xvEPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 78-OCM-3
March 1979
Air
Benzene Organic
Chemical Manufacturing
Emission Test Report:
Ethylbenzene/Styrene
Cos-Mar
Carville, Louisiana
-------
SOURCE TEST AT COS-MAR'S
ETHYLBENZENE/STYRENE PLANT
CARVILLE, LOUISIANA
Contract No. 68-02-2812
Work Assignment 11
EPA Technical Manager: Winton Kelly
Prepared for:
U.S. ENVIRONMENTAL PROTECTION AGENCY
.Emission Standards and Engineering Division
Emission Measurements Branch
Research Triangle Park, North Carolina 27711
TRW
ENVIRONMENTAL ENGINEERING DIVISION
-------
TABLE OF CONTENTS
Section Page
1.0 INTRODUCTION 1
2.0 SUMMARY OF RESULTS -2
3.0 PROCESS DESCRIPTION 22
4.0 LOCATION OF SAMPLING POINTS 23
5.0 SAMPLING AND ANALYSIS PROCEDURE 29
APPENDICES
A. COMPLETE RESULTS AND CALCULATIONS 36
B. GAS CHROMATOGRAPH RESULTS 41
C. LABORATORY RESULTS • 63
D. SAMPLING PROCEDURES 115
E. FIELD DATA SHEETS. '. 121
F. TEST LOG 154
G. PROJECT PARTICIPANTS 157
-------
LIST OF FIGURES
Number Page
2.1 Benzene Drying Column Vent Equipment 6
2.2 Modified Moisture Train 7
2.3 Caustic Scrubber Equipment 8
2.4 Catalyst Mix Tank Equipment 9
2.5 Benzene/Toluene Column Vacuum Equipment 10
4.1 Benzene Drying Column Vent Equipment 25
4.2 Caustic Scrubber Equipment 26
4.3 Catalyst Mix Tank Equipment 27
4.4 Benzene/Toluene Column Vacuum Equipment 28
-------
LIST OF TABLES
Numbei
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
2.10
2.11
P
Results of Gas Sampling Analysis - Test Point #1 ....
Results of Gas Sampling Analysis - Test Point #2 ....
Summary of Results Species Anaylsis - Organic
Condensate at Points 1 and 2
Results of Gas Sampling Analysis - Test Point #4 ....
Results of Gas Sampling Analysis - Test Point #5 ....
Summary of Results Caustic Scrubber Liquid Analysis
(Point #6)
Results of Gas Sampling Analysis - Test Point #7 ....
Results of Gas Sampling Analysis - Test Point #8 ....
Results of Gas Sampling Analysis - Test Point #9 ....
Results of Gas Sampling Analysis - Test Point #10. . . .
Gas Flow Calculations for Locations 5, 7, and 10 ....
'age
11
12
13
14
15
16
17
18
19
20
21
-------
1.0 INTRODUCTION
Under the Clean Air Act, The U.S. Environmental Protection Agency
(EPA) is required to establish National Emission Standards for Hazardous
Pollutants for emissions that have been found to cause adverse health effects.
Benzene has been listed as a hazardous pollutant and studies have been ini-
tiated to develop background information. The test program at this facility
was conducted to collect emission data from ethylbenzene-styrene production
for these studies.
Testing was conducted at the Cos-Mar, Carville, Louisiana No. 2 Ethyl-
benzene-Styrene plant during June 19-30 and July 10-14, 1978, By TRW Environ-
mental Engineering Division personnel under contract to EPA. Testing was
coordinated and observed by a representative of Monsanto Research Corporation,
also under contract to EPA.
The purpose of testing was to obtain data before and after control
devices (if present) for total organics and specifically, benzene from the
following systems:
- Benzene Drying Column Vent Equipment
- Alkylate Degasser Vent Equipment
- Catalyst Mix Tank Equipment
- Benzene/Toluene Column Vent Equipment
The number of locations sampled, a system description, summary of
specific test methods, sample point locations, and a results presentation are
given separately in Section 2 of this report for each of the above systems.
General test methods are presented in Section 5. Detailed sample location
descriptions are included in Section 4. Complete data summaries, sample
calculations, and field data are included in appendices A through G.
-------
2.0 SUMMARY OF RESULTS
2.1 BENZENE DRYING COLUMN EQUIPMENT
The benzene drying column vent system is illustrated in Figure 2.1. A
condenser is used to further remove organics from the vent streams from the
column reflux condenser and reflux decanter. The exit stream from this
condenser is routed to the plant flare system. Sampling was conducted before ,
and after this condenser at Points 1 and 2 as indicated in Figure 2.1.
Gas samples were collected simultaneously into flexible plastic bags
(Tedlar) at each location. The system pressure (80 psig) was used to fill
the bags. Samples were extracted through existing side-tap valves on the
piping. During initial sampling attempts, condensate was observed in the
sample bags. In an attempt to avoid this problem, a condenser-knockout system
was added prior to the sample bag. This system consisted of the equipment
normally used for moisture determination operated in an ice bath (see Figure
2.2). This modification prevented the appearance of condensate in the sample
bags.
A two-phase liquid was observed in the condenser train after sampling.
The lower phase was water and was used for calculation of the stream
moisture content. The top phase represented condensed organics and, possibly,
liquid entrainment from the stream sampled. It was observed that liquids were
entrained in the flowing streams at both points. It is not possible to predict
how much of the collected condensate is attributable to trapped entrained liquids
and how much was due to condensation.
This liquid was recovered and analyzed. However, no attempt was made to
combine these results with the vapor sample results to obtain total stream
organic content.
Analyses were conducted on the bag samples to determine nonaromatics as
GI - GS (species) and aromatic compounds, specifically, benzene, toluene,
xylene, ethyl benzene, and styrene.
An orifice flow meter was located near'Point 2. This meter was to be
used for flow data, but the presence of entrained liquids in the stream
resulted in erratic differential pressure levels and prevented-the measurement
of flow rates.
-------
The results of gas sampling are presented in Tables 2.1 and 2.2. As
discussed previously, these data represent the vapor phase compounds present
after a condenser (i.e., impingers) and are potentially less than the concen-
trations existing in the streams. This would be due to the difference
between the source and the final condenser train temperature.
The vapor sampling results are presented for C-j - 65 nonaromatic
compounds and the indicated aromatics on the basis of benzene equivalents
and as the specified compound. Two results are presented for C4 because two
defined peaks were present with retention times near that of butane. Addi-
tional studies were not performed to identify these compounds specifically.
The results presented for total hydrocarbons by total HC analyzer are
not representative of true concentrations. Because of the high organic
concentration, the detector of the instrument was saturated. In other words,
the concentration was beyond the analyzer range, and all results represent
the upper limit measurable. A dilution apparatus was not available; therefore,
the sample concentration could not be reduced to a level where valid results
could be obtained on a total HC analyzer. The total hydrocarbons by summation
of the individual species results should represent the total organics concen-
tration in each stream.
On Run 2-2, the results for CB and C$ are significatly different than
the results for the other two runs. Analysis records were rechecked to
comfirm the results. No reason is known for the apparently different results.
The total organics in the condensate are presented in Table 2.3.
2.2 ALKYLATE DEGASSER VENT EQUIPMENT
The alkylate degasser vent equipment is illustrated in Figure 2.3.
The vent stream from the degasser is first scrubbed with a polyethylbenzene
solution and then is routed to a caustic scrubber along with a vent from
other equipment. During the test period, there was no flow from the other
equipment (Test Point 3); therefore, no measurements were performed at that
location. After caustic scrubbing, the vent stream is routed to the plant
flare system. The caustic solution charge to the scrubber vessel is on a
batchwise basis. During sampling, no fresh caustic was added.
Gas samples were collected from the streams before and after the caustic
scrubber. Grab samples of the caustic liquid were also collected during gas
stream testing. The grab samples of caustic liquid were analyzed for
-------
Na+ and Cl" and the results are listed in Table 2.6. Integrated samples were
collected in flexible bags for organic compound analysis and inert gas
analysis. In addition, samples were collected from the gas stream using a
condenser train for moisture content determination. Finally, gas samples were
extracted using the procedures given in Section 5 for HC1 determination. The
flow rate at the scrubber outlet was measured using a vane anemometer attached
to the outlet of the vent to atmosphere. (Note: For sampling purposes, the
scrubber outlet stream was not routed to the flare system. Instead, it was
bypassed through an existing vent to atmosphere.) The inert gas analysis can
be used to estimate the inlet flow rate using a nitrogen balance.
The results of gas stream analysis are summarized in Tables 2.4 and 2.5.
The results of caustic liquid analysis are presented in Table 2.6.
At Point 4, there are significant differences in the results for 04, C$,
benzene,and ethyl benzene between the two runs. Also, at Point 5, there are
significant differences in (4, C^s.and benzene. The analysis records were
rechecked to verify the results. One reason for the differences may be due to
the fact that the two runs were separated by about two weeks.
2.3 CATALYST MIX TANK EQUIPMENT
The catalyst mix tank vent equipment is illustrated in Figure 2.4. The
equipment included is the catalyst storage vessel and the mix tank where
catalyst is dissolved in a polyethlybenzene solution. The vent gasses are
essentially nitrogen purge streams that are used to inert the system. The
vent stream is scrubbed with fresh polyethylbenzene solution prior to discharge
to the atmosphere.
Samples were collected for organic species analysis and to determine HC1
content of the vent stream. The vent flow rate was determined using a vane
anemometer.
Sampling was conducted during catalyst mixing only, which is a batch
process activated on an as-needed basis.
The results of testing are summarized in Table 2.7.
2.4 BENZENE-TOLUENE COLUMN VACUUM EQUIPMENT
The vacuum equipment serving the benzene-toluene separation column is
illustrated in Figure 2.5. The vent stream from the reflux condenser and
accumulator is first passed through a brine chilled-condenser to remove
organics prior to the steam ejector. The ejector exit stream is passed
-------
through a vacuum condenser to a hotwell. The hotwell vent stream is then
passed through a final brine-chilled condenser prior to exhaust to the
atmosphere.
Samples were collected from the noncondensable stream (Point 8) prior
to the ejector for organic species analysis. Sampling for moisture content
was attempted but was not successful because of the vacuum present at this
location.
Samples were collected for organic species analysis, moisture content,
and inert gas analysis at the hotwell vent before and after (Points 9 and 10,
respectively) the final condenser. Flow rate measurements were performed
using a vane anemometer at the vent to the atmosphere (Point 10). Because
of the low flow rate and low pressure drop across the condenser, it was not
possible to sample simutaneously at Points 9 and 10. Therefore, these were
sampled sequentially.
The results of sampling are summarized in Tables 2.8 through 2.10.
There were significant differences in the results for Cs, 64, and C$ between
runs at Points 8, 9, and 10. The analysis records were checked to verify
the results. One reason for the differences may be that the runs were
sampled on different days.
-------
From
Other
Equipment
'••'- : ; Vent
Overheads ! •!- ;
1 :
OV
c
Ref 1ux
Condenser
C
Decanter Vent
Reflux 1
Reflux
Decanter
Water
Benzene Drv-ijig
Column :
;-...; .': f"-Q ' Test Location
Vent.To ;
J. Flare Manifold
cw
Condensate
Vent Condenser
FIGURE 2.1. 'Benzene Drying Column Vent Equipment.
-------
TO SAMPLE
LOCATIONS
;Teflon'Sample Line To
Rigid Container System-
©
GREENBURG-SMITH
IMPINGERS
FIGURE 2.2. Modified Moisture Train.
-------
Vent To Flare Mam'foLc
Caustic Feed
:(Batfch;
From Other
Equipment
(Off During Test)
Jest Location
From
Settling Tank
Caustic Scrubber
(ECM-002)
Degasser Vent
Scrubber
(ECM-001)
Caustic Drain
(Batch Empty)
Figure
2.3. • Caustic scrubber equipment. . ; : ;
• .' " i ' • '
Vent
Degasser
(AS-1008)
-------
Vent
Vent
Catalyst
Storage
Catalyst Mix
Tank
Catalyst Mix
Tank Scrubber
(ECM-003)
j Drain
To Sewer
Return From
Scrubber
Catalyst Mix
A Vent To
^-'Atmosphere
Drain To Sewer
(Capped During Test)
I Polyethy:! benzene
• Feed
f-0 :
Test Location
Figure 2.4. Catalyst mix tank equipment.
-------
15015 Steam
Noncondensables
iCW
Qverheads ^
1
Reflux
Steam
Ejector
Condensibles
Accumulator5
Vent Condenser
Brine
Reflux .
Condenser &
Accumulator
B/T Col umn
f— O • Test Location
Vacuum Condense?
To Main
Hotwell
CW
iBrine T
HotweU*
^
i Vent To
'Atmosphere
Hotwell Vent
Condenser
(ECM-005)
Condensate
Hotwell No. 1
Figure 2.5. Benzene/toluene column vacuum equipment.
-------
RUN NO.
DATE
TIME
SPECIES ANALYSIS
C-l
C-l
C-2
C-2
C-3
C-3
C-4
C-4
C-5
C-5
C-6
C-6
BENZENE
ETHYLBENZENE
TOTAL HYDROCARBONS
BY:
SPECIES SUMMATION
TOTAL HC ANALYZER
INERTS AND FLOW DATA
HgO, % by volume
N2, % by volume
02 > X by volume
C0£f % by volume
Flow rate, scfm,*dry
HC1 , ppmv
TOTAL
n/m = not measured
* = 20°C atm. , wet :
1A
7/13/78
1045-1055
ppmv as ppmv as
compound benzene
11,025 2,613
92,828 41,257
3,740 2,413
348,970 266,389
840,306 641,455
9,552 9,552
23,808 28,564
2«,185 28,185
1,358,414 1,020,428
n/m
0 0
n/m
IB
7/13/78
1203-1213
ppmv as ppmv as
compound benzene
8,925 2,155
94,160 41,849
3,857 2,488
381 ,957 291 ,570
854,196 652,058
7,671 7,671
30,087 36,097
13,529 13,529
18.71 12,86
1,394,401 1,047,389.86
n/m
0 0
n/m
1C
7/13/78
1437-1447
ppmv as ppmv as
compound benzene
7,350 1,742
77,040 34,240
3,390 2,187
347,230 265,061
774,332 591,093
9,408 9,408
42,645 51,164
28,185 28,185
1,289,580 983,080
n/m
0 0
n/m
Table 2.1. Results of°Gas-Sampling Analysis-Test Point #1
Vent Condenser Inlet
11
-------
RUN NO.
DATE
TIME
SPECIES ANALYSIS
C-l
C-l
C-2
C-2
C-3
C-3
C-4
C-4
C-5
C-5
C-6
C-6
BENZENE
ETHYLBENZENE
TOTAL HYDROCARBONS
BY:
SPECIES SUMMATION
TOTAL HC ANALYZER
INERTS AND FLOW DATA
H20, % by volume
NZ. * by volume
02 . * by volume
C(>2, % by volume
Flow rate, scfm.*dry
HC1, ppmv
TOTAL
n/m = not measured
* = 20°C atm. , wet
2-1
7/13/78
1045-1055
ppmv as ppmv as
compound benzene
73,500 17,417
194,027 86,234
2,104 1,357
149,377 114,028
291,677 222,654
2,316 2,316
17,660 21,188
16,911 16,911
747,572 482,105
n/m
n/m n/m
2-2
7/13/78
1203-1213
ppmv as ppmv as
compound benzene
. 78,750 18,661
178,809 79,750
2,992 1,885
164,936 125,905
307,302 234,582
50,657 50,657
746,935 896,143
23,675 23,675
1,554,056 1,430,979
n/m
n/m n/m
2-3
7/13/78
1437-1447
ppmv as ppmv as
compound benzene
68,250 16,173
155,982 69,325
2,571 1,659
147,574 112,652
281,260 214,702
2,460 2,461
37,085 44,493
23,675 23,675
578,857 485,140
n/m
n/m n/m
Table 2.2 Results of Gas Sampling Analysis- Test Point #2
Vent Condenser Inlet
12
-------
LOCATION
RUN NO.
COMPOUND
C5
BENZENE
TOLUENE
ETHYLBENZENE
STYRENE
VOLUME OF
CONDENSATE (ml)
GAS VOLUME
SAMPLED (liters)
CONDENSER INLET
1
9.7
90.3
<0.1
<0.1
<0.1
95
10.0
2
WEIGHT %
8.3
91.7
<0.1
<0.1
<0.1
92.5
10.0
(POINT 1)
3
12.9
87.1
<0.1
<0.1
<0.1
82
10.0
CONDENSER OUTLET (POINT 2)
1 2
WEIGHT %
(INSUFFICIENT
3 1.5
10.0 10.0
3
SAMPLE)
1.5 '
10.0
Table 2.3. Summary of Results of Species Analysis- Organic
Condensate at Points 1 and 2.
13
-------
RUN NO.
DATE
TIME
SPECIES ANALYSIS
C-1
C-l
C-2
C-2
C-3
C-3
C-4
C-4
C-5
C-5
C-6
C-6
BENZENE
ETHYLBENZENE
TOTAL HYDROCARBONS
BY:
SPECIES SUMMATION
TOTAL HC ANALYZER
INERTS AND FLOW DATA
HgO, % by volume
NZ» % by volume
02 t % by volume
C02> % by volume
Flow rate, scfm,*dry
HC1 , ppmv
TOTAL
n/m = not measured
* = 20°C atm., wet
4-1
6/28/78
1045-1100
ppmv as ppmv as
compound benzene
8,099 1,919
8,300 3,689
--
504 385
3,454 2,637
5,970 5,970
252,188 302,565
59,601 59,601
338,116 376,766
n/m
19.69
42.6
7.12
0
53.0
69.41
4-2
7/11/78
inn-inafi
ppmv as ppmv as
compound benzene
8,050 1,908
11,355 5,047
__
5,546 4,234
64,646 49,348
4,713 4,713
64,527 77,417
14,374 14,374
173,211 157,041
n/m
22.77
42.6
7.78
0
53.7
73.15
AVERAGE
ppmv as ppmv as
compound benzene
8,075 1,914
9,828 . 4,382
-r-
3,025 2,310
34,050 25,993
5,342 5,342
158,358 189,991
36,988 36,988
255,666 266,920
n/m
21.23
42.6
7.45
0
53.4
71.28
Table 2.4. Results .of.Gas Sampling Analysis - Test Point #4
Caustic Scrubber.Inlet
14
-------
RUN NO.
DATE
TIME
SPECIES ANALYSIS
C-l
C-l
C-2
C-2
C-3
C-3
C-4
C-4
C-5
C-5
C-6
C-6
BENZENE
ETHYLBENZENE
TOTAL HYDROCARBONS
BY:
SPECIES SUMMATION
TOTAL HC ANALYZER
INERTS AND FLOW DATA
H20, X by volume
N2» X by volume
02> X by volume
C02> X by volume
Flow rate, scfm*
HC1, ppmv
TOTAL
n/m = not measured
* = 20°C atm. , wet
5-1
6/28/78
1045-1100
ppmv as ppmv as
compound benzene
288 68
4,427 1,968
--
300 229
1,956 1,493
3,881 3,881
48,012 57,603
56,670 56,670
115i534 121,912
n/m n/m
17.3
43.2
3.08
176.1
0
63.58
5-2
7/1 1 / 7 A
1031 - 1046
ppmv as ppmv as
compound benzene
3.325 788
4,367 1,941
._
2,377 1,814
8,715 6,653
2.062 2,062
42,284 50,731
9,856 9,856
72,986 73,845
n/m . n/m
18.2
n/m
n/m
176.2
0
--
AVERAGE
ppmv as ppmv as
compound benzene
1 ,806 428
4,397 1,955
~
1,338 1,022
5,336 4,073
2,972 2,972
45,148 54,167
33,263 33,263
94,260 97,860
n/m n/m
17.75
43.2
3.08
176.2
0
64.03
Table 2.5. Results of Gas Sampling Analysis - Test Point £5
:> Caustic Scrubber Inlet
15
-------
RUN NO.
DATE/TIME
COMPOUND
(ORGAN ICS)
TOTAL
Na+,mg/ml
TOTAL
m-ppm by
Cl vv weight
1
6/28/78
161
34,000
2
6/28/78
159
3
6/28/78
bRun
43
6/28/78
BLANK RUN
0.03
<*Run 4 was used as a blank and calculated into the results,
"Insufficient sample collection.
TABLE 2.6. Summary of Results of .Caustic Scrubber Liquid
Analysis Test Point 6.
16
-------
RUN NO.
DATE
TIME
SPECIES ANALYSIS
C-l
C-l
C-2
C-2
C-3
C-3
C-4
C-4
C-5
C-5
C-6
C-6
BENZENE
ETHYLBENZENE
TOTAL HYDROCARBONS
BY_:
SPECIES SUMMATION
TOTAL HC ANALYZER
INERTS AND FLOW DATA
t^O, % by volume
N2» t> by volume
02 . % by volume
C02» % by volume
Flow rate, scfm*
HC1 , ppmv
TOTAL
n/m = not measured
* = 2QQC atm. , wet
7-1
6/21/78
1005-1020
ppmv as ppmv as
compound benzene
49.4 11.7
7.2 3.2
60.1 26.7
2.5 1.6
.70 .53
33.2 33.2
153.1 76.93
n/m
7.4
70.4
7-2
6/21/78
1030-1050
ppmv as ppmv as
compound benzene
27.8 6.59
7.0 3.11
65.5 29.11
1.2 ,77
.3 .23
57.2 57,2
159.0 97.01
n/m
7.0
74.2
7-3
6/22/78
inm.iT|i
ppmv as ppmv as
compound benzene
17,0 4.03
4.6 2.04
43.7 19.42
2.3 1.48
7-0 7,0
74.6 33.97
n/m
6.6
68.7
Table 2.7.
Results of Gas Sampling Analysis - Test Point #7
Catalyst Mix Tank'Vent
17
-------
RUN NO.
DATE
TIME
SPECIES ANALYSIS
C-l
C-l
C-2
C-2
C-3
C-3
C-4
C-4
C-5
C-5
C-6
C-6
BENZENE
ITTLJVI DCU7CUC
tlnlLbtNZtNL
TOTAL HYDROCARBONS
BY:
SPECIES SUMMATION
TOTAL HC ANALYZER
INERTS AND FLOW DATA
HgO, % by volume
NZ» * by volume
02, 2 by volume
CC>2, % by volume
Flow rate, scfin,*dry
HC1 , ppmv
TOTAL
n/m = not measured
* = 20°C atm. , wet
8-1
7/13/78
1202-1213
ppmv as ppmv as
compound benzene
78,750 18,661
34,240 15,218
6,848 3,044
18,701 12,065
15,626 11,928
6,513 6,513
54,114 54,114
214,792 121,543
n/m
n/m
8-2
7/13/78
1630-1631
ppmv as ppmv as
compound benzene
78,750 18,661
45,683 20,290
7,609 3,323
7,692 4,931
25,653 20,277
108 83
4,523 4,523
14,438 17,322
62,006 62,006
246,462 151,416
n/m
n/m
8-3
7/13/78
1654-1655
ppmv as ppmv as
compound benzene
115,500 27,370
50,409 22,404
9,987 4,439
22,675 14,629
32,466 24,783
174 132
5,138 5,138
103,341 123,984
36,076 36,076
375,766 258,955
n/m
n/m
Table 2.8. Results of.Gas Sampling Analysis - Test Point. #8
Noncondensables to Steam Ejector
18
-------
RUN NO.
DATE
TIME
SPECIES ANALYSIS
C-l
C-l
C-2
C-2
C-3
C-3
C-4
C-4
C-5
C-5
C-6
C-6
BENZENE
ETHYLBENZENE
TOTAL HYDROCARBONS
BY:
SPECIES SUMMATION
TOTAL HC ANALYZER
INERTS AND FLOW DATA
H20, % by volume
N2> X by volume
Q£ » X by volume
C0£t % by volume
Flow rate, scfm,*dry
HC1 , ppmv
TOTAL
n/m = not measured
* = 20°C atin. , wet
9-1
6/22/78
1631-1646
ppmv as ppmv as
compound benzene
34,528 8,182.
45,756 20,336
14,536 6,460
2,390 1,542
33,585 25,637
16,000 •• 16,000
85,245 102,273
27,245 27,245
259,285 207,675
28.41
29.0
6.25
21.45
n/m
85.11
' 9-2
6/23/78
0950-1005
ppmv as ppmv as
compound benzene
30,163 7,148
39,840 17,707
11,952 5,312
17,630 11,374
26,046 19,882
10,880 10,880
63,377 76,037
30,439 30,439
230,327 178,779
20.95
20.6
2.68
25.2
n/m
69.43
9-3
6/26/78
1058-1115
ppmv as ppmv as
compound benzene
34,630 8,206
37,840 16,818
8,632 3,836
22,766 14,688
39,196 29,921
9,355 9,355
24,427 24,427
176,846 107,251
57.12
n/m
Table 2.9 Results of Gas Sampling Analysis - Test Point #9
Inlet of Hotwell Vent Condenser -
19
-------
RUN NO.
5-1
5-2
10-1
10-2
10-3
7-1
7-2
7-3
AVERAGE
FT/MIN
2,123
2,083
137.0
86.7
99.3
87.2
82.4
78.4
Ts
95
84
80
70
75
85
85
85
AREA
ANEMOMETER
(F2)
.0873
.0873
.0873
.0873
.0873
.0873
.0873
.0873
FLOW
ACFM
185.3
181.8
12.0
7.6
8.7
7.6
7.2
6.8
SCFM
20°C, 1 atm
176.1
176.2
11.6
7.6
8.7
7.4
7.0
6.6
DSCFM
145.7
144.2
9.6
6.4
7.1
Table 2.11. Gas Flow Calculations Test Points 5, 7, 10.
20
-------
RUN NO.
DATE
TIME
SPECIES ANALYSIS
C-l
C-l
C-2
C-2
C-3
C-3
C-4
C-4
C-5
C-5
C-6
f.-6
BENZENE
ETHYLBENZENE
TOTAL HYDROCARBONS
BY:
SPECIES SUMMATION
TOTAL HC ANALYZER
INERTS AND FLOW DATA
HgO, % by volume
N2» I by volume
02 t % by volume
COg. % by volume
Flow rate, scfm *
HC1 , ppmv
TOTAL
n/m = not measured
* = 20°C atm. , wet
10-1
6/22/78
1631-1646
ppmv as ppmv as
compound benzene
37,667 . 8,926
46,849 20,822
13,998 6,221
2,410 1,555
36,049 27,518
15,543 15,543
37,374 44,840
21 ,608 21 ,608
211,498 147,033
16.07
19.7
5.54
18
11.6
59.31
10-2
6/26/78
0920-0935
ppmv as ppmv as
compound benzene
29,002 6,824
39,176 17,412
10,624 4,722
17,630 11,374
25,507 19,471
4,960 4,960
73,242 87,873
32,694 32,694
232,835 185,330
17.82
35.3
8.12
19.5
7.6
80.74
10-3
6/26/78
1015-1040
ppmv as ppmv as
compound benzene
26,880 6,370
29,216 12,985
7,968 3,541
16,126 10,404
24,547 18,738
5,748 5,748
No Data No Data
No Data No Data
110,485 57,786
19.1
n/m
n/m
n/m
8.7
. 76.89
Table 2.10 Results of Gas -Sampling Analysis - Test Point #9
Outlet from Hotwell-Vent Condenser
21
-------
3.0 PROCESS DESCRIPTION
The Cos-Mar plant has two integrated ethylbenzene/styrene units
producing ethylbenzene by benzene alkylation and styrene by ethylbenzene
dehydrogenation. Both units use the conventional process technology
described in available literature. The testing was performed on the No. 2
unit, which had a rated capacity of 700(1.0)6 ib/yr styrene. During all
periods of actual testing, the No. 2 unit was operating between 90 and 100
percent of its capacity.
22
-------
4.0 LOCATION OF SAMPLING POINTS
The locations from which samples were collected in each system tested
have been identified in Chapter 2. In this section, the physical config-
uration of each location is described and the general stream characteristics
are discussed. The location identification numbers correspond to those
referenced throughout this report.
4.1 BENZENE DRYING COLUMN VENT EQUIPMENT
(Points 1 and 2 on Figure 2.1)
Both points on the benzene drying column were at elevated pressure.
Point 1 was approximately 90 PSIG and Point 2 was 63 PSIG. All pipe was
schedule 80 pipe. The pressure of the stream was used to push the sample
throught the sampling apparatus. No pump was needed. As can be seen from
Figure 4.1, the plant had installed a flow measuring orifice.in the pipe
before Point 2. This orifice was ineffective due to the entrained liquid
discussed earlier.
Prior to sampling, the residual liquid in both Points 1 and 2 was
purged for a minute or two to purge accumulated liquids from the sample
taps. The sample line was then connected and the sample was taken.
4.2 CAUSTIC SCRUBBER EQUIPMENT
(Points 4, 5, and 6 on Figure 2.3)
Points 4 and 5 represent the inlet and outlet of the caustic scrubber.
Both.Points 4 and -5 were under slight positive pressure. Normally, the.
scrubber outlet stream is routed to the plant flare system; however, for
testing, this stream was diverted through an existing vent to atmosphere.
A sampling line was connected directly to Point 4 while it was inserted into
the 2-inch diameter hole for Point 5 (Figure 4.2). Point 6 was a liquid
sample point.
23
-------
4.3 CATALYST MIX TANK EQUIPMENT
(Point 7 on Figure 2.4)
Point 7 was the vent for the catalyst mix tank. The vent was at the
70-foot level and is normally exhausted 5 feet in the air from the platform.
At TRW's request, the vent was manipulated so the vent end was accessible
to the samplers on the platform. The pipe was 4 inches in diameter and was
sampled by inserting the sampling line into the open pipe (Figure 4.3).
Before sampling could occur, the line below (drain to sewer) had to be
closed to force the flow up the vent pipe.
4.4 BENZENE/TOLUENE COLUMN VACUUM EQUIPMENT
(Points 3, 9, 10 on Figure 2.5)
Point 8 was on the noncondensable line and was normally under a vacuum
of 27 inches of mercury. An existing valve was used to extract the sample.
Points 9 and 10 were at essentially barometric pressure. An existing
valve was used to extract a sample at Point 9, and the sample line was
inserted into the open pipe at Point 10. See Figure 4.4 for the equipment
setup at Points 8, 9, and 10.
24
-------
1/40' Above Ground
Reflux &
Decanter Vent
30' Level ..
Teflon Sample Line .
2" Di.a.
To Flare
Teflon Sample Line
25
From
Vent
Condenser
FIGURE 4.1-- Benzene drying column vent equipment
-------
0
85' Level
From
Scrubber
Degasser Vent
b=o
16"
3" Dia.
Valve
Sample Line
V Teflon
\ ...To Caustic Scrubber
Pipe! [
••Sample
Line
Valve
Liquid Sample Valve
From Caustic
Scrubber
To Air
Caustic Drain
26
Anemometer
Caustic
Scrubber
FIGURE 4.2 Caustic'scrubber equipment.
-------
Sample Line
70'
Ground
4" Pice
A»«»'""M>t*
To Air
iFrom Catalyst
Mix Tank
Anemometer
27
FIGURE 4.3.. .Catalyst-mix tank equipment.
-------
Valve
emp L
[idicator
t
From
Hotwel1
If
j Sample
/ Line
Y^? "Anemometer
Sample Line
3"
27"
Did.
Vacuum . "'
Jo Hotwell
Condenser
Vent To
Atmosphere
- 2" Dia.
Brine Inlet
1" Pipe
Hotwell Vent
Condenser
(ECM-005-V>
FIGURE 4.4. Benzene/toluene column vacuum" equipment.
28
-------
5.0 SAMPLING AND ANALYSIS PROCEDURE
5.1 BAG SAMPLING AND ORGANIC SPECIES ANALYSIS
The sampling system used for collection of a gas sample into a flexible
container is shown in Figure 5.1. In all cases the flexible bag was placed
inside a rigidly sealed container. The methods used to fill the bags
depended on the pressure of the source gas. At Points 1 and 2, bags were
filled by source pressure. At Points 4, 5, 7, 9, and 10, an explosion-proof
pump was used to evacuate the rigid container to about 20" Hg vacuum. At
Point 8, the rigid container was evacuated to approximately 29" Hg vacuum
prior to sampling. The container was evacuated and leak checked before
being placed at the test site. Once at the test site, the Teflon sampling
lines were connected to the sample points, and the values were opened slightly
to produce the desired sampling rate (2 liters per minute). Once a sufficient
sample was extracted, the container was removed to the laboratory where it
was immediately given to the analyst for processing on the gas chromatograph.
Sampling rates were set based on flow valve setting calibrations prior to
testing. In some cases, the variability in source pressure (absolute) rigid
container flow restriction or imprecision of initial flow settings prevented
continuation of sampling for the desired interval. Because of these variations,
the duration of actual sampling varies from run to run.
In all tests except those at Points 1 and 2, the sample line was connected
directly to the valve or open pipe that was to be sampled. At Points 1 and 2,
the bag sampling systems were connected to the outlet of a condenser train
where a side stream was removed.
29
-------
The bag was run on four separate gas chromotagraphs;
1) Benzene and higher molecular weight hydrocarbons.
-Dual FID Shimadzu GC Mini 1 with 5 percent OV-101 and Bentone
34 - 100/120 mesh
2) Total hydrocarbons
-AID Portable FID GC with no column
3) Cj - C6 (Low molecular weight HC).
-Dual FID Shimadzu GC Mini 1 with a poropak Q column.
4) Stationary Gases - 02, N2, C02, CO
-Shimadzu 3BT Dual Thermal Conductivity with a Chromosorb 102 and
molecular seive 5x columns.
The gases were run on these instruments consecutively and the results
were recorded on a linear recorder with a integrator. In all cases, the
analysis was done within 4 hours of sampling.
The audit gases were supplied by EPA and were the only standards avail-
able during the first week of testing. After the first week, standards "for
benzene and toluene were shipped to the site. As can be seen from the audit
gas report, the concentrations of the audit gases and the standards were
very close. The ethyl benzene and styrene standards were not available during
the test period. Liquid samples were used to determine the retention time.
This provided information as to the presence of ethyl benzene and styrene but
was not definitive as to the amount.
5.2 MOISTURE
A condenser train (Figure 5.2) was used to obtain moisture results.
The procedures used are described in EPA Method 4 (40CFR60 Appendix B)
except that larger impingers were used, followed by a silica gel absorbent.
Concurrent with collection of the water in the samples, organics were condensed
as evidenced by the formation of a lighter liquid layer over the water in the
impingers. The two liquid layers were separated, with the water phase being
used for moisture content in the source stream and the organic layer being
retained for analysis. At Sample Locations 4, 5, 7, 8, 9, and 10 the organics
collected were analyzed and reported in Appendix C, but are not reported in
the summary of results because the organics were included in the integrated
bag sample results.
At Points 1 and 2, the bag samples were extracted after the condenser,
and thus the collected organic condensate is a portion of the sample. The
contribution of the liquids is summarized in Chapter 2.
30
-------
5.3 HYDROGEN CHLORIDE
HC1 was determined at Points 4, 5, and 7 using IN NaOH as a collection
media and a silver nitrate/potassium thiocyonate titration analysis. The
complete procedure is included in Appendix E. During collection of the
sample, using a modified EPA Method 6 train, organics were condensed. Analysis
for HC1 was performed on the water fraction after separation of the phases.
No attempt was made to analyze the organic layer for HC1. To determine if
dissolved organics in the water would interfere with HC1 determination,
benzene was added to blank solutions and analyzed. No interference was
observed.
5.4 ORGANIC LIQUID SAMPLE ANALYSIS
The samples were run an1 a Shimadzu 6AM gas chromatograph equipped with dual'
column FID with a two-meter glass column packed with 3-percent OV-101 on 80/
100 mesh Gas Chrom R. A linear integrating recorder was used to record data.
Standards of the aromatics of interest were run in the concentration
range of 1 percent, 10 percent, and 50 percent in pentane. Each concentration
is run at an appropriate range and sensitivity of .32V and 103 for the 50
percent. The resulting integrated area for each standard in each concentration
range was obtained from the chart recorder. The integrated area for each
aromatic of interest for each sample was obtained. A linear relationship
between area and concentration is assumed for each concentration range (0.1 to
4 percent), (4 to 20 percent), (20 to 100 percent) and the following relation-
ship used to calculate the unknown concentration:
. . . ... /<*\ = Area of Unk x Cone, of Std.
Concentration of unknown (%) Area of Std.
The samples were stored for approximately four months prior to analysis.
A long sample holding time and room temperature storage introduces several
uncertainties into the analysis for aromatic compounds.
Aromatics are only slightly soluble in water, so aqueous samples held in
nongas-^tight containers at ambient temperatures are subject to possible
losses. Similar problems are possible in the nonaqueous samples since losses
of components are a function of their vapor pressures at ambient conditions.
Thus, losses are more pronounced in the lower boiling constituents, making
relative concentrations change in the total sample.
Chemical changes, especially in unsaturated compounds like styrene, can
result in oxidation. This can further affect relative concentrations since
the oxidation products are not extracted as efficiently in extraction steps.
31
-------
Samples such as those from Point 8 containing less than 5 milliliters
are particularly susceptible to headspace losses because the large gas-to-
liquid ratio in the container allows saturation of a larger gas volume with
vapor. This could change the relative concentrations in a small liquid volume.
5.5 CAUSTIC SOLUTION SAMPLE ANALYSIS
The liquid collected at Point 6 was analyzed by the procedure described
above for organics content.
Total sodium and chloride were determined using atomic absorbtion.
5.6 VOLUMETRIC FLOW RATE
A Rochester G-694 vane anemometer was used to determine the velocity of
the vent streams at Points 5, 7, and 10. Since the pipes were smaller in
diameter than the anemometer, an expansion adaptor was fabricated. (The area
of the anemometer face was used to calculate volumetric flow rate.) The
calibration data show that the length of the expansion section was not enough
to have fully developed flow at the anemometer; therefore, a correction was
used. The anemometer was calibrated using this configuration. The calibra-
tion apparatus and results are presented in Figure 5.6.1. The gas flow calcu-
lations are presented in Appendix A.
32
-------
COARSE
SAMPLING
VALVE
PROBE
co
GO
FLOW
METER
QUICK
DISCONNECT
CONNECTORS
QUICK DISCONNECT
CONNECTORS
VACUUM
GAUGE
FIGURE 5.1. Modified gas- sampling apparatus.
-------
TO SAMPLE
LOCATIONS
co
GREENBURG-SMITH
IMPINGERS
FIGURE 5.2. Moisture train
-------
oo
en
FIGURE 5.3. Anemometer Calibration Apparatus With
Correction Results.
aFt/min Adjusted from Standard Pitot Readings.
See Anemometer check with calculations of 1 7/8"
Readinq and 4" reading to the Anemometer Flows.
(Appendix A)
------- |