NC0004952_Meeting Notes_20030903

NPDES DOCUMENT !;CANNING COVER SHEET
NPDES Permit:
NC0004952
Celanese Shelby Facility
Document Type:
Permit Issuance
Wasteload Allocation
Authorization to Construct (AtC)
Permit Modification
Complete File - Historical
Engineering Alternatives (EAA)
Correspondence
Owner Name Change
V
/led j S
•A,
�iYf``YH1 Pfw..a
Instream Assment (67b)
Speculative Limits
Environmental Assessment (EA)
Document Date:
September 3, 2003
Thies document is printed on reuse paper - ignore any
content an the reYex.se side

NC 000 f i,S t.
Final Conference Call Minutes - Celanese/Shelby Fibers Superfund Site
Purpose: Discuss "Talking Points - Proposed Modifications to Groundwater Treatment System"
prepared by Kubal-Furr & Associates, August, 2003
Call Date: September 3, 2003
Attendees: Ken Lucas (EPA)
Jerry Kubal (Kubal-Furr)
Pem Carter (Ticona Fibers/Shelby)
John Kuhn (Ticona Fibers/Shelby)
Steve Olp (CNA Holdings)
Dave Mattison (North Carolina Superfund Section)
Everett Glover (Earth Tech, Inc.)
John Blanchard (Black & Veatch)
Background: Kubal-Furr submitted "Talking Points - Proposed Modifications to Groundwater Treatment
System" to the EPA. Ticona is proposing shutting down the groundwater treatment system currently
treating groundwater recovered from the "inner -tier" extraction wells. Instead, the extracted groundwater
would be pumped to the existing wastewater treatment plant (WWTP) where it would be treated in
combination with the process water generated by the ongoing industrial operations.
Minutes:
1. J. Kubal provided an overview of the document and site background in general. There were initially two
records of decision (RODs) prepared for the site. The first ROD required pumping and treating
groundwater from the "inner tier" and "outer tier" groundwater extraction wells under Operable Unit (OU)
1. The second ROD required source removal and remediation of nearby creeks under OU2. The "outer tier"
portion of OU1 and all of OU2 have been completed and delisted. Groundwater is currently being extracted
from the inner tier wells and treated by the existing groundwater treatment system.
The primary contaminant being treated, in terms of concentration, is ethylene glycol. Other groundwater
contaminants include volatile organic compounds (VOCs).
A separate WWTP treats process water from ongoing industrial operations at the site. When the OU1 ROD
was prepared, it was thought that a separate groundwater treatment system would be necessary to prevent
overloading of the existing WWTP being used to treat industrial process water. The primary concern was
that the biological treatment used in the WWTP "prefers" ethylene glycol over other contaminants being
treated; introducing additional ethylene glycol from the contaminated water would adversely impact the
treatment of other contaminants in the industrial process water.

J. Kubal noted that, due to limited groundwater quantities extracted from the inner tier monitoring well
network, the WWTP can treat the contaminated groundwater along with the industrial process water.
Ticona's intent is to eliminate the "redundancy" of operating both the groundwater treatment system and the
industrial WWTP.
J. Kubal noted that biofouling of monitoring wells, limited groundwater yield from the saprolite aquifer,
and the adverse affect of iron and manganese on submersible pumps has progressively reduced the
efficiency of the inner tier groundwater extraction system. Kubal-Furr has replaced various components of
the groundwater extraction system to address these issues, with little or no success.
He noted that the proposed modifications would not address these issues.
J. Kubal presented Attachments A, B, and C of the "Talking Points" and opened the call for questions:
Questions/Answers:
Q1. K. Lucas: Why were the outer tier wells shut down?
Al: The outer tier wells were included in OU1 due to the potential risk to nearby residents drinking water
from groundwater wells. Those residents have since been connected to the Cleveland County water supply,
institutional controls have been put in place to prevent future installation and use of groundwater wells, and
existing wells have been plugged.
Q2: K. Lucas: Why didn't they pump extracted water to the WWTP and treat it there to begin with, instead
of constructing a separate groundwater treatment system? Why propose it now?
Al: For several reasons: 1) Having a separate groundwater treatment system would allow easier tracking of
the groundwater contamination (E. Glover). 2) As discussed above, there was a concern that combining the
extracted groundwater with the industrial process water would "overload" the biological treatment system.
At the time the ROD was signed, the industrial process used larger volumes of ethylene glycol than are
currently used. With the smaller volumes of ethylene glycol currently used by the industrial process, the
WWTP would not be overloaded by adding the extracted groundwater to the waste stream (Olp).
Q3: K. Lucas: How will the performance of the groundwater treatment remedy be tracked under this
proposal?
A3: The existing monitoring well network would continue to be sampled.
Q4: K. Lucas: What would happen if the existing industrial process were shut down? How would the
groundwater be treated?
A4: The existing groundwater treatment system would be restarted, if needed.
Q5: K. Lucas: Kubal-Furr had proposed shutting down the groundwater extraction system and replacing
that remedy with monitored natural attenuation (MNA). Kubal-Furr still considers MNA to be a viable
option. Will MNA parameters continue to be collected to support potential MNA in the future?
A5: Yes.

Q6: K. Lucas: Will water being piped from the extraction wells to the WWTP be monitored?
A6: It will be monitored; one method would be to collect samples through a sampling port.
Q7: J. Blanchard: The Talking Points note that the contaminated groundwater contains the same suite of
contaminants present in the industrial wastewater, which is treated by the WWTP. Are the contaminants
exactly the same? Can Kubal-Furr provide a table comparing the contaminants/concentrations present in
the groundwater versus those in the industrial wastewater being treated by the WWTP? We'd like to be able
to see where the differences are, and determine if there are any impacts.
A7: Yes, this information can be provided.
Q8: K. Lucas: How often are the monitoring wells currently sampled?
A8: The inner tier wells are sampled quarterly; site wide, approximately 12 wells are sampled semi-
annually. Kubal-Furr will provide a sampling matrix table from out of the semiannual report to avoid
confusion about which and how many wells are sampled.
Other Minutes:
1. K. Lucas noted that the proposed change must be clearly spelled out. The scope, cost, and performance
of the proposed change must be compared to the scope, cost, performance of the remedy specified in the
ROD. This information is needed to determine whether this would be a "fundamental change" requiring a
ROD amendment and public meeting, or would be a lesser change, requiring only an explanation of
significant differences (ESD).
2. K. Lucas noted that community concerns must be considered. For example, the community may now
think that "everything is being taken care of' by the existing groundwater treatment systems. Pumping
"Superfund" water to the WWTP, where some of the community members may work, may cause concern
because of the community's perception of Superfund.
3. K. Lucas noted that financial assurances would have to be put in place to ensure that treatment would
continue if the industrial plant were shut down.
4. E. Glover noted that Mark McIntire of the North Carolina NPDES Water Quality Section had stated that
this proposed change would not require changes to the NPDES permit. D. Mattison noted that the NPDES
permit typically does not address Superfund-type contaminants. Therefore, a separate monitoring system
may be required to address the contaminants present in the extracted groundwater.
5. D. Mattison suggested that a public meeting be held even if only an ESD is required. This would ensure
that the community is aware of the changes and may minimize their concerns.
6. D. Mattison noted that Kubal-Furr must demonstrate that this remedy is not simply a way to treat the
groundwater contaminants by "dilution."

7. K. Lucas will determine if any required permits/permit changes must be in place prior to issuance of a
ROD amendment, or if they can be obtained as part the modifications.
8. J. Blanchard requested that Kubal-Furr clearly spell out how all numbers were derived in Attachment C.
In addition, J. Blanchard requested that Kubal-Furr provide a detailed description of the existing WWTP
treatment system. The attachments currently only note "biological" treatment. Kubal-Furr agreed to provide
this information through a "feasibility -like" document.
9. Several callers expressed concern about the ability of the WWTP to replace the pH adjustment,
coagulation, iron removal, biological treatment, air stripping, and carbon adsorption functions performed
by the existing groundwater treatment system. This must be clearly explained and demonstrated.
10. D. Mattison noted that the comparison between the proposed changes and the original remedy should
include contaminants treated by the WWTP and the groundwater treatment system, monitoring
requirements for the WWTP and the groundwater treatment system, and a comparison of the treatment
technologies. This is in addition to those parameters requested by K. Lucas (scope, cost, performance).
11. J. Blanchard confirmed that there have been no ESDs or ROD amendments at the site to date.
12. S. Olp noted that they would like to have the proposed changes in place by October 2003.
13. A site visit will be conducted on September 18-19th.
Action Items:
1. J. Blanchard will prepare and distribute draft meeting minutes to all callers.
2. All callers will review and comment on meeting minutes; J. Blanchard will make final changes and issue
final minutes.
3. J. Blanchard, K. Lucas, and D. Mattison will meet to discuss and prepare comments. K. Lucas will
transmit the comments to Celanese (Olp) by Friday September 12` •
4. Kubal-Furr will then submit revised proposal addressing comments provided by K. Lucas.

Post Office Box 273210
Tampa, FL 33688-3210
[phone]
FAX [phone]
Mr. Kenneth Lucas
U.S. EPA, Region 4
61 Forsyth Street •
Atlanta, GA 30303
Dear Ken:
Kubal-Furr & Associates
Environmental Management Services
August 14, 2003
Post Office Box 80247
Simpsonville, SC 29680-0247
[phone]
FAX [phone]
As you requested, attached is a hard copy version of the document entitled: "Talking
Points —Proposed Modifications to Ground -Water Treatment System —Celanese Fiber Operations
Site —Shelby, North Carolina" dated August 2003. I've also provided to you via email, a pdf
version of this same document.
The purpose for preparing the Talking Points is to provide both the EPA and the DENR with
sufficient information for a conceptual understanding of current conditions and proposed
modifications to the existing ground -water treatment system. It's not intended to be an exhaustive
presentation of issues at this point but one that will allow further discussion as to the appropriate
mechanism for effectuating this change at the Shelby facility.
Please give me a call after you've had a chance to review this information so that we can discuss
the next steps in the process. I would suggest we try and schedule a conference call with yourself,
DENR and Celanese representatives at your earliest convenience.
Sincerely,
// Via email //
Jerry E. Kubal, P.G.
cc: David Mattison, DENR
Steve Olp, Celanese -Acetate
PEM Carter, Ticona-Shelby
Everett Glover, EarthTech

Talking Points —Proposed Modification to Ground -Water Treatment
System —Celanese Fiber Operations Site —Shelby, North Carolina
Background
Ticona, an operating subsidiary of Celanese, owns and operates a polyester and
engineering plastics production facility in Shelby, North Carolina. CNA Holdings, Inc.
(CNA), also a subsidiary of Celanese, retains management of environmental matters for
the corporation
Operations at the site began in 1960 by Fiber Industries, Inc. (FII). Celanese
Corporation bought the facility from FII in 1983, renaming it Celanese Fiber Operations
(CFO). The site has alternately been known as HNA Holdings, Inc., Hoechst Celanese
Corporation and Celanese Fiber Operations, the latter being the site name as listed on the
NPL.
CNA currently operates a long term remedial action (LTRA) system at the CFO site,
designated as Operable Unit 1 (OU-1) in the Record of Decision (ROD), which consists
of two ground -water extraction and treatment systems identified as the Inner Tier and
Outer Tier systems. OU-1 became operational in August 1989. A second operable unit,
OU-2, consisted of excavation, incineration, stabilization and reburial of treated sludge
and other waste materials.
Current Ground -Water Treatment System
On April 21, 1998, OU-2 and the Outer Tier portion of OU-1 were shut down as part
of a partial delisting petition approved by the EPA effective April 17, 1998. Currently,
active ground -water recovery and treatment is only taking place at the Inner Tier, which
was designed to extract contaminated ground water from around a former source area
remediated as part of OU-2. A process flow diagram showing the current ground -water
treatment system is summarized on Attachment A.
Recovered Inner Tier ground water is treated biologically in a sequencing batch
reactor (SBR) which is capable of processing approximately 2500 gallons of water per
day (gpd). The Inner Tier is currently not capable of sustaining this flow rate due to a
combination of the physical make-up and geochemistry of the aquifer and biofouling of
the well screens. Therefore, even though the Outer Tier is not actively recovering ground
water, Outer Tier well OT-1R is still pumped to provide make-up water for the ground-
water treatment system and for process water used for back -flushing filters.
The remainder of the Outer Tier system is being maintained in "stand-by status" as part
of the delisting agreement and can be brought on line within a relatively short period of
time if it is found necessary to provide hydraulic control along the property boundary.
The Outer Tier wells and treatment system's functionality is assured by performing
routine monthly and quarterly system checks such as energizing the well pumps, transfer
Ticona/Shelby-Talking Points 1 August 2003

pumps and blowers, verifying heater operation, etc. and performing maintenance
activities as required by the routine system checks.
Water extracted from the Inner Tier, including make-up water as required from the
Outer Tier, is first routed to an equalization tank. At this point the pH is adjusted and
coagulant added to facilitate iron removal. Sludge generated during the iron removal
stage and from the SBR are routed to a filter press where they are dewatered into a cake,
tested annually for the full TCLP list of constituents and subsequently disposed of off -site
as a non -hazardous waste.
Following treatment in the SBR, the treated ground water is routed through an air
stripper followed by carbon adsorption. At this point, the treated ground water is routed
to the existing industrial wastewater system where it enters a series of polishing ponds
and ultimately discharges through the plant's NPDES Outfall 001, continuing on to
Buffalo Creek.
Ticona is currently permitted to discharge up to a monthly average of 800,000 gpd of
treated wastewater through Outfall 001 Representative flows used for the recent permit
renewal are 450,000 gpd as dry weather flow and between 600,000 and 750,000 gpd as
wet weather flow. An input of 2500 gpd from the ground -water treatment system
amounts to slightly more than one half of one percent of the dry weather flow.
Proposed Modifications to Ground -Water Treatment System
The proposed modifications would be to discontinue use of the Inner Tier treatment
system as presented on Attachment A and route the Inner Tier water to the headworks of
the plant's industrial wastewater treatment plant (WWTP). The Outer Tier system would
be maintained in standby in accordance with the Partial Delisting Agreement. A process
flow diagram of the system after the proposed changes is shown on Attachment B.
The source of the ground -water impact being addressed is the former disposal of plant
process wastes. This has impacted ground water with generally the same suite of
chemicals that are currently processed in the WWTP. Thus, the existing treatment plant
should be able to satisfactorily treat the recovered ground water and a preliminary
evaluation was performed to estimate the treatment that would be received by a combined
wastewater stream consisting of the industrial wastewater and the extracted ground water
(Attachment C).
The analysis was based on a "projected" wastewater stream constructed by combining
the loadings from the industrial wastewater and the extracted ground water using the
most recent data available (January 2003 thru April 2003). The analysis presented in
Attachment C demonstrates the viability of redirecting the small stream of extracted
ground water from the ground -water treatment unit to the head of the plant's industrial
wastewater treatment plant for treatment. It is our opinion that this modification can be
made without adversely affecting compliance with the NPDES permit.
Ticona/Shelby-Talking Points 2 August 2003

The physical changes needed to divert the combined Inner Tier influent would be to
repipe the collection header to the headworks of the WWTP. If needed, a small booster
pump station would be added to provide the needed lift to the WWTP. The preliminary
evaluation suggests that this will not be needed. In this scenario, the Outer Tier well(s)
that are currently being pumped for make-up water for the Inner Tier treatment system
would be idled, and the entire ground -water treatment system would be taken off line and
placed in a stand-by mode in the event that it was needed in the future.
Preliminary discussions have been held with state water quality personnel to identify
actions needed to address NPDES issues, and it appears that it would be a simple process
to make the change from their perspective. They would need to see some data
(Attachment C) to assure themselves that the change would not result in NPDES
violations, but a permit modification would not be needed. Additionally, an Authorization
to Construct (ATC) would not be needed for a simple piping change.
Ticona/Shelby-Talking Points 3 August 2003

Attachment A.
Current Process Flow Diagram—Ticona, Shelby, NC
Outer Tier
Extraction
Wells
CI m
.5
Q
'
cr.o
2LL
c
Inner Tier
Extraction
Wells
Equalization
Tank
pH adjustment/
Coagulant
addition
Iron Removal
.2500 gpd
Sequencing
Batch Reactor
(Biological)
Air Stripper
Sludge/Non-
hazardous/
Off -site disposal
Flow Varies
(Ticona/KOSA
Process
Waters J
v
0-
0
0
0
Lci
Wastewater
Treatment Plant
(Biological)
2500 gpd
Carbon
Adsorption
Polishing Ponds
A/B/C
NPDES Outfall 001
.0
0
0
0
0
0
0
Buffalo Creek

c
Attachment B.
Proposed Process Flow Diagram—Ticona, Shelby, NC
Outer Tier
Extraction
Wells
I
1
1
1
1
,;7
)
Maintain in 1
1Stand-by Mode /
\
'
If Pumping Resumed
1 Air Stripper
(
Inner Tier
Extraction
Wells
)
ill<
(current avg < 1500 gpd)
Varies
Flow Rate To Be Determined
Ticona/KOSA
Process
Waters
-0
a.
0)
0
0
N.
co
lr
Wastewater
Treatment Plant
(Biological)
Carbon
Adsorption
Polishing Ponds
A/B/C
NPDES Outfall 001
Z
n.
CD
0
0
0
0
v
Buffalo Creek

Attachment C.
Ticona Combined Effluent Evaluation
As a preliminary evaluation of the viability of incorporating the inner tier groundwater into the plant's
industrial wastewater treatment plant, an evaluation was performed to estimate the treatment that would
be received by a combined wastewater stream consisting of the industrial wastewater and the extracted
groundwater. As a basis for this evaluation, relevant data on the two wastewater streams from the period
of January 2003 through April 2003 were reviewed. This timeframe was selected since it represented the
most recently available data. The analysis was based on a "projected" wastewater stream constructed by
combining the loadings from the industrial wastewater and the extracted groundwater during the period of
January 2003 and April 2003. The data are summarized in the following table.
Daily
Wastewater
Influent
Daily
Wastewater
Effluent'
Daily
Groundwater
Daily Combined
Influent
(Projected)
flow (MGD)2
minimum
0.196
0.265
0.0002
0.198
maximum
0.834
0.824
0.003
0.836
average
0.492
0.512
0.0018
0.494
median
0.489
0.529
0.0019
0.491
chemical oxygen
demand (COD)
(lbs/day)2
minimum
1433
111
10
1510
maximum
11,954
744
152
12,069
average
4881
330
86
4965
median
4479
321
86
4581
biochemical
oxygen demand
(BOD5) (lbs/day)
minimum
-
12
-
-
maximum
-
76
-
-
average
-
36
-
-
median
-
35
-
-
— The data on the effluent includes the impacts of discharge of the treated groundwater into the wastewater
polishing ponds after aerobic biological treatment of the groundwater.
2 - The values represent the minimum, maximum, etc. of the daily values for type of information, and are not
directly additive across the rows.
The plant routinely measures BOD5 and COD on the effluent, and COD on the influent. The control
parameter in the plant's NPDES permit is BOD5. In order to estimate the impacts of adding the
groundwater stream to the wastewater stream at the head of the wastewater treatment plant, the data were
evaluated to estimate the overall removal efficiency of the treatment plant, and a relationship between
BOD5 and COD. The overall efficiency was calculated by comparing the influent COD (lb/day) of the
plant's industrial wastewater and the effluent COD (lb/day) of the entire flow that discharges through the
NPDES effluent point. This simplified approach does not account for the residency time of the
wastewater in the system, or the effects of precipitation and storm water on the open wastewater basins;
however, when calculating an average efficiency over approximately 4 months of daily data, it is
expected to give an approximate estimate of the efficiency of the overall wastewater system. This
analysis estimated that the average COD removal efficiency of the treatment system was 92.83%.
Since the control parameter for the NPDES permit is BOD5, the BOD5 and COD data were compared to
determine whether there was a predictable correlation between the two parameters. Comparison of the
Att C/Ticona/Shelby-Talking Points 1 August 2003

Attachment C.
Ticona Combined Effluent Evaluation
data showed appreciable scatter. The BOD5:COD ratio ranged from 0.0472 to 0.2468 with an average
value of 0.1061 and a median value of 0.1. As part of this evaluation, the BOD5:COD ratios were plotted
against the COD to see whether the ratio was a function of COD. This evaluation also resulted in
appreciable data scatter with the best -fit line showing a slightly decreasing trend at higher COD values.
However, this trend was not considered significant enough to warrant using different ratios for different
COD ranges. As a result of the scatter in the data, use of some standard deviation around the mean was
selected as a conservative approach for estimating the BOD5:COD ratio that would be used' for predicting
the impact of adding the groundwater to the head of the wastewater treatment plant. For this analysis, a
value representing the mean plus two standard deviations was selected. This resulted in a BOD5:COD
ratio of 0.1796 being used for analysis of the combined wastewater stream.
The plant's NPDES permit (NC0004952) contains discharge limits for BOD5 of 83 lbs/day as a monthly
average, or 203 lbs/day for a daily maximum. Using a BOD5:COD ratio of 0.1796, this establishes an
allowable COD of 462 lbs/day for a monthly average or 1130 lbs/day for a daily maximum to remain in
compliance with the permit. As shown in the following table, the predicted COD levels are below the
allowable levels to maintain permit compliance applying a conservative conversion factor based on a
value two standard deviations above the mean of the measured data.
Daily
Wastewater
Influent
Daily
Groundwater
Daily
Combined
Influent
(Projected)
Daily
Combined
Effluent
(Projected)
Monthly
Average
Combined
Effluent
(Projected)'
flow (MGD)
minimum
0.196
0.0002
0.198
maximum
0.834
0.003
0.836
average
0.492
0.0018
0.494
median
0.489
0.0019
0.491
chemical oxygen
demand (COD)
(lbs/day)
minimum
1433
10
1510
108
maximum
11,954
152
12,069
865
average
4881
86
4965
356
median
4479
86
4581
328
biochemical
oxygen demand
(BOD5) (lbs/day)
minimum
19
54
maximum
155
80
average
64
64
median
59
61
— Based on an overall average COD removal efficiency of approximately 93% and a BOD5:COD ratio of
0.1796
In summary, the analysis presented in the preceding paragraphs demonstrates the viability of redirecting
the small stream of extracted groundwater from the groundwater treatment unit to the head of the plant's
industrial wastewater treatment plant for treatment. It is our opinion that this modification can be made
without adversely affecting compliance with the NPDES permit.
Att C/Ticona/Shelby-Talking Points 2
August 2003

Figure 3. Proposed Modification to Groundwater
Treatment Process
Groundwater from
Tier I and Teir II
Extraction Wells
Air Stripper
r nular Activated
Carbon
New Pipeine
Sludge Removed t
Ho4dino Tank
and Disposal
New piping to allow groundwater to be
treated in WWTP
Ticona/KOSA
Industrial and Sanitary
Wastewater
10.1 Equalization Basins
Extended Aeration
Basins with Brush
Aerators
Clarifiers
r
Polishing Pond A
Polishing Pond B
Polishing Pond C
7
Pressure Sand
Filtration
NPDES Discharge
Sludge Removed
to Settling Basin
1
Sludge Removal
and Disposal

Q
x
. LEGEND:
E A R T H` T E C III
N
1
0' 100' 200'
SCALE
OT-6 4 0- EXISTING OUTER TIER EXTRACTION WELL
IT-5 . EXISTING INNER TIER EXTRACTION WELL
" DECOMMISSIONED LINE OR STRUCTURE
NEW LINE
ASB AERATION SLUDGE BASIN
EQB EQUALIZATION BASIN
CL CLARIFIER
FLOW DIRECTION
OCT 2003
FIGURE 4
GROUNDWATER TREATMENT
SYSTEM MODIFICATION
KOSA/TICONA FACILITY
SHELBY, NORTH CAROLINA
40064.30