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

-------