United States
          Environmental Protection
          Agency
              Off ice of
              Emergency and
              Remedial Response
EPA/ROD/R03-91/109
December 1990
&EPA
Superfund
Record of Decision
          Whitmoyer Laboratories
          (Operable Unit 3), PA

-------
50272-101
REPORT DOCUMENTATION '- REPORT Na i
PAGE EPA/ROD/R03-91/109
4. TMeendSubMe
SUPERFUND RECORD OF DECISION
Whitmoyer Laboratories (Operable Unit 3) , PA
Third Remedial Action - Final
7. AI0IOM


•*• SponBonnQ OrQBflUBelon Nuiw mo ABBFOMS
U.S. Environmental Protection Agency
401 M Street, S.W.
Washington, D.C. 20460
XRedptenrs Accession No.
S. ReponOete
12/31/90
6.


10. ProlMvTHUWorkUnHNo.
11. CoMneUC)orOnm(G)No.
(G)
13. Type ol Report A Period Covered
800/000
M.

 16. M»»*ct(LMt:200»ord*)
   The 22-acre Whitmoyer Laboratories site is an inactive laboratory  facility in Jackson
   Township, Lebanon County, Pennsylvania.  Land use in the area  is predominantly
   agricultural with adjacent wetlands areas.  In addition, part  of the site lies within
   the 100-year floodplain of the  Tolpehocken Creek-Union Canal.  An  estimated 20
   residences in the vicinity of the  site use the underlying aquifer  as their drinking
   water supply.  From 1957 to  1964,  Whitmoyer Laboratories, Inc., produced organic
   arsenicals onsite.  In 1964, the new owners, Rohm & Haas, began storing concentrated
   wastes in a concrete vault,  and until 1971 conducted onsite ground water pumping and
   treatment and ocean dumping  of  wastes.  In 1977, sludge from ground water treatment
   was placed in lagoons in the eastern area of the site.  Between 1978 and 1982, the
   site changed ownership twice, and  then in 1985, a RCRA site closure plan was filed.
   In 1986,  EPA began providing bottled water to residents with ground water
   contaminated by site activities.   A public water supply line extension is currently
   being designed and will be constructed as part of a removal action.   When the site
   was abandoned in 1987, very  little of the RCRA closure plan had been implemented.
   From 1988 to 1990, EPA removed  approximately 800 abandoned drums and laboratory

   (See Attached Page)
 17. I
   Record of Decision - Whitmoyer Laboratories  (Operable  Unit 3),  PA
   Third Remedial  Action - Final
   Contaminated Media:   soil,  sediment, gw
   Key Contaminants:  VOCs (benzene, PCE, toluene, xylenes),  other organics  (phenols)
   o. kfcnmera/open-EncMTenne   metals (arsenic)
   c. COSATlFWd/Oroup

10. Security CteM (This Report)
None
20. Security O«ss(Thl*Pege)
None
21. No. of Page*
84
22. Pita*
(Set ANSU30.18)
                                    See DufructfofM on fltwrae
OPTIONAL FORM 272 (4-77)
(FormMty NtlS-H)
DoputrnMit of Conviwrov

-------
EPA/ROD/R03-91/109
Whitmoyer Laboratories (Operable Unit 3),  PA
Third Remedial Action - Final

Abstract (Continued)

wastes offsite.  A 1989 Record of Decision (ROD)  provided for remediation of hazardous
concentrated liquids, including laboratory wastes, abandoned at the site as Operable
Unit 1 (OU1).   A December 1990 ROD provided a remedy for vault and lagoon wastes,
miscellaneous  products and feedstocks, tanks, process vessels, and other onsite
structures as  OU2.  This ROD addresses contaminated onsite soil, sediment, and ground
water as OU3.   The primary contaminants of concern affecting the soil, sediment, debris,
and ground water are VOCs including benzene,  TCE, and PCE; other organics including PAHs;
and metals including arsenic.

The selected remedial action for this site includes  excavating and fixation of
approximately 61,000 cubic yards of soil/sediment with contaminant concentrations above
the "principal threat" action levels using an iron-based or other fixation process,
followed by offsite disposal; using biological treatment for approximately 5,600 cubic
yards of soil/sediment with organic concentrations above the "principal threat" organic
chemical action levels for the heavily contaminated soil either prior to or following
fixation, followed by offsite disposal; excavating and consolidating onsite in the vadose
zone approximately 39,000 cubic yards of soil/sediment having contaminant concentrations
below the "principal threat" action levels but above ground water-based unsaturated soil
clean-up targets, followed by capping with low-permeability materials; placing onsite in
the vadose zone approximately 16,000 cubic yards of soil/sediment with concentrations
below the ground water-based unsaturated soil action levels; soil capping any remaining
contaminated surface soil that contains arsenic concentrations greater than 21 mg/kg and
other disturbed areas, as needed; backfilling, grading, and revegetating the excavated
areas; demolishing onsite structures, followed by salvaging non-hazardous debris and
offsite disposal of unsalvaged debris; onsite pumping and treatment of contaminated
ground water using physical, chemical, and possibly biological treatment, followed by
either onsite discharge to surface water,  reinjection into the aquifer, or both methods;
disposing of any treatment residuals offsite; conducting long-term ground water
monitoring; and implementing Institutional controls Including deed restrictions for the
remaining contaminated areas.  Additionally,  this ROD provides a contingency for ground
water if it becomes technically impracticable to achieve cleanup goals.  The contingent
remedy includes pumping from the perimeter area only to prevent migration of the
contaminant plume.  The estimated present worth cost for this remedial action is
$77,300,000, which includes annual O&M costs of $2,397,600 or $2,477,600  (depending on
the ground water discharge option chosen).

PERFORMANCE STANDARDS OR GOALS:  Soil action levels for saturated and unsaturated soil
and "principal threat" action levels were developed based on the threat due to
inhalation/ingestion and to ground water using MCLs, RCRA toxicity characteristic levels
(TCLP), and health-based criteria.  The target clean-up goal for surface soil is
arsenic 21 mg/kg.  The action levels for unsaturated soil include benzene 0.009 mg/kg,
PCE 0.051 mg/kg, TCE 0.017 mg/kg, and arsenic 450 mg/kg.  This will reduce the excess
cancer risk level associated with inhalation/ingestion exposure to 10"^.  The action
levels for saturated soil include benzene 0.002 mg/kg, PCE 0.012 mg/kg, TCE 0.004 mg/kg,
and arsenic 210 ug/kg based on protection of ground water.  Principal threat action
levels include benzene 10 mg/kg, PCE 14 mg/kg, TCE 10 mg/kg, and arsenic 1,000 mg/kg.
Ground water clean-up goals are based on Federal MCLs, proposed MCLs, and a 10-6 excess
cancer risk level and reference dose equal to 1 for noncarcinogens.  Chemical-specific
ground water clean-up goals include benzene 0.005 mg/1  (MCL), PCE 0.005 mg/1 (pMCL),
TCE 0.005 mg/1  (MCL), and arsenic 0.05 mg/1.   A State ARAR to remediate ground water to
background levels will be waived because of technical impracticability.  If the
contingent remedy is employed, an ARAR to comply with SDWA MCLs will also be waived
because of technical impracticability.

-------
                        RECORD OF DECISION
                        OPERABLE  UNIT THREE
                   WHITMOYER LABORATORIES SITE

                           DECLARATION
 SITE NAME AND LOCATION

 Whitmoyer Laboratories  Site
 Lebanon  County,  Pennsylvania.
This  decision document presents the selected remedial action for
the  third operable  unit of  the Whitmoyer Laboratories  Site in
Lebanon  County, Pennsylvania.   This  document was  developed in
accordance   with  the  Comprehensive   Environmental  Response,
Compensation  and Liability Act of 1980 (CERCLA),  as amended  by the
Superfund Amendments and Reauthorization Act of 1986  (SARA); and,
to  the  extent practicable,   the  National  Oil  and  Hazardous
Substances  Pollution Contingency Plan  (NCP).   This  decision is
based on the  Administrative Record for this site.

The Pennsylvania Department of  Environmental Resources staff have
verbally concurred with the selection of this remedy.

ASSESSMENT OP THE BITE

Pursuant to duly delegated authority,  I hereby determine, pursuant
to Section 106  of CERCLA,  42  U.S.C.  Section 9606, that actual or
threatened releases  of hazardous  substances  from this  site,  as
discussed in  "Summary of Site  Risks,"  Section VI,  if not addressed
by implementing the response  action  selected in  this  Record of
Decision, may present an imminent and substantial endangerment to
the public health, welfare, or  the environment.


DESCRIPTION OP  THE REMEDY

This operable unit  (OU)  is  the third  of  three operable units for
the site.  The  first operable unit (OU One) at this site involved
remediation of hazardous concentrated  liquids which were abandoned
at  the  site.    The  second   operable  unit   (OU  Two)  involved
remediation of  concentrated wastes abandoned in a concrete  vault;
concentrated  wastes abandoned  in two  groups of lagoons; outdated

-------
products and miscellaneous chemicals abandoned in the buildings;
and the buildings and related structures (tanks, process vessels,
etc.)  located  on  the site.   This  action (OU  Three)  addresses
contaminated soils and adjacent sediments,  and groundwater present
at the site. These media pose some of the principal threats at the
site.

The U.S. Environmental  Protection Agency (EPA)  has selected, and
the Pennsylvania Department of Environmental Resources staff have
verbally concurred on  the selection  of the  following  Remedial
Action  for  the third operable  unit (OU Three)  of  the Whitmoyer
Laboratories Site.  This action addresses the materials making up
the third  operable unit by extracting  and treating contaminated
groundwater  to   the  extent  described   herein  and  the  most
contaminated  soils/sediments;  consolidating  and  capping  less
contaminated  soils/sediments;  disposing   the  solid  treatment
residuals offsite,  such that the materials will  not  require any
long-term onsite  management;  and  discharging  treated groundwater
to either Tulpehocken Creek or the aquifer beneath the site.


The  major   components of  the  Selected Remedial  Action are  as
follows:

     *    Excavation of all moderately contaminated soils/sediments
          from  offsite and saturated  onsite  locations,  and all
          heavily   contaminated  onsite   and/or   offsite  soils
          [estimated volume « 116,000 cubic yards (CY)].

     *    Demolition of Buildings 4, 9,  11, and 14.

     *    Backfilling of  the  excavated  areas with  clean fill or
          lightly contaminated soil.

     *    Onsite fixation of the approximately 61,000 CY of heavily
          contaminated soils/sediments using an iron-based or other
          similar   fixation  process  that  provides  equivalent
          protection.

     *    Biological  treatment  of the  approximately  5,600  CY of
          soils  with organic  chemical  concentrations  above the
          heavily contaminated soil  action  levels either prior to
          or following the fixation  step.

     *    Consolidation    of    the    moderately    contaminated
          soils/sediments  on site above the groundwater table.


     *    Capping  of the  approximately 39,000  CY  of moderately
          contaminated    soils/sediments   having    contaminant
          concentrations above groundwater-based  unsaturated soil
          cleanup targets  with  low-permeability materials.

-------
      *     Soil  capping of  all soils/sediments  remaining on the
           surface following the excavation and consolidation steps
           that  are not capped  with  low-permeability materials and
           contain  greater  than  21  mg/kg   arsenic;  and  other
           disturbed areas as needed.

      *     Grading and revegetation of all areas affected by the
           soil/sediment remediation.

      *     Following the soils/sediments remediation, placement of
           deed restrictions on areas with  remaining contamination.

      *     Aggressive extraction of all groundwater from the aquifer
           beneath the site with concentrations above health-based
           levels  (e.g..  50   ug/1  arsenic)  until  the  maximum
           groundwater contaminant concentrations are all  less than
           health-based levels.

      *     Treatment  of the  extracted groundwater  in an onsite
           treatment   plant,  utilizing   physical,  chemical  and
           possibly biological  processes.

      *     Disposal of the treated water  by either discharging it
           to Tulpehocken Creek,  reinjecting  it into the  aquifer,
           or a combination of  the two methods.

      *     Salvaging nonhazardous demolition debris, as feasible.

      *     Disposal  of the  following in  offsite  landfill(s)  in
           accordance with all  applicable  regulations: all treated
           soils;  the groundwater treatment  residuals;   and  the
           demolition debris that is not salvaged.

The selected remedy is the  last of several phases in the long-term
remediation of  this  site and  will  be consistent with previously
selected site remedies.

It may potentially prove technically impracticable to achieve the
health-based groundwater cleanup  goals  under the selected remedy
for the groundwater.   If  information emerges  from the operation of
the  selected remedy  system that  strongly  suggests  that  it  is
technically impracticable to achieve the  cleanup goals throughout
the   contaainated  groundwater  plume   because  of  an   observed
"leveling-off"  of   contaminant  concentrations,   the  EPA,   in
consultation with  the Commonwealth of Pennsylvania,  intends  to
implement  a  contingent remedy in those  areas where  the cleanup
goals will not  be met.   The contingent remedy is similar to the
selected remedy, with the exception that groundwater would only be
extracted  in sufficient quantities to keep the non-attainment area
from growing.

-------
STATUTORY DETERMINATIONS

Both the selected remedy and the contingency remedy are protective
of human health and the environment, and are cost-effective.  EPA
believes that both the selected and contingent remedies will meet
all  Federal and  State  Applicable or  Relevant and  Appropriate
Requirements with the sole exception of the State action-specific
requirement to remediate groundvater to background concentrations.
Accordingly, I hereby waive the provisions of 25 PA Code, Chapter
75, Part 264 with respect to groundwater backgroud concentrations
due to technical impracticability.  Both remedies utilize permanent
solutions  and  alternative treatment technologies to  the maximum
extent  practicable  and  satisfy   the   statutory  preference  for
remedies which  employ  treatment that  reduces toxicity, mobility,
or volume as a principal element.

Because both the selected remedy and the contingency remedy for the
third operable unit will result in hazardous substances remaining
onsite above health-based levels,  a review under Section 121(c) of
CERCLA, 42 U.S.C.  9621 (c)  will be conducted within five years after
the  commencement  of remedial  action to  ensure that  the  remedy
continues to provide adequate  protection  of human health and the
environment.
Edwin BT'Erickson                  Date
Regional Administrator

-------
                               RECORD OF DECISION
                          WHITMOYER LABORATORIES SITE

                                DECISION SUMMARY
                    I.  SITE NAME, LOCATION, AND DESCRIPTION

 A.     SHE NAME AND LOCATION

 The Whitmoyer  Laboratories  Site (Site) is  located on  approximately 22 acres in  Jackson
 Township, Lebanon County, Pennsylvania, about 1 mile southwest of the Borough of Myerstown
 (see Figures 1 and 2).  The she lies between the Union Canal of Tulpehocken Creek and the
 Conrail (Reading) Railroad.  Fairlane Avenue forms the site's eastern boundary, while Creamery
 Street adjoins the site to the west.

 A food storage warehouse is  active in Building 18 on the site.  Land surrounding the site is
 predominantly farmland, with  scattered farmhouses.   A  Sterling Drug  factory is  located
 2,000 feet east of the site, while PJ Valves, a manufacturing plant,  is located about 1,500 feet
 to the south.  A large active limestone quarry, locally referred to as the Calcite Quarry,  is
 located approximately 1.5 miles west of the site.

 B.     TOPOGRAPHY.  SURFACE WATER. AND DRAINAGE

 Topographic relief on the site is moderate, varying in elevation from 493 feet in the southwest
 corner to 449 feet rn the northeast corner. The entire site drains to Tulpehocken Creek, with
 drainage being  roughly perpendicular to the Creek axis. Portions of the site are within the
 100-year flood plain of  Tulpehocken Creek-Union Canal.

 The Union Canal branches from Tulpehocken Creek just west of the site and rejoins the Creek
 near the site's eastern boundary. Myerstown is the first downstream community, at a distance
 of approximately 3/4th of a mile.  Tulpehocken Creek is a tributary to  and joins the Schuylkill
 River  near Reading, Pennsylvania.  The Schuylkill River flows into the Delaware River, which
 eventually empties into  the Atlantic Ocean. Tulpehocken Creek and the Schuylkill River serve
 as drinking water supplies and irrigation sources downstream of the site. The headwaters  of
 the section of Tulpehocken Creek which passes by the site originate approximately 3 miles  to
 the northwest.

 C.     GEOLOGY

The Whitmoyer Laboratories Site is located within the Lebanon Valley,  part of the Great Valley
 portion of the Valley and Ridge Physiographic Province. The valley is a  topographic expression
of the underlying, relatively easily eroded carbonate bedrock units. The site is underlain by
carbonate bedrock of the Ontelaunee Formation, the youngest member of the Ordovician Age
 Beekmantown Group.  A thin mantle of clayey residual soil overlies bedrock in the site vicinity
 Depths to bedrock in the site vicinity range from 0-19 feet, based on the Remedial Investigation

-------
BASE MAP IS A PORTION OF THE USOS RICHLAND.PA QUADRANGLE (75 MINUTE SERIES. I93S,PHOTOREVISEO 1969) CONTOUR INTERVAL 2O'
                                                                                    FIGURE   1
                                 LOCATION MAP
            WHITMOYER LABORATORIES SITE. LEBANON COUNTY. PA
IMUS

-------
SOURCE- GENERAL LAYOUT DRAWING, PREPAREDNESS, PREVENTION B CONTINGENCY PLAN, DEC 22, 1982
                         QENERAL_ ARRANGEMENT

              WHITMOVER LABORATORiES SITE! MYERSTOWN. PA
                                                                             SCALE M fEET
                                                                                FIGURE   2
    JIMUS
	l_l CORPORATON

-------
 (Rl).  The depth to bedrock is greatest in the vicinity of Tulpehocken Creek and the Union
 Canal.

 The Ontelaunee Formation is described in regional literature as a light to dark gray dolomite,
 which weathers to a dark grayish brown.  The Ontelaunee Formation strikes N60°E to N80°E
 predominancy, with an overall dip to the SE of approximately 30°. In the Myerstown area, this
 formation is approximately 500 feet thick.

 Soils in the area are primarily residual soils derived from weathering of the bedrock surface,
 with some alluvium adjacent to Tulpehocken Creek.  Based  on the Rl,  the soils  consist
 predominantly of silt and clay.  A thin veneer of organic-rich topsoil overlies the residual soils
 throughout much of the area. Fill material is present in several locations within the site property
 boundaries.

 D.     HYDROGEOLOGY

 The carbonate bedrock units underlying the Lebanon Valley form the major aquifer in the area
 The various formations present, although differing somewhat in water-yielding  capacity, are
 considered to form a single, large, heterogeneous, unconfined aquifer.   The  porosity of the
 carbonate aquifer  is almost entirely secondary,  with fractures  enlarged  through solution
 channeling forming the primary groundwater storage zones and migration pathways.

 Groundwater flow directions in the region generally follow topography, then follow stream flow
 direction  in valley  bottoms.  In the site area, portions  of the groundwater  flow both  m
 northeasterly and southeasterly directions, before generally following the course of the stream
 to the east-northeast.  Depth to groundwater ranges from  2 to  21 feet below  land  surface  at
 the site.

 Recharge to  groundwater in the carbonate  rack units  is  principally through precipitation
 infiltration, with additional recharge due to groundwater migration from adjacent rock units, and
 occasional surface water recharge during extended dry periods.

 Groundwater beneath the site is classified as a Class 2A aquifer, a current source of drinking
water.  The groundwater is used for both potable and industrial water supplies.  Approximately
 40 residences in the site vicinity have potable water supply wells tapping the aquifer.  Twenty
of these residences have been  placed on bottled water by EPA due to contamination of their
water supply from the site activities.  Large industrial users of groundwater include Sterling
 Drug, Inc., Quaker Alloy Casting Co., and P.J. Valves Company.

The  Myerstown  Water Authority  (Authority)  provides  potable water to  the  residents  of
Myerstown. One of the Authority's reserve wells, No. 8, taps the bedrock aquifer underlying
the site.  This  well is utilized during periods of high demand. To date, contamination from the
site has not been detected in this well.

An extension to the Myerstown Water Authority's water line has been designed by EPA for
those residents in the vicinity of the site whoses wells have been shown to contain arservc
contamiation.  The Whitmoyer Laboratories Private Study Group (WLPSG), a group of frome'

-------
site owners, have entered into a consent agreement with EPA to provide for the construction
of this extension.
E.     CLIMATOLOGY

The Whitmoyer Laboratories Site is located within the southeastern Piedmont Climatologies)
Division of Pennsylvania. Second Mountain, which rises 1,500 feet along the north border, and
South  Mountain, which rises 1,000 feet along the southern border, form the Lebanon Valley,
in which the site is located.  The Lebanon Valley has a humid continental climate. Due to the
valley's location, weather systems are typically modified  before reaching Lebanon  County.
Weather extremes are most often the result of unusually strong weather systems.

The average annual precipitation at the site is 42.3 inches. This  precipitation is mostly evenly
distributed throughout the year, with  slightly less precipitation occurring in the  winter.  The
average annual snowfall is 27 inches.   Evaporation at the  site is 36.3 inches;  thus,  net
precipitation is 6 inches.

In the summer, high temperatures are generally in the mid-80s and the lows near 60°F   During
the winter the highs average in the upper 30s and the lows in the 20s.  The prevailing wind is
from the northwest in  winter and from the west-southwest in summer.

F.     POPULATION AND ENVIRONMENTAL RESOURCES

Lebanon County, according to the 1980 census, has a population of 109,829, and is classified
by the Commonwealth of Pennsylvania as a "5th Class" county. The population of Myerstown
in 1984 was 3,270.  Populations  of 1,296 and 4,683 reside within 1 and 3 miles of the site,
respectively.

Portions of Tulpehocken Creek (Creek) adjacent to the  site contain very small open water
wetlands areas  consisting of small pockets along the riverine system of the Creek and Union
Canal. Fioodplain forest wetlands exist starting approximately 3.5 miles downstream of the site
The area has some habitat value, with opossum, raccoon, numerous fish, a water snake, and
various songbirds observed during a 1986 EPA site visit.

Tulpehocken Creek  has been proposed for inclusion on the Commonwealth of Pennsylvania's
scenic river system, with a  "priority 1A status."  This designation is for streams which "have
the most urgent need  for protection and immediate need for additional study," according to a
Pennsylvania Department of Environmental Resources (PADER) official.

                 II.  SITE HISTORY AND ENFORCEMENT ACTIVITIES

A brief chronology of site history  and enforcement activities follows.

1900 Circa   -An oil pipeline was constructed across the site.
1934     -    Whitmoyer Laboratories, Inc. (WLI) formed.
1957     -    WLI begins production of organic arsenicals.

-------
1964
1971
1977
1978
1982
1984

1985

1986

1987
1988
1989
1990 January

1990 February -
1990 May
1990 June
        Rohm & Haas buys WLI.  Concentrated wastes placed  in a concrete
        vault.  Groundwater pump-and-treat program initiated. Ocean dumping
        of wastes begins.
        Groundwater pump-and-treat and ocean dumping program terminated.
        Sludges from groundwater treatment consolidated in eastern lagoons.
Beecham Laboratories acquires WU.
Stafford Laboratories, Inc. purchases WLI.
Stafford Laboratories,  Inc. files  for bankruptcy.  Whitmoyer Laboratories Site
proposed for the National Priority List (NPL).
WLI files a RCRA Closure Plan with PADER, and changes its RCRA status from
a Treatment, Storage, or Disposal facility to a generator facility.
Whitmoyer Laboratories Site finalized on the NPL  EPA begins providing bottled
water to area residents with contaminated wells.
Stafford Laboratories, Inc. abandons facility, with very little, if any, of the RCRA
Closure Plan implemented.  EPA initiates the Remedial Investigation/Feasibility
Study (RI/FS).
EPA initiates an emergency response to remove abandoned drums from the site.
This work  continues into the summer of  1990.
EPA selects a  remedy for the concentrated liquids operable unit. Abandoned
laboratory wastes  are packaged  and  disposed by  EPA.  The Whitmoyer
Laboratories Site Rl Report is finalized.  Clarence W. Whitmoyer, former president
of WLI,  dies.  U.S.  Department  of Justice  files claim  against estate in  Dade
County, Florida.
        The concentrated  liquids  (first) operable  unit Remedial  Design  is
        completed.
        The Whitmoyer Laboratories Site FS report, which addresses the media
        making  up  the second operable  unit,  as  well  as  the groundwater
        medium, is finalized.  Two former site  owners,  Rohm  &  Haas and
        SmithKline Beecham, propose to EPA a separate remedial alternative for
        the  vault wastes, lagoon wastes,  miscellaneous products/feedstocks.
        contaminated soils/sediments, and  groundwater.
        The concentrated liquids Remedial  Action commences.
        The FS report which addresses the soils/sediment medium, is finalized
        The two former site owners, Rohm & Haas  and SmithKline Beecham.
        propose to  EPA a second, separate unique remedial alternative for the
        vault wastes, lagoon wastes, and miscellaneous  products/feedstocks
1990 September-
1990 September
1990 December
        The concentrated liquids Remedial Action is completed. Two. The two
        former site owners, Rohm &  Haas and SmithKline Beecham, provide
        additional information to EPA on their separate remedial alternatives, and
        propose a separate remedial  alternative for the site structures.

        Rohm and Haas and SmithKline Beecham enter into consent order with
        EPA under which they will extend public water services to residents
        affected by the Site.

        EPA selects a remedy for OU Two.
                                         8

-------
                       III.  COMMUNITY RELATIONS HISTORY

In accordance with Sections 113 and 117 of CERCLA, 42 U.S.C. Sections 9613 and 9617, EPA
held a public comment period from July 16, 1990 through September 14, 1990 for the third
operable unit Remedial Action described in the Remedial Investigation (Rl) and Feasibility Study
(FS) Reports released in April 1990, and the Soils Feasibility Study Report and Proposed Plan
released in July  1990.  The notice of availability of these documents was published in the
Harrisbura Patriot on  July 16, 1990 along with notice of the comment period and a public
hearing to be held concerning the third operable unit. The Rl and FS Reports and the Proposed
Plan were made available to the public in the Administrative Record  maintained in the EPA
Region  III office and at the Myerstown Public Library.  A public meeting was held on August
1, 1990 to outline the Preferred Remedial Action and to accept comments from  the attendees.
A transcript  of the public meeting was maintained in accordance with Section 117(a)(2) of
CERCLA, 42 U.S.C. Section 9617(a)(2). Written and verbal comments were received and are
addressed in the Responsiveness Summary which is attached.

All documents that form the basis for the selection of the remedial decisions contained in this
Record of Decision are included in the Administrative Record for this site and can be reviewed
or referred to for additional information.

                     IV. SCOPE AND ROLE OF OPERABLE UNIT

As with  many Superfund sites, the problems at the Whitmoyer Laboratories Site are complex
As a result, EPA is addressing portions of the site contamination using its emergency response
authorities, whereas  other portions are being addressed as a part of the remedial  program

A.      EMERGENCY RESPONSE ACTIONS

The approximately 800 drums and the laboratory wastes and chemicals  and production run
samples abandoned at the site were disposed as an  emergency response action.  A public
water supply line extension to residences with contaminated wells is currently being designed
and will be constructed as an emergency response action.  While the line is being designed.
affected residences are being supplied by EPA with bottled water.

B.     OTHER REMEDIAL ACTIONS

EPA has divided  the remaining remedial work into three operable units (OUs).  These are as
follows:

      *      OU One:     Concentrated liquids abandoned in tanks and process vessels

      *      OU Two:     Vault wastes, lagoon wastes, miscellaneous products/feedstocks.
                          and site structures

      *      OU Three:    Contaminated soils/sediments and groundwater

-------
EPA has already selected the cleanup remedy for OUs One and Two. The concentrated liquids
(OU One) pose a principal threat at  the site, because  of the potential for  direct contact;
tank/piping failure with subsequent contamination of Tulpehocken Creek; fire/explosion; and
tank failure from flooding.  This action  is in the Remedial Action stage.  This remediation was
completed in September 1990.

The concentrated wastes abandoned in a concrete vault; concentrated  wastes abandoned in
two groups  of lagoons; outdated products  and miscellaneous chemicals abandoned in the
buildings; and the buildings and related structures (tanks, process vessels, etc.) located on the
site (OU Two materials) also pose some of the principal threats to human health and the
environment from the site, because of  the following risks: possible ingestion or direct contact
with the materials; contaminant migration from the materials into the underlying groundwater
that is a source of drinking water for  local  residents; and contaminant migration to surface
water. A Record of  Decision  for OU  Two was signed by the EPA on December 17, 1990
Major features of this selected remedy include incineration followed by fixation of the high-
organic-content  (upper) vault  wastes, hazardous miscellaneous products/feedstocks, and
hazardous, combustible site structures; fixation of the low-organic-content (lower) vault wastes
and lagoon wastes; direct landfilling of the nonhazardous miscellaneous products/feedstocks
and site structures; surface cleaning of hazardous, incombustible, impermeable site structures,
and coating and sealing hazardous, incombustible, permeable site structures.  The Remedial
Design for OU Two is expected to commence in the near future.

The third OU addressed by this ROD includes contaminated soils and adjacent sediments; and
groundwater.  These  media also pose some of the principal threats to human health and
environment from the site.  The  purposes of this response for the soils/sediments are to prevent
current or future exposure to the these materials through treatment and/or containment, and
to reduce the migration of contaminants from the soils/sediments to groundwater and surface
water. The purposes  of this  response for groundwater are to clean  up the contaminated
groundwater to  health-based concentration levels if technically practicable, and to prevent
current or future exposure to groundwater exceeding health-based concentration levels. The
remedy for the third OU is currently  proposed as the final response action for the site.

                        V. OPERABLE UNIT CHARACTERISTICS

The site materials to  be remediated under OU Three are described as follows:

A.     CONTAMINATED SOILS/SEDIMENTS

As discussed in the Whitmoyer  Laboratories Site Rl Report, numerous chemicals and hazardous
substances were detected  in the contaminated soils/sediments in and  around the plant site
In the baseline risk assessment for the soils/sediments, arsenic contamination was determined
to  present  the  greatest  risk for the  exposure  scenarios   studied.   Other  primary  soil
contaminants  identified  include aniline, n-nitrosodiphenylamine,  tetrachloroethene  (PCE).
trichloroethene    (TCE),    total-1,2-dichloroethene,   benzene,   pyrene,   benzo(a)pyrene
benzo(b)fluoranthene, and  indeno(1,2.3-cd)-pyrene.  All of these contaminants are known or
                                          10

-------
 probable carcinogens.  Arsenic, PCE. and total-1,2-dichloroethene are also considered to be
 systemic toxicants.

 Contaminated surface soils at the Whitmoyer Laboratories Site pose an actual or potential threat
 of inhalation/ingestion  exposure if no remediation occurs.  EPA has determined that cleaning
 up the  contaminated surface soil to a concentration of 21 mg/kg of arsenic will reduce the
 excess  lifetime  cancer risk  posed  by the surface soils following remediation under the
 residential use (soil ingestion/inhalation) scenario to 1 x 1CT6.  This 21  mg/kg cleanup target
 was established for this site as part of the risk assessment conducted during the RI/FS.

 Surface and subsurface soils also pose a threat to groundwater.  The threat differs depending
 on whether the soils are above or below the groundwater table ("unsaturated" or "saturated",
 respectively).  EPA has determined that the soils should be  remediated to ensure that migration
 of soil contaminants to groundwater would not cause groundwater contaminant concentrations
 to exceed the groundwater cleanup targets cited below. Groundwater-based  cleanup  targets
 for saturated and  unsaturated soils are provided in Table  1.

 For the Whitmoyer Laboratories  Site, EPA has determined that soils/sediments which contain
 at least one contaminant whose leachate concentrations  would likely exceed 100 times the
 groundwater  cleanup targets are the principal threats from the  soil/sediment medium.  The
 corresponding soil concentrations are provided in Table 1. Nearly all of the "principal threat"
 (heavily contaminated) soils/sediments contain arsenic  in concentrations  greater than 1000
 mg/kg, the arsenic action level for treatment. Soils with arsenic concentrations greater than this
 level  will  likely exhibit the RCRA characteristic of arsenic toxicity based on  a statistical
 correlation between  total arsenic concentrations in soil and  arsenic concentrations in the
 resulting leachate. This correlation was developed  by EPA using the standard TCLP  testing
 procedures and is fully described in the Feasibiltiy Study report for this site.

 For this ROD, contaminated soils/sediments are defined as the contaminated soils and adjacent
 sediments that are contaminated with arsenic and/or organic chemicals above  remedial action
 levels.  The baseline risk assessment indicated that contaminated sediments alone do not
 appear to constitute  a  significant risk to human health and the environment.  As  a result, the
 only sediments addressed by this ROD are those sediments bounded by  soils contaminated
 above the remedial action levels.

The Rl surface soil data indicating arsenic and organic chemical contamination are presented
 on Figs. 3 and 4, respectively. These data demonstrate widespread onsite and offsite arsenic
contamination. Surface organic chemical contamination appears to  be limited to  onsite soils
The Rl subsurface soil data indicating arsenic and organic chemical contamination  in the 2-
foot to 6-foot depth interval are presented on Figs.  5 and  6, respectively.  On Figs.  7 and 8.
the Rl subsurface  soil  data indicating arsenic and organic chemical contamination at  depths
greater than 6 feet are  presented, respectively. The subsurface soil data indicate that arsenic
and organic   chemical contamination are  present in subsurface  soils  onsite.    Arsenic
contamination is relatively widespread in offsite subsurface soils, whereas only a limited degree
of subsurface  soil  organic chemical contamination in offsite areas was noted  during the Rl
                                          11

-------
                                TABLE 1
REMEDIAL ACTION LEVELS FOR CONTAMINATED SOIL/SEDIMENTS AND GROUNDWATER
                      WHITMOYER  LABORATORIES SITE
                     LEBANON COUNTY, PENNSYLVANIA
Parameter
Arsenic
Benzene
Trans- 1,2-dichloroethene
Tr ichloroethene
Tet rachloroethene
Pyrene
Benio(a)pyrene
Benzo(b)f luoranthene
N-nitrosodiphenylaaine
Aniline
lndeno(l ,2,3-cdlpyrene
1 , 1 -Diehloroethene
Methylene chloride
4-Chloroaniline
Soil Ingostion-BasedU)
Action Lav* Is (eg/kg)
Residential Use
lBlO-6 BCR(3)
21
14
>1,000<*)
38
8.2
1.6
0.13
0.93
85
73
0.56
-
-
-
Grounduater(2)
Contamination-Based Soil Action Levels (ng/kg)
No Dilution
(Saturated
Soils)
210(8)
0.002
0.037
0.004
0.012
0.033(0.66|(6)
2.1
0.72
0.030(0.66)16)
0.002(0.66)16)
4.2
-
-
-
Underflow
Dilution (Vadose
Zone Soils)
450(8)
0.009
0.16
0.017
0.051
0.140(0.66)(6)
8.9
3.1
0.13(0.66)(6)
0. 009(0. 66)(6)
18
-
-
-
Principal
Threat H)
1.000(8)
10*
200*
10*
14*
3.3
210
72
14*
12"
420
-
-
-
Groundwater (?)
Action Levels
(•9/1)
0.05
0.005
O.l(P)
0.005
O.OOS(P)
-
0.0002(P)
0.0002(P)
-
0.006" (0. 010)1')
0.0004(P)
0.007
0.005**
0.14**

-------
      1
IEMEDIAL ACTION  LEVELS  FOR CONTAMINATED SOIL/SEDIMENTS AND GROUNDWATER
fHITMOYER LABORATORIES  SITE
.EBANON COUNTY,  PENNSYLVANIA
•AGE TWO


I)  Based on models  used  in the HI,  action levels calculated for  commercial  use,  agricultural  use,  or
    limited construction  at the site are higher  than the residential use  scenario.
21  MCL or  10-« excess cancer  risk  in  GW,  converted to  soil  concentrations using  Koc and TOC  (=0.65%)
    distribution  based calculations.  Underflow  dilution equal  to 4.25
3)  ECR  Excess Cancer Risk
«)  Based  on leachate concentrations  exceeding RCRA  toxicity  characteristic  levels,  if  available, oc
    100 times  the 10-6 Excess  Cancer Risk  level  in GW.   Action levels  indicated by  an  * were  derived
    using TCLP  methodology.  All others  are  based on equilibrium  partitioning  using Koc and  TOC.
*>)  Noncarcinogen, Dose/RfD based (=1).
6)  Number shown  in  parentheses is practical quantitation (detection) limit.
')  MCL based  standards,  unless  indicated by a  •*.   A **  indicates that  no MCL  or  proposed MCL (P) is
    available.   The value  presented is  then based  on  a 1 x 10-* excess  cancer  risk  or a  Dose-Rfd ratio
    equal to 1.
i>  Arsenic levels are based on Rl-measured  partitioning data.

-------
                                                                                       LEOEND
                                                                                  r-	»
                                                                                  v	/
                                                                                  C5J^ iKMimi.
                                                                                        Mt*l«XMU
                                                                           **>'£• ' yjSZS "H««'» Uf ron W««H
                                                                                                   flCURE
SURFACE SOILS - ARSENIC ISOCONCENTRATION CONTOURS
    WHITMQYER LABORATORIES SITE. LEBANON CO. PA
• rtn
IMUS
OORMORATOM

-------
                                                                                                LCOENO
                 •OTf •  I MntwC co«loo«> inc rmitmiio ran
                       «fMUM nimriuiiOM ruoom

                      f (U. KOCOMCtlltRlTlOX MM.MS Ml *ve« M
                                                                                                KM
                                                                                               MOO
                                                                                               MM.OOO
                                                                                                           FICUHC   4
   SURFACE SOILS - ORGANIC CONTAMINATION

WHjTMQYER ^ABORATORtES SITE. LEBANON CO. Pft

-------
                                                                                                LEGEND
                                                                                   *<>T<' * CWTOM nwM.ru Mf ran
                                                                                        nutnuto. MB nmunwr
SUBSURFACE SOILS (2 TO 6 FEET) -ARSENIC ISOCONCENTRATION CONTOURS

                      LAeQRATQQjES SITE. LEBANON CO. PA
tout • fin
IMUS

-------
                                                                                               LEGEND
                                                                                              CONMLMTDll
                                                                                              CKXMItO IMOOH
                                                                                            ,  Mta (xauoco nan
                                                                                                  MMjri
                   f 
-------
                                                                                                        lEOENO
                                                                                            MOTE- i i
                                                                                                  •.miration MO ctntMTON
                                                                                                I M.L oooOMCf »m«no« wuic$ MC
                                                                                                                  |««/>«I
SUBSURFACE SOILS (GREATER THAN 6 FEET I -ARSENIC ISOCONCENTRATION CONTOURS

               WHITMOYER LABORATORIES SITE. LEBANON CO. Pft
*CMM m HIT
                                                                                                                  FtQUHE   7
IMUS

-------
                                                                                                  LECCNO
                                                                                             	f ICMIM i

                                                                                            *%%? "f*1
                                                                                                  — •«•
                       •MUM. U.MTMHOD
                      I. «*. *oew«c»i«tiuno«. M4.MU Mt
                       M«.tmmn rtn «iLO*nu< l-«/»«l
                                                                                                   ; en err-.-—,
                                                                                      1*0
SUBSURFACE SOILS (GREATER THAN 6 FEET) - ORGANIC CONTAMINATION
         WHITMOYER LABOHATOTOES SITE. LEBANON CO. PA
                                                                                                             FIGURE   g
IMUS

-------
 The estimated volumes of soils/sediments having at least one contaminant whose concentration
 exceeds the remedial action levels are provided in Table  2 for each  exposure scenario
 described above. An estimated 480,000 CY of soils/sediments present at the site have arsenic
 concentrations in excess of the background (1x10*6 excess lifetime cancer risk) concentration
 of 21  mg/kg. These soils  are located on an estimated 46 acres. Approximately 61,000 CY of
 soils/sediments contain at least one contaminant  whose concentration is  greater than the
 principal threat action levels.  Of this amount,  about 5,600  CY of the principal  threat soils
 contain contaminants exceeding the organic-based principal threat action  levels.

 B.     GROUNDWATER

 During the Rl, numerous contaminants were also detected in the groundwater in and around
 the plant site.  In the baseline risk  assessment for groundwater, arsenic  contamination was
 determined to  present the greatest risk for the exposure scenarios studied.  Other primary
 groundwater  contaminants  identified include antimony,  cadmium,  manganese,  aniline,  4-
 chloroaniline, PCE, TCE, total-1,2-dichloroethene, 1,1-dichloroethene, methylene chloride, and
 benzene. Arsenic,  aniline, PCE, TCE,  1,1-dichloroethene, methylene chloride, and benzene
 are classified as known or probable carcinogens.  Arsenic, antimony, cadmium, manganese,
 4-chloroaniline, PCE, total-1,2-dichloroethene, and 1,1-dichloroethene are classified as systemic
 toxicants.

 Contaminated groundwater at the Whitmoyer  Laboratories Site poses an  actual or potential
 threat to human health and the environment if no  remediation occurs. EPA has determined that
 contaminant concentrations in groundwater at  the Whitmoyer Laboratories Site  should be
 cleaned  to Federal  and State drinking water standards, where they are available for the
 particular contaminants.   These health-based  standards reduce  the risks  posed by  the
 contaminants to  acceptable levels.  When there  are no Federal or state drinking  water
 standards, EPA has determined that the excess lifetime cancer risk posed by each contaminant
 following remediation should be reduced to 1 x 10^.  This cancer risk level would  reduce the
 probability of contracting cancer as a result of direct exposure to  these contaminants  in the
 groundwater to one additional person  in one million, which is an acceptable level.  These
 cleanup targets were established for this site as part of the risk assessment conducted during
 the RI/FS.  Using this approach, EPA determined that cleaning up contaminated groundwater
 to the concentrations of 50 ug/l arsenic and 10 ug/l aniline will be protective of human health
 and the environment.  All groundwater containing other contaminant concentrations  above nsk-
 based levels have arsenic  and/or aniline concentrations above the levels cited above (i.e , the
 extent of contaminated groundwater is adequately defined by the aniline and arsenic criteria)

The estimated area! extent of contaminated groundwater using the  arsenic and aniline criteria
 is presented on Figure 9 and totals 215 acres.  The depth of contamination was not completely
defined during the Rl; an assumed depth of 500 feet from ground surface was used during the
 RI/FS.   This  assumption  results in an  estimated  volume of contaminated groundwater of
350,000,000 gallons.  The  estimated average arsenic, aniline  and PCE concentrations in this
groundwater are  17 mg/l, 6.4 mg/l, and 0.25 mg/l  respectively.   The estimated dissolved
quantities of these contaminants are 44,000 Ibs,  20,000 Ibs, and 730 Ibs, respectively.  These
quantities reflect only the dissolved portion of the  contaminants in the groundwater.  Substantial
amounts of the groundwater contaminants may also be adsorbed onto clays found within the

                                         12

-------
                                               TABLE 2
                       COMPARISON OP SOIL/SEDIMENT VOLUMES AND QUANTITIES OF
                   CONTAMINANTS CORRESPONDING TO EACH  SOIL REMEDIAL ACTION LEVEL
                     HHITMOYBR LABORATORIES  SITE,  LEBANON COUNTY, PENNSYLVANIA
Parameter
olume (Cubic Yards)
real Extent (Acres)
rsenlc Quantity (pounds)
emi -volatile Organics
uantity (pounds)
olatile Organics Quantity
pounds )
Soil Ingest ion-
Based Action Level
Residential Use
480,000
46*
560,000
9,800
78
Groundwater Contamination-Based
Soil Action Levels
No Dilution
(Saturated Soils)
189,000
28
516,000
9,800
78
With Underflow
Dilution (Vadose
Zone Soils)
100,000
15
440,000
4,700
45
Principal Threat
61,000
10
373,000
3,900
8
   Area  is not adequately defined
:R  Excess Cancer Risk

-------
                                                           tt€fi>\^
Wi
©
e/-*
                                                                                    OP
                                                                               or CIIMDIT
                                                                           AHILIHI CONTkMIHATIO«
                                                                     CMOUHDWATIR.  CASED OH LIMITED
                                                                  KVAILMLI OPPSITt MTk.
                                                                                  FKMJftC  9
   LOCATION OP CONTAMINATED GROUNOWATER

WHITMOYER LABORATORIES SITE. LEBANON COUNTY. PA
  IMUS

-------
bedrock fractures, and organic chemicals such as PCE and aniline could also be present in
the bedrock as nonaqueous phase liquids.

                            VI.  SUMMARY OF SITE RISKS

The objective  of this section is to estimate  the  potential for adverse health or environmental
effects incurred by human or ecological receptors exposed to the materials making up the OU
Three under the exposure scenarios established in the Rl Report for the Whitmoyer Laboratories
Site. This section characterizes the potential noncarcinogenic, carcinogenic, and environmental
risks associated with OU Three. EPA guidelines for the use of dose-additive models are used
to combine the risks for individual chemicals to estimate cumulative risks for the mixtures found
onsite, assuming the toxicological endpoints are the same.  This section summarizes the  risk
assessment presented in the Whitmoyer Laboratories  Site Rl Report, which was finalized in
November 1989.

A.     HUMAN HEALTH RISKS

For human health risks, both carcinogenic risk and the potential for noncarcinogenic effects are
presented.  Carcinogenic risk is evaluated by  determining the excess lifetime  cancer risks
(ELCRs) for actual or potentially exposed individuals. ELCRs are determined by multiplying the
contaminant exposure dose with the cancer potency factor (cancer slope factor). These risks
are probabilities that are generally  expressed in scientific notation  (e.g.. 1  x  1CT6). An ELCR
of 1  x 1CT6 indicates that, as a plausible upper bound, an individual has a one-in-one million
chance  of developing cancer as a result of site-related exposure to a carcinogen  over a 70-
year lifetime under the specific exposure conditions at a site.

Cancer  potency factors (CPFs) have been developed by EPA's Carcinogen Risk Assessment
Verification Endeavor workgroup for estimating lifetime cancer risks associated with exposure
to potentially carcinogenic chemicals. CPFs, which are expressed in units of (mg/kg-day)'7. are
multiplied by the estimated intake of a potential carcinogen, in mg/kg-day, to provide an upper
bound estimate of the ELCR associated with exposure at that intake level. The term "upper
bound"  reflects the conservative estimate of the risks  calculated from the CPF.  Use of  this
approach makes  underestimation  of the actual  cancer risk  highly  unlikely.  Cancer  potency
factors  are derived from the  results of human epidemiological studies  or  chronic animal
bioassays to which animal-to-human extrapolation  and uncertainty factors have been  applied

Potential concern for noncarcinogenic effects of a single contaminant in a single  medium is
expressed as  the hazard  quotient (HQ) [or the ratio of estimated intake derived from  the
contaminant concentration in a given medium to the contaminant's reference dose (RfD)). The
HQ is also referred to as the Dose/RfO ratio.  By adding the HQs  for all contaminants within
a medium or across all media to which a given population may reasonably be exposed, the
Hazard Index (HI) can be generated. The HI provides a useful reference point for gauging the
potential significance  of multiple contaminant exposures within a single medium or across
media.

Reference doses  (RfDs)  have been developed by EPA for indicating the potential for adverse
health effects from exposure to chemicals exhibiting noncarcinogenic effects.  RfDs, which are

                                          13

-------
expressed in units of mg/kg-day, are estimates of acceptable lifetime daily exposure levels for
humans, including sensitive individuals.  Estimated  intakes of chemicals from environmental
media (e.g. the amount of a chemical ingested from contaminated drinking water) can be
compared to the RfD. RfDs are derived from human  epidemiologicat studies or animal studies
to which uncertainty factors have been applied (e.g.. to account for the use of animal data to
predict  effects  in humans).  These uncertainty factors help ensure that the  RfDs will not
underestimate the potential for adverse noncarcinogenic health effects to occur.

The following risk summary is presented by medium for the OU Three media.
1.     Soils/Sediments

The contaminant concentrations and exposure pathways for the contaminated soils sediments
are briefly  described  above in Section  V.A. above.  The  major exposure pathways include
accidental ingestion/inhalation  by future site residents, present or future site workers, or farmers
cultivating  or  pasturing animals on fields adjacent  to the  site  presently or in the future;
consumption of crops/beef grown on/pastured on fields adjacent to the site presently or in the
future; and present or  future consumption of  groundwater contaminated by soil/sediment
leachate.

A conservative accidental ingestion/inhalation exposure scenario for residential use of the site
was developed.  Key assumptions include that children and  adults would ingest 200 mg/day
and 100 mg/day of soil, respectively.  Based on this scenario, an HQ of 470 and an ELCR of
1.1  x 10° was calculated for a reasonable worst-case exposure to the arsenic in soils from the
drum  burial area of the site.  Thus, under the conditions of the risk assessment, adverse
noncarcinogenic  health effects are possible (since  the  HQ is greater  than  1), and the
carcinogenic risk is greater than the CERCLA acceptable ELCR of between 1  x  1 Cr4 and 1  x
10'6.

Similarly, a conservative accidental ingestion/inhalation exposure scenario was developed for
commercial/industrial  use  of the site.  Key assumptions of this scenario include that  adults
would  ingest 100 mg/day of soil for 165 days/year and a 40-year working lifetime.   Based on
this scenario, an HQ of 18.2 and an ELCR of 1.87 x 1CT2 was calculated for a reasonable worst-
case exposure to soils from the drum burial area of the site.  Thus, adverse noncarcinogenic
health  effects are possible and there is a significant excess lifetime cancer risk under the
conditions  of the risk assessment.

Farm  workers tilling fields containing contaminated surface soils  adjacent to  the site can
potentially  inhale soil particulates.  A conservative inhalation model for this exposure was
developed  in the baseline  risk assessment.   Key assumptions of this model include that the
workers would inhale 1.3 cubic meters per hour of particulate-laden air for 12 hours/day and
10 days/year of a 40-year working lifetime.  Based on this scenario, an ELCR of 3.0 x 1CT2 was
calculated for an exposure to  average soil arsenic concentrations in soils from the Grumbine
field immediately north of the Whitmoyer  Laboratories property.  Thus, there is a significant
excess lifetime cancer risk under the conditions of the risk assessment.

                                           14

-------
 Consumption of vegetables grown on contaminated soils and  consumption of beef or dairy
 products raised on/produced on contaminated pastureland also pose potential risk to human
 health.  These exposures were also modeled in the baseline risk  assessment. Key assumptions
 for  the vegetable  consumption  scenario   include  a soil-vegetable partition   coefficient
 (contaminant concentration in plant/contaminant concentration  in soil) of 0.2, a 50 gram/day
 consumption rate, and a 70 kilogram (kg) receptor body weight.  Key assumptions for  the
 scenario involving consumption of beef or dairy products produced from animals pastured on
 contaminated fields include a soil-vegetation partition coefficient of  0.2, a  product-vegetation
 partition coefficient (contaminant concentration in milk or beef/contaminant concentration in soil)
 of 0.01  for beef and 0.001 for dairy products, and a 70 kg receptor body  weight.   Based on
 these scenarios,  an HQ of 13.8 and an ELCR of 2.1  x 1CT2  was calculated for  vegetable
 consumption; an  ELCR of 1.9 x 10^ was calculated for beef consumption; and an ELCR of 7
 x 10~5 was calculated for milk consumption for products  produced on soils  from the  Grumbine
 field having average soil arsenic concentrations. Thus, adverse noncarcinogenic health effects
 are possible and  there is a significant excess lifetime cancer risk  under the conditions of the
 risk  assessment.

 Precipitation that  has contacted contaminated soils could run off  and potentially contaminate
 surface water.  The surface-water pathway was not quantitatively analyzed  in the baseline risk
 assessment.

 Contaminated soils  are  contributing to  the  groundwater  contamination  at  the Whitmoyer
 Laboratories  Site  and would continue  to  contaminate groundwater  in  the  future  if  left
 unremediated.  The groundwater pathway is further discussed below.

 2.     Groundwater

 As identified in the Rl and described briefly in Section V.B, the groundwater at the site and
 downgradient of the site is highly contaminated.  Peak concentrations of arsenic and orgamcs
 (eg., tetrachloroethene) measured in the groundwater exceed Safe Drinking Water Act (SDWA)
 Maximum  Contaminant  Levels  (MCLs)  of  50 ug/l arsenic and 5  ug/1  tetrachloroethene
 (proposed) by a factor of about  3,000.  ELCRs and His for the onsite/near-site groundwater
 (residential use-reasonable worst case scenario) approach unity and exceed 6000, respectively
This  scenario assumes residential use of the  site and consumption of the most contaminated
groundwater at a rate of 2 liters/day for  70 years by a  70-kg  adult.  The risk  data indicate
potential adverse  carcinogenic and noncarcinogenic effects under this exposure scenario

B.     ENVIRONMENTAL RISKS

Based on  the aquatic biota  survey and fish tissue sampling  conducted  during the  Rl.  no
evidence of impacts on  the  ecosystem  from the site  was  observed.  Fish tissue arsenic
concentrations were below 2 mg/kg, the method detection limit. Sensitive benthic species. e.g..
stoneflies and mayflies, were found in downstream waters of Tulpehocken  Creek.  (There are
no endangered species or  natural resources of special concern in the vicinity of the site)
Thus, contamination from the materials making up OU Three do not appear to  be  impacting
the ecosystem currently.  As heavily contaminated groundwater continues to migrate otfsile over

                                          15

-------
time, contaminant contribution to surface  water  and sediment could potentially  affect  the
ecosystem in the future if no remediation occurs.

In summary, actual or threatened releases of hazardous substances from the materials making
up OU Three, if not addressed by implementing the response in this ROD, may present an
imminent and substantial endangerment to  public  health, welfare, or the environment.

              VII.  DESCRIPTION AND COMPARISON OF ALTERNATIVES

Based on the Rl risk assessment for the OU Three materials, EPA developed the following
remedial action objectives to protect human health and the environment:

       1.    Prevent human exposure (dermal contact, ingestion, inhalation) to soils/sediments
             having contaminants in concentrations greater than carcinogenic (ELCR greater
             than 1 x 10"6) and noncarcinogenic (Hazard  Index greater than  1) risk-based
             levels.

       2.    Prevent human exposure (dermal contact, ingestion, inhalation) to groundwater
             having contaminants in concentrations greater than MCLs (e.g.. 50 ug/l arsenic
             and 5 ug/l PCE), where available, or, if MCLs are not available, carcinogenic
             (ELCR greater than 1  x 1CT6) and noncarcinogenic (Hazard Index greater than
             1) risk-based levels.

       3.     Prevent migration (via  leaching) of contaminants in soils/sediments that would
             result in groundwater  contamination in excess of MCLs (e.g.. 50 ug/l arsenic
             and 5 ug/l PCE), where available, or carcinogenic/noncarcinogenic risk-based
             cleanup levels (ELCR greater than 1 x lO^/Hazard Index greater than 1).

       4.     Prevent migration of contaminants in soils/sediments (via runoff, flooding, erosion)
             or groundwater (via groundwater discharge) that would result in  surface-water
             contamination in excess of the more stringent of the Pennsylvania Water Quality
             Standards (e.g.. 50 ug/l arsenic) or Federal Ambient Water Quality Criteria, where
             available; or carcinogenic/noncarcinogenic risk-based cleanup levels.

       5.     Restore groundwater  contaminant concentrations to the MCLs  (e.g..  50 ug i
             arsenic), where available, or  carcinogenic/noncarcinogenic risk-based cleanup
             levels (ELCR greater than 1 x 1CT6/Hazard Index greater than 1) as soon as is
             technically practicable.

       6.     Comply with chemical-specific, location-specific, and  action-specific applicable
             or relevant and appropriate regulations (ARARs), including MCLs and RCRA lane
             disposal restrictions.
Based on data available in the Rl and FS Reports, the following OU Three materials will need
to be remediated to achieve the remedial action objectives:
                                          16

-------
       *      Contaminated Soils/Sediments - Estimated volume is 480,000 cubic yards.
       *      Contaminated Groundwater - Estimated volume is 350,000,000 gallons

The Superfund process requires that the alternative chosen to clean up a hazardous waste site
meet several criteria. The alternative must protect human health  and the environment, be cost-
effective, and meet the requirements of environmental  regulations.  Permanent solutions  to
contamination  problems  should be developed wherever possible.   These solutions should
reduce  the  volume,  toxicity, or mobility  of the contaminants.   Emphasis is also placed on
treating the  wastes at  the site,  whenever this  is possible,  and  on applying innovative
technologies to clean up the contaminants.

EPA studied  a variety  of technologies to see  if  they were applicable for use on the
contaminated soils/sediments  and groundwater.   The technologies  determined to  be most
applicable  to these  materials  were developed into  remedial alternatives.  Because  of the
different nature of each of these media, separate remedial alternatives for each medium were
developed.  These individual alternatives are presented and discussed below.   The remedial
alternatives developed  by the former site owners and  presented to the EPA are also described
and discussed.

A.     SUMMARY OF ALTERNATIVES - SOILS/SEDIMENTS

Alternatives  1 through 8 for the contaminated soils/sediments are numbered to correspond with
the numbers in the soils FS report (7/90).   Alternative 9 is the alternative presented by the
former site owners.   The alternatives are the following:

    Alternative 1: No Action
    Alternative 2: Soil  Capping
    Alternative 3: Consolidation/Clay Capping
    Alternative 4: Bulk Excavation/Offsite Landfill
    Alternative 5: Bulk Excavation/Fixation/Offsite Landfill
    Alternative 6: Bulk Excavation/Biological Treatment, Fixation/Offsite Landfill
    Alternative 7: Bulk Excavation/Incineration, Fixation/Offsite Landfill
    Alternative 8: In-situ Vitrification
    Alternative 9: Soil  Flushing

1. Alternative 1: NO ACTION

The Superfund Program requires that the "no action" alternative be evaluated at every site  to
establish a baseline  for comparison with the other alternatives. Under this alternative,  EPA
would take no actions other than performing reviews every 5 years. Under this alternative, there
would be no  deed restrictions or any other institutional controls.   Alternative 1 would not
comply  with the  groundwater relevant  and appropriate  Safe  Drinking  Water Act (SDWA)
Maximum Contaminant Level  (MCL) of 50  ug/l arsenic, and possibly not with the  pertinent
relevant and appropriate MCLs for organic chemicals. Alternative 1  would also not comply with
the CERCLA preference for a  remedy that  employs treatment to reduce  toxicity, mobility,  of
volume as a principal element.  While no capital costs would be incurred under this alternative
                                          17

-------
annual operation & maintenance (O&M) costs are estimated to be $3,600. This alternative has
a present-worth cost of $56,000. and can be implemented immediately.

2.   Alternative 2: SOIL CAPPING

Under Alternative 2, the entire area of surface soil contamination (arsenic concentration greater
than 21 mg/kg) (40 acres) would be left in place and capped with 1.5 feet of clean soil  and 6
inches of topsoil.  Erosion control  measures (e.g.. riprap) would  be applied to the stream
channel. These measures would require maintenance indefinitely.  Deed restrictions would  be
placed on areas where contaminated soils/sediments remain following remediation; and long-
term groundwater monitoring in compliance with the relevant and appropriate requirements of
40 CFR 264.117 and 5-year reviews would be conducted.  Alternative 2 would not comply with
the SDWA MCL for arsenic, and possibly not with the pertinent MCLs for organic chemicals
Alternative 2 would also not comply with the CERCLA preference for a remedy that employs
treatment to reduce toxicity, mobility, or volume as a principal  element. The estimated capital
cost of this alternative is $4,440,000.  Annual O&M costs are estimated to be $7,600. The
estimated present-worth cost of this alternative is $4,450,000. The estimated time to Implement
this alternative is approximately 2 years.

3. Alternative 3: IMPERMEABLE CAPPING

Under  Alternative 3,  Buildings  4, 9,  11, and 14 would be demolished and the pipeline/pump
station  would  be  temporarily abandoned  or relocated to  facilitate the  excavation  of
contaminated soils/sediments.  The demolition debris would either be  landfilled offsite m
compliance  with  all applicable regulations  or salvaged.   Saturated soils/sediments with
contaminant concentrations exceeding the groundwater-based soil action levels for saturated
soils and unsaturated soils with contaminant concentrations exceeding the groundwater-based
soil action levels for unsaturated soils would be consolidated in the vadose zone onsite The
excavation areas would  be backfilled with clean fill or lightly contaminated soil.   During
excavation of the saturated soils and stream  sediments, the groundwater table would  be
lowered by groundwater pumping, and Tulpehocken Creek/Union Canal would be temporarily
relocated.  The  extracted contaminated  groundwater would  be treated in the  groundwater
pump-and-treat system,  if present, or in a temporary  treatment system.  The  extracted
uncontaminated  groundwater would be  directly discharged to the creek.  The final stream
channel would  be located in  a similar position and with similar dimensions  as the current
channel.  Erosion control measures  would be implemented during the stream relocation
process. The onsite canal lock would be archived during remediation and would be reinstalled
at the  completion of activities.

Following consolidation of the excavated soils/sediments, the approximately 100.000 CY of soils
with contaminant concentrations above the  groundwater-based unsaturated soil action levels
would  be impermeably capped (e.g..  with clay). The cap would be designed to meet  the
relevant and appropriate RCRA landfill closure requirements in 40 CFR 264.310, which, among
other things, specify that the permeability  of  the  cap must  be less than or equal  to  the
permeability of the natural underlying materials at the site.   All  surface soils that are  not
impermeably capped and contain greater than 21 mg/kg arsenic would be capped with soil
All affected areas would be graded  and revegetated.

                                          18

-------
Following implementation of Alternative 3, deed restrictions would be placed on areas where
contaminated soils/sediments  remain.    Consistent  with  the  relevant  and  appropriate
requirements  of  40 CFR 264.117, long-term  O&M would  be  conducted to monitor the
groundwater around the consolidated wastes and to ensure the integrity of the cap, and 5-
year reviews would be conducted. Alternative 3 should comply with all applicable or relevant
and appropriate requirements (ARARs).

Alternative 3 would not comply  with the CERCLA preference  for  a remedy  that employs
treatment to reduce toxicrty, mobility, or volume  as a principal element. The estimated capital
cost of this alternative is $8,400,000.  Annual O&M costs are estimated to be $7,600.  The
estimated present-worth cost of this alternative is $8,300,000. The estimated time to implement
this alternative is approximately 2 years.

4. Alternative 4:  BULK  EXCAVATION/LANDFILL

Under Alternative 4, some or all of the contaminated soils would be  excavated and disposed
offsite.   The approximately 61,000 CY  of soils/sediments with contaminant concentrations
exceeding the "principal threat' action levels (see Table 1) would be disposed in an offsite
landfill in accordance with all applicable regulations. Since nearly all of these soils exhibit the
toxicity characteristic for arsenic, they would be disposed in a hazardous waste landfill.  The
approximately 39,000 CY of soils/sediments which contain contaminant concentrations less than
the "principal threat" action levels but greater than the groundwater-based unsaturated soil
action levels (see Table 1) would either be disposed offsite in an intermediate landfill (Options
A or B) or be consolidated in an onsite vadose zone and covered by an impermeable cap
(Option C). The impermeable cap would be designed to meet the  relevant and appropriate
requirements of  RCRA landfill closure in 40 CFR 264.310.  The balance of the  contaminated
soils would either be taken offsite to a less secure, nonhazardous landfill (Option A) or remain
onsite.   If these  soils  are left onsite  (Options  B and  C), saturated soils with contaminant
concentrations exceeding the  groundwater-based saturated soil  action levels (see Table 1)
would  be relocated to the vadose zone onsite. Following this consolidation, surface soils with
arsenic concentrations  greater than 21 mg/kg arsenic but with contaminant concentrations less
than the groundwater-based unsaturated soil action levels would be covered by a soil cap

Other activities required to implement this alternative include demolition of Buildings 4, 8, 9,11
and 14 and temporary abandonment or relocation of the pipeline/pump station. The demolition
debris would either be landfilled in  an offsite  landfill in accordance with all applicable
regulations or salvaged.  During excavation of the saturated soils and stream sediments, the
groundwater table would be lowered by groundwater pumping, and Tulpehocken Creek/Union
Canal  would be temporarily relocated.  The extracted  contaminated groundwater would be
treated in the groundwater pump-and-treat system, if present, or in a temporary treatment
system. The extracted uncontaminated groundwater would be directly discharged to the creek
The final stream channel would be located in a similar position and with similar dimensions
as the current channel. Erosion control measures would be implemented during the stream
relocation process. The onsite canal lock would  be archived during remediation  and would be
reinstalled at the completion of activities.  The excavated areas would be backfilled with clean
fill, covered with soil, graded, and revegetated.

                                          19

-------
 Option  A would comply with RCRA  landfill clean closure requirements, 40 CFR Part 264.
 Subpart N.  Therefore, deed restrictions would not be required for this option.  For Options B
 and C, deed restrictions would be used to control access to the contaminated soils remaining
 at the site.  Since contaminants remain onsite under Options B and C,  5-year reviews would
 be conducted.   Long-term groundwater monitoring  in compliance with the  relevant and
 appropriate requirements of 40 CFR 264.117 would also be conducted following implementation
 of Option C, since materials which pose a potential threat to groundwater would remain on site.

 None of the Alternative 4 options would comply with RCRA land disposal restrictions (LDRs,  40
 CFR Part 268).  These restrictions require RCRA hazardous wastes, such as the soils/sediments
 exhibiting the arsenic toxicity characteristic, to be treated prior to placement in a landfill. These
 restrictions apply to the soils/sediments after May 8, 1992.  Since it would be impossible to
 landfill the hazardous soils/sediments prior to this date (because of the time required for stream
 and pipeline relocation), this ARAR would not be met.  Alternative 4 complies with all other
 ARARs.

 Alternative  4 would not comply with  the CERCLA preference for a remedy that employs
 treatment to reduce toxicity, mobility, or volume as a principal element. The estimated capital,
 annual O & M, and net present-worth  costs associated with this  alternative are $82,000,000.
 34,000/year,  and  $80,000,000,  respectively, for  Option  A; $40,000,000, $7,600/year, and
 $39.000,000, respectively,  for Option  B;  and $34,000,000, $7,600/year, and  $33,000.000.
 respectively,  for Option C.  The  estimated time to implement this alternative is approximately
 2 years.

 5.  Alternative 5: BULK  EXCAVATION/FIXATION/OFFSITE LANDFILL

 Under Alternative 5, all saturated soils/sediments with contaminant concentrations exceeding
 the groundwater-based action levels for saturated soils would be excavated, as would all of
 those unsaturated soils with contaminant concentrations exceeding the "principal threat" action
 levels  (see  Table  1).    Additionally, the  unsaturated  soils/sediments with   contaminant
 concentrations  exceeding the groundwater-based action levels for unsaturated soils but less
 than the "principal threat" action levels would be  consolidated in  the vadose zone onsite  To
facilitate the excavation of these approximately 116,000 CY of contaminated soils/sediments
 Buildings  4, 9, 11,  and 14 would  be demolished and the  pipeline/pump station would  be
temporarily abandoned or relocated.  The demolition debris would either be landfilled off site in
compliance  with all applicable  regulations  or salvaged.  The excavation  areas would  be
 backfilled with clean fill or lightly contaminated soil.  During excavation  of the saturated soils
 and stream sediments,  the groundwater table would be lowered by groundwater pumping, and
Tulpehocken Creek/Union Canal would be temporarily relocated.  The extracted contaminated
 groundwater would be treated in the groundwater pump-and-treat system, if present, or m a
temporary treatment system.  The extracted uncontaminated groundwater would be directly
 discharged to the creek. The final stream channel would be located in a similar  position and
with similar  dimensions as  the current  channel.   Erosion control  measures  would  be
 implemented during the stream relocation process. The onsite canal lock would be archived
during remediation and would be reinstalled at the completion of activities.
                                          20

-------
 The approximately 61,000 CY ol 'principal threat' soils/sediments would be treated using iron-
 based  fixation or  a similar  process.  The  approximately  5,600 CY of  "principal threat"
 soils/sediments with organic  chemical  concentrations above the "principal threat" organic
 chemical soil action levels would also be fixated with activated carbon.  Following treatment,
 the  arsenic mobility of the treated wastes would  be reduced by  at least 90%.  The fixation
 would occur onsite in accordance with RCRA standards for miscellaneous treatment units (40
 CFR Part  264, Subpart  X).   The fixation  unit would be mobilized,  operated, and closed
 according to the requirements of 40 CFR 264.600, et sea.  The specific operating parameters
 for the fixation process would be determined in the Remedial Design phase through engineering
 design and analysis and the  competitive bidding process.  Because most of the "principal
 threat"  soils/sediments exhibit the RCRA characteristic of arsenic toxicity and Alternative  5
 constitutes treatment, RCRA Subtitle  C  is  applicable.  The RCRA land disposal restriction
 treatment standard  for these wastes is 5.0 mg/l arsenic, as measured by the EP Toxicity Test
 or TCLP.  (A national capacity extension for these wastes is in effect until May 8, 1992.) The
 fixation process should achieve this treatment standard.  The treated soils/sediments should
 no longer be RCRA characteristic wastes as the fixation process would prevent these materials
 from exceeding the TCLP limit for arsenic;  they would be considered residual  wastes  under
 Pennsylvania law (25 PA  Code, Chapter 75). All treated soils would be landfiiled off site in an
 intermediate (residual waste) landfill.   Offsite landfill disposal would comply with all ARARs.

 The  approximately 39,000 CY of soils/sediments site with contaminant concentrations less than
 the "principal threat1 action  levels but above the groundwater-based unsaturated soil  action
 levels would be consolidated onsite in the vadose zone and capped with impermeable material
 The  cap would be designed to meet the relevant and appropriate requirements of RCRA landfill
 closure in  40 CFR  264.310.  The remaining  approximately  16,000  CY  of excavated
 soils/sediments with contaminant concentrations less than the groundwater-based unsaturated
 soil action levels would be placed onsite in the vadose zone. All soils remaining on the surface
 after the impermeable cap is placed and which contain greater than 21 mg/kg arsenic.but less
 than the groundwater based unsaturated soil action levels, would be capped  with soil.  All
 affected areas would be  graded and revegetated.   Following implementation of Alternative 5,
 deed restrictions would be placed on areas where contaminated soils/sediments remain. Since
 contaminants remain onsite  under Alternative 5, 5-year reviews  would be  conducted.  Long-
 term groundwater monitoring in compliance with the relevant and appropriate requirements of
 40 CFR 264.117 would  also be conducted  following implementation of Alternative  5,  since
 materials which pose a potential threat to groundwater would remain  on site.

 Alternative  5 would  comply with the CERCLA preference for a remedy that employs treatment
 to reduce toxicity, mobility, or volume as a principal element. The estimated capital cost  of this
 alternative  is $28,000,000.  Annual O&M costs are estimated to be $7,600. The estimated
 present-worth  cost  of this alternative  is $27,000,000.  The estimated time to implement this
 alternative is approximately 3 years.


 6. Alternative 6:   BULK EXCAVATION/BIOLOGICAL TREATMENT, FKATION/OFFSITE LANDFILL

 Under Alternative 6, all saturated soils/sediments with  contaminant concentrations exceeding
the gvoundwater-based action levels for saturated soils would be  excavated, as would all of

                                          21

-------
those unsaturated soils with contaminant concentrations exceeding the "principal threat" action
levels  (see  Table  1).   Additionally,  the  unsaturated  soils/sediments  with contaminant
concentrations exceeding the groundwater-based action levels for unsaturated soils but less
than the "principal threat" action levels would be consolidated in the vadose zone onsite. To
facilitate the  excavation of these approximately 116,000 CY of contaminated soils/sediments,
Buildings  4, 9, 11, and  14 would be demolished  and the  pipeline/pump station  would be
temporarily abandoned or relocated. The demolition debris would either be landfilled offsite in
compliance with all applicable regulations  or  salvaged.  The excavation areas would be
backfilled with clean fill or lightly contaminated  soil.  During excavation of the saturated soils
and stream sediments, the groundwater table would  be lowered by groundwater pumping, and
Tulpehocken Creek/Union Canal would be temporarily relocated.  The extracted contaminated
groundwater would be treated in the groundwater pump-and-treat system, if present, or in a
temporary treatment system.  The extracted uncontaminated groundwater would be directly
discharged to the creek.   The final stream channel would  be located in a similar position and
with similar  dimensions  as the current channel.  Erosion control  measures  would  be
implemented during the stream relocation process.  The onsite canal lock woufd be archived
during remediation and would be reinstalled at  the completion of activities.

The approximately 61,000 CY of "principal threat" soils/sediments woulo be treated using iron-
based fixation  or  a  similar  process.  The approximately  5,600  CY of  "principal threat"
soils/sediments  with organic  chemical  concentrations  above the  "principal threat"  organic
chemical soil action levels would also be biologically treated, either before or after the fixation
step.  Following treatment, the arsenic mobility of the treated wastes would be  reduced by at
least 90%, and  an estimated 50 to  100 percent  of the organics would be destroyed.  The
fixation  and biological  treatment would occur onsite in  accordance with RCRA standards for
miscellaneous treatment units (40  CFR  Part 264,  Subpart X).   The  fixation and  biological
treatment units would be mobilized, operated, and closed according to the requirements of 40
CFR 264.600, et sea.  The specific operating parameters for the fixation  process  would be
determined in the Remedial Design phase through  engineering design and analysts and the
competitive bidding process.  A treatabitity study would be conducted prior to full-scale
irrr ementation to validate the biological treatment.  The specific operating  parameters for the
biological treatment process would be determined in the Remedial Design  phase through the
treatability study,  engineering design and  analysis, and  the competitive  bidding  process
Because most of the "principal threat" soils/sediments exhibit the RCRA characteristic of arsenic
toxicity and Alternative 6 constitutes treatment, RCRA Subtitle C is applicable. The RCRA land
disposal restriction treatment standard for these wastes is 5.0 mg/l arsenic, as measured by the
EP Toxicity Test or TCLP.  (A national  capacity extension for these wastes is in  effect until
May 8,1992.) The fixation/biological treatment process should achieve this treatment standard
The treated soils/sediments should no longer be RCRA characteristic wastes as the fixation
process would prevent these materials from  exceeding  the TCLP limit for arsenic; they would
be considered residual wastes under Pennsylvania law  (25 PA Code, Chapter 75).  All treated
soils would be  landfilled  off site in  an intermediate (residual waste)  landfill.  Offsite landfill
disposal would comply with all ARARs.
                                          22

-------
The approximately 39,000 CY of soils/sediments site with contaminant concentrations less than
the "principal threat" action levels but above the groundwater-based unsaturated soil action
levels would be consolidated onsite in the vadose zone and capped with impermeable material.
The cap would be designed to meet the relevant and appropriate requirements of RCRA landfill
closure in 40 CFR 264.310.

The remaining  approximately  16,000 CY  of  excavated soils/sediments with   contaminant
concentrations less than the groundwater-based unsaturated soil action levels would  also be
placed onsite in the vadose zone.  After the impermeable capping step is completed,  all soils
remaining on the surface which contain greater than 21  mg/kg arsenic would be capped with
soil.  All  affected areas would be graded and revegetated.  Following  implementation of
Alternative 6, deed restrictions would be placed on areas where contaminated soils/sediments
remain. Since contaminants remain onsite under Alternative 6,  5-year reviews would  be
conducted.  Long-term groundwater monitoring in compliance with the relevant and appropriate
requirements of  40  CFR 264.117 would  also be conducted following  implementation of
Alternative 6, since materials which pose a potential threat to groundwater would remain on
site.

Alternative 6 would comply with the CERCLA preference for a remedy that employs treatment
to reduce toxicity, mobility, or volume as a principal element.  The estimated capital cost of this
alternative is $28.000,000.  Annual O&M costs  are estimated to be $7,600.  The estimated
present-worth cost of this alternative is $25,000,000.  The estimated time to implement this
alternative is approximately 5 years.

7.  Alternative 7:  BULK EXCAVATION/INCINERATION, FIXATION/OFFSITE LANDFILL

Under Alternative 7, all saturated soils/sediments with contaminant concentrations exceeding
the groundwater-based  action levels for saturated soils would be excavated,  as  would all
unsaturated soils with contaminant concentrations exceeding the "principal threat" action levels
(see Table 1).  Additionally, the unsaturated soils/sediments with contaminant concentrations
exceeding the groundwater-based action levels for unsaturated soils but less than the "principal
threat"  action levels  would be  consolidated in the vadose zone onsite.  To facilitate the
excavation of these approximately 116,000 CY of contaminated soils/sediments, Buildings 4.
9, 11, and 14 would be demolished and the  pipeline/pump station would  be temporarily
abandoned or relocated. The demolition debris would  either be landfilled offsite in compliance
with all applicable regulations or salvaged. The excavation areas would be backfilled with clean
fill or lightly contaminated soil. During excavation of the saturated soils and stream sediments,
the groundwater table  would  be  lowered  by groundwater  pumping,  and  Tulpehocken
Creek/Union Canal would be temporarily relocated.  The extracted contaminated groundwater
would be treated in  the groundwater pump-and-treat system, if present,  or in a temporary
treatment  system.  The  extracted uncontaminated groundwater would be directly discharged
to the creek.  The final stream channel would be located in a similar position and with similar
dimensions as the current channel.  Erosion control measures would be implemented during
the stream relocation process. The onsite canal lock  would  be archived during remediation and
would be reinstalled at the completion of activities.
                                          23

-------
The approximately 61,000 CY of "principal threat' soils/sediments would be treated using iron-
based fixation or  a similar  process.  The  approximately 5,600  CY  of  "principal  threat"
soils/sediments with organic  chemical  concentrations above the "principal threat" organic
chemical soii action levels would first be thermally treated, followed by fixation with cement,
iron, or similar reagents.  Following treatment, the arsenic mobility of the treated wastes would
be  reduced by  greater  than  90%, and nearly  all  of  the organic  contaminants  would be
destroyed.  The fixation  treatment would occur onsite in a mobile unit, in accordance with
RCRA standards for miscellaneous treatment units (40 CFR Part 264, Subpart X).  The fixation
unit would be mobilized, operated, and closed according to the  requirements of 40  CFR
264.600, et sea.   These requirements are applicable to the soils/sediments to  be fixated,
because treatment and  disposal  of  hazardous waste (arsenic characteristic wastes)  are
occurring.  The specific operating parameters for the fixation process would be determined in
the Remedial  Design phase through  engineering design and analysis and the competitive
bidding process.  The thermal treatment would occur  onsite  in a mobile unit,  in accordance
with RCRA 40 CFR Part 264, Subpart  0 standards. The specific type of incineration process
(e.g..  rotary kiln)  would  be determined in the Remedial Design  phase through engineering
design and analysis and the  competitive bidding process.  The incineration  unit would be
mobilized, operated, and closed according to the requirements of RCRA Part 264 Subpart 0,
40 CFR 264.340  et sea. These requirements are  applicable to the soils/sediments to be
incinerated, because treatment and disposal of hazardous waste (arsenic characteristic wastes)
are occurring.  Specific  operating practices necessary to meet  the  performance objectives,
including  a 99.99 percent destruction and removal efficiency (DRE) of  stack emissions as
required by Subpart O of RCRA, would be determined through a trial burn at the site after the
installation of the incineration unit.  Specialized air pollution control equipment would be applied
during the incineration step to capture contaminants in the exhaust air and thus ensure
compliance with the relevant and appropriate NAAQS (40 CFR Part 50) and National  Emissions
Standards for Hazardous Air Pollutants (NESHAPS) (40 CFR Part 61, Subpart N).  A pilot-
scale  study would  be conducted  prior to  full-scale implementation to adequately evaluate
arsenic removal versus size and .cost for the air pollution control equipment.  The specific
operating  parameters for the thermal treatment process would be determined in the Remedial
Design  phase through  the pilot-scale study,  engineering  design  and  analysis, and  the
competitive bidding process.

Because most of the "principal threat' soils/sediments exhibit the RCRA characteristic of arsenic
toxicity and Alternative 7  constitutes treatment, RCRA Subtitle C is applicable.  The RCRA land
disposal restriction treatment standard  for these wastes  is 5.0 mg/l arsenic, as measured by the
EP Toxicity Test or TCLP. (A national capacity extension for these wastes is in effect until
May 8,  1992.)  The incineration/fixation treatment  process should  achieve  this treatment
standard.  The treated soils/sediments should no longer be RCRA characteristic wastes as the
fixation process would prevent these materials from exceeding the TCLP limit for arsenic; they
would be  considered residual  wastes  under Pennsylvania law (25 PA Code,  Chapter 75)  All
treated soils would be landfilled off site in an intermediate (residual waste) landfill.  Offsite
landfill disposal would comply  with all ARARs.

-------
The approximately 39,000 CY of soils/sediments site with contaminant concentrations less than
the "principal threat" action levels but above the groundwater-based unsaturated soil action
levels would be consolidated onsite in the vadose zone and capped with impermeable material
The cap would be designed to meet the relevant and appropriate requirements of RCRA landfill
Closure in 40 CFR 264.310.

The remaining approximately   16,000  CY  of  excavated soils/sediments with contaminant
concentrations less  than the groundwater-based unsaturated soil action levels would also be
placed onsite in the vadose zone.  After the impermeable capping step is completed, all soils
remaining on the surface which contain greater than 21  mg/kg arsenic would be capped with
soil.  All  affected areas  would be graded and revegetated.  Following implementation  of
Alternative 7, deed restrictions would be placed on areas where contaminated soils/sediments
remain. Since contaminants remain onsite  under Alternative 7,  5-year  reviews would  be
conducted.  Long-term groundwater monitoring in compliance with the relevant and appropriate
requirements of 40 CFR 264.117 would  also be  conducted following  implementation  of
Alternative 7, since  materials which pose a potential threat to groundwater would remain on
site.

Alternative 7 would comply with the CERCLA preference for  a remedy that employs treatment
to reduce toxicity, mobility, or volume as a principal element.  The estimated capital cost of this
alternative is $33,000,000.  Annual O&M costs  are estimated to  be $7,600.  The  estimated
present-worth cost of this alternative is $32,000,000. The estimated time to  implement this
alternative is approximately 3 years.

8. Alternative 8: IN-SITU VITRIFICATION

Under Alternative  8, all saturated soils/sediments with contaminant concentrations exceeding
the groundwater-based action levels for saturated soils would be excavated.  Additionally, the
unsaturated  soils  with contaminant concentrations exceeding the  groundwater-based action
levels for unsaturated soils would be consolidated in the vadose zone onsite. To facilitate the
excavation/consolidation of these approximately 116,000 CY of contaminated soils/sediments
Buildings  4, 9. 11, and  14 would be  demolished and  the pipeline/pump station  would be
temporarily abandoned or relocated. The demolition debris would either be landfilled offsite in
compliance  with all  applicable regulations or  salvaged.  The excavation  areas  would  be
backfilled  with clean fill or lightly contaminated soil.  During excavation of the saturated soils
and stream sediments, the groundwater table would be lowered by groundwater pumping, anc
Tulpehocken Creek/Union Canal would  be temporarily relocated.  The extracted contaminated
groundwater would  be treated  in the groundwater pump-and-treat system, if present, or m a
temporary treatment system.  The  extracted uncontaminated groundwater would be directly
discharged to the creek.  The final  stream channel would be located in a similar position and
with similar  dimensions  as the  current  channel.   Erosion control measures would  be
implemented during the stream relocation process. The onsite canal lock would be archivec
during remediation and would be reinstalled at the completion of activities.

The approximately 61,000 CY of "principal threat"  soils/sediments would be consolidated on site
These  soils/sediments would  then be heated in place using electricity  passing  throug-
electrodes until the mixture formed a pool of molten  glass.    Nearly  all of  the organ -

                                          25

-------
contaminants would be destroyed during heating, while the metal contaminants would become
trapped in the glass during the subsequent cooling step.  The arsenic mobility of the treated
soils/sediments might be reduced by approximately 90%. The vitrification would be conducted
with a mobile unit in accordance with RCRA standards for miscellaneous treatment units (40
CFR  Part 264, Subpart  X).  These standards have been  determined to be relevant and
appropriate to the vitrification step because placement is not occurring. The vitrification unit
would be mobilized, operated, and closed according to the requirements of 40 CFR 264 600,
et sea. A treatability study would be conducted prior to full-scale implementation to validate
the proposed treatment.  The specific operating parameters of the vitrification  unit would  be
determined in the Remedial Design phase through the treatability study, engineering design and
analysis, and the competitive bidding process. Specialized air pollution control equipment
would be applied during the vitrification step to capture contaminants in the exhaust  air and
thus  ensure compliance with the relevant  and appropriate NAAQS (40 CFR  Part 50) and
NESHAPS (40 CFR Part 61, Subpart N). Residuals from the air pollution control system would
be vitrified in subsequent batches.  Once the vitrification step  is completed, gravel would  be
placed around the vitrified material to direct  infiltration and  groundwater around  it. The gravel
would be covered by  a soil cap.

The approximately 39,000 CY of soils/sediments site with contaminant concentrations less than
the "principal threat"  action levels but above the groundwater-based unsaturated soil action
levels would be consolidated onsite in the vadose zone and capped with impermeable material
The cap would be designed to meet the relevant and appropriate requirements of RCRA landfill
closure in 40 CFR 264.310.

The  remaining approximately  16,000 CY  of  excavated  soils/sediments  with contaminant
concentrations less than  the groundwater-based unsaturated soil action levels would  also  be
placed onsite in the vadose zone. After the impermeable capping and vitrified materials capping
steps are completed, all soils remaining on  the surface which contain greater than 21  mg/kg
arsenic would be capped with soil.  All affected  areas would be graded and revegetated
Following implementation of Alternative 8, deed restrictions would be placed on areas where
contaminated soils/sediments remain. Since contaminants remain onsite under Alternative  8
5-year reviews would be conducted.  Long-term  groundwater monitoring in compliance with the
relevant and appropriate  requirements of 40 CFR 264.117 would also be conducted following
implementation of Alternative  8, since materials which pose a potential threat to groundwater
would remain on  site.

Alternative 8 would comply with the CERCLA preference for a remedy that employs treatment
to reduce toxicity, mobility, or volume as a principal element.  The estimated capital cost of this
alternative is $45,000,000.  Annual O&M  costs are estimated to be  $7,600.  The estimatec
present-worth cost of this alternative is $44,000,000.  The estimated time to implement tr.s
alternative is approximately 3  years.

Alternative 9: SOIL FLUSHING

[Note: The WLPSG submitted an initial soil washing (flushing) proposal to EPA in Februa^/
1990.  The former owners submitted supplemental information describing the former  owners
soil flushing  proposal in more detail to  EPA on September 14,  1990.   This alternative

                                          26

-------
 description, as well as the rest of this Record of Decision, incorporate the new supplemental
 information.]

 Under the former owners'  proposal (Alternative 9), a 3-year in situ  pilot-scale soil flushing
 demonstration program (treatability study) would be conducted.  In one  place in the former
 owners' proposal, it is stated that the objective of the pilot program is to determine if arsenic
 can be leached from the soil in place over an extended  period of time, as well as to provide
 an initial evaluation of sediment flushing.  In another place it is stated that the objective is to
 determine  whether arsenic will  leach  from the soils  at levels above  the  threshold for
 groundwater impact.  Leachate from the soil flushing  test will in theory be captured by the
 pumping  network installed  by the former  owners as  part of their  groundwater  remedial
 alternative.

 Once the 3-year study results are obtained, they would be studied to determine if the leachate
 concentration is above the threshold for  groundwater impact. This threshold is defined as not
 exceeding  SDWA MCLs at the point of exposure. A realistic exposure point is more than 500
 feet from the site boundary, according to the former owners.  If soil leachate concentrations are
 above the  former owner-defined threshold, soils and sediment flushing would be expanded
 If  the  leachate concentrations  are  below the former  owner groundwater impact threshold.
 surface soils  that  exceed  action  levels protective  of  human  health  would  be capped
 Institutional controls would be applied to the entire site  plus off-site areas that are capped
 and/or treated with soil flushing.

 To conduct the pilot-scale study, a percolation/leaching field would be installed in a 20-foot by
 20-foot, moderately contaminated  area.  A wall would be constructed around the perimeter of
 the leaching area to control the lateral migration of washing solution.  Horizontal dram pipes
 would  be installed under the soils to be leached to collect the leaching fluid.  Leachate
 produced during  the  study would  be treated  in the groundwater pump-and-treat  system
 Additional soil and sediment washing testing would be conducted in a mobile  laboratory on
 site.

Alternative  9 would not comply with the  groundwater relevant and appropriate requirement of
 50 ug/l arsenic for the entire groundwater plume area of attainment, and possibly not with the
pertinent relevant and appropriate MCLs for organic chemicals. This noncompliance occurs
because  a  remedial objective is to not exceed MCLs at a realistic exposure point. (A realistic
exposure point is  more than  500 feet from the site boundary, according to the former owners )
The groundwater plume closer to the site would be allowed to exceed MCLs at the completion
of the  remediation.  Additionally, since  only a small  area of soil contamination would be
addressed  during the 3-year pilot program,  the remaining area of  soil contamination would
continue to leach contaminants to the groundwater and  cause MCLs to be exceeded during
this period.

The former owners estimate the total cost of their demonstration program,  including two years
of O&M,  at $1,400,000.  The former owners estimate that the present worth cost of the  soil
maximum expansion, including 27 years  of O&M, at $7,300,000. This figure includes costs for
soil capping and institutional  controls.   Thus the total cost of Alternative  9  for maximum
expansion is $8,700,000. The maximum expansion would encompass an estimated 156 acres

                                          27

-------
 10.  COMPARATIVE ANALYSIS OF ALTERNATIVES - SOILS/SEDIMENTS

 The nine soil/sediment remedial action alternatives described above and the selected remedy
 were evaluated under the nine evaluation criteria in the NCP 40 CFR 300.430(e)(9) as set forth
 in "Guidance for Conducting Remedial Investigations and Feasibility Studies Under CERCLA"
 (EPA, October 1988), EPA Directive 9355.3-02 "Draft Guidance on Preparing Superfund Decision
 Documents:  The  Proposed  Plan  and  Record of Decision,"  and "Guidance  on Preparing
 Superfund Decision Documents: The Proposed Plan, The Record of Decision, Explanation of
 Significant Differences, and the Record of Decision Amendment" (EPA/540/6-89/007, July 1989
 Interim  Final).  These nine criteria can be further categorized into three groups: threshold
 criteria, primary balancing criteria, and modifying criteria, as follows:

 Threshold  Criteria

     *      Overall protection of human health and the environment
     *      Compliance with applicable or relevant and appropriate requirements (ARARs)

 Primary Balancing  Criteria

           Long-term effectiveness
           Reduction of  toxicity, mobility or volume through treatment
           Short-term effectiveness
           Implementability
           Cost

 Modifying Criteria

    *      Community Acceptance
    *      State Acceptance

These evaluation criteria,  which measure the overall feasibility and acceptability of the remedy.
relate directly to requirements in Section 121 of CERCLA, 42 U.S.C. Section 9621.  Threshold
criteria must be satisfied  in order for a remedy to be eligible for selection.  Primary balancing
criteria  are used  to  weigh major  trade-offs between  alternatives.   State and  community
acceptance are modifying criteria formally taken into account after public comment is received
on the Proposed Plan. The evaluations are as follows:

Overall Protection. Alternatives  3, 4, 5, 6,  7,  and 8 would provide adequate protection of
human  health and the  environment  by eliminating, reducing, or  controlling risk through
treatment, engineering controls, or institutional controls.  Alternatives 5, 6, and 7 would treat
the most heavily contaminated ("principal threat")  soils/sediments and dispose these treated
materials off site. Moderately contaminated soils/sediments would be consolidated on site The
contaminated soils/sediments remaining at the site  would be capped to reduce  the risks
associated with direct contact and minimize the migration of contamination to the groundwater
                                          28

-------
 Alternatives 5,  6, 7, and 8 would be more protective  than  the other alternatives, since the
 heavily contaminated soils would be treated to reduce toxicity. mobility, or volume. Alternative
 5 would be less protective than Alternatives 6, 7. and  8, since the organic chemicals in the
 soils/sediments would be fixated rather than destroyed.  There are risks associated with arsenic
 volatilization during the incineration or vitrification steps of Alternatives 7 and 8; these risks
 would be managed through the use of specialized air pollution control equipment.  Alternative
 3 would  be less protective of human health and the environment than Alternatives 5, 6. 7, and
 8, because the heavily contaminated wastes would remain untreated and there is the potential
 of containment system failure from sinkhole formation, erosion, or other causes.  Containment
 system failure could result in a substantial release of contaminants to groundwater.

 Alternative 2 would not be protective  of human health  and the environment, as precipitation
 infiltration would continue to cause substantial groundwater contamination.

 Alternative 9 would  be less protective of human health or the environment than Alternatives 5,
 6, 7, and 8. Under Alternative 9, either no remediation other than soil capping and institutional
 controls  would occur, or soil flushing would be implemented on the most  contaminated soils
 If only soil capping/institutional controls  occur, the  alternative  would  essentially  be the
 equivalent of Alternative 2, and would not be protective  of human health and the environment
 If the  maximum soil flushing expansion occurs, there  is a significant risk that some of the
 leaching solution would escape the  well capture  network  (because of  the  site's complex
 hydrogeology) and contaminate downgradient groundwater.  If soil flushing is implemented, the
 flushing  duration can  not be  specified,  because the former  owners  did  not specify  a
 soil/sediment cleanup level.  Rather, they specified that  soils/sediments would  be treated until
 the effect of leachate on groundwater would not exceed MCLs at the point of exposure  (A
 realistic exposure point is more than 500 feet from the site boundary, according to the former
 owners.)  Using the former owners' calculations, it would take  10,000 to 24,000 years for the
 maximum soil flushing  option for soils to be cleaned up to a point  where their leachate
 concentration  would  meet MCLs.   (See the discussion in the attached Responsiveness
 Summary.)  Thus, the  soil flushing treatment duration  would likely be very lengthy.  During
 this period, the groundwater capture network would have to be operated to protect current or
 potential future downgradient groundwater users.

 Under the former owners' proposal, soils/sediments would not be remediated to a point where
 contaminant concentrations in the entire aquifer would be at or below the groundwater cleanup
 levels presented in Section V above. Rather, they would  only be remediated until a point where
 groundwater woula  meet MCLs at  a point at least 500 feet beyond the site boundary.  In the
 absence  of MCLs, the former owners advocate the use of a 1 x 10"4 excess lifetime cancer risk-
 based cleanup level for carcinogens and a Hazard Quotient of 1 for non-carcinogens. EPA has
 determined that, in the absence of MCLs, a 1 x 1CT6 excess lifetime cancer  risk-based standard
 is appropriate  for groundwater.  Given the  site-specific conditions, EPA has determined  that
 reducing groundwater contaminant concentrations at the Whitmoyer Laboratories Site to only
 a 1 x 10""* excess lifetime cancer risk-based level in the absence of MCLs is not protective of
 human health.  Additionally, EPA has determined that contaminant concentrations should be
 reduced  below MCLs (where they exist for the contaminants)  in the entire aquifer, and  not just
for portions of the aquifer at a significant distance downgradient of the site.
                                          29

-------
 Under Alternative 9, only minimal actions would occur during the 3-year pilot-scale program.
 During these 3 years, groundwater contamination would continue to occur and the potential for
 surface runoff and inhalation/ingestion would continue to exist.

 The "no action" alternative is not protective of human health and the environment; therefore, it
 is not considered further in this analysis as an option for the soils/sediments.

 Compliance with ARARs. Alternatives 3, 5, 6, 7, and 8 would meet their respective ARARs.
 Alternative 6 would  comply with the  applicable LDRs for arsenic characteristic  wastes, the
 applicable RCRA miscellaneous treatment unit  standards, and the relevant and  appropriate
 RCRA closure and groundwater  monitoring requirements.  Also, the relevant and appropriate
 SDWA groundwater quality standards would be met.

 Alternative 4 would not be expected to comply with the LDR standards for arsenic characteristic
 wastes, since the heavily contaminated soils would not be completely landfilled by May 8,
 1992. There is a RCRA land disposal restriction  capacity extension until this date  - see  55 PR
 22520. Thus, an ARAR waiver would be required to implement Alternative 4. Alternative 2 would
 not comply with the relevant and appropriate groundwater quality standards.  Alternative 9
 would  also not  comply with the relevant  and appropriate groundwater quality standards,
 because the soils/sediments would not be cleaned up to a point where they would no  longer
 cause the entire aquifer (area of attainment) to meet these standards.  Rather, under Alternative
 9 soils/sediments would only be cleaned up to a point where groundwater would meet these
 standards at a point of exposure a minimum of 500 feet downgradient of the site.  Alternatives
 2, 3, and 4 would not conform with the CERCLA preference for treatment.

 Long-term Effectiveness and Permanence.  Alternative 6 would reduce the hazards  posed
-by the soils/sediments by fixating the arsenic in the most heavily contaminated materials and
 biologically treating the soils/sediments most heavily contaminated with organic chemicals  The
 long-term  risk of exposure to the treated soils/sediments would be reduced  by placing these
 materials in an offsite landfill. Potential future exposure to the less contaminated materials would
 be addressed by the following engineering and institutional controls: removing soils/sediments
 that can contaminate groundwater from the saturated zone and offsite locations; consolidating
 the  excavated materials;  capping the soils which present the potential for exposure  in the
 future; and placing deed restrictions on capped areas.

 Alternatives 7 and 8 would be slightly more protective than the Alternative 6, in that a slightly
 higher percentage of the organic contamination  in the "principal threat1 soils/sediments  would
 be destroyed.  On the other hand, Alternative 8 would be less protective than the Alternatives
 5, 6, and 7 since the treated wastes would not be contained in a landfill. Alternative 5  would
 be less  effective than Alternatives 6, 7, and  8  since the organic contaminants in  the most
 heavily contaminated soils would be immobilized rather than destroyed.

 Alternatives 5, 6, 7, and 8 would be less protective of human health and the environment than
 Options A (clean closure) and B of Alternative 4, because the approximately 39,000  CY of
 soils/sediments which contain contaminant concentrations less than the "principal threat"  action
 levels but  greater than the groundwater-based unsaturated soil action levels would remain on
 site, and there is the potential for the containment (capping) system to fail from sinkhole

                                           30

-------
formation, erosion, or other causes. The onstte containment system would require long-term
maintenance, and portions of it might need to be replaced in the future.   (A groundwater
monitoring system would be placed around the capped areas to assess the effectiveness of
the remedy.)  If deed restrictions are not effective, direct exposure to the wastes in the future
could result from construction activities.

Alternatives 5, 6, and 7 would be more effective in the  long-term than Alternatives 2 and 3,
since the most heavily contaminated ("principal threat") soils/sediments would be both treated
and landfilled off site.  Alternatives 5, 6, 7, and 8 would be more permanent than Alternative 4,
since the heavily contaminated soils would be  treated prior to offsite disposal.

Alternative  9 would be less effective in the long term than Alternatives 5, 6, 7, and 8.  Under
Alternative  9, either no remediation other than soil capping  and institutional controls would
occur, or soil flushing would be imptemented  on the most contaminated soils.   If only soil
capping/institutional controls occur, the  alternative would essentially be the  equivalent of
Alternative  2, and would  not be effective in  the  long-term.   If the  maximum soil flushing
expansion occurs, an undefined quantity of soils would be treated to reduce their contaminant
concentration.   This  quantity could  potentially exceed the quantity to be treated under
Alternatives 5, 6, 7, and 6.  For the soils to be treated under Alternative 9, treatment  would only
occur until a point where leachate concentrations would not cause groundwater  to  exceed
MCLs at an unspecified point at least  500 feet beyond the site boundary. In the absence of
MCLs, the  former owners advocate the use of a 1 x 1(H excess  lifetime cancer  risk-based
cleanup level for carcinogens and a  Hazard Quotient of 1  for non-carcinogens.   EPA has
determined that, in the absence of MCLs, a 1 x  1CT6 excess lifetime cancer risk-based standard
is appropriate for groundwater.  Given the site-specific conditions, EPA has  determined that
cleaning up groundwater  at  the Whitmoyer Laboratories  Site to a 1 x 1CT4 excess lifetime
cancer  risk-based level,  in  the  absence  of  MCLs,  is  not  protective of human  health
Additionally, EPA has determined that contaminant concentrations  should be reduced below
MCLs (where they exist for the contaminants) in the entire aquifer, and not just for  portions of
the aquifer at a  significant distance downgradient of the site.  Thus,  residual risks would be
higher under  Alternative 9 than under  Alternatives 5. 6, 7, and 8.

If the maximum soil  flushing expansion occurs, there is a significant risk that some of the
leaching solution escaping  the well capture  network  (because  of  the  site's  complex
hydrogeology) and contaminating downgradient groundwater. If soil flushing is implemented.
the flushing duration  can not be  specified, because the former owners did  not specify a
soil/sediment cleanup level.  Rather, they specified that soils/sediments would be treated until
the effect of leachate on groundwater would not exceed MCLs at the point of exposure  (A
realistic exposure point is more than 500 feet from the site boundary  according to the former
owners.)  Using the former owner calculations, it would take 10,000 to 24,000 years for the
maximum  soil flushing option for soils to  be cleaned  up to a point where  their leachate
concentration would  meet  MCLs.  (See the  discussion in the  attached  Responsiveness
Summary.) Thus, the soil flushing treatment duration would likely be very lengthy.  During this
period, the groundwater capture network would  have to  be operated to protect current or
potential future downgradient groundwater users.
                                          31

-------
 Reduction  of Toxictty, Mobility, or  Volume of the Contaminants Through Treatment.
 Alternatives 5, 6, 7, and 8 would treat the heavily contaminated soils to reduce toxicity, mobility,
 or volume.  Alternative 6 would reduce the mobility of the arsenic in the wastes by fixation with
 iron, and would reduce the toxicity of the soils/sediment heavily contaminated with organics by
 (biologically treating (destroying) these contaminants. Alternatives  7 and 8 would eliminate the
 toxicity of the organic contaminants  in the wastes by thermal destruction, and would reduce
 the mobility of arsenic by fixation for  Alternative 7 and encapsulation in a glass matrix for
 Alternative 8.  Alternative  5 would reduce the contaminant mobility  by fixation with  iron and
 activated carbon.

 It is uncertain what quantity of soils/sediments would  be treated under Alternative 9.   If no full-
 scale flushing is implemented, only a minor reduction of contaminant mobility and toxicity would
 be realized  during the 3-year treatability study. If the maximum expansion  is  implemented, the
 soils  would only be treated until a point where leachate concentrations would not  cause
 groundwater to exceed MCLs at an  unspecified point  at  least  500 feet  beyond  the site
 boundary.   Some additional reduction in mobility would result from the capture and treatment
 of contaminants.  However, the mobility of the  contaminants in the treated soils/sediments
 would remain higher under  Alternative 9 than under Alternatives 5, 6, 7, and 8.

 No  reduction in toxicity, mobility,  or volume is realized  for Alternatives 2,  3, and 4.   Disposal
 without treatment is  the least preferred option under  CERCLA.

 Short-term  Effectiveness.  Under Alternatives 5 and 6, there would  be a  minimal increase in
 short-term worker exposure risk during the excavation, treatment, and capping steps.  The
 community would also have a minimal short-term exposure risk from fugitive dust, erosion and
 runoff, and  transport of the treated wastes off site.  These risks would be minimized by safe
 operating practices.  Similarly, Alternatives 2, 3, and 4 would only pose minimal short-term risks

There is a potential risk associated with arsenic volatilization  under Alternatives 7 and 8   This
 risk would be reduced to acceptable  levels by the  use of specialized air pollution control
equipment.

All of  the alternatives except Alternatives 2 and 9  include a temporary stream relocation
program.  The Tulpehocken Creek  ecosystem  would be moderately  impacted during this
program. These effects would  be minimized by employing sound ecological practices.

Alternative 6 would be implemented within an estimated 5 years from the remedy selection date
This timeframe is  necessary for efficient  biological  treatment of the soils/sediments heavily
contaminated  with organic chemicals.  All other alternatives, with the exception of Alternative
9. would require much shorter timeframes (2 to 3 years).

Full-scale treatment under  Alternative 9 would not commence  for a minimum of 3 years. During
this time groundwater  contamination would continue  and  the potential for surface  runoff
contamination and direct  contact exposure would remain.   The timeframe for completion  of
Alternative 9 depends on whether or not  full-scale flushing is implemented,  and the target
cleanup level of the flushing effort if implemented.  If no full-scale flushing is implemented (only
soil capping  and placement of  institutional controls occurs),  the  remediation could be

                                          32

-------
completed 2 to 3 years after completion of the soil flushing demonstration program.  If soil
flushing is implemented, the flushing duration can not be specified, because the former owners
did not specify a soil/sediment cleanup level. Rather, they specified that soils/sediments would
be treated until the effect of leachate on groundwater would not exceed MCLs at the point of
exposure.  (A realistic exposure point is more than 500 feet from the site boundary, according
to the former owners.)  Using the former owner calculations, it would take 10,000 to 24,000
years for the maximum soil flushing option for soils to be cleaned up to a point where their
leachate concentration would meet MCLs. (See the discussion in the attached Responsiveness
Summary.)  Thus, the soil flushing treatment duration would likely  be very lengthy.

If the maximum soil flushing expansion occurs, there is a  significant risk that some of the
leaching  solution  escaping  the  well capture network  (because  of the  site's  complex
hydrogeology) and contaminating downgradient groundwater during implementation.  During
soil flushing, the groundwater  capture  network would have to be operated to protect current
or potential future downgradient groundwater users.

ImplementabilMy.  The various alternatives have few associated administrative difficulties that
coutd delay implementation. Permits would be required for the offsite disposal of the treated
materials (Alternatives  5, 6, 7, and 9)  or untreated wastes  (Alternative 4). The iron fixation
process utilized by Alternatives  5 and 6 and  the soil  flushing  program of Alternative 9 are
relatively unproven on a large scale.  However, remediation equipment and specialists are
readily available.   Biological treatment of soils heavily contaminated  with  arsenic is also
relatively unproven.  These processes have been  proven  on a small  scale  basis. The iron
fixation treatability  study conducted during the  RI/FS  indicated  that the iron fixation step
immobilized the soluble arsenic in the soils and allowed biological processes to occur. Bench-
scale optimization studies would be required prior to implementation of Alternatives 5, 6, and
7 to optimize reagent dosages and operating conditions. A 3-year pilot-scale treatability study
would be required prior to implementation of Alternative 9.  For Alternatives  7, and 8, treatment
equipment and skilled workers would  be available but limited.  For Alternative 4, hazardous
waste disposal facilities are available but limited.  The technology,  equipment, and specialists
required to implement Alternatives 2, 3, 4, 5, 6, and 9 would be readily available. For all of the
alternatives, monitoring of air  and water  during implementation  would  be required.   For
Alternatives 5 through  9. monitoring of the treated wastes would  also be  required.  Process
monitoring would be especially important for Alternative 9, since  heavily contaminated leaching
solution could escape the containment system  and contaminate groundwater.   Long-term
groundwater monitoring would  be required for  all of  the  alternatives except Option  A of
Alternative  4, to establish the continued viability of the  alternative.

Cost.  The lowest-cost alternative is Alternative 2 at $4,450,000. The highest cost alternative
is Alternative 4A, at $80,000,000.  The other FS  alternative costs are  presented in the alternative
description sections.

State Acceptance. The Commonwealth of Pennsylvania supports  the selection of Alternative
6.
                                          33

-------
 Community Accepter     A public meeting  on the Proposed Plan was held  August 1  in
 Lebanon County, Penr  .vania.   Comments received from  the public during the comment
 period are  referenced in the Responsiveness Summary attached to this Record of Decision.

 B.  SUMMARY OF ALTERNATIVES - GflOUNDWATER

 Alternatives 1 through 4 for the groundwater are numbered to correspond with the numbers in
 the first FS report (2/90).  Alternative  5 is the alternative presented by the former site owners.
 The alternatives are the following:

 *   Alternative 1: No Action
 *   Alternative 2: Plume Containment
 *   Alternative 3: Extraction  (As  > 1,000 ug/l)/  Physical.  Chemical,  (and  Biological)
    Treatment/Discharge
 *   Alternative 4: Extraction  of  All  Contaminated  Groundwater/Physical,  Chemical,  (and
    Biological) Treatment/Discharge
 *   Alternative 5: Phased Approach

 1. Alternative 1: NO ACTION

 Under the Superfund program, the "no action"  alternative is required to be evaluated at  every
 site to establish a baseline for comparison with the other alternatives. For this alternative, EPA
 would take no actions other  than monitoring  groundwater annually and performing reviews
 every 5 years. Alternative 1 would not comply with the relevant and appropriate groundwater
 quality ARARs ( §£., the SDWA MCL of 50 ug/l arsenic and the state hazardous  waste
 regulation requirement of cleaning up groundwater to background concentrations).  While no
 capital costs would be incurred under this alternative, annual operation & maintenance (O&M)
 costs are estimated to be $13,400.  This alternative has a present-worth cost of $200,000, and
 can be implemented  immediately.

 2.  Alternative  2: PLUME CONTAINMENT

 Under Alternative 2, a network of groundwater collection wells would be established near the
 perimeter of the  contaminated groundwater plume. These wells would be operated to only
 remove enough groundwater  from the aquifer  to keep the contaminant plume from growing
 Fracturing  methods  (e.g..  hydrofracturing  or blasting)  could be used to enhance the
 groundwater  removal  system.  An estimated 150 gallons per minute (GPM) of groundwater
would be removed.

The extracted groundwater would be treated  in an  onstte treatment plant constructed and
operated in compliance with 40 CFR 264.600 et sea. The treatment plant would utilize physical,
chemical and possibly biological processes.  Bench-scale studies would be conducted prior
to full-scale design,  to optimize the treatment process and determine if biological treatment
would be appropriate. An extensive aquifer testing program  would also be required prior  to
design of the treatment plant.   The treated water would either be discharged to Tulpehocken
Creek (creek discharge option),  reinjected into the aquifer (reinjection  option), or disposed
using  a combination of the two methods.  Treated water disposal would comply with all ARARs

                                         34

-------
 [££., Pennsylvania water quality standards (25 PA  Code, Chapter  93)  and Pennsylvania
 wastewater discharge standards (25 PA Code, Chapter 92)]. Treatment plant residuals would
 be considered  residual  wastes under Pennsylvania law  (25 PA Code, Chapter 75).  These
 residuals would be landfilled offsite in an intermediate (residual waste) landfill.  Offsite landfill
 disposal would  comply with all disposal ARARs.  If the treatment plant includes an air stripping
 unit, this unit would be operated to comply with the relevant and appropriate NAAQS (40 CFR
 Part 50) for ozone.

 As no attempt to actively clean up the aquifer would be made under Alternative 2, the relevant
 and appropriate requirements of the SDWA MCLs (40 CFR Part 141) and the state hazardous
 waste regulations requirement (25 PA Code, Chapter  75, Part  264) of cleaning up the entire
 aquifer to background concentrations would not  be met.  Thus, ARAR waivers  would be
 required to implement this alternative.   Compliance  with the state regulation is technically
 impracticable. Groundwater monitoring would be conducted during the remediation period in
 compliance with 40 CFR 264.101.  Residences with potentially affected potable water supply
 wells would  be  included in the monitoring program to ensure contaminated groundwater has
 not bypassed the containment system to pose an exposure threat.  Deed restrictions would be
 placed on the contaminated aquifer to restrict its use.

 Alternative 2 would comply with the CERCLA preference for a remedy that employs treatment
 to reduce toxicity,  mobility, or volume as a principal element.  The estimated capital costs of
 this alternative are $6,000,000 for the creek discharge option and $7,720,000 for the rejection
 option.  Annual  O&M costs are estimated to be $1,000,000 for the creek discharge option and
 $1,040,000 for the reinjection option.  The estimated present-worth costs of this alternative are
 $21,400,000 and $23,600,000 for the creek discharge and reinjection options, respectively  The
 estimated time to implement this alternative is approximately 2.5 years.

 3.  Alternative 3:  EXTRACTION (As > 1,000 ug/l)/PHYSICAL, CHEMICAL, (AND BIOLOGICAL)
                 TREATMENT/DISCHARGE

A network of  groundwater collection  wells would  be established  throughout  the most
contaminated part  of the contaminated groundwater plume.  These wells would be operated
to remove groundwater from the portion(s)  of the aquifer with arsenic concentrations of 1.000
ug/l or greater (i.e.,  the most contaminated groundwater). These portion(s) of the contaminated
aquifer contain  approximately 98% of  the contaminant mass.  Fracturing methods (e q
hydrofracturing  or blasting) could be  used  to enhance the groundwater removal system  An
estimated 300 gallons per minute (GPM) of groundwater would be removed.

The  extracted groundwater would be treated in an onsite treatment plant constructed and
operated in compliance with 40 CFR 264.600 et sea. The treatment plant would utilize physical
chemical and possibly biological processes.  Bench-scale studies would be conducted prior
to full-scale design, to optimize the treatment process and determine if biological treatment
would be appropriate. An  extensive aquifer testing program would also be required prior to
design  of the treatment plant.   The treated water would either be discharged to Tulpehocken
Creek (creek discharge  option), reinjected into the aquifer (reinjection option), or disposed
using a combination of the two methods.  Treated water disposal would comply with all ARARs
tea.. Pennsylvania water  quality  standards (25 PA  Code, Chapter 93)  and Pennsylvania

                                          35

-------
wastewater discharge standards (25 PA Code. Chapter 92)].  Treatment plant residuals would
be considered residual wastes under Pennsylvania law (25 PA Code, Chapter 75).  These
residuals would be landfilled off site in an intermediate (residual waste) landfill.  Offsite landfill
disposal would comply with all disposal ARARs. If the treatment plant includes an air stripping
unit, this unit would be operated to comply with the relevant and appropriate NAAQS (40 CFR
Part 50) for ozone.

As no attempt to actively clean up the portion of the aquifer with arsenic concentrations less
than 1,000 ug/l would be made under Alternative 3, the relevant and appropriate requirements
of the SDWA MCLs (40 CFR Part 141) (Table 1) and the state hazardous waste regulations (25
PA Code,  Chapter 75, Part 264) would not be met. Thus, ARAR waivers would be required to
implement this alternative. Groundwater monitoring would be conducted during the remediation
period in  compliance with 40 CFR 264.101.  Deed restrictions would  be placed on  the
contaminated aquifer to restrict its use.

Alternative 3 would comply  with the  CERCLA preference for a remedy that employs treatment
to reduce toxicity, mobility,  or volume as a principal element. The estimated capital costs of
this  alternative  are $12,900,000 for the  creek discharge  option and $16,000,000 for  the
reinjection option.  Annual O&M costs are estimated to be $2,020,000 for the creek discharge
option and $2,070,000 for the  reinjection option.  The estimated present-worth costs of this
alternative are $43,800,000  and $47,600,000 for the creek  discharge and  reinjection options.
respectively.  The estimated time to implement this alternative is approximately 2.5 years

4. Alternative 4:   EXTRACTION   OF  ALL   CONTAMINATED  GROUNDWATER/PHYSICAL
                 CHEMICAL,  (AND BIOLOGICAL) TREATMENT/DISCHARGE

The  goal  of Alternative 4 is to  return the entire contaminated aquifer to its beneficial use as
soon as is practicable.  To accomplish this, a network of groundwater collection wells would
be  established  throughout the contaminated groundwater  plume.   These  wells would be
operated to remove essentially  all groundwater from the aquifer with concentrations above the
remedial  action levels (e.g.. 50 ug/l arsenic).   Fracturing methods (e.g.. hydrofractunng or
blasting)  could  be used  to enhance the groundwater removal system.  An estimated 600
gallons per minute (GPM) of groundwater would be removed; however, the size of the treatment
system and associated pumping and piping systems  will be determined duing the Remedial
Design Phase of project implementation.  Although the Feasibility Study and Proposed Plan
specified certain pumping and treatment rates, optimization of the chosen system during design
will  ensure that the minimum required pumping  and treatment rates  will be utilized  for the
Remedial  Action.   Accordingly, capital and  operation and  maintenance  costs will  also be
minimized.

The  extracted groundwater would be treated in  an  onsite treatment  plant constructed  anc
operated  in  compliance with 40 CFR 264 et seq.  The treatment plant would utilize physical
chemical and possibly biological processes.   Bench-scale studies would  be conducted pncr
to full-scale design, to optimize the treatment process and  determine if biological treatment
would be  appropriate. An extensive aquifer testing program would also be required prior :c
design of  the treatment plant.  The treated water  would either be discharged to Tulpehocker
Creek  (creek discharge  option), reinjected into the aquifer  (reinjection option), or disposer

                                          36

-------
using a combination of the two methods. Treated water disposal would comply with all ARARs
[e.g.. Pennsylvania water quality  standards (25 PA  Code, Chapter  93)  and Pennsylvania
wastewater discharge standards (25 PA Code. Chapter 92)]. Treatment plant residuals would
be considered residual wastes under Pennsylvania law (25 PA Code, Chapter 75).   These
residuals would be landfilled off site in an intermediate (residual waste) landfill.  Offsite landfill
disposal would comply with all disposal ARARs. If the  treatment plant includes an air stripping
unit, this unit would be operated to comply with the relevant and appropriate NAAQS (40 CFR
Part 50) for ozone.

Alternative 4 would comply with all ARARs, including  the relevant and appropriate MCLs (40
CFR Part 141), with the sole exception of the relevant  and appropriate  state hazardous waste
regulation requirement (25 PA Code, Chapter 75, Part  264) of cleaning up the entire aquifer to
background concentrations.  Compliance with this regulation is technically impracticable   A
waiver for this ARAR would be required to implement Alternative 4. Deed restrictions would be
placed on the contaminated aquifer to restrict its use.

Alternative 4 would comply with the CERCLA preference for a remedy that employs treatment
to reduce toxicity, mobility, or volume as a principal element. The estimated capital costs of
this  alternative  are $15,600,000 for the creek discharge option and $19,800,000  for the
reinjection option.  Annual O&M costs are estimated to be $2,390,000 for the  creek discharge
option and $2,470,000 for the reinjection option.  The estimated present-worth costs of this
alternative are $52,300,000 and $57,700,000 for the creek discharge and reinjection options.
respectively.  The estimated time to implement this alternative is approximately 2.5 years.

Alternative 5: PHASED APPROACH

[Note: the former owners submitted an initial groundwater proposal to  EPA in February 1990
The  former owners submitted supplemental information  describing the  former  owners
groundwater  proposal in more detail  to  EPA on September  14, 1990.   This  alternative
description, as well as the rest of this Record of Decision,  incorporates the new supplemental
information.]

Under the former owner's proposal (Alternative 5), a 3-year large-scale pilot program would be
implemented. During the 3-year period, the pumping system and treatment  facility would be
initially operated. According to the former owners, the objectives of the pilot  program include
providing all the necessary data to select and design a final groundwater remedy;

In the first year of the 3-year program, an onsite groundwater recovery well network and
groundwater treatment system would be designed, installed, and tested. The recovery well and
treatment system would be operated during the second and third years of the program Also.
during the second year, offsite monitoring wells would be installed and tested.  In the third year
these wells would be monitored.

The  ecovery well and treatment system would be operated at  approximately 120 gpm  The
system operation would  only pump and  treat the  most  contaminated  groundwater.  The
extracted groundwater would be physically and chemically treated to remove contaminants
Portions of the groundwater would be reinjected into the aquifer or used for soil flushing, while

                                          37

-------
 the remainder would be discharged to Tulpehocken Creek.  Up to 40 wells would be installed
 to establish the otfsite monitoring well system.

 Other than  monitoring, offsite groundwater would be unaddressed; the contaminant plume
 would continue to grow during the 3-year period. The relevant and appropriate groundwater
 quality ARARs [MCLs (40 CFR Part 141)] (Table 1), and the state hazardous waste regulation
 requirement  (25  PA  Code, Chapter  75,  Part 264)  of  cleaning  up  the  entire  aquifer to
 background concentrations would not be met during the  pilot program.  Thus, ARAR waivers
 would be required to implement this alternative.  Deed restrictions would be placed on the
 contaminated aquifer to restrict its use during the pilot program.

 Once the 3-year pilot program is completed, the final groundwater remedy would be selected
 and the long-term (Phase II) groundwater remediation program implemented. According to the
 former owners, the objectives of the long-term  program  are to provide  for  removal and
 treatment of arsenic and other contaminants from groundwater that has become contaminated;
 to protect human health by preventing exposure to  groundwater above MCLs; to  serve as a
 collector system for soil flushing leachate; and to provide a source of water for the soil flushing
 program.  Specific groundwater remediation objectives (e.g, restore contaminant concentrations
 in the entire area of attainment to MCLs or 1 x 1CT6 excess lifetime cancer risk-based levels)
 were  not specified by the former owners.

 The former owners anticipate that the Phase  II program would be an expansion of the Phase
 I program, with a more extensive network of pumping wells. A long-term groundwater pumping
 rate of 300 to 400 gallons per minute is expected, with the actual rate determined based upon
 Phase I data.  The system would pump  and treat groundwater only from the most highly
 contaminated (undefined) part of the plume.  It is unclear how long the long-term pump-and-
 treat system would operate. In one place the former owners stated the system would operate
 up  to 27 years.   In another place they stated it would  be  operated until  groundwater
 remediation  goals were achieved.

 If groundwater is extracted from only the most contaminated portion of the plume (and is not
 extracted  from  the entire contaminant plume), Alternative 5 would  not comply with the
 groundwater relevant and appropriate requirement of 50 ug/l arsenic for the entire groundwater
 plume area of attainment, and possibly not with the pertinent relevant and appropriate MCLs
 for organic chemicals.  The portion of the aquifer unaddressed by  the extraction system that
 contained contaminant concentrations in excess of MCLs would be expected to grow over time


 The former owners estimate the total cost of  their pilot-scale program, including two years of
 O&M, at $5,500,000.  The former owners estimate that the present worth cost of the long-term
 groundwater pump-and-treat program,  including 27 years of O&M,  at $34,000,000.  Thus the
 former owners' estimated total cost of Alternative 5 is $39,500,000.

 6.   COMPARATIVE ANALYSIS OF ALTERNATIVES - GROUNDWATER

The five groundwater remedial action alternatives described above were evaluated using the
 nine criteria in the NCP. The  evaluations are as follows:

                                          38

-------
 Overall Protection. Alternative 4 would provide adequate protection of human health and the
 environment by collecting and treating the entire contaminated groundwater plume, although
 the remediation time may well exceed 30 years. Alternative 2 would be less protective, since
 it  is heavily dependent on  deed restrictions and monitoring to prevent  exposure.  Under
 Alternative 2. offsite groundwater quality would deteriorate in the near-term.   If  the plume
 containment (capture) system is incomplete, increasing amounts of contaminants would escape
 the containment system over time.

 Alternatives 3 and 5  would be  less  protective  than  Alternative 4, since not  all  of the
 contaminated groundwater would be extracted.  The groundwater contaminant plume would be
 allowed to grow over time.  Unlike Alternative 2, Alternatives .3 and 5 would remove the most
 heavily contaminated groundwater from the aquifer.  However, Alternative  2 would keep the
 contaminant plume from growing.

 Since the remedial action objectives (cleanup levels) under Alternative 5 were not specified, they
 can  not  be compared  with those for Alternatives 2-4.   For groundwater-based  soil cleanup
 levels, the former  owners advocate the use  of a 1 x 1CT4 excess lifetime  cancer  risk-based
 cleanup level for carcinogens and a Hazard Quotient of 1 for non-carcinogens in the absence
 of MCLs.  EPA has determined that, in the absence of MCLs, a 1 x 1CT6 excess lifetime cancer
 risk-based standard is appropriate for groundwater.  Given the site-specific conditions, EPA has
 determined that  cleaning up  groundwater at the Whitmoyer  Laboratories Site to  a 1 x 10~4
 excess lifetime cancer risk-based standard is not protective of human  health.

 The "no action" alternative is not protective of human health and the environment; therefore,  it
 is  not considered further in this analysis as an  option foMhe  groundwater.

 Regulatory Compliance.   Alternative 4 would meet all' of its applicable  and  relevant and
 appropriate requirements, including  the relevant and appropriate SDWA MCLs (40 CPR Part
 141) (Table 1), with the sole  exception of the state relevant and  appropriate requirement of
cleaning  up to background (25 PA  Code, Chapter  75, Part 264).   Since compliance with this
ARAR is technically impracticable, an ARAR waiver is justified.  Under Alternatives 2 and 5, the
entire contaminated groundwater plume would  not comply with the relevant and appropriate
SDWA MCLs, as well as the state requirement of cleaning up to background.  Under Alternative
3, the unremediated portion of the contaminated groundwater plume containing less than 1.000
ug/l would not comply  with the relevant and  appropriate SDWA MCLs, and the  entire plume
would not comply with  the state background requirement. Thus, a waiver o1 the SDWA  MCL
ARAR would be required to implement all of the alternatives except Alternative 4;  and a waiver
of  the state background requirement would be required for all of the alternatives.  A waiver of
the state background requirement is justifiable because of technical impracticability. AM of the
alternatives would conform with the CERCLA preference for treatment.

Long-term Effectiveness and Permanence.   If  Alternative 4  proved  to be  technically
practicable, risks from the groundwater would be virtually eliminated, although the remediation
timeframe may exceed 30 years.  Proper disposal of treatment residuals would prevent ft  -e
risks.   Alternative 2 would be less  effective in the same timeframe as  Alternative 4, as much
less contamination would be removed from the aquifer, and contamination  would continue to

                                          39

-------
 migrate off  site and deteriorate offsite  groundwater  quality.  Continued  expansion  of  the
 groundwater plume  would be controlled by the groundwater  extraction  system.   Under
 Alternative 3, migration of the most contaminated portion of the groundwater plume would be
 controlled by the extraction system.  This portion would be actively remediated.  However,  the
 less contaminated portion of the plume would be allowed to expand over time and pose future
 hearth risks.

 Since a final remedy would not be selected at this time  under  Alternative 5, the long-term
 effectiveness of this  alternative can not be completely discussed.  The pilot-scale program
 making up Alternative 5 is intended to evaluate the potential for both active remediation and
 containment of the contaminated groundwater. The former owners propose to only pump and
 treat groundwater from the most contaminated portion  of the groundwater plume.  Apparently
 the less contaminated portion of the plume would be allowed to expand over time and continue
 to pose future health risks.

 Reduction of Toxicrty, Mobility,  or  Volume  of  the Contaminants Through Treatment.
 Alternative 4 would reduce the toxicity, mobility, and volume of the contamination present in  the
 groundwater.  The  contaminant mobility would be reduced by the collection  system  and
 treatment  of the extracted  contaminants  to destroy the organic contaminants and immobilize
 the arsenic in a relatively insoluble chemical precipitate. Contaminant toxicity would be reduced
 by  organic destruction.  The  volume of contaminated groundwater would  be significantly
 reduced over time.

 Under Alternative 2, contaminant mobility would be reduced somewhat  by the  containment
system and the treatment of the extracted contaminants. However, the extraction system would
 be  less  aggressive  than  the  Alternative 4  system, and significantly  less  quantities of
contaminants would be removed for treatment in equivalent timeframes.  Contaminant toxicity
would be  reduced somewhat by the organic destruction.   There would be no  reduction in
contaminated groundwater volume, since active remediation of the groundwater plume would
 not be attempted.

 Under Alternative 3, approximately 50% of the groundwater plume containing approximately 98%
of the groundwater contaminants would be actively  remediated.  However, the volume of
contaminated groundwater would increase over time as the unaddressed groundwater plume
expands.

The former owners do not specify what portion of the  contaminated aquifer would be actively
remediated under their program  (Alternative 5).   An estimated 300 to 400 gallons per  minute
would be withdrawn, significantly  less  than the estimated 600  gallons per minute under
Alternative 4.  The former owners also stated that only the most  highly contaminated  pan ol
the plume would be pumped and treated. Apparently the less contaminated part of the plume
would be  unaddressed  by the  pump-and-treat program.   If this is the case, the volume of
contaminated groundwater would increase over time as the unaddressed plume expands

Short-term Effectiveness.  Alternative 4 would have a  high degree of short-term  effectiveness
Complete  remediation of the groundwater may take more than 30 years, however. Alternative
2 would be  less effective  in the short-term,  as contaminant concentrations  in the offsite

                                         40

-------
groundwater would  be allowed to increase over time.  If the  plume capture  system is
incomplete (which is a significant possibility considering the complex hydrogeologic setting of
the site), increasing amounts of contaminants would escape the containment system over time.
Discharge of groundwater contaminants to surface water could also increase over  time. The
remediation timeframe would be significantly longer than for Alternative 4.

Because Alternatives 3 and 5  allow the contaminant plume to increase over time,  these
alternatives are less reliable  than Alternative 4.  The remediation timeframe for Alternative 3 is
comparable to the  Alternative  4 timeframe  and less than the timeframe  for Alternative 2.
Because Alternative 5 does not specify the final groundwater remedy, its remediation timeframe
can not be discussed.

Impfementabillty.  There is  a concern  whether achieving the groundwater cleanup goals is
technically feasible.  The proposed groundwater treatment technologies for all of the alternatives
are highly reliable. The various  alternatives have few associated administrative difficulties that
could delay implementation.  Permits would be required for the offsite disposal of the treatment
residuals. The equipment, specialists, and treatment/disposal facilities necessary to  implement
the alternatives are readily available.  Bench-scale treatment optimization studies  would be
required prior to  implementation of all  of the alternatives to optimize reagent dosages and
operating conditions.  Additionally, an aquifer testing program would be required  for all  of the
alternatives to permit  more accurate estimation of the required treatment plant capacity.  For
all of the alternatives, monitoring of air, groundwater, surface water, and the treatment residuals
during implementation would be required.   Groundwater monitoring would be  especially
important for the  plume containment option (Alternative 2), since heavily groundwater  could
escape  the containment system and further contaminate groundwater.

Cost. The  estimated present-worth costs of Alternative 4 for the groundwater are  $52,300,000
and $57,700,000 for the creek discharge and reinjection options, respectively. Alternative 4 is
the highest cost alternative.  The lowest-cost alternative is Alternative 2 at $21,400,000 (creek
discharge option).  The other alternative costs are presented in  the alternative description
sections.

State Acceptance.  The Commonwealth of Pennsylvania supports the selection of  Alternative
4,  with the  contingent remedy of Alternative 2.

Community Acceptance.  A public meeting on the Proposed Plan was held  August 1  in
Lebanon County, Pennsylvania. Comments received from the public  during the comment
period are referenced in the  Responsiveness Summary attached to this  Record of  Decision

VIII.  THE SELECTED REMEDY

Based upon consideration of information available for Operable Unit Three of the Whitmoyer
Laboratories Site,  including the documents available in the Administrative Record, an  evaluation
of  the risks currently posed  by the site, the requirements of CERCLA, the detailed  evaluation
of  alternatives, and community input; both EPA  has selected the following alternatives as the
remedy  to be implemented for the operable unit.
                                          41

-------
A.  SOILS/SEDIMENTTS

Alternative 6 - Iron-based fixation of the "principal threat" soils/sediments; biological treatment
of the soils/sediments which contain organic chemicals in concentrations greater than the
"principal threat" organic action levels; onsrte consolidation of the soils which pose a threat to
groundwater but contain contaminant concentrations less than  the  "principal threat" action
levels, followed by  capping with impermeable materials;  and  soil  capping  of  the lightly
contaminated surface soils.  The "principal threat1 soils/sediments would be treated to comply
with RCRA land disposal restrictions; to reduce the arsenic mobility, as measured by the TCLP,
by at  least 90%; and to reduce the organic contaminant concentrations to the "principal threat"
soil action levels (biologically treated soils only), prior to  being landfilled at an offsite disposal
facility.

B.  GROUNDWATER

Alternative 4 - Extraction of all contaminated groundwater, followed by physical, chemical and
possibly biological treatment and  discharge of the treated water to  reinjection wells and/or
Tuipehocken Creek.  Based on current information,  this alternative appears to provide the best
balance of trade-offs among the groundwater alternatives with respect to nine criteria that EPA
uses  to evaluate alternatives.

There is a moderate degree of uncertainty over whether the selected remedy will  be able to
meet  the groundwater health-based cleanup levels identified by the EPA.  It may potentially
prove technically impracticable to achieve the health-based groundwater cleanup goals under
the selected  remedy for  the  groundwater.   It  will be  difficult to  predict  the ultimate
concentrations to which contaminants in the groundwater may be reduced until the Alternative
4 extraction system has been operating for some period of time.  If information emerges from
the operation of the Alternative  4 extraction system that strongly suggests that it is technically
impracticable to achieve the cleanup goals throughout the contaminated groundwater plume
because of an observed "leveling-off" of contaminant concentrations, the EPA, in consultation
with the Commonwealth of Pennsylvania, intends to select the contingent remedy of Alternative
2 in those areas where the cleanup goals will not be met.  Alternative 2 would involve extracting
groundwater from the perimeter of the targeted groundwater plume only in sufficient quantities
to keep the plume from spreading.  In other words, active remediation would not be attempted

The estimated present-value cost of this selected remedy is $77,300,000; as follows:

    Medium                                       Present-value Cost

Soils/Sediments                                    $25,000,000
Groundwater                                             $52.300.000

TOTAL COST                                            $77,300,000
                                          42

-------
The major components of the selected remedial action are as follows:

    *      Excavation of all moderately contaminated soils/sediments from offsite and saturated
           onsite locations, and all heavily contaminated soils [estimated volume = 116,000
           cubic yards (CY)].

    *      Demolition of Buildings 4, 9, 11. and 14.

    *      Backfilling of the excavated areas with clean fill or lightly contaminated soil.

    *      Onsite  fixation  of  the  approximately  61,000  CY  of  heavily  contaminated
           soils/sediments using an iron-based or other similar fixation process that provides
           equivalent protection.

    *      Biological treatment of the approximately  5,600 CY of soils with organic chemical
           concentrations above the heavily contaminated soil action  levels either prior to or
           following the fixation step.

    *      Consolidation of the moderately contaminated soils/sediments on site above the
           groundwater table.

    *      Capping of the approximately 39,000 CY of moderately contaminated soils/sediments
           having contaminant concentrations above groundwater-based unsaturated soil
           cleanup targets with low-permeability materials.

    *      Soil capping of all soils/sediments remaining on the surface following the excavation
           and  consolidation steps that are  not capped with low-permeability materials and
           contain greater than 21 mg/kg arsenic; and other disturbed areas as needed.

    *      Grading and revegetation of all areas affected by the soil/sediment remediation

    *      Following the soils/sediments remediation, placement of deed restrictions on areas
           with  remaining contamination.

    *      Aggressive extraction of all groundwater from the aquifer beneath the site with
           concentrations above health-based levels (e.g.. 50 ug/l arsenic) until the maxtmum
           groundwater contaminant concentrations  are alf less than  health-based levels.

    *      Treatment of the extracted groundwater in  an onsite treatment plant,  utilizing
           physical, chemical and  possibly biological processes.  The size of the treatment
           system and associated pumping and piping systems will be determined during the
           Remedial Design Phase of project implementation. Although the Feasibility Study
           and  Proposed Plan specified certain pumping and treatment rates, optimization of
           the chosen system  during design will ensure that the minimum required pumping
           and treatment rates will be utilized for the Remedial Action. Accordingly, capital and
           operation and maintenance  costs will also be minimized.
                                           43

-------
    *      Disposal  of  the  treated water by  either  discharging it to Tulpehocken Creek,
           reinjecting ft into the aquifer, or a combination of the two methods.

    *      Salvaging nonhazardous demolition debris, as feasible.

    *      Disposal  of  the  following  in off site landfill (s) in accordance with all applicable
           regulations:  the  treated  soils; the groundwater treatment residuals; and the
           demolition debris that is not salvaged.

The selected remedy is the last of several phases in the long-term remediation of this site and
will be consistent with previously selected site remedies.

The contingent remedy is similar to the selected remedy,  with the exception that groundwater
would only be extracted from the perimeter of the non-attainment area in sufficient quantities
to keep this area from growing. Additionally, contact with contaminated groundwater would be
restricted  through the  use  of deed restrictions and an  extensive groundwater  monitoring
program.

These actions will significantly reduce or eliminate the actual and potential threats to human
health and the environment  posed by the OU Three materials, and are consistent with EPA's
strategy for remediation of the site.

IX. STATUTORY DETERMINATIONS

Under its  legal authorities,  EPA's  primary responsibility  at Superfund sites is to undertake
remedial actions that are protective of human health and the environment.  In addition, Section
121 of CERCLA establishes several other statutory requirements and preferences.   These
specify that when complete,  the selected remedial action  (and the contingent remedial action)
for this site must comply with applicable or relevant and appropriate environmental standards
established  under  Federal and State environmental laws unless a statutory waiver is granted
The selected remedy and contingent remedy also must be cost-effective and utilize treatment
technologies or resource recovery technologies  to the maximum extent practicable.  Finally, the
statute includes a preference for remedies that permanently and significantly reduce the volume,
toxicity, or mobility of hazardous wastes.  The following  sections discuss  how the selected
remedy and contingent remedy for this site meet these statutory requirements.

A.  PROTECTION OF HUMAN HEALTH AND THE ENVIRONMENT

Based on the risk assessment developed for OU Three materials, actual or potential exposure
pathways  include groundwater  consumption,  dermal contact, accidental  ingestion,  and
inhalation.   The selected remedy addresses  these  risks (protects human  health and the
environment) by reducing the mobility of the arsenic in the heavily contaminated soils using an
iron-based or other  similar fixation  process;  destroying the organic contamination  in the
soils/sediments that are heavily contaminated with organic chemicals by biological treatment,
consolidating and  impermeably capping untreated soils which pose a threat to groundwater,
soil capping surface soils which pose only a direct contact threat; salvaging nonhazardous
demolition debris, as feasible; aggressively collecting and treating all contaminated groundwater

                                          44

-------
(i.e., restoring the contaminated aquifer); and disposal of the treated soils/sediments, the
groundwater treatment residuals, and the untreated (nonhazardous) demolition debris that is
not salvaged in offsite landfill(s) to further reduce contaminant mobility and access to these
materials.

The selected remedy will not pose any unacceptable short-term risks or cross-media impacts
to the site, the workers, or the community.  There will be some long-term risks associated with
leaving  moderately and lightly contaminated soils/sediments at the site.  However, these
materials wilt  be capped, and access to these materials will be restricted by placing deed
restrictions on areas where they are present following the remediation. Since metals cannot be
destroyed, there will be some long-term risk associated with the heavy metal (primarily arsenic)
content of the heavily contaminated soils/sediments. However, these soils/sediments will be
treated  prior  to  disposal  to reduce  the mobility  of  the  heavy metals, and  the treated
soils/sediments will be placed into an offsite landfill for proper long-term management.

The contingent remedy addresses the risks posed by the contaminated soils/sediments in the
same manner as  the selected  remedy.   However,  for the  groundwater medium, active
restoration of the aquifer would not occur; only measures intended to keep the contaminant
plume from expanding would occur.  The contingent remedy would protect human health and
the environment  by halting  plume expansion  and restricting  contact with  contaminated
groundwater through the use  of deed restrictions and groundwater monitoring.

B.  ATTAINMENT OF APPLICABLE OR RELEVANT AND APPROPRIATE REQUIREMENTS

The selected remedy will attain all applicable or relevant and appropriate requirements for the
site, the OU Three materials, and the actions that will be implemented, with the exception of
the state action-specific requirement to remediate groundwater to background concentrations
(25 PA  Code, Chapter 75, Part  264).  This ARAR would be waived because of  technical
impracticability. The  contingent remedy for groundwater would also not comply  with the
relevant and appropriate SDWA MCL chemical-specific ARAR.  This ARAR would be waived
because of technical impracticability if the contingent remedy is elected by the EPA in the
future.  The other  major ARARs include the following:

    *     Action-Specific ARARs - Soil/sediment excavation activities will be in accordance with
          Pennsylvania requirements for erosion control (25 PA Code, Chapter 102), and 4G
          CFR 264.31. Remedial activities will be conducted in compliance with the Fish and
          Wildlife Coordination Act (16 USC 661).  The fixation, biological treatment, anc
          groundwater treatment plants will be designed and operated in  accordance wnh
          RCRA Subtitle C miscellaneous treatment unit standards (40 CFR Part 264, Subpan
          X).  The treated  soils/sediments and  groundwater treatment residuals will  be
          monitored to ensure compliance with RCRA Subtitle C land disposal restrictions (40
          CFR Part 268)  (i.e., to ensure that the treated soils/sediments and groundwater
          treatment residuals are  nonhazardous),  prior to disposal  at an  approved offsite
          facility. Groundwater treatment will employ air monitoring as appropriate to ensjre
          compliance with  55  FR No.  120,  June 21, 1990  (page  25454).  Capping cf
          moderately contaminated soils/sediments onsite would be  conducted to  meet the
          relevant and appropriate requirements of RCRA landfill closure in  40 CFR 264 3iC

                                         45

-------
Materials transported offsite will meet the CERCLA otfsite disposal policy and comply
with Federal transportation regulations (40 CFR Parts 262 and 263; 49 CFR Parts
107 and 171-179)  and Pennsylvania  regulations (25 PA Code,  Chapter 263) for
material transport.  During contaminated soils/sediment removal and treatment, air
monitoring will be performed to ensure that any air emissions comply with Clean
Air Act (40 CFR Parts  50 and 61) and Pennsylvania air quality regulations (25 PA
Code, Chapters 123,127, and 131).  OSHA requirements (29 CFR Parts 1904, 1910.
and 1926) will be met for workers  engaged in remedial activities.    The offsite
landfill accepting the treated soils/sediments, unsalvaged demolition waste, and
groundwater  treatment  residuals will  comply  with RCRA Subtitle D and state
industrial (solid) waste management regulations. Treated groundwater disposal will
comply with all ARARs fe.g.. Pennsylvania water quality standards (25 PA Code,
Chapter  93)  and Pennsylvania wastewater discharge standards  (25 PA Code,
Chapter  92)].  Long-term groundwater monitoring in compliance with  40  CFR
264.117 will also be conducted following implementation of the selected remedy.

Chemical-Specific ARARs - RCRA Subtitle C and Commonwealth of Pennsylvania
requirements  for identification of characteristic hazardous wastes (40 CFR Part 261
and 25 PA Code,  Chapter 261,  respectively)  will  be complied with during  the
remediation of OU Three materials. Groundwater will be remediate to the Maximum
Contaminant  Levels specified in the Safe Drinking Water Act (40 CFR Part 141)  if
technically practicable.   Air emissions during remedial activities will be monitored
for compliance with Clean Air Act (40 CFR Parts 50 and 61) and Pennsylvania air
quality regulations (25 PA Code, Chapters 123,127, and 131).  Clean Water Act (40
CFR  Part 122) and Pennsylvania  (25 PA Code,  Chapter 92)  direct discharge
standards would be met by the groundwater remediation.

Location-Specific ARARs - Remediation of the contaminated soils/sediments will be
conducted in  accordance with the Federal Floodplains Management and Executive
Order (E.O. 11988).

Other Criteria, Advisories, or Guidance to be Considered- In determining acceptable
soil/sediment  and groundwater remedial action levels EPA used advisory levels and
guidelines that are "to-be-considered" for the remedial actions. These are-

       EPA-established reference doses for contaminants posing noncarcmogemc
       threats to human health
       EPA-established  carcinogenic  potency  factor for contaminants posing
       carcinogenic threats to human health
       Proposed primary drinking water standards  under the Safe  Drinking Water
       Act for tetrachtoroethene,  benzo(a)pyrene, indeno(1,2,3-cd)pyrene,  and
       benzo(b)1luoranthene

The selected  remedy  will also comply with the Commonwealth of Pennsylvania
guidance  document "Hazardous  Waste and Petroleum Products  Contamination
Cleanup  Projects" which  requires Best Available Technology for air strippers ana
other equipment designed to remove  volatile organic  chemicals from water

                               46

-------
          The selected remedy satisfies the CERCLA preference for remedies that incorporate
          treatment as a principal component.

C.  COST-EFFECTIVENESS

The estimated  present worth cost of the selected remedy for OU Three is $77,300,000.  EPA
and  the  Commonwealth of Pennsylvania believe the selected  remedy  is cost-effective  in
mitigating the  risks  posed by the OU Three materials in  a reasonable  period of time (an
estimated 5 years for the soils/sediments) and  meets all other requirements  of CERCLA.
Because groundwater concentrations in the entire area of attainment will be restored to health-
based levels if technically practicable, the majority of organic contaminants present in the OU
Three materials will be destroyed, the metal contaminants in the groundwater and "principal
threat" soils/sediments will be treated to reduce mobility, and the treated nonhazardous wastes
and demolition debris will be disposed in an appropriate landfill (or salvaged, as appropriate),
the selected remedy affords  a high  degree of long-term  effectiveness and  permanence.
Although  the  no-action  alternative,  the soil capping  alternative (Alternative 2),  and the
impermeable capping alternative (Alternative 3) can be implemented at lower costs than the
selected soil/sediment remedy, these alternatives do not provide for permanent treatment and
are not as effective in protecting  human health and  the  environment.  In addition, these
alternatives do not meet ARARs.  Although the  soil  flushing program (Alternative 9) can be
implemented at lower costs than the selected soil/sediment remedy, this alternative is not as
effective in protecting human health and the environment. In addition, the alternatives does not
meet ARARs.

The no-action alternative, the plume containment alternative (Alternative 2), the extraction and
treatment of the most contaminated groundwater alternative (Alternative 3), and the phased
approach (Alternative 5)  can be implemented at  lower costs than the selected groundwater
remedy.  However, these alternatives are less protective of human health and the environment.
do not meet the MCL ARAR or currently justify a waiver, and  do not attempt to restore the
contaminated aquifer to its beneficial uses as soon  as  may be technically  practicable.  The
contingent groundwater remedy is less expensive than Alternatives 3 and 5, yet more expensive
than the no-action alternative.  However, the no-action  alternative is not protective of human
health and the environment and does not meet ARARs or justify a waiver. (If the criterion for
election of the  contingent remedy is met, a waiver of the MCL ARAR would  be justified)

D.  PREFERENCE FOR TREATMENT AS A PRINCIPAL ELEMENT

By treating all of the OU Three soils/sediments that pose the principal threats to human health
and the environment and all contaminated groundwater, the selected  remedy addresses the
principal threats posed by the  OU Three materials through the  use of treatment technologies
Therefore, the statutory preference for remedies that employ treatment as a  principal element
is satisfied.

E.  UTILIZATION  OF  PERMANENT  SOLUTIONS  AND  ALTERNATIVE  TREATMENT  (OR
    RESOURCE RECOVERY TECHNOLOGIES) TO THE MAXIMUM EXTENT PRACTICABLE
                                         47

-------
EPA has determined that the selected remedial action represents the maximum extent to which
permanent solutions and treatment  technologies can be  utilized while  providing the best
balance among the other evaluation criteria.  Of the alternatives that are protective  of human
health and the environment and meet ARARs. EPA has determined that the selected remedy
provides the best balance of trade-offs in terms of long-term effectiveness and permanence;
implementability; short-term effectiveness;  reduction in toxicity, mobility,  or  volume through
treatment; state and community acceptance; and the CERCLA preference for treatment of the
soils/sediments and groundwater.

The selected remedy and contingent remedy addresses the principal threats posed by the OU
Three materials.  The remedy is protective of human health and the environment, meets ARARs
or a waiver is justified, incorporates treatment as a principal element, and is cost-effective. The
major tradeoffs that provide the basis for the selection decision are long-term effectiveness and
permanence;  reduction  of  toxicity,  mobility  or volume through  treatment;  short-term
effectiveness; and  cost.

Of the soil/sediment alternatives which are protective of human health and the environment and
meet ARARs. Alternative 5 is less effective in the long-term and more expensive to implement
than the selected remedy.   While Alternative 3 is less expensive than the selected remedy,
contaminant  toxicity,  mobility, or volume  is  not reduced  through  treatment.   Although
Alternatives 4A and 4B are more effective  in the long  term than the selected remedy, these
alternatives are more expensive to  fmplement and contaminant toxicity, mobility, or  volume is
not reduced  through treatment.  Alternative 4C is less effective in the long term and  more
expensive to implement than the selected remedy, and contaminant toxicity, mobility, or volume
is not reduced through treatment  There is less short-term risk associated with the selected
remedy than  for Alternative  7 (thermal treatment) and  Alternative 8 (vitrification).  These
alternatives are also significantly more expensive than the selected remedy, although they would
be slightly more  effective.  The soil flushing program (Alternative 9) can also  be implemented
at a tower cost than the selected soil/sediment remedy.  However, it  appears likely that the
proposed soil flushing program would be impracticably lengthy.  In-situ soil  washing,  as
proposed for the former owners, would also  have the  added disadvantages  o1 possible loss
of the leaching fluid to the complex groundwater system.  During the 3 years  that a soils
remedy decision would  be delayed, groundwater contamination would continue and the
potential for surface runoff contamination and direct contact with contaminated soils  would
continue to exist.   Alternative 9 would also not comply with the groundwater  MCL ARAR
Based on  the above  evaluations, the selected  remedy was  determined to be  the  most
appropriate  remedy for the soil/sediment medium of Operable  Unit Three at the Whitmoyer
Laboratories Site.

The selected remedy for groundwater is the  most protective of the groundwater alternatives
and is the only alternative which complies with the MCL ARAR.  (This ARAR would be waived
because of technical impracticability if the contingent  remedy is  elected by the EPA in  the
future.)   Of  the groundwater alternatives, the  selected  remedy attempts  to  restore  me
contaminated aquifer to its  beneficial  uses  as  soon as may  be  technically  practicable
Therefore, it would be the most effective alternative in the short  term.  Under Alternative 2
offsite groundwater contaminant  concentrations would increase in  the short-term    For

                                          48

-------
Alternatives 3 and 5, the contaminant plume would continue to grow in the short term. If the
selected remedy proves to be technically practicable,  it  would also be the most  effective
alternative in the long-term.  Alternative 2 would be less effective in the same timeframe as the
selected remedy, as  much  less contamination would  be removed from  the  aquifer, and
contamination would continue  to migrate off site and deteriorate offsite groundwater quality.
Under Alternative 3, the less contaminated  portion of the plume would be allowed to expand
over time and pose future health risks. The former owners stated that only the most highly
contaminated part of the plume would be pumped and treated under Alternative 5. Apparently
the less contaminated part of the plume would be unaddressed by the pump-and-treat program.
If this is the case,  the unaddressed plume of contaminated groundwater would continue to
expand over time and pose future health risks.  Under Alternative 5, the groundwater remedy
decision would be  delayed for 3 years, while a decision on whether to actively  remediate or
contain groundwater  is  being made.  During  the 3-year  pilot  program,  the plume will be
permitted to spread and pose future health risks. Thus, this alternative is not effective in the
short-term.  Since a final remedy would not be selected at this time  under Alternative 5, the
long-term effectiveness of this alternative is unclear. Sufficient information exists to evaluate the
groundwater alternatives options at this time. The remedial decision maintains the flexibility of
selecting a plume containment remedy in the future if cleaning up the aquifer to  health-based
levels proves technically  impracticable. Based  on the above evaluations, the selected remedy
was determined to  be the most appropriate remedy for the groundwater medium of Operable
Unit Three at the Whitmoyer Laboratories Site.  If cleaning up the aquifer to health-based levels
proves technically  impracticable, a provision for election of a contingency remedy (and the
waiver of the MCL  ARAR) is made.

X.  EXPLANATION OF SIGNIFICANT  CHANGES

The Proposed Plan for Operable Unit Three at the Whitmoyer Laboratories Site was released
for  comment in July 1990.  The Proposed Plan identified  EPA's preferred alternative.  EPA
reviewed all of the comments submitted during the public comment period.  Upon review of
these comments, it was determined that no  significant changes to the remedy, as it was
originally identified  in the Proposed Plan, were necessary.
                                          49

-------
                                                   Appendix A

                           APPLICABLE OR RELEVANT AND APPROPRIATE REQUIREMENTS
                                             OPERABLE UNIT THREE
                                         WHITMOYBR LABORATORIES  SITE
      Standard,  Requirement,
     Criterion or Limitation,

Chemical-Specific ARAKa

RCM Hazardous Waste
Identification

Haiardoua Watte Identification
Safe Drinking Water Act (SDMA)
Maximum Contaminant Levels
(NCLs)

Clean Air Act (CAR) Air
Emissions

CAA Air Emissions
Air Quality Regulations
Clean Water Act Discharge
Standards

Direct Discharge Standards
Water Quality Standards


Location-Specific ARARs

Federal Floodplairn Management
        Citation
40 CFR Part 261
          Description
RCRA regulations for hazardous
waste identification
                                                               Discueaioa
Soil/sediment and
groundwater media
25 PA Code, Chapter  261   Pennsylvania regulations for    Soil/sediment and
                         hazardous waste identification  groundwater media
40 CFR Part 141
40 CFR Part 50
40 CFR Part 61
25 PA Code,  Chapters
123, 127 and 131

40 CFR Part  122
25 PA Code,  Chapter 92
25 PA Code,  Chapter 93
Executive  Order  119BB
SDWA primary and secondary
drinking water standards
Rational Ambient Air Quality
Standards

National Emissions Standards
for Hazardous Air Pollutants
Pennsylvania air quality
regulations

Clean Water Act standards  for
direct discharges

Pennsylvania direct  discharge
standards

Pennsylvania water quality
standards
Soil/sediment and
groundwatec media


Soil/sediment and
groundwater media

Soil/sediment inciner-
ation and vitrifi-
cation options only

Soil/sediment and
groundwater media

Soil/sediment and
groundwater media

Soil/sediment and
groundwater media

Soil/sediment and
groundwater media
Federal floodplains  management   Soil/sediment medium
regulations                     only

-------
Appendix  A
APPLICABLE OR  RELEVANT AND APPROPRIATE REQUIREMENTS
OPERABLE  UNIT  THREE
HHITMOYER LABORATORIES SITE
Page Two
      Standard, Requirement,
     Criterion or Limitation

 action-Specific AKARa

 RCRA  Incineration
 RCRA Niacellaneoua Treatment
U.S. Department of
Transportation (DOT) Haste
Transportation

RCRA Hazardous Haste
Transportation

Haste Transportation
Occupational Safety and Health
Administration (OSIIA) Asbestos
Management

OSHA Horker Protection
        Citation
40 CFR Part 264
Subpart O


40 CPR Part 264,
Subpart X
49 CFR Parts 107 and
171-179
40 CFR Parts 262 and
263
          Description
                                                          RCRA regulations Cor  hazardous
                                                          vaste incinerators
RCRA regulations for
miscellaneous treatment units
DOT regulations for  Haste
transport


RCRA regulations for the
transport of hazardous waste
25 PA Code,  Chapter  263   Pennsylvania regulations Cor
                         waste transportation
29 CFR Part 1926
29 CFR Parts  1904,
1910, and 1926
RCRA Land Disposal Restrictions   40 CFR Part 268
RCRA Hazardous Waste Landfill
40 CFR Part  264,
Subpart N
OSHA regulations  for asbestos
removal
OSHA regulations  for  the
protection of  workers

RCRA restrictions on  the land
disposal of hazardous waste

RCRA requirements for solid
waste landfills
                                                              Discussion
Soil/sediment
incineration option
only

Soil/sediment
fiMation,biotreatment,
and vitriCication
options, and ground-
water treatment

Soil/sediment and
groundwater media


Soil/sediment and
groundwater media

Soil/sediment and
groundwater media

Soil/sediment medium
only


Soil/sediment and
groundwater media

Soil/sediment and
groundwater media

Soil/sediment and
groundwater media

-------
Appendix  A
APPLICABLE OR RELEVANT AND  APPROPRIATE REQUIREMENTS
OPERABLE  UNIT THREE
WHITMOYER LABORATORIES SITE
Page  Three
     Standard, Requirement,
     Criterion or  Limitation

 Hction-Specific ABARs Icont*d)

 RCRA Gioundwater Monitoring


 RCRA Solid Haste Landfill


 Solid Naste Nanagenent


 Erosion Control
 RCRA Landfill Closure and
 Post-Closure

 Fish and Wildlife Coordination
 Act
        Citation



40 CPR Part 264


RCRA Subtitle  D


25 PA Code, Chapter 75


25 PA Code, Chapter 102


40 CFR Part 264


16 USC 661
          Description
RCRA groundwater Monitoring
requirenents

RCRA requirements  for solid
waste landfills

Criteria Cor  siting and
operating landfills

Erosion control limits on
excavation activities

RCRA regulations Cor landfill
closure and post-closure

Governs modifications to
bodies of water
      Discussion
Soil/sediment and
groundwater media

Soil/sediment and
groundwater media

Soil/sediment and
groundwater media

Soil/sediment medium
only

Soil/sediment medium
only

Soil/sediment medium
only

-------
                            RESPONSIVENESS SUMMARY
                              OPERABLE UNIT THREE
                         WHITMOYER LABORATORIES SITE
             JACKSON TOWNSHIP, LEBANON COUNTY, PENNSYLVANIA

From July 16, 1990 through December  3. 1990, the  U.S. Environmental  Protection Agency
(EPA)  held   a  public comment  period  on  the  Proposed  Plan   and  the  Remedial
Investigation/Feasibility Study (RI/FS) for Operable Unit Three of the Whitmoyer Laboratories
Site in Lebanon County, Pennsylvania. This responsiveness summary summarizes comments
on the Proposed Plan and RI/FS pertinent to Operable Unit Three that were  expressed by local
officials, state officials, and other interested parties; and provides EPA  responses to the
comments.

This  responsiveness summary is divided  into the following sections:
       *      Overview

       *      Background on Community Involvement

       *      Summary of Comments Received during Public Comment Period and Agency
             Responses

       *      Remaining Concerns

A.     OVERVIEW

At the time of the public comment period, EPA had already endorsed a preferred alternative for
Operable Unit Three of the Whitmoyer Laboratories Site.  EPA's recommended alternative
addressed contaminated soils/sediments and groundwater at the site.  The preferred alternative
specified in the Record of Decision (ROD) for Operable Unit Three involves:

       *      Excavation  of all moderately contaminated soils/sediments from offsite and
             saturated onsite locations, and all heavily contaminated soils  [estimated volume
             = 116,000 cubic yards (CY)].

       *      Demolition of Buildings 4, 9,  11, and 14.

       *      Backfilling of the excavated areas with clean fill or lightly contaminated soil

       *      Onsite fixation of  the approximately 61,000  CY  of heavily  contaminated
             soils/sediments using an iron-based or other similar fixation process that provides
             equivalent protection.

       *      Biological treatment of the approximately 5,600 CY of soils with organic chemical
             concentrations above the heavily  contaminated soil action levels either prior to
             or following the fixation step.

-------
       *      Consolidation of the moderately contaminated soils/sediments on site above the
              groundwater table.

       *      Capping  of  the  approximately  39,000  CY  of  moderately  contaminated
              soils/sediments having contaminant concentrations above groundwater-based
              unsaturated soil cleanup targets with low-permeability materials.

       *      Soil capping  of all soils/sediments remaining on the surface following the
              excavation and consolidation steps  that are not capped with low-permeability
              materials and contain greater than 21 mg/kg arsenic; and other disturbed areas
              as needed.

       *      Grading and revegetation of all areas affected by the soil/sediment remediation.


       *      Following the soils/sediments remediation, use of institutional controls for areas
              with remaining contamination.

       *      Aggressive extraction of all groundwater from the  aquifer beneath the site with
              concentrations above  health-based levels  (e.g.,  50  ug/l  arsenic)  until  the
              maximum groundwater contaminant concentrations are all  less  than  health-
              based levels.

       *      Treatment of the extracted groundwater in an onsite treatment plant, utilizing
              physical, chemical and possibly biological processes.

       *      Disposal of the treated water by either discharging it to Tulpehocken Creek,
              reinjecting it into the aquifer, or a combination of the two methods.

       *      Salvaging nonhazardous demolition debris, as feasible.

       *      Disposal of the following in offsite landfill(s) in  accordance with all applicable
              regulations: the treated soils;  the groundwater treatment residuals; and the
              demolition debris that  is not salvaged.

It may potentially prove technically  impracticable  to achieve  the health-based  groundwater
cleanup goals under the selected remedy for the groundwater.  If information emerges from the
operation of  the  selected  remedy  system  that  strongly suggests that  it is technically
impracticable to achieve the cleanup goals throughout the contaminated groundwater plume
because of an observed "leveling-off" of contaminant concentrations, the EPA,  in consultation
with the Commonwealth of Pennsylvania, intends to implement a contingent remedy in those
areas where the cleanup goals will  not  be met.  The contingent remedy is similar to the
selected remedy, with the exception that groundwater would  only be extracted  from the
perimeter of the non-attainment area in sufficient quantities to keep this area from growing

Judging from  the comments received during the  public comment period, the Pennsylvania
Department of Environmental Resources (DER) strongly supports the preferred alternative.  The

-------
Jackson Township Board of Supervisors does not  support the preferred alternative  The
present owner of a property that contains some of the contaminated soils/sediments (Buckeye
Pipe Line Company) questions EPA's recommendation  of excavating, treating, and disposing
off  site  the  most contaminated  soils/sediments and  managing the  less  contaminated
soils/sediments  through engineering controls.  They also question the prudence and cost-
effectiveness  of an  aggressive aquifer  restoration program.   Rather,  they  support  the
implementation of a soil flushing alternative for the contaminated soils/sediments, and a plume
containment remedy for the contaminated groundwater.  The former site owners, Rohm & Haas
and  SmithKline  Beecham, support the alternatives of  pumping and treating only  the most
contaminated groundwater and flushing contaminants from the contaminated soils/sediments
using treated  groundwater. Under the  soil flushing alternative, contaminants mobilized during
the flushing would be captured in the groundwater extraction system.   These issues  are
discussed in Section  C below.

B.     BACKGROUND ON COMMUNITY INVOLVEMENT

There has been consistent community interest in the Whitmoyer site since its proposal for listing
in 1984,  Public meetings have attracted approximately 50 local residents as well as most major
local media.  In spite of this interest, however, there have been no attempts to organize any
formal special interest groups to address the Site or apply for a Technical Assistance Grant

With respect to  Operable  Unit Three, there has been little community interest in the remedy
proposed by EPA for this Operable Unit. Local officials, however, have expressed concern over
EPA's proposed plan for Operable Unit Three, though much of their concern was focused on
Operable Unit Two.

C.     SUMMARY OF COMMENTS RECEIVED DURING PUBUC COMMENT PERIOD AND
       AGENCY RESPONSES

Comments raised during the public comment period for  Operable Unit Three of the Whitmoyer
Laboratories Site are summarized below.  EPA responses to the comments are provided  The
comments are categorized by relevant  topics.
Impact on Surrounding Community

1.     The former Whitmoyer Laboratories, Inc. owners commented that the preferred alternative
      for remediating soils/sediments would  extremely disrupt the  community through the
      excavation, movement, and placement of 313,500 cubic yards (CY) of  material and
      transport of over 35,000 truck loads through the community.

      EPA Response: Using the FS estimates,  approximately 197,000 CY of materials (130.00C
      CY of clean soil, 48,000 CY of fill, 12,500 CY of clay, and 6,000 CY of ferric sulfate/iime
      would be brought to the site, and 70,000 CY of materials (68,000 CY of treated soils arc
      1700 CY of demolition debris) would be transported from the site. These quantities to;a>

-------
       to 267,000 CY.  Most of these shipments would likely occur in 15 or 20-CY trucks.
       Using 15-CY trucks, approximately 17,800 truck loads would be required to move this
       total quantity of material.  Assuming a 12-hour workday, 250 workdays per year, and a
       3-year remediation, this quantity equates to approximately 2 truck trips per hour.  Given
       the cleanup effort required, EPA does not consider this additional traffic to be extremely
       disruptive.   (Note: there  is  significant  current  truck traffic associated  with  the
       warehousing operation on site).

2.     The former owners commented that EPA did not assess the risks associated with
       implementation of the soil/sediment remedial alternative, including inhalation of fugitive
       dusts,  dermal contact with arsenic  laden dust,  the  increased  likelihood  of  traffic
       accidents on area roads, and erosion of soils during remediation.

       EPA Response: EPA disagrees. EPA has evaluated the long- and short-term risks from
       implementation of the preferred soil/sediment alternative. These risks were determined
       to be acceptable and will be further evaluated during  the Remedial Design  phase of
       project implementation.  For example,  fugitive dust emissions during demolition and
       excavation  are expected to be minor,  and appropriate  monitoring or ambient  air
       conditions will be conducted during the Remedial Action to ensure the acceptability of
       any  emissions.  If necessary,  air pollution control  measures, such as wetting these
       materials, will be implemented to minimize fugitive dust releases. A pathway for dermal
       contact with arsenic laden dust does not exist at the Site.  With respect to risks from
       transportation accidents, the selected remedy is equal to any other alternative meeting
       the threshold criteria for protection of human health and the environment.  Finally, the
       selected  remedy will be implemented in accordance with Pennsylvania erosion control
       requirements and all appropriate ARARs.

3      The former owners commented that, under EPA's preferred groundwater alternative, the
       groundwater withdrawal may result in the drying up or lowering of production rates of
       agricultural, livestock, industrial and commercial wells in the area.

       EPA Response:  EPA disagrees. It is unlikely that local wells will be affected by the
       proposed groundwater withdrawal program to any great extent.  The proposed Remedial
       Action  will  be  designed  so as not  to  significantly  lower  the  water table in the
       surrounding area. To ensure such a system will work properly, local conditions will be
       carefully  monitored  during the  implementation phase  of the project and  appropriate
       adjustments made during the Remedial Action to mitigate any such effects.

4.     The former owners commented that, under EPA's preferred soil/sediment alternative the
       Tulpehocken Creek ecosystem would unnecessarily be destroyed, and the historic locks
       and potentially significant archaeological areas in the northeast corner of the site wouic
       be threatened  unnecessarily.

       EPA Response:   Unfortunately,  it is  necessary to disrupt (but not destroy) a smai!
       portion of the Tulpehocken Creek ecosystem to carry out the soil/sediment remediation
       Mitigative measures as  described in the ROD will be implemented to minimize the
       temporary ecosystem damage.   During the remediation, the historic  locks will ce


                                           4

-------
       carefully dismantled and archived. Following completion of the soil/sediment excavation,
       the locks will be restored as a part of the Tulpehocken Creek/Union Canal restoration
       program.

 5.     The former owners commented that the implementation of the  preferred alternatives for
       Operable Units Two and Three would result in a massive engineering project which
       would indirectly disrupt the community.

       EPA Response: EPA does not believe the selected remedies for Operable Units Two and
       Three will significantly disrupt the community. (See the responses to Comments 1 and
       2.)  While the construction of several treatment plants  on  site will be significant,
       construction management practices would be used to efficiently schedule the work and
       minimize community disruption. For example, site structure demolition could occur prior
       to initiation of the soil/sediment excavation program.  Remediation of Operable Unit One
       (the concentrated liquids) has been scheduled for completion prior to other site remedial
       activities, and has already  been completed with minimal community disruption.
Appropriateness of Preferred Soil/Sediment Alternative

6.     The former owners commented that the 21  mg/kg arsenic surface soil/direct contact
       action level is overly conservative, because it is based on unrealistic exposure scenarios
       and the unjustified use  of the overly conservative 10** excess  lifetime cancer risk
       cleanup level.  They commented that a surface soil action level of 200 mg/kg arsenic
       would be more appropriate.

       EPA Response: EPA  believes the 21  mg/kg arsenic surface soil/direct contact action
       level is based on a reasonable maximum residential use exposure scenario given the
       existing and historical residential development patterns in the site vicinity.  Also, EPA
       believes that use of the 1 x 10"6 excess lifetime cancer risk cleanup level is appropriate
       and justified as described in the NCP and in EPA guidance on risk assessment

7.     The former owners commented that the groundwater dilution factor of 4.25 used to
       derive groundwater-based unsaturated soil cleanup levels is in error, and a factor of 5 1
       is more appropriate.

       EPA Response: The former owners base their conclusion on two assumptions that differ
       from the EPA assumptions. First, the former owners assert that percolating rainwater
       should be assumed to mix with the entire groundwater underflow beneath the site EPA
       on the other hand,  believes  that only mixing  with shallow  groundwater  should be
       considered.   EPA  believes its assumption  of  infiltration  mixing with only  shallow
       groundwater is more representative of site conditions based on the data collected during
       the Remedial Investigation and described in the Rl and FS reports for the site. Second
       the former owners suggest that the infiltration through the lagoon area should  not be
       considered, because there will be no contaminated soils present there following lagoon
       waste  removal.  As presented in the soils FS report (Figure 4-3), the lagoon area will

-------
       likely  be  utilized  for  soil/sediment  consolidation under the  preferred alternative.
       Therefore, EPA believes it is appropriate to include the estimated infiltration rate for the
       lagoon area in the dilution calculation.

8.      The former owners commented that no groundwater dilution factor was used to derive
       groundwater-based saturated soil cleanup levels, and that the use of a factor of 5.07 is
       more appropriate than  none at all.

       EPA Response:   The  former  owners base their  proposed  dilution  factor on the
       assumption  that  saturated soil leachate should  be assumed to mix with the entire
       groundwater  underflow beneath the site.  EPA, on the  other  hand,  believes that
       assumption of any mixing dilution is inappropriate for saturated soils.  Saturated soils
       at the Whitmoyer Laboratories primarily consists of alluvial materials, including significant
       quantities  of sands and gravels.  It is conceivable that water supply weits  could  be
       directly completed in the saturated soils.  Therefore, since it is possible for  receptors
       to directly ingest saturated soil leachate, EPA  believes it is more appropriate to not
       consider any potential  dilution  effects.

9.      The former owners commented that saturated zone attenuation factors should be applied
       when calculating groundwater-based cleanup levels, since site data indicate significant
       attenuation of contaminant concentration levels at increasing distances from the  site

       EPA Response:   The former owner argument is based on the  presumption that the
       remedial action levels should be achieved 500 feet downgradient of the site, and that
       significant attenuation should occur over this distance. This argument is in direct conflict
       with EPA policy.  EPA policy states that cleanup levels should be achieved for the entire
       groundwater plume (area of attainment), and not only at some point downgradient from
       the site. The approach proposed by the former owners would only be acceptable if the
       use of Alternate Control Limits (ACLs)  were justified. Adoption of ACLs is not warranted
       for the Whitmoyer Laboratories Site, based on site conditions ("Guidance for Remedial
       Actions for Contaminated Ground Water at Superfund Sites", OSWER Directive 9283 1 -
       2, December 1988).  Therefore, use of attenuation factors as proposed by the former
       owners is  inappropriate.

10.    The former owners commented that unsaturated  zone attenuation factors should also
       be applied when calculating groundwater-based unsaturated soil cleanup levels,  since
       site data indicate significant attenuation of contaminants with depth.

       EPA Response:  To  properly implement the former owner proposal, accurate vadose
       zone attenuation levels would have to be derived (sufficient information does not now
       exist), and cleanup levels would have to be derived for each interval (e.g., foot) the soil
       lies above the water table or bedrock. In other words, a higher attenuation factor would
       be appropriate for soils present 8 feet above the  water table/bedrock than soils only i
       foot above this level.  Implementation of such a system would be unworkable   The
       groundwater-based soil action levels derived  in the soils FS  report are based  on
       partitioning using either site-specific partitioning data or estimated partition coefficients
       based on the organic carbon content of site soils (the K^-f^ model, as presented m

-------
       "The Soil  Chemistry of Hazardous  Materials", Oragun,  J., 1988).   These  partition
       coefficients take into account contaminant sorption, which is believed to be the primary
       reason for the observed decreased arsenic concentrations at depth noted by the former
       owners.    EPA  believes the  methodology  used  to  derive the  groundwater-based
       unsaturated soil action levels is appropriately conservative for the protection of human
       health.

11.     The former owners commented that using the Organic Leaching Model (OLM) instead
       of  the KoC-foc  model  would have  resulted in more plausible  groundwater-based
       soil/sediment cleanup levels.

       EPA Response:  EPA believes the KoC-foc model is  more representative of site-specific
       conditions, since it incorporates the important variable of the total organic content (TOO)
       of the site soils.  Typically, the extent of organic chemical sorption directly correlates
       with the TOO content of soils (Dragun, 1988). The OLM model is generally only used
       when site-specific soil TOC data are not available and does not incorporate the use of
       TOC.

12.     The former owners commented that the main groundwater-based soil/sediment concern
       at  the site is not  the  possibility that  groundwater will  be contaminated by water
       percolating through contaminated soil, but  rather the opposite  mechanism of soils
       becoming contaminated by groundwater during periodic high groundwater levels  The
       site data suggest that contamination of the groundwater from soils containing arsenic
       is not  a contaminant source or pathway of concern.

       EPA Response:  EPA strongly disagrees with the former owners. The soils are one of
       the primary sources (if not the primary source) of continuing groundwater contamination
       During the RI/FS, soil/sediment leachate concentrations of 26,800 ug/l arsenic, over 500
       times the  groundwater cleanup level of 50 ug/l, were measured. Percolating rainwater
       with similar concentrations will continue to contaminate  groundwater if  no action or
       ineffective action is taken. If the soil (vault, lagoon, miscellaneous products/feedstocks.
       and site  structures) remediation(s) do not  take  place,  the continuing releases to
       groundwater from  these media will  substantially lengthen the time  required for
       groundwater remediation (if  not render it technically impracticable).

13.     The former owners commented that remediation of sediments is unjustified.

       EPA Response:  EPA disagrees.  Remediation of a portion of the sediments is justified
       as  it is impracticable to leave existing sediments adjacent to soils to be  excavated m
       place during the  preferred alternative soil excavation program. Therefore, these adjacent
       sediments will be excavated at the same time as the soils and managed in accordance
       with their  contaminant  levels.   However,  EPA  will not  be remediating  sediments
       downstream of the soil excavation area.

-------
14.    The former owners commented that the sediment exposure scenario presented in the
       soils FS report is unrealistic, and that the scenario presented in the Rl report is more
       appropriate.

       EPA Response: EPA agrees that the scenario presented in the Rl report is appropriate
       for evaluating exposures to the in-place sediments,  but not for evaluating excavated
       sediments.  The exposure scenario presented in the  FS report refers to potential
       exposure to the  sediments  once they are excavated  and managed (e.g., consolidated
       on site), and not to in-place exposure. Since we are dealing with excavated sedimants,
       the FS scenario  is appropriate.

15.    The former owners commented that the setting of independent action levels for heavily
       contaminated soils makes no sense.

       EPA Response:  EPA disagrees. The National  Contingency Plan (NCP)  requires the
       identification of the principal threats  posed by Superfund sites.  In this case, EPA has
       determined that those soils with at least one contaminant having a concentration above
       the heavily contaminated ("principal threat") soil action levels pose one of the principal
       threats at the site.  EPA expects to use treatment to address principal threats posed by
       a site, wherever practicable, and to use engineering controls, such as containment, for
       waste that poses a relatively low long-term threat or where treatment is impracticable
       (40 CFR 300.430).  EPA developed its soil/sediment remedial alternatives in accordance
       with these requirements.

16.    The former owners commented that EPA applied a totally arbitrary and scientifically
       insupportable uncertainty factor of 2.1 to derive a heavily contaminated soil action level
       of 1000 mg/kg for arsenic.

       EPA Response: EPA disagrees. EPA evaluated using either a total arsenic content action
       level or Toxicity Characteristic Leachate Procedure (TCLP) extract arsenic content action
       level for soils contaminated with  arsenic.  Because of the high cost of each TCLP
       analysis, it was decided to base the principal threat action level on a total arsenic basis
       rather than a teachable arsenic basis, using a statistical correlation between total arsenic
       concentrations found in the soil  and the concentrations of arsenic found in the leachate
       based on  the TCLP testing procedure.  Given the variability in  the soils TCLP data a
       "safety factor"  approach  was developed, with  the goal of no more  than 10% false
       negative.   (The  10% chance  of a false  negative  was  chosen using  engineering
       judgement.) This approach is  consistent with RCRA (40 CFR 261) requirements lor
       hazardous waste determinations. Using the statistical approach, a safety factor of 2 i
       is  necessary to  achieve  the goal of no  more than  10%  false  negatives.  It  was  not
       deemed appropriate to  apply  a safety factor  to the regression curve developed to
       estimate which soil arsenic concentrations corresponded with leachate concentration at
       the arsenic MCL, because the cost of a false negative (i.e., a groundwater concentration
       of 55 ug/l  arsenic in a portion of the aquifer) is not substantial.

17.    The former  owners commented that EPA's preferred alternative of treating arsenic
       contaminated soils by iron fixation will not work, as the apparent success noted in soils


                                          8

-------
      fixation treatability study is a product of the testing procedure and not the treatment
      itself.  Water is necessary for  the desired reaction to take place.  EPA's proposed
      treatment would occur in dry or slightly moist conditions.

      EPA Response: EPA agrees that water is necessary to complete the desired reaction.
      Water was added during the treatability study, and a liquid phase reaction was allowed
      to occur.  The selected remedy does not preclude adding water to the soil/fixative
      mixture if desired. The decision on whether or  not to add water to the mixture will be
      made during the remedial design phase.

      EPA currently believes that adding water during treatment is not essential.  One must
      look at the big picture. Once the soils/sediments are treated,  they will be placed in a
      landfill.  If the  soil/fixative mixture is  not wetted and the landfill cap is effective, the
      treated material will  never be wetted and  contaminants will not be  removed by
      percolating water. If the cap is  not completely effective and the treated soils/sediments
      become wetted, the desired reaction will occur and the arsenic present in the materials
      will be immobilized.  In either case, significant arsenic releases should not occur.

18.    The former owners commented that no meaningful evaluation of the effectiveness of the
      remedial alternatives has been  performed, since the chemical characteristics of arsenic
      relevant to the site  and the  soil  properties  governing  arsenic  behavior  were not
      completely characterized.  There is no  scientific basis to conclude the soil  remedy
      selected by EPA will be effective.

      EPA  Response:  EPA disagrees. Sufficient information exists  to support the selection
      of a  soils/sediment remedy.  The remedy selection  process followed by EPA tor the
      heavily arsenic-contaminated soils/sediments is typical of the Superfund remedy selection
      process.   Treatability  testing indicated that the selected iron fixation  remedy will be
      capable of achieving  the remedial objectives (e.g., complying with the RCRA LDRs) on
      a worst-case sample.  The iron fixation bench-scale study was conducted in accordance
      with EPA protocols. The treatability study data were validated using standard EPA data
      validation procedures. Successful fixation of arsenic compounds with iron has also been
      reported  by other researchers.  (See, for example, the American NUKEM data presented
      in the discussion of the RCRA "third-third" land disposal restrictions promulgation notice )
      Further optimization  of the treatment process will occur during the remedial  design
      phase of the project.

      The missing information cited by the former owners (e.g., the effects of calcium content,
      soil iron  oxide content, soil surface area, and soil  phosphorus  content on arsenic
      behavior in  soil)  would be extremely  difficult and  expensive to  collect.   Review of
      available data show it is extremely unlikely that collection of this information would result
      in  the discovery of  a more  viable (cost-effective) treatment for the  soil arsenic
      contamination.

 19.   The former owners expressed concern that the iron fixation technology may fail over
       long time periods.

-------
       EPA Response: EPA disagrees. The RI/FS treatability data indicate that small quantities
       of arsenic may be released from the iron-fixated wastes over extended periods upon
       exposure to the atmosphere.  However, this possibility will be precluded by placement
       of the treated waste in a capped and lined offsite landfill.

20.    The former owners commented that the six fixating agents tested by EPA in their soil
       fixation study do not represent all of the agents available for commercial use, and to that
       extent the study was incomplete.

       EPA Response: EPA disagrees  as to the ineffectiveness of the fixation process. The soils
       fixation study was phased  to allow the efficient evaluation of candidate fixating  agents.
       When promising results were  achieved with  the  low-cost iron fixative during the first
       phase, second phase work  concentrated on the validation of this agent.  No other
       potentially effective fixatives with costs comparable to the iron fixative were identified by
       EPA during the screening  of candidate fixatives.  If others were identified, they would
       have been tested during the study.

21.    The former owners commented that relatively high dissolved salt concentrations  may be
       generated during the fixation process.  These salt concentrations can enhance metals
       mobility in the  environment and therefore may defeat the whole purpose of the fixation
       process.

       EPA Response: EPA disagrees as to the ineffectiveness of the fixation process. The
       fixated wastes were demonstrated to be stable in the soils fixation study in the presence
       of three lixiviants: TCLP extraction  fluid (acetic acid), distilled water, and  background
       groundwater.  The stability of  the treated materials will  be further enhanced by their
       placement in a capped and lined landfill.

22.    The former owners commented that actual operation of the iron fixation plant would be
       a technical nightmare, since iron fixation treatment on this kind of scale is unproven
       For example,  crushing  and grinding of the  excavated  soils/sediments will result  in
       considerable noise and  dust.

       EPA Response: While EPA agrees that iron fixation treatment is innovative  and  has not
       been conducted on the scale  as outlined in  the  selected remedy, plant operation will
       not be a technical nightmare. The preliminarily designed treatment plant identified in the
       soils FS report included the  use of screens,  conveyors, pug mills, chemical silos, and
       standard air pollution control  equipment.  These pieces of equipment are routinely
       utilized by industry; their use would not be a technical nightmare. It is uncertain  whether
       crushing and grinding will be  required as  a  part of the plant flowsheet.  This will be
       determined during the final design.  However,  because very little of the soils and
       sediments  will require  crushing,   considerable noise will not  be generated  and
       engineering measures (e.g., wetting the crusher  feed if  necessary) will minimize  dust
       generation.

23.    The former owners commented that the EPA proposal of placing the treated wastes m
       an  intermediate landfill will   not  be  implementable. since  it is  unlikely that any


                                          10

-------
       intermediate landfill  will accept the treated material because of concerns about the
       efficacy of the treatment.  If EPA is unable to identify an intermediate landfill to accept
       the treated material and the material must instead be sent to a hazardous waste landfill,
       the cost of the preferred alternative will increase by up to $15 million.

       EPA Response: EPA disagrees. EPA believes the iron treatment will work and the treated
       material  will  be  accepted at intermediate landfills  based on  EPA contacts with such
       landfills during the RI/FS process.

24.    The former owners commented  that  EPA should not  select bioremediation as the
       treatment technology for soils heavily contaminated with  organic chemicals, since it is
       not certain that the treatment will work.

       EPA Response:   Bioremediation treatment is an innovative technology; it is not certain
       that it will work on the Whitmoyer Laboratories soils/sediments.  The majority of the
       primary organic contaminants (e.g., aniline and polycyclic aromatic hydrocarbons) have
       been effectively biodegraded in numerous other site applications.  It is possible, as
       explained in the Feasibility Study, that fixating the soils with iron prior to initiation of the
       biotreatment will reduce any arsenic inhibitory effects presented by the soils/sediments
       The viability of bioremediation will be confirmed during bench-scale optimization studies
       to be conducted during the remedial design for the contaminated soils/sediments.

       If biodegradation is successful, its implementation will be protective of human health and
       the environment; will comply with applicable or relevant and appropriate  regulations
       (ARARs) (e.g., RCRA land disposal restrictions); will be effective  in  the long-term and
       permanent; will involve treatment of the principal threats posed  by  this operable unit;
       will reduce toxicity, mobility, or volume  through treatment; will comply with the  CERCLA
       preference  for treatment; will be more cost-effective than  other  organic chemical
       treatment options (e.g., vitrification, incineration, and fixation with activated carbon); and
       will not present the incremental risks to human health and the environment posed by
       arsenic emissions during the thermal treatment (vitrification and incineration) options
       (These risks were determined to  be acceptable.)   The NCP requires  EPA to  consider
       using innovative technology when such technology offers the potential for comparable
       or superior treatment performance or implementability, fewer or lesser  adverse impacts
       than  other  approaches,  or lower costs  for similar  levels  of  performance than
       demonstrated technologies.  The  NCP  further requires EPA to use alternative treatment
       (i.e., innovative) technologies to the maximum extent practicable (40 CFR 300.430). The
       selection of the biotreatment for the heavily organic contaminated soils/sediments is in
       accordance with these requirements.

25.    The former owners commented that EPA did not explain where and how the soils to be
       biologically treated would be stored during the estimated 5 years the bioremediation
       process will take. They suggested that storage would not meet state ARARs.

       EPA Response: EPA disagrees. EPA believes that based on data  as set forth in the
       Feasibility Study, the soil/sediment excavation program will not be completed until nearly
       three years after its initiation. Thus, it is unlikely substantial quantities of soils/sediments


                                           11

-------
       would be stored for 4.5 years prior to treatment initiation.  Excavated soils/sediments
       awaiting  biotreatment would be stored (if necessary) in an  environmentally acceptable
       manner,  in compliance with state and'federal ARARs.

26.    The former owners commented that leachate generated during the biological treatment
       may contain chemical concentrations in excess of secondary drinking water standards,
       and that  there is the potential for ground water pollution if a liner failure occurs during
       landfarming.

       EPA Response: The biological treatment system will be designed to meet ARARs and
       to prevent groundwater pollution from the biotreatment operations.  It is possible that
       the remedial design will consist of biological treatment in a  bioslurry reactor vessel.  In
       this case, liner failure would  not be possible. However, a monitoring system will be
       designed and installed for detection of any such  failures and should a failure be
       detected, appropriate measures for remediation will be taken.

27.    The former owners commented that significant quantities of volatile organic chemicals
       would volatilize during soil/sediment storage and the biological treatment process.  This
       volatilization would raise significant air pollution problems.

       EPA Response:   EPA  disagrees.  All of the heavily contaminated ("principal threat')
       soils/sediments spread across the site  contain an estimated total of only 8 pounds
       (approximately one gallon) of volatile organic chemicals.  This figure includes those
       soils/sediments that will be and will not not be biologically treated.  Because of these
       small quantities, no adverse effects on human health and the environment would be
       expected.  Nonetheless, air pollution controls, if  required, will be specified during the
       remedial  design phase of the project.

28     The former owners commented that the estimated duration of the selected soil/sediment
       remedy is uncertain. In the Proposed Plan, a duration of 5 years is presented, whereas,
       in Appendix E of the FS,  a 9-year  duration is estimated.

       EPA Response: EPA estimates the  remediation will last 5 years,  once it is initiated.  The
       9-year figure presented in the soils FS appendix is an error.   This figure should  be 3
       years. The difference in durations  represent time required for bench-scale testing and
       construction, as well as the time for additional lifts if a multiple lift system is  specified
       during the remedial design phase.

29     The former owners commented that the limited availability of equipment and resources
       for performing bioremediation projects will result in delays  for the proposed treatment
       of the contaminated soils/sediments.

       EPA Response: EPA disagrees. As  discussed in the Feasibility Study Report, equipment
       and resources are readily available to  conduct the  proposed  Remedial Action While
       minor delays are possible, these will not have a significant effect on overall remediation
       of the site.
                                          12

-------
30.    The former owners commented that EPA should not have rejected soil flushing as a
       remedial technology, but rather should select this technology as the soil/sediment
       remedy.  The present owner of a property  that contains some of the contaminated
       soils/sediments (Buckeye Pipe Line Company) also stated that in situ leaching (i.e soil
       flushing) may be a better alternative.  The former owners stated that the soil flushing
       remedy will be completed well before groundwater cleanup levels are achieved.

       EPA Response: EPA gave serious consideration to the soil flushing alternative proposed
       by the former owners.  While this alternative is purportedly less costly than the selected
       remedy (according to the former owners' cost estimate), EPA believes that the selected
       remedy  is  cost-effective  in  mitigating  the   risks  posed  by  the  contaminated
       soils/sediments in a reasonable  period of time (an estimated 5 years) and meets all
       other requirements of CERCLA. Because the majority of organic contaminants present
       in the heavily contaminated soils/sediments will be destroyed, the metal contaminants
       in these materials  treated to reduce mobility, the treated material disposed  in  an
       intermediate landfill;  and the less contaminated soils/sediments will be managed using
       a combination of engineering and institutional controls, the selected remedy affords a
       high degree of long-term effectiveness and permanence.

       Under the soil flushing system, there is the  danger of mobilized contaminants  in the
       flushate to escaping the well capture  network and  substantially  further  degrading
       downgradient groundwater, considering the  karst setting and complex hydrogeology
       present at the site.   Using the former owners' calculations, it will take 96,000 years for
       percolating rainwater to wash out arsenic contamination from the soils to the  cleanup
       level of 50 ug/l, given the assumption that one-half of a pore volume is exchanged per
       year. The  former owners estimate that 5 pore volume exchanges per year would occur
       during soil flushing.  Given this assumption, the cleanup would occur ten times as fast
       as for no action; and the leachate from the most contaminated soil would reach the
       cleanup level of 50- ug/l arsenic in 10,000 years. This duration is obviously beyond the
       realm  of practicability.  (EPA estimates  that only 2 pore volumes per year could be
       transmitted by the  aquifer.  If this  is the case, the  cleanup duration would last an
       estimated 24,000 years.)

       In comparison, the soil/sediment remediation could be completed  in an estimated 5
       years.   The duration  required  for groundwater  cleanup  is uncertain.  However
       extrapolating the contaminant reduction rates  achieved during the 7-year pump-and
       treat program, the aquifer could be cleaned up to the arsenic standard of 50 ug/l m 13
       years. (During the pump-and-treat program, aquifer arsenic concentrations were reducec
       by one order of magnitude roughly every 3.7 years.)

31     The former owners commented that EPA's calculations indicated it would take about
       100,000 years for the leachate concentrations of the most contaminated soils  to reac*
       50 ug/l, whereas the former owners calculations showed it would take about 96 000
       years.  This apparent discrepancy shows the sloppiness in the soils FS report

       EPA Response: EPA disagrees.  For all intents and  purposes, these numbers are tre
       same.


                                          13

-------
32.    The former owners commented that up to 67% of the arsenic was removed from site
       soils in one pass during the soil washing treatability study, and implied that an ongoing
       soil flushing  involving continuous washing over a long period of time would achieve
       greater removals in a reasonable timeframe.

       EPA Response:  The soil washing data  using  aggressive leaching solutions is not
       comparable  with the former  owners'  proposed program that uses plain  water  as  a
       lixiviant. Using the former owners' own calculations, it will take between 10,000 years
       and 2- ,000 years to achieve the soil/sediment remedial action objectives. (See the
       response to  Comment 30 and Figure 1 of "Comments on the Proposed Plan for Soils.
       Sediments and  Ground  Water at the  Whitmoyer Laboratories Site,"  GeoTrans, Inc,
       September 1990).

Appropriateness of Preferred Groundwater Alternative

33.    The former owners commented that selection of EPA's preferred groundwater alternative,
       aggressive pumping-and-treating (Alternative 4), is not warranted, since the groundwater
       remediation will likely require  35,000 years and the groundwater plume is only growing
       at a rate of one  foot per year.  Even if the estimated cleanup duration of 160 years
       presented  in  the  FS report (February 1990) is accurate,  EPA's preferred groundwater
       alternative should be considered technically infeasible.  When the contingent remedy of
       plume containment is inevitably selected, the 600 gallon per minute (GPM) groundwater
       treatment plant will be grossly underutilized.

       The present owner of a property that contains some of the contaminated soils/sediments
       (Buckeye Pipe Line Company) commented that EPA's preferred alternative includes more
       wells than is necessary for protection  of public health and the  environment.  A less
       costly alternative for groundwater  that would afford  similar  protection could  involve
       Alternative 2  plus a set of wells to intercept groundwater  containing greater than 1.000
       ug/l arsenic.

       EPA Response:  The  former owner  calculations  of the 35,000-year groundwater
       remediation duration  and the groundwater plume  rate  of growth are based  on the
       conceptual model that all of  the groundwater fractures are completely filled with soils
       exhibiting the same partitioning behavior observed during  the bench-scale testing of site
       soils. A similar approach was presented in the groundwater FS report (February 1990)
       Based on the new partitioning data, EPA does not agree with this approach   if the
       fractures were completely filled with  soils  having  a  similar partitioning  behavior
       contaminant  concentrations measured in offsite wells and springs  (e.g., the Wagner
       residential  well and the PJ Valves Co. well)  in 1967 could  not  be explained   Since
       arsenic was reportedly first used at the site in 1957, it would have been impossible for
       the observed contamination to travel to the offsite observation points in only 10 years
       given the soil partitioning characteristics.  Therefore, EPA believes that the conceptual
       model of the bedrock fractures being only partially filled with soils is more accurate ci
       existing  site  conditions.  This model is supported by observations made during the
       drilling programs conducted by both Rohm & Haas and EPA.  During both programs


                                         14

-------
it was noted that some fractures were filled with soils while others were "clean"  This
groundwater  conceptual  model was  calibrated  using  site data, most notably the
groundwater concentrations noted in offsite wells in 1967, the assumed 10-year travel
time, and estimated groundwater flow rates.  Using these data, a "bulk Kg" (partition
coefficient)  of 5  is  estimated  for  the  aquifer.   (Standard Calculation  Sheet Re:
Groundwater  Modelling Calibration (reruns) 12/15/90). (This logic was similarly utilized
for the assumption that the soils had a Kg of 5 in the modeling presented in the FS
report.)  The  actual bulk Kg for the aquifer may be significantly  less than 5, as the
offsite contaminated wells and springs may have been significantly contaminated prior
to 1967.  A Kg lower than 5 would produce a predicted cleanup time frame lower than
160 years.  Given this conclusion, EPA still believes the modeling presented in the
groundwater FS report is a reasonable rough estimate.

The former owners pointed out that the change in arsenic Kg value with concentration
explains how  the groundwater became contaminated in a short period of time, but the
groundwater plume will now only grow at a rate of one foot per year. However,  much
of the plume growth in the 1960's occurred while arsenic concentrations (at the plume
perimeter) were approximately  1 mg/l or less.  At these concentrations plume growth
under the former owner model should have been very slow, and contamination should
not have reached outlying wells such as the Wagner well and the PJ Valves well.  Thus,
the former owner model does not appear to accurately represent  site conditions

Further support to the refined EPA model is provided by reviewing the historic data of
the onsite wells (e.g., well 7) and offsite well 16B.  Arsenic concentrations measured in
well 16B historically have been approximately  one-half the concentration measured m
well 7, despite the fact that well 16B is located approximately 600 feet east of well 7
This relatively small reduction would  not have been observed if bedrock fractures were
completely  filled with soil.

EPA also reviewed the data from the groundwater pump-and-treat program conducted
by Rohm & Haas.  During the period 1965-1969 when only wells 1-7 were in service.
the arsenic concentration in the extracted groundwater was reduced from 1160 mg/l to
150 mg/l.  (Overall the reduction in concentration was from 8500 mg/l at day 10 to 150
mg/l at day 1290.)  These data indicate that the arsenic concentration was decreasing
by  one order of  magnitude  every  1350 days  (3.7 years).    Using  this  rate,  the
groundwater  concentration  could   be reduced  from  212 mg/l  to 0.05  mg/l (the
groundwater arsenic cleanup level) in 13 years. The reduction in arsenic concentration
during the  pump-and-treat program  occurred  despite the fact that  infiltration through
contaminated soils/sediments and the groundwater treatment system itself continued to
add arsenic contamination to the groundwater. If the contaminant contribution sources
of the  soils/sediments  (as  well  as  the  vault and lagoon  wastes,   miscellaneous
products/feedstocks, and site structures) are eliminated as planned by EPA, groundwater
cleanup rates would be expected to be faster than they would otherwise be.

EPA does  not expect all contaminated  groundwater will be remediated  below the
cleanup levels in 13  years.   Some "tailing effect" is expected.  However, given me
existing database, EPA believes there is a good possibility that the contaminated aquifer


                                   15

-------
       could be restored to its beneficial uses in a reasonable timeframe, and is using this
       justification for its selection of Alternative 4 as the groundwater remedy. EPA recognizes
       there is uncertainty regarding the duration of the groundwater cleanup, but does not feel
       that writing  off the aquifer and implementing a containment system, as proposed by
       the former owners and Buckeye Pipeline Company (The Application of EPA's Pump and
       Treat Memorandum  and the  NCP to  the Whitmoyer Laboratories Superfund  Site",
       Whitmoyer Laboratories Private Study Group, June 1990) is warranted at this time  EPA
       also believes there is sufficient evidence to support the selection of an aggressive pump-
       and-treat program, as specified in Alternative 4, at this time. Even if this program is not
       completely  successful and  the contingent remedy of  plume containment must  be
       implemented for portions of the aquifer,  much of the aquifer should be restored  to its
       beneficial  uses  during  the  aggressive  program,  and  substantial  quantities  of
       contamination will be removed from the aquifer.

       The size of  the treatment system and associated pumping and piping systems will be
       determined  during the Remedial design phase of project implementation.  Although the
       Feasibility Study  and Proposed Plan specified certain pumping  and treatment rates,
       optimization of the chosen system during design will ensure that the minimum required
       pumping and  treatment rates  will  be utilized for the  Remedial Action.  Accordingly.
       capital and  operation and maintenance costs will also be minimized.

34.    The former owners  commented  that  after  the  first  year  of  the prior  pumping,
       concentrations of arsenic did not decrease in any discernable manner.

       EPA Response: The  data cited in the response to  Comment 33 above directly conflict
       with this conclusion.

35.    The former  owners commented that their proposal of the 3-year pilot program, which
       starts by treating the most contaminated areas first, should be selected by EPA, because
       it is consistent with EPA guidance on initiating an early response action and  using a
       staged  approach at  sites where available  data do not clearly define  the parameters
       necessary to optimize system  design.

       EPA Response: EPA's guidance on early response actions refer to the implementation
       of a response action to  prevent  the contaminant plume from  spreading  while  the
       investigation to select the remediation system progresses. The former owners approach
       does not prevent the contaminant plume from spreading.  The  time for implementing an
       early response action at the Whitmoyer Site was in 1986, when EPA's negotiations with
       the former owners began, and  has now past.  Sufficient information now exists to select
       the remediation system. While the database necessary to finish the remedial design of
       the  groundwater extraction and  treatment  systems  is not  complete, the required
       information will be collected during the remedial design phase.  EPA sees no justification
       for delaying selection of the ultimate remedy for groundwater.

-------
 36.    The former owners commented that there are not sufficient data to support the selection
       of a  groundwater remedy. Data gaps include the valence of arsenic at the site; the
       composition and  relative amounts of inorganic  and organic arsenic compounds in
       groundwater; the leaching  or  adsorption  characteristic characteristics of the  soil.
       bedrock, and fissure fillings; the impact of pumping on groundwater elevations and flow
       directions; and groundwater/surface-water relationships.

       EPA  Response: EPA agrees that more information on several of these items will need
       to be collected during the remedial design phase  of the groundwater remediation.
       However, EPA believes there  is sufficient information available at this point to support
       the selection of a groundwater remedy.

 37.    The former owners commented that the Rl failed to adequately determine the extent of
       groundwater contamination below 150 feet  in depth, and that the  assumption  that
       groundwater is contaminated to a depth of 500 feet  may lead to an overestimation of
       the extent of contamination.

       EPA  Response: EPA installed nine new wells to a depth  below 150 feet during the Rl
       Samples from the deepest installed well (107C), at a depth of 385 feet, were significantly
       contaminated with arsenic and aniline.  Existing well 4, at a depth of +335 feet, has also
       been historically demonstrated to be substantially contaminated. This well was used for
       the injection of contaminated water.   Given  these data,  it is reasonable engineering
       judgement to estimate that the aquifer is contaminated to a depth of 500 feet   Cost
       constraints limited the placement of deep  wells during the  Rl program.   Further
       placement of wells that define the extent of  groundwater contamination at  depth  will
       likely occur during the remedial design phase of the groundwater remediation.

 38.    The former owners commented that, under Alternative 3, the  volume of contaminated
       groundwater will decrease over time.

       EPA  Response: EPA disagrees.  If no action  is taken on the portion of the  plume
       containing between 50 ug/l and 1,000 ug/l arsenic, this plume will grow over time, at a
       rate significantly greater than the one foot per year averred by the former owners  (See
       the response to Comment 33.)

39.    The former owners  commented that  there is  virtually no support  for the assumed
       southern (or second) lobe of groundwater contamination.

       EPA Response: EPA  disagrees. There  are several contaminated wells that demonstrate
       the southern lobe  of  groundwater contamination, including the Wagner residential wen
       (RW0003),  the PJ  Valves Co.  wells (RW0007), the G. Hamish well, and  the  High well
       These last  two wells  were not sampled during the Rl, since valid data from these wells
       were  available from prior EPA and DER sampling efforts.

40.    The former owners commented that lowering  the water table beyond the area affected
       by the previous pumping program, as envisioned under EPA's  preferred groundwater
       alternative, might trigger the formation  of sinkholes.


                                         17

-------
       EPA Response: No sinkhole formation was noted during the 7-year groundwater pump-
       and-treat program conducted by Rohm & Haas.  Bedrock cavities were only infrequently
       encountered  during  the  Rl drilling  program.  As the  former owners  pointed  out,
       catastrophic collapse of the rock mass because of failure of rock voids is uncommon
       in the Whitmoyer Laboratories Site region.  Rather most of the sinkholes that do form
       are relatively small  ("Whitmoyer Laboratories Site - Background Information - Proposed
       Phase  I Remediation," Whitmoyer Laboratories Private Study  Group,  March 1990)  and
       (Telephone conversation between Bill Kochanov and John Trepanowski, 9/21/90). Given
       these considerations, it is unlikely that catastrophic sinkholes would form from operation
       of the aggressive groundwater extraction system.  Because of the rural agricultural site
       setting, even if sinkholes would form from groundwater extraction, the probability of
       significant economic damage resulting from  their formation is low.

41.    The former owners  implied that their proposed program would avoid overpumping which
       would dewater portions of the aquifer that are highly contaminated, but which would also
       leave arsenic bound to fracture fillings that could recontaminate groundwater; whereas
       EPA's preferred alternative would realize these results.

       EPA Response: The preferred alternative groundwater withdrawal system would  be
       flexibly  designed.   To reach  the contamination reduction  objective at least  two
       approaches could  be used to protect the water  table in the area.either alone or in
       combination with each other. Continual monitoring of water levels both on and off the
       Site  property would  allow feedback of aquifer conditions during remediation.   The
       system would be designed with pumps that could  be turned on or off depending upon
       groundwater conditions to  protect the adjacent wells and surrounding groundwater
       Additionally, a groundwater re-injection program could be operated in the Site  area to
       maintain the level of the surrounding water table.

42.    The former owners  question the efficacy of the groundwater treatment system presented
       in  the  FS  report.   The  present  owner of a  property that contains some of  the
       contaminated  soils/sediments (Buckeye Pipe Line Company)  commented that  the
       groundwater treatment system presented in the FS report is unnecessarily costly

       EPA Response: The exact groundwater treatment method is not specified in the selected
       remedy.  The  specific treatment method to  be  utilized will be engineered  during the
       remedial design phase. The preliminary  design presented in the FS was only provided
       as an example of a  potentially viable treatment system and for cost estimation purposes
       If the system  presented  by the former owners  or the modifications suggested  by
       Buckeye are determined to represent  the optimal treatment  system,  they  will  be
       incorporated into the final design.

43.    The former owners argue that groundwater cleanup standards based on 1CT6 excess
       lifetime cancer risks  are too stringent,  since the groundwater is not now used  for
       drinking water, and there is no reasonable expectation that groundwater in the area will
       be  a future drinking water source.
                                          18

-------
       EPA Response: The aquifer beneath the site is a current source of drinking water for
       many nearby residents who are not connected to the Myerstown public water supply.
       Additionally, the Myerstown public water supply system has one well (#8) located in the
       aquifer.  EPA classifies the aquifer as a Class 2 aquifer (current and potential sources
       of drinking water and waters having other beneficial uses).

44.    The former owners  claimed that EPA mischaracterized their groundwater proposal
       (Alternative 5) in the Proposed Plan for Operable Unit 3  by stating that their program
       would be operated to contain (and not actively remediate)  onsite groundwater.

       EPA Response: Epa's characterization comes directly from the former owners' June 1990
       submittal ("The Application of EPA's Pump and Treat Memorandum and the NCP to the
       Whitmoyer Laboratories Superfund Site", Whrtmoyer Laboratories Private Study Group,
       June 1990):

             "the ground water on-site  would  be contained by the WLPSG's
             proposed pump and treat  system."

45.    The former owners commented that different investigative techniques should have been
       used during the Remedial Investigation (Rl), including more comprehensive and frequent
       groundwater elevation surveys; pumping/injection tests at  higher rates and longer in
       duration; coring, downhole geophysics, and packer testing; and surface geophysics.

       EPA Response: There are several alternate investigatory approaches that can be utilized
       during an Rl. EPA focussed its Rl by having an extensive screening program, including
       an extended review of historical documents, interviews with former employees, a review
       of historical air photos, a fracture trace analysis, and an evaluation of existing wells for
       Rl usability. Based on the extensive database developed from the review of historical
       documents and  interviews with former employees and given the practical Rl cost and
       time constraints, completion of the work suggested  by the former owners during the Rl
       was evaluated and eliminated  as a  cost-reduction  step.  EPA believes that adequate
       investigation of site conditions for decision making  purposes with respect to selecting
       a remedy and commencing Remedial Action have been completed.  Much of this work
       (e.g., additional groundwater elevation surveys and  more extensive aquifer testing) will
       be conducted during the remedial design phase of the groundwater remediation.

46.    The former  owners commented that the  descriptions   of  the  aquifer  hydraulic
       characteristics in the Rl and PS reports are inconsistent.

       EPA Response: The differences in aquifer hydraulic characteristics presented in the  Rl
       and FS  reports  result from the use of two different approaches in estimating these
       characteristics.  These approaches differ because the process used in the evaluation
       of site conditions is  an evolving one which is continually refined and updated as data
       are continually analyzed.   EPA believes the approach used in the  FS report is more
       appropriate for the preliminary design of the groundwater extraction systems presented
       as a part of each  groundwater alternative  and followed logically from continuing data
                                          19

-------
       analysis in the Rl.  The Rl approach was utilized to calculate an alternative extraction
       rate for the FS sensitivity analysis.

47.    The former owners commented that the Rl data do not support the EPA assertion that
       the aquifer underlying the site is a single large heterogeneous  unconfined aquifer.

       EPA  Response:  The  Rl data  indicate  that the  fractured  limestone  aquifer  is
       heterogeneous and anisotropic. There is no evidence of a low permeability confining
       layer that could be considered a boundary between two or more separate aquifers.
       Both the presence of  contamination at all levels within the formation underlying the site
       and the similar hydraulic heads for wells open over different depth intervals at a given
       location support the  description of the fractured limestone as  a  single aquifer.   The
       response to pumping discrete, highly transmissive zones within an  otherwise  less
       transmissive aquifer would be expected to be similar to what was observed at the site.
       and does not indicate the presence of hydraulic barriers between fractures encountered
       at different depths.  Horizontal hydraulic conductivities of 1-2 orders of magnitude higher
       than  vertical hydraulic  conductivities are  not  uncommon. This difference does not
       indicate the existence of separate aquifers, however. The FS description of the aquifer
       as a single, heterogeneous, highly anisotropic aquifer best describes the conditions  at
       the site.

48.    The former owners commented that, with the deep well drilling technique used, drilling
       fluids and cross-contamination  from upper portions ~of the aquifer may have migrated
       downward via the wells, affecting water quality analyses.

       EPA Response:  EPA disagrees.   The well completion technique used is a standard
       operating procedure for site investigations. Grout seals were allowed to cure overnight
       before drilling proceeded. When drilling resumed, removed cuttings were dry, indicating
       waters from upper portion of the aquifer were not migrating down the borehole.

49     The  present   owner  of  a property that contains  some  of  the  contaminated
       soils/sediments (Buckeye Pipe Line Company) commented that  EPA has not presented
       adequate justification for the proposed remediation of  the  Buckeye  property.   The
       Buckeye  property is  not the primary source of the groundwater contamination at the
       site.  The proposed remediation will impact Buckeye operations.

       EPA  Response: Remediation of the contaminated soils  (and  lagoon wastes) on the
       Buckeye  property is  necessary to achieve the remedial action objectives for the site
       which include  the  elimination of a serious threat to groundwater from contaminated
       materials on the site.  EPA believes their selected remedy, which will likely impact
       pipeline operations, is the best overall approach for addressing site contamination

Risk Assessment

50.    The former owners  commented that the exposure  assumptions presented  in the
       baseline  risk assessment (found in the Rl report) did not use reasonable maximum
       exposure (RME) values.  Instead average and worst-case assumption values were  used


                                          20

-------
despite EPA guidance documents available at the time which specifically instructed that
the use of average and worst-case assumptions are not appropriate.  In using a worst-
case scenario, EPA overestimated the risks posed by the Whitmoyer site.

EPA Response: The Whitmoyer Rl was dratted and finalized in the summer and fall of
1989,  respectively.   The  guidance document referred to in the comments, "Risk
Assessment Guidance for Superfund, Volume I. Human Health Evaluation Manual' (EPA,
December 1989) is dated December 1989. This document was generally not available
to the public until February/March 1990.   Consequently, the Whitmoyer Rl and
groundwater FS reports were written before the December  1989 guidance document
was available.  The RI/FS was conducted per the guidance  available at that time.

In reality, the  evaluation of  the RME versus worst-case or average exposures has
marginal  impact on the risk analysis  results.   The evaluation of the  average and
maximum exposures provide a range of possible risk analysis estimates.  The RME is
an estimate that is within the range of possible exposures.  The guidance document
cited by the former  owners states as follows:

       In the past, exposures generally were estimated for  an average
       and an upper-bound exposure case, instead of a single exposure
       case (for both current and future land use) as recommended here.
       The advantage of the two case approach is that the resulting
       ranoe of  exposures provides some measure of the  uncertainty
       surrounding these estimates.  The disadvantage of this approach
       is that the upper-bound estimate of exposure may be above the
       range of  possible exposures, whereas the average  estimate is
       lower than  exposures potentially experienced  by much of the
       population.  The intent of the RME is to estimate a conservative
       exposure case (i.e.,  well above the average case) that is still within
       the range of possible exposures.  Uncertainty is still evaluated
       under this  approach.   However, instead of combining  many
       sources of uncertainty into average and upper-bound  exposure
       estimates, the variation in individual exposure variables is used to
       evaluate uncertainty.  In this way, the  variables contributing most
       to the  uncertainty  in  the  exposure  estimate  are more  easily
       identified.

It serves  no real purpose to  recalculate risk using the RME (i.e., a 95 percent upper
confidence limit  on  the arithmetic  mean) because an  evaluation  of exposures to the
average contaminant concentrations in the groundwater and  soils  at Whitmoyer (which
predict lower risks than the RME methodology) yields unacceptable risks (i.e., a health
index greater than 1 and excess lifetime cancer risks in excess of the 1 x KT4 -1 x icr
6 acceptable risk range).

In summary, the Whitmoyer Rl risk assessment was  conducted per the  guidance
available at the time. The Rl analysis conservatively provides  a range of risks potentially
incurred by receptors who may contact site contaminants. (Current and future exposure


                                  21

-------
       scenarios were  considered.)   The  risks  for average  contaminant concentration
       exposures, which are less than RME risks, are unacceptably high. Also, the failure to
       calculate RME risks in no way impacted the development of the cleanup standards for
       the Whitmoyer Laboratories Site (see the response to comment 53 below).

 51.    The former owners commented that many of the elements common to risk assessments
       (hazard identification,  dose-response  assessment,  exposure assessment,  and risk
       characterization)  appear to be missing  in the documents on which the Proposed Plan
       is based.

       EPA Response: EPA disagrees. The risk assessment for the Whitmoyer Laboratories Site
       clearly contains all of the cited elements.  The baseline risk assessment presented in
       the Rl report was outlined and composed using these elements.

 52.    The former  owners commented  that  there  is no  discussion  of  the uncertainties
       associated with  the  exposure assumptions  or  values  presented  in  the  exposure
       assessment and  risk assessment.

       EPA Response:  EPA disagrees. Uncertainties in the exposure assessment and risk
       assessment are discussed in several sections of the Whitmoyer Rl report (e.g., pages
       6-46, 6-49, 6-52,  6-53. 6-54, 6-94, and 6-95).

 53.    The former owners commented that, because risks for all media via all pathways have
       been estimated without incorporating reasonable activity patterns or data, the results of
       the  combined assumptions exaggerate  the potential risk.

       EPA Response: The exposure scenarios established for the Whitmoyer Site were based
       upon EPA guidance available at the time and upon the current and potential future use
       of the  Whitmoyer site.  Although site-specific population-activity-pattern  data were not
       collected, the assumptions made were reasonable for the stable rural community in the
       vicinity of the site.  Therefore,  EPA does not believe the risks are exaggerated by the
       assumptions made.  (Also see the  responses to comments 55 and 56).

54     The former owners commented that ingested arsenic should be treated  as a non-
       carcinogen and not as a carcinogen.  They stated that EPA treated ingested arsenic as
       a  non-carcinogen in their proposed RCRA corrective action  rule,  citing a  personal
       communication with Ms. S. Griffin of EPA.

       EPA Response: EPA believes treating ingested arsenic as a carcinogen is justified.  All
       current EPA guidance, including the EPA Integrated Risk  Information System (IRIS) and
       the  Health Effects Assessment tables, classify ingested arsenic as a carcinogen.  Ms
       Griffin  of EPA denies making the statement attributed to her by the former owners.

55.     The former  owners  commented  that  the  Whitmoyer risk  assessment  does not
       incorporate the arsenic  inhalation absorption factor of 0.3, as  presented in the most
       recent IRIS listing.
                                         22

-------
       EPA Response: EPA agrees with the former owners. The most recent IRIS information
       indicates that a 30 percent absorption factor is appropriate. However, the adjustment
       of the risk  analysis results  to reflect this new absorption factor would not alter the
       conclusion that contaminant concentrations (most notably arsenic) in the environmental
       media (e.g., groundwater, soils) at the Whitmoyer Laboratories Site present a potential
       human health threat under one or more of the exposure scenarios evaluated because
       this revised  calculations would still  indicate  unacceptable human  health  risks.
       Additionally, the soil/sediment and groundwater cleanup levels  are based on  the
       ingestion route of exposure (and drinking water standards),not inhalation.  Thus, the
       recent change in the absorption factor does  not affect the cleanup levels set for the
       soils/sediments and groundwater at the Whitmoyer Laboratories Site.

56.    The former  owners commented that some of the EPA exposure duration estimates are
       overestimated and do not represent reasonable durations. For example, for substances
       EPA considers carcinogenic, the exposure duration should be only 30 years as the
       potential residence time near the site, not the 70 years used in the soils FS report

       EPA Response: The 30-year exposure duration is the national upper-bound time (90th
       percentile)  at  one residence  suggested by the  new  December 1989 EPA  risk
       assessment  guidance.   By all  appearances, the community in  the vicinity of  the
       Whitmoyer site is a stable, rural community. (One example of this is that few residences
       with contaminated potable water supply wells  have changed ownership during the last
       26 years).  Consequently, a 70-year exposure timeframe is certainly possible for  this
       area, and was adopted  based on  professional judgement.   Given  the  contaminant
       concentrations detected in the environmental  media at Whitmoyer, the 30-year versus
       70-year timeframe  argument does  not  alter the conclusion that the  contaminant
       concentrations at Whitmoyer present a potential human health hazard because the risk
       assessment would predict unacceptable human health risks using either timeframe

57.    The  former owners commented that the failure to collect arsenic speciation data during
       the Rl made the risk assessment of limited value. The risk assessment assumption that
       all arsenic detected is inorganic arsenic may lead to a substantial overestimation of risk


       EPA Response: During RI/FS scoping, the former owners commented that all analyses
       for valences of arsenic (speciation) should be eliminated.  (Technical Assessment of
       the Need  for Immediate Removal of Vault/Lagoon Contents at Whitmoyer  Laboratories
       Site*, January 1987).  This comment was  incorporated  into the Rl.

       In general, organic arsenic compounds are less toxic than  inorganic forms of arsenic
       There are exceptions to this  rule, however. Presently, there are inadequate toxicology
       data to determine the carcinogenic potential for organic arsenic compounds.

       Assuming that  a substantial  percentage of the arsenic contamination detected during
       the Rl  is organic arsenic, the risk analysis based on the assumption that all arsenic
       present at the site is inorganic  may have overestimated the noncarcinogenic ana
       carcinogenic risks  calculated for the site.  For example, assuming that the percentage


                                         23

-------
       of organic arsenic is 50 percent and assuming that organic arsenic is  one-tenth as
       potent a carcinogen as inorganic arsenic, the predicted excess  lifetime cancer risk
       would  be approximately one-half the risk assuming  100 percent inorganic arsenic.
       However, given the very high arsenic  concentrations measured at Whttmoyer. the
       assumption that organic arsenic is prominent at the site does little to alter the overall
       conclusion that the Operable Unit 3 materials present substantial noncarcinogenic and
       carcinogenic risks, and require remediation.  The assumption of 100 percent inorganic
       arsenic adds a measure of conservatism to the risk assessment.

58.    The former owners commented that the Rl surface soil sampling strategy biased the risk
       estimates, because sampling concentrated on target locations suspected of containing
       high contaminant  concentrations, rather than using a grid or random sampling plan

       EPA Response: The sampling strategy used at the Whrtmoyer Laboratories site is typical
       of that used to investigate uncontrolled hazardous waste sites such as Superfund  sites
       In addition to samples collected to confirm or deny the potential presence of various
       source areas throughout the site, samples  were also collected away from potential
       source areas to address data gaps. It  should be noted  that in the "occurrence and
       distribution of contaminants in soils" and risk assessment presentations in the Rl report.
       EPA did not assume that chemical constituents are present continuously throughout the
       site at  "hot spot" concentrations, as maintained by the former owners.  The entire site
       was divided into five distinct areas based upon contaminant concentrations detected
       and source areas. Consequently, EPA did recognize that chemical concentrations do
       vary across the site and did not bias its  risk assessment as asserted by the former
       owners.

59.    The former owners commented that EPA's sampling strategy may have resulted in the
       inaccurate classification of large areas of the site as contaminated, when they do not
       pose a threat to human health and the environment.

       EPA Response: EPA believes that the site has been reasonably  classified given the
       existing data base. The areas of the site to be remediated will be more accurately
       delineated during  the remedial design and remedial action phases of the project   The
       action  levels presented in the ROD will be used to differentiate soils/sediments that
       require remediation from soils/sediments that do not.

60.    The former owners implied in their comments that the NCR requires EPA to minimize
       potential risks  at Superfund sites.

       EPA Response: The NCP does not require EPA to minimize potential risks. The NCP
       requires  EPA  to  select a  remedy  which is  protective of human  health  and  the
       environment [i.e.,  reduces  the risks (current  or future, actual or potential)  to a~
       acceptable level, for example the  1 x 10~6 excess cancer risk level].  EPA's selectee
       remedy meets  all  the requirements established  in the  NCP.

61.    The former owners commented that the use of a 1 x 10r6 excess cancer risk level  at :^e
       Whrtmoyer Laboratories Site is  not scientifically justified,   Furthermore, its use s


                                         24

-------
       inconsistent with EPA having adopted 10~* risk goals at other Superfund sites and in
       other regulatory programs. Also, the Whitmoyer arsenic cleanup levels are inconsistent
       with the arsenic cleanup levels proposed in the proposed  RCRA corrective action
       regulations.

       EPA Response: In the Superfund program, EPA establishes acceptable risk levels and
       cleanup goals on a site-specific basis.  Consequently it is anticipated that cleanup goals
       will vary based on site-specific conditions, the nature of the environmental problem
       under consideration, and the lexicological information available to the  risk  assessor
       The EPA has taken into account site-specific conditions at the Whitmoyer Laboratories
       Site (e.g., the residential and agricultural use of adjoining properties), and has decided
       that the 1  x  1CT6 excess cancer  risk level  goal is appropriate for the Whitmoyer
       Laboratories Site.  (See also Attachment 1.)

       It should be noted that the proposed RCRA corrective action regulations are  proposed
       only and not promulgated, and are subject to substantial change.

62.    The former owners commented that the groundwater-based soil action (cleanup) levels
       and groundwater action levels for the  Whitmoyer Laboratories  Site should be changed
       to reflect the recently published proposed drinking water standards for benzo(a)pyrene,
       indeno(1,2,3-cd)pyrene,  and benzo(b)fluoranthene.

       EPA Response:  EPA agrees.   The cleanup levels have been adjusted in the  ROD to
       reflect these new values.

D.     REMAINING CONCERNS

An issue that EPA was unable to address during remedial planning activities was how much
of the cleanup will be paid by potentially responsible  parties.  EPA was unable to address this
since negotiations are still unresolved.  EPA will inform the community if the negotiations  are
successfully concluded.
                                          25

-------