-
\
V
*"« PHOT*0
542R04008
Case Study of the Triad Approach: Expedited
Characterization of Petroleum Constituents and
PCBs Using Test Kits and a Mobile
Chromatography Laboratory at the Former
Cos Cob Power Plant Site
-------
-------
Case Study of the Triad Approach: Expedited
Characterization of Petroleum Constituents and
PCBs Using Test Kits and a Mobile
Chromatography Laboratory at the Former
Cos Cob Power Plant Site
Prepared by:
U.S. Environmental Protection Agency
Office of Superfund Remediation and Technology Innovation
Brownfields Technology Support Center
Washington D.C. 20460
In cooperation with:
U.S. Environmental Protection Agency Region 1 and Metcalf & Eddy
ussy
-------
-------
Notice
This material has been funded wholly by the United States Environmental Protection Agency (EPA)
under Contract Number 68-W-02-034. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
Copies of this report are available free of charge from the National Service Center for Environmental
Publications (NSCEP, P.O. Box 42419, Cincinnati, OH 45242-2419; telephone (800) 490-9198 or (513)
489-8190 (voice) or (513) 489-8695 (facsimile). Refer to document EPA 542-R-04-008. This document
can also be obtained electronically through EPA's Clean Up Information (CLU-IN) System on the World
Wide Web at http://cluin.org.
Comments or questions about this report may be directed to Dan Powell, EPA, Office of Superfund
Remediation and Technology Innovation (5102G), 1200 Pennsylvania Avenue NW, Washington, D.C.
20460; telephone (703) 603-7196.
-------
Foreword
This case study is one in a series designed to provide cost and performance information for innovative
tools that are available to support less costly and more representative site characterizations. Case studies
developed by the EPA Office of Superfund Remediation and Technology Innovation Office (OSRTI) are
designed to introduce project managers and technical team leaders the use of new technologies and novel
applications of familiar tools or processes. Case studies such as this one include detailed information
about the technologies and strategies used in this Triad project. Briefer project profiles for this and other
types of projects that exemplify how the concepts of the Triad approach have been used can be found at
www.triadcentral.org.
Acknowledgments
Special acknowledgement is given to U.S. EPA Region 1 and to the staff of Metcalf & Eddy for their
thoughtful suggestions and support in preparing this case study.
11
-------
Table of Contents
Notice ii
Foreword '. iii
Acknowledgements ....iii
CASE STUDY ABSTRACT ix
TECHNOLOGY QUICK REFERENCE SHEETS
Total Petroleum Hydrocarbons (TPH) and Polynuclear Aromatic Hydrocarbons (PAH) By
Ultraviolet Fluorescence (UVF) xi
Polychlorinated Biphenyls (PCB) By Gas Chromatography/Electron Capture Detector xii
EXECUTIVE SUMMARY 1
SITE INFORMATION 4
SITE LOCATION AND DESCRIPTION 4
SITE HISTORY AND USE 5
PROPERTY REUSE SCENARIO 6
CONCEPTUAL SITE MODEL 6
Initial Site Investigations 7
Current Site Use 10
Site and Regional Geology 10
MEDIA OF CONCERN .'. 11
CONTAMINANTS OF POTENTIAL CONCERN 11
EXPOSURE ROUTES AND RECEPTORS 13
WORK PLAN DEVELOPMENT 14
INITIAL WORK PLAN '14
Site-Specific Objectives for Initial Work Plan 14
Development of the Original Sampling Approach 15
Original Sampling Rationale 15
SYSTEMATIC PLANNING 16
The New Sampling Approach 17
Analytical Options 18
Developing Decision Logic 19
DEMONSTRATION OF METHODS APPLICABILITY 22
Initial Field Sampling Event Conducted in Support of the Demonstration of Methods
Applicability Study 23
FIELD INVESTIGATION 28
SAMPLE COLLECTION 28
SAMPLE ANALYSIS.. 29
SITELAB® UVF TEST KITS 29
EPA Region 1 Mobile Laboratory Polychlorinated Biphenyl Analyses 30
EPA Region 1 Mobile Laboratory XRF 31
in
-------
TABLE OF CONTENTS (Continued)
Section ' Page
ARSENIC RESULTS 32
PAH RESULTS 33
Correlating Field PAH Results with the Connecticut Residential Direct Exposure
Criterion 34
Understanding Field-based Results for PAHs by Reviewing Sample Chromatograms.... 35
Summary of Findings for Field-generated PAH Data 35
TPH RESULTS 36
Correlating Field-based Results for TPH with the Residential Connecticut Direct
Exposure Criterion 36
Understanding Field-based Results for TPH by Reviewing Sample Chromatograms 38
Summary of Findings for TPH 38
PCB RESULTS 38
LESSONS LEARNED 40
COST COMPARISON 42
BIBLIOGRAPHY 44
TABLES
Tables
1 PATHWAY RECEPTOR DIAGRAM
2 LIST OF CONTAMINANTS OF POTENTIAL CONCERN
3 SUMMARY OF ORIGINALLY PROPOSED SAMPLES
4 SUMMARY OF ANALYTICAL DATA FROM THE DEMONSTRATION OF METHODS
APPLICABILITY STUDY
5 COST COMPARISON BETWEEN A TRADITIONAL AND A TRIAD APPROACH
IV
-------
FIGURES
Figure
1 SITE LOCATION MAP
2 CONCEPTUAL SITE MODEL DIAGRAM
3 HISTORICAL AND INITIALLY PROPOSED SAMPLING LOCATIONS
4 PROPOSED SAMPLING LOCATIONS (TRIAD APPROACH)
5 SAMPLING LOCATIONS, FEBRUARY 2003
6 LOGIC DIAGRAM 1 - ARSENIC IN FLY ASH INVESTIGATION
7 LOGIC DIAGRAM 2 - TPH, PAH, AND PCB INVESTIGATION
8 ARSENIC RESULTS SUMMARY, EPA SW-846 METHOD 601 OB
9 FIXED LABORATORY TOTAL CARCINOGENIC PAHs vs. FIELD TEST KIT TOTAL PAHs
9A DEVELOPMENT OF A FIELD BASED ACTION LEVEL FOR CARCINOGENIC PAHs
WHEN NO HISTORICAL DATA IS AVAILABLE
10 TOTAL PAH RESULTS SUMMARY, SITE LAB FIELD TEST KITS
11 FIXED LABORATORY TPH vs. FIELD TEST KIT TPH
11A DEVELOPMENT OF A FIELD BASED ACTION LEVEL FOR TPH WHEN NO
HISTORICAL DATA IS AVAILABLE
12 TPH RESULTS SUMMARY, SITE LAB FIELD TEST KITS
13 PCB RESULTS SUMMARY, FIELD AND FIXED LABORATORY RESULTS
14 TOTAL PCB RESULTS SUMMARY, FIELD AND FIXED LABORATORY RESULTS
ENCLOSURES
Enclosures
1 SOME METHOD VALIDATION ISSUES FOR THE RCRA PROGRAM: THE FORMAL
VALIDATION PROCESS FOR NEW METHODS DEVELOPMENT AND
DEMONSTRATION OF METHOD APPLICABILITY FOR A SPECIFIC PROJECT
2 STATISTICAL PLOTS AND SUMMARY STATISTICS FOR PROJECT TARGET
ANALYTES
3 SELECTED CHROMATOGRAMS FOR SAMPLES USED TO DEVELOP FIELD-BASED
ACTION LEVELS
-------
ACRONYMS AND ABBREVIATIONS
95LCL
95UCL
ACM
bgs
BTSC
CLP
COPC
cPAH
CSM
CTDEP
CTDPH
DMA
DEC
EA
ECD
EDRO
EPTOX
EPA
ETPH
GC/ECD
GC/FID
GC/MS
GPS
ICAP
IOC
M&E
uL
mfl
mL
mg/kg
OEME
OVA
PAH
PCB
ppb
ppm
95 percent lower confidence level
95 percent upper confidence level
Asbestos-containing material
Below ground surface
Brownfields Technology Support Center
EPA Contract Laboratory Program
Contaminant of potential concern
Carcinogenic polynuclear aromatic hydrocarbon
Conceptual site model
Connecticut Department of Environmental Protection
Connecticut Department of Public Health
Demonstration of methods applicability
Residential direct exposure criterion
Environmental assessment
Electron capture detector
Extended Diesel-Range Organics
Extraction procedure toxicity test
U.S. Environmental Protection Agency
Extractable total petroleum hydrocarbon
Gas chromatography method using an electron capture detection system
Gas chromatograph/flame ionization detector
Gas chromatography/mass spectrometry
Global positioning system
Inductively coupled argon plasma
Inorganic carbon
Metcalf&Eddy
Microliter
Micrograms per kilogram
Million fibers per liter
Milliliters
Milligram per kilogram
Office of Environmental Measurement and Evaluation
Organic vapor analyzer
Polynuclear aromatic hydrocarbon
Polychlorinated biphenyl
Part per billion
Part per million
VI
-------
ACRONYMS AND ABBREVIATIONS (Continued)
QC Quality control
RAS Routine analytical services
SIM Selected ion monitoring
SOP Standard operating procedure
SPLP Synthetic precipitation leaching procedure
SVOC Semivolatile organic compound
TBA Targeted brownfields assessment
TCLP Toxicity characteristic leaching procedure
TPH Total petroleum hydrocarbons
TRC TRC Environmental Consultants
UV Ultraviolet
UVF Ultraviolet fluorescence
VOC Volatile organic compound
WAM Work assignment manager
XRF X-ray fluorescence
vn
-------
CASE STUDY ABSTRACT
Cos Cob Power Plant Site
Greenwich, Connecticut
Site Name and Location:
Cos Cob Power Plant
22 Sound Shore Drive
Greenwich, CT 06830
Period of Operation:
1907-mid 1960s
Operable Unit:
The property where the former Cos
Cob power plant was located is
approximately 9 acres in size.
Sampling and Analytical Technologies:
• Systematic planning process
• Dynamic work strategies
• Direct push soil sampling
• Field measurements of total polynuclear
aromatic hydrocarbons (PAH) by ultraviolet
fluorescence spectrophotometry
• Field measurements of total petroleum
hydrocarbons (TPH) by ultraviolet fluorescence
spectrophotometry.
• Field measurements for polychlorinated
biphenyls (PCBs) by gas chromatograph/
electron capture detector (ECD).
CERCLIS #:
None
Current Site Activities:
The Town of Greenwich,
Department of Public Works
has recently used the property
for storage of construction
materials.
The town of Greenwich
Connecticut is also evaluating
re-use and remediation
alternatives for the property.
Points of Contact:
Denise M. Savageau
Town of Greenwich
101 Fields Point Road
Greenwich, CT 06830
203-622-6461
James P. Byrne
EPA Region 1
1 Congress St., Ste. 1100
Boston, MA 02114-2023
617-918-1389
Kathleen M. Yager
EPA Technology
Innovation Program
11 Technology Drive
North Chelmsford, MA
01863
617-918-8362
Media and Contaminants:
Surface and subsurface soil site wide:
• Soil may be affected by TPH and PAHs from
petroleum spills on site.
• Contaminants associated with on-site disposal
of coal ash, including fly ash and slag.
Chemicals of potential concern (COPCs)
include PAHs and arsenic.
• Site soil may be affected by PCBs associated
with storage and use of transformers. Known
spills have occurred on the property.
• Asbestos is a COPC based on historical
occurrence and previous assessments but is not
addressed in this case study.
Ground-water:
• Groundwater was sampled in 1988, and no
contamination was found. Based on the lack of
water use and tidal influences in the area,
groundwater is not considered a medium of
potential concern at this time.
Technology Demonstrator:
siteLAB® ultraviolet
fluorescence kits for PAHs and
TPH. The EPA Region 1
laboratory provided single
column GC/ECD analyses in
the field for PCBs and
screening for metals using
XRF. See Technology Quick
Reference Sheets for additional
information.
Number of Samples Analyzed During the Investigation:
A total of 112 samples were collected and analyzed at an off-site laboratory for arsenic. A total of 93 samples were collected and
analyzed in the field for PAHs and TPH. Of the 93 samples, 23 were sent for comparative analysis for PAHs and 17 samples
were sent for comparative analysis for TPH at an off-site laboratory. A total of 103 samples were also analyzed for PCBs on-site
using a single column GC/ECD. Fifteen of these samples were also sent for off-site analyses of PCBs by a dual column GC/ECD
method.
Cost and Time Savings:
Use of the Triad approach for site characterization resulted in an estimated cost savings of approximately 35 percent when
compared with a traditional approach, assumed to involve two mobilizations and fixed laboratory analytical methods. In addition
to saving costs, use of the Triad approach increased the size and quality of the data set used to make decisions about the site. Site
characterization was achieved in a single mobilization lasting 1 week. Sufficient data necessary to make site decisions was
collected in a single Brownfield's funding cycle (1 year). Site characterization following a traditional approach would have
required multiple mobilizations taking place over 2 Brownfield's funding cycles (2 years).
Vlll
-------
Results:
The project was completed successfully and cost-effectively. The relatively high data density for the site allowed regulators and
stakeholders to make decisions with a high degree of certainty. Members of the project team were very satisfied with the time
and cost savings achieved following the principles of the Triad approach.
Description:
The Cos Cob Power Plant case study is an example of how site investigation can be streamlined using systematic planning,
dynamic work strategies, and field-based measurements. Work originally planned to be conducted over a two-year period was
compressed into a single 1-week mobilization. Data needed by the town of Greenwich, Connecticut was collected in a single
event. A preliminary conceptual site (CSM) model was evolved to near-maturity in real-time through field-based decision
making, providing stakeholders with the data they needed in a fraction of the time required when using a traditional phased
approach. Significant time and cost savings were realized when compared to a traditional multi-phased approach.
The Cos Cob Power Plant site was accepted by the U.S. Environmental Protection Agency (EPA) for the Targeted Brownfields
Assessment (TBA) program in June 2002. Metcalf & Eddy, Inc. (M&E) was retained in the fall of 2002 to carry out the work,
which was estimated to cost $75,000 for the field effort. The 9-acre waterfront property is the location of the former Cos Cob
Power Plant, which was a coal-fired trunk-line electrical generation facility. Operations were terminated in the 1960s. Historical
industrial activities have resulted in the potential increase of contamination such as asbestos, coal, ash, transformer fluids and
slag on site.
The plant was decommissioned in 1986; in 1987, the site was deeded to the town by the Connecticut Department of
Environmental Protection (CTDEP) with the understanding that the property would eventually be open to all residents of the
state. Since 2000, the Town of Greenwich, Department of Public Works has used the property for storage of construction
materials.
The Town of Greenwich has established a Cos Cob Power Plant Committee made up of citizens interested in redevelopment of
the site. Public interest is high and support is overwhelming for restoration of the property as a public recreational site.
Restoring the site will provide public access to important coastal resources and protect important fisheries and wildlife habitat
associated with the Cos Cob Harbor, Mianus River, and Long Island Sound. Proposed developments for the property include a
waterfront public park with walking trails, playing fields, picnic areas, and a boating facility. The site is located in an area where
most shoreline is privately owned; which limits public access to coastal resources. As a public site, public access and
recreational opportunities would be provided.
Based on a review of historical aerial photographs and conversations with town representatives, the area south of the former
powerhouse is almost entirely composed of fly ash from the former power plant. It is estimated that 22 to 35 feet of fly ash
material exists on and beneath this portion of the property. It is also estimated that 30 feet of fill material made up of coal, slag,
and ash is present beneath the northeastern portion of the site.
Remediation costs for the site will be included in the town's capital budget. Redevelopment will be financed through public and
private partnerships. The town faces some challenges in reaching closure, however. For example, the results obtained during the
February 2003 Triad-based TBA site investigation indicate that contamination from total petroleum hydrocarbons (TPH),
polynuclear aromatic hydrocarbons (PAHs), and arsenic was found at levels that generally exceed Connecticut residential direct
exposure criteria (DEC) to a depth of possibly 30 feet. Although elevated concentrations of TPH and PAH appear to be mainly
associated with releases of fuels and other hydrocarbons, lower concentrations of these constituents are also widespread at the
site at depths well below ground surface. As a result of the arsenic, PAH, and TPH contamination, reuse alternatives and
potential closure requirements may be revised because removal of soil necessary to meet residential DEC for these contaminants
would not be cost effective. Some viable alternatives for closure in support of reuse might include emplacement of fill material
over portions of the site or development of alternative closure standards based on a risk assessment and more realistic exposure
assumptions.
IX
-------
TECHNOLOGY QUICK REFERENCE SHEET #1
Total Petroleum Hydrocarbons (TPH)
And Polynuclear Aromatic Hydrocarbons (PAH)
By Ultraviolet Fluorescence (UVF)
Summary of Project-Specific Performance Information
Project Role:
Provide real time results
to guide dynamic
sampling activities for a
Triad investigation. Test
kit concentrations were
correlated to fixed
laboratory concentrations
to develop field-based
action levels.
Analytical Information Provided:
A total of 93 samples were collected, extracted, and analyzed for Total Polynuclear ,
Aromatic Hydrocarbons (PAHs) (Cl 1-C22 Aromatics) using a UV3100 fluorescence
detector. Raw fluorescence results were used in conjunction with the Total Petroleum
Hydrocarbon (TPH) forecasting guide to generate TPH Extended Diesel Range
Organics (TPH-EDRO) results. Ultraviolet Fluorescence (UVF) results for total PAHs
in 23 samples were compared to fixed laboratory results using SW-846 method 8270
operated in the selected ion monitoring (SIM) mode. UVF results for TPH in 16
samples were compared to fixed laboratory results using SW-846 method 8015M.
Project Cost and Time Savings
Total Cost (includes instrument rental,
consumables, and labor): $7,300
Total Cost Per Sample (includes instrument rental,
consumables, and labor): $78
Instrument Cost:
UV3100 Purchase Price:
$12,500
Rental: $300/day
$600/3 days
$ 900/week
$2,500/month
Consumables Cost:
Extraction Kits
(20 samples)= $300
5 kits used=$ 1,500
Calibration Kit=$200
2 kits used= $400
Labor Cost:
$487 Sample
Waste Disposal Cost:
Not available. Disposal
cost for small amounts of
methanol and extract are
assumed to be minimal.
Time Savings:
1 Year
Site characterization was
achieved in a single mobilization
lasting 1 week. Sufficient data
necessary to make site decisions
was collected in a single
Brownfield's funding cycle (1
year). Site characterization
following a traditional approach
would have required multiple •
mobilizations taking place over
2 Brownfield's funding cycles (2
years).
Site-Specific Precision and Accuracy Achieved:
Relative Percent Difference
RPD =
IA-BI
x 100
Throughput Achieved:
93 samples
(A+ B) / 2
Comparability (UVF result/fixed lab result) x 100
Total PAHs (Cl 1-C22 Aromatics)
Duplicate RPDs: 9% to 10%
Split Sample Comparability: Not Applicable
TPH
8% to 10%
43% to 186%
-------
TECHNOLOGY QUICK REFERENCE SHEET #1 (continued)
Total Petroleum Hydrocarbons (TPH)
And Polynuclear Aromatic Hydrocarbons (PAH)
By Ultraviolet Fluorescence (UVF)
.. ••••*.
Vendor Contact:
Steve Greason
978-363-2299
Vendor Information:
siteLAB Corporation
4 Crane Neck Street, West
Newbury,MA01985
877-748-3522
www.site-lab.com
Limitations on Performance:
The test kit reports total PAHs as opposed to specific PAH
compounds. TPH values are estimated as diesel-range
organics based on the raw sample fluorescence. The shelf life
of the calibration kits is 3 months after purchase.
Principle of Analytical Operation:
This test is based on the excitation of aromatic
compounds, such as benzene, toluene,
ethylbenzene, and xylenes (BTEX) and PAHs, in
petroleum hydrocarbons exposed to ultraviolet
(LTV) light. The compounds both absorb and emit
(fluoresce) energy at specific wavelengths, so an
energy filter is used to allow only wavelengths
specifically absorbed by aromatic compounds to
reach the cuvette containing the sample (or
sample extract). Fluorescence emissions from the
sample are then passed through another filter to
reach a photomultiplier detector. The result is a
total PAH, TPH, or BTEX concentration for each
sample, depending on how the instrument is
calibrated. Sample results are quantitated based
on a 5-point calibration curve generated using
certified standards of interest (PAH, TPH, BTEX,
or a combination) for a given project. Water
samples are analyzed directly after a simple 10X
or 100X dilution. Soil samples are extracted by
shaking for 2 minutes in methanol; then, the
extract is filtered and diluted for analysis.
Availability/Rates:
Test kits are commercially available as off-the-shelf products.
Associated extraction kits and UV fluorescence detectors are
available for purchase or rental from the manufacturer.
Power Requirements:
120 volts of AC power is required for the UV3100
fluorescence detector. The unit can be operated using a
vehicle cigarette lighter or portable generator.
Instrument Weight and/or Footprint:
Approximately 5 square feet of space is required to run
sample extract batches, analyze samples using the UV3100,
and download information from the UV3100 to a laptop
computer.
GEHBRAL PERFORMANCE INFORMATION
Known or Potential Interferences:
Interferences are limited because the excitation and emission filters limit fluorescence from non-target analytes.
Site-specific matrices also provide ratios of PAH compounds from site contaminants. Soil with high
concentrations of naturally occurring humic materials can produce low-level false positive results in rare cases.
Applicable Media/Matrices:
Soil/Water
Wastes Generated
Requiring Special Disposal:
Small volumes of methanol
used for sample extraction
and small volumes of sample
extract.
Analytes Measurable with
Expected Detection Limits:
Total PAHs-0.05
milligram/kilogram (mg/kg)
Extended Diesel-Range Organics
(EDRO) - 0.05 mg/kg
(values are wet weight)
Other General Accuracy/Precision
Information:
See www.site-lab.com for technology
evaluations, case studies, and detection
limit studies.
Rate of Throughput:
Samples can be prepared in batches of 10
with a batch preparation time of about 5
minutes. Analysis of individual samples
can be completed in about 30 seconds.
XI
-------
TECHNOLOGY QUICK REFERENCE SHEET #2
Polychlorinated Biphenyls (PCBs)
By Gas Chromatography/Electron Capture Detector (GC/ECD)
Summary of Project-Specific Performance Information:;
Project Role:
Provide real time screening
results for Polychlorinated
Biphenyls (PCBs). Results
were used to guide step-out
sampling for contaminant
delineation and select
samples for definitive fixed
laboratory analysis.
Analytical Information Provided:
103 soil samples were collected, extracted, and analyzed for PCBs using a single-
column Gas Chromatography (GC) instrument with an Electron Capture Detector •
(ECD) for tentative identification and semi-quantitation of PCB congeners. All field
samples with tentatively identified concentrations of PCBs were sent for offsite
definitive analysis using a dual column GC/ECD method. Confirmation samples were
analyzed following EPA Region 1 Standard Operating Procedures (SOP)
(PESTSOIL2.SOP).
Project Cost and Time Savings
Total Cost:
EPA Region 1 provided mobile laboratory services
at no cost to the project.
Total Cost Per Sample:
EPA Region 1 provided mobile laboratory services at no cost
to the project.
Instrument Cost:
ShimadzuGC-14with
ECD and data module.
Purchase price: $12,000
Rental costs were not
available because EPA
Region 1 provided mobile
laboratory services at no
cost to the project.
Consumables Cost:
Not available
Labor Cost:
EPA Region 1 provided
the mobile laboratory at
no cost to the project.
Waste Disposal Cost:
Not available. Disposal
cost for small amounts of
methanol, hexane, water,
and extract are assumed to
be minimal.
Time Savings:
1 Year
Site characterization was
achieved in a single mobilization
lasting 1 week. Sufficient data
necessary to make site decisions
was collected in a single
Brownfield's funding cycle (1
year). Site characterization
following a traditional approach
would have required multiple
mobilizations taking place over
2 Brownfield's funding cycles (2
years).
Site-Specific Precision and Accuracy Achieved:
Relative Percent Difference
RPD
IA-BI
(A+ B) / 2
x 100
Throughput Achieved:
103 samples
Comparability (Mobile lab result/fixed lab result) x 100
Duplicate RPDs:
Split Sample Comparability:
Total PCBs
14%to21%a
13%tol30%a
Notes:
" (mobile laboratory results in wet weight/ fixed laboratory results in dry weight)
Xll
-------
TECHNOLOGY QUICK REFERENCE SHEET #2 (continued)
Polychlorinated Biphenyls (PCBs)
By Gas Chromatography/Electron Capture Detector (GC/ECD)
General Commercial Information (information valid as of October 2003)
Vendor Contact:
U.S. EPA
Region I Mobile
Laboratory
Scott Clifford/
Dick Siscanaw
617-918-8327
Vendor Information:
U.S. EPA Region I
Laboratory
USEPA New England
Office of Environmental
Measurement and
Evaluation (OEME)
11 Technology Drive
N. Chelmsford, MA
01863-2431
Limitations on Performance:
Analyses were performed on a single-column GC/ECD .
instrument for tentative identification and semi-quantitation of
PCB congeners. Detections by the single-column method can
be false positives, and all detections found using the mobile
field laboratory were confirmed by off-site analysis using a
dual-column method. Confirmation samples were analyzed
following EPA Region 1 Standard Operating Procedures
(SOP) (PESTSOIL2.SOP).
Principle of Analytical Operation:
This analysis is based on a micro-extraction of a soil
sample and analysis of the extract by GC/ECD.
Approximately 1 gram of each sample aliquot was
weighed in a 4-milliliter vial and a mixture of
methanol, hexane, and reagent-grade water was
added to perform the extraction. Each sample was
placed on a vortex for approximately 1 minute and
then centrifuged. A portion of the resulting extract
was injected directly onto the gas chromatograph
column, and results were graphed by the ECD.
Concentrations were evaluated using an external
standard technique. Dilutions were made as
necessary to samples that exceeded the calibration
range.
Availability/Rates:
The EPA Region I mobile laboratory is available to assist in
analysis of samples for investigations at Brownfield's sites.
Requests should be submitted to the EPA Region I
Laboratory. The Region I Mobile Laboratory is also available
to support Superfund site cleanups.
Power Requirements:
The mobile laboratory provides all of its own power and is
completely self-contained. The mobile laboratory comes
complete with all reagents and tools to perform extraction and
a Shimadzu model GC with ECD was used for analysis.
Instrument Weight and/or Footprint:
Bench-top GCs generally weigh between 100 and 200
pounds, but can be less than 100 pounds. Laboratory space
required is controlled by the need for sample preparation and
extraction. Documentation can also increase the need for
additional space in the laboratory.
GENERAL PERFORMANCE INFORMATION
Known or Potential Interferences:
Interferences are limited, but false positives can occur. All samples with potential PCB detections identified from the
mobile laboratory were sent for confirmation analysis using a dual-column GC/ECD method according to EPA
Region 1 SOP (PESTSOIL2.SOP), which is based on EPA SW-846 Method 8082.
Applicable Media/Matrices:
Soil/Water
Wastes Generated
Requiring Special Disposal:
Small volumes of methanol
and hexane used for sample
extraction and small volumes
of sample extract.
Analytes Measurable with
Expected Detection Limits:
PCBs - 1 milligram per
kilogram (mg/kg) wet weight
Other General Accuracy/Precision Information:
Soil samples were analyzed for PCBs following
EPA Region I SOP "PCBs Field Testing for Soil
and Sediment Samples" (EIA-FLDPCB2.SOP)
Rate of Throughput:
Sample preparation is about 3 to 5 minutes per
sample. Analysis of individual samples can be
completed in 10 minutes.
Xlll
-------
-------
Cos Cob Power Plant Site
EXECUTIVE SUMMARY m—~——^—^—~~—i
The following case study was prepared by the Brownfields Technology Support Center (BTSC), within
the U.S. Environmental Protection Agency's (EPA) Office of Superfund Remediation and Technology
Innovation (OSRTI). The case study was developed as part of EPA's ongoing initiative to promote the
use of an integrated Triad approach to limit decision uncertainty at hazardous waste sites through the use
of sound science. The Triad approach, which consists of systematic planning, dynamic work strategies,
and real-time measurement technologies that include field-based analyses, is being promoted by OSRTI
and its partners as a viable method for streamlining site investigations.
The Cos Cob Power Plant site is located in the southeastern corner of Connecticut, and is adjacent to Cos
Cob Harbor. The site was the location of a former power plant, which has since been demolished. The
current owner of the site is the Town of Greenwich, Connecticut. The town plans to reuse the site and has
received a Targeted Brownfields Assessment (TBA) grant to assess potential reuse options. Reuse
alternatives proposed for the site include creation of a walking trail and wetlands and a series of playing
fields. EPA Region 1 requested assistance from the BTSC in an attempt to maximize the efficiency of the
TBA by applying the Triad approach. A scoping meeting was held between EPA Region 1 and the BTSC
during November of 2002. This initial planning meeting discussed a revised approach to site
characterization that would rely on the use of field-based measurement technologies and the Triad
approach.
A preliminary conceptual site model (CSM) was developed based on a review of existing data from
previous investigations. The CSM indicated that potential threats to human health and the environment
were essentially limited to those posed by direct contact with contaminated surface soil and sediment.
Contaminants of potential concern included asbestos, petroleum-related substances, polychlorinated
biphenyls (PCBs), and arsenic. A review of past use and analytical results showed that contamination
could extend to some depth beneath the site and that the distribution mechanism was known to be related
to placement of coal ash as fill across the site, petroleum spills, or to storage of transformers that
contained PCBs. It was agreed that a primarily field-based approach could be used to expand sampling
and analytical coverage at the site and that a dynamic work strategy would likely be beneficial to assist in
further delineation of contaminants at the site, particularly for PCBs.
The initial work plan developed by Metcalf & Eddy (M&E) (2002) called for limited authoritative
sampling and analysis, using a fixed off-site laboratory, at locations where historical releases were
expected. A second phase of work would have been required under the original project plan. A revised
"dynamic" sampling strategy based on the Triad approach was proposed by the BTSC that called for use
1
-------
of test kits, field methods and a grid sampling approach to affordably expand the extent and density of
information available to support decision-making. The revised work plan, developed by M&E (2003), in
cooperation with the BTSC and the EPA Region 1 laboratory, called for the use of ultraviolet
fluorescence test kits for analysis of total petroleum hydrocarbons (TPH) and polynuclear aromatic
hydrocarbons (PAH) and a field-based single column gas chromatography (GC) method using an electron
capture detection system for analysis of PCBs. To meet project sensitivity requirements, samples were
analyzed for arsenic at an off-site laboratory using trace inductively coupled argon plasma (ICAP) atomic
emission spectrometry. Analysis at a fixed laboratory was used to confirm field-based results for PCBs
for samples with positive detections. A modified GC/mass spectrometry method using the instrument in
the selected ion-monitoring mode was also to be used to refine the correlation between results for PAHs
from the field test kits. Assumptions concerning the appropriateness of limiting metals analysis to a
single analyte (arsenic) were confirmed by analyzing a percentage of the total soil samples collected for a
full suite of metals using an off-site ICAP method.
The field effort was completed in one week. Direct-push methods were used to collect soil samples from
1 -foot intervals across the site. Initially, samples from only the top two 1 -foot depth intervals were
analyzed in the field, and a small percentage of the samples collected were sent for off-site comparative
analyses. Deeper sampled intervals were analyzed selectively based on the presence of PCBs. EPA
Region 1 's mobile laboratory performed the field analyses for PCBs that were used to guide the
investigation. All samples that tentatively identified the presence of PCBs were sent for comparative
analysis at an off-site laboratory.
The town was interested in evaluating the potential need for remediation based on a comparison between
contaminant concentrations measured at the site and a residential reuse scenario. In Connecticut,
residential reuse criteria are generally applied to recreational reuse scenarios. Results obtained during the
investigation indicated that the top 2 to 3 feet of soil contained concentrations of TPH, PAHs, arsenic, and
PCBs that consistently exceed the residential reuse criteria. (Connecticut defines surface soil as 0 to 4
feet below ground surface [bgs].) A review of past disposal practices at the site suggested that the same
conditions could extend to as deep as 30 feet bgs, thereby rendering excavation and disposal of
contaminated material to meet residential standards cost prohibitive. Concentrations of contaminants of
potential concern (COPCs) are relatively low given the planned reuse for the site, suggesting that
modification of the reuse alternative or establishment of reasonable and protective action levels could
facilitate reuse with only minor amounts of cleanup required for hot spots that contain PCBs.
Field-based technologies and unaligned grid sampling were employed to increase site coverage and limit
decision uncertainty. A dynamic work strategy was used to further delineate PCB contamination,
focusing on areas where positive detections were above the DEC and some additional characterization
2
-------
was warranted to support estimating the cost of remediation. Estimated cost savings as compared with
the use of a more traditional, phased approach were calculated at approximately 35 percent. Time was
also saved in that the project was completed in a single investigation and TBA funding cycle.
-------
Cos Cob Power Plant Site
SITE INFORMATION m^^^^^—^—m^^^^^^^^^^^—
The Cos Cob Power Plant site is located in the southeastern corner of Connecticut and is adjacent to Cos
Cob Harbor. The site was the location of a former power plant, which has since been demolished. The
current owner of the site is the Town of Greenwich, Connecticut. The town is planning for reuse of the
site and has received a Targeted Brownfields Assessment (TEA) grant to assess potential reuse options.
The U.S. Environmental Protection Agency (EPA) Region 1 requested assistance from the Brownfields
Technology Support Center (BTSC) to increase the efficiency of the TEA by applying the Triad
approach, an integrated method to limit decision uncertainty at hazardous waste sites through sound
science using systematic planning, dynamic work strategies, and real-time measurement technologies that
include field-based analyses.
Site description and background information presented in this section were summarized from reports
provided by the Town of Greenwich, and from conversations and a site visit with town representatives
and Metcalf & Eddy (M&E), contractor to EPA Region 1, held on July 31, 2002. Limited information
was also drawn from an environmental database search and Sanborn map review (Environmental Data
Resources 2002), and a search of files at the Connecticut Department of Environmental Protection
(CTDEP).
SITE LOCATION AND DESCRIPTION
The former Cos Cob Power Plant site is located at 22 Sound Shore Drive in the Cos Cob section of the
Town of Greenwich, Connecticut, on the west bank of the Mianus River just south of the 1-95 bridge and
east of the Cost Cob Metro North train station (Figure 1). The portion of the former plant property owned
by the town is approximately 9 acres and is bordered to the north by Metro North transformer yards, to
the west by an electrical substation and transformer yard owned by Connecticut Light and Power
Company, to the southwest by a condominium complex, and to the east and south by Cos Cob Harbor
(Mianus River). The property was deeded to the town by the CTDEP in 1987, with the understanding
that the property would ultimately be open to all residents of the state.
An aerial photograph of the site provided by the Town of Greenwich predates demolition of the power
plant and was used to prepare figures for this report and to identify locations of structures that were
formerly present at the site. Demolition of the power plant began in 1999 and was completed in 2000.
The power plant, metal frame building, water towers, and aboveground oil tank have all been removed.
-------
No structures on the property are owned by the town, but the town public works department uses the
property to store construction materials. A large pile of soil currently occupies the area of the former
powerhouse. The town property is otherwise unused. Access to the property is limited to town •
employees and is restricted by a fence along the northern and western boundaries.
SITE HISTORY AND USE
The Cos Cob Power Plant was built in 1907 by the New York, New Haven, and Hartford Railroad to
provide power for electrification of the railroad. The coal-fired plant operated in this capacity until the
1960s. The plant was a multi-level concrete and metal building that housed boilers, transformers (known
to contain polychlorinated biphenyls [PCBs]), and other electrical generation and distribution equipment.
Ash and slag from power generation were disposed of on the grounds surrounding the plant; a large
aboveground oil storage tank was also formerly located on the site. The type of oil stored in the tank has
not been identified, however. Several additions were made to the building over the years before its use
was discontinued in the 1960s. In 1986, the idled plant was decommissioned by the Connecticut
Department of Transportation, and the property was transferred to the Town of Greenwich the following
year. Plant equipment, such as transformers and boilers, were left intact. The plant, which had been
essentially vacant for 20 years, continued to deteriorate and was the target of vandalism. One case of
vandalism resulted in the release of transformer oils to the ground. The town reported the spill to the
CTDEP, and contracted with Matin Environmental (Marin) to remediate the release area and to conduct a
limited investigation of site soils in 1998 (Marin 1998a). The town also retained Marin Environmental to
complete an inventory of transformers and breakers within the plant and sample oil within the equipment
to determine its PCB content (Marin 1997). Further information regarding the work by Marin is
presented below under initial site investigation. Transformer oils were subsequently sampled again by
Osprey Environmental in 1998 to support a request for bids issued by the town for removal and disposal
of the oils (Osprey 1998). These latter samples were analyzed for parameters other than PCBs that were
needed to estimate costs for off-site disposal, namely Resource Conservation and Recovery Act metals,
British thermal unit content, percent water, and total halogens.
In 1999, the town and CTDEP became increasingly concerned about the potential for releases of asbestos-
containing materials (ACM) from the deteriorating main building and smaller metal buildings. An
inspection of the main building in 1999 by Osprey Environmental, on behalf of the town, found ACM
both inside and outside the building. The types of ACM identified ranged from nonfriable transite,
galbestos, and roofing products, to boiler gasket materials and thermal insulation. Osprey Environmental
-------
evaluated the potential for a release of ACM to the environment and concluded that releases had likely
occurred from several areas.
On behalf of the town, Osprey Environmental notified EPA, CTDEP, and the Connecticut Department of
Public Health (CTDPH) of the likely releases. Osprey recommended the removal of ACM, with the
concurrence of the CTDPH. The ACM was removed, and the main building and smaller metal buildings
were demolished during 1999 and 2000.
PROPERTY REUSE SCENARIO
Reuse scenarios are not yet well established for the site. The property currently has deed restrictions
resulting from the agreement when the town received the property from the state of Connecticut.
Restrictions include: the property must be maintained as open space and accessible to all state residents.
Although some additional re-use scenarios have been suggested, the town has been unable to consider re-
use scenarios that do not meet the deed restriction criteria.
Currently, the town intends to redevelop the property as a waterfront public park. Possible uses of the
park that the town is considering include, walking trails, playing fields, picnic areas, and a shoreside
boating facility. The town has formed a committee that has actively sought community input regarding
redevelopment of the property. Responses to a town-wide questionnaire distributed in 2001 indicate that
the majority of residents prefer passive recreational uses such as walking trails and picnic areas.
CONCEPTUAL SITE MODEL
A CSM is a description of everything that is known about a site that is relevant to the decisions to be
made in support of proposed reuse alternatives. At the Cos Cob site, existing environmental sampling
data, geologic information, and preliminary reuse plans were reviewed to develop a preliminary CSM
diagram (shown on Figure 2). A more detailed review of existing chemical and historical information is
presented in the draft field task work plan (M&E 2003a). A brief description of the results of previous
investigations and the materials compiled in development of the preliminary CSM for the site is presented
below.
-------
Initial Site Investigations
TRC Environmental Consultants (TRC) conducted an environmental assessment (EA) of the site for the
Town of Greenwich in 1988 (TRC 1988). Surface and subsurface soil samples, limited samples of
surface water and sediment, and groundwater samples were collected, as follows:
• Ten surface soil samples and 19 soil boring samples were collected from nine locations at 0 to 12
inches bgs. Split spoon samples were collected at 2-foot intervals to 6 feet bgs, and at every 5
feet or change in strata throughout the depth of the borehole. A minimum of two samples were
taken from each borehole and samples were analyzed for PAHs, extraction procedure (EP)
toxicity metals, and PCBs. Samples from three locations were also analyzed for volatile organic
• compounds (VOC) because of the proximity to potential fuel oil contamination.
• Six subsurface test borings and three monitoring well borings were drilled on the property. Test
borings were advanced until no visible sign of contamination or waste material was observed.
• Three groundwater samples were collected from the three monitoring wells installed during the
EA. Samples were analyzed for metals, common leachate indicator parameters, metals, PAHs,
and VOCs. The wells are no longer in existence at the property.
• Two surface water and four sediment samples were collected from two locations bordering the
site as well as from two on-site locations. Samples were analyzed for metals, PCBs, inorganic
carbon, PAHs, and VOCs.
Soil was analyzed for extractable metals using the EP toxicity procedure, an acid-leach procedure that has
been superseded by the toxicity characteristic leaching procedure (TCLP). The use of EP toxicity was
consistent with CTDEP guidelines at that time (1988). Four metals were detected in EP toxicity samples:
arsenic, barium, lead, and silver. Based on the guidelines of that time, results did not suggest a need to
remove or remediate site soils. Key observations, as noted in the TRC report, are quoted from the EA
report document as follows:
"...approximately 22 to 35 feet of fly ash fill material exists in the southern portion of
the site, and approximately 30 feet of coal/slag/ash fill material exists in the northeastern
portion of the site. At two other areas on the site, namely at test borings B-l and B-3
(Figures 3 and 4), signs of fuel oil soil contamination were observed, both visually and
olfactory (i.e., oily appearance and odor) and through elevated readings with a portable
Organic Vapor Analyzer (OVA)." (TRC 1988).
Only one soil sample contained detectable levels of VOCs. TRC did not compare detected concentrations
of PAHs with current Connecticut residential direct exposure criterion (DECs), because none were in
place when the investigation was performed in 1988. Tables in Appendix B of the draft field task work
plan (M&E 2003a) compare the TRC results for soil samples to the current DECs. Surface soil samples
-------
S: -3, SS-5, B-2, and B-3 (Figure 3) exhibited concentrations of PAHs that exceed the current DECs.
Concentrations of PAHs reported by TRC for subsurface soil samples did not exceed the currently used
DECs.
Of the three groundwater samples collected, the concentration of zinc reported in two of the samples
exceeded the surface water protection criterion of 123 parts per billion (ppb). This suggested the potential
need for at least some surface water control measures as part of any remedy. No VOCs or PAHs were
detected, and no other metals were detected at levels that exceeded the DECs.
The TRC report prompted later concerns with respect to direct exposure to surface soils (PAHs in some
samples exceeded the residential DECs later established by CTDEP). Results for metals were not
conclusive because data for total metals in soils were not obtained, and EP toxicity results are not directly
comparable to DEC. The widespread presence of ash noted by TRC indicated that soils were likely to
contain levels of metals that exceed DEC. In addition, TRC noted fuel oil odors in some locations (B-l
and B-3), suggesting the possibility of petroleum releases. At these two locations, PAHs were also
detected in the soils.
Marin Transformer Inventory, 1997. The town retained Marin in 1997 to conduct a transformer survey
and PCB evaluation at the site (Marin 1997). The purpose was to identify the number and volume of
electrical transformers and aboveground storage containers within the plant. Seventy-four transformers
were identified on the site; 54 disconnect switches/circuit breakers were located within the power plant
building, and 11 aboveground storage containers (tanks, drums, and others) were found within the power
plant building. A total of 91 oil samples were collected from selected transformers, switches, circuit
breakers, and storage containers, and then analyzed for PCBs. Aroclor 1260, a common commercial PCB
mixture, was detected in five of the 91 oil samples at concentrations ranging from 2.2 to 12.0 parts per
million (ppm). Most of these detections were located in the area called "former transformer/breaker area"
shown on Figures 3 and 4. Others were from transformers located on the main level of the power plant
building in the northwestern corner of the site, in an equipment maintenance area. The EA report
recommended that fluids within the equipment at the site be removed and disposed of off site because the
equipment did not have secondary containment and thus did not comply with CTDEP aboveground
storage regulations.
Marin Limited Soils Investigation, 1998. The town also retained Marin to conduct an investigation of
soil in selected site locations, as requested by CTDEP in a letter from Ms. Lori Saliby dated November
8
-------
25, 1997. The CTDEP had received an anonymous report of historical disposal of dielectric fluid and had
also been notified by the town of a spill from a transformer that had apparently been overturned by
vandals.
The main areas of concern were: (1) an area in the basement of the building where the transformer was
overturned and soil staining was evident (the transformer spill area, identified on Figures 3 and 4 as the
remediated basement oil-stained area), and (2) several outdoor areas alleged to have been subject to
historical disposal of fluids (the A-, B-, and D-series Marin borings on Figures 3 and 4). An area "C" was
also identified (north of the Cos Cob plant, around one of the cinder block buildings), but was eliminated
from the investigation because it was not on the town's property. Shallow soil samples were collected in
the A, B, and D areas (depth range of 6 inches to 1.5 feet below ground surface [bgs]) and analyzed for
PCBs by EPA Method 8080 and for TPH by EPA Method 418.1 (Marin 1998a). No PCBs were detected
in any of the soil samples collected. TPH was detected in samples collected at locations A-l, A-2, B-2,
B-3, and D-3 at levels exceeding the Connecticut DEC for residential exposure under the Remediation
Standard Regulations of 500 ppm (that is, 500 milligrams per kilogram [mg/kg]). Marin recommended
that these areas be further investigated as part of a comprehensive site assessment, before the appropriate
course of action could be selected.
Marin's investigation of the transformer spill area confirmed the presence of TPH in excess of the DEC in
shallow soils in the basement. Samples were collected from the surface of the basement dirt floor to a
depth of 6 inches. PCBs were detected in one of five soil samples collected, at a concentration of
1.9 mg/kg. PCBs were not detected in the other four soil samples (the reporting limit for these samples
was 1.0 mg/kg). The Marin report recommended that the transformer spill area be remediated by
excavating the contaminated soil. Marin completed remediation for the town in April 1998. A letter from
Marin to CTDEP dated July 6, 1998, documented the remediation and presented the results of
confirmation soil sampling. The depth of the excavation ranged from 12 to 18 inches, and approximately
15 cubic yards of soil was removed and disposed of off site (Marin 1998b).
Osprey, 1999: Asbestos was found at levels that constitute a "release." Types of ACM identified
included nonfriable transite to galsbestos. The ACM was removed, and the buildings were demolished in
1999 and 2000.
-------
Current Site Use
The Department of Public Works has used the town's portion of the property to store construction
materials since the plant was demolished in 2000. Multiple piles of segregated construction materials
(gravel, soil, and piping) are located within the site boundaries. There are no structures on the site. The
topography is relatively flat, except near the stone and concrete sea walls or where piles of clean soil are
present and the ground surface slopes steeply up and then back down to Cos Cob Harbor. Most of the site
is vegetated only by grass or consists of vegetation free tightly compacted fill, except for the area along
the southern sea wall and the slag ash area, where trees are present.
Site and Regional Geology
Cos Cob Harbor is located in Greenwich, Connecticut, along the southwestern region of the state. The
regional bedrock is primarily composed of Harrison Gneiss. The Harrison Gneiss is characterized by
interlayered dark and light gray layers of medium grain size crystalline material with a strong foliation.
The major geological units which make up the Long Island Sound include bedrock, buried coastal-plain
sediments, continental glacier moraines, glacial-lake deposits, and marine delta sequences. Crystalline
bedrock and deltaic deposits associated with glacial Lake Connecticut, influence sedimentation in the
northern and western regions of the sound. Data from boring logs indicate black silt with little sand and
trace gravel to a depth of 30 to 35 feet below ground surface at the site. A gray marine silt is present at
depths below 35 feet (Figure 2).
Historic aerial photographs and information gathered from conversations with town representatives
revealed that the southernmost portion of the site, south of the former powerhouse, is almost entirely
composed of fly ash from the former power plant (Figures 3 and 4). This area was slowly filled in over
the years with ash from the power plant. Previous site investigations document the presence of ash to a
depth of near 30 feet bgs in this area. As shown on Figure 3, two distinct areas of ash cover the site. The
area northeast of the former power house is composed of slag ash derived from coal ash that remains in
the boiler, while the debris to the south is composed almost entirely of fly ash that was collected from the
plant's exhaust stacks.
10
-------
Cos Cob Power Plant Site
MEDIA OF CONCERN MBHHM^BM^^MH^HHHI^HBMMHMH^MB^^H
Table 1 provides a pathway receptor diagram developed by the project team that shows the primary
sources and potential receptors for the Cos Cob site. This table shows that surface soil and solids present
a potential risk to human health and the environment and represent the primary media of interest at the
site. Surface water and sediment are not considered media of interest because remediation under the
reuse scenario would likely involve PCB hot spot removal, along with capping and drainage
improvements for remaining arsenic, PAH, and TPH contamination. Groundwater at the Cos Cob site is
considered brackish and nonpotable; therefore, groundwater is not designated as a medium of concern in
this evaluation. Contaminant concentrations in groundwater are expected to be relatively low and
constituents immobile. Therefore, impacts from groundwater beneath the site to the harbor are considered
unlikely.
CONTAMINANTS OF POTENTIAL CONCERN
The contaminants of potential concern (COPC) at the site were identified through historical information
and documentation from previous investigations as petroleum hydrocarbons and related PAHs, metals
(primarily low levels of arsenic related to fly ash), and PCBs in locations where transformers had been
stored. In addition, asbestos near former buildings was identified as a potential concern. Table 2
provides a summary of COPCs at the site and their related DEC.
Metals. Metals, particularly arsenic, are contaminants of concern because of their presence in coal ash
and the documented use of ash as fill on the site. The DEC for arsenic is low (10 mg/kg), and arsenic is
commonly present in soils affected by coal ash at concentrations that exceed this level. During the EA
conducted at the site, soil samples were collected for analysis of metals by the EP toxicity procedure, a
leaching test that has since been superseded by the toxicity characteristic leaching procedure (TCLP) and
synthetic precipitation leaching procedure (SPLP). The EP toxicity procedure (EPTOX) does not give a
measurement of the total concentrations of particular metals in soil; consequently, the available historical
results are not directly comparable to the Connecticut DECs and additional analyses would be required to
meet project objectives.
PAHs. PAHs are also of potential concern, due to their presence in coal ash, as well as their presence in
petroleum oils. Previous assessments (TRC 1988) revealed low levels of PAHs (but that exceeded DEC)
in some surface soil samples. It is not known whether the PAHs are from coal, coal ash, petroleum
releases, nor is it known whether their presence is widespread or localized.
11
-------
Petroleum Hydrocarbons. Marin surface soil samples at locations A-l, A-2, B-2, B-3, and D-3 (Figure
3) exhibited concentrations of TPH at levels greater than DEC (Marin 1998a). These samples were
collected in areas suspected of being affected by a release. These areas were investigated by M&E on
December 9, 2002, during a field event designed to collect site-specific soil samples for a "demonstration
of methods applicability" study designed to constrain the use of a siteLab® test kit. No odor or visual
evidence of petroleum hydrocarbons was noted in the samples. It is considered likely that PAHs
represent the principle petroleum-related COPCs present in site soil. The DEC for carcinogenic PAHs
(cPAH) are much lower than that for TPH.
PCBs. The Marin letter report documents remediation of a PCB spill in the basement of the former
powerhouse (Marin 1998b). Although TRC's soil samples collected during the EA for PCB analysis did
not exhibit detectable levels of PCBs (TRC 1988), the EA was limited in extent and PCBs are suspected
to be present over at least the northern two-thirds of the site.
Asbestos. The likelihood that surface soil is affected by ACM is considered moderate, based on the
evaluation conducted by Osprey Environmental in 1998. Releases of ACM to the environment are likely
to have occurred before the ACM was abated and the buildings were demolished in 2000 (Osprey 1998).
Areas and contaminants of potential concern are summarized in the following list.
Site Area
Media of Potential Concern
Sitewide surface and
subsurface soils
Soils throughout the site may be affected by contaminants associated with on-site •
disposal of coal ash (both fly ash and slag or bottom ash) from the power plant.
Contaminants commonly found in areas where ash is used as fill include metals
(particularly arsenic) and PAHs. Ash extends to depths of up to 30 feet across
portions of the site.
Asbestos containing material (ACM) may also be mixed with site soils as a result of
possible releases from the deteriorated power plant buildings before they were
demolished in 2000.
Previous investigations detected petroleum hydrocarbons in several locations
(A-, B-, and D-series borings), and recommended further investigation of these
areas. Anecdotal evidence suggests the potential for PCB contamination as well,
although PCBs were detected in only one location previously, and that location has
since been remediated.
Surface and subsurface
soils in areas where
contaminants were
previously encountered
(Marin investigations)
12
-------
Site Area
Media of Potential Concern
Groundwater
Groundwater was sampled by TRC in 1988, and no contamination was encountered.
Groundwater is not considered a concern based on the TRC results. However,
because the TRC samples were collected more than 10 years ago, before the power
plant was demolished, it may be worthwhile to conduct a limited groundwater
sampling event at some future time, particularly if construction that will require
dewatering is contemplated.
EXPOSURE ROUTES AND RECEPTORS
A complete exposure pathway consists of four fundamental components: (1) a source and mechanism of
chemical release, (2) an affected environmental medium and a probable chemical migration process, (3)
an exposure point, and (4) an exposure route by which humans come into direct contact with the
chemical. If any of these components is missing, then the exposure pathway is incomplete and no
exposure can occur.
Results of the previous investigations indicate that human receptors are the risk drivers for the former Cos
Cob Power Plant. Potential exposure pathways for the site were selected based on current land use and
most probable future activities at the site, as well as an evaluation of potential transport or uptake
pathways.
Future adult and child recreational users may be exposed to COPCs in surface soil. Exposure to the
recreational user is limited to surface soil because planned recreational use of the area is not likely to
require disturbance of deeper (subsurface) soils. Potential surface soil exposure pathways for the future
recreational user are incidental ingestion of soil, dermal contact with soil, and inhalation of particulates
released from soil (Table 1). The surface water exposure pathway is considered incomplete for this study
because no surface water bodies or direct contact between ash and surface water are anticipated; this
pathway may need to be addressed further as site reuse is planned and implemented, however.
13
-------
Cos Cob Power Plant Site
WORK PLAN DEVELOPMENT ^••^^^^^•••••^••i^^^
The BTSC became involved with this project in the fall of 2002, when the EPA project manager and the
EPA Region 1 Technology Innovation Office (TIO) representative identified the site as a good candidate
for application of the Triad approach. The EPA Region 1 project manager arranged for assistance to be
provided by the BTSC and further directed M&E to develop a draft work plan for the TEA at the Cos '
Cob site. The draft work plan was developed to provide a summary of existing site information and
present an example for a "traditional" approach to site assessment under a TEA. Subsequent to a review
of that work plan and other site information, a project meeting was held at EPA Region 1 offices in
November 2002. As a result of that meeting, the work plan was revised to incorporate the use of field-
based analytical technologies, statistical sampling, and accelerated decision-making consistent with the
principles of the Triad approach. The initial work plan was never completed or formally submitted for
EPA review because it was decided early on to abandon it and move toward a plan based on the Triad
approach.
The objective for the BTSC was to work with EPA Region 1 and M&E to develop a revised work plan
and approach for the Cos Cob site that incorporated elements of the Triad approach. The Triad approach,
a framework for efficiently managing decision uncertainty, can be applied to reach project objectives
faster and in as few mobilizations as possible. The Triad approach is well suited to Brownfields projects
such as Cos Cob where budget and schedule are crucial to successful project completion. Further
information about the Triad approach is available on line at the Triad Resource Center website at
http://www.triadcentral.org, and in "Using the Triad Approach to Streamline Brownfields Site
Investigation and Cleanup" (EPA 542-B-03-002; June 2003).
INITIAL WORK PLAN
This section describes the activities originally identified under the TEA using a traditional phased
approach and fixed laboratory analyses. The original technical approach for this site was based on EPA's
statement of work, and clarifications and modifications to the scope were discussed at two scoping
meetings held between EPA and M&E (November 9, 1998, and March 26, 1999).
Site-Specific Objectives for Initial Work Plan
The original objective for this site was to conduct a site investigation to assess the nature and extent of
soil contamination, sufficient to allow development of a conceptual remedial plan and planning-level cost
14
-------
estimate for site remediation, consistent with the town's plans for site reuse as a recreational area. In
general, the original proposed site investigation would have focused on known or suspected areas where
contaminants of concern were anticipated to have been released. Field observations and analytical data
were to be collected to provide information on surface and subsurface conditions and to support
development of further assessment procedures and potential cleanup alternatives, if necessary. The goal
was to identify the presence or absence of potential hot spots; delineation was to be left to later phases'of
the project.
Development of the Original Sampling Approach
The original sampling plan called for collection of authoritative, judgmental samples to confirm the
presence or absence of contamination at locations where previous reports had implied potential source
areas. Contamination was primarily suspected to be related to two primary source types: (1) surface
spills of petroleum and PCBs that were expected to decrease with depth and, (2) coal fly ash containing
low levels of metals and spent petroleum hydrocarbons to a depth of nearly 30 feet bgs on some portions
of the site. The primary objectives of the original TEA for the former Cos Cob Power Plant property
were:
• Visually observe surface and subsurface soil conditions (particularly noting the presence of
ash).
• Use "field-based" methods (headspace and x-ray fluorescence [XRF]) to help identify
possible "hot spots" of petroleum or heavy metals contamination, and to select samples for.
laboratory analyses.
• Characterize site surface soil via off-site laboratory analyses for the identified COPCs.
• Compare analytical results for soil with DEC in a screening-level risk assessment.
• Develop additional assessment procedures and potential cleanup options, if warranted.
Original Sampling Rationale
The original sampling rationale called for collection of authoritative, judgmental soil samples at 20
locations (Figure 3). The rationale for each sample location and the off-site laboratory analytical suite
proposed for each location are presented in Table 3. The original sampling rationale was based on off-site
laboratory analysis only.
15
-------
SYSTEMATIC PLANNING
Comprehensive, up-front planning is essential to effectively complete any environmental project. Proper
planning will promote collection of data that will lead to defensible decisions. The residential DECs were
used as the cleanup goals and in selection of the appropriate analytical methods. By understanding the
questions that need to be answered project managers and team personnel can use systematic planning as a
tool to develop a roadmap to success.
In order to streamline the site characterization activities, project goals were examined and potential design
modifications identified before entering the field. Project objectives identified by the core technical team
for the site are provided in the box below.
Overall Project Objectives
1) Estimate the nature and extent of contaminants of potential concern (COPCs) at the site above the
state of Connecticut residential direct exposure criterion (Table 2).
2) Accomplish the above mentioned objective in a single mobilization to facilitate reuse planning
activities, while staying within budgetary constraints.
• After a review of the project objectives, the project team provided input on potential design modifications
to the originally proposed work plan regarding the following site activities:
• Redesign of the sampling and analysis program to increase the sampling coverage and
incorporate the use of appropriate field-based analytical technologies.
• Design and implementation of a "demonstration of methods applicability" study to optimize
method performance and develop preliminary field-based decision criteria.
• Focus the collection of QC samples where results would yield the highest value relative to
limiting decision errors.
• Implementation of the fie Id program.
• Data processing and refinement of decision criterion.
The BTSC project team participated during the project startup to assure integration of activities elemental
to the Triad approach. Activities discussed included evaluation of field method performance and
completeness of documentation. The BTSC team suggested that audits of mobile and fixed-labs during
startup be conducted by an outside party not directly responsible for data collection. A third party review
such as that conducted by the BTSC for this project was found to be helpful. Alternatively, for other
projects, it would be helpful to have a project chemist on site during startup/implementation.
16
-------
Detailed decision rules and focused quality control requirements were developed in cooperation with the
BTSC during the systematic planning process. The core technical team worked in conjunction with the
vendors of field-based analytical tools and other stakeholders to develop the final approach used at the
site. The following sections describe the results of the core technical team's efforts to revise the proposed
sampling and analysis approach for the site.
The New Sampling Approach
A random or non-nodal systematic grid-sampling program was developed to replace the original
authoritative sampling plan. Some aspects of the original plan were preserved such as sampling locations
selected based on historical information. Preserving aspects of the original sampling program and
increasing data density across the site using a random systematic grid, provided a larger number of
samples improving site coverage. Field-based sampling technologies were proposed to allow for this
increased sample density at no additional cost. By using field-based methods, a larger number of samples
could be collected and analyzed in a single mobilization.
The grid sampling approach used was developed based on an exposure area of 5,000 square feet. In the
project teams opinion this size was thought to be consistent with the typical size of an average residential
lot. Based on the size of the exposure area (5,000 square feet) and available project resources, a total of
72 grids were sampled during the project. The plan called for further delineation where PCB
contamination was found within any one of the grid sectors. At these grids, a "step out" sampling
approach was planned where the original location would be surrounded by 4 additional direct push
sampling locations 10 feet to the north, south, east, and west. Initial sampling points within each grid
were identified using a random sample locator program. Non-nodal grid sampling schemes, also known
as random systematic sampling, combines the benefits of both random (a sampling scheme best employed
when historical knowledge of contaminant distribution is limited) and systematic (a sampling scheme
used to locate and delineate "hot spots") sampling methods.
A track-mounted Geoprobe was proposed as the sampling platform to drive sample cores from 0 to 4 feet
bgs at each sampling location, and field analytical options were evaluated by BTSC staff and EPA Region
1 Mobile Laboratory personnel. The State of Connecticut considers the upper 4 feet of soil to be surficial
and where DEC action levels are applicable to the site.
17
-------
Analytical Options
The Triad approach encourages project teams to make maximum use of real-time measurements to
support decision-making on site. The use of field-based analytical technologies allows for the
investigation to focus on areas of elevated contamination detected in real time and helps to direct
additional field analyses and fixed lab comparison analyses in a single mobilization.
Connecticut DECs shown in Table 2 were used when selecting suitable field-based analytical methods for
the project. For example, the low DEC for arsenic of 10 milligrams per kilogram (mg/kg) was considered
when the project team decided that field-based XRF methods might not be sufficiently sensitive.
Typically XRF detection limits range from 10 to 100 mg/kg for most metals. It is generally considered •
acceptable when reporting limits or practical quantitation limits for analytical methods are at least 2 to 3
times lower than potentially applicable regulatory threshold limit values. This assures the utility of
results for decision-making purposes. Arsenic is known to be present in the fly ash, at relatively low
concentrations near the regulatory threshold of 10 mg/kg making it a potential risk driver. As a result, the
team agreed that off-site analysis for arsenic would be the preferred option using an ICAP atomic
emission spectroscopy method that has reporting limit of 1 mg/kg or less. The EPA Region 1 Mobile
Laboratory planned to take a portable XRF unit into the field to confirm the assumption that arsenic was
the primary metal of potential concern at the site and to rule out the presence of other metals that have
regulatory threshold limit values well within the working range of sensitivity for field portable XRF.
In addition to XRF for metals, options for analyzing TPH and PAHs in the field included various
immunoassay kits, ultraviolet fluorescence (UVF), gravimetric, and turbidometric methods. A single-
column gas chromatograph equipped with an electron capture detector (GC/ECD) was identified for use
by the EPA Region 1 Mobile Laboratory to evaluate the presence of PCBs. The EPA Region 1
Laboratory later identified the siteLAB® test kits, a UVF system, as the preferred method for TPH and
PAH analyses based on costs, ease of application, and successes at other sites with petroleum product
contamination.
The project team concluded that a demonstration of methods applicability (DMA) was needed for the
UVF method to assure the reliability of the method for application at the site. The DMA would also be
used to establish preliminary field-based action levels. The results of the DMA study are further
discussed in the Section titled Demonstration of Methods Applicability. A detailed description of the
methods as they were applied for the investigation is presented in the Section titled Sample Analysis.
18
-------
Developing Decision Logic
When decision logic is developed for sites where actions will be based primarily on the use of field-based
measurement technologies, it is often necessary to consider many factors, such as:
• Field observations or other data may suggest that there is the potential for similar, yet different,
analytes to yield similar responses when a test kit or other screening analytical method is used.
This issue can be identified and sometimes resolved through application of the DMA.
• If a significant bias is expected in the field analytical results, one strategy is to collect sufficient
comparison data (for example, splitting well-homogenized samples for analysis by both the field
and traditional methods) during a DMA, in early sampling events, or both. If a predictive
relationship can be identified between the field measurements on an analyte-specific basis, this
relationship can be used to guide decision making using thelield based methods.
When data distribution characteristics are normal or lognormal, a predictive relationship can be
established and field-based action levels developed (Enclosure 2). Statistical plots like those provided in
Enclosure 2 should be used initially to establish that the use of a particular field-based technology is
viable. The reliability of the preliminary field-based action levels developed during a DMA need to be
refined as more data is collected. Field-based action levels are initially selected based on a comparison
between more selective fixed lab results and those generated using a selective field-based method. Safety
factors are usually applied to assure the reliability of site decisions.
As work progresses, a sufficiently large comparison data set usually becomes available allowing action
levels to be revised and improved. '
Depending on the nature of project decisions, two or three decision intervals or safety factors are
commonly identified based on the results of the DMA. The most common breakdown is into three
intervals, as shown in the diagram on the following page: (1) an interval where it is judged that the field
data can be trusted to confidently declare areas "clean" (where no further action is needed), shown on left
side of the line; (2) an interval where field results can be trusted to confidently declare an area "dirty"
(where remedial action is needed), shown on the right side of the line); and (3) an interval where the field
results are considered ambiguous, and a confident decision of "clean" or "dirty" would require more data
to manage the decision uncertainty, shown in yellow in the central area of the line. Reasons for this
uncertainty may stem from sampling variability or from analytical uncertainty (imprecision or bias in the
field method), or both. When only two intervals are desired or, a single limit is proposed: data values
less than this value allow the area to be declared "clean," and data values greater than the limit are
accepted as indicating that the area is "dirty". In this case, a safety factor is applied. For example, in the
19
-------
figure below, the field-based action level with a safety factor built in might be set at 40 ppm when
deciding that the site is clean. Below this level sample results would be considered clean while results
above this level would be considered dirty.
Action Level
(AL) (ppm)
Confident Decision that
40 50 65
i
Confident Decision that
True Cone > AL
^ True Cone
-------
• Estimating the cost of a "false inaction" decision error (that is, incorrectly declaring an
area "clean" so that no further action is needed) must consider the ramifications to human and
ecological health of potential exposure to excessive contamination, as well as the social and
political costs that will be incurred when the error is discovered or suspected. It is often more
important to protect public and environmental health from potentially harmful health effects and
to error on the side of caution. On the other hand, this approach can be costly. Nonetheless,
since it can be prohibitively expensive in some scenarios to gather all the information needed to
ensure that decisions are entirely correct, it is possible to structure the decision-making process so
that substantial costs can be saved by judiciously deciding when relatively small errors on the '
side of caution can be accommodated. These decisions can be thought of as a kind of "safety
factor" that supports using field measurements and other types of nontraditional tools to achieve
significant cost savings while decisions remain protective of human health and the environment.
Managing decision uncertainty that stems from sampling variability can require collection of grab or
composite samples to obtain a more confident estimate of the mean concentration for the decision unit or
a more confident estimate of the boundaries of contamination. Managing decision uncertainty that stems
from analytical uncertainty requires first that sampling variability has been managed (so the
representativeness of samples is known). Then, samples that represent critical decision points are selected
for processing by more rigorous analytical methods to produce analyte-specific data, or data free of
excessive analytical bias or imprecision.
Usually, a DMA study as defined by EPA's Office of Solid Waste Methods Team, which manages the
SW-846 methods manual (see Enclosure 1) is designed and implemented initially to begin the process of
evaluating potential sampling and analytical method issues, as well as the comparability of the various
sampling and analytical methods under consideration. The results of the study are used to compute
appropriate safety factors and preliminary uncertainty limits for decision-making that should be applied at
a site. Differing safety factors may need to be developed for a specific monitoring and measurement
technology and type of decision. Uncertainty limits to support decision-making are used to establish
concentrations where stakeholders feel comfortable that a correct decision is being made.
A DMA is usually designed to evaluate the ability of a method to meet project-specific data needs (that is,
the specific contaminants and media of concern at a given site). The study considers the precision,
sensitivity, and bias of the field-based instrument technology so that an adequate safety factor can be built
into the overall decision uncertainty limits. Internal method quality control results, along with
investigative, replicate, and spiked samples analyzed in the field as well as off-site methods, are generally
used collaboratively to estimate the total uncertainty associated with a measurement such that realistic
safety factors can be developed. Uncertainty management measures or focused QC to support refinement
of decision criteria at the Cos Cob site included the following:
• Focusing collection of comparative analyses on a concentration range where fixed lab results
would be of the greatest benefit for managing decision uncertainty (i.e., in and around the field-
based action levels)
21
-------
• Comparative analysis for PAHs using off-site GC/mass spectrometry (MS) and selective-ion
monitoring (SIM)
• Comparative analysis for PCBs using off-site dual-column GC/ECD whenever positive field
screening results indicated the potential presence of PCBs
The number of data points is typically limited when a DMA is performed, making rigorous statistical
analysis less powerful. Judgment is therefore used to evaluate the comparison data used to construct
preliminary safety factors so that intolerable decision errors are avoided. As more is learned about a site
and the contaminant distributions more clearly defined, the preliminary safety factors can usually be
reduced as the standard deviation for results becomes smaller.
DEMONSTRATION OF METHODS APPLICABILITY
The project team and the BTSC discussed potential options for implementing the field program for the
Cos Cob site and developing decision criteria for field-based analytical methods. Decision logic diagrams
were developed for arsenic (Figure 6) as well as for TPH, PAH, and PCBs (Figure 7) and circulated to the
team for further refinement. A conservative initial action level of 100 ppm for TPH was identified based
on a comparison between field test kit and fixed laboratory results concerning when TPH concentrations
might exceed the DEC of 500 ppm. Another field-based action level of 50 ppm was established for PAHs
using the test kit-based on the professional judgment of the project chemist, given that the test kits report
total PAH and the lowest DEC for an individual cPAH compound is 1 ppm. Test kit concentrations of
TPH between 100 and 500 ppm, and of PAHs between 50 ppm and 500 ppm, were identified as regions
of analytical uncertainty had the highest potential for impacting decision-making. In these concentration
ranges, the project team decided to collect additional samples for off-site comparative analyses to refine
the decision criteria. Samples with concentrations below or above these ranges were considered "clean"
or "dirty", respectively, and only limited comparative analyses were deemed necessary.
EPA Region 1 Mobile Laboratory personnel indicated that the laboratory had the capacity to analyze all
samples in the 0- to 1- and the 1- to 2-foot intervals for PCBs. Consequently, the EPA Region 1 Mobile
Laboratory analyzed all grid samples for PCBs and sent samples for comparative analyses to an off-site
laboratory when results indicated the presence of PCBs. This approach alleviated the need for field-based
decision criteria or a DMA for PCBs.
As noted previously, off-site analysis of arsenic samples was selected to provide the best reporting limits
and data most appropriate for making project decisions. Therefore, no field-based decision criteria were
established for metals, and a DMA was not required. The off-site analyses were conducted following
22
-------
EPA SW-846 Methods 3050B/6010B. Nonetheless, the EPA Region 1 Mobile Laboratory still used
XRF to screen soil samples and evaluate the potential for metals other than arsenic to be present at
concentrations that might exceed the potentially applicable DECs. It was thought that the additional data
provided by the on-site XRF would be useful in evaluating other metals and identifying sample locations
with higher arsenic values. The XRF results were also used to identify samples for a full suite of analyses
for metals to confirm the assumption that arsenic was the primary inorganic constituent of potential
concern.
With PCBs and metals excluded, the DMA study was designed to assess the applicability of the siteLAB®
UVF test kits and to refine the TPH and PAH decision criteria for use during the field investigation.
M&E'developed a technical memorandum that summarized the proposed approach detailing sampling
locations and procedures to support the study (M&E 2003b). The field-based analytical technology
vendor (siteLAB®) participated in a day of analytical training for the samples collected during the study.
Samples used to conduct the DMA were collected in a single day from areas where contamination was
suspected to be present. The objective of the training session was not only to train the project team in the
, use of the kit, but also to analyze the samples collected from the site and submit them for analysis at the
off-site laboratory to establish a correlation between the field-based and fixed laboratory methods.
Initial Field Sampling Event Conducted in Support of the Demonstration of Methods
Applicability Study
Soil borings were advanced using a Geoprobe 5410 drilling rig on December 9, 2002. Soil samples were
collected continuously at each location using direct-push techniques. The soil borings were logged and
described, including any evidence of impact by oil and hazardous materials. Soil boring locations are
shown on Figure 5.
Seventeen soil borings (T-l through T-17) were completed from 0 to 4 feet bgs. Borings T-l through T-
11 were advanced in the fly ash area located in the southern portion of the site (Figure 5), where samples
(A-l and A-2) collected by Marin in 1998 indicated elevated TPH contamination. Borings T-l2 through
T-17 were advanced in the slag ash area located in the northeastern portion of the site (Figure 5), where a
sample (D-3) collected by Marin in 1998 indicated high TPH contamination. All samples except T-17
were divided into a 0- to 2-feet bgs interval and a 2- to 4-feet bgs interval. Each sample interval was
homogenized and placed into a sealed labeled Ziploc bag.
While advancing boring T-17, M&E personnel observed a distinct layer of ash at the 2- to 3-foot bgs
interval. Sampling personnel homogenized the 2 to 3 feet bgs interval as one sample and the 0 to 2 feet
23
-------
bgs and 3 to 4 feet bgs intervals as the second sample. Samples for analysis by the siteLAB® test kits
were selected based on the presence of visible staining. Coal and ash was present in the majority of
samples. Field personnel attempted to select samples that covered a range of contamination in the coal
and ash. Selecting samples across a range in concentration can provide a more robust data set for,
supporting preliminary correlations. The 2 to 3 feet bgs sample from boring T-17 was also selected since
it was composed primarily of ash. In addition, two samples that did not contain any coal or ash were
selected. A total of eight samples were selected for analysis by the siteLAB® test kits.
Test Kit Training Session and Sample Analysis
On December 12, 2002, a half-day training session was held at EPA's Office of Environmental
Measurement and Evaluation (OEME) laboratory in Chelmsford, Massachusetts. In attendance were the
EPA project manager, the M&E project manager, two M&E staff members scheduled to participate in the
main February field event to be conducted at the Cos Cob site, the OEME quality assurance chemist, the
OEME mobile laboratory chemist, and several others interested in learning about the kits. The training
was conducted by Mr. Stephen Greason, technical support manager, siteLAB® Corporation. Mr. Greason
provided an overview of the siteLAB® UVF methods, how they work, and hands-on instruction to the
M&E representatives. After the training, the eight samples brought back from the site were analyzed
using the UVF test kit. Mr. Greason summarized the results and transmitted them via e-mail to session
participants.
Laboratory Results
The eight samples selected for test kit analysis were also submitted to fixed laboratories for extractable
total petroleum hydrocarbons (ETPH) analysis using the CTDEP approved version of SW-846 method
8015, and for semi-volatile organic compounds (SVOC) analysis using the EPA Contract Laboratory
Program - Routine Analytical Services (CLP-RAS). Woods Hole Group conducted the ETPH analyses
and Ceimic conducted the CLP-RAS analyses. The results for ETPH and the target analyte PAHs are
presented in Table 4, along with the test kit results.
CLP-RAS Results. For the eight samples submitted, the laboratory diluted the extracts prior to analysis
based on sample pre-screening that indicated high concentrations of hydrocarbons that exceeded the CLP-
RAS method calibration range. For a number of the samples (T-l 0-2', T-3 0-2', T-4 0-2', and T-5 0-2'),
no PAHs were reported as detected, but reporting limits exceeded the DEC regulatory threshold limit
values because of over dilutions of the samples by the fixed laboratory. The detection limits were lower
24
-------
for the remaining samples (T-5 2-4', T-13 0-2', T-13 2-4', and T-17 2-3'), but they still exceeded the
DECs. After discussion among project team members, samples T-1 0-2', T-3 0-2', T-4 0-2', and T-5 0-2'
were reanalyzed using EPA SW-846 method 8270C operated in the selected ion monitoring (SIM) mode
to obtain lower reporting limits. The Connecticut residential DECs were not exceeded in the reanalyzed
samples where PAH concentrations were reported.
Results of the methods applicability study indicated at least two differing PAH data populations. Field-
based results for PAHs in one population, represented by samples T-1 and T-3, were between 700 and
850 ppm (Table 4). Individual PAH concentrations in these samples were at the DECs for
benzo(b)fluoranthene and benzo(a)pyrene (1.0 ppm), according to the EPA Region 1 laboratory analysis.
Field-based results for the second distinct population were near J.O ppm with some PAH concentrations
near the DEC in samples T-13 and T-17, according to the EPA Region 1 laboratory results. The
benzo(b)fluoranthene concentrations from the laboratory analyses in these samples ranged as high as 0.5
ppm. Results for the two samples at T-13 and the sample at T-17 were consistent with each other (55
ppm, 53 ppm, and 52 ppm). Taken as a whole, the results indicated that, for test kit total PAH
concentrations less than or equal to 50 ppm, it was unlikely that any individual PAH compound would
exceed its DEC. Therefore the preliminary decision criteria seemed to hold up under the initial empirical
test.
ETPH Results. For the eight samples analyzed, the test kit results were equivalent to or higher than the
ETPH results from the fixed laboratory, except for sample T-5 (0- to 2-foot depth). As observed for the
PAHs, two distinctly different sample types were identified. The correlation between the results for the
test kits and the laboratory were close for the five samples where the laboratory identified the petroleum
as "similar to fuel oil #6; absence of straight chain aliphatics indicated the product present was
weathered" (Table 4). Examples of chromatograms from these samples (T-4 0-2', T-13 0-2', and T-17 2-
3') are presented in Enclosure 3. It is considered probable that, for these samples, ETPH is dominated by
PAHs, and the PAHs are likely from coal and coal ash.
For three other samples (T-1, T-3, T-5 areas), the laboratory identified the petroleum product evidenced in
the sample chromatograms as "similar to high molecular weight components in the lube-oil range." The
chromatograms from these samples are presented in Enclosure 3. The test kits exhibited a high bias for
two of these samples (T-1 and T-3), reporting field-based concentrations of 3,300 to 4,000 ppm versus
laboratory results of 1,500 and 1,400 ppm. The reported laboratory concentration for Sample T-5, taken
from 0 to 2 feet bgs, was 760 ppm TPH, while the field-based method result was only 138 ppm (Table 4).
These three samples were located in an area of the site where previous investigations had reported
25
-------
elevated TPH concentrations. It is considered possible that this area of the site may be contaminated by
multiple petroleum releases as well as by coal ash suggesting to the project technical team that correlation
of cPAHs to the total PAH results of the test kits could be problematic. When numerous sources of
petroleum products are present various ratios and relative concentrations for cPAH should be anticipated.
Refining the Preliminary Decision Criteria
The DMA confirmed that samples with a total PAH concentration less than 50 ppm as reported from the
siteLAB® test kits would not be likely to exceed the individual DECs for any cPAH compounds. Based
on the trial results, samples with test kit total PAH concentrations on the order of 800 ppm began to have
individual cPAHs that exceeded DECs (Table 4). However, in light of the complexity of the PAH data
observed during the DMA, the project team incorporated a safety factor where additional confirmation
data would be collected in an attempt to continue to refine decision criteria while the major field effort
was under way. The safety factor effectively increased the upper limit of the uncertainty range to 500
ppm, so that the range of between 50 ppm to 500 ppm total PAH was identified as being of greatest
interest for submission of comparative fixed laboratory analysis during the main field investigation at the
site. Although the majority of comparative samples would be collected in this range of concentration,
several samples would also be collected for concentrations greater than 500 ppm and less than 50 ppm for
completeness.
For TPH, where the field test kit and fixed laboratory results had better correlations, the anticipated
decision criterion of 500 ppm was confirmed; it appeared that test kit values greater than 500 ppm TPH
would begin to exceed the DEC. Additionally, the DMA results indicated that, for test kit results between
100 and 500 ppm, comparative analyses should be performed during the investigation to refine the
correlation between field-based and fixed laboratory results. Although the test kit result was close to or
higher than the laboratory result for all but one sample (T-5, 0-2 feet bgs), this result seemed to suggest
that a false decision could be made based on test kit results below 500 ppm, (A subsequent data review
after the investigation suggested that the result for sample T-5, 0-2 from the fixed laboratory could have
been biased high by contaminant carryover from a previously run sample).
To summarize, the DMA study helped the project team define a window of decision uncertainty between
50 to 500 ppm for PAH and 100 to 500 ppm for TPH for the siteLAB® test kits. Comparative analyses
would therefore be submitted at greater frequencies in these concentration ranges during the main field
investigation. The study also found that there were two separate populations of results for both PAH and
TPH, indicating two basic different types of contaminated soil media at the site. The sample population
26
-------
containing the highest total PAH concentrations (more than 700 ppm) also contained the higher TPH
concentrations that were classified in the "lube oil" range (sample locations T-l and T-3). The other
population contained total PAH concentrations near 50 ppm and TPH concentrations less than 500 ppm
that were classified as hydrocarbons in the weathered "fuel oil #6" range.
Reporting limit study documentation provided by siteLAB®, and the successful analysis of samples within
the kit's default calibration range (0.05 to 1.5 ppm for PAH, 0.1 to 5 ppm for TPH) in the DMA, indicated
that the reporting limits provided by the field test kits were easily sufficient to meet project decision
objectives.
27
-------
Cos Cob Power Plant Site
FIELD INVESTIGATION MH^^M^^H^MMM^HMBHH^^M^MS
The main field sampling program of the TEA, which included on-site analysis using test kits and a mobile
laboratory, as well as collection of samples for off-site laboratory analysis, was conducted from February
3, 2003, through February 7, 2003. Personnel from M&E, the EPA Region 1 Mobile Laboratory, and the
BTSC were on site to conduct or observe field activities and to provide input into decisions about when to
send samples for off-site comparative testing.
Several of the field activities included a dynamic component that allowed data to be analyzed and
interpreted in the field to guide the investigation in real time. These types of field activities are an
integral part of the Triad approach. Before the team mobilized for site activities, a grid spacing network
was developed that would allow for a reasonably dense sampling design over the site given logistical
considerations such as layout and area. The result was an overlay that segregated the site into 72 grids
that were approximately 70 feet by 70 feet. A random location generator was then used to select
sampling locations within each grid (Figure 5) except were historical data and or visual observations
suggested an alternative location might be more representative of the presence of contamination.
SAMPLE COLLECTION
Samples collected during the field activities were obtained using a small track-mounted Geoprobe to drive
sample cores from 0 to 4 feet bgs. The Geoprobe was moved to the appropriate grid and the approximate
sampling location was identified by the location generator as necessary. Some sampling locations were
adjusted to account for steep slopes, debris, vegetation, or soil piles within each grid. Within some grids,
soil piles of clean soil prevented the collection of a representative sample. Sampling locations were
marked using stakes and were mapped on the last day of the field effort using global positioning system
(GPS) equipment.
When samples were collected, the cores were visually inspected, logged, and segregated into 1-foot
homogenized sample aliquots. Samples were then analyzed for TPH and PAHs using field test kits. The
on-site laboratory also conducted field analyses for PCBs and screened samples for metals using XRF.
Various PCB "hot spots" were delineated horizontally as they were identified by surrounding the point
location with four "step out" samples at a distance of 10 feet. Particular attention was paid to locations
where PCBs or high concentrations of PAHs were detected, and these "hot spots" were delineated as time
and resources allowed.
28
-------
Some sample locations were also sampled for ACM, but those results are not further discussed in this
case study.
SAMPLE ANALYSIS
With the exception of arsenic analyses which were analyzed at the off-site laboratory by EPA SW-846
method 3050B/6010B, the soil samples were analyzed in real time (that is, as the samples were collected)
in the field and samples were sent off site for fixed laboratory comparative analysis in accordance with
the focused quality control plan developed during the DMA. The use of field analyses allowed the project
team to collect more data points within the time and budget allotted under the TEA. Additional sample
collection in areas where PCB "hot spots" were found could also be delineated. Field analytical
techniques for the UVF test kits (TPH and PAH), GC/ECD (PCBs), and XRF (metals) are discussed in
the following sections.
siteLAB* UVF Test Kits
The siteLAB® test kits used a simple extraction technique where 5 grams of the homogenized sample
were placed into a small extraction jars and 10 milliliters (mL) of methanol added. The cap of the
extraction jar was tightened firmly and the samples were vigorously shaken for 2 minutes and allowed to
settle for an additional 2 minutes. The extract was forced through a 0.45-micron filter to remove solids
and a small portion (100 microliters [jiL]) of the extract was diluted with methanol to 100X. The 100
times dilution factor is recommended by the manufacturer to avoid saturating the UVF detector, an effect
known as "quenching."
After the sample was prepared, the diluted extract was placed in a cuvette and analyzed using the UVF-
3100 analyzer. The UVF-3100 uses a powerful lamp from which light is passed through an excitation
filter, producing specific wavelengths that are absorbed by aromatic hydrocarbons such as the PAHs
present in fuels and other petroleum mixtures. The wavelengths pass through the cuvette that contains the
sample and excite the PAH molecules, producing fluorescence in the sample. The resulting fluorescence
wavelengths are passed through an emission filter specific to PAHs, allowing only those PAH-specific
wavelengths to pass through to the photomultiplier/detector.
Concentrations of total PAHs were recorded on the UVF-3100, and the results were transferred to a
laptop spreadsheet. Using the raw fluorescence emitted from the sample, the UVF-3100 software also
29
-------
generated a value for extended diesel-range organics (EDRO) for transfer to the spreadsheet. The values
were corrected for dilution factors and reported in ppm.
A PAH-specific calibration solution was used to create five standards used during the calibration process
to analyze samples for total PAHs using the UVF-3100. The five standards used during the calibration
process contained concentrations of total PAHs at 0.05, 0.1, 0.5, 1.0, and 1.5 ppm. The resulting
fluorescence from each standard was recorded against the standard concentration to develop a 5-point
calibration curve. Samples analyzed that exhibited concentrations outside of the calibration curve were
diluted until the results were within the range of concentration covered by the calibration standards (0.05
to 1.5 ppm). The corresponding result was then multiplied by the dilution factor to calculate the
concentration in ppm. Field results for total PAH and TPH were reported on a wet-weight basis.
EPA Region 1 Mobile Laboratory Polychlorinated Biphenyl Analyses
Historical results for the Cos Cob site include a number of detections for PCBs. Twenty-seven sampling
locations were identified for PCB analysis based on historical data. The EPA Region 1 Mobile
Laboratory analyzed all samples collected from the 0- to 1- and 1- to 2-foot intervals for PCBs as samples
were collected. If any PCBs were detected, then subsequent-deeper intervals at the location were sampled
up to a total depth of 4 feet bgs. Additional samples were also analyzed as time permitted and at the
discretion of the M&E field team leader.
Soil was analyzed for PCBs in the field using the EPA Region 1 standard operating procedures (SOP) for
field analysis of soil and sediment samples (EIA-FLDPCB2.SOP). Approximately 1 gram of soil was
weighed and placed in a 4 mL vial for micro-extraction. Next, 200 uL of reagent-grade water, 800 uL of
methanol, and 1,000 |iL of hexane were added to the vial. The sample was then held on a vortex to
vigorously shake the sample and reagents for 1 minute as part of the sample extraction step, and the vial
was then placed in a centrifuge to separate the liquid extract from any residual solids. A portion of the
extract was removed from the centrifuged sample and injected directly into a Shimadzu GC-14A gas
chromatograph equipped with an electron capture detector (GC/ECD). The Shimadzu GC-14A uses a
steel silica coated column (Restek MXT-5) with a 30-meter length and 0.53 millimeter diameter.
Before samples were analyzed each day, a number of Aroclor standards were run, including Aroclor
1242, Aroclor 1248, Aroclorl254, and Aroclor 1260. The resulting chromatograms were then used as
external standards for comparison to chromatograms from soil samples collected during the Cos Cob
investigation. Based on the comparison of the external standards to actual sample chromatograms, PCBs
30
-------
in soil samples were tentatively identified. When the field GC/ECD indicated the presence of any PCB
congeners, the samples were sent for off-site comparative analysis by the EPA Region 1 laboratory using
the internal SOP (PESTSOIL2.SOP). The method used is based on EPA Method 8082 and employed a
high-resolution Hewlett Packard 5890 GC equipped with a dual-column BCD.
Reporting limits using the Shimadzu GC-14A in the field and following EPA Region 1 SOP were
adequate to support project decision-making. The residential DEC for total PCBs (1.0 ppm) compared
well with GC-14A reporting limits for Aroclors 1260 and 1254 of 0.5 ppm each, while the reporting limit
for Aroclors 1242 and 1248 were 1.0 ppm. Therefore, the single column GC/ECD used in the field could
identify the presence or absence of any PCB congener of potential interest to levels at or below the DEC.
.Based on historical site use information, samples-irom the southern most portion of the site were not
analyzed for the presence of PCBs because no record of transformer storage was evidenced. Field results
for PCBs were reported on a wet-weight basis. Drying of the samples was not deemed necessary, because
correction would have simply resulted in an increase in the apparent concentration. Since all positive
results were to be sent for comparative off-site analyses sample drying was not preformed, saving time
and money in performing the field analyses for the PCBs.
EPA Region 1 Mobile Laboratory XRF
In addition to sending samples from the 0- to 1-foot and 1- to 2-foot intervals to an off-site laboratory for
analysis of arsenic by EPA SW-846 Method 3050B/6010B, the EPA Region 1 Mobile Laboratory also
analyzed samples for metals using XRF technology. Results were reported on a wet-weight basis for
lead, arsenic, zinc, copper, and nickel.
Soil samples from the 0- to 1-foot interval were analyzed using a Niton 732 XRF instrument, equipped
with a cadmium-109 radioactive source and a silicon pin detector. Samples were collected in 1-gallon
Ziplocbags and thoroughly homogenized by manual kneeding. Additional sample preparation steps such
as drying or sieving were not employed. After homogenization, a portion of the sample was placed in an
XRF cup and pressed firmly against the Mylar film placed over the sample cup. Accurate XRF results are
best achieved when the sample cup is filled and the sample is in good contact with the film that covers the
cup. The XRF probe window then can be placed in direct contact with the sample cup for the most
accurate readings.
The XRF reporting limit for arsenic (60 mg/kg) exceeded the DEC of 10 mg/kg, and reporting limits for
the other metals were relatively elevated. A reporting limit of 40 mg/kg was used for lead, 70 mg/kg for
31
-------
zinc, 150 mg/kg for copper, and 200 mg/kg for nickel. Results were based on wet weight and provided
good screening information to help identify arsenic hot spots and indicate whether additional metals were
present at levels near the residential DECs. Select samples were sent for off-site analysis for a full suite
of metals. Because XRF reporting limits for several metals exceeded the DECs, the instrument was not
used to make any decisions in the field.
ARSENIC RESULTS
Arsenic is believed to be the metal that presents the greatest potential risk to human health and the
environment at the Cos Cob site based on past site use and disposal practices. A total of 112 samples
(excluding duplicates) were collected for arsenic analysis during the field effort in February 2003. Fifty-
six of the 72 grids were sampled successfully, and samples analyzed for the presence of arsenic for the 0-
to 1-foot and 1- to 2-foot intervals at each location. Laboratory analytical results for arsenic are provided
on a dry-weight basis in Figure 8. Individual sampling grids are color coded to display areas where: (1)
samples could not be collected, (2) the maximum arsenic result in samples collected from the top 2 feet
was reported as less than the DEC, (3) the maximum arsenic result in samples collected from the top 2
feet exceeds toe residential DEC, and (4) the maximum arsenic result in samples collected from the top 2
feet exceeds two times the DEC. Regardless of the calculated 95 percent upper confidence limit (95UCL)
for a specific analyte in a sample data set, CTDEP requires that any area with a sample result greater than
two times the DEC be remediated to limit exposure.
Elevated levels of arsenic were found commonly at two times the DEC in the top 2 feet of soil across the
site, with higher concentrations generally reported from the southern "fly ash" area. Further review of the
data also reveals that 55 of the 112 total results (49 percent) for arsenic exceeded the DEC of 10 mg/kg.
In addition, concentrations from 32 of the 112 results (29 percent) were greater than two times the DEC.
Summary statistics and statistical plots for arsenic are included in Enclosure 2. The summary statistics
include the detection frequency, mean, median, geometric mean, minimum detected result, maximum
detected result, standard deviation, variance, and the 95UCL. Probability plots, histograms, and box-and-
whisker-plots for both the untransformed and log-transformed data are also provided in Enclosure 2.
Review of the probability plots indicates that the arsenic results visually approximate a single lognormal
distribution.
The 95UCL for arsenic is approximately 30 ppm across the site. This concentration for arsenic is
relatively low considering the planned reuse for the property (recreational), but greater than residential
32
-------
risk-based criteria. This type of contamination is likely present to a depth of almost 30 feet bgs at the site
based on the CSM (the known distribution of coal and or fly ash at the site).
PAH RESULTS
Samples from the 0- to 1-foot bgs interval at each sample location were analyzed for PAH and TPH using
siteLAB® Test Kits. Based on the test kit results, a subpopulation of these samples was submitted for off-
site analysis for PAHs using a GC/MS method operated in the SIM mode. Samples from deeper intervals
were collected from (1 to 2 foot, 2 to 3 foot, and 3 to 4 foot intervals) at locations where total PAH
concentrations were found to be elevated. Analysis of the samples from the deeper intervals was
completed as time allowed and at the discretion of the M&E field team leader.
Given the limited resources for comparison analyses, off-site analysis of samples with total PAH results
less than 50 ppm or greater than 500 ppm was limited to two samples below and two samples above these
levels of concentration. Results found between 50 ppm and 500 ppm were considered to be in the range
of concentration where it was most important to continue to refine the relationship between results from
the field test kit and the off-site laboratory. Fifteen samples with total PAH results in this range were sent
for off-site analysis. Therefore, a total of 19 comparison sample pairs were collected and sent for off-site
analysis during the February 2003 sampling event. Comparison samples were chosen to represent a range
of values within the 50 ppm to 500 ppm window and from across the site to improve the spatial
representativeness of the comparison data set.
Summary statistics and statistical plots for the PAH results are provided in Enclosure 2. Usable data
points from the methods applicability study conducted in December 2002 have been included in the data
sets for PAHs. Specifically, PAH results for samples T-l 0-2', T-3 0-2', T-4 0-2', and T-5 0-2' from
December 2002 were included in the PAH data set.
Statistics and plots for PAHs were divided into two different categories: a summary of the results from
the siteLAB®field test kit for total PAHs, and a summary of the results for total cPAHs from the off-site
comparison analysis by EPA Method 8270C operated in the SIM mode. The summary statistics for both
PAH categories include the detection frequency, mean, median, geometric mean, minimum detected
result, maximum detected result, standard deviation, variance, and the 95UCL. Probability plots,
histograms, and box-and-whisker plots for both the untransformed and log-transformed data are provided
in Enclosure 2.
33
-------
Correlating Field PAH Results with the Connecticut Residential Direct Exposure Criterion
Field results did not appear to have a simple relationship to the DEC for cPAHs. This finding is not
surprising or unusual, considering the complexity of the hydrocarbon chemistry present at the site as
evidenced by the distinctly differing TPH patterns observed across the site. To better understand the
relationship between results from the test kit and laboratory, results for cPAHs from the fixed laboratory
were plotted in two-dimensional scatter plots for comparison with field-based results (Figure 9). Total
cPAH values were calculated by adding the sum of the five carcinogenic PAH compounds
(benz[a]anthracene, benzo[b]fluroanthene, benzo[a]pyrene, indeno[l,2,3-cd]pyrene, and
dibenz[a,h]anthracene) for results reported from the fixed laboratory analyses.
Results for the 19 comparison samples collected in February 2003 and the four reanalyzed sample results
from the DMA study were combined to create a 23-data-point comparison (Figure 9). Results for total
PAHs from the siteLAB® field test kits were plotted on the y-axis, and results for cPAHs were plotted on
the x-axis. Correlations were poor for several samples (J-9 2-3', T-l 0-2', T-3 0-2', F-8 0-1', and A-3 0-
1'), but review of the sample chromatograms indicated some potential explanations for the observed
discrepancies. These issues are discussed in more detail later in this case study.
After the analytical results for cPAHs had been reviewed, it was determined that the lowest total cPAH
result where an individual cPAH compound exceeded the DEC was found in sample F-8 3-4': The total
cPAH result for this sample was 4.03 ppm, while individual cPAH results for benz(a)anthracene (1.3
ppm) and benzo(a)pyrene (1.1 ppm) exceeded the DEC of 1.0 ppm. On this basis, a slightly more
conservative value of 4.25 ppm was then plotted on a best-fit regression line where the total PAH result
from the siteLAB® test kits was the dependent variable and the total cPAH result from the fixed
laboratory was the independent variable (Figure 9A). According to the best-fit line, a value of 4.25 ppm
for total cPAH corresponded to a siteLAB® field test kit value of 510 ppm. An additional 20-percent
safety factor was built into the kit value, and 400 ppm was estimated as the kit value where corresponding
cPAH compounds may begin to exceed the DEC of 1.0 ppm.
Field-based decision criteria for data from the siteLAB® field test kits was then applied to the results of
the February 2003 sampling event and used to prepare a map that displays locations of soil samples where
concentrations were expected to exceed the DEC. Figure 10 provides an overview of the site sampling
grids and indicates locations where total PAH values from the siteLAB® field test kits exceeded 400 ppm.
Elevated levels of total PAHs are indicated in the top several feet of soil across the site, with higher
concentrations found in the southern "fly ash" area. Further review of the data also reveals that of the 93
34
-------
total PAH results from the siteLAB* field test kits, 14 (15 percent) exceeded the DEC of 400 ppm, and
two of the 93 results (2 percent) exceeded two-times the decision criteria. In addition, of the 23
comparison samples, seven results from the siteLAB® test kit exceeded 400 ppm, while individual cPAH
values in nine samples exceeded the DEC of 1.0 ppm.
Understanding Field-based Results for PAHs by Reviewing Sample Chromatograms
A review of Figure 9 indicates a reasonable degree of correlation between the field and laboratory PAH
results for most of the samples, particularly those samples where the chromatograms indicate the presence
of weathered lubricating oil range hydrocarbons. There are, however, several exceptions to the good
correlation between field-based and laboratory results. In the case of sample J-9 2-3', for example, the
siteLAB® test kit indicated an elevated total PAH value, while the laboratory results indicated relatively
lower values for cPAHs. For samples F-8 0-1' and A-3 0-1', the siteLAB® field test kit indicated low
total PAH values and laboratory results indicated relatively higher values for cPAHs. In these cases, a
review of the sample chromatograms from the fixed laboratory GC analysis indicates a diversity of lighter
fuel-range hydrocarbons with varying concentrations of PAHs (see Enclosure 3). When the lighter fuel-
range hydrocarbons are present, the response of the field test kit for total PAHs appears less reliable
(given the kit calibration scheme used).
Summary of Findings for Field-generated PAH Data
Overall, the PAH data generated in the field using the siteLAB® test kit appear adequate for the decision
making purposes of this project, despite the difficult matrix. The reliability of any correlation between
field and laboratory results is complicated by the variety of hydrocarbon sources present at the site and
the complexity of hydrocarbon chemistry in general concerning cPAHs and their related action levels.
Although the highest concentrations of PAHs may be associated with releases of fuels and other
petroleum products (as opposed to the fly ash), the wide distribution of relatively low levels of cPAHs
across the southern portion of the property suggests that contamination levels that exceed the DEC for
cPAHs could persist well below ground surface in the southern portion of the site. As with arsenic, this
suggests that removal of soil to meet residential standards may not be feasible at the site and perhaps
alternative standards or modifying reuse alternatives should be considered.
35
-------
TPH RESULTS
TPH analyses were conducted along with the PAH analyses for the 0 to 1-foot interval for each sampling
location using a siteLAB® test kit. Based on the test kit results, a subpopulation of these samples was
submitted for off-site analysis of TPH using a gas chromatograph/flame ionization detector (GC/FID)
method (EPA Method 8015). Samples from deeper intervals (1 to 2 feet, 2 to 3 feet, and 3 to 4 feet) were
analyzed at locations where total TPH concentrations were found to be elevated. Analysis of the samples
from deeper depth intervals using siteLAB* instrumentation was completed as time allowed and at the
discretion of the M&E field team leader.
Given the limited resources for comparison testing, off-site analysis of samples for TPH field results less
than 100 ppm or greater than 500 ppm (the two decision criteria) were limited with two samples below
and two samples above these levels. Results found between 100 ppm and 500 ppm were considered to be
in the range of concentration where it was most important to continue to refine the relationship between
the results from the field test kit and the fixed laboratory. Five samples with total TPH results in this
range were sent for off-site analysis. Therefore, a total of nine comparison sample pairs were collected
and sent for off-site analysis during the February 2003 sampling event. Comparison samples were chosen
to represent a range of values within the 100 ppm to 500 ppm window and from across the site to improve
the spatial representativeness of the collaborative data set.
Summary statistics and statistical plots for the TPH results are provided in Enclosure 2. Usable data
points from the methods applicability study conducted in December 2002 have been included in the data
sets for TPH. Specifically, results for samples T-l 0-2', T-3 0-2', T-4 0-2', T-5 0-2', T-5 2-4', T-13 0-2',
T-13 2-4', and T-17 2-3' from December 2002 were included in the TPH data set. TPH statistics and
plots were divided into two categories: a summary of the results from the siteLAB® field test kit for TPH,
and a summary of the TPH results from the off-site analyses by EPA Method 8015. In general, direct
statistical comparisons are only valid if identical methods for quantification are used.
The summary statistics for the two TPH categories include the detection frequency, mean, median,
geometric mean, minimum detected result, maximum detected result, standard deviation, variance, and
the 95UCL. Probability plots, histograms, and box-and-whisker plots for both the untransformed and log-
transformed data are provided in Enclosure 2.
36
-------
Correlating Field-based Results for TPH with the Residential Connecticut Direct Exposure
Criterion
In addition to the standard summary statistics and statistical plots, TPH results from the siteLAB® field
test kit and from the fixed laboratory were plotted in two-dimensional scatterplots for comparison.
Results for eight of the comparison sample pairs collected in February 2003 (no field analysis was
completed for sample KK-2 0-1') and the eight sample pairs from the methods applicability study
conducted in December 2002 were combined to create a 16-data-point comparison (Figure 11). TPH
results from the siteLAB® field test kits were plotted on the y-axis, while results for TPH from the fixed
laboratory analyses were plotted on the x-axis. Correlations for several samples were poor, with some kit
values over-predicting the laboratory result (T-l 0-2', T-3 0-2'), while others tended to underestimate the
result from the fixed laboratory (T-5 0-2', and G-9 0-1'). Review of the sample chromatograms indicates
the reason for the discrepancies. Sample chromatograms are discussed in more detail later in this case
study.
After the analytical results for TPH had been reviewed, it was determined that the lowest field result for
total TPH where the corresponding fixed laboratory result exceeded the CTDEP residential DEC of 500
mg/kg was found in sample T-5 0-2'. Problems were found both for this sample and sample G-9 0-1'
associated with quantitation of the TPH results from the fixed laboratory. Problems associated with
quantitation of samples T-5 0-2' and G-9 0-1' are discussed below. Issues associated with the
quantitation of the results for these samples were, however, considered to be inconsequential relative to
constructing a best fit line and field-based action level. The DEC (500 ppm) was plotted as the fixed
laboratory value on the linear regression curve (Figure 11 A) and a corresponding field TPH value
computed. The best-fit line indicated that the corresponding value from the siteLAB® field test kit for
TPH was 668 ppm. Given the good correlation (R= 0.77) between the fixed and field results for the
majority of the samples, an additional 10 percent safety factor was built into the kit value, and a value of
600 mg/kg was estimated as the field test kit value where the corresponding result for TPH from the fixed
laboratory may begin to exceed the DEC of 500 mg/kg.
Based on the field data collected using the siteLAB® field test kits, a map was prepared showing the
estimated distribution of TPH results that exceeded the DEC (Figure 12). Figure 12 provides an overview
of the site sampling grids and indicates locations where TPH values from the siteLAB® field test kits
exceed the field action level of 600 mg/kg. Elevated levels of TPH are distributed across the site, with
higher concentrations found in the southern "fly ash" area. Further review of the data also reveals that, of
the 93 total TPH results from the siteLAB® field test kits, 32 (34 percent) exceeded the field-based
decision criterion of 600 mg/kg, and 21 of the 93 results (23 percent) exceeded two-times the field
37
-------
decision criteria. In addition, of the 16 comparison sample pairs, results for four from the siteLAB® test
kit exceeded 600 mg/kg, while fixed laboratory values for six samples exceeded the DEC of 500 mg/kg.
Understanding Field-based Results for TPH by Reviewing Sample Chromatograms
A review of Figure 11 indicates a good correlation among the majority of the samples. Chromatograms
for these samples generally indicated the presence of weathered lubricating oil-range hydrocarbons.
There are, however, a few exceptions to the good correlation, including samples T-l 0-2', T-3 0-2', T-5 0-
2' and G-9 0-1'. These outliers were also classified as lubricating oil-range hydrocarbons by the
laboratory (see Enclosure 3). These outliers likely reflect random variation in hydrocarbon composition
at the site, as well as uncertainties in analyte identification and quantitation by the field and fixed
laboratory methods. The difference in results could also be a function of high sub-sample heterogeneity
commonly referred to as the nugget effect. While this is a common occurrence at hazardous waste sites
and must be addressed where the decisions being made in the field may be in support of an excavation or
other treatment activity, at the Cos Cob site this was not the case. These excursions from the norm were
noted, but no action was deemed necessary because sufficient data that fit the linear regression were
available such that a general decision concerning nature and extent of TPH contamination could be made.
Under different circumstances such as those mentioned above corrective action might have been needed
to be considered. There are several ways to limit the impact of contaminant heterogeneity commonly
associated with the nugget effect on sample results. Project teams may need to consider increasing
sample volumes, conducting sample processing/homogenization, collecting composite samples, and other
sampling alternatives when the nugget effect is pronounced and site decisions require a higher degree of
certainty.
Summary of Findings for TPH
Results for TPH are similar to those reported for PAHs, as might be anticipated, although contamination
levels that exceed the residential DECs appear to be even more wide spread (Figure 12). TPH
concentrations suggest that removal of soil to meet residential standards across the site would be cost
prohibitive and that other alternative reuse plans or cleanup options should be considered.
PCB RESULTS
Samples were collected for analysis of PCBs in 32 grids that were also sampled for the other target
parameters. Additional "step-out" samples were collected where PCBs were tentatively identified by the
38
-------
mobile laboratory, as possible, to further assess the lateral and vertical extent of PCB contamination
within the schedule and budget constraints of the field program. Sample locations, depths, and results are
displayed in Figures 13 and 14.
Samples from 21 grids where aliquots were analyzed for arsenic, PAHs, and TPH were not subsequently
analyzed for PCBs. Samples were not analyzed for PCBs from grids: K3, K5, J3, J4, and J5, located at
the site entrance, nor in grids D6, D7, C2, C3, C4, C5, C6, C7, B2, B3 B5, B6, A2, A3, A4, and A5, in
the southern portion of the site. Samples for analysis of PCBs were not collected in these grids because
no evidence of transformer storage in these areas was indicated based on historical use records. In some
areas, PCB contamination could not be adequately delineated due to budget and time limitations. Those
areas Where data gaps remain include (grids F8, F9, and Gl 1) shown on Figure 14. Additional sampling
will likely be required in these areas should PCB contamination require remediation at the site.
The majority of PCB detections were tentatively identified as Aroclor 1260 by the mobile laboratory,
however, results for comparative fixed laboratory analyses, indicated the presence of Aroclor 1242,
Aroclor 1262, and Aroclor 1268 as well. Detections of Aroclor 1260 in samples from three locations
(sample G-l 1 1-2' and step-out samples F9N 0-1' and F9NW 1-2') analyzed in the field were significant
(10 ppm or greater). At sampling grids F8, F9, and Gl 1 where PCBs were detected at greater than 5 ppm
in the 0- to 1-foot or 1- to 2-foot intervals, the hot spots were further delineated vertically by sampling
subsequent deeper 1-foot intervals until results were non-detected or the 3 to 4 feet interval was reached.
The contamination was partially delineated horizontally by surrounding the location with an additional
four boreholes approximately 10 feet from the original borehole. If the 10-foot step-out (horizontal)
samples were analyzed and PCBs were detected, then the 10-foot step out location with the highest
concentration was surrounded with 5-foot step-out samples to further delineate the contamination. This
logic continued at the discretion of the M&E field team leader as time and budget permitted in an attempt
to delineate the approximate size of the identified PCB "hot spots."
Fourteen samples were sent for off-site comparative analysis for PCBs. At locations D2 0-1', D5 0-1',
and J12 0-1', samples analyzed in the field indicated the presence of PCBs. Subsequent comparative
analyses did not, however, detect PCBs in samples from these locations suggesting the presence of some
type of matrix interference. At locations F8 0-1', F9 0-1', G9 0-1', G10 0-1', and Gil 0-1, G13 0-1' and
H9 0-1' and step-out samples F9N 0-1', F9NN 0-1', and F9NW 0-1', off-site analysis confirmed the
results from the field laboratory for both non-detected values (submitted at a frequency of 10 percent of
all samples collected for PCB analysis) and reported positive results.
39
-------
Cos Cob Power Plant Site
LESSONS LEARNED •M^^^^^HH^^^^^MH^^^^^^^^MBB
Using the Triad approach successfully improved site coverage and the certainty with which the town can
i
now move forward with a reuse plan for the site within a single TEA funding cycle. The original
judgmental sampling plan was revised through the cooperative development of a systematic plan, which
called for the use of a dynamic work strategy and the use of field-based analytical measurements. The
planning process identified key ranges of concentrations and safety factors to guide data interpretation
and decision-making. Increased data density as a result of using field-based methods, enabled the project
team to more aggressively manage decision uncertainties.
In general, TPH values from the field test kits had greater reliability than those obtained for PAH values
primarily because of the diversity of hydrocarbon signatures found at the site. TPH results correlated very
well with fixed laboratory results and could be used to make subsequent decisions for the site. Results
from the field test kit for TPH provide a more powerful tool for making site decisions. Lighter fuel-range
hydrocarbons proved particularly troublesome when developing a predictive relationship between the
total PAH results from the siteLAB® field test kits and the results from the fixed laboratory for total
cPAHs. Analysis of the data indicated that developing a strong linear predictive relationship for decisions
based on total PAHs would be even more difficult. However, using safety factors and decision confidence
intervals (as described on page 20) allowed data for both PAH and TPH to provide sufficient maturation
of the CSM in relation to the potential issues related to site redevelopment.
Analytical results for arsenic in samples from the 0- to 1-foot and 1- to 2-foot intervals indicated that '
arsenic values consistently exceeded the DEC of 10 mg/kg. Additional sampling at greater depths is
expected to provide similar results based on the uniformity of fly and slag ash fill in the southern and
northern portions of the site.
Because the schedule of the field team was not sufficiently flexible to remain at the site until PCB
contamination could be completely delineated, several issues with PCBs at the site may remain. Schedule
and budget constraints did not allow for the mobile laboratory or the subcontract driller to remain on-site
as was necessary to complete the PCB delineation. Additionally, weather conditions at the site
deteriorated over the course of the investigation and a snowstorm in conjunction with schedule and
budget issues converged to make it necessary to demobilize prior to completion of the PCB delineation
phase.
40
-------
Some additional step-out sampling may be required to more adequately define the extent of PCB
contamination within grids F8, F9, and Gl 1 (Figure 14). Future sampling strategies should employ a
focused effort within these grids, where the 70- by 70-foot grids are further divided into 10-foot grid
sections and additional samples collected. Using this approach, sample locations where PCBs were
detected could be delineated with a higher degree of certainty, possibly using a field test kit, which might
improve sample throughput. This approach may also provide the additional data necessary to support a
case-by-case removal of PCB hot spots, should this approach be deemed prudent at the site. The
remaining data gaps for PCBs are minor, however, and can be easily addressed at the beginning of site
redevelopment rather than in a second field investigation.
Overall, the project successfully expanded the data available to the town to support decision making in a
single 1-week mobilization. Sites with multiple sources of differing forms of hydrocarbons can be
challenging when attempting to apply both laboratory and field-based methods for decision-making.
Laboratory methods and field-based methods must be well constrained and both types of data reviewed
carefully before predictive relationships are deemed adequate or sufficiently representative for making
site decisions. Based on the CSM developed for the site, it is clear that removal of soil necessary to meet
residential standards for arsenic, PAHs, and TPH could be cost prohibitive. PCB hot spots exceeding
residential standards could be further delineated, excavated and disposed of appropriately at relatively low
cost. Direct exposure to the remaining arsenic, PAH, and TPH contamination can be limited by placing a
clean soil cap over portions of the site and placing deed restrictions on the property to prevent digging
into the contaminated materials. Alternative reuse plans and the development of more realistic risk-based
action levels should also be considered, in the BTSC's opinion, before site redevelopment proceeds.
41
-------
Cos Cob Power Plant Site
COST COMPARISON ^^^^••••^^^^^^^^^^•^•BB^SSS
The TEA characterization at the Cos Cob site that was based on the Triad approach produced
considerable savings when compared with more traditional characterization approaches. Table 5 provides
a summary of the estimated cost of using a traditional phased approach versus the cost of the actual Triad
project. Using systematic planning, dynamic work strategies, and field-based measurement technologies
with limited off-site comparative analyses allowed for cost-effective site characterization with savings of
approximately 35 percent over a traditional plan. Cost savings could potentially range between 20 and 40
percent depending on actual findings and the number of necessary mobilizations that would been required
to sufficiently characterize the site using a traditional approach. Following the principles of the Triad
approach, the site was characterized to project decision confidence using a single mobilization by
providing greater data densities where needed to satisfy project goals. It is always difficult to estimate
what project costs would have been if a different plan had been followed. In this case, the original work
plan was abandoned in favor of a more dynamic, field-based strategy before cost scenarios for the original
plan had been fully calculated. Where available, existing and estimated costs developed to support the
original approach have been incorporated into this cost comparison.
This cost comparison assumed that a traditional approach would have required two mobilizations to
evaluate the nature and extent of contamination at the property. The costs associated with two
mobilizations would have also been accompanied by costs for developing two different work plans and
two sets of analytical suites. In comparison, Table 5 shows that while planning and field preparation
costs (including the DMA) under the Triad approach were higher for the first mobilization, the
completeness of the resulting data set eliminated the need for a second mobilization for site
characterization. A dynamic approach provided total site coverage using a fairly dense spatial grid, while
allowing the field team to react to areas with samples containing elevated levels of contaminants. Under
the original plan, the lower proposed sampling density may have missed areas of contamination. Any that
were encountered would likely not have been identified for further delineation until after demobilization
when all the analytical data had been received back from the laboratory and compiled into a report to
project decision makers. Delineation efforts would then have to be deferred to a subsequent mobilization.
A dynamic work plan, field kits for TPH and PAHs, and a mobile laboratory for PCBs analysis supported
rapid identification of contaminated areas for evaluating the nature and extent of the contamination, while
providing analytical cost savings. Field kits also provided the data necessary to choose comparison
samples that provided the best coverage of spatial variability and ranges of concentration. Resource
constraints limited the number of comparison sample pairs for TPH, PAH, and PCBs, so it was important
42
-------
to choose samples that would contribute most to developing correlations between field analytical and
fixed laboratory analytical techniques.
The 35 percent cost savings were realized despite the initial unfamiliarity of the project team with the
Triad. Because of this lack of familiarity with Triad concepts, the project required additional meetings,
consultations, and planning activities between the BTSC and the project team as the Triad approach was
adopted. Table 5 shows that the hours devoted to planning activities increased from 200 hours (estimated
for the planning for the original approach) to 345 actual hours. Additional consultations and reviews
concerning the Triad approach, which are included as a separate line item at the end of Table 5, are
estimated to have added $18,150 in labor and other direct costs to the Cos Cob project budget. This level
of planning and consultation would not be needed as.project staff develop familiarity with the Triad
approach, which would further increase the cost savings for future Triad projects.
43
-------
Cos Cob Power Plant Site
BIBLIOGRAPHY wmmmmmmmmmmmmmmmm^mmmii^^^^mmmm^^^fmSSm
EPA. 2000. SW-846 Test Methods For Evaluating Solid Waste, Physical/Chemical Methods. Draft
Update IVB, November 2000.
EPA. 2002. Guidance On choosing A Sampling Design For Environmental Data Collection For Use In
Developing A Quality Assurance Project Plan. EPA QA/G-5S.
EPA. 2003. "Using the Triad Approach to Streamline Brownfields Site Assessment and Cleanup."
Brownfields Technology Primer Series. Technology Innovation Office, Brownfields Technology
Support Center. EPA 542-B-03-002. June.
Marin Environmental, Inc. (Marin). 1997. "Transformer/Breaker - PCB Investigation, Cos Cob Power
Plant, Sound Shore Road, Greenwich, Connecticut." May.
Marin Environmental, Inc. (Marin). 1998a. "Limited Soils Investigation, Cos Cob Power Plant,
Greenwich, Connecticut." March.
Marin Environmental, Inc. (Marin). 1998b. Letter Regarding Contaminated Soil Removal and Disposal,
Cos Cob Power Plant, Greenwich, Connecticut. From Gerald M. Clark, Regional Manager,
Marin. To Lori Saliby, Pesticide, PCB, UST and Marine Management Division, Connecticut
Department of Environmental Protection. July 6.
Metcalf & Eddy (M&E). 2003a. "Draft Field Task Work Plan, Former Cos Cob Power Plant,
Greenwich, Connecticut. January."
Metcalf &Eddy (M&E). 2003b. "Draft Technical Memorandum, Former Cos Cob Power Plant,
Greenwich, Connecticut, Targeted Brownfields Assessment. Test Kit Trials-Investigation
Summary and Suggested Approach for Use of Test Kits." January 15.
Metcalf &Eddy (M&E). 2003c. "Targeted Brownfields Assessment, Draft Report, Former Cos Cob
Power Plant, Greenwich, Connecticut." June.
Osprey Environmental Engineering. 1998. Letter Regarding Transformer Oil Characterization-Cos Cob
Power Plant, Greenwich, Connecticut. From Timothy O. Myjak, Environmental Scientist,
Osprey. To Alan Monelli, Superintendent of Building Construction and Maintenance Division,
Greenwich Town Hall. November 2.
TRC Environmental Consultants (TRC). 1988. "Environmental Site Assessment Report, Cos Cob Power
Plant, Greenwich, Connecticut." April 28.
U.S. Environmental Protection Agency. 1998. "Statement of Work for the Targeted Brownfields
Assessment at the Cos Cob Site."
44
-------
TABLES
-------
-------
Primary Source
Primary
Release Mech.
Table 1
Pathway Receptor Diagram
Cos Cob Power Plant, Greenwich, Connecticut
Secondary
Source
Secondary
Release Mech.
Pathway
Exposure
Route
Receptor
Human Biota
Recreational Residential Terrest Aquatic
Ingestion
Dermal
0
0
0
0
o
o
o
o
Wind
Ingestion
Inhalation
Dermal
O
O
O
o
o
o
0
0
o
o
o
0
Surface
Water Run off
Creek Water
Creek Sediments
Water
w
Ingestion
Inhalation
ueimai
O
o
o
o
o
o
o
o
Fill
Solids/Liquids
Infiltration/
Percolation
h
Subsurface
Soil
Ingestion
Inhalation
Dermal
0
O
O
o
o
o
0
o
o
o
o
o
• Completed pathway
O Possible complete pathway (data required)
O Incomplete pathway
0901500I30207\g:\contracts\tio\oswer four\wa 13 - brwnflds\cos cob\coscobfinal_8-04\final documentVfmal tables\table lnew.doc
-------
Table 2
List of Contaminants of Potential Concern at the Cos Cob Power Plant Site,
Greenwich, Connecticut1
Substance
Residential Criteria in ppm
Metals
Arsenic
10
Polynuclear Aromatic Hydrocarbons
Acenaphthene
Acenaphthylene
Anthracene
Benz(a)anthracene3
Benzo(b)fluorathene3
Benzo(k)fluoranthene
Benzo(g,h,i)perylene
Benzo(a)pyrene3
Chrysene
Dibenz(a,h)anthracene3
Fluoranthene
Fluorene
Indeno( 1 ,2,3-cd)pyrene3
Naphthalene
Phenanthrene
Pyrene
1000
1000
1000
1
1
8.4
1000
1
84
I2
1000
1000
I2
1000
1000
1000
Polychlorinated Biphenyls (PCBs)
Total PCBs
1
Total Petroleum Hydrocarbons (TPH)
TPH
500
Notes:
ppm parts per million or milligrams per kilogram
Regulatory threshold limit values were obtained from the State of Connecticut Regulation
of Department of Environmental Protection. On-line Address:
http://dep.state.ct.us/wtr/regs/remediation/rsr.pdf Threshold limit values for
residential/recreational reuse are those listed in this table.
Criteria are based on the estimated reporting limit.
Carcinogenic polynuclear aromatic hydrocarbons used during the demonstration of
methods applicability study to constrain test kit method performance.
-------
Table 3: Summary of Originally1 Proposed Samples
Location
Rationale
Analyses
Near former transformer and TRC sample location SS-3, where
PAHs were detected in soils. Advance soil boring to 16 feet below
ground. Collect samples from 0 to 4 feet (surface soil) and at water
table (if encountered), or as selected based on "field screening"
(sic) and observations, for a total of two samples for laboratory
analysis. Screen each 4-foot interval using portable XRF unit
(arsenic, lead) and photoionization detector (PID) headspace
(petroleum). Install well only if free product (oil) contamination is
observed.
PCBs, extractable TPH
(ETPH), ACM (direct
subcontract labs)
Semi-volatile organic
compounds (SVOC), metals
(Contract Laboratory
Program-Routine Analytical
Services Labs [CLP-RAS])
2, 3,4, 5, 6
Near Marin B and D series borings, where PAHs and/or TPH were
detected in soils in excess of CTRSR soil standards.
Depth and sampling protocol same as for Boring 1. Total number
of samples for laboratory analysis = 10.
PCBs, ETPH, ACM (direct
subcontract labs)
SVOCs, metals (CLP-RAS)
7,8
In former transformer/breaker area, where PCB-containing
equipment is likely to have been located.
Depth and sampling protocol same as for Boring 1. Total number
of samples for laboratory analysis = 4.
PCBs, ETPH, ACM (direct
subcontract labs)
SVOCs, metals (CLP-RAS)
9,10,11
Near Marin A series borings, where TPH was detected in soils in
excess of CTRSR soil standards.
Depth and sampling protocol same as for Boring 1. Total number
of samples for laboratory analysis = 6.
PCBs, ETPH, ACM (direct
subcontract labs)
SVOCs, metals (CLP-RAS)
Distributed evenly throughout property, making sure to include
areas previously identified as containing ash.
Analyze for ACM and ash-related contaminants (SVOCs, metals)
only. Consider ETPH or PCBs only if oily contamination is
observed during field screening. Total number of samples =18.
12 through 20
ACM (direct subcontract
lab)
SVOCs, metals (CLP-RAS)
Notes:
Prepared using information provided in the Draft Field Task Work Plan For the Former Cos Cob
Power Plant, Greenwich, Connecticut (M&E 2003a) and following discussions with the M&E
project manager.
-------
Table 4
Summary of Analytical Data From the Demonstration of Methods Applicability Study
TBA Investigation - Former Cos Cob Power Plant Property - December 2002
LOCATION NAME
SAMPLE DEPTH (ft bgs)
DATE SAMPLED
LUTION FACTOR FOR SVOCs
COMMENTS
PAR AM ETER/AN ALYTE
T-l
0-2
12/9/2002
30
RAS lab
5
OEME
T-3
0-2
12/9/2002
30
RAS lab
5
OEME
T-4
0-2
12/9/2002
30
RAS lab
2
OEME
T-5
0-2
12/9/2002
30
RAS lab
2
OEME
EXTRACTABLE TOTAL PETROLEUM HYDROCARBONS - CTDEP Method (mg/kg)
Total Petroleum Hydrocarbons
Lab's qualitative identification
results (summarized from lab Form
Is):
1,500
Similar to high
MW components
in lube oil range
NA
1,400
Similar to high
MW components
in lube oil range
NA
450
Similar to fuel
oil #6; absence
of straight chain
aliphatics
indicates sample
is weathered
NA
760
Similar to high
MW components
in lube oil range
NA
SITELAB TPH/EDRO TEST KIT CONCENTRATION (ppm wet weight)
TPH/EDRO
3,294
4,001
409
138
Percent Difference, CTDEP vs. SITELAB methods for TPH
75
96
10
139
SEMIVOLATILE ORGANIC COMPOUNDS (only PAH results presented below) (mg/kg)
Naphthalene
2-Methylnaphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
Indeno( 1 ,2,3-cd)pyrene
Dibenz(a,h)anthracene
Benzo(g,h,i)perylene
Total PAHs (mg/kg)**
10U
10U
10U
10U
10U
10 U
10 U
10 U
10U
10U
10U
10U
10U
10U
10U
10U
10U
85
0.043 U
NA
0.047
0.047
0.043 U
0.25
0.065
0.54
0.85
0.42
0.50
0.76
0.22
0.59
0.42
0.16
0.45
5
10U
10U
10U
10U
10U
10 U
10 U
10 U
10 U
10U
10U
10U
10U
10U
10U
10U
10U
85
0.074
NA
0.083
0.043 U
0.043 U
0.48
0.12
1.0
1.5
0.64
0.68
1.0
0.31
0.77
0.50
0.17
0.52
8
12U
12U
12U
12U
12 U
12 U
12 U
12U
12 U
12 U
12 U
12U
12 U
12 U
12U
12U
12 U
102
0.14
NA
0.030
0.020 U
0.020 U
0.39
0.046
0.40
0.44
0.24
0.37
0.36
0.13
0.22
0.18
0.062
0.19
3
12U
12 U
12 U
12 U
12 U
12 U
12 U
12 U
12 U
12 U
12 U
12 U
12 U
12-U
12 U
12 U
12 U
102
0.044
NA
0.013 U
0.01 3 U
0.013 U
0.073
0.013 U
0.088
0.14
0.060
0.082
0.098
0.026
0.063
0.048-
0.016
0.058
1
-------
Summary of Analytical Data From the Demonstration of Methods Applicability Study
TBA Investigation - Former Cos Cob Power Plant Property - December 2002
(continued)
LOCATION NAME
SAMPLE DEPTH (ft bgs)
DATE SAMPLED
LUTION FACTOR FOR SVOCs
COMMENTS
T-l
0-2
12/9/2002
30
RAS lab
5
OEME
T-3
0-2
12/9/2002
30
RAS lab
5
OEME
T-4
0-2
12/9/2002
30
RAS lab
2
OEME
T-5
0-2
12/9/2002
30 2
RAS lab OEME
SITELAB Total PAHs TEST KIT CONCENTRATION (ppm wet weight)
Total PAHs
700
851
83
28.7
LAB SAMPLE ID
ETPH-CTDEP method
SVOCs-RAS and/or OEME
0212046-01
AOZ98
AA27888
0212046-02
AOZ99
AA27889
0212046-03
AOZAO
AA27890
0212046-04
AOZA1 AA27891
NOTES:
* - RAS laboratory data have not undergone data validation.
** - Total PAHs calculated by using 1/2 the detection limit for non-detects, and adding that value to any reported values.
U - analyate note detected at the listed concentration.
J - analyte was detected at a concentration below the reporting limit.
-------
Table 4
Summary of Analytical Data From the Demonstration of Methods Applicability Study
TEA Investigation - Former Cos Cob Power Plant Property - December 2002
(continued)
LOCATION NAME
SAMPLE DEPTH (ft bgs)
DATE SAMPLED
LUTION FACTOR FOR SVOCs
COMMENTS
PARAMETER/AN ALYTE
T-5 field duplicate
0-2
12/9/2002
10
RAS lab
2
OEME
T-5
2-4
12/9/2002
5
T-13
0-2
12/9/2002
10
T-13
2-4
12/9/2002
5
T-13
2-4
12/9/2002
Not submitted
field duplicate
T-17
2-3
12/9/2002
2
Mostly fly ash
CT Residential
DEC
EXTRACTABLE TOTAL PETROLEUM HYDROCARBONS - CTDEP Method (mg/kg)
Total Petroleum Hydrocarbons
Lab's qualitative identification
results (summarized from lab Form
Is):
Not submitted
19
Not enough
material for
qualitative
identification
300
Similar to fuel
oil #6; absence
of straight chain
aliphatics
indicates sample
is weathered
220
Similar to fuel
oil #6; absence
of straight chain
aliphatics
indicates sample
is weathered
350
Similar to fuel
oil #6; absence
of straight chain
aliphatics
indicates sample
is weathered
250
Similar to fuel
oil #6; absence
of straight chain
aliphatics
indicates sample
is weathered
500
SITELAB TPH/EDRO TEST KIT CONCENTRATION (ppm wet weight)
TPH/EDRO
Not submitted |
0.4 U
271
273
Not submitted
237
Percent Difference, CTDEP vs. SITELAB methods for TPH
—
192
10
22
—
5
SEMI VOLATILE ORGANIC COMPOUNDS (only PAH results presented below) (mg/kg)
Naphthalene
2-Methylnaphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
Indeno( 1 ,2,3-cd)pyrene
Dibenz(a,h)anthracene
Benzo(g,h,i)perylene
Total PAHs (mg/kg)**
4.3U
4.3U
4.3U
4.3U
4.3U
4.3U
4.3U
4.3U
4.3U
4.3U
4.3U
4.3U
4.3U
4.3U
4.3U
4.3U
4.3U
37
0.042
NA
0.015
0.01 3 U
0.01 3 U
0.13
0.024
0.18
0.21
0.11
0.10
0.13
0.052
0.093
0.066
0.020
0.073
1
1.8 U
0.86 J
.8U
.8U
.8U
0.36 J
.8U
.8U
.8U
0.27 J
.8U
.8U
.8U
.8U
.8U
.8U
.8U
14
0.82 J
2.2 J
3.8 U
3.8 U
3.8 U
2.1J
3.8 U
0.74 J
0.62 J
0.40 J
1.0 J
0.50 J
3.8 U
3.8 U
3.8 U
3.8 U
3.8 U
25
0.30 J
1.3 J
1.8 U
1.8 U
1.8 U
0.80 J
1.8 U
0.36 J
0.27 J
1.8 U
0.79 J
0.22 J
.8U
.8U
.8U
.8U
.8U
13
. 0.28 J
0.87 J
0.73 U
0.73 U
0.10 J
0.77
0.73 U
0.26 J
0.25 J
0.10 J
0.42 J
0.23 J
0.73 U
0.094 J
0.73 U
0.73U
0.099 J
5.7
1,000
474
,000
jOOO
,000
jOOO
,000
jpoo
^000
1
84
1
8.4
1
1
1
1,000
-------
Summary of Analytical Data From the Demonstration of Methods Applicability Study
TBA Investigation - Former Cos Cob Power Plant Property - December 2002
(continued)
LOCATION NAME
SAMPLE DEPTH (ft bgs)
DATE SAMPLED
LUTION FACTOR FOR SVOCs
COMMENTS
T-5 Held duplicate
0-2
12/9/2002
10
RAS lab
2
OEME
T-5
2-4
12/9/2002
5
T-13
0-2
12/9/2002
10
T-13
2-4
12/9/2002
5
T-13
2-4
12/9/2002
Not submitted
field duplicate
T-17
2-3
12/9/2002
2
Mostly fly ash
CT Residential
DEC
SITELAB Total PAHs TEST KIT CONCENTRATION (ppm wet weight)
Total PAHs
Not submitted |
0.2 U
55
53
Not submitted
52
LAB SAMPLE ID
ETPH-CTDEP method
SVOCs-RAS and/or OEME
Not submitted
AOZA6
AA27896
0212046-05
AOZA2
0212046-06
AOZA3
0212046-07
AOZA4
0212046-09
Not submitted
0212046-08
AOZA5
NOTES:
* - RAS laboratory data have not undergone data validation.
** - Total PAHs calculated by using 1/2 the detection limit for non-detects, and adding that value to any reported values.
U - analyate note detected at the listed concentration.
J - analyte was detected at a concentration below the reporting limit.
-------
TableS
Cost Comparison Between a Traditional Approach
and the Triad Approach
at the Cos Cob Power Plant
Task Description ;
TVs;&%« ' '-- - ', ,
' t&" ^" ^T ^-' '- 1 < ^
Background review,
project planning,
preparation of workplan,
site visit, and
discussions with town
officials
Background review,
project planning,
preparation of workplan,
site visit, and
discussions with town
officials for second
mobilization
First mobilization
Field Investigation
including pre- field work
and post-field paperwork
Second mobilization
Field Investigation
including pre-field work
and post-field
paperwork8
Sample analysis ",
interaction with labs, and
data validation
\*^ /^ ', >/ t f ,\
&<>, • J-
345
200
N/A
200
373
248
N/A
248
60
90
Labor
Costs at
$75/Hour
' *l ~"*.
$25,875
$15,000
N/A
$15,000
$27,975
$18,600
N/A
$18,600
$4,500
$6,750
Subcontractor
_ tifc.**"*
$1,500
N/A
N/A
N/A
$5,500 Driller
$600 IDW
$8,000 Driller
$600 IDW
' N/A
$8,000 Driller
$600 IDW
$4,800 Fixed-lab
A«,TGLP
$4,800 Fixed- lab
ETPH, PAHs
$630 Fixed lab
asbestos
$5,000 OEME
Mobile Lab PCBs,
XRF metals
$22,970 OEME
Fixed lab SVOC,
Metals, PCB,
ACM, and ETPH
analyses
O30C«*v
^'W
$700
$200
N/A
?'
t.
$200
$10,000
$6,000
N/A
$6,000
N/A
N/A .
N/A
N/A:
N/A
Estimated ?
FeeRewiw*1
:-*tm^
$2,607 ,
$1,520
N/A;, /,
$1,520
$4,347
$68
$3,260
$60
.N/A;.;
$3,260
$60
$930
$4S&
' ;f-fi
.$&i y ft
$500";p':^:
•r: '•;, f •
$2,972
Actual Cost Using
tts t^T|l«lm .*/
Estimated
Traditional Costs
$28,682
$16,720
' ^ N/A ,_ _ _ ;
$16,720
$47,822
$660?
$35,860
$660
N/A
$35,860
$660
$10,230
$5,280" - V '?
$693
, f,t*- -v --'-r'-V
$5,500
$32,692
-------
Table 5
Cost Comparison Between a Traditional Approach
and the Triad Approach at the Cos Cob Power Plant (continued)
,-„ %&i$&tris#$k,\$
'1 '':•• ^*V^7V>^*^**t'F ^
' J*/s * ' * "* 4 <* *i /
' '.£$<"-'" "''" ' :.
Sample analysis,
interaction with labs, and
data validation for
second mobilization
Data evaluation, TEA
report preparation, and
file closeout for the first
mobilization
Data evaluation, TEA
report preparation, and
file closeout for the
second mobilization
Consultations about use
of Triad approach
Totals
•f^'lUdbr -
-*:>«*'*• «i-
| Heurs ,;
N/A
90
122
180
122
180
200
N/A
1,100
1,436
i£ Libw ,f
% Costijiiit^ ^
* STSaioiii-,4
* ,;ix^ --;v>
N/A
$6,750
$9,158
$13,500
$9,150
$13,500
15,000
N/A
$82,500
$107,700
Subcontractor
£S ^pMi* :'
-1^,'cv^- -
N/A
$22,970 OEME
Fixed lab SVOC,
Metals, PCB,
ACM, and ETPH
analyses
N/A
N/A
N/A
N/A
N/A
N/A
$22,830
$63,140
ODCs*
N/A
N/A
$1,000
$1,000
$1,000
$1,000
$1,500
N/A
$14,200
$14,400
Estimated :
'Ft* Reserve '
^,atl0^»,,'.
^ <, .( * «
, ;'
N/A
$2,972
$1,015
$1,450
SI, 01 5
$1,450
$1,650
N/A
$11,652
$18,524
Actual Cost Using :•
;.; the Triad vs. ,.*
Estimated
Traditional Costs
WA
$32,692
$11,165
, '^
$15,950
; $11,165
$15,950
SI 8,1 50
N/A
$132,282
$203,764
Notes:
Costs associated with the Triad approach
TRADITIONAL
Estimated costs from a traditional approach
N/A Not applicable
A second mobilization and sampling event was assumed as a requirement under the traditional
approach.
Field analysis of TPH and PAHs using the Sitelab field test kits is included under other direct
costs (ODCs) for the first mobilization
ODCs: Includes copies, phone, sample shipment, computer time, field equipment rentals and
supplies including test kits and fluorescence detector for TPH and PAH field analyses, travel, per
diem, vehicle rental, gasoline etc.
-------
-------
FIGURES
-------
-------
CONNECTICUT
FORMER COS COB POWER PLANT
GREENWICH, CONNECTICUT
FIGURE 1
SITE LOCATION MAP
p taken from 7.5 Minute USGS
x>graphfc Map of the Stamford
mecticut Quadrangle. 1951. {
U.S. EPA REGION I
IN COOPERATION WITH
BROWNFIELDS TECHNOLOGY SUPPORT CENTER
-------
SOUTH
A
NORTH
SITE
SOUTHERN
BOUNDARY
A'
FORMER
TRANSFORMER YARD
FORMER
COS COB
POWER HOUSE
SITE
NORTHERN
BOUNDARY
DRAINAGE
DITCH
PCB
CONTAMINATION
TREES
FORMER
FUEL OIL
TANK
UJ
COS COB
HARBOR
MARINE SILT
NOT TO SCALE
(VERTICALLY EXAGGERATED)
FORMER COS COB POWER PLANT
GREENWICH. CONNECTICUT
FIGURE 2
CONCEPTUAL SITE MODEL DIAGRAM
- U.S. EPA REGION I
IN COOPERATION WITH
BROWNFIELDS TECHNOLOGY SUPPORT CENTER
-------
-TRANSFORMER YARD
-CAT TOWER
IOER BLOCK BUILDING
TRANSFORMER YARD
CRUSHED STONE
SLAG ASH
REMEDIATED BASEMENT
OIL STAINED AREA,
MARIN 1998-v
OVERHEAD ABANDONED
WOOD RAILROAD TRESTLE
OVERHEAD COAL
CONVERTER BELT
FORMER TRANSFORMER/
BREAKER AREA. MARIN 1997
— CONCRETE SEA WALL
A-1 +
B-1-f
IM-f
SD-4 X
SS-1M
LEGEND
ROADWAY
SOIL BOWNGS BY MARIN. 1998
SOIL BORINGS BY TRC. 1988
MONITORING WELL BY TRC. 1988
SEDIMENT SAMPLE BY TRC. 1988
SURFACE SOIL SAMPLE BY TRC, 1988
JUDGMENTAL BOWNG FOR SITEWIDE COVERAGE1
JUDGMENTAL BOWNG TO TARGET AREA OF CONCERN1
COS COB HARBOR
STONE SEA WALL
RQIEJS.
SAMPLE LOCATIONS PROPOSED BY
METCALF AND EDDY IN THE ORIGINAL
(PRE-TRIAD) WORK PLAN FOR THE
TARGETED BROWNF1ELDS ASSESSMENT
(METCALF AND EDDY 20O2)
SO1
50'
=*S
SCALE IN FEET
FORMER COS COB POWER PLANT
CREE.NWICH, CONNECTICUT
FIGURE 3
HISTORICAL SAMPLING LOCATIONS
U.S. EPA REGION I
IN COOPERATION WITH
BROWNFIELDS TECHNOLOGY SUPPORT CENTFR
-------
-TRANSFORMER YARD
-CAT TOWER
-CINDER BLOCK BUILDING
TRANSFORMER YARD
REMEDIATED BASEMENT
OIL STAINED AREA,
MARIN 1998
CRUSHED STONE
SLAG ASH
OVERHEAD COAL
CONVERTER BELT
STONE RETAINING WALL
CAT TOWER 8-4
FORMER TRANSFORMER/
BREAKER AREA, MARIN 1997
CONCRETE SEA WALL
yf o iff 100'
S&==StS
SCALE IN FEET
COS COB HARBOR
OVERHEAD ABANDONED
WOOD RAILROAD TRESTLE
LEGEND
• PROPOSED PROBABILISTIC SAMPLING LOCATION
A-1 + SOIL BORINGS BY MA^IN. (998
B-1 •+- SO|L BORINGS BY TRC. 1988
M-3 -4" MONITORING WELL BY TRC, 1988
SD-« X SEDIMENT SAMPLE BY TRC, 1988
SS-1 K SURFACE SOIL SAMPLE BY TRC. 1988
COLUMN AND ROW BOUNDARY SHOWN
FOR REFERENCE ONLY, NO ADDITIONAL
SAMPLING RECOMMENDED
1 2 3 4 5 6 7. 8 9 10 11 12 13 14
FORMER COS COB POWER PLANT
GREENWICH, CONNECTICUT
FIGURE 4
PROPOSED SAMPLING LOCATIONS
FRQM THE REVISED WORK PLAN
FOR THE TARGETED BROWNFIELDS
ASSESSMENT (TRIAD APPROACH)
U.S. EPA REGION I
IN COOPERATION WITH
6ROWNFIELDS TECHNOLOGY SUPPORT CENTER
-------
-TRANSFORMER YARD
-CAT TOWER
BLOCK BUILDING
TRANSFORMER YARD
REMEDIATED BASEMENT
-..-- J5| OIL STAINED AREA,
LIGHT & POWER
SUBSTATION
COS COB
POWER HOUSE
CRUSHED STONE
SLAG ASH
OIL TANK
DOCK AREA
OVERHEAD COAL
CONVERTER BELT
PROPERTY
BOUNDARY
STONE RETAINING WALL
FORMER TRANSFORMER/
BREAKER AREA, MARIN 1997
CONCRETE SEA WALL
STONE SEA WALL
OVERHEAD ABANDONED
WOOD R.R. TRESTLE
SAMPLE LOCATIONS GPS SURVEYED DECEMBER 2002
SAMPLE LOCATIONS GPS SURVEYED FEBRUARY 2003
GRID NOT SAMPLED DUE TO THE
PRESENCE OF DEBRIS PILES. STEEP
SLOPES. OR HEAVY VEGETATION.
COS COB HARBOR
1 2 3 4 5 6 7 "89 10111213 T4
FORMER COS COB POWER PLANT
GREENWICH. CONNECTICUT
FIGURE 5
SAMPLING LOCATIONS
FEBRUARY 2003
U.S. EPA REGION I
IN COOPERATION WITH
BROWNFIELDS TECHNOLOGY SUPPORT CEN
-------
FIGURE 6
LOGIC DIAGRAM 1:
ARSENIC IN FLY ASH INVESTIGATION
COS COB SITE
Start of site investigation
Divide site into 70 foot by 70 foot grid sectors
Identify and survey a random sampling location
within each grid sector
Drive sampling sleeve to a depth of 4 feet at each
location
Label each 1-foot interval and archive samples
collected from the 2 to 3 foot and the 3 to 4 foot
intervals for later analysis if required.
Homogenize each 1 foot sleeve sample
thoroughly. Send aliquots from the top of 0 to 1
foot interval and the 1 to 2 foot interval to an
offsite laboratory for arsenic analysis using
inductively coupled plasma/atomic emission
spectrometry (ICP/AES)
No
Is arsenic in the 0 to 1 or 1 to 2
foot interval greater than 10 parts
per million (ppm)?
Progressively analyze samples in 1 foot intervals
below and surrounding the hot spot until the
remaining contamination is less than 5 ppm.
Evaluate the potential for hot spot removal.
Evaluate reuse scenarios and
remedial alternatives as necessary.
-------
FIGURE?
LOGIC DIAGRAM 2:
TPH, PAH, AND PCB INVESTIGATION
COS COB SITE
Start of sampling program
Divide site into 70 foot by 70 foot grid sectors
Collect sample from the 0 to 1 foot interval and homogenize.
Analyze sample for total TPH and PAHs using field test kits.
Is the concentration greater than
100 parts per million (ppm) TPH
or greater than 50 ppm PAHs?
Yes
No
Conduct offsite analyses on those samples with
concentrations of TPH greater than 100 ppm but less
than 500 ppm, or PAHs greater than 50 ppm but less
than 500 ppm. Confirmation analyses are not needed
for samples that have concentrations greater than 500
ppm TPH or PAH. (Note 1)
/ No further field analysis is \
| recommended. Send 10 percent \
I of samples to an offsite J
\ laboratory for analysis J
If the TPH concentration is greater than 1000 ppm, or a potential PCB release
is suspected in the erid sector, analvze samples for PCBs in the field.
Have PCBs or potential PCB
compounds been detected in the
sample?
No
Yes
Send samples to an offsite laboratory for confirmation
of PCB results and continue step out sampling and
analysis until the nature and extent of the PCB
contaminant hot spot is constrained. (Note 2)
c
Evaluate Reuse Scenarios and Remedial Alternatives
Notes:
1. TPH and PAH test kit values for
comparison to applicable regulatory
thresholds will be adjusted based on
statistical analysis as the field data set is
compiled.
1. Based on historical data, PCB
concentrations greater than 2 ppm are not
expected. Because of this fact and the
concerns over PCB occurrence at the site,
my samples in which PCBs are detected
using the field-based method are
recommended for further off-site
laboratory analysis.
-------
-TRANSFORMER YARD
-CAT TOWER
CINDER BLOCK BUILDING
TRANSFORMER YARD
CRUSHED STONE
SLAG ASH
REMEDIATED BASEMENT
OIL STAINED AREA.
MARIN 1998-^>
FORMER TRANSFORMER/
BREAKER AREA. MARIN 1997
OVERHEAD ABANDONED
WOOD R.R. TRESTLE
2.3-
4.5-
LEGEND
-0 TO 1 FOOT BELOW GROUND SURFACE
-1 TO 2 FEET BELOW GROUND SURFACE
MAXIMUM ARSENIC RESULT
LESS THAN 10.0 mg/kq
FOR GRID SECTOR
MAXIMUM ARSENIC RESULT GREATER
THAN 10.0 AND LESS THAN 20.0
FOR GRID SECTOR
MAXIMUM ARSENIC RESULT
GREATER THAN 20.0 mgAg
FOR GRID SECTOR
GRID NOT SAMPLED DUE TO THE
PRESENCE OF DEBRIS PILES, STEEP
SLOPES, OR HEAVY VEGETATION.
NOTES:
SAMPLES WERE COLLECTED FROM THE 0-1 FOOT AND
THE 1-Z FOOT INTERVAL USING A GEOPROBE. THESE
SAMPLES WERE HOMOGENIZED AND THEN SUBMITTED TO
SEVERN TRENT LABORATORIES FOR ANALYSIS USING EPA
SW-846 METHODS 30506/60106. ANALYSIS WAS PERFORMED
USING INDUCTIVELY COUPLED PLASMA/ATOMIC EMISSION
SPECTROMETRY (ICP/AES).
THE STATE OF CONNECTICUT DEPARTMENT OF ENVIRONMENTAL
PROTECTION (COEP) RESIDENTIAL DIRECT EXPOSURE CRITERIA (1996)
FOR ARSENIC IS 10 mgAg. SPECIAL RESTRICTIONS CAN APPLY
WHEN RESULTS EXCEED TWO TIMES THE DIRECT EXPOSURE
CRITERIA.
STONE SEA WALL
50' 0 50'
i rt (=
SCALE IN FtET
1 2 3
4
5 6 78 9 10 11 12 13 14
FORMER COS COB POWER PLANT
GREENWICH. CONNECTICUT
FIGURE 8
ARSENIC RESULTS SUMMARY
EPA SW-846 METHOD 601 OB
EPA REGION I
IN COOPERATION WITH
BROWNFIELDS TECHNOLOGY SUPPORT CENTER
-------
FIGURE 9
2,000
M 1,800
'i
« 1,600
•3 1,400
| 1,200
•S S 1,000
i i 8°°
I £ 60°
S 40°
H
2 200
0
-200
Fixed Laboratory Total Carcinogenic PAHs vs.
Field Test Kit Total PAHs
Field =111.04 + 93.862* Lab
Correlation: r=.38610
Field Test Kit Action Level
With 20% Safety Factor Built In;
J-9 2-3'
Field Test Kit ; ;
Total PAH Concentration Indicates a Potential Exceedence
of One or More Carcinogenic PAH Compounds
T-3 0-2'
T-l 0-2' L*.
510 mg/kg
400 mg/kg
*
.f 20%; Safety Factor
'i
.f-
^•^^ r
i
• •" |-
3mg/kg|
r -/•
F-80-11
4.25 mg/kg
A-3 0-1'
-1
0
Total Carcinogenic PAHs in mg/kg
Fixed Laboratory Results On a Wet Weight Basis |»x 95% confidence
-------
Figure 9A
Development of a Field-based Action Level for cPAHs When No Historical Data is Available
1,000
.Confident Decision that True Concentration > AL
Initial cPAH Fixed Laboratory Action Level = 4.25 mg/kg
Field-based Total PAH Equivalent = 510 mg/kg
20 % Safety Factor
Final Field-based Action Level = 400 mg/kg
Including 20% Safety Factor
Confident Decision that True Concentration < AL
0
Total Carcinogenic PAHs in mg/kg | ^95% confidence
Fixed Laboratory Results On a Wet Weight Basis
-------
-TRANSFORMER YARD
-CAT TOWER
OVERHEAD ABANDONED
WOOD R.R. TRESTLE
BLOCK BUILDING
TRANSFORMER YARD
EMEDIATEO BASEMENT
OH. STAINED AREA, 46.
MARIN 1998^ 62.7
725.
81.1 ;
METAL][
FRAME j
BUILDING i
cos cost/ •'
HOUSE
OVERHEAD PIPE
LIGHT & POWER
SUBSTATION
CRUSHED STONE
SLAG ASH
DOCK AREA
01 TANK
OVERHEAD COAL
CONVERTER BELT
Lt_ PROPERTY
BOUNDARY
STONE RETAINING WALL
FORMER TRANSFORMER/
BREAKER AREA, MARIN 1997
CONCRETE
SEA WALL
SO 0 SO'
EX£SMM^!
SCALE IN FEET
COS COB HARBOR
0 TO 1 FOOT BELOW GROUND SURFACE
1 TO 2 FEET BELOW GROUND SURFACE
2 TO 3 FOOT BELOW GROUND SURFACE
3 TO 4 FEET BELOW GROUND SURFACE
TOTAL PAHs GREATER THAN 400 mg/\tg
FOR AT LEAST ONE OF THE 1-FOOT
INTERVALS IN THE GRID SECTOR
GRID NOT SAMPLED DUE TO THE
PRESENCE OF DEBRIS PILES. STEEP
SLOPES. OR HEAVY VEGETATION.
TOTAL PAH> GREATER THAN BOO mgAg
(TWO TIMES THE THRESHOLD VALUE
OF 400 mgA«) FOR AT LEAST ONE OF
1-FOOT INTERVALS IN THE GRID SECTOR
1. SAMPLES «ERE COLLECTED IN 1-FOOT INTERVALS USING A SEOPROBE
THESE SAMPLES WERE HOMOGENIZED AND THEN ANALYZED IN THE
FIELD USING SITE LAB TEST WDS AN A UV auORESCENCE DETECTOR.
THE STATE OF CONNECTICUT DEPARTMENT OF ENVIRONMENTAL
PROTECTION (CDEP) RESIDENTIAL DIRECT EXPOSURE CRITERIA (1996)
FOR ANY SINGLE CARCINOGENIC PAH COUMPOUNO IS 1 moA*
SPECIAL RESTRICTIONS CAN APPLY WHEN RESULTS EXCEED TWO
TIMES THE DIRECT EXPOSURE CRITERIA.
FORMER COS COB POWER PLANT
GREENWICH. CONNECTICUT
1 234567891011121314
FIGURE 10
TOTAL PAH RESULTS SUMMARY
SITE LAB FIELD TEST KITS.
EPA REGION I
IN COOPERATION WITH
BROWNFIELOS TECHNOLOGY SUPPORT CENTER
-------
FIGURE 11
-500
Fixed Laboratory TPH vs. Field Test Kit TPH
Wet Weight
Field =32.741 + 1.2706* Lab
Correlation: r = .76815
Field Test Kit Action Level
With 10% Safety Factor Build In
Field Test Kit
TPH Result indicates a Potential Exceedence-
ofTPH :.-''
10% Safety Factor
-500
0 500 1,000 1,500 2,000 2,500
TPH in mg/kg [
Fixed Laboratory Results on a Wet Weight Basis
3,000 3,500
95% confidence
-------
Development of a Field-based Action Level for TPH When No Historical Data is Available
2,400
«, 2,200
'a
ffl 2,000
0
400
600
800
TPHinmg/kg
1,000 1,200 1,400 1,600
^ 95% confidence
Fixed Laboratory Results on a Wet Weight Basis
Confident Decision that True Concentration > AL
Initial Fixed Laboratory Action Level = 500 mg/kg
Field-based Equivalent = 668 mg/kg
10 % Safety Factor
Final Field-based Action Level = 600 mg/kg
Including 10% Safety Factor
Confident Decision that True Concentration < AL
-------
-TRANSFORMER YARD
-CAT TOWER
CRUSHED STONE
SLAG ASH
OER BLOCK BUILDING
TRANSFORMER YARD
EMEDIATED BASEMENT
STAINED AREA.
4-72,5
364.3 i
289.9 f
187.2!
/.'•'I
cos COB}/'./]
POWER HOUSE
OVERHEAD PIPE
LIGHT & POWER
SUBSTATION
TRANSFORMER
YARD
DOCK AREA
Oft. TANK
PROPERTY
BOUNDARY
STONE RETAINING WALL
FORMER TRANSFORMER/
BREAKER AREA, MARIN 1997
! 100,2
8.5
' G;t
COS COB HARBOR
OVERHEAD ABANDONED
WOOD R.R. TRESTLE
OVERHEAD COAL
CONVERTER BELT
LESEMC
0 TO 1 FOOT BELOW GROUND SURFACE
1 TO 2 FEET BELOW GROUND SURFACE
2 TO 3 FOOT BELOW GROUND SURFACE
3 TO 4 FEET BELOW GROUND SURFACE
TPH GREATER THAN 600 mgA«
FOR AT LEAST ONE OF THE 1-FOOT
INTERVALS IN THE GRID SELECTOR
GRID NOT SAMPLED DUE TO THE
PRESENCE OF DEBRIS PILES. STEEP
SLOPES. OR HEAVY VEGETATION.
TPH GREATER THAN 1200 mg/ttg
(TWO TIMES THE THRESHOLD VALUE
OF 600 mg/kg)
NOTES:
1. SAMPLES WERE COLLECTED W 1-FOOT INTERVALS USING A GEOPROffi.
THESE SAMPLES WERE HOMOGENIZED AND THEN ANALYZED IN THE
FIELD USING SITE LAB TEST KITS AND A UV aUORSCENCE DETECTOR.
2. THE STATE OF CONNECTICUT DEPARTMENT OF ENVIRONMENTAL
PROTECTION (CDEP) RESIDENTIAL DIRECT EXPOSURE CRITERIA (1996)
FOR TPH IS 500 mjAs- SPECIAL RESTRICTIONS CAN APPLY
WHEN RESULTS EXCEED THO TIMES THE DIRECT EXPOSURE CRITERIA.
CONCRETE
SEA WALL
STONE SEA WALL
wo sff
- — ——
SCALE IN FEET
io
-------
-TRANSFORMER YARD
-CAT TOWER
OVERHEAD ABANDONED
WOOD RAILOAD TRESTLE
NDER BLOCK BUILDING
TRANSFORMER YARD
REMEDIATED BASEMENT
OIL-STAINED AREA.
MARIN 1998
COS COB
POWER HOU
LIGHT 4 POWER
SUBSTATION
CRUSHED STONE
SLAG ASH
TRANSFORMER
YARD
DOCK AREA
TANK
,50
.30
.00
,14
OVERHEAD COAL
CONVERTER BELT
STONE RETAINING WALL
SEE FIGURE 14
0,50 U
AT TOWER
FORMER TRANSFORMER/
BREAKER AREA. MARIN 1997
CONCRETE
SEA WALL
LEfiEHE
0 TO 1 FOOT BELOW GROUND SURFACE
1 TO 2 FEET BELOW GROUND SURFACE
2 TO 3 FEET BELOW GROUND SURFACE
3 TO « FEET BELOW GROUND SURFACE
TOTAL POLYCHLORINATED BIPHENYLS (PCS)
CREATES THAN 1.0 mgAg
FOR GWO SECTOR
GRID NOT SAMPLED DUE TO THE
PRESENCE Or DEBRIS PILES. STEEP
SLOPES. HEAVY VEGETATION. OR
THE LACK OF POTENTIAL PCS SOURCES.
SAMPLES WERE COLLECTED FROM t-FOOT INTERVALS
USING A GEOPROBE. THESE SAMPLES WERE HOMOGENIZED
AND THEN ANALYZED BY THE EPA REGION I LABORATORY IN
THE FIELD. SAMPLES KITH DETECTED CONCENTRATIONS Of PC8s
WERE THEN SENT FOR CONFIRMATION ANALYSIS.
THE STATE OF CONNECTICUT DEPARTMENT OF ENVIRONMENTAL
PROTECTION (CTDEP) RESIDENTIAL DIRECT EXPOSURE CRITERION (1996)
FOR TOTAL PCBt IS 1.0 mg/H SPECIAL RESTRICTIONS CAN APPLY
WHEN RESULTS EXCEED TWO TIMES THE DIRECT EXPOSURE
CRITERIA SPECIFIED BY CTDEP.
COS COB HARBOR
1 2 3 4 5 6 7 8 9 10 11 12 13 14
FORMER COS COB POWER PLANT
GREENWICH. CONNECTICUT
FIGURE 13
PCB RESULTS SUMMARY:
FIELD AMD FIXED LABORATORY RESULTS
EPA REGION I
IN COOPERATION WITH
BROWNFIELDS TECHNOLOGY SUPPORT CENTER
-------
R:\Cli«nta\COS COB\ Figure 14_to1ol PCB a«a«d zoom.d*g 0+/02/2004 Vsncant.Covonough DN
10
DETAIL BELOW
O
LEGEND
PREVIOUSLY COLLECTED SAMPLE
PROPOSED NEW CONFIRMATION SAMPLE
0 TO 1 FOOT BELOW GROUND SURFACE
1 TO 2 FEET BELOW GROUND SURFACE
2 TO 3 FEET BELOW GROUND SURFACE
3 TO 4 FEET BELOW GROUND SURFACE
TOTAL POLYCHLORINATED BIPHENYLS (PCB)
GREATER THAN 1.0 mg/kg
FOR GRID SECTOR
NOT SAMPLED DUE TO THE
PRESENCE OF DEBRIS PILES, STEEP
SLOPES, HEAVY VEGETATION. OR
THE LACK OF POTENTIAL PCB SOURCES.
FORMER COS COB POWER PLANT
GREENWICH, CONNECTICUT
FIGURE 14
PCB RESULTS SUMMARY:
FIELD AND FIXED LABORATORY RESULTS,
CAT TOWER AREA
U.S. EPA REGION I
IN COOPERATION WITH
BROWNFIELDS TECHNOLOGY SUPPORT CENTER
-------
ENCLOSURE 1
SOME METHOD VALIDATION ISSUES FOR THE RCRA PROGRAM:
THE FORMAL VALIDATION PROCESS FOR NEW METHODS DEVELOPMENT AND
DEMONSTRATION OF METHOD APPLICABILITY FOR A SPECIFIC PROJECT
-------
-------
Reference for this published article:
Lesnik, B. "Some Method Validation Issues for the RCRA Program", LC-GC, pp. 1048-1056.
October, 2000.
BLWP353.00
SOME METHOD VALIDATION ISSUES FOR THE RCRA PROGRAM:
THE FORMAL VALIDATION PROCESS FOR NEW METHODS DEVELOPMENT
AND DEMONSTRATION OF METHOD APPLICABILITY
FOR A SPECIFIC PROJECT
Introduction
I would like to thank the editors of LCGC for giving me the opportunity to discuss some
important issues regarding method development and use in the U.S. EPA's hazardous waste
program under the Resource Conservation and Recovery Act (RCRA), administered by the
Office of Solid Waste (OSW).
The term "validation" when applied to methods or data use has a plethora of different
meanings to different people. Thus, it can be a major source of confusion to laboratory analysts,
quality assurance officers, and data users alike, since they tend to have very different
interpretations of what it means. In this column, I will discuss two issues involving method
"validation", one for new methods developers and one for methods users. I will only use the
term "method validation" in the context of new methods development, and will use the term
"demonstration of method applicability" when discussing issues concerning the end use of
methods for generating effective environmental data, i.e., data that has been documented to be of
appropriate quality to be used in making environmental decisions. I will not touch the term "data
validation".
Validation of New Methods for Inclusion in SW-846
Test Methods for Evaluating Solid Waste (1), or SW-846, is the compendium of
analytical and test methods published by EPA's Office of Solid Waste (OSW) for use in
determining regulatory compliance under the Resource Conservation and Recovery Act (RCRA).
SW-846 functions primarily as a guidance document setting forth acceptable, although not
required, methods to be implemented by the user, as appropriate, in responding to RCRA-related
sampling and analysis requirements. SW-846 methods have a great deal of built-in flexibility as
explained in the Preface and Overview, Disclaimer and Chapter Two of the manual. However,
whether SW-846 methods or alternative methods are used for a RCRA analytical application, the
user must demonstrate the applicability of the methods selected for that application. Exceptions
to this requirement are the "method-defined parameters", e.g., Method 1311-Toxicity
Characteristic Leaching Procedure and Method 9095-Paint Filter Test, which must be performed
as written, since modifying these methods changes the regulation for which the determination is
Enclosure 1-1
-------
performed.
There seems to be an impression among methods developers and the regulated
community that there is some esoteric or mystical process that must be followed in order to get
an analytical method "approved" by regulatory agencies like the USEPA. In two guidance
documents (References 2-3), OSW is attempting to dispel these misconceptions, identify'some
basic principles, and present a logical approach to methods development that is currently
followed by OSW in developing methods for SW-846. This approach is based on sound
scientific principles, and methods developed according to this process should be acceptable for
use in other Agency programs as well as OSW.
In this column, I will discuss the two levels of validation for methods development that
are covered in Reference 2, initial "proof of concept" and a formal validation, either single
laboratory or multilaboratory. The key elements of this guidance are applicable to both new
methods submitted for potential inclusion in SW-846 or for a demonstration of applicability
using either existing SW-846 or alternative methods.
Levels of Validation
The RCRA Program looks at two "Levels of Validation" for new methods, proof of
concept validation and a formal validation process utilizing either a single laboratory or a
multiple laboratory study. Proof of concept is the preliminary stage of method development and
validation where a new method or concept is tested for its potential applicability for use in the
RCRA Methods Program and generally address elements 1 through 6 or 7 in Table 1. A method
taken through the proof of concept stage has only utilized spiked clean matrices to determine its
potential limits of performance as to scope and application, sensitivity, bias and precision,
repeatability, and limited optimization and interference testing. It has not been subjected to the
potential matrix effects and interferences that could be encountered in real world samples.
However, if a method performs poorly at the proof of concept stage, there is little point in
continuing its development, without significant modifications, if at all.
However, if a method performs well during the proof of concept stage, and the developer
wishes to submit it for inclusion in SW-846, the developer must perform a formal validation
addressing all of the elements in Table 1. The formal validation process includes a
multilaboratory study on real world samples at multiple concentrations based on the proposed
scope and application of the method. A successful multilaboratory study is important in that it
demonstrates that operators other than the developer can run the method, which is a critical
factor for a method to be included in a national methods manual.
Basic Principles
The RCRA method development approach utilizes three basic principles for either
demonstrating "proof of concept" (See elements 1 to 6 in Table 1) or for use in a formal
validation. These basic scientific principles are:
Enclosure 1-2
-------
1) Identify the scope and application of the proposed method, (What is this method
supposed to accomplish?)
2) Develop a procedure that will generate data that are consistent with the intended
scope and application of the method, and
3) Establish appropriate quality control procedures which will ensure that when the
proposed procedure is followed, the method will generate the appropriate data
from Step 2 that will meet the criteria established in Step 1.
A developer must also meet two other specific criteria before a method will be
considered for inclusion in SW-846: 1) Is there either an existing or anticipated RCRA
regulatory need for this method; and 2) Is it significantly different in principle or approach from
existing SW-846 methods?
Key Elements
OSW has identified eleven key elements essential to a sound method development and
validation effort. These are listed in Table 1 and should be addressed in the Method.
Development Study Plan.
1. Identification of Scope and Application and Regulatory Need
The key factor that a developer must establish before proceeding with a method
development project is a clearly defined scope and application for the proposed method. Factors
to be considered should include type of method, (i.e., screening or assay), applicable target
analytes, appropriate matrices, sensitivity, bias and precision, availability of equipment, and cost.
All of these considerations need to be written into a flexible study plan mat outlines the activities
necessary for a successful method development project. When establishing the scope and
application for a potential new method, it is also essential that the method developer identify that
there is a regulatory need for the method, either current or anticipated by the Program Office
involved. A method may be scientifically elegant, but it has very little value if there is no
application for its use in the EPA Program for which it is intended.
2. Quality Control Requirements
When developing a method, the developer needs to identify the appropriate quality
control procedures that must be performed to unequivocally demonstrate that the data generated
by the method will meet the objectives defined in the scope and intended application(s).
Examples of QC factors include appropriate calibration or tuning criteria, the need for replicate
analyses, appropriate surrogates, blanks and spikes. QC criteria specific to the particular method
should be well-documented and included in the QC section of the method as well as the method
development project report. General and technique-specific QC requirements should also be
Enclosure 1-3
-------
included in the study plan for the method development project during the planning stage and in
the project report.
3. Analytical Approach
In developing an analytical approach, the developer should keep in mind that the ultimate
goal of the project is to develop a method that will be published for general use by the analytical
community. Therefore, any analytical instrumentation or equipment used in the development of
the method needs to be commercially available to potential users at the time of the publication of
the method. OSW encourages the use of either conventional or innovative technology, provided
that it is demonstrated to be appropriate for the intended method application and provides data of
sufficient quality to satisfy the criteria delineated in the scope of the method. In order for a
method to be considered for inclusion is SW-846, it must be practical, i.e., has the potential for
general use in the environmental analytical community, address a RCRA regulatory or
monitoring need, and be significantly different from existing SW-846 methods.
4. Method/Instrument Sensitivity (Clean Matrix or Known Standard)
The method sensitivity for a proposed new method is influenced by several factors and
should be determined in a clean matrix or with a known standard to explore the limits of
sensitivity. These factors include the instrument detection limits, method quantitation limits, and
the analytical requirements for the proposed applications. RCRA regulations basically require
that an analyst demonstrate the ability to measure the analytes of concern in the matrices of
concern at the regulatory levels. Therefore, a method must exhibit analytical sensitivity
appropriate for its intended application, as delineated in the scope of the method, before it will be
considered for inclusion in SW-846. Many applications in the RCRA and other EPA programs
do not require the use of methods at their extreme limits of instrument or method sensitivity.
Method sensitivity is an important parameter in determining method applicability and will be
discussed in greater detail in that section of the column.
5. Method Optimization and Ruggedness Testing
After determining that the chosen analytical approach should work for its intended
application with appropriate sensitivity, the method developer should begin to optimize the
method and determine whether it possesses sufficient ruggedness to be considered for inclusion
in SW-846. This is also accomplished using known standards.
The initial parameters should be chosen according to the analyst's best judgment. These
are varied systematically (usually using Youden pairs as described in Reference 4) to obtain the
greatest response, least interference, greatest repeatability, etc. Developers must determine those
variables which should not be changed without adversely affecting method performance.
Potential operator-sensitive steps, e.g., color development time in colorimetric methods or other
timed reactions, also need to be identified at this stage.
Enclosure 1-4
-------
6. Accuracy (Bias), Precision, Repeatability (or Long-term Precision) in a Clean Matrix
Accuracy, which in most cases is measured as method bias, is defined as nearness to the
true value. Precision is defined as the dispersion of results around the mean value. Repeatability
(or long-term precision) is defined as the ability to reproduce a measurement from one week to
the next.
Bias is measured by determination of % recovery of target analytes spiked into the matrix
of concern. An acceptable spike recovery range for most method development applications is
from 80% to 120%. Precision is measured as relative % difference of target analyte
concentration(s) between duplicates or duplicate spikes, and should usually be <20%.
Repeatability is measured as long-term, e.g., weekly, precision, when the instrument is calibrated
using comparable standards, and on a different day should not vary by more than 15%.
These are key method performance factors which determine how a method can be used in
real world situations. The initial determination of bias, precision and repeatability should be
made in a spiked clean matrix which is similar to a real environmental matrix, but free from
interferences, e.g., reagent water, sand, soil (for inorganic methods), vermiculite or ash. These
values should be obtained using multiple replicates at both high and low spike concentrations.
7. Effect of Interferences
The determination of method interferences, both positive and negative is a key factor in
method development. It is a critical for the methods developer to determine the effects of
potential analytical interferences and to develop techniques to minimize or eliminate these
interferences. In chromatographic methods, interferences include coeluting peaks and/or analyte
degradation due to interaction with either the injector port, transfer line or column. In
spectroscopic methods, interferences can result from overlapping spectral lines causing either
positive or negative signal enhancement. In immunoassay methods, interferences include cross-
reacting compounds.
Method interferences should be determined in a spiked clean matrix. Developers should
determine the effects of interferences in a potential new method between target analytes and
other compounds reasonably expected to be present in waste matrices.
False negative rates, i.e., the percentage that a method generates a negative result when
the sample contains the target analytes at or above the action level and false positive rates, i.e.,
the percentage that a method generates a positive result when the sample contains the target
analytes below the action level, are critical factors which will determine the utility of a potential
method for its intended application.
Documentation of interferences should include any coelution of target analytes, any
enhancement or suppression of target analyte signals caused by interferences, any necessary or
optional cleanup procedures to minimize the effect of interferences, and any matrix-specific
Enclosure 1-5
-------
difficulties.
8. Matrix Suitability
t
The previous elements of the methods development process involved the use of either
known standards or target analytes spiked into clean matrices, designed to indicate potential
method performance in real RCRA matrices. Once the potential new method has passed.all of
the preliminary tests, it is now ready for the most important demonstration in the entire methods
development process, i.e., how it will perform in the real world matrices for which it is intended
to be used.
The method should be suitable for a variety of matrix types. Therefore, the developer
should choose appropriate RCRA matrices for the demonstration of method performance. By ,
matrix types, OSW refers to different matrices within a particular medium, e.g. water, soil and
ash. Appropriate RCRA water matrices include groundwater, TCLP leachate and wastewater,
while appropriate RCRA soil matrices include sand, loam and clay. Appropriate ash matrices
include bottom ash, fly ash and/or combined ash. The method should perform adequately in a
variety of spiked matrices and then in a variety of well-characterized natural samples or standard
reference materials (SRMs) when SRMs are available. Performance data including matrix type,
precision, bias, quantitation limits (see next section), and any other pertinent data should be
documented in appropriate tables. A summary of the single-laboratory performance data should
be included in tabular format in the method while the detailed performance data, including QC,
should be included in the supporting documentation.
9. Method Detection and Quantitation Limits
In a new method submission, OSW is most concerned about the performance of the
method in the RCRA matrices of concern. The developer should generate method quantitation'
limits (for assay methods) and method detection limits (for screening methods) for the analytes
of concern in the matrices of concern following the guidelines established in Chapter One of
SW-846 or other appropriate guidance. The practice of generating method detection limits
based on reagent water is not usually a very useful parameter for most RCRA methods.
Method detection and quantitation limits are usually based on a specific sample size. The
limits determined in clean matrices usually indicate the limits of the acceptable performance for
the method. Matrix effects may affect the achievable quantitation limits on real world samples.
However, the method quantitation limits for the target analytes in the target matrices must meet
the analytical requirements for its intended application, as defined in the scope of the method,
before it can be considered for inclusion in SW-846.
10. Laboratory Reproducibility (Multiple Operators and Multiple Laboratories)
The final stage in the method development process, prior to the submission of the method
for Agency review, is the determination of laboratory reproducibility. By reproducibility, OSW
Enclosure 1-6
-------
means that multiple operators and multiple laboratories should be able to obtain comparable
performance data on split real world environmental samples using the method. If certified SRMs
or secondary standards are available, they are the preferred samples for use hi the
multilaboratory validation study.
Since all of the previous elements involved single operators or single laboratories, it is
necessary to demonstrate that satisfactory method performance is not limited to the individual
operator or laboratory that developed the method. The minimum number of laboratories that are
needed to participate in a multilaboratory method validation is three, with preferably more.
Developers of new methods today need to do a limited multilaboratory evaluation and provide
the individual laboratory and summary performance data in the method submission. Developers
are encouraged to consult with the appropriate regulatory agency when planning a
multilaboratory study.
In order to minimize the number of variables involved in method validation, the
developer needs to follow a few simple guidelines to demonstrate appropriate multilaboratory
method performance. When validating a sample preparation method, the participating
laboratories should only perform the sample preparation procedure. The collected samples
should then be sent to one laboratory for analysis. The analysis should be done by a single
operator on a single instrument in a single batch to minimize analytical variability inherent to the
determinative method. Conversely, if a determinative method is to be validated, the .developer
should have a single operator perform all of the sample preparation operations in order to
minimize operator and laboratory variability inherent to the sample preparative procedures. The
sample extracts should then be split and sent to the laboratories participating in the validation
study for the analytical determination.
11. Document Submission and Workgroup Evaluation
i
When the method project is completed, the developer must assemble a package of
documents describing the project, and submit it to the Agency for review and evaluation. This
documentation package should include 1) draft copies, both hard and electronic, of the method in
an appropriate Agency format; 2) a supporting document describing the rationale behind the
methods development effort and how the key elements of the methods development project as
described in this document were addressed; 3) a data package containing both the raw and
summarized single laboratory and multilaboratory data; 4) any specific equipment diagrams and
chromatograms, spectra, etc. pertinent to the demonstration of appropriate performance for the
intended application of the method; 5) copies of any references listed in the method; and 6) any
method-specific quality control criteria.
The OSW Methods Team reviews the methods submission package for completeness and
quality, and then decides whether the method is ready to be sent to the appropriate SW-846
Methods Workgroup for review or back to the developer for additional work. OSW has several
standing SW-846 Methods Workgroups which meet formally each year to review the methods
packages submitted by developers for potential inclusion in SW-846. Workgroup members are
scientists from across the Agency who are qualified experts to evaluate the procedures and
Enclosure 1-7
-------
performance data submitted by the methods developers. Key evaluation criteria used by the
Workgroups to determine whether a submitted methods package is appropriate for inclusion in
SW-846 include: 1) Does the data package support the application and performance criteria
delineated in the scope and application of the method?; 2) Can the method be performed
routinely by the personnel available to an environmental laboratory?; 3) Is the equipment
commercially available?; 4) Is the method cost effective?
Demonstration of Method Applicability
Having discussed method validation for new methods developers in significant detail, I
will now address the primary validation issue for methods users, "demonstration of method
applicability". The RCRA Program does not have "reference methods" in its methods manual,
i.e., SW-846 methods are either method-defined parameters or guidance. Reference methods
have been taken to mean that, if one follows a published method as written and generates the
wrong data for an application, then nothing else needs to be done. For RCRA applications, the
regulated party must demonstrate that the methods used, whether they are published in SW-846
or alternative methods, must be appropriate to generate effective environmental data for that
application, i.e., data that has been documented to be of appropriate quality to be used in making
that environmental decision. Therefore, OSW allows sufficient flexibility in method selection
, and modification to be able to meet this requirement using the performance-based measurement
system (PBMS) approach. In a Federal Register Notice, EPA defines PBMS (Reference 5) as a
set of processes wherein the data quality needs, mandates or limitations of a program or project
are specified, and serve as criteria for selecting appropriate methods to meet those needs in a
cost-effective manner.
Using the PBMS approach, the operator must use some form of systematic planning to
establish the goals and data quality needs for the particular project and be able to answer the
following key questions which will help determine the appropriate methods to be used:
1) What is the purpose of this analysis? (Why are we doing this?)
2) How will the data be used? (What decisions will it support?)
3) How good does the data have to be, or what quality of data do we need to support
the decision?
Two separate factors are involved in demonstrating method applicability: 1)
demonstrating that the operator can perform the method properly in a clean matrix, and 2)
demonstrating that the method selected generates "effective data" in the matrix of concern. Item
1) is a laboratory or operator training or proficiency issue that I will not discuss in great detail
here. Item 2) involves the same basic scientific method questions as described above for method
development activities, and this section will focus on the key factors to be considered when
performing this demonstration.
Enclosure 1-8
-------
Operator Proficiency
Demonstration of operator proficiency in running the method(s) in question in a clean
matrix is important for two primary reasons. The first is a demonstration that the operator can
adequately perform the method and the second is that the method performs well hi a clean matrix
and therfore the analytical system is in control. Therefore, any analytical problems encountered
in the real world applicability testing can be addressed in terms of matrix suitability issues,
rather than operator or issues.
Determination of Method Applicability
Documentation of the determination of method applicability should be included in the
project planning documents including performance data from operator proficiency and matrix
suitability testing. Some of the key factors to consider, based on the data quality requirements
from the project planning documents, include selection of appropriate target analytes, analytical
matrix, method sensitivity, bias and precision, and reproducibility. It is usually not necessary to
perform a complete formal new methods validation project for this method user validation
process. However, the analyst must address many of the same key elements, but usually the
demonstration of applicability may require up to the proof of concept stage, depending on the
application. In some cases it may be a little more and in others a little less. Since we are dealing •
primarily with site-specific applications, there is little need for doing a multilaboratory validation
study.
In selecting appropriate target analytes, OSW only requires the analysis of target analytes
"that are reasonably expected to be present" at a site, and allows for the elimination of many
analytes on a long list through process knowledge. If the application is for a site survey, a wider
range of target analytes may be necessary than for a monitoring application for a well
characterized site, e.g., ground water monitoring wells around a landfill. It is frequently
appropriate for monitoring activities to use less expensive methods utilizing non-specific
detectors, e.g., GC/PID, for a few target analytes, and treating all positives as hits, thus
eliminating the need for confirmatory analyses. In some cases appropriate quantitative screening
methods, e.g., immunoassays may be used for very selective monitoring at appropriate action
levels for cleanup activities.
Demonstration of appropriate method sensitivity in the matrix of concern is a critical
element for the applicability demonstration. The RCRA Program does not require the formal
method detection limit (MDL) studies required by other EPA Offices, since many applications
operate at orders of magnitude above the quantitation or detection limits of the methods used. It
is necessary that the low calibration standard be selected to demonstrate that the method can
unequivocally determine whether the analytes of concern are present at the action level. For
analyses in the ppm range, a low standard of about 50% of the action limit is appropriate, while
for analyses in the ppb range, we recommend a factor of 0.1 times the action level for the low
standard. The analyst should select an appropriate calibration range to address the specific data
quality needs of the project. It need not be more than one or two orders of magnitude in many
Enclosure 1-9
-------
cases. The last calibration issue that I will mention is that for most RCRA applications, it is only
necessary to meet calibration requirements for the analytes of concern to the project, and not for
all of the analytes in the Table of Analytes list of a method. Another suggestion for
demonstrating appropriate method sensitivity is to spike the target analytes of concern into an
appropriate sample matrix free of target analytes at 70-80% of the action level. Run the spiked
sample through the analytical sequence and if recoveries of the target analytes are sufficient for
quantitation above the low standard, then method performance should be sufficiently sensitive
for its intended application.
Bias and precision are also important project-specific parameters. Bias is measured using
spike recovery of target analytes from the matrices of concern and precision from the analysis of
either duplicate or matrix spike duplicate (if no target analytes are present) samples. Poor
recoveries of target analytes generally indicate inappropriate sample preparatory conditions or
method selection, interferences, or interaction with the matrix. The analyst should identify the
problem and take steps to eliminate it. Poor precision at this point, since operator error has been
eliminated, generally indicates sample heterogeneity and needs to be addressed as a sampling or
sample handling issue. For projects involving highly contaminated sites, optimizing analyte
recoveries is much less important than for projects where the objective is a site cleanup to a low
action level. Sample reproducibility between operators and on repeated analyses of the same
sample should be within the specific limits of the project plan.
The number of samples necessary for the demonstration of applicability will vary with
sample type and proximity of the samples to the action level. Demonstrations for samples that
have concentrations of target analytes that are well below action levels can be completed with
relatively few samples. A demonstration for highly heterogeneous samples and samples with
target analytes very close to the action level will need many more analyses to complete the
demonstration and should be statistically calculated.
Summary
In this column, I have attempted to clarify the "validation" issues common to both new
methods developers and methods users. A full validation of a new method for inclusion in a
national methods manual involves many more steps, including a multilaboratory collaborative
study, than does an on-site "demonstration of method applicability". However the key elements
of scope and application, method sensitivity, bias and precision, reproducibility, and elimination
of interferences are common to both activities.
Enclosure 1-10
-------
Table 1: Key Elements for Regulatory Methods Development
Element 1: Identification of Scope and Application and Regulatory Need
Element 2: QA/QC Requirements
Element 3: Analytical Approach
Element 4: Method/Instrument Sensitivity (Clean Matrix)
Element 5: Method Optimization and Ruggedness Testing
Element 6: Accuracy, Precision and Repeatability (Clean Matrix)
Element 7: Effect of Interferences
Element 8: Matrix Suitability
Element 9: Quantitation and Detection Limits
Element 10: Laboratory Reproducibility (Multiple Operators and Multiple
Laboratories)
Element 11: Document Submission and Workgroup Evaluation
References
1. Test Methods for Evaluating Solid Waste, Physical/Chemical Methods, Third Edition,
(SW-846) including Updates, U.S. Environmental Protection Agency, Office of Solid
Waste and Emergency Response, Washington, DC, (1997).
2. OSW Guidance Document: "Guidance for Methods Development and Methods
Validation for the RCRA Program", October, 1995, http://www.epa.gov/SW-
846/methdev.pdf.
3. OSW Guidance Document: "Immunoassay Methods in SW-846: Recommended Format
and Content for Documentation Supporting New Submittals, July, 1995,
http://www.epa.gov/SW-846/immunoas.pdf.
4. W. J. Youden and E. H. Steiner (1975) Statistical Manual of the AOAC, Association of
Official Analytical Chemists, 1111 North 19th Street, Suite 210, Arlington, VA 22209.
Fifth printing in 1987.
5. 62 FR 52098, October 6, 1997.
Enclosure 1-11
-------
-------
ENCLOSURE 2
STATISTICAL PLOTS AND SUMMARY
STATISTICS FOR PROJECT TARGET ANALYTES
-------
-------
Summary Statistics
Cos Cob Power Plant
Greenwich Connecticut
Arsenic
601 OB
112
112
100
mg/Kg
23.2
9.80
10.8
0.75
152
34.3
29.7
Total Carcinogenic PAHs
8270C SIM
23
23
100
mg/Kg
2.42
2.37
1.73
0.225
5.89
1.69
3.15
Total PAHs
Field Test Kit
93
90
97
mg/Kg
180
61.8
64.5
0.1
1,860
269
235
TPH
8015
17
17
100
mg/Kg
773
411
425
13.8
3,040
802
1,190
TPH
Field Test Kit
93
92
99
mg/Kg
783
274
280
0.2
7,640
1,130
1,020
Aroclor 1260
Region 1 Mobile Lab
103
30
29
mg/Kg
1.35
0.25
0.40
0.14
56
5.72
2.47
All Aroclors
Region 1 Fixed Lab
135
13
10
mg/Kg
1.52
0.05
0.18
0.04
52
5.11
2.39
Enclosure 2-1
-------
-3
-20
Normal Probability Plot
Arsenic
Fixed Laboratory Results on a Dry Weight Basis
Arsenic
Fixed Laboratory Results on a Dry Weight Basis
Lognormal Probability Plot
Arsenic
Fixed Laboratory Results on a Dry Weight Basis
Box & Whisker Plot
Arsenic
Fixed Laboratory Results on a Dry Weight Basis
I1
•a
,3 -1
20 40 60 80 100 120 140 160
Concentration in mg/kg
Diy Weight
Histogram Untransfonned Data
Arsenic
Fixed Laboratory Results on a Dry Weight Basis
K-S d=.26232, p<.01 ; Lilliefors p<.01
Shapiro-Wilk W=. 62545, p=.00000
-0.4 0.0 0.4 0.8 1.2 1.6 2.0 2.4
-0.2 0.2 0.6 1.0 1,4 1.8 2.2
Natural Logarithm of the Concentration in mg/kg
Dry Weight
Histogram Logtransformed Data
Arsenic
Fixed Laboratory Results on a Dry Weight Basis
K-S d=. 12765, p<. 10; Lilliefors p< 01
Shapiro-Wilk W=.95317, p=.00062
100
80
60
40
20
0
-20
-40
-60
-20
1.0
1.5
2.0
2.5
Concentration in mg/kg
Dry Weight
Natural Logarithm of Ac Concentration in mg/kg
Dry Weight
a
Mean
MetmtSD
Mean±1.96*SD
Untransfonned Data
Logtransformed Data
-------
2.5
2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
-2.5
Normal Probability Plot
Total Carcinogenic PAHs
Fixed Laboratory Results on a Wet Weight Basis
lotai carcinogenic
Fixed Laboratory Results on a Wet Weight Basis
Lognomul Probability Plot
Total Carcinogenic PAHs
Fixed Laboratory Results on a Wet Weight Basis
Box & Whisker Plot
Total Carcinogenic PAHs
Fixed Laboratory Results on a Wet Weight Basis
-1000 0 1000 2000 3000 4000 5000 6000 7000
Concentration in ug/kg
Wet Weight
Histogram Untransformed Data
Total Carcinogenic PAHs
Fixed Laboratory Results on a Wet Weight Basis
K-S d=.19171, p> .20; Lilliefors p<,05
Shapiro-Wilk W-.91370, p=.04890
-1000 0 1000 2000 3000 4000 5000 6000
Concentration in ug/kg
Wet Weight
• Mean
I I MeaniSD
~T Mean±1.96'SD
2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0
Natural Logarithm of the Concentration in Ug/kg Wet Weight
Histogram Logtransformed Data
Total Carcinogenic PAHs
Fixed Laboratory Results on a Wet Weight Basis
K-S d=. 19474, p> .20; Lilliefors p<05
Shapiro-Wilk W=.9040I, p=.03061
Untransformed Data
Logtransformed Data
3.0 3.2 3.4 3.6 3.8
Natural Logarithm of the Concentration in ug/kg
Wet Weight
Enclosure 2-3
-------
-3
-2E5
100
90
80
70
1 60
•S 50
1 40
Z
30
20
10
0
Nonnal Probability Plot
Total PAHs
Site Lab Field Test Kit Results on a Wet Weight Basis
Total PAHs
Site Lab Field Test Kit Results.on a Wet Weight Basis
Lognormal Probability Plot
Total PAHs
Site Lab Field Test Kit Results on a Wet Weight Basis
-5E5
SE5
1E6
1.5E6
2E6
Concentration in ug/kg
Wet Weight
Box & Whisker Plot
Total PAHs
Site Lab Field Test Kit Results on a Wet Weight Basis
2E5 6E5 1E6 1.4E6 1.8E6
0 4E5 8E5 1.2E6 1.6E6 2E6
Concentration in ug/kg
Wet Weight
Histogram Untransformed Data
Total PAHs
Site Lab Field Test Kit Results on a Wet Weight Basis
K-S d=.25179, p<.01; Lilliefors p<.01
Shapiro-Wilk W=. 63605, p=.00000
8ES
6E5
4E5
2E5
-2E5
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5
Natural Logarithm of the Concentration in ug/kg
Wet Weight
Histogram Logtransformed Data
Total PAHs
Site Lab Field Test Kit Results on a Wet Weight Basis
K-S d=. 12035, p<. 15 ; Lilliefors p<.01
Shapiro-Wi'lk W=. 91421, p=.00001
-4E5
I
• Mean
I I MeaniSD
~T Mean±1.96*SD
Untransformed Data
Logtransformed Data
Natural Logarithm of the Concentration in ug/kg
Wet Weight
-------
3.0
2.5
2.0
1.5
1.0
0.5
-0.5
-1.0
-1.5
-2.0
-500
11
10
9
8
1 1
I 6
•8
I 5
I «
3
2
1
0
Normal Probability Plot
TPH
Fixed Laboratory Results on a Wet Weight Basis
TPH
Fixed Laboratory Results on a Wet Weight Basis
Lognormal Probability Plot
TPH
Fixed Laboratory Results on a Wet Weight Basis
Box & Whisker Plot
TPH
Fixed Laboratory Results on a Wet Weight Basis
2.0
1.5
1.0
0.5
$
> 0.0
I -05
a -1.0
-1.5
-2.0
-2.5
0 500 1000 1500 2000 2500 3000 3500
Concentration hi mg/kg
Wet Weight
Histogram Untransformed Data
TPH
Fixed Laboratory Results on a Wet Weight Basis
K-S d=.25247, p<,20; Lilliefors p<.01
Shapiro-Wilk W=.81338, p=.00311
-500 0 500 1000 1500 2000 2500 3000 3500
Concentration in mg/kg
Wet Weight
-3.0
2500
2000
1500
1000
500
-500
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6
Natural Logarithm of the Concentration in mg/kg
Wet Weight
Histogram Logtrsnsfoimcd Data
TPH
Fixed Laboratory Results on a Wet Weight Basis
K-S d-. 10453, p> .20; Lilliefors p> .20
Shapiro-Wilk W=93739, p=.28852
-1000
• Mean
I I MeaittSD
~TMean±l.%*SD
Untransformed Results
Logtransformed Results
1.0
1.5
2.0
2.5
3.0
3.5
Natural Logarithm of the Concentration in mg/kg
Wet Weight
Enclosure 2-5
-------
Normal ProbabilityPlot
TPH
Site Lab Field Test Kit Results on a Wet Weight Basis
TPH
Site Lab Field Test Kit Results on a Wet Weight Basis
Lognormal Probability Plot
TPH
Site Lab Field Test Kit Results on a Wet Weight Basis
-3
-1000 0
I
-2
1000 2000 3000 4000 5000 6000 7000 8000
Concentration in mg/kg
Wet Weight
Histogram Untransformed Data
TPH
Site Lab Field Test Kit Results on a Wet Weight Basis
K-S d=.24850, p<.01; Lilliefors p<.01
Shapiro-Wilk W=. 65315, p=.00000
01234
Natural Logarithm of me Concentration in mg/kg
Wet Weight
Histogram Logtransformed Data
TPH
Site Lab Field Test Kit Results on a Wet Weight Basis
K-S d=.14405, p<.05 ; Lilliefors p<.01
Shapiro-Wilk W=.87399, p=.00000
Box & Whisker Plot
TPH
Site Lab Field Test Kit Results on a Wet Weight Basis
3500
3000
2500
2000
1500
1000
500
0
-500
-1000
-1500
-2000
• Mean
I lMean±SD
T~ Mean±l.96*SD
Untransfonned Results
Logtransfonned Results
-1000 0 1000 2000 3000 4000 5000 6000 7000 8000
Concentration in mg/kg
Wet Weight
Natural Logarithm of the Concentration m mg/kg
Wet Weight
-------
I 2
-10
120
100
80
J
§
40
20
Aroclor 1260
Mobile Laboratory Results on a Wet Weight Basis
Normal Probability Plot
Aroclor 1260
Mobile Laboratory Results on a Wet Weight Basis
10
20
30
40
50
Concentration in mg/kg
Wet Weight
Histogram Untransfonned Data
Aroclor 1260
Mobile Laboratory Results on a Wet Weight Basis
K-S d=.41599, p<.01 ; Lilltefors p<.01
Shapiro-Wilk W=. 18441, p=0.0000
-10
10 20 30 40
Concentration in mg/kg
50
60
Lognormal Probability Plot
Aroclor 1260
Mobile Laboratory Results on a Wet Weight Basis
Box & Whisker Plot
Aroclor 1260
Mobile Laboratory Results on a Wet Weight Basis
|
60
-2
-3
-1.0 -0.6 -0.2 0.2 0.6 1.0 1.4 1.8
-0.8 -0.4 0.0 0.4 0.8 1.2 1.6 2.0
Natural Logarithm of the
Concentration in mg/kg
Wet Weight
Histogram Logtransformed Data
Aroclor 1260
Mobile Laboratory Results on a Wet Weight Basis
K-S d=.40686, jkoi ; Lilliefors p<.01
Shapiro-Wilk W=.56713, p= .00000
14
12
10
8
6
4
2
0
-2
-4
-6
-8
-10
-12
90
80
70
J 60
I 50
•s
| 40
1.
20
10
0
-1.5 -1.0 -0.5
0.5 1.0 1.5 2.0
Natural Logarithm of the
Concentration in mg/kg
Mean
MearrtSD
Mean±1.96*SD
UntnnisTonncd Data
Logtransformed Data
Enclosure 2-7
-------
6
5
4
> 3
I 2
0
-1
-2
160
140
120
•3. 100
C/l
•S 80
S
§ 60
Z
40
20
Normal Probability Plot
Aroclors 1016, 1221, 1232, 1242, 1248,1254,
1260, 1262, and1268
Fixed Laboratory Results on a Dry Weight Basts
L
-10 0 10 20 30 40 50
Concentration in mg/kg Dry Weight
Histogram Untransformed Data
Aroclors 1016, 1221,1232, 1242,1248,1254,
1260, 1262, andl268
K-S d=.38962, p<.01; Lilliefors p<.01
Shapiro-Wilk W=.28667, p=0.0000
Aroclors 1016,1221, 1232, 1242, 1248, 1254,
1260,1262, and!268
Fixed Laboratory Results on a Dry Weight Basis
60
-10 0 10 20 30 40 50
Concentration in mg/kg Dry Weight
60
3.0
2.5
2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
-2.0
Lognormal Probability Plot
Aroclors 1016, 1221,1232, 1242,1248,1254,
1260,1262, and1268
Fixed Laboratory Results on a Dry Weight Basis
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5
Natural Logarithm of the Concentration in
mg/kg Dry Weight
Histogram Logtransformed Data
Aroclors 1016,1221, 1232,1242,1248,1254,
1260, 1262, and!268
K-S d=.3257I, p<.01 ; Lilliefors p<.01
Shapiro-Wilk W-.76339, p=.00000
2.0
2
0
-2
-4
-6
-8
-10
-2.0 -1.5 -1.0 -0.5 0.0 0.5
1.5 2.0
Box & Whisker Plot
Aroclors 1016,1221,1232,1242,1248,1254,
1260,1262, and 1268
Fixed Laboratory Results on a Dry Weight Basis
• Mean
I I Mean±SD
T Mean±1.96*SD
U ntfan sformed Data
Logtransformed Data
Natural Logarithm of the Concentration
in mg/kg Dry Weight
-------
ENCLOSURE 3
SELECTED CHROMATOGRAMS FOR SAMPLES
USED TO DEVELOP FIELD-BASED ACTION LEVELS
-------
-------
3Bi
37i
36-|
35-E
34-1
33i
32-f
140-=
120=
DF12007.RAW
o
CO
•fr
Modified Method 8015
Sample A-30-1'
0302030-03
80-=
q- 2
DO23026.RAW
T
4
i I i i i i I i i i i | i i t i I i i i i | i i i i | i i i i | i i i i | i i i i | i i i i | i i i i | i i i i | i i i i | i i i i
8 10 12 14 16 . 18 20
LUBRICATING OIL STANDARD
I [ I I I I | I I I | [ |
24
i | | I I I I | I I I I | M I I | I I I I | I I I I [ I I I I | I I I I | I I I I | II I I | I I I I | I I I I | I I I I | I I I I
8 10 12 14 16 18 20
Enclosure 3-1
-------
04/24/2003 12:52 PAX 506 822 3286
WOODS HOLE GROUP
Quantisation Report
©002
Data Pile : C:\HPCHEM\1\DATA\FEB14\02030-OB.D
' Acq On : 15 Feb 2003 12:03 am
Sample : 0302030-08
MiBC : IX
MS Integration Parama: rteint.p
Quant Time: Feb 18 11:37 2003 Quant Results File: PAHTIS1.RES
Vial: 21
Operator: aja
Inst : HP 6890
Multiplr: 1.00
Method : 0;\ORGAMICS\METHODS\BNA2\2003\FRONT\PAHT1S1.M (RTE Integrate
Title : Semivolatiles by QC/M8
Last Update : Tue Feb IB 06:29:01 2003
Response via : Initial Calibration
Modified Method 8270C
Selective Ion Monitoring
Sample F-8 0-1 '
4.DO 5.00 6.00 7.00 6.00 9.00 10.00 11.00 IZ.'pO 13.00 14.00 ISiOO 16.00 17100 Iglob
02030-08.D PAHTIS1.M
Tue Peb 16 11:37:41 2003
Page 2
Enclosure 3-2
-------
42-
40-
38-
DF09012.RAW
34-
150-
100-
50-
Modified Method 8015
Sample 0-9 0-1'
1 23 4 5 67 8 9 10 11 12 13 14 15 16 17 18 19
. Enclosure 3-3
-------
HflZflRDOUS UflSTE Fax: 17812246548 flpp 28 2003 15:28
04/24/2003 12:82 PAX BOB 822 3288 WOODS HOLE GROUP
Quantitation Report
P. 05
Id 003
Data Pile : C:\HFCHEM\1\DATA\FBB14\02030-01.D
Aoq On : 14 Feb 2003 7:41 pro
Sample : 0302030-01
Misc : 2X
MS Integration Params: rteint.p
Vial; 12
Operator: aja
Inst : HP 6890
Multiplr: 1.00
Quant Time: Feb 18 10:27 2003
Quant Results File: PAHTIS1.RES
Method : O:\OROANICS\MBTHODS\BNA2\2003\FRONT\PAKTIS1.M (RTE Integrate
Title : Semivolatilee by GC/MS
Last Update : Tue Feb 18 06:29:01 2003
Response via ; Initial Calibration
--*---'
Modified Method 8270C
Selective Ion Monitoring
02030-01.D PAHTIS1.M
Tue Feb 18 10:31:50 2003
Page 2
-------
90-3
80-3
60-3
50-3
403
II
DD16019.RAW
0212046-01
Modified Method 8015
Sample T-l 0-2*
i i i i i i i i i i i i i i i i i i i i rrrrp~TTT]~TTTT| n i i | T T
2 46 8
io
12
14
16
18
20
yrm j
500-H
400H
300H
200-3
10H
SAE-30.RAW
o
CD
i.
Lubricating Oil Standard
JL
I I I I I I I I I I I I I I II I I I M I M M | I I I I | I I I I | I II I | I I I I | I ' I I | I I I I | I II
8 10 12 14 16 18
O)
I I | 1 I iT| 1 I I I | I I I I | I I I I | i I i i
2 4 6
Enclosure 3-5
-------
100^
40-
DD16016.RAW
o
CD
Modified Method 8015
Sample T-3 0-2'
0212048-02
500-H
400-
300-
200-
100-
fl
SAE-30.RAW
Si
10
12
14
16
18
20
Lubricating Oil Standard
OO
to
"I I I I [ I I I I | I I I I [ I I I I | I I I I | I I I I | I I I I I I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | M I I | I I I I | I I I I | I I I I | I I I I | I I I I |
2 46.8 10 12 . 14 - 16 18
I I M I'l I
-------
50-
45-
40-
35-
11
DD16017.RAW
o
CD
0212046-04
Modified Method 8015
Sample T-5 0-2'
I I I I | I I I I | I I I I | I I I I | I I I I I I I I II ITT TTI I I I | I I I I I
2 4 68
i | i i i i | i i i i | i i i i | i i i i | i i i i
10 12 14
16
i i | i
18
i i | i
20
500H
400-
300-
200-
100-1
K>
6
fl
SAE-30.RAW
o
CO
Lubricating Oil Standard
•' i i i | i i i i | i i i i | i
2
i i i i i i i i i i i i i i i i i i i r i | i i i t | i 'i i i | i i i i | i i i i | i
4 6 8 10 12
i i i i i i i i i i i i i i 'i i i i i i i i i i i r i i i i
14
16
18
Enclosure 3-7
-------
Office of Solid Waste
and Emergency Response
(5102G)
EPA-542-R-04-008
June 2004
www.epa.gov
www.cluin.org
Official Business
Penalty for Private Use $300
------- |