tr/EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report SO-SNF-2
February 1981
Air
Synthetic Fiber
Manufacturing
Emission Test Report
Monsanto Textiles
Company
Decatur, Alabama
-------
SOURCE TEST AT
MONSANTO SYNTHETIC FIBERS PLANT
DECATUR, ALABAMA
Contract No. 80-02-3545
Work Assignment 1
Project No. 80-SNF-2
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
P. 0. Box 13000
Research Triangle Park, North Carolina 27709
-------
GLOSSARY OF TERMS
AN Acrylonitrile
DMAC Dimethylacetamide
DMF Dimethylformamide
EMB Emissions Measurement Branch
EPA Environmental Protection Agency
FID Flame lonization Detector
GC Gas Chromatograph
HPLC High Pressure Liquid Chromatography
NIOSH National Institute for Occupational
Safety and Health
NSPS New Source Performance Standard
OAQPS Office of Air Quality Planning and Standards
OVA Organic Vapor Analyzer
PES Pacific Environmental Services, Inc.
SNF Synthetic Fiber Industry
TCD Thermal Conductivity Detector
VOC Volatile Organic Compound
-------
TABLE OF CONTENTS
Section Page
1 INTRODUCTION 1-1
2 SUMMARY AND DISCUSSION OF RESULTS 2-1
3 METHOD DEVELOPMENT 3-1
4 LOCATION OF SAMPLING POINTS 4-1
5 PROCESS DESCRIPTION 5-1
6 SAMPLING AND ANALYSIS METHODS 6-1
APPENDIX A - SAMPLE CALCULATIONS
APPENDIX B - FIELD DATA SHEETS
APPENDIX C - ANALYTICAL DATA
APPENDIX D - SAMPLING METHODS AND ANALYTICAL TECHNIQUES
APPENDIX E - QUALITY CONTROL AND QUALITY ASSURANCE
APPENDIX F - TEST LOG
APPENDIX G - PROJECT PARTICIPANTS
-------
1.0 INTRODUCTION
During the week of October 27-31, personnel from TRW's
Environmental Engineering Division conducted an emission source test for
volatile organic compounds (VOC) at the Monsanto Company's synthetic
fiber plant in Decatur, Alabama. Also present were representatives from
the Environmental Protection Agency's Emission Measurement Branch
(EPA/EMB) and Pacific Environmental Services Inc. (PES). PES is under
contract to EPA's Office of Air Quality Planning and Standards (OAQPS)
to assess the emissions of the synthetic fiber industry. This testing
was conducted under Contract #68-02-3545, Work Assignment 1, Project
080-SNF-2 to EPA/EMB.
Data was obtained to fulfill the following requirements:
(a) to provide process mass emission data for the development of a
New Source Performance Standard (NSPS) for the synthetic
fibers industry,
(b) to recommend a test method for determination of dimethyl-
acetamide (DMAC),
(c) to detect the presence of acrylonitrile.
The Monsanto facility produces an acrylic fiber utilizing DMAC as a
solvent in a wet spinning process. Emissions from the spinning, crimper,
tow and dope prep areas were tested for total organic compounds.
Forty-three separate emission points were either tested or estimated
based on test results of similar units. The only VOC detected was the
solvent DMAC. In analyzing for other VOC's, a major emphasis was placed
on the determination of the presence of acrylonitrile in the emissions
of the wet spinning process.
Four methods of testing were utilized at the Monsanto plant to
obtain a proposed method for VOC determination of the synthetic fiber
industry.
-------
2.0 SUMMARY AND DISCUSSION OF RESULTS
The test plan for sampling at the Monsanto synthetic fibers plant
called for testing the main spinning exhausts continuously for three
one-day periods. In addition, a screening of the other vents was to
take place in order to determine if continuous readings should be taken
at these locations. A secondary purpose was to evaluate alternative
methods at as many locations as possible and to determine the presence
of acrylonitrile from the main sources of emissions.
The mass emission results (Table 2.1) are separated according to
the various components of the process. Figure 4.1 displays an aerial
view of the roof vents and stacks that were sampled. Table 5.1 presents
a key to the vent numbers. Upon arrival at the plant, a quick screening
of the vents was undertaken in order to evaluate the test plan.
Measurements with a portable organic vapor analyzer (OVA) indicated the
greatest concentrations were being emitted from the north and south main
spinning exhaust stacks and from the basement vents A, B and C. Therefore,
continuous flame ionization detector (FID) analyzers were set up at the
north and south stacks and at basement vents A and B. The FID analyzer
(Horiba) set up at basement vent B was subsequently determined to be
inoperable. It was decided to operate an FID analyzer for one day only
at each of the basement vents. Moisture was not a problem at any of the
vents with the exception of the north and south spinning exhausts.
Heated lines were installed at these locations and performed adequately
at the south stack. The analyzer at the north stack had a lower bypass
flowrate which caused some condensation. The results during this period
were biased high due to this moisture. At the end of two days of sampling
at the south stack, the analyzer at that location was moved to the north
stack for analysis. The last two days of data at the north stack are
reported in Table 2.1.
-------
Table 2.1. SUMMARY OF RESULTS
Location
South stack
North stack
Basement vent A
Basement vent B
Basement vent C
Basement vent D
ro
i
Dope prep room vent A
Dope prep room vent B
Dope prep room vent C
Dope prep stack A
Dope prep stack B
Dope prep stack C
Dope prep stack D
Dope prep stack E
Crimper exhaust
Type B
Crimper exhaust
Type F
Crimper exhaust
Type G
Source If
Spinning exhaust
(1-16)
Spinning exhaust
(17-30)
BV A
BV B
BV C
8
RMV A
RMV B
RMV C
Room vent
Room vent
Hood vent
Room vent
Hood vent
52(17-19)
47(12-14)
36(9-11)
45(23-26)
34(27-30)
44(1-4)
54(20-22)
11(27-30)
Method of Sampling and Analysis
NIOSH Wet Impingement FIA OVA
(ppmv as DMAC) (ppmv as DMAC) (ppmv as DMAC) (ppmv as DMAC)
<0.1 9.4 9.9
11. 7C 10. 3e 24.2
10.7 44.8 12.0
18.2 25.3 18.7
2.7 <0.1 5.6
"** "*" ~™ ~~
3.9
3.2
3.5
.-
1.2
35.7 65.0
1.3
1.4 0.8
1.1
1.1
2.9 0.5
0.5
1.2
0.7
0.5
Source flow
(DSCMM)
Actual Estimated
4,778b
5.221b
1,008
761
1,594'
~~
470
522
536
—
110
„
"
120
429
186
—
—
"*"
--
--
110
110
120
120
429
429
429
429
Emission rate"
(Ibs/hr)
Actual Estimated
21.19
56.61
77.80 (Subtotal)
5.42
6.38
4.00
Not operating
15.80 (Subtotal)
0.82
0.75
0.68
2.25 (Subtotal)
0.06f
--
3.20
0.06f
0.06
3.38 (Subtotal)
0.08
0.069
0.06
0.56
0.099
0.239
0.139
0.04
-------
Table 2.1. Continued
ro
i
to
Location
Crimper exhaust
Type H
Crimper exhaust
Type I
Exhaust fan type D
Exhaust fan
Spinning room
celling exhaust
Tow cooler type E
Tow cooler
Source *
57(25-26)
21(15-16)
20(17-20)
10(21-24)
27(nc 16,17)
9(mc 18,19)
5
46(nc 1)
28(sp1n c)
30(mc 2)
26Jx 3)d
39°
8(basenent)
4
3
1
2
15(28)
14(29)
12(26)
13(30)
16(30)
17(25)
18(24)
19(23)
23(11.12)d
24(9) .
32(13,14)°
Method of Sampling and Analysis
NIOSH Wet Impingement FIA
(ppov as DMAC) (ppmv as OHAC) (ppnv as DMAC)
—
--
..
3.9
--
—
—
—
--
—
—
—
5.4
4.1
__ ... ..
..
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
..
...
..
OVA
(ppmv as DMAC)
0.5
1.1
1.3
—
3.1
--
--
—
--
--
3.1
3.1
•
5.3
1.9
1.1
0.9
0.5
--
--
--
--
—
—
—
..
—
--
Source flow
(DSCHH)
Actual Estimated
17
26
26
—
219
..
--
--
..
--
219
219
..
415
415
415
415
34
34
34
34
34
34
34
34
..
. ..
—
Emission rate*
(Ibs/hr)
Actual Estimated
0.004
0.019
0.01
Not operating
1.27 (subtotal)
0.38
Not operating
Not operating
Not operating
Not operating
Not operating
0.38!
0.38T
Not operating
1.14 (Subtotal)
1.00 --
0.76a
0.209
0.179
2.13 (Subtotal)
0.01 ~ .
O.OlL
o.oi;
O.OlI
O.OlL
O.OlI
O.OlI
o.or
O.OlJ
O.OlI
o.or
0.11 (Subtotal)
(continued)
-------
Table 2.1. Concluded
Method of Sampling and Analysis
Location
Source * NIOSH
(ppmv as DMAC)
Wet Impingement
(ppmv as DMAC)
FIA
(pprav as DMAC)
Sour
(0
OVA Actual
(ppmv as DMAC)
•ce flow Emission rate8
ISCMM) (lbs/hr)
Estimated Actual Estimated
Roof exhaust 56
Finishing stack 6 and 7
Condenser stack 59
Filter press hoods
Not operating
104.35 (Total plant)
Not tested
Not tested
Not tested
ro
i
aBased on FID concentration.
These flows are based on pHot tube measurements, the remainder are based on anemometer readings.
€Average of two runs of 12.6 and 10.7.
Operating periodically.
eAverage of two runs of 17.4 and 3.2.
Assume similar vents to have same flow and concentration.
^Assume similar vents to have same flow with a known concentration.
-------
Due to time constraints and a shortage of FID analyzers, the final
screening was performed by both an FID continuous analyzer (Beckman 402)
and the OVA analyzer. All of the objectives of the test plan were
satisfied with the exception of the gas chromatographic results. The
column chosen for DMAC separation (6' x 1/8" Teflon® packed with 0.2%
Carbowax 20M on Carbopak 80/100 mesh) was the same one utilized for
separation of a similar solvent at another synthetic fiber plant. No
results were obtained and it now appears that the column was poisoned
during the previous testing. A blank column was attempted with similar
results. More discussion of this problem is presented in Section 6.
No acrylonitrile (<.l ppm) was found during this test as would be
expected due to the low levels of solvent. The NIOSH (silica gel)
methods and experiemental wet impingement methods appear to have varying
results. Most of this is due to the low levels, which bordered on the
minimum detectable amounts, collected in the samples.
The remainder of this section is separated into the discussion of
the individual sections of this test. The methods used for testing and
their comparisons, limitations and definitions are contained in Section 3.
2.1 NORTH AND SOUTH MAIN STACKS
Table 2.1 presents the total average emissions for the north and
south stacks while Tables 2.2 and 2.3 report the hourly averages at each
location. Since most of the emissions from the spinning roof came from
the main exhaust spinning stacks, more sampling time was spent and
differing methods were performed here than elsewhere on the roof.
Each spinning cell located in the building is enclosed and ambient
air drawn in to remove the solvent (DMAC) concentrations to the main
exhaust stacks. These enclosure hoods are ducted together to enter
either the north or south main spinning exhaust stack. Thirteen
enclosure hoods are ducted into the south stack while fourteen are
ducted into the north stack. Both stacks were monitored by continuous
FID analyzers for two one-day periods. In addition, the auxiliary test
methods were conducted simultaneously for a one-half hour period. The
results appear to be quite accurate as no drift or noise problems were
encountered for the instrument during the test period. Subsequent
2-5
-------
Table 2-2. NORTH SOLUTION STACK - HOURLY AVERAGES
ro
en
Date
10/28/80
10/29/80
10/30/80
Time
1803-1853
1903-1953
2003-2053
2103-2153
2203-2253
2303-2353
0003-0033
1650-1740
1750-1840
1850-1940
1950-2040
2050-2140
2150-2240
2250-2340
2350-0040
0050-0140
0150-0240
0250-0340
0350-0440
0450-0540
0550-0640
0650-0740
0750-0840
Concentration3
38.8
37.5
36.5
36.3
36.0
35.8
35.5
25.0
22.3
21.3
23.5
23.0
22.0
22.0
22.0
22.5
21.8
22.3
22.8
22.5
22.8
22.2
25.2
Date Time
10/30/80 0928-1018
1028-1118
1128-1218
1228-1318
1328-1418
1428-1518
1528-1618
1628-1718
1728-1818
1828-1918
1928-2018
2058-2118
2128-2018
2228-2318
2328-0018
10/31/80 0028-0118
0128-0218
0228-0318
0328-0418
0428-0518
0528-0618
0628-0718
0728-0818
Concentration3
27.3
26.5
22.0
27.0
27.8
34.0
35.2
33.7
29.8
29.0
30.5
31.3
31.0
31.5
30.8
29.0
28.5
29.3
28.8
27.5
27.7
26.3
21,2
ppm as propane.
-------
Table 2-3. SOUTH SOLUTION STACK - HOURLY AVERAGES
ro
Date
10/28/80
10/29/80
Time
1250-1340
1350-1440
1450-1540
1550-1640
1650-1720
1740-1830
1840-1930
1940-2030
2040-2130
2240-2330
2340-2430
2440-0130
0140-0230
0240-0330
Concentration3
14.2
13.7
13.8
14.2
15.0
13.0
12.7
12.0
12.0
11.5
11.7
11.7
11.7
11.5
Date
10/29/80
Time
0340-0430
0440-0530
0540-0630
0640-0730
0740-0830
0840-0930
0940-1030
1040-1050
1108-1158
1208-1258
1308-1358
1408-1458
1508-1538
Concentration3
12.3
12.0
11.8
12.0
12.0
12.0
9.3
7.5
8.0
8.2
9.0
8.8
9.0
ppm as propane.
-------
calibrations during the test period showed only a one percent drift and
a similar signal to noise ratio. The results of silica gel and impingement
methods varied from the FID results. This is due in the most part to
the low levels collected which put the analysis close to the minimum
detectable limit of these methods. Flow measurements were obtained by
Federal Register Methods 1-4. A point of average velocity was chosen as
the sampling location for the FID's and alternate methods. Strip chart
results are located in Appendix B.6. Sample locations and diagrams
appear in Section 4.
2.2 BASEMENT VENTS (DOPE PREP AREA)
The second largest concentration of emissions came from the
ventilation of the basement area underneath the spinning machines. The
sampling took place by utilizing an FID continuous analyzer to obtain a
24-hour sample at each location. Flows were taken by utilizing an
anemometer due to the stack configuration. This method of flow measure-
ment is not as accurate as the pi tot tube but does give a good (+20%)
range for the flows. The mass emission rates ranged from 4.0 pounds per
hour to 6.4 pounds per hour for the three basement vents. It was observed
that the higher concentrations (ppm) were found at vents having lower
flow as less ambient air is introduced in these areas. The strip chart
results are located in Appendix B.6 and hourly averages in Table 2.4.
Sample locations and diagrams appear in Section 4.
2.3 CRIMPER EXHAUSTS
Each spinning machine has an accompanying crimper. Here the fiber
is steamed and crimped to give the fiber body. The steam is hooded and
exhausted to the roof by the crimper exhaust vents. Each crimper exhaust
was measured either by the continuous hydrocarbon analyzer or the portable
OVA analyzer. The readings were taken over short (5 min) time periods
and are, therefore, just screening results. Table 2.1 presents the
crimper exhausts which ranged from 0.5 to 2.9 ppm as DMAC. The numbers
in parenthesis under source numbers indicate the corresponding spinning
machine associated with each crimper. Velocities were measured by an
anemometer on one vent of each type on the roof. The other vents of
each type were assumed to have the same volumetric flow as the
2-8
-------
Table 2-4. BASEMENT VENTS - HOURLY AVERAGES
Basement Vent A
Basement Vent B
Basement Vent C
10
Date
10/28/80
10/29/80
Time
1400-1450
1500-1550
1600-1650
1700-1740
1803-1853
1903-1953
2003-2053
2103-2153
2203-2253
2303-2353
0003-0053
0103-0153
0203-0353
0303-0353
0403-0453
0503-0553
0603-0653
0703-0753
0803-0853
0903-0953
1003-1043
Concentration3
(ppm)
11.0
11.0
11.0
10.0
11.8
20.8
14.3
14.0
13.5
13.0
13.0
13.0
13.0
13.0
13.0
12.7
12.0
12.0
12.0
12.0
12.0
Concentration3
Date Time (ppm)
10/29/80 1113-1208 18.1
1213-1308 18.2
1313-1410 20.4
1415-1510 20.7
1515-1610 20.7
1615-1640 21.0
Date Time
10/29/80 1732-1822
1832-1922
1932-2022
2032-2122
2132-2222
2232-2322
2332-0022
10/30/80 0032-0122
0132-0222
0232-0322
0332-0422
0432-0522
0532-0622
0632-0722
0845-0945
Concentration3
(ppm)
7.3
7.0
6.5
6.5
6.0
5.5
5.5
6.5
5.5
6.0
5.5
5.5
5.5
4.8
6.4
a,
ppm as propane.
-------
measured type. All dimensions and fan types were the same within each
type group.
The OVA and Beckman 402 are both flame ionization detector (FID)
hydrocarbon analyzers. Both analyzers were calibrated with propane and
compared at the basement vents. The analyzers compared favorably during
this initial comparison. The differences in the results in Table 2.1
between the analyzer readings could be due to the fact that both analyzers
were operated on different days. Sample locations and diagrams are
presented in Section 4.
2.4 EXHAUST VENTS
The remaining vents on the main roof were room air exhaust vents
and tow cooler exhaust vents. The source number presented in Table 2.1
is a plant assigned number while the number in parenthesis is the spinning
machine number associated with the exhaust fan or tow cooling fan. A
representative vent from each vent type was tested either by continuous
hydrocarbon analyzer (Beckman 402) or the portable FID (OVA).
Vents of the same type and dimensions were assumed to have the same
concentrations and volumetric flow as the measured vents. Velocities
were taken by anemometer as mentioned before and provided good approxi-
mations of the actual volumetric flow. Hourly results are listed in
Table 2.5. Sample location diagrams and vent diagrams are in Section 4.
2.5 DOPE PREP ROOM VENTS
In order to complete the survey, it was decided to sample the room
vents and stacks emitting from the dope prep area. Three large room air
exhausts venting from the side of the dope prep building were tested
(adjacent to the roof top area, see Figure 4.1). These were analyzed
utilizing the continuous hydrocarbon analyzer (Beckman 402). The results
ranged from 3.8 to 4.2 ppm. Strip chart results are listed in Appendix B.6
and hourly averages in Table 2.6. Figures and diagrams are in Section 4.
2.6 DOPE PREP STACKS
A series of five stacks were present on the roof of the dope prep
area. Three of these vents A, B, and D (TRW nomenclature) were short
stacks which emitted room air from this section of the process. Only
2-10
-------
Table 2-5. MISCELLANEOUS VENTS - HOURLY AVERAGES*
#45 Crimper exhaust
#52 Crimper exhaust
Date
Time
Concentration3
(ppmv)
Date
Time
Concentration3
(ppmv)
10/30/80 1445-1510
3.1
10/30/80 1200-1310
1.5
#15 Tow cooler
#3 Fan room ceiling exhaust
ro
Date
Time
Concentration3
(ppmv)
10/30/80 1520-1540
1.1
Date
Time
Concentration3
(ppmv)
10/30/80 1431-1433
4.4
#27 Exhaust fan
#4 Fan room ceiling exhaust
Date
Time
Concentration1
(ppmv)
Date
Time
Concentration1
(ppmv)
10/30/80 1325-1350
4.2
10/30/80 1400-1425
5.8
ppm as propane.
-------
Table 2-6. DOPE PREP ROOM VENTS - HOURLY AVERAGES
Room Vent A
Date
10/30/80
Time
1700-1750
1800-1850
Concentration3
(ppmv)
4.2
4.9
Room Vent B
Date
Time
10/30/80 1332-1422
Concentration1
(ppmv)
3.7
ro
i
Date
Room Vent C
Time
Concentration'
(ppmv)
10/30/80 1225-1315
3.8
ppm as propane.
-------
one of these was measured. The other two room vents were assumed to
have the same flow and concentration for purposes of determining mass
emission rate. The two remaining stacks were hood exhausts serving
equipment cleaning areas. The results range from 1.3 ppm for hood "E"
to 65 ppm for hood "C." This wide range is quite possible because the
emissions are dependent upon the operation being conducted under the
hood at the time. The measurements on all of these stacks were taken
with the Beckman 402 continuous FID analyzer for concentration and an
anemometer for velocity. Strip chart results are listed in Appendix B.6
and hourly averages in Table 2.7. Sample diagrams are in Section 4..
2-13
-------
Table 2-7. DOPE PREP STACKS - HOURLY AVERAGES
ro
Date
10/30/80
Date
10/30/80
10/31/80
"B"
Time
0906-0936
urn
Time
1740-1830
1840-1930
1940-2030
2040-2130
2140-2230
2240-2330
2340-0030
0040-0130
0140-0230
0240-0330
0340-0430
0440-0530
0540-0630
0640-0730
0740-0830
Stack "C" Stack
Concentration3 Date Time Concentration3
(ppmv) (ppmv)
1.3 10/30/80 0941-1025 69
Stack
Concentration3
(ppmv)
3.8
4.7
3.0
2.2
1.0
0.8
0.6
0.6
0.8
0.7
0.4
0.6
0.5
0.2
0.6
ppm a.6 propane.
-------
3.0 METHOD DEVELOPMENT
In addition to field testing plants to obtain data for possible new
source emission standards, TRW contracted with EPA/EMB to develop or
utilize the best available method in order to obtain this data. After a
presurvey at the plant, it became apparent that the major VOC was DMAC.
Efforts at method development centered around the determination of DMAC
concentrations and acrylonitrile (AN).
A product specification guide to DMAC was obtained and served as a
guide. Extensive experience had been gained on a similar solvent
(dimethyl formamide) in a previous test and it appeared that the problems
would be similar. Like dimethyl formamide, DMAC adsorbed on metal, was
hygroscopic, and had similar physical constants. The only immediately
known method of analysis was the NIOSH S-254 silica gel adsorption tube
method. It was known that high pressure liquid chromatography (HPLC)
analysis was possible and adequate glass capillary separation with an
(GC/FID) had been shown (NIOSH method). In addition, continuous analysis
with a hydrocarbon analyzers would be possible if there were no interfering
organic compounds present. This information led to suggesting four
separate methods for consideration during this project. They were as
follows:
a) NIOSH Method S-254 (see Section 3-1 and Appendix E.5).
b) FID continuous analyzer method (see Section 3-2 and
Appendix E.6).
c) GC/FID Method (see Section 3-3 and Section 6-2).
d) Wet impingement method (See Section 3-4 and Section 6-2).
A detailed discussion of each method follows, and an overall
comparison of results is presented in Table 3.1.
-------
Table 3-1. METHOD COMPARISONS
Location
North Stack
North Stack
South Stack
Basement A
Basement B
Basement C
Dope Prep "C"
Date
10/28
10/29
10/29
10/29
10/29
10/30
10/31
Time
1630
1015
1700
900
1045
900
930
FID
(ppmv)
__a
29.1
15.0
12.0
18.1
6.4
69
NIOSH
(ppmv)
10.79
12.59
NDC
10.74
18.24
2.66
35.78
Wet impingement GC/FIDb
(ppmv) (ppmv)
3.2
17.4
9.41
44.8
25.3
<0.1
NDC
Data biased by H20 condensation.
^Column plugged no peaks apparent.
"Not determined.
3-2
-------
3.1 NIOSH METHOD S-254
The NIOSH method was the only method readily available from the
literature and was used during the presurvey visit. The method is
contained in Appendix D.I. Results from a sample taken during the
presurvey visit revealed no measurable quantity of DMAC, apparently due
to small sample size. It was decided that during the test a larger
sample would be taken. A one-half hour sample at maximum flow (^250 ml/min)
was obtained during the field testing. The results of the basement
vents appear to correlate to the FID results quite well while the results
at the main spinning stacks are low. These low results could be due to
the sampling position of the silca gel tube in the stack. At the main
spinning stacks, the NIOSH tubes could not be securely placed deep into
the stack because of the port configuration. The results from the
locations sampled by this method are presented in Table 3-2.
3.2 FID CONTINUOUS ANALYZER (SEE ALSO SECTION 6)
The gas from the purchased DMAC gas cylinder standard (uncertified)
was introduced to the flame ionization analyzer with favorable results.
The response factor of the standard was approximately that of propane.
Since DMAC attacks stainless steel, the instruments were modified by
removal of all excess stainless components (i.e., filters, valves) and
®
by replacement of these with Teflon -coated components. This appeared
to give good results both in the field and in the laboratory. The
advantages of the FID continuous analyzer are that it is continuous,
accurate, yields real time results and has a medium to low cost. The
OVA-FID is advantageous in producing more data to compare emission
points. This is evident in Table 3.3. The disadvantages of the FID
continuous analyzer are the necessity of having a gaseous phase sample
which requires heated analyzers and heated sample lines. Another disad-
vantage is the reliance on gas standards. No commercial manufacturer of
gas standards will certify a gas standard of DMAC. Therefore, the DMAC
standard was verified by an HPLC and the wet impingement method at the
TRW lab facility in Research Triangle Park, North Carolina.
3.3 GAS CHROMATOGRAPH/FID ANALYSIS
There were several reasons for attempting a GC/FID analysis of the
gas streams at Monsanto. The primary reason was to achieve another
3-3
-------
Table 3-2. NIOSH SAMPLING
Location
Stack "C"
North Stack
South Stack
Basement Duct C
Basement Duct B
Basement Duct A
North Stack
Run #
C-S-1
NSE-S-1
SSE-S-1
CBE-S-1
BBE-S-1
ABE-S-1
NSE-S-2
DMAC detected
(Mg)
0.98
0.27
N.D.a
0.07
0.40
0.25
0.31
Volume sampled
(ML)
8080
7389
7502
7751
6485
6860
7266
DMAC
Concentration
(ppmv)
35.78
10.79
N.D.
2.66
18.24
10.74
12.59
Emission rates
1.76
25.24
N.D.
1.90
6.21
4.85
29.45
Not determined.
-------
Table 3-3. OVA MONITORING
Location Type
Crimpers B
Crimpers F
Crimpers G
Crimpers H
Crimpers I
Exhaust Fan D
Spinning Room Exhaust
Tow Cooler
Exhaust Relief Vent
Source #
52
47
36
45
34
44
54
11
57
21
20
10
27
4
3
2
1
15
22
48
37
Concentration3
(ppm)
0.76
1.1
1.1
0.48
0.47
1.2
0.66
0.47
0.47
1.1
1.3
Not operating
3.1
5.3
1.9
1.1
0.94
0.48
0.4
0.6
0.5
ppmv as DMAC.
3-5
-------
method of analysis for DMAC. Another reason was to determine that [JMAC
was the only organic compound present in the emissions. The third
reason was to monitor for the presence of acrylonitrile. By utilizing
different columns, the second two goals were achieved (see Section (5.2.1).
The first goal was unobtainable due to a variety of reasons.
An attempt to pack a Teflon column with Carbopak C had been
successful in previous tests for dimethylformamide - a similar compound.
This column produced readings for the head space amounts of DMAC, there-
fore, it was assumed that this type of column would be capable of reading
smaller amounts of DMAC. At the time of the field test, the column
appeared to be inadequate as the smallest amount detectable was 1000 ppm.
An attempt to recondition the column gave similar results. Therefore,
since removing all the metal in the field GC was not achieved, it is
possible that the small injection amounts of low level DMAC were adsorbed
into the system. Glass columns and injection ports could have solved
this problem. However, none were available for field analysis.
3.4 WET IMPINGEMENT
This method utilizes the hygroscopic properties of the DMAC. The
solvent is extremely soluble in water. The basic method is EPA Method 4
with water as the collection media. In laboratory tests of gaseous
DMAC, the HPLC analysis showed all of the DMAC had been absorbed in the
first impinger. The field analyses of the silica gel backup, in all
cases, found no DMAC present. The field results for this method are
sporadic. An examination of the individual impinger results gives
evidence that there was contamination in two of the runs (see Table 3.4).
No determination has been made as to the source of this contamination.
This method has the advantage of utilizing water as a collection
media in a standard method sampling train. The analysis is done with
standards based in water which is stable and easily reproducible. The
disadvantage is that the analysis requires a research-grade HPLC which
is not universally found in all laboratories.
3-6
-------
Table 3-4. EXPERIMENTAL WET IMPINGEMENT METHOD
co
i
Concentration
detected
Location (mg/ml)
North Stack
North Stack
South Stack
1st Impinger
2nd Impinger
Basement A
1st Impinger
2nd Impinger
3rd Impinger
4th Impinger
Basement B
Basement C
0.014
0.102
0.044
0.016
0.137
0.011
0.005
0.139
0.132
N.D.C
Volume of liquid DMAC Volume of gas
sample (tug) sampled
(ml) Subtotal Total (scf)
27.0 4.04 1.25
30.0 3.06 1.83
32.0 1.41
33.0 0.53 1.93a 2.15
29.5 4.04
32.5 0.36 .
32.0 0.16 8.31° 1.92
27.0 3.75
38.0 5.02 2.08
—
DMAC
Concentration
(ppmv)
3.2
17.4
9.4
44.8
25.3
—
The concentration detected appears in impinger 1 and 3, therefore, suspect contamination.
""The concentration in impinger 4 is suspect contamination.
"Not determined.
-------
4.0 LOCATION OF SAMPLING POINTS
This section presents descriptions of the sampling locations with
sample points used during the emissions testing program at the Monsanto
synthetic fiber plant in Decatur, Alabama. Figure 4.1 shows a schematic
of the spinning roof and dope preparation roof vent plan where the
sampling was conducted.
4.1 NORTH AND SOUTH STACKS
Spinning exhausts 1-16 vented to the south stack, while spinning
exhausts 17-30 vented to the north stack. Each stack had identical
dimensions, including a cross-section diameter of 144 inches. A schematic
of one of the stacks, including traverse point locations, is present in
Figure 4.2. The sampling ports were located ten feet down from the top
of the stack. The ports were at 90 degree angles and were Tabled "South"
and "West." Continuous FID monitoring with heated lines were connected
to the west port. Samples for the integrated bag, NIOSH, moisture, and
experimental wet impingement methods were taken from the south port.
The sample was taken at the point inside the port with the highest
velocity during preliminary velocity traverses.
Condensate collected from the spinning exhausts drained into a
series of pipes which lead to a common point below the roof where a grab
sample was obtained. The analytical results are in Appendix C.I.
4.2 BASEMENT VENTS
Flow velocities at basement vents A, B, and C were determined by
anemometers. The 52" x 96" vent was divided into equal sections and
flow readings were taken at the center of the sections and averaged (see
Figure 4.3). Sample points for the Beckman 402 continuous monitor,
integrated bag samples, NIOSH, and experimental wet impingement were
taken at the point of average flow through Teflon sample lines.
-------
I
ro
30Ch
34D 29CP
13O HDCRX 27-30 28CH
HO panho
15D No.
16Q
12D 57°
17D 45O
18D s*
19D 1QB
Fan!
No
•Tnt.
Exhaust 20D
9® 52Q
,o"° '0
32« 47O Ffl*°
238 490
240 036
?*B .. .
Fanho
9«B . No-
44O
289
12V UnltO Fanh«
No.
49O
Exh. Fan
"'" n^'%J^
Brat.ClPLJ25rir
Heat 13 ^^
70- 24Dl-
_ 23CH
540 "0 o ao.
louse 20Di
•4 ,a
18Di
170*
16Di
ISCR
O 14Di
. Exh B ITTU
use 1 T-i
3 Hrrr-^ 12D1
HtalTZ^ 1]D|
usi ' ««-h
2 TTll
6|§1
5*
59 Qa
3D-
l" «• Exh.C ^
K J i&
g)56 |
a L*"
/^ North
V^Stack
Room Vent A
27O I
4O
Room Vent B
Room Vent C
26® 37Q [
309 (fV""*h
Wstack
aO
D -o
HIV Unit
aO
^
Oa Not Tested
(g) Operates Periodically
• Not Operating
Filter Press Hood I
Exhausts
V«L
if \T
"u 1
r° r°
ir r
D. ft
k) fo
r if
0. ft
r° 1°
jr f
Vent
•• OA a
•
Room OB
""" % ««" „ ...
OE Vent
Spinning Roof Vent Plan
Figure 4.1. Aerial view
Dope Preparation Roof Vent Plan
-------
co
Traverse
Point
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Traverse
Point Location
From Inside
of Wall (In.)
1.6
4.6
7.9
11.4
15.1
19.0
23.2
27.9
33.1
39.2
46.5
57.3
86.7
97.5
104.8
110.9
116.1
120.8
125.0
128.9
132.6
136.1
139.4
142.4
T
10'
30'
Roof
t
Flow
Stack Inlet
Ports
144"
South
West
Condensate
FIGURE 4.2 SOLUTION STACK SCHEMATIC
-------
Blower
Rool
©
96"
Exhaust Blower
Section X-X
Showing Location of Sample Points
Figure 4.3. Basement vent
-------
4.3 DOPE PREP ROOM VENTS AND STACKS
Prior to entering the spinning room, the fibers pass through a dope
preparation room. The fugitive emissions from the dope prep process
were fan forced from the room out of three room vents. The flows of the
room vents were taken by anemometers. The vents had a diameter of 3'7".
Twelve flow radings were taken from a 3 x 4 matrix as presented in
Figure 4.4. The Beckman 402 continuous monitor sample point was at the
center of flow.
The dope prep stack flows were determined by an anemometer flow
taken at Stack B in Figure 4.1. The dimensions of Stack B was a diameter
of 23" with a cross bar of 3" in the center. Eight points were used for
readings (see Figure 4.5). The flows on the other stacks were estimated
to be the same because of their same approximate size. Continuous FID
data was taken at the point of average velocity on stacks B, C, and D.
4.4 CRIMPER EXHAUSTS
After the fiber leaves the spinning room, it goes through a crimping
process and emissions exit through the crimper exhausts. Type B had two
dampers and two directions of flow. One was out the top of the cylindrical
vent with a 45" diameter, above the dampers. The other was out the
bottom, below the dampers. Ten anemometer reading points were taken at
the top and bottom of the vents (see Figure 4.6). Continuous monitor
and OVA readings were taken from the emissions exiting through the top
of the vent.
Type F exhausts had twelve anemometer reading points measured at
the top of a 50" diameter vent (Figure 4.7). FID and OVA data was taken
from emissions exiting the top of the vent. The fan was a butterfly
type with an area of 222 square inches, which was subtracted from the
area calculated in the flow results.
Type G exhausts had a rectangular screen with dimensions of
31 1/2" x 15 3/8". Anemometer readings were taken from eight points in
a 2 x 4 matrix. The flow area allowed for 10 percent screen blockage,
and the fan was a blower type (Figure 4.8). OVA data was taken at the
point of average velocity.
4-5
-------
3'7"
Showing Location of Sample Points
Room Vent
Figure 4.4. Dope prep room vent
-------
r
23"
Roof
Exhaust Stack
Section X-X
Showing Location of Sample Points
Figure 4.5. Dope prep stack
-------
45"
00
r
45"
Dampers
Sample Points
1-10
Sample Points
11-20
Roof
Exhaust Fan
Section X-X
Showing Location of Sample Points
Figure 4.6. Crimper exhaust vent, Type B
-------
50"
r
50"
Dampers
~\
Roof
Exhaust Fan
Section X-X
Shewing Location of Sample Points
Figure 4.7. Crimper exhaust vent, Type F
-------
Blower
Roof
15.375"
31.5*
Exhaust Blower
Section X-X
Showing Location of Sample Points
Figure 4.8. Crimper exhaust vent, Type G
-------
Type I exhausts had a rectangular screen with dimensions of 12" x
23". Anemometer readings were taken at eight points in a 2 x 4 matrix.
The flow area allowed for 10 percent screen blockage. The fan was also
a blower type (Figure 4.9). OVA data was taken at the point of average
velocity.
Type H exhaust vents had a rectangular screen measuring 9 1/2" x
17 1/2". Anemometer readings were taken at six points in a 2 x 3 matrix
(Figure 4.10). The fan was a blower type. OVA data was taken at the
point of average velocity.
4.5 EXHAUST FANS
All the exhaust fans were a Type D bell-shaped vent with a diameter
of 60 inches. The exhaust fans were vented out the bottom of the bell.
Thirteen anemometer readings were taken around this area (Figure 4.11).
FID data was taken here also.
4.6 SPINNING ROOM CEILING EXHAUST
All were the same type. Anemometer readings were taken at twelve
points around the dampers. The vents were cylindrical, having a diameter
of 56 inches and a height of 37 inches. Both FID and OVA data were
taken as indicated (Figure 4.12).
4.7 TOW COOLERS
All the tow coolers were of the E type, having a rectangular screen
vent measuring 9" x 11" with a fan type blower. Six anemometer readings
were taken in a 2 x 3 matrix. The area used to calculate the flow
values allowed for 20 percent screen blockage for the screen that
covered the vent (Figure 4.13). FID data was taken.
4.8 EXHAUST RELIEF VENTS
Exhaust relief vents were new vents, Type A design. They had
blower type fans and had rectangular screens with dimensions of 44" x
29 1/2". Anemometer readings were taken at twelve locations in a 3 x 4
matrix (Figure 4.14). OVA readings were taken at this location.
4-11
-------
ro
Roof
Exhaust Blower
12*
23"
Section X-X
Showing Location of Sample Points
Figure 4.9. Crimper exhaust vent, Type I
-------
Blower
Roof
9.5"
3) (4
.5) (6
17.5"
Exhaust Blower
Section X-X
Showing Location of Sample Points
Figure 4.10. Crimper exhaust vent, Type H
-------
Roof
60"
Exhaust Fan
Section X-X
Showing Location of Sample Points
Figure 4.11. Exhaust fan vent
-------
in
Roof
Exhaust Fan
Section X-X
Showing Location of Sample Points
Figure 4.12. Spinning room ceiling exhaust vent
-------
Blower
Motor
Exhaust Blower
9"
ii"
Section X-X
Showing Location of Sample Points
Figure 4.13. Tow cooler vent
-------
B1ower
Motor
T) (2,
5) (?) C6
7) oy Q9^
fii) (12)
44'
Exhaust Blower
Section X-X
Showing Location of Sample Points
Figure 4.14. Exhaust relief vent
-------
5.0 PROCESS DESCRIPTION
Emission measurements were made at the Monsanto facility in Decatur,
Alabama, in order to obtain data necessary to develop a New Source
Performance Standard for the synthetic fiber industry. This plant is
considered to employ process technology representative of modern
synthetic fiber plants utilizing the wet spinning process.
Figure 5.1 presents a flow diagram of the wet spinning process and
indicates the location of the process emission sources. The key
presented in Table 5.1 groups the sample locations according to the
process emission sources. The source number from Table 5.1 is the
number identifying the vent locations on the spinning and dope prep roof
in Figure 4.1.
5.1 PROCESS EQUIPMENT
The wet spinning process produces filaments from polymer chips.
The polymer chips are dissolved in the solvent at the dope prep area in
solution vessels. This solution is extruded through spinnerets into
precipitation baths. The spinning area is hooded and exhausts to the
solution stack system. The ceiling exhausts were located at various
locations in the process to allow for fugitive emission collection. The
filament is drawn through a series of crimper machines and the exhaust
of each crimper machine is vented through separate duct systems to the
roof.
-------
To Dope Prep Stacks
en
ro
To North & South
Stacks
l
To Crimper
Exhausts
t
Heat
"FLFLF^
b b
Drier
\
Dope
Preparation
Area
Coagulation
Tank
Mash
Tank
FIGURE 5.1 FLOW DIAGRAM
-------
Table 5-1. SAMPLE LOCATIONS GROUPED ACCORDING TO PROCESS EMISSION SOURCE
Source
Process equipment
Source
Process equipment
01
00
Stack A
Stack B
Stack C
Stack D
Stack E
RMV A
RMV B
RMV C
South Stack
North Stack
1
2
3
4
9
22
26
27
28
30
37
39
46
48
49
56
(Dope Prep)
Room Vent
Room Vent
Hood Vent
Room Vent
Hood Vent
Room Vent
Room Vent
Room Vent
(Spinning)
Spinning Exhaust (1-16)
Spinning Exhaust (17-30)
Spinning Ceiling Exhaust #1
Spinning Ceiling Exhaust #2
Spinning Ceiling Exhaust #3
Spinning Ceiling Exhaust #4
(Exhaust Fan)
Exhaust Vent MC 18,19
Exhaust Relief Fan
Exhaust Fan MC 3
Exhaust Fan MC 16,17
Spinning C Exhaust
Exhaust Fan MC 2
Exhaust Relief Fan
Exhaust Fan
Exhaust Fan MC 1
Exhaust Relief Fan
Exhaust Relief Fan
Room Exhaust
10
11
20
21
34
36
44
45
47
52
54
57
12
13
14
15
16
17
18
19
23
24
32
6
7
(Crimper)
Crimper 21,23,23,24
Crimper 27,28,29,30
Crimper 17,18,19,20
Crimper 15,16
Crimper 27,28,29,30
Crimper 9,10,11
Crimper 1,2,3,4
Crimper 23,24,25,26
Crimper 12,13,14
Crimper 17,18,19
Crimper 20,21,22
Crimper 25,26
(Tow Cool)
Tow Cool 26
Tow Cool 30
Tow Cool 29
Tow Cool 28
Tow Cool 27
Tow Cool 25
Tow Cool 24
Tow Cool 23
Tow Cool 11,12
Tow Cool 9
Tow Cool 13,14
(Finishing)
MEF South Blower
MEF North Blower
-------
6.0 SAMPLING AND ANALYSIS METHODS
This section presents general descriptions of sampling and analytical
procedures employed during the emissions testing project conducted at
the synthetic fiber manufacturing facility of the Monsanto Company,
Decatur, Alabama, during October 27-31, 1980. Details of sampling and
analysis procedures are contained in Appendix D.
6.1 EPA REFERENCE METHODS UTILIZED DURING TESTING OF THE MONSANTO
FACILITY
The following EPA Reference Methods were used during this emission
testing program. These methods are taken from "Standards of Performance
for New Stationary Sources," Appendix A, 40 CFR, Part 60.
• Method 1 - Sample and Velocity Traverses for Stationary Sources.
This method specifies the number and location of sampling
points within a duct, taking into account duct size and shape
and local flow disturbances. In addition, this method
discusses the pitot-nulling technique used to establish the
degree of cyclonic flow in a duct.
t Method 2 - Determination of Stack Gas Velocity and Volumetric
Flow Rate. This method specifies the measurement of gas
velocity and flow rate using a pi tot tube, manometer and
temperature sensor. The physical dimensions of the pitot tube
and its spatial relationship to the temperature sensor and any
sample probe are also specified.
• Method 3 - Gas Analysis for CCs, Cs, Excess Air and Dry Molecular
Weight. This method describes the extraction of a grab or
integrated gas sample from a stack and the analysis of that
sample for C02, CO, 02.
-------
o Method 4 - Determintion of Moisture Content j_n Stack Gses,
This method describes the extraction of a gas sample from a
stack and the removal and measurement of the moisture in that
sample by condensation in impingers. The assembly and
operation of the required sampling train is specified.
6.2 MODIFIED EPA METHODS UTILIZED DURING TESTING OF THE MONSANTO FACILITY
The following Modified EPA methods were used during this emission
testing project. The Integrated Bag Method is a modification of EPA
Method 110 proposed in the Federal Register (April 18, 1980). The
continuous monitor system setup was based on EPA Method 25A -
Determination of Total Gaseous Organic Concentration Using a Flame
lonization Analyzer, which is a draft method.
6.2.1 Integrated Bag Method (Modified EPA Method 110)
6.2.1.1 Sample Method. Figure 6.1 illustrates the sampling apparatus
used during integrated bag sampling. The collection system consisted of
a can which seals at a vacuum of 15" of mercury (Hg), a bag evacuated to
29" Hg, a flowmeter for metering gas sample, and Teflon tubing to
serve as a sample line.
The procedure was to evacuate the can with the outside self-sealing
valve. After a vacuum was achieved, the can vacuum would be checked by
placing a vacuum gauge on the outside valve and monitoring the pressure.
The can would be considered leak free if pressure change was observed
over a ten-minute period. The second step was to evacuate the bag to
29" Hg. The same leak check was made on the bag as the can. The
flowmeter and can were then transported to the sample site and the
sample system assembled as in Figure 6.1. The sample probe was placed in
the sample port at the point of maximum velocity in the gas stream and
the sample valve opened at the appropriate time to extract the sample.
The proper flow was maintained with adjustment of the flowmeter. The
sample was taken over a 30-minute time period. The bag was capped and
transported to the on-site TRW Mobile Environmental Analysis Laboratory
for analysis.
6.2.1.2 Analytical Method for Hydrocarbons. Although several
analytical techniques were attempted in the field to analyze the
6-2
-------
Ttflw®
i
co
Bulkhead
Sampling
ClMr Htx1glM .
Lid JL HowMettr
Disconnect
CvacMtH
Cans
FIGURE 6.1 EVACUATED CAN SAMPLING SYSTEM
-------
integrated bag samples for DMAC, no reliable results were obtained.
Shimadzu Mini 2 was setup with two different column configurations in
attempting to detect DMAC. The system was also adapted with two other
columns to determine if any hydrocarbons were present besides DMAC.
The columns used for the detection of DMAC were a Carbopak C and
®
a blank Teflon column. The conditions of operation of the Shimadzu
Mini 2 were as follows:
A. DMAC ANALYSIS
Column: 6' x 1/8" Teflon® column packed with 0.2%
Carbowax® 20M on Carbopak® 80/100 mesh
Column Temperature: 96°C
Injection/Detector Temperature: 218°C
Flowrates: He - 31.7 ml/min
H2 - 32.4 ml/min
Air - 300 ml/min
B. DMAC ANALYSIS
Column: 6' x 1/8" blank Teflon® column
Column Temperature: 100°C
Injection/Detector Temperature: 330°C
Flowrate: He - 30.6 ml/min
H2 - 31.6 ml/min
Air - 312 ml/min
C. HYDROCARBON ANALYSIS Cg-C-^
Column: 6' x 1/8" stainless steel column packed with 0.1%
Supelcoport 1000 on Carbopak® C 80/100 mesh
Column Temperature: 62.4°C
Injection Detector Temperature: 330°C
Flowrate: He - 31.1 ml/min
H2 - 30.2 ml/min
Air - 306 ml/min
D. HYDROCARBON ANALYSIS - LOW MOLECULAR WEIGHT
Column: 8' x 1/8" stainless steel column packed with
Porapak Q®, mesh 80/100
Column Temperature: 108°C
6-4
-------
Injection/Detector Temperature: 241°C
Flowrate: He - 30.1 ml/min
H2 - 31.1 ml/min
Air - 300 ml/min
The pressures at the inlet of the gas chromatograph remained constant
for the four analytical methods at:
He - 4.2 kg/cm2
H2 - 0.61 kg/cm2
Air - 1.1 kg/cm2
The results on the gas chromatograph analysis were recorded and
electronically integrated with a Shimadzu CR1A Chromatopak®.
6.2.1.3 Analytical Method for Stationary Gases. Stationary gas
analysis for molecular weight determination was conducted in the field
on the integrated bag sample with the Shimadzu 3BT gas chromatograph/
thermal conductivity detector (GC/TCD). The chromatographic conditions
were:
o Range - 32 millivolts
o Column temperature - 32°C
o Injector/detector temperature - 120°C
o Carrier gas - Helium (He)
o Back pressure (He) - 3.0 kg/cm2
Two 6 x 1/8" O.D. stainless steel columns packed with Chromsorb 102
and molecular sieve were used in series to separate carbon dioxide
(CCO, oxygen (CO, nitrogen (N2), and carbon monoxide (CO). The analysis
was recorded by the Chromatopak . Stationary gas percentages were
calculated for each sample based upon calibration factors determined
daily from the commercially supplied stationary gas standards and the
GC/TCD. An example chromatogram is attached (Figure 6.2).
6.2.2 Continuous Monitor Utilizing Flame lonization Detectors
EPA Method 25A was used in the detection of Total Gaseous Organics.
The equipment was prefabricated at the TRW facility in the Research
Triangle Park, North Carolina, before the test. The sampling systems
fabricated at the sample site were designed in three fashions. The
first sytems were set-up at sources of suspected high organic levels for
6-5
-------
CHROM ATOQRAPHIC CONDITIONS
Instrument: SMnadzu GC-3BT
Column - two 6' x 1/8" O.D. stainless steel 1n series packed with
Chromsorb 102 and Molecular Sieve
Range: 32 mw
Column Temp. - 32° C
Inj/Det Temp. - 120°C
Back Pressure - HE - 3.0 Kg/cm2
Carrier - Helium
USE - B - 1
Start 10.28.17.44
—I 0.25
0.46
r
1 I
r
'
1.56
D 1.81
Stop
C-RIA
Sapl I 00
File t 1
Kept I 167
Method 42
1
3
4
NAME
A1r
TIME
0.46
1.56
1.81
TOTAL
CONC
0
20.9978
79.0021
99.9999
MK
V
AREA
130165
24197
89922
244285
FIGURE 6.2 STATIONARY GAS CHROMATOGRAM
6-6
-------
long period sampling. The second systems were semi-portable sampling
set-ups to allow for short testing periods. The third system was a
portable system for an instantaneous readout at a specific time period.
Strip charts are presented in Appendix B.6.
6.2.2.1 Extended Test Period Continuous Monitor Systems
Fabricated at the North and South Solution Stack. The continuous monitor
systems located at the solution stacks are presented in Figure 6.3. For
detailed description, see Appendix D.3. The systems were fabricated
with 50 feet of heated sample line maintained at 200°F and attached to
the Beckman 400 hydrocarbon analyzers. The monitors were connected
into a recorder and calibrated during test runs. These systems were
designed for extended test periods over 24 hours. The operation and
calibration of the Beckman 400 is described in Appendix D.3.
6.2.2.2 Semi-Portable Continuous Monitor Systems Utilized at
the Basement Vents, Dope Prep Room Vents and Stacks, Crimper Exhausts,
Exhaust Fan, Spinning Room Ceiling Exhaust, and Tow Cooler Exhaust. Two
semi-portable continuous monitor systems were designed for short-term
testing of various sources at Monsanto. The first system was fabricated
for monitoring of the dope prep room vents and the filter press hood
vents. The system was designed with a Beckman 400 located centrally to
the position of the dope prep room vents. The semi-portable system was
set-up identical to the continuous monitor system at the solution stacks
except that the 50 feet of heated sample line was utilized to reach the
separate room vents from the central location.
The system set-up at the dope prep room vents was utilized at the
filter press hood vents, but due to the inconsistent timing of the
operation of the filter press hood vents, no data was obtainable.
The second semi-portable system was designed to roll about on the
spinning roof for monitoring of the basement vents, dope prep stack:;,
crimper exhausts, exhaust fans, spinning room ceiling exhausts, and a
tow cooler exhaust. The system was designed with a Beckman 402 on a
cart with wheels. This system allowed the hydrocarbon analyzer to be
rolled between sample locations without dismantling the system. This
allowed the system to run continuously whenever possible or to relight
6-7
-------
o>
I
00
Sample Probe
50 Ft. Heated Sample Line
Sample Pump
(Teflonboated Dlaphram)
o of n
Hydrocarbon
Free Air
FIGURE 6.3 CONTINUOUS MONITOR SET UP AT SOLUTION STACKS
-------
immediately to minimize warm-up time for the instrument. The system is
presented in Figure 6.4 with the calibration and operation of the
Beckman 402 described in Appendix D.3.
6.2.2.3 The Portable Analyzer Utilized for an Instantaneous
Concentration Readout at the Crimper Exhausts, Spinning Room Ceiling
Exhausts, and Exhaust Relief Vents. The portable system was utilized to
screen each type of emission source before the continuous monitor systems
were set-up and to obtain instantaneous concentration levels of sample
locations similar to vents tested with the semi-portable FID. The
system was a portable Century Systems Organic Vapor Analyzer (OVA)
hydrocarbon detector. The system was calibrated and operated according
to the method described in Appendix E.3.
6.2.2.4 The Calibration Gas System Designed for Transporting
Between Continuous Monitor Sample Locations. The calibration gas system
was designed with the ability to be transported from the different
locations on the spinning roof. The gas system was moved from the
different sample locations for calibrations, drift checks, the
recalibration of the Beckman 400s and the Beckman 402. The mobility of
the calibration system was achieved by the use of two cylinder carts.
One set of calibration gasses consisting of 98.5 ppm C^Hg and zero gas
were placed on one cart, while the second set consisting of 5.04 ppm
CgHg and zero gas were placed on the second cart. The calibration gases
were certified by the supplier to ±2 percent. The carts were transported
to the instrument that was operating at these particular ranges.
6.3 DMAC SAMPLIING AND ANALYSIS BY THE NIOSH METHOD NUMBER S254
6.3.1 Sampling Method
DMAC in the north and south solution stacks from the spinning
exhausts, the basement vents, and the stack "C" from the dope prep room
was sampled by the procedure described in the National Institute of
Occupational Safety and Health (NIOSH) Method Number S-254. The sample
train was a Sipin personal sampling pump with an in-line silica gel
tube. The tubes were supplied by Environmental Compliance Corporation
of Venetia, Pennsylvania, in accordance with the guidelines set forth in
the NIOSH method. The personal sampling pump was calibrated before the
6-9
-------
Sample Pump
(Tefloi© Coated Diaphram)
O O ffc I I
Sample Probe
FIGURE 6.4 SEMI-PORTABLE CONTINUOUS MONITOR SYSTEM
-------
test trip. The silica gel tube was placed in the gas stream and sample
was drawn for a 30-minute time period. Problems were encountered at
some sampling locations with the tubes clogging. These difficulties are
recorded on the field sheets which appear in Appendix B.4. The silica
gel tubes were capped-off after sampling, labeled, and stored under
refrigeration until transported back to the TRW laboratory facility at
Research Triangle Park, North Carolina for analysis.
6.3.2 Analytical Method
The analytical procedures for the DMAC study of the silica gel tube
sample were carried out according to the procedures described in NIOSH
Method S-254 found in Appendix E.I. The tubes were extracted with methanol
and 2 ml aliquots of the recovered sample were injected into a Varian 3700
gas chromatograph. The sample peak areas were compared against peak
areas from standard concentrations of DMAC. Peak area values were
recorded on the Shimadzu CR1-A Chromatopak . The results appear in
Appendix C.2. A sample chromatogram appears in Figure 6.5.
6.4 DMAC SAMPLING AND ANALYSIS BY THE EXPERIMENTAL WET IMPINGEMENT
METHOD
6.4.1 Sampling Method
During the preliminary method development phase of the synthetic
fibers emissions testing program, the high solubility of DMAC was noted.
Consequently, it was proposed to trap DMAC 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 Teflon sample probe, and
cleaning of all 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 H^O. The fourth
impinger was empty, and the fifth impinger held 15 grams of non-
indicating ACS reagent grade silica gel.
Recovery of the impinger solutions was performed with distilled
water. The flow rate of the sample train was 8 to 10 standard cubic
feet per hour (scfh). The sampling time was coincident with the silica
gel and integrated bag sampling methods at the north and south solution
6-11
-------
CHROM ATOGR APHIC CONDITIONS
Column - 3' x 1/8" Teflon packed with 0.21 CW 20 m on Carbopack C 80/100
Range: 10" /Attepvator: x 128
Column Temp. - 100°C Inj/Det Temp - 290° C
Flow Rate - He - 28 ml/min
Air - 300 ml/min
H2 - 30 ml/min
Back Pressure - He - 24 psig
02203 Front
02203 Front
Start 11.11.13.54.
Stop
C-RIA
SMPL #
File #
Rept #
Method
00
2
346
41
NAME
Methanol
DMAC
TOTAL
TIME
0.3
CONC
99.43
0.5699
99.9999
MK
E
T
AREA
40830388
234032
41064420
FIGURE 6.5 NIOSH METHOD CHROMATOGRAM
6-12
-------
stacks from the spinning exhausts and the basement vents. Field data
sheets recording the sample train operation are in Appendix B.5.
6.4.2 Analytical Method
The impinger solutions collected in the field for the experimental
wet impingement sampling train were kept refrigerated in the mobile
laboratory at subambient temperatures until sample analysis at TRW's
Research Triangle Park analytical laboratory. All field samples were
logged and recorded in the appropriate laboratory notebook. Liquid
volumes were measured and a two (2) ml aliquot of each sample was taken
for HPLC analysis. HPLC results are in Appendix C.I.
A Varian Instruments Model 5061 HPLC was utilized for the separation
and detection of the dimethyl formamide. The HPLC was coupled to a
variable wavelength UV-Visible detector. The chromatographic record of
the analyses is contained in Appendix C.I. An example chromatogram is
presented in Figure 6.6.
6.5 QUALITY CONTROL AND QUALITY ASSURANCE
A quality control measure for monitoring the progress of the project
was the maintenance of instrument and field analytical notebooks. The
notebook documentation is on file and available through the field project
manager. The certification of the propane gas standards used on the
test program were performed by the supplier, Scott Environmental
Technology, Inc. The analytical report for the calibration gases is
presented in Appendix E.I. A problem arose when the Scott Laboratory
could not perform certification tests on the DMAC gas standards. The
uncertified DMAC gas standards were certified by the TRW Analytical
Laboratory facility at Research Triangle Park, North Carolina. The
certification procedure performed was the collection and analysis on a
Varian HPLC instrument. The certificaton test results are presented in
Appendix E.2.
The calibration of the personal sample pumps utilized in the NISOH
sampling method was performed prior to the testing period at the TRW
facility. The calibration procedure was performed according to NIOSH
method recommendations. The results of the calibration of the two pumps
used at Monsanto are provided in Appendix E.3.
6-13
-------
CHBOMATOQRAPHIC CONDITIONS
InstruMnt - Varian 8700 HPLC
Coluain - NCH - 5 R.P.
Mobile Phase
MX H,0
70X Acetonltrlle
Flow Rate • 0.5 »l/Bln
Detector wavelength - 240 tm
DMAC
1
PK
12 P
TIME
3.70
AREA
13008
AREA %
17.48
Rack 1
File 1
V1al 13
Inj 1
ID f 65
FIGURE 6.6 WET IMPINGEMENT METHOD - HPLC CHROMATOGRAM
6-14
------- |