oEPA
United states	EPA 600/R-16/383 I December 2016 I www.epa.gov/research
Environmental Protection
Agency
Pilot-Scale Demonstration of In Situ Chemical Oxidation
Involving Chlorinated Volatile Organic Compounds
Design and Deployment Guidelines
Parris Island, SC - Marine Corps Recruit Depot Site 45 Pilot Study
¦66 loon
DESIGN
DEPLOYMENT
M4-N-S
(Total CV0C*|
NaMnO, (970lhs 40%; JtiQC. 2013)
NaMnO., (2451 Jbs40%; S^20J3)
NaMnO, (5070 Itys 40%; Spring 2014)
PERFORMANCE
EVALUATION
PROJECT REPORT
Office of Research and Development
National Risk Management Research Laboratory

-------
This page intentionally left blank.

-------
EPA 600/R-16/383
December 2016
Pilot-Scale Demonstration of In Situ
Chemical Oxidation Involving Chlorinated
Volatile Organic Compounds
Design and Deployment Guidelines
Parris Island, SC, Marine Corps Recruit Depot Site 45 Pilot Study
Scott G. Huling
U.S. Environmental Protection Agency
Office of Research and Development
National Risk Management Research Laboratory
Robert S. Kerr Environmental Research Center
P.O. Box 1198, Ada, OK, 74820
huling.scott@epa.gov 580.436.8610
Bruce E. Pivetz
CSS
452 Mountain View Drive
Lewiston, NY 14092
Ken Jewell
U.S. Environmental Protection Agency
Office of Research and Development
National Risk Management Research Laboratory
Robert S. Kerr Environmental Research Center
P.O. Box 1198, Ada, OK, 74820
Saebom Ko
National Research Council
Robert S. Kerr Environmental Research Center
P.O. Box 1198, Ada, OK, 74820
Office of Research and Development
National Risk Management Research Laboratory I Ground Water and Ecosystems Restoration Division

-------
Disclaimer
This document has been reviewed in accordance with U.S. Environmental Protection Agency policy and
approved for publication. Any mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
Acknowledgements
The Authors wish to acknowledge the collaborative support by the Parris Island Partnering Team (PIPT),
including L. Llamas (EPA, Region 4), M. Singletary, D. Owens (US Navy), T. Harrington, L. Donahoe (US Marines),
C. Wargo, A. Gerry, M. Amick (SC DHEC).
The EPA field and analytical staff contributing to this project include T. Lankford, L. Callaway, K. Hargrove,
R. Neill, K. Jones, and J. Groves (EPA, NRMRL, GWERD, Ada, OK), and P. Clark (EPA, NRMRL, LRPCD, Cincinnati, OH).
Key Words
Site characterization, ground water, CVOCs, permanganate, in situ chemical oxidation, ISCO design,
direct-push injection, rebound
Acronyms
ASTM	American Society of Testing Materials
bgs	below ground surface
CMF	contaminant mass flux
CSIA	compound-specific isotopic analysis
CSM	conceptual site model
CVOCs	chlorinated volatile organic compounds
DoD	Department of Defense
c-DCE	cis-l,2-dichloroethylene
t-DCE	trans-l,2-dichloroethylene
1,1-DCE	1,1-dichloroethylene
DNAPL	dense non-aqueous phase liquid
EPA	Environmental Protection Agency
GC/MS	gas chromatography / mass spectroscopy
GWERD	Ground Water and Ecosystems Restoration
Division
ICP - OES	inductively coupled plasma - optical
emission spectrometry
ID	inside diameter
ISCO	in situ chemical oxidation
MCRD	Marine Corps Recruit Depot
LRPCD	Land Remediation and Pollution Control
Division
MSDS	material safety data sheet
NAPL	non-aqueous phase liquid
NOD	natural oxidant demand
NOM	natural organic matter
NRMRL	National Risk Management Research Laboratory
OD	outside diameter
ORD	Office of Research and Development
ORP	oxidation reduction potential
PCE	perchloroethylene
PI, SC	Parris Island, South Carolina
PIPT	Parris Island Partnering Team
PPE	personal protection equipment
PUPS	Palmetto Utility Protection Service, Inc.
(Columbia, SC)
PV	pore volume
RAP	research action planning
RCRA	Resource Conservation and Recovery Act
RSKERC	Robert S. Kerr Environmental Research Center
SC DHEC	South Carolina Department of Health and
Environmental Control
SHC	Sustainable and Healthy Communities
TCE	trichloroethylene
TEAP	terminal electron accepting process
TOC	total organic carbon
Tt NUS	TetraTech-NUS Corp.
UIC	underground injection control
USGS	United States Geological Survey
UST	Underground Storage Tanks
VC	vinyl chloride

-------
Table of Contents
Disclosure	i
Acknowledgements	i
Abstract	v
1.	Introduction	1
1.1	In Situ Chemical Oxidation (ISCO)	3
1.1.1	ISCO Description	3
1.1.2	In situ chemical oxidation at sites with chlorinated volatile organic compounds (CVOCs) ...4
1.1.3	Advantages and limitations of the technology	4
1.2	Site Selection	5
1.3	Partial plume remediation - source reduction remedy	6
1.4	Binary mixtures of permanganate and chlorinated volatile organic compounds in ground water
samples: sample preservation and analysis																	.....8
1.5	Objectives of the ISCO demonstration at Parris Island Marine Corps Recruit Depot (Pi MCRD) ....8
2.	Detailed Site Description 																												9
2.1	Site location and history...																									9
2.2	Site conceptual model, previous investigations																	 10
2.2.1	USGS site characterization																						 10
2.2.2	TetraTech NUS (TtNUS) (2004; 2012)...																		 13
2.2.3	Aquifer characteristics.....																						 13
2.2.4	Seepage velocity.....																								 14
2.2.5	Conceptual Site Model	14
3.	Methods and Materials	15
3.1 Site characterization of source area targeted for ISCO	15
3.1.1	Aquifer cores	16
3.1.2	Ground water monitoring micro-wells	18
3.1.3	Direct-push injection well					18
3.1.4	Natural oxidant demand (NOD)	20

-------
3.2	ISCO approach, design, and critical analysis	20
3.2.1	Oxidant injection	20
3.2.2	ISCO approach	20
3.2.3	Injection strategy	21
3.2.4	Oxidant delivery methods - direct-push and injection wells	22
3.2.5	Monitoring well as an oxidant injection well	23
3.2.6	Top-down versus bottom-up injection methods	23
3.2.7	Outside-in oxidant injection	24
3.2.8	Injection pressure and overlying hydrostatic pressure	24
3.2.9	Direct-push disruption of aquifer lithology	25
3.2.10	Field-scale ISCO deployment	25
3.2.11	Natural oxidant demand and oxidant loading	27
3.3	Ground water sampling and analysis	28
3.3.1	Groundwater	28
3.3.2	Contaminant mass flux	30
3.3.3	Impact on natural attenuation	30
3.4	Health and safety plan and permits	32
3.4.1	Health and safety plan	32
3.4.2	Injection permit	32
3.4.3	Well permit	32
3.4.4	Dig permit	32
3.5	Oxidant handling	33
3.6	Design and operation of injection system/equipment	33
3.7	Analytical and Quality Assurance/Quality Control	35
3.8	Data analysis	36
4. Results	37
4.1	Soil cores	37
4.1.1	Visual inspection of core material	37
4.1.2	Total organic carbon	38
4.1.3	Metals and sulfur	39
4.1.4	CVOC	40
4.2	Oxidant delivery and impact on total CVOCs in ground water	41
4.2.1	First oxidant injection event (June 23-29, 2013)	41
4.2.2	Post-oxidation 1 CVOC concentrations	43
4.2.3	Second oxidant injection event (September 22-27, 2013)	47
4.2.4	Post-oxidation 2 CVOC concentrations	47
4.2.5	Third oxidant injection event (March 24-29, 2014)	49
4.2.6	Post-oxidation 3 CVOC concentrations	49
4.3	Massflux	51
4.4	Oxidant distribution and persistence	53

-------
4.5	General indicators of oxidation	54
4.5.1	Chloride (CI )	54
4.5.2	Ferrous iron (Fe+2)	59
4.5.3	Oxidation reduction potential (ORP)	64
4.5.4	Dissolved methane	64
4.5.5	pH	67
4.5.6	Metals	67
4.6	Oxidant delivery and impact on total CVOCs in aquifer solids	70
4.7	Molecular biology tools	71
4.8	Compound-specific isotope analysis (CSIA)	73
4.9	Cost analysis	76
4.9.1	Site characterization	76
4.9.2	Remediation	78
5.	Discussion	79
5.1	Injection equipment, design, and impact	79
5.1.1	Injection equipment	79
5.1.2	Incremental benefits of soil core sampling and analysis	80
5.1.3	Oxidant delivery design and methods	81
5.2	Critical analysis of oxidant loading	82
5.2.1	Estimating oxidant volume for injection	82
5.2.2	Contrasting ISCO design at site 45 with other treatment systems	83
5.3	Assessment of achieving objectives	85
5.4	Metals mobilization	85
5.4.1	Chromium (Cr)	85
5.4.2	Arsenic (As)	86
5.5	Natural attenuation	87
5.5.1	Proposed natural attenuation conceptual model	87
5.5.2	ISCO impact on natural attenuation	88
5.6	Contamination rebound	89
5.7	Sustainability	91
5.8	Recommendations	91
5.8.1	Proposed monitoring activities	91
5.8.2	Proposed ISCO activities	91
6.	Conclusions	93
7.	References	97
Appendices
A.	Oxidant injection pressure calculations	102
B.	Detailed description of injection equipment (schematics, manufacturers, part numbers, costs)	104
C.	Photographic compendium of ISCO activities at the site 45 ISCO demonstration project
(Parris Island, MCRD, SC)	109
D.	Pre-oxidation (baseline) soil core analytical results for total CVOCs	131
E.	Huling, S.G., Ross, R.R. and Meeker Prestbo, K. 2017. In situ chemical oxidation: permanganate
oxidant volume design considerations. Ground Water Monit. Remed. (37)1, Spring	153
F.	Recommended Ground Water Sampling Plan for PI MCRD Site 45	163
iv

-------
Abstract
A pilot-scale in situ chemical oxidation (ISCO) demonstration, involving subsurface injections
of sodium permanganate (NaMn04), was performed at the US Marine Corps Recruit Depot
(MCRD), site 45 (Parris Island (PI), SC). The ground water was originally contaminated with
perchloroethylene (PCE) (also known as tetrachloroethylene), a chlorinated solvent used in dry
cleaner operations. High resolution site characterization involved multiple iterations of soil core
sampling and analysis. Nested micro-wells and conventional wells were also used to sample
and analyze ground water for PCE and decomposition products (i.e., trichloroethyelene (TCE),
dichloroethylene (c-DCE, t-DCE), and vinyl chloride (VC)), collectively referred to as chlorinated
volatile organic compounds (CVOC). This characterization methodology was used to develop
and refine the conceptual site model and the ISCO design, not only by identifying CVOC
contamination but also by eliminating uncontaminated portions of the aquifer from further ISCO
consideration. Direct-push injection was selected as the main method of NaMn04 delivery due
to its flexibility and low initial capital cost. Site impediments to ISCO activities in the source area
involved subsurface utilities, including a high pressure water main, a high voltage power line,
a communication line, and sanitary and stormwater sewer lines. Utility markings were used in
conjunction with careful planning and judicious selection of injection locations. A portable, low
cost injection system was designed, constructed, and deployed at the site. The oxidant delivery
design and deployment methods were used to achieve aggressive, effective, and efficient
oxidant delivery and oxidation of CVOCs. Specifically, this included numerous injection locations,
a narrow radius of influence of the injected oxidant, short vertical screen injection intervals, low
injection pressure, outside-in oxidant injection, and a total porosity oxidant volume design.
Three oxidant injection events were carried out, where the oxidant loading was more aggressive
and the areal footprints were progressively larger. In this process, the oxidant was delivered
into the targeted zones, hydraulic control of the injected oxidant was maintained, the oxidant
persisted in zones where heavy oxidant loading was delivered, and significant CVOC destruction
was achieved. Ground water and aquifer material sampling and analysis involved an array of
parameters, including CVOC, iron, chloride, oxidation reduction potential (ORP), methane,
metals, dissolved oxygen, total organic carbon, oxidant demand, molecular biological tools,
and compound-specific isotopic analysis. Monitoring these parameters provided insight into
the impact of ISCO and ISCO treatment performance, but limited insight was provided by some
parameters. Significant reductions in post-oxidation CVOC concentrations in ground water
and soil, and a 92% and 76% reduction in total CVOC mass flux in shallow and deep micro-
wells, respectively, occurred as a result from the three oxidant injections. CVOC rebound was
determined in 3 of the 38 wells. At one depth interval in the source area, elevated post-oxidation
CVOC concentrations were measured in the soil, indicating the presence of PCE dense non-
aqueous phase liquids. This result suggests that rebound will continue and that subsequent ISCO
activities are needed, on a limited scale, in the source area to address CVOC rebound. Specific
details and guidelines are provided regarding continued monitoring and recommended ISCO
activities. The results of this study are intended to provide details and guidelines that can be
used by EPA and Department of Defense remedial project managers regarding ISCO remediation
at other sites.

-------
This research project involved the pilot-scale
deployment of in situ chemical oxidation (ISCO)
at the United States Marine Corps Recruit
Depot (MCRD) (Parris Island, SC), site 45. The
purpose of the project was (1) to advance the
state of the science of the ISCO technology
through improvements in ISCO design, methods,
techniques, and tools, and (2) to collaborate with
the US Environmental Protection Agency (EPA)
Region 4 Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA) program
and verify the effectiveness of the ISCO technology
at pilot-scale.
In a detailed feasibility analysis (TetraTech,
2012), it was determined that an ISCO remedy,
in conjunction with natural attenuation, could be
protective of human health and the environment
by meeting the cleanup goal of reducing the
toxicity, mobility, and volume of groundwater
CVOCs, reducing the high-concentration areas,
significantly reducing the expansion of the
plume, and permanently removing CVOCs in the
ground water. Overall, this would accelerate the
remediation process. Independently, based on
a critical analysis of site-specific conditions and
goals, it was determined that the health and safety
of project and facility personnel could be assured,
that ISCO deployment was appropriate, and stood
a moderately good probability to achieve the
treatment objectives. The ISCO pilot-scale study
was proposed by the U.S. Environmental Protection
Agency, Office of Research and Development,
National Risk Management Research Laboratory,
Ground Water and Ecosystems Restoration Division
(EPA, ORD, NRMRL, GWERD) at the Robert S.
Kerr Environmental Research Center (RSKERC)
(Ada, OK) through an EPA research action plan
(RAP). Subsequently, the proposed research was
approved by the EPA ORD through the research
planning process. The Parris Island Partnering
Team (PIPT) oversees the CERCLA program at the
Parris Island MCRD, and is comprised of the US
EPA Region 4, US Navy, US Marine Corps Recruit
Depot, and the South Carolina Department of
Health and Environmental Control (SC DHEC). The
PIPT determined that ISCO was an appropriate
source reduction remedy for site 45 and also
1

-------
approved the ISCO pilot-scale project. Therefore,
fortunate timing between the US EPA CERCLA
site remediation process and the EPA ORD RAP
project planning allowed collaboration and
synergy between these two programs. Overall,
the successful implementation and completion of
this pilot-scale demonstration was accomplished
through the collaboration of EPA, NRMRL, GWERD
(Ada, OK); Region 4 (Atlanta, GA); EPA, NRMRL,
Land Remediation and Pollution Control Division
(LRPCD), (Cincinnati, OH); the Department of
Defense including both the US Navy and the US
Marine Corps; and the South Carolina Department
of Health and Environmental Control (SC DHEC).
Through the execution of the pilot-scale ISCO
project, valuable data and information was
provided that satisfied the remedial steps in the
CERCLA feasibility study process, has advanced the
state of the science of the ISCO technology, and
will be used by the regulatory decision-makers
associated with the PIPT to determine whether
ISCO should be deployed as the final treatment
remedy for site 45.
This project's feasibility analysis included research
statistics provided by the Superfund Remedy
Report, which includes information and analysis
on remedies selected to address contamination at
Superfund sites (EPA, 2014).
The US EPA prepared a Superfund Remedy Report
that compiled data and information regarding
remedies selected that address contamination
at Superfund sites (EPA, 2014). In 2009-2011, a
critical analysis of the data indicates that decisions
regarding the Superfund remedial program
selected "treatment" for a large number of source
remedies, and in situ treatment continued to make
up an increasing portion of the selected treatment
technologies. On average, half of recent source
treatment decision documents included in situ
treatment. The EPA's analysis of remedy selection
from FY 2009 to 2011, and a comparison to early
data, shows that the Superfund remedial program
continues to select treatment at nearly 75 % of
Superfund sites over the life of the program.
In situ treatment for ground water continues
this overall upward trend, averaging 38 % of
decision documents addressing ground water.
Furthermore, in situ chemical oxidation (ISCO)
treatment continues to be one of the most
frequently selected in situ treatment technologies.
Concerning groundwater remedies (EPA, 2014), the
selection of pump and treat and monitored natural
attenuation (MNA) has decreased slightly, while
the selection of in situ treatment has increased.
The overall selection of in situ chemical treatment
remedies for groundwater remains steady and
more than half of the chemical treatment remedies
involved ISCO. These trends in technology selection
indicate the need for the continued development
of ISCO, a technology that has the ability to
transform contaminants in the subsurface while
minimizing the use of fossil fuel energy, chemicals,
and environmental impact (Siegrist et al., 2001;
2011; Huling eta!., 2006).
Task 4.1 of the sustainable and healthy
communities (SHC) research action plan (RAP)
involved assessment and management of
contaminated ground water to protect human
health and ecological services. Section 3.1.5.1
of the SHC research program involved ISCO, a
technology that meets the goals and objectives
of task 4.1. Contamination of aquifers with
organic chemicals remains a significant concern
even after decades of research into remedial
technologies and field applications. In recent
years, various methods of in situ chemical
oxidation have emerged involving the injection of
different oxidants into the subsurface, including
sodium or potassium permanganate (NaMn04
and KMn04, respectively), hydrogen peroxide
(H202), and sodium persulfate (Na2S208). ISCO
is a developing technology for hazardous waste
treatment in surface and subsurface systems and
is used to transform and destroy a wide array
of environmental and emerging contaminants
of national interest. These contaminants are a
national EPA priority and are regulated under
numerous environmental programs, including
CERCLA, RCRA, and UST; and millions of US
dollars are spent annually to treat ground water,
soil, and aquifer material. Engineering design
criteria and guidelines for ISCO deployment,
monitoring, and performance evaluation methods
still requires further development. For example,
understanding the fate and transport of injected
chemical oxidants, in conjunction with the targeted

-------
contaminants in the subsurface, are critical in the
successful development and deployment of this
treatment technology and is a core competency
at the EPA/NRMRL/Ground Water and Ecosystems
Restoration Division (Ada, OK). The ISCO project
at Parris Island Marine Corps Recruit Depot,
(Parris Island, SC) involved multiple EPA priority
research areas, including research on the rapid
chemical oxidation of chlorinated solvents. The
results of this research can be extended to other
priority contaminants including biofuels, endocrine
disrupting compounds, pharmaceuticals, and other
emerging contaminants, so the knowledge gained
from the Parris Island ISCO project can be applied
to a broader area of contaminant oxidation and
subsurface remediation.
Specific technical objectives of this project were
to establish an accurate conceptual site model
through high resolution aquifer core material
and ground water sampling and analysis, to
focus oxidant delivery to targeted zones through
improved resolution of contaminant distribution
data and information, to design and develop a low
cost, portable injection system, and to improve the
design of oxidant injection dosages (i.e., volume
and concentration) and delivery using spatial
distribution criteria. Satisfying these objectives
would permit an improved critical analysis of ISCO
deployment and treatment performance, avoid
potential treatment inefficiency and ineffectiveness
associated with poorly developed methods, and
ultimately advance the state of the science of the
ISCO technology.
1.1 In Situ Chemical Oxidation (ISCO)
1.1.1 ISCO Description
ISCO involves the introduction of a chemical
oxidant into the subsurface for the purpose of
transforming groundwater or soil contaminants
into less harmful chemical species (Siegrist et
al., 2001; Huling and Pivetz, 2006; Siegrist et al.,
2011). Several forms of oxidants are used for ISCO,
including sodium and potassium permanganate
(NaMn04, KMn04), hydrogen peroxide (H202),
and sodium persulfate (Na2S208). Sodium
permanganate (NaMn04) was selected as the form
of oxidant to use in the pilot-scale deployment
of ISCO at site 45. A detailed description of the
fundamentals of ISCO using permanganate is
described elsewhere (Siegrist etal., 2001; Huling
and Pivetz, 2006; Petri et al., 2011). Sodium
permanganate (40%; 1.36 g/mL) is formulated
and produced as a liquid, is a highly soluble form
of permanganate, and only requires dilution prior
to injection. Potassium permanganate (KMn04(s))
is a solid, and costs less than NaMn04, but must
be dissolved and rigorously mixed to assure
complete dissolution prior to injection. NaMn04
was selected rather than KMn04 to minimize the
use of mixing tanks and equipment, potential
oxidant exposures during mixing, labor associated
with oxidant handling and mixing, logistics in
staging oxidant mixing and injecting, and the
potential for permeability reduction near the
injection points and wells. Specifically, this helped
to assure the injection of more uniform oxidant
concentrations, to limit the on-site footprint of the
remedial activity, and to minimize the interruption
in through traffic and on-site activities at the dry
cleaner.
Successful ISCO is functionally dependent on
three basic requirements. First, there must be a
strong reaction between the oxidant and target
compound, indicating the importance of reaction
kinetics. The faster the reaction the better, as
this limits the contact time required between the
reactants. There must be physical contact between
the target chemical and the oxidant, emphasizing
the importance of good site characterization and
effective oxidant delivery to the specific targeted
zone. Finally, a sufficient quantity of oxidant must
be delivered to the targeted zone and must persist
for a sufficient reaction period to achieve the
treatment objectives. These basic tenets must be
satisfied for significant contaminant reduction to
occur at the site.
3

-------
1.1.2 ISCO at sites with chlorinated volatile organic compounds (CVOCs)
Reaction 1 (Table 1) is the 3-electron half reaction
for permanganate (Mn04") oxidation under most
environmental conditions (pH 3.5 to 12). One of
the reaction byproducts is MnOz, and in the pH
range of 3.5 to 12 it is a solid precipitate. Under
acidic conditions (pH <3.5), Mn in solution or
in colloidal form may be present in different
redox-dependent oxidative states (Mn+2 +4 +7).
Additionally, under strongly alkaline conditions, Mn
may be present as Mn+S. Under conditions where
pH is <3.5 and >12, 5-electron and 1-electron
transfer reactions occur, respectively (Table 1, half
reactions 2 and 3). Reactions 1 to 3 illustrate the
general permanganate reactions in the subsurface.
Overall, permanganate oxidation potentially
involves various electron transfer reactions
(reactions 1 to 3), but is generally considered
independent of pH in the range of 4 to 8.
Reactions 4 to 7 (Table 1) illustrate the oxidation of
perchloroethylene (PCE), trichloroethylene (TCE),
dichloroethylene (DCE), and vinyl chloride (VC),
respectively. These are chlorinated volatile organic
compounds (CVOCs) commonly found at site 45.
Examination of these balanced redox reactions
indicates that the oxidant demand is inversely
correlated with chlorine substitution. For example,
the stoichiometric requirements for PCE, TCE, DCE,
and VC are 1.33, 2.0, 2.67, and 3.33 mol KMn04/
mol contaminant, respectively. The reaction rate
between Mn04" and the CVOCs is fast (Yan and
Schwartz, 1999) (Table 2). Assuming conservatively
that [Mn04"] = 1 mM (120 mg/L), it is projected
that the half-lives of target CVOCs found at the
ISCO site range from less than an hour for TCE,
c-DCE, t-DCE, and 1,1-DCE, to 4.5 hours for PCE.
The average ground water temperature is > 20 °C
during summer months, and faster reaction rates
are expected for these CVOCs.
Table 1. Permanganate oxidation and CVOC mineralization reactions.
pH dependency of Mn04 reactions
Mn04 + 2 H20 + 3 e —> Mn02(s) + 4 OH	(pH 3.5-12)
Mn04 + 8 H+ + 5e—>Mn+2 + 4H20	(pH <3.5)
Mn04 + e —> Mn04 2	(pH >12)
CVOC stoichiometry
4 KMn04 + 3 C2CI4 + 4 H20-
~6 C02 + 4 Mn02 + 4 K+ + 8 H+ + 12 CI
2 KMn04 + C2HCI3—J
8KMn04 + 3C2H2CI2-
10 KMn04 + 3 C2H3CI-
~2 C02 + 2 Mn02 + 2 K+ + H+ + 3 CI
• 6 C02 + 8 Mn02 + 8 K+ + 2 OH + 6 CI + 2 H20 (6)
~6 C02 + 10 Mn02 + 10 K+ + 7 OH + 3 CI + H20 (7)
Table 2.
First order transformation of
chlorinated volatile organic compounds.

Reaction Rate
Half-Life, t1/2(2)
CVOCs
Constant, k(1)
(20 °C) (hours)

(20 °C) (s1)

PCE
4.5xl05
4.3
TCE
6.5x10"
0.3
c-DCE
9.2x10"
0.2
t-DCE
3.0xl02
<0.1
1,1-DCE
2.38xl03
<0.1
1 The estimated pseudo first-order reaction rate
constant involving Mn04 (1 mM, 20 °C)
(Yan and Schwartz, 1999).

2 The half-life is the time required for 50% loss
of CVOCs.

1.1.3 Advantages and limitations of the technology
The oxidation of CVOCs in ground water at the Parris Island (PI), MCRD site 45 involved relatively simple and
fast reaction kinetics between Mn04" and soluble CVOCs. This type of reaction does not require a catalyst.
NaMn04 is highly soluble, and high concentrations of the oxidant can be injected, where appropriate. The
long-term persistence of Mn04" in the subsurface permits both advective and diffusive transport, which is
generally associated with relatively good distribution of the oxidant and good contact between the oxidant
and the target contaminants. High concentrations of NaMn04 can result in a density greater than ground
water, causing density-driven vertical transport of the oxidant into the subsurface. This transport mechanism
can contribute to improved distribution and contact between the oxidant and CVOCs that may be present
in low-permeability materials. Specifically, the oxidant may migrate vertically and accumulate on or near
low permeability materials that are contaminated with CVOCs. This condition is desirable, given that CVOC
diffusion from low permeability materials, and oxidant diffusion into the low permeability materials can
result in a larger and effective reaction zone between the oxidant and the CVOCs.

-------
Permanganate has been successfully delivered
into a wide range of hydrogeologic environments:
sands, clays, sand-clay mixtures, alluvial materials,
fractured shale, and fractured bedrock. A
disadvantage of in situ technologies where fluids
are injected, including ISCO, is the potential for
hydraulic short circuiting of the injected oxidant.
Naturally occurring and man-made preferential
pathways result in non-uniform delivery of the
oxidant, and potentially the transport of the
oxidant into non-targeted zones. Consequently, the
failure to achieve adequate oxidant coverage and
distribution in the target zone results in incomplete
contact and limited reaction between the MnO "
4
and CVOCs. Significant efforts to characterize the
hydrogeology, CVOC distribution, and CVOCfate
and transport at site 45 helped to develop an
accurate conceptual site model that was critical in
the effective and efficient delivery of the oxidant.
MnOz(s) is the main reaction byproduct that may
accumulate near the injection well or at the dense
non-aqueous phase liquid (DNAPL) interface
(i.e., encrustment) resulting in mass transport
(permeability reductions) and mass transfer
limitations, respectively. Permeability reductions
may also result from the (1) injection of potassium
permanganate solids, such as incompletely
dissolved KMn04(s), and (2) entrapment of
C02(g) released from CVOC mineralization. Ion
exchange of Na+ in NaMn04 for divalent cations
in the aquifer matrix may disperse soil colloids
and also contribute to permeability reductions,
however, permeability reductions of this
nature are not often reported. These causes of
permeability reductions can be largely avoided
by adhering to design and operational guidelines.
Subsequent increases in the permeability due
to dissolution of C02(g) and KMn04(s) suggests
that these mechanisms are reversible under
ambient conditions. Permeability reduction can
also be avoided during oxidant recirculation in
ISCO by filtering re-injected fluids, by selection
of permanganate with low silicate content,
such as NaMn04, and by assuring adequate
mixing of the KMn04 solution before injection
(Luhrs etal., 2006). Well-development steps can
also be performed to more rapidly restore the
permeability of affected injection wells.
Other disadvantages potentially include secondary
drinking water standards for manganese and the
oxidant impact on microbial activity. The EPA has
established a secondary maximum contaminant
level (MCL) for drinking water for manganese
(0.05 mg/L), based on color, staining, and taste.
At site 45, there are no downgradient domestic
water wells where the ground water is being used
for domestic purposes. A short-term reduction in
microbial activity may result from the injection
of permanganate; however, this does not appear
to be permanent, and post-oxidation increases
in microbial numbers, activity, and contaminant
attenuation are often reported (Luhrs et al., 2006).
1.2 Site Selection
ISCO has been deployed over a wide range of geologic and hydrogeologic environments. A moderate to
high permeability subsurface is preferred, and sandy material is ideal. High permeability allows rapid
delivery of the oxidant into the subsurface, and therefore does not require high pressure or extended
periods of oxidant injection. A low degree of heterogeneity is preferred since this leads to a more uniform
distribution of the oxidant. Low organic content is also preferred since this limits the natural oxidant
demand and depletion of the oxidant by non-target reactants. At the PI MCRD ISCO site, the shallow
surficial aquifer extends to 18 feet below ground surface (ft bgs) and consists of fine to medium sand.
Localized silty, clayey lenses in the surficial aquifer are limited in areal extent and are not expected to be
functional confining units, typically associated with extensive low permeability materials. A peat and silty
clay layer occurs at 17-27 ft bgs and does effectively function as a local confining unit.
Ideally, the ground water table should be between 10-30 ft bgs, and the contaminated zone should
be no deeper than 65 ft bgs. This limits the vertical depth required for site characterization, remedial
investigations, and ISCO deployment (i.e., limits the depth of drilling and well construction materials/costs).
5

-------
At the PI MCRD ISCO site 45, the ground water
is 3-4 ft bgs and the water table fluctuates with
the tide (0.2-0.6 ft). The contaminated interval is
shallow, resulting in a limited overburden pressure
to off-set the injection pressure.
Fast and well-established reaction rates between
the oxidant and the target contaminant are ideal
and emphasizes the importance of oxidation
chemistry in ISCO. The reaction between
permanganate and the targeted chlorinated
ethenes of interest at the PI MCRD ISCO site
45 has previously been established (Yan and
Schwartz, 1999). Furthermore, the site should not
contain large quantities of dense non-aqueous
phase liquids (DNAPL). In general, the presence
of excessive DNAPL, especially mobile DNAPL,
would make it difficult to accomplish significant
contaminant mass reduction. Prior to ISCO
deployment, DNAPL removal is an important
pretreatment step; however, some DNAPL will
likely exist at most sites containing CVOCs.
Historical information regarding DNAPL handling
and releases, site characterization data, and
ongoing remedial efforts is useful to help assess
the relative quantity of DNAPL that could be in
the subsurface. In preliminary investigations, soil
core samples (n= 31) from site 45 were tested
for DNAPL using a fluorescent light screening
method and no information was found suggesting
direct evidence of DNAPL, neither was it found
in conjunction with other investigations (Tt NUS,
2012). However, it was projected that small
quantities of residual DNAPL were likely at the site,
given the PCE handling and release records and
the extensive nature of the CVOCs' ground water
plume.
Minimal access limitations at ISCO sites are
important regarding site characterization, including
soil coring, installation and sampling of micro-
wells, wells, and remediation, including utilization
of injection equipment, chemical storage vessels,
etc. Further, the site should not contain significant
above- or below-ground utilities that interfere
with the deployment of the technology. At site
45, multiple subsurface, surface, and overhead
impediments to ISCO required special design and
deployment considerations. In the subsurface,
these include a sanitary sewer pipe, storm sewer
line, an 8-inch high pressure water main, and a
high voltage power line. Surface impediments
include overhead steam lines, a drive-through at
the dry cleaner facility, street and parking traffic,
and constant foot traffic. As-built construction
maps were obtained and reviewed to help identify
the presence of subsurface utilities. Facility dig
permits and utility markings were used prior to all
subsurface activities, to confirm these locations.
The CVOC ground water contamination plume
had migrated under the drive-through of the
facility, Kyushu Street, an existing parking lot, and
all the utilities discussed above (Figure 1). Site
activities adhered to appropriate health and safety
requirements, and access to the US Marine Corps
Recruit Depot was through a security gate, where
identification of project personnel and project
purpose was required each day.
1.3 Partial plume remediation - source reduction remedy
Source area contaminant mass reduction, or partial plume remediation, was the goal of this pilot-scale
deployment of ISCO. The PCE source area located near the corner of Bldg. 192, which housed a new dry
cleaner facility, resulted in a CVOC ground water plume that extended to the southeast (Figure 1). As is
typically observed, the downgradient ground water plume was much larger than the source area, but lower
in concentration. It is well known that significant challenges exist in achieving maximum concentration limits
(MCLs) on a site-wide basis for CVOCs at DNAPL sites (McGuire et al., 2006; Krembs et al., 2010). These
challenges are partially attributed to contaminant mass transfer and transport limitations. However, ISCO
deployment in the source area, (1) limits the areal extent in which ISCO is deployed, (2) focuses the oxidant
delivery in the most contaminated zones, where contact between target and contaminant generally results
in fast and efficient reaction rates, (3) and reduces CVOC concentrations and mass flux of contaminants from
the source area. A reduction in mass flux leads to a reduction in the transport load of CVOC contaminants in
the downgradient direction that must ultimately undergo natural attenuation processes.

-------
10,000
"ai0 ¦>'
„s;A
^ Monitorwg well samp fed durog June 200S
¦ USGS Temporary wel sampled during March 2008
~ USGS Temporary well sampled during August 2007
A USGS Temporary well sampied durng June 2007
• Tetra Tech NUS. Inc., temporary well, 2005 (Mark Siadic,
Tetra Tech NUS, Inc., written cornmun . 2007)
«ns Concentration of tetrachloroethefse, in micrograms
per liter. [«, less than: J. estimated: H. analyzed out of
nolo ng lime: ND. not detected at unstated detection limit]
	10	 line of equ3l tetrachloroethene
concentration, in micrograms per liter.
Interval variable as shown. Dashed where
uncertain.
	 Storm sewer
	 Sanrtary »wer
Figure 1. Generalized distribution of tetrachloroethylene in ground water at Site 45,
Marine Corps Recruit Depot, South Carolina (2006-2008) (Vroblesky et at., 2009).

-------
In summary, the treatment objective was not to achieve site-wide MCLs for CVOCs, but rather to investigate
whether a cost-effective remedy could be achieved through ISCO deployment over a limited area. Ideally,
this could lead to a reduction in mass flux, and in conjunction with natural attenuation, achieve acceptable
CVOC concentrations at a downgradient compliance plane, eliminate exposure pathways, and reduce
risks beyond the compliance plane. Under this scenario, the total mass of contamination would not be
completely eliminated, but reduced in a cost-effective manner that limits or minimizes risk. In general,
natural attenuation is expected to be an integral component of ISCO remedies at DNAPL sites because
of the difficulty and technical challenges to achieve site-wide MCLs. Partial plume remediation does not
require 100% removal of contaminant mass, rather this approach recognizes existing technical and economic
remediation constraints, and limits exposure pathways and risks at critical site locations.
1.4 Binary mixtures of permanganate and chlorinated volatile organic
compounds in ground water samples: sample preservation and analysis
Often, ground water samples collected specifically to be analyzed for organic contaminants at ISCO sites may
contain the oxidant and the organic contaminants in a "binary mixture" (Huling et al., 2011; Johnson et al.,
2012; Ko et al.,2012). When organic contaminants and oxidants are commingled in a binary mixture, there
is significant potential for oxidative transformation of contaminants to occur after the sample is collected.
Consequently, the results of the sample analysis may be non-representative of in situ conditions at the time
of sampling, the quality of the ground water sample may be compromised, and false negative results may
occur.
An integral component of ISCO is the collection and analysis of representative ground water samples to
assess ISCO treatment performance (Ko et al., 2012). High quality analytical results are critical to support
site assessment and remedial performance monitoring, where ISCO is deployed. Therefore, the guidelines,
methods, and procedures presented in Ko etal. (2012) were used in this study to preserve binary mixture
ground water samples. The visual confirmation of Mn04" in ground water samples is possible due to the
characteristic pink/purple color of the oxidant. Under this condition, ascorbic acid was amended to the
sample to eliminate Mn04". Excess ascorbic acid amended to aqueous samples did not negatively impact the
quality of the aqueous samples, nor the analytical instruments used in the analyses (Johnson et al., 2012;
Ko etal., 2012).
1.5 Objectives of the ISCO demonstration at PI MCRD
Bench-, pilot-, and full-scale ISCO applications have been conducted using sodium permanganate and much
is already known about process fundamentals and field-scale deployment of this technology (Huling and
Pivetz, 2006; Petri et al., 2011). Through the development of new methods, equipment, and techniques, and
through utilization and adaptation of existing methods, the overall objectives were to make improvements in
both process fundamentals and technology deployment that would lead to improved ISCO effectiveness and
efficiency. The specific objectives of the proposed demonstration were (1) to significantly reduce the mass
of CVOCs in the source area, (2) to limit the areal extent of the CVOC concentration plume downgradient
from the source area and significantly reduce the mass flux of CVOCs from the source area, (3) to quantify
pre- and post-oxidation CVOC natural attenuation to assess the short- and long-term impacts of ISCO on
the natural attenuation process, and (4) to assess contaminant rebound and focus subsequent oxidant
delivery into specific target zones, if needed. The results of this study will provide a conceptual approach
and guidelines that can be used by the EPA and DoD remedial project managers when considering ISCO for
remediation at other sites.

-------
PI MCRD

rrj. an-
SOUTH
CAROLINA
CHARLESTON
2.1 Site location and history
The MCRD at Parris Island (PI), SC is located along
the southern coast of South Carolina, 1 mile south
of the city of Port Royal, It covers approximately
8,047 acres of dry land, salt marshes, saltwater
creeks, and ponds. The PI MCRD facility is the
reception and recruit training facility for the US
Marine Corps for enlisted men from states east
of the Mississippi River and for enlisted women
nationwide. Prior to 2001, the old dry cleaning
facility at site 45 was in a building located in the
Main Post area of PI MCRD, between Panama
Street to the north, Kyushu Street to the south, and
Samoa Street to the east (Figure 1). Four above-
ground storage tanks, put into place in 1988, were
situated along the northern side of the building.
Reportedly, one of the tanks was overfilled on
March 11, 1994, with PCE and an unknown
amount of the PCE solvent flowed into a concrete
catch basin (TtNUS, 2004). The PCE overflow
was not collected at that time, and heavy rainfall
subsequently washed the contaminant onto the
surrounding soil and into subsurface storm drains
(TtNUS, 2004). Additionally, miscellaneous solvent
spills resulted in releases to the sanitary sewer line
that were located in and near the old dry cleaner
facility. Subsequently, multiple lines of evidence
indicate that the PCE leaked from the broken
sewer pipe at a location approximately 50-75 yards
southwest of the dry cleaner facility (Vroblesky
et at., 2009). A new dry cleaner facility was
constructed to the southwest of the demolished
old dry cleaner building. Prior to construction,
a sanitary sewer manhole and a portion of the
sanitary sewer line was removed and replaced with
a diagonal section to bypass the newiy constructed
Building 192. The southeast corner of the new
dry cleaner facility, Building 192, was constructed
over the former manhole and sanitary sewer line.
In this report, "the plume" refers to the southern
CVOC plume emanating from the southeast corner
of Building 192, and the northern CVOC plume
emanates from the north side of the former old
dry cleaner facility (Figure 1).
Leakage of PCE was from the sanitary sewer
and served as the source of contamination for
the southern plume, and not from the new dry
cleaner facility as was once surmised (Vroblesky
et a!., 2009). Contaminated soil at the original spill
location was excavated, and an interim remedial
action was initiated. In early 2001, the main dry
cleaning building, solvent tanks, and other related
structures were demolished and removed from
the site. Currently, the site is mostly a vacant lot
covered with mowed grass and isolated shrubs
and trees. Physical features remaining at the site
consist of three above-ground extraction weil
housing units and one ground water treatment
system shed.
9

-------
2.2 Site conceptual model, previous investigations
Data and information from previous investigations were available for review, and included contaminant
distribution, hydrogeology, and other chemical and physical parameters (TtNUS, 2004; 2012; Vroblesky,
2007; 2008; 2009). This data was important to develop an accurate site conceptual model, to understand
site-specific contaminant fate and transport, and to guide additional site characterization activities.
2.2.1 USGS site characterization
The United States Geological Survey (USGS) performed a subsurface investigation of
ground water quality and levels which involved the monitoring of existing permanent
wells and installation and monitoring of additional permanent and temporary wells
(Vroblesky etal., 2007; 2008; 2009). Temporary borings were installed using direct-push
technology and a membrane interface probe (MIP) was used in temporary borings, which
provided information on the depth of the contamination. Collectively, the data and
information were used in the development of plan-view isocontour maps illustrating the
CVOCs, including PCE, TCE, c-DCE, and VC isopleths (Vroblesky et al., 2009). Examination
of the CVOC ground water contamination isopleths indicated that the source of the
southern plume originated near the corner of the new dry cleaner facility (Bldg. 192)
(Figure 1) (TCE, c-DCE, VC isopleth maps not included). However, several lines of evidence
indicate that the contaminant source in the southern plume was due to the release of
PCE into the sanitary sewer from the old dry cleaner facility, and discharge of the PCE
from an incompetent section of the sanitary sewer line, was constructed of vitrified clay,
in the vicinity of the new dry cleaning facility (Vroblesky et al., 2009). USGS (2009; pg.
28) states "A sewer-inspection camera was used to examine the integrity of the sewer
line in 2007. The camera revealed that although the existing pipe between the former
and new dry-cleaning facilities contained no collapsed sections, it contained many cracks,
and grass roots extended into the pipe. Thus, it is highly probable that the abandoned
sanitary sewer is leaky". Several lines of evidence suggest the source of the plume is in
the immediate vicinity of the southeast corner of Building 192: the CVOC plume emanates
from the southeast corner of Building 192 (Figure 1); the highest CVOC concentrations
measured in ground water and soil samples collected during baseline characterization
were from the area immediately adjacent to MW-25, approximately 10 ft from the corner
of Bldg. 192; and the occurrence of parent and decomposition chlorinated products
downgradient from a source at the southeast corner of Bldg. 192, are consistent with the
natural attenuation of CVOCs. The USGS investigation also identified ground water table
elevations in the lower surficial aquifer, and the presence of subsurface sanitary and storm
sewer lines and the roles they play in the fate and transport of CVOCs in the southern
plume (Figures 2-3) (Vroblesky etal., 2009).
The USGS conducted a detailed assessment of natural attenuation processes in the
southern plume and revealed that the surficial aquifer in the southern plume is anaerobic
at most locations, with the predominant terminal electron accepting process (TEAP) being
iron reduction in the shallowest sediment (Vroblesky et al., 2009). In deeper sediment
containing the main body of contamination, the predominant TEAP appears to be sulfate
or iron reduction; however, the presence of methane, the high degree of contaminant
dechlorination, and a H2 value (31 nM) in the range of methanogenesis at one well
indicates that methanogenic zones probably also are present. In the deepest part of the
surficial aquifer, near the peat layer, the predominant TEAP probably is methanogenesis,
by virtue of the abundance of available natural organic matter.

-------
PAMS MW01-SL
3.46
STS12
.3^1)
mW5MW21-SL
x/ 3 -93 j

PAMS-MWD7 SL
M, 2B2 .|
PA14JMW14-SJ
271
STS14
3E3?
PHM&-MW23 a
267,
WU-4S-MW03-SI
270-,-"
FA^SMWC6-SL
270	.
MMS MW1G-SL
- 229
RAH5-MW3f-5L		—
£¦ 2 74			.| 3W I	STSli
/	I	OJ)
/ / .--^PAMS-MWIML	Vi
< ] '& 3 £2	I
NW4S-MW13-5T
/WlM5-MW19 5L--2-lTOJ5f V
^ j i	asp /L •) 4'—* /
^ MWMSl
MMS-MW2M1
I	£TiC=
iosei
MHI5P*W2&-5L
2«
HAM5-MW1B SL
		 220
M.MS-MW17S1.
1.97
0 5 10 15 20 Meters
-2.3— - Water-level contour, June 27.2008. Interval is 0.1 foot. Datum is North American Vertical Datum of 1B3S
(NAVD 83). Dashed where inferred. Arrows show approximate directon of groundwater movement
PAMS-WW2C-SL
211
Monitonng well identifier; ana water level n feet abowe NAVD 8S, June 27. 20QS, low fcde
Storm sorter, manbote or drain, arc location identifier. Une cashsc where hstoncai or uncertain location.
Storm sewer is high! gmed in yellow where it is known or suspecteo to be below the water table. Nimber in parentheses
is est/nated aftitude of invert in feet atove NAVD 83.
wu-.is-mw::u
3.17 ¦£
Ne« Dry-aeanng
Fadity
W)-«-MW2S-a
"¦—	32? /
¦Nifc am.
t RM-45-V,WlS-Sl
ISC

MWOS-Sl
5T3C2
c.ei
= 333
1 -OSt
RW-4S-MW27-5L
. 110
1UL -ee:
EXPLANATION
11
Figure 2. Ground water levels in the lower surficial aquifer at Site 45, Marine Corps Recruit
Depot, South Carolina (Vroblesky et al., 2009).

-------
Mew Ory-Gearlng
Facitfty
SAS01
'STSB1
La* offices
{Btitdng 233)
EXPLANATION
ST9DE
— Storm sewer, rr\anhole or drain, and location identrfier.
Sewer dashed where hisMrcal or uncertain location.
,	Arrows show direction of" drainage at low tide
¦" — — San tary sewer mafvhcie. and location identfier.
Sewer dashed where historical Arrows show direction
of drainage
Section of storm sewer that is known or suspected to be
lower than the water table.
12
Figure 3. Storm and sanitary sewers at Site 45, Marine Corps Recruit Depot, South Carolina
(Vroblesky et a!., 2009).

-------
2.2.2 TtNUS (2004; 2012)
The surficial aquifer underlying Site 45 consists of sandy Pliocene to Holocene
sediments at an average depth of approximately 18 ft (TtNUS, 2004). In general,
the water table encountered within these heterogeneous sediments is shallow
and is typically encountered at a depth of 3 to 4 ft bgs, at the site. Ground water
is expected to preferentially migrate through the higher-permeability sandy
sediments within the surficial aquifer. Because of their limited areal extents, the
localized silty/clayey lenses found within the surficial aquifer are not expected to
function as significant confining units. However, localized hydraulic effects have
been observed because of silt and clay (TtNUS, 2004). Recharge to the surficial
aquifer is likely to occur, primarily through infiltration of precipitation.
2.2.3 Aquifer characteristics
Slug tests and aquifer tests were conducted in the area associated with the
former dry cleaner building (TtNUS, 2004). Some of these tests were conducted
within approximately 120 ft from the proposed ISCO test areas and were useful
in conducting a preliminary assessment of the oxidant and reagent fate and
transport in the study area. A summary of the slug tests performed in wells in
the upper surficial aquifer (3-7' bgs) and lower surficial aquifer (9-14 ft bgs) are
summarized (TtNUS, 2004; Table 3-6). The hydraulic conductivity arithmetic and
geometric means in the upper surficial aquifer were 12 and 8 ft/day (4.2xl03
cm/s and 2.8xl0"3 cm/s) (n=8), and in the lower surficial aquifer they were 2.7
and 2.1 ft/day (l.OxlO3 cm/s and 7.4xl0"4 cm/s) (n=6), respectively. Based on
these values, this aquifer material would be classified as clean sand or silty sand.
Aquifer test results involving the pumping of well RW-3 were reported; 8
observation wells screened in the upper and lower surficial aquifer were used
during the test (TtNUS, 2004; Table 3-7). Although the methods used are
generally for confined aquifers, the aquifer core data (i.e., boring log) indicate
that the surficial aquifer is unconfined, but the drawdown patterns more closely
represent a confined or leaky-confined aquifer. This may be the result of the
presence of the relatively finer-grained sediments (silty-sand) within the upper
portion of the shallow aquifer, in comparison to the deeper sediments (fine
sand). Evidence supporting the occurrence of this clay layer was observed in
aquifer cores collected during subsequent site characterization activities, as
discussed in section 5.5.1. (Proposed natural attenuation conceptual model),
below. The average transmissivity was 230 ft2/day and, assuming an average
thickness of 15 ft, the overall hydraulic conductivity of the shallow aquifer
sediments is 15.3 ft/day (5xl03cm/s). The well (RW-3) that was pumped was
screened in the upper and lower surficial aquifer (4-16 ft bgs). Given that the
water table is approximately 3.5 ft bgs, and the bottom of the surficial aquifer is
16 ft, a more accurate estimate of the aquifer thickness and average hydraulic
conductivity is 12.5 ft and 18.4 ft/day (6.6xl03 cm/s), respectively. The results of
the slug tests indicate that the majority of water captured by RW-3 was from the
upper surficial aquifer.
The surficial aquifer extends to a depth of approximately 17 ft bgs (TtNUS, 2004).
The peat and clayey materials found underlying the surficial aquifer sediments
throughout the site, at depths ranging from 17 to 27 feet (bgs), are expected
13

-------
to function locally as a confining unit to ground water flow (Tt NUS,
2004). Based on the results from previous laboratory hydraulic
conductivity testing, of six samples from this unit, the geometric mean
vertical hydraulic conductivity for this confining unit is 0.00166 ft/day
(5.8 xlO"7 cm/sec) (TtNUS, 2004). This, in combination with an average
thickness of 5 to 6 ft, indicates that the unit significantly restricts
vertical ground water flow.
The low-permeability materials in the 17-27 ft bgs interval are deeper
than what was targeted in the ISCO demonstration, and serves as a
vertical impediment to downward transport of contaminated water
or oxidant solutions. The aquifer cores collected during the site
characterization investigation associated with the ISCO study extended
to 16 ft bgs, and did not penetrate the confining unit that separates the
surficial and deep aquifers. Overall, site characterization efforts and
other subsurface investigations associated with the ISCO study at site
45 did not extend beyond 16 ft bgs.
2.2.4	Seepage velocity
Given the following parameter values: porosity 0.32, K = 8 ft/d
(geometric mean), dh/dl = 0.005, the seepage velocity in the upper
surficial aquifer is 46 ft/yr (clean to silty sand). The lower surficial
aquifer is comprised of darker material. Given the following parameter
values: porosity 0.32, K = 2 ft/d, dh/dl = 0.006, the seepage velocity is
14 ft/yr (silty sand). Wells in the deep aquifer (17-27 ft bgs) penetrate
farther into the dark, silty clay and organic-rich aquifer materials and
given the estimated porosity of 0.32, K (1 ft/d), and dh/dl (0.004), the
seepage velocity is estimated to be 4.6 ft/yr.
2.2.5	Conceptual site model (CSM)
The data and information provided by the USGS and TtNUS
investigations (1) identified the areal and vertical extent of the
southern ground water plume, (2) provided important hydrogeologic
parameters, and (3) development of probable fate and transport
processes. Collectively, this information served as the basis for a CSM
upon which to base additional detailed site characterization activities
used in the ISCO design.
14

-------
3. Methods and Materials
3.1 Site characterization of source area targeted for ISCO
Additional site characterization was needed
to augment previous work, to improve the
resolution of areal and vertical extent of the CVOC
contamination. The driver for additional data was
(1) to establish a high resolution baseline CVOC
concentrations to compare with post-oxidation
data, which is required to assess treatment
performance, and (2) to identify high priority
specific contamination zones {i.e., locations and
vertical intervals) for oxidant delivery. It is equally
important to identify zones of contamination,
where oxidant delivery is critical, as it is to identify
non-targeted zones, where oxidant delivery is not
needed. Through this approach, the oxidant can be
delivered to the high priority, highly contaminated
zones of the aquifer.
CVOC analyses of ground water samples represent
an integrated measure of contamination mass in
the subsurface, including aqueous, sorbed, and
NAPL phases. Development of generalized ground
water quality isocontour maps is useful in the
delineation of source area(s), assessment of CVOC
fate and transport, and in evaluating treatment
performance. Additionally, analysis of CVOCs
in aquifer core samples can be used to identify
discrete zones of vertical and areal contamination.
Through a critical analysis of ground water and
aquifer core CVOC analytical results, an accurate,
high resolution CSM can be developed. The CSM
can be used to develop an oxidant injection
strategy and design for the targeted delivery of the
oxidant into high priority contamination intervals.
15

-------
3.1.1 Aquifer cores
Multiple rounds of aquifer core collection, sampling, and analysis took
place over the course of this study. As CVOC concentration data sets
were assimilated, data gaps were identified and frozen cores (-15 °C)
were sampled (i.e., subcores collected, extracted, and analyzed) at
different locations along the core. Also, additional cores were collected
at different locations and intervals. Throughout this process, the EPA-
RSKERC continued to address data gaps and to refine the CSM.
Initially, aquifer cores were collected along six transects (T1-T6), spaced
approximately 33 ft apart along the longitudinal axis of the projected
ground water plume (Figure 4). Cores were collected at three locations
(A, B, and C) along each transect. Core location "B" was collected along
the longitudinal axis of the ground water plume. Core locations "A"
and "C" were located 20 ft laterally from core location "B". Two cores
were collected, 4 ft in length, at each location which extended from
approximately 8-16 ft below ground surface (bgs), (i.e., 8-12 ft bgs, 12-
16 ft bgs). Aquifer cores were collected in transparent acetate sleeves,
allowing a visual inspection of the core. The cores were frozen on-site,
by placing the cores on dry ice, and transported to the EPA-RSKERC
where they were stored (-15 °C) until analyzed. A cement/bentonite
mixture (4% bentonite, by weight) was pressure injected, bottom-
up, into each abandoned core location to seal the exploratory boring
preventing it from becoming a potential preferential pathway.
The frozen aquifer cores were partially thawed and subsamples were
collected and analyzed for CVOCs, metals, and total organic carbon
(TOC). Sub-cores analyzed for CVOCs were collected at specific
intervals, extracted with methanol (MeOH), and the MeOH extracts
were analyzed via gas chromatography/mass spectroscopy (GC/MS).
The soil cores were visually examined for general lithology (sand, silt,
clay), stratigraphy (layering, lenses, geochemistry), and permeability.
The hydraulic conductivity of subsamples collected from the cores was
measured in a lab permeameter (ASTM method D2434).
An additional transect (TO) (Figure 4) of aquifer cores, and several
additional source area core locations were collected near the southeast
corner of Bldg. 192, sampled, and analyzed. Through this effort,
numerous sub-core samples provided valuable CVOC concentration
data and information, leading to the development of an accurate CSM
regarding the CVOC distribution in the source area. This information
was used to design oxidant injection locations and intervals.
16

-------
New
Dry Cleaning
Facility
Former
Dry
Cleaner
MW25-SL
MW23-SL
5 STS14
MW16-SL
MW05-SU
0viWl9-SL
MW05-SL,
MW20-SI
MW26-SL"
Panama Street
MW27-
0
'STS22
Former
Temporary
Lodging
200 Feet
50 Meters
C Total CVOC's ground water plume
°	Location of background cores
j^/iW04-SL permanent well sampled in FY2007,
and abbreviated identifier,
o	Permanent well not sampled in FY2007.
q Location of aquifer cores on transects T0-T6
EXPLANATION
_ — . Soil core transects T0-T6
—	 Storm sewer. Dashed where uncertain. Green triangle
indicates drain or manhole.
D STS22 Manhole in storm sewer sampled in FY2007, and
abbreviated identifier.
Figure 4, Conceptual model of the CVOCs ground water plume and aquifer core transect locations. Transects
TO - T6, centered on the approximate longitudinal axis of the CVOCs ground water plume, involve 3 aquifer
locations (A, B, C) extending from left to right (see transect TO). The ordinate of the longitudinal transect is
MW 25-SL located at the southeast corner of the new dry cleaning facility. Transects are located approximately
16.5 ft (TO), 33 ft (Tl), 66 ft (T2), 99 ft (T3), 132 ft (T4), 165 ft (T5), and 198 ft (T6) downgradient from
MW 25-SL.
17

-------
3.1.2 Ground water monitoring micro-wells
Four transects of ground water monitoring micro-well clusters were installed to
assess fate and transport of the contaminants and injected oxidant (Figure 5). The
micro-wells were installed at two depths, shallow and deep. Transect Ml includes
5 monitoring clusters, and transects M2-M4 includes 3 monitoring clusters.
This results in 14 monitoring locations along four transects, and 28 monitoring
points total. Four sets of sentry micro-wells were also installed to assess the fate
and transport of Mn04" and CVOCs on the periphery of the study area. Micro-
wells were constructed with PTFE tubing, and 21 inch stainless steel GeoProbe
screens; they were clustered at depths approximately 7-10 ft and 10.25-12.25 ft
bgs. A 1 / inch GeoProbe rod was driven to depth, and an assembly consisting of
Z* inch stainless steel screen, 21 inch in length, was attached to 5/16 inch (I.D.)
PTFE tubing and inserted into the rod to the bottom. The rod was then removed
and the surficial aquifer material collapsed around the screen and PTFE tubing.
The upper 3-4 ft of the hole was sealed with a cement/bentonite mixture to the
ground surface. The micro-wells were covered by a traffic-rated vault box set in
concrete, and set for 4 months before they were sampled. Well development
procedures involved the removal of 10-20 pore volumes of the micro-well
tubing and screen using a peristaltic pump. The turbid purge water produced
during development, and the purge water produced during each ground water
sampling event, was collected and stored in the investigation derived waste (IDW)
storage drum at the site 45, and transferred to the Horse Island Hazardous Waste
Processing facility.
3.1.3 Direct-push Injection well
A temporary direct-push injection well (Inj.-l, Deep) was installed in the source
area, approximately 7 ft to the northwest of MW-25. The objective of the well
was to more aggressively deliver oxidant into the source area underlying the
suspected leaking sanitary sewer line, now decommissioned. The hole was pre-
punched (= 3 in) over the length of the solid pipe and the well was installed using
direct-push. Inj.-l, Deep was constructed vertically with a screened interval over
11.5-15 ft bgs (2.5 inch outside diameter (OD), 1.99 inch ID, stainless steel, 0.010
inch slot size, wire wrapped). Natural collapse of the aquifer material above the
screen (= 1 ft) was followed by a cement-bentonite grout to within 2 ft from the
surface. A traffic-rated, flush-mounted vault cover was installed. Prior to pre-
punching, aquifer cores were collected from 5-15 ft bgs. These cores were frozen
on-site, transported back to the GWERD facility, and stored at -15 °C. Partially
frozen sub-cores were collected, extracted with MeOH, and analyzed for CVOCs,
as described in section 3.1.1. Upon completion of the well, and prior to oxidant
injection, ground water samples were collected and analyzed for CVOCs.
Originally, it was proposed to construct the injection well at an angle to direct
the oxidant under the corner of the building. However, it was discovered that a
thick concrete layer extended 5 ft from the building in the general vicinity of the
SE corner of Bldg. 192. The exact thickness could not be determined but "mat-
slab" foundations are common at the MCRD and involve a wide-based footing-
foundation. Specifically, the foundation extended approximately 5 ft from the
exterior wall of Bldg. 192. Consequently, the injection well could not
be constructed at an angle and was constructed vertically.

-------
19
M3-N-S
M3-N-D
M3-5-S
M3-S-D
M3-N-S
M3-N-D
M3-5-S
M3-S-D
New Dry
Cleaner Facility
] Nested micro-well cluster
~ Sentry micro-well cluster
O 2" Well
T2-S-S
T2-S-D
T2-N-5
T2-N-D
M4-S-S
M4-S-D
O MW-31, SL
Tl-5-5
Tl-S-D
OMl-Mid-S
Ml-Mid-D
MW-25 O
fc*1. CA C
Ml-SB-S
Ml-SB-D
New Dry
Cleaner Facility
Key
~	Nested micro-well cluster
~	Sentry micro-well cluster
- Monitoringtransects(Ml-M4)
Figure 5. U.S. MCRD (Parris Island, South Carolina) Site 45 (A) well and micro-well numbers
and locations; (B) plan-view schematic.

-------
3.1.4 Natural oxidant demand (NOD)
The permanganate injected into the subsurface will react with a range of species in the
aquifer material, including natural organic matter (NOM), reduced inorganic species, and
the organic chemicals that are the intended target of the oxidant (Mumford et al., 2005;
Huling and Pivetz, 2006; Urynowicz, 2008; Urynowicz etal., 2008). In some cases, excessive
NOD may preclude the feasibility of ISCO due to cost and to projections involving limited
reaction with the target compounds. Therefore, the NOD is often measured as a preliminary
screening tool to assess the general feasibility of ISCO. A standardized 48-hour test method
was used (ASTM, 2007) to measure the NOD. In general, the ASTM test method involves
the use of a complete-mix test reactor allowing full contact between oxidant and reactants.
Due to these physical differences between lab and field conditions, NOD results generally
represent higher NOD values than might be encountered in the field.
The NOD was measured using aquifer core material in the central portion of the plume,
at location T2-B. Approximately 35 g of wet soil was collected at depths of 5, 7, 9, 11, and
13 ft bgs and was amended with a NaMn04 solution (5 g/L NaMn04). The test reactors
were shaken periodically throughout the day and the NaMn04 was replenished in the
test reactors when the solution was pink or clear over a 48 hour period. The NaMn04 was
replenished one time in test reactors containing soil from 5, 7, 9, and 11 ft bgs, and four
times in the reactor containing soil from 13 ft bgs. At the end of the testing period, the NOD
was estimated to be < 2.7 g/kg in the 5-11 ft bgs interval, and approximately 5.0 g/kg in the
13 ft bgs interval. Given that TOC concentrations are generally proportional to NOD, these
results are consistent with the TOC data which exhibited increasing TOC concentrations
with depth (refer to Section 4.1.2 Total Organic Carbon).
3.2 ISCO approach, design, and critical analysis
3.2.1	Oxidant selection
Sodium permanganate was selected as the oxidant to use in this ISCO project for several
important reasons. As described in section 1.1.2 above, fast reaction rates occur between
the permanganate anion (Mn04") and the chlorinated ethenes found at site 45. Half-lives
have been estimated to be less than an hour for TCE, c-DCE, t-DCE, 1,1-DCE, and 4.5
hours for PCE (20 °C). Since the average ground water temperature is > 20 °C at site 45
during the summer months, faster reaction rates and shorter half-lives are expected for
these CVOCs. Furthermore, no catalyst is required, NaMn04 is highly soluble, long-term
persistence of NaMn04 will result in good contact with the target contaminants near
NAPL interfaces and in low permeability conditions, preliminary results indicated that the
natural oxidant demand was low, and oxidant delivery was not projected to be limited
due to reported values of hydraulic conductivity.
3.2.2	ISCO Approach
There were three objectives of the first injection event: (1) assure that the newly
designed injection equipment could be operated in a manner that protected the health
and safety of the field crew and the general public, (2) assess the fate and transport of
the oxidant, and (3) assess the impact of ISCO activities on the CVOC concentrations and
mass flux.

-------
Previously, an unknown number of subsurface exploratory borings had been performed
at site 45 by several organizations and it was unclear to what extent the abandoned
borehole locations had been properly grouted and sealed. Potentially, old unsealed
exploratory borings could significantly contribute to "daylighting", where the oxidant
migrates upward along the former borehole location and seeps out at the ground surface.
Numerous subsurface utilities are known to occur in the source area, including an 8-inch
high pressure water main, high voltage power line, communication lines, sanitary sewer,
and a stormwater sewer. Although the proposed oxidant injection intervals were below
these buried utilities, it was unclear whether the injected oxidant could migrate though
preferential pathways, enter into utility line trenches, and short-circuit the intended
targeted zones. The overhead high pressure steam lines also represented a hazard when
raising and lowering the mast height of the direct-push rig (GeoProbe 6610DT). Subsurface
structures from former buildings were probable, as per discussion with long-term MCRD
employees, and as observed via GeoProbe rejection in areas known to be utility-free.
Given these details and complexities in site conditions, the primary objective was to
ensure the health and safety of the field crew and the general public. Regarding oxidant
fate and transport, the objective was to confirm that the oxidant could be successfully
delivered to the targeted zone as designed, to assess oxidant persistence, and determine
whether hydraulic control could be achieved. Essentially, it was critical to ensure that the
oxidant did not seep into the sanitary or storm sewers, or seep into non-targeted areas.
Consequently, the first oxidant injection event involved a small-scale, low volume oxidant
loading approach designed to limit potential risk, to gather oxidant fate and transport
information, and to assess the impact on the CVOCs. Assuming it could be confirmed as a
result of the first oxidation event that the oxidant was injected safely, and was successfully
delivered to the targeted zone(s), subsequent aggressive remediation involving greater
oxidant loading was planned.
This approach integrated well with the overall treatment approach involving multiple
oxidant injections. Consequently, in order to assess the impact of ISCO activities on the
CVOC concentrations and mass flux, multiple monitoring events were used to quantify
treatment performance, to understand the fate and transport of CVOCs, and to guide
subsequent injections. Therefore, ground water sampling of all monitoring wells was
performed prior to and following each oxidant injection event.
3.2.3 Injection strategy
The soil core analytical results, presented in section 4.1 below, indicated that the main
source of CVOC contamination was near the sanitary sewer line southeast of Building
192, at a depth of 8-14 ft bgs, and 8-12 ft bgs for the downgradient portion of the plume.
Therefore, the targeted oxidant injection intervals occurred over these zones. The CVOC
data from the discrete, non-composited soil cores was used to identify specific CVOC
contamination intervals. This soil core data and information was critical to the continued
development of a conceptual site model, documenting and refining the CVOC source
areas, and to guide the design of the first oxidant injection event. Post-oxidation ground
water monitoring of CVOC concentrations was also used to assess treatment performance
and to guide subsequent injections. Persistent CVOC concentrations, measured as a
result of post-oxidation ground water sampling, are indicative of high contaminant mass,
insufficient delivery and persistence of oxidant, and/or slow oxidant or CVOC mass transfer
and mass transport. Three sequential oxidant injection events targeting these zones was
needed to address CVOC persistence. Additionally, pre- and post-oxidation ground water
monitoring was performed to help assess and improve treatment performance.
21

-------
3.2.4 Oxidant delivery methods - direct-push and injection wells
The most commonly used methods for oxidant delivery are through direct-push injection
probes and conventional ground water injection wells (Krembs, 2008). Direct-push
injection provides wide flexibility, both in terms of injection location and vertical interval
(Huling and Pivetz, 2006). During field deployment, significant flexibility is needed in
oxidant delivery to avoid the site utilities, to evaluate the ability of the subsurface system
in accepting the oxidant solution, and to acquire new site characterization data and
information generated during ongoing field activities. In shallow, unconsolidated porous
media, direct-push injection can be fast and cheap, and multiple injection locations are
possible. However, there is direct-push depth limitation of approximately 100 ft bgs,
which may prevent the use of this technology in deeper contamination zones. Also, it
may not be possible in some geologic environments where rocks, cobbles, or boulders
prevent the injection probe from advancing into the subsurface (Huling and Pivetz, 2006).
But these limitations do not apply at site 45. Due to the limited contact area between the
injection tip and aquifer material, and the shallow injection depths, low injection rates
were projected.
Injection wells are designed and constructed for the effective delivery of oxidant into
the subsurface. The continuous slotted well screen and coarse sand pack associated with
the conventional injection well facilitate the flow of liquid from the screen, providing
excellent contact between the well and aquifer material, and a high area of contact
between the injected fluids and the aquifer (Payne et al., 2008). Specifically, the large
contact area between the sand pack and the native materials allows the oxidant to more
easily penetrate into the adjacent aquifer materials at appreciable oxidant injection rates.
Often, these injection wells are constructed with long-screened intervals of 10-15 ft
(Simpkin et al., 2012) relative to direct-push intervals of 2-5 ft. Longer screened intervals
(>15 ft) are not uncommon, however, long well screens are generally discouraged for
conventional injection wells due to the risk of preferential flow of the oxidant into zones
exhibiting greater permeability. The use of packers may help isolate sections of a long
screen and therefore deliver oxidant over partial sections of the screened interval.
Oxidant transport along the sand pack, and around the packer, may limit the ability to
deliver oxidant to specific zones, however (Payne, 2008). A bentonite seal constructed
above the sand pack is designed to prevent vertical transport of the injectate along the
disturbed aquifer materials adjacent to the well. The final design for the injection well
screen length must be appropriate for site-specific conditions.
Overall, greater rates of oxidant injection and oxidant injection volumes can be delivered
into injection wells than into direct-push injection tips. This is mainly attributed to
the high level of contact area between the screen interval and the aquifer materials.
Disadvantages of conventional injection wells include (1) higher capital costs for the
construction of the injection wells, (2) oxidant injection being restricted to the same
injection location, (3) oxidant injection is restricted to the same injection interval, unless
packers can be successfully deployed, (4) long-screened intervals in heterogeneous
aquifer materials being vulnerable to disproportionate oxidant delivery into high-
permeability zones, (5) large radii of influences often leading to heterogeneous spatial
oxidant delivery and distribution, as a result of heterogeneity and anisotropy.
For this ISCO demonstration, direct-push was selected as the main method of oxidant
delivery at site 45 due to spatial flexibility, lower initial capital cost, and the disadvantages
of injection wells, presented above. Utilizing the existing site infrastructure, an existing

-------
monitoring well was used to augment oxidant injection, discussed in section 3.2.5 below.
Additionally, a newly constructed, temporary direct-push injection well was used to
assist in the delivery of a heavy oxidant dosage into the source area. This well was a
conventional well but was advanced into the subsurface using direct-push, and is therefore
a hybrid of each. Overlapping, narrow ROIs (2-5 ft) and short injection intervals (2 ft) were
used to assure oxidant delivery to the targeted zones. The equipment used for oxidant
injection using the direct-push, the conventional injection well, and the temporary, direct-
push injection well was mobilized for each injection event.
3.2.5	Monitoring well as an oxidant injection well
Utilizing existing infrastructure at hazardous waste sites is an important component
of remedial design and can lead to minimizing remedial cost and improving remedial
efficiency and effectiveness. Construction details of ground water monitoring wells are
different than the construction details of injection wells (Driscoll, 1986; Payne, 2008).
However, monitoring wells are often constructed in ideal source area locations, indicating
their unique potential to deliver oxidants into the heart of the source area. Despite
differences in construction details between monitoring and injection wells, monitoring
wells are capable of delivering large quantities of oxidant into the subsurface. Assuming
monitoring wells are constructed in appropriate locations and screened over appropriate
intervals, they may successfully serve this application. The dual purpose role of a
monitoring well serving as an injection well is becoming more common and accepted
given the scientific rationale for their use, the compressed remediation schedules, and the
potential for time and cost savings.
At the PI MCRD ISCO site, monitoring well MW-25 was located in the source area near
the southwest corner of Building 192 and was screened over the targeted contaminated
interval of 10-15 ft bgs. MW-25, the most contaminated well at the site, represented a
convenient opportunity to inject oxidant into a highly contaminated, utility-congested
source area. Given the numerous potential impediments causing direct-push refusal,
and the associated hazards of direct-push in this zone, the use of MW-25 as an injection
well (1) reduced the need and risk of additional oxidant delivery subsurface borings in an
already congested utility zone, and (2) permitted convenient delivery of a high oxidant
loading into the heart of the source area. MW-25 was incorporated into the injection
plan and was successfully used in all three oxidant injection events, and in post-oxidation
monitoring events.
3.2.6	Top-down versus bottom-up injection methods
Two oxidant injection approaches used with direct-push oxidant injection technology
include, (1) the top-down and (2) the bottom-up. The top-down approach involves
advancing the injection tip to the first depth interval, delivering the oxidant, driving to the
next depth, and delivering the oxidant, etc. This sequence of events continues until the
final targeted depth is reached. Subsequently, the direct-push rod and injection tip string
is removed and the hole is sealed with an appropriate mixture of bentonite and cement.
The bottom-up approach starts at the lowest injection depth where oxidant injection is
initiated. The drill string is moved upward to a shallower injection interval, and oxidant
injection continues. This sequence of events continues until the shallowest targeted
depth is achieved. A potential disadvantage of the bottom-up oxidant injection approach
is that there is hydraulic short circuiting of the oxidant around the injection tip into the
underlying collapsed, or open hole. The underlying collapsed, or open borehole serves as
23

-------
a proximal preferential pathway. Short circuiting in this manner would occur in a
downward direction since there is only a short (2-4 inch) segment of the injection
tip separating the open hole below from the injection interval at the injection tip.
Pressurized oxidant injection may potentially erode and collapse the porous media near
the injection tip. Consequently, erosion of the porous media adjacent to the injection
tip would allow short circuiting of oxidant between the injection interval and the open/
collapsed hole below. Anecdotal evidence of this condition has been observed, where
the rate and volume of oxidant delivered during the bottom-up method is greater than
when using the top-down injection method. During the top-down injection method,
soil erosion due to injected fluids does not occur at the top end of the injection string,
or in the upward direction since the direct-push hole is tightly sealed by the direct-push
assembly. Therefore, the top-down injection configuration results in less risk of oxidant
short circuiting and greater certainty that the oxidant is delivered at the targeted interval.
The top-down injection method was used at the PI MCRD ISCO site.
3.2.7	Outside-in oxidant injection
ISCO application over the entire plume footprint, including the source area and
downgradient plume, is generally infeasible given the scope of such activities and the
associated cost. Rather, ISCO is most commonly used as a partial plume remediation
technology, where injection locations are designed to target the source area in an
effort to achieve maximum contaminant oxidation and mass reduction, and to reduce
contaminant concentrations at a downgradient compliance plane. A critical analysis of
partial plume remediation indicated that partial DNAPL removal from the source zone
is likely to lead to large reductions in plume concentrations and mass, and a reduction
in the longevity of the plume. Furthermore, when the mass discharge from the source
zone is linearly related to the DNAPL mass, it is shown that partial DNAPL depletion
leads to linearly proportional reductions in the plume mass and concentrations (Falta et
al., 2005a; 2005b). Therefore, a common strategy is partial source treatment to a level
allowing subsequent passive management or natural attenuation (Stroo et al., 2014).
The outside-in oxidant injection strategy is used in source area applications to minimize
the lateral displacement of contaminated ground water. When more than two adjacent
injection locations are designed, the outside-in approach targets injection locations on
the periphery of the targeted zone, then moves inward. Through this process, oxidant
injection in the central portion of the plume would still displace ground water, but would
likely contain the oxidant solution and lower concentrations of contaminants, thus
minimizing contaminant dispersal. This strategy also helps to limit the development of
stagnation zones and hydraulic interference between injection locations.
3.2.8	Injection pressure and overlying hydrostatic pressure
In saturated porous media, the overlying hydrostatic and overburden pressure plays an
important role in counterbalancing the oxidant injection pressure at the injection tip and
screened interval of the injection well. During oxidant injection into shallow aquifers,
this is especially important due to the lack of hydrostatic and overburden pressure. In
a simple homogeneous aqueous system, approximately 2.3 ft of water is required to
counterbalance 1 lb/in2 of injection pressure. In porous media, the weight of porous
media plays a role as overburden pressure. The density and weight of dry and wet
aquifer material above the injection point, relative to the weight of water, can be used to
estimate the maximum injection pressure prior to failure or breakout of the injected

-------
liquid (Appendix A. Oxidant Injection Pressure Calculations - Role of Hydrostatic and
Overburden Pressure). These calculations represent ideal conditions, and do not take into
consideration existing discontinuities in porous media, either naturally occurring or man-
made. Discontinuities represent weaknesses in the fabric of the lithology and may cause
breakout at low injection pressures.
3.2.9	Direct-push disruption of aquifer lithology
The horizontal hydraulic conductivity (Kx) is typically greater than the vertical hydraulic
conductivity (Kz); this is attributed to the orientation of clay minerals in unconsolidated
aquifer materials (Freeze and Cherry, 1979). The vertical hydraulic conductivity of many
aquifers is approximately 10 % of the horizontal hydraulic conductivity (Payne et al.,
2008). GeoProbe pneumatic hammering of the direct-push rod assembly used at the
PI MCRD site 45 resulted in the disturbance of the unconsolidated, saturated porous
media adjacent to the injection string (i.e., GeoProbe rod and injection tip). Under
this condition, it was projected that the lithologic integrity of the porous media, and
the associated hydraulic properties, were altered in a way that permitted preferential
pathways along the direct-push string assembly. Consequently, during injection activities
at the PI MCRD ISCO site, a settling period of 15-30 minutes was permitted after the drill
string was advanced to the targeted injection interval. It was projected that the settling
period would allow the porous media to re-settle, to partially re-establish the lithologic
properties, and to limit the preferential pathway along the direct-push assembly. Although
re-settling and repositioning of the aquifer particles in this manner unlikely achieved pre-
disturbed aquifer material density values, empirical observations indicated that oxidant
daylighting was limited when the settling period preceded injection, relative to immediate
pressurization and injection.
3.2.10	Field-scale ISCO deployment
The oxidant injection strategy included (1) the injection of progressively greater quantities
of the oxidant into the source areas with each successive oxidant injection event, and
(2) the complete oxidant coverage of the source area. The source area was defined as
zones within the aquifer that contained the greatest CVOC mass. The two main source
areas included the PCE release area from the sanitary sewer near the southeast corner of
Building 192, and along the longitudinal axis of the plume. Three oxidant injection events
took place, where 970 lbs, 2450 lbs, and 5070 lbs of 40% NaMn04 were injected during
the 1st, 2nd, and 3rd oxidant injection events, respectively.
The first round of NaMn04 oxidant injection in the ISCO pilot study was carried out in June,
2013. Seventeen (17) 5-gallon pails of 40% remediation grade (Rem-Ox) NaMn04
(57 lbs 40% NaMn04/pail; 970 lbs 40% NaMn04) (Carus Chemical, Peru, IL) was injected
into the subsurface using direct-push injection and using the existing monitoring well
MW-25 as an injection well (Figure 6). The second oxidant injection event (September,
2013) was more aggressive than the first as greater quantities of oxidant were injected
over a larger area (Figure 6). During the second injection event, 43, 5-gallon pails of 40%
NaMn04 (57 lbs 40% NaMn04/pail; 2451 lbs 40% NaMn04) were delivered using direct-
push injection, the existing monitoring well, MW-25, and a newly constructed direct-push
injection well, Inj.-l, Deep. The third injection event (March, 2014) involved the injection
of 89, 5-gallon pails of 40% NaMn04 (57 lbs 40% NaMn04/pail; 5073 lbs 40% NaMn04)
using direct-push, MW-25, and Inj.-l, Deep.
25

-------
Wl-Nb-S
(Fig. 6.A)
[Total CVOCsJ

M4-S-8
(Fig. 6.B)
3 mos.	X-
-6 mos.-
Oxidant
Injection - 1
Oxidant
Injection - 2
	X	
Oxidant
Injection - 3
11 mos.-
Pre-oxidation
GW Sampling
i \
Post-oxidation 1
GW Sampling
) \
1.25 mos.
Post-oxidation 2
GW Sampling
4.75 mos.
Post-oxidation 3 Post-oxidation 4
GW Sampling GW Sampling
3 mos,
8 mos.
4?

V	V	to	if-	^
nP' .n>'	.j.*6,	. &
$0'	vO
J?


&
> ^ ^
%v
&
& &
~ ^ J
Figure 6. (A) Oxidant injection locations for injection 1 (June, 2013), injection 2 (September, 2013),
and injection 3 (March, 2014). Oxidant injection transects (0X1 through OX-4) were used in the first
injection, radial distribution of MnO,-, represented by small white circles, ground water flow direction,
and nomenclature for injection locations (A-F). The ordinate of the axis system {i.e., (0, 0) is well
MW 2b which is used as an injection well. The large circles near Bldg. 192 represent oxidant injected
into wells in the source area (MW-25; Inj.-l, Deep), and the small purple circles represent oxidant
from direct push injection. Tl, T2, and T3 represent soil core sampling transects depicted in Figure 4.
(B) Pre- and post-oxidation ground water sampling and oxidant injection timeline.

-------
The oxidant was injected using an outside-in approach, where the oxidant was injected
on the perimeter and progressively moved towards the center of the plume. In the case
where two adjacent injection points were used, as in injection event 2, the outside-in
approach was not used. The oxidant was injected starting at the lower concentration
end of the plume and moved towards the higher concentration source area. The area
surrounding these wells and the first two transects was considered the high-strength
source area, likely containing PCE-DNAPL; the mass of oxidant used here accounted for
approximately 30% of the total oxidant mass injected. During each injection event, several
injection locations were not amenable to injection due to proximity to utilities, rejection of
the direct-push string, or daylighting. Minor adjustments to the injection locations, depths,
and volumes were made in the field to account for these contingencies. The subsurface
utilities were marked prior to each injection and a 2 ft buffer distance on either side of
the utility was maintained. The accurate marking of utilities by Palmetto Utility Protection
Service, Inc. (PUPS) (Columbia, SC) and by the MCRD was critical to ensure that utilities
were avoided during oxidant injection and soil coring activities.
3.2.11 Natural oxidant demand (NOD) and oxidant loading
NOD tests are performed to satisfy different objectives, including (1) as a screening tool to
assess the general feasibility of ISCO, and (2) as a design parameter in the oxidant loading
design. Oxidant loading, or dosage, involves the oxidant mass delivered {i.e., oxidant
volume x concentration) per mass of aquifer material in the targeted zone, or the mass of
oxidant per mass of aquifer material (g NaMn04/kg aquifer material). One approach used
in the design of oxidant loading involves measuring the NOD of the aquifer material in
laboratory tests, applying different adjustment factors to the laboratory-determined NOD
value, and estimating the oxidant dosage for field delivery. These adjustment factors are
applied by ISCO practitioners and are generally based on previous experience, site-specific
contaminant and hydrogeologic parameters, and possibly other factors. The final oxidant
loading involves the delivery of a "proportional" quantity of the NOD into the subsurface.
In this study, the NOD tests were conducted and the results were used as a preliminary
screening tool to determine whether the NOD was within a reasonable range of oxidant
demand values and to assess the general feasibility of the technology. The NOD results
were also used as a general check in the oxidant loading. The average oxidant demand was
approximately 2.7 g/kg (n=4) and 5.0 g/kg (n=2) in the shallower aquifer material (5-11
ft bgs) and deeper aquifer material (13 ft bgs), respectively. Rough correlations between
soil TOC and NOD have previously been established with some success (Honning et al.,
2007; Xu and Thompson, 2009). Given the increase in TOC with depth at site 45 (Section
4.1.2 below), the increase in NOD is consistent with previous correlations reported
in the literature. In one study, where the 48 hour NOD test was performed using 274
samples from 46 sites, a histogram indicated that a NOD of < 5 g/kg was measured in
approximately 70% of the samples measured, and > 5 g/kg at the remaining sites (Petri et
al., 2011, and references therein). These results suggested that the average value of the
NOD in the shallow aquifer was lower than the range of values measured at 70% of the 46
sites tested.
The oxidant loading was based on delivering NaMn04 into the two source areas
defined from the pre-oxidation soil and ground water monitoring data and information.
Specifically, this included the area adjacent to the southeast corner of Bldg. 192, where
PCE was reported to have leaked from the broken sewer pipe and where the source of the
27

-------
CVOC ground water plume originates, and along the longitudinal axis of the CVOC plume. The incremental
volume of oxidant solution to be injected at each location was based on a radial flow, cylindrical, total
porosity porous media conceptual model (Eqn. 1).
Vox,DANT=nRO|2nh	(Eqn. 1)
Where,
Voxidant~ volume of oxidant (gal)
ROI = radius of influence (ft)
r| = total porosity
h = vertical injection interval (ft)
The first oxidant injection event involved a small-scale, low volume oxidant loading approach designed
(1) to limit potential risk, (2) to gather oxidant fate and transport information, and (3) to assess the impact
on the CVOC concentration in the ground water. A small range in NaMn04 concentration (18.9-19.6 g/L)
was delivered using direct-push injection during all three oxidation events. Assuming a range in soil bulk
density (100-120 lbs/ft3) in the targeted zone, the radial-flow, total porosity, and a cylindrical, porous
media conceptual model) the oxidant loading (i.e., oxidant volume x concentration) delivered to the
idealized pore volume, was projected to be
3-4 g NaMn04/kg aquifer solids.
Based on the average NOD values measured in the 5-11 ft bgs interval (i.e., 2.7 g NaMn04/kg aquifer
material), the oxidant loading was projected to exceed the laboratory-measured NOD. The persistence
of unreacted oxidant residual in ground water would be vulnerable to transport through advection and
diffusion beyond the immediate targeted zone. Thus, it was projected that the injected oxidant would
cover a larger aquifer volume than the radial flow, cylindrical, total porosity porous media conceptual
model. Given the projected moderate costs in oxidant and labor required to deliver 3-4 g NaMn04/kg
aquifer solids, a decision was made by the Parris Island Partnering Team to move forward with pilot-scale
deployment of ISCO and to further evaluate the feasibility of ISCO as a final remedy.
3.3 Ground water sampling and analysis
Additional site characterization data and information was initiated to augment the previous work by
the USGS and to refine the conceptual site model. Specifically, the CVOC distribution in the source area
and along the longitudinal axis of the plume, where the majority of CVOCs were surmised to occur was
quantified.
3.3.1 Ground water
A comprehensive pre-oxidation baseline ground water sampling event was performed prior to injection
of the oxidant. Ground water samples were collected from 39 micro-wells, MW-25, Inj.-l, Deep, and
MW-31SL (Figure 5). Post-oxidation ground water sampling was performed, involving the same set of
micro-wells and other wells, to assess oxidant fate and transport, treatment performance, and to guide
subsequent injections. The samples were analyzed for an array of parameters using either standard EPA
methods, or procedures developed and approved for use at the R.S. Kerr Environmental Research Center
and referred to as R.S. Kerr standard operating procedures (RSKSOP) (Table 3) (please refer to section 3.7
for information on the analytical methods). Laboratory-based analyses were performed at RSKERC as well
as contract laboratories. Field analysis was performed by the EPA field crew.

-------
Molecular biological tools (MBT) and compound-specific isotope analyses (CSIA) were
performed by the contractors at Microbial Insights and Microseeps, Inc., respectively.
Prior to ground water sample collection, a YSI multi-parameter probe with a flow through
cell was used to collect field parameters, including pH, dissolved oxygen (DO), oxidation
reduction potential (ORP), temperature, specific conductivity, and turbidity. Ground water
samples were collected when field parameters stabilized (<10% variability).
During ground water sampling, some ground water samples contained visible
concentrations of permanganate, in binary mixture samples, as per the pink-purple color.
Previous studies indicate that without appropriate preservation, the quality of the ground
water sample may be compromised, and false negative results may occur. Therefore, the
guidelines, methods, and procedures presented in Ko et al. (2012) were used in this study
to preserve binary mixture ground water samples.
Table 3. Ground water parameters and methods used in analysis of samples.
Parameters Method
pH, turbidity, temperature, ORP, DO
YSI Multi-parameter system with flow-through cell
CVOCs (PCE, TCE, cis-l,2-DCE,
trans-l,2-DCE, 1,1-DCE, VC)
EPA Method 8260B/C GC/MS
Metals
EPA Method 6010C (ICP-AES); EPA Method 200.7
Dissolved methane gas
(1»RSKSOP-194/175,Rev. 5
Ferrous and total iron
EPA Method 3500-Fe D
Chloride and sulfate
EPA Method 6500
(refer to RSKSOP-276, Rev. 4 below)
Compound specific isotope analysis
(CSIA)
(2) analyses performed by Microseeps, Inc.
220 William Pitt Way, Pittsburgh, PA 15238
Molecular biology tools (MBT)
(2) analyses performed by Microbial Insights,
2340 Stock Creek Blvd., Rockford TN 37853
111 There is no existing EPA method for dissolved methane gas. Please refer to Kampbell and Vandegrift, 1998
in the reference section.
121 CSIA and MBT were measured in six wells in the source area (Ml-mid-shallow and -deep, Ml-NA-shallow
and -deep, and Ml-SA-shallow and -deep).
EPA Method 8260B - Volatile organic compounds by purge and trap GC/MS.
EPA Method 6010C (EPA SW-846) and EPA Method 200.7 - Inductively coupled argon plasma with atomic
emission spectrometry.
RSKSOP-194, Rev. 4 - Gas Analysis by Micro Gas Chromatograph.
RSKSOP-175, Rev. 5 - Sample Preparation and Calculations for Dissolved Gas Analysis in Water Samples Using a
GC Headspace Equilibration Technique.
EPA Method 3500-Fe D - Phenanthroline method.
RSKSOP-276, Rev. 4 - Determination of Major Anions in Aqueous Samples Using Capillary Ion Electrophoresis
with Indirect UV Detection and Empower 2 Software.
29

-------
3.3.2 Contaminant mass flux
Contaminant mass flux is a quantitative parameter describing the contaminant mass
transported through a prescribed plane over time (Eqn. 2). Contrasting mass flux
estimates before and after ISCO deployment is a useful metric to critically assess changes
in contaminant mass transport across the site, and ISCO performance. The calculation
is based on Darcy's Law, where the ground water velocity, total CVOCs concentration,
and aquifer cross-sectional area are used to estimate mass flux. Site-specific parameters
used in these calculations are provided in Table 4. Given the clusters of micro-wells in
the shallow and deep zones (Figure 5), four mass flux transects across the CVOC plume
were established as zones MF-1 through MF-4 (Figure 7). The ground water velocity
associated with transect MF-1 was corrected to include directional differences between
the ground water flow direction and the coordinate system at the MF-1 transect (i.e.,
the y-direction ground water flow component does not go through MF-1; the x-direction
is perpendicular, Vx = VD cos 33.6°, Vx = 0.833 VD). The distances between monitoring
locations in transects MF-2 through MF-4 were calculated based on the relative
differences between the ground water flow direction and the coordinate system
(i.e., cos 33.6° = VD/VY, VY = 1.2 VD). Micro-well location M3-N-S did not function correctly
during ground water sampling and a reliable ground water sample was unobtainable.
The concentration of CVOCs from Tl-N-S was used as a substitute for M3-N-S in mass
flux calculations for MF-3.
Contaminant Mass Flux (CMF) = Vs x [CVOC] x Area (Eqn. 2)
Where,
Vs = VD/n = Kx(dH/dL)/n
Vs = Seepage velocity (ft/day)
[CVOC] = Total CVOC concentration (ng/L)
A = cross sectional area (depth x width)
VD = Darcy velocity (ft/d)
r| = porosity
K = hydraulic conductivity (ft/d)
dH/dL = hydraulic gradient (ft/ft)
3.3.3 Impact on natural attenuation
As reported above in section 2.2.1 (USGS Site Characterization), an abundance of site
specific indicator parameters strongly suggest that natural attenuation processes are
playing a significant role in the attenuation of CVOCs at the site. The surficial aquifer is
anaerobic at most locations with the predominant terminal electron accepting process
being iron reduction in the shallowest sediments, and sulfate or iron reduction in the
deeper sediment, including methanogenesis (Vroblesky et al., 2009). Molecular biology
tools and contaminant specific isotope analysis were utilized to assess the impact of
ISCO on microbial populations and natural attenuation processes.
30

-------
Table 4. Parameters used in the contaminant mass flux calculations.

Aquifer
Hyd. Cone.
Hyd. Grad.
Cross sectional area
Porosity
Zone
(K)'1'
(dH/dL) (ft/ft)'1'
(A) (ft2)'2'
n(1)
Shallow (7-10)
4.30E-03
0.006
171 (MF-1)
0.3



153 (MF-2)
0.3



168 (MF-3)
0.3



183 (MF-4)
0.3
Deep (10-12.25)
1.00E-03
0.004
114 (MF-1)
0.3



102 (MF-2)
0.3



112 (MF-3)
0.3



122 (MF-4)
0.3
1	Hydraulic properties used in these calculations were based on previous investigations
(Vroblesky et at., 2009).
2	Width of mass flux transect: MF-1 (57'); MF-2 (51'); MF-3 (56'); MF-4 (61') (Figure 7).
> Ground water (low direction
Mass llu.N transects (MF-1 through MF-4 located in
both the shallow (7-10*) anU deep zones (10-12,25')),
Total ICVOCs]
(Ug/L)
188000
•64000
¦60000
H 78000
!72000
968000
-64000
Jeoooo
¦156000
-52000
-^44000
40000
-	30000
-	32000
-	28000
-24000
— 20000
18000
12000
6000
4000
J0
Figure 7. Pre-oxidation baseline contaminant mass flux transects (MF-1 through
MF-4) used in the shallow and deep zones of the aquifer in the ISCO study area.
Due to clustering of shallow and deep micro-wells at the field site, a similar set
of transects was used for both the shallow and deep zones of the aquifer.

-------
3.4 Health and safety plan and permits
3.4.1 Health and safety plan
A project-specific and site-specific health and
safety plan (HASP) was prepared in 2008, prior to
field activities carried out at Site 45. The HASP was
prepared in accordance with the requirements
of the EPA - Robert S. Kerr Environmental
Research Center (EPA-RSKERC). It was reviewed
and approved by the RSKERC Safety, Health and
Environmental Management Program Manager
and was provided to the Parris Island Partnering
Team (PIPT) as part of the in-situ chemical
oxidation pilot study work plan. The PIPT provided
comments to the draft HASP based on their
individual site-specific knowledge. A Material
Safety Data Sheet (MSDS) was provided by the
manufacturer (Carus Corp., Peru, IL) of the sodium
permanganate and included as an attachment
to the HASP. A presentation "The Safe Use and
Handling of Permanganate Products", prepared by
Carus Corp. was included as an attachment to the
HASP. The HASP was updated as needed prior to
each new stage of the study. In 2013, the HASP was
revised, updated, and re-formatted to meet new
HASP guidelines adopted by the US EPA National
Risk Management and Research Laboratory
(NRMRL). Each member of the field crew reviewed
the HASP prior to beginning field work, and daily
safety meetings were held to discuss current site
conditions.
The HASP included detailed information on a
range of important topics: project description,
field activities, laboratory activities, physical
hazards summary, personal protective equipment
(PPE) summary, equipment requirements,
chemicals to be used, waste management, sample
management, spill response, authorized personnel,
staff concurrence, job hazard analysis, controls, and
PPE, emergency telephone numbers and hospital
information, a generic emergency response plan,
emergency recognition and prevention, electrical
safety, physical hazards, personnel roles, lines of
authority, and communication procedures during
an emergency, procedures for emergency medical
treatment and first aid, and emergency contacts.
The job hazard analysis covered all the tasks for
each stage of the field and laboratory work.
3.4.2	Injection permit
The injection of permanganate during the field
study required the issuance of an underground
injection control (UIC) permit by the South Carolina
Department of Health and Environmental Control
(SC DHEC), Ground-Water Protection Division. The
application was submitted by MCRD environmental
staff to SC DHEC on behalf of the RSKERC ISCO
team. Once the UIC permit application was
approved by SC DHEC, a construction permit
to install the direct-push rods was issued by SC
DHEC and the permanganate injections were
subsequently fully permitted to commence.
3.4.3	Well permit
The installation of monitoring wells, including
micro-wells and direct-push temporary well, at
Site 45 required the approval of SC DHEC, Bureau
of Water. Information on the location, type, and
construction details of the wells to be installed
was provided to the SC DHEC for technical review
and approval. Once SC DHEC staff members were
satisfied with the proposed well installation,
approval for construction was given by the SC
DHEC. Following the monitoring well installation,
SC DHEC required the submission of SC DHEC Form
1903, Water Well Record.
3.4.4	Dig permit
Submission and approval of a Ground Penetrating
Activity Permit Request, or the "dig permit",
was required by MCRD for all activities that
disturbed the subsurface. A dig permit form was
prepared and distributed to all relevant MCRD
departments (i.e., those with a stake in subsurface
activities). Each stage of investigation involving
soil coring, well installation, and oxidant injection
activities required that a dig permit be prepared
and approved prior to commencement of field
activities.
32

-------
3.5 Oxidant handling
The 40% sodium permanganate oxidant
(permanganate) delivered to the site was a
hazardous chemical; it is a strong oxidizing agent
and can react violently with oxidizable materials
and was handled in accordance with the proper
safety requirements. Before permanganate oxidant
injections began, MCRD placed a metal hazardous
materials storage unit at the site for secure storage
of the 40% liquid sodium permanganate. Upon
delivery of the 40% sodium permanganate, by
Carus Corp., the 5-gallon buckets of 40% sodium
permanganate were placed in the storage unit.
The storage unit was placarded with "oxidizing
material" signs and was locked. Individual 40%
sodium permanganate 5-gallon pails were
removed one at a time from the storage unit, as
they were needed, and moved a short distance
to the 350-gallon injection tank. The appropriate
personal protective equipment (PPE), including a
face shield, plastic gloves, and a rubber apron over
a Tyvek suit, was used during the transport and
transfer of the 40% sodium permanganate. The
oxidant was transferred from the 5-gallon pail to
the mixing tank, where it was mixed with water
to create the dilute (~1%) injection solution. The
40% sodium permanganate solution was metered
into the mixing tank via a peristaltic pump to avoid
potential contact and exposure. The much lower
oxidant solution concentration (~1%) was less
hazardous and appropriate PPE consisted of Tyvek
suits, vinyl gloves, and eye protection.
Potential release and exposure to the 1% solution
was mitigated through the presence of shut-off
valves, pressure relief valves with return lines
to the mixing tank, pressure gauges, secondary
containment, and high pressure-rated and
corrosion-resistant hoses in the injection manifold
and distribution lines. A dilute ascorbic acid
solution was kept in pressurized spray bottles
near each of the permanganate handling locations
for neutralization of minor spills and leaks of the
diluted oxidant solution.
3.6 Design and operation of injection system/equipment
The selection of direct-push as the main method
of oxidant delivery in the pilot study (see section
3.2.4 above, Oxidant delivery methods - direct-
push and injection wells) required that an oxidant
injection system be designed, constructed, and
operated as a part of the pilot-scale study project.
Preliminary calculations indicated that the three
proposed oxidant injection events, if contracted
with an ISCO vendor, would significantly exceed
the research budget. Therefore, the design criteria
for the injection system included low cost, and
ease of mobility, so that it could be transported
back and forth between the EPA, GWERD facility
(Ada, OK) and the MCRD site 45 (Parris Island, SC).
The injection system was designed and constructed
by the GWERD staff at the GWERD facility and
the details of the injection system, including a
schematic and description of injection pallet
components, component manufacturers, part
numbers, cost, and relevant description details are
included in Appendix B.
The cost of the injection system for injection
arm (i.e., header) was $16,000. Three arms were
designed and constructed in the system, allowing
the oxidant to be injected at three locations
simultaneously. All parts were constructed
with corrosion-resistant components allowing
the system to be used with strong oxidants,
or other corrosives. These include reductants,
acids, solvents, or surfactants, if needed. Each
injection arm, or header, was equipped with a flow
dampener, pressure release valve and return line,
pressure control valve, pressure gauge, flowmeter/
totalizer, emergency cutoff ball valve, and a 45-60
ft long, high pressure injection hose. The injection
hose was designed to fit inside the GeoProbe
rod as a separate unit and attached directly to
the injection tip. The rationale of this design
was to maintain hydraulic control of the oxidant.
Alternatively, oxidant can be injected down
through the GeoProbe rod but there is a greater
potential for leaks to develop, risk personnel
exposure.

-------
The oxidant injection occurred
at low pressures, resulting in low
to average injection rates (0.5-
1.5 gpm) per location. However,
at three injection locations, the
overall injection rate was tripled
(1.5-4.5 gpm). This was due to
(1) the shallow injection intervals
and limited overburden pressure,
and (2) to the subsurface utilities
and the potential for other
preferential pathways, such as
improperly sealed exploratory
borings. Only low injection
pressures (<10-15 psi) were used
throughout the demonstration.
The use of high oxidant
injection pressure can lead to
hydraulic short circuiting and
breakout of the injected oxidant
into preferential pathways.
Consequently, an unintentional
and disproportionately high
volume of oxidant may be
transported into non-targeted
zones.
The on-site monitoring well, MW-
25, and the direct-push injection
well (Inj.-l, Deep) are 2-inch
wells constructed with screened
intervals approximately 10-15
feet bgs. Oxidant injection into
these wells was accomplished
with above-ground, 55-gallon
HDPE drums, using a peristaltic
pump with Phar-Med tubing
{% inch) and PTFE pipe (Figure
8). All components in contact
with the oxidant were corrosion
resistant. The fittings between
the injection hose and the 2-inch
wells involved PTFE couples, %
inch nipples, and hose clamps.
Slow oxidant injection rates
(0.5-1.0 gpm) were used to
limit mounding of oxidant and
ground water and to reduce
the potential for discharge
into nearby storm and sanitary
sewers.
§
/
A I' A
h * | *
Figure 8. Schematic of Injection apparatus used in the delivery
of oxidant into injection wells MW 25 and Inj.-l, Deep.

-------
3.7 Analytical and Quality Assurance/Quality Control
Upon collection, ground water samples collected
for CVOC analysis were preserved with HCI.
Aquifer subcores were collected from aquifer
cores, placed in 40 mL vials, and amended with
Methanol (MeOH) (15 mL). The MeOH-soil slurry
was shaken intermittently and allowed sufficient
contact time for CVOC extraction (>48 hr). The
MeOH extracts of the soil cores were analyzed for
CVOCs at the EPA GWERD, using EPA Method 8260C
with headspace sample introduction using EPA
Method 5021A as implemented in GWERD analytical
method RSKSOP-299/2 "Determination of Volatile
Organic Compounds (Fuel Oxygenates, Aromatic
and Chlorinated Hydrocarbons) In Water Using
Automated Headspace GC/MS." All quality control
(QC) results met criteria established in RSKSOP-299/2.
MeOH extracts were further prepared by adding 250
piL of the methanol/water mix to a vial containing 10
mL of boiled milli-Q water and two grams of sodium
chloride. CVOC analytical results were reported per
dry weight of soil extracted.
Metals analysis of ground water samples was
performed via inductively coupled plasma, optical
emission spectrometry (ICP OES) (Perkin Elmer,
Model Optima 3300 DV, Norwalk, CT), using EPA
Method 200.7 as implemented in RSKSOP-213,
Rev. 5 "Standard Operating Procedure for Operation
of Perkin Elmer Optima 3300DV ICP". All QC results
met criteria established in RSKSOP-213/5. Metals
analysis of soil core samples collected at site 45
were also analyzed by RSKSOP-213/5 after the soil
samples were extracted using EPA Method 3051A
as implemented in GWERD method RSKSOP-180/3
"Standard Operating Procedure for Total Nitric
Acid Extractable Metals from Solids and Sludges
by Microwave Digestion". Here, representative soil
samples were digested, using nitric acid (40 mL, 10%)
in a microwave oven (40 min; 150 °C; 1000 kPa).
Loss of the on-site analytical contract at the EPA
GWERD during the course of this demonstration
required that analytical support be provided
by an off-site laboratory. Shealy Environmental
Services (Columbia, SC), accredited by the National
Environmental Laboratory Accreditation Program,
was contracted and carried out similar GC/MS and
ICP OES analytical methods. They analyzed ground
water and MeOH extracts using equivalent methods
of analysis (EPA method 8260B; EPA Method 6010C,
(water only)) and QA/QC requirements.
Chloride (CI ) and sulfate (S042 ) in ground water
samples were analyzed using EPA Method 6500 as
implemented in GWERD method RSKSOP-276/4,
"Determination of Major Anions in Aqueous
Samples Using Capillary Ion Electrophoresis with
Indirect UV Detection and Empower 2 Software". All
quality control results met the criteria established
in RSKSOP-276, Rev. 4. Ferrous iron (Fe+2) and total
Fe in ground water was measured using the EPA
Phenathroline method (EPA Method 3500-Fe D).
Prior to ground water sample collection, a YSI multi-
parameter probe with a flow through cell was used
to collect field parameters, including pH, dissolved
oxygen, oxidation reduction potential, temperature,
specific conductivity, and turbidity. Ground water
samples were collected when field parameters
stabilized.
Aquifer sub-cores collected from aquifer cores for
TOC analysis were treated with hydrochloric acid
to remove inorganic carbon and then analyzed by
RSKSOP-120, Rev. 3 for organic carbon by combustion
and subsequent detection and quantitation of carbon
dioxide using a LECO CR-412 carbon analyzer with an
infrared detector.
Ground water samples collected for dissolved
methane analysis by RSKSOP-194/175, Rev. 5 were
placed in 60 mL serum bottles, without headspace
and preserved with HCI. A headspace of helium was
created in the sample bottle and shaken to allow the
gases to equilibrate between the aqueous and gas
phases. A gas-tight syringe was used to withdraw a
portion of the headspace and injected into a micro-
gas chromatograph for separation of the gases and
detection and quantitation by a thermal conductivity
detector.
Molecular biology tools (Microbial Insights, 2017)
and compound-specific isotope analysis (CSIA) (Pace
Analytical, 2017) were analyzed in ground water
samples collected from six micro-wells. Analyses
included total eubacteria (qEBAC), Dehalococcides
(qDHC), and DHC functional genes (TCEa reductase,
BAV1 vinyl chloride reductase (BVC), and vinyl
chloride reductase (VCR). The CSIA for carbon on PCE,
TCE, c-DCE, and vinyl chloride were analyzed. These
micro-wells included both the shallow and deep
micro-wells at locations Ml-S-A, Mi-Mid, and Ml-
N-A. Samples for analysis by molecular biology tools
(MBT) and compound-specific isotope analyses (CSIA)
35

-------
were collected by the Navy and submitted to contract
laboratories for analysis. Sample preservation and
handling of samples analyzed using MBT (qPCR or
quantitative polymerase chain reaction) and CSIA
were performed by the contractors at Microbial
Insights and Microseeps, Inc., respectively. The data
was made available to the EPA. The quality assurance
and quality control procedures established by these
laboratories for these analyses were followed.
As required by US EPA QA policy, a Quality Assurance
Project Plan (QAPP) was prepared and approved
for this project prior to collection of data and
implemented without significant deviations. A QAPP
describes the technical and QA/QC activities of an
environmental research project that is implemented
to ensure that the results will satisfy the intended
used of the data.
Field sampling quality control included the collection
of duplicates (both ground water and aquifer
subcores) and blanks for ground water including field,
equipment, and trip blanks. Analytical quality control
included positive controls (calibration checks, matrix
spikes, and second source checks), negative controls
(blanks), and duplicates. Data quality acceptance
was initially determined by the analyst using the
procedure's acceptance criteria. The Principal
Investigator ultimately determined whether or not
data was usable for the project. All data used in this
report satisfied QA/QC requirements.
CVOCs in both the ground water and aquifer cores
were considered a critical parameter needed to
meet the project objectives. As mentioned earlier
in this section, all QC criteria performed at the EPA
GWERD were met for RSKSOP-299/2 (EPA Method
8260C). These included method blanks which should
not have analyte concentrations above the Method
Detection Limit (MDL), calibration checks and second
source standards should be within +/-20% of their
known value, matrix spikes and laboratory control
spikes should have 70-130% recovery, and laboratory
duplicates should have a Relative Percent Difference
of <20%. Surrogates were also added to all samples
and QC samples with a required recovery of 80-120%.
The MDL and QL (Quantitation Limit) were 0.5 and
0.09 - 0.22 pig/L in the water samples, and 0.004 and
0.02 mg/kg in the aquifer core samples (assumes 25
ml total MeOH extract (MeOH + soil water) and 25
g dry soil). Note, the range in QL for water samples
reflects the 6 CVOCs detected (i.e., VC, 1,1-DCE, trans-
1,2-DCE, cis-l,2-DCE, TCE, PCE). Sample results below
the QL were considered to be estimated.
Samples analyzed for CVOCs by Shealy Environmental
Services also met the QC criteria for EPA Method
8260B based on the criteria in the DOD Quality
Systems Manual for Environmental Laboratories,
Version 4.2. These included method blanks which
should not have analytes detected above % the
RL (Reporting Limit) and > 1/10 the amount in any
sample, for calibration checks the average Response
Factor for VOC SPCCs (System performance check
compounds) should be >0.30 for chlorobenzene
and >0.1 the remainder, second source standards
should be within +/-20% of true value, matrix
spike and laboratory control spikes should be
trans-l,2-Dichloroethene 60-140%, cis-1,2-
Dichloroethene 70-125%, 1,1-Dichloroethene 70-
130%, Tetrachloroethene 45-150%, Trichloroethene
70-125%, Vinyl chloride 50-145%. Surrogates were
also added to all samples and QC samples with a
required recovery of l,2-Dichloroethane-d4 70-120%,
Toluene-d8 85-120%, Bromofluorobenzene 75-120%,
and Dibromofluoromethane 85-115%. The MDL and
QL (Quantitation Limit) for undiluted samples were
0.25 and 0.5 pig/L in the water samples, respectively.
The MDL and QL in soil were 0.002-0.02 mg/kg and
0.025 mg/kg, respectively (assumes 25 ml total
MeOH extract (MeOH + soil water) and 25 g dry soil).
Sample results below the QL were considered to be
estimated.
3.8 Data analysis
Statistical analysis involving the 95% confidence interval (i.e., 95% CI = Avg. ± standard error (S.E.), where S.E. =
(t005 x std. dev.) / (n)0 5; std. dev. = standard deviation; t005 = two tail t-values at the 0.05 level of significance and
n-1 degrees of freedom) in post-oxidation [CVOCs]SOIL involved the two-tailed , standard t-tables 0.05 level of
significance.
Isocontour plots of chemical parameters were prepared for the study area using Surfer* 7.0 (Golden Software).
Linear kriging was used for the gridding method, and filled contours over similar concentration ranges for
specific parameter data sets were used. Post maps showing well locations were used based on an (X, Y)
coordinate system where the ordinate (i.e., 0, 0) was selected as well MW 25-SL.

-------
4. Results
A visual perspective of site activities is provided through numerous site photographs in Appendix C
(Photographic compendium of ISCO activities at the site 45ISCO demonstration project (Parris Island,
MCRD, SCJ). A short narrative is provided for each photo to give background information of the ISCO and
ISCO-related site activities.
4.1 Soil/aquifer cores
4.1.1 Visual inspection of core material
Visual inspection of the subsamples collected from the soil and aquifer cores revealed distinct geochemical
conditions and layering. The shallow, 3-10 ft bgs, aquifer material was comprised of light-colored sandy
material. This depth interval was sometimes overlain by a distinct orange-colored sandy layer, indicating
the presence of iron oxides resulting from exposure to air during low water table events (Figure 9). The
light-colored sandy material graded into a darker sandy material with depth containing greater levels of
silt, clay, and organic matter. Aquifer cores collected from 16 ft and lower were very dark and consistently
exhibited pieces of decaying wood of natural origin (i.e., not building or commercial debris).
Figure 9. (See above.) Photo of soil cores and depth below
ground surface.
37

-------
4.1.2 Total organic carbon
Total organic carbon (TOC) was analyzed in aquifer cores collected at transects Tl-B, T3-B, and T5-B (8-16
ft bgs) (Figure 4) over 0.5-1.0 ft intervals. The average TOC values were lower in the 8-12 ft bgs interval
than in the 12-16 ft bgs interval (see summary below). A distinct correlation between TOC and depth was
not established though (Figure 10), suggesting that there is not a distinct, high organic, lithologic layer in
the subsurface over this interval. Rather, a general increasing trend in TOC concentration with depth is
observed. Examination of aquifer cores at 16 ft bgs indicated an abundance of organic materials, such as
wood or peat, underlying the contamination zone. The darker color of aquifer material in the deeper cores
was partially attributed to the organic matter, which is characteristically dark.
Transect-Location
Interval
TOC (%)
Number of samples analyzed
Tl-B
8-12 ft
0.120
9

12-16 ft
0.227
8
T3-B
8-12 ft
0.112
7

12-16 ft
0.259
6
T5-B
8-12 ft
0.107
10

12-16 ft
0.281
6
d)
U
£
3
CO
"O
c
3
O
i
o
tt)
CD
-C
cL
OJ
Q
-16
y = -12.46x - 9,78; R* = 0.42
0 00	0 10	0.20	0.30	0.40	0 50
Total Organic Carbon (%)
Figure 10.
TOC concentration with
depth from aquifer samples
collected at transect
locations Tl-B, T3-B, and
T5-B (refer to Figure 4 for
transect locations).

-------
4.1.3 Metals and Sulfur
Sub-cores were collected from the 8-12 ft and 12-
16 ft bgs cores in replicate, extracted, and analyzed
for metals via ICP OES. Results of these analyses
indicate that iron, manganese, and sulfur increase
with depth (Figure 11). An abundance of iron and
sulfur at the site suggests that naturally occurring
minerals containing these elements likely serve as
a source of terminal electron acceptor in reductive
dehalogenation natural attenuation mechanisms.
Under reduced conditions, these geochemica!
minerals would provide an abundance of dissolved
species in the ground water. Upon oxidant
injection however, it is projected that a significant
increase in the oxidation potential would alter the
geochemistry resulting in a decline in the solubility
of these species.
OT1
¦ T2
AT3
XT4
* T5
0T6
Unear Regression
y =-0.0007x - 8.78, R? - 0.53
T2-B*4and T 2-B-4b data not
included in tho rogrossion analysis
\
5000 10000 15000 20000 25000
Iron Concentration (mg/Kg)
e
O
5
o
07
CD
s
OT1
¦ T2
AT3
XT4
* T5
OT6
Linear Regression
y =-0.095x - 8.57; RJ = 0.67
,00

T2-B-4 and T2-B-4bdata not
Included intho regression analysis
\
100	150	200
Manganese Concentration (mg/Kg)
m
on
¦ T2
AT3
XT4
* T5
OT6
Linear Regression
y =-0.0011x-9.47; R2-0.65
30000
T2-B-4andT2-B-4bda!a not
included in ttie regression analysi

—30O6-
10000 15000 20000 25000
Sulfur Concentration (mg/Kg)
30000
Figure 11. Concentration of Iron, manganese, and sulfur in soil with depth.

-------
4.1.4 CVOCs
Aquifer cores were collected at the locations
illustrated in transects T-l through T-6 (Figure 4;
Appendix D, Figure D.l). Sub-cores were collected
at a minimum of five locations for each core at
the following depths: 8 ft, 10 ft, 12 ft, 14 ft, and
16 ft bgs. Sub-core aquifer samples (~25-40 g
dry weight) were extracted with methanol and
analyzed for CVOCs via GC/MS. These total-extract
samples represent CVOCs found in both the
aqueous and solid (adsorbed) phases. Based on
a contaminant mass distribution analysis, it was
determined that NAPLs were not present in these
subsamples. Total CVOCs were mostly limited
to the 8-12 ft bgs interval (Appendix D, Figures
D.2-D.3).
Additional site characterization in the source area
was performed approximately 20-22 ft upgradient
of transect T-l, closer to the decommissioned
sanitary sewer line near the south-southeast
corner of Building 192 (Figure 4; Appendix D,
Figure D.l). Aquifer cores were collected along a
transect (T-0) located very near the corner of the
dry cleaner building, and adjacent to well MW-
25 (Figure 4). Cores were also collected (1) at the
same location where injection well Inj.-l, Deep
was installed, and (2) at other locations near the
decommissioned sanitary sewer line. The total
CVOCs analytical results of cores collected along
transect T-0 and in the source area clearly indicated
that the total CVOCs concentrations in these areas
were the highest at the site, the contamination
extended over a broader vertical interval extending
from =7.5-15 ft bgs, that the contamination was
predominantly distributed along the longitudinal
axis, and that lateral distribution was limited
(Appendix D, Figures D4-D5). Furthermore, based
on a contaminant mass distribution analysis of the
total CVOCs concentration data, it was concluded
that PCE DNAPL was present in the aquifer core
collected at the Inj.-l, Deep location (i.e., PCE
DNAPL was likely when PCE concentrations were
greater than 225 mg/kg).
The aquifer core collection and analysis provided
data and information used to refine the conceptual
site model. Specifically, that contamination was
predominantly constrained between 8-12 ft bgs,
except in the source area where it extended over
a broader interval (=7.5-15 ft bgs). Consequently,
the oxidant injection strategy was designed
to focus oxidant injections into the 8-12 ft bgs
contaminated interval, and that more aggressive
oxidant loading was needed in the source area at
the southeast corner of Bldg. 192.
Based on the general lithology performed for
well PAI-45-MW-28D, loose sand existed at
approximately 6-11 ft bgs (Vroblesky et al., 2009;
Figure 17). The loose sand was underlain by silty
sand that extended down to 16 ft bgs. The loose
sand and silty sand were expected to have high
and low hydraulic conductivity, respectively.
Conceptually, the general interface area between
these two zones would represent a lower
boundary that limited downward contaminant
migration and distribution. Given that the source
of contamination was introduced above this
elevation, the vertical migration would have been
impeded by the silty sand layer. This conceptual
model was consistent with ground water samples
collected and analyzed in temporary wells collected
above and below this zone (Vroblesky et al., 2009).
Further, as discussed below, aggressive natural
attenuation conditions in the lower surficial aquifer
were also projected to limit the transport of CVOCs
into the lower surficial aquifer. Overall, given the
elevation of the TCE release from the vitrified clay
pipe, the high hydraulic conductivity of aquifer
material < 12 ft bgs, and the aggressive natural
attenuation fate mechanisms and lower hydraulic
conductivity occurring > 12 ft bgs, aquifer core
CVOC concentrations were consistent with TCE
presence, persistence, and rapid transport in the
8-12 ft bgs interval.
40

-------
4.2 Oxidant delivery and impact on total CVOCs in ground water
The oxidant loading design predominantly included
direct-push delivery of Mn04" at 15 g/L (17.9 g/L
NaMn04), in 2-ft increments into the 8-10 ft bgs
and 10-12 ft bgs intervals. Assuming a range in
bulk density (1.61-1.93 kg/L, 100-120 lbs/ft3), the
oxidant loading calculated within the design radius
of influence was 3-3.6 g NaMn04/kg, but the actual
oxidant loading delivered into the subsurface
ranged slightly higher (3.0-3.9 g/kg). The oxidant
loading was greater than the measured oxidant
demand of <2.7 g/kg (see Section 3.1.4 Natural
Oxidant Demand), suggesting some oxidant
residual persisted and was transported in the
ground water over a limited time frame. During
the third and final injection event, a higher oxidant
concentration (30 g/L Mn04") was delivered to
the centerline of the CVOC contaminant plume,
where high concentrations of CVOC were known
to occur. Given the repeated delivery of oxidant
in the high concentration CVOC source areas of
the site during oxidant injection events 1-3, it is
likely that the oxidant demand was satisfied and
that downgradient oxidant drift occurred in some
areas. In the source area located at the southeast
corner of Building 192, the highest concentrations
of CVOCs were measured during baseline ground
water and soil monitoring events. Consequently,
greater oxidant loading in the source area was
achieved by injecting higher concentrations (36.5
g/L) and volume of Mn04" into the wells (MW-25;
Inj.-l, Deep) present in the source area.
The density of the NaMn04 oxidant solution at
2% (i.e., 20 g/L) is slightly greater than water
and would have a mild downward transport
component. It is reasonable to assume that the
oxidant injected in the source zone at a higher
average concentration (i.e., 3.65%) also underwent
vertical transport and eventually rested on or near
the lower permeable materials at 15 - 16 ft bgs.
Conceptually, this is ideal since it is a known hot
spot area, the oxidant will naturally migrate into
the low permeable media with time, and oxidant
transport will be limited due to low permeability.
Overall, this helps to assure oxidant delivery
into the source zone, long term persistence
of the oxidant, and good contact between the
oxidant and contaminated media, an important
requirement for successful chemical oxidation.
4.2.1 First oxidant injection event (June 23-
Sodium permanganate was injected at multiple
direct-push locations and depths near the corner
of Building 192, and into MW-25 (Figures 6, 8, 12).
The injection interval was 2 ft and the injection
depths reported below represent the top of the
interval. Specifically, there were six injection
locations along the first oxidation transect,
OX-1, at four depths (6 ft, 8.5 ft, 11 ft, 13.5 ft
bgs) per location. On the other three transects
(OX-2, OX-3, OX-4) there were two depths (8 ft,
10 ft bgs) (Figure 12). Not all of the projected
injection locations were successful or precisely
located due to proximity of subsurface utilities,
daylighting, and refusal of the injection tip. Refusal
was attributed to competent, but unknown
subsurface material possibly including a former
building pier, construction debris, etc. An oxidant
I, 2013)
solution was injected (0.5 gpm) into MW-25 using
a 55-gallon drum, a peristaltic pump, and a 2 in
x 2 in flexible coupling pumping system (Figure
8). A total of 17, 5-gallon pails of NaMn04 (57
lbs, 40% NaMn04/pail) were injected. During the
first oxidant injection event, approximately 2205
gallons ([NaMn04]AVG = 18.1 g/L) and 480 gallons
([NaMn04]AVG = 13.6 g/L) were injected using
direct-push technology and the existing monitoring
well MW-25, respectively. The sanitary and
storm sewers in the vicinity of the injection zone
were routinely inspected over the course of the
oxidant injection activities. The absence of visual
observation of Mn04" in these sanitary and storm
sewers indicates that there were no releases into
these conduits.
41

-------
A)
ft(bgs)
-- 3
Key	n
Permanganate Injection
Micro-well (24" SS screen) Q
Ground Water Flow Direction
-- 6
Ml	M2	MB
7.5' 7,5'	15'	15'
«	^	X	
M4
30'
Ox-1 Ox-2 Ox-3	Ox-4
-- 12
New Dry
Cleaner Facility
~
~
Key
Permanganate injection transect (Ox-1, -2, •
3, -4, See Inset below for injection locations)
Nested micro-well cluster
Sentry micro-well cluster
Monitoring transects (M1-M4)
Inset: Injection locations schematic
(refer to Table 1)
A B C D E F
Ox-1
42
Figure 12. Conceptual model of permanganate injection in the south CVOC plume at Parris
Island MCRD site 45. A) Cross section schematic of permanganate injection and micro-well
screened intervals; B) Plan view schematic of ground water plume, oxidant injection, and
ground water monitoring micro-well transects.

-------
As a result of the first oxidant injection event, it was apparent that (1)
a significant volume and quantity of oxidant could be injected into the
contaminated interval(s), (2) the oxidant was delivered and was contained
within the specific targeted zone, and (3) there was no oxidant solution
leakage into the nearby sanitary or storm sewers. Wells in the southern
plume were sampled on August 5-9, 2013, and indicated that NaMn04
persisted (0.05 to < 1 g/L) in several wells in the source area (MW-25, Mi-
mid-shallow, M2-mid-deep), indicating that the NaMn04 had not completely
reacted. NaMn04 was not detected in the downgradient sets of micro-wells
along the longitudinal axis of the plume (M3-mid-shallow/deep; M4-mid-
shallow/deep) (Figure 5) where oxidant loading had not occurred. This result
suggested that rapid transport of oxidant down the longitudinal axis of the
plume did not occur.
4.2.2 Post-oxidation 1 CVOC concentrations
A schematic illustrates the base map locations of the micro-well clusters,
shallow and deep, and MW-25 (Figure 13). This is the base map used in
Figures 6-7, illustrating oxidant injection locations and mass flux transects,
respectively. The pre-oxidation concentrations of CVOC in ground water
in the shallow (7-10 ft) and deep (10-12.25 ft) zones are represented
by isocontours and provide useful information regarding CVOC fate and
transport trends in the ISCO study area (Figure 14a). These results confirmed
the conceptual model: that the source of the CVOCs is located near the
corner of the new dry cleaner building, ground water flow is from the
northwest to the southeast, and that total CVOC plume is limited laterally.
These results were consistent with previous investigations indicating that
the source of PCE was from the old dry cleaner facility; the PCE entered the
sanitary sewer system and then leaked from the sanitary sewer located at
the corner of the new dry cleaner facility (Vroblesky et al., 2009). Faster
transport, greater plume length, and more widespread distribution of the
CVOCs in the shallow zone relative to the deeper zone is partially attributed
to higher hydraulic conductivity and hydraulic gradient in the shallow zone,
and aggressive natural attenuation processes in the deeper zones.
The areal extent of oxidant distribution during the first oxidant injection
event was limited relative to the plume footprint. Because the oxidant was
injected in the source area, significant CVOC destruction was achieved as
observed by contrasting the pre- and post-oxidation CVOC isocontours
(Figures 14a and 14b). Results of the first oxidant injection indicated that the
oxidant was successfully delivered to the targeted zones, hydraulic control of
the injected oxidant was achieved, and the oxidant persisted in the source
zone. Furthermore, based on a comprehensive ground water monitoring
program, significant CVOC destruction was evident through lower CVOC
concentrations.
43

-------
Ml-Nb-S

M1-NS-S
T-
M2-N-3

-20
-30-
-40-
-50
M2-Mid~S
+
M2-3-3
M1-S6-S +
T1-S-S
T1-N-S
I—
10
M3-Md-S
T
M3-S-S
T2-S-S
W3-N-S
—I
20
T
30
—I—
40
T2-N-S
M4-N-S
+
Mrl-Mkl-i
~I	
60
50
M1-NM>
20-r-^—
10-
o-H-
-f-
.20-
-30-
-40-
-60-
+
M2-N-D
+
M2-Mld-D
4-
M2-S-D
Ml-Sb-D
T
T1-8-D
+
10
T1-M-0
+
M3-MH-D
+
M3-S-D
+
—I—
2D
—r
30
T2-3-D
+
M3-N-0
W4-S-D
40
—n
50
T2- +-D
M4-N-D
+
M4-Wid-D
+
—I—
60
44
Figure 13. Shallow (7-10') (top figure) and deep (10.25-12.25') (bottom figure) well locations
and names at Parris Island, SC Marine Corps Recruit Depot (MCRD) Site 45.

-------
MVNb-S
M1-Na-S

M2-S-S
M2-N-S
M1-Sb-S
-30
-10
T1-S-S
TV

M3-N-S
'°°Oo
uws
10000
M3-S-S
M4-N-S
M4-M6-S
Total CVOCs
(Mg/L)
Issooo
I 34000
180000
_] 76000
T2-N-S *72000
Tsaooo
164000
leoooo
156000
152000
146000
144000
140000
136000
132000
128000
124000
120000
' 16000
12000
8000
4000
0
-50
10	20
T2-S-S
30	40
M4-S-S
50
60	70

10
Ml-Na-D
10000
M2-N-D

-20
M1-Sb-D
M2-S-D
-30
-40
-50
T1-S-D
T1-N-D

M3-N-D
M3-Mid-D
M3-S-D
10	20
T2-S-D
30	40
M4-S-D
50
M4-N-D
M4-Md-D
Total CVOCs
(Mg/D
Issooo
184000
180000
176000
172000
T2-N-D H 68000
164000
60000
56000
52000
48000
44000
40000
36000
32000
28000
24000
20000
16000
12000
8000
4000
0
60
45
Figure 14.A. Pre oxidation shallow (February, 2013) (7-10 ft bgs) (top figure) and deep
(10-12.25 ft bgs) (bottom figure) total chlorinated volatile organic compounds (|Jg/L)
(replicates) at Parris Island, SC MCRD Site 45.

-------
M1-Nb-S
10
M1-«a-S
M2-W-S - 0

M2-Mjd-S
nW-Sa-S
»20
M1-Sb-S
M2-S-S
-30
-40
-BO
T1-S-S o
T1-N-S
| f
*°Oo
Sooo
12000
'saoo
M3-Mid-S
\ '6,
M3-S^
M3-f^S
— 0	T2-N-S
lOOO
M4-N-S
\
M4-M6-S
T2-S-S
0	10 20	30 40
M4-S-S
90

eo
Total CVOCs
(yg/L)
88000
84000
80000
76000
72000
68000
64000
60000
56000
52000
48000
44000
40000
36000
32000
28000
24000
20000
16000
12000
8000
4000
0
70
M1-Nb-D
20
¦MI-ffeD
M2-N-D
Ti-r+p
OX*
s*o
-20
-30
-40
-50
l-Mid-D
'eooo
M1-Sb-D
T1-S-0
10
20
*°Oo
M3-S-D
T2-S-D
\ +
30	40
M3-N-D
vzooo
M2-S-0	M3-Md-D

Total CVOCs
T2-N-D
M4-N-D
M4-Mi(J-D
*000
M4-S-D
so eo
(Mg/D

88000

84000

80000

76000

72000

68000

64000

60000

56000

52000

48000

44000

40000

36000

32000

28000

24000
P|
20000
16000
12000
8000
4000
0
Figure 14.B, Post-oxidation 1 (August, 2013) shallow (7-10 ft bgs) (top figure) and deep
(10-12.25 ft bgs) (bottom figure) total chlorinated volatile organic compounds (|jg/'L)
at Parris Island, SC MCRD Site 45.

-------
4.2.3 Second oxidant injection event (September 22-27, 2013)
Based on the results and conclusions from the first injection event, the strategy
developed for the second oxidant injection was to more aggressively target the
source area near the corner of Building 192 and along the longitudinal axis of the
downgradient plume, where the majority of the CVOCs mass resided. Sodium
permanganate was injected at multiple direct-push locations and depths in the source
area and along the longitudinal axis of the plume (Figure 6). Additionally, a greater
oxidant dosage was injected into MW-25 and into the newly constructed injection well
(Inj.-l, Deep) (Figure 8). Overall, 43, 5-gallon pails of NaMn04 (57 lbs, 40% NaMn04/
pail) were injected.
Direct-push injection. Oxidant was injected along eight transects on the longitudinal
axis of the plume, consisting of two injection locations per transect (14 locations), and
at two additional single locations in the source area and downgradient from M3-mid
(Figure 6). At each transect, the oxidant was injected approximately 7 ft apart and at
two depth intervals (8-10, 10.5-12.5 ft bgs). Approximately 135 gallons of NaMn04
oxidant solution was injected, resulting in a theoretical 3 ft radius of influence (ROI).
This ROI ideally assumes a cylindrical shape, a 2 ft vertical interval, and a porosity =
0.3. The locations of these transects were selected to lie between previous oxidant
injection locations and to extend further down the plume. Relocation of some of
the pre-determined oxidant injection locations was required due to the proximity of
subsurface utilities.
Injection wells. MW-25, an existing monitoring well used as an injection well, is
screened 10-15 ft bgs in both the loose sand (6.5-11.5 ft bgs) and in the underlying
silty sand (11.5-17.5 ft bgs) (USGS, 2009). It was estimated that the permeability in
the loose sand is approximately four times greater than in the underlying silty sand
(TetraTech, 2004). It was projected that the oxidant volume injected into MW-25
during the first injection (480 gallons; [NaMn04]AVG =13.6 g/L) was predominantly
delivered into the overlying loose sand, but was also introduced into the silty sand.
During the second oxidant injection event, an increase in the oxidant dosage was
delivered into MW-25 (640 gallons; [NaMn04]AVG = 37 g/L) to achieve an approximate
4-5 ft radius of influence (i.e., assuming a 5 ft screened interval,
porosity = 0.3).
The direct-push injection well (Inj.-l, Deep) was installed in the source area,
approximately 7 ft to the northwest of MW-25. The objective was to deliver more
oxidant into the source area adjacent to and underlying the suspected leaking sanitary
sewer line. The construction of this well, is described above in section 3.1.3. A heavy
oxidant loading (125 gallons; [NaMn04]AVG = 111 g/L) was successfully injected into this
well to achieve approximately a 2-3 ft radius of influence.
4.2.4 Post-oxidation 2 CVOC concentrations
Contrasting Figures 14a and 14c, it is evident that significant and more widespread
CVOC destruction was achieved through the second oxidant injection. Results of the
second oxidant injection confirmed the findings from the first oxidant injection event:
(1) the oxidant was successfully delivered to the targeted zones, (2) hydraulic control
of the injected oxidant was achieved, (3) the oxidant persisted in zones where a heavy
oxidant loading was delivered, and (4) that significant CVOC destruction was achieved.
47

-------
M1-Nb-S
Total CVOCs
10
M1-^S	—m*'
M2-M6-S
!\1?-Sa-S
M3-N-S
-20
M1-Sb-S
-30
Tt-S-S
-40	O
M2-S-S
T2-N-S
10000
M3-Md-S
">°00
M3-S-S
M4-f^S
M4-WW-S
(Mg/L)

88000

84000

80000

76000

72000

68000

64000

60000

56000

52000

48000

44000

40000

36000

32000

28000

24000

20000

16000

12000

8000

4000

0
-50
10
20
T2-S-S
30	40
M4-S-S
SO
60
70
M1-NMD
20
M1-Na-D
M2-N-D
M2-KW-D
-20
M2-S-D
M1-SW)
-30
-40
T1-S-D
T1-N-D 0
M3-N-D
T2-N-D
M3-MW-D
M3-S-D
M4-N-D
M4-Mid-D
Total CVOCs
(Mg/L)
80000
84000
80000
76000
72000
68000
64000
60000
56000
52000
48000
44000
40000
36000
32000
28000
24000
20000
16000
12000
8000
4000
0
-50
10
20
T2-S-D
30	40
M4-S-D
50
60
48
Figure 14.C. Post-oxidation 2 (February, 2014) shallow (7-10 ft bgs) (top figure) and deep
(10-12.25 ft bgs) (bottom figure) total chlorinated volatile organic compounds (pg/L) at
Parris Island, SC MCRD Site 45.

-------
4.2.5 Third oxidant injection event (March 24-29, 2014)
Based on the results from the first two oxidant injection events, the ISCO
strategy during the third oxidant injection was to aggressively target the
source area(s) and to deliver the oxidant into zones where CVOCs had
persisted. Again, this included the southeast corner of Building 192 and
along the longitudinal axis of the CVOC plume. Greater oxidant dosages
of sodium permanganate were delivered at multiple direct-push locations
and depths in the source area and along the longitudinal axis of the plume
(Figure 6). Greater oxidant dosage was also injected into MW-25 and Inj.-l,
Deep (Figure 8). Overall, there were 87, 5-gallon pails of NaMn04 (57 lbs,
40% NaMn04/pail) injected, which was greater than the mass of oxidant
injected during the first two events combined. Given the repeated delivery
of oxidant in the high concentration CVOC source areas during oxidant
injection events 1-3, it is likely that the oxidant demand was satisfied and
that oxidant residuals in some areas would drift downgradient.
Direct-push injection. Oxidant was injected at 14 transects along the
longitudinal axis of the plume, consisting of 3-5 injection locations
per transect and at two additional locations in the source area and
downgradient from M3-mid (Figure 6). Approximately 7900 gallons
([NaMn04]AVG = 26 g/L) was injected using direct-push technology. At
each transect, the oxidant was injected approximately 7 ft apart and at
two depth intervals (8-10, 10.5-12.5 ft bgs). Approximately 135 gallons
of NaMn04 oxidant solution was injected per interval, resulting in a
theoretical 3 ft ROI. This ROI ideally assumes a cylindrical shape, a 2
ft vertical interval, and porosity = 0.3. The locations of transects were
selected to lie between previous oxidant injection locations and to extend
further down the plume. The oxidant was injected using an outside-in
approach, where the middle injection was last. Specific transect locations
targeted high CVOC concentration zones in the subsurface, and the oxidant
was injected starting at the lower concentration end of the plume and
moved towards the higher concentration source area.
Injection wells. An aggressive oxidant dosage was delivered into MW-25
(640 gallons, [NaMn04]AVG = 41 g/L) to achieve an approximate 4-5 ft ROI,
and into Inj.-l, Deep (218 gallons, [NaMn04]AVG = 41 g/L) to achieve an
approximate 3-4 ft ROI.
4.2.6 Post-oxidation 3 CVOC concentrations
By contrasting Figures 14.A and 14.D, it is evident that significant and
widespread CVOC destruction was achieved by the third oxidant injection
event. Noticeably absent are the high concentration isocontours near the
source area and at the M3-Mid-shallow micro-well monitoring location,
where CVOC concentrations were historically high and persistent. Results of
the 3rd oxidant injection confirmed the findings from the first two injection
events: (1) the oxidant was successfully delivered to the targeted zones,
(2) hydraulic control of the injected oxidant was achieved, (3) the oxidant
persisted in zones where the heavy oxidant loading was delivered, and
(4) that significant CVOC destruction was achieved.
49

-------
Ml-Nb-S
10
-20
-30
-40
-50
M1-Na-S
M2-N-S T1_N^


m
M2-Mid-S
I

M&N-S
T2-I
Sa-S



M2-S-S
M1-Sb-S
M3-Mid-S

M4-N-S
T1-S-S
M3-S-S
T2-S-S
M4-S-S
M4-N/W-S
Total CVOCs
(Mg/L)
180000
176000
172000
[68000
164000
160000
156000
152000
148000
144000
140000
136000
132000
I28000
124000
120000
'16000
'12000
8000
4000
0
10	20	30	40
50
60
70
M1-Nb-D
20
10
M1-Na-D
M2-N-D
T1-N-D
0
¦(W-Sa-D
M2-NM-D
M3-N-D
T2-N-D
-20
M2-S-0
WB-Mid-D
Ml-Sb-D
-30
-40
T1-S-D
M3-S-D
M4-N-D
IVM-Mid-D
Total CVOCs
(Mg/L)
I80000
176000
[72000
168000
164000
160000
[56000
152000
148000
144000
140000
136000
132000
128000
124000
120000
j16000
J 12000
8000
4000
0
-50
10
20
T2-S-D
30	40
M4-S-D
50
60
50
Figure 14.D. Post-oxidation 3 (June, 2014) shallow (7-10 ft bgs) (top figure) and deep
(10-12.25 ft bgs) (bottom figure) total chlorinated volatile organic compounds (|Jg/L) at
Parris Island, SC MCRD Site 45.

-------
4.3 Mass flux
The CVOC mass flux (refer to Eqn 2 in Section 3.3.2
Contaminant mass flux) is an estimate of the VOC
mass in ground water, passing through a plane.
The mass flux is defined, or quantified, by the
micro-wells used in each of the four monitoring
transects (Figure 7). It was evident that the
majority of the CVOC mass flux was attributed to
the shallow interval (7-10 ft bgs) rather than the
deeper interval (10-12.25 ft bgs) (Figure 15.A).
This was attributed to lower CVOCs concentration,
slower transport, and smaller cross sectional area
associated with the deeper interval. Below the
12.25 ft bgs elevation, the CVOC concentrations
in the aquifer material were significantly limited
(refer to section 4.1.4, above), and the hydraulic
conductivity was lower due to the increase in clay
content. The baseline mass flux results indicated
that focusing the oxidant in the 7-12 ft bgs interval
should efficiently destroy the most mobile and
greatest mass of CVOCs at the site, ideally, this
approach was projected to significantly reduce the
CVOC concentrations and have the greatest impact
on limiting CVOC transport from the site.
The oxidant injected during the first oxidant
injection event was focused oniy in the areas
represented by mass flux transects 1 and 2. The
oxidant was not injected near mass flux transects
3 and 4 and therefore, the greatest reduction in
CVOCs occurred in the areas of mass flux transects
1 and 2 (Figure 15.A). A significant reduction in
mass flux was not observed in the deeper interval,
suggesting that greater oxidant delivery was
needed in that interval during subsequent injection
events. An unexpected increase in mass flux was
estimated at MF-3. During oxidant injection, the
displacement of ground water by the injected
oxidant can temporarily change ground water flow
lines and directions. This can potentially result in
changes in ground water CVOC concentrations in
side and downgradient directions. The spike in
CVOC mass flux in transect MF-3 was attributed to
an increase in [CVOCs] in downgradient micro-well
M3-Mid-S. Overall, a 22.6% and 2.5% reduction in
mass flux was measured in the shallow and deep
intervals respectively (Table 5).
Figure 15. A.
Pre-oxidation (baseline)
and post-oxidation 1 CVOC
mass flux across transects
MF 1 through MF-4 in
shallow and deep zones
in the study area (refer to
Figure 7 for locations of
mass flux transects).
6000
5000
>¦4000
TO
M
3.
X 3000
2000
1000

N T/ "b k
~ # # # #
I I Pre-Oxidation (baseline)
I _ I Post-Oxidation 1
3 cn ^
1/1 ^
dm
psn ffln
N 'V ?>
# # # #
Mass Flux Transects
51

-------
Table 5. Summary of mass flux results for baseline and post-oxidation events 1-3.
Mass Flux Transect
MF-1
Total CVOCs Mass Flux (|ig/day)
MF-2 MF-3
MF-4
Total mass flux decline (%)(1)
Shallow





Baseline
5730
3280
2960
2330

Post-Oxid. 1
850
1740
5900
2570
22.6
Post-Oxid. 2
770
1000
3550
1940
49.3
Post-Oxid. 3
210
240
90
670
91.5
Deep





Baseline
690
590
170
250

Post-Oxid. 1
740
540
200
180
2.5
Post-Oxid. 2
560
220
150
120
50.4
Post-Oxid. 3
40
245
50
80
75.5
111 The total mass flux decline is relative to baseline (pre-oxidation conditions).
The second oxidant injection event caused the continued decline in total CVOC mass flux in the shallow
mass flux transects MF-1 and MF-2, and significant declines at MF-3 and MF-4 (Figure 15.B). This was
attributed to the more aggressive and widespread oxidant injection associated with the second oxidation
event. A greater impact was also observed at the deep interval mass flux transects relative to the first
injection, but the magnitude was small compared to the shallow interval. Approximately a 50% decline in
mass flux was achieved in both the shallow and deep intervals after the second oxidation event (Table 5).
More oxidant was delivered in the targeted source zones in the third injection event than in both the first
and second injections combined (Figure 6; section 4.1.3). Correspondingly, a significant and widespread
decline in CVOC mass flux was observed in both the shallow and deep intervals (Figure 15.C). Overall, a
91.5% and 75.5% reduction in total CVOC mass flux occurred as a result from oxidant injections 1-3 (Table 5).
Figure 15. B.
Pre-oxidation (baseline)
and post-oxidation 2 CVOC
mass flux across transects
MF-1 through MF-4 in
shallow and deep zones
in the study area (refer to
Figure 7 for locations of
mass flux transects).
6000
5000
>¦4000
00
3
^ 3000
2
in
(/I
ra
2 2000
1000
> JV <> ,?> ->
Pre-Oxidation (baseline)
I I Post-Oxidation 2
M KO 2
_ m ^
w
£ LA IN S

-------
Figure 15. C.
Pre-oxidation (baseline)
and post-oxidation 3 CVOC
mass flux across transects
MF-i through MF-4 in
shallow and deep zones
in the study area (refer to
Figure 7 for locations of
mass flux transects).
6000
5000
>4000
TO
73
X 3000
2000
1000
<>  <>
J> $ ^ 4 4
Pre-Oxidation (baseline)
Post-Oxidation 3
IX>	o>
LL
js \ 'V ft

-------
over the length of the injection zone. The post-
oxidation 3 ground water sampling event occurred
three months after the third oxidant injection and
revealed oxidant persistence over the length of the
injection zone. Permanganate was measured in 6
wells (MW-25; Inj.-l, Deep; Ml-Mid-shallow, Ml-
Mid-deep, Ml-S-A-shallow; M4-Mid-deep). Finally,
the post-oxidation 4 ground water sampling event
(Feb. 27-28, 2015) occurred > 11 months after the
third permanganate injection and permanganate
was found in six wells (MW-25; Ml-SA-shallow;
Ml-Mid-shallow; Ml-Mid-deep; M2-S-deep; M4-
Mid-deep).
It is evident from these results that permanganate
persisted mainly in the source area near the
southeast corner of Building 192, where heavy
oxidant dosages were delivered near the suspected
PCE release point(s) along the sanitary sewer line.
Ideally, oxidant persistence in this area results
in long-term contact between the oxidant and
the contaminated media. Oxidant persistence
along the longitudinal axis, over the length of the
injections, was also evident from ground water
monitoring results. Long term persistence of the
oxidant resulted in improved contact between
the oxidant and contaminated media, as a result
of enhanced CVOC mass transfer and transport
processes: (1) CVOCs slowly dissolving from DNAPL,
(2)	CVOCs desorbing from aquifer solids, and
(3)	CVOCs diffusing from low permeability
silty media. Further, oxidant diffusion into low
permeability media can be improved through long
term oxidant persistence.
Given the repeated delivery of oxidant and the
potential for overlapping ROIs during oxidant
injection events 1-3, it is probable that the oxidant
demand was satisfied, that downgradient drift
of oxidant residual occurred, and that oxidant
distribution occurred over a greater area than
depicted by the ROIs. Excessive oxidant persistence
may allow oxidant transport into non-targeted
areas. At the time of the last ground water
sampling event (Feb. 27-28, 2015), permanganate
had not been detected in downgradient well MW-
31 SL, which is 75 ft downgradient of monitoring
transect M4 (i.e., the furthest downgradient
location where oxidant was injected).
4.5 General indicators of oxidation
The indicator parameters including chloride (CI"), ferrous iron (Fe+2), and oxidation reduction potential
(ORP) were measured prior to and after the oxidant injections. These parameters potentially provide
insight regarding the impact of oxidant injection on total CVOC oxidation and dechlorination, or CI" release,
and change in the redox potential in the subsurface.
4.5.1 Chloride (CI )
Dechlorination of the CVOCs results in the release of CI" ions in solution. Background CI" levels in surficial
aquifers near marine environments typically exhibit elevated concentrations of CI". Further, elevated CI"
levels in the study area can also be attributed to the long term dechlorination of CVOCs through natural
attenuation reductive dechlorination mechanisms known to occur at this site as described by Vroblesky
et al. (2009). Contrasting pre-oxidation CI" isocontours in the study area with post-oxidation isocontours
revealed marginal evidence of the formation of CI" ions released as a result of total CVOC oxidation. It is
evident that the pre-oxidation baseline concentration of CI" in the shallow aquifer interval is lower than
in the deeper interval. However, despite significant reductions in total CVOCs (Figures 14.A-D), a stark
contrast was not observed between pre-oxidation and post-oxidation CI" isocontours (Figures 16.A-D).
Consequently, CI" data resulting from post-oxidation ground water sampling provided limited insight into
ISCO treatment performance due to elevated background levels of CI".

-------
M1-Nt>-S



Chloride ion
10 M1-Na-S
M2-N-S TWWS


(mg/L)
M2-Mid-S
1

M3-N-S
T2-N-S

1500
1400
1300
1200
lJ?-Sa-S




1100
1000
900
-20
M2-S-S
M1-Sb-S
M3-Mid-S

M4-N-S

800
700
600
+
-30




500
400
T1-S-S
M3-S-S

M4-Mid-S

300
200
100
0
-50-j
T2-S-S
M4-S-S



0	10
M1-Nb-D
20	30	40	50	60	70
20
10
M1-Na-D
M-MM89
o 4
fW-
-20
M2-N-D
-30
^0
-50
500 M2-Md-D
%
M1-S0-0
M2-S-D
T1-S-D
T1-N-D
%
M3-McJ-D
M3-S-D
T2-S-D
M3-N-D
%
T2-N-D
M4-N-D
IW-Mld-D
10
20	30
40
M4-S-D
50	60
Chloride ion
(mg/L)
1500
- 1400
11300
1200
Jl100
11000
900
800
- 700
600
500
400
300
200
100
0
55
Figure 16.A. Pre-oxidation shallow (7-10') (top figure) and deep (10,25-12.25') (bottom figure)
chloride ion (CI) concentrations (mg/L) at Parris Island, SC MCRD Site 45 (2/19/13 - 2/21/13).
Note: there was an outlier data point in Tl-N-S (763 mg/L) that was deleted from the data set.
All 3 samples that followed were <100 mg/L.

-------
Ml-Nb-S
10
M1-IMa-S
M2-N-S
T1-W-S

Msa-S
M2-WS«J-S
-20
M1-Sb-S
M2-S-S
-30
-40
T1-S-S
-60
M3-Md-S
M3-S-S
10	20
T2-S-S
30	40
M3-N-S
M4-S-S
50
T2-
M-S
M4-f+S
M4-Mid-S
Chloride ion
(mg/L)
\ 1500
-1400
1300
»1200
1100
= 1000
-900
800
700
600
-500
400
300
200
100
0
60	70
M1-Nb-D
10
Ml-Na-D

fW-
-20
-30
-40
-50
M2-N-D
600
M2-Mid-D
%
M2-S-D
M1 -Sb-D *-
T1-S-D
T1-N-D
M3-N-D
%
M3-Mid-D
M3-SD
T2-S-D
%
M4-S-D
T2-N-D
M4-N-D
M4-Mid-D
Chloride ion
(mg/L)
; 1500
1400
¦	1300
-1200
I 1100
i 1000
¦	900
-800
700
-600
500
-400
- 300
200
100
0
10	20
30
40	50
60
56
Figure 16.B, Post-oxidation 1 shallow (7-10') (top figure) and deep (10.25-12.25') (bottom
figure) chloride ion (CI) concentrations (mg/L) at Parris Island, SC MCRD Site 45.

-------
M1-Nt>S
10
M1-Na-S
M2-N-S T1"N"S

iW-Sa-S
M2-Mld-S
-20
M1-Sb-S
M2-S-S
-30
^to
-50
T1-S-S
M3-Mid-S
M3-S-S
10
20
T2-S-S
30	40
M3-N-S
M4-S-S
50
T2-N-S
M4-N-S
M4-Mld-S
60
70
Chloride ion
(mg/L)
1500
1400
11300
' 1200
11100
11000
-900
800
-700
600
- 500
400
300
200
100
0
M1-Nb-D
20
10
M-Mwaas
0 ft
-20
-30
^10
-50
M1-Na-D
M2-N-D
M2-MKJ-D
$Oo
\
M2-S-D
M1-Sb-D
T1-S-D
T1-N-D
%
m-Md-D
M3-S-D
T2-S-D
B4D
10
20
30
40
%
WW-S-D
50
T2-N-D
M4-N-D
M4-Md-D
Chloride ion
(mg/L)
¦11500
™ 1400
11300
ll200
¦1100
J 1000
-I 900
J 800
1700
600
I 500
400
300
200
100
0
60
57
Figure 16.C. Post-oxidation 2 shallow (7-10') (top figure) and deep (10.25-12.25') (bottom
figure) chloride ion (CI) concentrations (mg/L) at Parris Island, SC MCRD Site 45.

-------
M1-Nb-S
10
M1-Na-S
M2-N-S
nW-Sa-S
M2-Mict-S
-20
-30
-40
-50
M1-Sb-S
T1-S-S
M2-S-S
T1-N-S

M3-S-S
T2-S-S
— —i	~ _i_
10	20	30	40
M3-N-S
M4-S-S
60
T2-N-S
M4-N-S
M4-KM-S
Chloride ion
(mg/L)
1500
M1400
" J 1300
<| 1200
11100
1000
•[ 900
800
J 700
-j 600
-j 500
'400
300
200
100
0
60	70
M1 -Nb-D
20
10-
Ml-Na-D
u-mtea
ot
iflfi-
M2-N-D
-20
-30
^J0
-50

%
M2-Mid-D
M2-S-D
M1-Sb-D
T1-S-D
T1-N-D
%
M3-Md-D
M3-S-D
%
T2-S-D
10	20	30	40
M3-N-D
T2-N-D
M4-N-D
M4-Mid-D
500
M4-S-D
50	60
Chloride ion
(mg/L)
31500
— 1400
-1300
1200
j 1100
1000
1900
jj 800
1700
¦600
- 500
400
300
200
100
0
58
Figure 16.D, Post-oxidation 3 shallow (7-10') (top figure) and deep (10.25-12.25') (bottom
figure) chloride ion (CI) concentrations (mg/L) at Parris Island, SC MCRD Site 45.

-------
4.5.2 Ferrous iron (Fe+2)
The presence of naturally occurring organic matter in the aquifer
material was a general indicator that reducing conditions existed at
the site. This was especially the case at depth, where an abundance
of organic material, such as decaying wood material and wood chips,
was visibly present in the aquifer cores. Further, sewage inputs into
the subsurface were known to occur as a result of leaking sanitary
sewer line(s). Sewage odors emanating from the work area during the
collection of aquifer cores provided anecdotal evidence supporting
the presence of sewage residuals in the subsurface. Ferrous iron,
a byproduct of ferric iron (Fe+3) reduction, is a general indicator of
reducing conditions and was projected to be present in the ground
water, especially since the aquifer material sampled and analyzed
contained up to 10 g/kg (Figure 11). Under increasingly reduced
conditions, iron solubility increases and soluble ferrous iron was
measured in abundance under pre-oxidation, or baseline, conditions
in the ground water (Figure 17.A). Ferrous iron concentrations
progressively declined with increasing applications of oxidant (Figures
17.B-D). The redox shift towards oxidized conditions was expected
to be temporary, however, given the significant quantity of organic
materials and the abundance of iron in the subsurface. Nevertheless,
the absence of ferrous iron in ground water along the longitudinal axis
of the plume and in the source area indicated the significant impact of
the oxidant injected in these areas.
59

-------
M1-Nb-S
10
M1-Na-S

l\W-Sa-S
M2-N-S
£
T1-N-S
M2-Mid-S
tvm
-20
M1-Sb-S
M2-S-S
-30
-40
-50
T1-S-S
0	10
M1-Nb-D
20 I
M3-N-S
12
M3-MiiJ-S
M3-S-S
~
T2-S-S
20
30
40
M4-S-S
50
T2-N-S
WI4-N-S
M4-Md-S
60
70
Ferrous iron
(mg/L)
26
125
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
Ferrous iron
10
M1-Na-D
0 ^
M2-N-D
M2-Mjd-D
M
-20
-30
-40
-50
M1-Sb-D
<1
M2-S-D
T1-S-0
If
T1-N-D

1/
M3-Md-D
M3-S-D
T2-S-D
M3-N-D
10
20
30
40
M4-S-D
50
M4-N-D
M4-M6-D
12
T2-N-D
(mg/L)

26

25

24

23

22

21

20

19

18

17

16

15

14

13

12

11

10

9

8

7
—
6

5

4

3

2
60
60
Figure 17.A. Pre-oxidation shallow (7 10') (top figure) and deep (10.25-12.25') (bottom
figure) ferrous iron (Fe+2) concentrations (mg/L) at Parris Island, SC MCRD Site 45 (2/19/13 -
2/21/13).

-------
M1-Nb-S
10
M1-N3-S
M2-N-S

ld?-Sa-S
M2-Mid-S
-20
M1-Sb-S
M2-S-S
-30
-40
-50
T1-S-S
T1-N-S
\+
'0
0	10
M1-M3-D
20
M3-Mid-S
20
M3-S-S
20
20
30
T2-S-S
M3-N-S
40
M4-S-S
50
Ferrous iron
(mg/L)
'o
T2-N-S
M4-N-S
M4-Mid-S
60
70

26
24
22
20
18
16
14
12
10
8
6
4
2
0
10
M1-Na-D
M1WNCB
0 !

M2-N-D
M2-A/M-D
-20
-30
-40
-50
10
M1-Sb-D
M2-S-D
T1-S-D
\6
T1-N-D
M3-Md-D
M3-N-D
M3-S-D
10
T2-S-D
10
20
30
40
M4-S-D
50
T2-N-D
M4-N-D
M4-Md-D
Ferrous iron
(mg/L)
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
-I9
8
7
6
5
4
3
2
1
0
60
61
Figure 17.B. Post-oxidation 1 shallow (7-10') (top figure) and deep (10.25-12.25') (bottom
figure) ferrous iron (Fe+2) concentrations (mg/L) at Parris Island, SC MCRD Site 45.

-------
Ml-Nb-S
10
M1-N3-S
m
fJl9-Sa-S
-20
-30
-40
-50
M2-N-S
10
M2-Mid-S
M1-Sb-S
T1-S-S
M2-S-S\0
T1-N-S
f
M3-Md-S
10
M3-S-S
10
20
30
T2-S-S
I
M3-N-S

T2-N-S
M4-S-S
M4-N-S
10
M4-Mid-S
Ferrous iron
(mg/L)
26
24
22
20
18
¦I 16
114
12
10
8
6
4
2
0
40
50
60
70
M1-Nb-D
20
10
M1-Na-D
M1WWB
0—
iW-Sa-D
M2-N-D
M2-Mid-D
-20
-30
^0
-50
10
Ml-Sb-D
M2-S-D
T1-S-D
T1-N-D
M3-Mid-D
M3-S-D
10
20
T2-S-D
30	40
M3-N-D
>0
Ferrous iron
(mg/L)
T2-N-D
M4-N-D
M4-NW-D
M4-S-D
50	60

26

24

22

20

18

16

14

12

10

8

6

4

2

0
62
Figure 17.C. Post-oxidation 2 shallow (7-10') (top figure) and deep (10.25 12.25') (bottom
figure) ferrous iron (Fe+2) concentrations (mg/L) at Parris Island, SC MCRD Site 45.

-------
Ml-Nb-S
10
M1-N3-S
M2-Mid-S
#-Sa-S
M2-N-S
>0
-20
M2-S-S
M1-Sb-S
-30
-40
T1-S-S
T1-N-S
M3-Mid-S
M3-S-S
M3-N-S
Ferrous iron

(mg/L)






—
26



24
T2-
M-S

22



20



18



1.6



14
M4-N-S


12



10



8
10


6
M4-Mid-S


4
/ ->v \ \


2


!	
0
-50
o	10
M1-Nb-D
20
20
T2-S-S
30	40
M4-S-S
50	60
70
10
M1-Na-D
MWMW2B
OH
-rW-Sa-D
It
-20
M2-N-D
M2-Mid-D
M1-Sb-D
!
-30
T1-S-D
^tO
M2-S-D
70
T1-N-D
"O
M3-Mid-D
+
M3-S-D
M3-N-D
T2-M-D
M4-M-D
M4-Mid-0
Ferrous iron
(mg/L)
6
4
I—12
0
-50
10	20
T2-S-D
	
30	40
M4-S-D
50	60
Figure 17.D. Post-oxidation 3 shallow (7-10') (top figure) and deep (10.25-12.25') (bottom
figure) ferrous iron (Fe+2) concentrations (mg/L) at Parris Island, SC MCRD Site 45.

-------
4.5.3 Oxidation reduction potential (ORP)
Pre-oxidation measurements of ORP were only measured in three shallow and deep
micro-well clusters along monitoring transect Ml, including Ml-S, Mi-Mid, and Ml-N. The
pre-oxidation ORP values indicated that the ground water was under reducing conditions
(6.6-66.6 mV) for the three nested well pairs (Table 6). A significant post-oxidation increase
in ORP was measured in all wells along monitoring transect Ml, indicating the impact of
nearby oxidant injection.
Table 6. Summary of ORP measurements in micro-wells near the source area.
Sampling	ORP (mV)
event(1) I Ml-S-A-D Ml-S-A-S Ml-Mid-D Ml-Mid-S Ml-N-A-S Ml-N-A-D
Baseline
22
44
29
67
45
7
Post-Oxid. 1
31
113
392
629
65
14
Post-Oxid. 2
415
484
591
617
263
121
Post-Oxid. 3
478
658
523
651
30
56
111 Baseline pre-oxidation (6/21/13), Post-Ox 1 (8/6/13), Post-Ox 2 (2/13/14), Post-Ox 3 (6/24/14)
4.5.4 Dissolved methane
Pre-oxidation elevated concentrations of dissolved
methane were measured in the ground water,
indicating methanogenic conditions along the
longitudinal axis of the plume in the shallow
(7-10 ft bgs) aquifer (Figure 18.A). Despite this
distinct zone of elevated dissolved methane,
concentrations were lower and dispersed in the
deeper (10.25-12.25 ft bgs) interval.
Multiple lines of evidence have been presented
indicating that the PCE release in the southern
CVOCs plume at site 45 originated from a leak in
the sanitary sewer line at the southeast corner
of Building 192. Given this observation, it is
reasonable to assume that raw sewage would also
have been released at the same location(s) of the
broken sewer line. The continuous discharge of
sewage into the subsurface at an elevation of 5-6
ft bgs would have introduced large quantities of
biodegradable organic matter. The organic matter
associated with raw sewage is readily degradable
and would have resulted in the depletion of
dissolved oxygen and rapid onset of anaerobic
and methanogenic conditions. Introduction of the
dissolved organic matter in the highly permeable
aquifer material would follow the southeastern
ground water flow direction previously described.
Therefore, the high levels of methane measured
in the shallow interval are generally consistent
with this conceptual model. The aquifer material
directly above the 8-12 ft bgs interval exhibited
a lower permeability and may potentially limit
vertical dispersal of the methane. Conceptually, it
is surmised that the lower permeability materials
may serve to form a "methane trap" where gas-
phase methane may accumulate.
Lower post-oxidation 3 concentrations of methane
were measured in the shallow interval (Figure
18.B). This trend in dissolved methane may be
an indication that (1) the heavy and repeated
oxidant dosages applied in the subsurface may
have impacted methanogenic microbial activity,
(2) chemical oxidation of easily biodegradable
organic matter lowered the availability of organic
substrate and thus the formation of the methane
byproduct, and/or (3) the methane may itself have
been oxidized as a result of chemical oxidative
treatment.

-------
Mt-Nb-S
M2-N-0
Methane
(Mg/D
2800
2600
2400
2200
2000
1800
1800
1400
1200
1000
800
600
400
200
0
T1-J+0
M1-Sb-D
4
M4-N-D
+

Methane
(Mg/D
2800
2600
2400
2200
2000
1800
1600
1400
1200
1000
800
BOO
400
200
0
+
0-
65
Figure 18.A. Pre-oxidation shallow (7-10') (top figure) and deep (10.25-12.25') (bottom figure)
methane ([Jg/L) at Parris Island, SC MCRD Site 45 (February, 2013).

-------
M1-Nb-S
10
M1-Na-S
M2-N-S
o
T1-N-S

rJl?-Sa-S
M2-Md-S
•20

M1-Sb-S
M2-S-S
M3-Md-S
-30
^to
-50
T1-SS
M3-S-S
1000
20
10
0	10
M1-Nb-D
M1-Na-D
20
T2-S-S
—I	1	—
30	40
M2-N-D
T1-N-D
OH
ftfl-Sa-D
M2-MkJ-D
-20
M3-N-S
Ml-Sb-D
M2-S-D
M3-Mcf-D
-30
¦40
T1-S-D
M3-S-D
T2-N-S
1000
M4-N-S
M4-Md-S
1OQo
M4-S-S
50
Methane
(Mg/L)
70
M3-N-D
+
M4-N-D
M4-Md-D

2800
2600
2400
2200
2000
1800
1600
1400
1200
1000
800
600
400
200
0
T2-N-D
Methane
(Mg/L)
2800
2600
2400
2200
2000
1800
1600
1400
1200
1000
800
600
400
200
0
-50
10	20
T2"S"D	M4-S-D
30	40	50	60
Figure 18.B. Post-oxidation 3 shallow (7-10') (top figure) and deep (10,25-12.25') (bottom
figure) methane (|jg/L) at Parris Island, SC MCRD Site 45 (June, 2014).

-------
4.5.5 pH
The pre-oxidation pH was measured in micro-well clusters Ml-S-A, Ml-mid, and Ml-
N-A and the pH ranged from 5.5 to 5.8. Post-oxidation pH was measured in micro-
wells and wells where there was no residual permanganate, the pH ranged from 4.4-
6.5 (post-oxidation 1), 5.4-7.4 (post-oxidation 2), and 5.3-7.3 (post-oxidation 3). An
elevated pH (pH 9) was measured in a limited number of wells where permanganate
persisted during the post-oxidation ground water sampling event.
4.5.6 Metals
An increase in heavy metals concentration in ground water may result from heavy
metals impurities contained in the permanganate, as well as in situ mobilization of
pre-existing redox- or pH-sensitive heavy metals by the oxidant. Chromium (Cr) and
arsenic (As) have historically been the impurities of concern. Field investigations
generally reveal that these metals attenuate through various mechanisms and
within acceptable transport distances, but monitoring and assessment is needed for
confirmation. Ground water samples were collected in all nested micro-wells and
analyzed for metals via ICP OES.
Chromium. The concentration of total chromium increased in the ground water, from
background concentrations in all wells that were below the quantitation limit (10
Hg/L), to elevated concentrations in a few wells (Figure 19). The elevated chromium
concentrations were measured in wells where residual Mn04" had persisted and
spatial correlation between elevated chromium and elevated ORP was also evident
(Figure 20). Approximately 11 months after the third oxidant injection, the post-
oxidation 4 ground water sampling event (Feb. 27-28, 2015) occurred and the
concentration of chromium had declined in M3-mid-shallow from 1500 ng/L to 95
Hg/L, and had declined in M4-mid-shallow from 950 ng/L to 17 ng/L. MW-31 SL
is located approximately 75 ft downgradient of the M4 monitoring transect and
the concentration of chromium there was < 10 ng/L. Empirical evidence provided
by the ORP ground water data indicates that MW-31 SL was representative of a
downgradient well in the ISCO area. Prior to oxidant injection (June 19, 2013), the
ORP was 63 mV in MW-31 SL. After the oxidant injections, the ORP in MW-31 SL
was 236 mV and 152 mV during the post-oxidation 3 (June 23, 2014) and 4 (Feb.
27-28, 2015) monitoring events, respectively. The increase in the ORP between pre-
and post-oxidation indicated that MW-31 SL was downgradient and hydraulically
connected with ground water in the ISCO area.
Arsenic. The background pre-oxidation concentration of arsenic in the ground water
was measured in the monitoring network at the site (Figure 21). Given the lack
of waste management activities involving heavy metals in this area, it is assumed
that this is naturally-occurring arsenic. The impact of ISCO on lowering the arsenic
concentrations in the ground water is evident by examining the post-oxidation
3 and 4 ground water arsenic isocontours (Figure 21). Arsenic species in ground
water are functionally dependent on the pH and the redox state of the aquifer and
thus significant changes were the result of the elevated ORP (Figure 20) measured
in the ground water between the pre-oxidation and post-oxidation conditions.
Consequently, the attenuation of the arsenic in the ground water occurred as a result
of ISCO activities. The geochemistry associated with arsenic attenuation is discussed
below (Section 5.4, Metals Mobilization).
67

-------
P\~?\	
M1-Na-S
M2-N-S
T1-N-S
M2-Md-S
500
M3-IM-S
- 0
T2-N-S
Ms&s
-20
M1-Sb-S
M2-S-S
M3-ftfljd-S
%
-30
-40
%
T1-S-S
TV
'Ooo
M3-S-Sop
M4-N-S
M4-Mid-S
Chromium
(Mg/D
12000
1900
1800
-	1700
1600
1500
1400
1300
1200
1100
-	1000
900
800
700
600
500
400
300
200
100
0
-50
^0
T2-S-S
10
20
30
M4-S-S
50
. , -
60
70
Figure 19. Post-oxidation 3 concentration of chromium (|Jg/L) (pink isocontours).
M1-Nb-S
_J	|	>
















ORP
10
M1-Na-S
M2-N-S T1-W-S



(mV)
% M2-Md-S
\ " v '

M3-N-S
T2-
N-S

700
600
ld9-Sa-S
400




500
400
-20
-30
M2-S-S
M1-SO-S
%
MWvtd-S

M4-N-S


300
200



400



100
0
-40
T1-S-S
400
N&S-S
M4-Md-S









-100
-50



<00



		TC-S-S .
	i		
M4-S-S




0 10
1 ! [
20 30 40
50
60 70


68
Figure 20. Post-oxidation 3 oxidation reduction potential (ORP) (mV) (yellow isocontours).

-------

10
Ml-Na-S
t
M2-W-S
i
^fe-Md-S
T1-M-S
¦
M3-N-S
Arsenic
(Mg/L)
T2-M-S
^9-Sa-S %
2S
¦ e
	fl
-20 *
-X
M1-Sb§
M£&s
M3-Mid-S

M4-N-S
T1-S-S
-40
-50
£
M3-S-S
T2-S-S
M4-MlckS
M4-S-S
10
20
30
40
50
60
70


0

T1'S-S
+
Arsenic
(Mg/L) io
MgrS-S
*l5>
25
M4-MO-S
*0
T^S-S
100
I.
85
80 |
75
ld?-$a-S
65
55
50 -20
45
40 |
35
» 401
20
15
10
5
0
8
T1-N-S
Tl-S-gSf
+

T2-SS
30	40
Arsenic
(Mg/D
M4-S-S
50
T2-N-S
100
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
0
Figure 21. (Top Figure) Pre-oxidation (February, 2013) shallow (7-10') arsenic concentration
([jg/L) at Parris Island, SC MCRD Site 45. Similar sample locations were also collected in
the deep aquifer as in the shallow aquifer. All ground water samples collected in the deep
(10.25-12.25') zone of the aquifer were reported as non-detect for arsenic. (Bottom Left
Figure) Post-oxidation 3 (June, 2014) shallow (7-10') arsenic concentration (|jg/L) at Parris
Island, SC MCRD Site 45. (Bottom Right Figure) Post-oxidation 4 (February/March, 2015)
shallow (7-10') arsenic concentration ((Jg/L) at Parris Island, SC MCRD Site 45.

-------
4.6 Oxidant delivery and impact on total CVOCs in aquifer solids
Oxidant residuals from ISCO were visually observed
in most of the post-oxidation soil cores collected
at site 45. Manganese dioxide (MnOz(s)) is
characteristically dark in color, and permanganate
is characteristically purple in color. Due to the
clear acetate sleeves used to collect soil cores,
the purple oxidant and the dark staining of the
MnOz(s) in the aquifer material could be observed,
indicating the presence of the oxidant and oxidant
residuals. Although these visual indicators were
often observed in soil cores removed from the
subsurface, comprehensive mapping of the oxidant
and/or oxidant residuals in the soil cores was not
conducted.
The macro-, meso-, and micro-scale
heterogeneities in lithology and depositional
mixtures of sand, silt, clay, and total organic
carbon impact the hydraulic conductivity, the
transport of ground water, the predominant
advective transport pathways involving aqueous
phase CVOCs, and the transport and distribution
of DNAPL in the subsurface. Similarly, variability
in the TOC concentration in aquifer material will
significantly alter the extent of CVOCs sorbed
to the aquifer solids. These heterogeneities in
aquifer composition and hydraulic properties
ultimately impact the distribution of the CVOCs.
Contrasting pre- and post-oxidation soil CVOC
concentrations for the purpose of performance
evaluation must recognize the inherent variability
in subsurface systems and in the distribution
of the CVOCs. Consequently, simple empirical
differences between pre- and post-oxidation
CVOC concentrations are used here as one line of
evidence to assess treatment performance.
Post-oxidation soil cores were collected in high
CVOC concentration areas and were consistent
with locations where the permanganate oxidant
was injected. Soil cores were not collected in
downgradient areas of the plume, where CVOC
concentrations were limited and where oxidant
injection did not occur. Sub-cores were collected
from the 17 core locations and analyzed using
similar procedures as used during the pre-
oxidation soil sampling program. Based on the
(X, Y) coordinate system used in the study,
distance calculations were performed to quantify
the distance between the pre- and post-oxidation
points using the distance formula (Eqn. 3).
Distance (d) = [(Xj-XJ2 + (Y^YJ2)]172	Eqn. 3
Where,
X1 and Y1 = the X, Y coordinates for the first point
X2, Y2 = the X, Y coordinates for the second point
d = distance between the two points
The distance calculations were made to contrast
the vertical distribution of the concentration of
CVOC in soil ([CVOC]SOIL) in post-oxidation sampling
locations with 2-5 nearby pre-oxidation sampling
points (Appendix D, Figures D.6-D.22). These
sampling points ranged from 2.2 to 25.6 ft away
from the post-oxidation sampling location. As per
Eqn. 3, the average distance between pre- and
post-oxidation soil core locations was 11 ft (n=47).
The figures represent contrasting [CVOC]SOIL profiles
with increasing distance from the source zone.
Visual examination of Figures D.6-D.22 indicate
that the post-oxidation [CVOC]SOIL are consistently
lower than the pre-oxidation [CVOCs]SOIL profiles.
There is one isolated depth interval where
elevated post-oxidation [CVOC]SOIL were measured.
Specifically, at the (-5, -4) sampling location (-14
ft bgs interval), the highest CVOC concentration
was measured (576 mg/kg) (Appendix D, Figure
D.6) indicating the presence of DNAPL. This result
is consistent with a complex DNAPL distribution
conceptual model, often observed at DNAPL sites,
where nearby soil samples are contaminated
with CVOC, but do not contain DNAPL (Appendix
D, Figures D.6-D.23). The CVOC concentrations
(0.3-10.4 mg/kg), measured in the same core at
higher intervals (6.5-12 ft bgs), contained much
lower CVOC concentrations. The (-5, -4) soil sample
location was from the most contaminated portion
of the source zone. Two other nearby sampling
locations, (0, -5) (35.3-50.1 mg/kg) and (11, -3)
(10.6-18.9 mg/kg), also exhibited persistent
CVOCs concentrations at the 12-14 ft bgs interval
indicating CVOC persistence at this location.
Correspondingly, additional oxidant injection and
long-term persistence of permanganate in this

-------
general vicinity will be required for the
continued oxidation and depletion of CVOCs
in the source area.
The bar chart inset in Figures D.6-D.22 illustrates
the average pre- and post-oxidation [CVOCs]SOIL
for each core location, where the average value
was based on the post-oxidation soil sample depth
interval (i.e., ~ 8- 14 ft bgs). The post-oxidation
concentrations of total CVOCs are generally lower
than the pre-oxidation concentrations. However,
sampling locations close to and upgradient of the
ordinate (0, 0), at MW 25-SL, indicated sporadic
[CVOC]SOIL ranging from very low (1.1 mg/kg total
CVOCs) to elevated (101.3 mg/kg total CVOCs),
suggesting a CVOC source area. Specifically, in the
X-distance range of -5 to 11 ft from MW 25-SL (i.e.,
5 ft upgradient to 11 ft downgradient), the average
[CVOC]SOIL was 20.6 mg/kg and the 95% confidence
interval was high (95% CI = 0 - 60.2 mg/kg),
indicating significant variability. High variability
in total CVOC concentrations is consistent with
DNAPL source areas and is in agreement with
the conceptual site model for site 45. Much
lower [CVOC]SOIL variability was measured in the
downgradient direction (i.e., X-distance > 11 feet)
from this zone, where the average [CVOC]SOIL is
1.29 mg/kg and the 95% CI indicated much lower
variability in total CVOC concentration (95% CI =
0.90-1.60 mg/kg).
Overall, contrasting post-oxidation and pre-
oxidation CVOC concentrations (Figures D.6-D.22)
indicated a declining trend in post-oxidation CVOC
concentrations. The declining trend is attributed to
the successful delivery of permanganate oxidant
into these areas and the subsequent oxidation of
the CVOCs. The [CVOC]SOIL data provide additional
evidence of the destruction of CVOCs in the
soil media associated with the source area and
the longitudinal axis of the plume. Although it
is probable that chemical oxidation of CVOCs
occurred in the aqueous, sorbed, and DNAPL
phases of contaminated media, differentiation
between the CVOC mass oxidized between these
phases was not possible.
4.7 Molecular biology tools
Under aerobic conditions, oxygen is the most energetically favorable terminal electron acceptor (TEA) in
the biodegradation of organic compounds. Given the limited solubility of oxygen in water, dissolved oxygen
is rapidly depleted, resulting in anaerobic conditions (Huling et al., 2002). Under anaerobic conditions,
however, the biochemical oxidation of organic compounds also occurs (Lovley and Phillips, 1986;
Hutchins et al., 1998). The sequential order of TEA utilization under anaerobic conditions is nitrate (N03"),
manganese (Mn(IV)), ferric iron (Fe(lll)), sulfate (S042-), and carbon dioxide (C02). Although biodegradation
of BTEX has been correlated with this sequential utilization of TEA under field conditions (Borden et al.,
1995), the distribution of the terminal electron accepting processes is highly dynamic in both time and
space (Vroblesky and Chapelle, 1994). In anaerobic conditions, certain bacteria utilize the chlorinated
ethenes: PCE, TCE, DCE, and VC, as electron acceptors in biotic reductive dechlorination processes. The net
result is the sequential dechlorination of PCE and TCE, through daughter products DCE and VC, to non-
toxic ethene, which volatilizes or can be further metabolized biochemically.
The microbial species Dehalococcoides (sp.) is capable of complete sequential dechlorination of PCE to
ethylene, and Dehalobacter (sp.) can also dechlorinate chloroethenes. Dechlorination of CVOCs can be
carried out by mixed cultures of dechlorinators under anaerobic conditions (Bradley, 2003); however,
Dehalococcoides (Dhc) (spp.) is the only known bacteria that completely dechlorinates PCE and TCE to
non-toxic ethene. Site-specific analysis of Dhc in ground water provides information regarding whether
sufficient Dhc are present on site and potentially capable of carrying out this desirable biodegradation
pathway. Further analysis of the functional Dhc genes responsible for encoding enzymes that transform
the CVOCs also provides information regarding the microbial characteristics of the subsurface microbial
71

-------
system. Microbial Insights Inc., a commercial laboratory, was contracted to analyze ground water sample
filters for Dhc, and the enzymes responsible for the CVOC biotransformation pathway. Specifically,
qTCE and qVC are the functional genes in Dhc that encode reductive dehalogenases for TCE and VC.
The presence of TCE reductase indicates the ability to reduce TCE to DCE and VC. The presence of VC
reductase indicates the potential for reductive dechlorination of VC to ethene, and the absence of
VC reductase suggests that VC may accumulate (Microbial Insights Inc., 2014). Dehalococcoides sp.
strain BAV1 couples growth with the reductive dechlorination of VCto ethene (Krajmalnik-Brown et
al., 2004). Identification of these genes, qTCE and qVC, provides information on potential sustainability
of the reductive dechlorination mechanism at the site. The comprehensive analysis of these microbial
characteristics (Microbial Insights, CENSUS"), involving the enumeration of specific organisms or genes
encoding specific biological functions, provides a direct avenue to investigate the potentials and limitations
to implementing corrective action plan decisions and to target a variety of organisms involved in the
reductive dechlorination pathway. Further, total Eubacteria is a measurement that targets universal
regions of bacterial 16S rRNA genes to provide a broad index of total bacterial biomass.
Pre-oxidation (i.e., baseline) and post-oxidation ground water samples were collected from six micro-wells
in the source area and analyzed for Dehalococcoides, the enzymes TCE-A reductase, BAV1 VC reductase,
VC reductase, and total Eubacteria (cells/mL) (Table 7). The pre-oxidation microbial enzyme markers
Dehalococcoides TCE-A reductase, BAV1 VC reductase, and VC reductase were notably absent as they were
mostly reported as U (not detected). The same condition existed in the post-oxidation samples from five
of the wells. The pre- and post-oxidation populations of total Eubacteria were measurable but a definitive
trend between pre- and post-oxidation conditions was not established. For example, in four wells the total
Eubacteria populations increased but in two wells it decreased. Lu et al. (2006) suggested that a density of
107 Dhc cells/L is necessary for reductive dehalogenation natural attenuation rate that provides significant
degradation in a reasonable time frame. The ground water samples collected from the micro-wells located
in the source zone do not meet this general requirement and therefore no firm conclusions are possible
regarding the impact of ISCO on biotic natural attenuation processes in the source zone.
Table 7. Pre- and post-oxidation molecular biology tool analysis (Microbial Insights)(1) for micro-wells
Ml-Mid-shallow and deep, Ml-NA-shallow and deep, and Ml-SA-shallow and deep. The baseline
(pre-oxidation) sampling event occurred on 6/21/2013; oxidant injections 1 and 2 occurred on
6/23-29/2013 and 9/22-27/2013, and post-oxidation 2 (post-ox 2) ground water sampling occurred
2/13/2014.
Micro-well
ID
Sample
Date
Sampling
Event
Dehalococcoides
(cells/mL)
tceA
Reductase
(cells/mL)
BAV1 VC
Reductase
(cells/mL)
VC
Reductase
(cells/mL)
Total
Eubacteria
(cells/mL)
Ml-Mid-S

Baseline
0.5 U
0.5 U
0.5 U
0.5 U
3,330


Post-ox 2
0.5 U
0.5 U
0.5 U
0.5 U
1,930
Ml-Mid-D

Baseline
0.3 U
0.3 U
0.3 U
0.3 U
479


Post-ox 2
1.7
1.7
1.7
1.7
3,390
Ml-NA-S

Baseline
0.5 U
0.5 U
0.5 U
0.5 U
5,770


Post-ox 2
0.7 U
0.7 U
0.7 U
0.7 U
9,290
Ml-NA-D

Baseline
0.5 U
0.5 U
0.5 U
0.5 U
437


Post-ox 2
0.5 U
0.5 U
0.5 U
0.5 U
1,200
Ml-SA S

Baseline
0.5 U
0.5 U
0.5 U
0.5 U
693


Post-ox 2
0.8 U
0.8 U
0.8 U
0.8 U
21,200
Ml-SA D

Baseline
0.3 U
0.3 U
0.3 U
0.3 U
3,390


Post-ox 2
0.6 J
0.7 U
0.7 U
0.7 U
2,590
'U - analyte not detected above reporting limit; J - concentration estimated.

-------
The MBT results were derived from samples collected in micro-wells in the heart of the source area.
This suggests that limited quantities of the microbial indicators may have been measured as a result of
inhibitory levels of CVOCs known to occur in these micro-wells. Specifically, the pre-oxidation average
CVOC was 31.4 and 32.2 mg/L in the three shallow and deep wells, respectively. Subsequently, three
aggressive injections of NaMn04 in the source area near these wells caused additional microbial toxicity.
Therefore, it is proposed that the low concentrations of pre- and post-oxidation microbial indicators:
Dehalococcoides, TCE-A reductase, BAV1 VC reductase, VC reductase, and total Eubacteria are consistent
with highly toxic conditions for microbial populations both from elevated concentrations of CVOCs, or
NaMn04, or both. Future efforts to assess the impact of ISCO on biotic natural attenuation should include
the collection of ground water samples in a downgradient direction from the source area. In a previous
study at Site 45, several groundwater samples collected near, within 50-200 ft, the source area contained
Dehalococcoides sp. at >107 cells/L. This suggests that sufficient quantities of Dehalococcoides sp. were
present and robust biodegradation of the CVOCs to ethylene is possible in downgradient portions of the
plume (Vroblesky et al., 2009).
4.8 Compound-specific isotope analysis
CSIA can potentially distinguish destructive
chemical oxidation transformation reactions
from non-destructive mechanisms (e.g., dilution,
displacement, dispersion, desorption, advection,
etc.). Stable isotope analysis of carbon involves
measurement, quantifying, and contrasting of
the relative abundance of the naturally occurring
stable isotopes of carbon, 13C and 12C, in organic
chemicals (US EPA, 2008). Since the chemical
bonds associated with 13C are stronger than
12C, 13C destructive reactions are slower and
permit the abundance of 13C to increase over
time in the parent compound, relative to 12C. In
contrast, dilution, displacement, and desorption
mechanisms that are non-destructive have
little impact on the distribution of 13C and 12C
in an aqueous solution containing the organic
compounds. Consequently, destructive reactions
such as chemical oxidation and biodegradation
will preferentially destroy 12C, resulting in a higher
quantity of 13C, relative to 12C. Typically, the
abundance of 12C is much greater than 13C, which
represents approximately 1% of the total naturally
occurring organic carbon.
In order to ensure inter-laboratory comparability
and accuracy, the ratios of 13C and 12C are
expressed relative to an international standard.
Therefore, for a "sample" compound, the data are
(CSIA)
reported as Rsample= (13Csample/12Csample) relative to
the ratio of the international standard (Rstd) where
Rstd = 13Cstd /12Cstd. Measured values are reported as
613C (Eqn. 4) (US EPA, 2008).
513C (%o) = [13Csample/12CsampJ-[13Cstd/12Cstd] x 1000
[13Cstd/12Cstd]	(Eqn. 4)
Since the resulting 6 values are very small (e.g., for
613C, typically < 0.05), they are generally multiplied
by 1000 for convenience and reported as parts per
thousand or "per mill", indicated by the symbol %o
(US EPA, 2008). In the common delta (6) notation,
the deviation of the stable isotope value of the
sample from the standard will be either negative or
positive. A negative value means that the sample
is depleted in its 13C-content, relative to the 13C/12C
content of the standard whereas a positive sign
implies an enriched 13C-content (US EPA, 2008).

-------
To be significant, the extent of fractionation must be greater than the total analytical
uncertainty. In addition, the observed difference in the values of 613C must exceed the
spatial and temporal variability introduced by different sources of contamination at
the site, by the mixing of ground water flow lines, and by what are typically the minor
effects of processes such as sorption or volatilization. The total analytical uncertainty
for 613C analyses was estimated to be approximately ± 0.5%o (US EPA, 2008). As a
result, the observed fractionation must be at a minimum > l%o. To ensure reliable
interpretation, it was recommended that fractionation on the order of 2%o be used as
a criterion for positive identification of degradation in order to minimize the possibility
of an erroneous interpretation (US EPA, 2008).
Pre-oxidation (i.e., baseline) and post-oxidation ground water samples were collected
from six micro-wells in the source area and analyzed for CVOCs; and CSIA was
performed for the corresponding CVOCs from the same micro-wells (Table 8). CSIA
values were not reported by the contract laboratory for some ground water samples
and therefore comparisons between pre- and post-oxidation conditions were not
possible in these cases (these cells are shaded in gray in Table 8). As indicated by
others (US EPA, 2008), spatial and temporal variability in CVOCs and CSIA can result
from analytical variability, the heterogeneous distribution of CVOCs at the site,
the mixing of ground water flow lines, and by the variability in transport and fate
processes such as sorption or volatilization. Further, non-equilibrium conditions
resulting from oxidant injection and incomplete oxidant reaction can also introduce
variability in CVOCs and CSIA results. Consequently, as recommended by US EPA
(2008), fractionation values of < 2 %o between pre- and post-oxidation were omitted
from the pool of results (these cells are shaded in blue in Table 8). An increase in the
abundance of 13C, relative to 12C, is an indication of the chemical oxidation of CVOCs.
This condition occurred in six of the samples analyzed, indicating evidence of CVOC
destruction in the source area (Table 8, shaded in yellow). Conversely, there were four
ground water samples where both the CVOC concentration and the 13C/12C increased
(Table 8, shaded in red). One possible explanation for this occurrence could be that
pre-oxidation conditions involved aggressive biodegradation that was responsible for
lower CVOCs and CSIA. After ISCO activities, a temporary decline in biodegradation
due to ISCO, in conjunction with CVOCs influx due to rebound and upgradient CVOC-
contaminated ground water, could have raised both the CVOC concentration and the
CSIA values. The post-oxidation CVOC and CSIA data represent a highly transient non-
equilibrium ground water chemistry condition resulting from disturbances in ground
water flow paths from fluids injection, chemical oxidation destruction of CVOCs,
inhibitory response of biotic processes to the highly oxidative conditions, spatial
variability in CVOC distribution, major changes in redox and concentration gradients,
etc. Overall, the CSIA results are variable and represent ground water in a highly mixed
and disturbed condition. Consequently, due to these complexities, it was difficult to
establish firm and definitive trends in the analytical results. Nevertheless, it can be
concluded that in several wells located close to the oxidant injections (i.e., Ml-Mid-S
and D, Ml-NA-S), where significant reductions in CVOCs were measured, the CSIA data
provides proof of concept that PCE, TCE, DCE, and VC were destroyed.
74

-------
Table 8. Pre- and post-oxidation CSIA (Microseeps, Inc.) and CVOCs summary table for micro-
wells Ml-Mid-shallow and deep, Ml-NA-shallow and deep, and Ml-SA-shallow and deep.
The baseline (pre-oxidation) sampling event occurred on 6/21/13; oxidant injections 1 and 2
occurred on 6/23-29/2013 and 9/22-27/2013, and post-oxidation 2 (post-ox 2) ground water
sampling occurred 2/13/2014(1'2).
Micro-well
ID
Sample
Date
Sampling
Event
PCE
613C/12C
(%o)
TCE
613C/12C
(%o)
Cis 1,2-DCE
613C/12C
(%o)
VC
613C/12C
(%o)
Ml-Mid-S
6/21
13
Baseline
-28.52
-29.81
-29.41
-23.26
2/13
14
Post-ox 2
-24.03
-22.39
-18.68
-
CVOCs
6/21
13
Baseline
67100
10800
7310
1180
(Hg/L)
2/13
14
Post-ox 2
6400
1400
2300
61
Ml-Mid-D
6/21
13
Baseline
-27.66
-29.9
-29.97
-
2/13
14
Post-ox 2
-24.87
-15.1
-14.09
-
CVOCs
6/21
13
Baseline
76300
3990
536
1.7
(Hg/L)
2/13
14
Post-ox 2
13000
3900
39000 Q
40 U
Ml-NA-S
6/21
13
Baseline
-
-
-25.53
-25.85
2/13
14
Post-ox 2
-24.23
-24.54
-23.86
-16.80
CVOCs
6/21
13
Baseline
1670
1110
154
22.3
(Hg/L)
2/13
14
Post-ox 2
80
44
58
9.4
Ml-NA-D
6/21
13
Baseline
-14.81
-14.75
-25.28
-
2/13
14
Post-ox 2
-23.68
-23.62
-25.08
-17
CVOCs
6/21
13
Baseline
2050
627
3630
5
(Hg/L)
2/13
14
Post-ox 2
92
48
6100
U
Ml-SA-S
6/21
13
Baseline
-18.91
-29.55
-43.87
-
2/13
14
Post-ox 2
-24.64
-25.04
-28.39
-
CVOCs
6/21
13
Baseline
60
47
34
0
(Hg/L)
2/13
14
Post-ox 2
99
58
1500
u
Ml-SA-D
6/21
13
Baseline
-23.58
-24.87
-32.27
-
2/13
14
Post-ox 2
-24.85
-26.42
-31.15
-
CVOCs
6/21
13
Baseline
2640
2780
2160
2
(Hg/L)
2/13
14
Post-ox 2
6000
4200
3600
U
1	Shading description for cells:
light gray - CSIA data not reported for either baseline or post-oxidation sampling event 2, or both, preventing a comparison of data;
light blue - data eliminated due to low fractionation criteria (< 2%o);
yellow - aggressive oxidation, indicated by CSIA where 13C becomes more abundant relative to 12C
(i.e., 13C/12C is less negative as per Eqn 4 in section 4.8);
red - increase in both CVOC and CSIA, suggesting possible influx of CVOCs.
2	U - contaminant not detected above reporting limit; J - concentration estimated; Q- surrogate recovery failure.
75

-------
4.9 Preliminary Cost Analysis
The cost of remediation is of great interest to the site stakeholders {e.g., federal,
state, and potentially responsible party (PRP) project managers). Lowering the cost of
remediation is an important consideration in the overall remedial goal; however, the
successful outcome is also contingent upon achieving the remedial goals. For example,
greater costs are justified when a more expensive remedial option is effective and
achieves the remedial goal, relative to a less-costly remedial alternative that falls short
of achieving the remedial goal. Further, greater remedial costs may also be considered
relative to a desired time frame for remediation.
The two sections below present a preliminary cost analysis involving actual costs of
the pilot-scale ISCO demonstration. The pilot-scale demonstration of ISCO is expected
to be the first stage of a final remedy at site 45. Therefore, the costs presented
are intended to be used to project additional costs associated with a final ISCO
remedy. As will be discussed below in section 5.8 Recommendations, assuming the
recommended limited-scale ISCO and monitoring activities are adopted, projected
costs for ISCO deployment as a final remedy will be limited. The costs presented below
do not include travel costs. The site 45 pilot-scale ISCO demonstration originated as
a research study and travel costs for EPA research staff traveling from Oklahoma to
South Carolina would be unlikely to be incurred assuming remediation deployment
involved local or regional expertise. Labor and analytical costs represented the primary
costs for the site characterization. The primary remediation costs were divided among
injection system capital costs, oxidant, labor, and analytical costs.
4.9.1 Site characterization
The collection and analysis of aquifer cores and ground water samples were carried
out to develop and refine the site conceptualization model.
Cores. The initial site characterization for the pilot study consisted of the collection of
soil cores along six transects of the ground water plume, involving three locations at
each transect (Figure 4). Pre-oxidation soil cores were collected to better understand
contaminant distribution in aquifer materials at the site, and post-oxidation soil
cores were collected and analyzed to assess treatment performance, and to quantify
the remaining contaminant residuals. The soil cores were collected using Geoprobe
equipment, owned by RSKERC, so no capital costs were incurred. Each soil core
characterization event incurred costs for items such as core sleeves and caps, dry ice to
preserve the cores for shipping, and other miscellaneous supplies. The bulk of the cost
incurred for the site soil characterization involved labor of the personnel during the
field work, and in analytical costs for analyzing the cores.
Micro-wells. Micro-wells were installed to use as sentry wells, to better understand
contaminant distribution in ground water, and to help assess the ISCO process and
treatment performance. The costs incurred for these wells included micro-well
screens, tubing, traffic-rated vault boxes, and miscellaneous supplies, such as grout.
The bulk of the cost involved labor by the RSKERC field team and research staff.

-------
Analytical. Analytical costs for the site characterization activities at site 45 were incurred for analysis
of chlorinated compounds, metals, and field parameters in ground water samples, and for chlorinated
compounds in soil samples. The laboratory ground water and soil analyses for the site characterization were
predominantly conducted using contract staff at RSKERC. These costs depended on the level of analyst and
the time spent for the analyses, and cannot be easily quantified. A number of field parameters for ground
water were measured in the field, incurring costs for sample containers, reagents, disposable PPE, and the
labor required for sample collection and field analyses. During the course of this project, the RSKERC was
unable to retain the on-site analytical contract. Consequently, off-site analyses of ground water and soil
samples at Shealy Environmental Inc. (Columbia, SC) were conducted.
Personnel. Labor costs were the second of the two major categories of expenses for the site characterization
activities at site 45. The duration (in days) of the work conducted for site characterization and the level of
personnel involved have been estimated (Table 9). The monetary cost for the labor performed for the site
characterization activities is difficult to calculate, due to different rates for different personnel.
Table 9. The duration of work days, the number of personnel utilized, and the personnel work days
associated with the site characterization and remediation activities for the pilot-scale ISCO
demonstration at Site 45
. . Duration of Number of Personnel days
work (days) Personnel utilized (days)
Site Characterization
Initial site reconnaissance
1
Research staff 1
Research staff 1
Soil core collection
4
Research staff 2
Field crew 4
Research staff 8
Field crew 16
Micro-well installation
5
Research staff 2
Field crew 2
Research staff 10
Field crew 10
Total days for site characterization
10

Research staff 19
Field crew 26
Remediation
Baseline sampling
1.5
Research staff 2
Research staff 3
First injection
9
Research staff 2
Field crew 3
Research staff 18
Field crew 27
First post oxidation sampling
1.5
Research staff 1
Field crew 1
Research staff 1.5
Field crew 1.5
Second injection
5
Research staff 1
Field crew 2.5
Research staff 5
Field crew 12.5
Second post oxidation sampling
1.5
Research staff 1
Field crew 1
Research staff 1.5
Field crew 1.5
Third injection
7
Research staff 1
Field crew 2
Research staff 7
Field crew 14
Third post oxidation sampling
1.5
Research staff 1
Field Crew 1
Research staff 1.5
Field Crew 1.5
Post-oxidation soil core sampling
2
Research staff 1
Field crew 2
Research staff 2
Field crew 4
Total for remediation
29

Research staff 39.5
Field crew 62
77

-------
4.9.2 Remediation
The costs for the remediation portion of the pilot-
scale ISCO deployment at site 45 included the
capital costs of the injection equipment, injection
process equipment (peristaltic pumps), oxidant,
analytical, and personnel costs. The cost of the
injection unit was $15,400, the supporting injection
process equipment was approximately $4,500, and
the total capital cost was about $19,900. The total
cost of the oxidant injected was approximately
$26,000.
Injection system. An injection system was
constructed specifically to be used for the pilot-
scale ISCO demonstration at Site 45 (Appendix B
-Oxidant Injection System). The ISCO approach was
to carry out three oxidant injection events involving
increasing oxidant loading for each successive event,
and rigorous monitoring between oxidant injections.
Consequently, a portable injection system was
required that could be mobilized, set up on-site,
safely and effectively deployed, and demobilized.
The design included sizing the injection system
(e.g., dimensions, injection solution volume, weight
loaded/empty), three-manifold injection arms,
safety considerations, portability, corrosion resistant
components from inlet to outlet, etc. The injection
system was designed to allow additional injection
lines to be added. For example, two injection lines
were used during the first oxidant injection, and
three injection lines during the third injection. The
cost of the components for the injection system,
including one injection arm, was $15,400. Additional
injection lines could be added (i.e., > 3 arms), with
the cost being approximately $2,860 per injection
arm. The field injection process utilized additional
equipment such as a peristaltic pump and tubing
to transfer the 40% oxidant into the mixing tank
($2,946), Geoprobe drive caps (custom-made in-
house), top-down injection tools and drive rod
($1,211), and oxidant neutralizer and absorbent
socks ($351), for a total cost of $4,508.
Oxidant. The oxidant was purchased from Carus
Corp. (Peru, IL), and included 5-gallon pails of
40% liquid sodium permanganate (i.e., Carus
Corp., Remox-L, 57 lb Jerrican). The oxidant cost
approximately $2.50-$2.70/lb of 40% NaMn04
(i.e., $143-$154/5-gallon Jerrican). The cost of
the oxidant plus shipping charges resulted in an
approximate actual cost of $175 per five-gallon
pail. The number of 40% NaMn04 pails and cost for
the 1st, 2nd, and 3rd oxidant injections was 17 pails
($2,975), 43 pails ($7,525), and 89 pails ($15,575),
respectively. The total cost of oxidant, including
freight, was $26,075.
Analytical. Analytical costs for the remediation
activities of the pilot-scale study at site 45 were
incurred primarily for analyses of chlorinated
compounds and field parameters in ground water.
Additional analytical costs occurred throughout
the study, involving the analyses of soil cores used
to further delineate CVOC concentrations in the
source area. The additional site characterization
was needed to help focus oxidant injections in the
source area. The laboratory ground water analyses
for the ISCO baseline sampling were conducted
using contract staff at RSKERC. The laboratory
ground water analyses for the subsequent ISCO
performance monitoring were conducted using
Shealy Environmental (Columbia, SC). The basic
cost for CVOC analysis of ground water samples and
methanol extracts of soil was $50/sample. A number
of field parameters for ground water were measured
in the field, incurring additional costs for sample
containers, reagents, disposable PPE, and the labor
required for sample collection and field analyses.
Personnel. Labor costs contributed significantly to
the total cost of the remediation activities at site 45.
The duration of the work and the level of personnel
involved in remediation and monitoring the
remediation (i.e., additional site characterization)
are approximated and summarized (Table 9). The
cost for the labor performed for these activities
will vary based on the hourly rate of personnel,
therefore, "personnel-days" equivalence is reported
and can be used to estimate cost based on local
and/or regional hourly rates.
The duration of the first injection trip was lengthy
since the research and field crew were conducting
the initial set-up, operation, and monitoring
procedures for the injection equipment for the first
time. This was required for the safe and effective
deployment of the oxidant injection process. The
total volume of oxidant solution injected during
the first, second, and third injections was 2688,
4450, and 8814 gallons, respectively. One method
of analyzing manpower efficiency is to examine
the ratio of the volume of oxidant injected and the
combined (i.e., research and field crew) personnel-
days. Given the personnel-days (Table 9), the
combined oxidant injection efficiency was 60, 254,
and 419 gallons oxidant/personnel-day for the 1st,
2nd, and 3rdinjections, respectively.

-------
5.1 Injection equipment, design, and impact
5.1.1 Injection equipment
The cost of the oxidant injection system was
approximately $15,400, which included one
injection arm. Each injection arm, including the
45 ft injection hose and the injection tip, cost
$2,860 each, excluding the GeoProbe rod
(Appendix B. Detailed description of injection
equipment (schematics, manufacturers, part
numbers, cost)). The injection pallet proved
to be compact and portable, making it easy to
accommodate transport of the system between
the RSKERC facility and the ISCO site. The pallet has
a 4 ft x 4 ft footprint and was easily transported
on a flatbed trailer or on the bed of a 1-ton truck.
The injection arms are easily removed from the
injection pallet, allowing for easy storage and
transport when not in use. A temporary catchment
system was constructed for the injection pallet,
serving as secondary containment. This was
performed for added safety, to contain the oxidant,
if there were an unexpected catastrophic spill. The
injection pallet was moved to different locations
during injection events to accommodate large
distances between the source area injection
locations and other injection locations further
downgradient. At little additional cost, a longer
injection hose (60 ft vs 45 ft) was purchased,
allowing greater reach between the injection pallet
and the injection location. Even longer hoses could
79
have been ordered that would have been able to
access all the injection locations without moving
the injection pallet. Once mobilized and set up,
the injection pallet was moved between injection
locations using multiple methods, including a
fork lift, a Tommy-lift on the 1-ton truck, and the
GeoProbe sonic rig that the field crew mobilized
for a subsequent project on the second leg of the
field trip.
The addition of the 2nd and 3rd arms on the
injection pallet allowed injection to occur at
three locations simultaneously. This contributed
significantly to the overall efficiency in delivering
the oxidant into the subsurface. Each arm was
individually equipped with a flow dampener,
pressure relief valve, pressure control valve and
return line, injection pressure valve, flow meter,
and emergency shut-off valve. This design allowed
each of the injection arms and tips to be operated
individually in the delivery of the specific oxidant
dosage for each injection location. An injection
tip could be advanced to another injection
interval or moved from one location to another
while simultaneously injecting into the other two
injection tips, thus being efficient, effective, and
flexible.

-------
5.1.2 Incremental benefits of soil core sampling and analysis
Ground water samples represent an integrated
measure of CVOCs in the subsurface. For example,
CVOC concentrations in ground water spatially
represent the sorbed and DNAPL mass distributed
nearby in the subsurface. Specifically, CVOCs
partition into the ground water from the aquifer
solids and DNAPL. However, it is difficult to
refine in detail the distribution of CVOCs in the
subsurface using ground water samples alone.
This is especially true in the vertical scale, given
the screen lengths of monitoring wells that tend
to represent the average CVOC concentration.
Short screened intervals, such as the 2 ft stainless
screened micro-wells used in the study area at
this site, help to limit the vertical averaging of long
screened intervals.
Collection and analysis of soil cores for CVOC
can help to refine the CVOC conceptual model,
especially in the vertical scale. Examination of
historical ground water concentration data from
monitoring wells at the site suggested that CVOC
contamination extended from 3-18 ft bgs. This
observation is due to the wells' screen lengths,
which occur over this vertical interval. Constructing
ground water monitoring wells with screen lengths
of 5-10 ft, or greater, is common. Consequently, the
length of the well screen can preclude an accurate
vertical discretization of CVOCs present in the
aquifer solids or as DNAPL. At this site, the majority
of the CVOC contamination was found within a
loose sand layer. Consequently, wells partially
screened over the loose sand layer, containing
CVOC contamination, would suggest that
contamination occurred over the entire length. A
5 ft well screen is considered to be relatively short,
but these wells partially screened across a highly
contaminated interval may give a false impression
that contamination extends over a greater vertical
interval than it actually does. The benefit of the soil
core data was that it allowed greater differentiation
between contaminated and uncontaminated
intervals and therefore established a relatively
narrow contaminated interval. This information
allowed the development of an oxidant delivery
strategy that predominantly targeted the 8-12 ft
bgs contaminated interval rather than the 3-18 ft
bgs interval. Thus, the soil core data provided the
scientific basis to reduce the vertical interval over
which ISCO should be deployed from 15 ft (i.e.,
3-18 ft bgs) to 4 ft (i.e., 8-12 ft bgs). This is nearly
a 75% reduction in the vertical interval requiring
treatment. If oxidant were injected outside the
targeted interval of 8-12 ft bgs, the oxidant would
have served a limited purpose. This is especially
the case at lower depths where the NOD was high
due to the elevated organic material, the CVOC
concentrations were low or non-detect, and the
oxidant transport and persistence would have
been limited. It is evident that this modification
in the ISCO design resulted in an overall cost
savings to the pilot-scale ISCO demonstration
project. In addition to the cost savings attributed to
purchasing a smaller quantity of oxidant, significant
cost savings resulted in a reduction in the labor
costs associated with injecting a much lower
volume of oxidant. The investment in additional
site characterization involving soil core sampling
and analysis provided a high value dividend, a
refined conceptual site model. This allowed ISCO
activities to be focused over a specific targeted
interval which limited remedial costs.
80

-------
5.1.3 Oxidant delivery design and methods
The oxidant delivery design and methods used
in the pilot-scale ISCO demonstration at site 45
resulted in more effective and efficient oxidation
of CVOCs. Specifically, narrow ROIs, short vertical
screened injection intervals, low injection pressure,
outside-in oxidant injection, and total porosity
oxidant volume design were used to help assure
oxidant delivery to the targeted zones. The narrow
ROIs required additional labor due to the greater
number of injection locations required to cover
the source areas, but did not add significantly to
the overall injection time. The short 2-ft vertical
screened injection intervals required the injection
tip to be advanced in 2 ft increments at the same
injection location. Relative to longer injection tips,
this additional injection time at each injection
location was marginal. However, the short
(2-ft) injection increments limited the potential
of injecting oxidant across vertical intervals with
contrasting hydraulic conductivity. Collectively,
these oxidant delivery methods and design
were beneficial as they helped to limit the role
of preferential pathways and the unintentional
delivery of oxidant into non-target areas, and to
increase the probability and confidence that the
oxidant was delivered to the targeted zones.
the injection point, were used to calculate the
maximum injection pressure (Appendix A). Given
the shallow injection depths, and hydrostatic and
overburden pressure to counter-pressure the
oxidant injection pressure, low injection pressures
were calculated. For the injection intervals of
8-10, 10-12, and 12-14 ft bgs, maximum injection
pressures were 4.8, 5.6, and 6.4 lb/in2 (psi)
respectively. Therefore, it was evident that low
injection pressures were critical for effective
oxidant delivery at site 45 to limit the potential for
breakout and daylighting of the oxidant. Because of
the low flow rate under this low oxidant injection
pressure, an informal trial and error method was
used to test higher injection pressures. As a result,
an operational injection pressure of < 10 lb/in2
was developed as a guideline. Due to the frictional
head loss of the injected fluids in the injection
hose, connectors, and injection tip, the downhole
pressure was less than the gauge pressure on the
injection arm. The < 10 lb/in2 injection pressure
permitted a higher flow rate of oxidant, while
preventing breakout and daylighting of the oxidant.
Higher injection pressures (i.e., > 10 lb/in2) were
avoided as this could have resulted in daylighting
of the injected solution.
The role of preferential pathways during oxidant
injection and remedy failure can be correlated
with high injection pressure (Suthersan etal.,
2011). High injection pressures can also promote
fracturing, fluid movement along the fractures,
and compromise the ability to uniformly distribute
the oxidant. In these cases, delivery of oxidant into
the subsurface does not equate to appropriate
distribution. Parameter values for density of dry
soil and saturated soil, the thickness of the vadose
zone, and the height of the saturated zone above
81

-------
5.2 Critical analysis of oxidant loading
Concepts and fundamentals used to design the volume of oxidant
injected into a targeted treatment zone often involves porosity. A
detailed critical analysis of oxidant loading has been presented in a
manuscript entitled, "In Situ Chemical Oxidation: Oxidant Volume
Design Considerations" and is included as Appendix E (Huling et al.,
2016a). A summary of the salient technical issues from the manuscript,
involving oxidant loading design, will be presented here. Also, ISCO
design parameters used at site 45 will be contrasted with conventional
ISCO design parameters to clarify the unique and aggressive oxidant
loading design approach used.
5.2.1. Estimating oxidant volume for injection
The targeted treatment zone (TTZ) defined here refers to the
contaminated volume of porous media in a source area requiring
oxidative treatment. Adequate coverage of the oxidant in the TTZ
requires the delivery of a sufficient volume of oxidant, containing a
sufficient mass of oxidant, to achieve the treatment objectives. An
estimate of the volume of oxidant to inject to fill the pore volume in
the TTZ can vary depending on the assumptions used in the calculation.
Assuming the simplified radial-flow, cylindrical, porous media
conceptual model (Eqn. 1) (refer to section 3.2.11 Natural oxidant
demand and oxidant loading), two methods used to estimate the
volume of oxidant are contrasted, involving the mobile porosity and
the total porosity. The oxidant injected into the subsurface will fill the
pore spaces in the unconsolidated porous media. The total porosity (r|)
of unconsolidated porous media is defined as the volume of voids (Vv)
relative to the total volume (VT) of aquifer material (n=Vv/VT). However,
unconnected, poorly connected, and dead-end pores are responsible
for a fraction of water in the porous media but minimally contribute to
fluid displacement.
Therefore, the concept of total porosity was broadened to include
effective porosity, and refers to that fraction of the total volume
of pore space where pore fluid can be readily displaced; i.e., the
interconnected pore volume or void space in an aquifer that
contributes to fluid flow or permeability. Specifically, the fraction of
the total porosity that contributes to advective flow and transport of
ground water in aquifers is the mobile porosity (0M), and the portion of
the void space that does not contribute to the advective flow of ground
water, and behaves as immobile or slowly moving ground water, is the
immobile porosity (Qf) (Payne et al., 2008). The total porosity is the sum
of mobile and immobile porosity (r| = 0M + 0,). The selection of mobile
porosity or total porosity in designing the volume of oxidant to be
injected into the TTZ can have major implications to ISCO.
82

-------
5.2.2 Contrasting I SCO design at site 45 with other treatment systems
Krembs et al. (2010; 2011) compiled methods and results of field applications of ISCO case
studies, specifically involving permanganate ISCO sites. Several oxidant injection design
parameters were compiled including the design and observed ROI (ft), oxidant dosage (g/kg),
number of pore volumes delivered, number of delivery events, and duration of delivery events
(days) (Table 10). The ISCO design parameters from the MCRD site 45 ISCO project were also
compiled to contrast with the ISCO database used by Krembs et al. (2010; 2011) (Table 10).
A critical analysis of the median design parameters reported by Krembs et al. (2010; 2011) is
preceded by the caveat that the number of case studies for the median value of each design
parameter varied, and therefore only general observations are possible.
Table 10. A compilation of the median value for ISCO design parameters from field
application case studies (Krembs et al., 2011) contrasted with design details used in
the pilot-scale ISCO demonstration at Site 45.
Design Parameter
Median Value Reported
Krembs etal. (2011)(1)
Site 45 ISCO project
Parris Island, SC
Design ROI (ft)
14 (n=29)
2-4
Observed ROI (ft)
25 (n=ll)
= 2-4
Oxidant dose (g/kg)
-si-
rs!
II
c
-sJ;
d
3.8-5.2'21
Number of pore volumes delivered
0.16(n=32)
1
Number of delivery events
2 (n=65)
3
Duration of delivery events (days)
4 (n=45)
5.5 (n=3)(3)
Vertical injection interval (ft)
NA(4»
2
111 The median design value and number of case study sites is reported.
121 The range in oxidant dose was estimated, assuming a bulk density of 1.2-1.6 g/cm3.
131 Average duration of delivery events = site mobilization, de-mobilization, and oxidant delivery.
141 The vertical injection interval was not reported in Krembs et al. (2011).
Radius of Influence (ROI), Oxidant Dosage, and Pore Volume. The observed ROI (25 ft) is
considerably greater than the design ROI (14 ft) (Table 10) suggesting that oxidant distribution
was more extensive than designed. A firm explanation cannot be provided regarding this
anomalous difference, but some speculation is warranted. The difference could be attributed to
a number of factors, including an increase in the median ROI value simply based on the weighted
values of the reduced number of case studies reported. Assuming preferential pathways, a
disproportionate volume of oxidant could have been delivered into high permeability layers
present within the screened interval. Non-uniform oxidant transport, perhaps unidirectional, and
breakthrough in monitoring wells under these conditions may account for the misinterpretation
and the large ROI discrepancy. Regardless of the design method or value of porosity used, the
volume of oxidant required to achieve a 25 ft ROI, relative to a 14 ft ROI, would involve greater
than 3x more oxidant volume. Heterogeneities in aquifer hydraulic properties invite vulnerability
to disproportionate transport of the oxidant in preferential pathways. Smaller ROIs have several
advantages including lower probability that preferential pathways will play a role in oxidant
transport, greater potential for hydraulic control, greater accuracy in the spatial emplacement
of the oxidant, and greater confidence that the oxidant can be delivered to the ROI. However, a
reduction in the ROI translates into smaller injection well spacing, leading to the installation of
additional injection wells or more direct-push injection locations.
83

-------
Natural oxidant demand impact on oxidant volume
design. The design of ISCO oxidant loading, defined
here as mass of oxidant per mass of soil (i.e., units
of g oxidant/kg aquifer material), is sometimes
based on natural oxidant demand (NOD) values
measured in bench-scale studies. The NOD by
soil and aquifer material for permanganate often
involves a 48-hour test procedure (ASTM D 7262-
07, 2007). However, the results from these tests
must be qualified, given that the reaction between
Mn04" and the reactants responsible for the NOD
varies with time, the reactants are often comprised
of multiple reactive components that exhibit
varying reactivity (i.e., fast and slow reaction), the
measured values of NOD may increase substantially
with longer testing periods, and the reaction of
Mn04" is concentration dependent (Mumford et
al., 2005; H0nning et al., 2007; Urynowicz, 2008
Urynowicz et al., 2008; Xu and Thomson, 2009;
Cha et al., 2012). More quantitative methods have
been developed that utilize NOD results and can be
used to project oxidant distribution and/or dosages
(Heiderscheidt et al., 2008; Borden et al., 2010; Cha
and Borden, 2012). Cha et al. (2012) reported that
most of the NOD in soil and aquifer samples was
slow reacting, and that results from the 48-hr NOD
measurements are poor predictors of total NOD and
cannot be used accurately to estimate long-term
Mn04" consumption. Once the fast NOD fraction is
rapidly consumed, the remaining Mn04" may persist
for weeks to months, diffusing into low permeability
zones where contaminants may reside. Use of
these NOD results in field-scale applications must
recognize that these laboratory measured oxidant
demand values, whether based on short-term or
long-term testing procedures, are derived from the
total aquifer solids, including both the mobile and
immobile porosity fractions of the media.
In one study, 50 samples of aquifer solids were
analyzed from 12 different facilities in the US, where
ISCO with Mn04" was being considered as a possible
remedial alternative (Cha et al., 2012). The total
oxidant demand, measured over a long period (up
to 41 days) in 80% of the samples exhibited a broad
range in total NOD, between 0.24 and 18.8 g Mn04"
/kg soil; and the median value was 3.33 g/kg, and
the overall range in values was 0.2 to 150 g/kg. This
range of values was similar to previously reported
ranges (Mumford et al., 2005; Huling and Pivetz,
2006; H0nning et al., 2007; Urynowicz, 2008; Xu
and Thomson, 2009). The Mn04" dosage generally
applies to the mass of Mn04" delivered per mass
of aquifer material within the ROI, in the TTZ. The
median value of Mn04" dosage reported by Krembs
et al. (2010; 2011) is low (0.4 g/kg) (n = 24) (Table
10), relative to these values. Assuming that long-
term persistence of Mn04" was needed to address
contaminant mass transport and mass transfer
limiting processes, contrasting the Mn04" dosage
with total NOD values reported in the literature
suggests that median range values used at many
ISCO sites have been under-designed, either in
terms of oxidant volume or concentration.
The pore volume (PV) represents the volume
of voids within the TTZ, spatially defined by the
ROI and vertical interval. The median number of
permanganate pore volumes delivered (PV = 0.16;
n=32 sites) (Table 10), as reported by Krembs et
al., (2010; 2012), was less than the pore volume
delivered (PV = 1.0) at site 45 (Table 10). The
low pore volume of oxidant delivered appears
inconsistent with both the ROI and the oxidant
loading (Table 10). The volume of oxidant required
to achieve a 25 ft ROI, relative to a 14 ft ROI, would
involve greater than 3x more oxidant volume.
Further, the post-injection oxidant dispersal would
require significant persistence and transport to
achieve coverage in the remaining 0.84 pore volume
in a TTZ where oxidant was not initially delivered.
Finally, the median oxidant dosage was low (0.4 g/
kg) and it would be unlikely to persist long enough
to broadly disperse within the TTZ while satisfying
both the fast and slow acting NOD.
Overall, contrasting the ISCO design parameters
reported by Krembs et al. (2010; 2011) and the
pilot-scale ISCO demonstration at site 45 reveals
significant differences. The median values for
oxidant dosage (0.4 g/kg) and pore volume delivery
(PV = 0.16) associated with the ISCO field site data
base (Krembs et al. 2010; 2011) suggest a less
aggressive, low oxidant volume ISCO design that
appears to be consistent with a mobile porosity
ISCO design. Such designs could result in incomplete
oxidant distribution within the TTZ, insufficient
oxidant dosage within the TTZ, and short duration
oxidant persistence. Conversely, the more aggressive
ISCO design at site 45 (Table 10), based on the total
porosity oxidant volume design, involved a greater
oxidant dosage (3.8-5.2 g/kg) and pore volume
delivery (PV = 1), permitting long duration oxidant
persistence (> 1 year) in some areas.

-------
5.3 Assessment of achieving objectives
An accurate conceptual site model (CSM) was established by first examining existing data and information.
This was followed by an aggressive pre-oxidation aquifer core and ground water sampling and analysis
effort, designed to augment existing information and to fill data gaps. Multiple iterations in gathering
additional site characterization data and information continued throughout the ISCO deployment, in an
effort to develop and refine the CSM. This information was used to identify the source areas, which were
subsequently targeted through the focused delivery of oxidant. A low cost, portable injection system was
designed, constructed, and successfully deployed in the delivery of oxidant during three injection events.
Post-oxidation ground water samples were collected from the nested, on-site micro-wells and used to assess
the fate and transport of the permanganate and CVOCs. This information was used to guide subsequent
oxidant dosages and injections into the source zones. Post-oxidation soil sampling and analysis was executed
after the third and final ISCO deployment. Contrasting the pre- and post-oxidation CVOC soil data served
as a metric by which to critically assess ISCO treatment performance. Overall, significant reduction in the
mass of CVOCs in the ISCO targeted areas was achieved through the three oxidant injection events. This
was apparent through the significant reductions in the CVOC concentration evident by contrasting CVOCs
isocontours (Figures 14a and 14d) and the overall 91.5% and 75.5% reduction in total CVOC mass flux in the
shallow and deep surficial aquifer (Figure 15.c).
5.4 Metals mobilization
There are two main mechanisms for increasing concentrations of metals in the ground water during ISCO:
(1) the KMn04 or NaMn04 provided by the manufacturer may contain elevated levels of the heavy metals,
and (2) mobilization of pre-existing redox- or pH-sensitive heavy metals (in situ) by the oxidant (Huling
and Pivetz, 2006). The content of heavy metals in permanganate is dependent on the type and source of
the oxidant. Because of the manufacturing process, NaMn04 has lower concentrations of heavy metals
than KMn04. Both forms of the oxidant are manufactured in the US, Germany, and China. In 2006, Carus
Chemical, a manufacturer of permanganate in the US provided analytical data for heavy metal impurities
in their products. The remediation-grade KMn04 was developed by Carus Chemical to contain minimal
quantities of metal impurities.
Chromium (Cr) and arsenic (As) have historically been the impurities of concern. Due to the low maximum
contaminant level (MCL) in drinking water, established by EPA for these metals (0.1 mg/L total Cr MCL, 0.01
mg/L As MCL) (U.S. EPA, 2002), injection of technical-grade KMn04 may exceed the MCL for these elements.
Generally, natural attenuation of these metals has been achieved within acceptable transport distances
and time frames. Due to the possibility of exposure pathways and potential receptors, monitoring of these
parameters may be needed under some conditions. A site-specific evaluation of the potential impact of
heavy metals should be conducted to assess whether ground water monitoring for these metals is needed
(Huling and Pivetz, 2006).
5.4.1 Chromium (Cr)
Enhanced transport of pre-existing, or naturally occurring, redox- or pH-sensitive metals may occur
when deploying ISCO using permanganate. Changes in ORP and pH may change the fate and transport
of chromium. Depending on pre-ISCO site conditions, permanganate ISCO can drop the pH of a system
to below 3, or raise it to above 10 (Petri et al., 2012, and references therein). Further, highly oxidizing
conditions involving permanganate ISCO can raise the system ORP to as high as +800 mV (Siegrist et al.,
2001). Oxidation of Cr(III) to Cr(VI) by Mn04", and the subsequent mobilization, have been demonstrated in
the laboratory (Li and Schwartz, 2000; Chambers et al., 2000b). Additionally, Cr(VI) and Ni mobilization has
been observed under field conditions, where Mn04" has been injected (Crimi and Siegrist, 2003). Several
field studies have reported anomalously high post-oxidation concentrations of Cr(VI), but natural attenuation
of Cr(VI) was observed (Crimi and Siegrist, 2003), and cleanup concentrations have been achieved within
acceptable transport distances and time frames (Chambers et al., 2000a).
85

-------
The ground water monitoring data for chromium (section 4.5.6 Metals) indicated that chromium
mobilization occurred as a result of injecting permanganate at site 45, but that attenuation mechanisms
were efficient and effective, and that chromium persistence and transport was limited. Cr+S undergoes
natural attenuation through several mechanisms (McLean and Bledsoe, 1992; Palmer and Puis, 1994),
including adsorption to MnOz(s) and various other mineral species containing iron. Given the abundance of
manganese and iron mineral species at the site (Figure 11), and the abundance of MnOz(s) resulting from
permanganate ISCO, elevated concentrations of chromium were not projected. Further, Cr+S attenuation
mechanisms were expected to play a major role, where chromium transport would be temporary,
attenuation rapid, and transport distance limited. This pilot-scale study provided an excellent opportunity
to assess whether metals mobilization was an issue, and whether attenuation occurred at a rate that was
acceptable prior to implementing a final ISCO remedy.
5.4.2 Arsenic (As)
The two main forms of arsenic are arsenite (As(lll))
and arsenate (As(V)), and specific arsenic species are
dependent on redox conditions and pH (Table 11).
It is apparent that numerous arsenic species may
occur in ground water under varying pH and redox
conditions. The reduced form, arsenite, is more
toxic, soluble, and mobile than the oxidized form,
arsenate. There are several mechanisms in which arsenic removal from the ground water may occur as
a result of the oxidation of aquifer materials. Complex arsenic chemistry and variability in subsurface
geochemical conditions contribute to the varying roles of removal mechanisms, and consequently the
relative role of each mechanism may be difficult to quantify (Huling et al., 2016b).
Table 11. Stability of arsenic species (Vu et al., 2003).
Reducing Conditions Oxidizing Conditions
PH
As(lll)
PH
As(V)
0-9
H3As03
0-2
H3As04
10-12
H2As03
3-6
H2As04
13
HAs032
7-11
HAs042
As(III) and As(V) adsorb to iron (Fe) minerals found in aquifer material through complexation reactions,
which form various As-Fe species (Sun et al., 1999; Vu et al., 2003; Akai et al., 2004). Under most
environmental systems (i.e., pH 6-9), As(V) adsorbs more strongly than As(lll), and Fe(lll) complexes
more arsenic than Fe(ll). Complexation and adsorption of arsenic species involve electrostatic attraction
between the anionic form of arsenic species (Table 11) and the cationic forms of iron and other positively
charged sorption sites and surfaces. Through this mechanism, oxidized aquifer materials, containing
greater quantities of ferric iron, have greater potential for the arsenic adsorptive removal mechanism than
reduced aquifer materials.
Arsenic removal may occur by co-precipitation with iron and other metals. Specifically, in ground water
containing Fe(ll) and As(lll), a shift towards oxidative conditions would result in the oxidation of As(lll) to
As(V), Fe(ll) to Fe(lll), the precipitation of ferric iron (Fe(lll)) and co-precipitation, or coagulation, of As(V)
(Johnston and Heijnen, 2001). Once Mn04" is applied to aquifer materials, the predominant manganese
reaction byproduct is Mn02(s), which is a stable form of manganese. Further, MnOz(s) readily oxidizes
As(lll) to As(V), which becomes adsorbed to the hydroxyl group on the MnOz(s) surface (Oscarson et al.,
1981; Sun et al., 1999), forming birnessite ((MnO)2AsOOH) (Manning et al., 2002).
Microbial mechanisms responsible for the reduction of arsenate to arsenite can result in the enrichment
of arsenic in ground water (Akai et al., 2004). However, oxidative conditions resulting from the application
of permanganate, persulfate, or H202, yield antiseptic conditions that are detrimental to microbial species
and activity. Consequently, inhibition of microbial activity that carries out reductive processes, limits
the arsenic dissolution process. The aquifer materials at site 45 exhibit high concentrations of naturally
occurring iron (Fe 2-10 g/kg) and manganese (Mn 20-70 mg/kg) (Figure 11), and were supplemented with
MnOz(s) through the NaMn04 ISCO processes, suggesting that these arsenic attenuation mechanisms could
play a significant role.

-------
5.5 Natural attenuation
5.5.1 Proposed natural attenuation conceptual model
Based on the available data and information, several lines of evidence indicate that natural attenuation of
the CVOCs was occurring in both the source area and downgradient zones of the aquifer. Attenuation of
CVOCs occurs via biotic and abiotic reductive dehalogenation, and through other fate mechanisms (US EPA,
1998). The naturally occurring total organic carbon (TOC) (Figure 10) in the aquifer material is an available
source of substrate material supporting biotic activity and serves as an important parameter, impacting
CVOCfate and transport, potential electron donor materials, and the natural attenuation processes in the
subsurface. A general increasing trend in TOC concentration with depth was observed, but a correlation
between TOC and depth was not established (Figure 10), suggesting that there is not a distinct, high
organic, lithologic layer in the subsurface over this interval. Nevertheless, the presence, abundance,
and availability of TOC over the profile supports biotic activity and reducing conditions. The presence
of terminal electron acceptors in the deeper zone including solid phase iron, manganese, and sulfate, is
projected to support reductive transformation processes (Huling et al., 2002); these appear to be more
abundant in the deeper zone (Figure 11). The occurrence of reduced conditions is evident by the presence
of ferrous Fe (Figure 17.A) and the presence of sulfides, hydrogen, and methane, measured in the ground
water at depth (Vroblesky et al., 2009). High sulfate concentration (112-130 mg/L) in downgradient wells
indicates that the sulfate reducing condition is outcompeting methanogenesis. Hydrogen measurements
(1.1-3.4 nM/L) in six of the seven wells are indicative of sulfate reducing conditions (Vroblesky et al., 2009).
The increase in acid-extractable Fe with depth
suggests that Fe reduction could be a predominar
terminal electron acceptor process to a depth
of 16 ft bgs. However, acid extractable Fe is
not necessarily a measure of bioavailable ferric
iron, so the actual availability of microbially-
reducible ferric iron is not known. The co-
existence of ferrous iron, sulfide, and methane
at the site is an indication that multiple terminal
electron accepting processes are occurring
simultaneously, either at different locations of
the site, or at different micro-sites in the same
subsurface location. Lower redox in the deeper
zone (Vroblesky et al., 2009) could result in faster
rates of reductive dechlorination (i.e., natural
attenuation) of the CVOCs.
Although the surficial aquifer is considered
unconfined, drawdown patterns more closely
represent a confined or leaky-confined aquifer
(TtNUS Rl report, 2004). This may be the result
of the presence of the relatively finer-grained,
silty-sand sediments within the upper portion
of the shallow aquifer, in comparison to the
deeper sediments of fine sand (Figure 22). This
observation is consistent with aquifer cores
collected and visually examined by EPA-RSKERC
staff, which indicated that a thin clay layer exists
in the 3-6 ft bgs interval. Specifically, hydraulic
conductivity measurements using ASTM method
87

Hydraulic Conductivity (cm/s)
0 0 002 0.004 0 006 0 008 0.01 |
u
2
0) 4 -
&


0J
I

"t 6 •
I

W
,

T3
r*


1 8'


0


5
I 1°
CD


-C
Q.
S 12
ti



14
n-t1

16 •
u 	

1
Figure 22. Hydraulic conductivity values
in aquifer material collected from Parris
Island, SC, MCRD, site 45 using a laboratory
permeameter (ASTM Method D2434).

-------
D2434 were performed at EPA-RSKERC, using aquifer cores collected at site 45 (Figure 22). It is evident
that there is a thin, low permeability clay layer at approximately 3-6 ft bgs, and that a highly conductive
zone co-exists at and just below that depth, contributing to CVOC persistence and transport. Further, the
permeability decreases with depth, in conjunction with the increased clay content of the aquifer solids. The
low permeability of aquifer media with depth would result in slow ground water transport conditions and
would contribute to restricted transport and dispersal of CVOCs at this depth.
Despite the significant source strength of CVOCs at the corner of the dry cleaner building (Figure D.2-D.3
in Appendix D), low concentrations or the absence of CVOCs in deeper aquifer intervals at downgradient
transect locations (T1-T6) (Figure D.4-D.5 in Appendix D) are an indicator that natural attenuation
mechanisms play a significant role in the fate of CVOCs with depth. Overall, multiple lines of evidence
suggest that permeability, ground water velocity, and CVOC transport decreases with depth and that biotic
and abiotic natural attenuation of CVOCs is greater in the deeper zone than the shallow zone.
5.5.2 ISCO impact on natural attenuation
ISCO can potentially change or affect several parameters and mechanisms associated with natural
attenuation. These impacts may enhance or inhibit, be significant or negligible, or be short-term or
long-lasting. NaMn04 is an antiseptic (i.e., biocide) and will kill and inhibit microbial activity at much
lower concentrations than what was injected into the subsurface at site 45. Typically, microbial toxicity
is evaluated in simple laboratory systems that generally permits excellent contact between oxidant and
microbes, therefore, results project a high impact of the oxidant on microbial activity at low oxidant
concentrations. In subsurface systems, there are heterogeneities in aquifer material that cause short
circuiting, dead zones, low-flow areas, and micro-niches that result in incomplete contact between the
injected oxidant and microbial populations in aquifer material. This allows microbial populations to survive
rigorous applications of oxidants during ISCO. The same heterogeneous, anisotropic condition that limits
the ability to uniformly deliver oxidant into the subsurface, is probably also the same condition that
prevents the injected oxidant from impacting microbial activity in subsurface systems. Therefore, laboratory
studies conducted in ideal, complete-mix systems provide insight into the microbial toxicity but do not
represent actual non-ideal systems, which permit microbial survival and activity under harsh oxidative field
conditions. These basic differences between laboratory and field conditions help explain discrepancies
between microbial toxicity and inhibition results reported from laboratory studies and the seemingly low
impact of ISCO on microbial activity at field-scale (Luhrs et al., 2006).
The oxidation reduction potential (ORP) is a general indicator of oxidation conditions in the subsurface. The
generally accepted sequential order of TEA utilization, under aerobic and anaerobic conditions is:
o,
¦
2
Aerobic
H
nitrate reducing
Fe+3, Mn+4
iron, manganese reducing
sulfate reducing Hi methanogenesis
It is generally accepted that methanogenesis is the least efficient TEAP. Iron and sulfate reducing
conditions are the most favorable for reductive dehalogenation processes for the CVOCs at site 45. These
transformations are more energetic and efficient, relative to methanogenesis. The injection of NaMn04
oxidant into the subsurface will result in a strong increase in ORP and shift the TEAP to the left, killing or
inhibiting some microbial species. Examination of the process residuals suggests that ISCO would shift
the predominant terminal electron accepting process from an inefficient one (methanogenesis) to more
efficient processes, such as Fe+3 and S042" reduction, and provide a sustained long-term source of TEA. The
injection of oxidants into the subsurface will result in the rapid oxidation of aquifer sediments and increase
the relative bioavailability and abundance of TEA. For example, ferrous iron is oxidized to ferric iron,
sulfides to sulfate, and so on. In general, the short-term impact of ISCO was projected to negatively impact
natural attenuation, but the long-term impact is projected to enhance and help sustain natural attenuation,
given the shift in electron accepting conditions, and the long-term bioavailability and abundance of TEAs.

-------
Some oxidation reactions are acid producing, so pH changes can occur under some conditions. Since pH
is widely known as the master variable in water chemistry, this parameter has direct and indirect effects
on biotic and abiotic attenuation processes. However, the ground water is well-buffered at site 45 and
pH changes in the post-oxidation ground water were not significant. Oxidation of organic contaminants
by different oxidants, and formation of simpler organic molecules, results in an increase in the solubility,
bioavailability, and biodegradability of the reaction byproducts (Allen and Reardon, 2000; Klens et al., 2001;
Droste et al., 2002; Chapelle et al., 2005; Luhrs et al., 2006; Rivas, 2006). For example, many post-oxidation
reaction byproducts are identified as carboxylic acids, chloroacetic acids, and other unregulated compounds.
Finally, ISCO was deployed in the source area and along the longitudinal axis of the plume at site 45; this
resulted in a physical separation between the ISCO area and other downgradient and side-gradient areas
where ISCO had minimal impact on natural attenuation processes.
In summary, the large volume of the NaMn04 solution injected into the subsurface was broadly dispersed.
As a result, it is projected that (1) there will be a localized decline in microbial activity near the injection
points and farther away from the injection points, where there was direct contact between the NaMn04
and microbial populations, (2) the inhibitory effects on microbial activity will diminish with time, leading
to microbial rebound after the oxidant is reacted, and (3) increased bioactivity will result from improved
bioavailability of trace constituents, greater abundance of easily oxidized substrates, lower concentrations of
challenging chemicals, increased levels of TEAs, and more efficient TEAPs.
5.6 Contamination rebound
Rebound is a site condition where contaminant concentrations in ground water decrease as a result of
remediation, but then increase over time. In general, this term should be applied to parent compounds
only (i.e., PCE at site 45); this is due to the potential for other mechanisms to alter the concentration of
decomposition products.
In ISCO, rebound has been attributed to several mechanisms and conditions, including the slow post-
oxidation equilibrium of contaminant mass transfer (i.e., the dissolution of NAPL residuals and desorption of
residual contaminants in solid-phase media), and the back diffusion of contaminants from low permeability
media, etc. Other complexities in contaminant rebound include long-term persistence of oxidants, or
impact on hydraulic conductivity by entrapment of 02(g), especially with H202-based ISCO. Various methods
have been used to quantify rebound. For example, in one case it was considered to have occurred with
concentrations increased greater than or equal to 25% over the post-treatment monitoring period (McGuire
et al., 2006). Assuming one year of post-treatment data, the geometric mean of the first half of the year
would be compared to the second half of the year; for two years of post-treatment data, the geometric
mean of the first year would be compared to the second year. In Krembs et al. (2010), rebound was
determined when the increase in the concentration of total contaminants of concern (COCs) in groundwater
during the post-ISCO monitoring period was greater than 0.25 of the pre-ISCO baseline value (Eqn. 5).
[COCsjj
Year post-ISCO
- [COCs]
lowest post-ISCO
[COCs]
>0.25
pre-ISCO baseline
(Eqn.5)
Rebound at site 45 may be attributed to several potential mechanisms associated with the incomplete
oxidation of CVOCs. These mechanisms have a cumulative overall effect on CVOC concentrations and
include, but are not limited to: (1) the non-uniform distribution of injected oxidant, and the heterogeneous
distribution of CVOCs (i.e., simply not delivering a sufficient volume of oxidant for adequate contact with the
TTZ), (2) the back diffusion of CVOCs from low permeability materials, (3) the accumulation of Mn02(s) at
DNAPL interfaces, which impede CVOC mass transfer and CVOC and Mn04" mass transport, (4) insufficient
delivery of oxidant to the contaminated media (i.e., the mass of oxidant delivered was inadequate to satisfy
the natural oxidant demand and to oxidize the CVOCs), and (5) slow mass transfer associated with the
dissolution of DNAPL residuals and desorption of residuals on solid-phase media.

-------
Overlapping oxidant ROIs, total porosity oxidant volume design, heavy oxidant loading, and multiple
oxidant injections into the source zone were components of the ISCO design at site 45, to ensure aggressive
oxidant delivery to the CVOC targeted zones. Other oxidant injection strategies were used to limit the
role of heterogeneous porous media and to improve oxidant coverage in the TTZ, including low injection
pressure, short injection intervals, and small ROIs at numerous injection points. However, the impact
of heterogeneities on oxidant transport cannot be completely avoided and non-uniform distribution of
oxidant occurred at site 45. The inability to achieve ideal contact between oxidant and the TTZ can be
amplified due to the heterogeneous distribution of CVOCs. Consequently, the oxidation of CVOCs may
range from being highly effective in areas where good contact was achieved, to less effective in other areas
where good contact was not achieved. Persistent CVOC ground water concentrations can be used as an
empirical measure of incomplete contact between the oxidant and the TTZ, and the identification of hot
spots of CVOCs, DNAPL, and NOD.
The lithology of the aquifer material at site 45 ranges from highly permeable sand to sandy clay, and the
study area is underlain by a peat-clay material. This lithology suggests that the downward transport of
PCE DNAPL would be limited, and that diffusion transport of the CVOCs would dominate in the lower
permeability material, relative to advective transport. Oxidant delivery into higher permeability material
would sweep through relatively quickly via advection. However, diffusion, a much slower transport
process would dominate CVOC and oxidant transport in the lower permeability material. Specifically,
both the transport of permanganate into the low permeability material and transport of CVOCs from the
low permeability materials would be limited. Overall, CVOC oxidation may also be limited in these zones,
allowing CVOCs to slowly diffuse and rebound in the ground water.
The accumulation of MnOz(s) at NAPL interfaces in the source area at site 45, may have occurred and is
projected to slow mass transfer (i.e., dissolution of the DNAPL components in the ground water). In some
respects, this result may be desirable in that the CVOC transport distance may be significantly reduced
under this condition, when considering the role of natural attenuation (i.e., the CVOC plume length and
dimensions are limited, static, and/or declining).
The PCE concentrations measured in the pre-oxidation, post-oxidation 3, and post-oxidation 4 ground
water monitoring events were used to assess rebound of PCE at site 45, using Eqn. 5. The following data
sets were used for the parameters in Eqn. 5 for all ground water micro-wells, MW-25, and MW-31 SL.
Eqn. 5 parameters
Ground water monitoring data sets
[COCs] pre-ISCO baseline
Pre-oxidation baseline [PCE] (Feb. 20, 2013)
[COCs] lowest post-ISCO
Post-oxidation 3 [PCE] (June 23, 2014)
[COCs] 1 Year post-ISCO
Post-oxidation 4 [PCE] (Feb. 27, 2015) (11 mos. post ox)
The results of this analysis indicated that 3 of the 37 wells evaluated showed rebound. These wells were
either in the source area (Ml-S-A-shallow, Ml-S-A-deep) or were along the centerline of the plume (M2-S-
shallow). The post-oxidation performance evaluation metrics used in this study indicated that reductions in
soil and aqueous CVOC concentrations and in mass flux have been achieved.
However, based on the elevated concentrations of CVOCs measured in the source area soil samples, further
CVOC rebound is probable. One depth interval where elevated post-oxidation [CVOC]SOIL were measured
(i.e., (-5, -4) sampling location, -14 ft bgs interval) exhibited the highest CVOC concentration (576 mg/kg)
(Appendix D, Figure D.6), indicating the presence of DNAPL. The oxidant loading into the source area was
insufficient to fully oxidize this material. Consequently, it is projected that rebound would occur in the
source area and subsequent ISCO activities are recommended to target the source area and to address
CVOC rebound.

-------
5.7 Sustainability
The ISCO design and deployment associated with this study was aimed at achieving a sustainable solution
through various design steps and approaches. First, the remedy is expected to be fully integrated with a
final natural attenuation remedy since it is projected that ISCO will not achieve the MCLs for CVOCs. Due
to the high resolution site characterization method and the development of an accurate site conceptual
model, clean portions of the aquifer were eliminated from further consideration for oxidant delivery. The
most contaminated portions of the aquifer were then targeted to receive appropriate oxidant dosages,
thus achieving good oxidation reaction and oxidant efficiency. These aspects of the ISCO design and
deployment helped to limit the overall amount of oxidant used, and indirectly the amount of fossil fuels
consumed and the project's carbon footprint.
5.8 Recommendations
It is recommended to continue ground water monitoring and to carry out additional, but limited, ISCO
activities.
5.8.1	Proposed monitoring activities
It is recommended to perform semi-annual ground water monitoring for CVOCs, metals, Mn04", and
indicator parameters (i.e., pH, oxidation-reduction potential (ORP), specific conductance (SC), dissolved
oxygen (DO), and temperature) to assess Mn04" persistence, metals mobilization and attenuation, and
CVOC destruction and rebound. A proposed ground water monitoring plan has been included as Appendix
F. Dissolved methane gas, ferrous and total iron, chloride, and sulfate are optional parameters that could
be measured and may provide useful background data and information to assess natural attenuation.
5.8.2	Proposed ISCO activities
It is recommended that continued, but limited, ISCO activities be deployed to address the CVOC residuals
at the site. The basis for recommended ISCO activities includes the following:
(1)	It is assumed that there are CVOC residuals in the source area located near the suspected point
of origin. The point of origin is in the area of the former sanitary sewer that existed adjacent to
the southeast corner of Bldg. 192.
(2)	It is assumed that CVOC rebound will continue to occur, mainly in the source area, and that
the CVOC concentration could eventually exceed acceptable levels. It is also assumed that
CVOC concentrations may generally rebound downgradient from the source area. Additional
ground water monitoring is needed to assess this potential occurrence and determine whether
additional ISCO activities are needed.
(3)	It is assumed that periodic injection of the permanganate oxidant in the source area will prevent
the re-development of the CVOC plume and will limit and/or prevent the capture of CVOCs in
the downgradient storm sewer.
91

-------
Regarding source area ISCO, it is recommended
to periodically inject oxidant into 3-5 injection
wells located in the source area. Two of these
wells are existing and were used as injection
wells during the ISCO pilot study (i.e., MW-25
and Inj.l-deep). The other 1-3 proposed wells
are located close to the existing wells but serve
to inject the oxidant across a broader footprint
in the source area. Collectively, these weils
would deliver oxidant into the source area
across a broader lateral direction, transverse to
the ground water and CVOC plume transport
direction (Figure 23). Given these stationary
injection wells, oxidant injection would involve
more simplified mobilization and oxidant
delivery than the direct-push injection method.
The specific details of the recommended design
include 4 ft ROIs, used to cover a broader
oxidant footprint in the source area; screened
intervals of 4-5 ft are needed to address the
suspected CVOC residuals that occur in the
deeper source area (Table 12). Inj.-l, Deep is
located upgradient from MW-25, and therefore,
post-injection oxidant transport will occur
along the longitudinal axis of the PCE source
zone. This aggressive oxidant loading delivery
is designed to continue targeting the precise
locations where the highest [CVOCs]SOIL were
measured. lnj.-2 and -3 are located transversely
to the longitudinal axis of the PCE source zone
to address PCE that may have migrated laterally
from the source zone. However, these weils are
not located very far from the suspected release
point, given the limited lateral distribution of
CVOC in the soil and ground water near this
location.
Previous oxidant loadings to the source area,
during injection events 1-3, were aggressive
and are expected to have met the oxidant
demand of the aquifer material. Therefore,
the proposed oxidant concentration (20 g/L)
in this design is expected to achieve long-
term oxidant residuals needed to address the
long-term source of CVOCs that exists in this
area. The overall cost of oxidant using this
oxidant injection strategy is estimated to be
approximately $3,000. It is projected that 2-3
additional oxidant injections may be needed
but should be based on continued ground
water monitoring results.
New Dry
Cleaner Building
-
Drive-through 	>

\
k ^
•
IA //
V
o



o
Kyushu Street

'
)

Sanitary sewer {manhole cover)
Fence
Monitoring (injection} well MW25
Q Direct push injection well tnj.~l*D (11-15 ft bgs)
Proposed source area injection wells (10-15 ft bgs}
(lnj.-2, -3, -4)
Proposed downgradient injection well(8-12 ft bgs)


Figure 23. Existing and proposed oxidant injection well
locations for recommended continued oxidant injection
activities at site 45.
Table 12. Summary of oxidant (NaMnOJ volume and
loading for the long-term ISCO activities. Injection wells
lnj.-2, lnj.-3, and lnj.-4 are proposed and depicted in
Figure 23.
Injection Wells
NaMn04
NaMn04
Projected

Volume
Mass
total cost of

(gal)
(lbs)
oxidant (1|($)
MW-25, lnj.-2, lnj.-3(2)
600(1800)
360

Inj.-l, deep
480 (480)
96
$3080
1	Carus Chemical method of estimating involves the following steps: divide
lbs of NaMn04 by 0.4 since it is 40% NaMn04; since each pail of 40%
contains 57 lbs, divide the weight in step 1 by 57 then round up to the next
number, multiply the round number of pails by 57 lbs/pail, then $2.70 / lb.
2	Assumes (1) a 4 ft ROI, (2) 5 ft screened interval for MW-25, lnj.-2, -3, and -4,
a 4 ft screened interval for Inj.-l, deep, (3) 20 g/L NaMn04, and
(4) $2.70/lb NaMn04.

-------
6. Conclusions
A pilot-scale ISCO demonstration involving subsurface injections of sodium permanganate was performed
at the Marine Corps Recruit Depot, site 45 (Parris Island, SC). The ground water at the site is contaminated
as the result of the release of tetrachloroethyiene (PCE), a chlorinated solvent used in dry cleaner
operations. The treatment objective was to evaluate the feasibility of ISCO as an effective remedy, over
a limited area, to reduce CVOC concentrations and mass flux in conjunction with natural attenuation.
The results of this study are intended to provide details and guidelines that can be used by EPA and DoD
remedial project managers regarding ISCO for remediation at other sites.
93

-------
A preliminary site conceptual model was based on
ground water samples collected from temporary
and permanent ground water wells and was
effective in delineating the plan view distribution
of CVOCs. The well screen lengths and placement
of these wells tended to average the vertical
distribution of CVOCs. Therefore, the vertical
extent of CVOC contamination was subsequently
quantified and the conceptual site model was
refined through the analysis of soil cores collected
at locations transverse to the ground water flow
and along the longitudinal axis of the CVOC plume
transport direction. These data were augmented
to fill in data and information gaps by sampling
and analyzing archived cores that were frozen
and stored at EPA RSKERC. Additional soil cores
were collected and analyzed to further refine the
CSM, as further data gaps were identified. High
resolution site characterization also involved
the installation of nested micro-wells used to
sample and analyze ground water. This overall site
characterization methodology was critical in the
continued refining of the CSM and the ISCO design,
not only by identifying CVOC contamination, but
also by identifying and eliminating uncontaminated
portions of the aquifer from further ISCO
consideration. Throughout this process, ISCO
activities were focused on the most contaminated
portions of the aquifer, resulting in efficient and
effective ISCO deployment. Overall, both the soil
and ground water data were critical in further
developing and refining the conceptual site
model, the ISCO design, and in assessing the ISCO
treatment performance.
Sodium permanganate was selected as the oxidant,
and direct-push was selected as the main method
of oxidant delivery, due to its flexibility and low
initial capital cost. Numerous impediments and
subsurface utilities existed in the source area
where the oxidant was to be injected, including
an 8-inch high pressure water main, a high voltage
power line (230 V), a communication line, a
sanitary sewer, a stormwater sewer, and the
drive-through to the dry cleaner. An existing
monitoring well and a newly constructed,
temporary direct-push injection well were used
in the oxidant injection. Utilizing these wells
augmented the delivery of a heavy oxidant load
into the source area.
A portable, low cost injection system was designed,
constructed, and deployed at the site. All parts of
the system, from inlet to outlet, were constructed
with corrosion-resistant components, and included
a 150-gallon tank, a teflon diaphragm pump,
three oxidant injection arms, and an injection
hose designed to fit inside the GeoProbe rod as a
separate unit and attach directly to the injection
tip. This design was used to help maintain hydraulic
control of the oxidant and to limit the potential
for leaks and the risk of exposure. Several oxidant
delivery designs and methods used in the ISCO
deployment resulted in an aggressive, effective,
and efficient oxidation of CVOCs. Specifically, this
includes narrow ROIs (3-5 ft), short vertical screen
injection intervals, low injection pressure, outside-
in oxidant injection, and a total porosity oxidant
volume design. The injection tip was designed with
a 2 ft screened interval. The top-down approach
involved advancing the injection tip to the first
depth interval, delivering the oxidant, driving to
the next depth, and delivering the oxidant, etc. This
configuration resulted in less risk of oxidant short-
circuiting and greater certainty that the oxidant
was delivered to the targeted interval. The outside-
in oxidant injection strategy was used to minimize
the lateral displacement of contaminated ground
water. Due to the shallow water table and injection
depths, low injection pressures and flow rates were
required to minimize the potential for daylighting
of the oxidant and to limit the role of preferential
pathways.
The volume of oxidant solution to be injected
at each location was based on a radial-flow,
cylindrical, total porosity porous media conceptual
model. This design method was expected to result
94

-------
in the delivery of a large volume of oxidant over
a large volume of contaminated aquifer media.
Based on the average NOD values measured in the
5-11 ft bgs interval {i.e., <2.7 g NaMn04/kg aquifer
material), and accounting for natural variability, the
actual oxidant loading (3.8-5.2 g/kg) was expected
to exceed the majority of the NOD measured in
the subsurface. Due to heterogeneities in aquifer
characteristics, heavy oxidant loading, overlapping
ROIs, and multiple oxidant injections, it was
projected that the oxidant would be transported
beyond the projected ROIs in the targeted
treatment zones. In contrast, the mobile porosity
design method sometimes used at ISCO sites was
not used at site 45; it is based on the delivery of
oxidant to the most conductive portions of the
aquifer, where ground water transport is fastest.
It was projected that using the mobile porosity
design method would be less aggressive, involve a
smaller injection volume and oxidant footprint, and
leave some zones oxidant-free within the targeted
treatment zone.
Administrative requirements associated with ISCO
activities at site 45 included the preparation of
health and safety plans, dig permits, injection and
monitoring well construction permits, and oxidant
injection permits. Coordination of these plans
and permits with the responsible parties, EPA,
and state regulatory agencies was an important
component in the remediation approval and
deployment process.
Three oxidant injection events were carried out
where the oxidant loading and areal footprint
were progressively larger. Given the proximity
of the targeted treatment zones with surface
and subsurface utilities, it was necessary to first
establish the oxidant could be delivered safely,
and without migration into non-targeted zones.
The measurement of NaMn04 in nearby wells
and micro-wells indicated that oxidant injection
was successful: the oxidant was delivered into the
targeted zones, hydraulic control of the injected
oxidant was achieved, the oxidant persisted
in zones where a heavy oxidant loading was
delivered, and that significant CVOC destruction
was achieved. Overall, a 91.5% and 75.5%
reduction in total CVOC mass flux occurred in
the shallow and deep zones, respectively, as a
result from oxidant injections 1-3. Contrasting
numerous post-oxidation and pre-oxidation CVOC
concentrations in soil samples indicated a declining
trend in post-oxidation [CVOCs]SOIL, relative to pre-
oxidation [CVOCs]SOIL. A CVOC rebound analysis
indicated that rebound has occurred in 3 of 37
wells and micro-wells at the site. At one depth
interval in the source area, elevated post-oxidation
[CVOCs]SOIL was measured and indicated the
presence of DNAPL. For this reason, it is probable
that CVOC rebound will continue to occur in the
source area, but would likely play a lesser role in
the downgradient portions of the plume, where
lower CVOC mass and mass flux were measured.
Subsequent ISCO activities are recommended
to target the source area and to address CVOC
rebound.
The permanganate persisted mainly in the source
area near the southeast corner of Building 192,
where heavy oxidant dosages were delivered to
the suspected PCE release zone. Long-term oxidant
persistence in this area is desirable as it permits
the long-term contact between the oxidant and
(1) CVOCs slowly dissolving from DNAPL, (2) CVOCs
desorbing from aquifer solids, and (3) CVOCs
diffusing from low permeability, silty media.
A significant post-oxidation increase in ORP
was measured in the source area and along the
longitudinal axis of the plume, where the oxidant
was injected. The redox shift towards oxidized
conditions is projected to be temporary, given
the significant quantity of organic materials in the
subsurface that favor reduced conditions. The
reduction in ferrous iron concentrations in ground
water along the longitudinal axis of the plume
and in the source area indicates the significant
95

-------
impact of the oxidant injected in these areas. Despite
significant reductions in total CVOCs, a stark contrast
between pre- and post-oxidation chloride isocontours
was not observed. This was attributed to the high
background chloride levels in the ground water,
typical of surficial aquifers near marine environments,
and a source of CI" attributed to long term reductive
dechlorination at the site. The concentration of
total chromium increased in the ground water
from background concentrations in a few wells. A
significant decline in the post-oxidation concentration
of chromium was measured. Given the abundance of
manganese and iron mineral species at the site, the
continued attenuation of chromium is projected to be
rapid and elevated concentrations to be temporary.
The attenuation of background, naturally occurring
arsenic occurred as a result of ISCO activities. Multiple
attenuation mechanisms associated with ISCO
geochemistry indicated that the arsenic species in the
ground water were immobilized, or adsorbed onto the
aquifer solids as a result of oxidation.
Firm conclusions were not possible from the MBT
results regarding the impact of ISCO on biotic processes
in the subsurface. Further, it was difficult to establish
firm and definitive trends in the CSIA analytical results.
Nevertheless, it can be concluded that in several
wells located close to the oxidant injections, where
significant reductions in CVOCs were measured, the
CSIA data provides proof of concept that PCE, TCE,
c-DCE, and VC were destroyed.
Recommendations are provided to perform semi-
annual ground water monitoring for CVOCs, metals,
Mn04", and indicator parameters to assess Mn04"
persistence, CVOC destruction and rebound, and
metals attenuation. It is recommended to periodically
inject oxidant into 3-5 injection wells located in the
source area; two of these wells are existing and used
during the ISCO pilot study. The other three proposed
wells are located close to the existing wells and serve
to inject the oxidant across a broader footprint in
the source area. Specific details and guidelines are
provided regarding these recommendations.
96

-------
Akai, J., Izumi, K., Fukuhara, H., eta!., 2004. Mirieralogic arid geomicrobiological investigations on
groundwater arsenic enrichment in Bangladesh. Applied Geochemistry 19, 215-230.
ASTM (American Society of Testing Materials). 2007. ASTM D7262-07 Standard test method for estimating
the permanganate natural oxidant demand of soil and aquifer solids. ASTM International,
West Conshohocken, PA, USA. 5p.
Borden, R.C., C.A. Gomez, and M.T. Becker. 1995. Geochemical indicators of intrinsic remediation,
Ground Water, (33)2, 180-189.
Bradley, P.M., 2003, History and ecology of chloroethene biodegradation—A review: Bioremediation
Journal, (7)2, 81-109.
Chambers, J., A. Leavitt, C. Walti, C.G. Schreier, J. Melby, and L. Goldstein. 2000a. In situ destruction of
chlorinated solvents with KMn04 oxidizes chromium. In: Chemical Oxidation and Reactive Barriers,
The Second International Conference on Remediation of Chlorinated and Recalcitrant Compounds.
Wickramanayake, G.B., A.R. Gavaskar, and A.S.C. Chen (Eds.), Battelle Press, Columbus, OH. pp. 49-55.
Chambers, J., A. Leavitt, C. Walti, C.G. Schreier, and J. Melby. 2000b. Treatability study: Fate of chromium
during oxidation of chlorinated solvents. In: Chemical Oxidation and Reactive Barriers, The Second
International Conference on Remediation of Chlorinated and Recalcitrant Compounds.
Wickramanayake, G.B., A.R. Gavaskar, and A.S.C. Chen (Eds.), Battelle Press, Columbus, OH. pp. 57-65.
Crimi, M.L. and R.L. Siegrist. 2003. Geochemical effects on metals following permanganate oxidation of
DNAPLs. Ground Water. 41(4):458-469.
Driscoll, F.G., 1986, Groundwater and Wells, 2nd ed.: Minnesota, Johnson Division, 1089 p.
Falta, R.W., Basu, N,, Rao, S.P. 2005a. Assessing the impacts of partial mass depletion in DNAPL source
zones : I. Analytical modeling of source strength functions and plume response, J. Contam. Hydrol.
78 (2005)259-280.
Falta, R.W., Rao, S.P., Basu, N. 2005b. Assessing impacts of partial mass depletion in DNAPL source zones:
II. Coupling source strength functions to plume evolution, J. Contam. Hydrol. 79 (2005) 45- 66.
Honning, J., Broholm, M.M., and Bjerg, J.L. 2007. Quantification of potassium permanganate consumption
and PCE oxidation in subsurface materials, J. Contam. Hydrol. 90:221-239.
Huiing, S.G., B. Pivetz, and R. Stransky. 2002. "Terminal Electron Acceptor Mass Balance: NAPLs and
Natural Attenuation". J. Environ. Engin., 128(3), 246-252.
Huiing, S.G. and B. Pivetz. 2006. "In Situ Chemical Oxidation - Engineering Issue". US Environmental
Protection Agency, National Risk Management Research Laboratory, R.S. Kerr Environmental Research
Center, Ada, OK. EPA/600/R-06/072.
Huiing, S.G., Ko, S., and Pivetz, B. 2011. Ground water sampling at ISCO sites - binary mixtures of volatile
organic compounds and persulfate. Ground Water Monit. Remed. (31)2, Spring 72-79.
Huiing, S.G., Ross, R.R. and Meeker Prestbo, K. 2017. In situ chemical oxidation: permanganate oxidant
volume design considerations. Ground Water Monit. Remed. (37)1, Spring.
97

-------
Huling, J.R., Huling, S.G., and Ludwig, R. 2016b. {In Preparation) Enhanced Adsorption of Arsenic on
Aquifer Solids and Soil: Oxidative Treatment and Feasibility Assessment. Wat Res.
Hutchins, S.R., D.E. Miller, and A. Thomas. 1998. Combined laboratory/field study on the use of nitrate for
in situ bioremediation of a fuel-contaminated aquifer. Environ. Sci. Technol., (32)12, 1832-1840.
Johnston, R. and Heijnen, H. 2001. "Safe Water Technology for Arsenic Removal." In: Ahmed, M.F. et.
al. [Eds], Technologies for Arsenic Removal from Drinking Water. Bangladesh University of Engineering
/Technology, Dhaka, Bangladesh.
Johnson, K.T., Wickham-St. Germain, M., Ko, S. and Huling, S.G. 2012. "Binary Mixtures of Permanganate
and Chlorinated Volatile Organic Compounds in Groundwater Samples: Sample Preservation and
Analysis." Ground Water Monit. Remed. 32(3), Summer 84-92.
Kampbell, D.H. and S.A Vandegrift. 1998. Analysis of Dissolved Methane, Ethane, and Ethylene in Ground
Water by a Standard Gas Chromatographic Technique. J. of Chromatographic Science. 36: 253-256.
Ko, S., Huling, S.G., and Pivetz, B. 2012. Ground Water Sample Preservation at In Situ Chemical Oxidation
Sites - Recommended Guidelines, EPA Ground Water Issue Paper. US Environmental Protection
Agency, National Risk Management Research Laboratory, R.S. Kerr Environmental Research Center,
Ada, OK. EPA/600/R-12/049.
Krajmalnik-Brown, Holscher, T., Thomson, I.N., Saunders, F.M., Ritalahti, K.M., Loffler, F.M. 2004. Genetic
Identification of a Putative Vinyl Chloride Reductase in Dehalococcoides sp. Strain BAV1,
Appl. Environ.Microbiol. 70(10): 6347-6351.
Krembs, F.J., Siegrist, R.L., Crimi, M.L., Furrer, R.F., and Petri, B.G. 2010. ISCO for ground water remediation:
Analysis of field applications and performance. Ground Water Monit. Remed., 30(4) 42-53.
Krembs, F.J., Clayton, W.S., Marley, M.C. 2012. Chapter 8 Evaluation of ISCO field applications and
performance. In: In Situ Chemical Oxidation for Remediation of Contaminated Groundwater;
Siegrist, R.L.; Crimi M. L.; Simpkin T.J. (Eds). Springer Science and Business Media, LLC,
New York, New York. A volume in SERDP/ESTCP Remediation Technology Monograph Series,
C.H. Ward (Series Ed). Pgs. 319-354.
Li, X.D. and F.W. Schwartz. 2000. Efficiency problems related to permanganate oxidation schemes.
In: Chemical Oxidation and Reactive Barriers, The Second International Conference on Remediation
of Chlorinated and Recalcitrant Compounds. Wickramanayake, G.B., A.R. Gavaskar, and A.S.C. Chen
(Eds.), Battelle Press, Columbus, OH. pp. 41-48.
Los Angeles Regional Water Quality Control Board - In Situ Remediation Reagents Injection Working Group.
2009. Technical Report: Subsurface Injection of In Situ Remedial Reagents Within the Los Angeles
Regional Water Quality Control Board Jurisdiction.
http://www.swrcb.ca.gov/rwqcb4/water_issues/programs/ust/guidelines/.
Lovley, D.R. and E.J.P. Phillips. 1986. Availability of ferric iron for microbial reduction in bottom sediments
of the freshwater tidal Potomac River. Appl. Environ. Microbiol., (52)4, 751-757.
Lu, X., D.H. Kampbell, and J.T. Wilson. 2006. Evaluation of the role of Dehalococcoides organisms in the
natural attenuation of chlorinated ethylenes in groundwater: U.S. Environmental Protection Agency
EPA/600/R-06/029, 101 p.
Luhrs, R.C., R.W. Lewis, and S.G. Huling. 2006. "ISCO's Long-Term Impact on Aquifer Conditions and
Microbial Activity." In, Proceedings of the 5th International Conference on Remediation of Chlorinated
and Recalcitrant Compounds, Monterey, CA, USA, May 22-25, Paper D-48.
Manning A.B., Fendorf E.S., Bostick B. and Suarez L.D. (2002) "Arsenic (III) Oxidation and Arsenic (V)
Adsorption Reactions on Synthetic Birnessite." Environ. Sci. Technol. 36(5), 976-981.
McLean, J.E. and B.E. Bledsoe. 1992. "Behavior of Metals in Soils." US EPA Ground Water Issue Paper
EPA/540/S-92/018, Pp. 25.

-------
Microbial Insights, 2017. http://www.microbe.com/how-it-works-census/
Mumford, K.G., Thomson, N.R., Allen-King, R.M., 2005. Bench-scale investigation of permanganate natural
oxidant demand kinetics. Environ. Sci. Technol. 39 (8), 2835-2849.
Oscarson, D.W., Huang, P.M., Liaw, W.K. 1981. Role of manganese in the oxidation of arsenite by freshwater
lake sediments. Clay and Clay Minerals 29(3) 219-225.
Pace Analytical, 2017. https://www.pacelabs.com/environmental-services/energy-services-forensics/csia/
fundamentals-of-csia.pdf
Palmer, C.D., Puis, R. 1994. Natural Attenuation of Hexavalent Chromium in Ground Water and Soils.
US EPA, Office of Research and Development, USEPA/540/5-94/505.
Payne, F.C., Quinnan, J.A., and Potter, S.T., 2008, Remediation Hydraulics, Florida, CRC Press; Taylor &
Francis Group LLC, 408 p.
Petri, B., Thomson, N.R., Urynowisz, M.A. 2011. Chapter 3 - Fundamentals of ISCO using permanganate,
In: In Situ Chemical Oxidation for Remediation of Contaminated Groundwater; Siegrist, R.L.,
Crimi, M.L., Simpkin T.J. (eds). Springer Science and Business Media, LLC, New York, New York.
A volume in SERDP/ESTCP Remediation Technology Monograph Series, C.H. Ward (Series ed).
Pgs. 89-146.
Ptak, T. Piepenbrink, M., Martac, E. 2004. Tracer tests for the investigation of heterogeneous porous media
and stochastic modelling of flow and transport—a review of some recent developments, J. Hydrol.
294, 122-163.
Russo, D. 2012. Numerical analysis of solute transport in variably saturated bimodal heterogeneous
formations with mobile-immobile-porosity, Advances in Water Resources (47) 31-42.
Siegrist, R.L., Urynowicz, M.A., West, O.R., Crimi, M.L., Lowe, S.L. 2001. Principles and Practices of In Situ
Chemical Oxidation Using Permanganate. Battelle Press, Columbus, OH, USA, 336 p.
Simpkin, T.J., Palaia, T., Petri, B.J., and Smith, B. 2012. Chapter 11 Oxidant delivery approaches and
contingency planning, In: In Situ Chemical Oxidation for Remediation of Contaminated Groundwater;
Siegrist, R.L., Crimi, M.L., Simpkin, T.J. (eds). Springer Science and Business Media, LLC, New York, New
York. A volume in SERDP/ESTCP Remediation Technology Monograph Series, C.H. Ward (Series ed).
Pgs. 449-480.
Simunek, J., Jarvis, N.J, van Genuchten, M., Gardenas, A. 2003. Review and comparison of models for
describing non-equilibrium and preferential flow and transport in the vadose zone, J. Hydrol. 272,
14-35.
Sun, X., Doner, H.E., and Zavarin, M. 1999. Spectroscopy study of Arsenite [As(III)] oxidation on
Mn-substituted goethite. Clay and Clay Minerals 47(4), 474-480.
Suthersan, S., J. Horst, D. Nelson, M. Schnobrich. 2011. Insights from years of performance that are
shaping injection-based remediation systems, J. of Remediation, Spring, 9-25.
TetraTech NUS, 2004. Remedial Investigation RCRA Facilities Investigation for Site /WMU 45 Former MWR
Dry Cleaning Facility (Contract No. N62467-94-D-0888).
TetraTech NUS, 2012. Remedial Investigation Addendum for Site/SWMU 45 - Former MWR Dry Cleaning
Facility Marine Corps Recruit Depot Parris Island, South Carolina Contract Task Order 0335 April 2012.
TetraTech, 2012. Feasibility Study for Site 45 - Former Morale, Welfare and Recreation Dry Cleaning
Facility, Marine Corps Recruit Depot Parris Island, South Carolina Contract Task Order 0335, April 2012.
Urynowicz, M.A. 2008. In-situ chemical oxidation with permanganate: Assessing the competitive
interactions between target and non-target compounds. Soil Sediment Contam. 17:53-62.
Urynowicz, M.A., Balu, B., Udayasankar, U. 2008. Kinetics of natural oxidant demand by permanganate in
aquifer solids. J. Contam. Hydrol. 96:187-194.
99

-------
US EPA, 1998. Technical Protocol for Evaluating Natural Attenuation of Chlorinated Solvents in Ground
Water, US EPA Office of Research and Development, 600-R-98-128, 1998.
US EPA, 2008. A Guide for Assessing Biodegradation and Source Identification of Organic Ground Water
Contaminants using Compound Specific Isotope Analysis (CSIA), EPA 600/R-08/148, December 2008.
US EPA, 2012. GWERD Quality Assurance Project Plan for the In Situ Chemical Oxidation Using
Permanganate in the Remediation of Contaminated Groundwater at Solid Waste Management Unit 45,
Marine Corps Recruit Depot, Parris Island, South Carolina.
US EPA, 2013. Health and Safety Plan for the In Situ Chemical Oxidation Using Permanganate in the
Remediation of Contaminated Groundwater at Solid Waste Management Unit 45, Marine Corps
Recruit Depot, Parris Island, South Carolina.
US EPA, 2014. Superfund Remedy Report, 14th Ed. EPA 542-R-13-016, November 2013, Solid Waste and
Emergency Response.
US EPA GWERD RSKSOP 299.2. Determination of Volatile Organic Compounds (Fuel Oxygenates, Aromatic
and Chlorinated Hydrocarbons) in Water Using Automated Headspace Gas Chromatography/Mass
Spectrometry (Agilent 6890/597 Quadrupole GS/MS System).
US EPA GWERD RSKSOP 259.1. Determination of Volatile Organic Compounds (Fuel Oxygenates,
Aromatic and Chlorinated Hydrocarbons) in Water Using Automated Headspace Gas Chromatography/
Mass Spectrometry (TEKMAR 7000 HS-Varian 2100T GC/MS System-ION Trap Detector).
US EPA GWERD RSKSOP 215.3. Quality Control Procedures for General Parameters Analysis Using Manual
Methods for pH, Conductivity, Salinity, Total Dissolved Solids, and Total Suspended Solids.
US EPA GWERD RSKSOP 180.5. Total Nitric Acid Extractable Metals from Solids and Sludges by
Microwave Digestion.
US EPA GWERD RSKSOP 120.3 Determination of Total Carbon and Total Organic Carbon in Solids using the
LECO CR-412 Carbon Analyzer.
Vroblesky, D.A. and F.H. Chapelle. 1994. Temporal and spatial changes of terminal-electron accepting
processes in a petroleum hydrocarbon contaminated aquifer and the significance of biodegradation.
Wat Resour. Res., (30)5, 1561-1570.
Vroblesky, D. 2007. "Preliminary Results: USGS Investigation of Site 45 MCRD, Parris Island, April-July, 2007.
PowerPoint presentation given to the Parris Island Environmental Team on July 30, 2007.
Vroblesky, D. 2008. Progress Report for US Geological Survey FY 2007 Activities and Work plan for FY 2008
Field Activities at Site 45, Marine Corps Recruit Depot, Parris Island, South Carolina. U.S. Geological
Survey, 720 Gracern Road, Suite 29, Columbia, SC, 29210-7651.
Vroblesky, D.A., Petkewich, M.D., Landmeyer, J.E., and Lowery, M.A., 2009, Source, transport, and fate of
groundwater contamination at Site 45, Marine Corps Recruit Depot, Parris Island, South Carolina:
U.S. Geological Survey Scientific Investigations Report 2009-5161, 80 p.
Vu, K.B., M.D. Kaminski, and L. Nunez. Review of Technologies for Contaminated Groundwater.
ANL-CMT-03/2Argonne Natl. Laboratories, Argonne, IL 60439.
Xu, X., and Thompson, N.R. 2009. A long-term bench-scale investigation of permanganate consumption by
aquifer materials. J. Contam. Hydrol. 110:73-86.
Yan, X., and G.E. Schwartz. 1999. Oxidative degradation and kinetics of chlorinated ethylenes by potassium
permanganate. J. Contam. Hydrol. 37, 343-365.
100

-------
Wkim* JHPV/7 i4fl
Mi
B^l f V H	i^H7 y	J
Appendices
A.	Oxidant injection pressure calculations	102
B.	Detailed description of injection equipment (schematics, manufacturers, part
numbers, costs)				104
-
C.	Photographic compendium of ISCO activities at the site 45 ISCO demonstration
project (Parris Island, MCRD, SC)			109
D.	Pre-oxidation (baseline) soil core analytical results for total CVOCs	131
E.	Huling, S.G., Ross, R.R. and Meeker Prestbo, K. 2017. In situ chemical
oxidation: permanganate oxidant volume design considerations.
Ground Water Monit. Remed. (37)1, Spring				 153
F.	Recommended Ground Water Sampling Plan for PI MCRD Site 45	163

-------
Appendix A
Oxidant Injection Pressure Calculations - Role of Hydrostatic and
Overburden Pressure
(Los Angeles Regional Water Quality Control Board - In Situ Remediation
Reagents Injection Working Group, 2009)
The rate that an aquifer can accept fluids and the lateral migration of these fluids
before reaching structural failure is significantly influenced by the vertical acceptance
rate. Maximum injection pressure can be estimated by the density of the dry soil and
saturated soil, the thickness of the vadose zone, and the height of the saturated zone
above the injection point using the following equation (Eqn. A.l):
Pmax = KPdry 8 hdry + Psat 8 hsat ) " Pwater 8 hsat 1 PS' (OI" dynes/cm2) (Eqn. A.l)
Where:


P
max
Pressure maximum

Pdry
Density dry soil - vadose zone

Psat
Density saturated soil

g
Gravitational acceleration

h
dry
Height dry or thickness of vadose zone above
the injection point
hsat
Height saturated of saturated zone above the
injection point
O
¦water
Density water

psi
Pounds per square inch (lbs/in2)

cm2
Centimeters squared

It is important to note, there are several units conversions that make these
calculations clearer:
dyne = g-cm/s2 so the units of g-cm/cm2-s2 = dyne/cm2;
1 cm H20 x 980 = dyne/cm2;
dynes/cm2 x 0.1 = pascals,
and psi x 6.89 = kPa (or psi = kPa / 6.89).
102

-------
It is recommended that for injection applications a 60 percent safety factor be applied to the maximum
calculated pressure as part of the derivation of Pinjection (Payne, 2008). As fluids are injected into an aquifer
the pressure applied to deliver these fluids is expressed upward against the effective hydraulic conductivity
and the downward gravitational force of the water mound. Commonly the vertical hydraulic conductivity
of many aquifers is approximately 10 percent of horizontal hydraulic conductivity and can be used as the
effective hydraulic conductivity. The vertical acceptance is then determined by the relationship between
pressure and the effective hydraulic conductivity as the vertical mounding expands. The following equation
can be used to express this relationship between effective hydraulic conductivity and vertical mounding:
Q/A = K (P. . -p t g h)/h (Eqn. A.2)
effective x injection "water 0 "	x '	'
Where:
Q/A the flow rate applied over the area of the expanding mound.
Vertical flow ceases as the mound height (h) reaches the pressure limit or the
selected "not to exceed" injection pressure (Payne, 2008).
K „ . vertical hydraulic conductivity
effective	'	'
^injection inJection pressure (60% of the allowable injection pressure)
p , Density of water
r water	*
g Gravitational acceleration
h mound height above water table
Example Calculations for site 45 at the MCRD (Parris Island, SC)
A spreadsheet was used to calculate the injection pressures that the aquifer could accept fluids before
reaching structural failure. The following parameter values were used in these calculations:
1.5 g/cm3 (usually 1.1 - 1.6 g/cm3, higher values for sandy material)
1.9 g/cm3 (using a pdry = 1.5 g/cm3; a particle density = 2.65 g/cm3;
total porosity (r|) was estimated (r| = l-(pBULK/ppD) and assumed to be saturated
with water
981 cm/s2
91.4 cm
214-336 cm (8-10, 10-12, 12-14 ft bgs injection intervals)
1 g/cm3
The maximum injection pressures were 4.8, 5.6, and 6.4 lbs/in2 for the 8-10, 10-12, and 12-14 ft bgs
intervals, respectively.
Pdry
psat
h
dry
hsat
103

-------
Appendix B
Oxidant Injection System
This section provides the details of the oxidant injection system used in the ISCO demonstration at site 45,
Parris Island, MCRD, SC. It consists of a detailed schematic of the oxidant injection pallet (Figure B-l) and
a detailed description of injection pallet components, component manufacturers, part numbers, cost, and
relevant description details (Table B-l). Other miscellaneous and/or optional equipment and supplies are
aiso provided that were used for backup contingencies or health and safety matters. The estimated cost
to construct the injection pallet involving one injection arm was approximately $15,400, and the cost per
injection arm (i.e., components 15-27 in Figure B-l) was $2860.
Figure B-l. Oxidant injection schematic. Refer to Table B-l for complete listing of system
components, manufacturer, part number, cost, and description.
Note: the oxidant in jection line is "threaded"
through 3-5 GeoProbe rods to achieve the
appropriate injection depth intervals.
(Not to scale)
HD—
Injection arms 1, 2 and 3
(arms 2 and 3 missing for clarity)
104

-------
Table B-l. Injection system details (components, manufacturer, part numbers, cost, and
description). Miscellaneous and optional supplies and equipment associated with oxidant
injection system and activities are included. Oxidant injection system components follow the
component number identified in the Oxidant Injection System Schematic (Figure B-l, above).
"poly" and "SS" refer to polypropylene and 316 stainless steel, respectively. Blue Monster pipe
compound was used to join and seal threaded parts. Screw clamps were used to join tubing.
Component
Manufacturer
Part number
Cost
Description
1. RemOx-L57 lb Jerrican (pail)
Carus Chemical
2195-110-500
$143/pail
40% sodium



($2.50/1 b NaMn04)
permanganate.
2. PTFE tubing (50 ft)
Chemfluor, or
T-96000-28
$399/50 ft
5/8" O.D. Tubing either

Cole Parmer
T-06605-46
$162/25 ft
fits inside of Phar-Med




tubing and serves as




the oxidant transfer




line, or serves as a




pressure relief return




line to oxidant tank.
3. Phar-Med pump tubing (25 ft)
Cole Parmer
T-06508-82
$226
Phar-Med pump tubing




for peristaltic pump w/




0.5" I.D.
4. Peristaltic pump
Cole Parmer
S-77410-10
$1886
Oxidant transfer
l/P variable speed, brushless



peristaltic pump w/
process drive

S-77600-82
$824
Phar-Med corrosion
l/P high performance pump



resistant tubing.
head




5. PTFE tank (150gal) and rod
GeoProbe
207104
$2979
150 gal corrosion
rack pallet



resistant tank on




skid mounted metal




pallet. Outlet 2" F-NPT




polypropylene.
6. Between tank and hose
Grainger, Banjo


The 1.5" poly ball valve
2"-1.5" poly reducer M-NPT

1MKF1
$3.70
opens the tank allowing
1.5" poly ball valve F-NPT

1MKK7
$75
oxidant flow to the
1.5" poly male adapter M-NPT

1DPN1
$3.70
arms.
1.5" poly nipple M-NPT

1MJZ8
$4.40

1.5" poly Tee F-NPT

1MKH3
$12.60

Right side



The right side of Tee
1.5"-1" poly bushing

1MKC3
$3.90
involves a check valve
1"- %" poly bushing

1MKB8
$3.20
leading to a water hose
%" poly in-line ball valve

3ELV6
$18.60
adapter for cleaning
ZA" poly nipple NPT

1MJZ5
$1.70
the lines and arms.
%" hose to pipe adapter

4 KG 88
$5.00

Left side




1.5" poly male adapter M-NPT

1DPN1
$3.70
The left side leads to




the pump inlet.

-------
Table B-l (continued). Injection system details (components, manufacturer, part numbers, cost,
and description).
Component
Manufacturer
Part number
Cost
Description
7. Pureflex Ultraflex PTFE (1)
convoluted hose with polypro
braid (1.5"x2') and quick connect:
one end SS JIC w/ SS 1.5" F-NPT;
other end SS JIC w/ SS 1.5" F-NPT
Corrosion Fluid
Products Corp.
(hose)
Custom
$563
Corrosion resistant
hose from tank to
pump.
8. 1.5"xl.5"SS nipple M-NPT
on both ends of 1.5"xl.5" SS
coupling
Grainger
1XBC2
$14.60

9. 1.5" SS coupling F-NPT
Grainger
6JK17
$27.30

10. Neptune 7000 series metering
pump; SS dosing head, check
valves, and teflon diaphragm, 300-
gph @ 30-psi performance w/ 1%
NaMn04 solution; 1-HP, 115/230-
volt, 1-phase motor.
Phoenix Pumps
7250-N3-
100753
$4768
Corrosion (oxidant)
resistant pump. Inlet is
1/5" SS M-NPT/check
valve; outlet is 1.5" SS
M-NPT/check valve.
11. 1.5"xl.5" SS nipple M-NPT
on both ends of 1.5"xl.5" SS
coupling
Grainger
6JK17
1XBC2
¦

12. Pureflex Ultraflex PTFE (1)
convoluted hose (1.5"x4') w/ poly
braid: one end SS JIC w/ SS 1.5"
Female-NPT; other end SS JIC with
SS 1.5" Female NPT
Corrosion Fluid
Products Corp.
(hose)
Grainger
Custom
$683
Corrosion resistant
hose from pump to
header.
13. 1.5" SS male adapter (quick
coupler) SS F-NPT (cam/groove)
1.5"x4" SS nipple M-NPT
Festenal
Zoro
0400659
G2961147
$36.80
$26.40
Quick connect between
hose and 3-way cross.
14. SS 1.5" F-NPT threaded cross
1.5"xl"SS bushing M-NPT
Grainger
1LVF2
1LUR7
$82.30
$20.80
45° or 90° depending
on the angle for arm 1,
2, or 3.
15. l"x4"SS nipple M-NPT
SS elbow (45° or 90°) F-NPT
Zoro
Grainger
G2812022
6JK43
$20.20
$15.30

16. 1" poly F-coupler x M-NPT
1" poly M-adapter quick couple
1" Tee FPT poly F-NPT
Banjo
Banjo
Grainger
1DPJ9
1DPK4
1MKH1
$7.60
$3.60
$7.50

106

-------
Table B-l (continued). Injection system details (components, manufacturer, part numbers, cost,
and description).
Component | Manufacturer | Part number | Cost | Description
17. Flow dampener
4" PVC end cap
4"x6" long PVC schedule 40
4" to 2" PVC reducer
2" to 1" PVC bushing
1" PVC male adapter
Grainger
5WPW6
WKJ4
5WPN9
22FJ17
$5.30
$10.90
$1.90
$0.60
Provides flow
dampening due to
pump surge action.
18. Pressure relief valve system
l"-3/4" poly reducing bushing
M-NPTxF-NPT
%" 150 psia poly relief valve w SS
pressure gauge
1" poly Tee F-NPT to
3/8" barb fitting for return line
%" poly Tee F-NPT
Grainger
Valworx
Sprayer Depot
Sprayer Depot
Grainger
1MKB7
36472
2312034PP
36472
3XVP2
$3.20
$45
$26.40
$45
$6.70
Measures delivery
pressure and pressure
relief valve that directs
oxidant back to oxidant
tank as a safety feature.
19. %" Pressure control valve
1" poly nipple M-NPT
1" poly diaphragm valve, 2-way
1"- %" poly reducing bushing
Grainger
1MJZ6
33Z902
1MKB8
$2.3
$143.
$3.2

Controls pressure from
pump to downhole
pressure.
20. Injection pressure system
%"-l" poly reducer M-NPT
%" poly Tee F-NPT
1/4" poly reducing bushing
SS pressure gauge
Grainger
Grainger
Grainger
Valworx
1MJZ5
1MKG9
1MKB3
36472
$1.70
$5.60
$3.10
$45
Measures the
downhole injection
pressure.
21. %"x6" poly pipe nipple
Grainger
3DTJ7
$2.80
A 10" nipple at this
location would be
better for flow meter
operation.
22. %" flowmeter/totalizer
Grainger
1XPR8
$708
Measure flow and
volume (totalizer)
in each arm of the
injection system.
23. 1"-%" SS hex bushing F-NPT/
M-NPT
1" SS ball valve w/ handle option
1"- %" SS coupling to %" bevel
hose fitting M-NPT
Grainger
Zoro Tools Inc.
Grainger
1LTF1
WMY4
$7.30
$44
Flow control valve for
quick shut-off.
24. Pureflex Ultraflex PTFE (1)
convoluted hose with polypro
braid (%"x 45') SS JIC ends w/ %"
SS male NPT adapters
Corrosion Fluid
Products Corp.
(hose)
Custom
$1308
Corrosion resistant
hose from injection
header to oxidant
injection tip.

-------
Table B-l (continued). Injection system details (components, manufacturer, part numbers, cost,
and description).
Component | Manufacturer | Part number | Cost | Description
25. Drive cap

Custom made

Drive cap allows the
GeoProbe to drive
the injection string
(rods + injection tip,
i.e., oxidant hose is
threaded through
rods.)
26. 2.25" x 4' GeoProbe drive rod
(GeoProbe, Salina, KS)
GeoProbe
204766
$139
GeoProbe rod is used
to drive the injection
tip; Oxidant delivery
hose is threaded inside
the rod.
27. 2.25"x2' top-down injection
tool. Removable injection tip head
(GeoProbe threads) allows to
clean out injection tip
Environmental
Services
Products
INJP225-2
$405
Oxidant injection tip
and drive head.
28. Electric cord for 115V
Hardware store

o
*—1
-oo-
Electric cord did not
come with the pump.
Other Miscellaneous and/or Optional Equipment and Supplies
Component | Manufacturer | Part number
Cost
Description
Alternative oxidant injection tip
2.25"x2' retractable injection tool
Environmental
Services
Products
INJP7K-2
$602
Retractable oxidant
injection tip.
Spill contingency
Pig hazard, materials adsorb sock
New Pig
(Tipton, PA)
124CR
(3"x46")
$139
Absorptive socks.
Oxidant neutralizer
Ascorbic acid (bulk, 55 lbs)
Nextag Inc. /
My Spice Sage

$212
Food-grade oxidant
neutralizer.
Micro-well screen
21" SS screen; 3/16' hose bar
GeoProbe
AT-8717S
$65
Stainless steel micro-
well screen.
Spare parts kit
Includes spare parts for Phoenix
7000 series pump
Pump Locker
004334 1
$1028
Field contingencies
Pump.
Extra oxidant injection hose
Pureflex Multiflex PTFE
convoluted hose with polypro
braid (%"x 60') SS JIC ends w/SS
%" male NPT adapters
Corrosion Fluid
Products
Custom
$1889
A 60' length allows
greater flexibility in
depth and location w/o
moving the injection
pallet.
111 Hose is full vacuum rated; operating pressures 1.5" 250 psi; %" 400 psi.
108

-------
Appendix C
US. MARINE CORPS RECRUIT DEPO
EASTERN RECRUITING REGION
PARRIS ISLAND m SOUTH CAROI VA
COMMANDING GENERAL Sgr ScHGEAN' AJOfi
BGEN L E. REVNOLDS	SGTMAJ 6. W BUC
Photographic compendium of ISCO activities
at the site 45 ISCO demonstration project
(Parris Island, MCRD, SC).
Photo 1. Front gate of the Parris Island
Marine Corps Recruit Depot at Parris Island, SC.
Photo 2. Site 45 looking west from the middle of Kyushu Street, and the new Dry Cleaner building (Bidg 192)
is on the right.
109

-------
Photo 3. Site 45 looking upgradient from the middle of Kyushu Street towards the new Dry Cleaner building. The approximate
centerline of the plume is from the near corner of the dry cleaner building and extends in the southeastern direction.
110

-------
Photo 4. Looking north from the southeast corner of the new Dry Cleaner building and laundry pick-up drive-through. The source
area of the plume is the general area in the southeast corner of the building (left side of drive-through).
111

-------
Photo 5. Utiiity notification prior
to subsurface activity at Site 45
involved detection of utilities and
marking. Marked utilities included
a high pressure 8 inch water main
(blue), communication line (orange),
high voltage (230V) power line (red),
storm and sanitary sewers (green). The
approximate mid-line of the CVOCs
plume was marked by the research staff
(yellow).
Photo 6. Retrieving soil cores using the GeoProbe equipment along transects extending across (perpendicular) the
ground water plume.
112

-------
Photo 7. Soil core collection using a GeoProbe (background), and soil core processing (capping,
cutting, sealing, and freezing) for shipment to RSKERC facility for analysis.
Photo 8. Preparation of cemerit-bentonite grout using GeoProbe grout pump. The grout was
pumped into the bottom of boreholes where soil cores were collected. A tremie tube and a
"bottom-up" method was used to properly seal abandoned soil coring locations.

-------
Photo 9. Preparation and staging of materials used in the installation of the micro-wells.

Photo 10. Close-up of the micro-well screen and tubing.

-------
Photo 11. Installing the micro-well well screen and tubing by threading it through the Geoprobe rod to the target depth

-------
Photos 12-13. Closeup of the installation of
the micro-well well screen and tubing through
the Geoprobe rod. The GeoProbe rod and
tubing is filled with water to prevent heaving
and sand-bindirig of the tubing inside the rod.
The tubing and screen are held in place as the
GeoProbe rod is retrieved.
116

-------
Photo 15. Looking upgradient from the middle of Kyushu Street towards the new Dry Cleaner building. The approximate
centerline of the plume (yellow) is from the corner of the dry cleaner building and extends in the southeastern direction.
The open, flush-mounted micro-wells (M3 mid shallow and M3 mid deep) are shown in the immediate foreground.
117

-------
Photos 16-17. Post-installation development of micro-wells was carried out by pumping
ground water at a slow rate. A few liters of ground water was generally required to "develop"
the micro-wells until stabilized genera! parameters and clear ground water could be
achieved.
118

-------
Photo 18. Storage of 5 gal pails (jerricans) of 40%
liquid sodium permanganate in the hazardous waste
storage and containment unit.
Photo 19. The injection tool with holes (left) for oxidant delivery into the subsurface is connected with corrosion-resistant
stainless steel fittings. The corrosion-resistant, oxidant delivery line (blue) is Pureflex UItraflex PTFE (45-60 ft in length) and
is a convoluted hose with polypro braid.
119

-------
Photo 20. The oxidant
injection line is strung
through each of the GeoProbe
rods needed to achieve the
target depth at each location.
The direct push injection tip,
GeoProbe rod, and injection
hose connected prior to
injection.
I -w-—-*•~«»»
Photo 21. The oxidant injection system under a rain/shade canopy with 2 injection arms and secondary
containment.
120

-------
Photo 22. Three (3) oxidant injection arms in use at site 45. Each is equipped with flow dampener, pressure gauge, pressure
relief valve and return line, flow control valve, pressure gauge, in-line flow totalizer, and on the end is the (not shown)
emergency shut-off valve.
121

-------
Photo 23. The oxidant injection system delivering oxidant to 3 injection locations
122

-------
Photo 24. Using a peristaltic pump to transfer a specific volume of 40% liquid sodium permanganate into
the mixing tank. Water was introduced into the mixing tank at high flow rates to assure mixing of the sodium
permanganate and to achieve the targeted oxidant solution concentration needed for injection. A secondary
containment system was constructed under the oxidant injection system.
123

-------
Photo 25. Access restrictions along Kyushu Street while oxidant handling and injection activities were ongoing.
124

-------
Photo 26. Oxidant injection at one location in the drive-through to the dry cleaner.
Photo 27. Simultaneous injection of oxidant into three locations on Kyushu Street while avoiding subsurface
utilities. Facing the west, these injection locations straddle the communication line and flank the south side
of the 8" high pressure water main and 230V high voltage line.

-------
Photo 28. Rain/shade canopy used
for injection pallet while oxidant is
delivered to three locations.
Photo 29. Direct-push at an angle of the injection tip to access the subsurface corner of building 192 in the drive-through
of the dry cleaner. This technique was used to avoid the "mat apron" concrete foundation extending 5 ft from the building.
126

-------
Photo 30. Direct-push of oxidant under corner of the Bidg. 192 into the source area associated with the former leaking
sanitary sewer line in this area. This was performed while avoiding the communication lines (orange-marked utility box is the
communication line), and the overhead steam line supported by the green post in the background of the photo.
127

-------
Photos 31-32. Pumping oxidant into well MW 25 and Inj. 1-Deep at approximately 2-4% NaMnO

-------
Photo 33-34. Decontamination of GeoProbe rig prior to departure; capture of investigation derived
wastewater for subsequent testing and disposal.
129

-------
Photo 35. Ground water sampling at micro-wells involved "binary mixtures" of ground water and the oxidant solution that
was injected as seen by the pink-purple color in the tubing. Ground water samples exhibiting presence of NaMnO.; were
neutralized using ascorbic acid.
Photo 36. The color of the sodium permanganate solutions at different concentrations were prepared to
provide a field "quick test" to approximate permanganate concentrations.
130

-------
Appendix D
Pre- and post-oxidation soil CVOCs concentrations
Pre-oxidation (baseline) and post-oxidation soil core analytical results for total CVOCs. Pre-oxidation soil core
transects are located approximately 16.5 ft (TO), 32 ft (Tl), 65 ft (T2), 98 ft (T3), 131 ft (T4), 164 ft (T5), and
197 ft (T6) downgradient from the X,Y ordinate at MW 25-SL. The X-Y coordinate system used in the pre- and
post-oxidation soii core iocations is depicted in Figure D.l below. There were three core locations (A, B, C) on
each pre-oxidation soil core transect T0-T6. Core B was on the longitudinal axis of the plume, and cores A and
C were located approximately 20 ft on the south and north sides, respectively. The vertical distribution of pre-
oxidation total CVOCs at Transects T0-T6 are presented (Figures D2-D5).
Bldg. 192
New Dry
Cleaner Building
00
3
o
01
>
Kyushu
North
Figure inset: soil core layout
for soil transects T0-T6,
20 ft
20 ft
'r*4
9

X dir
	 Sanitary sewer
	Sanitary sewer (decommissioned)
Sanitary sewer (manhole cover)
	 Fence
O	USGS monitoring well (MW25)
O	Micro-well transect locations(Ml-M4)
—** X, Y ordinatesystem
Foundation/footing
Ground water flow direction
_ Core transecti oca tionsT0-T6
(see inset)
Figure Dl. The X, Y-coordinate system used to illustrate the relative locations and distances between
monitoring well MW 25-SL, micro-well transects M1-M4, and soil core location transects T0-T6
specified in the following figures.
131

-------
-Q-T3-A
-C-T3-B
-A-T3-C
4 6 8 10
Concentration (mg/kg)
-D-T4-A
-0-T4-B
-A-T4-C
4 6 S 10
Concentration (mg/kg)
w-io
¦c
4 6 8 10
Concentration (mg/kg)
6	8 10
Concentration (mg/kg)
-16 &
-16 QF
o

Figure D2. Vertical distribution of pre-oxidation total VOCs at Transects T1-T4 (refer to Figure
Dl). Transects are located approximately 33 ft (Tl), 66 ft (T2), 99 ft (T3), 132 ft (T4)
downgradient from MW 25-SL.

-------
-0-T5-B
-14
14
8
12
?
4
10
-12
-13
-14
14
0
4
8
10
12
b
Concentration (nig/kg)	Concentration (nig/kg)
Figure D3. Vertical distribution of pre oxidation total VOCs at Transects T5-T6 (refer to Figure
Dl). Transects are located approximately 165 ft (T5) and 198 ft (T6) downgradient from IV1W
25-SL.
133

-------
-4
OQ
-6 -
72.8
~ 2* northsideofMW 25 (0.2)
A 2' southside of MW 25 (0,-2)
Q Inj.-l, shallow (-4, -2)
O Inj.-l, deep (-7, -2)
Based on phase distribution calculations. PCE
DNAPL presence likely when [PCE] > 225 mg/kg
83 31.0
102.7
58.56
A
285.8
118 34.9
261.5
o
338.7
o
172.9
o
328.5
-16
Total VOCs: PCE, TCE, c-U-DCE, VC
-18
384.7
0 50 100 150 200 250 300 350 400
Total [CVOCs] (mg/kg)
Figure D4. Vertical distribution of pre -oxidation total CVOCs in aquifer cores collected in the
source area. Aquifer cores were collected adjacent to MW 25-SL and 7 ft upgradient. when
installing injection wells Inj.-l, Shallow and Deep (refer to Figure 4). Based on a contaminant
mass distribution analysis of the total CVOCs concentration data, PCE DNAPL was likely
present in the aquifer core material when PCE concentrations were greater than 225 mg/kg.

-------
0-
-4 -
[]
-16 fk
-18
-B-
0.0
-B- 2' northside of MW 25 (0, 2)
—©— 2' southside of MW 25 (0, -2)
—TO A
—TO B
— TOC
10.0 20.0 30.0 40.0 50.0 60.0
Total CVOCs Concentration (mg/kg)
Figure D5. Vertical distribution ot pre-oxidation total CVOCs at: Transect TO (reter to Figure D.l
above, and Figure 4). The TO transect is located approximately 16.5 ft downgradient from
MW 25-SL.
The contrast between pre- and post-oxidation vertical distribution of total CVOCs concentrations in soil is
evident by plotting post- oxidation [CVOCs]SOiL with nearby aquifer locations of pre-oxidation [CVOCs]SOIL
(Figures D6-D23). The post- oxidation [CVOCs]SOIL are illustrated in purple and pre-oxidation [CVOCs]SOIL are in
red. The figures represent contrasting CVOCs profiles with increasing distance from the source zone. Sampling
locations close to the ordinate at MW 25-SL and upgradient of the ordinate (0, 0) indicated a strong CVOCs
source strength. The value in parentheses following the (X, Y) coordinate represents the radial distance (feet)
from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report). The bar chart inset
illustrates the average pre- and post-oxidation [CVOCs]SOILfor each core location where the average value was
based on the post-oxidation soil sample depth interval, and the post-oxidation [CVOCs]SOIL bar is purple.
135

-------
¦	(*5,-4)
¦	MW 25 N <0, 2) {7.8")
¦	MW25S(2, -2) {5.4'J
¦	tnj.-l, 5 (-4, -2} (2.2')
¦	Inj.-l, D(-7, -2) (2.8')
-is E3	1	1	1	1	1	1	1
0.0	100,0 200,0 300.0 400.0 500.0 600.0 700.0
Total CVOCsConcentration (mg/kg)
Figure D6. The depth-dependent vertical distribution of post-oxidation ICVOCsW (purple square
symbol) at location (-5, -4) contrasted with nearby pre-oxidation [CVOCs]SD|L (red symbols). The
(-5, -4) coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated in
Figure D1, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report).
The bar chart inset illustrates the average pre- and post-oxidation [CVOCs]SOjL. The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
-4
-7)
-H-MW25N (0, 2) (9.5')
-©- MW 25 S (2, -2) (5.8')
-A-lnj.-l, S (-4,-2) (5.1')
—0-Inj.-l, D (-7, -2) (6.4')
-14 [
-16
100








¦ 1-3.-/J


eg
M
"at
¦ WW ?!» N (D. i) |9.5't


S
¦ MW25S<2. >


5 2 DO



_7
• InJ.-l.SK CU'J










c
¦ Inj. l.Of-7. 21 (6.4')


>



ISO



(S
100



50
^ •3 s'



VV |



-18 0-
0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0 450.0
Total CVOCs Concentration (mg/kg)
Figure D7. The depth-dependent vertical distribution of post-oxidation [CVOCsko-.: (purple square
symbol) at location (-3, -7) contrasted with nearby pre-oxidation [CVOCs]S0|L (red symbols). The
(-3, -7) coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated in
Figure Dl, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report).
The bar chart inset illustrates the average pre- and post-oxidation [CVOCs]SO|L« The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
-a-{-2,1)
-3- MW 25 N {0, 2) (2.2')
-©- MW 25 S (2, -2} (3.6')
—A— Inj.-l, S (-4, -2) (3.6')
-O-Inj.-l, D (-7, -2) (5.8')

300



250
¦<*2, i)


'ox
"m
200
¦	MW 25 N (0.2)(2.2'J
¦	MW 25 S (2, -2) {3.6')


B
¦ Inj.-l,S {-4, -2) {3.6')


O

¦ lnj.-l4 D|-7, -2) (5.8 )


3J
W
O
>
SJ
ISO



"w
,c
H
100




50
0
*3




100.0 150.0 200.0 250.0 300.0 350.0 400.0
Total CVOCsConcentration (nig/kg)
450.0
Figure D8. The depth-dependent vertical distribution of post-oxidation [CVOCs]SOil (purple square
symbol) at location (-2, 1) contrasted with nearby pre oxidation [CVOCs]Soil (red symbols). The
(-2, -1) coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated in
Figure D1, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report).
The bar chart inset illustrates the average pre- and post-oxidation [CVOCs]SOil- The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
o
u
eS
t:
3
C/5
¦o
-4
-6
-10
o
£
£
4>
Cfi
-E
•w
c.
o
O
-12
-14
-16
-18
0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0
Total CVOCs Concentration (mg/kg)
400.0 450.0
~~B" (0, -5)
-B-MW25N (0, 2) (7')
MW 25 S {2, -2) (3.0')
Inj.-l, S (-4, -2) (5.0')
Inj.-l, D (-7, -2) (7.6')
Figure D9. The depth-dependent vertical distribution of post-oxidation [CV0Cs]S0|L (purple square
symbol) ) at location (0, -5) contrasted with nearby pre-oxidation I CVOCsL % (red symbols). The
(0, -5) coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated in
Figure Dl, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report).
The bar chart inset illustrates the average pre- and post-oxidation rCVQOsB-m, The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
-4
c -10
-B-(5,1)
-B-MW25N (0, 2) (5.1')
-©- MW 25 S (2, -2) (5.8')
-A- Inj.-l, S (-4, -2) (9.5')
-0- Inj.-l, D (-7, -2) {12.4')
—5K-T0-B (16.5, 0) (11.5')
¦	15.11
¦	MW 25 N(0. 2)15.1')
¦	MW 25S(2,-2| (S.8-|
¦	Inj.-l, 5 (-4, -2) (9.5'|
¦	Inj.-l, E (-7,-21(12.4')
TO-B (16.S, 0) (H-S'l
50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0
Total CVOCsConcentration (mg/kg)
450.0
140
Figure D10. The depth-dependent vertical distribution of post-oxidation [CVOCsMi (purple square
symbol) at location (5, 1) contrasted with nearby pre oxidation [CVOCs]SOJL (red symbols). The (5, 1)
coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated in Figure D.l,
above). The value in parentheses following the (X, Y) coordinate represents the radial distance (feet)
from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report). The bar chart
inset illustrates the average pre- and post-oxidation [CVOCs]SO|L. The average value for each core
location was based on the post-oxidation soil sample depth interval.

-------
I", -3>
TO-A (16.5,-IS)(14.1')
TO-B (16.5,0) (6.3')
¦ MW25N(Q, 2) (12,1)
Hi MW 25 5 (2, -2) (11.0')
= -10
¦g "12 E]
0.0
-B-(ll,-3)
-B-T0-A (16.5, -16) (14.1')
-0-TO-B (16.5, 0) (6.3')
-A- MW 25 N (0, 2) {12.1')
MW 25 S (2, -2) (11.0')
10.0	20.0	30.0	40.0	50.0	60.0
Total CVOCsConcentration (mg/kg)
70.0
Figure Dll. The depth-dependent vertical distribution of post-oxidation [CV0Cs]SOil (purple square
symbol) at location (11, -3) contrasted with nearby pre-oxidation [CVOCs]SOIL (red symbols). The
(11, -3) coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated in
Figure Dl, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report).
The bar chart inset illustrates the average pre- and post-oxidation [CVOCs]SQ)L. The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
*TG-A(16.5. -16) <10.5 )
Total CVOCs Concentration (mg/kg)
142
Figure D12, The depth-dependent vertical distribution of post-oxidation [CVOCs]SOiU (purple square
symbol) at location (11.2, -25) contrasted with nearby pre-oxidation [CVOCskr, (red symbols). The
(11.2, -25) coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated
in Figure Dl, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report).
The bar chart inset illustrates the average pre- and post-oxidation [CVOCs]S0jL. The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
-4
§J -8
3
CZ2
¦o
5 -10
o
O
it
— -12

-------
e-{34,5)
¦Q—TO-B (16.5, 0) (18.2')
¦e-T0-C(16.5,16.2) (20.8')
-A-Tl-B (32,0) (5.4')
2.0	4.0	6.0	8.0	10.0	12.0	14.0
Total CVOCs Concentration (mg/kg)
Figure D14. The depth-dependent vertical distribution of post-oxidation [CVOCs]Soil (purple square
symbol) at location (34, 5) contrasted with nearby pre oxidation [CVOCs]SoIl (red symbols). The
(34, 5) coordinates are based on the ordinate (i.e., (0, 0) located at MW 25-SL as illustrated in
Figure Dl, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report).
The bar chart inset illustrates the average pre- and post oxidation [CVOCs]SOil- The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
(40,0)
Tl-A(32,-20} (21,5')
TIB (32,0) (8.0')
Tl-C (32,20) |21.5')
-0- (40, 0)
-B-Tl-A (32, -20) (21.5')
-6-T1-B (32, 0) (8.0')
-A-Tl-C (32, 20) (21.5')
o.o
2.0	4.0	6.0	8.0	10.0	12.0
Total CVOCs Concentration (mg/kg)
14.0
Figure D15. The depth-dependent vertical distribution of post-oxidation [CV0Cs]SDil (purple square
symbol) at location (40, 0) contrasted with nearby pre-oxidation [CVOCs]SOil (red symbols). The
(40, 0) coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated in
Figure D.l, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report).
The bar chart inset illustrates the average pre- and post-oxidation [CVOCs]SO|L. The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------

-16 0
-18	
0.0	2.0	4.0	6.0	8.0	10.0	12.0	14.0
Total CVOCs Concentration (mg/kg)
Figure D16. The depth-dependent vertical distribution of post-oxidation [CVOCs]Soil (purple square
symbol) at location (44.5, -6.8) contrasted with nearby pre-oxidation [CVOCsfct (red symbols). The
(44.5, -6.8) coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated
in Figure D.l, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report),
The bar chart inset illustrates the average pre- and post-oxidation [CVOCs]SO|L. The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
¦ Tl-B (32,0) (16.6')
Figure D17, The depth-dependent vertical distribution of post-oxidation [CVOCs]SDil (purple square
symbol) at location (48,5, -2) contrasted with nearby pre-oxidation [CV0Cs]SOil (red symbols). The
(48.5, -2) coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated in
Figure D.l, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report)
The bar chart inset illustrates the average pre- and post-oxidation [CVOCs]so;r The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
¦ T2-C165,17) (18.0')
148
Figure D18. The depth -dependent vertical distribution of post-oxidation [CVOCs]Soil (purple square
symbol) at location (52.5, 4) contrasted with nearby pre oxidaiion [CVOCs]SQ|L (red symbols). The
(52,5, 4) coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated in
Figure Dl, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report).
The bar chart inset illustrates the average pre - and post-oxidation [CVOCs]Soil- The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
e- (62, 0)	|
o
a
Q-T2-B (65,0) (3.0') g
U
2
0-12-0(65,17) (17.3') -
c -10
¦	(62.0)
¦	T2-B(65,OJ|3.[y]
¦	T2-C (6S, 17) (173')
-16 0
[]
-18
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0
Total CVOCs Concentration (mg/kg)
9.0
149
Figure D19. The depth-dependent vertical distribution of post-oxidation I CVOCs W (purple square
symbol) at location (62, 0) contrasted with nearby pre-oxidation [CVOCs]SOil (red symbols). The (62,
0) coordinates are based on the ordinate (I.e., (0, 0) located at MW 25-SL as illustrated in Figure
Dl, above). The value in parentheses following the (X, Y) coordinate represents the radial distance
(feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report). The bar
chart inset illustrates the average pre- and post-oxidation [CVOCs]Si.;... The average value for each
core location was based on the post-oxidation soil sample depth interval.

-------
B-(67, 4.5)
0-T2-B (65, 0} (4.9')
e-T2-C(65,17) (12.7')
= -10
3f
E
(67.4.5)
t2-8(IS5,0|(4.9'|
T2-CI65, t7||12.7'l
16 o
[]
-18
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0
Total CVOCs Concentration (mg/kg)
9.0
Figure D20. The depth-dependent vertical distribution of post-oxidation iCVOCsI .-, (purple square
symbol) at location (67, 4.5) contrasted with nearby pre-oxidation [CVOCs]SQ|L (red symbols). The
(67, 4.5) coordinates are based on the ordinate (i.e., (0, 0) located at MW 25-SL as illustrated in
Figure D1, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report).
The bar chart inset illustrates the average pre- and post-oxidation [CVOCs]SO|L. The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
Total CVOCs Concentration (mg/kg)
Figure D21. The depth-dependent vertical distribution of post-oxidation [CVOCs|Si,r (purple square
symbol) at location (71.5, 0) contrasted with nearby pre oxidation [CVOCs]SO|L (red symbols). The
(71.5, 0) coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated in
Figure Dl, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report).
The bar chart inset illustrates the average pre- and post-oxidation rcVQCskm, The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
e-(73, 4.5)
-B-T2-B (65, 0) (9.2')
-©-T2-C (65,17) (14.8')
¦ <73,45)
• T2-B (65, 0) {9.2'J
¦T2-C{65, 17) (14 S')
-18 		1	1						1	
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0
Total CVOCs Concentration (nig/kg)
Figure D22. The depth-dependent vertical distribution of post-oxidation [CVOCs]SO|L (purple square
symbol) at location (73, 4.5) contrasted with nearby pre-oxidation [CVOCs]SQ|L (red symbols). The
(73, 4.5) coordinates are based on the ordinate {i.e., (0, 0) located at MW 25-SL as illustrated in
Figure Dl, above). The value in parentheses following the (X, Y) coordinate represents the radial
distance (feet) from the post-oxidation soil sample location (as per Eqn 3, section 4.6 of the report).
The bar chart inset illustrates the average pre- and post oxidation [CVOCs]SOil- The average value for
each core location was based on the post-oxidation soil sample depth interval.

-------
Appendix E
Publication involving ISCO oxidant volume design considerations
Huling, S.G., Ross, R.R, and Meeker Prestbo, K. 2017. In situ chemical oxidation: permanganate oxidant
volume design considerations. Ground Water Monit. Remed. (37)1, Spring.
Groundwater	NGWA
Monitoring&Remediation vHssc
Explore this journal >
Technical Note
In Situ Chemical Oxidation: Permanganate Oxidant
Volume Design Considerations
by Scott G. Huling, Randall R. Ross, Kimberly Meeker Prestbo
First published: 19 January 2017
DOl: 10.1111/gwmr.12195
Article impact statement: Two approaches in designing the volume of oxidant to inject for ISCO are critically analyzed and tielp to clarify the
advantages and limitations.
153

-------
MonTtoring&Remedlation	Technical Note
In Situ Chemical Oxidation: Permanganate
Oxidant Volume Design Considerations
by Scott G. Huling, Randall R. Ross, and Kimberly Meeker Prestbo
Abstract
Contaminant rebound and low contaminant removal are reported more frequently with in situ chemical oxidation than other in situ tech-
nologies. Although there are multiple causes for these results, a critical analysis indicates that low oxidant volume delivery is a key issue. The
volume of oxidant injected is critical and porosity of the aquifer matrix can be used to estimate the pore volume. The total porosity (0T) is the
volume of voids relative to the total volume of aquifer material. The mobile porosity (6>J is the fraction of voids that readily contributes to fluid
displacement, and is less than 6J leading to smaller estimates of oxidant volume. Injecting low-oxidant volume may result in inadequate oxidant
distribution and postinjection dispersal within the radius of influence, insufficient oxidant contact and oxidant loading, and incomplete treat-
ment; whereas, greater oxidant volume achieves a greater oxidant footprint and may involve risk that the injected oxidant may migrate into
nontarget areas and displacement of contaminated groundwater. Design guidelines and recommendations are provided that could help achieve
more effective technology deployment, reduce the role of heterogeneities in the subsurface, and result in greater probability the oxidant is
delivered to the targeted treatment zone.
Introduction and Background Information
The U.S. Environmental Protection Agency (US EPA)
Superfund remedial program continues to select in situ
chemical oxidation (ISCO) as one of the most frequently
selected in situ treatment technologies (US EPA 2014).
These trends in technology selection indicate the need for
the continued development of ISCO, a technology that has
the ability to transform contaminants in the subsurface
while minimizing the use of fossil fuel energy, chemicals,
and enviromnental impact (Siegrist et al. 2001,2011; Huling
and Pivetz 2006). A review of ISCO design and performance
was performed involving 242 case studies including 83 sites
where permanganate was used for the remediation of chloro-
ethanes (Krembs et al. 2010,2011). The median reduction in
contaminant concentration in groundwater using permanga-
nate was 51% with contaminant rebound observed at 78% of
the sites. In another review of ISCO case studies, a median
reduction in total chlorinated volatile organic compounds
(CVOCs) concentrations of 72% was observed at 12 ISCO
sites, but rebound was more prevalent relative to bioremedia-
Article impact statement Two approaches in designing the
volume of oxidant to inject for ISCO are critically analyzed and
help to clarify the advantages and limitations.
Published 2017. This article is a U.S. Government work and is in
the public domain in the USA.
doi: 10.1111/gwmr. 12195
tion, thermal, or solvent/cosolvent treatment (McGuire et al.
2006). In another remediation survey at dense nonaqueous
phase liquids sites, the occurrence of rebound was reported
to be more prevalent at ISCO sites compared to sites imple-
menting other technologies (GeoSyntec Consultants 2004).
The consistency in the conclusions from these surveys
regarding ISCO treatment performance possibly suggests a
systematic cause and effect in technology design, deploy-
ment, and performance results. The cause of contaminant
rebound and low contaminant removal reported in these
cases is attributed to multiple causes and mechanisms.
Here, we propose that low oxidant volume delivery is a
key issue. The objectives of this manuscript are to contrast
and critically analyze two methods used in estimating the
volume of oxidant to inject in the targeted treatment zone
(TTZ), clarify the impact of this important design param-
eter, and recommend injection design options that limit
the role of heterogeneities and its negative impact on oxi-
dant distribution. Several forms of oxidant can be used in
ISCO but the focus of this study is potassium and sodium
permanganate (KMn04 and NaMn04). A detailed descrip-
tion of the fundamentals of ISCO using permanganate is
described elsewhere (Siegrist et al. 2001, 2011; Petri et al.
2011). ISCO is often used as a source reduction remedy in
TTZs where efficient oxidation can be achieved, and is not
generally applied over the entire footprint of the ground-
water plume. Therefore, the TTZ defined here refers to
the contaminated volume of porous media in the source
area requiring oxidative treatment to achieve the treatment
objective.
NGWA.org
Groundwater Monitoring & Remediation 1

-------
Importance of Oxidant Volume
The delivery of a sufficient volume of oxidant is required
to achieve adequate coverage of the oxidant in the TTZ. The
oxidant volume must contain sufficient oxidant mass to
achieve the oxidant loading and treatment objectives. The
combination of both oxidant volume and oxidant concen-
tration (i.e., oxidant loading) is required to address both
oxidant distribution in the TTZ, and to target contaminant
and noncontaminant oxidant losses including the natural
oxidant demand (NOD) (Mumford et al. 2005; Urynowicz
et al. 2008; Xu and Thomson 2009; Cha et al. 2012). The
focus of this critical analysis is on the immediate contact
between injected oxidant and aquifer media. The postinjec-
tion oxidant transport in the downgradient direction could
result in greater contact between oxidant and aquifer solids,
and is dependent on hydrogeologic conditions and oxidant
loading. For example, the aquifer volume contact efficiency
(Ev) is an empirical measure of the oxidant impact on aqui-
fer solids relative to the total volume of the treatment zone
(Cha and Borden 2012). In an evaluation of aquifer char-
acteristics and ISCO injection system design numerical
simulations of oxidant injection, transport, and reaction in
porous media indicate that increases in Ev are functionally
dependent on oxidant loading, persistence, and advective
transport. Simulation results indicated that the mass and
volume of permanganate injected had the greatest impact on
Ev and contaminant treatment efficiency (Cha and Borden
2012). These parameters were also identified by others as a
key aspect for ISCO design (Bachiochi et al. 2014).
Estimating Pore Volume
Porosity is a design parameter used to estimate the pore
volume (PV) within the radius of influence (ROI). The total
porosity (#T) of unconsolidated porous media is the volume
of voids (I\.) relative to the total volume (FT) of aquifer
material (Or = Vv/VT). In unconsolidated porous media, the
volume of oxidant injected into the subsurface required
to fill the pore spaces within the ROI can be estimated as
0T x Froi (i.e., PV), ideally where FR0I is the total volume
within the ROI and the vertical interval.
A fraction of the water in porous media is attracted to
the surfaces of the solids through forces of molecular attrac-
tion and is functionally dependent on the surface area of the
sediment minerals (Marsily 1986). Unconnected, poorly con-
nected, and dead-end pores are responsible for the fraction of
water in porous media that do not contribute to fluid displace-
ment. Thus the concept of porosity is expanded to include
effective porosity which is linked to the displacement of pore
fluid rather than to the percentage of the volume occupied by
the pore spaces. Payne et al. (2008) reported that a fraction of
the total porosity that contributes to advective flow and trans-
port of groundwater in aquifers is the mobile porosity (0,J.
and the portion of the void space that does not contribute
to the advective flow of groundwater behaves as immobile
or slowly moving groundwater is the immobile porosity (^).
The total porosity is the sum of mobile and immobile poros-
ity (0.^ = (), +()). Decisions between the selection of mobile
porosity and total porosity in designing the volume of oxi-
dant to be injected into the TTZ is an important distinction
and can have major implications in ISCO.
Critical Analysis of Oxidant Volume Estimation
Methods
Mobile Porosity Method
Payne et al. (2008) reported that standard aquifer test-
ing protocols obtain the average hydraulic conductivity
(K) which combines the high and low K, and consequently
understates the actual flow velocities. This implies that the
breakthrough of a solute in groundwater, whether it is a
tracer, contaminant, or oxidant will occur faster than pre-
dicted if based on the average K. The "actual groundwater
velocity," determined by tracer studies and used to esti-
mate the mobile velocity, is significantly greater than the
average velocity (Payne et al. 2008). Tracer study design,
implementation, and data interpretation can be complex
and very few sites yield the "ideal" tracer distributions
that allow easy analysis (Payne et al. 2008). Complexities
include tracer selection, tracer injection rates, monitoring
network, tracer detection and analysis, retardation, reactive
transport, and tracer test interpretation including variabil-
ity in tracer breakthrough due to localized aquifer hetero-
geneities, poor or skewed recovery, depth-integration, and
other factors (Ptak et al. 2004; Payne et al. 2008; Suthersan
et al. 2014).
A summary of tracer test results (n = 15) conducted in
various aquifer materials yielded estimated values of mobile
porosity ranging from 0.0008 to 0.18 (Table 1) (Payne et al.
2008). The mobile porosity of sand and gravel aquifers was
estimated to be less than 0.1, and it was suggested that mobile
porosity values ranging from 0.02 to 0.10 would be more
appropriate than using the often recommended 0.20 value as
the effective porosity (Payne et al. 2008). In another summary
of 9U values estimated through tracer tests in alluvial aquifers
(;? = 73), the 9U was <0.09 at 50% of the sites, and <0.15 for
about 80% of the sites (Suthersan et al. 2014). It was con-
cluded that a small portion of the total pore space meaning-
fully participated in flow and advective solute transport.
Site-specific tracer tests which are used to quantify 9U
involve measuring the volume of tracer injected (Vol.mj50)
to reach 50% of the maximum observed breakthrough con-
centration (Cmax/2) in the dose response (i.e., monitoring)
wells (Equation 1; Figure 1) (/? = vertical injection interval).
Ideally, the CMAX/2 should be nearly half the injected tracer
concentration (CQ) however, low tracer injection volume,
long distances between injection and dose response wells,
and structural failure of the aquifer matrix (Payne et al.
2008) contributes to low tracer recovery (i.e., QAx/2
-------
Table 1
Mobile Porosity Values Determined by Tracer Tests, and
Total Porosity and Density of Aquifer Materials
Materials
Mobile Porosity
(LJ/L3)
Location/Aquifer
Poorly sorted
sand/gravel
0.085
Quebec, Canada
Poorly sorted
sand/gravel
0.04 to 0.07
Central Valley, CA
Poorly sorted
sand/gravel
0.09
North TX, Ogalalla
Fractured
sandstone
0.001 to 0.007
N.T, Passaic
formation
Alluvial
formation
0.0102
Los Angeles, CA/
Gaspur
Glacial outwash
0.145
Northern N.T
Weathered
mudstone regolith
0.07 to 0.10
Northern MO
Alluvial
formation
0.07
Sao Paulo, Brazil
Alluvial
formation
0.07
Phoenix, AZ
Silty sand
0.05
Savanah R., SC
Fractured
limestone
0.0008 to 0.001
Trifels formation
Alluvium sand/
gravel
0.017
West TX
Alluvial poorly
sorted sand/gravel
0.003 to 0.007
North TX/Ogalalla
Alluvial sand/
gravel
0.11-0.18
Central CO/Cherry
Creek
Siltstone,
sandstone,
mudstone
0.01-0.05
Central CO/Denver
Formation
Materials
Total Porosity
(M/M) (0T)
Soil Density
(Pbulk) (kg71113)
Gravel
0.24 to 0.38 3
2,000 to 2,350 4
Coarse sand
0.31 to 0.46 3
1,400 to 1,900 4
Fine sand
0.26 to 0.53 3
1,400 to 1,900 3
Silt
0.34 to 0.61 3
1,300 to 1,920 4
Clay
0.34 to 0.60 3
600 to 1,800 4
Glacial tills
0.20 4
1,700 to 2,300 4
Silts and clays
(inorganic)
0.29 to 0.52 5
600 to 1,800 4
Silts and clays
(organic)
0.66 to 0.75 3
500 to 1,500 4
Peat
0.60 to 0.80 6
100 to 300 4
Notes: Total porosities reflect the typical range for each material considering
compaction and sorting. Soil bulk density (pBXJLK) refers to the common ranges of
density for unsaturated conditions and various degrees of compaction.
'Percentage (%) values reported by Payne et al. (2008) were changed to units of
length (L3/L3) for contrast to total porosity values.
2Payne et al. (2008).
3Zheng and Bennet (1995).
4Perloff and Baron (1976).
5Holtz and Kovacs (1981).
6Bear (1988).
is intrinsically extended from one distinct layer of limited
scale, across the vertical injection (i.e., screened) interval
involving varying geologic and hydrogeologic character-
istics exhibiting a range in permeability (Figure 1). Under
this condition, it is proposed that estimates of mobile poros-
ity correspond to cross-sections of the formation with the
highest permeability and the immobile porosity with the
lowest permeability. Consequently, extrapolation of mobile
porosity concepts across lengthy vertical intervals intro-
duces complexity in assessing the specific role of porosity
in groundwater and solute transport as it invites the role of
other parameters that impact groundwater transport, includ-
ing differing depositional processes and materials.
It has been proposed that 6U can be used to determine the
injection volume to achieve adequate reagent coverage at a
given radial distance from an injection well (Suthersan et al.
2014). Assuming a simplified radial flow-cylindrical porous
media conceptual model, the 0, v the vertical injection inter-
val, and the ROI, the volume of oxidant (f oxidant »m) can be
estimated (Equation 3) (Payne et al. 2008).
K
OXIDANT,t)M
= n rop e h
(3)
Tracer Testing, Mobile Porosity, and Contaminant Distribution
The most permeable and highly conductive aquifer
media characterized through tracer testing is predominantly
responsible for estimates of 0,v but may not correspond
with the majority of contamination. Rigorous site charac-
terization is needed to validate and establish a correlation
between contaminated intervals in the TTZ and the intervals
involved in tracer transport. Slow, steady diffusion of con-
taminants into low permeable materials may have occurred
over decades accounting for significant contaminant storage
relative to more permeable portions of the aquifer (Chapman
and Parker 2005; Stroo et al. 2012) represented by mobile
porosity. Specifically, "immobile pores" involving either
lower permeability lenses or layers at the macro-scale, or
"immobile pores" in which groundwater transport is slow
or impeded at the micro-scale may not contribute to tracer
transport, but may be highly contaminated.
Total Porosity Method
Total porosity measurements of aquifer media using labo-
ratory methods (Danielsonand Sutherland 1986) are relatively
simple and low cost, or field geophysical methods (resistivity,
neutron and gamma-gamma radiation) can be used (Marsily
1986). Total porosity can also be calculated from site-specific
measurements or estimates of the bulk density (pBULK) and
particle density (pPD) of aquifer material (Equation 4). Values
of total porosity tabulated from main measurements over a
wide range of aquifer media (Table 1), and that exhibit simi-
lar compositional characteristics to the TTZ can be used for
oxidant volume design. Assuming the simplified radial flow-
cylindrical porous media conceptual model, total porosity, the
vertical injection interval, and the ROI, the volume of oxidant
(^ oxidantot) can be estimated (Equation 5). This method does
not differentiate between mobile and immobile pores.
^t 1 PbvuJ Pm
f OXIDANTS =" ROP 0t/7
(4)
(5)
NGWA.org
S.G. Huling et al./ Groundwater Monitoring & Remediation 3

-------
Length (L)
E
Tracer detection (CMAX) Tracer injection (C0)
V V
K2 » K1 ~ K3 > K4
Screen length (h)
Coarse sand
Silty sand
Figure 1. Schematic of tracer breakthrough in a heterogeneous porous media aquifer and mobile porosity (0M) calculations. The
tracer volume (Vol. ) is measured where 50% breakthrough of the maximum tracer concentration (CMAX;2) occurs (refer to Equa-
tions 1 and 2).
In this method, it is assumed the injected oxidant is even-
tually dispersed throughout the ROIPV. Consequently, long-
term oxidant persistence and dispersal into the spectrum of
lesser connected pores is critical. Oxidant dispersal involv-
ing the groundwater velocity continuum between mobile and
immobile pores is dependent on different reactions, advec-
tion and diffusive transport mechanisms, ISCO design, and
site conditions. In application of these principles to ISCO,
soluble chemicals exchange between zones of mobile and
immobile porosity. Preferential flow in structured media lias
been described using a variety of models (Simunek et al.
2003). Dual-porosity and dual-permeability models both
assume that the porous medium consists of two interacting
regions where solute exchange in groundwater, for example,
between mobile pores and immobile pores, occurs through a
rate-limiting diffusion process (Russo 2012). In some cases,
oxidant persistence is insufficient, either due to reaction or
transport, to effectively penetrate porous media through dif-
fusion (Goldstein et al. 2004) suggesting that the eventual
dispersal assumption may be invalid under some conditions.
Oxidant Delivery Strategies to Improve ISCO Effectiveness
Relative to long injection screened intervals, short injec-
tion screened intervals have a lower probability of being
screened across lithologic layers exhibiting a wide range in
hydraulic conductivity. Therefore, short injection intervals
limit the risk in delivering a disproportionate volume of oxi-
dant into higher permeability layers and greater probability
of injecting the oxidant into discrete zones within the TTZ
(Figure 2). Where feasible, the length of a well screen in an
injection well should be no more than 10 to 15 ft (3.05 to
4.57 m), particularly in heterogeneous formations and where
treating highly contaminated source zones (Simkins et al.
2011). Shorter injection well screens (<10 to 15ft) (<3.05
to 4.57m), and direct-push injection using short injection
intervals (2 to 4 ft) (0.61 to 1.22 m) can further limit the role
of preferential pathways. Overall, a combination of short-
screened injection intervals, a greater number of injection
locations, and smaller ROI's per injection location reduces
the risk of delivering excessive oxidant volumes into prefer-
ential pathways (Figure 2) and results in greater probability
that the oxidant is delivered to the TTZ. Areal coverage of
the TTZ with multiple, overlapping oxidant ROIs is criti-
cal to improve contact between oxidant and contaminated
media.
Oxidant breakout into higher permeability zones,
mounding, surfacing, and transport beyond the TTZ in gen-
eral, can result from excessive injection pressure. The rate
that an aquifer can accept fluids and the lateral migration
of these fluids before reaching structural failure is signifi-
cantly influenced by the vertical acceptance rate. Maximum
injection pressure can be estimated by the density of the dry
soil and saturated soil, the thickness of the vadose zone, and
the height of the saturated zone above the injection point
(Los Angeles Regional Water Quality Control Board - In
Situ Remediation Reagents Injection Working Group 2009).
Adhering to these injection guidelines will improve oxidant
delivery to the TTZ.
Assuming the contamination is mainly in the most per-
meable zones as defined by tracer tests, the mobile poros-
ity oxidant volume design could be an effective approach.
However, assuming contamination is present in both high
and low permeability materials suggests that oxidant injec-
tion into high permeability layers limits oxidant delivery into
4 S.G. Huling et al./ Groundwater Monitoring & Remediation
NGWA.org

-------
(A)
NaMn04 Injection
/
K2 » K1 ~ K3 > K4

-



~





Sand ^1

Coarse sand ^2
9
B Sand K3

| Silty sand K4
(B)
A
NaMnQ,
/
K2 » K1 ~ K3 > K4
K1
K2
Sand
K3
Silty sand
(C)
(D)
/
K2 » K1 - K3 > K4

—



~





Sand


Coarse sand ^2


B Sand K3

Silty sand
K2 » K1 - K3 > K4
e sand
Figure 2. Schematic of idealized conceptual model illustrating the role of heterogeneities, and the impact of screen length, radius of
influence, and injection spacing on the distribution of the permanganate oxidant. (A) The long well screen in heterogeneous lithology
accounts for disproportionate oxidant distribution; (B to D) alternatively, advancing short, direct-push injection well screens, using
small radii of influence and close injection locations, achieves more effective oxidant distribution (oxidant density effects excluded).
low permeability TTZs, and/or misses the target altogether.
Chemical oxidation of contaminants in, or near, low perme-
ability materials requires that the oxidant be delivered into
or near lower permeability zones and persist for timeframes
consistent with steep concentration gradients and diffusive
transport (Cavanagh et al. 2014). Periodic batch delivery of
oxidant could be used to address the back diffusion of con-
taminants. Alternatively, slow but constant oxidant delivery
methods have been developed for sites with low conductiv-
ity lithology involving gravity delivery and/or constant head
injection designs (Pac et al. 2014; Lulirs et al. 2015).
Critical Analysis of Mobile and Total Porosity Oxidant
Injection Volume Estimation Methods
Contrasting tabulated values of 9U and 0T (Table 1; Equa-
tions 3 and 5) indicate that a larger volume of oxidant is
estimated using total porosity vs. mobile porosity. Consider
the illustrative example of an oxidant volume ISCO design
where assumed values for 9U (0.10) and 0T (0.35) are used
to estimate oxidant volume and where ROI= 10ft (3.05 m),
and /? = 5ft (1.52 m). The volume of oxidant required to
achieve the ROI across the screened interval using 0 and
0T is 1170gal (4428L) and 4110gal (15.556L). respectively.
Delivering a larger volume of oxidant associated with 0T
translates into greater oxidant handling, preparation, and
labor resulting in higher remedial cost.
The potential for groundwater displacement and oxidant
transport beyond the design ROI and TTZ into nontargeted
zones are risks that should be evaluated. Since ISCO is usu-
ally deployed in a source area, oxidant transport beyond
the ROI into lesser-contaminated aquifer conditions, and/
or downgradient transport during the postinjection drift
phase could also occur. In these cases, chemical oxida-
tion of contaminants would continue to occur, potentially
in nontargeted zones with less contamination. Despite the
seemingly benign consequence of oxidant transport beyond
the immediate ROI in this case, adjustments to the ISCO
design would need to consider the loss of oxidant from
the ROI. Nearby receptors beyond the TTZ and potential
discharge areas (i.e., seeps, creeks, water bodies, utility
corridors, etc.) should be identified to determine whether
migration of oxidant beyond the contaminated TTZ could
occur and what steps, if any, to implement given potential
contingencies.
NGWA.org
S.G. Huling et alj Groundwater Monitoring & Remediation 5

-------
Contaminants are distributed between aqueous, solid,
and nonaqueous phases in the subsurface and can be esti-
mated using equilibrium partitioning calculations (Feen-
stra et al. 1991; Newell and Ross 1992; Cohen and Mercer
1993). Under non-aqueous phase liquid (NAPL)- or heavily
contaminated source area conditions, the contaminant mass
may be present in all three phases but the majority of con-
taminant mass is generally found in the NAPL and solid
phases relative to the dissolved aqueous phase contaminant
mass. Under NAPL-free conditions, based on the hydropho-
bicity of CVOCs and the presence of natural organic mat-
ter in aquifer material, the majority of contaminant mass is
adsorbed on the solid phase material. Assuming the major-
ity of the CVOCs mass in the ROI is immobilized either as
residual NAPL or through adsorption onto aquifer solids,
the displacement of the dissolved CVOC contaminant mass
in the groundwater is projected to be limited relative to the
total contaminant mass.
Critical Analysis of Permanganate ISCO Design
Parameters
Krembs et al. (2010, 2011) compiled permanganate
ISCO design parameters including the design and observed
ROI (ft), oxidant dosage (g/kg), number of PVs delivered,
number of delivery events, and duration of delivery events
(days) (Table 2). A critical analysis of these design param-
eters reported is preceded by the caveat that the number of
case studies for the median value of each design parameter
varied, and therefore only general observations are possible.
Radius of Influence
The ROI can be estimated by measuring appreciable
concentrations of the oxidant in monitoring wells located
in different directions from the injection location. When the
observed ROI (25 ft) (n = 11) (7.62 m) is considerably greater
than the design ROI (14ft) (;? = 29) (4.27 m) (Table 2), this
suggests that oxidant distribution was more extensive than
designed. However, given these ROIs, the volume of oxi-
dant required to achieve a 25-ft ROI (7.62 m), relative to
Table 2
A Compilation of the Median Value for ISCO Design
Parameters from Field Application Case Studies
(Krembs et al. 2010,2011)
Design Parameter
Median Value Reported
Krembs et al. (2010, 2011)1
Design ROI (ft)
14 (h = 29)
Observed ROI (ft)
25 (h= 11)
Oxidant dosage (g/kg)
0.4 (h = 24)
Number of pore volumes
delivered
0.16 (h = 32)
Number of delivery events
2 (h = 65)
Duration of delivery events (days)
4 (h = 45)
Vertical injection interval (ft)
NA2
'The median design value and number of case study sites is reported.
2The vertical injection interval was not reported Krembs et al. (2010, 2011).
a 14-ft ROI (4.27 m), would involve injecting greater than
three times more oxidant volume. A firm explanation can-
not be provided regarding this anomalous difference but
some speculation is warranted. For example, groundwater
quality parameters (i.e., ORP, DO, temperature, conductiv-
ity) are sometimes used as indirect indicators of oxidant
ROI. Spikes in these parameter values can occur in ground-
water after the oxidant has fully reacted, yet the effects, or
the residuals of the oxidant, can still be measured yielding
false-positive results for the ROI.
In some cases, the hydraulic conductivity (K) profile
measured across screened intervals in groundwater wells
exhibits an order of magnitude change over short vertical
sections of the screened interval (Zlotnik and Zurbuchen
2003). Under this condition a disproportionate volume of
oxidant could be delivered into high K layers (Figure 1),
and could also yield an ROI artifact. Injection and monitor-
ing well screens representative of discrete vertical intervals
would be useful to diagnose this condition and to obtain a
more accurate assessment of oxidant transport and ROI.
Due to heterogeneities in aquifer hydraulic properties,
ROIs on the order of 25 ft (7.62 m) require the injection of a
large volume of oxidant which invites vulnerability to dis-
proportionate transport of the oxidant in preferential path-
ways. Decreased ROIs translate into smaller injection well
spacing and would include installation of additional injec-
tion wells or more direct-push injection locations. Smaller
ROIs have several advantages including lower probability
that preferential pathways will play a role in oxidant trans-
port, greater potential for hydraulic control, greater accuracy
in the spatial emplacement of the oxidant, and greater confi-
dence that the oxidant can be delivered to the designed ROI.
Natural Oxidant Demand
ISCO oxidant loading, defined here as mass of oxidant
per mass of soil (i.e., g oxidant/kg aquifer material) is some-
times designed based on NOD values measured in bench-
scale studies. The NOD for permanganate often involves
a 48-h test procedure (ASTM 2007). This batch test does
not differentiate between mobile or immobile pores. Judi-
cious interpretation of test results is warranted given that
the reaction between Mn04~ and reactants varies with time,
there are multiple reactive components that exhibit varying
reactivity, the NOD may increase substantially with longer
testing periods, and Mn04~ reaction is concentration depen-
dent (Mumford et al. 2005; Honning et al. 2007; Urynow-
icz 2008; Urynowicz et al. 2008; Xu and Thomson 2008,
2009; Cha et al. 2012). Most of the NOD in soil and aquifer
samples is slow reacting, and 48-h NOD measurements are
poor predictors of total NOD and cannot accurately estimate
long-term Mn04~ consumption (Cha et al. 2012). However,
an alternative, quick and economical permanganate chemi-
cal oxidant demand test has been developed to estimate the
maximum permanganate NOD for aquifer materials (Xu
and Thomson 2008). Further, results from a broad range
in aquifer materials indicated excellent linear relation-
ship between the maximum NOD and the 7-day NOD test
indicating that results could be used to support perman-
ganate ISCO site screening and design (Xu and Thomson
2009). Once the fast NOD fraction is rapidly consumed, the
6 S.G. Huling et alj Groundwater Monitoring & Remediation
NGWA.org

-------
remaining MnO, may persist for weeks to months, and dif-
fuse into lower permeability zones where contaminants may
reside. Samples of aquifer solids from 12 sites (;? = 50) in the
U.S. were analyzed for NOD over a long period (up to 41
days). In 80% of the samples, a broad range in NOD (0.24 to
18.8g IVTnO! /kg soil; median value = 3.33 g/kg) was mea-
sured and the overall range was 0.2 to 150 g/kg (Cha et al.
2012). This was similar to previously reported NOD ranges
(Mumford et al. 2005; Huling and Pivetz 2006; Honning
et al. 2007; Urynowicz 2008; Xu and Thomson 2009). The
median value of IVT nO, dosage reported by Krembs et al.
(2010, 2011) is low (0.4 g/kg; ;? = 24) (Table 2) relative to
long-term NOD values. Assuming that long-term persis-
tence of MnO, was needed to address contaminant mass
transport and mass transfer limiting processes, this low
IVT nO, dosage applied suggests that the median range value
used at many ISCO sites represents an under-design either
in terms of oxidant volume or concentration. Overall, this
low median value of oxidant dosage would unlikely persist
long enough to broadly disperse within the ROI and TTZ
while satisfying both the fast- and slow-acting NOD.
Pore Volume
A PV represents the volume of voids contained within
the ROI spatially defined by the ROI, vertical interval, and
porosity. The median number of permanganate PVs delivered
(PV=0.16; n=32 sites) (Table 2) (Krembs et al. 2010, 2011)
is significantly less than a single PV (PV= 1.0) and suggests
an under-design of oxidant volume. In this case, significant
postinjection oxidant persistence and dispersal would be
required to achieve both (1) coverage in the remaining 0.84
PV of the ROI where oxidant was not initially delivered and
(2) to address long-term NOD requirements. Postinjection
oxidant dispersal of this magnitude is unlikely.
Summary
The median values reported for oxidant dosage (0.4 g/
kg) and PV delivery (PV=0.16) (Table 2) (Krembs et al.
2010, 2011) suggests a less aggressive, low oxidant vol-
ume ISCO design which may be consistent with a mobile
porosity ISCO design. This design appears likely to result
in incomplete oxidant delivery, insufficient oxidant dosage
within the ROI, and short duration oxidant persistence. A
more aggressive ISCO design based on total porosity would
involve a greater oxidant dosage (>3 g/kg) (i.e., Cha et al.
2012) and PV delivery (PV= 1) permitting greater distri-
bution and longer duration oxidant persistence. It is noted
that low oxidant concentration could have contributed to the
low oxidant dosage, but oxidant concentration data were not
reported by Krembs et al. (2010, 2011), and could not be
evaluated.
Other Causes of Rebound and Low Contaminant
Removal
Current ISCO deployment is increasingly reflective of
an iterative design involving repeated oxidant injections
with intermittent monitoring diagnostics used to guide
subsequent oxidant injections that target persistent, high
NGWA.org
concentration source areas. Further, depletion of the "more
accessible" contamination may be followed by an ISCO
approach characterized by smaller oxidant batch injections,
or long-term, slow oxidant delivery targeting the back dif-
fusion of contamination (Pac et al. 2014; Luhrs et al. 2015).
In the early years of ISCO design, short duration and/or
single injection deployment (i.e.. Table 2, number of deliv-
ery events = 2; n = 65) may not have fully recognized the
technical challenges associated with hydrogeologic hetero-
geneities, oxidant distribution limitations, high contamina-
tion zones, contamination in low permeable or fractured
systems, high oxidant demand zones, limited oxidant per-
sistence, etc. Consequently, rebound and low contaminant
removal through postoxidant injection groundwater moni-
toring would be a probable outcome. Projects involving
these designs may be reflected in ISCO treatment perfor-
mance conclusions in the ISCO surveys (GeoSyntec Con-
sultants 2004; McGuire et al. 2006; Krembs et al. 2010,
2011).
Groundwater samples collected specifically to be ana-
lyzed may contain the injected oxidant and organic contami-
nants in a "binary mixture" (Huling et al. 2011; Johnson et
al. 2012; Ko et al. 2012). Oxidative transformation of con-
taminants in the sample after it is collected, causes false-
negative results. Upon complete reaction of the injected
oxidant in the subsurface, the potential for false-negative
results is eliminated and the contaminants present in
groundwater samples are subsequently detected and quanti-
fied. Essentially this condition contributes to a determina-
tion of "rebound" although the contaminants may have been
present all along. Given the relatively recent development
in groundwater sample preservation guidelines (Ko et al.
2012),	the interpretation of groundwater quality under this
condition is likely reflected in the case study survey results
and statistics that report rebounding contaminant concentra-
tions at ISCO sites.
Summary
Critical reviews of in situ remediation technology treat-
ment performance indicate that ISCO, compared to other
in situ technologies, exhibits a higher degree of chemical
rebound and a lower reduction in contaminant concentra-
tions (GeoSyntec Consultants 2004; McGuire et al. 2006;
Krembs et al. 2010, 2011). Although several mechanisms
may provide a reasonable explanation for these observa-
tions, low oxidant volume delivery at ISCO sites is likely
to be a key factor for rebound and the low reduction in con-
taminant concentrations.
In conjunction with porosity, a simplified radial flow,
cylindrical, porous media conceptual model is often used to
estimate the volume of oxidant required to achieve a ROI
over a specific vertical interval. Mobile and total poros-
ity values used in these calculations yield a wide range in
estimates of oxidant volume. The mobile porosity method
specifically recognizes that groundwater readily moves in
"mobile pores" and can be measured using tracer tests. How-
ever, tracer tests are vulnerable to lithology-dependent vari-
ability in hydraulic conductivity where tracer transport in the
most permeable sections of the media will skew estimates of
S.G. Huling et al.I Groundwater Monitoring & Remediation 7

-------
mobile porosity downward. Consequently, a smaller volume
of oxidant is estimated resulting in a smaller oxidant foot-
print potentially leaving some TTZs oxidant-free within the
ROI. Total porosity, often calculated or measured in the lab
does not differentiate between mobile and immobile pores.
This method involves a larger oxidant volume to inject, a
larger oxidant footprint, and would likely impact ISCO field
activities due to greater oxidant handling, preparation and
labor. Displacement of contaminated groundwater and oxi-
dant migration beyond the ROI and TTZ are risks that must
be evaluated. In source areas, phase distribution analysis
suggests that the majority of contaminants are immobilized
in soil and NAPL which limits the relative displacement
of contaminants. Several oxidant injection methods and
designs can be used to reduce the impact and risk of pref-
erential pathways on oxidant delivery and provide greater
probability that the oxidant is delivered to the ROI.
Acknowledgments
The authors acknowledge Dr. Kiyoung Cha (National
Research Council, US EPA, Ada, Oklahoma) for technical
input on this manuscript. The U.S. Environmental Protec-
tion Agency, through its Office of Research and develop-
ment, funded and managed the research described here. It
has not been subjected to Agency review and therefore does
not necessarily reflect the views of the Agency, and no offi-
cial endorsement should be inferred.
References
ASTM. 2007. ,4STMD 7262-07. Standard Test Method for Esti-
mating the Permanganate Natural Oxidant Demand of Soil and
Aquifer Solids. West Conshohocken, Pennsylvania: ASTM Inter-
national, 04.09.
Bachiochi, R., L. D'Aprile, I. Innocenti, F. Massetti, and I. Vergi-
nelli. 2014. Development of technical guidelines for the applica-
tion of in-situ chemical oxidation to groundwater remediation.
Journal of Cleaner Production 11, no. 15: 47-55.
Bear, J. 1988. Dynamics of Fluids in Porous Media. New York:
Dover Publications.
Cavanagh, B.A., P.C. Johnson, and E..T. Daniels. 2014. Reduction
of diffusive contaminant emissions from a dissolved source in a
lower permeability layer by sodium persulfate treatment. Envi-
ronmental Science and Technology 48: 14582-14589.
Cha, K.Y., M.L. Crimi, M.A. Urynowicz, and R.C. Borden. 2012.
Kinetics of permanganate consumption by natural oxidant
demand - Modeling approaches and statistical analysis. Environ-
mental Engineering Science 29, no. 7: 646-653.
Cha, K. Y., and R.C. Borden. 2012. Impact of injection system design
on ISCO performance with permanganate — mathematical mod-
eling results. Journal of Contaminant Hydrology' 128: 33—46.
Chapman, S.W., and B.L. Parker. 2005. Plume persistence due to
aquitard back diffusion following dense nonaqueous phase liq-
uid source removal or isolation. Water Resources Research 41:
W12411.
Cohen, R., and J. Mercer. 1993. DNAPL site evaluation. EPA 600/R-
93/022. Washington, DC: U.S. Environmental Protection Agency.
Danielson, R.E., and P.L. Sutherland. 1986. Porosity. InMethods of
Soil Analysis Part 1: Physical and Mineralogical Methods, 2nd.
Monograph 9 ed., ed. A. Klute. Madison, Wisconsin: American
Society of Agronomy/Soil Science Society of America.
Feenstra, S., D.M. Mackay, and J.A. Cherry. 1991. A method for
assessing residual NAPL based on organic chemical concentra-
tions in soil samples. Gmundwater Monitoring Review 11, no.
2: 128-136.
GeoSyntec Consultants. 2004. Assessing the Feasibility> of
DK4PL Source Zone Remediation: Review of Case Studies.
Port Hueneme, California: Naval Facilities Engineering Ser-
vices Center.
Goldstein,K.J., A. Vitolins, D. Navon, B.L. Parker, S. Chapman, and
G.A. Anderson. 2004. Characterization and pilot-scale studies for
chemical oxidation remediation of fractured shale. Remediation
Journal 14, no. 4: 19-38.
Holtz, R.D., and W.D. Kovacs. 1981. An Introduction to Geotechni-
calEngineering. Englewood Cliffs, New Jersey: Prentice Hall.
Honning, J., M.M. Broholm, and PL. Bjerg. 2007. Quantification of
potassium permanganate consumption and PCE oxidation in sub-
surface materials. Journal of Contamination Hydmlogy 90: 221.
Huling, SG. and B. Pivetz. 2006. In-situ chemical oxidation -
Engineering issue. EPA/600/R-06/072. Ada, Oklahoma: US
Environmental Protection Agency, National Risk Management
Research Laboratory, R.S. Kerr Environmental Research Center.
Huling, S.G., S. Ko, and B. Pivetz. 2011. Ground water sampling at
ISCO sites - Binary mixtures of volatile organic compounds and
persulfate. Groundwater Monitoring and Remediation 31, no. 2
Spring: 72-79.
Johnson, K.T., M. Wickham-St. Germain, S. Ko, and SG. Huling.
2012. Binary mixtures of permanganate and chlorinated volatile
organic compounds in groundwater samples: Sample preserva-
tion and analysis. Groundwater Monitoring and Remediation 32,
no. 3 Summer: 84-92.
Ko, S., Huling, S.G., and Pivetz, B. 2012. Ground water sample
preservation at in-situ chemical oxidation sites - Recommended
guidelines, EPA Ground Water Issue Paper. EPA/600/R-12/049.
Ada, OK: US Environmental Protection Agency, National Risk
Management Research Laboratory, R.S. Kerr Environmental
Research Center.
Krembs, F.J., R.L. Siegrist, ML. Crimi, R.F. Furrer, and BG. Petri.
2010. ISCO for ground water remediation: Analysis of field
applications and performance. Groundwater Monitoring and
Remediation 30, no. 4: 42-53.
Krembs, F. J., W.S. Clayton, and M.C. Marley. 2011. Chapter 8 Evalua-
tion of ISCO field applications and performance. In In Situ Chemi-
cal Oxidation for Remediation of Contaminated Gmundwater.
SERDP/ESTCP Remediation Technology Monograph Series, C .H.
Ward (Series ed), ed. R.L. Siegrist, ML. Crimi, and T.J. Simpkin,
319-354. New York: Springer Science and Business Media, LLC.
Los Angeles Regional Water Quality Control Board - In Situ Remedi-
ation Reagents Injection Working Group. 2009. Technical report:
Subsurface injection of in situ remedial reagents within the Los
Angeles Regional Water Quality Control Board Jurisdiction.
http://www.swrcb.ca.gov/rwqcb4/water_issues/programs/ust/
guidelines/ (accessed September 16,2009).
Luhrs,R., S.Andrews, andP. J.Dugan.2015.UseofISCOasacosteffec-
tive long term remedy, http://www.caruscorporation.com/page/
remediation/carus-remediation-newsletters (accessed October
26,2015).
Marsily, G. 1986. Quantitative Hydrogeology. San Diego, CA: Aca-
demic Press, Harcourt Brace Jovanovich.
McGuire, T.M., J.M. McDade, and C.J. Newell. 2006. Performance
of DNAPL source depletion technologies at 59 chlorinated
solvent-impacted sites. Groundwater Monitoring and Remedia-
tion 26, no. 1: 73-84.
Mumford,KG.,N.R. Thomson,andR.M. Allen-King. 2005. Bench-
scale investigation of permanganate natural oxidant demand
kinetics. Environmental Science and Technology 39: 2835.
8 S.G. Huling etal.l Groundwater Monitoring & Remediation
NGWA.org

-------
Newell, C. aiidR.R. Ross. 1992. Estimating potential for occurrence
of DNAPL at superfund sites, OSWER Publication 9355.4-07FS.
Washington, DC.
Pac, T.J., R.W. Lewis, and E.C. Gyles. 2014. Constant head injection
for enhanced in-situ chemical oxidation. Remediation Journal
25: 71-83.
Payne, F.C., J.A. Quinnan, and S.T. Potter. 2008. Remediation
Hydraulics. Boca Raton, Florida: CRC Press; Taylor & Francis
Group LLC, 408 p.
Perloff, W.H., and W. Baron. 1976. Soil Mechanics: Principles and
Applications. New York: John Wiley & Sons.
Petri, B., N.R. Thomson, and M.A. Urynowisz. 2011. Chapter 3 -
Fundamentals of ISCO using permanganate. In In Situ Chemi-
cal Oxidation for Remediation of Contaminated Gmundwater. A
volume in SERDP/ESTCP Remediation Technology Monograph
Series, C.H. Ward (Series ed), ed. R.L. Siegrist, M.L. Crimi, and
T..T. Simpkin, 89-146. New York: Springer Science and Business
Media.
Ptak, T., M. Piepenbrink, and E. Martac. 2004. Tracer tests for the
investigation of heterogeneous porous media and stochastic mod-
elling of flow and transport—A review of some recent develop-
ments. Journal of Hydrology' 294: 122-163.
Russo, D. 2012. Numerical analysis of solute transport in vari-
ably saturated bimodal heterogeneous formations with mobile-
immobile-porosity. Advances in Water Research 47: 31—42.
Simunek, J., N..T. Jarvis, M. van Genuchten, and A. Gardenas.
2003. Review and comparison of models for describing non-
equilibrium and preferential flow and transport in the vadose
zone. Journal of Hydrology 272: 14-35.
Siegrist, R.L., M.A. Urynowicz, O.R. West, M.L. Crimi, and S.L.
Lowe. 2001. Principles and Practices of In Situ Chemical Oxida-
tion Using Permanganate. Columbus, Ohio: Battelle Press, 336
pp.
Siegrist, R.L., M. Crimi, and R.A. Brown. 2011. Chapter 1: In Situ
Chemical Oxidation: Technology description and status. Inln Situ
Chemical Oxidation for Remediation of Contaminated Ground-
water. A volume in SERDP/ESTCP Remediation Technology
Monograph Series, C.H. Ward (Series ed), ed. R.L. Siegrist, M.L.
Crimi, and T..T. Simpkin, 1-32. New York: Springer Science and
Business Media.
Simkins, T.J., T. Palaia,BG. Petri, andB.A. Smith. 2011. Chapter 11
Oxidant delivery approaches and contingency planning. In Situ
Chemical Oxidation for Remediation of Contaminated Ground-
water. A volume in SERDP/ESTCP Remediation Technology
Monograph Series, C.H. Ward (Series ed), ed. R.L. Siegrist, M.L.
Crimi, and T..T. Simpkin, 449^180. New York: Springer Science
and Business Media.
Stroo, H.F., et al. 2012. Chlorinated ethene source remediation: Les-
sons learned. Envimnmental Science and Technology 46, no. 12:
6438-6447.
Suthersan, S., C. Divine, E. Cohen, andK. Heinze. 2014. Tracer test-
ing: Recommended best practice for design and optimization of
in situ remediation systems. Groundwater Monitoring and Reme-
diation 34, no. 3: 33—40.
Urynowicz, M.A. 2008. In situ chemical oxidation with permanga-
nate: Assessing the competitive interactions between target and
nontarget compounds. Soil and Sediment Contamination 17: 53.
Urynowicz, M.A., B. Balu, and U. Udayasankar. 2008. Kinetics of
natural oxidant demand by permanganate in aquifer solids. Jour-
nal of Contaminant Hydrology 96, no. 1^1: 187-194.
US EPA. 2014. Superfund Remedy Report, 14th Ed. EPA 542-R-13-
016. November 2013, Washington, DC: Solid Waste and Emer-
gency Response.
Xu, X., and N.R. Thomson. 2008. Estimation of the maximum con-
sumption of permanganate by aquifer solids using a modified
chemical oxygen demand test. Journal of Envimnmental Engi-
neering 134,no. 5: 353-361.
Xu, X., and N.R. Thomson. 2009. A long-term bench-scale investi-
gation of permanganate consumption by aquifer materials. Jour-
nal of Contaminant Hydrology 110: 73-86.
Zheng, C., and G.D. Bennet. 1995. Applied Contaminant Transport
Modeling. New York: VanNostrandReinhold.
Zlotnik, V.A., and B.R. Zurbuchen. 2003. Estimation of hydraulic
conductivity from borehole flowmeter tests considering head
losses. Journal of Hydrology' 281: 115-128.
Biographical Sketches
S. G. Hilling, corresponding author, is at U.S. Envimnmental
Protection Agency, Office of Research and Development, National
Risk Management Research Laboratory, Robert S. Kerr Environ-
mental Research Center, P.O. Box 1198, Ada, OK 74820; (580)
436-8610; (580) 436-8615; huling.scott(a)epa.gov
R. R. Ross is at U.S. Environmental Protection Agency, Office of
Research and Development, National Risk Management Research
Laboratory, Robert S. Kerr Environmental Research Center, P.O.
Box 1198, Ada, OK 74820.
K. Meeker Prestbo is at U.S. Environmental Protection
Agency, Office of Environmental Cleanup, Region 10, 1200 Sixth
Avenue, Suite 900, Seattle, II. I 98101.
NGWA.org
S.G. Huling et al./ Groundwater Monitoring & Remediation 9

-------
Appendix F
Recommended Ground Water Sampling Plan for PI MCRD Site 45
Background
It is recommended to perform post-oxidation ground water sampling and analysis to continue the
assessment of treatment performance of ISCO at site 45. Recommended guidelines are provided that are
consistent with methods used in previous ground water monitoring events.
Ground Water Sampling
Post-oxidation ground water samples would include 39 wells and micro-wells (Table Fl; Figure Fl). It is
recommended that ground water samples be analyzed for CVOCs, metals and NaMn04 (i.e., Mn04"), and
other optional ground water parameters listed below (i.e., dissolved methane gas, ferrous and total iron,
chloride, sulfate). Prior to ground water sample collection, a YSI multi-parameter probe with a flow through
cell should be used to collect field parameters, including pH, dissolved oxygen, oxidation reduction potential,
temperature, specific conductivity, and turbidity. Ground water samples could then be collected when
field parameters stabilize. Previously, a peristaltic pump (MasterFlex Easy Load II, L/S, model 77200-62;
MasterFlex PharMed-24 tubing; MasterFlex speed Controller) was used and connected to the micro-well and
pumped at 100-300 mL/minute. Similar methods to retrieve a sample from the micro-wells is recommended.
A ground water purge log is included below, and could be used to keep track of readings until the parameters
stabilize. The purge log includes a record of time, temperature (°C), specific conductivity (mS/cm), DO
(mg/L), pH, ORP (mV), and field comments.
Parameters
Method
CVOCs (PCE, TCE, cis-l,2-DCE, trans-l,2-DCE, 1,1-DCE, VC)
EPA Method 8260B GC/MS
Metals
EPA Method 6010C (ICP-AES)
Dissolved Methane Gas
RSKSOP-194/175, Rev. 5
Ferrous and total iron
EPA Method 3500-Fe D
Chloride and sulfate
(1»EPA Method 6500 (refer to RSKSOP-276,
Rev. 4 below).
EPA Method 8260B - Volatile organic compounds by purge and trap GC/MS.
EPA Method 6010C (EPA SW-846) - Inductively coupled argon plasma with atomic emission Spectrometry.
RSKSOP-194, Rev. 4 - Gas Analysis by Micro Gas Chromatograph (see Table 3).
RSKSOP-175, Rev. 5 - Sample Preparation and Calculations for Dissolved Gas Analysis in Water Samples Using a GC
Headspace Equilibration Technique (see Table 3).
EPA Method 3500-Fe D - Phenanthroline method.
RSKSOP-276, Rev. 4 - Determination of Major Anions in Aqueous Samples Using Capillary Ion Electrophoresis with
Indirect UV Detection and Empower 2 Software.
(1)There is no existing EPA method for dissolved methane gas. Refer to Kampbell and Vandegrift, 1998.
163

-------
Table F-l. Micro-well and 2"
sampling event (total wells =
monitoring wells to be sampled during the post-oxidation 3
39).
Micro-well Monitoring Transect


Micro-well Number

Ml
Ml-SB-S
Ml-SB-D
Ml-SA S
Ml-SA D
Ml-Mid-S
Ml-Mid-D
Ml-NA-S
Ml-NA-D
Ml-NB-S
Ml-NB-D
M2
M2-S-S
M2-S-D

M2-Mid-S
M2-Mid-D

M2-N-S
M2-N-D
M3
M3-S-S
M3-S-D

M3-Mid-S
M3-Mid-D

M3-N-S
M3-N-D
M4
M4-S-S
M4-S-D

M4-Mid-S
M4-Mid-D

M4-N-S
M4-N-D
T1

Tl-S-S
Tl-S-D

Tl-N-S
Tl-N-D

T2

T2-S-S
T2-S-D

T2-N-S
T2-N-D

2 inch wells

MW-25, Inj-
1, Deep, MW-31 SL

] Nested micro-well cluster
~ Sentry micro-well cluster
O Y Well
New Dry
Cleaner Facility
Ml
M2
M3
M4
M2-S-S
M2-S-D
Tl-S-S
Tl-S-D
T2-5-S
T2-S-D
Figure Fl. Conceptual layout of 2 inch wells and micro-wells at the U.S. MCRD Parris Island, SC Site 4b.

-------
New Dry
Cleaner Facility
Soil Core (SC) transects (SC-1 thru SC-4)
Approximate locations of
transects M-l thru M-4
Figure F2. Conceptual model ot approximate locations ot soil core and micro-weil transects used
in soil core sample collection.

-------
Sample Handling, Labeling, Packaging and Shipping, and QA/QC
Ground water samples. Ground water samples should be collected in pre-labeled 40 mL vials, preserved
with HCI for CVOCs and HN03 for metals, and stored in ice chests with blue ice. Upon completion, samples
will be transported/shipped overnight to the analytical laboratory (i.e., Shealy Environmental Services,
Columbia, SC), using the appropriate chain of custody forms.
The presence of NaMn04 in a ground water sample can be visually determined as either a pink or purple
color (Johnson et al., 2012). Assuming the presence of Mn04" is ascertained in the ground water sample, it
is recommended to (1) note the presence and the nature of the color in the ground water purge log notes,
(2) collect a sample to measure the NaMn04 concentration via spectrophotometer analysis (i.e., wavelength
= 525 nm), and (3) neutralize the sample with ascorbic acid, as described in Johnson et al. (2012) or Ko et al.
(2012).
Johnson, K.T., Wickham-St. Germain, M., Ko, S. and Huling, S.G. 2012. "Binary Mixtures of Permanganate and
Chlorinated Volatile Organic Compounds in Groundwater Samples: Sample Preservation and Analysis."
Ground Water Monit. Remed. 32(3), Summer 84-92.
Ko, S., Huling, S.G., and Pivetz, B. 2012. Ground Water Sample Preservation at In-Situ Chemical Oxidation Sites-
Recommended Guidelines, EPA Ground Water Issue Paper. US Environmental Protection Agency, National Risk
Management Research Laboratory, R.S. Kerr Environmental Research Center, Ada, OK. EPA/600/R-12/049.
QA/QC. Laboratory, field, equipment, and trip blanks, will be used for aqueous samples analyzed for CVOCs,
metals, CI", S042", Fe+2, and FeT; it is recommended that 10-15% duplicates for aqueous samples should be
collected and analyzed. Method blanks, practical quantitation limits, method detection limits, matrix spikes,
and percent recoveries are to be used and reported by the analytical contractor. The analytical contractor
will issue a case narrative and discuss whether samples adhered to the quality assurance management plan
and SOPs.
Investigation-Derived Waste
Less than 1 drum of contaminated ground water will likely be produced as investigation derived waste
(IDW). The contaminated ground water is produced as a result of the micro-well and 2 in well purging
prior to sample collection. The production, handling, and storage of IDW should be communicated and
coordinated with the Installation Restoration Program Manager, Lisa Donohoe at (843) 228-2779.
Description of Operating Procedures for the YSI MPS 5600
The ground water flow will be diverted through a flow cell equipped with a YSI 5600 multi-parameter
probe. The rate of pumping will be approximately 100-300 mL/minute. The YSI probe is used to track the
stabilization of pH, oxidation-reduction potential (ORP), specific conductance (SC), dissolved oxygen (DO),
and temperature. In general, the following criteria are used to determine when parameters have stabilized:
a pH change of less than or equal to 0.02 units per minute, an oxidation-reduction potential change of
less than or equal to 0.002 V per minute, and a specific conductance change of less than or equal to 1%
per minute. These criteria are initial guidelines; professional judgment in the field is used to determine
on a well-by-well basis when stabilization has occurred. The time-dependent changes in geochemical
parameters recorded by the YSI probe are logged by the handheld instrument and recorded for each well
on the ground water purge log sheet (see below). Once stabilization occurs, the final values for pH, ORP,
specific conductance, dissolved oxygen, and temperature are recorded. After the values for pH, ORP, SC, DO,
and temperature have been recorded, the flow cell is disconnected. For background information, the EPA
NRMRL SOP for field analytical QA/QC (NRMRL-GWERD-23-0) is available.

-------
Parris Island, SC MCRD Site 45 - ISCO Project: Ground Water Purge Log
Well ID:	_ Date:_	
Start Purge Time:	 Initial Purge End Time:_
Purge Rate:	
End of Sampling Time;	
Weather Conditions:	
Time
Temp
(9C)
Specific
Cond.
(mS/cm)
TDS
(g/L)
DO
(mg/L)
PH
ORP
(mV)
Comments
































































































167

-------
SEPA
United States
Environmental Protection
Agency
Office of Research and Development (8101 R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300

-------