United States
Environmental Protection
Agency
Xpert Design and
Diagnostics' (XDD) In Situ
Chemical Oxidation Process
Using Potassium
Permanganate (KMnO4)
Innovative Technology
Evaluation Report
SUPERFUND INNOVATIVE
WHHQUOGY WALUATW
-------
EPA/540/R-07/005
May 2007
Xpert Design and
Diagnostics' (XDD)
In Situ Chemical Oxidation
Process Using Potassium
Permanganate (KMnO4)
Innovative Technology Evaluation Report
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
Notice
The information in this document has been funded by the U.S. Environmental Protection Agency
(USEPA) under Contract Number 68-C-00-179 to Science Applications International Corporation
(SAIC). It has been subjected to the Agency's peer and administrative reviews and has been
approved for publication as an EPA document. Mention of trade names or commercial products does
not constitute an endorsement or recommendation for use.
-------
Foreword
The U.S. Environmental Protection Agency is charged by Congress with protecting the Nation's land,
air, and water resources. Under a mandate of national environmental laws, the Agency strives to
formulate and implement actions leading to a compatible balance between human activities and the
ability of natural systems to support and nurture life. To meet this mandate, EPA's research program
is providing data and technical support for solving environmental problems today and building a
science knowledge base necessary to manage our ecological resources wisely, understand how
pollutants affect our health, and prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory (NRMRL) is the Agency's center for
investigation of technological and management approaches for preventing and reducing risks from
pollution that threatens human health and the environment. The focus of the Laboratory's research
program is on methods and their cost-effectiveness for prevention and control of pollution to air, land,
water, and subsurface resources; protection of water quality in public water systems; remediation of
contaminated sites, sediments and ground water; prevention and control of indoor air pollution; and
restoration of ecosystems. NRMRL collaborates with both public and private sector partners to foster
technologies that reduce the cost of compliance and to anticipate emerging problems. NRMRL's
research provides solutions to environmental problems by: developing and promoting technologies
that protect and improve the environment; advancing scientific and engineering information to support
regulatory and policy decisions; and providing the technical support and information transfer to ensure
implementation of environmental regulations and strategies at the national, state, and community
levels.
This publication has been produced as part of the Laboratory's strategic long-term research plan.
It is published and made available by EPA's Office of Research and Development to assist the user
community and to link researchers with their clients.
Sally C. Gutierrez, Director
National Risk Management Research Laboratory
-------
Abstract
Xpert Design and Diagnostic's (XDD)potassium permanganate in situ chemical oxidation (ISCO)
process was evaluated underthe EPA Superfund Innovative Technology Evaluation (SITE) Program
at the former MEC Building site located in Hudson, New Hampshire. At this site, both soil and
groundwaterare contaminated with chlorinated volatile organic compounds (VOCs). The VOCs are
primarily perchloroethylene (PCE), trichloroethylene (TCE), and cis-1,2-dichloroethylene, (cDCE).
Three saturated stratigraphic zones, occurring between 6 and 25 feet (1.8 to 7.6 m) below land
surface (bis) and within an approximate 1,200 ft2 (111.5 m2) area, were targeted for ISCO treatment.
Little [320 Ib (145 kg)] potassium permanganate was able to be injected into the shallow, gravelly
sandy zone, whereas 1,500 Ibs (680 kg) and 1860 Ibs (845 kg) were injected into the intermediate
peat and deep, silty sand layers, respectively. The average soil concentrations of PCE decreased
by 96 percent and 88.5 percent, in the peat and deep layers, respectively. The average soil TCE
concentrations decreased by 92 and 98 percent, in the peat and deep layers, respectively. However,
cDCE exhibited a no change (+1 percent) and strong increase (+2,570 percent) in the peat and deep
layers. The average final ground water concentrations were 746, 612, and 3,090 ug/L PCE, TCE and
cDCE, respectively, which were below the site specific remediation performance standards of 750,
5,500, 17,500 ug/L. NO chlorinated ethylenes were measurable in samples with visible potassium
permanganate but potassium permanganate was not evenly injected into the target formation.
IV
-------
Contents
Notice ii
Foreword iii
Abstract iv
Tables viii
Figures ix
Abbreviations and Acronyms x
Acknowledgments xiii
Executive Summary ES-1
1.0 Introduction 1
1.1 Background 1
1.2 Brief Description of the SITE Program 4
1.3 The SITE Demonstration Program and Reports 4
1.4 Purpose of the Innovative Technology Evaluation Report (ITER) 4
1.5 Technology Description 5
1.6 Key Contacts 5
2.0 Technology Applications Analysis 6
2.1 Key Features of the XDD ISCO Process 6
2.1.1 Oxidant 6
2.1.2 Portable Oxidant Delivery System 7
2.1.3 Oxidant Injection Points 7
2.1.4 Monitoring Wells 9
2.2 Operability of the Technology 10
2.3 Applicable Wastes 11
2.4 Availability and Transportability of Equipment 11
2.5 Materials Handling Requirements 11
2.6 Site Support Requirements 11
2.7 Limitations of the Technology 12
2.8 ARARS for XDD's ISCO Process 12
2.8.1 CERCLA 14
2.8.2 RCRA 14
2.8.3 CAA 15
2.8.4 CWA 15
2.8.5 SDWA 15
2.8.6 OSHA 15
2.8.7 State and Local Requirements 16
-------
3.0 Economic Analysis 17
3.1 Introduction 17
3.2 Conclusions 20
3.3 Factors Affecting Estimated Cost 20
3.4 Issues and Assumptions 20
3.4.1 Site Characteristics 20
3.4.2 Design and Performance Factors 21
3.4.3 Financial Assumptions 21
3.5 Basis for Economic Analysis 21
3.5.1 Site Preparation 21
3.5.2 Permitting and Regulatory Requirements 22
3.5.3 Capital Equipment 22
3.5.4 Startup and Fixed Costs 23
3.5.5 Labor 24
3.5.6 Consumables and Supplies 26
3.5.7 Utilities 28
3.5.8 Effluent Treatment and Disposal 28
3.5.9 Residuals Shipping and Disposal 28
3.5.10 Analytical Services 29
3.5.11 Maintenance and Modifications 29
3.5.12 Site Restoration 29
4.0 Technology Effectiveness 30
4.1 Introduction 30
4.1.1 Project Background 30
4.1.2 Project Objectives 30
4.2 Site Description 33
4.2.1 Site Location and History 33
4.2.2 Site Lithology and Hydrogeology 33
4.3 Pre-Demonstration Activities 35
4.3.1 Preliminary Sampling 35
4.3.2 MeOH Extraction Studies 35
4.3.3 Treatability Study 36
4.3.4 CPT/MIP Characterization 38
4.4 Demonstration Activities 43
4.4.1 Injection of Oxidant 43
4.4.2 Soil and Groundwater Sample Collection 45
4.4.3 Process Monitoring 47
4.5 Performance and Data Evaluation 49
4.5.1 Soil Results 49
4.5.2 Groundwater Results 58
VI
-------
4.5,3 Quality Assurance/Quality Control 65
4.5.4 Analytical QC Results 69
4.5.5 Audits 72
5.0 Other Technology Requirements 73
5.1 Environmental Regulation Requirements 73
5.2 Personnel Issues 73
5.3 Community Acceptance 74
6.0 Technology Status 75
6.1 Previous Experience 75
6.2 Ability to Scale Up 75
7.0 References 76
VII
-------
Tables
Table Page
ES-1 Summary of Results ES-3
2-1 Properties of KMnO4 7
2-2 Federal and State ARARS for XDD's ISCO Process 13
2-3 Groundwater Cleanup Criteria Comparison 16
3-1 Treatment Design for Hypothetical Site 17
3-2 Cost Estimate for a Full-Scale Application 19
3-3 Estimated Labor Costs 24
3-4 Estimated Miscellaneous Supplies Costs 27
3-5 Estimated Residuals Shipping/Disposal Costs 28
3-6 Estimated Analytical Costs 29
4-1 Demonstration Objectives 32
4-2 Remediation Performance Standards 33
4-3 Injection Summary 44
4-4 Summary of Laboratory Analyses 46
4-5 Shallow Zone Soil Results for PCE, TCE, cDCE, and VC - Baseline Vs Final 50
4-6 Middle Peat Zone Soil Results for PCE, TCE, cDCE, and VC - Baseline Vs Final . ... 51
4-7 Bottom Zone Soil Results for PCE, TCE, cDCE, and VC - Baseline Vs Final 52
4-8 Statistical Summary for Soil Sample Pair Analysis 53
4-9 Averaged Soil Results (Excluding Non-Detects) for PCE, TCE, cDCE, and VC
- Baseline Vs. Final 54
4-10 Averaged Soil Results (Including Non-Detects) for PCE, TCE, cDCE, and VC
- Baseline Vs. Final 55
4-11 Humic Acid Results for Middle Peat Zone - Baseline Vs. Final (wt/wt %) 57
4-12 Groundwater Results for Target Ethenes - Baseline Vs. Final 59
4-13 Shallow Zone Groundwater Results for Target Ethenes - Baseline Vs. Final 60
4-14 Intermediate Zone Groundwater Results for Target Ethenes - Baseline Vs. Final .... 61
4-15 Deep Zone Groundwater Results for Target Ethenes - Baseline Vs. Final 62
4-16 Groundwater Results for Non-Critical VOCs - Baseline Vs. Final 63
4-17 KMnO4 Vs. PCE, TCE, and cDCE Concentrations in Groundwater- All Events 64
4-18 Metals Summary Results for Groundwater (mg/l) 66
4-19 Acetone, Bromide, and Bromoform Summary Results for Groundwater (ug/l) 67
4-20 Project Accuracy for Soil VOC Analyses 68
4-21 Project Accuracy for Groundwater VOC Analyses 69
4-22 Accuracy for Soil VOC - Spike Results - Baseline Event 69
4-23 Precision for Soil VOC - Spike Results - Baseline Event 70
VIII
-------
4-24 Samples Analyzed Outside Holding Times 70
4-25 Accuracy for Soil VOC - Spike Results - Post-treatment Event 70
4-26 Precision for Soil VOC - Spike Results - Post-treatment Event 71
4-27 Accuracy for Groundwater - Post-treatment VOCs 71
4-28 Precision for Groundwater - Post-treatment VOCs 71
Figures
Figure Page
1-1 Former MEC Building Source Area 2
1-2 Time Line of Demonstration Events 3
2-1 Injection Well Cluster Schematics 8
2-2 Multi-Chamber Well Schematic 9
3-1 Hypothetical ISCO Site 18
4-1 Map of Study Area Showing Demonstration Soil Boring and Well Locations 31
4-2 Cross Section Showing Stratigraphic Zones of Study Area 34
4-3 Maximum MIP PID Response for Shallow Zone Soil (0-18 ft) - Baseline Vs. Final .... 39
4-4 Maximum MIP PID Response for Deep Zone Soil (18-36 ft) - Baseline Vs. Final 40
4-5 Maximum MIP ECD Response for Shallow Zone Soil (0-18 ft) - Baseline Vs. Final .... 41
4-6 Maximum MIP ECD Response for Deep Zone Soil (18-36 ft) - Baseline Vs. Final 42
4-7 Map of KMnO4 Injection Points 43
4-8 Estimated Dispersion of KMnO4 104 Days After First Injection 47
4-9 Estimated Dispersion of KMnO4 245 Days After First Injection and 125 Days
After Second Injection 48
4-10 Estimated Dispersion of KMnO4 280 Days After First Injection and 160 Days
After Second Injection 48
IX
-------
AQCR Air Quality Control Regions
AQMD Air Quality Management District
ARARs Applicable or Relevant and Appropriate Requirements
ASTM American Association of Testing and Materials
ALSI Analytical Laboratory Services Inc.
Aries Aries Engineering, Inc.
As Arsenic
bgs Below ground surface
BFB Bromofluorobenzene (tune performance compound for SW 8260)
CA Chloroethane
CAA Clean Air Act
Cl-ethenes Chlorinated ethenes
CERI Center for Environmental Research Information
CERCLA Comprehensive Environmental Response, Compensation, and Liability Act
CFR Code of Federal Regulations
CPT Cone penetrometer technology
CWA Clean Water Act
Cr Chromium
DCA 1,1-dichloroethane
DCE 1,1-dichloroethylene
°C Degrees Celsius
cDCE cis-1,2-dichloroethene
DNAPL Dense non-aqueous phase liquid
DPT Direct push technology
ECD Electron capture detector
EW Evaluation well
EPA U.S. Environmental Protection Agency
ft2 Square feet / square foot
ft3 Cubic feet / cubic foot
Gal Gallons
GC/MS Gas chromatography/mass spectroscopy
g/L Grams per liter
G&A General and administrative
GW-1 Groundwater criteria for NHDES
GMP Groundwater Management Permit
Hg Mercury
H2O2 Hydrogen peroxide
HSWA Hazardous and Solid Waste Amendments
ICAL Initial calibration standard
In Inch
ID Inner diameter
IDW Investigation derived waste
ISCO In situ chemical oxidation
ITER Innovative Technology Evaluation Report
-------
(Cont'd)
IW Injection well
KMnO4 Potassium permanganate
Kg Kilogram
KVA Kilovolt amperes
LLC Limited liability corporation
L Liter
LL Lower limit (of a specified confidence interval)
LCS Laboratory control sample
LRL Laboratory reporting limit
LDR Land disposal restriction
LRPCD Land Remediation and Pollution Control Division
MEC formerly Nashua Electric Motors
MeOH Methanol
MSDS Material Safety Data Sheet
MS/MSD Matrix spike/matrix spike duplicate
MCLs Maximum contaminant levels
MCLGs Maximum contaminant level goals
MIP Membrane interface probe
m Meter
m2 Square meters
MDL Method detection limit
MnO2 Magnesium oxide
MnO4 Permanganate
NH New Hampshire
mg/L Milligrams per liter
ml Milliliters
MV Millivolt
MW Monitoring well
NAAQS National ambient air quality standards
NCP National Oil and Hazardous Substances Pollution Contingency Plan
NPDES National Pollutant Discharge Elimination System
NPDWS National primary drinking water standards
NRMRL-C! National Risk Management Research Laboratory (EPA-Cincinnati)
NSCEP National Service Center for Environmental Publications
NPDWS National primary drinking water standards
NHDES New Hampshire Department of Environmental Services
OSHA Occupational Safety and Health Administration
ORD Office of Research and Development (EPA)
OSWER Office of Solid Waste and Emergency Response
OSC On-scene coordinator
O&M Operation and maintenance
OD Outside diameter
ORP Oxygen reduction potential
PCE Perchloroethene or tetrachloroethene
PID Photoionization detector
PPE Personal protective equipment
PQL Practical quantitation limit
XI
-------
(Cont'd)
PE Professional engineer
PG Professional geologist
PVC Polyvinyl chloride
POD Portable oxidant delivery system
POTW Publicly owned treatment works
QAPP Quality assurance project plan
QA Quality assurance
QC Quality control
Rfs Response factors
RDL Reporting detection limit
RPD Relative percent difference
RSD Relative standard deviation
RPM Remedial project manager
RPS Remediation Performance Standard
RCRA Resource Conservation and Recovery Act
R&D Research and development
SDWA Safe Drinking Water Act
SAIC Science Applications International Corporation
SARA Superfund Amendments and Reauthorization Act
Se Selenium
SOD Soil oxidant demand
SOP Standard operating procedure
SW-846 Test methods for evaluating solid waste, physical/chemical methods
SWDA Solid Waste Disposal Act
SITE Superfund Innovative Technology Evaluation
1,1,1-TCA 1,1,1-Trichloroethane
TCE Trichloroethene
TER Technology Evaluation Report
TSA Technical Systems Audit
TSR Technical Systems Review
TOC Top of casing
TSCA Toxic Substances Control Act
TSD Treatment, storage, and disposal
UL Upper limit (of specified confidence interval)
(jg/Kg Micrograms per kilogram
ug/L Micrograms per liter
uV Microvolts
USEPA United States Environmental Protection Agency
VP Vertical profile
VC Vinyl chloride
VOCs Volatile organic compounds
XDD Xpert Design and Diagnostics, LCC
yd3 Cubic yards
Zn Zinc
XII
-------
This report was prepared under the direction of Michelle Simon, the EPA Technical Project Monitor
for this SITE demonstration at the National Risk Management Research Laboratory (NRMRL) in
Cincinnati, Ohio. Both Dr. Simon and Paul de Percin (retired) served as EPA NRMRL Project
Monitors for the demonstration. Joseph Evans and Joseph Tillman of Science Applications
International Corporation (SAIC) served as the SITE work assignment managers for the
implementation of demonstration field activities and completion of associated reports.
Field sampling and data acquisition was conducted by Mr. Evans, Mr. Tillman, John King, Andrew
Matuson, Mark Hasting, Jillian Loughlin, and Hakim Abdi (all of SAIC). Data validation was conducted
by Mr. Evans and Rita Schmon-Stasik (SAIC); and a statistical evaluation of data was conducted by
Dr. Maurice Owens, also of SAIC.
The Demonstration required the combined services of Xpert Design and Diagnostics, LLC (XDD), and
SAIC. Aaron Norton served as the primary developer contact throughout the Demonstration. Also,
George Holt of Aries Engineering, provided valuable insight regarding site specific information
throughout the project duration. The cooperation and efforts of these organizations and individuals
are gratefully acknowledged. The cooperation and assistance of Bill Archibald, the contact for the
former MEC Building site, was invaluable and greatly appreciated.
Cover Photographs: Clockwise from top left are 1) Installation of Multi-Chamber Monitoring Well; 2) Jugs of
granular KMnO4 (20 L); 3) KMnO4 in deep silt zone core sample (post-treatment sampling event) 4} Tanker (left)
of pre-mixed KMnO4 oxidant solution (used for first injection) and water tanker (right) used for batching
operation 5) dark purple KMnO4 in well purge water; 6); Adding granular KMnO4 during batch operation, and
7) Back end of XDD Portable Oxidant Delivery (POD) trailer-mounted unit, showing distribution manifold and
flow meters.
Xiil
-------
Introduction
This report summarizes the findings of an evaluation of the
Xpert Design and Diagnostic's (XDD) in situ chemical
oxidation (ISCO) process at the former MEC Building site
located in Hudson, New Hampshire. At this site, both soil
and groundwaterare contaminated with chlorinated volatile
organic compounds (VOCs) due to releases from a former
underground concrete storage tank. The tank was
removed in May 1997.
Site contamination occurs as a dense non-aqueous phase
liquid (DNAPL) at about 23-25 ft (7 to 7.6 m) below land
surface (bis) and as dissolved phase VOCs in soil and
groundwater, primarily from 6-25 ft (1.8 to 7.6 m) bis. The
VOCs consist primarily of chlorinated ethenes, including
perchloroethylene (PCE), trichloroethylene (TCE), cis-1,
2-dichloroethylene, (cDCE), vinyl chloride (VC), and
toluene. Ethane compounds, such as 1,1,1-trichloroethane
(1,1,1-TCA) and 1,1-dichloroethane (1,1-DCA), are also
present. The soil contamination is concentrated in three
stratigraphic zones occurring between 6 and 25 feet (1.8 to
7.6 m) below land surface (bis); including an uppergravelly
silty-sand zone (6-13 ft bis), a thin peat zone (12-14 ft), and
a fine sandy-silt zone (14-26 ft). The DNAPL occurs at the
interface of the sand-silt zone and a basal till. The three
stratigraphic zones were targeted for ISCO treatment.
Project Objectives
There were two primary objectives for the demonstration;
one for soil and one for groundwater. The soil objective
was to determine if the XDD's process could reduce
concentrations of the chlorinated ethenes PCE, TCE,
cDCE, and VC by 90% over the course of the
demonstration comparing pre-treatmentto post-treatment.
The 90% reduction objective was to be determined on a
"paired sample evaluation" (i.e., pre-treatment versus post-
treatment samples collected in close proximity to one
another). The comparison was to be conducted for each
soil zone and for all three zones combined.
The primary objective for groundwater was to determine if
the XDD's process could reduce concentrations of specific
VOCs to below their respective 0.5% solubility limit based
on post-treatment results. VOCs having such a
Remediation Performance Goal (RPG) included PCE (750
ug/I), TCE (5,500 ug/I), and cDCE (17,500 ug/I). The
groundwater objective also specified that 90% of the
samples evaluated from post-treatment sampling were
required to meet this goal (i.e., If 10 pre-treatments
samples were above the limit to start, at least 9 post-
treatment results would have to be below the limit).
Treatment Design
XDD's treatment design included installation of three well
clusters within an approximate 1,200 ft2 area to serve as
injection points for potassium permanganate (KMnO4), the
oxidant chosen by XDD for treating the site VOCs. The
three well clusters were installed to form a triangle, each
cluster being about 18 feet apart and each consisting of a
well screened in each of the three stratigraphic zones
targeted for treatment. Each well cluster injection had an
estimated diameter of influence was 20 feet. Thus,
injections were to overlap laterally.
Oxidant Injection & Monitoring
Injection of KMnO4into all three well clusters began in early
June 2005. Just prior to this first injection the EPA SITE
Program sampled soils and groundwater to establish
baseline VOC concentrations. Soon after the start of
oxidant injection, there was noted short-circuiting of KMnO4
to the surface in two of the well clusters. This was
evidenced by KMnO4 oozing from surface cracks shortly
after Injecting the oxidant into deep wells screened at 20-
25 ft (6.1-7.6 m) bis. Due to suspected failed well seals,
ES-1
-------
the two well clusters in question were replaced and a
second injection was conducted in October of 2005 into
replacement wells. When combining the two injections,
just over 15,000 gallons {5,640 L) and approximately 3,700
Ibs (1,680 kg) of KMnO4 was injected into the three well
clusters combined, at concentrations ranging from 25 to 40
g/l. Of the total mass of KMnO4 injected, approximately
1,860 Ibs (840 kg) was injected into the deep zone,
approximately 1,500 Ibs (680 kg) was injected into the
middle peat zone, and approximately 320 Ibs {145 kg) was
injected into the shallow zone.
Following the injection phase of the project, groundwater
was monitored to gain insight into the ongoing
effectiveness of the process. The monitoring included
noting the visual presence of KMnO4, tracking of a bromide
tracer used forthe first injection, and analyzing samples for
target VOCs and metals. For each eventa total of 15 wells
were sampled {five for each of the three zones). Following
the initial baseline event and immediately preceding the
second injection, a second groundwater baseline was
needed to establish baseline VOC concentrations for the
two replacement injection well clusters. Two additional
intermediate events were conducted after both injections
but well before the final post-treatment soil and
groundwater sampling event.
The final post-treatment sampling event for both soils and
groundwater was conducted in March 2006 about nine
months after the first injection and just over five months
after the second injection. At this time KMnO4 was still
visible in the deep zone groundwater and was observed in
deep zone soil cores. However, KMnO4 was not observed
in soil or groundwater collected from the shallow or middle
peat zones. In fact KMnO4 was visually observed in just
one peat zone well (IW-3i) following oxidant injection and
was never visually observed in shallow zone groundwater
samples collected after the injections.
Results - Introduction
The two primary objectives were evaluated using the VOC
analytical results from the pre-treatment (baseline) and
post-treatment (final) sampling events. Table ES-1
provides a general overall summary of these results. For
determining if the soil contaminant removal efficiency was
90% or greater on a paired soil sample basis (i.e., the
primary objective), eligible sample pairs were selected for
statistical evaluation. (Note: due to high laboratory
reporting limits resulting from field methanol extractions of
soils, some sample pairs lacked quantified data results and
so some of these pairs were not deemed eligible for
evaluation).
The primary objective for groundwater was to evaluate the
effectiveness of XDD's process in reducing concentrations
of PCE, TCE, and cDCE in groundwater to below their
corresponding RPS of 750 ug/l, 5,500 ug/l, and 17,500
ug/l. The QAPP specified that to meet this objective more
than 90% of eligible samples had to meet those regulatory
criteria, as to reject the null hypothesis (i.e., have pre-
treatment concentrations reduced to below the RPS based
on post-treatment sampling results). However, there were
fewer than expected instances where criteria were
exceeded prior to treatment. As a result, only those
instances where such criteria were exceeded could be
evaluated statistically to determine if the objective was met.
Soil VOC Results
The soil VOC data was highly variable. High sample
concentration variability at DNAPL sites is commonly
observed, particularly at sites with heterogeneous
lithologies. Of the four critical VOCs evaluated (PCE, TCE,
cDCE, and VC), TCE had the best overall percentage of
sample pairs showing reductions > 90% (40.9% for all
zones combined). However, VC fared the best for
percentage of eligible pairs showing contaminant
decreases of any magnitude. For all zones combined,
82.1% of eligible sample pairs for VC showed contaminant
reductions from baseline to final sampling; including 88.9%
in the shallow zone and 70% in the peat zone.
Soils were also statistically evaluated on an average pre-
treatment versus post-treatment basis, using the
hypothesis test suggested in the Quality Assurance Project
Plan (QAPP) that the process will remove 90% of
contamination. For this evaluation, all pre-treatment and
post-treatment soil data was used so that pre- and post-
treatment sample population was equal. This meant
including non-detect values, which were assigned a value
of one half (Vz) the laboratory reporting limit. The null
hypothesis tested is that the contamination removed does
not exceed 90%. This null hypothesis was tested for 16
sets of analyte-zone combinations (i.e., the four
combinations for PCE would be PCE-shallow zone, PCE-
peat zone, PCE-deep zone, and PCE-all three zones
combined). Accepting the null hypothesis results in a
finding that the process does not meet demonstration
objectives; rejecting the null hypothesis results in a finding
that the process meets the demonstration objective. In
using a two-sided 95% confidence interval to test a one-
sided hypothesis, the null hypothesis could not be rejected
for any of the 16 combinations. Thus, the objective was
not met for any combination.
ES-2
-------
Table ES-1. Summary of Results.
Soil VOC Treatment
PCE
Shallow Zone
Middle (Peat) Zone
Deep Zone
All Zones Combined
5 of 17 eligible pairs show a decrease; 4 of 17 eligible pairs show a decrease > 90%
7 of 9 eligible pairs show a decrease; 4 of 9 eligible pairs show a decrease > 90%
11 of 15 eligible pairs show a decrease; 6 of 15 eligible pairs show a decrease > 90%
23 of 41 eligible pairs show a decrease; 14 of 41 eligible pairs show a decrease > 90%
Paired
Analysis
TCE
Shallow Zone
Middle (Peat) Zone
Deep Zone
All Zones Combined
6 of 14 eligible pairs show a decrease; 4 of 14 eligible pairs show a decrease > 90%
6 of 7 eligible pairs show a decrease; 4 of 7 eligible pairs show a decrease > 90%
1 of 1 eligible pair shows a decrease; and 1of T eligible pair shows a decrease > 90%
13 of 22 eligible pairs show a decrease; 9 of 22 eligible pairs show a decrease > 90%
cDCE
Shallow Zone
Middle (Peat) Zone
Deep Zone
All Zones Combined
14 of 26 eligible pairs show a decrease; 4 of 26 e_ligible pairs show a decrease > 90%
14 of 19 eligible pairs show a decrease; 6 of 19 eligible pairs show a decrease > 90%
1 of 2 eligible pairs show a decrease; 0 of 2 eligible pairs show a decrease > 90%
29 of 47 eligible pairs show a decrease; 10 of 47 eligible pairs show a decrease > 90%
VC
Shallow Zone
Middle (Peat) Zone
Deep Zone
All Zones Combined
16 of 18 eligible pairs show a decrease; 8 of 18 eligible pairs show a decrease > 90%
7 of 10 eligible pairs show a decrease; 2 of 10 eligible pairs show a decrease > 90%
There were no eligible pairs in the deep zone.
23 of 28 eligible pairs show a decrease; 10 of 28 eligible pairs show a decrease > 90%
PCE
Shallow Zone
Middle (Peat) Zone
Deep Zone
All Zones Combined
TCE
Shallow Zone
Middle (Peat) Zone
Deep Zone
All Zones Combined
Averaged
Results cDCE
Shallow Zone
Middle (Peat) Zone
Deep Zone
All Zones Combined
VC
Shallow Zone
Middle (Peat) Zone
Deep Zone
All Zones Combined
Number of
Samples
30
27
30
87
30
27
30
87
30
27
30
87
30
27
30
87
Average
Pre-Treafment
Concentration (MQ/Kg)
4.721
21120
194,523
75,259
962
2,825
306
1,314
4,400
17,217
126
6,904
524
4,700
109
1,677
Average
Post-Treatment
Concentration (|jg/Kg)
54,395
1105
15i,586
23,440
7,099
291
102
2,573
3,983
15,601
741
6,471
118
6,693
103
2,154
% Change
+ 1,050
-94.8
-93.5
-68.9
+ 638
-89.7
-66.7
+ 95.8
-9.48
-9.39
+ 488
-6.27
-77.5
+ 42.4
-5.5
+ 28.4
ES-3
-------
Table ES-1. Summary of Results (Continued).
Groundwater VOC Treatment
Analysis of Individual
Wells Having VOCs
above Criteria
Analysis of Zone
having concentrations
of PCE above Criteria
VOC exceeding criteria & Pre-Treatment
Well which exceeded Concentration
(M9/I)
PCE in IW-1s 5,200
(Shallow Zone)
PCE in 6,830
(Shallow Zone)
PCEinEW-4d 2,310
(Deep Zone)
TCEinIW-1s 11,200
(Shallow Zone)
TCE in 7,090
(Shallow Zone)
cDCEinIW-1s 31,000
(Shallow Zone)
cDCEin!W-1i 50,100
(Middle Peat Zone)
PCE in the five Shallow 2,450
Zone Wells (avg. of 5 values)
Post-Treatment Remediation
Concentration Performance
(|jg/l) Standard (|jg/l)
1,090/1,360 750
Avg. = 1,225
4,190/5,560 750
Avg. = 4,875
<0.5/<0.5 750
Avg. = < 0.5
572 / 766 5,500
Avg. = 669
4,810/7,430 5,500
Avg. =6,120
3,630/3,620 17,500
Avg. = 3,625
20,100/16,700 17,500
Avg. = 18,400
1,390 750
(avg. of 10 values)
Goal
Attained
No
No
Yes
Yes
No
Yes
No
No
Other Observations
KMnO4 visually observed in deep zone samples for all deep wells after the first injection up
through final event sampling; indicating good presence and persistence in deep zone.
KMnO4 in
Groundwater
KMnO4 not observed in any shallow zone samples after either injections; but detection of
bromide tracer in the same wells indicates possible exhausting of oxidant by high SOD.
KMnO4 observed in just one intermediate (peat) zone sample on one occasion (i.e., about 4
months after first injection event at the point of injection). But bromide tracer present. This
also indicates possible exhausting of oxidant by high SOD.
Metals in There is no evidence indicating that metals were mobilized by XDD's process. On an
Groundwater average basis, concentrations of most of the metals analyzed showed little change
throughout the five groundwater sampling events.
Estimated Cost
(XDD ISCO Process)
Using a hypothetical site having characteristics similar to the demonstration site, a size of
about 100,000 ft3 and 3,700 yd3 (i.e., 3.7 times the volume treated for the demonstration),
the cost for implementing a 10-injection well ISCO treatment system and monitoring
effectiveness over a 1-month period is estimated at $139,000
ES-4
-------
However, it should be noted when using a two-sided 95%
confidence interval to test a one-sided hypothesis, the
hypothesis is actually being tested at the 97,5% level.
Testing the hypothesis at the 90% hypothesis testing level
results in rejecting the null hypothesis (i.e., meeting the
objective) in one instance; namely PCE in the peat zone.
PCE had a measured 94.8% decrease in the peat zone;
from a 21,120 M9/kg baseline average to a 1,105 ug/kg
post-treatment average.
Soil was also qualitatively evaluated via cone penetrometer
technology (CRT) and membrane interface probe (MIP)
surveys. A total of 18 CRT /MIP survey points were
completed prior to baseline sampling. During this survey,
CPT/MIP logs were used to identify specific subsurface
zones and the highest concentration areas to target for soil
sampling. Although the final survey was scaled down,
results were fairly consistentwith the quantitative laboratory
data. In the deep zone, there was a slight decrease of the
PID signal in the vicinity of the two CPT/MIP points located
within the more contaminated portion of the DNAPL plume.
However, in the shallow zone there was a marked increase
of the PID signal at those same locations.
Groundwater - VOC Results
On an individual well basis, there were only seven
instances where any of the three critical VOCs exceeded
their respective RPS prior to treatment. Of these seven
instances VOCs were reduced to below the criteria in three
cases and were reduced to very close to the criteria in two
other instances. However, statistically, the null hypothesis
was not rejected for any of the three compounds.
On a per zone basis, there were apparent reductions from
baseline to final in the shallow zone for PCE (43%), TOE
(60%), and cDCE (61%), however the null hypothesis was
accepted because out of five instances where the pre-
treatment value exceeded the criteria, reductions to below
the criteria occurred in just three cases. The shallow zone,
in fact, was the only zone of the three that had an average
pre-treatment concentration for a particular contaminant
above the criteria. When averaging results for all shallow
zone wells, PCE averaged 2,450 ug/l in the shallow zone
prior to treatment, which exceeded its RPS of 750 ug/l.
PCE averaged 1,390 ug/l following treatment.
Averaged results in the intermediate peat zone showed
sharp increases in PCE and TCE, and a sharp decrease in
cDCE. The majority of this variation, however, was
attributable to one well which was closest to the source
area. The only VOC measured pre-treatment above its
criteria in an intermediate well was cDCE, which did show
a significant reduction from pre- to post-treatment (from
50,100 ug/l to an average of two sample results of 18,400
ug/l). However, this final result was still above its criteria of
17,500 ug/l). For the intermediate zone the null hypothesis
was not rejected for each of the three contaminants.
Averaged results of all deep zones wells showed no
appreciable change in pre- vs. post-treatment
concentration for the three VOCs. For the deep zone the
null hypothesis was not rejected for each of the three
contaminants. However PCE, the only VOC measured
pre-treatment above its criteria in a deep well, did show a
significant reduction to below the criteria (from 2,310 ug/l
to < 0.5 ug/l).
When all zones are combined as a single sample set (i.e.,
representing all site groundwater) PCE was the only VOC
of the three that had a baseline average above its criteria
of 750 ug/l. The average PCE baseline concentration was
1,020 ug/l and the average final PCE concentration was
746 ug/l. For the combined zone scenario, the null
hypothesis was accepted for each of the three
contaminants, including PCE.
Groundwater - KMnOA
Potassium permanganate, KMnO4, is visually observable
in groundwater at relatively low concentrations (i.e., 1
ppm). Therefore, following the initial injection of the
oxidant, KMnO4 was visually monitored in the groundwater
during sample collection. There were two observations of
note regarding the presence of KMnO4
1. KMnO4 was visually seen in all deep zone wells
following the first injection, but was seen in only
one well on one occasion in the intermediate zone
(a well receiving injected KMnO4 about four
months earlier). KMnO4 was not observed in any
shallow zone well at any time.
2. In most instances, target VOCs were reduced to
below detectable levels when KMnO4 was visually
present.
Due to its higher capacity to receive fluids, more KMnO4
was injected into the deep zone relative to the peat zone
and, especially, the shallow zone. Taking this into account,
and knowing that the peat zone contained substantial
humic material, the two observations above imply that the
mass of KMnO4 injected into the shallow and peat zones
may not have been sufficient to overcome high SOD
suspected in these two zones. The detection of a bromide
tracer ion at some measurable concentration in all shallow
and peat zone wells following the first injection imply that
the initial injected oxidant solution (containing KMnO4) was
indeed injected to those areas.
ES-5
-------
On the other hand, the mass of KMnO4 injected into the
deep zone appears to have been more than sufficient to
treat deep zone groundwater, especially since the KMnO4
persisted long after target VOCs were apparently oxidized.
This observation is more fully supported by the soil oxidant
demand (SOD) test for the soils at the M EC site, performed
by XDD (2005). They found that the shallow, gravelly
sands required 4.6 to 21,0 g KMnO4 per Kg sand; 104.7 to
146.9 g KMnO4 per Kg peat; and 1.8 to 5.6 g KMnO4 per
Kg deep silty sand.
Groundwater- Metals Results
There were no data suggesting that metals were mobilized
due the XDD process. On an average basis, most of the
metals analyzed did not change significantly in
concentration throughout the five groundwater sampling
events. Some metals {e.g., As, Be, Cr, Mg, and Zn)
actually showed decreased concentrations in the shallow
zone from baseline to final sampling events. Substantial
increases in average manganese (Mn) concentrations in
the peat and deep groundwater zones for the second
baseline sampling event is attributed to the influx of Mn
from the initial injection of KMnO4 into a nearby injection
well cluster. Increases of Cr and Mg in the deep zone
groundwater following both the first and second injections,
and an increase in Se in the deep zone groundwater
following the second injection were not sustained as the
average concentrations of these three metals reverted
back to an average value very close to baseline levels.
Costs of applying the XDD ISCO process were estimated
for a larger scale hypothetical site that was approximately
3.7 times the treatment area of the demonstration site. The
cost to install a 10-injection well ISCO treatment system,
utilizing XDD's POD to deliver oxidant to approximately
100,000 ft3 (3,700 yd3 ) of DNAPL-contaminated soil and
groundwater, and monitor effectiveness over a 1-month
period is estimated at $139,000. If further treatment were
required (i.e., re-injection), thus extending the treatment
period, the cost would increase by a considerable amount.
The largest cost categories for the application of the XDD
ISCO technology at a site having characteristics similar to
those described for the hypothetical site are 1)
consumables and supplies (41%) and 2) labor (21%),
together accounting for 62% of the total cost. The other
major costs, as estimated, include startup and fixed
(18%),and analytical services (9.7%).
ES-6
-------
1.0
Introduction
This Section provides background information about this
Superfund Innovative Technology Evaluation (SITE)
demonstration, the SITE Program, discusses the purpose
of this Innovative Technology Evaluation Report (ITER),
and describes the Xpert Design and Diagnostic's (XDD) in
situ chemical oxidation (ISCO) process using potassium
permanganate (KMnO4) to treat chlorinated ethenes in soil
and groundwater. Key contacts are listed at section's end
for inquiries regarding additional information about the
SITE Program, XDD's technology, and the Demonstration.
1.1 Background
XDD's ISCO process using KMnO4 was demonstrated
under the SITE Program from May 2005 to May 2006 at
the former MEC Building site in Hudson, New Hampshire.
The specific area treated at the former MEC site was a
relatively small 1,200 ft2 (114 m2) area that was
characterized as being contaminated with a dense non-
aqueous phase liquid (DNAPL) and dissolved-phase
volatile organic compounds (VOCs), primarily chlorinated
ethenes.
The source of the contamination was a concrete holding
tank located adjacent to what is referred to as the "former
MEC Building". MEC (formerly Nashua Electric Motors)
repaired and rebuilt electric motors at the property from the
mid-1970s up until approximately 1990. The subsurface
concrete holding tank, which had reportedly overflowed
during theirtenure, caused the contamination of underlying
soil and groundwater. The tank was removed in May 1997.
The general contaminant source area is defined in Figure
1-1, which is from a Remedial Action Plan prepared in
January2001 (Aries Engineering, 2001). The approximate
80 ft long x 50 ft wide x 26 ft deep plume (24 m x 15 m x 7
m) depicted in Figure 1-1 was generated from 1) initially
conducting vertical profiles (VPs) of soil and groundwater
VOC headspace data via a Photovac Gas Chromatograph
(GC); 2) comparing those data to a subset of samples that
were analyzed in a laboratory; and 3) calculating a ratio
between the laboratory-analyzed data and headspace data
to generate estimated VOC concentrations. Aries
conducted this characterization in May 1998. Soil and
groundwater at the site are contaminated, primarily with
perchloroethylene (PCE), trichloroethylene (TCE), cis-1,2-
dichloroethylene, (cDCE), and vinyl chloride (VC).
The USEPA SITE program conducted a preliminary site
assessment at the former MEC during 2003. Sampling
and analysis of monitoring wells in the vicinity of the former
MEC building revealed VOCs in excess of GW-1
standards. The highest concentrations detected were near
the former tank spillage. GW-1 standards are essentially
equivalent to federal drinking water standard maximum
contaminant levels (MCLs). Specifically the New
Hampshire Department of Environmental Services
(NHDES) regulates VOCs at the site that had exceeded
their respective GW-1 numeric cleanup standards at least
once in 2003. These include the following VOCs:
•• Perchloroethene (PCE)
» Trichloroethene (TCE)
•• cis-1,2 Dichloroethane (cDCE)
Vinyl Chloride (VC)
>• Toluene
> 1,1,1-Trichloroethane (1,1,1-TCA)
» 1,2-Dichloroethane (1,1 -DCA)
The areal extent of the pilot-scale treatment system's
injection and monitoring wells was approximately1,200 ft2
(-35 ft x 35 ft) or 114 m2- The locations of the oxidant
injection wells and the monitoring wells are provided in
Section 4.0 as Figure 4-1.
-------
FORMER MEG BUILDING
> 2001 ARIES ENGINEERING. INC W061E(4)1rt)1
Figure 1-1. Former MEC Building Source Area (Source: Aries Engineering Inc., January 2001).
-------
The demonstration of XDD's ISCO process was initiated in
May of 2005 with pre-treatment cone penetrometer
technology and membrane interface probe (CPT/MIP)
characterization, followed by baseline soil sampling and the
installation of three clusters of injection wells and three
multi-chambered monitoring wells aligned parallel with the
general northeasterly groundwaterflow direction. The well
installation was followed by baseline groundwater
sampling. The demonstration concluded in March of 2006
with final post-treatment soil and groundwater sampling.
Figure 1-2 is a time line of major events that occurred
during the demonstration.
Both injection well clusters and multi-chamberwell clusters
were screened at three separate depth intervals; 4-9 ft
(1.2-2.7 m) bis, 13-14 ft (4-4.3 m) bis, and 20-25 ft (6.1-7.6
m) bis to target an upper gravelly sand zone, a thin zone of
peat, and a lower sandy-silt zone, respectively. The
injection wells also served as monitoring wells. Thus, there
was a total of 15 monitoring points; five for each of the
three soil horizons.
The injection wells were spaced approximately 15-20 ft
(4.6-6.1 m) apart from each another to provide adequate
coverage of the demonstration area and provide overlap of
a calculated 10-foot (3 m) radius of influence.
There were two primary objectives of the demonstration:
1) to determine if the XDD ISCO process could remove
90% of target VOCs from paired soil samples in all three
horizons, and 2) to determine if the process could reduce
concentrations of critical VOCs in groundwaterto below the
following Remediation Performance Standards (RPS):
PCE
TCE
cDCE
VC
750 ug/l
5,500 ug/l
17,500 ug/l
no Standard
Baseline Soil
Sampling
May 12-19
(30 Borings and
87 VOC Samples)
Installation
of Injection (IW)
/ wells and Multl-
Chamber Wells
(May 20-22)
First Injection
1,760 IDS of KMnO,
injected. Most (88%)
injected into IW-3 cluster
due to failed seals in
IW-1 & IW-2 clusters
(June 6-10)
Replacement of
Injection well
clusters IW-1 and
IW-2 (6 wells)
June 27-30
Second Injection
1,930 IDS of KMn04
injected injected into
IW-1 & IW-2 clusters
(October 3-6)
Second
Baseline GW
Sampling
Sept. 19-21
(15 we Is)
XDD Water
Injection Test
onNewlW
Well Clusters
(Sept. 8)
Final
Post-Treatment
Soil Sampling
March 27-31
(30 Borings and
89 VOC Samples)
Intermediate
GW Sampling
December 6-8
(12 MWs)
Intermediate
GW Sampling
February 7-8
(15 MWs)
Final
Post- Treatment I
GW Sampling
March 14 & 16
(15MWS2X) I
(30 Samples) .
Final
CPT/MIP
Survey
March 14-15
(5 points)^
' May ' June ' July 'August' Sept. ' Oct. ' Nov. ' Dec. ' Jan. ' Feb. ' March'
2005
Figure 1-2. Time Line of Demonstration Events.
2006
-------
1.2 Brief Description of the SITE Program
The U.S. Environmental Protection Agency's (EPA) SITE
Program was established by EPA's Office of Solid Waste
and Emergency Response and the Office of Research and
Development (ORD) in response to the 1986 Superfund
Amendments and Reauthorization Act, which recognized
a need for an "Alternative or Innovative Treatment
Technology Research and Demonstration Program." The
SITE Program is administered by the ORD National Risk
Management Research Laboratory (NRMRL) in the Land
Remediation and Pollution Control Division (LRPCD),
headquartered in Cincinnati, Ohio. The SITE
Demonstration Program encourages the developmentand
implementation of:
1. Innovative treatment technologies for hazardous
waste site remediation, and
2. Monitoring and measurement.
In the SITE Demonstration Program, the technology is
field-tested on hazardous waste materials. Engineering
and cost data are gathered on the innovative technology so
that potential users can assess the technology's
applicability to a particular site. Data collected during the
field demonstration are used to assess the performance of
the technology, the potential need for pre- and
post-processing of the waste, applicable types of wastes
and waste matrices, potential operating problems, and
approximate capital and operating costs.
1.3 The SITE Demonstration Program and
Reports
In the past technologies have been selected for the SITE
Demonstration Program through annual requests for
proposal (RFP). EPA reviewed proposals to determine
promising technologies for use at hazardous waste sites.
Several technologies also entered the program from
current Superfund projects, in which innovative techniques
of broad interest were identified for evaluation under the
program. More recently, EPA has selected sites that would
require innovative technologies for clean-up.
Once the EPA has accepted a proposal, cooperative
arrangements are established among EPA, the developer,
and the stakeholders. Site owners and Developers are
responsible for implementing, operating and/or main tain ing
their innovative systems at a selected site, and are
expected to pay the costs to transport equipment to the
site, operate and/or maintain any equipment on-site during
the demonstration, and remove the equipment from the
site. EPA is responsible for project planning, sampling and
analysis, quality assurance and quality control, preparing
reports, and disseminating information.
Results of Demonstration projects are usually published in
three documents: the SITE Demonstration Bulletin, the
Technology Capsule, and the ITER. The Bulletin describes
the technology and provides preliminary results of the field
demonstration. The Technology Capsule provides more
detailed information aboutthe technology, and emphasizes
key results of the field demonstration. The ITER provides
detailed information on the technology investigated, a
categorical cost estimate, and all pertinent results of the
field demonstration. A Technology Evaluation Report
(TER) is sometimes prepared, but is available by request
only. The TER serves as verification documentation
contains a more comprehensive presentation of the
analytical data (i.e., raw data packages, etc.) collected
during the demonstration. For the demonstration of the
XDD ISCO process, this ITER is intended for use by
remedial managers for making a detailed evaluation of the
technology for a specific site and waste.
1.4 of the Innovative Technology
Evaluation (ITER)
This ITER provides information on the XDD in situ
chemical oxidation process for treating primarily chlorinated
ethenes in soil and groundwater. The ITER includes a
comprehensive description of this demonstration and its
results and is intended for use by EPA remedial project
managers, EPA on-scene coordinators, contractors, and
other decision-makers carrying out specific remedial
actions. The ITER is designed to aid decision-makers in
evaluating specific technologies forfurtherconsideration as
applicable options in a particular cleanup operation.
To encourage the general use of demonstrated
technologies, the EPA provides information regarding the
technology applicability to specific sites and wastes. The
ITER includes technology-specific information on cost,
advantages, disadvantages, and limitations; and discusses
desirable site-specific characteristics. Each demonstration
evaluates the performance of a technology treating a
specific waste matrix. Characteristics of other wastes and
other sites may differfrom the characteristics of the treated
waste; therefore, a successful field demonstration of a
technology at one site does not necessarily ensure
applicability to other sites. Only limited conclusions can be
drawn from a single field demonstration.
-------
1.5 Technology Description
In situ chemical oxidation involves the introduction of a
chemical oxidant into the subsurface for the purpose of
transforming ground water or soil contaminants into less
harmful chemical species. The process is non-selective;
therefore, any oxidizable material reacts. Consequently,
the natural humic content of the soil that contacts oxidant
is an important measured parameter.
XDD's ISCO process utilizes several oxidants. Specific to
this demonstration, KMnO4 was used to treat soil and
groundwater contaminated with chlorinated VOCs. The
KMnO4 is the less expensive of the two common MnO4
oxidants (the other being NaMnO4). KMnO4 is widely and
commonly used in the wastewater treatment industry and
therefore is readily available from several manufactures.
Granular KMnO4 is packaged and shipped in 20 Liter
plastic jugs, each weighing about 55 pounds. KMnO4 is
mixed in large tanks at desired dosages with potable water,
filtered to remove solids (e.g., silica) and metered for
injection through a manifold and high-pressure hoses.
There are several ways to inject oxidant to the desired
contaminant location. The most common method, as was
the case during the demonstration, is via traditional
injection wells (2 inch ID is preferred). However XDD has
utilized Geoprobe push points, infiltration galleries, and
even trenches and pits for shallow zone treatments. For
bedrock applications, they have utilized wells equipped with
packers for targeting specific fractured bedrock zones.
Other manufacturers have injected in situ oxidants by
hydraulic or pneumatic fracturing.
As shipped in granular form, KMnO4 is a strong oxidizer. It
is a known irritant to the respiratory system, highly
corrosive to the skin, and potentially fatal if swallowed.
When handling the dark purple crystalline solid during
batching operations, XDD personnel wear respirators and
chemical-resistant clothing (e.g., coated tyvek) to avoid
breathing dusts and dermal contact. Residuals potentially
generated during a full-scale ISCO treatment are
contaminated drill cuttings generated from well installation,
purge water from well development and sampling of
monitoring wells, and personal protective equipment.
1.6 Key Contacts
Additional information regarding XDD's technology and the
SITE Program can be obtained from the following sources:
Demonstration Technology Contact
Ken Sperry, Branch Manager.
Xpert Design and Diagnostics, LLC
22 Marin Way, Unit# 3
Stratham, NH 03885
Phone: (603) 778-1100
Fax: (603) 778-2121
Web Site: www.xdd-llc.com
E-mail: sperry@xdd-llc.com
The SITE Program
Michelle Simon, Ph.D., P.E.
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
26 West Martin Luther King Drive
Cincinnati, OH 45268
Phone:(513)569-7469
Web Site: www.epa.gov/ord/SITE
E-mail: simon.michelle@epa.gov
Information on the SITE Program is available through the
following on-line information clearinghouses:
• The SITE Home page (www.epa.gov/ORD/SITE)
provides general program information, current
project status, technology documents, and access
to other remediation home pages.
The OSWER CLU-ln electronic bulletin board
(http://www.clu-in.org) provides information on
innovative treatment and site characterization
technologies while acting as a forum for all waste
remediation stakeholders.
-------
2.0
This section addresses the general applicability of the
Xpert Design and Diagnostic's (XDD) 1SCO process to
sites having soil and groundwater contaminated with
chlorinated VOCs. The analysis is based on observations
made during the SITE Program Demonstration, and from
additional information received from XDD (the developerof
a specialized ISCO treatment process). The results of this
SITE Demonstration are presented in Section 4.0 of this
report. XDD had the opportunity to discuss the
applicability, other studies, and performance of the
technology in Appendix A.
2,1 Key of the XDD
There are four key features comprising XDD's ISCO
process. These include the following:
* Oxidant
•I Portable Oxidant Delivery (POD) System
$ Oxidant Injection Points
* Monitoring Wells
Each of these key features is discussed in the following
paragraphs.
2.1.1 Oxidant
Depending on the contaminants to be treated XDD utilizes
either persulfate (S2OX) or permanganate (MnO4) as the
oxidizing agent. Potassium permanganate (KMnO4) was
the chosen oxidant at the former MEC Building site due to
its ability to economically treat chlorinated ethenes. MnO4
is a very stable oxidant and can persist for several months
in the subsurface. This stability makes it a good choice for
subsurface applications (i.e., fewer injection events, fewer
wells to treat the target area, and the ability to more
effectively penetrate into low permeability zones). XDD
utilizes two forms of MnO4 to treat organic contaminants,
1) sodium permanganate (NaMnO4) and 2) potassium
permanganate (KMnO4). Each has its distinct advantages
and disadvantages as discussed below.
NaMnO4
Higher concentrations of MnO4 can be injected in the
sodium form as compared to the potassium form. The
most significant drawback to NaMnO4 is cost (NaMnO4 is
five to seven times more expensive than KMnO4)
NaMnO4 is handled in a liquid form which is preferred over
the solid (i.e., granular) KMnO4 from a health and safety
and materials handling perspective. The application of this
liquid oxidant is simpler than injecting KMnO4, which has to
be pre-mixed with potable water at the desired
concentration prior to injection. But because of cost, a
precise knowledge of where contaminant(s) are situated is
needed to implement a surgical injection strategy.
KMnO4
KMnO4 is preferred over NaMnO4 when the subsurface
lithology is complex or when soils with high humic content.
This is because NaMnO4is much more expensive and thus
must be precisely injected to accessible zones and not be
exhausted by naturally occurring organic material. An
assessment of SOD is needed to determine how much
reagent may be wasted on oxidizing benign organic
compounds.
KMnO4 is a dark purple/bronze solid, produced in either a
crystalline or free-flowing powder form. It is shipped in 20
Liter plastic jugs and has a reported shelf life of one year.
It is readily manufactured due to its common use in the
wastewater treatment industry and usually contains silicon
dioxide as an anti-caking agent. Consequently, the created
oxidant solution is filtered prior to injection to remove such
unwanted solids that may clog well screens.
-------
Table 2-1. Properties of KMnO^.
^Appearance;
Molecular Wt
Spec, Gravity
Bulk Density:
Flash Point:
Storage:
Shelf Life:
Handling:
Hazard
Overview:
HMIS Ratings
Water Soluble:
Dark purple/bronze odorless solid in
crystalline or free-flowing powder form.
158.04
2.7 @ 15 °C
90-100Ibs/ft3
Not combustible
Strong Oxidizer: Store in cool, dry, non-
freezing area away from direct sunlight,
intense heat, & combustible materials.
One year® 13-32°C
Avoid dermal, gloves/goggles suggested
Strong Oxidizer
Health hazard - 3 (moderate)
Fire hazard - 0 (minimal)
Physical hazard - 0 (minimal)
6.38g/100 cc @ 20 "C
Sources: CHEM ONE Product Specification, rev. 2/23/04; Material
Safety Data Sheet, re. 7/31/03; XDD Personal Comm., 01/26/06.
2.1.2 Portable Oxidant Delivery System
XDD's Portable Oxidant Delivery ("POD") System is an
array of equipment mounted inside a utility trailer (13 by 7
ft wide, not including a trailer hitch assembly). The POD
can be towed with a heavy-duty pickup truck. The following
primary equipment is contained within the POD.:
Two 300-gallon (1,140 L) chemical oxidant
batching tanks which can be pumped in unison to
transfer oxidant to injection wells.
Dual oxidant metering pumps (for injecting KMnO4
and NaMnO4simultaneously)
Process Equipment (i.e., high durable/chemical-
resistant totalizer flow meters) for monitoring flow
rates, temperature, and pressure.
Filtration system to remove particulate matter in
the injection fluid stream.
Manifold designed to simultaneously deliver
oxidant in up to 6 injection wells.
During the demonstration, XDD performed batch
operations within a bermed area in proximate vicinity to the
POD. Batching was conducted in two larger chemical
tanks, each having a capacity of 500 gal. (1,890 L). Jugs,
each containing about 20 Liters of granular KMnO4, were
emptied into these tanks and mixed with potable water
pumped from a water tanker trailer at the desired liquid
KMnO4 concentration. The KMnO4/watersolution was then
mixed for M>-1 hour with a sump pump, routed through two
cannister filters connected in series, and pumped onto the
two 300-gallon chemical tanks within the POD unit. A
piston pump within the POD pumps the oxidant through a
PVC manifold assembly equipped with pressure valves and
totaliser meters. The POD has the capability of routing
oxidant to up to six injection wells simultaneously.
2.1.3 Oxidant Injection Points
Injection points are required to properly deliver oxidant to
the desired locations by means of the most effective and
economical method possible. The most common type of
injection points are injection wells that are screened within
an aquifer at the targeted contaminated zone. At the
former MEC Building site, XDD's treatment system
consisted of three injection well clusters. Each cluster
consisted of three wells screened at three different depths;
thus, there was a total of nine injection points. The
injection wells targeting the top and bottom zones were
fitted with 5-ft (1.5 m) long screens and the injection wells
targeting the thin middle peat zone were fitted with 1-ft (0.3
m) long screens. All wells were flush mounted since they
were located in an area immediately adjacent to a docking
bay that was frequented by heavy trucks and forklifts. The
locations of all demonstration wells is shown in Figure 4-1.
Injection wells require fittings for connecting a high
pressure hose to the injection wellheads. XDD uses
common materials and parts for these fittings, which are
available at local hardware stores. As a result, existing
monitoring wells can be adapted for injection as long as
they are screened at the proper interval and are equipped
with the appropriate slotted screens. For the
demonstration XDD used % -inch ID high pressure hoses
to injected into both 1-inch ID and 2-inch ID injection wells.
-------
Figure 2-1 presents example schematics of injection wells
comprising an injection well cluster that was used for the
demonstration (IW-1 and IW-2, specifically). These wells
are of traditional design and can be used for monitoring
following injection of oxidant.
Thenumberand locations of injection wells are dependent
on a site's characterization. XDD has indicated that the
preferred injection well construction would be comprised of
a 2-inch inner diameter (ID) well having screen lengths
preferably not in excess of 10 feet in length.
It should be noted that there are direct-injection techniques
that can be used to deliver oxidants which do not require
an injection well. A geoprobe, equipped with drive rods of
1.25-inch O.D./0.625-inch l.D. or2.125-inch O.D./1.5-inch
I.D., can be utilized to provide a temporary casing (i.e.,
drive point) to allow the oxidant to be pumped subsurface
without installing a well. Also, oxidant can be injected into
infiltration galleries, excavated trenches, hydraulic or
pneumatic fractures and pits.
Sump
n
"
0
2'
"
^""Bentonite
- 0,5-2'
"4- Grout 2-1 3'
Sump'
Grout-1.5-3'
15
20
25
Sump
Small layer of grout
atop bentonite
Bentonite
-0.5-3.5'
Filter Pack (#2
sand) 3.5 - 9'
Screened Interval
4-9' 10 slot
Filter Pack (#2 sand)
12.5-13'
Screened Interval
13-14' 10 slot
Bentonite
13-19.5'
Filter Pack (#2 sand)
= 19.5 -25'
Screened Interval
20-25' 10 slot
4.25"
Figure 2-1. Injection Well Cluster Schematics.
-------
2.1.4 Monitoring Wells
Monitoring wells are a necessary component for monitoring
the effectiveness of an ISCO process. In many instances,
existing monitoring wells can be utilized for the monitoring,
dependent on their location. Also, in some instances (as
was the case for the Demonstration), the injection wells
can serve a dual purpose (i.e., as monitoring points). It
should be noted that typically the injection wells are spaced
or grouped in such a manner that they alone would not
provide adequate coverage of a contaminant plume.
For the SITE Demonstration, there were two types of wells
installed for monitoring; traditional wells (as shown on
Figure 2-1) and "multi-chamber wells" (Figure 2-2). The
multi-chamberwells were essentially a 3- in-1 design; each
1-inch OD casing containing three separate 7/16-inch ID
wells set at different depths. All in all, there was a total of
five monitoring wells for each of the three stratigraphic
zones monitored (for a total of 15 wells).
BLS*(ft) Ground Surface
Screen Mesh
4'-9'
(multiple sections)
Bentonite
Pellet Socks
(13'-IS1)
Bentonite
Pellet Socks
(IT-W)
Formation
Collapse
(19' - 25')
Screen Mesh
20' -25'
(multiple sections)
Enlargement of
Multi-Chamber
Inner/Outer Casing
1"OD
\
7/16" ID
Figure 2-2. Multi-Chamber Well Schematic.
-------
The number of monitoring wells that would be needed for
a remediation project using XDD's process is highly site-
specific. At a minimum, monitoring wells are typically
required upgradient, downgradient, and lateral to the area
to be treated. The former MEC Building site has been
monitored continuously for an extended period of time to
track the slow dissolution of the DNAPL and dissolved
organic contaminant plume. The Site Demonstration
focused on monitoring treatment effectiveness within a
small area side-gradient of the main DNAPL source zone.
2.2 Operability of the Technology
The implementation of XDD's technology is broken down
into an eight-step process, listed and discussed below.
1. Site Evaluation
2. Treatability Testing
3. System Design
4. Injection Well Installation
5. Pilot Testing {Water Injection Tests)
6. System Setup
7. Batching of Oxidant
8. Oxidant Injection & Monitoring
9. Equipment Decontamination
10. Monitoring Treatment Effectiveness
XDD will sometimes conductsite characterization activities
themselves; but more commonly will evaluate existing
characterization data for the site. But site evaluation is an
important aspect, as XDD typically conducts treatability
testing to determine various aspects of the contaminated
soils that are to be treated. For example, prior to the
demonstration XDD conducted a laboratory treatability
study in order to evaluate the effectiveness of their ISCO
process on contaminated soil from the site. The objectives
of this treatability study were to:
Assess the overall feasibility of using ISCO to
meet the site-specific cleanup goals;
Estimate the soil oxidantdemand forthe majorsoil
units (i.e., gravelly sand, peat layer, and sandy-
silty) in the treatment area;
Develop site-specific data necessary to design an
ISCO field pilot test and/or full-scale application.
Injection well installation is performed by an outside
contractor. The injection wells can typically serve a dual
purpose (i.e., as monitoring wells once oxidant is injected).
Also, existing monitoring wells may be able to be retrofitted
for injection purposes. XDD sometimes conducts water
injection testing on prospective injection wells, a practice
they are utilizing more frequently. As an example, for the
demonstration project at the former MEC Building site,
XDD performed a small-scale water injection test at a
targetflow rate to monitorwell seal and injection pressures
of newly-installed injection wells that served as
replacements for wells that had to be abandoned due to
failed seals.
The following tasks comprise the basic setup of XDD's
equipment (i.e., the system set-up step).
Oxidant delivery/storage of KMnO4
Setup of spill guards for batch area
Potable water delivery (tanker)
Drop-off of rented generator
Hookup of batch operation to POD
Connect injection distribution system to wells
The batching process is labor intensive. For example the
granular KMnO4 used during the demonstration is
containerized in 20 L jugs. The granular oxidant needs to
be mixed with potable water to form an oxidant solution at
the desired concentration. To do this, the jugs were
manually transported to the top of a large mixing tank,
where the jug contents were dumped into the tank
equipped with a sump pump for mixing. Tyvek and
respirators were worn during this process as a
precautionary health and safety measure.
To inject oxidant, the KMnO4 solution is pumped through
one or more cannister filters (two filters, connected in
series, were used during the demonstration). The filtration
is needed to remove any unwanted material from the
solution. Such material includes a small amount of silica
(i.e., 1% or less of silica is mixed in with the granular
KMnO4 as an anti-clogging agent).
During XDD's injection process, equipment
decontamination consists of periodic flushing of cannister
filters, which are used to remove unwanted material (e.g.,
silica solids) from the oxidant solution. Following an
injection event, XDD conducts a thorough flushing of the
oxidantfrom the POD and related equipmentcomponents.
To accomplish this, a neutralizing solution is used, that
consists of a 5:1:1 mixture of potable water, vinegar, and
hydrogen peroxide (H2O2), respectively. The neutralizer is
pumped through the POD components, through the high
pressure hoses and down into the injection wells (the small
amount of injected neutralizer has a negligible effect on the
subsurface).
Monitoring of treatment effectiveness is highly site-specific
and very dependent on regulatory requirements as to the
degree of monitoring required. This endeavor may or may
10
-------
not be conducted by XDD, largely depending on whether
they are serving as the site remediation contractor.
Regardless, XDD would have some input into evaluating
treatment results. In most cases monitoring of an ISCO
process would involve at minimum a pre-treatment and
post-treatment sampling event for all affected media (i.e.,
soils and groundwater). Typically groundwater would
require continued post-treatment monitoring for an
indefinite period, largely dependent on state and local
regulatory requirements.
2.3 Applicable
ISCO is primarily used to treat contaminated soil and
groundwater. Permanganate (MnO4) has been widely used
and has been shown to completely mineralize several
common chlorinated VOCs to yield innocuous end
products. MnO4 is more effective for treating chlorinated
ethenes; particularly tetrachloroethene (PCE) and
trichloroethene (TCE) are most readily oxidized. However,
significant degradation of chlorinated ethanes, such as
1,1,1-trichloroethane (1,1,1-TCA), has also been observed.
In addition to organics, MnO4also reacts with natural soil
organic matter and reduced metal oxides (e.g., iron and
manganese oxides).
With respect to applicable soil types, high clay soils would
be expected to be less amenable to ISCO since plugging
could cause the flow of groundwater to divert around areas
of contamination. Deeper contamination would require
more costly well installation.
2.4 Availability Transportability of
Equipment
The XDD ISCO process can theoretically be implemented
anywhere injection and monitoring wells can be installed,
which would include any location that can be accessed by
a drill rig, or geoprobe, or other direct push technology
(DPT) equipment. Since all-terrain drill rigs and DPT
equipment are available, most locations would be
accessible.
At the former MEC Building site, the injection system
consisted of three injection well clusters, in which each
cluster contained three separate wells screened in a
shallow (4-9 ft), middle (13-14 ft), and deep (20-25 ft)
stratigraphic zones thathad been previously characterized.
Nine additional small diameter wells were installed at three
locations for monitoring purposes only. These wells were
screened at the same depth intervals the injection wells
were and were constructed with readily available
construction materials typically used for well installation.
The primary components for the KMnO4 injection
equipment used during the demonstration were contained
on a single trailer (the POD). The POD can be easily
mobilized with a heavy duty pickup truck. Granular KMnO4
was shipped to the site in 20 Liter plastic jug-like
containers, each weighing approximately 55 pounds. The
jugs were unloaded from a truck on wooden pallets and
stored in an adjacent warehouse.
During the demonstration XDD's system required periodic
monitoring of basic groundwater parameters. The
equipment used for these activities (e.g., water level
indicators, multi-parameter water quality meters, etc.) are
portable and can be easily shipped or transported to a site.
2.5 Handling Requirements
Materials handling requirements for XDD's ISCO process
are largely dependent on the type of oxidant used. The
considerably less expensive KMnO4 oxidant used during
the demonstration requires a somewhat rigorous materials
handling relative to the NaMnO4 oxidant. Granular KMnO4
is shipped in 20 Liter plastic jugs, each jug weighing
approximately 55 pounds. During the demonstration the
jugs were manually carried from a loading dock to XDD's
batch set up area. To mix the oxidant with potable water
the jugs had to be lifted via a ladder to the top of a mixing
vessel and emptied into a tank. Due to the health and
safety concerns (inhalation on KMnO4 dust), a respirator
and coated tyvek were worn during this operation.
NaMnO4is shipped in liquid form and thus can be delivered
in a large tanker, from which the oxidant can be pumped
directly to the POD.
Because drilling operations are involved to install the
injection wells and possibly monitoring wells, there is the
potential of handling hazardous residual materials.
Examples would include drumming of soil cuttings, purge
water, and decontamination water.
2.6 Support Requirements
During the Demonstration, the equipment and supplies for
the XDD ISCO process encompassed an approximate 80
ft x 30 ft area (i.e., 2,400 ft2 or 220 m2). The majority of
space is needed for the batching process, which requires
a large water supply for diluting granular KMnO4 (a 5,000
gallon tanker was used for the demonstration), and a
bermed area adjacent to the water supply that contains two
500 gal. (1,890 L) polypropylene mixing tanks, three
transfer pumps and two cartridge filters connected in
series. The trailer-mounted POD and generator take up a
11
-------
very small amount of space (i.e., approximately 200 ft or
18.5 m2).
For in situ oxidation to take place effectively in
groundwater, it must be technically and economically
feasible for the oxidant to contact the contaminated media
(i.e., both soil and groundwater). This can be a tradeoff of
cost of drilling multiple wells for injecting KMnO4 versus the
cost of alternative treatments involving fewer wells (e.g.,
pump and treat systems). The sites where ISCO is to be
used must be able to provide good access for a drill rig or
DPT equipment.
XDD typically rents a gasoline ordiesel powered generator
(e.g., 70 KVA) to supply electric power to the POD. The
peristaltic pumps needed for low flow groundwater
sampling of shallow wells can operate off a car battery.
Both of these items are easily rented and are usually
available locally.
Electrical power may also be needed to supply lighting to
an on-site trailer and a security light, and possibly for a
phone and facsimile hookup. During the demonstration
project, use of space within the former MEC Building was
invaluable for preparing soil and groundwater samples
during variable weatherconditions encountered throughout
the demonstration sampling events. Withoutthe use of the
building, a trailer would likely have been rented for that
purpose. Other than electricity, a water source is needed
for occasional decontamination activities (e.g., rinsing out
the individual jugs of KMnO4). Since XDD usually rents the
water tanker during an injection event, that could also
serve as a source for water used for other purposes.
2.7 Limitations of the Technology
Because permanganate reacts with natural soil organic
matterand reduced metal oxides, contaminantdegradation
rates can be adversely affected by the presence of these
competing species. Humic and fulvic acids (collectively
referred to as humates) are highly susceptible to
permanganate oxidation. Humic-containing soils are
categorized as having a high soil oxidant demand (SOD),
a term used for measuring how much oxidant the natural
material in a soil could potentially use up, thus depriving
the oxidant's intended use for treating organic
contaminants.
Chemical oxidation also has the potential to mobilize
valence sensitive toxic metals, such as Arsenic (As),
Chromium (Cr), Selenium (Se), Zinc (Zn), and Mercury
(Hg), from soils into groundwater. As a result, sites that
contain elevated levels of metals contaminants as well as
organic contaminants should be monitored for such
mobilization potential.
The porosity of the soil and ground water flow can affect the
rate and overall effectiveness of oxidation. Low porous
and low conductive soils would hamper injection and retard
the mobility of the oxidant. Sites having heterogeneous
stratigraphy will not have consistent hydraulic properties.
Small-scale high permeability zones can act as localized
conduits for dissolved-phase VOCs, thus making the
contaminants a hard target for injected MnO4.
Some organic compounds are not degraded as readily by
permanganate as chlorinated ethenes are. For example,
toluene is known to degrade in the presence of MnO4, but
at relatively slow rates. Compounds, such as toluene and
harder to degrade ethanes, are commonly found at sites
having ethene contamination. Such was the case for the
former MEC Building site.
2.8 for
This subsection discusses specific federal environmental
regulations pertinent to the operation of the XDD ISCO
process, referred to as Applicable or Relevant and
Appropriate Requirements (ARARs). ARARs include the:
(1 )Com prehensive Environmental Response,
Compensation, and Liability Act (CERCLA); (2) Resource
Conservation and Recovery Act (RCRA); (3) Clean Air Act
(CAA); (4) Clean Water Act (CWA); (5) Safe Drinking
Water Act (SOW A), and the (6) Occupational Safety and
Health Administration (OSHA) regulations. These six
general ARARs and state requirements for the former MEC
Building site are discussed in the following subsections.
Specific ARARs that may be applicable to the XDD ISCO
process are identified in Table 2-2,
12
-------
Table 2-2. Federal and State ARARs for XDD's ISCO Process.
Process
Activity
Waste
Charac-
terization
Waste
Processing
Storage of
auxiliary
wastes
Determination
of cleanup
standards
Waste
disposal
ARAR
RCRA: 40 CFR
Part 261 (or the
state equivalent)
RCRA: 40 CFR
Part 264 (or the
state equivalent)
CAA: 40 CFR
Part 50 (or the
state equivalent)
RCRA: 40 CFR
Part 264
Subpart J (or the
state equivalent)
RCRA: 40 CFR
Part 264
Subpart I (or the
state equivalent)
SARA: Section
121(d)(2)(ii);
SDWA: 40 CFR
Part 141
RCRA: 40 CFR
Part 262
CWA: 40 CFR
Parts 403 and/or
122 and 125
RCRA: 40 CFR
Part 268
Description
Standards apply to
the identification and
characterization of
wastes. Chemical
and physical
properties of waste
determine suitability
for ISCO treatment.
Standards apply to
treatment of wastes
inatreatmentfacility.
Regulations govern
toxic pollutants,
visible emissions and
particulates.
Regulation governs
the standards for
tanks at treatment
facilities.
Regulation covers
the storage of waste
materials generated.
Standards apply for
treatment of surface
water or groundwater
that is to be used for
drinking water
supplies.
Standards that
pertain to generators
of hazardous waste.
Standa rds for
discharge of
wastewater to a
POTW or to a
navigable waterway.
Standards regarding
land disposal of
hazardous wastes
Basis
Chemical and physical properties of waste
determine its suitability for treatment by an
ISCO process.
Not likely to be applicable or appropriate for
the XDD ISCO process.
During well installation and oxidant
batching/injection, any off-gas venting (i.e.,
from buildup of VOCs, etc.) must not
exceed limits set for the air district of site.
(Not likely to occur.)
Storage tanks for liquid wastes (e.g.,
decontamination waste) must be placarded
appropriately, have secondary containment
and be inspected daily.
Potential hazardous wastes remaining after
treatment (i.e., drill cuttings) must be
labeled as hazardous waste and stored in
containers in good condition. Containers
should be stored in a designated storage
area and storage should not exceed 90
days unless a storage permit is obtained.
Remedial actions are required for
groundwater to meet MCL goals (MCLGs)
or MCLs established under the SDWA.
Standards apply to surface & groundwater
sources that may be used as drinking
water.
Potential hazardous waste generated by
ISCO process is limited to drill cuttings,
well purge water, PPE, and wastes
generated from decontamination.
Applicable and appropriate for well purge
water and decontamination wastewater
generated from drilling process. The ISCO
process does not generate wastewater.
Applicable for off-site disposal of auxiliary
waste (e.g., drill cuttings).
Response
Chemical and physical analyses
must be performed to determine if
waste is a hazardous waste.
When hazardous wastes are
treated, there are requirements for
operations, record keeping, and
contingency planning.
Off-gases may contain volatile
organic compounds or other
regulated substances, although
levels are likely to be very low.
If storing non-RCRA wastes, RCRA
requirements may still be relevant
and appropriate.
Applicable for RCRA wastes;
relevant and appropriate for non-
RCRA wastes.
Remedial actions for surface and
groundwater are required to meet
federal MCL goals (MCLGs) or
MCLs established underSDWA; or
in the case of the former MEC site
the NHDES site-specific criteria.
Generators must dispose of wastes
at facilities permitted to handle the
waste. Generators must obtain an
EPA ID number prior to disposal.
Discharge of wastewater to a
POTW must meet pre-treatment
standards; discharges to a
navigable waterway must be
permitted under NPDES.
Hazardous wastes must meet
specific treatment standards prior
to land disposal, or be treated using
specific technologies.
13
-------
2.8.1 CERCLA
The CERCLA of 1980 as amended by the Superfund
Amendments and Reauthorization Act (SARA) of 1986
provides for federal funding to respond to releases or
potential releases of any hazardous substance into the
environment. As part of the requirements of CERCLA, the
EPA has prepared the National Oil and Hazardous
Substances Pollution Contingency Plan (NCP) for
hazardous substance response. The NCP is codified in
Title 40 CFR Part 300, and delineates the methods and
criteria used to determine the appropriate extentof removal
and cleanup for hazardous waste contamination. SARA
states a strong statutory preference for remedies that are
highly reliable and provide long-term protection. It directs
EPA to do the following:
• Use remedial alternatives that permanently and
significantly reduce the volume, toxicity, or the
mobility of hazardous substances, pollutants, or
contaminants;
« Select remedial actions that protect human health
and the environment, are cost-effective, and
involve permanent solutions and alternative
treatment or resource recovery technologies to the
maximum extent possible; and
• Avoid off-site transport and disposal of untreated
hazardous substances or contaminated materials
when practicable treatment technologies exist
[Section 121(b)].
In general, two types of responses are possible under
CERCLA: removal and remedial actions. Superfund
removal actions are conducted in response to an
immediate threat caused by a release of a hazardous
substance. Many removals involve small quantities of
waste of immediate threat requiring quick action to alleviate
the hazard. Remedial actions are governed by the SARA
amendments to CERCLA. As previously stated, these
amendments promote remedies that permanently reduce
the volume, toxicity, and mobility of hazardous substances
or pollutants.
The XDD ISCO process could possibly be part of a
CERCLA remedial action since the toxicities of the
contaminants of concern are intended to be reduced by
chemical destruction or alteration. Remedial actions are
governed by the SARA amendments to CERCLA. On-site
remedial actions must comply with federal and more
stringent state ARARs. ARARs are determined on a site-
by-site basis and may be waived under six conditions: (1)
the action is an interim measure, and the ARAR will be met
at completion; (2) compliance with the ARAR would pose
a greater risk to health and the environment than
noncompliance; (3) it is technically impracticable to meet
the ARAR; (4) the standard of performance of an ARAR
can be met by an equivalent method; (5) a state ARAR has
not been consistently applied elsewhere; and (6) ARAR
compliance would not provide a balance between the
protection achieved ata particularsite and demands on the
Superfund remedial projectmanager(RPIV1)forothersites.
These waiver options apply only to Superfund actions
taken on-site, and justification for the waiver must be
clearly demonstrated.
2.8.2 RCRA
RCRA, an amendment to the Solid Waste Disposal Act
(SWDA), is the primary federal legislation governing
hazardous waste activities. It was passed in 1976 to
address the problem of how to safely dispose of the
enormous volume of municipal and industrial solid waste
generated annually. Subtitle C of RCRA contains
requirements for generation, transport, treatment, storage,
and disposal of hazardous waste, most of which are also
applicable to CERCLA activities. The Hazardous and Solid
Waste Amendments (H SWA) of 1984 greatly expanded the
scope and requirements of RCRA.
RCRA regulations define hazardous wastes and regulate
their transport, treatment, storage, and disposal. These
regulations are only applicable to the ISCO process if
RCRA defined hazardous wastes are present. Hazardous
wastes that may be present include contaminated soil
cuttings and purge water generated during well installation
and development, and the residual wastes generated from
any groundwater sampling activities {e.g., PPE and purge
water). If wastes are determined to be hazardous
according to RCRA (either because of a characteristic or a
listing carried by the waste), essentially all RCRA
requirements regarding the management and disposal of
this hazardous waste will need to be addressed by the
remedial managers.
Wastes defined as hazardous under RCRA include
characteristic and listed wastes. Criteria for identifying
characteristic hazardous wastes are included in 40 CFR
Part 261 Subpart C. Listed wastes from specific and
nonspecific industrial sources, off-specification products,
spill cleanups, and other industrial sources are itemized in
40 CFR Part 261 Subpart D. RCRA regulations do not
apply to sites where RCRA-defined wastes are not present.
Unless they are specifically delisted through delisting
14
-------
procedures, hazardous wastes listed in 40 CFR Part 261
Subpart D currently remain listed wastes regardless of the
treatment they may undergo and regardless of the final
contamination levels in the resulting effluent streams and
residues. This implies that even after remediation, treated
wastes are still classified as hazardous wastes because
the pre-treatment material was a listed waste.
For generation of any hazardous waste, the site
responsible party must obtain an EPA identification
number. Other applicable RCRA requirements may
include a Uniform Hazardous Waste Manifest (if the waste
is transported off-site), restrictions on placing the waste in
land disposal units, time limits on accumulating waste, and
permits for storing the waste.
Requirements for corrective action at RCRA-regulated
facilities are provided in 40 CFR Part 264, Subpart F and
Subpart S, which also generally apply to remediation at
Superfund sites. Subparts F and S include requirements
for initiating and conducting RCRA corrective action,
remediating groundwater, and ensuring that corrective
actions comply with other environmental regulations.
Subpart S also details conditions under which particular
RCRA requirements may be waived for temporary
treatment units operating at corrective action sites and
provides information regarding requirements for modifying
permits to adequately describe the subject treatment unit.
2.8.3 CAA
The CAA establishes national primary and secondary air
quality standards for sulfur oxides, particulate matter,
carbon monoxide, ozone, nitrogen dioxide, and lead. It
also limits the emission of 189 listed hazardous pollutants
such as vinyl chloride, arsenic, asbestos and benzene.
States are responsible for enforcing the CAA. To assist in
this, Air Quality Control Regions (AQCR) were established.
Allowable emission limits are determined by the AQCR, or
its subunit, the Air Quality Management District (AQMD).
These emission limits are based on whether or not the
region is currently within attainment for National Ambient
Air Quality Standards (NAAQS).
The CAA requires that treatment, storage, and disposal
facilities comply with primary and secondary ambient air
quality standards. The most likely air emissions that would
be anticipated with an activity associated with ISCO would
be VOC emissions generated during drilling activities. The
ISCO process also uses a generator during injections.
However, these potential emissions would typically be very
low concentrations and are easily monitored on-site.
2.8.4 CWA
The objective of the CWA is to restore and maintain the
chemical, physical and biological integrity of the nation's
waters by establishing federal, state, and local discharge
standards. If treated water is discharged to surface water
bodies or Publicly Owned Treatment Works (PGTW), CWA
regulations will apply. A facility desiring to discharge water
to a navigable waterway must apply for a permit under the
National Pollutant Discharge Elimination System (NPDES).
NPDES permits include waste discharge requirements.
Discharges to POTWs also must comply with general
pretreatment regulations outlined in 40 CFR Part 403, as
well as other applicable state and local requirements.
Since XDD's chemical oxidation process is in situ and
purge water generated during the demonstration was
containerized and properly disposed of, CWA criteria did
not apply for this demonstration.
2.8.5 SDWA
The SOW A of 1974, as most recently am ended by the Safe
Drinking Water Amendments of 1986, requires the EPA to
establish regulations to protect human health from
contaminants in drinking water. The legislation authorized
national drinking water standards and a joint federal-state
system for ensuring compliance with these standards.
The National Primary Drinking Water Standards (NPDWS)
are found in 40 CFR Parts 141 through 149. Parts 144 and
145 discuss requirements associated with the underground
injection of contaminated water. If underground injection
of wastewater is selected as a disposal means, approval
from EPA or the delegated state for constructing and
operating a new underground injection well is required.
The contaminated groundwater at the former MEG Building
site is not considered a source of drinking water. However,
if the groundwater was to be used for drinking purposes
while providing no additional treatment, the quality of the
water would need to meet NPDWS. These are much more
stringent than State of New Hampshire (NH) 0.5% aqueous
solubility cleanup criteria specific to the former MEC
Building site. Table 2-3 provides a comparison between
these site specific criteria and federal drinking water
standards.
2.8.6 OSHA
CERCLA remedial actions and RCRA corrective actions
must be performed in accordance with the OSHA
requirements detailed in 20 CFR Parts 1900 through 1926,
especially Part 1910.120, which provides for the health and
safety of workers at hazardous waste sites. On-site
15
-------
construction activities at Superfund or RCRA corrective
action sites must be performed in accordance with Part
1926 of OSHA, which describes safety and health
regulations for construction sites. State OSHA
requirements, which may be significantly stricter than
federal standards, must also be met.
Table 2-3. Groundwater Cleanup Criteria Comparison.
Analyte
PCE
TCE
cDCE
VC
Toluene
1,1,1-TCA
1,1 -DCA
0.5% aqueous solubility
- State Regulatory Goal
for the former MEC
Building site (ug/L)
750
5,500
17,500
NA
2,750
6,800
27,500
National Primary
Drinking Water
Standards (ug/L)
5
5
70
2
1,000
200
—
If working at a hazardous waste site, all personnel involved
with the installation of wells and implementation of the XDD
treatment process are required to have completed an
OSHA training course and must be familiar with all OSHA
requirements relevant to hazardous waste sites. Workers
on hazardous waste sites must also be enrolled in a
medical monitoring program. The elements of any
acceptable program must include: (1) a health history, (2)
an initial exam before hazardous waste work starts to
establish fitness for duty and as a medical baseline, (3)
periodic examinations (usually annual) to determine
whether changes due to exposure may have occurred and
to ensure continued fitness for the job, (4) appropriate
medical examinations after a suspected or known
overexposure, and (5) an examination at termination.
For most sites, minimum personal protective equipment
(PPE) for workers will include gloves, hard hats, steel-toe
boots, and tyvek coveralls. Depending on contaminant
types and concentrations, additional PPE may be required,
including the use of air purifying respirators or supplied air.
For an ISCO process, XDD personnel utilized coated tyvek
and respirators during the batching operation to minimize
inhalation and dermal exposure to the strong oxidizer used.
Noise levels would potentially be high only during drilling
activities involving the operation of a drill rig or DPT probe.
During these activities, noise levels should be monitored to
ensure workers are not exposed to levels above a time-
weighted average of 85 decibels over an eight-hour day.
Workers are required to wear hearing protection at noise
levels above 85 decibels. The levels of noise anticipated
are not expected to adversely effect the community, but
this will depend on proximity to the treatment site.
2.8.7 State and Local Requirements
State and local regulatory agencies may require permits
prior to implementing an ISCO technology and/or for
specifically treating chlorinated ethenes. Specific to the
State of New Hampshire, XDD was required to acquire an
injection permit prior to implementing their treatment.
Most federal permits will be issued by the authorized state
agency. NH requires that a Groundwater Management
Permit be issued to a site owner or legally responsible
person to remedy contamination associated with the past
discharge of regulated contaminants, and to manage the
use of the contaminated groundwater. The state may also
require a Treatment, Storage, and Disposal (TSD) Permit
for on-site storage of hazardous waste for greater than 90
days. An air permit issued by the state AQCR may be
required if air emissions in excess of regulatory criteria, or
of toxic concern, are anticipated (highly unlikely for this
technology). Wastewater discharge permits may be
required in the unlikely event that wastewater were to be
discharged to a POTW. If remediation is conducted at a
Superfund site, federal agencies, primarily the USEPA, will
provide regulatory oversight. If off-site disposal of
contaminated waste is required, the waste must be taken
to the disposal facility by a licensed transporter.
For the Demonstration, there were cleanup standards for
both the soil and groundwater. The groundwaterstandards
were based on the solubility of DNAPL and were
presented in Table 2-2 as compared to U.S. EPA's MCLs
fordrinking water. The State of NH soil standards, referred
to as S-1 Soil Standards, are as follows:
PCE
TCE
cDCE
VC
Toluene
1,1 -TCA
2 mg/Kg
0.8 mg/Kg
2 mg/Kg
0.4 mg/Kg
100 mg/Kg
42 mg/Kg
16
-------
Section 3.0
Economic Analysis
3.1 Introduction
The purpose of this economic analysis is to estimate costs
for chlorinated ethene remediation of soil and groundwater
utilizing a specialized application of in situ chemical
oxidation (ISCO) developed by Xpert Design and
Diagnostic's (XDD) LLCof Stratham, New Hampshire. The
estimated costs for this technology are based on
information provided by XDD, and observations made by
SITE Program personnel during the demonstration.
The areal extent of the Demonstration treatment system's
injection wells, monitoring wells, and soil borings was
roughly 1,200 ft2 (114 m2). Based on groundwater levels of
about 3 ft bgs (as measured in June 2005, the vertical
extent of contamination at the former MEC site was
characterized to be roughly 22 ft (i.e., 3-25 ft bgs).
Therefore the volume of groundwater and saturated soil
targeted for treatment was approximately 27,000 ft3(1,000
yd3). This size can be considered pilot-scale.
For this cost estimate a larger hypothetical site is used to
better represent full-scale remediation. Figure 3-1
illustrates a hypothetical site, having an areal extent of 100
ft x 32 ft (-32,000 ft2), or about % of an acre. Basic
characteristics of this hypothetical site are as follows:
* Similar to the former MEC Building site,
contamination consists of chlorinated compounds
(e.g., PCE, TCE, cDCE, and VC). Contaminant
concentrations and cleanup goals are the same;
* Site characterization studies has identified a long,
narrow DNAPL plume, approximately 100 ft long
and 25 ft wide (30.5 x 7.6 m) is slowly dispersing
from the source area and is following an easterly
groundwater flow pattern;
* The saturated soils occur from about 18-50 ft (5.5-
15.2 m) bgs and overlay a relatively impervious
bedrock. Therefore, aquifer thickness at this
specific location is approximately 32 ft (9.8 m). The
volume of saturated aquifer material to be treated
is roughly 100,000 ft3 (100 ft x 32 ft x 32 ft).
Site geology is similar to the demonstration site,
but does not contain the thin peat layer that
required an extraordinary effort to treat. Instead a
loamy soil overlies the bedrock. There is a slight
horizontal hydraulic gradient of ~ 0.04 ft/ft;
Relative to the plume, there are one upgradiant
and two downgradiant monitoring wells. However,
there are no wells set within the immediate plume
area. Therefore, oxidant injection wells are
available for monitoring.
The specific treatment design parameters for the
hypothetical site are summarized in Table 3-1.
Table 3-1. Treatment Design for Hypothetical Site.
Injection Wells
No. of wells
Well spacing
Well head ID
Screen length
Screened int.
Radius of infl.
Diam. Of infl.
10
-18 ft
2 in.
10ft
30-40 ft
10ft
20ft
Oxidant Injection
KMnO4 concentration
Flow Rate/per well
Volume injected
Mass injected
Area treated
Volume treated
Residence Time
40g/L
2gpm
56, 000 gal
(212,00(71.)
1 4,000 Ibs
(635 kg)
100,000ft2
37,000 yd3
< 1 month
17
-------
Diameters of
Influence
Injection/
Monitoring
Well
100ft
50ft
Figure 3-1 Hypothetical ISCO Site
As shown in Figure 3-1, the plume is positioned at a slight
angle and is situated between 35-45 ft bgs. Groundwater
is at about 18 ft bgs, thus the estimated thickness of the
saturated zone is approximately 32 ft (9.8 m). XDD in
some instances could use well clusters to inject oxidant at
various depths (this was done during the demonstration).
Howeverforthis hypothetical site setting the screens at 30-
40 ft (9.1-12.2 m) bgs, should enable delivery of oxidant to
the shallowest occurrences of the plume while downward
flow of oxidant should enact contact with the deepest
occurrences of the plume (i.e., entire plume thickness).
Although the plume is depicted as narrow (i.e., about 25 ft
or 7.6 m wide), dissolved phase organic compounds occur
in the groundwater surrounding the plume. Overlapping
radii of influences (depicted by the circles on Figure 3-1)
are a conservative way of completely enveloping the entire
plume and at the same time directing oxidant to
contaminated groundwater surrounding the plume.
The costs associated with implementing the XDD's ISCO
process at this hypothetical site have been broken down
into 12 cost categories that reflect typical cleanup activities
at Superfund sites. They include the following:
(1) Site Preparation
(2) Permitting and Regulatory Activities
(3) Capital Equipment
(4) Start-up and Fixed
(5) Labor
(6) Consumables and Supplies
(7) Utilities
(8) Effluent Treatment and Disposal
(9) Residuals Shipping, & Disposal
(10) Analytical Services
(11) Maintenance and Modifications
(12) Site Restoration/Demobilization
Table 3-2 presents a categorical breakdown of the
estimated costs for implementing XDD's ISCO process at
the hypothetical site, assuming a single injection eventand
monitoring consisting of a pre-injection sampling of soil
and groundwater, one intermediate groundwatersampling,
and a final post-treatment sampling of soil and
groundwater. As with all cost estimates, there are
associated factors, issues, and assumptions that caveat
specific cost values. The major factors that can affect
estimated costs are discussed in subsection 3.3. Specific
issues and assumptions made regarding site
characteristics are incorporated into the cost estimate.
They are discussed in subsection 3.4.
18
-------
Table 3-2. Cost Estimate for Full-Scale Application.1
Cost Category
1. Site Preparation
Site Clearing
Shipment of KMnO4
2. Permitting & Regulatory Activities
GW Management Permit
Other Regulatory Requirements
3. Capital Equipment
Storage Building (1 0 ft x 1 5 ft x 8 ft)
Bladder Pumps/Tubing
Pump Flow Regulator
4. Startup & Fixed
Treatability Study {oxidant dosing)
Injection Well Installation 3
5. Labor
Permit Preparation Costs
System Design
Well Installation Oversight
KM nO4 Injection (XDD)4
GW Evaluation Sampling (3 events)
Soil Evaluation Sampling (2 events)
Site Restoration/Demobilization
Report Preparation
6. Consumables and Supplies
Granular KMnO4
Potable Water (tankers)
Rental - Geoprobe + mob. /demob.
Rental - Steam Cleaner
Rental -XDD POD
Rental - Pick-Up Truck
Rental - Generator (70 KVA)
Rental - Pumps (1 sump; 2 centrifugal)
Rental - 500 gal. Poly Tanks w/mixer
Rental - H&S Equipment (PID & CGM)
Rental - GW Sampling Equipment
Other Miscellaneous Supplies 5
7. Utilities
8. Effluent Treatment & Disposal
9. Residuals Shipping & Disposal
Contaminated Solids 6
Contaminated Liquids 6
10. Analytical Services
VOCs in Soil
VOCs in Groundwater
Sample Shipments
11. Maintenance & Modifications
12. Site Restoration (Borehole grouting)
Quantity
NA
1
1
NA
1
5
1
3
10
4
40
50
120
60
120
20
40
14,000
56,000
6
6
1
1
1
1
2
3
6
NA
NA
NA
4
15
60
30
23
NA
2,400
Units
Each
Each
Each
Each
Each
Each
Each
Samples
Each
Hours
Hours
Hours
Hours
Hours
Hours
Hours
Hours
Ibs
Gallons
Day
Day
Week
Week
Week
Week
Week
Week
Day
NA
NA
NA
Drums
NA
Each
Each
Each
NA
Feet
Unit Cost $ Total $/Category
$0
$1,000
$1,000
$0
$1,000
$600
$1,100
$1,500
$2,000
$80
$80
$80
$60
$60
$60
$60
$80
$2.30
$0.10
$1,400
$150
$4,500
$225
$700
$135
$105
$350
$122
$1,700
NA
NA
$90
$145
$130
$150
$50
$0
$2
Total
$1,000
$0
$1,000
$1,000
$1,000
$0
$5,100
$1,000
$3,000
$1,100
$24,500
$4,500
$20,000
$29,900
$320
$3,200
$4,000
$7,200
$3,600
$7,200
$1,200
$3,200
$56,300
$32,200
$5,600
$8,400
$900
$4,500
$225
$700
$135
$210
$1,050
$730
$1,700
$0 $0
$0 $0
$2,500
$360
$2,175
$13,500
$7,800
$4,500
$1,150
$0 $0
$4,800 $4,800
Estimated Cost $139,000
% of Total
0.7
0.7
3.7
18
21
41
0
0
1.8
9.7
3.4
100
Based on the hypothetical site characteristics described in subsection 3.1, one injection application, and a treatment period of at least 1 month.
Cost value totals are rounded to three significant digits; % of total values are rounded to two significant digits.
! Includes installing 2 inch ID PVC wells to 40 ft bgs using a geoprobe with DPT. Well completion materials consist of road boxes.
1 Includes one day to set up and five days for KMnO4 injection.
' Includes such items as PPE, well connect fittings, calibration gases, and shipping charges for rental equipment.
' Solids include drill cuttings, residual sample, sample liners, visqueen, and PPE. Liquids include well development water and sample purge water.
19
-------
The basis for costing each of the individual 12 categories
in Table 3-2 is discussed in detail in subsection 3.5. Much
of the information presented in that subsection has been
derived from observations made from the SITE
demonstration that was conducted overan approximate 11
month period at the former MEC Building site in Hudson.
NH. Other cost information has been acquired through
subsequent discussions with XDD representatives,
information gathered from www.XDD-llc.com, and
researching current estimates for specific cost items
related to the technology.
It should be emphasized that the cost figures provided in
this economic analysis are "order-of-magnitude" estimates,
generally + 50% 1-30%.
3.2 Conclusions
(1) The cost to install a 10-injection well ISCO
treatment system, utilizing XDD's POD to deliver
oxidant to approximately 100,000 ft3 (3,700 yd3) of
DNAPL-contaminated soil and groundwater, and
monitor effectiveness over a 1-month period is
estimated at $139,000. If further treatment were
required (i.e., re-injection), thus extending the
treatment period, the cost would increase by a
considerable amount.
(2) The largest cost categories for the application of
the XDD ISCO technology at a site having
characteristics similar to those described for the
hypothetical site are 1) consumables and supplies
(41%) and 2) labor (21%), together accounting for
62% of the total cost. The other major costs, as
estimated, include startup and fixed (18%),and
analytical services (9.7%).
(3) The cost of implementing XDD's process may be
less or more expensive than the estimate given in
this economic analysis, depending on several
factors. Such factors may include the depth and
areal extent of the contaminated media, site
geology, contaminant concentration levels, the
level of site preparation required, number of
injection/monitoring wells needed to be installed,
and the cleanup goals and process monitoring
required by a regulatory agency.
3.3 Factors Affecting Estimated Cost
There are a number of factors that could affect the cost of
treating soils and groundwater contaminated with
chlorinated VOCs using XDD's ISCO technology. It is
apparent that the number of injection wells required to
deliverthe oxidantto affected areas, the quantity of oxidant
needed, and the number of wells required for monitoring
the treatment have very significant impacts on up-front
costs. The contaminant distribution pattern has the
largest impact on the number of injection wells required to
attain a sufficient area of oxidant coverage to contact and
degrade the contaminants to acceptable levels. Soil
humic content is a critical factor for determining the types
and quantity of oxidants used, which would obviously
directly affect treatment cost as well.
Site hydrogeology is an important cost consideration,
since it can dictate how well the oxidant will work, the
ease of which oxidant can be injected and how many
injection points may be needed; regardless of whether
those points are injection wells, drive points, infiltration
galleries, orsimple pits. The amountand spacing needed
for injection and the desired screened depth intervals
determine if such injection methods were feasible. In
general, increased drilling locations and deeper drilling
directly lead to increased drill footage costs, increased
decontamination costs, and increased costs for well
construction materials.
3.4 Issues and Assumptions
This section summarizes the major issues and
assumptions used to estimate the cost of implementing
the ISCO at full-scale. In general, the assumptions are
based primarily on information provided by XDD and
observations made by SAIC during the SITE
demonstration.
3.4.1 Site Characteristics
Site characteristics are an important consideration for
deciding whether an ISCO process is an appropriate
remedyfortreating chlorinated ethenesataparticularsite.
In general in situ processes are more appropriate when
contaminated soils and groundwater are at depth; which
makes excavation and disposal too costly. Site geology
must be well defined to determine areas of high humic
content. Specific aspects of the groundwaterflow regime
(i.e., flow direction and rate), as well as plume delineation,
need to be known prior to installing oxidant injection wells.
The following specific assumptions have been made
regarding the site characteristics of the hypothetical site.
1. The target contaminated soil hotspots have been
characterized with CPT/MIP technology, as was
the case with the demonstration.
2. Groundwater occurs at about 18 ft (5.5 m) bgs
and has a well-defined flow pattern to the north.
3. Groundwater at the site is contaminated primarily
with chlorinated ethenes (PCE and TCE) as both
DNAPL and a dissolved phase mobile plume.
Also present as breakdown products are cDCE
and VC. Concentrations of these compounds
20
-------
well exceed their respective cleanup standards.
Other contaminants {e.g., metals) either are not
present at the site or have inconsequential
concentrations.
4. Site soils are assumed to have a low to moderate
humic content, ensuring that the majority of the
injected KMnO4 will react with contaminants and
not with naturally occurring humic material.
5. The site is easily accessible and secured;
therefore there is no need to install special access
roads, security fencing, etc. is not required. Also,
there are no overhead, surface, or underground
impediments that would interfere with borehole
drilling and oxidant injection.
3.4.2 Design and Performance Factors
The most important design aspects of ISCO processes is
the selection of oxidant, oxidant dosage, and the injection
well network (i.e., the number, depth, and areal pattern of
injection wells) required for optimum treatment. Injection
well spacing is usually dictated by the site geology.
The following assumptions are made regarding the
injection well network installed at the hypothetical site.
(1) Ten injection wells, spaced ~ 18 ft {5.5 m)apart in
two rows, will provide sufficient coverage of
contaminated media above action levels.
(2) All injection wells are constructed of 2 inch ID PVC
casing and 10 ft-long, 40 slot well screens. The
well screens are set at 30-40 ft bgs, above the top
of bedrock at about 50 ft {15 m) bgs. Wells are
flush mounted with road boxes.
(3) As was the case with the demonstration site, push-
points are not feasible. Therefore, a hollow stem
auger (HSA) rig is required to drill 4% inch ID
boreholes and set the wells.
(4) Potassium permanganate (KMnO4) is used as the
oxidizing agent of choice. A total of 15,000 gal.
{56,800 L) of KMnO4 is to be injected once into the
contaminated area {1,600-1,700 gallons or6,100-
6,400 L per well), at a concentration of 40 g/L.
(5) Treatment duration is assumed to be at minimum
of one month. Additional treatment (i.e., re-
injection) is not considered for costing purposes,
however monitoring is mandated for one year.
3.4.3 Financial Assumptions
All costs are presented in Year 2006 U.S. dollars (unless
otherwise noted) without accounting for interest rates,
inflation, or the time value of money. Insurance and taxes
are assumed to be fixed costs lumped into the specific
costs under the "Startup and Fixed" category. Any
licensing fees and site-specific royalties passed on by the
developer, for use of any proprietary injection equipment
or methods, would be considered profit. Those fees are
not included in the cost estimate.
3.5 for Economic Analysis
In this section, each of the 12 cost categories that reflect
typical clean-up activities encountered at Superfund sites,
are defined and discussed. Combined, these 12 cost
categories form the basis for the detailed estimated costs
presented in Table 3-1. The labor costs are grouped into
a single labor category (subsection 3.5.5).
3.5.1 Site Preparation
Site preparation includes all activities necessary for
preparing the site for installing injection wells and injecting
of the KMnO4 oxidant. Included in this setup phase are
non-labor cost for conducting any clearing and/or
regrading of the site, hooking up utilities, providing for the
storage of oxidant, and staging of equipment and
supplies. Each of these site setup cost components is
discussed in the following paragraphs.
3.5.1.1 Site Clearing
Although the XDD ISCO process does not require a large
amount of setup space, such in situ processes typically
involve subsurface drilling in some form. General
requirements apply to all drilling operations. All
impediments and surface, overhead and underground
obstacles must be identified. These would include trees,
utility lines and piping, sewers, drains, and landscape
irrigation systems.
No site clearing of above-ground obstructions or regrading
was necessary at the demonstration site, however, a
survey of potential underground utility lines was required
to obtain a "dig safe" clearance before drilling activities
could be initiated. For this economic analysis, an
assumption is made that for the hypothetical site that
there is no impediments in the area of the proposed ISCO
injection and monitoring operations. Thus, there is no
associated cost for site clearing.
3.5.1.2 Site Setup
Site setup typically includes making arrangements to
secure a trailer, hooking up utilities, etc., and shipping
supplies. If the treatment is being implemented at an
active facility (such as the former MEC Building site),
there may be no need for a site trailer. For the
hypothetical site, this also will be the case. However, due
to the large quantity of oxidant to be used a small building
or shed will be needed to store the many jugs of KMnO4
oxidant away from facility workers.
21
-------
Because XDD is able to powerthe POD, pumps, and other
equipment with a large portable generator, there is no need
for hooking up electricity. As a result, the non-labor costs
associated with site setup phase would most only include
shipment of oxidant (as explained below).
Shipment of Oxidant
After the injection wells have been installed, the KMnO4 is
shipped in bulk and properly stored at the site. (Note:
granular KMnO4 has a reported shelf life of one year and
should be stored in a dry place). The KMnO4 is packaged
and shipped in 20 Liter (5.3 gal.) jugs. Each bucket
contains approximately 55 pounds of KMnO4. For the
hypothetical site, it is anticipated that a minimum of 250
buckets will be required to ship approximately 14,000 Ibs
(635 kg) of KMnO4. For large quantities such as this, the
buckets are stacked and secured on wooden pallets that
can be transferred to the batching area with a forklift.
XDD commonly uses a specific carrierforshipping oxidant
and has estimated thatshipping approximately 7,800 Ibs of
granular KMnO4 to the former MEC Building site cost
approximately $900. Forth is costestimate, a shipping cost
of $1,000 will be used.
3.5.2 Permitting and Regulatory Requirements
3.5.2.1 Permitting Requirements
Several types of permits may be required for implementing
a full-scale ISCO remediation. Permits required may
depend on the type and concentration of the contamination,
the regulations covering the specific location, and the site's
proximity to residential neighborhoods. For example, the
New Hampshire Department of Environmental Services
(NHDES) requires a site owner or legally responsible
person to remedy contamination associated with the past
discharge of regulated contaminants, and to manage the
use of contaminated groundwater.
A Groundwater Management Permit (GMP) from the State
is required for such endeavors. The application for the
GMP must be prepared and stamped by a professional
engineer (PE) or a professional geologist (PG) licensed in
the State. There is an application fee of $1,000.
Specific to ISCO processes there is in most cases, if not
all, a requirement to obtain an injection permit. There may
or may not be a fee for this. For the former MEC Building
site, XDD was required to apply forand acquire an injection
permit from the NHDES. However, there was no actual
permit fee. Note that other states may require a fee.
3.5.2.2 Other Regulatory Requirements
The costs incurred for receiving approval from regulatory
agencies to install a treatment system may include those
associated with preparing site characterization reports and
the feasibility study for treatment system design, and
attending meetings with regulators for discussing
comments and supplying related documentation for
acquiring approval for implementing the treatment
technology.
Depending upon the classification of the site, certain
RCRA requirements may have to be satisfied as well. For
active Superfund sites it is possible that the technology
could be implemented under the umbrella of existing
permits and plans held by the site owner or other
responsible party (e.g., the GMP). Certain regions or
states have more rigorous environmental policies that
may result in higher costs for permits and verification of
cleanup. Added costs may result from investigating all
regulations and policies relating to the location of the site;
and for conducting a historical background check for fully
understanding the scope of the contamination. Specific to
ISCO technologies, states may require that injection wells
be properly abandoned/decommissioned if they are no
longer to be used.
Due to the very site-specific nature of these costs, an
assumption will be made that sufficient pre-existing site
information exists for regulators to allow initiation of
treatability and/or pilot-scale field studies. As a result, no
further costs regarding regulatory requirements will be
incurred.
3.5.3 Capital Equipment
In many instances ISCO processes consist of one-time
applications at a particular contaminated area. As a
result, the equipment for injection is required for a short
period of time and is thus rented (see 3.5.6). Even if
multiple injection applications are necessary, they would
not occur immediately following an initial injection.
Therefore, the equipment would be re-mobilized and
rented again for additional injection events.
For the particular hypothetical site example used for this
economic analysis, it is assumed that the only capital
equipment required is an outdoor shed to store oxidant,
and dedicated bladder pumps used for groundwater
monitoring before and after treatment. (Note: a peristaltic
pump was used forcollecting demonstration groundwater
samples because water was pulled from less than 25 ft
bgs. Since groundwatercontamination atthe hypothetical
site is below the maximum sample depth of peristaltic
pumps (i.e., > 25 ft bgs), other alternatives must be used.
The storage shed must be large enough to contain
approximately 250 jugs of KMnO4. If the injection were to
take place in cold temperatures, the shed would need to
22
-------
be heated to prevent freezing of the granular oxidant.
However, for this scenario, this is not the case). The
installation of the prefabricated shed for the hypothetical
site is estimated to be $1,000,
Although bladder pumps and tubing are relatively
expensive, once installed these dedicated pumps 1) assure
that there is no air/water contact during sampling; and 2)
they eliminate the need to decontaminate sample collection
equipment between wells and reduce the chance of cross-
contamination or the introduction of decontamination
chemicals into the groundwater. In essence, much of the
capital expenditure related to the use of dedicated bladder
pumps is recouped by reduced labor costs.
Teflon bladder pumps with stainless steel housings can be
purchased for about $500. Along with associated tubing,
the cost of each pump is about $600. Five of the ten
injection wells are to be sampled to monitor treatment
effectiveness. Therefore, the total cost for five bladder
pumps is estimated to be $3,000. A pump flow controller,
estimated to cost $1,100, is required to regulate
compressed air as a cycle of pulses that corresponds to a
desired groundwater flow rate out of the well. The high
rental cost of this equipment justifies its purchase if used
for several sampling events. The total cost for capital
equipment is estimated to be approximately $4,100.
3.5.4 Startup and Fixed Costs
Startup and fixed costs include those costs that must be
incurred before treatment can commence. They are one
time non-recurring costs in which labor is typically
imbedded within a flat fee fora task. Based on information
provided by XDD and SITE demonstration observations,
these costs for full scale ISCO applications include
primarily 1) initial treatability testing; 2) installation of
injection wells and, if needed, 3) pilot-scale testing.
Estimates for all three of these startup and fixed costs are
discussed in the following subsections.
3.5.4.1 Treatability Study
Initial treatability testing is necessary for determinating
whether the technology is feasible a particular site. XDD
performed a two-part bench-scale treatability study prior to
the demonstration; 1) A dosing study determined
permanganate dosing level and exposure period for
evaluating process effectiveness and assessment of
feasibility of attaining site-specific cleanup goals; 2) A soil
oxidantdemand (SOD) study to estimate SOD forthe three
major soil units. The results of the treatability study helped
XDD develop a site-specific strategy for applying ISCO
treatment at the former MEC Building site.
Per XDD, the SOD determination is always conducted.
The cost for XDD treatability studies run approximately
$1,500 per sample. Because there are no distinct soil
zones at the hypothetical site, an assumption is made that
three samples of the site soils are sufficient for
determining SOD, even though the site is four times larger
than the demonstration site. Therefore, treatability study
costs are estimated at $4,500.
3.5.4.2 Injection Well Installation
The amount and location of oxidant injection points and
the number of points required for monitoring
effectiveness is highly site-specific. The number of
injection wells required to produce an adequate ISCO is
relatively high. Close injection well spacing is typically
needed in heterogeneous soils to ensure adequate
coverage. For the hypothetical site a total of 10 injection
wells, each screened from 30-40 ft (9.1-12.2 m) is
assumed sufficient to inject the volume of oxidant needed
to treat 3,700 yd3 of contaminated material.
XDD prefers to inject into 2-inch ID wells installed within
a 4Vz or 6-inch ID borehole. It should be noted that direct
push technology (DPT) can be used to install wells and
has the advantage of minimizing generation of
contaminated soil cuttings. However, using traditional well
installation methods involve auguring of larger boreholes
that result in larger diameter annular spaces between the
well casing and borehole sides. This makes for easier
installation of a filter pack and bentonite seal. For the
hypothetical site, each of the 10 injection wells is to be
installed using HSA to drill 41/2-inch boreholes.
An all inclusive well installation cost of $50 per foot will be
assumed for this cost estimate (i.e., cost to include
mobilization, demobilization, drilling, well materials and
well development). The $50 rate would correlate to
drilling costs of $2,000 per well (i.e., there are 10 wells
total to be installed at a depth of 40 ft bgs. Therefore, the
estimated total injection well drilling cost is $20,000.
The total startup and fixed costs for this economic
analysis is thus estimated to be approximately $24,500.
3.5.4.3 Pilot-Scale Testing
In addition to the bench-scale treatability study of
formation material and sample groundwater, pilot-scale
testing may be appropriate in certain instances. XDD
sometimes conducts water injection testing on prospective
injection wells, a practice they are utilizing more
frequently. As an example, forthe demonstration project
at the former MEC Building site, XDD performed a small-
scale water injection test at a target flow rate to monitor
well seal and injection pressures of newly-installed
injection wells. The costs of such field tests are almost
exclusively labor (see 3.5.5).
23
-------
3.5.5 Labor
This subsection describes the core labor costs that are
associated with the XDD ISCO technology. The hourly
labor rates presented are loaded, which means they are
intended to include base salary, benefits, overhead, and
general and administrative (G&A) expenses. Travel, per
diem, and standard vehicle rental have not been included
in these figures. The labor tasks have been broken down
into subcategories that represent distinct phases of
technology implementation (Table 3-3).
Table 3-3. Estimated Labor Costs.
Category
Permit Preparation
System Design
Injection Well Installation
Oxidant Injection
Treatment Monitoring
Site Restoration/Demobilization
Report Preparation
Totals
Hours
4
40
50
120
180
20
40
454
Cost
$320
$3,200
$4,000
$7,200
$10,800
$1,200
$3,200
$29,920
3.5.5.1 Permit Preparation
As discussed in subsection 3.5.2 there is a fee for certain
permits. However, researching permit requirements and
preparing applications takes time and thus incurs a labor
cost. If a groundwater management permit and injection
permit are to be required for the hypothetical site the
applications would have to be submitted by either a
professional engineer (PE) or professional geologist (PG)
licensed in the particular state that the site is located in.
Assuming that both applications are prepared and
submitted by a PE or PG in two hours at $80/hr, permit
preparation costs can be estimated at $320.
3.5.5.2 System Design
In most all instances system design is conducted by the
remediation contractor, in this case XDD. The information
that XDD needs for system design are the 1) amount of
contaminants present (i.e., mass and extent) and 2) Site
hydrogeological information. For the SITE demonstration
the system design was a joint effort among EPA, SAIC,
XDD, the site owner and their remediation contractor.
Although the demonstration project was smaller in scale
than the remediation scenario used for this economic
analysis, the experimental design for evaluating treatment
effectiveness was relatively complex.
Per XDD, it takes roughly 1-2 people any where from 2-5
days to complete a proposed treatment strategy,
depending on complexity. For this cost estimate, the
primary labor task relating to system design would involve
a senior level scientist or engineer reviewing existing site
characterization data for the hypothetical site, reviewing
results from SOD treatability testing of site samples,
calculating oxidant mass and concentration, and
determining location, depth, and numberof injection wells
needed to adequately treat the contaminated DNAPL
plume. Assuming a 40-hr week sufficient to accomplish
this task, and a labor rate of $80/hour, an estimated labor
cost of $3,200 would be incurred for system design.
3.5.5.3 Injection Well Installation Oversight
Although drilling and well installation labor activities are
performed by a drilling contractor, the remediation
contractor (e.g., XDD) at a site would be responsible for
logging boreholes, monitoring for VOCs and explosive
conditions, and ensuring that well construction and
installation is conducted in accordance with design
specifications. It should be noted that since the
hypothetical site is adequately characterized, geologic
descriptions would not be required when drilling the
boreholes for injection wells.
During the demonstration, it took approximately four days
fora subcontracted driller to mobilize to the site, install six
injection wells to an average drilling depth of 16 ft using a
hollow stem auger (HSA), conduct flush mount well
completion with road boxes, develop the wells, and
demobilize. Assuming that the drilling subcontractor is
local and that there are only minor drilling impediments,
the time required to mobilize, install 10 injection wells 40
feet through unconsolidated sediments with similar well
completion, and demobilize is estimated to take
approximately one week. Assuming 10-hour days are
required forsubcontractoroversight, a geologist's laborat
an $80/hour rate would result in $4,000 in oversight labor.
3.5.5.4 Oxidant Injection
Assuming the site is within a % day or less driving
distance (i.e., the demonstration site), mobilization and
equipment setup can be done in one day by two people.
The following tasks comprise the basic setup of
equipment.
Oxidant delivery/storage of KMnO4
Setup of spill guards for batch area
Potable water delivery (tanker)
Drop-off of rented generator
Hookup of batch operation to POD
Connect injection distribution system to wells
24
-------
Injection of KMnO4orotheroxidant, as observed during the
demonstration, involves three basic steps; 1) a batching
operation in which granular KMnO4 is mixed with potable
water to form an oxidant solution at the desired
concentration; 2) pumping the KMnO4solution from one or
more mixing vessels through a series of cannister filters
and onto the POD; and 3) pumping oxidant from the POD
down the wells via high pressure hoses.
During the demonstration, the injection rates for individual
wells were varied on a regular basis to control seepage of
oxidant. Flow rates during the second injection event
ranged from 1.7 to 2.6 gpm for individual wells, the higher
injection rates were typically for the deeper wells. The
injection event took two people four days to inject a little
over 1,900 Ibs (862 kg) of KM nO4 into a total of 10 injection
wells. At this same rate, it would take the same two
individuals 28 days to batch, transfer, and inject over seven
times that amount of KMnO4 (14,000 Ibs or 635 kg).
However, all of the wells at the hypothetical site are set in
a formation similar to the demonstration site's deep zone,
which was able to take a large volume of oxidant (i.e., XDD
was able to inject about 480 Ibs (218 kg) of oxidant daily
into the deep wells. Assuming XDD could inject via their
POD the same amount into six wells at the hypothetical site
simultaneously, they could inject a total of about 2,900 Ibs
(1,320 kg) of oxidant daily.
At this rate, XDD could inject all 14,000 Ibs of oxidant in
five full days. If a day of mobilization and setup is included,
the injection event could be completed in six days and
would take two people and a total of 120 hours to
complete. Thus, at a $60 per hour rate for the estimated
120 labor hours, the labor cost for the KMnO4 injection
phase would total $7,200.
3.5.5.5 Treatment Monitoring
As previously discussed, the contamination in the soil and
groundwater at the hypothetical site has been assumed to
be fully characterized prior to installation of the XDD ISCO
injection wells. For monitoring the technology
effectiveness on both the soil and groundwater, at leasttwo
soil and groundwatersampling events are required. A pre-
injection "baseline" event is typically desired to assess pre-
treatment concentration levels for both soil and
groundwater. Intermediate events, conducted after
treatment, but prior to the estimated treatment completion
time, are used to determine the rate of treatment progress
and to assess any trends. Final post-treatment monitoring
is required to determine if treatment goals have been met.
For the hypothetical site, a reasonable scenario for
collecting treatment verification samples would be to collect
groundwater samples from five of the 10 injection wells a
week or two before the KMnO4 injection date to establish
a true pretreatment groundwater baseline. At about the
same time approximately 10 soil samples, randomly-
selected within the 100 ft x 32 ft x 32 ft defined
contaminated zone, would be collected to establish a
baseline soil concentration. Following the first two weeks
of treatment, an intermediate groundwatersampling of the
same five wells would be conducted to evaluate
groundwater contaminant trends. Finally, after at least
one month has passed since the end of the injection date,
a final post-treatment sampling of both soil and
groundwater would be conducted to verify treatment
effectiveness. These six sampling events are
summarized as follows.
1. Pre-lnjection Soil (Baseline)
2. Pre-lnjection Groundwater (Baseline)
3. Post-Injection Groundwater (Intermediate)
4. Post-Treatment Soil (Final)
5. Post-Treatment groundwater (Final)
Groundwater Sampling
For this cost analysis, it will be assumed that the two-
person sampling team can mobilize to the site, setup,
purged and sample the five wells, ship the samples to an
off-site laboratory, and demobilize in one 10-hour day.
Therefore, each of the three groundwatersampling events
would incur 20 hours of technician labor at $60/hr; or
$1,200. Thus, from the time just prior to KMnO4 injection
until one year following the KMnO4 injection, a labor cost
of $3,600 would be incurred for groundwater sampling.
Soil Sampling
Collection of soil samples to determine the technology
effectiveness was difficult and time consuming during the
demonstration. Two events, a pre-treatment baseline and
final post-treatment, are deemed sufficient so long as
enough samples are collected to provide a statistically
significant sample set.
A geoprobe was used for collecting soil samples during
the final sampling event at the demonstration site. A total
of 30 samples was collected from each of three lithologic
zones (for a total of 90 samples). However, Although the
hypothetical site is larger there is only one contaminated
zone. Therefore, an assumption is made that 30 soil
samples from this zone prior to treatment and again one
month after treatment will provide an adequate statistical
sample set for evaluating treatment effectiveness.
For the demonstration, sampling at 30 locations took
approximately four days, however three discrete depth
zones were sampled. Since there is essentially one
affected zone at the hypothetical site, it is assumed that
samples from the 30 locations can be collected in three
10-hour days. Thus, combining baseline and post-
treatment verification sampling (i.e., 60 samples), six 10-
25
-------
hour days are assumed sufficient to complete all soil
sampling. Other than the geoprobe crew, two technicians
at $60 per hour can collect, prepare, and ship the on-site
methanol-extracted soil samples in this time period. This
would equate to $1,200 of labor per day or a total of $7,200
of total labor for soil sampling.
Therefore, total treatment monitoring labor costs
(groundwater plus soil) is estimated at $10,800.
3.5.5.6 Site Restoration/Demobilization
For the XDD ISCO process, the primary activity for
demobilizing from the demonstration site involved a
thorough flushing of the oxidantfrom the POD and related
equipment components. To accomplish this, XDD uses a
neutralizing solution that consists of a 5:1:1 mixture of
potable water, vinegar, and hydrogen peroxide (H2O2),
respectively. The neutralizer is pumped through the POD
components and all of the high pressure hoses and
injected down the injection wells {the small amount of
neutralizer has a negligible effect on the subsurface).
For the demonstration, the demobilization effort took
approximately one day for two people to complete.
Therefore, it is assumed that demobilization will take two
people, working a ten-hour day (20 hours total) to
demobilize. At a loaded rate of $60/hr per technician, the
total demobilization labor cost is estimated at $1,200.
3.5.5.7 Report Preparation
Labor for report preparation can be quite variable,
depending upon the complexity of the data and client
expectations. Forthis economic analysis an assumption is
made that XDD will prepare a brief final report containing
the basic information and data needed for a third party to
evaluate. The report would include the following
information at a minimum.
Brief description of the site (including a site map);
Re-summarizing of the treatability study results;
Injection Summary (volume, mass, dosages, etc.)
Map of Injection points, wells, and borings
Groundwater sample results for all three events;
Soil sample results for both events;
Results of quality assurance samples;
Brief interpretation of results and conclusions.
It is estimated that it would take a senior level engineer or
scientist a week (i.e., 40 hours) to prepare such a report.
At a loaded rate of $80/hr, report preparation is estimated
at $3,200.
3.5.6 Consumables & Supplies
Consumables and supplies for implementing a full-scale
ISCO process comprise a significantamountof total costs.
They are estimated at $56,350 for this economic analysis.
Consumable and supply costs can be segregated into
three separate subcategories: 1) Consumables (i.e.,
materials consumed or used up); 2) Equipment Rentals;
and 3) Miscellaneous Supplies. Each of these
subcategories is discussed separately in the following
subsections.
3.5.6.1 Consumables
The two major consumable items associated with XDD's
ISCO process are 1) the oxidant itself (KMnO4) and 2)
potable water that is needed to mix with the oxidant to
form an injectable solution at a desired oxidant
concentration. Note that water is considered a
consumable in this instance (as opposed to a utility).
KMnO,,
The injected KMnO4is considered a consumable because
it is consumed after injection. During the demonstration,
XDD injected approximately 15,000 gal. (57,000 L) of
KMnO4 solution into about 1,000 yd3 of contaminated
material. The 3,700 yd3 of contaminated material at the
hypothetical site is 3.7 times greater in volume. Therefore,
there would need to be about 56,000 gal. (212,000 L) of
oxidant/water solution injected at a similar dosage, or
5,600 gal. of solution injected per the 10 injection wells.
The bulk purchase price of treated granular KMnO4 is
reported by XDD to be $2.30 a pound. Using a KMnO4
concentration of 30 g/L, approximately 14,000 Ibs (7 tons)
of granular KMnO4 would be required to produce 56,000
gallons of oxidant solution. At the $2.30/lb price, the cost
of granular KMnO4 is estimated to be $32,200.
Potable Water
The potable water that is used to make up the oxidant
solution is also a consumable item. During the
demonstration XDD utilized 5,000 gal. (18,900 L) tankers
to supply potable water at the site at a cost of $0.10 per
gallon. This cost included delivery and use of the tanker.
The cost of 56,000 gallons of potable water is therefore
estimated at $5,600.
3.5.6.2 Equipment Rental Costs
For this economic analysis equipment rentals include
costs for non-capital equipment required to perform four
basic functions. These include 1) Direct Push Technology
(DPT) for conducting baseline and post-treatment soil
sampling; 2) Equipment for injecting oxidant; 3)
Equipment for health & safety monitoring during field
activities; and 4) Equipment for conducting groundwater
sampling. Each of these rental categories is discussed
separately as follows:
26
-------
Soil Sampling
Final event soil sampling at the demonstration site was
accomplished with DPT employed by a geoprobe.
Although this work is almost always subcontracted out, the
use of the geoprobe is costed out on a daily basis (i.e.,
rented equipment). The cost of the geoprobe, including a
daily mobilization charge for a local drilling company, is
estimated at $1,400/day. During the demonstration, a
geoprobe was able to collect 90 discrete soil samples from
30 locations in about four days. If 30 locations are
sampled at a depth of about 40 feet bis to collect both
baseline and post-treatment soil samples at the
hypothetical site, it will be assumed that soil sampling can
be completed in six days. Therefore, the total geoprobe
rental cost is estimated at $8,400.
Directly associated with geoprobe soil sampling is steam
cleaning of DPT samplers and push rods to prevent cross
contamination between boreholes. Steam cleanerrental is
estimated at $150 per day, or $900 for the six-day event.
Oxidant Injection
The primary rented item for injecting oxidant is XDD's
Portable Oxidant Delivery System (POD). XDD's POD is
a trailer-mounted unit specifically designed for injecting and
monitoring delivery of oxidant at up to six injection points
simultaneously. The unit's primary components include
batch tanks, mixers, a metering pump, a distribution
manifold, and chemically-resistant digital flow meters that
measure flow rates and total flow volumes. The POD can
be rented on a daily basis for $1,500 or $4,500 weekly.
During the demonstration, XDD was injected up to 1,000
gal. (3,800 L) of oxidant solution into deep zone wells {i.e.,
25 ft bis) in a 10-hour day. Since the POD can deliver
oxidant to six wells at a times, it is assumed that XDD
could inject 6,000 gal. (22,700 L) of oxidant solution per
day at the hypothetical site. Therefore, rental of the POD
is estimated at $4,500.
Directly associated with the POD is a rented pickup truck,
used to transport the POD and supplies, and to mobilize
personnel to and from the site on a daily basis. The truck
is rented out at $225 per week. Other rented equipment
needed by XDD for injection include a 70-KVA generator
($700/week), transfer pumps (a combined $135/week), and
two 500 gallon mixing tanks (a combined $210/week).
Together, these items add up to an additional $1,270/week.
Health and Safety Monitoring
A photoionization detector (PID) and combustible gas
indicator (CGI) are standard health and safety
requirements for drilling operations (i.e., well installation
and soil sampling). The PID can also be used to screen
drill cuttings to aid in the determination of disposal options.
Drilling operations should consist of just two events; 1)
one week to install wells immediately followed by three
days of soil sampling, and 2) three days of post-treatment
soil sampling. A PID and CGI can be rented for $200 a
week and $150 a week, respectively (i.e., a combined
$350 per week. A conservative time frame estimate for
use is three weeks for both events. Therefore, the cost
of renting these field monitoring instruments is
conservatively estimated at $1,050.
Groundwater Sampling
Groundwater sampling during the demonstration was
conducted with a rented peristaltic pump. However, as
previously discussed, purchased bladder pumps are to be
used at the hypothetical site for sampling groundwater
below 25 ft bgs (see 3.5.3). Sampling equipment is still
required for low flow purging and sampling of wells. The
largest rental cost is for use of a multi-parameter water
quality meter. This instrument combines the
measurement capabilities of several instruments and can
be rented for $100 per day (includes a flow-through cell).
In addition to the multi-parameter meter, a water level
indicator is required for recording water levels during the
low-flow purging/sampling. A water level indicator can be
rented for $22/day, thus the combined rental equipment
needed is $122/day. For the six days required (i.e., 3
events x 2 days), total rental costs are estimated at $730.
3.5.6.3 Miscellaneous Supplies
Miscellaneous supplies could include a whole array of
items that by themselves are relatively insignificant in
cost, but combined would add up to a sizable cost
needing consideration. These items may include, but are
not limited to, the following items shown in Table 3-4.
Table 3-4. Estimated Miscellaneous Supplies Costs.
Item
Soil Sample Liners
Decon. Containment
Personal Protective Equip.
Calibration gases/solutions
Well couplings
Rental Shipping charges
Unit Cost
$10/Each
$100/Each
$20/day
50/Each
$2/Each
$80/Each
No.
60
1
20
2
10
6
Total Estimate
Total
$600
$100
$400
$100
$20
$480
$1,700
As shown in Table 3-4, total miscellaneous supplies are
estimated at $1,700, the largest component being soil
sample liners used for the geoprobe sampling.
27
-------
3.5.7 Utilities
The predominant utility required forXDD's oxidant injection
is the electricity required to operate the POD and
associated pumps. Electricity is also required for powering
an aircompressorduring groundwatersampling. Certainly,
the proximity of the demonstration site to a readily available
facility would make this a minor issue. However, at a
remote site, logistics can get complicated. It may even be
necessary to use a gasoline or diesel powered generator
if electrical hookup is not practicable.
A small additional electrical cost may be needed to supply
lighting to the trailer and a security light, and possibly for a
phone and facsimile hookup. Otherthan electricity, a water
source may be needed for occasional decontamination
activities; however, those costs are considered negligible.
During the demonstration, both electricity and water was
available from the former MEC Building. Forthis economic
analysis, the same is assumed for the hypothetical site.
Thus, no costs for utilities are included.
3.5.8 Effluent Treatment and Disposal
For this technology there is no effluent. Therefore, it is
assumed that there will be no effluent treatment and
disposal expense. Disposal of small amounts of
decontamination wastewater generated from cleaning
sampling equipment is considered negligible and therefore
no costs are included for this category.
3.5.9 Residuals Shipping and Disposal
The ISCO technology generates essentially no waste
streams per se. However, investigation derived wastes
(IDW) are typically generated when installing injection and
monitoring wells, and when collecting soil and groundwater
samples for evaluating treatment effectiveness.
The IDW generated for an ISCO remediation can be
subdivided into two broad categories; 1) Waste Solids, and
2) Waste Liquids. During the demonstration, waste solids
consisted of soil cuttings from HSA drilling for well
installation, residual soil sample material, and
miscellaneous waste solids (i.e., visqueen and wood from
the decontamination pad, PPE, used sample liners, etc.).
Waste liquids consisted of well development water,
sampling equipment decontamination water, and purge
water generated during groundwatersampling.
For the hypothetical site it is assumed that DPT will be
used for both well installation and soil sampling (i.e., not
HSA). Therefore, no soil cuttings will be generated.
However, there will be residual soil sample material. Since
only about % of one drum of residual sample material was
generated for the 90 samples collected during the
demonstration final event, it is assumed that just one drum
of residual material will be generated from the 60 samples
collected for the hypothetical site sampling scenario. As
was the case with the demonstration, three additional
drums of waste solids (e.g., sample liners, visqueen, PPE,
etc.) will be assumed generated for the hypothetical site
scenario. Disposal of waste solids cost $90 per drum.
Therefore, the total cost of disposing the four waste solid
drums is estimated at $360.
Based on the demonstration IDW, generation of waste
liquids is anticipated to substantially exceed waste solid
generation in volume. During the demonstration about2.6
drums of well development water was generated for six 2-
inch ID wells representing about 90 feet of water column.
The water column for each of the 2-inch ID hypothetical
site wells is 22 feet, thus a total 220 feet of water column
should generate about 370 gallons of development water,
or about seven 55-gallon drums.
Also, during the demonstration steam cleaning of DPT
equipment used to bore about 1,200 feet of soil generated
four 55-gallon drums of decontamination water. For the
hypothetical site a total of 60 samples will be collected
from an approximate 35-ft depth, which equates to
roughly 2,100 ft. of boring footage. This would equate to
1.75 times the demonstration volume of decontamination
water generated and would require seven drums for
containment. Purge water from both baseline and post-
treatment events should add another drum. All in all there
are an estimated 15 liquid drums requiring disposal.
Disposal of waste liquids cost $145/drum. Therefore,
disposal costs for the 15 drums of waste liquids are
estimated at $2,175. Total disposal costs for the
hypothetical site (solids + liquids) is estimated at
approximately $2,530. It should be noted that there is a
minor charge for pickup and manifesting, which will be
considered negligible for this cost estimate. Table 3-5
below summarizes the estimated disposal costs.
Table 3-5. Estimated Residuals Shipping/Disposal Costs.
Item
Unit Cost
No.
Total
Waste Solids
Residual Soil Sample
Miscellaneous Solids
$90/Drum
$90/Drum
1
3
$90
$270
Waste Liauids
Well Development Water
Decontamination Water
Sample Purge Water
$145/Drum
$145/Drum
$145/Drum
7
7
1
Total Estimate
$1,015
$1,015
$145
$2,535
28
-------
3.5.10 Analytical Services
The level of testing required to substantiate successful
ISCO treatment at full scale (i.e., at the hypothetical site) is
assumed to be significantly scaled down from the SITE
Demonstration sampling plan. For this cost analysis, a
treatment period of one month is assumed and the two-
event soil and three-event groundwatersampling schedule
discussed previously (see 3.5.5.4) is considered adequate
to monitor and evaluate treatment effectiveness.
Although the site owner or the site owner's contractor
would likely collect these samples, the state or local
regulatory agency may require in dependent analysis of the
samples by an outside laboratory (especially for final
post-treatment samples).
It is also being assumed thatfor both soil and groundwater
the only required analytical parameter is for volatile organic
compounds (VOCs); for soil SW-846 Method 5035/8260B
(including total solids analysis for soil samples to report
results on a dry weight basis) and SW-846 8260B for
water. VOC analyses are essential since specific VOCs
are the target contaminants. Other analyses, such as
metals, are considered optional (i.e., metals' analyses may
provide insight as to whether the ISCO process mobilizes
metals or not, however metals are not considered a
concern at the hypothetical site.
Table 3-6 provides an estimate for the cost of analytical
samples for the hypothetical site sampling scenario.
Table 3-6. Estimated Analytical Costs 1
Item
Unit Cost
No.
Total
Soils 2
Baseline Event
Post-treatment Event
$130/Sample
$130/Sample
30
30
$3,900
$3,900
Groundwater
Baseline Event
Intermediate Event
Post-treatment Event
$150/Sample
$150/Sample
$150/Sample
10
10
10
$1,500
$1,500
$1,500
Sample Shipments
Soil3
Groundwater
$50/Shipment
$50/Shipment
20
3
Total Estimate
$1,000
$150
$13,450
1 Costs include QC sample analyses.
2 Total solids analysis included for reporting dry we ght soils
3
-------
4.0
4.1 Introduction
4.1.1 Project Background
XDD's ISCO process was evaluated under the EPA SITE
Program at the former MEC Building site located in
Hudson, NH. Soil and groundwater at this site are
contaminated with DNAPL and dissolved-phase
contaminants, such as TCE, cDCE, PCE, 1,1,1-TCA,
1,1-DCA , toluene and VC. The contamination originated
from releases from an underground concrete holding tank.
The tank was removed in March 1997. Concentrations
within the site soils were measured prior to the
demonstration as high as the percent range. The overall
goal of the study was to evaluate the ability of XDD's ISCO
process to reduce levels of specifically targeted organic
compounds in contaminated soil by 90% and to reduce
targeted groundwater contaminants to below remediation
performance standards set specifically for the site.
This pilot-scale study was initiated in August 2004 (pre-
demonstration activities) and concluded in May 2006. The
study focused on a small contaminated area located just
downgradient of the former concrete tank excavation site.
Figure 4-1 shows this demonstration study area, including
locations of the demonstration injection wells (IW),
evaluation wells (EW), and soil boring locations. The areal
extent of this area was about 1,200 ft2 (370 m2). The
vertical extent of site contamination is roughly 22 ft or 6.7
m {i.e., 3-25 ft bgs). Therefore the volume of groundwater
and saturated soil targeted for treatment was 27,000 ft3
(8,200 m3or 1,000 yd3).
A total of two soil and five groundwater sampling events
were conducted during the demonstration. A pre-treatment
baseline and post-treatment final soil sampling event were
conducted in which approximately 90 soil samples were
collected from three stratigraphic zones. For groundwater,
emphasis was placed on the final post-treatment sampling
in which 15 wells were samples twice to produce 30
sample results. A day of well recovery was spaced
between the two sampling rounds. Both the soil and
groundwater samples were analyzed forVOCs and other
parameters of interest. Results of VOC analyses were
evaluated statistically to determine if primary project
objectives were met.
4.1.2 Project Objectives
Specific objectives for this SITE demonstration were
developed and defined prior to the initiation of field work.
These objectives were subdivided into two categories;
primary and secondary. Primary objectives are those goals
that support the developer's specific claims for the
technology demonstrated. These objectives are usually
evaluated using both descriptive and inferential statistical
analyses and require quantitative results to draw
conclusions regarding technology performance.
Secondary objectives are also in support of developer
claims, however, the data analysis associated with these
objectives are considered less rigorous. Secondary
objectives pertain to information that is useful, and do not
necessarily require the use of quantitative results to draw
conclusions regarding technology performance.
Critical data support primary objectives, and non-critical
data support secondary objectives. Critical measurements
were formally evaluated against the demonstration target
level using statistical hypothesis tests that are summarized
in subsection 4.5.
Table 4-1 presents the two primary and four secondary
objectives of the demonstration, and summarizes the
method(s) by which each was evaluated. Objectives 1-4
are addressed in this section. Objective 5 was not
evaluated because adequate permeability data was not
acquired from the CPT/MIP survey. The cost estimate
(Objective 6), is discussed in Section 3.
30
-------
Former MEG Building
N
27
IW-1
Cluster
26
IW-2
Cluster
30
LEGEND
= Baseline Boring Location
= Boring/Multi-Chamber Well Location
(Note: Borings 7, 8, & 11 are also Evaluation Wells
EW-2, EW-3, and EW-4, respectively)
= Injection Well Cluster well, each consisting of
shallow (s), intermediate (I), and deep (d) wells.
Scale
29.
10ft
20
10
23
IW-3
Cluster
11
Figure 4-1. Map of Study Area Showing DemonstrationSoil Boring and Well Locations.
31
-------
Table 4-1. Demonstration Objectives.
Objective
Description
Method of Evaluation
Primary Objectives
Objective 1 Determine if there is a statistically significant
reduction of VOCs in soil (specifically chlorinated
ethenes) over the period of the demonstration;
and specifically show that the XDD oxidation
process can remove 90% of the VOCs from soil
for all three contaminated soil horizons.
Determine for each horizon and each individual compound
(PCE, TCE, cDCE and VC) the removal efficiency by
comparing the analysis of soil samples taken before and
after the demonstration test period (baseline and final
sampling events). Note: XDD's claim was specific to
chlorinated ethenes but the SITE Program also assessed the
impact of the technology on chlorinated ethanes.
Objective 2 Determine whether the XDD oxidation process
can reduce concentrations of contaminants in the
groundwater to below the 0.5% solubility limit of
750 ug/L, 5,500 ug/L, and 17,500 ug/L for PCE,
TCE, and cDCE, respectively. These remediation
performance standards are not for drinking water;
they were provided by the state of New Hampshire
to specifically address this site.
Determined by analysis of groundwater samples collected
during the final post-treatment sampling episode; in which
each sample is considered spatially and temporally
separated such that a statistically significant number of
samples are collected for analysis.
Secondary Objectives
Objective 3 Assess the Impact of the Organic Matter in the
Peat and its effect on Oxidant Depletion.
Collect samples of peat material at each boring location
during baseline and final sampling events and determine on
average if there is a depletion of humic content in samples
due to the ISCO process.
Objective 4 Evaluate the potential for mobilization of valence
sensitive toxic metals (i.e., As, Cr, Se, Zn) into the
groundwater system.
Sample and analyze groundwater within and downgradient
of the source area for metals at designated periods
throughout the course of the demonstration.
Objective 5 Evaluate the effect of MnO4 on permeability
reductions in the contaminated media due to the
formation of MnO2. Rapid buildup of MnO2 can
occur when treating with MnO4 where high levels
of DNAPL saturation occur, leading to pore
plugging and an overall reduction in permeability.
Performing in situ permeability tests before and after
treatment within the three stratigraphic zones, (i.e., fill
material, peat, and sandy-silt).
Objective 6 Collect and compile information and data
pertaining to the cost of implementing the XDD
ISCO process.
Acquire cost estimates from past SITE experience, and from
the developer (XDD). Cost treatment for full-scale treatment
of similar contaminated material. Breakdown estimates into
12 cost categories that reflect typical cleanup activities at
Superfund sites (See Section 3).
32
-------
4.2 Site Description
4.2.1 Site Location and History
The approximate four-acre former MEC property contains
a 36,000 square foot former MEC building and associated
paved and unpaved parking and storage areas. MEC
(formerly Nashua Electric Motors) repaired and rebuilt
electric motors at the property from the mid 1970s up until
approximately 1990. During this time frame, releases from
an underground concrete holding tank has contaminated
soil and groundwater in the vicinity of the southeast corner
of the former MEC Building.
Aries Engineering, Inc. (Aries) is the remediation contractor
for the former MEC Building site. Per Aries, the concrete
holding tank was excavated in May of 1997. A total of
1,175 gallons (4,400 L) of liquids and five 55-gallon drums
of sludge was removed from the tank (personal
communication between Aries and SAIC, June 2006).
Currently, Aries is conducting monitoring activities at the
site in accordance with the site New Hampshire
Department of Environmental Services (NHDES)
Groundwater Management Permit (GMP), which was
issued December 5, 1997 and renewed on January 17,
2003. There are specific remediation goals for VOC
contaminants at the former MEC Building site, both for
soils and groundwater. These goals are referred to as
Remediation Performance Standards (RPS) and are
presented in Table 4-2.
Table 4-2. Remediation Performance Standards.
VOC
PCE
TCE
cDCE
VC
Toluene
1,1,1-TCA
1,1 -DCA
Soil Standards
(mg/Kg) '
2
0.8
2
0.4
100
42
3
Groundwater
Standards (ug/l) 2
750
5,550
17,500
—
2,570
6,800
27,500
1 State of NH S-1 standards (Aries Engineering, Inc. July 2001).
2 State of NH 0.5% agueous solubility goal ( SAIC, May 2005).
Attainment of the RPS via treatment, such as the XDD
ISCO process, may allow for the utilization of natural
attenuation as a follow-on treatment option (personal
communication between Aries and SAIC, June 2006).
4.2.2 Site Lithology and Hydrogeology
4.2.2.1 Lithology
Characterization of the demonstration study area lithology
has resulted from several field investigation efforts; most
notably the drilling of numerous soil borings and installation
of several injection and monitoring wells. Site-specific
lithology and geology were acquired via descriptions of 87
soil samples collected from the 30 soil borings shown in
Figure 4-1. An approximate equal number of samples
were collected from each of the three stratigraphic zones
targeted for the ISCO treatment. The shallowest sample
collection intervals were 6-7 feet below land surface (bis).
(The upper 0-6 ft bis consists of asphalt underlain with
rubble and debris; this material was not sampled).
Figure 4-2 is a generalized cross section showing distinct
stratigraphic zones comprising the demonstration study
area. These include an: 1 )upper debris zone, 2) the three
zones that were sampled for characterization and chemical
analysis, 4) a basal till, and 5) underlying bedrock. Brief
descriptions of these zones are as follows:
Upper Fill/Debris Layer 0-6 ft bis (Not Sampled)
The upper 6 feet of the demonstration site consists of fill
material, including large chunks of concrete, rocks,
plywood, shredded wood, etc. This zone was notsampled.
Top Gravelly Sand Zone 6-13 ft bis (Sampled)
Just below the upper fill material, there is a zone of loose
gravelly-sand, with silt; which has an olive to grey color.
This zone extended to about 6-13 feet bis. Samples from
this zone were saturated since groundwater was typically
just 3>2-4 feet bis. Wood fragments were frequently
encountered in this zone and hydrocarbon odors were
noted for several boring samples. A total of 30 baseline
and 30 final soil samples were collected from this zone.
Middle Peat Zone 12-14 ft bis (Sampled)
At this approximate depth interval there exists a relatively
thin peat zone, black in color, that varies in thickness from
0.2 ft. to 3.2 ft (0.6-0.98 m). At several locations the peat
was party composed of shredded wood fibers or wood
chunks; and contained twigs, leaves, and bark. The peat
occurred as shallow as 9 ft bis and as deep as 15.8 ft bis.
Based on baseline and final borehole drilling, the peat was
absent in boring location No. 30. Boring 30 is located on
the south edge of the study area (Figure 4-1), not far from
the former tank location. The peat may have been
removed during excavation of the holding tank and
surrounding soil material. A total of 27 baseline peat
samples and 29 final peat samples were collected from
this zone.
33
-------
sw
Feet
BLS IW-11
5 Feet
IW-1d
EW-2
EW-3
IW-3d
NE
10.
15.
20-
25"
Peat
J8,9'
Grawelly-Sandy
CSravelly
Coarse
Sand
24.7'
Fill
¥ery Fine
Sand - Silt
8.5'U
Grawelly
Coarse
Sand
Stiff Clayey-Silt
(till)
Fill
Grawelly-Sandy
Very Fine
- Silt
Figure 4-2. of Study
34
-------
Lower Sandy-Silt Zone 14-26 ft bis (Sampled)
A sandy-silt zone, olive-yellow to grey in color, extends
from the bottom of the peat down to a basal till layer
(anywhere from about 10-28 ft or 0.3-8.5 m bis). Typically
this zone grades from a very course or gravelly-sand to a
finer sandy-silt to pure silt. Product (i.e., DNAPL) was
commonly observed in this zone. A total of 30 baseline
and 30 final soil samples were collected from this zone.
Basal Till Layer 26-30 ft (Not Sampled)
This layer was not sampled but provided a reference point
for collecting lower zone samples (i.e., the most
contaminated portion of the overlying sandy silt was
sampled). PID readings typically were the highest just
above this till layer.
Bedrock 30 ft (Not Sampled)
Bedrock occurs roughly at 30 ft (9.1 m) bis and reportedly
consists of a highly weathered quartz-biotite schist.
Bedrockwas encountered when drilling IW-3d ataround 30
ft (9.1 m) bis and at boring location No. 9 at about 31 ft bis.
4.2.2.1 Hydrogeology
Site overburden groundwater generally flows in a
northeasterly direction across the entire former MEC
Building area. Horizontal hydraulic gradient, based on
Aries's 2003 and earlier site groundwater data, is
approximately 0.04 feet/foot (ft/ft) or 0.12 m/m between two
monitoring wells located in the immediate vicinity of the
study area. In addition, site vertical gradients have been
measured by Aries (e.g., Aries observed vertically upward
hydraulic gradients ranging between 0.04 ft/ft to 0.1 ft/ft in
a monitoring well couplet upgradient of the study area).
4.3 Pre-Demonstration Activities
For this demonstration, a significant amount of time and
effort was devoted to pre-demonstration activities (Phase
1). These activities included preliminary sampling for site
characterization (e.g., monitoring well installation; and soil
and groundwater sampling), laboratory methanol (MeOH)
extraction studies on soil samples, a treatability study
conducted by the ISCO developer (XDD), and cone
penetrometer technology and membrane interface probe
(CPT/MIP) survey. Each of these pre-demonstration
activities is discussed in the following subsections.
4.3.1 Preliminary Sampling
SAIC conducted pre-demonstration sampling during the
latter part of August, 2004. The purpose of the preliminary
sampling was twofold:
2,
the target VOCs within the context of the lithoiogic
and hydraulic regimes of the site media and;
Acquire samples for an XDD treatability study.
1.
Gain a better understanding of the distribution of
Three contaminated soil zones (gravelly sand, peat, and
silty sand) were sampled at each of three locations and
three sets of nested monitoring wells (nine total) were
installed at the same locations. Eight soil samples (3
gravel/sand, 2 peat, and 3 silty sand) were collected from
separate boreholes near an existing well located in the
immediate vicinity of the demonstration site. Also, one
groundwater sample was collected from that same well.
Results of the pre-demonstration sampling indicated that
contaminant levels at the site were sufficient for purposes
of the demonstration. Samples collected from an existing
monitoring well indicated that concentrations of cDCE were
well above the RPS. However, TCE and PCE were below
these standards. Although based on just sampling of one
well, the data suggest that oxidation may have been
in-process in the sampled groundwater, as the more
chlorinated compounds were detected at lower
concentrations (below the RPS) and one of the known
breakdown products (e.g., cDCE) was at a much higher
concentration.
4.3.2 MeOH Extraction Studies
Based upon previous experience with peat and several
published laboratory studies, it had been anticipated that
VOCs would be difficult to extract from the site peat
material using the standard 1:1 MeOH to soil ratio. As a
result, a series of MeOH extraction studies was conducted
on peat and other material collected from the site, with
native concentrations of contaminants. A detailed
description of the extraction study procedures and results
are presented in the QAPP (SAIC, May 2005).
The difficulty with a 1:1 MeOH to soil extraction occurs
because the contaminants reach an equilibrium with peat
and MeOH. Thus, significant contamination remains on
the peat even after two or three subsequent extractions. In
order to find an optimum extraction procedure for the peat
material, a set of experiments was performed by extracting
the peat in MeOH following standard SW-846 protocol (i.e.,
soxhlet extractions). Additional studies varied the
procedure by using sonication and heating. Resulting data
suggested that much more contamination was present in
the peat than that suggested by the initial test results.
Further studies reinforced the notion that sonication and
heating of the MeOH extract removed some additional
VOC contamination from the peat, other then what was
found in the original extraction using standard SW-846
procedures. However, in every case it appeared that some
contamination remained in the peat material. Even after
35
-------
the final extraction, the procedure continued to recover and
remove some portion of the contamination. Overall the
results of the extraction study suggested that standard
SW-846 procedures would not be robustenough to remove
the VOCs present in the peat at the site.
An additional extract sample, extracted in a 10 to 1 ratio of
MeOH to peat instead of the usual 1 to 1 ratio, resulted in
removal of almost all the contaminants of interest in the
very first extraction. This suggested that the equilibrium
between the MeOH and peat was affected by increasing
the MeOH volume in contact with peat material.
Subsequentstudies were then undertaken to determine the
ideal MeOH to peat ratios for the demonstration.
Results of these subsequent studies suggested that a
higher concentration of MeOH to peat was needed for the
peat instead of the standard 1:1 ratio. The drawback of
this increased ratio is increased detection limits.
Results of these pre-demonstration analytical studies were
implementedforthedemonstration. Forthedemonstration
a 5:1 milliliters (ml) of MeOH to grams of soil, was used for
the middle peat and lower sandy-silt zones. A 1:1 ml of
MeOH to grams of soil was used for the top gravelly-sand
zone.
4.3.3 Treatability Study
A laboratory treatability study was conducted by XDD prior
to the demonstration in order to evaluate the effectiveness
of their ISCO process on contaminated soil from the site.
The objectives of this treatability study were to:
Assess the overall feasibility of using ISCO to
meet the site-specific cleanup goals;
Estimate the soil oxidant demand (SOD) for the
major soil units (i.e., gravelly sand, peat layer, and
silty sand) in the treatment area;
Develop site-specific data necessary to design an
ISCO field pilot test and/or full-scale application.
The treatability study is detailed in a separate report
prepared by XDD (XDD, January 2005). Permanganate
(MnO4) was the chosen oxidant for the study. MnO4 is an
oxidizer used extensively in wastewater treatment and
drinking water purification processes. MnO4 is a very
stable oxidant and can persist for several months in the
subsurface. This stability makes it a good choice for
subsurface applications (i.e., fewer injection events, fewer
wells to treat the target area, and the ability to more
effectively penetrate into low permeability soil zones).
MnO4 also reacts with natural soil organic matter and
reduced metal oxides (e.g., iron and manganese oxides,
etc.). Contaminantdegradation rates are influenced by the
presence of these competing species. SOD is one
measure of these additional demands that consume
oxidant. Because the SOD exerts a competitive demand
on the oxidant, it must typically be satisfied to ensure
complete oxidation of the VOC compounds.
The individual soil samples that were used in the SOD
testing were combined into three composite soil samples
representative of the three major soil units at the site: 1)
gravelly sand, 2) peat, and 3) sandy silt.
Laboratory testing consisted of two separate evaluations:
1. SOD testing for evaluating utilization rates of
variable MnO4concentrations in a soil/groundwater
system, and
2. Testing the destruction efficiency of the VOCs.
SOD Testing
Eight representative soil samples were evaluated for the
SOD procedure, which involved preparing eight separate
40 ml VOA vials (batch reactors) with approximately 10
grams (g) of each soil sample. This produced eight vials
for each of the eight soil samples (64 total).
The reactor vials were then dosed with KMnO4 oxidant
solution at four different concentrations; approximately 2,
5, 10, and 20 grams per liter (g/L) KMnO4 prepared using
distilled water. Two reactor vials were prepared at each
oxidant concentration for duplicating analyses on each soil
sample. Controls, consisting of four MnO4 solution
concentrations (plusfourduplicate control samples with no
soil added), were created at the start of the test and
analyzed to evaluate if there were losses of MnO4 over the
duration of the test.
All the reactor vials and controls were allowed to equilibrate
over a fourteen-day test period. On Days 3, 7, 10, and 14,
aqueous samples from each vial were measured by a
colorimeter for oxidant solution concentration change. A
second dose of oxidant was applied to peat samples on
Day 17 and samples from those peat reactors were
analyzed for oxidant concentration changes on Days 24,
28, and 34. XDD (January 2005) reported SOD test results
as follows:
XDD Averaged SOD Results (g/Kg)
Zone Sample 1 Sample 2 Sample 3
Shallow Gr. Sand 21.0 4.6 8.8
Middle Peat 104.7 — 146.9
Deep Siity Sand 5.6 1.8 2.3
36
-------
VOC Destruction Efficiency Testing
VOC destruction efficiency testing involved placing a
specific volume of groundwater/soil in a concentrated MnO4
solution; and then placing the solution into a batch reactor
(40-ml glass VOA vial) for a specified exposure period.
Approximately 38 ml of 20 g/Loxidant{as KMnO4) solution
was prepared for site groundwater {from MW-10) and 10
grams of each soil sample were added into each batch
reactor. Six identical batch reactors for each soil
composite were prepared (18 reactor vials total) as follows:
Four reactor vials to be sacrificed at four different
time intervals for the analysis of VOCs;
One duplicate reactor for VOCs analysis; and
One reactor vial used to monitor changes in
oxidantconcentration throughoutthe testduration.
In addition to the above reactors, three "no oxidant" control
reactors (one persoil composite) were prepared identically
to the other reactors. The controls were used to assess
losses in VOC concentrations due to mechanisms other
than chemical oxidation (e.g., volatilization, biodegradation,
etc.). To minimize volatilization losses, all batch reactors
were completely filled with either oxidant solution or
groundwater to zero headspace.
In the peat composite, MnO4 appeared to be very effective
in destroying high levels of cDCE and VC. Also, it appeared
that most of the cDCE was destroyed after the first dose of
oxidant. PCE and TCE were not evaluated due to very
low/non-detect results in the control samples. Toluene
exhibited reasonable reduction rates in the peat composite
even though this compound typically reacts slowly with
MnO4. Interestingly, the chlorinated ethanes (1,1,1-TCA
and 1,1-DCA) also appeared to exhibit significant
reductions. Available literature suggests that these
compounds do not react directly with MnO4. However,
there may be other reactions occurring in the presence of
the oxidant that may cause degradation (i.e., other reactive
species may be forming, other abiotic processes, etc.).
Nearly complete destruction of cDCE, VC, and toluene was
observed in the shallow composite. TCE, present at
relatively low concentrations in the aqueous control, was
nonetheless reduced to non-detectable levels. PCE was
not detected in control samples, thus it was not evaluated.
Partial destruction of 1,1,1-TCA, and 1,1-DCA was also
observed in the aqueous phase of the shallow composite.
However, soil concentrations of 1,1-DCA were higher in all
of the treated reactor vials as compared to the soil control
sample (approximately 2 to 7 times higher). This trend for
1,1-DCA was also observed in the deep composite soil
samples (discussed in the next section).
Nearly complete destruction of cDCE, VC, and toluene was
observed in the soil and aqueous phases of the deep
composite. Significant reduction of TCE was also
observed in the Deep Composite tests. The Deep
Composite control was the only sample in the study to
exhibit significant aqueous TCE concentrations. The
aqueous TCE concentration in the control sample was
1,780 ug/L. This was significantly higherthan the aqueous
baseline TCE level (40.4 ug/L). Based on this, it appeared
that the aqueous phase TCE in the control resulted from
soil desorption. Although not certain, it seemed further
supported by the fact that the deep composite soil control
was the only control that exhibited detectable TCE on the
soils, although relatively low (152 ug/Kg).
Consistent with other tests, significant destruction of
1,1,1-TCA, and 1,1-DCA was also observed in the aqueous
phase samples. However, soil concentrations of 1,1-DCA
in the Deep Composite test appeared to increase
(approximately 3 to 8 times higher) as compared to the
control sample (similar to the Shallow Composite test).
Summary of Treatability Testing
Soils were mixed together to create three separate soil
composites representative of each of the three major
stratum at the site (i.e., shallow gravelly sand composite,
peat composite, and deep silty sand composite samples).
The soil composites were dosed with approximately 20 g/L
KMnO4solution. The peat composite was dosed twice due
to the high oxidant utilization rates.
PCE was not present in any of the soil composite reactors,
and as such could not be evaluated. The remaining VOCs
(including TCE, cDCE, VC, 1,1,1-TCA, 1,1-DCA, and
toluene) were present in one or more of the samples and
were therefore evaluated for destruction efficiency. Only
cDCE and toluene were detected in the studies at levels
above the anticipated clean-up goals for the site. These
two compounds were reduced to below the clean-up levels
during the treatability study.
Aqueous TCE, cDCE, and VC were effectively destroyed
in all tests as anticipated. Even in the peat composite
where SOD was very high, the cDCE and VC were
destroyed rapidly and during the initial dose of oxidant.
This is a positive result since there was some concern that
VOCs would be difficult to treat in the presence of such a
high oxidant demand.
Chlorinated ethanes (1,1,1-TCA and 1,1-DCA) are typically
not thought to be directly oxidizable by MnO4. However,
test results indicated that there was up to 75% reduction of
1,1,1-TCA and 64% reduction of 1,1-DCA in the aqueous
phase deep composite sample. As anticipated, reduction
of these two ethanes in the peat composite samples were
37
-------
slightly less efficient (49% reduction of 1,1,1-TCA and 41%
reduction of 1,1 -DCA in the aqueous phase). The actual
mechanisms for destruction of these compounds in the
presence of MnO4 are not fully known, but this behavior
has been observed at other sites.
Toluene also was significantly reduced (aqueous phase
reduction of >99% in the Deep and Shallow Composites,
and up to 62% in the Peat Composite). Toluene is known
to degrade at relatively slow rates in the presence of MnO4.
Based on the results published by XDD's chemical
oxidation laboratory testing report (XDD, January 2005)
and the summarized results noted above, ISCO was
determined to be an effective treatment for the majority of
the targeted VOCs at the site. Although the SOD is very
high in the peat zone, significant reduction was observed
for many of the target VOCs.
4.3.4 CPT/MIP Characterization
Cone Penetrometer Technology (CPT) and Membrane
Interface Probe (MIP) was utilized for pre- and post-
treatmentcharacterization of the demonstration study area.
CPT is basically a soil conductivity logging tool used to
interpret lithology. The MIP is used to determine position
and approximate concentration of VOCs (i.e., the MIP is
not quantitative, but its detector response can be used at
a particularsite to estimate soil concentrations). These two
logging tools, developed by Geoprobe Systems, can be
combined into the same probe to collect subsurface
information (Christy, no date).
A total of 18 CPT/MIP survey points was completed prior to
baseline sampling. During this survey, CPT/MIP logs were
used to identify specific subsurface zones and the highest
concentration areas to target for soil sampling. Following
both XDD injections, a post-treatment CPT/M IP survey was
conducted that was significantly scaled down from the
baseline survey. A total of 5 CPT/MIP survey points were
completed to adequate depths. The reduced number of
post-treatment survey locations makes comparison to the
baseline data difficult. Kriging contour lines are
dramatically different due to the absence of data.
Figure 4-3 shows the maximum response of the MIP's
photo-ionization detector (PID) in micro volts (uV) for a
below-surface depth range of 0-18 ft (5.5 m), both for
baseline and final post-treatment. Figure 4-4 shows the
maximum response of the MIP's PID in uV for a below-
surface depth range of 18-36 ft (5.5-11 m), both for
baseline and final post-treatment. The baseline event plots
are collectively based on 14 MIP baseline points and the
final event plots are based on 5 final MIP points. Each MIP
survey point identified as a blue cross on the plots.
The more detailed baseline map shows that about 20% of
the study area plume extends beneath the former MEC
Building. This generally favors the use of an in situ
treatment (especially for soil), since excavation near and
beneath such a permanent structure is undesirable. The
baseline map also reveals a concentrated circular-shaped
plume, approximately 60 ft in diameter, whose center is in
the immediate vicinity of the former concrete holding tank
location. The highest uV response of 2.00E +06 uV
comprises an approximate 30 ft-diameter area. The
response rapidly dissipates beyond this area. Comparison
of baseline to final PID shallow zone plots (Figure 4-3)
show a potential increase of the maximum PID signal (vs.
baseline) at locations 2-3 and 2-5. In the deep zone
(Figure 4-4), there is a slightdecrease of the maximum PID
signal (vs. baseline) in the vicinity of locations 2-4 and 2-5.
In addition to the PID detector, the MIP is also equipped
with an electron capture detector (ECD), which is
responsive to chlorinated compounds. Figures 4-5 and 4-
6 show the maximum response of the ECD in uV for below-
surface depth ranges of 0-18 ft (5.5 m)and 18-36 ft (5.5-11
m), respectively. These baseline and final event plots are
also collectively based on the 14 MIP points (baseline) and
5 MIP points (final) identified as blue crosses.
Evaluation of the shallow zone baseline ECD plot (Figure
4-5) reveals a concentrated circular-shaped plume, similar
in size and area shown by the PID shallow zone map. The
only difference is increased intensity (i.e., an upper uV
response of 1 .OOE +07 uV for the ECD versus a response
of 2.OOE +06 uV for the PID). It should be noted that the
baseline plot in Figure 4-5 may be misleading in showing
what appears to be a second plume, detached from the
main plume that surrounds the former MEC building. The
contours that comprise the shape of this apparent plume
are interpolations from the single MIP point4-2, which may
only be an isolated contaminated spot. Comparison of the
baseline and final plots show no appreciable difference in
the maximum ECD response before and after injection
The deep zone baseline ECD plot (Figure 4-6) also shows
a concentrated plume emanating from the contaminant
source area. However, the shape of the plume is more
elliptical than circular and it is oriented in a southwest to
northeast pattern. This is consistent with groundwaterflow
direction and is similar to the Aries plume characterization
(refer back to Figure 1-1). Possibly, the deeper zone map
is showing DNAPL that has settled along the top of the
basal till layer and bedrock and is very slowly dispersing to
the northeast. When comparing the baseline and final
plots, there is no appreciable difference between the
maximum ECD response before and after injection.
38
-------
101940
101920-
101900
101880
101S60
101B40
101820-%!-
PID Response. uV
2.00E +06
2.00E+06
2.00E +06
2.00E +06
1.00E+06
1.00E+06
1.00E+06
9.00E +05
7.00E +05
5.00E +05
3.00E+05
5.00E +04
101940
101920
101900
101860
101840
3Sf
1061700
1061640
1061640 1061660 1061680
Baseline
Figure 4-3. Maximum MIP PID Response for Shallow Zone Soil (0-18 ft) - Baseline Vs. Final.
1061S60 1061680
Final
1061700
-------
101S40-
101920-
101SOO-
101880-
101860
101840-
101820^
PID Response, uV
2.00E +06
2.00E +06
2.00E +06
2.00E +06
1.00E+06
1.00E+06
1.00E+06
9.00E +05
7.00E +05
5.00E +05
3.00E +05
5.00E +04
101S40
101920-
101900-
101880-
101860-
101840-
1061640 1061660 1061680
Baseline
1061700
1061640
1061660
1061680
1061700
Final
Figure 4-4. Maximum MIP PID Response for Deep Zone Soil (18-36 ft) - Baseline Vs. Final.
-------
101940-
101920-
101900-
101880-
101860
101840-
101820'
ECD Response, uV
1.00E +07
1.00E+07
1.00E+07
1.DOE+07
9.00E +06
8.00E +06
7,OOE +06
6.00E +06
5.00E +06
4.00E +06
3.00E +06
2.00E +06
1.00E +06
2.00E +05
101940-
101920-
101900-
101880-
101860-
101840
1061640
1061660
1061680
1061700
1061640
1061660
1061680
1051700
Baseline
Final
Figure 4-5. Maximum MIP ECD Response for Shallow Zone Soil (0-18 ft)-Baseline Vs. Final.
-------
101940-
101920-
101900
ECD Response. uV
1 .OOE +07
1.OOE+07
1.OOE+07
1. OOE +07
9.00E +06
8.00E +06
7.00E+06
6.00E +06
5.00E +06
4.00E +06
3.00E +06
2.00E +06
1.OOE+06
2.00E +05
101940-
101920-
101900-
101880
101850
101340
1061640 1061660 1061580 1061700
Baseline
1061640
1061660
1061680
1061700
Final
Figure 4-6. Maximum MIP ECD Response for Deep Zone Soil (18-36 ft) - Baseline Vs. Final.
-------
4.4 Demonstration Activities
4.4.1 Injection of Oxidant
XDD developed a 3-well cluster injection strategy for
treating the chlorinated ethenes in all three soil zones at
the demonstration site. Originally, a single injection event
was planned. However, the originally-installed pre-packed
one inch ID injection wells for two of the clusters had failed
seals. They were replaced by traditional 2-inch ID wells.
Thus, a second injection event was required.
Figure 4-7 is an illustration showing the locations of the
three injection well clusters (IW-1, IW-2, and IW-3), the
estimated radii of influence (10 feet for each), and
summarizes injected volumes of KMnO4 for each of the
wells and for each injection. Table 4-3 provides a more
detailed injection summary, including mass of KMnO4
injected perXDD's injection logs.
All injection wells were constructed of like materials and
had similar well diameters (IW-3 cluster wells are 114-inch
ID and IW-1 and IW-2 cluster wells are 2-inch ID). Well
screen lengths for the shallow (s) and deep (d) zones were
five feet and the screen length for the intermediate (I) a
peat zone was one foot. Well cluster spacing was roughly
18 feet.
Former MEC Building
Approximate location
- Former Concrete
Holding Tank
3-6,
Total of ~ 4,700 gallons of
KMn04 injected @ 36-40 g/L
- 1s-122 gallons
-11- 2,190 gallons
- 1d- 2,360 gallons
2nd
3-6,
Total of ~ 4,700 gallons of
Kyn04 injected @ 36-40 g/L
1st
(June 6-10,2005)
Total of ~ 4,800 gallons of
KMnO4 injected @ 25 g/L
- 3s - 647 gallons
-31-1,830 gallons
- 3d - 2,360 gallons
- 2s -',
-2i-1,960" gallons
- 2d - 2,350 gallons
LEGEND
) = Multi-Chamber Well Location
| = Injection Well Cluster well, each consisting of
shallow (s), intermediate (i), and deep (d) wells.
\= Estimated Radius of Influence - injected KMn04
Scale
10ft
Figure 4-7.
Injection Points.
43
-------
Table 4-3. Injection Summary
Zone
Shallow
Intermediate
(Peat)
Deep
Injection Well
IW-1s
IW-2s
IW-3s
Total
IW-1i
IW-2i
IW-3i
Total
IW-1d
IW-2d
IW-3d
Total
All Zones Combined Totals
First Injection
June 6-10, 2005-
second
Volume
(gallons)
3
27
647
677
33
152
1,827
2,012
222
251
2,362
2,835
5,524
Mass
(Pounds)
1
8
208
217
10
46
597
653
60
76
752
888
1,758
Second Injection
October 3-6, 2005
Volume
(gallons)
122
386
—
508
2.186
1.957
4,143
2,359
2,355
—
4,714
9,365
Mass
(Pounds)
25
78
—
103
451
403
854
486
486
—
972
1,929
Both Injections
Combined
Volume
(gallons)
125
413
647
1,185
2,219
2,109
1,827
6,155
2,581
2,606
2,362
7,549
14,889
Mass
(Pounds)
26
86
208
320
461
449
597
1,507
546
562
752
1,860
3,687
44
-------
4.4.2 Soil and Groundwater Sample Collection
Demonstration sampling and analysis began in May of
2005 with baseline soil sampling and was followed by an
initial baseline groundwater sampling event, a second
baseline groundwater sampling event, two intermediate
groundwater sampling events, and a final event for both
soil and groundwater (events and dates are listed below).
* Baseline Soil (May, 2005);
4 First Baseline for Groundwater (June, 2005);
* Second Baseline for Groundwater (Sept. 2005);
* Intermediate Groundwater (December, 2005);
* Intermediate Groundwater (February, 2006);
* Final Groundwater and Soil (March, 2006)
4.4.2.1 Soil Sampling
The two soil sampling events included a baseline event to
establish pre-treatment soil concentrations for VOCs and
soil humic content, and a final event to determine post-
treatment soil concentrations for those same parameters.
Each is discussed in detail below.
Baseline Soil Sampling
The baseline soil sampling event was initiated on May 12,
2005 completed May 19, 2005, thus preceding the first
injection event of early June 2005 by approximately three
weeks. Soil collection was accomplished with a mini-sonic
rig equipped with a direct push sampler. Boreholes were
drilled at a total of 30 closely-spaced locations within the
small area where a DNAPL plume had been identified (see
Figure 4-1). Logs from a CPT/MIP survey were used to
target hotspots at depth from which to collect soil samples.
A total of 87 of the planned 90 soil samples were collected
(i.e., 30 from each of three depth zones were planned;
however, peat samples were not able to be collected at
three borehole locations).
During baseline sampling, samples collected from the
shallow zone gravelly silty-sand material were field
extracted in a 1:1 MeOH to soil ratio by volume. Samples
collected from the intermediate peat layer were field
extracted in a 5:1 MeOH to soil ratio by volume. However,
there had still been uncertainty related to the deeper zone
silt as to which extraction ratio would be the most suitable.
As a result of this uncertainty, a field study (similar to the
pre-demonstration MeOH extraction study described in
4.3.2) was conducted during the baseline event to ensure
that field extractions of the deep zone sandy silt material
would be complete.
Three samples from three boreholes were extracted in the
normal fashion (1:1 MeOH to soil ratio) and three
duplicates were extracted and analyzed as was done in the
pre-demonstration extraction study (5:1 MeOH to soil ratio),
following previously described protocols. Data was
compared to determine if the extraction efficiency was
significantly increased using a 5:1 ratio of MeOH to soil
compared to a 1:1 ratio. In order to avoid losing data from
either increased detection limits in the 5:1 extraction or not
obtaining the actual concentration of organics as may be
lost from a 1:1 extraction, all lower horizon samples
collected during baseline sampling were extracted in the
field using both protocols (5:1 and 1:1 ratio). This required
all lower horizon samples to be collected in duplicate.
Three selected samples had both extracts analyzed.
Results of these analyses were used to determine if
additional duplicate analyses were required and to
determine the better extraction ratio for the final event.
Final Soil Sampling
The final soil sampling event was conducted in late March
of 2006, approximately ten months following the first
injection event and about eight months following the
second injection event. A total of 90 soils samples were
collected with a geoprobe equipped with a macrocore
sampler. The final event boreholes were drilled as close as
possible to the baseline borehole locations (i.e., from 0.3 to
2 feet away) and soil samples were collected from the
approximate same depth interval as were the baseline soil
samples. A 1:1 MeOH to soil ratio was used for extracting
the top zone samples and a 5:1 MeOH to soil ratio was
used for extracting peat and bottom zone samples.
4.4.2.2 Groundwater Sampling
An initial baseline groundwater sampling event was
conducted soon after the soil baseline event (i.e., the first
week in June 2005). Due to failed seals in injection well
clusters IW-1 and IW-2, all three wells for those clusters
had to be replaced, and a second baseline groundwater
event was conducted in September 2005. This event also
served as an intermediate event to evaluate groundwater
treated from the initial oxidant injection. Two more
intermediate groundwater sample events were conducted
following the second oxidant injection to evaluate target
contaminant trends in groundwater.
Groundwater was sampled at a low-flow rate using a
peristaltic pump. Where possible the purged groundwater
was routed through a flow-through cell so parameters could
be monitored with a multi-parameter direct read probe and
deemed stabilized prior to sampling. In some instances
this procedure could not be conducted due to an
insufficient volume of well water available for purging. This
was the case with the small diameter multi-chamber wells,
especially those set in the low recovery shallow and peat
formations. In instances where the flow-through cell and
probe could not be used, groundwater samples were
collected directly into the sample container following a
purge of anywhere between 250 mis to 550 mis of water.
45
-------
4.4.2.4 Laboratory Analyses
Table 4-4 summarizes the laboratory analyses conducted
on soil and groundwater samples collected during each
sampling event. Critical and non-critical measurements
were conducted per the following discussion.
Critical Measurements
Critical measurements for the study included VOCs and
moisture for soils, and VOCs for groundwater. Based on a
pre-demonstration methanol extraction study (see 4.2.2),
different MeOH to soil extraction ratios were used on
different soil types that characterized each of the three
zones. A 1:1 MeOH to soil extraction ratio was used for
the shallow zone (dominated with gravelly silty-sand
material). For the middle peat zone and the underlying
deeper sandy-silt zone (where silt was prevalent), a 5:1
MeOH to soil extraction ratio was used.
To perform the field extraction procedure, the analytical
laboratory had prepared pre-weighed sample containers
containing the proper amount of MeOH for solvating
approximately 5 grams of soil from each of the three soil
horizons. The pre-weighed containers were shipped to the
field site. Wide-mouth 4-ounce jars were used for 5:1
MeOH to soil ratio extracts and 40-ml vials were used for
1:1 MeOH to soil ratio extracts. This prevented extracting
a particular soil type in the wrong volume of MeOH solvent.
Because soil samples for VOC analyses were placed
directly into jars containing MeOH, additional sample
material from each core was also placed in a separate
container for moisture analysis. This measurement was
considered critical since the results were used to present
the VOC data on a dry-weight basis.
Groundwater samples were also analyzed for VOCs. Due
to the target compounds being chlorinated VOCs, the
samples were not preserved with acid. As a result, sample
holding times were reduced from 28 to 7 days.
Non-Critical Measurements
Non-critical measurements for soils included humate
analysis of the peat material that was located about 12-14
bis. A total of 27 samples was analyzed for each the
baseline and the final sampling events. In addition
groundwater was also measured for total metals and
specifically for bromide since that compound was a tracer
within the KMnO4 oxidant solution used for the first
injection.
Table 4-4. Summary of Laboratory Analyses.
Parameter
Test
Method
Method
Type
NO. SAMPLES ANALYZED PER EVENT ''
Baseline
Second
Baseline
First
Intermediate
Second
Intermediate
Final
Soil
VOCs
% Moisture
Hu mates
SW 5035/8260
SW 3540
CA Humic Residue
Purge & Trap, GC/MS
Dessication
Separation/TOC analyzer
87
87
27
—
—
—
Groundwater
VOCs
Bromide
Total Metals
KMn04
SW 5035/8260
EPA 300
SW 30 10/60 10
NA
Purge & Trap, GC/MS
Ion Chromatography
Acid Digestion, ICP
Visual Observation
15
15
15
—
15
15
15
15
—
—
—
12
12
12
12
—
—
—
15
15
15
15
89
89
27
30
30
30
30
; Does not include QC samples.
46
-------
4.4.3 Process Monitoring
XDD was ultimately responsible for monitoring its system
parameters during the injection phase of the
demonstration. XDD's monitoring during injection
consisted primarily of adjusting the flow rate of oxidant
based on injection pressure readings. As a qualitative way
of estimating oxidantdispersalthroughouteach of the three
soil zones, the SITE Program visually monitored KMnO4
during groundwater sampling rounds coupled with
laboratory analysis of bromide. Bromide was used as a
tracer for the first injection and was not detected in any
baseline groundwater samples collected prior to oxidant
injection (the bromide data for all events, including the
baseline sampling, is presented in Subsection 4.5).
Figure 4-8 is an illustration showing the estimated
dispersion of the KMnO4 (shown in purple) and the bromide
concentration for each well (shown adjacent to the well
screen), approximately 104 days following the initial
injection into the IW-3 well cluster. As illustrated in Figure
4-8, although KMnO4 was injected into all three zones,
there was no visual evidence of the oxidant in any shallow-
zone wells 104 days following injection. Also, the only
visual evidence of KMnO4 in the intermediate (peat) zone
was at IW-3i (the point of injection). However, there are
relatively high concentrations of bromide in IW-3 and EW-
3 cluster wells in the absence of visual KMnO4. This infers
that the oxidant solution may have dispersed radially as
expected (i.e., to and beyond the designed 10-foot radius
of influence) but that consumption of KMnO4 by VOCs
and/or naturally occurring humic material (especially in the
peat zone) may have occurred during this time period.
Figures 4-9 and 4-10 are similar illustrations showing the
estimated dispersion of the KMnO4 and bromide
concentrations at time intervals after the second injection.
Both illustrations show that visual KMnO4 has, due to the
second injection, dispersed throughout the entire area of
the deep zone. However, visual KMnO4 is absent in both
the shallow and intermediate (peat) zones. Although
bromide diminishes in time from its injection during the first
injection event, its persistence in all wells at some
concentration indicates that the oxidant solution had
dispersed throughout the desired treatment area. The
visual absence of KMnO4 may indicate its consumption by
the contaminant and/or natural humic material as
previously discussed.
It should be noted that it had been planned to inject KMnO4
into all three injection well clusters during the initial June
2005 injection event; however an apparent failure of well
seals for the IW-1 and IW-2 well clusters resulted in short-
circuiting and necessitated the shutdown of injection into
those wells. Only a very small volume of KMnO4 oxidant
was actually injected into those wells during the first
injection event (hence the low concentrations of bromide
depicted in Figure 4-8 for the IW-1 wells). Consequently,
those prepacked cluster wells were removed and replaced
with traditional wells, and a second injection for the IW-1
and IW-2 clusters was conducted in October of 2005.
10
15
20
25
IW-1 Cluster
Q\A7 (June '05 Injection cane
OVV due to faulty wells)
ets 1i 1d 1s
)
)
V
73
0.1
Horizontal Scale
Bllflri ^ 20ft .
^ r
\ EW-2 EW-
24.5
(/ 0.1
23.4 WO
1 KMn04, 620
Estimated
Dispersion \ ^~
Pattern^^
235
IW-3 Cluster MC
(June ' 05 injection of Kmn04 ' ' •—
with Bromide tracer)
3 3s 3i 3d EW-4
^ 1800
\
JVflsb
/
I
^^J
"S
2140
, Bromide,
\ ^/^ 1 ,000 mg/l
\
v Groundwater
\ Flow ->•
\
^
22
54
535
Figure 4-8. Estimated Dispersion of KMn04 104 Days After First Injection.
47
-------
10
15
20
25
IW-1 Clu
Q\A/ (October '05 Inje
O V V newly installed
2 1i 1
V
7.6
0.06J
Horizontal Scale
ster no ft
rrtinnintn ^ *•»* ll W,
wells) "^ *""
\ 1s EW-2 EW-
I
I:
0.1J
67 \~1100
KMnO4, NS
Estimated
3.8 DiSPerf°n305
Y
0.6
IW-3 Cluster Kip
(June ' 05 injection of KmnO, ' N •--
with Bromide tracer)
3 3s 3i 3d EW-4
|
450 1
\
^ ^30
^
^
150
Bromide,
> 1 ,000 mg/l
y
Groundwater^^w
Flow ^
4.6
570
42.5
Figure 4-9. Estimated Dispersion of KMn04 245 Days After First Injection and 125 Days after Second Injection.
10
15
20
25
IW-1 Clu
Q\A/ (October '05 Inje
OVV newly installed
et ... .
.s 1i 1
V
10.3
0.08J
Horizontal Scale
ster 20 ft
rtinnintn ^ *•« ' l 'W
wells) "^ *"
\ 1s EW-2 EW,
4.9
KMnO4,
Estimated
Dispersion
0.55 J
^-'~
59.8 ^ 990
X
5.85 305
0.2 J
IW-3 Cluster KJ p
(June '05 injection of Kmn04 INt
with Bromide tracer)
3 3s 3i 3d EW-4
I
fl
520 1
" - ^ 960
N
\
\
X
X _
1.9J
Bromide,
x ^^ > 500 mg/l
s
\
\
1
s
s
s
^
Groundwater w
Flow ^
43.9
169
63.5
Figure 4-10. Estimated Dispersion of KMn04 280 Days After First Injection and 160 Days after Second Injection.
48
-------
4.5 Performance Evaluation
This subsection presents the performance data obtained
during theXDD ISCO SITE Demonstration conducted from
August 2004 to May 2006. Subsection 4.5.1 summarizes
the soil analyses results soil (target VOCs in all three soil
zones and humic fraction for the peat zone). Subsection
4.5.2 summarizes groundwater analyses results for target
VOCs and metals. A detailed statistical evaluation of the
soil and groundwater VOC data is also included.
Subsection 4.5.3 summarizes data quality assurance.
4.5.1 Soil Results
Soils were evaluated as pre- and post-treatment sample
pairs to determine if reductions of VOCs had occurred due
to the ISCO treatment. Baseline (pre-treatment) and final
(post-treatment) soil samples were collected on a "paired
sample" basis; final samples were collected in the
approximate same location as baseline samples (i.e.,
within 0.3 feet to 2 feet laterally and within about one foot
vertically). Both baseline and final soil samples were
analyzed for VOCs and humate content. The soil VOC
data was highly variable. High sample concentration
variability at DNAPL sites is commonly observed,
particularly at sites with heterogeneous lithologies.
4.5.1.1 VOC Results for Soil
The primary objective for soil was to determine if XDD's
ISCO process could reduce concentrations of PCE, TCE,
cDCE, and VC by 90% from baseline to final sampling
events (Objective 1). Tables 4-5 through 4-7 present
results of the baseline (pre-treatment) soil sampling versus
the final (post-treatment) soil sampling for each of the four
critical VOCs (PCE, TCE, cDCE, and VC) for the shallow,
peat (middle or intermediate) and deep zones, respectively.
Based on pre-demonstration extraction studies, shallow-
zone soils were field-extracted using a 1:1 MeOH to soil
ratio by volume; the middle peat and bottom zone soils
were extracted using a 5:1 MeOH to soil ratio by volume.
It should be noted that due to the large number of non-
detect results, due to high laboratory reporting limits
resulting from the MeOH field extractions, the soil data was
analyzed (i.e., evaluated) three separate ways by:
1. Analysis of soil sample pairs;
2. Analysis of averaged soil results using only
detected values; and
3. Analysis of averaged soil results using all values.
Each analysis is discussed separately as follows.
Analysis of Soil Sample Pairs
Although there were 30 sample pairs planned for each of
the three stratigraphic zones (i.e., a planned total of 90
pairs), not all pairs could be evaluated for treatment
performance at their sampling location. This was primarily
due to many of the pairs consisting of two non-detect
values (i.e., values below the laboratory reporting limit or
LRL). In other instances a quantified value was paired with
a non-detect result having a reporting limit exceeding the
quantified value. In other instances (i.e., for the peat zone),
there was no baseline peat sample to pair with the post-
treatment peat sample.
As a result of not being able to evaluate a significant
number of soil sample pairs, a subset of "eligible pairs" was
selected for descriptive evaluation and included the
following:
Sample pairs that met the original criteria set forth
in the QAPP (i.e., where the baseline value was at
least ten times the LRL);
Sample pairs not meeting the QAPP criteria, but
either consist of two quantified values or have one
quantified value and one estimated J value;
Sample pairs that contain one non-detect value
(i.e., < the LRL); and the other value is a quantified
or estimated J value that is significantly higher or
lower than that reporting limit value.
Table 4-8 presents a statistical summary of the soil data
based on the eligible sample pairs, as just described. For
each of the four critical VOCs and for each of the three
zones, table 4-8 summarizes the results of the eligible pair
analysis from two perspectives:
1. Eligible pairs that show decreases in critical VOC
concentration of any magnitude, and;
2. Eligible pairs that show decreases in critical VOC
concentration > 90% (i.e., the primary soil
objective).
As inferred in Table 4-8, it is apparent that few of the
eligible paired sample sets for any of the four critical VOCs
showed concentration decreases > 90% from baseline to
final sampling events. Aside from the single TCE deep
zone sample pair (which did show a > 90% reduction), TCE
in the middle peat zone was the only instance where a 90%
reduction was measured for more than Vz (i.e., 50%) of the
eligible sample pairs evaluated. Of the four VOCs, TCE
had the best overall percentage of sample pairs showing
reductions > 90% (i.e., 40.9% for all zones combined).
With respectto the percentage of eligible pairs thatshowed
contaminant decreases of any magnitude, VC fared the
best. For all zones combined, 82.1% of eligible sample
pairs for VC showed contaminant reductions from baseline
to final sampling; including 88.9% in the shallow zone and
70% in the peat zone.
49
-------
Table 4-5. Shallow Zone Soil Results for PCE, TCE, cDCE, and VC - Baseline Vs Final 1
Parameter-*
Boring No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
PCEU
Baseline
<52.4
7,710
<61.5
234
88
90.5
41,000
34.9 J
<53.1
20,000
<52.8
< 114
<60.8
4,660
31.8 J
26,400
604
21.7 J
<59.4
287
<64.9
24.2 J
<63.7
<65.3
41. 3J
138
8,920
<59.9
21,000
9,940
jg/kg)
Final
110
<56.1
<54.9
275,000
11,100
103
20.3 J
183
<57.5
1,160
<54.3
<58.1
<57.7
29,100
201
393,000
25.3 J
<54.8
24.9 J
1,650
116
26.3 J
<60.5
<53.7
1,850
799,000
684
<58.1
107,000
11,200
TCE(k
Baseline
<26.2
757
<30.7
<53.5
<26.8
22.3 J
1,860
<26.1
<26.5
4,110
<26.4
<57.2
<30.4
158
<28
3,160
165
<59.1
<59.4
<52.2
<64.9
<65.8
<63.7
<65.3
<55.2
25.1 J
963
<59.9
1,700
15,500
9/kg)
Final
<30.3
<28
<27.4
19,600
1,500
22.0 J
<29.3
65
<28.8
243
<27.2
<29.0
<28.9
2,050
21.6 J
58,700
<29.3
<27.4
<30.6
549
34.7
<29.6
<30.3
<26.8
375
112,000
137
<29.1
10,900
6,570
cDCE f
Baseline
3,080
2,350
<30.7
16,500
10,400
1,030
6,850
1,740
<26.5
5,700
<26.4
<57.2
7,530
1,090
1,710
1,950
10,100
19.0J
26.0 J
67.2
476
8,900
19.4J
61.5
367
4,290
9,010
3,460
12,800
22,400
M9/kg)
Final
891
278
67.3
9,630
38,300
287
5,860
179
17.4J
11,600
<27.2
<29.0
<28.9
15,400
18.6 J
4,610
1,270
75.4
1,190
1,240
1,090
335
30.6
21. 4 J
904
4,320
6,100
19.9J
13,100
2,620
VC(M
Baseline
158
261
<30.7
35 J
1,410
138
280
2,060
<26.5
476
<26.4
<57.2
331
770
166
504
<54.1
37.5
254
76.8
417
3,020
<31.9
79
1,900
19.3 J
19.4 J
2,270
98.5
833
3/kg)
Final
<30.3
111
<27.4
<28.3
638
<29.9
<29.3
<31.2
<28.8
208
<27.2
<29.0
<28.9
137
<29.4
< 142
<29.3
<27.4
528
427
289
<29.6
<30.3
<26.8
654
<287
<29.9
<29.1
<131
<28.8
1 Sam pies field extracted with a 1:1 MeOHtosoil ratio by volume. J= Estimated value (i.e., value is above the detection limit but below the laboratory reporting limit).
50
-------
Table 4-6. Middle Peat Zone Soil Results for PCE, TCE, cDCE, and VC - Baseline Vs Final 1
Parameter-*
Boring No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30*
PCE(M
Baseline
<385
<298
<332
26,600
157 J
936
<863
<374
—
<560
<378
<291
<667
132,000
<865
77,400
243,000
<702
<693
<419
307 J
<888
<364
<307
—
2,550
34,900
<405
48,000
—
g/kg)
Final
<421
<396
<404
3,380
644 J
<549
141 J
509 J
<408
338 J
<581
<494
<515
1,660
<606
<635
<788
<925
<728
<558
<599
<821
<594
<542
521
4,800
999
<378
12,100
47,400
TCE(k
Baseline
< 192
< 149
< 166
674
< 189
< 175
<431
< 187
—
<280
< 189
< 145
<667
5,130
<865
6,770
40,900
<702
<693
<419
<535
<888
<364
<307
—
1,850
13,600
<405
3,380
—
9/kg)
Final
<210
< 198
<202
2,350
<454
<549
<234
<348
<204
<226
<290
<247
<258
<466
<303
<317
<394
<462
<364
<279
<299
<410
<297
<271
120 J
433
254
< 189
1,190
4,590
cDCE (
Baseline
977
97.8 J
<166
21,900
35,800
453
41,300
<187
—
<280
<189
<145
1,730
8,020
280 J
17,100
51,600
1,650
1,710
<210
4,430
10,800
270
<153
—
38,300
181,000
3,180
43,600
—
jg/kg)
Final
413
< 198
<202
24,400
1,090
2,410
19,700
<348
<204
413
<290
208 J
<258
2,230
662
467
7,650
1,410
207 J
272 J
379
461
1,990
<271
62,000
35,000
159,000
< 189
162,000
28,300
VC(Mc
Baseline
< 192
< 149
< 166
1,690
8,270
< 175
55,100
< 187
—
<280
< 189
< 145
212J
<750
<432
7,850
< 5,980
<351
328 J
<210
1,050
3,160
< 182
< 153
—
21,800
10,400
782
11,500
—
i/kg)
Final
<210
<198
<202
2,490
1,920
<549
21,300
<348
<204
<226
<290
<247
<258
<466
<303
<317
7,310
<462
<364
<279
<299
<410
<297
140 J
60,900
16,200
36,000
<189
92,400
<171
1 Sam pies field extracted with a 5:1 MeOH to soil ratio by volume. *At boring 30, peat material absent, but a sand sampled due to high PID reading.
— no sample collected. J = Estimated va ue (i.e., value is above the detection limit but below the laboratory report ng limit).
51
-------
Table 4-7. Bottom Zone Soil Results for PCE, TCE, cDCE, and VC - Baseline Vs Final 1
Parameter-*
Boring No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
PCE(Mc
Baseline
348
<150
<200
4,740,000
330 J
553
<174
243
188 J
2,950
<169
<128
1,590
249,000
1,260
<162
<153
301
4,980
<185
2,580
<167
998
<148
<182
<144
1,910
384 J
13,100
814,000
i/kg)
Final
<362
111 J
<369
31,400
<333
166 J
<357
29,600
<395
<310
<385
<326
<310
<364
<536
<382
< 1,180
5,240
<349
2,930
<313
<409
<296
<330
159 J
<319
205 J
<286
15,800
288,000
TCE(k
Baseline
<151
<150
<200
6,000
<169
<158
<174
<163
<201
<122
<169
<128
<133
<178
<167
<162
<153
<131
<146
<185
<166
<167
<156
<148
<182
<144
<173
<389
<229
< 1,570
9/kg)
Final
<181
<177
<184
<181
<166
<169
<178
139 J
<197
<155
<192
<163
<155
<182
<268
<191
<588
<193
<175
<163
<156
<204
<148
<165
<185
<160
<211
<143
<170
<447
cDCE (
Baseline
< 151
< 150
<200
< 149
< 169
< 158
< 174
234
<201
< 122
< 169
< 128
< 133
< 178
< 167
< 162
< 153
< 131
< 146
< 185
< 166
< 167
< 156
< 148
< 182
< 144
< 173
503
<229
< 1,570
jg/kg)
Final
<181
<177
<184
<181
<166
<169
<178
<223
<197
<155
<192
<163
19,500
<182
<268
<191
<29.3
<193
<175
<163
<156
<204
<148
<165
<185
<160
<211
<143
<170
237 J
VC(M
Baseline
< 151
< 150
<200
< 149
< 169
< 158
< 174
< 163
<201
< 122
< 169
< 128
< 133
< 178
< 167
< 162
< 153
< 131
< 146
< 185
< 166
< 167
< 156
< 148
< 182
< 144
< 173
<389
<229
< 1,570
3/kg)
Final
< 181
< 177
< 184
< 181
92.3 J
< 169
< 178
<223
102 J
< 155
< 192
< 163
< 155
< 182
<268
< 191
<588
< 193
< 175
< 163
< 156
<204
< 148
< 165
< 185
< 160
<211
< 143
< 170
<447
1 Sam pies field extracted with a 5:1 MeOHtosoil ratio by volume. J = Estimated value (i.e., value is above the detection limit but below the laboratory reporting lim it).
52
-------
Analyte Zone 2
Shallow
PCE Peat
Deep
All Combined
Shallow
TCE Peat
Deep
All Combined
Shallow
cDCE Peat
Deep
All Combined
Shallow
vc Peat
Deep
All Combined
LL = Lower limit of confiden
Table 4-8.
Eligible
Pairs
17
9
15
41
14
7
1
22
26
19
2
47
18
10
0
28
3e interval. I
Statistical Summary for Soil Sample
Decreases
#Dec. %Dec. 95% LL 95% UL
5 29.4 12.4 55.4
7 77.8 44.2 95.9
11 73.3 44.8 90.3
23 56.1 40.9 71.3
6 42.9 20.6 68.8
6 85.7 44.6 99.3
1 100 5.0 100
13 59.1 38.5 79.6
14 53.8 32.5 71.8
14 73.7 50 89
1 50 2.5 97.5
29 61.7 47.8 75.6
16 88.9 67.5 98
7 70 38.1 91.3
o
23 82.1 64.3 92.7
JL = Upper limit of confidence interval. Dash
'air Analysis.
Decreases > 90%
# Dec. % Dec. 95% LL 95% UL
4 23.5 8.5 48.9
4 44.4 16.9 74.9
6 40 19.1 66.8
14 34.1 19.6 48.7
4 28.6 10.4 61.1
4 57.1 22.5 87.1
1 100 5.0 100
9 40.9 20.4 61.5
4 15.4 5.4 32.5
6 31.6 14.7 57.4
0 0.0 0.0 77.6
10 21.3 9.6 33
8 44.4 23.6 67.5
2 20 3.7 60.3
o
10 35.7 19.2 57.6
led line = no calculation possible.
Analysis of Averaged Soil Results Using Detected Values
In addition to analyzing soils on a paired sample basis, soil
results were also evaluated on an average pre-treatment
(baseline) versus post-treatment (final) basis. Table 4-9
presents averaged results for the four critical VOC
compounds (PCE, TCE, cDCE, and VC), using all
quantified and estimated ("J") values. Average values are
shown for the three individual lithologic zones and for all
zones combined. The number of values averaged are
shown in parenthesis next to each average value.
When excluding non-detected results, on an all zones
combined basis, PCE is the only VOC of the four critical
compounds for which there is shown a relevant decrease
in average concentration (i.e., 63.5%). The averaged data
suggests that there were reductions of PCE concentrations
in the peat and deep zones, but an increase in PCE
concentrations in the shallow zone.
On a per zone basis and accounting for a decreased
sample size, the best overall reductions appear to have
been achieved in the middle peat zone. On average there
are relatively large decreases shown for PCE (95.6%) and
TCE (91.6%). A very small average reduction is shown for
cDCE (1.03%) and VC is shown to increase on average by
161%. For the shallow zone, two VOCs showed increased
average concentrations (PCE and TCE) and two VOCs
showed decreased average concentrations (cDCE and
VC). For the bottom zone, PCE and TCE were also shown
to have decreased average concentrations and cDCE was
shown to have increased average concentrations. VC
could not be evaluated due to the lack of baseline data
above LRLs.
Analysis of Averaged Soil Results Using All Values
To statistically determine whether the null hypothesis is
rejected or not (i.e., the primary objective), all soil results
comprising baseline and final pairs were evaluated (Table
4-10). By doing this, the sample population for baseline
and final samples were the same for each of the three
zones. For example, if non-detect results are included,
there are 30 baseline and 30 final sample results for the
shallow and deep zones; and there are 27 baseline and 27
final sample results for the middle peat zone (Note: there
were three borings at which peat was encountered at 27 of
the 30 borings sampled during the baseline activity).
53
-------
Table 4-9. Averaged Soil Results (Excluding Non-Detects) for PCE, TCE, cDCE, and VC - Baseline Vs. Final.
Analyte
PCE
TCE
cDCE
VC
Zone
Shallow
Peat
Deep
All Combined
Shallow
Peat
Deep
All Combined
Shallow
Peat
Deep
All Combined
Shallow
Peat
Deep
All Combined
Average Concentration, ug/Kg (sample No.)
Baseline
7,433 (19)
56,585 (10)
324,151 (18)
139,187 (47)
2,584 (11)
10,329 (7)
6,000 (1)
5,617 (19)
5,074 (26)
23,210 (20)
369 (2)
12,435 (48)
651 (24)
10,179 (12)
- (0)
3.827 (36)
Final
81,578 (20)
2,509 (10)
37,361 (10)
50,756 (40)
14,184 (15)
869 (5)
139 (1)
10,345 (21)
4,424 (27)
22,970 (21)
9,869 (2)
12,431 (50)
374 (8)
26,518 (9)
97 (2)
12.729 (19)
Decrease /
Increase
Increase
Decrease
Decrease
Decrease
Increase
Decrease
Decrease
Increase
Decrease
Decrease
Increase
No Change
Decrease
Increase
Increase
% Change
+ 998
-95.6
-88.5
-63.5
+ 449
-91.6
-97.7
+ 84.2
-12.8
-1.03
+ 2,570
-0.03
-42.5
+ 161
+ 233
When including non-detected results on an all zones
combined basis, PCE is still the only VOC of the four
critical compounds for which there is shown a decrease of
any magnitude in average concentration (i.e., 68.9%). The
other compound showing a decrease, although slight, is
cDCE (6.27%). Averaged data suggests that there were
considerable reductions of PCE concentrations in the peat
material (94.8%) and deep zone (93.5%), but an increase
in PCE concentrations in the shallow zone (1,050%).
On a per zone basis and accounting for a decreased
sample size, the best overall reductions appear to have
been achieved in the middle peat material and in the deep
zone. Forthe peat material, on average there are relatively
large decreases shown for PCE (94.8%) and TCE (89.7%)
and a small average reduction shown for cDCE (9.39%).
VC is shown to increase on average by 42.4%. For the
bottom zone, PCE and TCE were also shown to have
considerable decreased average concentrations (93.5%
and 66.7%, respectively). VC showed a slight decrease
(5.5%). But cDCE, a known breakdown product, was
shown to have increased by 488% on average. For the
shallow zone, two VOCs showed increased average
concentrations (PCE at 1,050% and TCE at 638%) and two
VOCs showed decreased average concentrations (cDCE
at 9.48% and VC at 77.5%).
The hypothesis test suggested in the QAPP is that the
technology will remove more than 90% of contamination.
The null hypothesis tested here is that the contamination
removed does not exceed 90%. That is, we test the null
hvoothesis:
Ho:
post - treatment constituent mass
pre - treatment constituent mass
aaainst the alternative hvoothesis
Ha:
post - treatment constituent mass
pre - treatment constiuent mass
>-0.9
<-0.9
Accepting the null hypothesis would be a finding that the
process does not meet demonstration objectives. If the
null hypothesis is rejected for a given set of experimental
circumstances, then the technology meets demonstration
objectives for those experimental circumstances.
54
-------
Table 4-10. Averaged Soil Results (Including Non-Detects) for PCE, TCE, cDCE, and VC - Baseline Vs. Final.
Analyte
PCE
TCE
cDCE
VC
Zone (Sample No)
Shallow (30)
Peat (27)
Deep (30)
All Combined (87)
Shallow (30)
Peat (27)
Deep (30)
All Combined (87)
Shallow (30)
Peat (27)
Deep (30)
All Combined (87)
Shallow (30)
Peat (27)
Deep (30)
All Combined (87)
Average Concentration, |jg/Kg
Baseline
4,721
21,120
194,523
75,259
962
2,825
306
1,314
4,400
17,217
126
6,904
524
4,700
109
1,677
Final
54,395
1,105
12,586
23,440
7,099
291
102
2,573
3,983
15,601
741
6,471
118
6,693
103
2,154
Decrease /
Increase
Increase
Decrease
Decrease
Decrease
Increase
Decrease
Decrease
Increase
Decrease
Decrease
Increase
Decrease
Decrease
Increase
Decrease
Increase
%
Change
+ 1,050
-94.8
-93.5
-68.9
+ 638
-89.7
-66.7
+ 95.8
-9.48
-9.39
+ 488
-6.27
-77.5
+ 42.4
-5.5
+ 28.4
95% '
LL
-166
-99.9
-103.5
-117
-310
-100
-109
-156
-73.1
-68.2
-514
-51.1
-92.3
-91.8
-38.8
-83.3
95%1
UL
+ 2,270
-89.6
-83.5
-21.2
+ 1,590
-79.0
-25.0
+ 348
+ 54.2
+ 49.5
+ 1,490
+ 38.6
+ 62.6
+ 177
+ 28.4
+ 140
1 Values rounded to three sicmificant diaits.
Table 4-10 provides the basis for testing the null hypothesis
for each of the four analytes for each of the three zones
and for all zones combined. As previously discussed for
each of these combinations, Table 4-10 provides an
estimate in the percentage change or reduction in
contamination in the third column from the right. In the last
2 columns the table also provides the lower and upper
limits of the 95% confidence interval for the actual or
population percentage change.
For example, consider TCE for the Peat zone. Table 4-10
shows a percentage change of -89.7% with a 95%
confidence interval of (-100%, -79%).
Now for each of the 16 sets of analyte-zone combinations,
the null hypothesis is accepted if the upper limit of the 95%
confidence interval is greater than -90.0%. For instance, in
the case of TCE in the Peat zone, the upper limit of the
confidence interval is -79.0% which is greaterthan -90.0%.
Thus, we fail to reject the null hypothesis for TCE in the
Peat zone. Based on this hypothesis testing framework,
we fail to reject the null hypothesis for each analyte in each
zone and for each analyte in all zones combined.
We hasten to add, however, that we are using a two-sided
95% confidence interval to test a one-sided hypothesis,
and therefore, the hypotheses are actually being tested at
the 97.5% level. Testing these hypotheses at the 90%
level, yields the same outcomes with one exception. At the
90% hypothesis testing level, the null hypothesis would be
rejected for PCE in the Peat zone. The null hypothesis
would be accepted for all other cases.
Statistical methods for confidence intervals provided in the
tables for the soil results are provided as follows.
Statistical Methods for Confidence Intervals
In this section, confidence intervals are presented for three
kinds of population quantities:
•• population proportion for a binomial distribution;
•• population mean values based on normal
distribution theory; and
•• ratios of population means based on normal
distribution asymptotic theory.
55
-------
Proportion
For a binomial distribution population proportion, the
sample proportion is given by:
number of successes
P =
where n is the number trials or sample pairs. In case n >
30, a 95% confidence interval for the population, proportion
p is given by Equation 1 (Eq. 1) where t(
(0.975, n - 1)
is the
upper 97.5 percentile of the student-t distribution with (n -
1)degrees of freedom. This confidence interval places 2.5
percentage points for both the upper and lower endpoints
of the confidence interval.
One can multiply the confidence interval expression by 100
to translate the proportional expression into percentage
points. When n < 30, the two-sided confidence intervals
were taken from tables of exact confidence intervals for a
binomial distribution, (see Table A-22 of Natrella, 1963).
Mean Value
The sample mean value is given by Eq.2. The two-sided
95% confidence interval for the population mean value is
given by Eq. 3, where /j is the population mean value and
S x in Eq. 4 is the sample estimate of the variance or the
square of the sample standard deviation.
Ratio of Mean Values
Define the proportion of change of in place contamination
from the baseline or pretreatment measurements to the
final or post-treatment measurements as:
PR =
post-treatment constituent mass
pre-treatment constituent mass
-1
where pR with the pretreatment and post-measurement
sample mean values by Eq. 5. From Hansen, et al.
(1993),the sample variance of
P
R
is approximated by Eq. 6. The two-sided 95% confidence
interval for pR is approximated by Eq. 7. Of course, one
can multiply the confidence interval expression by 100 to
translate the proportional expression into percentage
points.
P r~ * (0575/1-1) " P '- P j- £(OP75/!-l)
Y = • X^ Y where n = sample size, and X, are the
/ . i measurements for each sample for / = 1 ,..., n.
" i-l
- ST - Sx
V^ v^
•"I 1 TC ' 0
^yj _ A _ ^ / v" y\
« ~ 1 i_l
^ where X denotes the sample mean of the pre-treatment measurements;
— where Y denotes the sample mean of the post-treatment measurements, and
1
V R = X / Y with n, post-treatment measurements, and n, pre-treatment measurements
Equation (2)
Equation (3)
Equation (4)
Equation (5)
56
-------
S2 •
_JT2 72 JT-7_
where ^ ^ VfJT ^) C7 F)
^ »-l ti !
- s* . / / - sx .
PS ~7='^(0.975,K-1) ^ PS ^ PS. + ~^'%S75,K-1)
Equation (6)
Equation (7)
4.5.1.2 Humate Results for Soil
To assess the impact of the organic matter in the peat with
respect to its effect on oxidant depletion, samples were
collected from the middle peat zone during both baseline
and final sampling events (Objective 3). Table 4-11
presents the results of the humic acid analyses, which
reflects the amount of humate material in a soil. Instead of
an expected potential depletion of humic material, there is
shown a measured increase. The average of 27 baseline
values was 2.6 % and the average of 26 final values was
6.0 %, an increase of 230%. Also shown in the table is that
for all borings in which there are paired values, the final
value is always greater with the exception of Boring No. 1.
There is no obvious explanation forthe measured increase
in humic acid content from baseline to final.
Table 4-11. Humic Acid Results for Middle Peat Zone - Baseline Vs. Final (wt/wt %)'
Boring No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Baseline
3.1
1.1
0.8
0.6
1.4
< 0.1
2.1
< 0.2
3.7
1.1
<0.1
3.7
3.6
6.4
Final
2.07
2.65
1.35
8.33
7.59
9.58
4.02
4.29
4.27
6.08
5.11
7.02
8.45
Boring No.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Baseline
5.8
4.4
3.1
4.5
2.4
4.1
6.4
1.0
0.3
3.6
0.5
5.6
< 0.1
—
Final
10.5
8.21
11.3
5.53
4.57
8.04
7.32
3.40
10.1
3.33
7.29
1.74
4.77
1 Values rounded to maximum three sign if leant dig its. Dashed line = lack of peat material to analyze.
Using Yz the report ng I im it for non-detect values (i.e., < values), the average of all baseline values = 2.6% & average of all final values = 6.0%.
57
-------
4.5.2 Groundwater Results
4.5.2.1 Target Ethene VOCs
The primary objective for groundwater was to evaluate the
effectiveness of XDD's ISCO process in reducing
concentrations of the PCE, TCE, and cDCE in the
groundwater to below their corresponding Remediation
Performance Standards (RPS) of 750 ug/L, 5,500 ug/L,
and 17,500 ug/L, respectively. The QAPP specified that to
meetthe primary groundwaterobjective (Objective 2) more
than 90% of eligible groundwater samples had to meet
those regulatory criteria, as to reject the null hypothesis
(i.e., concentrations reduced to below the RPS).
The hypothesis test proposed in the QAPP is based on the
population proportion, call it p, of sample pairs that meet
regulatory goals. The null hypothesis to be tested is:
H0: p < or = 0.90
against the alternative hypothesis
HA: p > 0.90.
Accepting the null hypothesis would be a finding that the
process does not attain regulatory goals. For any
experimentin which the sample proportion of pairs meeting
regulatory goals do not exceed 0.90, the null hypothesis
would be accepted. (Note that this does not imply that the
proportion being, for example, 0.901 would be sufficient to
reject the null hypothesis).
It should be noted that, beginning with the initial baseline
sampling event and continuing through intermediate
sampling events, there was a total of up to 15 samples
collected from 5 different wells at the three distinct
horizons. However, the sample number for the final post-
treatment event was increased to a total of 30. This
change in strategy was due to a desire to increase the
sample population and thus provide better statistics.
To accomplish the increased sample number for the final
event without having to install additional wells, samples
were collected from the existing 15 wells on two different
days {2x15 = 30). A day (i.e., 24 hours) separated these
two sampling rounds so that, based on the horizontal
gradient, samples were considered independent both
spatially and with respect to temporal location. In other
words, the second set of samples was considered separate
and independent from the first set of samples.
It also should be noted that since there were two injection
events, there are two groundwater baseline events. Wells
influenced by the first injection (IW-3, EW-3 and EW-4 well
clusters) correlate to the April 2005 groundwater sampling
event and wells influenced by the second injection (IVV-1
and EW-2 well clusters) correlate to the September 2005
groundwater sampling event.
Table 4-12 presents results of the baseline (pre-treatment)
groundwater sampling and the two rounds comprising the
final (post-treatment) groundwatersampling. Within Table
4-12, the cells for baseline values that were measured
above the RPS are blocked (e.g., for well cluster IW-1,
shallow zone, the PCE value of 5,200 ug/L is blocked
because it is above the RPS of 750 ug/L).
The primary groundwater objective was evaluated three
different ways; including:
1. Baseline vs. final results for each individual
well;(i.e., the baseline pre-treatment value vs. the
average of the 2 final post-treatment values);
2. On an overall sample set basis (i.e., the average
of all 15 baseline pre-treatment values vs. the
average of all 30 final post-treatment values); and
3. On a pergroundwaterzone basis (i.e.,the average
of 5 baseline pre-treatment values for each of the
three zones vs. the average of 10 final post-
treatment values for each of those three zones).
Individual Well Evaluation
As shown in Table 4-12, on an individual well basis, there
are only seven instances where any of the three ethene
compounds exceeded theirrespective RPS priorto oxidant
injection. Therefore, there are just seven eligible baseline
samples to evaluate (Note: seven pairs do not give
sufficient resolution to test the null hypothesis at the 95%
confidence level). However, when averaging the two final
result values to attain a final post-treatment concentration,
such a reduction occurred in three of seven instances (i.e.,
PCE in EW-4, deep; TCE in IW-1 shallow; and cDCE in
IW-1 shallow). In two other instances (cDCE in IW-1
intermediate, and TCE in EW-2 shallow) reductions are
measured close to the RPS. For the individual well
evaluations, the null hypothesis is accepted for each of the
three contaminants (i.e., the objective was not met).
Total Sample Set Evaluation
For the entire sample set, PCE was the only critical VOC
that on average was above its respective RPS. On an
overall contaminant average basis, PCE did show an
approximate 27% reduction to very slightly below the RPS
(from 1,020 ug/L to 746 ug/L). Apparent reductions are
also shown for the two other critical compounds having
RPSs, TCE and cDCE. TCE is shown to have an
approximate 51% reduction (from an average baseline
concentration of 1,260 M9/L to an average final
concentration of 612 ug/L. For cDCE there is an
approximate 58% reduction (from an average baseline
concentration of about 7,400 ug/L to an average final
concentration of approximately 3,090 ug/L). it should be
noted that standard deviations are relatively high when
observing the total groundwater sample set, especially for
the baseline sample set. This is largely due to a high
variability of contaminant concentrations among the three
zone depths samples.
58
-------
Table 4-12. Groundwater Results for Target Ethenes
Well
Cl^ter Zone2
Shallow
IW-1 Intermediate
Deep
Shallow
IW-3 Intermediate
Deep
Shallow
EW-2 Intermediate
Deep
Shallow
EW-3 Intermediate
Deep
Shallow
EW-4 Intermediate
Deep
Average 3
Standard Deviation 3
Coefficient of Variance 3
95% UL3
95% LL 3
PARAMETER, Ren
PCE, 750
Baseline Final
5,200
16.4 J
208
81.7
3.6 J
369
6,830
86.5
10.6
141
<5.0
45.9
2.6
1.8
2,310
1,020
2,130
2.09
2,200
-160
1,090
1,360
1,120
1,310
545
5,370
718
637
9.1
7.3
<0.5
<2.5
4,190
5,560
17.8
55.5
3.0
1.5
80.1
284
3.6
8.2
<0.5
<0.5
0.75
1.6
<0.5
<0.5
<0.5
<0.5
746
1,530
2.05
1,320
174
- Baseline Vs. Final.
nediation Performance Standarc
TCE, 5,500
Baseline Final
11,200
62.1
1.0
63
<5.0
0.75
7,090
148
< 1.0
290
<5.0
<0.5
2.5
<0.5
2.1
1,260
3,300
2.62
3,080
-568
572
766
892
1,950
<0.5
<0.5
677
620
6.3
6.4
<0.5
<0.5
4,810
7,430
97.4
228
<0.5
<0.5
58.7
194
10.4
38.3
<0.5
<0.5
3.3
7.7
<0.5
0.37 J
<0.5
<0.5
612
1,590
2.6
1,210
17.7
(RPS) in |jg/L1
cDCE, 17,500
Baseline Final
31,000
50,100
1.7
4,590
18.4
<0.5
15,500
6,220
< 1.0
3,750
75.1
<0.5
345
15.3
1.6
7,440
14,500
1.95
15,500
-604
3,630
3,620
20,100
16,700
<0.5
0.34 J
2,810
1,890
392
614
<0.5
<2.5
9,510
13,600
4,490
7,770
<0.5
<0.5
341
894
24.8
105
<0.5
<0.5
978
5,300
8.0
6.9
<0.5
<0.5
3,090
5,320
1.72
5,080
1,110
1 Blocked values are those exceeding the RPS. Two sampling rounds were conducted for the final sampling event (March 14 & 16, 2006)
2 Approximate depths for zones (i.e., screened intervals in feet bis) were as follows: Shallow (4-9), Intermediate (13-14), and Deep (20-25)
Values are rounded to a maximum three significant digits. For Non-detect values (i.e., < values), % the reporting limit was used for averaging.
J = estimated value. UL = Upper Limit of 95% confidence interval. LL = Lower I im it of 95% confidence interval.
59
-------
Groundwater Zone Evaluation
On a per groundwater zone basis the results are based on
a small subset of the overall data set (i.e., there are just 5
baseline pre-treatment values for each of the three zones
versus the average of just 10 final post-treatment values
for each of those same three zones).
Shallow Zone: Standalone results for shallow zone
groundwater are presented in Table 4-13. On average,
there appears to be reductions in concentrations from
baseline to final sampling for all three ethene compounds
(i.e., PCE - 43%, TCE - 60%, and cDCE - 61%). Also
shown in Table 4-13, there are five instances where the
baseline value for any of the three ethene compounds
exceeded their respective RPS prior to oxidant injection.
Of these five instances, the average of the two final event
values is below the RPS in two cases (i.e., TCE in wells
IW-1sand cDCE in IW-1s).
For the shallow zone evaluations, the null hypothesis is
accepted for each of the three contaminants (i.e., the
objective was not met).
Table 4-13. Shallow Zone Groundwater Results for Target Ethenes - Baseline Vs. Final.
PARAMETER, Remediation Performance Standard (RPS) in ug/L1
Well ID Zone2
IW-1s Shallow
IW-3s Shallow
EW-2s Shallow
EW-Ss Shallow
EW-4s Shallow
Average 3
Standard Deviation 3
Coefficient of Variance 3
95% UL3
95% LL 3
PCE, 750
Baseline Final
5,200
81.7
6,830
141
2.6
2.450 I
3,300
1.35
6,550
- 1 652
1,090
1360
718
637
4,190
5560
80.1
284
0.75
1.6
1,390
1,920
1.38
2,770
20
TCE, 5,500
Baseline Final
11,200
63
7,090
290
2.5
3,730
5,150
1.38
10,100
-2670
572
766
677
620
4,810
7,430
58.7
194
3.3
7.7
1,510
2,520
1.67
3,320
-290
cDCE,
Baseline
31,000
4,590
15,500
3,750
345
11,000
12,500
1.13
26,600
-4 510
17,500
Final
3,630
3,620
2,810
1,890
9,510
13,600
341
894
978
5,300
4,260
4,240
1.00
7,290
1 220
1 Blocked values are those exceeding the RPS. Two sampling rounds were conducted forthe final sampling event (March 14 & 16, 2006).
2 Approximate depths for zones (i.e., screened intervals in feet bis) were as follows: Shallow (4-9), Intermediate (13-14), and Deep (20-25)
3 Values are rounded to a maximum three significant digits. For Non-detect values (i.e., < values), % the reporting limit was used for averaging.
J = estimated value. UL = Upper Limit of 95% confidence interval. LL = Lower I im it of 95% confidence interval.
Intermediate (Peat) Zone: Standalone results for
intermediate zone groundwater are presented in Table 4-
14. On average, there appears to be substantial increases
for PCE and TCE; and a decrease in cDCE. The increases
in PCE and TCE concentration averages are almost
exclusively due to drastic increases in those compounds in
well IW-1i (the IW-1 cluster is closest to the source area).
In similar fashion, the decrease in average cDCE can be
mostly attributed in a substantial decrease on that
compound in well IW-1i.
Also shown in Table 4-14, there is one instance where the
baseline value for one of the ethene compounds exceeds
its respective RPS prior to oxidant injection (i.e., cDCE in
IW-1 i). However the average of the two final event values
is 18,400 ug/L, which is above the RPS of 17,500 ug/L.
Forthe intermediate zone evaluations, the null hypothesis
is accepted for each of the three contaminants (i.e., the
objective was not met).
60
-------
Table 4-14. Intermediate Zone Groundwater Results for Target Ethenes - Baseline Vs. Final.
PARAMETER, Remediation Performance Standard (RPS) in ug/L1
Well ID Zone2
IW-1i Intermediate
IW-3i Intermediate
EW-2i Intermediate
EW-3i Intermediate
EW-4i Intermediate
Average 3
Standard Deviation 3
Coefficient of Variance 3
95% UCL 3
95% LCL 3
PC
Baseline
16.4 J
3.6 J
86.5
< 5.0
1.8
22.2
36.5
1.65
67
-23
E, 750
Final
1,120
1,310
9.1
7.3
17.8
55.5
3.6
8.2
< 0.5
< 0.5
253
509
2.01
617
- 111
TCE, 5,500
Baseline Final
62.1 892
1,950
< 5.0 6.3
6.4
148 97.4
228
< 5.0 10.4
38.3
< 0.5 < 0.5
0.37 J
43.1 323
64.2 634
1.49 1.96
123 777
-37 -131
cDCE,
Baseline
50,100
18.4
6,220
75.1
15.3
11,300
21,900
1.94
38,400
- 15,900
17,500
Final
20,100
16,700
392
614
4,490
7,770
24.8
105
8.0
6.9
5,020
7,540
1.50
10,400
- 374
1 Blocked value is that exceeding the RPS. Two sampling rounds we re conducted for the final sampling event (March 14 & 16, 2006).
2 Approximate depths for zones (i.e., screened intervals in feet bis) were as follows: Shallow (4-9), Intermediate (13-14), and Deep (20-25).
3 Values are rounded to a maximum three significant digits. For Non-detect values (i.e., < values), % the reporting limit was used for averaging.
J = estimated value. UL = Upper Limit of 95% confidence interval. LL = Lower I im it of 95% confidence interval.
Deep Zone: Standalone results for deep zone
groundwaterare presented in Table 4-15. The average of
all wells show no appreciable change in contaminant
concentration from baseline to final sampling. However,
that generalization is misleading for PCE, which shows an
increase in well IW-1d but decreases in the other four
wells. There are appreciable PCE concentration
decreases in IW-3d and EW-4d where average baseline
values of 369ug/L and 2,310 ug/L were reduced to below
detection limits in final event samples. The 2,310 ug/L
average value for EW-4d was the only instance where the
baseline value for PCE exceeded the RPS prior to oxidant
injection.
For the deep zone evaluations, the null hypothesis is
accepted for each of the three contaminants (i.e., the
objective was not met).
61
-------
Table 4-15. Deep Zone Groundwater Results for Target Ethenes - Baseline Vs. Final.
PARAMETER, Remediation Performance Standard (RPS) in ug/L 1
Well ID Zone2
IW-1d Deep
IW-3d Deep
EW-2d Deep
EW-3d Deep
EW-4d Deep
Average 3
Standard Deviation 3
Coefficient of Variance 3
95% UL3
95% LL 3
PCE, 750
Baseline Final
208
369
10.6
45.9
2,310
589
973
1.65
1,800
-619
545
5370
< 0.5
< 2.5
3.0
1.5
< 0.5
< 0.5
< 0.5
< 0.5
592
1,690
2.85
1,800
- 615
TCE, 5,500
Baseline Final
1.0 < 0.5
< 0.5
0.75 < 0.5
< 0.5
<1.0 < 0.5
< 0.5
< 0.5 < 0.5
< 0.5
2.1 < 0.5
< 0.5
0.8 0.4
0.8 0.3
0.89 0.90
1.79 0.58
-0.09 0.12
cDCE, 17,500
Baseline Final
1.7 < 0.5
0.34 J
< 0.5 < 0.5
< 2.5
<1.0 < 0.5
< 0.5
< 0.5 < 0.5
< 0.5
1.6 < 0.5
< 0.5
0.86 0.36
0.73 0.31
0.85 0.88
1.77 0.58
-0.05 0.13
1 Blocked values are those exceeding the RPS. Two sampling rounds were conducted for the final sampling event (March 14 & 16, 2006)
Approximate depths for zones (i.e., screened intervals in feet bis) were as follows: Shallow (4-9), Intermediate (13-14), and Deep (20-25)
3 Values are rounded to a maximum three significant digits. For Non-detect values (i.e., < values), % the reporting limit was used for averaging.
J = estimated value. UL = Upper Limit of 95% confidence interval. LL = Lower I im it of 95% confidence interval.
4.5.2.2 Non-Critical VOCs
In addition to the critical ethene compounds (PCE, TCE,
and cDCE), there are three additional VOCs having an
associated RPS for the former MEC building site. These
included toluene and two ethane compounds 1,1,1
Trichloroethane (1,1,1-TCA), and 1,1-Dichloroethane (1,1-
DCA). Table 4-16 presents results of the baseline
(pre-treatment) groundwater sampling and the two rounds
comprising the final (post-treatment) groundwater
sampling. Toluene was the only compound measured at
baseline above the RPS in any of the wells (i.e., EW-2).
Final toluene concentrations for EW-2 were measured
above the RPS of 2,570 ug/L.
4.5.2.3 KMnO4
KMnO4 is visually observable in groundwater at relatively
low concentrations (i.e., 1 ppm). Therefore, following the
initial injection of oxidant solution KMnO4 was visually
monitored in all ground water samples collected through the
final post-treatment event. Table 4-17 presents the critical
VOC results of all demonstration groundwater events with
respect to the presence of KMnO4 (i.e., sample results
shaded purple are those in which KMnO4was observed in
the groundwater during sample collection). There are two
general observations that are apparent regarding the visual
presence of KMnO4
1. KMnO4 was seen in all deep zone wells following
the first injection, but seen in only one well on one
occasion in the intermediate zone (a well receiving
injected KMnO4 about four months earlier).
KMnO4 was never observed in upper zone wells.
2. In most instances, target VOCs were reduced to
below LRLs when KMnO4 was visually present.
Knowing that there was more KMnO4 injected into the deep
zone relative to the shallow and intermediate peat zones,
and knowing that the peat zone contained substantial
humic material, the two observations above infer that the
mass of KMnO4 injected into the shallow and peat zones
may not have been sufficient or was not delivered to all
desired localities. However the mass of KMnO4 injected
into the deep zone appears to have been more than
sufficient to treat the deep zone groundwater, especially
since the KMnO4 persisted long after the target VOCs were
apparently oxidized. It is also possible that the deep zone
benefitted from downward seeping oxidant that had been
targeted for the peat zone.
62
-------
Table 4-16. Groundwater Results for Non-Critical VOCs - Baseline Vs. Final.
Well PARAMETER, Remediation Performance Standard (RPS) in |jg/L 1
Cluster
ID Zone
Shallow
IW-1 Intermediate
Deep
Shallow
IW-3 Intermediate
Deep
Shallow
EW-2 Intermediate
Deep
Shallow
EW-3
Intermediate
Deep
Shallow
EW-4
Intermediate
Deep
_Avejrag_e 3
Toluene, 2,570
Baseline Final
43.1
1,160
1.5
37.5
645
<0.5
1,510
5,770
<0.5
223
234
<0.5
154
90
9.4
659
7.7
5.6
1,580
1,250
6.2
12.2
221
121
1,630
1,780
0.39 J
2.4 J
197
296
5,660
9,250
3.1
1.1
6.9
3.1
232
467
0.24 J
<0.5
270
166
5.6
16.2
4.2
4.3
773
1,1,1-TCA, 6,800
Baseline Final
69.7 62.9
61.0
<10 166
163
0.63 0.68
1.20
61.5 159
130
<5 2.1
11.2
0.54 2.9
7.5
1,290 200
346
125 4.7 J
14.7
0.41 J < 0.5
<0.5
47.5 59.1
58.5
< 5 < 0.5
1.0
0.39 J <0.5
<0.5
<0.5 <0.5
0.24 J
<0.5 <0.5
<0.5
<0.5 3.5
3.0
107 48.7
DCA, 27,500
Baseline Final
<10 12.6
10.8
709 235
192
< 0.5 < 0.5
<0.5
29.3 123
57.3
6.0 92.6
93.4
<0.5 <0.5
<2.5
916 159
251
234 79.5
199
<0.5 <0.5
<0.5
84.8 38.4
48.2
21.1 7.5
18.5
<0.5 <0.5
<0.5
14.8 34.9
67.3
2.9 0.49 J
0.86
<0.5 <0.5
<0.5
135 57.5
1 Blocked values are those exceeding the RPS. Two sampling rounds were conducted for the final sam pi ing event (March 14 & 16, 2006).
Approximate depths for zones (i.e., screened intervals in feet bis) were as follows: Shallow (4-9), Intermediate (13-14), and Deep (20-25)
3 Values are rounded to a maximum three significant digits. For Non-detect values (i.e., < values), % the reporting limit was used for averaging.
J = estimated value.
63
-------
Table 4-17. KMnO^ Vs. PCE, TCE, and cDCE Concentrations in Groundwater - All Events
Well
ID1
IW-1
IW-3
EW-2
EW-3
EW-4
Event
Zone
S hallow
Inter
Deep
Shallow
Inter.
Deep
S hallow
Inter.
Deep
Shallow
Inter.
Deep
Shallow
Inter.
Deep
PARAMETER (Units), Remediation Performance Standard (RPS) in |jg/L
PCE (lJg/L), 750
IW-3 Injection IW-1 /IW-2 njection
(June 6-10) (Oct. 3-6)
I I
1"
BL
6/05
220
< 10
897
81.7
3.6 J
369
6 820
68.8
12.7
141
< 5.0
45.9
2.6
1.8
2310
2nd BL
9/05
5 200
16 4J
208
295
< 10
< 10
6 830
86.5
10.6
201
< 10
< 1.0
0.59
< 1.0
< 0 5
1"
Int.
12/05
126
13 4
1.7
...
...
< 0.5
9 760
10.1
< 0.5
83
5.9
< 0.5
2.2
1.6
< 0 5
2nd
Int.
2/06
343
102
56 9
175
7.6
<0.5
11 600
52.2
9.9
256
15.9
<0.5
3.2
2.2
<0 5
Final
3/06
R1
1 090
1 170
545
718
9.1
< 0.5
4 190
17.8
3.0
80.1
3.6
< 0.5
0.75
< 0.5
< 0 5
R2
1 360
1 310
5370
637
7.3
< 2.5
5 560
55.5
1.5
284
8.2
< 0.5
1.6
< 0.5
< 0 5
TCE (lJg/L), 5,500
IW-3 Injection IW-1 / IW-2 In ection
(June 6-10) (Oct. 3-6)
I I
1" BL
6/05
854
< 10
1.6
63
< 5.0
0.75
4 410
144
0.37 J
290
< 5.0
< 0.5
2.5
< 0.5
2 1
2nd BL
9/05
11 200
62 1
1.0
467
< 10
< 10
7 090
148
< 1.0
696
9.4 J
< 1.0
2.0
0.62 J
< 0 5
1" Int.
12/05
638
12 3
< 0.5
...
...
< 0.5
1? 800
28.3
< 0.5
113
18.1
< 0.5
6.0
2.2
< 0 5
2nd
Int.
2/06
748
932
3 0
103
12.5
< 0.5
14 600
136
31.4
203
62.7
< 0.5
9.5
3.7
< 0 5
Final
3/06
R1
572
892
< 0.5
677
6.3
< 0.5
4 810
97.4
< 0.5
58.7
10.4
< 0.5
3.3
< 0.5
< 0 5
R2
766
1 950
< 0 5
620
6.4
< 2.5
7 430
228
< 0.5
194
38.3
< 0.5
7.7
0.37 J
< 0 5
cDCE (lJg/L), 17,500
IW-3 Injection IW-1 /IW-2 n ection
(June 6-10) (Oct. 3-6)
I I
1" BL
6/05
6 500
2 180
< 0.5
4,590
18.4
< 0.5
2 540
9,660
0.35 J
3,750
75.1
< 0.5
345
15.3
1 6
2nd BL
9/05
31 000
50 100
1.7
3,600
< 10
7.5 J
15 500
6,220
< 1.0
6,510
137
< 1.0
176
8.8
< 0 5
1" Int.
12/05
3 840
77 800
0.32 J
...
...
< 0.5
74 700
11,300
< 0.5
1,950
129
< 0.5
740
40.9
< 0 5
1 1st or 2nd in ( ) denotes the injection that particular well is associated with based on its prox mity to in ected KM nO4. Dashed line indicates no sample collected. J = estimated v;
= Intermediate Event; R1 = Round 1 conducted 3/14/06; R2 = Round 2 conducted 3/16/06. Blocked cells indicate value above RPS. Purple indicates KmnO4 visually observed
2nd
Int.
2/06
1 620
73 800
73.7
390
623
< 0.5
73 400
8050
135
838
203
< 0.5
983
21.3
< 0 5
Final
3/06
R1
3 630
70 100
< 0.5
2,810
392
< 0.5
9 510
4,490
< 0.5
341
24.8
< 0.5
978
8.0
< 0 5
R2
3 620
16 700
0.34 J
1,890
614
< 2.5
13 600
7,770
< 0.5
894
105
< 0.5
5,300
6.9
< 0 5
i ue. BL = Baseline Event. Int.
-------
4.5.2.4 Metals
Chemical oxidation has the potential to mobilize valence
sensitive toxic metals, such as Arsenic (As), Chromium
(Cr), Selenium (Se), Zinc (Zn), and Mercury (Hg), from
soils into groundwater. As a result, sites that contain
elevated levels of metals contaminants as well as organic
contaminants should be monitored for such mobilization
potential. In order to assess the mobilization potential of
these compounds by KMnO4, groundwater samples were
analyzed for metals for each event (Objective 3).
Results for the toxic metals noted above are shown in
Table 4-18. It should be noted that significant mercury
concentrations were not detected during pre-demonstration
studies therefore additional mercury monitoring was not
continued for demonstration sampling.
On an average basis, most of the metals analyzed did not
change significantly in concentration throughout the five
groundwatersampling events. Some metals {e.g., As, Be,
Cr, Mg, and Zn) actually showed decreased concentrations
in the shallow zone from baseline to final sampling events.
There was however a fairly substantial increase in
manganese (Mn) in the intermediate (peat) and deep
groundwater zones for the second baseline sampling
event. This average increase can be attributed to the influx
of Mn from the initial injection of KMnO4 into the IW-3
cluster. The resulting elevated levels of Mn in those IW-3
cluster wells and nearby wells raised the overall average
concentration of all wells samples for those zones.
There was also an increase of Grand Mg in the deep zone
groundwater following both the first and second injections,
and an increase in Se in the deep zone groundwater
following the second injection. However, for the following
sampling event, the average concentrations of these three
metals reverted back to an average value that was very
close to their respective original levels.
4.5.2.5 Additional Observations
There is one other noted observation which may be of
some importance. Table 4-19 shows concentrations of
acetone bromide, and bromoform. As previously
discussed in subsection 4.4.3, bromide was used as a
tracer ion in the KMnO4 solution used for the first injection.
It was noted during the intermediate groundwatersampling
that bromoform formation was occurring in several of the
wells. Because this was thought to be a possible concern
and appeared to be a by-product of the use of the bromide
tracer ion some investigation was performed as to why this
was occurring.
An interesting aspect of this investigation was that bromide
showed up in several wells at significant levels (i.e., not just
as residuum from the first preempted injection event).
where there had been no indication of any visible KMnO4.
Therefore it is possible that the KMnO4 had seeped into
these upper wells (based on the presence of the bromide
ion) but has been exhausted leaving no excess KMnO4for
continued oxidation.
A build-up of acetone was noted between the baseline
sampling period and first injection suggesting oxidation of
organic compounds was occurring. This was particularly
noted in the deep wells where it appears, based upon
reductions in chlorinated ethenes that the oxidation
process was working the most effectively. What was also
noted was the increased concentration of bromoform in
upper, middle and lower wells between the baseline and
first injection and the subsequent decrease between the
first injection and second baseline sampling round.
Bromoform is not a naturally occurring compound. In fact,
the formation of bromoform is the result of a reaction
between acetone and bromide (an artifact of the use of the
bromide tracer ion in the KMnO4 solution). The reduction
of bromoform over time suggests that this compound was
not a concern at the end of the demonstration. The
formation of bromoform suggests that oxidation was
occurring in all three horizons, with the greatest oxidation
reactions occurring in the deep wells.
4.5.3 Quality Assurance/Quality Control
Quality Assurance (QA)may be defined as a system of
activities the purpose of which is to provide assurance that
defined standards of quality are met with a stated level of
confidence. A QA program is a means of integrating the
quality planning, quality assessment, quality control (QC),
and quality improvementefforts to meetuserrequirements.
This includes all actions taken by project personnel, and
the documentation of laboratory performance and, when
appropriate, field performance. The objective of a quality
assurance program is to reduce measurement errors to
agreed upon limits and to produce results of acceptable
and known quality. The laboratory data was generated
under EPA-approved method guidelines to ensure that the
measurement systems employed were in control. The
following sections provide information on the use of data
quality indicators, limitations on data use and a detailed
summary of the laboratory QC analyses associated with
select project measurements.
65
-------
Table 4-18. Metals Summary Results for Groundwater (mg/l). 1
Parameter
Arsenic
Beryllium
Cadmium
Chromium
Magnesium
Manganese
Selenium
Zinc
Zone
Shallow
Intermediate
Deep
Shallow
Intermediate
Deep
Shallow
Intermediate
Deep
Shallow
Intermediate
Deep
Shallow
Intermediate
Deep
Shallow
Intermediate
Deep
Shallow
Intermediate
Deep
Shallow
Intermediate
Deep
1" BL 6/05
0.32 (5)
0.07 (5)
0.003 (5)
0.003 (5)
0.001 (5)
0.001 (5)
0.00 (5)
0.00 (5)
0.001 (5)
0.166(5)
0.05 (5)
0.003 (5)
27.6 (5)
15.6(5)
8.6 (5)
4.5 (5)
8.1 (5)
0.8 (5)
0.004 (5)
0.006 (5)
0.001 (5)
0.133(5)
0.06 (5)
0.013(5)
2nd BL 9/05
0.29 (5)
0.11 (5)
0.0 (5)
0.001 (5)
0.00 (5)
0.00 (5)
0.00 (5)
0.004 (5)
0.00 (5)
0.071 (5)
0.18(5)
0.72 (5)
13.4(5)
11.8(5)
12.3(5)
9.5 (5)
503 (5)
825 (5)
0.003 (5)
0.037 (5)
0.00 (5)
0.041 (5)
0.008 (5)
0.032 (5)
SAMPLE EVENT
1" Int. 12/05 2nd Int. 2/06
0.3 (4)
0.18(4)
0.00 (4)
0.002 (4)
0.001 (4)
0.00 (4)
0.00 (4)
0.00 (4)
0.00 (4)
0.157(4)
0.08 (4)
1.37(4)
31.8(4) 16.7(5)
16.0(4) 11.2(5)
29.9 (5) 9.6 (5)
16.8(4) 16.5(5)
43.6 (4) 34.2 (5)
159(5) 102(5)
0.001 (4)
0.007 (4)
68.4 (4)
1.13(4)
0.07 (4)
0.00 (4)
'' Values are averages of the # of wells sampled in ( ) and rounded to a maximum three significant digits. Averages include
values and 0.0 is used for non-detected values except in cases where the LRL significantly exceeded quantified values. —
collected.
Final
0.125(10)
0.133(10)
0.048(10)
0.00(10)
0.03(10)
0.00(10)
0.00(10)
0.00(10)
0.00(10)
0.053(10)
0.164(10)
0.123(10)
17.3(10)
11.8(10)
14.5(10)
15.5(10)
65.8(10)
39.4(10)
0.003 (10)
0.007(10)
0.002(10)
0.037(10)
0.07(10)
0.04(10)
estimated (J)
No metals sample
66
-------
Table 4-19, Acetone, Bromide, and Bromoform Summary Results for
Zone
Shallow
Intermediate
Deep
Parameter *
Acetone
Bromide
Bromoform
Acetone
Bromide
Bromoform
Acetone
Bromide
Bromoform
" BL 6/05
15.5(5)
0.0 (5)
0.00 (5)
45(5)
0.0 (5)
0.00 (5)
0.0 (5)
0.0 (5)
0.0 (5)
' Values are averages of the # of wells sampled in ( ) and
values and 0.0 is used for non-detected values except in
collected.
2nd BL 9/05
13.6(5)
538 (5)
5.9 (5)
96.8 (5)
380 (5)
941 (5)
165(5)
582 (5)
1 ,450 (5)
rounded to a
cases where
SAMPLE EVENT
1"Int. 12/05
12.4(4)
574 (4)
3.5 (4)
32(4)
78.9 (4)
1.1 (4)
84.2 (5)
63(5)
213(5)
Groundwater (|jg/l) 1
2nd Int. 2/06
< 20 (4)
419(5)
1 1 .6 (3)
192(3)
298 (5)
1.0(2)
41.8(5)
31.1 (5)
20.6 (4)
maximum three significant digits. Averages include
the LRL significantly exceeded quantified values. —
Final
9.6 (5)
462(10)
5.8 (6)
168(8)
312(10)
0.0 (6)
68(10)
9.3(10)
34.3(10)
estimated (J)
No sample
4.5.3.1 Data Quality Indicators
To assess the quality of the data generated during this site
investigation, two important data quality indicators are of
primary concern: precision and accuracy. Precision can be
defined as the degree of mutual agreementof independent
measurements generated through repeated application of
the process under specified conditions. Accuracy is the
degree of agreement of a measured value with the true or
expected value.
Precision was assessed by laboratory spiked duplicates
and sample duplicates. In the case of spiked duplicates,
precision is evaluated by expressing, as a percentage, the
difference between sample and sample duplicate results.
The relative percent difference (RPD) is calculated as:
RPD =
(Maximum Value-Minimum Value) x 100
(Maximum Value+Minimum Value) 12
To determine and evaluate accuracy, known quantities of
select target analytes were spiked into selected field
Other data quality indicators influence whether a
measurement is considered valid. A sample
measurement must be reproducible and comparable.
Comparability expresses the extent to which one data set
can be compared to another. To generate comparable
results, laboratory standard methods that are widely
accepted were used. Data must also be representative of
field conditions. Representativeness refers to the degree
with which analytical results accurately and precisely reflect
samples. Equipment used to provide data for this project
was tested for accuracy through the analysis of calibration
check standards and laboratory control samples (LCSs).
To determine matrix spike recovery, the following equation
was applied:
% Recovery
where
x 100
= Analyte concentration in spiked sample
= Analyte concentration in unspiked sample
= Analyte concentration added to sample
To determine the % recovery of LCS analyses or spiked
blanks, the following equation was used:
% Recovery =
Measured Concentration
Theoretical Concentration
x 100
actual conditions present at the locations chosen for
sample collection.
4.5.3.2 Conclusions and Data Quality Limitations
A review of critical analytical sample data and associated
QC analyses was performed forXDD samples. Duplicate
spiked sample analyses and LCSs were used to assess
precision and accuracy as discussed below. Details of the
individual sample results with respect to surrogate spike
67
-------
recoveries, internal standards, dilutions, etc. are presented
in later subsections, along with information on calibrations,
blanks, performance check standards, and other QC
measurements.
The critical parameter reviewed was VOCs, specifically
cDCE,TCE, PCE.aswellasVC, 1,1-DCA, 1,1,1-TCA and
toluene in pre-and post-treatmentsoils and post-treatment
groundwaters. The results of the measurements designed
to assess the data qualityforeach method are summarized
below, along with a discussion of any impact on data
quality.
Soil VOC Samples
Accuracy: Matrix spike duplicates were analyzed with
pre-and post-treatment soil sample MeOH extracts. Table
4-20 presents average recoveries for both events.
Table 4-20. Project Accuracy for Soil VOC Analyses.
Compound
cDCE
TCE
PCE
VC
1,1-DCA
1,1,1-TCA
Toluene
Pre-treatment Avg.
Recovery (%)
97
105
74
127
112
107
106
Post-treatment
Avg. Recovery (%}
111
119
112
88
120
121
117
Average recoveries were within the QAPP specified criteria
of 80-120% (VC 75-125%) with the exception of the
baseline recovery of VC and PCE, and the post-treatment
recovery of 1,1,1-TCA. The baseline recoveries were
impacted by the spiking levels relative to the native sample
concentrations; spike levels were adjusted during the
post-treatment event to better accommodate the native
contaminant concentrations.
Precision: Duplicate spiked samples were used to assess
method precision for VOCs in soils. Where spiked levels
were appropriate, average RPD values for matrix
spike/matrix spike duplicate {MS/MSD) results were
generally less than 20%.
Comparability: Comparability expresses the extent to
which one data set can be compared to another. The
methods used were clearly specified in the QAPP and
reviewed before samples or data were collected.
Representativeness: Representativeness refers to the
degree with which a sample exhibits average properties of
the site at the particular time being evaluated. The
collection of representative samples was discussed
previously in the QAPP. Representative samples were
collected as per QAPP approved procedures.
Detection limits: Detection limits for this project are defined
as the reporting limit; the concentration determined by the
lowest calibration standard as determined by the
laboratory. Detection limits were adjusted as necessary
based on matrix and the need for dilution. Due to the
increased MeOH required for extracting peat, reporting
limits were adjusted in order to meet project criteria. The
typical reporting limit in soil for each of the critical
compounds was 250 ug/kg, based on a 1g to 5ml methanol
dilution as determined during the pre-demonstration
analytical extraction study. Detection limit objectives were
generally achieved for all samples; however, some
samples did not meet these objectives due to high
concentrations of one of the critical compounds thereby
diluting the other compound concentrations or due to
matrix interferences as occurred in several of the peat
samples.
Groundwater VOC Samples
Accuracy: Matrix spike duplicates were analyzed to assess
accuracy of the VOC analyses. Average recoveries for the
groundwater MS/MSDs are summarized in Table 4-21 for
VOC spikes from the post-treatment event. Two of the 30
samples were selected for spiking and one had poor or no
recovery of several critical compounds. The bottom
sample of Well IW-3 had no spike recovery of cDCE, TCE,
and VC and very low recovery of PCE. The results were
the same for the spike and spike duplicate. The sample
was noted to be thick in consistency and purple in color
and although the sample and MS/MSD were analyzed at a
dilution, the recovery was still non-compliant. Results from
this sample (GW-F-IW3-B[2]) should be used cautiously;
data for cDCE, TCE and VC should be rejected and data
for PCE should be considered an estimated value
potentially biased low. The other spiked sample was
analyzed at two different spiking levels due to varying
native sample concentrations.
Precision: Duplicate spiked samples were used to assess
method precision for groundwater VOCs. Maximum RPD
values ranged from 2-34%; the outliervalues {RPD values
>20%) were associated with the spike having
inappropriately low spike levels.
Comparability: Comparability expresses the extent to
which one data set can be compared to another. The
methods used were clearly specified in the QAPP and
reviewed before samples or data were collected.
Representativeness: Representativeness refers to the
68
-------
degree with which a sample exhibits average properties of
the site at the particular time being evaluated.
Representative samples were collected as per QAPP
approved procedures.
Table 4-21. Project Accuracy for
Groundwater VOC Analysis.
Compound
cDCE
TCE
PCE
VC
1,1 -DCA
1,1,1-TCA
Toluene
Post-treatment Avg. Spike Recovery (%)
59
50
57
47
110
110
100
4.5.4 Analytical QC Results
4.5.4.1 Soil VOC Analyses
A total of 87 pre- and 89 post-treatment soil samples were
collected and analyzed for VOCs. The samples were
collected from 30 locations, at three depths from each
location, except in instances where sample material was
not available. All solid samples were extracted into
methanol and the extract analyzed in accordance with
SW846 Method 8260B. The discussions below provide
details on the results of calibrations, surrogate recoveries,
internal standard results, holding times and blank results
for each event.
Soils Baseline Event
Accuracy and precision were assessed by the analysis of
spiked duplicates. An LCS was also analyzed with each
MS/MSD from a secondary source standard. All LCS
(secondary source standard) samples were within the
QAPP specified control limits of 80%-120% recovery. If the
MS/MSD samples failed to meet these limits for any of the
critical compounds then individual results were examined
to determine cause forfailure and appropriate recourse for
corrective action. Because all LCS recoveries met
appropriate QC limits it was determined that the cause of
MS/MSD failures were matrix specific.
As noted in Table 4-22 PCE and cDCE had low recovery
values in one of the MS/MSD samples. This was due to
the high native concentration of both these analytes
compared to the spiking concentration used by the
laboratory. When this situation occurs, natural variability of
the analytical instrumentation will overwhelm the spike
recovery detection and subsequently cause lowand/orhigh
calculated spike recoveries. Because the laboratory was
unable to use higher spiking concentrations and because
LCSs met accuracy specifications as noted in the QAPP,
rerunning and or re-spiking of the MS/MSD samples was
not performed. In addition, in these same samples all
other critical compounds were within acceptable spiking
limits. Therefore, no further action was taken in regards to
MS/MSD recoveries for both analytes noted above. In
addition, one MS/MSD recovery for VC was slightly above
acceptable limits (control limits: 75-125%). But because
this was not too far above acceptable recovery limits and
because LCS recoveries were considered acceptable, no
further action was taken.
Table 4-22. Accuracy for Soil VOC - Spike Results -
Baseline Event.
Compound
cDCE
TCE
PCE
VC
1,1 -DCA
1,1,1-TCA
Toluene
#of
Spikes
8
8
8
8
8
8
8
Spike Recovery, %
Avg.
97
105
74
127
112
107
106
Range
64-113
96-115
23-114
121-135
106-123
95-113
97-115
#of
Outside
Limits
2
0
2
2
1
0
0
Precision for volatile organics was assessed by the
analysis of duplicate MS/MSDs performed on selectproject
samples to determine the reproducibility of the
measurements. The RPD between the spiked sample
concentrations was compared to the objectives given in the
QAPP. RPD values for all but two critical compounds were
less than the QAPP specified objective of 20%. The two
critical compounds not meeting specifications were PCE
and cDCE. As noted above, these were the same
compounds not meeting accuracy specifications. The
likely reason for these out-of-bound specifications was
because of the high native concentration for both these
analytes compared to the spiking concentration used by
the laboratory. When this situation occurs, natural
variability of the analytical instrumentation will overwhelm
the spike recovery detection and subsequently cause low
and/or high calculated spike recoveries along with a
greater RPD variability. For the same reasons noted
69
-------
above in the accuracy discussion re-running of the spiked
compounds was not considered necessary. Critical
compound RPDs are reported in Table 4-23 which shows
all other critical compounds met QAPP specifications.
Table 4-23. Precision for Soil VOC - Spike Results -
Baseline Event.
Compound
cDCE
TCE
PCE
VC
1,1 -DCA
1,1,1-TCA
Toluene
RPD
Range, %
2-42
4-8
3-67
4-8
1-8
4-16
4-5
# of Values
4
4
4
4
4
4
4
# of Outside
Limits
1
0
1
0
0
0
0
Surrogate standards were added to all samples prior to
analysis. Mostsample surrogate recoveries met laboratory
specified control limits. The few outliers noted were in
samples that when re-analyzed at necessary dilutions,
generally had compliant surrogates (BL-7-P[12],
BL-30-SI[23], BL-27-P[13.5]). One sample (BL-28-SI[24.5])
had multiple non-compliant surrogate recoveries
approximately 10-20% above control limits; this was not
expected to have a significant impact on the sample data.
Soils Final Post-Treatment Event
Soil sample methanol extracts were analyzed by SW846
8260B. Several samples were analyzed past the 14 day
hold time; these samples are listed in Table 4-24. Since
most samples missed by only one day, and since the soils
were preserved in methanol at the time of collection, the
missed holding times are not considered to have an impact
on data quality.
Table 4-24. Samples Analyzed Outside Holding Times
Sam pie ID /
Days Past 14
F29-SA-I8) / 1
F04-SA-J8) 14
F04-P-M5) / 1
F04-SI-(27)/1
F25-SI-(24.5) / 1
F25-SA-(8)/4
Sam pie ID /
Days Past 14
F25-P-(13) / 1
F23-SI-(24) / 1
F23-SA-I8) / 1
F28-SI-(24)/1
F23-P-J11) / 1
F02-P-(13)/ 1
Sam pie ID /
Days Past 14
F02-SI-(25) / 1
TB033006 / 1
TB2033006 / 1
MTB033006/ 1
MTB2033006 / 1
MTB032906/ 1
Calibration requirements for SW8260B were generally met
for all analyses with few minor exceptions not expected to
impact data quality; for example the ICAL from April 12,
2006 had an RSD for 1,1 -DCA of 15.7% and for 1,1,1 -TCA
the RSD was 16%. VC reporting detection limits were
raised when the compound was notdetected in the 0.5 ppb
standard of the ICAL. Two samples were analyzed beyond
the 12-hour BFB clock. The initial analysis of sample
F05-SI-[24.5] missed by four minutes and the dilution was
analyzed 41 minutes after expiration of the BFB period. A
diluted analysis of F11-P-[11.5] was started 3 minutes after
the 12-hour clock expired. The tunes and continuing
calibration subsequent to these analyses met criteria so
the impact on data quality is considered minimal.
Method, trip and MeOH trip blanks were analyzed with the
samples. Several method blanks had low-level
contamination of toluene and PCE at estimated
concentrations below detection limits. A MeOH trip blank
(MTB032906) also had an estimated concentration of
toluene (10.6 J ug/kg) below the detection limit.
Each sample was spiked with surrogate standards; all
recoveries were within control limits exceptfor F14-SA-[8],
which had a slightly elevated surrogate recovery
{toluene-d8 = 148%, upper control limit = 146%). But this
did not affect data quality. All internal standards met
criteria.
LCSs and MS/MSDs were analyzed as required. Several
LCS results had recoveries slightly above the 80-120%
QAPP guideline (75-125% forVC). In particular, samples
from each of the three intervals from F09, F23, F05, F18,
F20 and F11 along with F02-SA were associated with LCS
analyses having recoveries up to 20% above control limits
for one or more VOCs. Nine spiked duplicate pairs were
analyzed with post-treatment samples (Table 4-25).
Table 4-25. Accuracy for Soil VOC - Spike Results -
Post Treatment Event.
Compound
cDCE
TCE
PCE
VC
1,1 -DCA
1,1,1-TCA
Toluene
No. of
Spikes*
18
18
14
18
18
18
14
Spike Recovery, %
Avg.
111
119
112
88
120
121
117
Range
95-130
105-148
95-136
62-110
108-152
105-153
95-142
No.
Outside
Limits
5
4
4
3
4
5
4
* This column presents the # of spikes that were spiked at an appropriate
ievei given the native sample concentration of the compound, relative to the
# of spiked samples prepared; e.g.. there were 18 spikes prepared but only
14 had levels of PCE suitable given the PCE concentrations in the sample.
70
-------
The average recovery for 1,1,1-TCA was above the upper
control limit specified in the QAPP; results for this
compound may be considered slightly biased high. Most
of the spiked recovery outliers were associated with two
spiked duplicate pairs (F11-P and F25-SI), which had
elevated recoveries for all target compounds except VC.
The MS/MSD pairs were also used to assess overall
precision for the project. RPD results are summarized in
Table 4-26. Only one spiked pair had an RPD value that
exceeded the QAPP criteria of 20%.
but had non-complaint recovery values (0%) for cDCE,
TCE and VC and very low recovery (2%) for PCE.
Table 4-26. Precision for Soil VOC - Spike Results -
Post-treatment Event.
Compound
cDCE
TCE
PCE
VC
1,1 -DCA
1,1,1-TCA
Toluene
RPD
Range, %
1-7
0-6
0-22
0-13
0-8
0-8
1-10
No. of
Values
9
9
7
9
9
9
7
No. Outside
Limits
0
0
1
0
0
0
0
Table 4-27. Accuracy for Groundwater - Post
Treatment VOCs.
Compound
cDCE
TCE
PCE
VC
1,1 -DCA
1,1,1-TCA
Toluene
No. of
Spikes1
4
4
4
4
6
6
4
Spike Recovery, %
Avg.
59
50
57
47
110
110
100
Range
0-119
0-100
2-113
0-95
103-121
104-119
93-109
No.
Outside
Limits
1
1
1
1
1
0
0
1 The number of spikes includes those with a spike level appropriate to the
native contaminant concentration.
Precision was assessed through these same spiked
duplicates, as shown in Table 4-28. Of the 18 RPD values
assessed, all but 2 were within lab limits.
4.5.4.2 Groundwater VOCs - Post-treatment Results
Fifteen wells were sampled at two different times for the
final groundwater event. The 30 samples were analyzed in
accordance with SW846 Method 8260B. All samples were
analyzed within holding times. Calibration requirements,
including GC/MS tuning with BFB and initial and continuing
calibrations, were met with one exception. One initial
calibration for toluene exceeded the 15% RSD criteria
slightly at 16%; this had no impact on data quality.
Surrogate recovery and internal standards criteria were met
for all samples. All method, trip and field blanks were free
from contamination.
Two samples were analyzed as spiked duplicate pairs.
Sample GW-F-IW1-M[1] was spiked at 500 ppb and then
re-analyzed at a spiking level of 10,000 ppb due to elevated
native concentrations of cDCE, TCE, PCE, VC and
toluene. At the appropriate spiking levels, all recovery
values were within control limits. Sample GW-F-IW3-B[2]
was noted to be thick and colored and was analyzed at a
dilution. The MS/MSD was analyzed at the same dilution
Table 4-28. Precision for Groundwater - Post-
treatment VOCs.
Compound
cDCE
TCE
PCE
VC
1,1 -DCA
1,1,1-TCA
Toluene
Maximum
RPD
34
6
16
22
2
3
17
No. of
Values
2
2
3
2
3
3
3
No. Outside
Limits
1
0
0
1
0
0
0
71
-------
4.5.5 Audits
4.5.5.1 Field Audit
A Technical Systems Audit (ISA) of the field activities at
the FormerMEC Building Site; Hudson, NH was conducted
May 13, 2005. There were no Findings requiring corrective
action. However, there were two observations noted. The
first observation was that the required number of baseline
peat zone samples (30) was not achieved during baseline
sampling. During baseline sampling using a sonic rig, peat
sample material was not collected at boring locations #s 9,
25, and 30, and was thought absent at these three
locations. In addition the peat material at boring # 2 was
fragmented and a portion of that samples included non-
peat material. As a result, there was at most 27 peat
samples instead of the desired 30.
In response to this observation, results were tabulated and
statistical parameters including mean, and 95% confidence
intervals were calculated. Based upon actual results it was
determined that the 27 samples collected during baseline
would satisfy the project objective. Even prior to this
calculation it was determined that a 90% completeness
objective would likely meet project objectives. The goal for
post-demonstration sampling was to collect at least as
many samples as collected during baseline. This indeed
was the case, however peat was found in borings drilled
adjacent to baseline borings#s 9 and 25 using a geoprobe.
The peat was again absent at boring # 30 (the closest
boring to the excavated tank area). It is postulated that the
peat had been excavated in the vicinity of boring 30. As for
the partial peat sample at boring #2, the auditor advised the
field manager to try to note the estimated composition of
sample material if such an occurrence happened again.
(This only happened once during baseline sampling).
The second observation noted some minorinconsistencies
in record keeping, including a missing calibration check of
the field balance used to measure this mass of soil sample
used for the field extraction procedure and an unsigned
logbook for previous days records. In response to this
concern the SAIC Project Manager met with some of the
same site personnel while working on another project and
explained and discussed what should be included in field
notebooks. This issue is often a continuing concern that
needs re-training on a regular basis.
4.5.5.2 Lab Audit
A Technical Systems Review (TSR) was performed at
Analytical Laboratory Services, Inc. (ALSI) in Middletown,
PA on May 25, 2005. In general ALSI met all QAPP and
method specifications and was expected to generate data
of known quality that meet project and data quality
objectives. The few issues to be addressed were either
resolved on-site or corrective action was expected to be
initiated promptly. These corrective actions included:
Initiation of the tracking of surrogate standards by,
at a minimum, lot # and date opened, either in the
standards logbook or on the log sheets where
medium level soil sample vials are prepared. ALSI
was requested to provide copy of documentation
to indicate that this corrective action had been
complete.
Initiation of the tracking of VOC gas standard by,
at a minimum, lot # and date opened, either in the
standards logbook or in the analytical logbook.
ALSI was requested to provide copy of
documentation to indicate that this corrective
action had been complete.
72
-------
5.0
Technology
5.1 Environmental
Requirements
Reg u lati on
State and local regulatory agencies may require permits
prior to implementing an in situ chemical oxidation
technology. Most federal permits will be issued by the
authorized state agency. The specific permits required
may depend on the type and concentration of the
contamination, the regulations covering the specific
location, and the site's proximity to residential
neighborhoods. For example, the New Hampshire
Department of Environmental Services (NHDES) requires
a site owner or legally responsible person to acquire a
Ground water Management Permit (GMP) from the State to
remedy contamination associated with the past discharge
of regulated contaminants, and to manage the use of
contaminated groundwater. The application for the GMP
must be prepared and stamped by a PE or a PG licensed
in the State. There is an application fee for this.
Since oxidant is injected underground, an injection permit
may also be required. For the former MEC Building site,
XDD was required to apply for and acquire an injection
permit from the NHDES. However, there was no actual
permit fee. Note that other states may require a fee. In
some cases permitting fees may be waived for
government-conducted research type projects.
An air permit issued by the state Air Quality Control Region
may be required if it is anticipated that the air emissions
from potential surface venting are in excess of regulatory
criteria, or are of toxic concern. Wastewater discharge
permits may be required if any wastewater generated from
well purging and equipment decontamination activities
were to be discharged to a POTW. If remediation is
conducted at a Superfund site, federal agencies, primarily
the U.S. EPA, will provide regulatory oversight. If off-site
disposal of contaminated waste (contaminated drill
cuttings) is required, the waste must be taken to the
disposal facility by a licensed transporter.
Section 2 of this report discusses the environmental
regulations that may apply to XDD's ISCO process.
5.2 Personnel
The number of personnel required to implement the XDD
ISCO process is dependent on the size of the treatment
system and the time desired for the installation. Initially,
drilling and well installation laboractivitiesare performed by
a drilling contractor. These activities typically involve a
minimum of two contractor personnel assigned to a drill rig
or geoprobe (head driller and helper). There may be an
additional contractor representative who conducts well
completion and development following well installation
(which can be conducted at the same time that additional
wells are being installed).
During well installation activities at a remediation site, the
site remediation contractor would be responsible for
logging boreholes, monitoring for VOCs and explosive
conditions during drilling of boreholes, and ensuring that
well construction and installation are conducted in
accordance with design specifications. These activities
would require the services of at least one individual
(preferably a geologist). XDD may or may not be present
for such activities.
Based on the demonstration study requirements, XDD
appears to need a minimum of two individuals to conduct
the oxidant injection. The oxidant batching process is a
labor intensive operation involving the shipment and
handling of numerous 20 L jugs of granular oxidant.
Estimated labor requirements for a full-scale treatment
system are discussed in detail in Section 3 of this report.
73
-------
Personnel are also required for sample collection and
groundwater monitoring. During the demonstration
sampling events, one to two SITE team members was
required to conduct field measurements during low-flow
well sampling via a multi-parameter instrument, collect
groundwater samples with a peristaltic pump, and prepare
the samples for shipment to an off-site laboratory.
Personnel present during sample collection activities at a
hazardous waste site must have current OSHA health and
safety certification. Specific to the XDD ISCO process and
chemical oxidation processes in general,the primaryhealth
and safety issue is personal protection from strong
oxidizers. From a health and safety and materials handling
perspective, NaMnO4 is handled in a liquid form and is
preferred to solid (i.e., granular) KMnO4. The application of
this liquid oxidant is simpler than injecting KMnO4, which
has to be pre-mixed with potable water at the desired
concentration prior to injection.
Per the MSDS for KMnO4, at a minimum, eye protection
and protective gloves are strongly recommended to be
worn during handling of oxidants. During the
demonstration, the batching process involved emptying
numerous 20 Liter jugs of granular KMnO4 into large batch
tanks. Due to the potential inhalation of particulate, XDD
also utilized full-face air-purifying respirators and
chemically-resistant tyvek in addition to the standard Level
D protection. Therefore the batch operation was
conducted in Level C. (Generally speaking, for most sites,
PPE for workers will include steel-toed shoes or boots,
safety glasses, hard hats during drilling operations, and
chemical resistant gloves).
Noise levels would be a short-term concern during drilling
operations and may be of concern during injection phases,
particularly near the piston pump that is contained within
the XDD POD. Noise levels should be monitored for such
equipment to ensure that workers are not exposed to noise
levels above the time weighted average of 85 decibels over
an 8-hour day. If this level is exceeded and cannot be
reduced, workers would be required to wear hearing
protection and a hearing conservation program would need
to be implemented.
5.3 Comimuinity Acceptance
The short-term risk to the community from implementing
this technology is minimal since the oxidant is injected into
the ground. In fact storage of the oxidant is of more
concern since granular oxidant can be combustible if not
stored properly.
The level of environmental disturbance of a site would be
dependent on the number of wells required and the
locations of those wells. For example, if injection or
monitoring wells were required to be installed in an area
having vegetation, some clearing of the vegetation may be
necessary to access the best injection or monitoring points.
This may affect habitat areas to some extent.
Other than the intermittent noise generated during drilling
and, the relatively minor level of noise generated during
oxidant injection is offset by the benefits of site
remediation.
74
-------
6.0
6.1 Previous
Chemical oxidation's use in the wastewater treatment
industry dates back many years. In the past 10 years or
so, the technology has been applied to remediation at
hazardous waste sites. Specific to the use of KMnO4 for
treatment of DNAPL and dissolved phase chlorinated
organic contaminants at hazardous waste sites, case
studies have been documented at least as far back as
1996 (EPA, 1998).
Xpert Design and Diagnostics, LLC (XDD) was founded in
1997 as a provider of innovative soil and groundwater
remediation technologies. Per its web site, XDD has
experience with a wide range of ISCO applications at
several sites. Their Portable Oxidant Delivery (POD) unit,
which was used during the demonstration, has been
operating since 2003. XDD staff is reported to have
worked on several hundred sites throughout the United
States, Canada, Europe and Asia for private industrial and
federal sector clients.
6.2 Ability to Scale Up
Based on the nature of the technology, theoretically there
is no limit to the areal extent of application of an ISCO
process as long as the oxidant injection design (i.e.,
injection well spacing, screened interval, and oxidant
dosage, etc.) is adequate. The areal extent of XDD's ISCO
treatment system implemented at the former M EC Building
site was roughly 1,200 ft2 (113 m2). This is considered a
pilot-scale application of the technology.
The pilot-scale demonstration is also not considered to be
a typical remediation, not only due to its small size but also
because of the atypical soil profile (i.e., the thin peat layer).
XDD has reported on their web site applications of their
ISCO process thatare much largerthan thatdetailed in this
report. One example includes a 16-injection point system,
covering an area of approximately 1,400 ft2, (130 m2) and
utilizing automated batching and dosing equipmentcapable
of injecting up to 240 liters of oxidant per minute. A second
example included treatment of an approximate 56,000 ft2
(5,200 m2) section of an approximate 3,300 ft (100 m) long
VOC plume.
XDD reports considerable research forthe development of
an advanced oxidation process, referred to as X-Ox. This
process is intended to use persulfate alone, or used in
conjunction with a proprietary transition metal to treat
petroleum hydrocarbons, MTBE, chlorinated solvents, coal
tar residues, PCBs, and energetic compounds.
75
-------
Section 7.0
References
Aries Engineering, Inc. 2006. 2006 Draft Preliminary
Remediation Budget Estimate Update, Former MEC
Building Site, Hudson, New Hampshire. June 2006.
Aries Engineering, Inc. 2004. 2003 Annual Groundwater
Monitoring Summary Report, Former MEC Building Site,
Hudson, New Hampshire. January 2004.
Aries Engineering, Inc. 2001. Remedial Action Plan Re-
Evaluation, Former MEC Building Site, Hudson, New
Hampshire. July 2001.
Aries Engineering, Inc. 2001. Revised Remedial Action
Plan, FormerMEC Building Site, Hudson, New Hampshire.
January 2001.
Chem One. 2004. Product Specification for KMnO4.
Revised February 23, 2004.
Christy, Thomas M. A Permeable Membrane Sensor for
the Detection of Volatile Compounds in Soil. No Date.
Hansen, Morris H., Hurwitz, William N., Madow, William G.
1993. Sample Survey Methods and Theory: Volume II,
Theory. John Wiley & Sons, Inc., New York, 1993.
Natrella, Mary Gibbons, Experimental Statistics, National
Bureau of Standards Handbook 91. U.S. Government
Printing Office, Washington, D.C. 1963.
New Hampshire Department of Environmental Services.
2001. Risk Characterization and Management Policy, rev.
April 3,2001,www.des.state.nh.us/orcb/doclist/pdf/rcmp.pdf
SAIC. 2005. Quality Assurance Project Plan - XDD In-Situ
Chemical Oxidation Process at the Former MEC Building
Site, Hudson, NH; May 2005 Revision 1.
U.S. Environmental Protection Agency. 2006. EPA
Groundwater and Drinking Water- Current Drinking Water
Standard Web Page
www.epa.gov/safewater/mcl.htmltfmcls. June 2006.
U.S. Environmental Protection Agency. 2002. Guidance on
Data Quality Indicators. EPA G5i. Washington, D.C. July
2002.
U.S. Environmental Protection Agency. 1998. Field
Applications of In Situ Remediation Technologies -
Chemical Oxidation. EPA/542-R-980-008. September
1998.
U.S. Environmental Protection Agency. 1996. Test
Methods for Evaluating Solid Waste, Physical/Chemical
Methods, SW-846 CD ROM, which contains updates for
1986, 1992, 1994, and 1996. Washington, D.C.
U.S. Environmental Protection Agency. 1991. Engineering
Bulletin - Chemical Oxidation. EPA/540/2-91/025 October
1991.
Xpert Design and Diagnostics LLC. 2006 Equipment Fee
Schedule.
Xpert Design and Diagnostics LLC. 2005. Chemical
Oxidation Laboratory Testing Report - Former MEC
Building Site, Hudson, NH. January 2005.
Xpert Design and Diagnostics LLC Web site.
www.xdd-llc.com
76
------- |