finishing Forum and Exposition, 28", San Diego,
CA.
Evans, U.R., 1960. The Corrosion and Oxidation of Metals: Scientific Principles and Practical
Applications. London, Edward Arnold (Publishers) Ltd.
6-3

References
Feltcorn, E. and R. Breeden, 1997. Reactive barriers for uranium removal. Ground Water
Currents. December 1997, No. 26.
Gavaskar, A., B. Sass, N. Gupta, J. Hicks, S. Yoon, T. Fox, and J. Sminchak, 1998. Performance
evaluation of a pilot -scale permeable reactive barrier at former naval air station Moffett Field,
Mountain View, California. Battelle Memorial Institute, prepared for the Naval Facilities
Engineering Service Center, Port Hueneme, CA.
Gavaskar, A., N. Gupta, B. Sass, R. Janosy, and J. Hicks, 2000. Final design guidance for
application of permeable reactive barriers for groundwater remediation, Columbus, OH.
Battelle for Air Force Research Laboratory.
Ground -water Remediation Technology Analysis Center (GWRTAC), 1997a. Phytoremediation.
TE-98-01, October 1997.
Ground -water Remediation Technology Analysis Center (GWRTAC), 1997b. Electrokine tics.
Technology Overview Report TO-97-03.
Gu, B., D.B. Watson, L. Wu, D.H. Philips, D.C. White, and J. Zhou, 2002. Microbiological
characterization in a zero-valent iron reactive barrier. Environmental Monitoring and
Assessment, Vol. 77, pp. 293-307.
Haggerty, G.M and R.S. Bowman, 1994. Sorption of chromate and other inorganic anions by
organo-zeolite. Environmental Science and Technology, Vol. 28, Issue 3, pp. 452-458.
Harek, J.F., S. Wilkis, and I. Winters, 2004. Composition and process for removing arsenic
and selenium from aqueous solution. U.S. Patent no. 6,800,204.
Hasset, D.J., D.F. Pflughoeft-Hasset, L.V. Heebink, 2003. Leaching of CCBs: observations from
over 25 years of research. 2003 International Ash Utilization Symposium, Center for Applied
Energy Research, University of Kentucky, Paper #76.
Hem, J.D., 1989. Study and interpretation of the chemical characteristics of natural water.
USGS Water -Supply Paper 2254, 264 p.
Irwin, R.J., 1997. Environmental contaminants encyclopedia. National Park Service,
Water Resources Division. Fort Collins, CO.
ITRC (Interstate Technology and Regulatory Council) 2005. Permeable reactive barriers:
lessons learned/new directions. Interstate Technology and Regulatory Council, Permeable
Reactive Barriers Team, PRB-4, Washington, D.C., Available on the Internet at
%'%'W.IlI'C`wcb.org.
Jang, M., S.H. Min, and J.K. Park, 2005. In situ permeable barriers using adsorptive media
for remediation of high levels of organic and inorganic arsenic contaminated groundwater.
American Water Resources Association — Wisconsin Section, 29th Annual Meeting.

References
Keeley, J.F., 1989. Chapter 1. Introduction. In Transport and Fate of Contaminants in the
Subsurface, USEPA Seminar Publication, EPA/625/4-89/019.
Kjeldsen, P. and I.A. Fuglsang, 2000. Demonstration program on reactive barrier technologies
using zero-valent iron. FZK/TNO International Conference on Contaminated Soil, Leipzig,
Germany, September 18-22, 2000, pp. 943-950.
Li, Lin, C.H. Benson, and E.M. Lawson, 2005. Impact of mineral fouling on hydraulic behavior
of permeable reactive barriers. Ground Water, Vol. 43, No. 4, pp. 582-596.
Li, Z., 2006. Chromate transport through surfactant -modified zeolite columns. Ground Water
Monitoring & Remediation, V. 26, No. 3, pp. 117-124.
McConchie, D., M. Clark, C. Hanahan, and R. Fawkes, 1999. The use of seawater neutralized
bauxite refinery residues (red mud) in environmental remediation programs. In Proceedings,
Global Symposium on Recycling, Waste Treatment and Clean Technology, San Sebastian,
Spain, Vol. 1, pp. 391-400.
McGregor, R. G., S. Hansler, D. W. Blowes, and E. Laratta, 2002. The use of a PRB to treat
groundwater impacted by coal -combustion by-products. In Proceedings, Third International
Conference on Remediation of Chlorinated and Recalcitrant Compounds, Monterey, Calif.
Columbus, Ohio: Battelle Press. Paper 2A-02.
McRae, C. W. T., D. W. Blowes, and C. J. Ptacek, 1999. In situ removal of arsenic from
groundwater using permeable reactive barriers: a laboratory study. In Proceedings: Mining
and the Environment, Vol. 2, Sudbury, Ontario, pp. 601-609.
Morrison, S. F., C. E. Carpenter, D. R. Metzler, T. R. Bartlett, and S. A. Morris, 2002. Design
and performance of a permeable reactive barrier for containment of uranium, arsenic, selenium,
vanadium, molybdenum, and nitrate at Monticello, Utah. In Handbook of Groundwater
Remediation Using Permeable Reactive Barriers: Applications to Radionuclides, Trace Metals,
and Nutrients, D. L. Naftz, S. J. Morrison, J. A. Davis, and C. C. Fuller, eds., Elsevier Science,
pp. 372-401.
Munro, L.D., M.W. Clark, and D. McConchie, 2004. A BauxsolTM-based permeable reactive
barrier for the treatment of acid rock drainage. Mine Water and the Environment, Vol. 23,
pp.183-194.
Naftz, D.L, S.J. Morrison, J.A. Davis, and C.C. Fuller, 2002. Handbook of groundwater
remediation using permeable reactive barriers: applications to radionuclides, trace metals,
and nutrients. Academic Press, New York.
National Research Council, 1994. Alternatives for ground water cleanup. National Academy
Press, Washington, D.C.
Ott, N. 2000. Permeable reactive barriers for inorganics. prepared for USEPA, littp://www.clLl-
a "Loma.
6-5

References
Phifer, M.A., R.L. Nichols, F.C. Sappington, J.L. Steimke, and W.E. Jones, 2005. GeoSiphonTM
ground water remediation system hydraulics. Environmental Geosciences, March 2005, V. 12,
No. 1, pp. 29-44.
Powell, R.M. and P.D. Powell, 1998. Iron metal for subsurface remediation. The Encyclopedia
of Environmental Analysis and Remediation. Robert A. Myers, ed., John Wiley & Sons, Inc.,
New York, 8:4729-4761.
Powell, R.M. and R.W. Puls, 1997. Hitting the bull's-eye in groundwater sampling. Pollution
Engineering, June 1997, pp. 50-54.
Powell, R.M., R.W. Puls, D. Blowes, J. Vogan, R.W. Gillham, P.D. Powell, D. Schultz,
R. Landis, and T. Sivavec, 1998. Permeable reactive barrier technologies for contaminant
remediation. U.S. EPA Report, EPA/600/R-98/125.
Powell, R.M., R.W. Puls, S.K. Hightower, and D.A. Sabatini, 1995. Coupled iron corrosion
and chromate reduction: mechanisms for subsurface remediation. Environ. Sci. Technol.
29:1913-1922.
Puls, R.W., 2006. Arsenic research at GWERD. Presentation to the EPRI Workshop on
Remediation of Inorganic Contaminants, July 11-12, 2006, Baltimore, MD.
Puls, R.W., R.M. Powell, C.J. Paul and D. Blowes, 1998. Ground water remediation of
chromium using zero-valent iron in a permeable reactive barrier. In Field Testing of Innovative
Subsurface Remediation Technologies, American Chemical Society Symposium, 13-17 April
1997, San Francisco, CA, 1998 (Paper also available from NTIS as EPA/600/A-98/108. Order
P1398-155088).
Redwine, J.C., 2001. Innovative technologies for remediation of arsenic in soil and groundwater.
In W.R. Chappell, C.O. Abernathy, and R.L. Calderon eds., Arsenic Exposure and Health Effects
IV, Proceedings of the Fourth International Conference on Arsenic Exposure and Health Effects,
18-22 June, 2000, San Diego, CA, USA, Elsevier Science & Technology.
Redwine, J. C., J.R. Howell, R.J. Donahoe, and L. Yang, 2004. In situ chemical fixation of
arsenic in soil: laboratory and field results. In Proceedings, Air and Waste Management
Association 97th Annual Conference & Exhibition, Indianapolis IN, paper # 548.
Reynolds, G.W., J.T. Hoff, and R.W. Gillham, 1990. Sampling bias caused by materials used
to monitor halocarbons in groundwater. Environ. Sci. Technol. 24: 135-142.
RTDF, 2000. Summary of remediation technologies development forum. Permeable Reactive
Barriers Action Team Meeting. Hilton Melbourne Airport, Melbourne, Florida, February 16-17.
Sass, B.M. and D. Rai, 1987. Solubility of amorphous chromium(III)-iron(III) hydroxide solid
solutions. Inorganic Chemistry, 26: 2228-2232.
Sculley, J. C., 1975. The Fundamentals of Corrosion. Pergamon Press, New York, NY.

References
Snoeyink, V. L. and D. Jenkins, 1980. Water Chemistry. John Wiley and Sons, New York, NY.
Sylvester, P., D. Patel, and T. M611er, 2006. The removal of arsenic and other species by
Arsenx"'', a hybrid anion exchange material. Presentation to the EPRI Workshop on Remediation
of Inorganic Contaminants, July 11-12, 2006, Baltimore, MD.
Taylor, S.W., and P.R. Jaffe, 1990. Biofilm growth and the related changes in the physical
properties of a porous medium 1, experimental investigation. Water Resources Research, Vol.
26, pp. 2153-2159.
Thomasser, R. M., and J. V. Rouse. 1999. In situ remediation of chromium contamination of soil
and ground water. Presentation for the American Wood Preservers Association, May 1999,
Conference on Assessment and Remediation of Soil and Ground Water Contamination at Wood
Treating Sites.
Thullner, M. L. Mauclaire, M.H. Schroth, W. Kinzelbach, and J. Zeyer, 2002. Interaction
between water flow and spatial distribution of microbial growth in a two-dimensional flow
field in saturated porous media. Journal of Contaminant Hydrology, Vol. 58, pp. 169-189.
Tokunaga, S. and T. Hakuta, 2002. Acid washing and stabilization of an artificial arsenic -
contaminated soil. Chemosphere Vol. 46, pp. 31-38.
Tokunaga, S., S. Yokoyama, and S.A. Wasay, 1999. Removal of arsenic(III) and arsenic(V)
ions from aqueous solutions with lanthanum(III) salt and comparison with aluminum(III),
calcium(II), and iron(II) salts. Water Environment Research, Vol. 71, No. 3, pp. 229-306.
U.S. Department of Energy (USDOE), 2005. Performance assessment and recommendations
for rejuvenation of a permeable reactive barrier: Cotter Corporation's Canon City, Colorado,
uranium mill. DOE-LM/GJ816-2005.
litt )://www.liii.tioe. ov/docu[IIerus/4 land/ESL rc Torts/canon city rh. df
U.S. Environmental Protection Agency (USEPA), 1992. Review and recommendations related to
chemical data used in the corrective action regulatory impact analysis (CARIA). Agency memo
dated February 14, 1992.
U.S. Environmental Protection Agency (USEPA), 1996. Soil screening guidance: technical
background document. Office of Solid Waste and Emergency Response, EPA/540/R-95/128.
U.S. Environmental Protection Agency (USEPA), 1997. Technology alternatives for the
remediation of soils contaminated with As, Cd, Cr, Hg, and Pb. Office of Research and
Development Engineering Bulletin, EPA/540/S-97/500.
U.S. Environmental Protection Agency (USEPA), 1998. Permeable reactive barrier
technologies for contaminant remediation. EPA/600/R-98/125.
U.S. Environmental Protection Agency (USEPA), 1999a. Coal -derived humic acid for removal
of metals and organic contaminants. Ground Water Currents, Issue No. 31.
6-7

References
U.S. Environmental Protection Agency (USEPA), 1999b. Pilot -scale testing of a surfactant -
modified PRB. Ground Water Currents, Issue No. 31.
U.S. Environmental Protection Agency (USEPA), 2000. In situ treatment of soil and
groundwater contaminated with chromium. EPA/625/R-00/005.
U.S. Environmental Protection Agency (USEPA), 2001. Phytoremediation of Contaminated
Soil and Ground Water at Hazardous Waste Sites. EPA/540/S-01/500.
U.S. Environmental Protection Agency (USEPA), 2002a. Field applications of in situ
remediation technologies: permeable reactive barriers. Office of Solid Waste and Emergency
Response, iiiip /www eli�i.{zov/tio/`IL)Wllload/ildf/field',il)1prh.l)di'
U.S. Environmental Protection Agency (USEPA), 2002b. Economic analysis of the
implementation of permeable reactive barriers for remediation of contaminated groundwater.
EPA 600/R-02/034.
U.S. Environmental Protection Agency (USEPA), 2002c. Long-term performance of permeable
reactive barriers using zero-valent iron: an evaluation at two sites. EPA/600/S-02/001.
U.S. Environmental Protection Agency (USEPA), 2003. Capstone report on the application,
monitoring, and performance of permeable reactive barriers for ground -water remediation:
vol. 1, performance evaluations at two sites. EPA 600/R-02/045a.
U.S. Environmental Protection Agency (USEPA), 2004. Issue paper on the environmental
chemistry of metals.
U.S. Environmental Protection Agency (USEPA), 2005a. Partition coefficients for metals in
surface water, soil, and waste. Office of Research and Development, EPA/600/R-05/074, July
2005.
U.S. Environmental Protection Agency (USEPA), 2005b. Superfund site progress profile:
Tonolli Corp. 7�Gii/cursites/t•. iiinfo.cfm?id=0301087.
U.S. Geological Survey (USGS), 2006. Mineral commodity summaries.
lit tp:l/niinei als.ustgs.gov/minerals/pubs/mcs/
Watson, D., G. Baohua, W. Goldberg, S. Dunston, and E. Rasor, 1998. Installation and design of
two reactive barriers for treatment of uranium and other contaminants at the S-3 pond site, Oak
Ridge Y-12 plant. Subsurface Barrier Technologies Conference: Engineering Advancements and
Application Considerations for Innovative Barrier Technologies, 26-67 January 1998, Tucson,
AZ. International Business Communications, Southborough, MA, 1998.
Wilkens, J., S.H. Shoemaker, W.B. Bazela, A.P. Egler, R. Sinha, and J.G. Bain, 2003. Arsenic
removal from groundwater using a PRB of BOF slag at the DuPont East Chicago (IN) Site.
Presented at the Research Technology Demonstration Forum Permeable Reactive Barriers
Action Team Meeting, Niagara Falls, NY, Oct. 16.
.:

References
Wright, J. and J. Conca, 2002. Remediation of groundwater contaminated with Zn, Pb and Cd
using a permeable reactive barrier with Apatite H. RTDF PRB Action Meeting November 6-7,
2002.
Zhang, W.X., 2003. Nanoscale iron particles for environmental remediation: an overview.
J. of Nanoparticle Research, V.5, pp. 323-332.
Zhang, Y., M. Yang, X.M. Dou, H. He, and D.S. Wang, 2005. Arsenate adsorption on an
Fe-Ce bimetal oxide adsorbent: role of surface properties. Environ. Sci. Technol. V. 39,
pp. 7246-7253.
6-9

A
PRB REACTIVE MEDIA DESCRIPTIONS
Field Tested Reactive Media
Zero-Valent Iron
The most commonly used barrier medium is zero-valent iron (ZVI) (ITRC, 2005). The iron
used in most PRBs is comprised of a mixture of ductile and cast iron cuttings and borings that
can be obtained from a number of industries. Originally used for its ability to degrade organic
compounds, the material has become increasingly utilized for its ability to immobilize a wide
range of inorganics.
Oxidation of Fe" to Fee+ is a thermodynamically favorable reaction under most environmental
conditions and forms the basis for the technology. The Fe° donates the electrons necessary for
other reduction reactions to occur, such as reductive dechlorination of chlorinated solvents and
chemical alteration of redox sensitive inorganics such as Cr, As, Se, and U. In the presence of
sulfur, some metals may form metal -sulfides. Others may be immobilized by adsorption onto
oxidized iron surfaces such as magnetite.
Morrison et al. (2002) report successful treatment of uranium, selenium, arsenic, molybdenum,
vanadium, and nitrate with a ZVI PRB in Monticello Canyon, Utah. Manganese was not treated
successfully. Groundwater pH within the PRB increased from a maximum of 6.8 in the influent
to 10. Laboratory column tests using ZVI have shown that arsenic can be successfully reduced
to levels below the detection limit (limit not reported) from concentrations as high as 10 mg/L
(Bain et al., 2002).
Organic Matter
Organic matter is commonly used to degrade organic contaminants such as perchlorate
(Craig, 2004). Several different materials have been used, including activated charcoal (GAC),
cottonseed meal, peat moss, lignite, humite, compost, wood chips, leaves, molasses, and whey.
The purpose is to promote biological activity that will act to destroy a contaminant (organics)
or create conditions that are optimal for contaminant immobilization (inorganics).
A common method of metal stabilization uses sulfate reducing bacteria (SRB) to reduce
available sulfate (SO4Z-) to sulfide species (HS- or SZ-):
SO4z- + 2CH20 -> H 2 S + 2HCO
A-1

PRB Reactive Media Descriptions
Mee+ + H2S --> MeS + 2H+
Bicarbonate (HCO,-) is also formed, which acts to buffer pH. The species S2_ is able to react with
metals to form metal sulfide solids. These solids are usually less soluble than a metal's hydroxide
phase, which makes the technology attractive in remediation designs.
A combination of municipal compost, leaf compost, and wood chips has been used at the Nickel
Rim Mine Site in Ontario, Canada to remediate nickel, iron, and sulfate in groundwater (Benner
et al., 1997).
Phosphate -Based Precipitation
Biogenic apatite, known as Apatite IITM, has been used in a PRB application at Success Mine,
Idaho (Wright and Conca, 2002). The material is mostly porous amorphous hydroxyapatite
(calcium phosphate), and is soluble enough to release phosphate ions at concentrations exceeding
the solubility limit of many metal -phosphate phases. The media is also proposed to reduce
contaminant concentrations by three other processes: 1) cations can substitute for Ca, 2)
oxyanions can replace structural PO4'-, and 3) anions F- and CY can exchange with OH-.
The affinity for cations is proposed as:
UO22+>Pb2+>Th4+>Cd2+>Mn2+-Zn 2+>Cu2+_SbO+-Hg2+>Ni2+>Sr2+>B a2+.
Removal of oxyanions has proven less successful, but follows the approximate order of
preference:
V 04,->MoO42->SeO32->AsO43->CrO42->TcO4 .
The method, termed phosphate induced metal stabilization (PIMS), has been demonstrated to
immobilize Cd, Zn, Pb, U, and Pu by precipitating stable metal -phosphate phases or other low -
solubility phases. Major effects of the barrier include substantial downgradient reductions of Cd,
Zn, and Pb, accompanied by a shift in pH from -4.5 upgradient to 7 downgradient (Bostick et al.,
1999).
Limestone and Hydrated Lime
Limestone (CaCO) has been used to treat a variety of contaminants due to its ability to buffer
pH and to provide a ligand (CO,'-) for solid formation. Limestone rnaterials have been used
extensively for reducing the effects of acid mine drainage (USEPA, 2002a). Buffering of pH is a
beneficial by-product of using limestone because some contaminants are effectively immobilized
by hydroxide formation (e.g., Be(OH)2). Others, such as lead, can form insoluble carbonates.
Hydrated lime, Ca(OH)21 can raise pH so that metal hydroxides such as Be(OH)2 are more
effectively precipitated.
A-2

PRB Reactive Media Descriptions
A mixture of limestone, organic matter, and sand was used at the Nickel Rim Mine Site in
Ontario, Canada. The limestone provides a pH buffer, the organic matter stimulates microbial
activity, and the sand provides increased permeability.
Zeolites and Surfactant -Modified Zeolites
Zeolites are hydrated alu mi no silicates with cage -like structures that are commonly used
as sorptive substrates due to their large surface areas and high cation exchange capacities.
Hydraulic characteristics make zeolites extremely effective filtration materials. The
predominantly negative surface charge of zeolites makes them much more selective for cations
than for anions (Haggerty and Bowman, 1994). Adsorption onto zeolites and surfactant -modified
zeolites is pH -dependent.
Surface charge of zeolites can be altered with cationic surfactants to promote sorption of anions,
in addition to cations. Hexadecyltrimethylammonium (HDTMA) is a common and low-cost
surfactant that is commonly used for this purpose. Field demonstrations have utilized zeolites for
removal of Sr-90, and some zeolites have been suggested as an immobilization mechanism for
boron (McGregor et al., 2002). Recent experiments at the University of Alabama have shown
that HDTMA-treatment of clinoptilolite greatly improves adsorption of As, Cr, Mo, Se, and V,
reduces adsorption of Ba, K, Na, and Sr, and slightly improves B adsorption (Donahoe, 2005,
2006b). However, another study showed a decrease in Cr immobilization effectiveness for a
HDTMA modified zeolite after flushing a column with 400 pore volumes of clean water (Li,
2006). Based on this decrease, the author suggested that caution should be exercised when using
SMZs in a PRB.
Basic Oxygen Furnace Slag
Basic oxygen furnace slag (BOF) is a mixture of material ranging in grain size from silt to fine
gravel and is the nonmetallic'waste by-product of steel production generated at several steel
plants. It contains various oxides and silicates of iron, calcium, magnesium, and aluminum. It is
commonly used as aggregate in roadbed and other construction projects. The material has a high
sorptive capacity and can buffer pH from acidic to alkaline conditions (ITRC 2005).
BOF has been used in a PRB to treat arsenic -impacted groundwater at an industrial site in
Chicago in 2002. Data from the first two years of operation indicate removal of arsenic from
concentrations greater than 1 mg/L to less than 0.001 mg/L within the barrier (Wilkens et al.,
2003). BOF is also capable of removing zinc, lead, and sulfate from solution (Blowes et al.,
2006)
Sodium Dithionite (NaS Od and Polysulfide Compounds
Sodium dithionite is a chemical reductant that is capable of converting ferric oxides and
hydroxides in soils to ferrous iron (Fey' -> Fe"). The re -oxidation of ferrous iron to ferric iron
has the ability to reduce other metal contaminants to immobile forms. Laboratory experiments
have shown that the effectiveness of this technology is limited to easily reducible metals, such
as uranium (Gavaskar et al., 1998). Amonette et al. (1994) and Cummings and Booth (1997)
have demonstrated sodium dithionite to be an effective treatment to reduce aqueous chromium
concentrations.
A-3

PRB Reactive Media Descriptions
Polysulfide compounds (S,z-, S4Z-, SSZ-, S1z-) are also capable of chemically reducing inorganic
pollutants such as Cr". The reduced sulfur compounds are readily oxidized by means of reducing
Cr`1 to Cr". Calcium polysulfide (CASCADE®) has been shown to effectively reduce chromium
concentrations at a wood treatment facility in Ukiah, California (Thomasser and Rouse, 1999).
BauxsolTM ViromineTm Acid-B Extra TM
BauxsolTM, marketed by Virotec, is a manufactured dry red solid composed of different minerals,
including hematite, boehmite, gibbsite, sodalite, quartz, cancrinite, brucite, calcite, diaspore,
ferrihydrite, gypsum, hydrocalumite, hydrotalcite, p-aluminohydrocalcite, portlandite, minor
aragonite, and other trace minerals (Clark, 2000; McConchie et al., 1999). Mixtures of sand
and BauxsolTM have been tested in column experiments for their ability to remove metal
contaminants (Munro et al., 2004), indicating metal removal by M(OH)' precipitation and other
processes such as sorption. ViromineTM and Acid-B ExtraTM are also marketed by Virotec as
sorbents for contaminants.
Other Proposed Reactive Media for Inorganic Constituents
HumasorbTm
HumasorbTM, developed by ARCTECH, Inc., is a water insoluble lignite -derived humic acid
with the ability to sorb metal cations, organic contaminants, and radionuclides (USEPA, 1999a).
Amorphous Ferric Oxide
Amorphous ferric oxide (AFO) has been tested at pilot scale for uranium removal at the
Department of Energy Fry Canyon, Utah site (Feltcorn and Breeden, 1997). It was shown to be
less effective at uranium treatment than zero-valent iron and bone char. Amorphous ferric oxide
has been shown in laboratory studies to remove large amounts of arsenic and selenium from
solution by adsorption. It is likely that AFO can sequester large amounts of various trace
elements in a barrier application, and that the effectiveness will be pH -dependent (ITRC, 2005).
Dissolved Oxygen Barriers
Certain redox-sensitive elements may be successfully immobilized by an oxidizing zone, rather
than a reducing zone. For example, it is well -documented that As(III) is poorly sorbed to aquifer
material compared to As(V). In aquifers that are mildly reducing or only weakly oxidizing, it
may be appropriate to create an oxidizing zone that will convert less reactive species such as
As(III) to the more reactive form.
There are limitations to this concept. An aquifer with abundant organic material may
consume any oxygen that is introduced at a faster rate than it can be replenished. Additionally,
replenishment of oxygen is an added maintenance and expense that is preferably avoided in
passive treatment systems. The best use of this technology may involve an aquifer with an

PRB Reactive Media Descriptions
oxygen consumption rate that will not readily scavenge dissolved oxygen. Slow -release
technologies are available (ORCO) that allow continued release of dissolved oxygen over
extended periods of time. Other possibilities include the direct introduction of oxygen through
a sparging system.
Diatomaceous Earth
Diatomaceous earth is a fossilized mineral, formed from the accumulation of silica -based diatom
skeletons. Jang et al. (2005) have shown that hydrous ferric oxide (HFO) impregnated diatomite
showed better arsenic removal capacities than ZVI or granular activated carbon.
Activated Alumina
Activated alumina is produced as a porous, dehydrated alumina oxide. It has a surface
complexation affinity for cations and anions, greater than basic oxygen furnace slag. McRae et
al. (1999) found that a mixture of 10 percent basic oxygen furnace slag, 20 percent activated
alumina with agricultural limestone, and silica sand had a high capacity to remove arsenic from
groundwater in column experiments.
Ion Exchange Resins
A hybrid ion exchange material is prepared by modifying a traditional ion exchange resin.
Iron oxide nanoparticles are impregnated into the surface of the ion exchange resin. The newly
created surface has unique properties for adsorbing contaminants from water. For example, one
ion exchange resin, marketed under the name ArsenX" ' by Purolite, can adsorb both As(III) and
As(V), and a number of other constituents including P, V, Ra, U, F, Cr, Mo, Se(VI), and possibly
Sb (Sylvester et al., 2006). The same manufacturer manufactures another ion exchange resin (S-
108) designed to immobilize boron.
ADSORB/ATm (Titanium Oxide -Based Adsorbent)
ADSORBIATM is a titanium -based media produced by DOW that is marketed for its ability to
adsorb arsenic. According to the manufacturer, it removes both arsenate and arsenite over a wide
range of pH without the need for pretreatment.
SORBPLUST44 Adsorbent (Mg -Al oxide)
SORBPLUSTM is a Mg -Al oxide sorbent developed by ALCOA for removal of chelated nickel,
copper, cobalt, hexavalent chromium, and metal-complexed cyanides from waste streams.
Evanoff et al. (1992) found that chromium concentrations in chemical film rinse water could
be reduced from as much as 20 mg/L to consistently less than 0.2 mg/L by the sorbent.
A-5

PRB Reactive Media Descriptions
Ferrous Sulfate
Ferrous sulfate (FeSO4.71-1,0) has also been used in chromium remediation designs. The
oxidation of ferrous iron (Fe") to ferric iron (Fe") is proposed to reduce Cr6+ to Cr". A field
application of ferrous sulfate to treat chromium at the Townsend Saw Chain site in Pontiac,
South Carolina, showed that, despite ultimate reduction of chromium concentrations in the
aquifer, initial concentrations of chromium increased due to displacement of sorbed chromium
by excess sulfate ions (USEPA, 2000).
Ferrous sulfate heptahydrate has been used successfully by Southern Company Services, Inc.
to treat arsenic -contaminated soils (Redwine et al., 2004). It has been shown that ferrous iron
precipitates as ferrihydrite in the subsurface and strongly adsorbs arsenic that would otherwise be
mobile. Batch tests, column tests, and a field scale demonstration have proven that newly formed
iron compounds and adsorbed arsenic are stable over time.
Ferrous sulfate has been tested as an additive to ash sluice water, and preliminary experimental
results have suggested that it can reduce the leaching potential of As, B, Cr, Mo, Ni, and V from
coal fly ash samples (Donahoe, 2006b).
FORAGERTm Sponge
Dynaphore, Inc. has developed a cellulose sponge by the name of FORAGERTM Sponge that
is open -celled cellulose containing iminodiacetic acid groups that chelate metal cations. The
sponge polymer also contains tertiary amine salt groups that can bind anionic contaminants.
According to Dynaphore, Inc., it can be designed for site -specific needs, such that a contaminant
of interest can be targeted for removal. For example, the sponge can be pre -loaded with ferric
iron to facilitate precipitation of ferric arsenate.
The technology was demonstrated at the National Lead Industry Site, Pedricktown, NJ, in
a mobile pump -and -treat system (Ott, 2000). The demonstration found that reductions in
contaminants were observed for Cu (>94% removal), Cd (-89% removal), and Pb (-96%
removal), but not for Cr (-32% removal).
Rare Earth Elements
Rare earth elements, particularly of the lanthanide series, have been studied for their ability to
remove arsenic and selenium from solution. Tokunaga and Hakuta (2002) found that salts and
oxides of lanthanum and cerium were effective in immobilizing arsenic in soil. Tokunaga et al.
(1999) reported that lanthanum ions outperformed salts of aluminum, polyaluminum chloride,
calcium, and ferric iron for removal of pentavalent arsenic from solution. Harck et al. (2004)
have patented a process for removing arsenic and selenium from solution using a concentrate of
lanthanum oxide and various oxides (U.S. Patent 6,800,204, issued 2004). Because of their high
costs, these elements are used mostly as minor additives to an iron -based sorbent. Lanthanum -
and cerium -doped iron oxide minerals exhibit high sorptive capacity for both arsenic and
selenium, but Zhang et al. (2005) report that Fe-Ce exhibited the highest removal capacity
for arsenic over the widest pH range (pH 3 to 7).

PRB Reactive Media Descriptions
KanchanTm Arsenic Filter
The KanchanTM Arsenic Filter has the ability to remove arsenic, iron, and pathogens from water
through filtration. It is constructed of PVC pipes, iron nails, brick, sand, and gravel. Arsenic is
removed by adsorption to rusted iron nails inside the filter. Solubilized iron and pathogens are
then removed by physical straining in a fine sand layer. The average arsenic removal efficiency
is between 90 and 93 percent. Modifying the use of iron nails may be an effective and cheap
media for PRB emplacement. The filter is jointly implemented by the Massachusetts Institute
of Technology (MIT), the Nepal -based non -governmental organization Environment and Public
Health Organisation (ENPHO), and the Rural Water Supply and Sanitation Support Programme.
Granular Ferric HydroxideO
USFilter has developed a ferric iron based media called GFH@ (Granular Ferric Hydroxide@)
that can remove arsenate, arsenite, phosphate, antimony, selenium, copper, and chromium from
solution. According to the supplier, this media is currently being tested for water supply wells in
Phoenix, AZ and has successfully removed arsenic to levels below 10 ppb.

B
PRB CASE STUDIES
The following pages present, in no particular order, summaries of PRB case studies from around
the world that involve remediation of inorganic constituents. PRB case study 1 is the only
example involving a CCP site. Note that some case studies describe applications that were
unsuccessful in remediating one or more constituents. Most case studies can be found on the
internet. References are given as sources of supplemental information.
CCP Landfill, Ontario, Canada
Reference: McGregor et al., 2002; Blowes et al., 2006
Site background: An unlined CCP landfill is underlain by a thick sand and gravel aquifer.
The landfill covers an area of 19 acres (7.7 Ha). Groundwater velocity in the aquifer ranges
from 23 to 165 ft/yr (7 to 50 m/yr). A plume containing As (up to 0.42 mg/L), B (35 mg/L),
Cr (0.19 mg/L), Mo (0.97 mg/L), Se (1.75 mg/L), V (1.0 mg/L), and SO, (2,000 mg/L) was
mapped downgradient from the site heading in the direction of a lake.
Contaminants: As, B, Cr, Mo, Se, and V
Reactive media: A) Wood chips, bottom ash, surfactant modified zeolite, ZVI, and boron ion -
exchange resin; B) ZVI, surfactant modified zeolite, bottom ash, and boron ion -exchange resin;
and C) ZVI, surfactant modified zeolite, and bottom ash.
Demonstration: A PRB was installed in a trench 42 feet (13 m) in length, 6 feet (2 m) wide,
and 12 feet (4 m) deep. The trench was divided along its length into three treatment zones, each
containing a different mixture of reactive media. All of the zones contained a surfactant modified
zeolite, which was hypothesized to have potential for immobilizing boron, and two of the
three zones contained a small percentage of boron ion -exchange resin. The final hydraulic
conductivity of the barrier was approximately 3 x 10-' cm/s, a factor of 10 to 15 greater than
the surrounding aquifer.
After nearly four years of monitoring the following observations were made:
• Reactive media mixture A was very effective in reducing concentrations of As, B, Cr, Mo,
and V, and moderately effective for Se.
• Mixture B was less effective than mixture A for arsenic and boron.
• Mixture C was less effective than mixture A for arsenic, and ineffective for boron.
• None of the mixtures reduced sulfate concentrations appreciably, although sulfate was not a
focus of this study.

PRB Case Studies
The authors concluded that the boron ion -exchange resin was more effective than surfactant
modified zeolite in immobilizing boron, and suggested that the adsorption surfaces of the zeolite
may have become saturated after only seven pore volumes.
Former DOE Mill Site, Monticello, UT
Reference: Morrison et al. (2002); []ILI)://www.r(df.or­
Site background: The former Monticello mill site was built in 1942 and operated as a uranium
and vanadium ore -processing mill. Beneath the repository, two aquifers exist —a perched alluvial
aquifer and the regional Burro Canyon aquifer beneath the alluvial aquifer. The perched aquifer
was contaminated by mill tailings prior to construction of the repository.
Contaminants: U, Se, V, As, Mn, Mo, Fe
Reactive media: ZVI
Demonstration: A funnel -and -gate system was constructed downgradient of the contaminant
plume. Two slurry walls funnel the contaminated plume through the gate, containing a PRB of
zero-valent iron. The south slurry wall is 240 feet (73 m) long and the north wall is 97 feet
(30 m) long; both are constructed of a bentonite and soil slurry mix. The barrier was built by
driving steel sheet piling into the bedrock, forming a box approximately 100 feet (30 m) long
by 8 feet (2.4 m) wide. The soils were replaced by ZVI and gravel. A downgradient gravel pack
2 feet (0.6 m) wide contains an air sparging system designed to remove Mn and Fe if
concentrations become too high.
Concentrations of As, Se, U, and V have been reduced to non -detectable levels (detection level
not reported). Concentrations of Fe increase as groundwater passes through the barrier, although
Fe concentrations are lower than expected and are well within acceptable risk ranges.
Savannah River Site TNX Area, Aiken, SC
Reference: Phifer et al. (2005); littp://www.rtdLoi-t,
Site background: The TNX Area at the Savannah River Technology Center was used for pilot -
scale testing and evaluation of various chemical processes associated with the Savannah River
Site. Contamination has been detected in the water table aquifer. This aquifer is approximately
35 to 40 feet (11 to 12 m) thick and is comprised of interbedded sand, silty sand, and relatively
thin clay layers. The aquifer has a horizontal hydraulic conductivity of 65 ft/d (2.3 x 10-Z cm/s),
a vertical hydraulic conductivity of 30 ft/d (1.1 x 10-' cm/s), an effective porosity of 0.15, a pore
velocity of 3 ft/d (0.9 m/d), and a horizontal gradient of 0.007.
Contaminants: NO3
Reactive media: ZVI
ONE

PRB Case Studies
Demonstration: The TNX GeoSiphon''m Cell is a large -diameter (8 feet, 2.4 m) well containing
ZVI. The cell passively induces flow using a siphon to the nearby Savannah River. So far,
effective treatment of nitrate has been observed.
Haardkrom Site, Kolding, Denmark
Reference: Kjeldsen and Fuglsang (2000); hitp://www.Lclf.oF
Site background: The site formerly hosted an electroplating facility. The plating process
involved the use of chromium, nickel, zinc, and TCE. Contaminants of major concern are Cr
`'+
and TCE, and concentrations range from 8-110 mg/L and 40-1,400 µg/L, respectively. Due to
aquifer heterogeneity, concentration levels vary significantly from point to point. The aquifer is
less than 6.6 feet (2.0 m) below ground surface and is not continuous through the site. The upper
6.5 to 10 feet (2.0 to 3.0 m) consists of a low permeability, heterogeneous mixture of sandy and
clayey loam interspersed with local lenses of sandy layers.
Contaminants: Cr, TCE
Reactive media: ZVI
Demonstration: Laboratory experiments showed chromate reduction capacities of 1 to 3 mg Cr"
per g Fe°. The designers accordingly set the dimensions of the trench and barrier to accommodate
all of the Cr" in the contaminant plume. A continuous trench system PRB was installed. The
PRB is 164 feet (50 m) long, 3.3 to 9.8 feet (1.0 to 3.0 m) deep, and 3.3 feet (1.0 m) thick.
Bypass trenches and recirculation pipes were installed to increase water flow through the low
permeability, heterogeneous aquifer. Results suggest the design is not effectively controlling the
uneven distribution of Cr, and the chromate removal capacity has been exhausted due to
heterogeneous loading of the PRB.
Y-12 Site, Oak Ridge National Laboratory, Oak Ridge, TN
Reference: Watson et al. (1998); hfip:Hww�v.i,ttit'.org.
Site background: Originally this site was a disposal pond for a DOE laboratory from 1952 to
1981. The area was capped in 1983, but both the groundwater and surface water are considered
to be contaminated. The site has a very low permeability and the soil is mostly unconsolidated
clay with overlying fractured shale. It is approximately 10 to 15 feet (3 to 5 m) to groundwater,
where the aquifer is approximately 10 to 20 feet (3 to 6 m) thick.
Contaminants: HNO3, U, Tc
Reactive media: ZVI
Demonstration: This system consists of two separate PRBs. The first PRB was installed in
November of 1997 and consists of a continuous trench system 225 feet (69 m) long, 2 feet
(0.6 m) wide, and 22 to 30 feet (6.7 to 9.1 m) deep. It is filled with approximately 80 tons of
MA

PRB Case Studies
zero-valent iron and emplaced parallel to groundwater flow. The second PRB is a funnel -and -
gate system designed to direct groundwater flow into a concrete vault to test treatment with
different kinds of media. The total cost for the system was approximately $1,000,000.
The system was enhanced in 1999 to improve treatment efficiency by extending the trench
system by approximately 100 feet (30 m). This served to increase the groundwater treatment
zone to other affected areas on the site. This enhancement was also an excellent deterrent to
changes in groundwater flow that resulted in PRB bypass.
U.S. Coast Guard Support Center, Elizabeth City, NC
Reference: Puls et al. (1998); litip://www.r[df,or,,
Site background: A groundwater plume containing hexavalent chromium and TCE exists near
a former electroplating shop that operated until 1984.
Contaminants: Cr(VI), TCE
Reactive media: ZVI
Demonstration: The PRB is a continuous wall design with dimensions of 150 feet (46 m) in
length, 24 feet (7.3 m) deep and 2 feet (0.6 m) thick. It was installed using a continuous trencher
method and filled with a ZVI reactive material. Continued monitoring of the site showed total
chromium removal within the first six inches of the wall.
Fry Canyon Site, Fry Canyon, UT
Reference: Feltcorn and Breeden (1997); liitl)://www.rt.tll`.org
Site background: Fry Canyon Site, UT, is an abandoned uranium upgrader site. The water table
is located 8 to 9 feet (2.4 to 2.7 m) below ground surface. The shallow aquifer is comprised of
colluvial material, with a groundwater flow rate of approximately 1.5 ft/d (0.5 m/d).
Contaminants: U
Reactive media: ZVI, AFO, PO4
Demonstration: Field -scale demonstration is underway, testing performance of three funnel -
and -gate barriers: ZVI, AFO, and PO4. Objectives of the demonstration include: 1) hydrologic
and geochemical characterization of the site, 2) design, installation, and operation of the three
barriers, and 3) evaluation of barrier performance.
Each barrier is 7 feet (2.1 m) wide, 3 feet (1.0 m) thick, and 4 feet (1.2 m) deep. Approximately
110 ft' (3.1 m') of material was used in each barrier. Groundwater velocities through the barriers
are approximately 4.5 ft/d (1.4 m/d). The ZVI and PO, barriers are removing over 99 percent of
U, but the AFO reached uranium breakthrough after about 1,000 pore volumes.
I' "

PRB Case Studies
Bodo Canyon, Durango, CO
Reference: ht1p:Ilww\,a.r1d1'.c7; Bronstein, 2005
Site background: Uranium mill tailings were relocated to the Bodo Canyon disposal cell in the
fall of 1990. Contaminated seeps developed downgradient shortly after construction.
Contaminants: As, Mo, Se, U, and V
Reactive media: ZVI, copper wool, steel wool
Demonstration: Four PRBs were installed in Bodo Canyon as a pilot -scale demonstration to
treat contaminated groundwater and test the efficiency of PRBs for remediation of metals and
uranium. In order to compare different designs, four PRBs were installed near the retention pond,
each containing a form of zero-valent iron media.
Results:
• As reduced from up to 186 µg/L to 2.2 µg/L
• Mo reduced from 1180 µg/L to 359 µg/L
• Se reduced from 337 µg/L to 5.9 µg/L
• Gasses (HZ and CH,) built up in PRB and required venting
Rocky Flats Environmental Technology Site (Solar Ponds Plume), Golden,
CO
Reference: litti)://www.rtdf.orL,
Site background: At the Rocky Flats Environmental Technology Site in Golden, CO, past waste
storage practices have resulted in groundwater contaminated with nitrate and uranium. The Solar
Ponds were drained and sludges removed by 1995, but contaminated groundwater has migrated
downgradient to a nearby stream.
Contaminants: NO,, U
Reactive media: ZVI and wood chips
Demonstration: Bench scale studies were conducted at the University of Waterloo. The
groundwater collection system extends approximately 1,100 feet (340 m). Excavations were
performed at variable depths between 20 to 30 feet (6 to 9 m) below ground surface and
approximately 10 feet (3 m) into underlying clay. The barrier consists of HDPE panels. The
concrete treatment cell is divided into two sections. Treatment media occupies the lower 10 feet
(3 m) of each section. The first cell contains a mixture of sawdust and leaf mold with 10 percent
ZVI by weight. The second cell is filled with ZVI. Water exiting the treatment cell typically
contains less than 5 mg/L nitrate (from 140-170 mg/L) and less than 1 pCi/L uranium (from 20-
28 pCi/L).
M.

PRB Case Studies
Nickel Rim Mine Site, Sudbury, Ontario, Canada
Reference: Benner et al. (2000); i7ttl�:ll�� a+w_.► tdl.iyr
Site background: Nickel Rim was an active mine from 1953 to 1958. The contamination has
resulted from 40 years of oxidation of a tailings impoundment on site. The site is underlain by an
aquifer 10 to 26 feet (3.0 to 7.9 m) thick composed of glacio-fluvial sand. The aquifer is confined
to a narrow valley, bounded on both sides and below by bedrock. Groundwater velocity is
estimated to be 49 ft/yr (15 m/yr).
Contaminants: Ni, Fe, and SO4. Initial concentrations were 2,400-3,800 mg/L SO4, 740-
1,000 mg/L Fe, and up to 10 mg/L Ni.
Reactive media and construction: Organic carbon (mixture containing municipal compost, leaf
compost, wood chips, and pea gravel to increase hydraulic conductivity) as a continuous barrier.
Demonstration: A continuous PRB was installed in August 1995 using a cut and fill method.
The reactive barrier is 50 feet (15 m) long, 14 feet (4.3 m) deep, and 12 feet (3.7 m) wide, for
a total of 8,400 ft3 (240 m3) of media. Coarse sand buffer zones were installed upgradient and
downgradient of the reactive material, and the PRB was capped with 12 in. of clay to minimize
entry of surface water and oxygen. Remediation was accomplished by sulfate reduction and
metal sulfide precipitation.
Nine months after installation, sulfate concentrations had decreased to 110-1,900 mg/L and
iron concentrations decreased to <1-91 mg/L. Dissolved Ni decreased to <0.1 mg/L within
and downgradient of the PRB. Groundwater pH increased from 5.8 to 7.0 across the barrier,
and the PRB, converted the aquifer from acid -producing to acid -consuming. Significant hydraulic
heterogeneities exist in the barrier, probably due to poor mixing of the media, and/or air pockets
from installation.
Tonolli Superfund Site, Nesquehoning, PA
Reference: USEPA, 2005b
Site background: The Tonolli Corporation operated a battery recycling plant and a lead
smelting plant at the site from 1974 until 1986. Elevated levels of dissolved metals are attributed
to both waste sources and anthropogenic sources, including dumping of battery acid and acid
mine drainage effect from spoil piles. The contaminants are located in a coal mine spoil at 0 to
19 feet (0 to 5.8 m) and in alluvium from 74 to 113 feet (23 to 34 m). Maximum concentrations
of these contaminants encountered were 328 µg/L of Pb, 77 µg/L of Cd, 313 µg/L of As,
1,130 µg/L of Zn, and 140 µg/L of Cu.
Contaminants: Pb, Cd, As, Zn, and Cu
Reactive media: Limestone
II

PRB Case Studies
Demonstration: In 1998, a continuous trench PRB was installed. A trackhoe was used to
excavate a trench, approximately 3 feet (1 m) wide, 20 feet (6 m) deep, and 1,100 feet (340 m)
long. Trench boxes were installed parallel to a creek. Lead concentrations are being reduced to
less than performance standards (not reported). As and Sb have shown increases in concentration
downgradient of the landfill.
Public School, Langton, Ontario, Canada
Reference: Baker et al. (1998);
lrlt 11 www Science l�watex loo calrese{irci�i �li'ermeableReactiveBarrie� 511'hc sph to Tretitnickit/
Phosphate. T reattnent.html
Site background: Not reported.
Contaminants: PO4, NO3
Reactive media: Fe/Ca oxides, limestone, wood chips
Demonstration: This funnel -and -gate PRB was installed on a septic system on school property
to remove nitrates and phosphates. After six years of operation, decreases in both contaminants
were noted in samples from downgradient wells.
Chalk River Laboratories, Ontario, Canada
Reference: l�tt p Jfw w w . rttl l'.�>a
Site background: In the early 1950s, a pilot plant was operated at Chalk River for the purpose
of decomposing and reducing the volumes of ammonium nitrate solutions that contained mixed
fission products. Some of these solutions were released into pits lined with crushed limestone.
In 1998, Atomic Energy of Canada, Ltd. installed a wall and curtain PRB to remove Sr-90 from
groundwater. The site is underlain by sands derived from granitic gneiss. The saturated thickness
of sandy aquifer ranges from 16.4 to 42.6 feet (5.0 to 13 m).
Contaminants: Sr-90
Reactive media: Clinoptilolite (Zeolite)
Demonstration: The PRB is a steel cut-off wall, a curtain of zeolite to treat the water, and a
subsurface bypass drainage system for non -contaminated, overlying groundwater. A granular
curtain of 153.4 yd' of 14x50 mesh clinoptilolite (zeolite) is positioned in front of the cut-off
wall. The curtain is 6.6 feet (2.0 m) long, 36.1 feet (11 m) wide, and 18 feet (5.5 m) deep.
The PRB cost a total of $300,000.
Groundwater outflow meets Canadian drinking water standards and the PRB has retained
100 percent of the contaminant since 1998. The wall and curtain have exhibited good
performance chemically and physically and require almost no cost for routine monitoring
of performance and to adjust capture zone dimensions.

PRB Case Studies
Large Experimental Aquifer Program (LEAP) Demonstration Facility,
Portland, OR
Reference: ltttlK4A :�Lw.it(r orr17; Haggerty and Bowman (1994); USEPA (1999b)
Site background: The LEAP facility is a PRB demonstration facility, located in Portland, OR.
Contaminants: Cr(VI), PCE
Reactive Media: Surfactant modified zeolites (SMZ)
Demonstration: The barrier construction is a hanging barrier in a perforated metal frame. This
site was created and intentionally contaminated for research purposes. Overall, the retardation
factors for each contaminant were on the order of 50. The system was designed for a sorption
method of remediation.
DuPont Site, East Chicago, IN
Reference: Wilkens et al. 2003; ITRC, 2005
Site background: DuPont purchased the site from Grasselli Corp. in 1928. The site is 440 acres
(180 ha) and was a diversified chemical manufacturing facility. This site is the first full-scale
permeable reactive barrier site to remove arsenic from groundwater using basic oxygen furnace
slag.
Contaminants: As (1-2 mg/L)
Reactive media: Basic oxygen furnace slag (BOF)
Demonstration: In 2002, a full-scale PRB was installed that consisted of 100 percent BOF slag.
The continuous trench PRB is 2,000 feet (610 m) long and 35 feet (11 m) wide, consisting of two
parallel trenches to achieve desired width. Results have shown a decrease from 1 to 2 mg/L to
<0.001 mg/L in the effluent.
Gilt Edge Mine, SD
Reference: McConchie et al. (1999)
Site background: The Gilt Edge Mine site is an open pit, cyanide heap leach gold mine,
developed in both oxidized and highly sulfidic ore bodies. Mining operations for gold, copper,
and tungsten were conducted since 1876. Currently the site hosts 150 million gallons of acidic,
heavy metal -laden water in three open pits and millions of cubic yards of acid -generating waste
rock.
Contaminants: As, Cd, Co, Cu, Pb, Zn, NO3, and SO4.

PRB Case Studies
Reactive media: ViromineTMAcid-B ExtraTM mixed as 10 percent (weight) with waste rock.
Demonstration: A trial was conducted for waste rock remediation. A series of 200-L drums
were filled with waste rock and Acid-B ExtraTM mixture (10%, 5%, and 2%) to determine
effective treatment at the lowest possible application rates. Results from the drum trial show
that 6 to 7 percent is the optimal addition rate for the reactive media. Leachate pH was raised
from 1.92 to 7.21, arsenic was reduced from 23,000 µg/L to <5 µg/L, iron was reduced from
19,000,000 ppb to 33 ppb, and concentrations of other trace metals (Sb, Cd, Cr, Pb, and Ni)
were all reduced to near or below detection limits (limits not reported).
A shallow lined trench containing about 20 cubic meters of waste rock mixed with 10 percent
(weight) Acid-B extra reagent was constructed. The leachate was sampled monthly, but monthly
data was not available. Ongoing three-year results from the trench trial are shown below. Data
is reported from Virotec, and has been validated by CDM Federal Programs Corporation.
• Leachate pH was raised from 1.93 to 7.9
• As was reduced from 35,000 µg/L to <4 µg/L
• Fe from 21,000,000 µg/L to 210 µg/L
• Sb from 500 µg/L to <10 µg/L
• Cd from 630 to <10 µg/L
• Cr from 390 µg/L to <10 µg/L
• Pb from 390 µg/L to <10 µg/L
• Mn from 34,000 to <10 µg/L
• Ni from 1,600 to <10 µg/L
100 D Area, Hanford Site, Benton County, WA
Reference: Naftz et al., 2002; Bronstein, 2005;
ilttp-.//www.rL-d1,cri InublicllaecrttL�ar�lpihsumn�sfproiilc.cfm?mid=43
Site background: The Hanford Site created plutonium from 1943 until the 1980s as part of the
Manhattan Project. The site has been divided into four separate NPL sites, one of which is named
the 100 D Area. A hexavalent chromium plume was detected in groundwater, which is about
85 feet (26 m) below ground surface. Average groundwater velocity is approximately 1 ft/day
(0.3 m/d).
Contaminants: Cr(VI)
Reactive media: Sodium dithionite injection
Demonstration: Not truly a PRB, because no solid media is used, this demonstration utilizes
"in situ redox manipulation" to create a reducing zone in the area of injection. In 2003, sodium
dithionite with a potassium carbonate/potassium bicarbonate buffer was injected into the
e

PRB Case Studies
hexavalent chromium plume. The redox zone was created by 65 injection wells over a
length of 2,000 feet (610 m). The zone is located parallel to the Columbia River. Chromium
concentrations in 59 of 66 wells are below the detection limit (0.008 mg/L). Performance
monitoring is complicated by seasonal fluctuations associated with the river and preferential
flow paths created by well installations.
Success Mine and Mill, Wallace, ID
Reference: Conca and Wright, 2006; Bronstein, 2005
Site background: The Success Mine and Mill site, located in Northern Idaho, was the
largest metals loader in the Ninemile Creek drainage area of the Coeur d'Alene mining district.
Groundwater contamination results from drainage of a tailings/waste rock pile that is 1,200 feet
(370 m) long and 150 feet (46 m) high. Hydraulic conductivity of the sand and gravel aquifer is
1.7 x 10-' ft/s (5.2 x 10-' cm/s). A shallow bedrock (quartz monzonite) aquifer exists below the
sand and gravel aquifer and has a hydraulic conductivity of approximately 5.6 x 10"5 ft/s
(1.7 x 10-' cm/s).
Contaminants: Pb, Zn, Cd, SO,, and NO,
Reactive media: Apatite IITM
Demonstration: Phosphate induced metals stabilization (PIMS) was used at this site in the
form of a 13.5 feet (4.1 m) high, 15 feet (4.6 m) wide, and 50 feet (15 m) long PRB. The PRB
is constructed of two cells, each measuring 8.0 feet (2.4 m) high, 6.5 feet (2.0 m) wide, and
45 feet (14 m) long. One cell contains 100 percent Apatite IITM, and the other contains 50 percent
Apatite IITM and 50 percent gravel. The PRB was keyed into underlying bedrock. A hydraulic
drain was installed upgradient of the PRB to direct flow into the cells. Cadmium concentrations
were reduced from 0.436 mg/L upgradient of the wall to <0.002 mg/L downgradient, Pb from
0.658 mg/L to <0.005 mg/L, and Zn from 68 mg/L to 0.034 mg/L. pH is buffered from 4.9 to
6.9 through the wall.
After 3.5 years, less than 40 percent of the media has been spent and the wall remains effective.
The PRB is anaerobic and creates conditions optimal for sulfate reducing bacteria. The wall is
expected to provide treatment for Cd and Pb for up to 30 years, but only a few years for Zn. Zinc
is currently being attenuated in sulfide phases.
Cyprus AMAX Minerals Company/AMAX Realty Development, Inc.,
Carteret, NJ
Reference: Bronstein, 2005
Site background: This site is a former copper smelting facility. Groundwater is contaminated
with Cu, Ni, Se, and Zn. The groundwater discharges into a nearby estuary.
Contaminants: Cu, Ni, Zn

PRB Case Studies
Reactive media: Dolomitic limestone and powdered sodium carbonate
Demonstration: A 685-foot (209-m) long and 45-foot (14-m) deep trench was installed in 1993,
and extended by 200 feet (61 m) in 2000. The trench was filled with 2,600 tons of dolomitic
limestone and 20 tons of sodium carbonate. Downgradient wells have shown increases in Ni and
Zn since barrier emplacement. Selenium concentrations have decreased slightly from 2.5 mg/L,
and no results have been reported for Cu.
E.I. DuPont, Newport Superfund Site, DE
Reference: Bronstein, 2005
Site background: The site is currently occupied by a paint pigment production facility, a
chromium dioxide production facility, two industrial landfills, and a baseball diamond. The site
was added to the NPL list in 1990 after elevated concentrations of barium, cadmium, cobalt,
lead, manganese, nickel, zinc, and volatile organic compounds were found in the 1970s and
1980s.
Contaminants: Mn, Ba, Cd, Cu, Ni, Pb, and Zn
Reactive media: Sand, calcium sulfate, ZVI, and magnesium carbonate
Demonstration: Batch scale studies revealed that a mixture of sand, calcium sulfate, ZVI,
and magnesium carbonate would decrease concentrations of Zn, Mn, and Ba. A field
demonstration PRB measuring 2,200 feet (670 m) long, 18 inches (46 cm) wide, and 20 feet
(6 m) deep was installed in 2002. Barium concentrations were reduced from 8,000 to 1,000 µg/L;
Zn concentrations were reduced from 1,000 to <9 µg/L, and Mn concentrations were reduced
from 26,000 to 900 µg/L or less. Manganese concentrations remained elevated due to reducing
conditions created inside the PRB. The lifetime of the PRB has been estimated at 600 years.
Projected savings in comparison to pump -and -treat have been estimated at $13 million.
Universal Forest Products, Inc., Granger, IN
Reference: Ott, 2000
Site background: Spills and leaks from a wood treatment plant contaminated local groundwater
with Cr(VI), Cu, and As. Pump -and -treat was used for off -site contamination. No on -site
remediation was implemented, creating the need for further remedial action.
Contaminants: Cr(VI), Cu, and As
Reactive media: Calcium polysulfide
Demonstration: A combination of pump -and -treat and PRB technologies was implemented
in 1995. Groundwater was pumped from a recovery well and treated with 29 percent calcium
polysulfide in a series of pipes, then discharged to a bag filter. Cr(VI) is reduced to Cr(III) and
B-11

PRB Case Studies
precipitated from solution as Cr(OH),. Treated water was then reinjected by a horizontal
infiltration pipe. The site remediation has been completed after five years and two months
of remediation. No results are reported for Cu or As.
Cotter Corporation Uranium Mill, Canon City, CO
Reference: USDOE, 2005
Site background: Groundwater at this former uranium -ore milling site is locally contaminated
with molybdenum and uranium at concentrations of approximately 4.8 mg/L (Mo) and 1.0 mg/L
(U). Groundwater flows through unconsolidated sand, gravel, and silt. Bedrock consists of
claystone, sandstone, and coal. The groundwater flux is estimated at 1 gallon per minute at the
site.
Contaminants: Mo, U
Reactive media: ZVI
Demonstration: A 30-foot (9.1-m) wide, 7-foot (2.1-m) high funnel -and -gate PRB was
installed in June 2000. The PRB consists of approximately 80 tons of ZVI. Part of the barrier
was excavated in 2004 because of deteriorating performance. The ZVI was clogged with
precipitates, including calcium carbonate, iron oxides, and sulfide minerals. Performance
deterioration was indicated by groundwater mounding at the upgradient side of the barrier and
increased concentrations of Mo within the barrier over time. It was recommended to install a
pretreatment zone composed of gravel and ZVI.
Columbia Nitrogen Site, Charleston, SC
Reference: hitp://www.epi.,(,,ovY.idiLlrescai'cli/wEiste/i-ese�ti-cli 01. )d1; Bronstein, 2005; Puls, 2006
Site background: Oxidation of pyrite cinders has contaminated groundwater with arsenic and
other heavy metals. The spent pyrite resulted from extensive phosphate fertilizer production
between 1905 and 1972.
Contaminants: Pb, Cd, As, acidic pH
Reactive media: Organic carbon (30%), ZVI (20%), limestone (5%), and pea gravel (45%)
Demonstration: A pilot -scale PRB was constructed on site in 2002. The PRB measures 30 feet
(9.1 m) long, 12 feet (3.7 m) deep, and 6 feet (2 m) wide. The study is designed to assess the
performance of organic carbon and ZVI in treating arsenic and other heavy metals, the longevity
of organic carbon -based systems, and the long-term reactivity and hydraulic performance of the
barrier. Two years of sampling revealed effective removal of Pb, Cd, and As, and buffering of
pH from <4 to >6.
B-12

Export Control Restrictions
Access to and use of EPRI Intellectual Property is granted with
the specific understanding and requirement that responsibility
for ensuring full compliance with all applicable U.S. and
foreign export laws and regulations is being undertaken by
you and your company. This includes an obligation to ensure
that any individual receiving access hereunder who is not a
U.S. citizen or permanent U.S. resident is permitted access
under applicable U.S. and foreign export laws and
regulations. In the event you are uncertain whether you or
your company may lawfully obtain access to this EPRI
Intellectual Property, you acknowledge that it is your
obligation to consult with your company's legal counsel to
determine whether this access is lawful. Although EPRI may
make available on a case -by -case basis an informal
assessment of the applicable U.S. export classification for
specific EPRI Intellectual Property, you and your company
acknowledge that this assessment is solely for informational
purposes and not for reliance purposes. You and your
company acknowledge that it is still the obligation of you and
your company to make your own assessment of the applicable
U.S. export classification and ensure compliance accordingly.
You and your company understand and acknowledge your
obligations to make a prompt report to EPRI and the
appropriate authorities regarding any access to or use of EPRI
Intellectual Property hereunder that may be in violation of
applicable U.S. or foreign export laws or regulations.
© 2006 Electric Power Research Institute (EPRI), Inc, All rights reserved.
Electric Power Research Institute and EPRI are registered service marks of
the Electric Power Research Institute, Inc.
®Printed on recycled paper in the United States of America
The Electric Power Research Institute (EPRI)
The Electric Power Research Institute (EPRI), with major locations in
Palo Alto, California, and Charlotte, North Carolina, was established
in 1973 as an independent, nonprofit center for public interest
energy and environmental research. EPRI brings together members,
participants, the Institute's scientists and engineers, and other leading
experts to work collaboratively on solutions to the challenges of electric
power. These solutions span nearly every area of electricity generation,
delivery, and use, including health, safety, and environment. EPRI's
members represent over 90% of the electricity generated in the
United States. International participation represents nearly 15% of
EPRI's total research, development, and demonstration program.
Together -..Shaping the Future of Electricity
Program:
Groundwater Protection and Coal Combustion Products
Management
ELECTRIC POWER RESEARCH INSTITUTE
3420 Hillview Avenue, Palo Alto, California 94304-1338 • PO Box 10412, Palo Alto, California 94304-1338 USA
800,313.3774 • 650.855.2121 • askepri@epri,com • www.epri.com
1012584