OPTIMIZATION SUPPORT EVALUATION
HAVERTOWN PCP SITE
HAVERTOWN, PENNSYLVANIA
Report of the Optimization Support Evaluation
Site Visit Conducted at the Havertown PCP Superfund Site
August 5, 2003
\
-------
This page is intentionally left blank.
-------
Office of Solid Waste EPA 542-R-04-033
and Emergency Response March 2004
(5102G) www.epa.gov/tio
clu-in.org/optimization
Optimization Support Evaluation
Havertown PCP Site
Havertown, Pennsylvania
-------
This page is intentionally left blank.
-------
NOTICE
Work described herein was performed by GeoTrans, Inc. (GeoTrans) for the U.S. Environmental
Protection Agency (U.S. EPA). Work conducted by GeoTrans, including preparation of this report, was
performed under S&K Technologies Prime Contract No. GS06T02BND0723 and under Dynamac
Corporation Prime Contract No. 68-C-02-092, Work Assignment ST-1-08. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
-------
EXECUTIVE SUMMARY
An Optimization Support Evaluation (OSE) involves a team of expert hydrogeologists and engineers,
independent of the site, conducting an evaluation of site operations. It is a broad evaluation that
considers the goals of the remedy, site conceptual model, above-ground and subsurface performance, and
site exit strategy. The evaluation includes reviewing site documents, visiting the site for up to 1.5 days,
and compiling a report that includes recommendations to improve the system. Recommendations with
cost and cost savings estimates are provided in the following four categories:
• improvements in remedy effectiveness
• reductions in operation and maintenance costs
• technical improvements
• gaining site closeout
The recommendations are intended to help the site team identify opportunities for improvements. In
many cases, further analysis of a recommendation, beyond that provided in this report, may be needed
prior to implementation of the recommendation. Note that the recommendations are based on an
independent evaluation and represent the opinions of the evaluation team. These recommendations do
not constitute requirements for future action, but rather are provided for the consideration of all site
stakeholders. This OSE report pertains to conditions that existed at the time of the OSE site visit, and
any site activities that have occurred subsequent to the OSE site visit are not reflected in this OSE report.
The Havertown PCP site is located in Havertown, Haverford Township, Delaware County, in
southeastern Pennsylvania. The site contamination was first discovered in 1962 when the Pennsylvania
State Department of Health became aware of contaminants in Naylors Run and attributed the source of
PCP and dioxin contamination to the disposal practices of a nearby wood treating facility. Investigation
and remedial activities have included the removal of existing waste from the site, treatment of effluent
discharging from the existing storm sewer, installation of an oil/water separator, off-site disposal of
generated wastes, installation of a cap over soils impacted by dioxin, and operation of an interim P&T
system, which began operation in 2001. This OSE focused on the operating interim P&T system but also
considers potential enhancements to this system for addressing deeper ground water contamination as
part of the final remedy for Operable Unit 3 (OU3).
The evaluation team observed a conscientious and knowledgeable site team (i.e., EPA and its contractor)
that has effectively managed this interim remedy, provided a number of much needed plant
improvements, and initiated the remedial investigation for OU3. The observations and recommendations
contained in this report are not intended to imply a deficiency in the work of either the system designers
or operators but are offered as constructive suggestions in the best interest of the EPA, the public, and the
facility. These recommendations have the obvious benefit of being formulated based upon operational
data unavailable to the original designers.
The recommendations to improve remedy effectiveness include the following:
• The site team should continue with their efforts to investigate and properly seal the abandoned
sewer line that has allowed contaminated ground water to migrate downgradient and then surface
as a seep in a residential area. The site team should make this effort and the remediation of the
associated soils and sediments their first priority.
-------
• Further delineation of the contaminant plume is suggested, especially to the south and at depth.
Locations and depths for seven new monitoring wells are proposed.
• Once the plume is delineated, a target capture zone should be determined and the site team
should conduct a thorough capture zone analysis. A preliminary analysis has been conducted as
part of this report and is provided in Section 4.2.2.
• Modifications should be made to O&M practices to reduce system downtime. Foremost among
the suggested items is to bypass one or two of the UV/Oxidation units.
Implementation of these recommendations might cost $375,000 to $405,000 in capital costs, but these
costs may be offset by savings associated with cost-reduction recommendations. Recommendations to
reduce cost include the following:
• Although the site team suggests bypassing one UV/Oxidation unit, process data suggests that two
UV/Oxidation units can be bypassed. As a further option, the site team could consider bypassing
all three UV/Oxidation units and relying on GAC as the primary treatment technology for
organic compounds. Annual savings of approximately $32,000 per year might be possible with
little or no capital costs if a second UV/Oxidation unit is bypassed. An additional $15,000 per
year might be saved if the UV/Oxidation system is bypassed altogether.
• The majority of the annual O&M costs are due to labor. Reductions in both project management
and sampling labor are likely possible, especially in the near future as contractor modifications
further simplify the operation of the system. Potential annual savings of $40,000 per year may be
feasible within one or two years.
Recommendations for technical improvement include modifying the system to improve treatment
capacity and making minor modifications to the equalization tanks. Recommendations for site closeout
include focusing the remedy on plume containment in the near term rather than aquifer restoration and
considering options for addressing contamination that may be outside of the current capture zone.
A table summarizing the recommendations, including estimated costs and/or savings associated with
those recommendations, is presented in Section 7.0 of this report.
11
-------
PREFACE
This report was prepared at the request of EPA Region 3 as part of a project to optimize the Region's
pump and treat (P&T) systems that are jointly funded by EPA and the associated State agency. The
effort was made possible with the help of the Office of Superfund Remediation and Technology
Innovation. The project contacts are as follows:
Organization
Key Contact
Contact Information
USEPA Office of Emergency and
Remedial Response
(OSRTI)
Kathy Yager
11 Technology Drive (ECA/OEME)
North Chelmsford, MA 01863
phone: 617-918-8362
fax: 617-918-8427
yager.kathleen@epa.gov
USEPA Region 3
Kathy Davies
USEPA REGION 3
1650 Arch Street
Philadelphia, PA 19103-2029
215-814-3315
davies.kathy@epa.gov
USEPA Region 3
Norm Kulujian
USEPA REGION 3
1650 Arch Street
Philadelphia, PA 19103-2029
215-814-3130
kulujian.norm@epa.gov
GeoTrans, Inc.
(Contractor to USEPA)
Doug Sutton
GeoTrans, Inc.
2 Paragon Way
Freehold, NJ 07728
(732) 409-0344
Fax: (732) 409-3020
dsutton@geotransinc.com
in
-------
TABLE OF CONTENTS
EXECUTIVE SUMMARY i
PREFACE iii
TABLE OF CONTENTS iv
1.0 INTRODUCTION 1
. 1 PURPOSE 1
.2 TEAM COMPOSITION 2
.3 DOCUMENTS REVIEWED 2
.4 PERSONS CONTACTED 2
.5 SITE LOCATION, HISTORY, AND CHARACTERISTICS 3
1.5.1 LOCATION AND BRIEF HISTORY 3
1.5.2 POTENTIAL SOURCES 4
1.5.3 HYDROGEOLOGIC SETTING 4
1.5.4 RECEPTORS 5
1.5.5 DESCRIPTION OF GROUND WATERPLUME 5
2.0 SYSTEM DESCRIPTION 7
2.1 SYSTEM OVERVIEW 7
2.2 EXTRACTION SYSTEM 7
2.3 TREATMENT SYSTEM 8
2.4 MONITORING PROGRAM 9
3.0 SYSTEM OBJECTIVES, PERFORMANCE AND CLOSURE CRITERIA 10
3.1 CURRENT SYSTEM OBJECTIVES AND CLOSURE CRITERIA 10
3.2 TREATMENT PLANT OPERATION STANDARDS 11
4.0 FINDINGS AND OBSERVATIONS FROM THE OSE SITE VISIT 12
4.1 FINDINGS 12
4.2 SUBSURFACE PERFORMANCE AND RESPONSE 12
4.2.1 WATER LEVELS 12
4.2.2 CAPTURE ZONES 12
4.2.3 CONTAMINANT LEVELS 14
4.3 COMPONENT PERFORMANCE 16
4.3.1 EXTRACTION SYSTEM WELLS, PUMPS, AND HEADER 16
4.3.2 OIL/WATER SEPARATOR AND EQUALIZATION TANK 16
4.3.3 METALS REMOVAL SYSTEM 16
4.3.4 UV/OX SYSTEM 16
4.3.5 PEROXIDE DESTRUCTION UNIT, CLAY FILTER, AND GAC 17
4.3.6 EFFLUENT TANK AND DISCHARGE 17
4.3.7 SLUDGE HANDLING TANKS AND EQUIPMENT 17
4.3.8 SYSTEM CONTROLS 18
4.4 COMPONENTS OR PROCESSES THAT ACCOUNT FOR MAJORITY OF ANNUAL COSTS 18
4.4.1 UTILITIES 18
4.4.2 NON-UTILITY CONSUMABLES 19
4.4.3 LABOR 19
4.4.4 CHEMICAL ANALYSIS 19
4.5 RECURRING PROBLEMS OR ISSUES 19
4.6 REGULATORY COMPLIANCE 20
iv
-------
4.7 TREATMENT PROCESS EXCURSIONS AND UPSETS, ACCIDENTAL CONTAMINANT/REAGENT RELEASES
20
4.8 SAFETY RECORD 20
5.0 EFFECTIVENESS OF THE SYSTEM TO PROTECT HUMAN HEALTH AND THE ENVIRONMENT . 21
5.1 GROUND WATER 21
5.2 SURFACE WATER 21
5.3 AIR 21
5.4 SOILS 22
5.5 WETLANDS AND SEDIMENTS 22
6.0 RECOMMENDATIONS 23
6.1 RECOMMENDATIONS TO IMPROVE EFFECTIVENESS 23
6.1.1 PROPERLY SEAL ABANDONED 12-INCH SEWER LINE 23
6.1.2 IMPROVE PLUME DELINEATION TO THE SOUTH AND VERTICALLY 24
6.1.3 EVALUATE PLUME CAPTURE ONCE PLUME is DELINEATED 25
6.1.4 TAKE MEASURES TO FURTHER REDUCE SYSTEM DOWNTIME 26
6.2 RECOMMENDATIONS TO REDUCE COSTS 26
6.2.1 USE FEWER UV/OXIDATION UNITS 26
6.2.2 EVALUATE AREAS TO REDUCE LABOR COSTS 28
6.3 MODIFICATIONS INTENDED FOR TECHNICAL IMPROVEMENT 28
6.3.1 CONTINUE IMPROVING TREATMENT PLANT TO FACILITATE OPERATION AND POTENTIALLY
INCREASE CAPACITY 28
6.3.2 MAKE PIPING CHANGES TO BETTER USE THE SECOND EQUALIZATION TANK 29
6.4 CONSIDERATIONS FOR GAINING SITE CLOSE OUT 29
6.4.1 ADAPT P&T SYSTEM TO Focus PRIMARILY ON COST-EFFECTIVE CONTAINMENT WITH
DECREASED EMPHASIS ON RESTORATION 29
6.4.2 POTENTIAL OPTIONS FOR IMPROVING CAPTURE 30
6.5 SUGGESTED APPROACH TO IMPLEMENTATION 31
7.0 SUMMARY 32
List of Tables
Table 7-1. Cost summary table
List of Figures
Figure 1-1. The Havertown PCP Site and the Surrounding Area
Figure 1-2. Area Immediately Surrounding the Former NWP Facility
Figure 1-3. Extent of PCP Contamination in the Shallow Zone (Approximately 5 to 20 feet bgs)
Figure 1-4. Extent of PCP Contamination in the Deep Zone (Approximately 35 to 60 feet bgs)
Figure 4-1. Potentiometric Surface for Shallow Ground Water Under Pumping Conditions (Approximately 5
to 20 feet bgs)
Figure 4-2. Potentiometric Surface for Deep Ground Water Under Pumping Conditions (Approximately 40 to
60 feet bgs)
-------
1.0 INTRODUCTION
1.1 PURPOSE
During fiscal years 2000, 2001, and 2002 Remediation System Evaluations (RSEs) were conducted at 24
Fund-lead pump and treat (P&T) sites (i.e., those sites with pump and treat systems funded and managed
by Superfund and the States). Due to the opportunities for system optimization that arose from those
RSEs, EPA Region 3 is expanding efforts to optimize its Fund-lead remedies. Region 3 requested that
GeoTrans conduct RSEs at two of its Fund-lead P&T systems: Havertown PCP and Greenwood
Chemical. Because GeoTrans has a business relationship with Tetra Tech, the contractor at these two
facilities, Optimization Support Evaluations (OSEs) were conducted in place of the RSEs. The OSE
process is identical to the OSE process, but the name change indicates the business relationship between
GeoTrans and Tetra Tech.
The Remediation System Evaluation (RSE) process (and therefore the OSE process) was developed by
the US Army Corps of Engineers (USAGE) and is documented on the following website:
http://www.environmental.usace.armv.mil/library/guide/rsechk/rsechk.html
The evaluation involves a team of expert hydrogeologists and engineers, independent of the site,
conducting a third-party evaluation of site operations. It is a broad evaluation that considers the goals of
the remedy, site conceptual model, above-ground and subsurface performance, and site exit strategy. The
evaluation includes reviewing site documents, visiting the site for 1 to 1.5 days, and compiling a report
that includes recommendations to improve the system. Recommendations with cost and cost savings
estimates are provided in the following four categories:
• improvements in remedy effectiveness
reductions in operation and maintenance costs
• technical improvements
gaining site closeout
The recommendations are intended to help the site team identify opportunities for improvements. In
many cases, further analysis of a recommendation, beyond that provided in this report, might be needed
prior to implementation of the recommendation. Note that the recommendations are based on an
independent evaluation and represent the opinions of the evaluation team. These recommendations do
not constitute requirements for future action, but rather are provided for the consideration of all site
stakeholders. This OSE report pertains to conditions that existed at the time of the OSE site visit, and
any site activities that have occurred subsequent to the OSE site visit are not reflected in this OSE report.
The Havertown PCP site was selected by EPA Region 3 based the potential to improve the effectiveness
of the remedy to protect human health and the environment and to reduce the annual costs of operating
the remedy. This report provides a brief background on the site and current operations, a summary of the
observations made during a site visit, and recommendations for changes and additional studies. The cost
impacts of the recommendations are also discussed.
-------
1.2
TEAM COMPOSITION
The team conducting the evaluation consisted of the following individuals:
Rob Greenwald, Hydrogeologist, GeoTrans, Inc.
Peter Rich, Civil and Environmental Engineer, GeoTrans, Inc.
Doug Sutton, Water Resources Engineer, GeoTrans, Inc.
The evaluation team was also accompanied by Kathy Davies and Norm Kulujian from EPA Region 3.
1.3
DOCUMENTS REVIEWED
Author
USEPA
USEPA
Calgon Carbon Corp.
Calgon Carbon Corp.
URS
Tetra Tech
Tetra Tech
Tetra Tech
Tetra Tech
Date
8/1989
9/30/1991
4/2001
4/2001
6/13/2001
10/2002-7/2003
8/5/2003
8/5/2003
8/8/2003
Title
Focused Feasibility Study
Operable Unit 2 Record of Decision
Operation and Maintenance Manual, Volume I:
RAY OX Advanced Oxidation System
Operating Manual: Mobil Model 6 Granular
Carbon Adsorption System
Preliminary Draft Operations and Maintenance
Manual: Havertown PCP Superfund Site
Project Meeting Agendas
NAPL Recovery Data
Process Monitoring Data
Supplementary Data and Figures
1.4
PERSONS CONTACTED
The following individuals associated with the site were present for the visit:
Jill Lowe, Remedial Project Manager, EPA Region 3
Mindi Snoparsky, Hydrogeologist, EPA Region 3
Harish Mital, Project Manager and Engineer, Tetra Tech
Andy Westerbaan, Hydrogeologist, Tetra Tech
J.B. Moore, Project Engineer, Tetra Tech
Dale Gant, Plant Operator, Tetra Tech
-------
1.5 SITE LOCATION, HISTORY, AND CHARACTERISTICS
1.5.1 LOCATION AND BRIEF HISTORY
The Havertown PCP site is located in Havertown, Haverford Township, Delaware County, in
southeastern Pennsylvania. The site is located approximately 10 miles west of Philadelphia and is
surrounded by a mixture of commercial establishments, industrial companies, parks, schools, and private
homes. The location of the site and the surrounding area are depicted in Figure 1-1. The entire
Havertown PCP site is approximately 12 to 15 acres, and is comprised of a former wood-treatment
facility operated by National Wood Preservers (NWP), the Philadelphia Chewing Gum Company (PCG)
manufacturing plant adjacent to the wood-treatment facility, Naylors Run (a creek that drains the area),
and neighboring residential and commercial properties. A more detailed plan of the site is shown in
Figure 1-2.
The NWP facility is the primary source of the contamination for the site. The NWP facility was first
developed as a railroad storage yard, and in 1947 it began operation as a wood-preserving facility. The
site contamination was first discovered in 1962 when the Pennsylvania State Department of Health
became aware of contaminants in Naylors Run and linked the source of contamination to the NWP waste
disposal practices. The following four media have been documented as being affected by contaminants
originating from the site:
on-site soils
on-site and downgradient ground water
surface water in storm sewers and off-site in Naylors Run
air on-site and in the vicinity of the catch basin on Naylors Run
Investigation and remedial activities have included the removal of existing waste from the site, capping
of impacted soil and debris on-site, treatment of effluent discharging from the existing storm sewer,
installation of an oil/water separator and off-site disposal of generated wastes. Other investigation and
remedial activities conducted at this site include the following:
1972 - The Pennsylvania Department of Environmental Resources (PADER) identified
contaminated ground water discharging from a storm sewer into Naylors Run
1976 - The EPA initiated activities and commenced containment operations under Section 311
of Clean Water Act in response to request from PADER
1982 - The site was placed on the National Priorities List
1982-84 - PADER and EPA conducted subsequent inspections and found many deficiencies with
the containment operations, and EPA issued a unilateral Administrative Order against
NWP which required NWP to perform various abatement activities
1987-89 - PADER performed the Remedial Investigation/Feasibility Study for Operable Unit One
(OU1), and the EPA issued an interim Record of Decision (ROD) for OU1 which
addressed the cleanup of wastes currently staged on the site from previous investigative
actions and included interim remedial measure of designing and installing an oil/water
separator
1988 - A catch basin was installed in Naylors Run to trap the discharge from the storm pipe by
EPA's Emergency Response Team to prevent the release of PCP-contaminated oil into
Naylors Run
-------
1991 - The EPA initiated the Remedial Investigation/Feasibility Study for Operable Unit Two
(OU2), and issued a ROD for OU2 that addressed the treatment of the shallow ground
water aquifer
1994 - A cap completed to cover soils impacted by dioxin
2001 - A P&T system was completed and began operation under Operable Unit 2 (OU2) as an
interim remedy for shallow ground water
The principal threat at the site remains the ground water contamination by PCP and other contaminants
of concern that occurred as a result of wood-treatment operations at NWP, including dioxins and
dibenzofurans, fuel oil and mineral spirits components, heavy metals, certain volatile organic
compounds, and phenols. This OSE report pertains to the ground water contamination, the operating
interim P&T system, potential alternative approaches to addressing the ground water contamination, and
site conditions that directly affect the ground water remediation.
1.5.2 POTENTIAL SOURCES
The use of an injection or disposal well to collect the spent wood-treating solutions containing PCP and
oil was allegedly the major source of contamination to the environment from 1947 until 1963 when
disposal practices at NWP changed. It is estimated that up to one million gallons of spent solution may
have been disposed of by this method. The uncontrolled disposal methods and typically poor
housekeeping practices prior to 1963 resulted in significant contamination of the ground water
surrounding the site. The presence of the light non-aqueous phase liquid (LNAPL) in wells R-2 and
HAV-02 continue to contribute to the dissolved ground water contamination. An estimated volume of
6,000 gallons of NAPL (approximately 2 inches thick) is present over the ground water in this area.
During the ongoing OU3 investigation (OU3 pertains to the remediation of the sediment contamination in
Naylors Run, the deep ground water aquifer which is fractured bedrock, and surface water and sediment
contamination due to runoff from on-site soils) an abandoned sanitary sewer line was found which may
act as a preferential pathway for contaminant transport to the Residential Open Space area southeast of
Achille Road on Naylors Run.
1.5.3 HYDROGEOLOGIC SETTING
The site is located in the Piedmont Upland section of the Piedmont Physiographic Province. Bedrock in
the site vicinity consists of metamorphic rocks of Wissahickon Formation. The site is in an area that is
relatively flat compared to the surrounding countryside. Much of the area's original topography has been
altered by cut and fill activities on both the NWP plant site and the PCG property. The site lies
approximately 300 ft above mean sea level (MSL). The depth to ground water beneath the site ranges
from approximately 23 ft below ground surface in the Young's Produce Store area to approximately 0.5
ft below ground surface in the Rittenhouse Circle area. Ground water in the vicinity of the site flows in
an easterly direction, with some unknown portion discharging to Naylors Run. For the most part, surface
runoff across the NWP site enters artificial drainage channels before discharging into Naylors Run.
According to the 1991 OU2 ROD, ground water at the site flows in an easterly direction and occurs in
two major zones. The upper zone consists of surficial soils and weathered schist saprolite. In general,
this upper zone extends to approximately 30 feet below ground surface. The lower zone consists of
highly fractured and jointed schist bedrock, with water movement occurring along interconnected
fractures. The bedrock aquifer receives some of its recharge from the downward flow through the
overburden aquifer, but upward flow from the bedrock to the overburden exists further downgradient,
-------
presumably as ground water discharges to Naylors Run. Some portion of the ground water in the deep
aquifer likely travels under Naylors Run via fractures and discharges further downgradient to the
southeast.
According to the ROD, the ground water velocity in the shallow aquifer is approximately 85 ft per year,
presumably based on aquifer testing and measurement of the hydraulic gradient. At the time of the ROD,
however, the extent of contaminant migration suggested that other factors, such as adsorption, were
limiting the rate of contaminant transport. Ground water flow velocity in the deep aquifer, within the
underlying fractured bedrock, is estimated to be 25 ft per year according to the ROD.
The presence of underground sewers that intercept the water table has altered ground water flow patterns
and contaminant transport at the site. Prior to site remedial activities, contaminated ground water
infiltrated into a 30-inch sewer line near the PCG property. Although this sewer line reduced the amount
of contamination that could migrate downgradient, this contaminated water discharged directly to
Naylors Run. As part of OU2, EPA lined the 30-inch sewer line to reduce its ability to collect and
discharge contaminated water. A collector drain, which is a component of the interim ground water
remedy, was installed in this general area to intercept and extract contaminated ground water in the
shallow zone (i.e., overburden).
In 1999, the site team identified a contaminated seep adjacent to Naylors Run (location indicated on
Figure 1-1). By the Summer of 2003, the site team found an abandoned 12-inch sewer line that has likely
been facilitating the transport of contaminated ground water downgradient and causing the seep. PCP
concentrations of over 500 ug/L have been observed at this downgradient location even though
upgradient monitoring points (such as the CW-12 cluster) show lower or undetectable concentrations.
During the OSE visit, the site team showed the OSE team the seep area where impacted water was
surfacing and indicated that addressing this sewer line, seep, and resulting contamination is a top priority.
1.5.4 RECEPTORS
The area potentially affected by contaminants released from the Havertown PCP site is almost entirely
developed. The primary contamination source (NWP) is located in midst of several commercial
establishments and surrounded by a mix of private homes, schools, stores, parks, and industrial facilities.
Consequently, humans potentially make up the most important receptor group. Additional environmental
receptors may include vegetation, aquatic biota, wildlife and domestic animals, and agricultural or garden
products.
Because the contaminated soil has either been removed or covered with a cap, the primary routes of
exposure to site-related contaminants would include extraction and use of contaminated ground water,
direct contact with contaminated ground water that has seeped to the surface, or construction activities
beneath the cap. With the exception of the above-mentioned seep caused by the 12-inch sewer line,
recent surface water sampling indicates low (i.e., at or near MCLs) contaminant concentrations in
Naylors Run, which is a substantial improvement compared to the 1,200 ug/L that had been detected in
Naylors Run near the PCG property during the initial investigations.
1.5.5 DESCRIPTION OF GROUND WATER PLUME
Figure 1-3 shows the extent of PCP contamination near the NWP and PCG properties in the shallow
aquifer (overburden), and Figure 1-4 shows the extent of PCP contamination in the same area for a
deeper aquifer interval (fractured bedrock). The highest detected levels of PCP at the time of the OSE
are 13,000 ug/L in R-2, 30,000 ug/L in HAV-02, and 8,700 ug/L in HAV-04 (March 2003 sampling
-------
event). Wells R-2 and HAV-02 are located just downgradient of the site along Eagle Road, and well
HAV-04 is located approximately 450 feet downgradient of the site. PCP contamination at further
downgradient locations is relatively low compared to upgradient. The CW-9 and CW-13 clusters (see
Figure 1-1) are the only clusters downgradient of the collector drain with detectable PCP concentrations
since sampling began at downgradient locations in October 2002.
Dioxin is also present in the ground water. The last comprehensive ground water sampling round for
dioxin was conducted in 2000, and the highest concentrations were found in CW-2D, HAV-02, and NW-
1-81, which are located either on the former NWP property or immediately downgradient of it. In
general, the highest dioxin concentrations are found in the wells with the highest PCP concentrations.
Dioxin concentrations in wells downgradient of the collector drain (see Figure 1-2 for well locations) are
below the MCL of 0.03 ppt (0.00003 ppb).
-------
2.0 SYSTEM DESCRIPTION
2.1
SYSTEM OVERVIEW
The ground water extraction system consists of 4 recovery wells located along Eagle Road and a 180-foot
long collector drain installed to the top of bedrock (16 feet bgs) downgradient of the PCG facility . The
collector drain is located adjacent to the existing storm sewer line to collect contaminated shallow ground
water for treatment at the treatment plant. The purpose of the collector drain is to effectively capture the
plume of contaminated water in the shallow aquifer as it flows to the east. The recovery wells collect
impacted ground water and LNAPL near the source area. The system was constructed by one contractor
in 2001, operated by another contractor through July 2002, and has since been operated by a third
contractor.
The treatment system consists of an oil/water separator (OWS), chemical precipitation system, an
advanced oxidation process system, and granular activated carbon (GAC) system, sequentially. The
treated water is discharged into Naylors Run.
The following table summarizes the contaminant concentrations in the plant influent for June 2003. The
effluent standards are also provided.
Parameter
Cobalt
Iron
Manganese
Benzene
Trichloroethylene
Pentachlorophenol
Phenanthrene
Dioxins/Furans (ppq***)
June 2003 Influent Concentration
(ug/L)
135
5,320
10,300
6.9
9.9
2600
25*
315.5**
NPDES Monthly
Average Effluent
Standard
(ug/L)
53
300
50
1
5
1
3
<4.4
* Concentration below the detection limit. The data comes from previous sampling event on 5/29/2003
** No data during June 2003 sampling event. The data comes from previous sampling event on 5/29/03
***ppq - parts per quadrillion or picogramsper liter
2.2
EXTRACTION SYSTEM
The ground water extraction system includes 4 recovery wells labeled RW-1 through RW-4 and the
collector drain. With the exception of RW-3, which is located on the PCG property directly downgradient
of R-2, the recovery wells are on or adjacent to NWP property in the vicinity of the hot spot at well R-2.
-------
They range from 26 to 28 feet deep (i.e, to the interface of the overburden and fractured bedrock). These
wells have QED AP-4 pneumatic submersible pumps. Only 1 of the 4 wells, RW-1, is currently pumping
over 2 gpm, and two of the wells (RW-3 and RW-4) are not pumping at all. The wells are in need of
redevelopment, which was planned for late August 2003.
Three extraction wells (EW-1, EW-2, and EW-3) were drilled as part of the OU2 effort, but have not
been used for ground water recovery purposes. R-2, a 4-inch monitoring well located on the former
NWP property, is manually bailed. Eight events have taken place since October 2002, recovering
approximately 19 gallons of product.
2.3 TREATMENT SYSTEM
The primary components of the treatment system are listed below.
Free product separation and storage
oil/water separator with the capacity of 100 gpm that gravity drains water to the equalization
tanks
5,900 gallon free product tank with level transmitter to store product draining from the oil/water
separator and bailed from R-2
two 200 cfm capacity GAC units to treat vapors vented from the oil/water separator and free
product tank
Flow equalization, metals removal, and solids handling
two 2,800 gallon equalization tanks
50 gpm equalization tank discharge pump with standby unit
three 800 gallon oxidation and mixing tanks with chemical and polymer addition
lamella plate clarifier unit with flash mix and flocculation chambers with a 55 gpm capacity
double diaphragm sludge pump at the oil/water separator and clarifier
upflow sand filter with a 50 gpm capacity
4,500 gallon sludge holding tank to store the sludge from the oil/water separator, the clarifier,
and sand filter
filter press with 20 cubic foot capacity
UV/Oxidation
2,000 gallon metals removal effluent surge tank and 45 gpm RayOx feed pumps (two including
one standby)
RayOx system including three 30 KW UV reactors in series
peroxide destruction unit with 2,000 pounds of GAC and 2,000-pound clay filter (currently off-
line)
two parallel cartridge filters
GAC
two 4,000-pound GAC units arranged in series
• two 3,000 gallon effluent tanks
In addition, the system includes a 450 gallon sump with a pump for recycling filter press filtrate, decant
water from the sludge thickener and sand filter backwash water; an air compressor (to operate the
recovery well extraction pumps); and controls with remote access and data storage capabilities.
-------
The treatment system was designed to treat 40 gpm. The flow from the recovery wells and the collection
drain is directed to the oil/water separator and the above-mentioned treatment components prior to being
discharged to Naylors Run.
2.4 MONITORING PROGRAM
Process monitoring consists of monthly influent and semi-monthly effluent sampling for VOCs, SVOCs,
metals, and dioxins/furans. Eight process samples are collected and analyzed quarterly to monitor the
performance of individual treatment components, with analysis of each sample dependent on its location
relative to individual treatment components.
Ground water monitoring includes sampling 10 wells quarterly, 10 wells semi-annually, and about 55 to
60 wells annually (planned). The last comprehensive ground water sampling event was conducted in
October 2002 when approximately 33 wells were sampled. At least 7 of the wells that were not sampled
were either dry or had product. All ground water samples (approximately 100 per year) are analyzed for
VOCs, SVOCs, and metals. About 50 of those 100 samples are analyzed for dioxins/furans.
-------
3.0 SYSTEM OBJECTIVES, PERFORMANCE AND CLOSURE
CRITERIA
3.1 CURRENT SYSTEM OBJECTIVES AND CLOSURE CRITERIA
The 1989 ROD for OU1 identified three operable units (OUs) and associated remedies:
OU1 - the removal of existing hazardous wastes from the site and the installation of an oil/water
separator for removing some contaminants from Naylors Run
OU2 - the remediation of the shallow ground water aquifer (overburden), considered as an interim
action for ground water
OUS - the remediation of the sediment contamination in Naylors Run, the deep ground water aquifer
(fractured bedrock), and surface water and sediment contamination due to runoff from on-site
soils
Additionally, a cap was installed in 1994 (between OU1 and OU2) as a removal action.
This optimization evaluation primarily pertains to the OU2 remedy but also considers potential options
for a final ground water remedy. As stated in a 1991 ROD, OU2 includes an interim remedy for shallow
ground water. The objective of this interim remedy is to collect and treat the shallow ground water that
flows into Naylors Run and initiate the remediation of the shallow ground water. The primary purpose is
to protect children that might play in Naylors Run. This is not a final remedy for ground water because it
only addresses shallow ground water and not both shallow and deep ground water. The information
collected during this interim remedy will help in the planning of the OUS remedy, which would include
remediation of the contaminated sediment and a final remedy for ground water.
The selected interim OU2 remedy includes the following major features:
Installation of free product recovery wells on the NWP property to recover product from the
shallow aquifer
• Rehabilitation of the existing storm sewer line to reduce infiltration of contaminated ground
water from the shallow aquifer to the storm sewer
Installation of a ground water collection drain adjacent to the existing storm sewer line and to the
top of bedrock to extract contaminated ground water in the overburden
• Installation of a ground water treatment plant at NWP to perform chemical precipitation, either
powdered activated carbon treatment or an advanced oxidation treatment, and finally granulated
activated carbon (GAC) polishing.
Concurrent with operating the OU2 remedy, the site team is conducting the OUS Remedial Investigation.
The objectives are to evaluate sources, nature, extent, migration pathways, and receptors of site-related
contamination in deep ground water. These efforts include the addition of 13 monitoring wells (vertical
well pairs CW-9 through CW-14 and single well CW-15), sampling of site-wide wells, sampling of
surface water, pumping tests, ground water flow modeling, and an evaluation of the recovery system.
10
-------
3.2
TREATMENT PLANT OPERATION STANDARDS
The design influent concentration and effluent concentration standards for the primary contaminants of
concern are summarized in the following table.
Contaminants of
Concern
PCP
Phenanthrene
Benzene
Trichloroethylene
TCDD (dioxins/furans)
TSS
Cobalt
Iron
Manganese
Oil & Grease
Design Influent
Concentration
llmg/L
lug/L
lOmg/L
270 mg/L
13.8 mg/L
3,500 mg/L
NPDES Monthly Average
Effluent Standard
lug/L
3ug/L
lug/L
5ug/L
< 4.4 ppq
No standard
53ug/l
300 ug/L
50ug/L
15 mg/L
11
-------
4.0 FINDINGS AND OBSERVATIONS FROM THE OSE SITE VISIT
4.1 FINDINGS
The evaluation team observed a conscientious and knowledgeable site team (i.e., EPA and its contractor)
that has effectively managed this interim remedy, provided a number of much needed plant
improvements, and initiated the remedial investigation for OU3. The observations provided below are
not intended to imply a deficiency in the work of the system designers, system operators, or site
managers but are offered as constructive suggestions in the best interest of the EPA and the public.
These observations obviously have the benefit of being formulated based upon operational data
unavailable to the original designers. Furthermore, it is likely that site conditions and general knowledge
of ground water remediation have changed over time.
4.2 SUBSURFACE PERFORMANCE AND RESPONSE
4.2.1 WATER LEVELS
Water levels collected in both the shallow zone (overburden) and deep zone (bedrock) in March 2003
have been used by the site team to generate potentiometric surface maps. These maps are presented in
Figures 4-1 and 4-2. Approximately 29 measurement points are used to generate the shallow
potentiometric surface map, and approximately 14 measurement points are used to generate the deep
potentiometric surface map. Both figures show ground water flow to the southeast, and both figures
show the influence of the collector drain. As expected, the influence of the collector drain in the shallow
zone is more predominant than the influence in the deeper zone. This is partially due to the use of water
level measurement points within the drain when generating the shallow potentiometric surface.
Nevertheless, the influence of the drain on the deeper water level measurements suggests that, under
pumping conditions, some water in the deeper aquifer is entrained by the drain. These two figures, when
compared with each other, can be used to interpret vertical ground water flow, but no easily discernible
trend is recognized.
These water levels only pertain to the conditions during the Spring of 2003, and more specifically to
March 2003. Water levels conducted at different times may suggest seasonal variation in both horizontal
and vertical ground water flow.
4.2.2 CAPTURE ZONES
As part of this report, the OSE team has used the March 2003 potentiometric surface maps generated by
the site team to interpret the horizontal capture in both the shallow and deep zones. The interpreted areas
of capture are depicted in Figures 4-1 and 4-2, with the shallow capture zone more extensive than the
deeper capture zone. Therefore, ground water that flows up from within the deeper capture zone should
be within the shallower capture zone. These preliminary analyses consider only one round of water
levels and do not rigorously consider vertical flow. Because water levels may change seasonally
affecting both horizontal and vertical flow, a more thorough capture zone analysis is merited.
No particular target capture zone appears to have been documented by the site team. This is likely
because OU2 is an interim remedy and because the collector drain could not be extended as far to the
12
-------
south as originally planned due to the residences. Nevertheless, the surface water contamination in
Naylors Run has decreased, and the interpreted shallow capture zone appears to provide capture for the
most contaminated portions of the site, including the contamination at CW-4 and CW-5. This contrasts
with a more thorough capture zone evaluation conducted by the site team with the water levels collected
during December 2002 (not provided in this report). The deeper capture zone interpreted from the March
2003 data is also fairly extensive but does not appear to include the contamination at CW-4 and CW-5.
A water budget analysis can also be conducted to provide a simplistic evaluation of capture. The
following parameter values are relevant.
• According to the 1991 ROD, the ground water velocity in the shallow aquifer is approximately
85 feet per year (0.23 feet per day). Assuming this ground water velocity was calculated using a
porosity of 0.30, this translates to a Darcy velocity of approximately 25.5 feet per year (0.07 feet
per day).
• Conservatively, the saturated thickness of the overburden (and perhaps the upper portion of the
fractured bedrock) is 35 feet.
• Approximately 20 gpm (3,860 ft3 per day) is extracted from the trench.
Assuming limited infiltration from precipitation or from the underlying formation, the amount of water
extracted by the drain is equal to the amount of water flowing through a given cross-section of the
aquifer. This equality can be expressed as follows {Q=V,fVb), where Vd is the Darcy velocity, Wis the
width of the cross-section, and b is the saturated thickness. This equation can be rearranged to solve for
the width.
Q__ 3,860 %
VJ>~ 0.07%,x 35ft ~ ' J
This result suggests that the width of capture is approximately 1,575 feet; however, this result is based on
the simplifying assumption that there is no infiltration from precipitation or from the underlying
formation. Much of the overlying ground is paved, which means that there is likely little contribution
from precipitation. However, the potentiometric surface for the deeper zone indicates influence from the
collector drain. Therefore, the capture zone width is likely smaller than the 1,575 feet because the
effective capture zone depth is likely greater than the 35 feet that was assumed in the water budget
calculation. In general, the water budget analysis supports the interpretation from the potentiometric
surface maps that capture is fairly extensive across the site. Neither of the above evaluations, however,
confirm complete capture across the site, especially in areas where contamination has not been fully
delineated (i.e., vertically and to the south).
Water quality sampling from downgradient wells can also be used to evaluate capture. The October 2002
sampling event included sampling of clusters CW-9 through CW-14 and sampling of CW-15. With the
exception of CW-9S, all of the shallow samples had undetectable concentrations of PCP. All of the
shallow samples had concentrations of dioxins under the MCL of 0.03 ppt. The elevated PCP
concentration at CW-9S (70 ug/L) suggests that contamination either reached this location prior to
operation of the interim OU2 ground water remedy, that capture is incomplete and PCP contaminated
ground water continues to migrate from the source area to this location, or that CW-9S is located inside
the capture zone. It is reasonable to assume that this contamination at CW-9S was present prior to
pumping, or, based on Figure 4-1, it is reasonable to conclude that CW-9S is within the capture zone. To
evaluate capture in the future, the trend in CW-9S can be analyzed, but the results may be inconclusive.
13
-------
If the trend shows decreasing concentrations toward background, this will indicate that CW-9S is outside
of the capture zone and that capture is successful. If concentrations remain the same or increase, this
could mean that CW-9S is either within the capture zone or that capture is not successful. Therefore,
evaluation of capture should not depend heavily on sampling results from CW-9S.
With the exception of CW-9D and CW-13D, the downgradient wells in the October 2002 ground water
sampling event had undetectable concentrations of PCP and dioxin levels below the MCL. As with the
shallow zone, this contamination may have been present before operation of the OU2 system began. If
capture is sufficient, the concentrations should decrease to background (i.e., zero) overtime. If the
concentrations consistently remain elevated or increase, then capture is likely insufficient.
A primary concern regarding capture is the presence of an abandoned 12-inch sewer line that is likely
acting as a conduit for contaminated ground water. This sewer line apparently ends near Naylors Run in
the location of "impacted seeps" indicated on Figure 1-1.
In general, the interim remedy has functioned as designed by substantially reducing contaminant
transport into Naylors Run. Capture in the shallow zone may or may not be complete, but a number of
lines of evidence suggest that the capture zone is fairly extensive and extends into the deeper zone.
Addressing the abandoned sewer line, first discovered in Summer 2003, presents the most severe issue
with respect to contaminant migration and further impacts to Naylors Run.
4.2.3 CONTAMINANT LEVELS
Ground water sampling in October 2002 provides an indication of the extent of ground water
contamination. The most impacted areas are beneath the former NWP property and the PCG property.
Concentrations downgradient of these properties are relatively low and even undetectable in most
locations. Ground water sampling has not been conducted long enough to evaluate a trend toward
cleanup. However, it is reasonable to conclude that concentrations upgradient of the collector drain have
not decreased substantially. NAPL remains a continuing source in this area, minimal background flow is
present to flush the contaminants toward the collector drain, and the recovery wells have been fairly
unsuccessful at recovering NAPL.
The ground water sampling is also relatively incomplete in terms of delineating the plume, particularly to
the south and in the vertical direction. The wells in the CW-4 and CW-5 clusters are the southern most
wells near Eagle Road, and as shown in Figures 1-3 and 1-4, the PCP concentrations at these wells
exceed 1,000 ug/L. With no additional wells to the south, the PCP plume is not delineated. This lack of
delineation affects the preliminary capture zone analysis conducted above because the degree of capture
cannot be determined without full delineation. As depicted in Figure 1-4, a number of deep wells with
screened intervals down to approximately 50 feet below ground surface have concentrations exceeding
1,000 ug/L. Without deeper wells, the PCP plume cannot be considered delineated vertically. In EW-1
through EW-3 (extraction wells located upgradient of the collector drain that have never been used for
pumping), the concentrations are also above 1,000 ug/L. These wells are screened from approximately
40 to 80 feet bgs, whereas the collector drain is completed to approximately 16 feet bgs. With this
relatively long screened interval, the vertical extent of contamination cannot be determined. The
contamination may only be present near the tops of these relatively long screened intervals, but it may
also be present at deeper levels. In addition, these wells may be acting as conduits, allowing
contamination at approximately 40 feet to migrate to depths near 80 feet. The upward gradient in this
area due to the operation of the collector drain makes this downward migration unlikely, but this
possibility should not be eliminated without deeper sampling.
14
-------
The concentrations entering the treatment plant are below the design values. The influent concentrations
and the design concentrations are compared in the following table.
Parameter
PCP
Dioxins/Furans
Iron
Manganese
Oil and Grease
Total VOCs
Design Influent Concentration
(ug/L)
11,000
1
270,000
13,800
3,500,000
No design value
Actual Influent
Concentration* (ug/L)
3,900
0.0015
8,400
11,000
15,000
<100
* Average influent concentration between September 2002 and June 2003 rounded to two significant digits
The mass removal rate for PCP, assuming the above concentration and an extraction rate of 20 gpm, has
been approximately 1 pound per day as shown in the following equation.
20 gal 3,900 ug 3.785L 1440min. 2.2 Ibs 0.94 Ibs
:—x x —x x
mm.
gal
day
10 ug day
Mass removal of other VOCs and SVOCs (including benzene, trichloroethylene, and naphthalene) is also
occurring, but at a much reduced level compared to the mass removal of PCP. The following table
presents those constituents in the influent that are above the discharge standards and require treatment.
Parameter
Benzene
Trichloroethylene
Naphthalene
PCP
Phenanthrene
Oil and Grease
Aluminum
Cobalt
Iron
Lead
Manganese
Maximum Influent Concentration
(ug/L)
12
15
1,800
9,500
400
61,200
2,470
227
43,500
8.6
12,300
NPDES Monthly Average
Effluent Standard (ug/L)
1
5
30
1
3
15,000
360
53
300
5
50
* Influent data from September 2002 through June 2003
Sampling data indicate that effluent concentrations of all the primary contaminants of concern are below
the effluent standards with the exception of manganese. The site team determined that the manganese
was leaching from the clay filter. The clay has since been taken offline, and exceedances of manganese
have not occurred since.
15
-------
4.3 COMPONENT PERFORMANCE
4.3.1 EXTRACTION SYSTEM WELLS, PUMPS, AND HEADER
The extraction system consists of the collector drain and the four recovery wells. The collector drain is
constructed like a french drain in a trench that is approximately 2 feet wide, 180 feet long, and
approximately 15 to 20 feet deep (completed to the top of bedrock). Water from the drain enters a four
foot sump where a Grundfos electric submersible pump directs water to the treatment plant in a 2-inch
pipe. Potential flow is limited by a 1-inch connection located in the piping near the drain. Four
piezometers are completed within the trench to provide an indication of drawdown. The high and low
level settings for transducers in the trench are not known, and there is not sufficient pumping capacity to
reach the low level. The drain sump is reportedly full of solids and requires cleaning.
The recovery wells are completed to the top of bedrock, which is about 30 feet bgs. They operate well
below expected recovery rates, with only one well pumping over 2 gpm. None of the wells produce
much product. The majority of recovered product is collected manually from R-2, which is located
adjacent to RW-2. All of the wells are in need of rehabilitation.
Three 6-inch wells (EW-1 through EW-3) were drilled, but were not converted to extraction wells and
have not operated. These wells are screened from 40 to 80 feet bgs just upgradient of the trench.
The extraction system piping is equipped with a leak detection system, but this system frequently sets off
alarms due to precipitation inflow (not associated with leaks). These alarms shut down the extraction,
causing downtime. The site team has therefore disconnected the leak detection system, allowing the
system to avoid unnecessary downtime during rain events.
4.3.2 OIL/WATER SEPARATOR AND EQUALIZATION TANK
The oil/water separator receives an influent flow of about 20 to 25 gpm typically. Minimal LNAPL is
recovered by the separator system. The free product tank is extremely oversized for this application; the
site team has added a fill line and level transmitter to make the tank easier to use. The two 2,800-gallon
equalization tanks receive water from the oil/water separator. The site contractor has added a level
transmitter within the first tank and a site tube for local level reading. The equalization tank effluent
pump basket strainers and ball check valves require daily cleaning to keep the plant operational.
4.3.3 METALS REMOVAL SYSTEM
The three 800-gallon tanks allow for chemical addition and mixing. Transfer from one tank to the next is
gravity fed, and the first tank is the rate-limiting step for the entire treatment plant. Although the plant is
reportedly designed for 40 gpm, the first tank begins to overflow (or at least splash over) at just under 40
gpm. The other components of the metals removal system, including the other tanks, chemical addition,
lamella clarifier, and sand filter operate as expected. Nevertheless, they require frequent cleaning. The
May 2003 process sampling indicates that the effluent from the clarifier meets the metals discharge
standards.
4.3.4 UV/OX SYSTEM
Water from the sand filter discharges to a surge tank where a 50% hydrogen peroxide solution is added to
the process water for a hydrogen peroxide concentration of 200 ppm. The water from this surge tank is
then pumped through the UV/OX system, which consists of three 30 KW reactors arranged in series. The
16
-------
May 2003 process sampling indicates that the effluent from the first UV reactor meets all of the organic
discharge standards with the exception of PCP. The PCP concentration at this location in May 2003 was
2.3 ug/L compared to the discharge standard of 1 ug/L, which could easily be removed by remaining UV
reactors or simply GAC units alone.
Maintenance of the UV/OX system involves replacement of the seals and bulbs and periodic cleaning of
the surge tank and reactors. The site team plans to add bag filters between the sand filter and the surge
tank to remove solids that pass through the sand filter. This should reduce the amount of solids entering
the surge tank and the UV/OX system.
The UV/OX system has been the primary reason for the above-average downtime. Between April and
July 2003, plant uptime was only 64%. Main problems include lamp failures, a blower motor failure, and
an insufficient inventory of replacement parts. The blower motor failure alone resulted in a four-day
shutdown because the part was not on hand.
4.3.5 PEROXIDE DESTRUCTION UNIT, CLAY FILTER, AND GAC
The peroxide destruction unit (PDU) operates as intended, reducing the hydrogen peroxide concentration
to 10 ppm or less. The site team has replaced the 2,000 pounds of carbon in this unit once and has added
the capability to backwash the PDU. The clay filter has been taken offline due to the leaching of
manganese that had built up over time and was causing periodic exceedances of the manganese discharge
standard. The clay filter was not needed or helpful in reaching organic treatment standards. Cartridge
filters are used to further reduce solids prior to the GAC.
The two 4,000-pound GAC units have been replaced once since the plant began operation in 2001. The
replacement was not due to organic contaminant loading or to pressure build up. It was due to the
potential leaching of manganese that had built up over time and was potentially contributing to
exceedances of the manganese discharge criteria.
Frequent backwashing of the GAC and PDU have contributed to the system downtime. This is primarily
because the current design of the treatment plant does not allow for storage of the waste water generated
from the backwashing. As a result, extraction needs to be temporarily stopped to allow continued
treatment to provide capacity in the equalization tanks.
4.3.6 EFFLUENT TANK AND DISCHARGE
Prior to discharge via pumping to Naylors Run, the process water is emptied into two 3,000-gallon
effluent tanks that are located outside of the plant building. As with other treatment components, the
effluent tanks and pumps require frequent cleaning due to biological build up.
4.3.7 SLUDGE HANDLING TANKS AND EQUIPMENT
Sludge from the system clarifier, oil/water separator, and the sand filter is thickened in a 4,500-gallon
sludge holding tank prior to being dewatered in a filter press. The filter press is operated about once per
week. Approximately 10 cubic yards of sludge is disposed of every 8 to 12 weeks. The sludge is
hazardous because it is listed waste (F032) and contains dioxins. This sludge and the recovered product
(in liquid form) is shipped and incinerated.
17
-------
4.3.8
SYSTEM CONTROLS
The site team has added a number of flow meters, high and low level controls, and sensors to facilitate
plant operation. It has also added a PC, software for trending analyses, and a DSL with PC Anywhere
software for remote access and viewing. Original documentation on the control system is limited.
4.4
COMPONENTS OR PROCESSES THAT ACCOUNT FOR MAJORITY OF
ANNUAL COSTS
Annual O&M costs for year two of the current contract (2003-2004) are approximately 591,000 per year,
which marks a decrease of $68,000 compared to the O&M costs for year one of the current contract
(2002-2003). The following table presents an estimated breakdown of Year-2 annual costs as provided
by the site contractor. A cost of $50,000 was incurred in Year 1 and is expected in Year 2 for major
equipment changes and plant improvements, but this $50,000 cost is not included in the table because
such repairs/improvement should not be indicative of future O&M. All presented costs are approximate.
Item Description
Labor: Project management and reporting, technical support, and design of plant
improvements
Labor: Plant operator (one full-time operator, one part-time engineer to upgrade the
remedy)
Labor: Process and ground water sampling and data analysis
Utilities: Electricity (assumes two operating UV/OX units)
Utilities: Water, phone, garbage, and cap maintenance
Non-utility consumables: chemicals
Non-utility consumables: UV lamps/accessories (assumes two operating UV/OX units)
Non-utility consumables: GAC
Chemical Analysis
Routine maintenance items
Waste disposal
Estimated Cost
$75,000 per year
$180,000 per year
$75,000 per year
$55,000 per year1
$15,000 peryear
$70,000 per year
$16,000 per year
$20,000 per year2
N/A3
$30,000 per year4
$55,000 per year4
Total Estimated Cost
$591,000
1 Site contractor estimated $75,000 using three UVOX units, $55,000 is a rough estimate for two UVOX units (some of the
electricity costs are due to pumps/motors and will not decrease with bypassing a UV/OX unit).
2 Site contractor estimated $40,000 in Year 1 for one replacement of two units and no replacements in Year 2. The OSE team
estimates $20,000 per year for one replacement of one unit.
3 Chemical analysis is conducted by EPA and costs are not assigned to the site.
4 Costs are incurred periodically, and the presented costs represent averages between Year 1 and Year 2.
4.4.1
UTILITIES
The primary contributor to utility costs is the electricity required to power the three UV/OX reactors. In
Year 1, three UV/OX units were used at a total of 90 KW, which likely translates to an electricity cost of
approximately $75,000 per year. The remainder of the electric costs are likely due to the process pumps,
blower motors, and compressor. In Year 2 and future years, the site team plans to reduce the number of
UV/OX units to two, which should reduce electricity costs to approximately $55,000 per year (some of
the electricity costs are due to pumps/motors and will not decrease with bypassing a UV/OX unit). The
remaining utilities include garbage pickup, maintenance of the cap, phone, and public water.
18
-------
4.4.2 NON-UTILITY CONSUMABLES AND DISPOSAL COSTS
Non-utility consumables represent approximately $106,000 per year in O&M costs for Year 2.
Chemicals for the metals precipitation and peroxide for the UV/OX system are the dominant
consumables costs, and ferric sulfate (for metals removal) is the dominant chemical cost. Of the
remaining $36,000 per year, approximately $16,000 is need for UV/OX maintenance parts. It should be
noted that the above table assumes replacement of one GAC unit per year, primarily due to solids build
up. If solids fouling does not occur with the clay filter removed, GAC replacements may be less
frequent. Because of the extremely low contaminant concentrations entering the GAC, frequent
replacements are not expected due to loading of organic contaminants. Waste disposal costs are about
$55,000 per year and are mainly for disposing of sludge and recovered oil as hazardous waste.
4.4.3 LABOR
Labor contributes more than 50% of the total O&M costs at this site, and is divided into three primary
categories: plant operation, PM/reporting, and ground water sampling/data analysis. Between Year 1 and
Year 2, plant operation has been reduced by $20,000 for an Year-2 cost of $180,000. This cost includes
one full-time operator and one part-time engineer that is contributing to system improvements. The
PM/reporting corresponds to approximately $6,500 per month and includes preparation and attendance at
the monthly site meetings, engineering support for improving the treatment plant, and preparing monthly
Discharge Monitoring Reports. The sampling costs of $75,000 per year include the cost of the NPDES,
quarterly process sampling, and ground water sampling. The majority of these sampling costs are related
to the ground water sampling and associated data analysis because of the labor and equipment involved
in low-flow sampling.
4.4.4 CHEMICAL ANALYSIS
Chemical analysis costs are not incurred by the site because EPA conducts the analysis through the
Contract Laboratory Program.
4.5 RECURRING PROBLEMS OR ISSUES
The previous manganese exceedances represent a recurring issue but have been addressed by removing
the clay filter and replacing the carbon in the PDU and GAC units. Another recurring issue that has been
addressed was the leak detection system response to precipitation inflow. The leak detection system has
been shut down to avoid unnecessary downtime during rain events.
The limiting capacity of the first mixing tank in the metals removal system is also a recurring issue.
Although the extraction rate is generally between 20 and 25 gpm, the process rate of the treatment plant
is generally higher due to backwashing of filtrate from the filter press, and other recycled water or water
added from outside the system. If the extraction rate is increased in the future, the limiting capacity of
this tank will be problem.
Maintenance issues with the UV/OX system cause excessive down time. At the time of the OSE, it was
practice to shutdown the entire treatment plant if one of the three UV/OX reactors was not working.
Therefore, when the blower on the third reactor failed, this caused system-wide downtime. System-wide
downtime was also caused by repeated lamp failures on the second reactor.
19
-------
In general, a number of treatment plant features have required modification or replacement. The absence
of level transmitters, flow meters, and flow controls at various parts of the plant (presumably due to cost-
cutting measures during design and installation) have hampered O&M of the plant and general trouble
shooting. The site team has worked to improve many of these features; nonetheless, issues with original
design and construction (such as limited space and lack of as-built drawings and control system
documentation) repeatedly arise and require both time and expense to address.
4.6 REGULATORY COMPLIANCE
With the exception of the manganese exceedances, which have now been addressed, the system regularly
meets its discharge criteria.
4.7 TREATMENT PROCESS EXCURSIONS AND UPSETS, ACCIDENTAL
CONTAMINANT/REAGENT RELEASES
No process excursions, accidents, or reagent releases were reported during as part of this optimization
effort.
4.8 SAFETY RECORD
One accident has occurred at the treatment plant. The RPM, while visiting the plant, tripped on the door
jamb and fell down a step resulting in a knee injury. The concrete pad has since been modified to reduce
the tripping hazard, and safety paint has been applied to draw attention to corners, steps, and the door
jamb.
20
-------
5.0 EFFECTIVENESS OF THE SYSTEM TO PROTECT HUMAN
HEALTH AND THE ENVIRONMENT
5.1 GROUND WATER
The interim ground water remedy has accomplished many of its intended goals. The lining of the 30-inch
storm sewer and the installation of the collector drain has reduced transport of contaminated water into
Naylors Run. Many of the surface water samples now have non-detect results because this portion of the
remedy has functioned as intended.
Despite these success, the interim remedy is likely not suitable as the final remedy without modifications.
First and foremost, the abandoned 12-inch sewer line that was first discovered in 2003 is likely
facilitating transport of PCP and dioxin from ground water to a downgradient location. PCP
concentrations in seeps as high as 750 ug/L have been detected. Human and other receptors can be
exposed to this contamination if it is not controlled or removed.
The current remedy may also not provide adequate capture. Current site conditions (with the exception
of the 12-inch sewer line) suggest that PCP is not reaching surface water in detectable amounts; however,
if the plume containment is not adequate, contaminated ground water may impact surface water in the
future. The interpreted capture zone, based on fairly preliminary analysis of the March 2003 ground
water elevation data (See Section 4.2.2), appears fairly extensive. However, the interpreted capture zone
based on other data is somewhat less extensive. Furthermore, the vertical extent of the capture zone is
difficult to evaluate and the plume is not fully delineated in both the horizontal (especially to the south)
and vertical directions.
It is understood that many of these items were not intended as part of OU2 and that they will be
addressed as part of OU3, which represents the final remedy for ground water.
5.2 SURFACE WATER
Conditions in Naylors Run have improved substantially due to the operation of the OU2 interim remedy;
however, portions of Naylors Run still have substantial impacts likely due to a 12-inch sewer line
providing a conduit for contaminated ground water to reach the surface and discharge into Naylors Run.
Seep concentrations of PCP have been as high as 750 ug/L and sediment PCP concentrations have been
as high as 6,200 ug/kg. Dioxin concentrations are also elevated. This area is indicated on Figure 1-1.
Proper sealing of the abandoned sewer line and localized remediation efforts as part of OU3 should be
able to address this contamination.
5.3 AIR
Due to previous removal actions, air quality is no longer adversely affected by the site.
21
-------
5.4 SOILS
Soils beneath the NWP cap are impacted, but the cap prevents direct contact with these soils. Other soils
downgradient of the cap and along the water table are also likely impacted due to the migration of free
product and the ability of the free product to smear as the water table rises and recedes. These soils
should not be a threat to human health unless there are invasive activities in these areas. Efforts should
likely be taken to delineate these areas and provide institutional controls to notify construction workers
or others that may be involved in these invasive activities.
A greater threat to human health and the environment are the surface soils that are impacted by PCP and
dioxin from the contaminated water that is accumulating in the 12-inch abandoned sewer and discharging
to the surface. These soils are located in a Residential Open Space and are accessible to homeowners in
the area.
5.5 WETLANDS AND SEDIMENTS
The discussion in the above sections regarding the impacted seep also apply to the sediments in this area.
Conditions in the wetlands and sediments will be addressed as part of OU3.
22
-------
6.0 RECOMMENDATIONS
Cost estimates provided herein have levels of certainty comparable to those done for CERCLA
Feasibility Studies (-307+50%), and these cost estimates have been prepared in a manner consistent with
EPA 540-R-00-002, A Guide to Developing and Documenting Cost Estimates During the Feasibility
Study, July 2000.
6.1 RECOMMENDATIONS TO IMPROVE EFFECTIVENESS
The recommendations in this section are not intended to imply deficiency of the OU2 remedy. It is
recognized that the OU2 remedy is an interim measure to address shallow ground water contamination
and its impacts on Naylors Run. Many of these recommendations in this section are offered not as
improvements to OU2 but as considerations for OU3. Some of these recommendations (particularly
6.1.1), however, should be implemented as soon as possible and potentially as part of an emergency
response.
6.1.1 PROPERLY SEAL ABANDONED 12-INCH SEWER LINE AND REMEDIATE SURFACE SOILS
NEAR THE SEEP
The site team is already addressing this issue. The first step involves locating the upgradient portion of
this abandoned sewer line immediately downgradient of the collection drain. It is thought that the line
may have been broken or disturbed during construction of the collection drain, allowing ground water to
infiltrate it. Once the upgradient portions are located, they can be evaluated to confirm that this line is
the cause of the impacted seeps further downgradient. Once confirmed, the line and the associated
bedding material should be plugged to prevent further infiltration and transport of contaminated water.
With the line plugged, the seep area can be evaluated to confirm that the discharge has stopped, and then
the contaminated soils and sediments can be addressed.
This recommendation is of primary importance and should be done as soon as practicable. Because this
area is considered a Residential Open Space (ROS) and is available to nearby residents, the contaminated
area should be clearly marked to warn residents of the contamination.
The abandoned line has been identified at possibly 9 feet bgs. Assuming that the line can be found, a dye
test and possibly a video inspection can be used to confirm the connection to the ROS area. Once
confirmed the line and bedding can be plugged with a cement/bentonite slurry. The length of the line to
be plugged beyond the upgradient end would be determined by the condition and accessibility of the line
between the collection drain and the ROS area. The exact scope of this work cannot be determined at this
point but an expenditure of $50,000 or less might be sufficient to eliminate this preferential pathway.
Once the pathway has been eliminated, contaminated soil and sediment can be delineated and removed
from the ROS; assuming that the amount of material to be removed is less than 500 cubic yards
remediation might be accomplished for about $150,000. Note these cost estimates were not rigorously
determined by the OSE team, and higher costs are possible (estimates that suggest higher costs should be
thoroughly reviewed).
23
-------
6.1.2 IMPROVE PLUME DELINEATION TO THE SOUTH AND VERTICALLY
Monitoring well clusters CW-4 and CW-5 include the southern most wells at the site, and they are
contaminated with PCP concentrations over 1,000 ug/L. An additional monitoring well cluster should
likely be placed further south with monitoring wells at the shallow and deep intervals. For cost-
effectiveness, wells at the intermediate elevations can likely be omitted. The optimization team suggests
placing this single cluster (with a shallow and deep well) approximately 250 feet south of CW-4. This
should place the well on Lawrence Road directly to the south of CW-4. It is expected that shallow and
deep monitoring wells at this location will yield samples with concentrations of PCP and dioxin that are
undetectable or below MCLs. This would delineate the plume to the south. If concentrations are low,
but above the MCLs, perhaps the southern boundary of the plume can be determined by extrapolation
between CW-4, CW-5, and this new cluster. If extrapolation is unacceptable, then the next likely place
to locate a monitoring well cluster is on Achille Road, directly south of CW-4 and the new cluster.
To delineate the plume vertically, deeper wells should be installed in multiple locations (perhaps up to
four locations). In addition to delineating the plume, these wells would provide monitoring points to
confirm that the final remedy (when implemented) prevents contamination from migrating at depth. Due
to the high concentrations found as deep as 50 feet below ground surface, it is likely that contamination
is also present from 50 to 60 feet and possibly from 60 to 70 feet. Therefore, for delineation, the
appropriate depths of the wells might be from 70 to 80 feet below ground surface, but a more appropriate
depth may be determined by the site team after further reviewing historical data and well logs. The well
logs from the construction of the extraction wells may be helpful. Alternatively, drilling techniques
where different depths are sampled as drilling occurs (to determine a final screen interval), using packers,
can be utilized.
Although the deeper portions of the extraction wells could be sampled to help determine the extent of
vertical contamination, the results would likely be inconclusive. For example, if these deeper samples
are contaminated, the contamination could be due to the following reasons:
• the lower portion of the screened interval is hydraulically connected to the upper portion and
contamination from shallower intervals is being drawn down during sampling
• the wells have historically acted as conduits for downward migration of contamination, but this
condition is not representative for other parts of the site
• contamination has migrated to this depth naturally and may be at this depth (or greater depths at
other parts of the site)
Because the results would be potentially inconclusive, sampling of the deeper portions of the extraction
wells is not recommended.
The following are suggested locations for installing deep wells.
Downgradient of the collector drain beyond the interpreted zone of capture (perhaps 100 to 200 feet
downgradient of the interpreted capture zone) - The collector drain captures contamination from the
overburden and likely from the shallower portions of the bedrock. A deep well would help determine if
contamination is present at a depth of 70 to 80 feet and migrating beneath the collector drain. To
evaluate capture at slightly shallower depths, this location would also be appropriate for a well screened
at a shallower interval (perhaps 40 to 50 feet).
24
-------
Co-located with CW-9 - CW-9D is approximately 50 feet deep and has PCP concentrations above MCLs.
In addition, CW-9D is downgradient of CW-4D and CW-5D, which have PCP concentrations as high as
5,200 and 1,700 ug/L, respectively. Co-locating a deep well (approximately 70 to 80 feet deep) with
CW-9D should vertically delineate contamination in this area. Furthermore, the enhanced CW-9 cluster
(i.e., with wells at three depths) should provide a downgradient monitoring point to confirm that the final
remedy (when implemented) provides capture of the contamination found at CW-4D and CW-5D. If
capture is adequate, concentrations at the three CW-9 wells should decrease to background (i.e., non-
detect) overtime.
Co-located with CW-13 - CW-13D is the deepest downgradient point with contamination above the
MCL. It had a concentration of 44 ug/L in October 2002. This deep well should provide vertical
delineation of this contamination. Furthermore, along with the new deep wells at CW-9 and
downgradient of the collector drain, this enhanced CW-13 cluster (i.e., with wells at three depths) should
provide downgradient monitoring points (assuming deep ground water flow is in the same direction as
ground water flow at shallower intervals) to enhance the capture zone evaluation of the final remedy
(when implemented).
Co-located CW-6D - Of the four proposed locations, this is the only location upgradient of the collector
trench. If samples from this new well have low or undetectable concentrations, then this location would
help vertically delineate contamination. In addition, over time it could be monitored to determine if
contamination is migrating downward.
In general, the proposed well locations have been chosen to both vertically delineate contamination and
to serve as performance monitoring points for the final remedy. The main presumption is that
contamination at depth will continue to migrate to the southeast (the deep hydraulic gradient can be
evaluated with these new wells) and that as long as these performance monitoring points remain clean or
decrease to background levels, then the plume is adequately contained and further vertical delineation is
not necessary. If this presumption is not sufficient justification, then an additional deep well could be
added at another location such as co-location with CW-2 or CW-3.
In summary, seven monitoring wells have been suggested. Two wells would be part of a cluster to the
south with a shallow and deep well (40 to 50 feet), one is 50 feet deep and downgradient of the collector
drain, and the other four are 80 feet deep. If contamination is found further to the south, an additional
well may also be needed. The depths suggested are preliminary, and drilling techniques where different
depths are sampled as drilling occurs (to determine a final screen interval), using packers, could be
utilized. The installation and two rounds sampling of the seven proposed wells should cost
approximately $120,000 to $150,000, including a work plan, oversight by a geologist, and a report
describing the delineation of the plume and a proposed target capture zone. Adding annual sampling
these wells to a sampling program would likely increase annual costs by up to $5,000.
6.1.3 EVALUATE PLUME CAPTURE ONCE PLUME is DELINEATED
Once the site team has sufficiently delineated the plume both to the south and vertically, a capture zone
analysis should be conducted. The capture zone analysis should include a clear statement of the target
capture zone and consist of converging lines of evidence such as those discussed in Section 4.2.2. These
lines of evidence include a water budget analysis, interpretation of capture using the potentiometric
surface maps under pumping conditions, and trend analyses in wells expected to be downgradient of the
capture zone.
25
-------
A ground water flow model with particle tracking could also be applied at this site to evaluate capture.
For evaluating horizontal capture this does not appear necessary, but for evaluating the vertical extent of
capture it may be helpful because the model could be used to help understand vertical ground water flow.
Furthermore, because additional extraction is likely necessary to provide adequate capture, a flow model
could be helpful in evaluating various pumping scenarios. This is particularly crucial since the treatment
plant has a limited capacity (under 40 gpm), and increasing this capacity significantly could be costly.
The estimated cost for the capture zone analysis without the model is approximately $20,000, including
documentation of this analysis in a report. The development of a well-calibrated flow model, and its use
in evaluating both capture and various extraction scenarios might cost $35,000.
6.1.4 TAKE MEASURES TO FURTHER REDUCE SYSTEM DOWNTIME
The maintenance problems associated with keeping three UV/Oxidation units operating continuously has
resulted in excessive downtime of the system. Using three units in series to meet effluent standards is
unnecessary and the site team is already considering bypassing one unit. Bypassing any number of the
units will allow a standby unit to be available in case of the unexpected but common downtime of one
operating UV/Oxidation unit. In addition, bypassing two units or all three units (i.e., using GAC for
primary organics treatment) may also be more cost-effective (see section 6.2.1). For the purpose of
reducing system downtime, we recommend that at least two of the units be bypassed. One UV/Oxidation
unit could be kept as a standby unit, and the third unit could be either kept at the site in reserve or
removed for use at another site. Even with additional flow likely from the final remedy one UV/oxidation
unit followed by GAC polishing will be a conservative system. The site should also maintain a sufficient
parts inventory to allow timely repair and/or replacement of system parts. Maintaining this inventory
will reduce the likelihood that downtime will occur due to backlogs in obtaining parts.
A few other issues contribute to downtime. One issue, which is discussed further in Section 6.3.2,
involves modifying the plant to provide storage of backwash waste water. The other pertains to the lack
of documentation associated with the system controls. A number of alarms cause system shut downs,
and in the absence of good documentation for these controls, the site team is in the process of learning
the logic behind the various alarms. This issue will be addressed over time given that the site team
continues its efforts to become familiar with the system.
There are no added cost to implementing these steps. There are likely cost reductions associated with
reducing the number of UV/Oxidation units, but these potential saving are discussed in Section 6.2.1.
6.2 RECOMMENDATIONS TO REDUCE COSTS
6.2.1 USE FEWER UV/OXIDATION UNITS
The site team is considering bypassing one of the UV/Oxidation units which will save about $20,000 per
year in electricity, $8,000 in lamps and accessories, and $4,000 per year in hydrogen peroxide. These
savings are already reflected in Section 4.4 of this report. Based on the May 2003 process sampling
event, greater than 99.9% of the organics mass removal is achieved in the first UV/OX unit.
26
-------
Parameter
Benzene
Trichloroethylene
Naphthalene
PCP
Phenanthrene
Maximum Influent
Concentration
9/2002 -6/2003 (ug/L)
12
15
1,800
9,500
400
5/2003
Influent
Concentration
(ug/L)
9.4
13
24
2,900+
25
Effluent After First
UV/OX Unit 5/2003
(ug/L)
ND
0.078 B
ND
2.3 J
0.02 B
NPDES Monthly
Average Effluent
Standard (ug/L)
1
5
30
1
3
B - contaminant found in laboratory blank
J - estimated value
Although the effluent of the first UV/Oxidation unit had a PCP concentration (2.3 ug/L) greater than the
discharge standard, the two 4,000-pound GAC units can easily reduce this concentration below that
standard without increasing the GAC replacement frequency. Furthermore, at 99.9% removal, one
UV/Oxidation unit would reduce 9,500 ug/L of PCP (i.e., the maximum influent concentration between
September 2002 and June 2003) to 9.5 ug/L, which can also easily be reduced below discharge standards
with the GAC units without affecting the GAC replacement frequency. Therefore, two UV/Oxidation
units can be bypassed without sacrificing effectiveness at an additional savings of approximately $32,000
per year relative to the costs reported in Section 4.4 of this report. As mentioned in Section 6.1.4 system
downtime can be reduced by using one of the bypassed units when the operating unit has failed or is
undergoing maintenance.
It is also possible to consider completely bypassing the UV/Oxidation units. For each of the organic
contaminants found in the plant influent above discharge limits, the following table presents the average
influent concentrations between September 2002 and June 2003, the estimated mass loading at a
maximum flow rate of 40 gpm, a common GAC adsorption isotherm, the ratio of GAC usage to mass
loading, and the estimated GAC per year usage assuming no UV/Oxidation units are operating.
Parameter
Benzene
Trichloroethylene
Naphthalene
PCP
Phenanthrene
Average Influent
Concentration
9/2002 - 6/2003
(ug/L)
8.72
9.86
277
3,936
2.63
Estimated Mass
Loading
at 40 gpm
(Ibs/year)
1.5
1.7
48.5
688.8
0.5
Isotherm*
0.016xC1/a4°
0.028xC1/a
-------
soluble and would be removed by removing particles to which it has already sorbed. Any dissolved
dioxin would be easily sorbed to GAC.
As is evident from the above table, it is estimated that less than 4,000 pounds of GAC would be required
per year to provide treatment in the absence of the UV/Oxidation units. At 40 gpm, a 4,000-pound GAC
unit would provide approximately 25 minutes of residence time, which should be more than adequate for
effective contaminant removal. Ideally, this would translate to replacement of the lead GAC unit once
per year. Practically, however, removal efficiency may decrease as fouling from solids increases.
Therefore, it is reasonable to assume that two replacements of the lead vessel may be required each year.
This is potentially one more unit replacement than anticipated with UV/Oxidation as the primary
treatment. Using GAC alone would therefore save an additional $15,000 per year (approximately) by
eliminating the costs for the final UV/Oxidation unit and the 2,000 pound PDU unit and including the
cost of an additional 4,000-pound GAC replacement.
Prior to fully eliminating the UV/Oxidation units, it is recommended that a thorough pilot test be
conducted. At this time, to avoid a major change to the treatment process, we recommend only bypassing
two UV/Oxidation units with continued consideration of bypassing the UV/Oxidation units completely
once a final remedy extraction system is determined.
6.2.2 EVALUATE AREAS TO REDUCE LABOR COSTS
The contractor has evaluated and implemented many improvements to the treatment system over the past
year. Monthly site meetings have clearly been very useful in expediting changes and keeping all parties
up to date. However, total labor costs for running this system are about $330,000 per year which is
higher than similar systems. As continued improvements make the system easier to operate, the site team
should have a goal of reducing operating labor costs and project management costs. For example, the
monthly site meetings and associated progress reporting can, within a year or two, be reduced to a
quarterly frequency, likely saving approximately $25,000 per year. Reporting and development of work
plans should also become easier over time because site activities will be better understood and less will
change on a monthly basis. As knowledge of the treatment plant increases and the need for system
improvements decreases, the level of engineering support (and associated costs) will decrease. The
operator will also be available to conduct other tasks such as NPDES and performance monitoring
sample collection, which is currently part of the sampling task. A total labor cost reduction goal of about
$40,000 per year from the current costs could be used within 1 to 2 years. Adding OU3 flow (if that
occurs as part of the OU3 ROD) may require some minor treatment system modifications and associated
management and labor costs initially but should not increase long term annual labor costs. The site team
has shown decreased labor costs between Year 1 and Year 2 of $25,000, demonstrating that labor costs
will decrease over time and that the suggested further decrease of $40,000 may be possible of the next
couple of years.
6.3 MODIFICATIONS INTENDED FOR TECHNICAL IMPROVEMENT
6.3.1 CONTINUE IMPROVING TREATMENT PLANT TO FACILITATE OPERATION AND
POTENTIALLY INCREASE CAPACITY
The treatment system is cramped and performing maintenance is difficult due to the lack of working
space. However, the contractor has already added level transmitters and sight tubes to monitor
operations and many other improvements to operations. The plant continues to have downtime problems
(see Section 6.1.4) and may need minor improvements to handle additional flow if pump and treat is
28
-------
selected as the remedy for OU3 and additional pumping is required. The first metals removal process
tank is currently the limiting unit for hydraulic capacity. Modifying or replacing this tank with a slightly
larger one should be considered. The other reaction tanks may also need modification. The head space
above the tanks and the general working space is limited and will complicate this task, but a solution
should exist such that the current treatment plant can accommodate the necessary flow for OU2, and
possibly OU3 if pump and treat is selected as the OU3 remedy. The tanks are all fiberglass reinforced
plastic. As an option, a fiberglass repair contractor (tank, pipe, marine, automotive, etc.) should be able
to modify the tanks to add freeboard. A section could be added to the top of the tank by using fiberglass
cloth and resin around the joint between the existing tank and the new section. The repair would be
watertight and, due to space limitations, would likely be easier than replacing the tank. Fiberglass repair
should cost less than $10,000, but tank replacement might cost up to $25,000.
6.3.2 MAKE PIPING CHANGES TO BETTER USE THE SECOND EQUALIZATION TANK
The equalization tank piping should be modified to include a line for potential influent flow directly to
equalization tank #2. Recycle lines should also be added so that minor maintenance can be accomplished
at equalization tank #1 without shutting down the extraction system. These piping changes should also
help the current equalization tanks provide some or all of the capacity necessary to store the wastewater
from backwashing. This effort should require less than $10,000 in capital costs.
6.4 CONSIDERATIONS FOR GAINING SITE CLOSE OUT
6.4.1 ADAPT P&T SYSTEM TO Focus PRIMARILY ON COST-EFFECTIVE CONTAINMENT
WITH DECREASED EMPHASIS ON RESTORATION
Ground water extraction as part of source removal efforts at RW-1 to RW-4 has had limited effectiveness
in terms of mass removal and progress towards restoration, and the existing treatment system limits the
volume of ground water that can be extracted to under 40 gpm. Rehabilitation of the recovery wells, or
other aggressive removal technologies may reduce mass but would not likely significantly shorten the
duration of the pump and treat system. A relatively large area with NAPL and high concentrations at
depth are present and would need to be addressed. Aggressive remediation (other than pumping at the
recovery wells) would therefore require substantial additional cost and interruption of above-ground
activities (e.g., along Eagle Road, at Swiss Farms, and at Young's Produce). The oil and grease that
comprises much of the NAPL would interfere with aqueous phase reactions (for dehalogenation or
oxidation) and would serve as a scavenger of chemical oxidation chemicals.
For the final remedy, the site team should consider focusing on hydraulic containment of a targeted
capture zone, even if it means reconsidering ground water extraction in the source area. LNAPL
collection by product skimming in select locations could be considered and some ground water removal
could be added over time if containment flow rates allow. No matter what ground water extraction
scheme is applied at the site, restoration to ground water cleanup standards will take many decades to
achieve and capture should not be compromised in the mean time to extract contamination from the
source area.
By focusing on containment only over a long period of time, monitoring, management, and data analysis
should be simplified, allowing for O&M costs to be "minimized" for an effective system. Ground water
monitoring, for example, would be used to demonstrate capture, and monitoring of wells within the heart
of the plume could be decreased in frequency or eliminated altogether. Ground water monitoring could
likely be reduced to sample 50 or 60 samples per year rather than 100 samples per year, yielding potential
29
-------
savings of approximately $30,000 per year. We suggest that EPA and State funds will be utilized in the
most effective manner by containing this site to eliminate the current risks to human health and the
environment. Based on the reported costs in Section 4.4 and the recommendations made in Section 6.2, a
target cost for long-term O&M of a containment-only system might be $500,000 or less per year to be
achieved within 1 to 2 years of the final remedy becoming operational and functional. After considering
the potential savings in Section 6.2, implementing this recommendation might save an additional $30,000
relative to the costs presented in Section 4.4 of this report.
6.4.2 POTENTIAL OPTIONS FOR IMPROVING CAPTURE
Capture of the fully delineated plume may require additional measures. To the south, there are elevated
concentrations at the CW-4 and CW-5 clusters and delineation is not complete. Downgradient of the
collector drain, there are PCP concentrations as high as 44 ug/L. The extent to which capture needs
improvement will not be fully known until completion of recommendations 6.1.2 and 6.1.3.
To the south, installation of another downgradient collection drain does not appear to be feasible due to
access considerations. Therefore, containment may require additional extraction wells south of the
existing drain. These wells will likely have to be screened across the shallow and deep zones to allow
the drawdown necessary to achieve flow rates required for containment. Pumped ground water from
these wells can likely be combined in a header line and added to the flow from the collection drain. The
1-inch pipe that restricts flow near the drain should be replaced. The current treatment system will likely
have adequate capacity for this additional flow, especially if the mixing tanks are modified as suggested
in Section 6.3.1. The additional extraction and should not significantly increase the annual O&M costs.
Downgradient of the collector drain, it is uncertain if remediation or capture is required. To date,
remediation efforts have not focused on the contamination at CW-9D and CW-13D. EPA appears to
have a few options for addressing this downgradient contamination.
• Monitor existing and suggested monitoring wells to determine if concentrations are naturally
attenuating
• Conduct temporary extraction events at this location to see if concentrations can be permanently
reduced
• Install permanent extraction capabilities near CW-13D and pipe that extracted water to the
upgradient extraction system (i.e., the trench and any new upgradient extraction wells)
• Inject nutrients or chemical reagents to enhance degradation of that contamination
Because this downgradient contamination appears fairly isolated, it would be difficult to monitor the
effectiveness of any of the above approaches without either relying on CW-9D and CW-13D alone or
installing additional monitoring wells. The decision on how to proceed may depend on the results of
sampling some of the recommended wells (see Section 6.1.2). Assuming no additional contamination is
found at depth, it might be most appropriate to monitor these downgradient wells (perhaps semi-
annually) for two years to determine if there is a trend. If there is a decreasing trend, it may be
reasonable to let that contamination attenuate. If there is persistent contamination, it may be reasonable
to attempt one year of quarterly temporary extraction events. If concentrations are increasing and/or the
temporary extraction events are unsuccessful, then either permanent extraction or the injection of
nutrients/reagents should be considered. Permanent extraction would likely require well over $100,000
in capital costs. The injection of nutrients/reagents might include lactate, molasses, or even nano-scale
30
-------
iron (ARS Technologies or PARS Environmental) to foster reductive dechlorination. It should be noted,
however, that the injection of nutrients/reagents (especially nano-scale iron) for enhancing the reductive
dechlorination of PCP is in a experimental stage. Therefore, if this approach is taken, appropriate testing
and piloting should be conducted prior to full-scale implementation. The costs for these options would
vary but would might be under $100,000 for a few injection events in a one or two existing wells over the
course of a year, especially if monitoring is limited (i.e., limited to the injection well and/or one
additional downgradient well).
6.5 SUGGESTED APPROACH TO IMPLEMENTATION
Addressing the abandoned sewer line preferential pathway (Section 6.1.1) is ongoing and is of the
greatest importance for risk minimization. These activities should be completed as soon as possible. If
the preferential pathway is eliminated plume delineation and completion of plume capture evaluation
(Sections 6.1.2 and 6.1.3) can be completed as part of the OU3 activities. We have identified several
delineation issues to consider in Section 6.1.2.
There is strong evidence to support bypassing two of the UV/Oxidation units (Section 6.2.1)
immediately. In addition to achieving cost savings without any negative impact on protectiveness
bypassing these units will improve the poor system downtime record (Section 6.1.4).
The remaining recommendations are related to or can be considered and implemented in concert with
selecting an OU3 remedy.
31
-------
7.0 SUMMARY
The evaluation team observed a conscientious and knowledgeable site team (i.e., EPA and its contractor)
that has effectively managed this interim remedy, provided a number of much needed plant
improvements, and initiated the remedial investigation for OU3. The observations and recommendations
contained in this report are not intended to imply a deficiency in the work of either the system designers
or operators but are offered as constructive suggestions in the best interest of the EPA and the public.
These recommendations have the obvious benefit of being formulated based upon operational data
unavailable to the original designers.
Recommendations to improve effectiveness include proceeding with the investigation and proper
abandonment of the sewer lines that led to downgradient contamination and remediation of the associated
soils and sediments. Other effectiveness recommendations include delineating the contaminant plume,
conducting a capture zone evaluation, and reducing system downtime. Specific items for each of these
recommendations are provided. Recommendations for reducing cost include bypassing two of the
UV/Oxidation units and reducing O&M labor. Recommendations for technical improvement include
modifying the metals removal mixing tanks to allow increased flow, and modifying the equalization tanks
to improve flexibility in operation and reduce downtime. Finally, recommendations associated with site
closeout include focusing the remedy on containment rather than restoration and a discussion of options
for potentially improving containment to the south and/or downgradient of the current extraction system,
if determined to be necessary after the recommended capture zone analysis is completed.
Table 7-1 summarizes the costs and cost savings associated with each recommendation. Both capital and
annual costs are presented. Also presented is the expected change in life-cycle costs over a 30-year
period for each recommendation both with discounting (i.e., net present value) and without it.
32
-------
Table 7-1. Cost Summary Table
Recommendation
6.1.1 Properly Seal
Abandoned 12-Inch Sewer
Line and Remediate Surface
Soils Near the Seep
6.1.2 Improve Plume
Delineation to the South and
Vertically
6.1.3 Evaluate Plume
Capture Once Plume is
Delineated
6.1.4 Take Measures to
Further Reduce System
Downtime
6.2.1 Use Fewer
UV/Oxidation Units
6.2.2 Evaluate Areas to
Reduce Labor Costs
6.3.1 Continue Improving
Treatment Plant to Facilitate
Operation and Potentially
Increase Capacity
6.3.2 Make Piping
Changes to Better Use the
Second Equalization Tank
6.4.1 Adapt P&T System
to Focus Primarily On
Cost-Effective Containment
with Decreased Emphasis on
Restoration
6.4.2 Potential Options for
Improving Capture
Reason
Effectiveness
Effectiveness
Effectiveness
Effectiveness
Cost
Reduction
Cost
Reduction
Technical
Improvement
Technical
Improvement
Site Closeout
Site Closeout
Additional
Capital
Costs
($)
$200,000
$120,000
to
$150,000
$55,000
$0
$0
$0
< $10,000
to
$25,000
$10,000
$0
Estimated
Change in
Annual
Costs
($/yr)
$0
$5,000
$0
$0
($32,000)
($40,000)
$0
$0
($30,000)
Estimated
Change
In Life-cycle
Costs
($)*
$0
$285,000
$55,000
$0
($960,000)
($1,200,000)
< $10,000
to
$25,000
$10,000
($900,000)
Estimated
Change
In Life-cycle
Costs
($)**
$0
$216,000
$55,000
$0
($516,000)
($646,000)
< $10,000
to
$25,000
$10,000
($484,000)
Contingent on results from implementing 6. 1.2 and 6. 1.3
Costs in parentheses imply cost reductions.
* assumes 30 years of operation with a discount rate of 0% (i.e., no discounting)
** assumes 30 years of operation with a discount rate of 5% and no discounting in the first year
33
-------
FIGURES
-------
FIGURE 1-1. THE HAVERTOWN PCP SITE AND THE SURROUNDING AREA.
GROUNDWATER
TREATMENT PLANT
CW-14S
*CW-14D
HAVERTOWN PCP
SUPERFUND SITE CAP
SWISS FARMS
YOUNG'S PRODUC
FOR WELL LOCATIONS IN
THIS AREA SEE FIGURE 1-2
LEGEND:
CW-7D EXISTING MONITORING
9 WELL
400
SCALE IN EEET
800
CW-7D
&
CW-7S
PHILADELPHIA
GUM COMPANY
\
COLLECTOR
TRENCH
CW-13'S
CW-13D
(Note: Tliis figure is developed from site figures provided by Tetra Tech, June 2003.)
-N-
NAYLOR'S RUN
CW-12S Jj
CW-12D ;fr
,// i
LOCATION OF IMPACTED
GROUNDWATER SEEPS
CW-15S
-------
FIGURE 1-2. AREA IMMEDIATELY SURROUNDING THE FORMER NWP FACILITY.
-N-
GROUNDWATER
TREATMENT PLANT
CW-1S
CW-11
CW-1D
NW-6-81
CW-21
<
HAVERTOWN PCP CW-2D
SUPERFUND SITE CAP
SWISS FARMS
CW-2S
RW-1
NW-1-81
LEGEND:
CW-3D EXISTING MONITORING
® WELL
MW-23
RECOVERY WELL
CW-3D
CW-3I
CW-3S
CW-8S
$
CW-8D
YOUNG'S PRODUCE / /'' HAV-02,
>^ \ RW-4/ / / PHILADELPHIA GUM COMPANY
R-4
CW-4S
CW-4I
CW-4D
CW-6I
CW-6D
EW-1
CW-6S
EW-2®
COLLECTOR
TRENCH /
«
NAYLOR'S RUN
HAV-04
PZ-'
PZ-2
HAV-05
HAV-07
CW-5S
CW-5I
CW-5D
CW-9D
CW-9S
0 125
SCALE IN FEET
250
CW-10D
(Note: This figure is developed from site figures provided by Tetra Tech, June 2003.)
-------
FIGURE 1-3. EXTENT OF PCP CONTAMINATION IN THE SHALLOW ZONE (APPROXIMATELY 5 TO 20 FEET BGS).
GROUNDWATER
TREATMENT PLAN
HAVERTOWN PCP
SUPEREUND SITE CAP
YOUNG'S PRODUCE-
LEGEND:
A_ FREE PRODUCT
^ >1,000 ppb
Q 100 ppb to 1,000 ppb
10 ppb to 100 ppb
o
O
SYMBOLS ARE BASED ON THE MAXIMUM DETECTED CONCENTRATION
FROM SAMPLING EVENTS BETWEEN 2000 AND MARCH 2003. EXTENT
OF PCP CONTAMINATION IS NOT NECESSARILY INDICATIVE OF EXTENT
OF DIOXIN CONTAMINATION.
400
SCALE IN FEET
800
LOCATION OF IMPACTED
GROUNDWATER SEEPS
(Note: Tliis figure is generated from base maps and sampling data provided by Tetra Tech.)
-------
FIGURE 1-4. EXTENT OF PCP CONTAMINATION IN THE DEEP ZONE (APPROXIMATELY 35 TO 60 FEET BGS).
-N-
GROUNDWATER
TREATMENT PLAN
HAVERTOWN PCP
SUPERFUND SITE CAP
-~^^ -~1,000 ppb
Q 100 ppb to 1,000 ppb
(^J) 10 ppb to 100 ppb
O <™ ppb
<>,
fery(
$r
o 1
K 1
^
0
s
LOCATION OF IMPACTED
GROUNDWATER SEEPS
SYMBOLS ARE BASED ON THE MAXIMUM DETECTED CONCENTRATION
FROM SAMPLING EVENTS BETWEEN 2000 AND MARCH 2003. EXTENT
OF PCP CONTAMINATION IS NOT NECESSARILY INDICATIVE OF EXTENT
OF DIOXIN CONTAMINATION.
400
SCALE IN FEET
800
(Note: This figure is generated from base maps and sampling data provided by Tetra Tech.)
-------
FIGURE 4-1. POTENTIOMETRIC SURFACE FOR SHALLOW GROUNDWATER UNDER PUMPING CONDITIONS (APPROXIMATELY 5 TO 20 FEET BGS).
CW-14S
o
o
INTERPRETED
CAPTURE ZONE
INTERPRETED
CAPTURE ZON
400
SCALE IN FEET
800
-N-
(Note: TMs figure is developed from a potentiometric surface map generated by Tetra Tech using March 2003 groundwater data.)
-------
FIGURE 4-2. POTENTIOMETRIC SURFACE FOR DEEP GROUNDWATER UNDER PUMPING CONDITIONS (APPROXIMATELY 40 TO 50 FEET BGS).
CW-7D
CW-14D
o
o
ro
^-®
CW-5D / CVJ~
flr
o
en
CM
INTERPRETED
CAPTURE ZONE-
------- |