Final Report:
Independent Design Review
Celanese Fiber Operations
Superfund Site
Shelby, North Carolina
EPA Region 4
-------
Solid Waste and EPA-542-R-11 -001
Emergency Response February 2011
(5203P) www.epa.gov
Final Report:
Independent Design Review
Celanese Fiber Operations
Superfund Site
Shelby, North Carolina
EPA Region 4
-------
TABLE OF CONTENTS
ABBREVIATIONS II
PREFACE Ill
1.0 INTRODUCTION 1
1.1 PURPOSE 1
1.2 TEAM COMPOSITION 2
1.3 DOCUMENTS REVIEWED 2
1.4 PERSONS CONTACTED 3
1.5 BASIC SITE INFORMATION AND SCOPE OF IDR 3
2.0 FINDINGS 5
2.1 FINDINGS RELATED TO SITE CONCEPTUAL MODEL 5
2.2 FINDINGS RELATED TO EXISTING REMEDY 10
3.0 RECOMMENDATIONS 16
3.1 SUGGESTED ADDITIONAL CHARACTERIZATION 16
3.2 SUGGESTED REMEDIAL STRATEGY (FROM A TECHNICAL PERSPECTIVE) 19
3.3 POTENTIAL SCHEDULE 21
ATTACHMENT A: FIGURES A-l
ATTACHMENT B: FIGURES AND TABLES FROM THE CONCEPTUAL SITE MODEL
REPORT AND SEMI-ANNUAL REPORTS B-l
-------
ABBREVIATIONS
BSD Explanation of Significant Differences
GRUB Glycol Recovery Unit Distillation Bottoms
IDR Independent Design Reviews
IT inner tier
MNA monitored natural attenuation
NCDENR North Carolina Department of Environment and Natural Resources
OT outer tier
OU operable units
P&T pump and treat
ROD Record of Decision
RP responsible party
RSE remediation system evaluations
TCE trichloroethene
11
-------
PREFACE
This report was prepared part of a pilot project conducted by EPA's Office of Superfund
Remediation and Technology Innovation (OSRTI). The objective of this pilot project is to
conduct independent, expert reviews of soil and ground water remedies that are jointly funded by
EPA and the associated State agency. The reviews are ideally conducted during the pre-design,
design, or redesign stage so that independent perspectives on the remedy are provided before
costs are incurred for implementing and operating the remedies. The project contacts are as
follows:
Organization
Key Contact
Contact Information
EPA OSRTI
Kathy Yager
U.S. EPA Region 1 - New England
Regional Laboratory
11 Technology Drive
Mail Code: ECA
North Chelmsford, MA 01863-2431
phone:617-918-8362
yager.kathleen@epa.gov
Tetra Tech EM, Inc.
(Contractor to EPA)
Carla Buriks
Tetra Tech EM Inc.
1881 Campus Commons Drive
Suite 200
Reston, VA 20191
phone: 703-390-0616
carla.buriks@ttemi.com
GeoTrans, Inc.
(Contractor to Tetra Tech EM,
Inc.)
Doug Sutton
GeoTrans, Inc.
2 Paragon Way
Freehold, NJ 07728
phone: 732-409-0344
dsutton@geotransinc.com
in
-------
1.0 INTRODUCTION
1.1 PURPOSE
During fiscal years 2000 and 2001 independent reviews called remediation system evaluations
(RSEs) were conducted at 20 operating Fund-lead pump and treat (P&T) sites (i.e., those sites
with P&T systems funded and managed by Superfund and the states). Due to the opportunities
for system optimization that arose from those RSEs, OSRTI has incorporated RSEs into a larger
post-construction complete strategy for Fund-lead remedies as documented in OSWER Directive
No. 9283.1-25, Action Plan for Ground Water Remedy Optimization. OSRTI has since
commissioned RSEs at additional Fund-lead sites with P&T systems.
Lessons learned from the RSEs conducted to date indicated potential value in conducting
independent reviews during the pre-design, design, or redesign stage of a remedy so that the
independent perspective can be considered before a remedy is implemented and operated. In
addition, there is potential for the same type of process to benefit responsible party (RP) sites. As
a result, OSRTI's Technology Innovation and Field Services Division is conducting pilot
Independent Design Reviews (IDRs) at Fund-lead and RP sites that are nominated by the EPA
Regions. These sites are typically at one of the following stages:
• Pre-design - The Region is in the process of determining the remedy for the site. The
Remedial Investigation is typically completed and the site team is evaluating the
feasibility of several remedial options.
• Design Stage - The Region has selected a remedy and documented it in a Record of
Decision (ROD). The Region is at any phase of the Remedial Design stage.
• Redesign Stage - The Region has selected, designed, and may have implemented a
remedy but the remedy is being reconsidered. Based on data collected to date, the Region
is considering new remedial options or is designing a new remedy for the site.
An DDR involves a team of expert hydrogeologists and engineers that are independent of the site,
conducting a third-party evaluation of remedy selection or remedy design. It is a broad
evaluation that considers the goals of the remedy, site conceptual model, available site data,
performance considerations, protectiveness, cost-effectiveness, and closure strategy. The
evaluation includes reviewing site documents, potentially visiting the site for one day, and
compiling a report that includes recommendations. Recommendations with cost and cost savings
estimates are provided in the following categories:
• Site strategy
• Initial recommendations
• Next-step recommendations
-------
The recommendations are intended to help the site team identify opportunities for improvements.
In many cases, further analysis of a recommendation, beyond that provided in this report, may be
needed prior to implementation of the recommendation. Note that the recommendations are
based on an independent evaluation by the IDR team, and represent the opinions of the IDR
team. These recommendations do not constitute requirements for future action, but rather are
provided for consideration by the Region and other site stakeholders.
The Celanese Fiber Operations Superfund site (the "Celanese site", or "site") was selected by
EPA OSRTI based on a nomination from EPA Region 4. The Celanese site is located in Shelby,
North Carolina. Operation of the P&T system was discontinued on a trial basis for two years to
evaluate monitored natural attenuation as a potential ground water remedy. The system has not
been restarted. EPA Region 4 is requesting a third-party review to help determine if the system
should be restarted, if a modified system should be restarted, or if another type of ground water
remedial approach is warranted. This report provides a brief background on the site, a summary
of observations made from the document review, and recommendations regarding future
modifications to the existing remediation system (including associated investigation activities).
The cost impacts of the recommendations are also discussed.
1.2 TEAM COMPOSITION
The IDR team consists of the following individuals:
Name
Doug Sutton
Kathy Yager*
(not present)
Affiliation
GeoTrans, Inc.
U.S. EPA OSRTI
Phone
732-409-0344
617-918-8362
Email
dsutton@geotransinc.com
yager.kathleen@epa.gov
* Project coordinator(s)
1.3
DOCUMENTS REVIEWED
The following documents were reviewed during the IDR process. The reader is directed to these
documents for additional site information that is not provided in this IDR report.
• 3rd Five-Year Review - August 2001
• 4th Five-Year Review - August 2006
• Explanation of Significant Differences - April 2004
• OU1 Semi-Annual Report January 2007 to June 2007 - September 2007
• OU1 Semi-Annual Report July 2007 to December 2007 - March 2008
• Institutional Controls/Watering Agreements
• Conceptual Site Model Report - September 2007
• Risk Assessment - August 2007
-------
Access Database through April 17, 2008
Region 4 Comments on OU1 Semi-Annual Report - June 4, 2007
Draft Region 4 Response to Comments Dated August 14, 2007 - September 2007
Region 4 Comments to Status Update and Conceptual Path Forward - August 9, 2007
North Carolina Department of Environment and Natural Resources (NCDENR)
Comments to Draft Risk Assessment - September 19, 2007
NCDENR Comments to Conceptual Site Model Report - December 5, 2007
NCDENR Comments to OU1 Semi-Annual Report July 2007 to December 2007 - April
14, 2008
1.4 PERSONS CONTACTED
The following individuals associated with the site were present for the visit:
Name
Beth Walden
David Jenkins
David Mattison
Steve Simpson
Everett Glover
Bryon Dahlgren
Joe Harrigan
Affiliation
EPA Region 4 (Remedial Project Manager)
EPA Region 4 (Hydrogeologist)
NCDENR
Celanese
AECOM (Celanese Contractor)
AECOM
AECOM
1.5 BASIC SITE INFORMATION AND SCOPE OF IDR
The Celanese site began operation in 1960 and currently operates as a polyester and engineering
plastics production facility. CNA Holdings, Inc., a subsidiary of Celanese, retains management
of environmental matters for the corporation and is responsible for environmental work
regarding the site. The Celanese site is approximately 450 acres located six miles south of
Shelby, North Carolina in south-central Cleveland County. The site is bordered to the west by
Highway 198 and Lavender Road to the south. The site consists of a main plant production area,
wastewater treatment area, former waste disposal areas, and recreational/wooded areas. The plant
production area is covered with buildings and paved or graveled areas. Toward the wastewater
treatment area, the site consists of grassy areas and roads. The recreation area is wooded and has
no structures present. The site is currently fenced and has controlled access points. Current land
-------
use surrounding the site includes mainly rural residential areas and agricultural. A site location
map is presented in Figure 3-1 of the Conceptual Site Model Report (see Attachment A).
The site has been divided into two operable units: OU1 addresses contaminated ground water,
and OU2 addresses contaminated soil and source materials. This independent review focuses on
ground water contamination associated with OU1 and only considers impacted soils as they
would potentially affect OU1 as a continuing source of contamination for ground water. The
OU1 remedy consists of two P&T systems. The inner tier (IT) system addresses contamination in
the immediate vicinity of the former source areas and was shutdown on a trial basis in
accordance with a 2004 Explanation of Significant Differences (BSD) to evaluate monitored
natural attenuation (MNA) in the absence of remedy pumping. The outer tier (OT) system
addresses the plume near the downgradient property boundary and was shutdown in accordance
with the National Priority List delisting process in 1998.
This independent review aims to determine if the IT P&T system should be restarted, if a
modified system should be restarted, or if another type of ground water remedial approach is
warranted.
-------
2.0 FINDINGS
The IDR team reports the following findings from the document review that are pertinent for the
recommendations that have been developed as part of the IDR process and described in Section
3.0 of this report.
2.1 FINDINGS RELATED TO SITE CONCEPTUAL MODEL
Hydrogeology
Site geology consists primarily of low permeable saprolite overlying bedrock. The saprolite is
generally thickest beneath the plant and thins toward the east and in the vicinity of the adjacent
streams. In some locations the saprolite has pinched out and the creeks flow across the bedrock.
Ground water occurs in the saprolite under water table conditions and in fractures in the bedrock.
The direction of groundwater movement in the shallow saprolite zone is to the east, northeast,
and southeast from upgradient areas along NC 198 toward discharge areas along unnamed
tributaries of Buffalo Creek and is shown on Figure 3 of the Semi-Annual Report January to June
2007 (see Attachment A). Direction of groundwater movement in the deeper saprolite and upper
bedrock zone is in the same general direction. Figure 6 of the Semi-Annual Report January to
June 2007 and Figures 4-2, 4-3, and 4-4 of the Conceptual Site Model Report (see Attachment
A) present cross-sections of the site prepared by the site contractor. The figures from the two
reports generally agree, but there are a few discrepancies, including the zone in which
monitoring well O-59 is screened and the presence of a transition zone between the saprolite and
bedrock. The Site Conceptual Model Report indicates that this transition zone likely has a higher
hydraulic conductivity than the saprolite or bedrock.
The bedrock fracturing reportedly decreases in intensity with depth below the top of rock. In the
saprolite, groundwater generally parallels the ground surface slope. This is also the case in
shallow bedrock. In the deeper bedrock, below the elevation of local creek bottoms, groundwater
flow becomes more regional, although discharge to the major creeks appears still dominant. The
potentiometric surface in the saprolitic aquifer slopes to the northeast, east, and southeast toward
Buffalo Creek. Based on a generalized hydraulic gradient of 0.036 feet/foot, an average
hydraulic conductivity of 5.13 x 10"4 centimeters per second (cm/sec) from slug tests, and an
effective porosity of 0.3, the groundwater seepage velocity was calculated by the site team at
0.17 feet/day (63 feet/year) in the saprolite. The hydraulic conductivity at some locations is as
high as 10 feet per day based on slug tests and pumping tests.
Extent of Contamination
The site investigation began in 1981 and determined that the burn pits and Glycol Recovery Unit
distillation bottoms (GRUB) disposal areas located near the center of the property were the
probable source of groundwater contamination. These areas are shown in Figure 3-3 of the
Conceptual Site Model Report. The groundwater contaminants include ethylene glycol,
-------
trichloroethene (TCE), diethylene oxide, DOWTHERM™A constituents (1,1-biphenyl and
biphenyl ether), and others (several volatile organic compounds, metals, and phenols in much
less extent).
The remedial action objective for the former GRUB disposal pits was reportedly to remove the
GRUB plus 1 foot of soil on all sides, and was not intended to be a clean closure in the disposal
area. This resulted in leaving contaminated soil in place above the groundwater table.
Approximately 4,549 tons of GRUB sludge, stained soils, and associated contaminated soils
were excavated and incinerated from the source areas in 1991. Additionally, approximately
3,259 tons of burn pit materials and plastic chips and 39 cubic yards of stream sediment were
excavated. The incinerator ash, burn pit material, plastic chips, and stream sediments were
solidified with 10% to 20% by weight of cement and backfilled into the excavated former GRUB
and burn pit areas. OU-2 was delisted in April 1998.
After OU1 was operational, localized areas of impact were identified at well F-55 (ethylene
glycol and DOWTHERM™A) and at wells TD-3/TD-4 (TCE). The Conceptual Site Model
Report states that these are interpreted as localized issues not related to the source area identified
for remediation under OU1 and OU2. These locations are located in the downgradient portion of
the process area, upgradient of the waste disposal areas addressed by OU1 and OU2.
Ground water contaminants with ongoing detections at the site include but are not limited to the
following:
• ethylene glycol
• diethylene oxide (i.e., 1,4-dioxane)
• 1,1-biphenyl
• biphenyl ether
• TCE
In addition, iron, manganese, and arsenic are found in elevated concentrations, presumably due
to the reducing ground water conditions at the site caused by the presence of ethylene glycol.
In July 2007, ethylene glycol was detected above the standard of 14,000 ug/L at the following
three wells:
• F-55 at 6,680,000 ug/L
• IT-6 at 1,011,000 ug/L (average between two samples)
• V-23 at 8,280,000 ug/L
IT-6 and V-23 are located in the vicinity of the former burn pits and GRUB disposal area. F-55 is
located in the saprolite upgradient of these locations in the downgradient portion of the process
area. Ethylene glycol is not detected in downgradient wells HH-48 and HH-77. Ethylene glycol
has a very low organic carbon partitioning coefficient; therefore, sorption to soil and associated
transport retardation is expected to be negligible. Ethylene glycol degrades rapidly in aerobic
environments and can also degrade in anaerobic environments.
-------
Data from 2006 and 2007 indicate that diethylene oxide is present in many of the site monitoring
wells above the standard of 7 ug/L with the highest concentrations occurring in monitoring wells
in the saprolite in the vicinity of the former burn pits and GRUB disposal area at a maximum
concentration of approximately 3,120 ug/L. From this location, it appears that the plume may
split into two or three arms with contamination migrating to the northeast near monitoring wells
AA-54, KK-55, P-58, PZ-12, to the east near monitoring wells 11-65, and to the southeast near
monitoring wells OT-2R, OT-3, NN-105, and T-58. In all three areas, it appears that
concentrations are the highest in deep saprolite or shallow bedrock. The pathways to the
northeast and east may be one larger plume, but it is difficult to tell given the absence of
monitoring points between these two pathways in the deep saprolite or shallow bedrock. A
separate diethylene oxide plume appears to originate from the F-55 and G-50 area that migrates
south toward 1-57. Diethylene oxide is not detected in downgradient wells HH-48 and HH-77.
Figure 1 illustrates the apparent diethylene oxide migration pathways. Diethylene oxide has a
very low organic carbon partitioning coefficient; therefore, sorption to soil and associated
transport retardation is expected to be negligible. In addition, diethylene oxide does not readily
degrade; therefore, diethylene oxide is expected to behave essentially as a conservative tracer
with no sorption or degradation.
Diethylene oxide appears to be present in the bedrock and deep bedrock based on sampling
results for the LL, MM, and NN clusters between 2004 and 2006. However, these detections of
diethylene oxide in these clusters may be the result of pulling down contamination while
installing the wells. Each of these three well clusters was installed in February 2004 through
shallower zones with elevated diethylene oxide concentrations, and the first round of sampling
was conducted in June 2004. Another round of sampling was conducted in August 2004 and a
third round of sampling was conducted in September 2006. For all three clusters, all wells (with
the exception of NN-105) have shown consistent and substantial declines that are much faster
than would be expected from an established plume at this distance from the source area at this
site. Based on the reported slow recovery of these wells during sampling, it is apparent that these
wells have limited flow of water through them and it is reasonable that contamination brought
down during the drilling process is slowly being flushed from the vicinity of the well screen.
Sampling for 1,1-biphenyl and biphenyl ether appears to be limited to F-55 and wells and the
PEW wells, which are in the general vicinity of F-55. The concentrations for 1,1-biphenyl are
highest in F-55 and are approximately one order of magnitude above the standard.
TCE is primarily detected in two different locations: a presumed but unidentified source area
near monitoring wells TD-3 and TD-4 in the process area and approximately 1,000 feet
downgradient of the site at HH-48 (saprolite) and HH-77 (shallow bedrock). The concentrations
at TD-3 and TD-4 in July 2007 were 416 ug/L and 2,280 ug/L, respectively. TCE in TD-3 has
been generally declining since it was first detected in 1998, and TCE concentrations in TD-4
generally increased since 2001 and appear to have begun declining since 2006. The
concentrations at HH-48 and HH-77 in July 2007 were 85.9 ug/L and 360 ug/L, respectively.
TCE has been present in HH-48 at comparable concentrations since 1989, and TCE
concentrations have generally increased at HH-77 to the present value since 1989. TCE at other
site monitoring wells was not detected (at an appropriate detection limit), was detected below
standards, or was detected slightly above standards. For example, TCE was detected at PEW-4
-------
and CC-33 at 15.7 ug/L and 3.8 ug/L, respectively, relative to a standard of 2.8 ug/L. TD-4 is
screened near the bottom of the saprolite and has a higher TCE concentration than TD-3, which
is screened approximately 30 feet above TD-4. TCE is not vertically delineated in this area. TCE
degradation products (cis-l,2-dichloroethene and vinyl chloride) are either not detected or are
detected at very low concentrations at both the TD-3/4 cluster and the HH cluster indicating that
there is little or no TCE degradation occurring in or upgradient of these areas.
Potential Receptors
Potential complete exposure pathways for humans to contaminated ground water identified in
RI/FS included ingestion of contaminated groundwater downgradient of the site. In 1996, 100%
of adjacent offsite residents considered to be potentially at risk were connected to the Cleveland
County water system and all individual domestic wells were abandoned. Although this might
eliminate current ground water receptors, a potential pathway for future ground water receptors
might still exist.
Potential ecological receptors for contaminated ground water are the unnamed tributaries to
Buffalo Creek that run along the northern and southern portions of the site.
Remedy Objectives
The ROD objectives of the ground water remedy are to
• Control migration of contaminated water
• Remove contaminated water
• Treat and discharge the extracted water
• Attain North Carolina 2L Ground Water Quality Standards
The following table presents the North Carolina 2L Ground Water Quality Standards for
constituents identified in the ROD and for constituents identified subsequent to the ROD.
Constituent
Standard
Inorganics:
Barium
Chromium
Iron
Lead
Manganese
Nickel
Selenium
2,000 ug/L
50 ug/L
300 ug/L
15 ug/L
50 ug/L
100 ug/L
50 ug/L
Organics
Acetone
Benzene
1,1-Biphenyl
bis(2-ethylhexyl) phthalate
Carbon tetrachloride
700 ug/L
lUg/L
350 ug/L
2.5 ug/L
0.269 ug/L
-------
Constituent
Chlordane
Chlorobenzene
Chloroform
Chloromethane
Diethylene oxide (1,4-dioxane)
1 , 1 -Dichloroethane
1 , 1 -Dichloroethene
cis- 1 ,2-Dichloroethene
trans- 1,2-Dichloroethene
Ethylene glycol
Methyl ene chloride
2-Methylnaphthalene
Naphthalene
Phenol
Tetrachloroethene (PCE)
Trichloroethene (TCE)
Vinyl chloride
Standard
O.lng/L
50ng/L
70ng/L
2.6 ug/L
7 ug/L
70 ug/L
7 ug/L
70 ug/L
100 ug/L
14,000 ug/L
4.6 ug/L
14ng/L
21 ug/L
300 ug/L
0.7 ug/L
2.8 ug/L
0.015 ug/L
Additionally, a Class C surface water standard for secondary recreation of 110 ug/L of
diethylene oxide has been established by NCDENR for streams adjacent to the site.
Description of Remedy
The ground water remedy consists of two P&T systems: an inner tier (IT) system designed to
address contamination near the former GRUB disposal area and burn pits and an outer tier (OT)
system designed to control plume migration near the downgradient property boundary.
The IT system consists of nine extraction wells that are each 2 inches in diameter and completed
in the shallow saprolite. The extracted water is pumped to an equalization tank and is then
treated by an activated sludge sequencing bioreactor, air stripping, and carbon adsorption. The
treated water is discharged to the facility's water treatment plant NPDES permitted outfall. The
IT system was temporarily shutdown in April 2004, in accordance with an ESD, to evaluate the
potential for MNA in the absence of remedy pumping.
The OT system consists of 10 extraction wells that are each 6 inches in diameter and completed
in the deep saprolite and shallow bedrock. The extracted water is pumped to an equalization tank
and is then treated by air stripping and carbon adsorption. The treated water is discharged to the
facility's water treatment plant NPDES permitted outfall. The OT system was shutdown in
accordance with the National Priority List delisting process in 1998.
Monitoring Program
The ground water monitoring program includes quarterly sampling from several wells for
constituents of concern and for MNA parameters. Results are reported semi-annually. Tables 1
-------
and 2 from the semi-annual reports outline the monitoring program. These tables are included in
Attachment A of this report for reference.
2.2 FINDINGS RELATED TO EXISTING REMEDY
Contaminant Fate and Transport
The contaminant fate and transport for three pertinent site contaminants (ethylene glycol, TCE,
and diethylene oxide) are discussed in this section. The discussion is limited to these parameters
because they are the likely drivers for site remediation. Understanding the fate and transport of
these contaminants depends heavily on an understanding of the ground water flow at the site.
The site contractor prepared a ground water model to help understand ground water flow at the
site. Findings regarding hydraulic data and this numerical model are discussed first.
Hydraulic Data and Ground Water Flow Model
The ground water flow model was calibrated using the water level data from October 2005. This
data set for the shallow saprolite indicates a hydraulic head gradient of 0.015 feet per foot below
the process building, 0.03 feet per foot from the treatment ponds to the downgradient edge of the
property (and also on the northern and southern edges of the process area), and 0.023 feet per
foot from the western boundary to the eastern boundary. Ground water beneath the process area
and immediately to the east of the process area appears to flow to the north, east, and south. The
hydraulic head gradient for the deep saprolite is relatively similar, and the hydraulic head
gradient for the shallow bedrock is relatively uniform at 0.017 feet per foot. Vertical ground
water flow seems to be downward from the saprolite to the bedrock near the process building and
is almost absent at downgradient locations.
The ground water flow model appears to generally simulate these ground water flow patterns in
the shallow saprolite, deep saprolite, and shallow bedrock. The model calibration statistics are
reasonable, and a sensitivity analysis was conducted. However, there are some limitations to the
model development that might affect how it is used for forecasting and/or understanding ground
water flow at the site.
• Hydraulic conductivity zones and values appear to vary significantly over relatively small
distances. Furthermore, the hydraulic conductivity zones and values do not appear to
agree with the slug or pump test results. For example, in the shallow saprolite, the area
among the ponds was assigned a hydraulic conductivity of 0.3 feet per day, but the slug
test results for O-25, N-29, and W-33 (which are in this area) were 1 foot per day, 2 feet
per day, and 8 feet per day, respectively. Therefore, one would expect the model to have
a higher hydraulic conductivity in this area. Similarly, in the deep saprolite hydraulic
conductivities of 5.3 feet per day (OT-1 pump test) and 10 feet per day (Q-33 slug test)
are located in model zones with a hydraulic conductivity of 1.6 feet per day.
• The low hydraulic conductivity from the IT wells may be the result of well fouling or
well construction. These wells operated as 2-inch extraction wells that had a relatively
low yield. It is likely that they had some degree of iron fouling given the high levels of
10
-------
iron in the aquifer in that vicinity. In addition, extraction wells are typically larger in
diameter, so the small diameter calls into question the quality of the well construction. If
significant well fouling is present, the aquifer hydraulic conductivity in this area may be
higher than that measured by the slug tests.
• The simulated contours appear to be substantially more complex than the contours
interpreted from actual measurements suggesting that the use of many hydraulic
conductivity zones in a single layer with large ranges in hydraulic conductivity (e.g., 0.2
feet per day to 50 feet per day), are complicating the contours in an attempt to match
water levels at individual wells.
• The conceptual model includes a transition zone between the saprolite and bedrock that is
likely higher in hydraulic conductivity than the saprolite or bedrock. This zone could
therefore be considered a preferential pathway for contamination. The zone is observed in
several monitoring wells and is depicted as continuous across the site, but the hydraulic
conductivity of the zone has not been tested, and the zone is included in the deep
saprolite zone in the model rather than as a distinct zone that may have a higher hydraulic
conductivity.
• The model was only calibrated against the water levels from October 2005 using recharge
and stream discharge as the primary stresses. A ground water flow model would ideally
be calibrated under multiple stresses (e.g., pumping and non-pumping conditions). Given
that a P&T system was active at this site, it would seem feasible to also calibrate the
model against a set of water levels from a time when the system was operating to
demonstrate that the chosen parameters also apply under a different pumping scenario. It
would also be appropriate to conduct extended aquifer tests using system extraction wells
to refine hydraulic conductivity estimates.
• The model report does not discuss the conductance used for the river and drain boundary
conditions in the model. These parameters can significantly change how much water is
discharged from the aquifer to the rivers and drains, and combined with recharge and the
constant head boundaries, are the only parameters in the absence of pumping that help
determine the calibrated value of the hydraulic conductivity. The conductance of the
streams close to the site relative to that of the other streams and Buffalo Creek could
control how much of the ground water from the site discharges to the tributaries that are
close to the site versus Buffalo Creek. Given that the rivers are the ultimate discharge
point for all water in the model, which river receives the majority of the water and
contamination will control how far the plume spreads before discharging to surface
water.
• The model report cites that optimal parameter settings for some hydraulic conductivity
zones were not selected because it caused the simulated values for the layers 4 and 5 to
deviate from actual values. However, in another part of the report, it is stated that the
measured values, particularly in layer 5, may not be reliable.
11
-------
In summary, based on the model developed, it is difficult to accept the modeling results without
some reservation. For example, although the calibrated value for a wide portion of the modeling
domain is 2 feet per day and this parameter is relatively sensitive to change in this value, this
hydraulic conductivity value is dependent on the value of the recharge selected and the
conductance of the river/stream beds. A different recharge rate and/or a different riverbed
conductance could result in a reasonable change in this calibrated parameter.
Diethylene Oxide
Based on water level measurements and stream gauge measurements it is readily apparent that
the core interval of diethylene oxide migrating in the deep saprolite/shallow bedrock discharges
to the stream that runs north/south through the eastern edge of the Celanese property. The
elevations of the screen intervals for contaminated wells AA-54, PZ-12, and P-58 are over 720
feet above mean sea level, and the screen interval for contaminated well 11-65 is between 700
and 705 feet above mean sea level. The stream elevation is approximately 700 feet above mean
sea level near 11-65 and diethylene oxide is consistently observed in surface water where the
diethylene oxide plume is expected to discharge. Diethylene oxide concentrations in surface
water have been as high as 67.5 ug/L, which is below the surface water standard for the site of
110 ug/L. There is a general upward gradient from bedrock to saprolite at the MM/AA cluster
location, and there is generally an upward gradient from shallow bedrock to saprolite at the II
cluster. There is a downward gradient in bedrock at the LL cluster, so additional information at
this location or downgradient of this location would be beneficial to demonstrate ultimate
discharge of the plume to the stream. Additional monitoring over time is likely needed to
confirm decreasing concentrations in bedrock wells to support the case that the observed bedrock
contamination is associated with well installation.
To the southeast of the site, the highest diethylene oxide concentrations suggest migration toward
and along the stream that runs through residential area with ground water restrictions. It is
unclear how far the diethylene oxide migrates before discharging to the stream and if the plume
is still present on the far side of the ground water restriction area. Sampling results indicate that
diethylene oxide contamination has not migrated to HH-48 and HH-47 suggesting that these
wells are not in a primary migration pathway for contaminants migrating from the GRUB area.
Ethylene Glycol
In contrast to the diethylene oxide, ethylene glycol appears to be relatively localized in the
source area. This is likely due to the relatively rapid degradation of ethylene glycol; however, it
may also be due, in part, to the extent of the sampling for ethylene glycol. Ten of the monitoring
wells along the preferential pathway for diethylene oxide have not been sampled for ethylene
glycol.
TCE
The interpreted TCE plume also differs from that of diethylene oxide plume. With the exception
of HH-48 and HH-77, the presence of TCE appears limited compared to that of diethylene oxide.
Unlike ethylene glycol, this is not likely due to TCE degradation because TCE degradation
12
-------
products are not present in significant concentrations at any of the sampled monitoring wells.
The apparent limited extent of TCE may partially be due to retardation of TCE since it has the
potential to adsorb to soil more strongly than diethylene oxide, but it would be unlikely for the
TCE retardation factor to be more than a factor of 2 at this site given the expected low organic
carbon fraction of the soil. The relatively limited sampling for TCE may be a factor. Seven of the
wells sampled along the preferential pathway marked by diethylene oxide have not been sampled
for TCE, and another four were sampled as far back as 2001 or earlier. It is unclear when the
TCE release occurred. TCE was first detected in a sample collected from TD-3 in 1998 but was
not detected in TD-4 until 2001. By contrast diethylene oxide was present in some downgradient
locations by 1998. The limited extent of the TCE plume, may be, in part, due to a relatively later
release compared to that for diethylene oxide.
It is unclear if the TCE identified at HH-48 and HH-77 result from a release to ground water
from the Celanese process area. The following are arguments for and against the TCE at HH-48
and HH-77 being linked to the Celanese process area.
Arguments against a link
The TCE detected at HH-48 was detected as early as 1988. It is unclear when or where
the TCE release occurred that resulted in the contamination at TD-3. The first sample
collected from TD-3 was collected in 1998. Evidence for a relatively late release is that
TCE was not detected in TD-4 in 1998. It was first detected in TD-4 in 2001. Therefore,
TCE contamination was identified far downgradient of the site 10 years prior to the first
TCE samples onsite and before detections were present in TD-4.
Diethylene oxide, which is more mobile than TCE and likely released in greater
quantities has not been detected in HH-48 and HH-77. These two wells appear to straddle
the general vertical zone where diethylene oxide would be detected if it was migrating in
this direction. The absence of diethylene oxide in these two wells suggests that these
wells are not along the preferred path for contaminant transport at this site and that the
TCE might be better explained by a local source.
Given that the TCE source is upgradient of the diethylene oxide source, it is expected that
TCE would follow the same migration pathways as the diethylene oxide. Based on
diethylene oxide sample results, the deep saprolite/shallow bedrock/transition zone is
likely the preferential pathway for contaminant transport at the site. As such, it is unlikely
that TCE would have migrated vertically into and through bedrock to the HH wells faster
than it would through the transition zone. In addition, TCE concentrations at the HH
wells first appeared the HH-48 in the saprolite before appearing in HH-77 in the bedrock,
lending further evidence for a localized TCE source at the surface that migrated deeper
over time rather than migrating upward from bedrock.
Potential Data Gaps or Uncertainties
There are concentrations onsite that are high enough to explain the observed
concentrations at HH-48 and HH-77. Although the detected TCE concentrations in TD-3
13
-------
and TD-4 are on the order of 1 to 2 mg/L, these wells are not necessarily at the source
area, and areas onsite with higher concentrations are likely.
There are very few, if any, samples collected from the transition zone between the source
area and HH-48 and HH-77.
A set of realistic transport parameters for the site can result in sufficiently fast transport
times for TCE to have reached HH-48 by 1988. For example, a hydraulic gradient of
0.02, a hydraulic conductivity as low as 5 feet per day in the transition zone, and a
porosity of 0.2 would yield an unretarded transport velocity of approximately 180 feet per
year. This would allow TCE to reach from the TD-3/TD-4 location to the HH-48/HH-77
location within 20 years. The facility began operation in 1960.
Remedy Effectiveness
The IT system appears to have had minimal effectiveness when operating. The wells were small
diameter and had very low yield relative to the estimated amount of water flowing through that
portion of the aquifer. As a result, the mass recovery and source control offered by the system
were likely negligible. The Fourth Five-Year Review indicated that the IT system pumped
approximately 1.6 million gallons over a three year period, which translates to an average
extraction rate of 1 gallon per minute. By contrast, using a hydraulic conductivity of 2 feet per
day (global hydraulic conductivity value from the model), a hydraulic gradient of 0.02, a
saturated thickness of approximately 50 feet, and a source area width of at least 500 feet, the
amount of water flowing through the system would be approximately 5 gallons per minute.
Typically, extraction should exceed this flow rate by a factor of two or more, and in this case, the
extraction is lower by a factor of five.
It is difficult to evaluate the effectiveness of the OT system because little water quality data and
system operational parameters are available to evaluate the hydraulic control it provided. The
diethylene oxide plume has migrated beyond the OT system, but this may have occurred prior to
system installation and operation. The OT system was apparently sufficiently effective to be
delisted in 1998, but this is before diethylene oxide was identified as a potential contaminant of
concern.
The evaluation of an MNA remedy depends on the constituent of concern being evaluated. With
respect to diethylene oxide, MNA does not appear to be a viable remedy because the
contaminant does not readily degrade and has demonstrated the ability to migrate over 1,000 feet
from the source area at concentrations that are 50 times greater than the cleanup standard.
With respect to TCE, there is insufficient information to evaluate MNA because many of the
critical monitoring wells (i.e., those marked by diethylene oxide transport and along ground
water flow paths) have not been sampled for TCE. Finally, there is a lack of information
regarding TCE distribution in the source area and between HH-48 and HH-77. The
concentrations at TD-3 have declined and the concentrations at TD-4 appear to have peaked, but
little is known about the locations of these wells relative to the source of TCE. A shift in the
ground water flow direction may be the cause for the change in concentrations in these wells.
14
-------
Alternatively, the TCE source may have attenuated and the TCE at these wells is migrating from
these points to locations downgradient. Regardless, there is little evidence for degradation of
TCE near TD-3/TD-4 and at HH-48/HH-77. If additional sampling of key wells substantially
increases the known TCE plume extent, then it is likely that MNA would not be a suitable
remedy for TCE. On the other hand, if sampling confirms a limited TCE extent, then MNA may
be appropriate for the TD-3/TD-4 TCE plume once the source has been addressed.
With respect to ethylene glycol, it is likely that MNA is an appropriate remedy due to the
potential for this compound to degrade; however, sampling of key monitoring wells is lacking. It
is unclear if aerobic degradation of ethylene glycol could lead reduced dissolved oxygen
concentrations in nearby streams and/or if increased metals concentrations from reducing
conditions caused by ethylene glycol degradation could lead to metals exceeding surface water
standards.
15
-------
3.0 RECOMMENDATIONS
3.1 SUGGESTED ADDITIONAL CHARACTERIZATION
Diethylene oxide has migrated further and at higher levels above standards than any other
contaminant onsite, and the plume is not delineated horizontally. The ground water flow model
suggests that all ground water discharges to streams in the vicinity of the site, but given the
limitations to model calibration at this complex site, additional information is needed to confirm
this fate of the diethylene oxide and consequently delineate the extent of the plume. Additionally,
the southeastern portion of the diethylene oxide plume appears to run parallel with a stream
running through a residential area. It is likely that diethylene oxide is discharging to this stream,
but additional information is needed to help confirm the extent of the plume along this stream.
Based on the results, it may be necessary to expand the restrictions regarding ground water use in
the area, particularly if it is found that the diethylene oxide continues to migrate in the subsurface
further downstream than the area currently covered by ground water use restrictions.
Suggested Characterization Efforts to the Northeast to Delineate Diethylene Oxide
The north/south running stream (gauged at locations SW-3 and SW-7) is the likely discharge
point for diethylene oxide migrating to the north and east of the site. However, the steep
topography and thick brush leading down to the stream and up the valley on the opposite side
make it impractical to install monitoring wells to delineate the plume with water quality data
alone. Therefore, it is practical to attempt to delineate the plume by demonstrating that it
discharges to this stream and therefore does not continue migrating through the subsurface to
another discharge point further downgradient. Sufficient evidence is discussed in section 2.2 of
this report to demonstrate that this stream is the discharge point for the diethylene oxide;
however, a few data gaps remain. The following is suggested:
• Continue sampling bedrock wells in the LL, MM, and NN clusters to confirm continued
decreasing trends and supporting evidence that observed contamination in bedrock is
associated with well installation rather than significant vertical migration. If continued
monitoring cannot support this hypothesis, then additional data will be needed to
determine the fate of diethylene oxide observed in deep bedrock because some of these
deep wells are almost 200 feet deeper than the stream bed.
• Install a bedrock monitoring well at the II cluster location to vertically delineate the
plume and to demonstrate an upward gradient within bedrock at this location given that
there appears to be a downward gradient at the LL cluster. To the degree possible,
measures should be taken to minimize the potential for pulling down contamination while
drilling. A telescoping well may be appropriate with a total depth of approximately 120
feet below ground surface. The well should be sampled and analyzed for site
contaminants and water levels should be measured and compared with other locations to
16
-------
evaluate the vertical hydraulic gradient. The cost for installing this well might be
$45,000, including work plans and contractor oversight.
• Install one or two new stream gauges upstream of SW-3 and include stream gauge
measurements from existing and new stream gauges in development of potentiometric
surface maps and potentiometric surface cross-sections.
Suggested Characterization Efforts to the Southeast
The diethylene oxide plume to the southeast (i.e., near OT-2R) would be best delineated by
several monitoring wells placed within and potentially downgradient of the residential area with
restricted ground water use. Given that ground water in this area is expected to discharge to the
stream, an alternative approach to delineation could be to collect and analyze samples from
ground water along the stream bank to determine where diethylene oxide is discharging to the
stream and at what distance downstream the diethylene oxide concentrations become
undetectable. A sampling methodology should be developed to ensure that ground water, and not
surface water, is sampled. If the results are inconclusive, delineation with monitoring wells
would be needed. In addition, if the plume extends beyond the current ground water restriction
area, potable well sampling and use restrictions would likely need to be expanded downgradient.
Assuming ground water sampling at 10 stream bank locations and surface water sampling at 5
locations, the cost for delineation activities might be approximately $25,000, including planning.
Suggested Sampling for Delineation of Other Constituents
The distributions of TCE and ethylene glycol at the site are not well understood due to a lack of
sampling from appropriate locations. The diethylene oxide plume helps highlight preferential
pathways for contaminant migration, and many of the wells with elevated diethylene oxide
concentrations have not been sampled for TCE or ethylene glycol. The following existing
monitoring wells should be sampled and analyzed for TCE (and volatile organic compounds),
DOWTHERM™ A, ethylene glycol, dissolved oxygen, oxygen demand, and metals because
elevated diethylene oxide concentrations are present or because they are transition zone wells.
- F-55
- G-50
- 1-57
- M-44
- P-58
- Z-78
- W-23
- AA-54
- BB-18
- DD-58R
- GG-39
- 11-65
- KK-55
17
-------
- LL-110
- LL-175
- LL-295
- MM-128
- MM-170
- MM-280
- NN-105
- NN-240
- PZ-3
- PZ-4
- PZ-12
- PZ-13
- OT-1R
- OT-2R
- OT-3
- The new deep well installed at the II cluster.
Assuming six wells can be sampled by a two person crew for $2,000 per day and analysis costs
of $250 per sample, this additional sampling could likely be accomplished for under $25,000 per
event. A minimum of two events would be suggested.
Suggested Characterization for the TCE Source Zone
The TCE source zone has not been identified, and the plume is not vertically delineated beneath
TD-4. A source zone investigation utilizing existing information and direct-push sampling is
recommended to identify the source area and determine if it is practical to remove. It is also
recommended that two new bedrock monitoring wells (one below and one approximately 100
feet downgradient and deeper than TD-4) be installed and sampled to vertically delineate TCE in
this area. The TRIAD approach with dynamic work planning and field-based decisions should be
considered when planning and executing all aspects of this characterization to minimize the
number of field events and to provide for timely characterization. Assuming a file review, three-
day direct-push event, plus installation and sampling of two monitoring wells, the cost for these
activities would likely be approximately $50,000, including planning.
Based on the results from the above activities, the site team should revisit the site conceptual
model and evaluate if additional important data gaps are missing with respect to the TCE source
and migration pathways. If additional data gaps are missing, additional characterization may be
required prior to selecting an appropriate remedial measure.
A number of limitations were mentioned with regard to the ground water flow model, but it is
unclear if these limitations should be addressed to reliably use the model for evaluating
contaminant fate and transport. Some of the limitations are fundamental, including the presence,
extent, and parameters of the transition zone as well as the conductance of the river and drain
boundary conditions. It is unclear if sufficient information is present to help accurately adjust the
model accordingly. Additional information would likely be required, and even if the model were
18
-------
updated, it is likely that ultimate decisions regarding contaminant fate and transport and site
remedy would be based on continuing water quality data. An updated model may be appropriate
for evaluating a P&T capture zone or to design an extraction system, but this is not an immediate
concern for the site.
3.2 SUGGESTED REMEDIAL STRATEGY (FROM A TECHNICAL PERSPECTIVE)
Selecting an appropriate remedial measure for the site is highly dependent on the results of the
recommended characterization, on the interpretation of regulations by the site team, and on the
development of appropriate standards for diethylene oxide. Remedy selection is discussed in
context of each of the three primary contaminants, beginning with diethylene oxide.
Diethylene Oxide
Diethylene oxide is a difficult contaminant to address. It cannot readily be removed via air
stripping or carbon adsorption and is generally resistant to bioremediation in typical subsurface
conditions or in typical bioreactors. To date, the discharge of diethylene oxide plume to nearby
streams has not resulted in an exceedance of the surface water criteria, but diethylene oxide
remains in the subsurface above the 7 ug/L ground water criteria which serves as an ARAR for
this site. There are two predominant remediation options for this constituent. One is to identify
and remove the remaining source material and allow the existing plume to attenuate. The other is
to declare source removal impractical and to capture the core of the plume, allowing the toe of
the plume to attenuate. Some combination of source removal and ground water extraction may
also be appropriate to allow aquifer restoration in a timely manner.
The GRUB area was reportedly used as far back as the mid to late 1960s, and the earliest
diethylene oxide sampling (in 1998) suggests that plume had migrated at least as far as the OT
system. Assuming diethylene oxide reached ground water in 1970 and the plume reached its
current extent by 1998, suggests that transport took less than 30 years. It is reasonable to
conclude that in the absence of a continuing source cleanup via natural flushing and dispersion
will take substantially longer than initial plume transport.
Figure 2 provides a flow chart to illustrate the potential process for arriving at a remedy for
diethylene oxide. Consistent with current findings in literature, this flow chart assumes that in-
situ remediation will not be appropriate for providing reliable plume control. Source removal,
source control, and active plume remediation are considered as is the potential need to expand
institutional controls for ground water usage in the vicinity of the site is also indicated.
Long-term source control or active plume remediation with ground water extraction and
treatment has the limitation of needing to provide cost-effective treatment. The facility's
activated sludge treatment plant provides some treatment of diethylene oxide through
cometabolism with tetrahydrofuran, which is reportedly present in the facility's waste water.
However, it is unclear if this degree of treatment would be sufficient to meet standards. In
addition, use of the facility's activated sludge system would require the system to continue
operating for the duration of the source control remedy, which would very likely be longer than
30 years. If additional treatment for diethylene oxide is needed, the most successful and reliable
19
-------
currently available treatment process is advanced oxidation. This is typically implemented with
UV/oxidation or with oxidation from reacting ozone and hydrogen peroxide. The implementation
of either of these approaches would highly depend on the quality of the extracted water. The
performance of these technologies is greatly influenced by the presence of iron and solids, pH,
total organic carbon, and other parameters. In addition, if bromide is present then there is
potential to create bromate (a carcinogen) as a byproduct. These complications associated with
long-term water treatment are further motivating factors for identifying and removing the source
material in an attempt to avoid ground water extraction and treatment.
TCE
Additional sampling for TCE may indicate that it has an attenuating source and a plume of
limited extent that is controlled by degradation associated with reducing conditions from the
ethylene glycol plume. Alternatively, additional sampling may indicate a continuing source and
preferential pathways that allow TCE to migrate from the site. If the former is true, then TCE
may be appropriately addressed by continued monitoring to document the declining source and
the controlled plume. If the latter is true, then source area remediation or some form of plume
control may be appropriate. The form of plume control would depend on how the site team
chooses to address the other contaminants. For example, if the site team chooses to control
diethylene oxide migration with a P&T system, then the TCE could likely also be controlled with
the same system.
Regardless of the source, the TCE in the vicinity of the HH cluster should be delineated as it may
extend downgradient of the restricted ground water use area. However, if the source is
determined to be local to the HH cluster (rather than site-related), these delineation activities
would presumably not be the responsibility of Celanese. The results from the suggested sampling
of existing wells and from the TCE source investigation should provide adequate information to
conclude if the source is site-related or local to the HH cluster.
Ethylene Glycol
Additional sampling for ethylene glycol may confirm that this constituent is degrading
sufficiently to limit plume migration. Further sampling may also indicate that the reducing
conditions fostered by the ethylene glycol are instrumental in controlling or helping to control
the TCE plume. Alternatively, additional monitoring may suggest that degradation of the organic
carbon associated with the ethylene glycol may eventually adversely affect the dissolved oxygen
content or metals concentrations in nearby streams.
20
-------
3.3 POTENTIAL SCHEDULE
It is recommended that additional testing and study for potential remedies be postponed until the
results from the above-noted sampling are available and the site conceptual model updated. The
following schedule might apply:
Conduct the suggested characterization
Report findings and update site conceptual model
Update/expand institutional controls as needed
Select a ground water remedy (source remediation,
source control, and the potential the need for active
plume remediation)
Design and plan for the ground water remedy
Implement the ground water remedy
Evaluate effectiveness and modify remedy accordingly
By 12/31/09
By 3/1/10
By 3/3 1/10
By 6/30/10
By 12/30/10
Begin by 3/3 1/11
Ongoing
21
-------
ATTACHMENT A:
FIGURES
Figure 1. Migration Pathways Based on Diethylene Oxide Concentrations >70 ug/L
Figure 2. Flow chart Illustrating Potential Process for Arriving at a Remedy for Diethylene
Oxide
A-l
-------
Legend
DIETHYLENE OXIDE* 0.007
DIETHYLENE OXIDE >0.007
a SURFACE WATER LOCATION
MONITORING WELL
EXTRACTION WELL
PEW WELL
* PIEZOMETER
| | PARCEL WITH WATER RESTRICTION
— CREEK
| | INDUSTRIAL POND
| | RECREATION POND
| | BUILDING'STRUCTURE
| |PROPERTY BOUNDARY
500 250 0 500
^^B=^^^^H Feet
NOTES:
it: CONCENTRATION UNIT: rr^'L
(21 BOLD-FACED FONT: DETECTED
CONCENTRATION ABOVE NC 2L STANDARD
(3) 0.007 rug/LAS NC 2L STANDARD FOR
DIETHYLENE OXIDE
If, HH AND ELLIOT WELL DATA IS FROM APRIL 2C
SAMPLING EVENT
Figure 1 - Migration Pathways Based on
Diethylene Oxide Concentrations > 70 ug/L
-------
Characterize source area and determine practicability
of source removal
Delineate plume
Update/expand institutional
controls for ground water
usage in the vicinity of the
site as needed
Estimate time frame for plume attenuation in the
absence of a continuing source
Is active ground water
treatment required for
source control, plume
control, or accelerated
plume remediation?
Is passive remedy
meeting remedial
objectives in a
timely manner?
Yes
Determine target capture zone and design
an appropriate extraction system to
provide control, estimate influent
concentrations
Proceed with an approved
passive remedy and evaluate
performance
Design, implement, and operate an appropriate
pre- or post- treatment system to address
diethylene oxide prior to discharge to facility
water treatment plant
No
Does existing treatment
process have adequate
hydraulic capacity to accept
water from the new treatment
system?
Does existing treatment
plant adequately and
reliably treat diethylene
oxide?
Design, implement, and operate an appropriate
treatment system to address all contaminants of
concern at the required hydraulic capacity
Develop standard for discharge to surface water
from treatment plant
Operate new extraction system with existing
treatment components with minor modifications
recognizing that an alternative will be needed if
operation of the existing treatment facility is
discontinued or modified
Figure 2. Flow Chart Illustrating Potential
Process for Arriving at a Remedy for
Diethylene Oxide
-------
ATTACHMENT B:
FIGURES AND TABLES FROM THE CONCEPTUAL SITE MODEL REPORT AND
SEMI-ANNUAL REPORTS
Figure 3-1 Site Location Map
Figure 6 Hydrogeological Cross Section A-A', January 2007
Figure 4-2 Cross Section A-A'
Figure 4-3 Cross Section B-B'
Figure 4-4 Cross Section C-C'
Figure 3-3 Contamination Source Locations and OU-2 Excavation Extent Map
Table 1 Quarterly Groundwater Sampling Plan — CERCLA Parameters
Table 2 Quarterly Groundwater Sampling Plan — MNA Parameters
B-l
-------
O
SOURCE: USGS 7.5 MIN. QUADRANGLES
BLACKSBURG NORTH, NC -1971
GROVER, NC-1993
SCALE
Earth Tech
A Tyco International Ltd. Company
FIGURE 3-1
SITE LOCATION MAP
CNA HOLDINGS. INC./TICONA
SHELBY. NORTH CAROLINA
SEPTEMBER 2007
79750
-------
850
840
830
820
810
800
790
780
770
760
750
740
730
720
710
700
690
680
670
Plant Area
Wastewater Treatment Area
*:'>
H. . ' .'.
• • • .*•• • --j • .*•: • • *-ii-.: ••.*•• ii ••.*••• -^-- • ••• •
- .v- • "v^jr- • '.v- • £i-'.v- • >.oS-v- • '^w^.'-'.^
v '• -f ' ' O- V V '• V 'O r '. V ' Or'. V '; '• <&' • v '•
'• vD-88 -.\v.' •'•
r-*-
<.: ••••j
v.' • > •. • v.' •
I'.'*.:?'.""'.'*.
v.' • > -.'• v; •
:.'-V:'> ^V:'
.'.\: v."
v; • > •.'•
• v; • >_ •
''•. •*' •';
.. */~:-
v; • > •.'•
• v; • >-/.v; • -v.1 • >_•
''.-•"yv"'-X:^
._•• v; • '• v; • y •
?'••<:!'••<::
.. */~."
v.' • > •.'•
.. */~."
v.' • > •.'•
''.'i:'"-::
V .' • > •.
•-.- ..*.,-.
• • v; • > •
•:'.'-r':>
v-\:V
•'• y .'•'>•
• y ' • v •
;.:•.'•; :^;>y;?yi..
• y .' • > -."'ly .' • • y ." • > -/^
•.'•••'':'; si.-.f'.'O.f'.'j; Jf'.-.f.'jr^s
"H • • : -"H • 'H • • : -"H • • : 'A "^^
•. - ^-.: ^: -V - ^-.-. - X: ^: •. - X: ^: •. - XOs
• v.' • > -."'ly.' • • y." • > -."• y." • > -."• y." • ' y." •
• ': *' •\J ^1 \ ^' •'' \ ^' •' ^ ^ \ ^' •' ^ ^ \ "*' •'' \ ^ •^ ^ \^
•1'- X: ^: -1!'- '^.-1'- X: ^: -1'- X: ^: -1'- '^.-1 '• "^.: ^: -1 "• X: ^ -1 "• '^.-1 • i-.-
• v.' • > -."'ly.' • • y." • > -."• y." • > -."• y." • ' y." • > •."" y." • > •."" y." • " y." • ^
•?-X: ^'.X-"'.'*.: :v1'.\: ^•:-'!vi.-'!'.\: y/-'!'.'*.: y/^'.X-'i'.x:
• y.' • v •. '|y.' • • y.' • v •.'• y.' • v •.'• y.' • • y." • v -.'• y.' • > -."• y.' • • y." • ^
:.'-V:'> ?':r I:'-'-r':^ ^V:'> T-'-V:'^V: > ^•'••V:'> ^-V:':•'•V:'.
;J'. *.: V.-'-r' '^.-H' *.: y.-'-J' *.: y.-'-J' '*.-H' '*.: V:'•"!'. *.: ''"••*. '• '*..*. y*:.
• v; • -T -.;|v^ • v; • > •.'• v; • > •.'• v; - • v; • > -.'• v; - > •.'• v; - • v; • > -|- v; • > -.
:.'-V:> ?':?':':'•'• r':'^ ^V:'> ^^^ ^V:'> ^'V:':•'• ^:> ^'^
;J'. *.: V.-'-r' '^.-H' *.: y.-'-J' *.: y.-'-J' '*.-H' '*.: V:'•"!'. *.: iv''. '*.-H' '•
• v.1 • >.-.''Iv.1 • -v.1 • >.-.''.v.1 • >.-.''.v.1 - • v; • >-.''.v.1 - y -/.v.1 - • v;
•" '• '^.: ^/ 'f '• '*.•" '• X: ?':'•'! '• '^.: ?':'•'! '• '*..". '• '•'-.'• ?:'". '• '^.: ?:'". '• '*..". '• '•
• v;. '> -'-iv;. '• y;. '> • '• v;. 'y •/• v;. '• v;. 'y - '• v;. V •/ v; - • v;
;V:'^'-V:':V:'|^V:'|^V:':^^V>.^.V:'>.^V:':>
•J'. '^.: ^.- f. '• X'H'. '^.: ?: '. '• '^.: ^.- '. '• '\-H'. '^.: 'T: '. '• '^.: ^.- '. '• '\-H'. '•
• v;. '> -'-iv;. '• y;. '> • '• v;. 'y •/• v;. '• v;. 'y - '• v;. V •/ v; - • v;
^*/* V-r :':'-'• r':'* T-'-r :'* ^.'••'':'^V:'> T-'"* :'> ^* ^^
;J'. '*.: HV'. X.-H'. X.: H'-J'. X.: H'-J'. \.-H'. \: '^'H'. \.: H'J'. \-H'. '•
• v •. >_-.'-|v; • • v; • >_-.''.v; • >.-.''.v; - • v; • >-.''.v; - y -.''.v; - • v;
.
•!'• v
-j- y '
.. *.•..•••• *.
v.' • > •. • v.' •
-.-*.•.:
• v.' • > •
.. *-.-.:^- *.-.• J-.-^.'-.- J-.-^.'-.- *.'.- *.•..".• J-.-^'-.- *.'.- *.
> - '• V ' . y • '• V ' . • V.
•r.v.-.'i
ii"-.*.:
.'•• •.• •
'• v; •>.
-..*.-?
' y.' • >
> ".'•"*•> &\*'."'.\ *'•"> ".'•"*•> ".''•*'-"f ."••*'•> ".'•"*•> *."••*'-"f ."••*'•>^.'-^- > ".'-
.. *_.;":•
y.' • > •. '
\- - - «-.-. :
.'• y; .'>_•.
.'f • *.' • :
. - *-.-.:-.. «..-. - «..- ^;-.. -i..- ^;-.. *.-.. *.• r." . • *.-. ;•.. V . •
y.' • > •.'" y.' • ' y.' • > •.'" y.' • > •.'" y.' •' y." • > •.' y.' • V •."' y.' •' y."
r -.'• y." - -r -.'• y.' • • y.'
".- *.•.:••• V-- *.••••
,'• y;/y •.'• y; •"• y.' •"> •.
,'• v."§> •.'• y.1 • • y.1 • >
?•'•:*; ^V:V-V:">
••iv^y^'.viv^V:
,'• y.' • > •.'• y.' • • y.' • > •.
. j v..
• "SJ'- '• ". •
• yjto* '•' v- •
$;,*• *;s:
'•".:• i:-'±'-".:\
• y.' - > •/ y.' -
850
840
830
820
810
800
790
780
r
,770
760
750
740
730
720
710
700
690
680
670
1 inch equals 500 feet
• Line of equal fluid potential (1 msl) Screen
(dashed where inferred)
Potentiometric Surface I 1 SaProlite
Inferred direction of groundwater fc=^| Bedrock
movement
^) EarthTech
A Tyco International Ltd. Company
FIGURES
HYDROGEOLOGICAL CROSS SECTION A-A'
JANUARY 2007
CNA Holdings, Inc./Ticona
Shelby, North Carolina
SEPTEMBER 2007
79750
-------
850
800
750
o>
u_
(0
I
HI
700
600
550
500
A
A' (East)
TD-3
X
4=
$
X
4=
$
T
T
T
T
LEGEND
Ground Surface
Casing
Sap Screen
Deep Sap Screen
TZ Screen
Shallow BR Scrn
Intermed BR Scrn.
Deep BR Scrn.
TopofTZ
Top of BR
Fracture Locations
Shallow Sap GWEIev.
Deep Sap GW Elev.
TZ GW Elev.
Shallow BR GW Elev
Intermed BR GW Elev.
Deep BR GW Elev. j
HH-7]
o
2
T3
a)
CD
NOTES:
(1) Groundwater elevations measured in January 2007
Well ID placement (2) TD-3, TD-4, TI-1, and TI-2 well screen intervals not
identified
shallow BR
EARTH
TEC
A tijca INTERNATIONAL LTD. COMPANY
FIGURE 4-2
CROSS SECTION A-A1
CNA Holdings, Inc./Ticona
Shelby, North Carolina
September 2007
79750
-------
B' (North)
850
800
750
"5 700 -I
0)
(0
650
600
550
500
OO-95
X
=1=
$
X
=1=
$
v
T
T
T
LEGEND
Ground Surface
Casing
Shallow Sap Screen
Deep Sap Screen
TZ Screen
Shallow BR Scrn
Intermed BR Scrn.
Deep BR Scrn.
TopofTZ
Top of BR
Fracture Locations
Sap GW Elev.
Deep Sap GW Elev.
TZ GW Elev.
Shallow BR GW Elev
Intermed BR GW Elev.
Deep BRGW Elev.
Bedrock
OO-150
OO-218
hallow Sap
shallow BR
Well ID placement
NOTES:
(1) Groundwater Elevations measured in January 2007
EARTH
TEC
A tl/CO INTERNATIONAL LTD. COMPANY
FIGURE 4-3
CROSS SECTION B-B1
CNA Holdings, Inc./Ticona
Shelby, North Carolina
September 2007
79750
-------
850
C (South)
C' (North)
800
750
700
re
600
550
X
4=
X
4=
LEGEND
Ground Surface
Casing
Shallow Sap Screen
Deep Sap Screen
TZ Screen
Shallow BR Scrn
Intermed BR Scrn.
Deep BR Scrn.
TopofTZ
Top of BR
Fracture Locations
Shallow Sap GW Elev.
Deep Sap GW Elev.
TZ GW Elev.
Shallow BR GW Elev
Intermed BR GW Elev.
Deep BR GW Elev.
1MN-105
Transition
Zone
Bedrock
NN-240
NN-280
500
LL-110
PZ-12
MM-128
MM-190
LL-195
LL-295
shal
ow Sap
NOTES:
(1) Groundwater Elevations measured in January 2007
Well ID placement
shallow BR
EARTH
TECH
A tlfCO INTERNATIONAL LTD. COMPANY
Figure 4-4
CROSS SECTION C-C1
CNA Holdings, Inc./Ticona
Shelby, North Carolina
September 2007
79750
-------
Y-38
Y-74
H-59
PEW-1
H-79
EarthTech
A Tyco International Ltd. Company
Legend
® WELL LOCATION
I PROPERTY BOUNDARY
OU1 BURN AREA
OU1 GRUB AREA
100 50 0
100
Feet
FIGURE 3-3
CONTAMINATION SOURCE LOCATIONS
AND OU-2 EXCAVATION EXTENT MAP
CNA Holdings, Inc./Ticona
Shelby, North Carolina
SEPTEMBER 2007
79750
-------
Table 1
Quarterly Groundwater Sampling Plan - CERCLA Parameters
CNA Holdings/Ticona Facility - Shelby, North Carolina
Earth Tech Project No. 79750
C-49
F-55
J-29
K-28
P-58
T-35
V-23
AA-54
CC-33
EE-58
FF-34
FF-62
GG-61
HH-48
HH-77
TD-3
TD-4
PEW-1
PEW-3
PEW-4
ITCI
ITEF
OTCI
OTEF
January 2007
O
o
X
X
en
O
o
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
SVOCs
X
X
X
X
X
Field Parameters1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0)
Q.
E
ro
CO
o
o
MS/D
DUP
DUP, MS/D
EB
FB
Inner Tier Shut Off
Inner Tier Shut Off
Outer Tier Shut Off
Outer Tier Shut Off
April 2007
O
o
X
X
en
O
g
X
X
SVOCs
X
X
X
X
Field Parameters1
X
X
X
X
0)
Q.
E
ro
CO
o
o
DUP
Inner Tier Shut Off
Inner Tier Shut Off
Outer Tier Shut Off
Outer Tier Shut Off
July 2007
O
o
X
X
en
O
o
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
SVOCs
X
X
X
X
Field Parameters1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0)
Q.
E
ro
CO
o
o
DUP
EB
MS/D
Inner Tier Shut Off
Inner Tier Shut Off
Outer Tier Shut Off
Outer Tier Shut Off
October 2007
O
o
X
X
en
O
o
X
X
SVOCs
X
X
X
X
Field Parameters1
X
X
X
X
0)
Q.
E
ro
CO
o
o
Inner Tier Shut Off
Inner Tier Shut Off
Outer Tier Shut Off
Outer Tier Shut Off
" Field Parameters - depth to groundwater, dissolved oxygen,
oxidation-reduction potential, pH, specific conductance,
temperature, turbidity
TOC - total organic carbon
VOCs - volatile organic compounds
SVOCs - semi-volatile organic compounds
QC - quality control
DUP - duplicate
EB - equipment blank
FB-field blank
MS/D - matrixspike/matrix spike duplicate
Tables 1H 2007 Report.xls
Page 1 of 1
9/25/2007
-------
Table 2
Quarterly Groundwater Sampling Plan - MNA Parameters
CNA Holdings/Ticona Facility - Shelby, North Carolina
Earth Tech Project No. 79750
IT-1
IT-2
IT-3
IT-4
IT-5
IT-6
IT-7
IT-8R
IT-9
F-55
G-50
G-88
J-29
K-28
N-29
O-25
Q-33
S-1
TI-1
TI-2
U-38
V-23
V-65
W-23
X-32
CC-33
ITCI
ITEF
OTCI
OTEF
January 2007
o
o
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0
o
><
0
<
ffi
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
MNA Parameters1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Field Parameters2
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0)
Q.
E
ro
CO
o
o
DUP
Inner Tier Shut Off
Inner Tier Shut Off
Outer Tier Shut Off
Outer Tier Shut Off
April 2007
O
o
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0
o
><
0
<
ffi
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
MNA Parameters1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Field Parameters2
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0)
Q.
E
ro
CO
o
o
DUP
DUP
DUP
Inner Tier Shut Off
Inner Tier Shut Off
Outer Tier Shut Off
Outer Tier Shut Off
July 2007
O
o
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0
o
><
0
<
ffi
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
MNA Parameters1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Field Parameters2
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0)
Q.
E
ro
CO
o
o
DUP
IMS/D
FB
Inner Tier Shut Off
Inner Tier Shut Off
Outer Tier Shut Off
Outer Tier Shut Off
October 2007
O
o
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0
o
><
0
<
m
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
MNA Parameters1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Field Parameters2
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0)
Q.
E
ro
CO
o
o
Inner Tier Shut Off
Inner Tier Shut Off
Outer Tier Shut Off
Outer Tier Shut Off
1 - MNA Parameters - methane, total iron, total manganese,
total alkalinity, nitrate-nitrogen, sulfate, ethyl alcohol, acetate
2 - Field Parameters - depth to groundwater, dissolved oxygen,
oxidation-reduction potential, pH, specific conductance,
temperaturte, turbidity
TOC - total organic carbon
QC - quality control
DUP - duplicate
FB-field blank
MS/D - matrixspike/matrix spike duplicate
Tables 1H 2007 Report.xls
Page 1 of 1
9/25/2007
------- |