United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 80-SNF-1
February 1981
Air
Synthetic Fiber
Manufacturing
Emission Test Report
E.I. duPont de Nemours
and Company
May Plant
Camden, South Carolina
-------
SOURCE TEST AT DUPONT
MAY PLANT
CAMDEN, SOUTH CAROLINA
\
Contract No. 80-02-3545
Work Assignment 1
Project No. 80-SNF-l
Technical Manager: Winton Kelly
Prepared for:
U.S. Environmental Protection Agency
Emission Standards and Engineering Division
Emission Measurement Branch
Research Triangle Park, North Carolina 27711
TRW
Environmental Engineering Division
Post Office Box 13000
Research Triangle Park, North Carolina 27709
-------
TABLE OF CONTENTS
Section Page
1.0 INTRODUCTION 1
2.0 SUMMARY AND DISCUSSION OF RESULTS 3
3.0 PROCESS DESCRIPTION AND OPERATION 14
4.0 TEST LOCATIONS 16
5.0 SAMPLING AND ANALYTICAL PROCEDURES 28
APPENDICES
A. COMPLETE RESULTS AND CALCULATIONS
B. FIELD DATA SHEETS
C. ANALYTICAL DATA
D. PROCESS INFORMATION
E. SAMPLING PROCEDURES
F. QUALITY CONTROL/QUALITY ASSURANCE
G. TEST LOG
H. PROJECT PARTICIPANTS
-------
1.0 INTRODUCTION
Personnel from TRW Environmental Engineering Division conducted an
emission source test under Contract #68-02-3545 for the Environmental
Protection Agency-Emission Measurement Branch at E. I. DuPont de Nemours
and Company Inc., May Plant, Camden, South Carolina. The emission
source testing was conducted over a three week period during the weeks
of September 22, September 29 and October 6, 1980. All emission testing
under this study was conducted at the ORLON^portion of the DuPont's
May Plant.
This facility was tested in order to gather data for the following
purposes: 1) to provide characterization of process exhaust gas streams
in support of a possible New Source Performance Standard (NSPS) for the
Synthetic Fibers Industry (SNF); 2) to gather data for recommendation of
a preferred test method for the detection and analysis of the dimethyl
formamide (DMF) which is used as a solvent in the dry spinning method of
acrylic fiber production; and 3) to detect the presence of acrylonitrile
in the process exhaust gas streams.
All sampling and analysis performed at the DuPont facility was
conducted by TRW field personnel. Selected samples were returned to
TRW's Research Triangle Park (RTP) laboratory for subsequent and further
analysis by TRW's analytical laboratory personnel. Plant operating data
was gathered by plant personnel under the direction of Mr. C. Reid Earnhart
of DuPont and was transmitted to TRW through personnel for Pacific
Environmental Services Inc. (PES), Durham, North Carolina, also under
contract to EPA for this study. The source test effort was conducted
under the direction of Mr. Winton Kelly - Emission Measurement Branch -
Technical Manager.
-------
Due to matters of confidentiality, the process exhaust gas streams
tested will be identified by code. The key to the code will be contained
within the Emission Standards and Engineering Division's (ESED), EPA
confidential file.
Section 2.0 will summarize and discuss the results of testing by
test location. Section 3.0 will describe the overall process operation.
Section 4.0 will deal with the physical description of the sampling
locations. Section 5.0 will fully describe the various sampling and
analytical techniques utilized to collect and determine the data presented.
Appendix A will contain the complete analytical results. The remainder
of the appendices will contain supportive data and supplemental
information, such as laboratory methods development and field test logs.
-------
2.0 SUMMARY AND DISCUSSION OF RESULTS
A total of nine (9) process exhaust gas streams were sampled and
analyzed by various sampling and analytical techniques during the emission
source test at DuPont's May Plant. These exhaust gas streams were
estimated by the NSS contractor, in cooperation with plant personnel to
contain the highest solvent concentrations. For convenience of testing
and results reporting, these nine groups were divided into three
functional groups and designated as Group I, Group II or Group III test
locations. These functional groups were chosen for reasons of proximity
of one test location to another and to aid in process data collection
along natural breaks in the fiber processing operation. These groupings
were as follows:
• GROUP I - WDX
CRX
• GROUP II - SSX
DMFX
DRY
• GROUP III A - SEI
SCX
• GROUP III B - PSB
WSX
(NOTE: The individual test locations are labeled by three or four
letter codes for reasons of confidentiality.)
At each test location, four sampling and analytical techniques were
utilized to characterize the process gas streams for solvent
concentration. These sampling and analytical techniques included wet
impingement, integrated bag, continuous monitor and silica gel tube
-------
absorption. The continuous monitor was originally set up to be the
primary method of collection and analysis since it could perform
continuously and indicate process variables as they accured. The wet
impingement method and silica gel tube adsorption method were to confirm
the continuous data. Due to unforeseen variables in the emissions such
as high moisture and concentrations, it became apparent that the process
could not be sampled continuously. The condensed water contained almost
all of the DMF and the instruments were not heated. The solution was to
collect the samples in two phases; a condensed or liquid phase and a
non-condensed or gaseous phase. The other methods were run one time
only at each location as confirmatory data. In addition to solvent
concentration, volumetric flow rate and moisture content of the process
gas streams were determined on a daily basis by appropriate sampling
techniques. The presence of acrylonitrile was screened on the same gas
chromatograph column as was used to quantify solvent (DMF) concentration.
A complete description of sampling and analytical techniques may be
found in Section 5.0.
2.1 CONCENTRATION, FLOW RATE AND MASS EMISSIONS
The test results of solvent concentration, flow rate, and solvent
process emissions are summarized in Table 2.1 by test location and test
method. The solvent concentration and the solvent mass emission rates
vary considerably between sampling/analytical techniques at the same
test location. The concentrations and mass emission rate data should not
be taken at face value without considering the advantages and
disadvantages of each method as implemented in the field. Section 2.2
discusses in depth the qualification of the data presented in Table 2.1.
Example calculations used to reduce the sampling and analytical data may
be found in Appendix A.2. The complete results for each sampling and
analytical method may be found in Appendix A.I.
2.2 METHODS: QUALIFICATIONS AND RECOMMENDATIONS
Because of the wide variation in operating conditions (moisture,
temperature, solvent concentration), the range of results and the different
types of sampling methods utilized in the field at the various test
locations, a point by point discussion of the results by test
-------
TABLE 2.1 - SUMMARY OF RESULTS BY TEST LOCATION - DUPONT (MAY PLANT) CAMDEN, S.C.
LOCATION
WDX
CRX
SSX
DMFX
DRX
METHOD
Wet Impingement3
Integrated Bag
Continuous Monitor
Continuous .Monitor
Total0
Silica Gel Tube
Wet Impingement
Integrated Bag
Continuous Monitor3
Silica Gel Tube
Wet Impingement
Integrated Bag
Continuous Monitor?
Conti nuous .Moni tor
Total0
Silica Gel Tube
Condensate return water
Wet Impingement
Integrated Bag
Continuous Monitor?
Continuous .Monitor
Total0
Silica Gel Tube
Wet Impingement
Integrated Bag
Continuous Monitor?
Conti nuous .Moni tor
Total0
Silica Gel Tube
CONCENTRATION
(PPMV as DMF)
—
5.95
5.44
60.4
28.53
44.6
11.93
14.7
36.26
238
3.60
14.7
541
788
1.6 x 101 mg/nU>
170
4.03
12.9
233
121
598
8.90
3.98
1265
388
VOLUMETRIC FLOWRATE
(DSCMM)
487
487
487
487
487
767
767
767
767
12.8
12.8
12.8
12.8
12.8
...
29.5
29.5
29.5
29.5
29.5
276
276
276
276
276
(DSCFM)
17224
17224
17224
17224
17224
27096
27096
27096
27096
453
453
453
453
453
...
1043
1043
1043
1043
1043
9771
9771
9771
9771
9771
EMISSION RATES
(Ib DMF)(kg DMF)
hr
—
1.14
1.05
11.60
5.48
13.5
3.61
4.45
10.98
1.20
.02
.07
2.73
2.80
3.98
2.13
1.98
.05
.15
2.77
2.92
1.41
65.26
.97
.43
137.8
138.23
42.34
hr
—
.52
.48
5.27
5.75
2.49
6.14
1.65
2.02
4.99
.55
.008
.03
1.24
1.27
1.81
.97
.90
.02
.07
1.26
1.33
.64
29.66
.44
.20
67.6
67.8
19.25
(Ib DMF)
Ib product
—
6.21 x 10- 5
5.72 x 10-5
6.32 x 10-4
6.89 x 10-4
2.98 x 10-4
7.35 x 10-4
1.97 x 10-4
2.42 x 10-4
5.98 x 10-4
2.89 x 10-4
4.82 x 10-6
1.78 x 10-5
6.59 x 10-4
6.77 x 10-4
9.59 x 10-4
5.13 x 10-4
4.78 x 10-4
1.20 x 10-5
3.61 x 10-5
6.68 x 10-4
7.04 x 10-4
3.40 x 10-4
1.57 x 10-2
2.37 x 10-4
1.04 x 10-4
3.32 x 10-2
3.33 x 10-'"
1.02 x 10-2
Ldss sample due to freezing.
Gaseous phase seen at FID.
cCondensate liquid by HPLC analysis.
Total is gaseous and condensate of continuous monitoring system.
-------
TABLE 2.1 - SUMMARY OF RESULTS BY TEST LOCATION - DUPONT (MAY PLANT) CAMDEN, S.C.
(Continued)
LOCATION
SCI
sex
PSX
WSX
METHOD
Wet Impingement
Integrated Bag
Continuous Monitor.
Continuous Monitor
Total
Silica Gel Tube
Wet Impingement
Integrated Bag
Continuous Monitor?
Continuous Monitor
Total
Silica Gel Tube
Wet Impingement
Integrated Bag
Continuous Monitor?
Continuous Monitor
Total
Silica Gel Tube
Wet Impingement
Integrated Bag
Continuous Monitor3
Total
Silica Gel Tube
CONCENTRATION
(PPMV as DMF)
600.2
123. 15
138. 3
8073
4753
2.77
6.93
7.81
6.5
<.01
481
120.6
9.22C
1011
351
N.D.d
20.05
9.16d
N.D.
VOLUMETRIC FLOWRATE
(DSCMM)
335
335
335
335
335
337
337
337
337
337
135
135
135
135
135
129
129
129
129
(DSCFM)
11830
11830
11830
11830
11830
11899
11899
11899
11899
11899
4761
4761
4761
4761
4761
4545
4545
4545
4545
EMISSION RATES
(Ib DMF)(kg DMF)
hr
79.4
16.3
18.29
1067
1085.29
628
0.37
0.92
1.04
.86
1.9
25.63
6.43
.49
53.87
54.36
18.70
N.D.
1.02
,47
N.D.
hr
36.1
7.40
8.31
49.5
493.3
286
0.17
0.42
.47
.39
11.65
2.92
.22
24.49
24.71
8.50
N.D.
.46
,21
N.D.
(Ib DMF)
Ib product
5.06 x 10-3
1.04 x 10-3
1.16 x 10-3
6.80 x 10-2
6.91 x 10-z
4.00 x 10-2
2.36 x 10-5
5.86 x 10-5
6.62 x 10-5
5.50 x 10-5
1.21 x 10-«
1.63 x 10-3
4.09 x 10-4
3.12 x 10-5
3.43 x 10-3
3.46 x 10-3
1.19 x 10-3
N.D.
N.D.
N.D.
N.D.
N.D.
Gaseous phase seen at FID.
Condensate liquid by HPLC analysis.
GTotal is gaseous and condensate of continuous monitoring system.
Not determined.
-------
location is required. In general, the integrated bag method results may
be disregarded at all of the test locations at the DuPont (May Plant)
because of sample collection problems. More precisely, the integrated
bag collected the sample into two phases, a liquid phase and a gas
phase. This was due to condensation in the bag. The analysis of the
collected sample by GC/FID accounted for only the gaseous component of
the sample. Due to the high solubility of the dimethyl formamide (DMF)
in water, any condensate collected had the effect of scrubbing the
solvent (DMF) from the gaseous phase. The liquid phase (the condensate)
remained in the bag adhering to the bag walls as droplets. Therefore,
the condensate was not injected into the gas chromatograph, and
consequently the analytical results are considered to be low and invalid.
2.2.1 WDX
At the WDX test location, the test results showed low moisture
(2.8%), ambient temperature (85°F) and low solvent concentration. Four
sampling and analytical methods were utilized at this test location (see
Table 2.1 for results). No analytical results were determined for the
wet impingement method, at this location, due to sample preservation
problems, which precluded any analysis and as previously mentioned the
GC/FID data was invalid.
The average range of the two remaining methods was 9.06 Ib/hr with
a range of 5.48 Ib/hr to 12.6 Ib/hr. The wet impingement method (which
was lacking) may have validated and confirmed the other two methods.
The operating conditions encountered: low moisture, low concentration
and ambient temperature allow, in TRW's opinion, for the application of
all proposed test methods. The continuous monitoring data was collected
over three days while the silica gel tube was only collected for a
one-hour period. See Appendix A for complete results.
2.2.2 CRX
The CRX test location was similar to the WDX test location in
operating conditions. These operating conditions were low moisture
(3.9%), low concentration and ambient temperature (80°F). All four
sampling and analytical methods were implemented (see Table 2.1).
The integrated bag method should be disregarded according to the
previously discussed rationale.
-------
The continuous monitor method may be disregarded due to a negligible
condensate fraction sample size. The liquid fraction collected in the
sample line and condensate jar prior to the flame ionization analyzer
was judged to be negligible at this location in the field and was not
collected. However, because of the solubility of the solvent in HpO,
the contribution to the total solvent collection of the continuous
monitor system may have been significant. Without an analysis of the
condensate and rinse, the results of the continuous monitor system
should be suspect. They represent only the gaseous phase as the
integrated bag does.
The comparison between the wet impingement and NIOSH-silica gel
tube method was favorable. The average emission rate determined from
the two methods ranged from 0.98 Ib/hr to 13.5 Ib/hr. On the bassis of
limited data collected in the field, either method, the wet impingement
and/or the silica gel tube can be considered equally valid.
2.2.3 SSX
The operating conditions encountered at the SSX test location posed
both sampling and analytical problems. The operating conditions present
were high moisture (17% saturation @135°F), suspected high concentration
and temperatures in the 125-150°F range. The liquid-gas phase problem
was heightened by the saturated stack conditions.
The integrated bag method can be disregarded by the previously
indicated rationale. The average emission rate for the test location
based upon the three other methods was 2.64 Ib/hr, with a range from
1.20 to 3.98 Ib/hr. The three test methods yield results within the
same range, but recommendation of one method over another would be
tenuous at this time.
The saturated moisture conditions and the stack configuration lead
to the sampling difficulty of entrained moisture. The conditions were
such that the water vapor would condense within the stack forming water
droplets (see Figure 4.3). The orientation of the sample line/sample
probe within the stack during sampling would consequently bias the
amount of solvent collected given the high solubility of DMF. Evidence
indicates that entrained moisture was, in fact, encountered. During the
second moisture train sampling run, moisture in excess of saturation was
collected (Run 0SSX-WI-2; see Appendix B).
8
-------
Assuming entrained moisture was encountered with all the methods
tested, the wet impingement train would appear to be the best alternative
since the sampling media is water. The results from the SSX test location
by wet impingement are lower than the other two methods. If the
condensate return water results are added to the wet impingement results,
the resultant value is roughly equivalent to the average of the continuous
monitor and silica gel tube methods. This may indicate that entrained
moisture was not a problem with the wet impingement method.
2.2.4 DMFX
The operating conditions during the testing at the DMFX location
were low moisture (4.130, moderate temperature (110°), and moderate
concentrations.
The integrated bag method can be disregarded according to the
previously discussed rationale.
The continuous monitor method yielded a slightly higher emission
rate than either the wet impingement or silica gel tube method. The
average emission rate for the three methods was 2.1 Ibs/hr with a range
from 1.41 Ibs/hr to 2.92 Ibs/hr.
All three sampling and analytical methods compared favorably based
on the limited amount of field data.
2.2.5 DRX
The operating conditions during testing at DRX location were high
moisture (15.4%), high concentration and high temperature (144°F). No
sampling problems were encountered due to entrained moisture because the
moisture content of the stack gas was below the saturation point at a
stack temperature of approximately 144°F.
The integrated bag method can be disregarded according to the
previously discussed rationale.
The wet impingement and silica gel tube methods compare favorably.
Their calculated emission rates average 53 Ibs/hr with a range of
42.3 Ibs/hr for the silica gel tube method and 65.3 Ibs/hr for the wet
impingement method.
The continuous monitor method yielded a considerably higher emission
rate. The calculated emission rate was 138 Ibs/hr of solvent as determined
by the continuous monitor method. This value appears to be biased high.
-------
The emission rate in terms of pounds of solvent per pound of product
-2
(fiber), 3.32 10 was greater than the loss incured in this portion of
the process calculated from the plant supplied process data of residual
solvent content in the product (fiber). Assuming that the only avenue
for solvent to escape are air and water emissions, the air emission rate
-2
should not have exceeded 1.3 x 10 Ibs of solvent per pound of product
(fiber) produced. This rationale assumes that the plant analytical
method used to measure the residual solvent content was valid and the
fiber used for analysis was representative.
2.2.6 SCI
The operating conditions during sampling at the SCI test location
were low moisture (3.3%), moderate temperature and high concentration.
The high concentration coupled with the sample location proved to be
difficult to adequately assess the data from the various methods.
The integrated bag method can be disregarded according to the
previously mentioned rationale.
The other three methods showed a range from 79.4 Ibs/hr for the wet
impingement method to 1,085 Ibs/hr for the combined results of continuous
monitor method. The average rate for the three methods was 597 Ibs/hr.
This range was judged to be excessive and based upon the limited amount
of data points, selection of one method over another is not possible at
this time.
A possible rationale for the disparity of results, is the sample
location. It is the opinion of TRW that the sample location selected
was possibly unrepresentative of inlet flue gas. The sample was taken
from a condensate relief valve at a dead end header. It was supposed
that the pressure would be sufficient to gain continuous flow without
interferring condensation. However, due to the low flow through the
system condensate was present at the port. A small amount of condensate
would greatly influence (bias) the sampling and analytical results.
Secondly, the port configuration (see Figure 4.6) may have biased one
method over another during the simultaneous methods testing. That
method which was aligned with the downward facing leg of the union was
observed to collect a substantially greater amount of condensate.
Because of the high solvent concentration of condensate at this location,
10
-------
the orientation of the sample train in relation to the sample port would
have biased one method high. The orientations of the sample trains
during the comparative test run were not recorded. Because of the wide
variation in emission results by the various analytical methods for this
sample location, the plant attempted to verify the solvent mass balance
data collected during the test period in order to determine an inlet
concentration. This data has been included in the ESED confidential
files and has been evaluated separately.
2.2.7 SCX
The operating conditions at the SCX test location were low moisture
(2.8%), moderate temperature, and low concentration. The integrated bag
method can be disregarded according to the previously mentioned rationale.
The silica ge tube method showed less than,.l ppm DMF.
The other two methods, the continuous monitor method and the wet
impingement method, averaged 1.14 Ibs/hr. The range was .37 Ibs/hr for
the wet impingement method to 1.9 Ibs/hr for the continuous monitoring
method.
All the methods utilized at this location resulted in lower than
expected emission rates. The expected solvent concentration according
to plant supplied information was approximately 20 ppmv. All methods
performed during the sampling period were lower than this value. The
sampling system may have lead to a result biased low. The sample for
all four methods was extracted from the stack through a 8-foot length of
stainless steel piping connected to 75 feet of heat traced teflon
sampling line (see Figure 4.6). The possibility existed that the solvent
condensed in the stainless steel pipe before entering the heat traced
line. Due to problems of access, this situation could not be confirmed
or rejected.
2.2.8 PSX
The operating conditions during the testing at PSX test location
were low moisture (2.4%), ambient temperature (85°F) and moderate solvent
concentration.
Plant supplied data. Appendix D.
11
-------
The integrated bag method results can be disregarded according to
the previously mentioned rationale.
The average emission rate for the other three methods was 41.2 Ibs/hr
with a range between methods from 18.7 Ibs/hr to 53.9 Ibs/hr.
The wet impingement method and the silica gel methods were comparable
based upon the one sampling run performed. The continuous monitor
method reported higher results than either the wet impingement or silica
gel tube method. These results may be biased high due to the estimated
condensate collection time and estimated condensate sample volumes. In
addition, the condensate reported as collected at the PSX test location
includes condensate collected at WSX test location and condensate
collected during extensive instrument (FID) line-out. Due to the proximity
of results for the silica gel tube method and the wet impingement method
and suspect operation of the continuous monitor at this location, an
estimated emission rate of 22.2 Ibs/hr is the best approximation.
2.2.9 WSX
The operating conditions at the WSX test location were low moisture
(1.2%), ambient temperature (85°F) and suspected low solvent concentration.
All the samples at this location were not collected, as this test
location was intended to be sampled on a "grab basis," time permitting.
Consequently, only the continuous monitor method and the integrated bag
method were utilized. The results listed in Table 2.1 are very limited
and should not be considered representative of the process because
sufficient testing and process data is lacking.
2.3 PRESENCE OF ACRYLONITRILE (AN)
The presence of acrylonitrile was determined by gas chromatograph
analysis in the field. In general no acrylonitrile was found to be
present. The only exception to this was at the inlet to the recovery
unit. At this location, peaks were determined to be acrylonitrile and
quantified in the 0.5 to 1.5 ppm range. This result confirms expectation
that if AN were present, it would occur at the location with the highest
DMF concentrations and then only in low amounts (for complete results,
see Appendix C.I - Field Chromatograms #276-278).
12
-------
2.4 MASS BALANCE CONSIDERATIONS
In general, the exhaust gas streams were low in concentration. It
is suspected that due to the solubility of DMF in water, the mass emission
rate attributable to the water portion of the process streams was equal
to or greater than the gaseous portion of the process exhaust stream.
The concentration of DMF in the process water streams, recovered or
exhausted, was measured routinely by plant personnel at several points.
The routine plant analysis of liquid process streams was not
duplicated by TRW. The historical accumulation of plant water analysis
data was deemed sufficient for purposes of the study and was considered
to be more representative than are grab samples that could have been
collected during the field test study. If a mass balance of solvent is
desired, the historical plant data of solvent concentration of the
process water feed and exhaust streams, and the gaseous exhaust streams
(measured by TRW during the field test) should be sufficient to approximate
an adequate mass balance around the acrylic fiber spinning operation.
13
-------
3.0 PROCESS DESCRIPTION AND OPERATION
The DuPont May Plant utilized the dry spinning method to produce
acrylic fiber. The suspension polymerization technique was used to
produce polyacrylonitrile used to dry spin acrylic fiber. The solvent
used to spin the fiber was dimethyl formamide (DMF). The exhaust gas
streams sampled were in the fiber spinning and fiber treating sections
of the production process.
A generalized flow diagram is outlined in Figure 3.1. Test locations
are indicated by their respective codes and approximate locations along
the fiber processing line.
A more complete diagram of the dry spinning process may be found in
the Source Category Survey Report, Phase I NSPS, February 14, 1980,
prepared by Pacific Environmental Services (PES).
14
-------
sex
Rec.
PSX
3*-SCI
CRX
WDX
SSX .
DMFX-
DRX •
Polymer
Preparation
Solution
Preparation
Spinning
Washing/
Drawing
Finishing
Polymer
Recovery
USX
Solvent
Recovery
Product
Figure 3.1 General Process Flow Diagram.
15
-------
4.0 TEST LOCATIONS
This section discusses the physical arrangement of the various test
locations that were sampled during the solvent emission study at DuPont.
A generalized schematic of each test location is provided as reference.
Pertinent information bearing on the reliability of the test data as
collected is provided.
4.1 WDX
The WDX test location was located on the rooftop of the main fiber
spinning building. The exhaust serving this test location measured
32 inches by 32 inches. Five holes for test ports were drilled into the
duct by plant personnel to accommodate an S-type pi tot tube for flue gas
velocity measurement and volumetric flowrate determination. A 5 x 4
sampling matrix was laid out for the velocity measurement according to
Federal Register specifications - EPA Method 1. The test ports were
located four equivalent diameters (Dg) upstream from the nearest flow
disturbance. A single hole located 12 inches below those ports used for
the velocity traverse was used for the continuous monitor sample line.
The gas samples taken for method comparison were withdrawn from the
stack at the point of average velocity through tubing temporarily mounted
on the stack. Approximately 50 feet of unheated 1/4" tubing connected
the test location to the continuous monitor station (Figure 4.1).
4.2 CRX
The CRX test location was located on the rooftop of main fiber
spinning building. The exhaust duct from this process measured 42 inches
by 58 inches. Five holes for test ports were drilled into the duct by
plant personnel to accommodate an S-type pi tot tube for flue gas velocity
measurement and volumetric flowrate determination. A 5 x 5 sampling
16
-------
1
2
3
4
5
Traverse Points
32"
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
.
•
Point Distance(a), in.
3.2
9.6
16.0
22.4
28.8
Distance(b), in.
4.0
12.0
20.0
28.0
18
Velocity and
Sampling Ports
Flow
Disturbance
Figure 4.1 WOX Exhaust Stack.
32".
O
o
20'
Continuous
Monitoring
Port
Roof
\
-------
matrix was laid out for velocity measurement according to Federal Register
specification - EPA Method 1. The test ports were located two equivalent
diameters (D ) from the nearest downstream flow disturbance. One of
these five ports was used for the comparative methods testing. An
attempt was made to locate the sampling probe line at the point of
average velocity for each of the test methods utilized. Ports not in
use were closed with duct tape to prevent dilution of the exhaust gas
stream. Approximately 50 feet of unheated 1/4" tubing connected the
test location to continuous monitor station (Figure 4.2).
4.3 SSX
The SSX test location was located on the rooftop of the fiber
finishing building. The two test ports were located 15 inches downstream
from he nearest flow disturbance on a straight run of circular duct.
The test ports were located 90° apart from each other. Six points were
measured on each of two velocity traverses according to Federal Register -
EPA Method 1 specifications. All samples were withdrawn from either of
the two sample ports through 1/4 inch sampling line. The sample for
continuous monitor station was drawn through a series of four standard
size impingers and approximately 35 feet of unheated sample line. The
SSX exhaust duct was capped with an inverted cone which served to knock
out moisture from the exiting flue gas. A rim around the top of the
duct collected the condensed moisture, which was then routed into a
return line on the outside of stack. The return line lead to the
process sewer (Figure 4.3).
4.4 DMFX
The DMFX test location was located on the rooftop of the fiber
finishing building. The exhaust duct was 102 inches in height and
10 inches x 12 inches in width. A 4 x 3 sampling matrix was used for
the velocity measurement and volumetric flowrate determination according
to Federal Register - EPA Method 1 specifications.
The four ports used for velocity measurement were located two
equivalent diameters (D ) downstream and six equivalent diameters (Dg)
upstream from the nearest flow disturbances. All samples were collected
from one of these four velocity ports. The continuous monitor station
18
-------
Traverse Points
1
42"
i
Point
1
2
3
4
5
•
•
•
•
•
•
a
D1stance(a), 1
5.8
17.2
28.8
40.2
51.8
•
•
•
•
LJ
b
18'
n. D1stance(b), 1n.
4.2
12.
21.
29.
37.
6
0
4
8
Velocity and '
Sampling Ports J
28'
Flow '
1
58" _
O
o o o o o
1
i
A. i
IV
o
Exhaust
V Duct
k
20'
Roof
\
Disturbance " \ s\±
Figure 4.2 CRX Exhaust Stack.
-------
Traverse Points
Point Distance(d). in.
1
2
.3
4
5
6
0.5
1.75
3.5
8.5
10.25
11.5
Electrical
Sampling and
Velocity Ports
Motor
Fan
Condensate Return Line
Platform
Roof
Figure 4.3 SSX Exhaust Stack.
-------
was connected to the DMFX exhaust through approximately 30 feet of 1/4"
tubing. A condenser was used in line to collect condensate prior to and
after the continuous monitor pump (Figure 4.4).
4.5 DRX
The DRX test location was located on the rooftop of the fiber
finishing building. The stack extended 16 feet above rooftop level.
The stack was constructed out of stainless steel and was equipped with
orifices to measure volumetric flowrate on either side of a venturi.
The pressure sensed by the installed orifices was transmitted through a
differential pressure cell (D ) to the control room. The volumetric
flowrate was recorded continually on a stripchart in the control room.
The volumetric flowrate recorded by plant equipment was deemed sufficient
for purposes of the study and therefore no manual velocity measurements
were conducted at this location by TRW personnel. The flowrates reported
by the plant were corrcted for moisture to a dry basis for this report.
A copy of the stripchart measuring volumetric flowrate is included in
Appendix D which is maintained in ESED confidential files.
The four ports used for moisture determination and solvent sampling
were located mideway around the stack on four adjoining sides of the
duct. These ports were two equivalent diameters (D ) from the nearest
downstream flow disturbance. The appropriate sample lines were connected
to the stack with brass or stainless steel fittings. No probe was
extending into the stack. Each of the four comparative methods were run
simultaneously, but out of different ports located around the perimeter
of the stack. The continuous monitor station was connected to the DRX
exhaust stack through approximately 50 feet of 1/4" Teflon ^tubing
(Figure 4.5).
4.6 SCI
The SCI test location was located beneath the recovery unit platform.
Samples were extracted from the condensate relief drain valve, located
at the end of the 24 inch header main, leading to the base of the recovery
unit. The relief valve was outfitted with a stainless steel four way
union fitting which served as a manifold for the simultaneous solvent
sampling. The port was under positive pressure. A condenser was used
on line to collect condensate between the sample port and the continuous
21
-------
12"
22"
Sampling and
Velocity Ports
10"
Traverse Points
Point Distance (a), in. Distance (b). In.
1
2
3
4
1.5
4.5
7.5
10.5
1.4
5.0
8.3
54"
- 10"
Exhaust
Duct
O
Roof
8.5'
Motor
Platform
Figure 4.4 DMF Exhaust Stack.
-------
ro
co
Air Vants,
24" _
45"
PUtfora
Roof
16'
Fan
Front V1aw
Side View
Figure 4.5 DRX Exhaust Stack.
-------
monitor station. The continuous monitor station was connected to the
sample port through approximately 25 feet of tubing (Figure 4.6).
Volumetric flowrate was determined by plant monitors and recorded on a
circular stripchart recorder. Copies of the stripcharts are in Appendix 0
which is maintained in ESED confidential files. The plant reported
volumetric flowrates were corrected for moisture, to a dry basis for
this report. The location of the pressure measuring devices used to
calculate the volumetric flowrate were less than optimum (see Figure
4.6). The accuracy of the plant supplied volumetric flowrate was not
determined. The reported volumetric flowrate was in the normal range
and was comparable to historical operational data.
4.7 SCX
The SCX test location was located at the base of the recovery unit
platform. Gas samples were extracted from the SCX exhaust stack at
approximately the 75 foot level of the unit through an eight-foot section
of stainless steel piping. The host plant installed a valve and tee
assembly so that a heated sample line could be connected closer to the
source than the previously used sampling system had allowed. Direct
access to the stack was not available as no platform existed on the
tower. The heated sample line was anticipated to be sufficient for the
collection of a representative gas sample. The eight-foot section of
uninsulated pipe connecting the stack and the heated sample line was a
possible depository for condensation. Tests to judge the adequacy of
the sample handling system were not possible and the heated sample line
as used was the most practical means of conducting the gas sample to the
continuous monitor station.
All samples (moisture and solvent concentration tests) were collected
from the exit of the heated sample line at a stainless steel four-way
union fitting, which was used as a sampling manifold. A condenser was
installed in line after the sampling manifold and before the continuous
monitor pump to collect accumulated condensation.
4.8 PSX AND WSX
The PSX and WSX test locations were located on the rooftop of the
main fiber production building. The PSX and WSX exhaust streams merged
into a common stack before exiting to the atmosphere through a 45-foot
24
-------
PO
in
Sample Manifold Fitting
Located at
Points A and B
SCX Sample Port
Heated Sample Line
~75'
^-\-
Underground
Exhaust Ducts
Recovery
Unit
SCI Sample Port
Monitoring
Station
Figure 4.6 SCI and SCX Exhaust Diagram.
-------
tall stack. Each exhaust stream was sampled prior to entering the
common exhaust stack. The WSX test location was of secondary importance
to the emissions study and was sampled on a time available basis.
Therefore, comparative testing and extensive continuous monitoring was
not performed. Accurate flowrates were determined.
The sample ports for the PSX test location were in a 13 foot run of
straight duct, measuring 24 inches wide by 28 inches deep. The ports
were located 5.6 equivalent diameters (D ) downstream from the nearest
flow disturbance and 4.6 equivalent diameters upstream from the nearest
flow disturbance. A 5 x 5 sample point matrix was used for flue gas
velocity measurement and volumetric flowrate determination. The same
ports were used for solvent, concentration measurements (Figure 4.7).
The sample ports for the WSX test location were located in a 30-foot
straight run of circular overhead duct work measuring 24 inches in
diameter. Two sample ports oriented 90° apart from each other were
located 10 equivalent diameters (D ) downstream from the nearest flow
disturbance and 2.5 equivalent diameters (D ) from the nearest upstream
flow disturbance. A six point velocity traverse was utilized at each
WSX port according to Federal Reigster - EPA Method 1 specifications for
velocity measurement and voplumetric flowrate determination. Integrated
bag samples and moisture train measurements were withdrawn from the
exhaust stream through the same ports (Figure 4.7).
26
-------
Exhaust Stack
ro
PSX Traverse Points
24"
WSX Traverse Points
24"
WSX
Exhaust
Duct
Roof
Figure 4.7 WSX and PSX Test Locations.
-------
5.0 SAMPLING AND ANALYTICAL PROCEDURES
This section will discuss the sampling and analytical procedures
utilized in the field and the laboratory under Task 1, Project 80-SNF-l
source test at DuPont May Plant, Camden, South Carolina. Sampling
methods and analytical procedures will be discussed separately.
5.1 SAMPLING
A total of six field sampling methods were performed in the field
during the DuPont May Plant source test. Several of the sampling
methods were implemented according to standard procedures, hence only a
brief discussion will be included herein. A few other sampling methods
were modifications of standard methods, unique to this field test and
consequently will be discussed in more elaborate detail.
5.1.1 Velocity Measurement and Volumetric Flowrate Determination
At every test location sampled at DuPont, a velocity measurement
for a volumetric flowrate determination was either made directly or
recorded indirectly by plant process control equipment. In the former
case, the measurement was sited at a location according to EPA Reference
Method 1 - Sample and Velocity Traverse for Stationary Sources, Federal
Register, Volume 42, No. 160, August 18, 1977. Specifications of
individual test locations including velocity traverse matrices,
equivalent duct diameters, and dimensions are illustrated in
Section 4.0 - Test Locations. The execution of the velocity measurement
and flowrate determination was performed in accordance with EPA
Reference Method 2 - Determination of Stack Gas Velocity and Volumetric
Flowrate (Type-S pitot) Federal Register, Volume 42, No. 160, August 18,
1977. The average gas velocity in the stack or duct was determined from
the gas density (see 5.1.2 and 5.1.5.1) and from the measurement of the
28
-------
average velocity head with a S-type pitot tube. The plant flowrate
determination was recorded by plant process control equipment operating
on a pressure transducing principle. The plant equipment had been
previously calibrated against a S-type pitot tube, hence the flowrate
data should be considered valid and adequate for the purposes of this
study.
5.1.2 Integrated Bag - Evacuated Can Technique
Gaseous samples were collected at each test location for solvent
concentration and molecular weight determination. A TRW modified
integrated gas sampling train was utHized. The integrated gas sampling
train used an evacuated can - Tedlar^bag system to collect the gas
sample. The evacuated can method was adapted from EPA Reference
Method 3 - Gas Analysis for Carbon Dioxide, Oxygen, Excess Air and Dry
Molecular Weight, Federal Register, Volume 42, No. 160, August 18, 1977,
and proposed EPA Method 110 - Determination of Benzene Emissions from
Stationary Sources.
The evacuated can method was used to collect a given quantity of
sample into a bag. This method uses the negative pressure from an
evacuated can connected to a sample bag can as the mechanism for
obtaining a controllable sample flow. A diapham pump is used to evacuate
the can, which is equipped with two self-sealing quick-disconnect valves,
to 29" Hg. A leak check of the sample train is performed by connecting
a vacuum gauge to one of the quick-disconnect valves. If the pressure
does not drop more than 1" Hg in 30 minutes, the can system is considered
to be leak-free. The sampling bags were checked for leaks before and
after each sampling run by filling the bags with nitrogen (N2) and
allowing the bags to remain overnight. If no leaks were observed and
chromatographic analysis proved negative (blank - see 5.2), the sample
bag was placed in the sample can and readied. The sampling train was
assembled as in Figure 5.1. The sample flow from the source was
initiated by opening the needle valve between the two cans. The sample
flow was monitored with a flow meter, recorded and adjusted periodically
during the sample run (Appendix C.2 - Field Data Sheets). The sample
flow will remain constant until the evacuated can reaches a low vacuum
level. A one-hour integrated gas sample was taken at a flowrate of two
29
-------
CO
o
Teflon"
Probe
Swage! ok
Bulkhead
Tedlar®
Sampling
Bag
Clear Plexiglas
Lid i. Flow Meter
Vacuum Gauge
.Quick
Disconnect:
Evacuated Storage
Cans
Figure 5.1 EVACUATED CAN SAMPLING SYSTEM
-------
cubic feet per hour. When the appropriate test time was completed, the
valve was closed between the cans and the bag immediately capped off and
removed from the can. The integrated bag was appropriately labeled and
transported immediately by the sampler to the laboratory for analysis.
5.1.3 Continuous Hydrocarbon Monitors (FID)
Solvent cncentration was measured on a continuous basis at each
test location. Three separate continuous hydrocarbon monitors operating
on the detection principle of flame ionization (FID) were used. Two
Beckman Model 400 analyzers and a Beckman 402 analyzer were used. A
potentiometric stripchart recorder was used to obtain a permanent record
of the solvent concentration. In addition to the stripcharts, continuous
monitoring field data sheets were maintained by field personnel during
test runs. These continuous monitoring field data sheets are in
Appendix B.4. The stripcharts are on file in the TRW project file
80-SNF-1D. The hydrocarbon analyzers were operated according to the
manufacturer's operating instructions and the draft EPA Method
25A - Determination of Total Gaseous Organic Concentration using a Flame
Ionization Analyzer. The analyzers were calibrated based upon propane
(C~Hg) standards of certified concentrations. Support gases for fuel,
combustion air and zero gas were of equivalent grade or better than
those specified in the referenced protocol. The continuous monitor
stations, are indicated in Figures 5-2, 5-3, 5-4 and 5-5. The essential
components of each station were the analyzers, a coated diaphragm pump,
sample line, support calibration gases, bypass flowmeters, and moisture
condensers. Pertinent information pertaining to the operation of the
continuous monitors is listed in Table D (Appendix A.I).
5.1.4 NIOSH Silica Gel Tube Method #S-255
A gas sample for method comparison was collected at each test
location by the recommended NIOSH sample method. The gas sample was
drawn simultaneously with the other gas sampling methods. A 1/2 to
1 hour sample was withdrawn from the appropriate flue gas stream into a
glass tube 7 cm long with a 6-mm O.D. and a 4-mm I.D., containing two
31
-------
CRX
oo
ro
Distance To CRX Approx. 50 Ft.
Teflon"
Diaphragm
Pump
Condenser
Ice Bath
Distance To WDX Approx. 50 Ft.
Condenser
Ice Bath
Teflon®
Diaphragm
Pump
WDX
Figure 5.2 CONTINUOUS MONITORING STATION
SCHEMATIC (GROUP I LOCATION)
-------
KEY
1. Pump
2. Condenser with
Ice Bath
Teflon
Sample
Line
Method 5 Impingers
Distance To DRX Approx. 50 Ft.
Distance to SSX Approx. 35 Ft.
Distance To DMFX Approx. 30 Fl.
Dry Gas
Meter
Figure 5.3
CONTINUOUS MONITOR STATION SCHEMATIC
(GROUP II LOCATIONS)
33
-------
OJ
Condenser
In Ice Bath
SCX
V
Heated
Sample
Line
Teflon
Line
-. T
V/W Teflon Diaphragm Pump
Beckman
400
FID
Flow Meter
< SCI
Distanc To SCI Approx. 25 Ft.
w
Teflon Diaphragm Pump
Beckman
400
FID
Flow Meter
Dry Gas Meter
Figure 5.4 CONTINUOUS MONITOR STATION
SCHEMATIC (GROUP III A LOCATIONS)
-------
Distance To WSX System Port Approx. 20 Ft.
Distanc To PSX System Port Approx. 35 Ft.
Teflon Diaphragm
Pump
CO
en
Beckman
402
FID
Flow
Meter
1)
2)
WSX Sample Port
PSX Sample Port
Figure 5.5
CONTINUOUS MONITOR STATION
SCHEMATIC (GROUP III B LOCATION)
-------
sections of 20/40 mesh silica gel separated by a 2-mm portion or urethane
foam. The absorbing section contained approximately 150 mg of silica
gel, the backup section, approximately 75 mg. A 3-mm portion of urethane
foam was placed between the outlet end of the tube and the backup section.
A plug of glass wool was placed in front of the absorbing section. A
calibrated personnel sampling pump (SIPIN Model SP-15) collected the gas
sample at an approximate f'lowrate of 250 milliliters per min (ml/min).
Calibration data and field data sheets are in Appendix B.4.
5.1.5 Moisture Determination
5.1.5.1 Moisture Determination-Conventional
The moisture of the process exhaust gas streams was collected
during each day of testing in order to determine dry molecular weight
and volumetric flowrate on a dry standard basis. EPA Reference
Method 4 - Determination Content in Stack Gases, Federal Register,
Volume 42, No. 160, August 18, 1977 was the method utilized. A gas
sample is extracted at a constant rate from the source, moisture is
removed from the gas stream by condensation through an ice bath, and the
moisture determined either volumetrically or gravimetrically. Complete
results are given in Appendix A.I, Table A. Moisture field data sheets
are contained in Appendix E5.5.
5.1.5.2 Experimental Wet Impingement Sample Train
During the preliminary development phase of the synthetic fibers
emissions testing program, the high solubility of dimethyl formamide
(DMF) was noted. Consequently, it was proposed to trap DMF in impinger
solutions of high purity water. A standard Method 4 moisture train was
selected. Slight modifications were necessary. The modifications
included addition of a fifth midget impinger, no stopcock grease, a tube
sample probe, and cleaning of glassware with methylene chloride. The
principle of operation was similar to that of the conventional moisture
train. Impingers 1, 2, and 3 contained 15 ml of distilled H20. The
fourth impinger was empty and the fifth impinger held 15 grams of non-
indicating ACS reagent grade silica gel. Recovery of the impingers
solution was performed with distilled water. The flowrate of the sample
train was 8 to 10 standard cubic feet per hour (scfh). The sampling
time coincided with the other comparative methods, which ranged from
36
-------
30 to 60 minutes depending on the expected solvent concentration. Field
data sheets recording the sample train operation are in Appendix B.5.
5.2 ANALYTICAL PROCEDURES
Several analytical techniques were used in the field and laboratory.
They included gas analysis for dimethyl formamide by gas chromatograph/
flame ionization detection (GC/FID), liquid analysis for dimethyl
formamide by high performance liquid chromatography (HPLC) and silica
gel analysis for dimethyl formamide by GC/FID.
5.2.1 Field Analysis (Shimadzu)
The integrated bag samples were analyzed in the field by gas
chromatography/flame ionization detector (GC/FID) for DMF concentrations
and gas chromatography/thermal conductivity detector (GC/TCD) for
stationary gas concentrations. A Shimadzu Mini 2 equipped with a
6' x 1/8" Teflon column packed with 0.2% Carbowax 20M on Carbopack
80/100 mesh was used to determine the DMF concentrations in the
integrated bag. The GC conditions were as follows:
• Column temperature - 96° isothermal
t Injection/detector temperature - 220°C
• Flowrates: Air - 405.4 ml/min
H2 - 31.25 ml/min
He - 38.96 ml/min
• Pressures: Air - 1.1 kg/cm2
H2 - 1.1 kg/cm2
He - 4.0 kg/cm2
The chromatographic results were recorded and electronically
D
integrated with a Shimazdu CR1A chromatopak. The chromatopak
operating parameters used during the field analysis were:
• Width - 2
• Slope - adjusted automatically
• Drift - 0
• Minimum peak area - 100
• T-DBL (parameter alteration time) - 1
t Lock - 0
• Stop time - 30 min.
37
-------
• Attenuation - 2
• Speed - 10
• Spl wt (sample weight) - 100
• Is-wt (weight of internal standard) - 1
• Range - 10
The complete field chromatographic record is in Appendix C.I for
inspection. Figure 5.6 is an example DMF chromatogram from the field
analysis. Gas samples were injected into a six-port gas sampling valve,
with ports 3 and 4 as sample loop and ports 5 and 6 connected to the
column and the detector through nickel tubing. Teflon ^tubing was used
for the sample loop. Stainless steel attachments were restricted as
much as possible so as to minimize adsorption of DMF onto metal surfaces.
The gas sampling valve remained in the load position one minute and
20 seconds before each injection to give the sample ample time to flush
through the loop. A vacuum pump was used to pull the gas from the
Tedlar^bags through the sample loop. The vacuum pump was not needed
for the standards because they were in pressurized cylinders. After
each sample, the sample loop was blanked by flushing ambient air through
the loop with the vacuum pump.
The external standard method was used for calibration. In this
method, a standard was run and the response of the detector per unit of
the standard was determined by the following:
Calibration factor = Concentration of amount of standard
response or area count of standard
Then the standard was run again as an unknown to obtain the
percentage of the unknqwn to the known. After the standards, the samples
were run again and the amount of the unknown in the sample was calculated:
Amount of unknown = Calibration factor x response of unknown
The standards were run before and after every fourth sample. The
standards were purchased from an outside vendor as a quality control
procedure. However, the vendor could not certify the standards prior to
shipment for use in the field. The calibration gases were, therefore,
used as approximate values and the concentration electronically recorded
on the chromatogram are only approximations. A post-field certification
38
-------
G.C. Conditions
Column - 6' x 1/8" teflon
packed with .2% carbowax 20m
on carbopack C, 80/100 mesh.
Column Temp. - 96°C
Inj./Dect. Temp. - 220°C
Flow Rate
Air - 405.4 ml/min.
He - 38.96 ml/min.
\\2 - 31.25 ml/min.
Retention Time (e.t.)-2.36 min.
0123 4
TIME (minutes)
Figure 5.6 DMF Chromatogram (Shlmadzu).
39
-------
of the standards was performed internally by TRW using a high performance
liquid chromatography technique (see Section 5.2.4).
Stationary gas analysis for molecular weight determination was
conducted in the field on the integrated bag sample with the Shimadzu
3BT gas chromatography/thermal conductivity detector (GC/TCD). The
chromatographic conditions were:
• Range - 32
• Column temperature - 32°C
• Injection/detector temperature - 120°C
• Carrier gas - helium (He)
• Back pressure He - 3.0 kg/cm2
Two 6 foot x 1/8" O.D. stainless steel columns packed with
Chromsorb 102 and molecular sieve were used in series to separate carbon
dioxide (C00), oxygen (09), nitrogen (N9), and carbon monoxide (CO).
fib
The analysis was recorded by the chromatopak^ Stationary gas per-
centages were calculated for each sample based upon daily determined
calibration factors from the commercially supplied stationary gas
standards and the GC/TCD. An example chromatogram is attached (Figure
5.7).
5.2.2 Wet Impingement - HPLC Analysis
The impinger solutions collected in the field for the experimental
wet impingement sampling train were kept refrigerated at 40°F until
sample preparation at TRW Research Triangle Park analytical laboratory.
All field samples were duly logged and recorded in the appropriate
laboratory notebook. Liquid volumes were measured and a 2 milliliter
(2 ml) aliquot of each sample was taken for HPLC analysis. Complete
HPLC results are in Appendix A.I, Table B.
A Varian Instruments Model 5067 high performance liquid chromatograph
(HPLC) was utilized for the separation and detection of the dimethyl
formamide. The HPLC was coupled to a variable wave length UV-visible
detector and maintained at a wavelength of 240 nm. Integration and
retention times were determined electronically with a Varian CDS-11L
integrator (Figure 5-8). Samples were injected automatically via a
Varian Series 8000 autosampler.
40
-------
TOTAL
G.C. Conditions
GC-3BT
Oven Temp. - 32°C
Inj./Det. - 120°C
Range 32 mv
He - Backpressure 3.0 kg/on'
• -H LO
voco
O t—• i
C-RIA
SMPL I 00
FILE # 1
REPT #1587
Method 44
1
1
3
4
NAME
AIR
0-2
N-2
TIME
0.47
1.18
1.61
1.85
CONC
0
20.6493
76.0943
MK
V
AREA
134336
304
25068
93390
96.7436
253100
5
co
t— i
O
Oslo
co"""1
«-H 4-»
l/l •»-» r*
UO to 1
1
c
a
X
o
CO
t— 1
f— t
Nitrogen
1
\J
1
0 1 2
TIME
Figure 5.7 Stationary Gas Chromatogram GC/TCD Analysis,
-------
Analysis Conditions
Mobile Phase
30% H90
7Q% ACetonitrile
Column
MCH-5 R.P.
Flow Rate
0.5 ml/rnin.
Dectector Wavelength
240 nm
R.T.
6.55
Area
20307
Area %
98.69
001-1
TIME (minutes)
Figure 5.8 HPLC Sample Chromatogram.
-------
The samples were analyzed by high performance liquid chromatography
utilizing a 5 micron, monomeric C,g reverse phase column maintained at
30°C. The mobile phase was an isocratic mixture of 30% water and 70%
acetonitrile flowing at a rate of 0.5 ml/min. The dimethyl formamide
was detected with a variable Uv-Vis detector set at a wave length of
240 nanometers. The retention time of dimethyl formamide was 6.54 minutes.
Quantitation was based on the peak areas as determined by an electronic
integrator with the chromatograms displayed on a strip chart recorder.
A record of chromatograms is maintained in Appendix C.2.
The samples were analyzed in the three batches corresponding to the
three weeks of the field testing. The detected concentrations of dimethyl
formamide varied considerably between sample runs. Consequently,
appropriate DMF standards were run and dilutions made as necessary.
Dilutions were required frequently to adequately quantify the DMF
concentration in the condensate samples collected from the condensers of
the continuous monitor system. The collection efficiency of the wet
impingement sampling train was very good with an approximate collection
efficiency of +95% in the first impinger. On no set of impinger solutions
was any detectable quantities of DMF found in the silica gel impinger of
the sample train.
5.2.3 NIOSH Method S-255
A Varian 3700 gas chromatograph with a flame ionization detector
was used to analyze the silica gel absorption tube samples at the RTP
laboratory. The operating conditions of the instrument were:
• Air - 300 ml/min
• Hydrogen (Hp) ~ 30 ml/min
• Helium (He) - back pressure 24 psig
• Oven temperature - 100°C isothermal
A .2% Carbowax 1500 on Carbopak C 80/100 mesh packed Teflon column
(31 x 1/8") was utilized to separate the DMF. The retention time was
2.4 minutes.
All samples were desorbed in 2 ml of methanol and agitated for
30 minutes before transfer to 1 ml autosampler vials. All samples were
injected twice using the Varian 8000 autosample coupled with the CDS-111
data system. Appropriate standards were run before and after each
43
-------
analysis. The analytical results are in Appendix A.I (Table F). The
chromatographic record of the analysis is contained in Appendix C.3. An
example chromatogram is attached (Figure 5.9).
5.3 QUALITY CONTROL AND QUALITY ASSURANCE
A quality control device for monitoring the progress of the project
was the maintaining of an instrument and a field analytical notebook.
The notebook documentation is presented in Appendix F.2 and Appendix F.3.
The certification of the propane gas standards used were performed by
the supplier, Scott Environmental Technology, Inc. The analytical
report for the calibration gases is presented in Appendix F.4. A problem
arose when the Scott Laboratory could not perform certification tests on
the DMF gas standards. The letter describing this problem is presented
in Appendix F.4. The uncertified DMF gas standards were certified by
the TRW Analytical Laboratory. The certification procedure performed
was the collection of the DMF from the standard gas and analysis of a
Varian HPLC. The collection is achieved by drawing a known amount of
standard gas through a wet impingement system (modified Method 6 train
which used H20 for the collection media). The analysis was a comparison
of known DMF concentrations in water compared with the extracted DMF by
standard HPLC procedures.
The calibration of the personal sample pumps utilized in the NIOSH
sampling method were performed at the TRW facility according to the
recommended NIOSH Method S-255 presented in Appendix E.5. The results
of the calibration of the two pumps used at DuPont are provided in
Appendix F.5.
The calibration for the two dry gas meters and the pi tot tube used
are presented in Appendix F.6, F.7, and F.8.
44
-------
en
G.C. Conditions
Column - 3' x 1/8" teflon
packed with .2% carbowax 20m
on carbopack C, 80/100 mesh.
Oven Temp. - 100°C Isothermal
Flow Rate
Air - 300 ml/mln.
H2 - 30 ml/m1n.
He - backpressure 24 psig
001-1
R.T. - 6.55
Area - 20,307
002-2
R.T. - 6.54
Area - 48,999
I
003-3
R.T. - 6.52
Area - 89,741
TIME
Figure 5.9 Silica Gel Test Chromatogram (DMF) - Varlan 3700.
-------
APPENDIX A.I
COMPLETE RESULTS
-------
TABLE A - VOLUMETRIC FLOWRATE COMPLETE RESULTS BY LOCATIONS
OUPONT (MAY PLANT) CAMDEN, S.C.
LOCATION CRX
Moisture (%) 2.86
Mole Fraction Dry (Md) .971
Molecular Weight Dry (MWd) 27.09
Molecular Weight (MWs) 26.83
Velocity (Vs) (fpm) 1685
Volumetric Flow rate (Qs)
(ACFM) 28502
(DSCFM) 27096
(DSCMM) 767
WDX
3.91
.961
27.00
26.65
2591
18422
17224
487
SSX
17.61
0.824
27.59
25.91
789
619
453
12.8
DMFX
4.97
.952
27.54
27.078
1426
1188
1043
29.5
DRX SCI SCX PSX
15.4 3.3 2.76 2.42
— .976
— 27.86
— 27.62
— 1073
11555 12233 12233 5013
9771 11830 11899 4761
277 335 337 13.5
WSX
1.21
.988
28.22
28.10
1485
4666
4545
129
-------
Table B HPLC ANALYTICAL RESULTS - DUPONT (MAY PLANT) CAMDEN, S.C.
FIELD SAMPLE I.D.
CRX-WI-1A
CRX-WI-1B
CRX-WI-1C
CRX-WI-1D
CRX-WI-1E
BL-WI-1A
BL-WI-1B
BL-WI-1C
BL-WI-10
BL-WI-1E
SSX-WT-1 (1)
SSX-WT-1 (2)
SSX-WT-1 (3)
SSX-WT-1 (4)
SSX-WI-1A
SSX-WI-1B
SSX-WI-1C
SSX-WI-1D
SSX-WI-1E
SSX-WI-1F
LAB #
02006
02001
02008
02015
02010
02011
02012
02013
02049
02002
02003
02018
02020
02015
02016
02017
02019
02046
02014
SAMPLE VOLUME
(ml)
31
32
33
32
—
16
32
32
33
--
367
237
17
20
31
33
33
34
--
33
1.069
6.368
8.905
.005
.005
.005
.005
.005
.005
.005
2.564
2.394
5.872
.005
2.220
9.458
1.469
1.384
.005
4.05
DMF
CONCENTRATION
(mg/ml )
x 10"2
x 10"3
x 10~2
x 10"3
x 10"1
x 10"1
x 10~2
x 10~2
DMF
COLLECTED
(rag)
33.14
2.04
.29
.01
.01
.01
.01
.01
.01
.01
940.99
5.67
.09
.01
68.8
31.2
4.85
.471
.01
1.34
-------
Table B HPLC ANALYTICAL RESULTS - DUPONT (MAY PLANT) CAMDEN, S.C.
(Continued)
FIELD SAMPLE I.D.
OMFX-WI-3A
DMFX-WI-3B
DMFX-WI-3C
DMFX-WI-3D
OMFX-WI-3E
SSX-WT-3A
SSX-VT-3B
SSX-WT-3C
SSX-WT-30
DMFX-CJ-2
ORX-CJ-2A
DRX-CJ-2B
DRX-CJ-3A
DRX-CJ-3B
OMFX-CJ-3A
DMFX-CJ-3B
LAB #
02033
02034
02035
02036
--
02021
, 02022
02039
02040
02026
02029
02030
02024
02025
02027
02028
SAMPLE VOLUME
(ml)
22
8
15
8
—
462
80
19
17
N.D.a
257
13
242
20
74
11
DMF
CONCENTRATION
(mg/ml)
6.105 x 10°
1.3 x 10"1
2.0 x 10"2
1.23 x 10~2
<.005
2.545 x 10°
6.2 x 10~3
<.005
<.005
2.4 x 10"1
1.1429 x 10*1
4.7534 x 10°
1.447 x 10°
3.361 x 10°
2.577 x 10°
4.369 x 10'1
DMF
COLLECTED
(nig)
134
1.04
.3
.098
<.01
1176
.496
<.01
<.01
N.D.
2937
61.8
350
67.2
190.7
4.81
a,
Not determined.
-------
Table B HPLC ANALYTICAL RESULTS - OUPONT (MAY PLANT) CAMDEN, S.C.
(Continued)
FIELD SAMPLE I.D.
BL-CJ-1
ORX-CJ-1A
DRX-CJ-1B
OMFX-CJ-1A
WDX-CJ-1
DRX-WI-2A
DRX-WI-2B
DRX-WI-2C
ORX-WI-20
ORX-WI-2E
PSX-WI-3A
PSX-WI-3B
PSX-WI-3C
PSX-WI-3D
PSX-WI-3E
LAB i
02001
01998
02000
01999
02074
02041
02042
02043
02044
02086
02087
02088
02089
02090
SAMPLE VOLUME
(ml)
50
161
14
8
61
22
23
20
2
—
23
23
22
9
—
DMF
CONCENTRATION
(mg/ml)
<.005
1.131 x 10*1
2.992 x 10°
7.803 x 10°
5.196 x 10°
9.204 x 10°
5.211 x 10°
2.051 x 10°
1.378 x 10°
<.005
1.450 x 101
2.088 x 10°
1.660 x 10"1
2.096 x 10"2
<.005
DMF
COLLECTED
(ing)
•c.Ol
1821
41.9
62.4
317
202
119.8
41.0
2.76
<.01
333.5
48.0
3.65
.189
<.01
-------
Table B HPLC ANALYTICAL RESULTS - DUPONT (MAY PLANT) CAMDEN, S.C.
(Continued)
FIELD SAMPLE 1.0.
PSX-CJ-1B
PSX-CJ-2B
PSX-CJ-3B
SCX-CJ-1
SCX-CJ-2
SCX-CJ-3
SCI-CJ-1
SCI-CJ-2
SCI-CJ-3
Blank O.I.
SSX-CB-1
DMFX-CJ-2
SSX-WT-2A
SSX-WT-2B
SSX-WT-2C
SSX-WT-2D
LAB »
02075
00076
02077
02071
02072
02073
02068
02069
02070
02001
02023
02026
02031
02032
02038
02037
SAMPLE VOLUME
(ml)
19
12
22
12
6
46
18
82
81
272
524
18
502
152
21
21
DMF
CONCENTRATION
(ing/ml)
4.312 x 10*1
4.9606 x 10+1
6.8671 x 10+1
2.575 x 10'1
8.586 x 10"1
8.941 x 10"1
6.367 x 10*1
4.0174 x 102
2.2465 x 102
<.005
1.6155 x 101
2.4 x 10"1
1.903 x 10°
7.6 x 10"3
<.005
<.005
DMF
COLLECTED
(n>g)
819
595
1511
3.09
5.15
41.1
1146
32943
18197
<.01
8465
4.32
955
1.16
<.01
<.01
-------
Table B HPLC ANALYTICAL RESULTS - OUPONT (MAY PLANT) CAMDEN, S.C.
(Concluded)
FIELD SAMPLE I.D.
SCX-WI-1A
SCX-WI-1B
SCX-WI-1C
SCX-WI-1D
SCX-WI-1E
SCI-WI-2A
SCI-WI-2B
SCI-WI-2C
SCI-WI-20
SCI-WI-2E
LAB i
02091
02092
02093
02094
02095
02096
02097
02098
02099
02100
SAMPLE VOLUME
(ml)
14
10
16
8
~
23
24
19
15
--
DMF
CONCENTRATION
(mg/ml )
1.3345 x 10"1
2.756 x 10~2
6.0155 x 10" 3
3.9147 x 10"3
<.005
9.9674 x 10°
6.622 x 10"1
5.656 x 10"2
<.005
<.005
OMF
COLLECTED
(nig)
1.87
.28
.10
.03
<.01
229
15.9
1.07
<.01
<.01
-------
Table C - SUMMARY CONTINUOUS MONITORS (F.I.D.) RESULTS - DUPONT (MAY PLANT) CAMDEN, S.C.
RUN 1
LOCATION CONCENTRATION
PPMV as PPMV as
Propane DMF
CRX
WDX
DMFX
SSX
DRX
SCI
sex
PSX
WSX
14.4
3.38
6.33
2.0
3.23
92.2
7.83
3.28
a
25.4
5.96
11.2
3.5
5.69
162
13.8
5.8
—
RUN 2
CONCENTRATION
PPMV as PPMV as
Propane OMF
6.69
3.34
14.03
2.29
2.27
51
2.77
8.42
5.17
11.8
5.89
24.7
4.03
4.0
90
4.9
14.8
9.1
RUN 3
CONCENTRATION
PPMV as PPMV as
Propane OMF
3.95
2.54
1.60
2.83
1.28
92.4
2.7
4.0
5.2
6.96
4.48
2.82
4.99
2.26
162
4.8
7.0
9.2
AVERAGE
CONCENTRATION
PPMV as PPMV as
Propane DMF
8.35
3.09
7.32
3.56
2.26
78.5
4.43
5.23
5.2
14.7
5.44
12.9
4.17
3.98
138
7.81
9.22
9.16
Not sampled
-------
Table D CONTINUOUS MONITORS - PARAMETERS AND RESULTS BY INDIVIDUAL RUNS - DUPONT (MAY PLANT) CAMDEN, S.C.
RUN f
CRX-1
CRX-2
CRX-3
AVERAGE
WDX-1
WDX-2
WDX-3
AVERAGE
SSX-1
SSX-2
SSX-3
AVERAGE
DMFX-1
DMFX-2
OMFX-3
AVERAGE
SAMPLE
PRESSURE
(PSIA)
4
4
4
2.75
2.75
2.75
2.75
2.75
2.75
4
4
4
BYPASS
SAMPLE
FLOW
SCFH(LPM)
3.0
3.0
3.0
7.0.
7.0
6.0
8.0
(3.6)
(4.6)
1.5
2.5
2.5
FUEL
PRESSURE
(PSI)
20
20
20
20
20
20
20
20
20
20
20
20
AIR
PRESSURE
(PSI)
10
10
10
10
10
10
10
10
10
10
10
10
VOLUME
(Vm)
(FT3)
7.74
7.6
12.8
19.3
17.5
25.5
18.0
26.11h
24.77h
3.38
8.33
8.95
TIME
(HR)
2.58
2.5
4.25
2.75
2.5
4.25
2.25
3.17
3.58
2.25
3.33
3.58
FID CONCENTRATION3
PPM C3Hg PPM DMF
14.4
6.69
3.95
8.35
3.38
3.34
2.54
3.09
19.92
2.29
2.83
8.35
6.33
14.03
1.60
7.32
25.4
11.8
6.96
14.7
5.96
5.89
4.48
5.44
35.1
4.03
4.99
14.7
11.2
24.7
2.82
12.9
CONDENSATE FRACTION5
DMF COLLECTED
(mg) (PPMV)
NGC
NG
NG
N.D.
See
See
See
317
947
956
1176
1026
62
4
196
129
—
—
N.D.
Average
Average
Average
d 60.4
624
435
564
541
.4 218
.32e 6e
258
238
EMISSION RATES1
(1b OMF) (kg DMF)
hr hr
7.69
3.57
2.11
4.45
12.75
12.74
12.47
12.65
3.32
2.21
2.88
2.80
2.87
.36e
3.02
2.92
3.50
1.62
.96
2.02
5.80
5.79
5.67
5.75
1.51
1.01
1.31
1.27
1.21
.10e
1.38
1.33
(1b DMF)
Ib product
—
—
2.42 x 10-«
—
—
6.42 x 10-4
6.77 x 10-«
—
—
7.04 x 10-«
aDoes not Include condensate fraction.
Condensate fraction by HPLC analysis.
Negligible.
Composite of 3 runs.
eNot included within average.
Condensate fraction of PSX and WSX combined.
^Continuous Monitor not run on WSX-1 due to analysis difficulties.
Measured by dry gas meter.
Includes both gaseous (FID) and condesnate (HPLC) fractions.
-------
Table D CONTINUOUS MONITORS - PARAMETERS AND RESULTS BY INDIVIDUAL RUNS - DUPONT (MAY PLANT) CAMDEN, S.C.
(continued)
RUN #
DRX-1
DRX-2
DRX-3
AVERAGE
SCI-1
SCI-2
SCI-3
AVERAGE
SCX-1
SCX-2
SCX-3
AVERAGE
PSX-1
PSX-2
PSX-3
AVERAGE
WSX-19
WSX-2
WSX-3
AVERAGE
SAMPLE
PRESSURE
(PSIA)
4
4
4
3
3
3
4
4
4
4
4
4
N.D.
4
4
BYPASS
SAMPLE
FLOW
SCFH(LPM)
4
4
4
(4)
(4)
(4)
2.5
1.75
1.75
4
4
4
N.D.
4
4
FUEL
PRESSURE
(PSI)
20
20
20
20
20
20
20
20
20
20
20
20
N.D.
20
20
AIR
PRESSURE
(PSI)
10
10
10
10
10
10
10
10
10
10
10
10
N.D.
10
10
VOLUME
(Vm)
(FT3)
18.1
15.7
16.3
29.54h
53.52h
24.28h
6.25
8.45
7
6.16f
12. 9f
18.8
N.D.
TIME
(HR)
4.5
3.9
4.06
2.5
5.17
4.0
2.5
4.83
4.0
1.54
3.23
4.7
N.D.
.83
.33
FID CONCENTRATION3
PPM C3H8 PPM DMF
3.23
2.27
1.28
2.20
92
51
92
78.5
7.83
2.77
2.7
4.43
3.28
8.42
4.0
5.23
N.D.
5.17
5.2
5.2
5.69
4.0
2.26
3.98
162
90
162
138
13.8
4.9
4.8
7.81
5.8
14.8
7.0
9.16
N.D.
9.1
9.2
9.16
CONDENSATE FRACTION6
DMF COLLECTED
(mg) (PPMV)
1863
2999
417
1760
1146e
32943
18197
25560
3.03
5.15
41. 4e
4.12
819
595f
1511f
975
N.D.
... f
...f
N.D.
1223
2267
304
1265
460e
7281
8865
8073
5.8
7.2
69. 4e
6.5
1533
564f
954f
1011
N.D.
— f
__.f
N.D.
EMISSION RATES1
(1b OMF)
hr
133.82
247.39
33.41
138.23
82. 24*
974. 58
1193.52
1085.29
2.61
.95
9.86*
1.9
81.99
29.88
51.20
54.36
N.D.
.46
.47
.47
(kg DMF)
hr
60.84
112.45
15.16
67.8
37.39*
442.98
547.52
493.3
1.18
•74e
4.49
.86
37.27
13.58
23.29
24.71
N.D.
.21
.21
.21
(1b OMF)
Ib product
...
...
—
3.33 x 10-2
...
—
—
6.91 x 10-2
—
—
...
1.21 x 10-4
...
—
—
3.46 x 10-3
N.D.
N.D.
N.D.
N.D.
aDoes not Include condensate fraction.
Condensate fraction by HPLC analysis.
Negligible.
Composite of 3 runs.
eNot included within average.
Condensate fraction of PSX and WSX combined.
^Continuous Monitor not run on WSX-1 due to analysis difficulties.
Measured by dry gas meter.
Includes both gaseous (FID) and condesnate (HPLC) fractions.
-------
Table E INTEGRATED BAG METHOD (GC/FID) COMPLETE RESULTS - DUPONT (MAY PLANT) CAMDEN, S.C.
RUN tf CHROMATOGRAM3
REPORT #
WDX-IB-1 1307
1308
1313
1314
Average
WDX-IB-3 1418
1419
1420
Average
WDX Average
AREA COUNTS
4200
4213
3406
3783
3901
1469
1482
1498
1483
CONCENTRATION11
(PPMV as OMF)
9.2
9.3
7.5
8.3
8.6
3.2
3.3
3.3
3.3
5.95
Appendix C
bBased upon HPLC standard verification (1000 ppm).
-------
Table E INTEGRATED BAG METHOD (GC/FID) COMPLETE RESULTS - DUPONT (MAY PLANT) CAMDEN, S.C.
(Continued)
RUN *
CRX-IB-1
CRX-IB-2
CRX-IB-3
CRX Average
CHROMATOGRAM*
REPORT (K
1320
1321
1322
1323
Average
1346
1347
1348
Average
1412
1413
1414
Average
AREA COUNTS
2995
3375
3599
3206
3294
6097
8446
8582
7708
4968
5505
5175
5216
CONCENTRATION6
(PPMV as DMF)
6.6
7.4
7.9
7.1
7.3
13.4
18.6
18.9
17.0
10.9
12.1
11.4
11.5
11.93
Appendix C
Based upon HPLC standard verification (1000 ppm).
-------
Table E INTEGRATED BAG METHOD (GC/FID) COMPLETE RESULTS - DUPONT (MAY PLANT) CAMDEN, S.C.
(Continued)
RUN IP
DRX-IB-1
DRX-IB-2
DRX-IB-3
•
DRX-IB-3
CHROMATOGRAM3
REPORT #
1522
1523
1524
1525
1526
Average
1594
1595
1596
1597
1598
Average
1643
1644
1645
1646
1647
1648
Average
AREA COUNTS
8737 -
9484
8706
9349
9131
9081
1014
1193
1072
1132
1182
1119
1049
1586
1811
2059
2387
2625
1920
CONCENTRATION6
(PPMV as DMF)
19.2
20.9
19.2
20.6
20.1
20.0
2.2
2.6
2.4
2.5
2.6
2.5
2.3
3.5
4.0
4.5
5.3
5.8
4.2
aAppend1x C
Based upon HPLC standard verification (1000 ppm).
-------
Table E INTEGRATED BAG METHOD (GC/FID) COMPLETE RESULTS - DUPONT (MAY PLANT) CAMDEN, S.C.
(Continued)
RUN #
DMF-IB-1
DMF-IB-2
DMF-IB-3
OMF Average
CHROMATOGRAM3
REPORT #
1507
1508
1509
Average
1578
1579
1580
Average
1638
1639
1640
Average
AREA COUNTS
2582
3708
3710
3333
1421
1511
1496
1476
654
779
740
724
CONCENTRATION6
(PPMV as OMF)
5.7
8.2
8.2
7.3
3.1
3.3
3.3
3.2
1.4
1.7
1.6
1.6
4.03
3Appendix C
Based upon HPLC standard verification (1000 ppm).
-------
Table E INTEGRATED BAG METHOD (GC/FID) COMPLETE RESULTS - OUPONT (MAY PLANT) CAMDEN, S.C.
(Continued)
RUN #
SSX-IB-1
SSX-IB-2
SSX-IB-3
SSX Average
CHROMATOGRAM3
REPORT #
1500
1501
1502
1503
Average
1574
1575
1576
Average
1635
1636
1637
Average
AREA COUNTS
4320
3049
3614
3366
3587
580
641
655
625
516
762
828
702
CONCENTRATION6
(PPMV as DMF)
9.5
6.7
8.0
7.4
7.9
1.3
1.4
1.4
1.4
1.1
1.7
1.8
1.5
3.6
aAppendix C
Based upon HPLC standard verification (1000 ppm).
-------
Table E INTEGRATED BAG METHOD (GC/FID) COMPLETE RESULTS - DUPONT (MAY PLANT) CAMDEN, S.C.
(Continued)
RUN f CHROMATOGRAM3
REPORT #
SCI-IB-1 1707
1708
1709
1710
1711
Average
SCI-IB-2 1800
1801
1802
1803
Average
SCI-IB-3C 1868
1869
1870
1871
Average
SCI Average
AREA COUNTS
35495
46084
50252
44962
45389
44436
63282
65889
70057
70666
67474
132432
149372
138021
139294
139780
CONCENTRATION5
(PPMV as OMF)
78.1
101.4
110.6
99.0
99.9
97.8
139.3
145.0
154.2
155.5
148.5
291.5
328.7
303.8
306.6
307.7
123. 15d
aAppend1x C
Based upon HPLC standard verification.
GSuspected data - No blank performed - See Appendix C.
Average of two sampling runs (Run #SCI-IB-3, omitted).
-------
Table E INTEGRATED BAG METHOD (GC/FID) COMPLETE RESULTS - DUPONT (MAY PLANT) CAMDEN. S.C.
(Continued)
RUN # CHROMATOGRAM3
REPORT #
SCX-IB-1 1735
1736
1737
1738
1739
Average
SCX-IB-2 1780
1781
1782
1783
1784
Average
SCX-IB-3 1850
1851
1852
1853
Average
SCX Average
AREA COUNTS
700
822
570
669
433
639
0
0
0
0
0
0
7370
. 8457
9843
9578
8812
CONCENTRATION6
(PPMV as OMF)
1.5
1.8
1.3
1.5
1.0
1.4
0
16.2
18.6
21.7
21.1
19.4
6.93
a Appendix C.
Based upon HPLC standard verification.
-------
Table E INTEGRATED BAG METHOD (GC/FID) COMPLETE RESULTS - DUPONT (MAY PLANT) CAMDEN, S.C.
(Continued)
RUN f CHROMATOGRAM3
REPORT *
PSX-IB-1 1723
1724
1725
Average
PSX-IB-2 1785
1786
1787
1788
Average
PSX-IB-3 1865
1866
1867
Average
PSX Average
AREA COUNTS
20317
29583
29448
26449
8297
10576
11553
11095
10380
123596
130622
128402
127540
CONCENTRATION6
(PPMV as DMF)
44.7
65.1
64.8
58.2
18.3
23.3
25.4
24.4
22.8
272.0
287.5
282.6
280.7
120.56
aAppendix C.
Based upon HPLC standard verification (1000 ppm).
-------
Table E INTEGRATED BAG METHOD (GC/FID) COMPLETE RESULTS - DUPONT (MAY PLANT) CAMDEN, S.C.
(Concluded)
RUN i CHROMATOGRAMa
REPORT i
WSX-IB-1 1813
1814
1815
1816
1819
Average
WSX-IB-2 1858
1859
1860
Average
WSX Average
AREA COUNTS
4481
5906
5111
5649
8943
6018
11461
12749
12486
12232
CONCENTRATION5
(PPMV as DMF)
9.9
13.0
11.2
12.4
19.7
13.2
25.2
28.1
27.5
26.9
20.05
aAppendix C
Based upon HPLC standard verification (1000 ppra).
-------
TABLE F - NIOSH METHOD S-255 SAMPLING AND ANALYSIS RESULTS
SAMPLE ID #
WDX -
CRX -
SSX -
DRX -
DMF -
PSX -
SCI -
sex -
SGT
SGT
SGT
SGT
SGT
SGT
SGT
SGT
TIME
(min.)
35
60
22
46
60
60
15
39
SAMPLE COLLECTION
(counts) (factor - ml/cnts
19799
34261
14120
25580
33594
33455
9416
22048
.427
.364
.364
.427
.364
.427
.364
.427
SAMPLE VOLUME
DMF
-) (cc) (nig)
8514
12471
5140
10327
12228
14285
3427
9414
.72
1.35
12.09
12.67
4.43
14.95
48.63b
<.la
CONCENTRATION
(mg/ro ) (ppmv)
84.6
108
2352
1160
362
1047
14190
...
28.5
36.3
788
388
121
351
4953
<.01
"Not detected
Value exceeds recommended value for NIOSH method.
-------
APPENDIX A. 2
SAMPLE CALCULATIONS
-------
Table A.2. SAMPLE CALCULATIONS
SAMPLE CALCULATION TO PROVIDE THE VOLUMETRIC FLOWRATE
(STANDARD CONDITIONS) AT DMFX LOCATION DURING TEST #2
AH
(Tm + 460)
1. Vmstd = 17.64 x Vm (PB + I376T _
.02
Vmstd = 17.64 (4.940) (29.76 + 13.6)
(75 + 460)
Vmstd = 117,64) (29 76) (4.940)
oob
Vmstd = 4.847 ft3
2. V . . = 0.0474 x Vw = ft3
wstd
Vwstd=°-1943ft3
V = B
wstd Dws
V + V
mstd wstd
(0.1943)
(4.847 + .1943)
= 3.85%
-------
Table A.2. Continued
4.
Md = .962
5. MWd = (% C02 X «j) + %02 x f^) H- (%N2 x f|j)
MWd = (0% X g_) + (20.82 x gjg) + (77.07 +
0 + 6.662 + 21.58
MWd = 28.242
6. MW = MWd X Md + 18(l-Md)
(28.242) (.962) + 18 (1-9.62)
(28.242) (.962) + 18 (.038)
(28.242) (.962) + .684
Mw = 27.853
?
Ts
7. Vs = KpCp KAP If Ps M.
Vs = (.84) (85.49) (.387)
Vs = 23.082 ft./sec.
Vs = 1384.926 ft./min.
8. .123 (Vs) (As) (Md) (Ps) =
(Ts + 460)
.123 (1384.926) (144) (.962) (29.77)
571.8
SCFM = 1228.58
W/ \
1571.8 \
\2B.76 (27.853)^
-------
Table A.2 Continued
Nomenclature
B = Water vapor in the gas stream, proportion by volume.
K = 85.49 ft/sec Pitot tube constant for English system.
C = .84 Pitot tube calibration coefficient dimensionless.
P
%M = Percent moisture by volume.
Md = Mole fraction dry gas.
MW = Molecular weight of stack gas - wet basis (Ib/lb mole).
MWd = Molecular weight of gas - dry basis (Ib/lb mole).
PD = Barometric pressure in inches of mercury.
D
P = Absolute pressure of the dry gas meter in inches of mercury (Hg).
P = Absolute pressure of the stack in inches of mercury (Hg).
TM = Temperature of the dry gas meter degrees Fahrenheit.
V = Stack gas velocity (ft/sec).
V xj = Volume of gas at standard conditions sample measured
mstd by dry gas (ft3).
V . . = Volume of water vapor in the gas sample, corrected
wsta to standard conditions (ft3).
AH = Average orifice pressure differential (in. H20).
-------
Table A.2. Continued
9. Sample calculation for response factor of propane
to DMF.
Response Factor (Fg) = p^pane Responsi (F )
Where:
Fl = Average Area^ounts (DMF)(10(? ppm)
_ Average Area Counts Propane
2 402 ppm
=1.76
As measured of a Shimadzo Mini 1 GC/FID.
-------
Table A.2 Continued
10. Sample calculation to provide the concentration level of the
WDX location by the silica gel tube method.
(Example of WDX-SGT)
™m HMF - .72 mg y counts y 1000 ml Y 1000 £ ,
PpmDMF - X X ~ X 3 X ('
19799 counts n .427 ml
= 28.53 ppmv
Where:
.72 mg = Weight of DMF collected in the sample.
19799 counts = Number of sample pump counts over the sample period.
.427 ml = Number of ml/count from the sippin pump calibration data.
1000 = Conversion factor from ml to liters.
1000 = Conversion factor from liters to cubic meters.
0.335 = Conversion factor of DMF from mg/m3 to ppm from the Bureau
of Mines tabulation table for a gas with MW = 73, 25°C
and 760 mm Hg2.
28.53 ppmv DMF = Concentration level of DMF in the gas stream by
volume.
11. Sample calculation to provide the concentration level of DMF
in the gas stream at DMFX location.
CFID (RF) = CDMF
Where:
CFID = Concentration level at DMFX by the FID Continuous
Monitors is 6.33 ppm as propane.
RF = Response factor of C3Hg to DMF is 1.76.
Cnur = Concentration level of DMF at the DMFX location.
Unr
(6.33) (1.76) = 11.12
CDMF = n-12 PpmV
-------
Table A.2. Continued
12. Sample calculation to provide the production rate compared
in pounds of DMF emission to the pounds of product.
(Example Day 1 at the WDX)
1b DMF 1.14 1b DMF y 4 hr. y Test Period
Ib Product hr. * Test Period 71990 Ib Product
= 6.33 X 10"5 Ib DMF
Ib Product
Where:
1.14 Ib. DMF Emission rate of DMF per hour at the WDX location.
hr.
4 hr. = Time length of test period.
17990 Ib = Ib. product processed over the test period.
13. Sample calculation to provide the concentration level by the
wet impingement method.
(Example of SCX-WI-1)
r _ 1.87 mg 35.42 mg .335
LWI " 9.767 ft3 iF~"^ 1
CWI = 2.27 ppmv DMF
Where:
CWT = Concentration of DMF as determined by the wet impingement
W1 method.
1.87 mg = Detected weight of DMF in the sample.
9.767 ft.3 = Volume of dry gas sampled at standard conditions -
68°F, 29.92" hg.
35.42 mg = Conversion factor from mg/ft3 to mg/m3.
.335 = Conversion factor of DMF from mg/m3 to ppm from the
Bureau of Mines tabulation table for DMF with a
MW = 73, 25°C, and 760 mm Hg2.
-------
Table A.2. Concluded
14. Sample calculation for ppm concentration of gas stream to
emission rate in Ib/hr.
(Example)
lb - fo RO « m-\ 2.99 mgi v 798.4 m3 v 60 min. v 1 kg v 2.2 1b
hr. " C9'58 ppn° ^ X min. X ~~hF~ X 10g mg X ~Ig~
= 3.019 Ib. DMF
hr
9.58 ppm = DMF concentration level by volume.
2.99 mg _ Conversion factor for DMF from the Bureau of Mines
in3 at standard, temperature, and pressure.
798.4 m3 _ Flow in dry standard cubic meters per minute (DSCMM).
min
— rr^-' - Conversion factor of minutes to hours.
hr
1 kg _ Conversion factor of milligrams to kilograms.
10b mg
2.2 Ib Conversion factor of kilograms to pounds.
kg
3.019 Ib _ Mass Emission Rate of DMF from gas stream.
hr
------- |