&EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 79-RBM-4
Air
Rubber Products
Styrene-Butadiene
Rubber Manufacture
Emission Test Report
General Tire and Rubber
Company
Mogadore Chemical
Plant
Mogadore, Ohio
-------
SOURCE TEST AT GENERAL TIRE'S
STYRENE BUTADIENE RUBBER PLANT
Mogadore, Ohio
Contract No. 68-02-3545
Work Assignment No. 5
Technical Manager: Terry Harrison
Prepared for:
United States Environmental Protection Agency
Emission Standards and Engineering Division
Emission Measurement Branch
Research Triangle Park, North Carolina 27711
By:
TRW
ENVIRONMENTAL ENGINEERING DIVISION
Post Ottice Box 13000
Research Triangle Park, North Carolina 22209
-------
TABLE OF CONTENTS
Page
LIST OF TABLES Hi
LIST OF FIGURES iv
GLOSSARY OF TERMS vii
SAMPLE POINT IDENTIFICATION KEY viii
1.0 INTRODUCTION ..... 1-1
2.0 SUMMARY. . 2-1
2.1 Plant Process Summary . 2-1
2.2 Results Summary 2-3
2.3 Hydrocarbon Emission Rate Per Gallon of SBR Latex
Processed 2-5
3.0 DISCUSSION OF RESULTS 3-1
3.1 Total VOC Emissions 3-1
3.2 Mass Balance of Styrene Around the Slowdown Tank. . 3-6
3.3 Mass Balance of Styrene Around the Steam
Stripper 3-12
3.4 Peak Characterization of Butadiene and Styrene
Emissions 3-16
4.0 PROCESS DESCRIPTION 4-1
5.0 SAMPLE LOCATIONS 5-1
5.1 Reactor Slowdown Tank System (BDT #4) 5-1
5.2 Central Vacuum System (CVS) 5-14
5.3 Styrene Stripper System 5-15
-------
TABLE OF CONTENTS (continued)
Page
5.4 Product Storage Tanks (PST) 5-20
5.5 Shaker Screens 5-23
6.0 SAMPLING AND ANALYTICAL PROCEDURES 6-1
6.1 Sampling Procedures 6-1
6.2 Analytical Procedures 6-7
6.3 Continuous Monitoring at the Slowdown Tank Outlet,
Central Vacuum System Outlet, and Styrene
Stripper Vacuum Outlet with a Flame lonization
Detector 6-11
6.4 Quality Assurance and Quality Control 6-16
6.5 Audit Samples 6-20
n
-------
LIST OF TABLES
Number Page
2-1 Summary of Average Gaseous Emission Results From
the Test Locations at the General Tire Facility
in Mogadore, Ohio 2-4
3-1 Test Results from the Central Vacuum System Outlet
(Sample Point 1). . 3-3
3-2 Test Results from the Central Vacuum System
(Sample Point 1) 3-4
3-3 Test Results from the Fugitive Emissions of the
Shaker Scree'ns (Sample Point 3) . . . • 3-7
3-4 Test Results from the Steam Stripper Exhaust
(Sample Point 2) 3-8
3-5 Test Results from the Storage Tank Vent (Sample
Point 11) 3-9
3-6 Test Results from the Blowdown Tank Outlet (Sample
Point 1A) . , 3-14
3-7 Test Results from the Blowdown Tank Outlet (Sample
Point 1A) 3-15
6-1 Audit Results 6-26
-------
LIST OF FIGURES
Number
2-1
3-1
3-2
3-3
3-4
3-5
4-1
4-2
5-1
5-2
5-3
5-4
5-5
5-6
5-7
5-8
5-9
5-10
Styrene/Butadiene Rubber Process
Mass Balance of Styrene Around the Slowdown System
on 3/27/80
Mass Balance of Styrene Around the Slowdown System
on 4/1/80
Mass Balance of Styrene Around the Steam Stripper
on 4/1/80
Butadiene Emissions
Styrene Emissions
Process Flow Diagram
Plant Layout
Rx/Blowdown System Schematic . .
Rx/Blowdown Tank Outlet - Sample Point
Location 1A
Rx/Blowdown Building Schematic (Plan View) ....
Sample Train Schematic .....
Slowdown Tank #4
Central Vacuum System (Floor Plan View).
Central Vacuum System - Exhaust Piping: Sample
Point #1
Vacuum System Flow Measurement (Side View) ....
CVS Sample Train Schematic (Point #1)
Styrene Stripper System Function Diagram
Page
2-2
3-11
3-13
3-17
3-18
3-19
4-3
4-4
5-3
5-4
5-5
5-8
5-9
5-10
5-11
5-12
5-13
5-17
IV
-------
LIST OF FIGURES (continued)
Number
5-11
5-12
5-13
5-14
5-15
5-16
6-1
6-2
6-3
6-4
6-5
6-6
6-7
6-8
6-9
6-10
Styrene Stripper System Condenser (Side View). . .
Sample Train Styrene Stripper Exhaust - Sample
Point #6 ..
Product Storage Tank - Aerial View . . •
Storage Tank: Sample Point #11 (Side View). . . .
Shaker Screens ....
Shaker Screens - Plan View - 2nd Floor Level . . .
Flow Line
GC/FID Gas Sampling System
FID Results at the Slowdown Tank Outlet on
March 27, 1980
FID Results at the Central Vacuum System Outlet
on March 31, 1980
FID Results at the Central Vacuum System Outlet
on April 1, 1980
FID Results at the Styrene Stripper Outlet on
April 2, 1980
Instrument Linearity for Benzene
Instrument Linearity for Styrene
Degradation Study - Hydrocarbons - Bag Sample -
Mogadore, Ohio
Degradation Study - Hydrocarbons - Bag Sample -
Mogadore, Ohio
Page
5-18
5-19
5-21
5-22
5-24
5-25
6-9
6-10
6-12
6-13
6-14
6-15
6-18
6-19
6-21
6-22
-------
LIST OF FIGURES (concluded)
Number Page
6-11 Degradation Study - Benzene - Bag Sample -
Mogadore, Ohio 6-23
6-12 • Degradation Study - Latex (Styrene) - Mogadore,
Ohio - March 25, 1980 - April 3, 1980 6-24
-------
GLOSSARY OF TERMS
EMB - Emission Measurement Branch
EPA - Environmental Protection Agency
SBR - Styrene/Butadiene Rubber
VOC - Volatile Organic Compounds
NSPS - New Source Performance Standards
NESHAPS - National Emission Standards For Hazardous Pollutants
CVS - Central Vacuum System
FID - Flame lonization Detector
GC - Gas Chromatograph
TC - Thermal Conductivity
ppm - parts per million
Rx - Reactor
GPM - Gallons Per Minute
BDT - Slowdown Tank
SSS - S.tyrene Stripper System
PST - Product Storage Tank
ss - Shaker Screens
Er. - Equivalent Diameters
D - Differential pressure
psi - pounds per square inch
QA - Quality Assurance
TBC - 4-tertiary Butylcatechol/4-tertiary Butylpyrocatechol
SD - Standard Deviation
MeCl - Methylene Chloride
vn
-------
SAMPLE POINT IDENTIFICATION KEY
Point Sample Identification
1A Reactor/Slowdown Tank (BDT-4) Exhaust
1 Central Vacuum System (CVS) Exhaust
2 Styrene Stripper Vacuum System Exhaust
3 Shaker Screens Exhaust
4 Slowdown Tank Inlet Latex Stream
5 Slowdown Tank Outlet Latex Stream
8 Styrene Stripper Inlet Latex Stream
9 Styrene Stripper Outlet Latex Stream
11 Product Storage Tank Vent
vm
-------
1. INTRODUCTION
Under the supervision of the Emissions Measurement Branch (EMB) of
the Environmental Protection Agency (EPA), TRW Environmental Engineering
Division personnel conducted a study of the volatile organic emissions
produced by the Styrene/Butadiene Rubber (SBR) Industry at General
Tire's facility in Mogadore, Ohio.
Information obtained from the February 4, 1980 presurvey of the
plant in Mogadore was used to develop a test plan. The field sampling
and analysis was performed from March 24 through April 4, 1980.
The purposes of this test were: 1) to determine the total volatile
organic emissions of the plant; 2) to study the mass balance of styrene
around the blowdown tank and styrene stripper; 3) to obtain data
concerning the VOC emissions to provide design criteria for possible
control devices; and 4) to determine the degradation rates of organic
compounds in the gas and latex samples. The specific compounds under
analysis include: butadiene, benzene, toluene, ethyl benzene, xylenes,
styrene, CO^, 02, and N2- The sampling points are described in
Section 5; with the sampling and analytical preparations and procedures
being described in Section 6.
This information was collected and may be used in the development
of possible New Source Performance Standards (NSPS), the National
Emission Standards for Hazardous Pollutants (NESHAPS).
-------
2. SUMMARY
The latex produced at the General Tire plant in Mogadore, Ohio, was
a butadiene specialty latex. This project studies the residual styrene
and butadiene emissions resulting from the formation of SBR. The
testing encompassed a two-week period. The first week involved sampling
and analysis of the blowdown system. The second week involved sampling
of the styrene stripper system, shaker screens, and storage tank. For a
clear understanding of the summary discussion, the plant process must be
understood (Section 4). A brief overview of the plant and the process
are described below.
2.1 PLANT PROCESS SUMMARY (Figure 2-1)
The latex is prepared in a reactor from raw materials. When the
latex in three reactors reached completion the contents were emptied
into a blowdown tank. The emissions from filling the blowdown tank were
channeled through a condenser and vented through the central vacuum
system. The latex left the blowdown tank system and was filtered
through a vibrating screen which removed any congealed latex. The
screens were open to the atmosphere inside the building, and a gas
sample was collected at this location. The latex then flowed into a
large pre-blend tank for storage. When needed, the latex was
transferred to the steam stripper for removal of the remaining residual
styrene and other hydrocarbons. The emissions generated by this process
were vented through the steam stripper vacuum system. Samples of these
gaseous emissions were collected along with samples of the latex before
and after steam stripping. If further product treatments were required,
the latex was routed to the blend tank. Otherwise, the finished latex
product was transferred to the product storage tank. A vent on the
storage tank permitted gas sampling data collection.
-------
Gas Vent
Condenser
Hew-
Down
Tank
D-
J_
Blow-
Down
Tank
D-
Blow-
Dawn
Tank
D-
D-
I
Blow-
Down
Tank
Central
Vacuum
System
Section 3.1.1
Gas Vent
Shaker
Screens
Section 3.1.2
Gas
Shaker
Screens
Pre+Blend
Tank
Gas Vent
Steam Strlppe
Vacuum System
Section 3.1.3
Steam
trlppe
Steam
trlppe
LEGEND
Latex Sampling Point
Gas Sampling Point
Gas Vent
Section 3.1.4
11
Product
Storage
Tank
Blend Tank
Figure 2-1. Styrene/Butadiene Rubber Process
-------
2.2 RESULTS SUMMARY
The objectives for the test included: 1) the quantification of
volatile organic emissions produced by normal process operation
(Section 2.2.1); 2) measurement and characterization of emissions
produced from a reactor dump prior to treatment by the control device
(Section 2.2.2); and 3) the determination of a material balance of
residual styrene between the blowdown tank and the steam stripper
(Section 2.2.3).
2.2.1. Volatile Organic Compound (VOC) Emissions
The summary Table 2-1 lists the average hydrocarbon emissions and
flow data from the test period. The results from the blowdown outlet on
the table are not VOC emission results, because the emissions were not
directly vented to the atmosphere, and will be discussed in a later
section. The results from the central vacuum system are VOC emission
results and include the hydrocarbon concentrations from the condensate
collected in the gas stream. They represent the average of a total
reactor dump-degassing cycle. However, there were other processes in
the plant that vented through the central vacuum system so that the
totals do not necessarily represent only the blowdown of the reactors
studied. The cycle lasted approximately 8 hours, during which time the
temperatures and emission concentrations varied from the average
considerably. The results from each test are discussed separately in
Section 3^
The results from the shaker screen are expressed in concentration
units only because no flow measurements could be made. The results
indicate very low levels of hydrocarbon emissions from the shaker
screens.
The results from the steam stripper are the average emissions
collected from the vent during an 8-hour period. The individual tests
(Section 3) were consistent so that the averages are representative of
the emissions associated with the operating process. The results from
the storage tank are the average emissions collected from the vent
during an 8-hour period. The individual tests (Section 3) were
2-3
-------
Table 2-1. SUMMARY OF AVERAGE GASEOUS EMISSION RESULTS
FROM THE TEST LOCATIONS AT THE GENERAL TIRE
FACILITY IN MOGADORE, OHIO
Process Data
Flue gas temperature (°F)
Flue gas area (ft2)
Flow rate (SCFM)
Vacuum pressure (in.Hg) Absolute
Emission rate (Ib/hr)
Analytical Results (ppmv as
compound)
1. Butadiene
2. Benzene
3. Toluene
4. Ethylbenzene
5. Xylene
6. Styrene
Total hydrocarbon
Volume (%)
Measured hydrocarbon as
compound
°2
N2
co2 .
H2°
Total (%)
Hydrocarbon emission/
production rate (Ib/gal)
Slowdown out let8 '9
151
f
111.98
12.7
55.2
128,881
61.6
231
432
1,137
1,189
132,632
13.3
5.27
76.99
1.55
1.51
98.62
0.037
Central vacuum
system '9
112
0.087
105.13
30.8
64.3
54,758
23.4
15.3
362
71.0
2,701
57,931
5.76
10.65
77.71
0.64
1.51
96.27
0.043
Steam stripper
outlet
60
0.09
7.51
29.0
1.2
361
2.91
12.9
718
458
4,131
5,684
0.57
19.10
74.21
0.15
1.51
95.53
0.00050
Storage tank
outlet
91
0.087
6.38
29.0
0.20
55.8
0
40.0
27.6
7.74
196
327
0.03
19.56
77.01
0.006
1.51
101.71
0.000082
Shaker .
screen
. 103
NA
NA
29.0
ID
39.2
0
0
0
0
24.8
-64
0.01
19.65
76.53
ND
1.51
97.70
ID
Average of 11 runs from two days. Results should
be reviewed from individual analyses and not from
an average (see sub-summaries).
reviewed from individual analyses and not from an
average (see sub-summaries).
cAverage of 3 runs from one day.
Represents one run.
'Concentration values are on a dry basis.
No emission point to atmosphere.
JAverage of two runs.
^Condensate values not included see sub-summary tables.
Pounds of hydrocarbon emissions per gallon of latex
processed.
'Average rate of two days tested.
NA - No measurements parameters available.
ID - Insufficient data to calculate rate
2-4
-------
consistent so that the averages are representative of the emissions
associated with the operating process. The emissions from the steam
stripper and storage tank are considerably less than the emissions from
the central vacuum system. The end result of the processes tested was a
total average emission rate of 57.5 Ib/hr hydrocarbons found under
normal operating conditions. This total figure does not include the
blowdown outlet emission rate because this gas stream was routed to the
central vacuum system for emission into the atmosphere.
2.2.2 Blowdown Tank Outlet
The hydrocarbon emissions from the blowdown tank were not vented
directly to the atmosphere (Table 2-1). However, a study of the
emissions is important in a study of possible control devices. The
emissions from the blowdown tank are averaged in Table 2-1. This is the
average of tests from 2 days of sampling, and does not accurately
reflect the emission rate. During the period of a reactor dump the
emissions were high and decreased significantly over the next several
hours of the degassing period. The emission profile is discussed at
length in Section 3. The results from this location are much clearer if
r
each sample is discussed separately.
2.2.3 Styrene Mass Balance
The third analytical problem was to study a mass balance of styrene
around the blowdown tank and the steam stripper. The mass balance was
determined by calculating the total styrene entering the tank and
leaving the tank. Styrene entered the tanks in the latex which had a
known flow rate. A sample of the entering latex was analyzed for the
residual styrene content by gas chromatography (Section 6.2). The
styrene left the tanks either in the latex or as a gas. Samples of gas
outlets and latex outlets were analyzed by gas chromatography for the
styrene content. Flow rates were monitored for both the gas samples and
latex samples. The results show mass balances for both tests of the
blowdown tank and all three tests of the steam stripper. The tables in
Section 3.2 represent the mass balance results of the two systems.
2.3 HYDROCARBON EMISSION RATE PER GALLON OF SBR LATEX PROCESSED
The production rate of SBR latex was calculated. The hydrocarbon
emission rate were determined by testing. The calculations of the
2-5
-------
emission/production rate was determined for the blowdown tank system,
the central vacuum system outlet, the styrene stripper outlet, and the
storage tank outlet. The hydrocarbon emission rate utilized the results
of the tests at the sample locations in the process. The production
rate of SBR latex was supplied by plant personnel.
The production rate of the blowdown tank outlet and central vacuum
system was determined from the amount of latex produced in the three
reactors dumping into blowdown tank #4 (BDT #4). The time period of
testing this process was from the initial reactor dumping into BDT #4
until the latex was transferred from BDT #4 to the shaker screen system.
The emission rate of hydrocarbon was averaged because the source of the
emission was a batch operation. The emission/production rate of the
2 days of monitoring the process through BDT #4 was 0.037 pounds of
hydrocarbon per gallon of latex produced (Ib/gal) for the blowdown tank
outlet (see sample calculation in Appendix A); and 0.043 Ib/gal for the
central vacuum system outlet.
The emission/production rates of the steam stripper outlet and
storage tank outlet were determined by the emissions from the continuous
process through the stripper system that filled the storage tank tested.
The emissions were determined over an 8-hour period during the operation
of the stripper. The production rate used was the average gallons of
latex feeding into the stripper system during the test period. The
results of the emission/production rate calculation was 5.10 x
-4 -5
10 Ib/gal at the steam stripper outlet and 8.2 x 10 Ib/gal at the
storage tank outlet.
2-6
-------
3.0 DISCUSSION OF RESULTS
The primary objectives of the test at the Mogadore facilty were to
determine the total volatile organic emissions (Section 3.1), to observe
a mass balance of styrene around the steam stripper (Section 3.2), a
mass balance of styrene around the blowdown tank (Section 3.3), to
characterize the butadiene emission peak associated with a reactor dump
for the purpose of design criteria for control devices (Section 3.4),
and rates of organic compounds in the gaseous and latex samples
(Section 3.5).
3.1 TOTAL VOC EMISSIONS
Gaseous samples were collected at the following emission points:
the central vacuum system outlet (Section 3.1.1), the fugitive emission
of the shaker screens (Section 3.1.2), the steam stripper vacuum system
outlet, (Section 3.1.3), and the product storage tank (Section 3.1.4).
The process, pertaining to the emission locations will be discussed,
briefly for each location. A note of caution should be made against
using the average results to represent the VOC emissions (Table 2-1);
the individual analytical results show that the emissions reflect the
cyclic process pattern around the blowdown tank process. A modified
sample location (#1A) was maintained at the blowdown tank outlet
(Section 3.1.5) to obtain concentration results of the gas stream before
the condenser in line to the emission point at the central vacuum system
exhaust.
3.1.1 Central Vacuum System Outlet (Sample Location #1)
In the plant process the raw styrene and butadiene were combined in
reactors to produce the SBR latex. Normally, three batches would dump
into one blowdown tank. There were four blowdown tanks in the Mogadore
facility. The emissions from all the blowdown tanks were channeled
-------
through a condenser and then vented to the atmosphere through the
central vacuum system. Gas samples were collected at the outlet of the
central vacuum system during two sets of the reactor dumps and degassing
cycles. Samples were collected following the modified EPA Method 110
(Appendix G). Temperature and flow measurements were taken throughout
the course of the test (Section 6). The first reactor dump degassing
cycle lasted approximately 8^ hours. During that time the emission
temperatures rose and then decreased. The flow rate also increased and
then decreased (Table 3-1). Over the course of the test the total
hydrocarbon emission rate showed dramatic changes: 47.8 Ib/hr for the
first hour, 89.7 Ib/hr and 90.6 Ib/hr average over the next 5 hours.
For individual hydrocarbons, the results show a gradual decrease in
concentration except for styrene, which shows a marked increased in
concentration over time (Table 3-1). The hydrocarbon measurements were
performed by GC/FID analysis (Section 6.2) the stationary gas analyses
were performed by GC/TCD (Section 6.2).
The second monitored reactor dump and degassing cycle lasted
approximately 8 hours. During the first day of testing, liquid
condensate was observed in the sample line, therefore the integrated bag
sampling method was modified by placing a condensate trap in the
sampling line to remove the liquid condensate that came through the
sampling line. The collected condensate was analyzed by the direct
injection technique into the GC/FID (Section 6.2). The condensate had a
water phase and an organic phase. The organic phase contained
31.91 percent total hydrocarbons by volume with the remainder being an
inorganic solution, and the water phase contained 0.15 percent total
hydrocarbons by volume (Table 3-2). The condensate hydrocarbon values
were added to the average gaseous hydrocarbon values to give the total
hydrocarbon emission rate reported for the test.
Temperature and flow measurements were taken throughout the course
of the test (Section 6.4). During the test the temperature rose
steadily. The flow rate increased in the first hour and decreased after
that (Table 3-2). The comparison of the flow rates for the first test
and the second test shows that the flows on the second day were
considerably lower.
3-2
-------
Table 3-1. TEST RESULTS FROM THE CENTRAL VACUUM
SYSTEM OUTLET (SAMPLE POINT 1)
Date
Time
Run number
Integrated bag number
Process Data
Flue gas temperature (°F)
Flue gas area (ft2)
Flow rate (SCFM)
Emission rate (Ib/hr)
Analytical Results3 (ppmv as
compound)
1. Butadiene
2. Benzene
3. Toluene
4. Ethylbenzene
5. Xylene
6. Styrene
.Total hydrocarbon
Volume (%)
Measured hydrcarbon as compound
°2 •
CO,
Total (%)
3/27/80
0906-1002
1
A
85
0.087
134.41
47.8
75.424
8.6
16.7
601
122
1,928
78,100
7.8
6.48
58.14
0.73
1.51
74.66b
3/27/80
1002-1102
1
B
123
0.087
163
89.7
62,512
3.6
15.6
577
168
2,147
65,423
6.5
5.17
80.48
0.43 .
1.51
94.09
3/27/80
1102-1236
1
C
155
0.087
131
90.6
69.670
1.0
1.3
309
113
2,767
72,862
7.3
7.48
72.80
1.41
1.51
90.50
3/27/80
1236-1731
1
D
107
0.087
101.30
82.2
70,062
13.2
11.1
323
104
4,338
74,851
7.5
8.38
76.10
0.68
1.51
94.17
Average0
115
0.087
117.83
80.8
69,687
9.3
10.4
381
115
3,518
73,721
7.4
7.62
74.01
0.79
1.51
91.31
aThese data represent those compounds expected to be present and identified as that compound by virtue
of having similar retention times to those retention times of these known gases. The retention times of
the known (reference) gases were determined by laboratory injection under similar GC operations.
The total X of gases analyzed only total 74.66%. Since 25S> is unaccounted for, these numbers should
be considered suspect.
cThe average is a "time weighted average" of the concentration value and Ib per hour values.
3-3
-------
Table 3-2. TEST RESULTS FROM THE CENTRAL
VACUUM SYSTEM (SAMPLE POINT 1)
Date
Time
Run Number
Integrated Bag Number
Process Data
Flue Gas Temp. (°F)
Flue Gas Area (Ft2)
Flow Rate (SCFM)
Emission Rate (Lb/Hr)
4/1/80
0728-0855
2
A
84
0.087
98.31
62.3
4/1/80
0900-1035
2
6
98
0.087
130.52
88.7
4/1/80
1040-1355
2
C
118
0.087
72.78
16.0
4/1/80
1405-1535
2
D
121
0.087
89.19
5.3
Averagec
108
0.087
92.44
37.4
Analytical Results3
(ppmv as compound)
1. Butadiene
2. Benzene
3. Toluene
4. Ethylbenzene
5. Xylene
6. Styrene
Total Hydrocarbon
72552
28.4
23.0
398
21.2
1599
74622
77919
11.7
10.7
366
27.6
1734
80069
21562
35.2
17.8
223
23
2056
23919
3494
36.5
15.3
245
18.9
1610
5419
39329
37.6
20.3
344
27.1
1884
42142
Volume %
Measured Hydrocarbon as
Compound
°2
N2
co2
H20
Total %
Unaccounted for %
7.5
13.0
77.81
1.2
1.51d
101.01
—
8.0
11.6
79.75
.38
1.51
101.21
—
2.39
15.51
83.93
.37
1.51
103.71
—
.54
12.55
81.15
.19
1.51
95.91
4.09
4.13
13.68
81.40
0.49
1.51
101.21
--
aNot representative if interested in control device fabrication - concentration varies widely from average.
Does not include condensate results - At this location 480 ml condensate collected in 0810 hrs. ; condensate
contains 32.06% hydrocarbons by volume.
cTime weighted average over test period.
Based on saturation at analyses temperature.
3-4
-------
The total emission rate for the second test period reflect the
process pattern. When the reactor dump began, the emission rates of the
first 2 hours were 62.3 Ib/hr and 88.7 Ib/hr. Only one sample was
collected over the next 4 hours, and the emission rate was 16.0 Ib/hr.
The last hours of the degassing period showed 5.3 Ib/hr (Table 3-2).
This increasing and then decreasing emission rate was seen both days
that the central vacuum system was tested. On the second day of testing
both the flow rate and measured hydrocarbon concentrations were lower
than the first day. A possible reason for this change was a process
interruption in the central vacuum system. A noticable decrease in
hydrocarbon concentration coincides with the system shut-down.
During the second week of testing (4/1/80), the CVS developed an
operational problem for several hours. A valve on the knockout pot
prior to the pumps malfunctioned, allowing entrained liquid from the
knockout pot past the vacuum pumps and into the exhaust line. During
this period (0920 hrs - 0945 hrs), the additional pressure from the
liquid may have caused inaccurate pressure readings at the D cell and
magnehelic. The system malfunction was evidenced by a white latex-like
rain exiting from the system atmospheric exhaust. Corrective action by
plant maintenance personnel solved the problem quickly and no further
system disruptions were encountered. A second significant process
difference between the 2 days was the condensate associated with the gas
emissions on the second day.
The emissions associated with blowdown tanks and the central vacuum
system vary widely from the averages. In the summary table (Table 2-1)
the central vacuum system emission results are the average of the
results from both days of testing. The results from the second day
include the results of the condensate analyses. A note should be made
that the real emissions follow a pattern of high levels during the
reactor blowdown with decreasing levels as the degassing process
continues. The average would not reflect this pattern.
3.1.2 Fugitive Emissions of the Shaker Screens (Sample Location #3)
The second emission point in the process was above the shaker
screens. The latex flowed from the blowdown tanks through a series of
3-5
-------
screens which separated out any congealed latex. The screens were open
to the atmosphere so no flow measurements were obtained. The
concentration of hydrocarbons being emitted from the screens were very
low, (Table 3-3) amounting to 64 ppm by volume total. Only one sample
was collected from the surface above the screens. The sample was
collected following the modified EPA Method 110 (Appendix G), and was
analyzed by GC/FID (Appendix F).
3.1.3 Steam Stripper Vacuum System Exhaust (Sample Location #2)
As the latex proceeded, the next stage was the pre-blend tank.
From there the latex flowed to one of two steam strippers. Steam was
forced through the latex to remove any residual hydrocarbons, primarily
styrene. The emissions from both steam strippers were vented through
the steam stripper vacuum system. Three gas samples were collected at
the steam stripper vacuum system outlet following the modified EPA
Method 110 (Appendix G). Flow rates and temperatures were monitored
during the sampling period (Section 6.4). The temperatures and the flow
rates remained consistent throughout the test. The total hydrocarbon
concentrations in each samples were found to be close to the average
5600 ppm by volume (Table 3-4). The flow rate was quite low and the
average hydrocarbon emission rate was 1.2 Ib/hr. The averages appear
representative of the conditions and emissions during operation of the
steam stripper vacuum system.
3.1.4 Product Storage Tank Outlet (Sample Location #11)
The final emission point analyzed was the outlet above the product
storage tank. Three gas samples were collected at the outlet following
the modified EPA Method 110 (Appendix G). Flow and temperature
measurements were made throughout the sampling period (Section 6.4).
The temperature and flow rates were consistent in each test. The
average flow rate was low, 6.38 SCFM average (Table 3-5). The total
hydrocarbon concentration in each sample was approximately 330 ppm (by
volume). Therefore, the average hydrocarbon emission rate was
0.20 Ib/hr (Table 3-5). The averages appear to be representative of the
conditions associated with the product storage tank.
3.2 MASS BALANCE OF STYRENE AROUND THE BLOWDOWN TANK
The mass balance around the blowdown tank was based on a comparison
of residual styrene found in the latex at the blowdown tank outlet plus
3-6
-------
Table 3-3. TEST RESULTS FROM THE FUGITIVE
EMISSIONS OF THE SHAKER SCREENS
(SAMPLE POINT 3)
Date
Time
Run number
Integrated bag number
Process Data
Flue gas temperature (°F)
Flue gas area (ft2)
Flow rate (SCFM)
Emission rate (Ib/hr)
Analytical Results3 (ppmv as compound)
1. Butadiene
2. Benzene
3. Toluene
4. Ethylbenzene
5. Xylene
6. Styrene
Total hydrocarbon as compound
Volume (%)
Measured hydrocarbon as compound
°2
N2
.C°2
H20
Total (X)
Unaccounted for (%)
4/2/80
1730-1800
1
A
103
NA
NA
NA
39.2
0
0
0
0
24.8
64
0.006
19.65
76.53
NO
1.51
97.71
2.29
aThese data represent those compounds'expected to be present and
identified as that compound by virtue of having similar reten-
tion times to those retention times of these known gases. The
retention times of the known (reference) gases were determined
by laboratory injection under similar GC operations.
3-7
-------
Table 3-4. TEST RESULTS FROM THE STEAM STRIPPER
EXHAUST (SAMPLE POINT 2)
Date
Time
Run number
Integrated bag number
Process Data
Flue gas temperature (°F)
Flue gas area (ft )
Flow rate (SCFM)
Emission rate (Ib/hr)
Analytical Results9 (ppmv as
compound)
1. Butadiene
2. Benzene
3. Toluene
4. Ethyl benzene
5. Xylene
6. Styrene
Total hydrocarbon as compound
Volume (%)
Measured hydrocarbon as compound
°2
CO,
H20
Total (*)
Unaccounted for (%)
4/2/80
1217-1250
1
A
65
0.09
8.02
0.6
318
8.6.
12
574
363
3,392
4,668
0.47
19.26
76.41
0.15
1.51
97.81
2.19
4/2/80
1250-1510
1
B
60
0.09
8.23
1.5
368
3.4
14.4
801.5
494
4,884
6,566
0.66
18.76
71.46
ND
1.51
92.41
7.59
4/2/80
1510-1638
1
C
57
0.09
6.16
0.9
366
0
10.9
638.5
437
3,209
4,662
0.47
19.50
77.80
ND
1.51
99.31
0.69
Average
60
0.09
7.51
1.2
366
2.9
12.9
718
458
4,131
5,684
0.57
19.10
74.21
0.15
1.51
95.53
4.47
These data represent those compounds expected to be present and identified as that
compound by virtue of having similar retention times to those retention times of these
known gases. The retention times of the known (reference) gases were determined by
laboratory injection under similar GC operations.
The average is a "time weighted average" of the concentration values and Ib per hour
values.
3-8
-------
Table 3-5. TEST RESULTS FROM THE STORAGE
TANK VENT (SAMPLE POINT 11)
Date
Time
Run number
Integrated bag number
Process Data
Flue gas temperature (°F)
Flue gas area (ft )
Flue rate (SCFM)
Emission rate (Ib/hr)
Analytical Results6 (ppmv as
compound)
1. Butadiene
2. Benzene
3. Toluene
4. Ethylbenzene
5. Xylene
6. Styrene
Total hydrocarbon as compound
Volume (X)
Measured hydrocarbon as
compound
°2
N2
CO,
H20
Total (X)
Unaccounted for (X)
4/2/80
1147-1202
1
A
55
0.087
6.84
0.2
63.8
0
38.4
26.1
8.9
210
348
0.03
19.29
76.33
0.06
1.51
97.21
2.79
4/2/80
1245-1345
1
B
96
0.087
6.33
0.2
56.7
0
40.4
24.5
13.1
195
328
0.03
19.08
74.19
0.06
1.51
94.91
5.09
4/2/80
1540-1640
1
C
95
0.087
6.32
0.2
52.9
0
39.9
31.1
2.1
196
322
0.03
20.10
80.07
ND
1.51
101.71
--
Average
91
0.087
6.38
0.2
55.8
• 0
40.0
27.6
7.7
196
327
0.03
19.56
77.01
0.06
1.51
98.17
1.83
These data represent those compounds expected to be present and identified as that
compound by virtue of having similar retention times to the retention times of these
known gases. The retention times of the known (reference) gases were determined by
laboratory injection under similar GC operations.
The average is a "time weighted average" of the concentration values and Ib per hour
values.
3-9
-------
the gaseous emissions of styrene found at the outlet of the blowdown
tank to the residual styrene found in the latex at the inlet of the
blowdown tank (Figure 3-1). Three reactors flowed latex into the
blowdown tank. The gallons of residual styrene in each reactor were
calculated from the total volume of latex and the percent of residual
styrene per gallon of latex (Appendix B). The percent of residual
styrene was determined by direct injection of the latex sample into a
GC/FID as prescribed in the EPA Draft method for the Determination of
Residual Styrene (Appendix F). A sample of latex from each reactor was
analyzed (Appendix B). The total pounds of residual styrene entering
the blowdown tank was 32.09 (Table B-5).
The styrene left the blowdown tank either as a gas or by remaining
in the latex. The residual styrene content in the latex leaving the
blowdown tank was determined by GC/FID analysis of a prepared latex
sample as described in the EPA Draft method for the
Determination of Residual Styrene. The total pounds of styrene leaving
the blowdown tank in the latex were the results of the GC/FID analyses
(Appendix C) multiplied by the (7.7 pounds/gallon density) and the total
pounds of latex leaving the blowdown tank (Appendix B).
The styrene content leaving the blowdown tank as a gas was
difficult to measure. Condensation of the gas occurred around the pump
so that condensate traps were placed before and after the pump. The
condensate samples were analyzed by direct injection into a GC/FID in
the same manner as a latex sample. The analytical results multiplied by
the volume of condensate gave the production of styrene per sampled
volume of gas; therefore, the weight per sample was multiplied by the
ratio of the total gas volume with the sampled gas volume to give a
result of pounds of styrene per sample period (Appendix B).
The concentration of styrene collected in the bag samples was
determined by GC/FID analysis (Section 6.2). Conversion of ppm styrene,
by volume, to total pounds of styrene appears in Appendix B. The pounds
of styrene leaving the blowdown tank as a gas was considered to be the
sum of the styrene in the bag samples plus the styrene in the condensate
samples (Figure 3-1). The total styrene, then, leaving the blowdown
tank as either gas or latex was 61.5 Ibs (Figure 3-1).
3-10
-------
Reactors
Condenser
before pump
.24 Ibs.'
Slowdown
Tank
Pump
Condenser
after pump
21.3 lbs.b
Latex Sample
8.0 lbs.a
On a dry basis.
On a wet basis - no dry basis was possible due to only volatile hydrocarbons being present.
Bag
Sample
31.9
Figure 3-1. Mass balance of styrene around the blowdown
system on 3/27/80.
-------
A second mass balance determination was performed around the
blowdown system (Figure 3-2). Analysis of the latex samples from the
three reactors showed 57.8 Ibs of styrene going into the blowdown tank
(Appendix B, Table B-6). The pounds of styrene leaving the blowdown
tank was 156.4 Ibs. The latex samples were analyzed by draft procedures
that presents the results on a dry basis. An attempt to calculate the
gas and condensate from the wet basis to a dry basis is not applicable
due to the analysis method and sample composition. One unusual result
was that no residual styrene was found in the latex sample collected at
the outlet of the blowdown tank. We can suggest no possible explanation
even after double checking all analytical systems. All the latex
samples and the condensate samples were analyzed by GC/FID following the
EPA Draft method for the Determination of Residual Styrene (Appendix F).
The gas samples were also analyzed by GC/FID (Section 6.2).
Analysis of the gaseous samples and condensate samples for their
styrene content by GC/FID permitted identification of all the
hydrocarbon species present in the samples (Tables 3-6 and 3-7). The
results shown for the blowdown tank (Tables 3-6 and 3-7) indicate the
concentrations of hydrocarbons that would be vented to the atmosphere if
the condenser was not in-line prior to the central vacuum system
(Section 3.1). Some rough estimations of emissions after the condenser
may be made by assuming that the condensate traps used for sampling
simulated the plant's condenser system, and then looking at only the
gaseous portion of the sample. However, no attempt was made to
duplicate condenser operating conditions. On the first day of sampling
the average gaseous hydrocarbon concentration was 4.4 percent by volume.
The condensate associated with the gas stream was collected during the
8-hour sampling period. A total of 454 ml was collected (Table A-9)
which contained 22.73 percent hydrocarbons by volume. On the second day
of testing the average total gaseous hydrocarbon concentration was
20.69 percent by volume. A total of 3553 ml of condensate was collected
over the 8-hour sampling period (Appendix B) and contained 1.72 percent
hydrocarbons by volume.
3.3 MASS BALANCE OF STYRENE AROUND THE STEAM STRIPPER
Latex samples were collected at the inlet and the outlet of the
steam stripper. The mass balance of styrene was determined by comparing
3-12
-------
CO
t->
CO
Reactors
9.8"
Ibs.
45.5 a
Ibs.
2.5"
Ibs.
Blowdown
Tank
Pump
Condenser
after pump
102.4 lbs.b
Latex Sample
0 1bs.a
On a dry basis.
On a wet basis - no dry basis was possible due to only volatile hydrocarbons being present.
Bag
Sample
54.034 b
Figure 3-2. Mass Balance of styrene around the
blowdown system on 4/1/80.
-------
Table 3-6. TEST RESULTS FROM THE SLOWDOWN TANK OUTLET
(SAMPLE POINT 1A)
Date
Tine
Run number
Integrated bag number
Process Data
Flue gas temperature (°F)
Flue gas area (ft2)
Flow rate (SCFM)
Emission rate (lb/hr)d
Analytical Results3 (ppmv as
compound
1. Butadiene
2. Benzene
3. Toluene
4. Ethyl benzene
5. Xylene
6. Styrene
Total hydrocarbon as compound
Volume (%)
Measured hydrocarbon as
compound
°2
N2
co2
H2o e 55°
Total (X)
3/27/80
905-1004
1
A
e
f
65
23.4
41,580
2.2
219
138
73
617
42,630.4
4.3
5.41
78.84
2.01
1.51
92.07
3/27/80
1007-1225
1
B
59.44
14.1
25,799
0.2
207
161
103
1,011
27,281
2.7
4.81
78.84
1.23
1.51
89.09
3/27/80
1230-1425
1
C
15.10
5.0
37,154
0
215
186
109
1,125
38,789
3.9
5.65
77.53
0.75
1.51
89.34
3/27/80
1438-1620
1
D
57.35
30.4
59,026
7.1
392
358
195
1,713
61,691
6.2
8.44
58.58
2.61
1.51
77.34
3/27/80
1630-1650
1
E
116.72
112.0
104,372
71
1,291
998
543
3,227
110,501
11.0
9.98
70.63
1.33
1.51
94.45
3/27/80
1705-1740
1
F
142.17
47.8
25,505
153
1,648
1,394
1,403
3,372
34,475
3.4
13.43
82.68
0.34
1.51
101.36
Average
57.42
31.69
41,198
16.3
404
335
237
1,413
43,603
4.4
6.74
74.04
1.45
1.51
88.14
aThese data represent those compounds expected to be present and identified as that compound by virtue of having similar
retention times to the retention times of these known gases. The retention times of the known (reference) gases were
determined by laboratory injection under similar GC operations.
As noted below, varying percentages (from 77 to 101 percent) of the total volume of gases present were accounted for
in the analysis procedures for 0,, N_, CO,, and hydrocarbons. Therefore, the reader should be careful in
interpreting these results.
GThe average is a "time weighted average" of the concentration values and Ib per hour values.
No emission vent to atmosphere, therefore, results is the gas stream concentration rate to the condenser in line to the
central vacuum system.
eNo data - dial thermometer in gas stream installed for 4/1/80 test day at sample point 1A.
No vent to atmosphere but flow determined with an orifice installed with a differential pressure gauge.
BThe total hydrocarbon emission rate of 31.6 was calculated with the addition of the 8.27 Ib/hr in the sampled
condensate (see Table A-9). The time weighted average of hydrocarbon emission rate without the condensate was 23.3 Ib/hr.
3-14
-------
Table 3-7. TEST RESULTS FROM THE SLOWDOWN TANK OUTLET
(SAMPLE POINT 1A)
Date
Tine
Run number
Integrated bag number
4/1/80
0726-1824
2
A
4/1/80
0827-0944
2
B
4/1/80
0945-1159
2
C
4/1/80
1204-1420
2
0
4/1/80 Average0
1437-1603
2
E
Process Data
Flue gas temperature (°F)
140
135
145
161
165
151
Flue gas area (ft )
Flow rate (SCFM)
Emission rate (lb/hr)d
Analytical Results8 (ppmv as
compound)
1. Butadiene
2. Benzene
3. Toluene
4. Ethylbenzene
5. Xylene
6. Styrene
Total hydrocarbon as compound
Volume ffl
Measured hydrocarbon as compound
°2
N2
H°0
Total (%)
Unaccounted for (X)
e
175.
32.8
143,148
68.
132.
162
56.
912
144,479
14.
75
2
5
1
4
2.38
75.96
2.38
1.51
96.
3.
63
37
181
177
275,239
72
91
399
180
1,907
277,889
27
5
83
1
1
119
-
.79
.7
.5
.1
.8
.14
.65
.0
.51
.11
-
87.
31.
216.994
119
47.
429
213
2,249
220,051
22.
3.
79.
1.
1.
108.
—
46
4
5
0
76
80
69
51
76
196
.23
312,355
138
40
736
435
2,996
316,700
31
.47
.8
.1
.7
3.76
79.80
1.69
1.51
118.46
-
-
221.64
55.7
61,388
95.2
22.0
726
10,391
2.945
78,845
7.9
3.76
79.80
1.69
- 1.51
94.66
5.34
166.54
68. 5f
216,563
107
57.
530
2,037
2,366
221,661
22.
3.
79.
1.
1.
109.
~-
9
2
81
95
66
51
13
aThese data represent those compounds expected to be present and identified as that compound by virtue of having
similar retention times to the retention times of these known gases. The retention times of the known (reference)
.gases were determined by laboratory injection under similar GC operations.
As noted below, varying percentages for 0., N_, CO. and hydrocarbons. Therefore, the reader should be careful
in interpreting these results.
cThe average is a "time weighted average" of the concentration values and Ib per hour values.
No emission vent to atmosphere, therefore, results are the gas stream concentration rate to the condenser in
line to the central vacuum system.
eNo vent to atmosphere but flow determined with an orifice installed with a differential pressure gauge.
The total hydrocarbon emission rate of 78.8 was calculated with the addition of the 10.3 Ib/hr in the
sampled condensate (see Table A-9). The time weighted average of hydrocarbon emission rate without the
condensate was 68.5 Ib/hr.
3-15
-------
the pounds of residual styrene entering the steam stripper in the latex
to the styrene leaving the steam stripper in the gaseous emissions and
in the latex. Three separate mass balances were determined
(Figure 3-3). The residual styrene in the latex samples was determined
by GC/FID analysis (Appendix F). The total amount of styrene entering
and leaving the stripper in latex was calculated from the analytical
results (Appendix C) and the total volume transferred. The styrene
escaping the stripper as a gas was collected as an integrated bag sample
(modified EPA Method 110) and analyzed by GC/FID. In all three test
periods correlation between styrene into and out of the system was found
(Figure 3-3). In the first test 1.95 Ibs was calculated going in, and
1.16 Ibs was calculated going out (Appendix B). In the second test,
1.82 Ibs was calculated going into the stripper with 1.80 Ibs coming out
(Appendix B). In the third test, 0.72 Ibs styrene was calculated going
into the stripper with 0.51 Ibs coming out (Appendix B).
3.4 PEAK CHARACTERIZATION OF BUTADIENE AND STYRENE EMISSIONS
During the transfer of latex from the reactor to the blowdown tank
the butadiene emissions increased and decreased significantly while the
styrene emissions over the same period increased steadily. The transfer
period lasted approximately 30 minutes. These peak emissions are
significant for the development of possible control systems. To
characterize the emissions over one of these transfer periods,
integrated grab bag samples were collected every 5 minutes following the
modified EPA Method 110 (Appendix D). Analysis of the gas samples was
performed by GC/FID (Section 6.2). The results (Table A-12) indicate
that the butadiene concentration rose from 143,000 ppm by volume to
312,000 ppm by volume in the first 20 minutes. In the following
5 minutes the butadiene concentration dropped to 61,000 ppm by volume
(Figure 3-4). The styrene concentrations, over the same 25-minute
period, rose from 1,200 ppm by volume to 3,200 ppm by volume
(Figure 3-5).
3-16
-------
1.95 Ibs.
Bag Sample
.361 Ibs.
Steam Stripper
Time: 1217 - 1307
Run A: 1.95 Ibs. In 1 1.16 Out
Latex Sample
.796 Ibs.
1.82 Ibs.
Bag Sample
.642 Ibs.
Steam Stripper
Time: 1420 - 1520
Run B: 1.82 In » 1.80 Out
.72 Ibs.
Bag Sample
.108 Ibs.
Steam Stripper
Time: 1638 - 1700
Run C: .72 In » .51 Out
Latex Sample
.40 Ibs.
Figure 3-3. Mass Balance of Styrene Around
the Steam Stripper on 4/1/80
3-17
-------
o
to
"i
V
•5
3
3
300,000
280,000
260,000
240,000
220,000
200,000
180,000
160,000
140,000
120,000
100,000
80,000
60,000
40,000
20,000
-_ Intergrated over
Sampling Period
1:30 1:35 1:40 1:50 1:55 2:00
Time, hours
Figure 3-4. Butadiene Emissions
3-18
-------
VI
c
o
V)
V)
1
0)
3.200
3,000
2,800
2,600
2,400
2,200
2,000
1,800
1,600
1,400
1,200
1,000
800
600
400
200
_Bj£_6_- Intergrated over
Sampling Period
1:30 1:35 1:40 1:45 1:50 1:55 2:00
Time, hours
Figure 3-5. Styrene Emissions
3-19
-------
4. PROCESS DESCRIPTION
The General Tire plant in Mogadore, Ohio, produces styrene
butadiene rubber latex for paper, textile coating, and other uses. The
process and equipment used in this production are described below. A
process flow diagram and plant layout are shown at the end of this
description (Figure 4-1 and Figure 4-2).
The process has three major steps: reaction, blowdown, and
stripping. In the reaction step, styrene, butadiene, and small amounts
of other monomers are added to a water solution of soap, catalysts,
modifiers, and other chemicals and reacted to form SBR latex. The
Mogadore plant has several stirred tank reactors for SBR latex use, each
about 6000 gallons in size with water jacketing for cooling. The
reactions are batch run and take 8 to 16 hours. Completion of reaction
is determined by time of reaction and amount of heat removed. The
reaction sequencing, including monomer addition, temperature profiles,
reacting time and cooling rates, are computer controlled.
After the completion of each reaction batch, the reactors are
dumped into a blowdown tank for removal of residual monomer. The
Mogadore plant has four blowdown tanks, each able to receive a dump from
any of the reactors. During a cycle on a blowdown tank, the tank
receives three reactor batches over a 2 to 6-hour period. During this
period, monomers are flashed off using vacuum from a central vacuum
system. After all three batches are dumped, steam, nitrogen, and vacuum
are used to further remove residual monomers from the latex in the
blowdown tank. Steaming time and analysis of latex samples are used to
determine completion of blowdown. The latex is then pumped from the
blowdown tank through shaker screens for removal of coagulated rubber,
and then to storage in which the final step is stripping. The residual
-------
styrene content of the latex is brought to within production
specifications using a four-stage process involving cyclone separation
and vacuum steaming. In this process, steam and latex are combined and
pumped into a cylone, where the heat from the steam and expansion into
the vacuum drive off styrene. The latex spins out to the side of the
vessel and drains out, where it is preheated, mixed again with steam and
pumped into the next stage cyclone. The overhead vapors are cooled to
condense out water and pulled out by a separate vacuum pump.
Process monitoring data and plant charts are provided in
Appendix D. They present the timing of the process during testing;
also, the vacuum and temperature profiles of the central vacuum system
on test days.
4-2
-------
HRNOMER ft SOAP SOLUTION
FEED LINES
do
•#
*CENTRAl
fc
ft)
VACUUM
(0
GENERAL TIRE. MOGADORE, OHIO
PROCESS FLOW DIAGRAM
SBR LATEX
(b-SBR
LATEX
BATCH REACTORS
5^'STEAM
EVACUATION
LINES
FROM SBR
REACTORS
FROM
BENTA
PROCESS
SHAKER
SCREENS
STRIPPING
AREA
CYCLONE
SEPARATORS
WITH PRE-HEATERS
f
OWN / — }
WITH
:T PRE-
FILTERS S
>
1
_i
6,
t
r
L ,
^rf
, 1
/ >
\
S.
BM-1
LJ^t-J]
I
' / v
^ n
1
_2
A
^
/ X
' - —•- — V-
r
^rJ
_j
\
1
!
TO
\ STORAGE
^ TO
STRIPPING
AREA VACUUM
TO
STORAGE
FROM
STORAGE
Figure 4-1. Process Flow Diagram
4-3
-------
PLANT LAYOUT
1
/
IBM
w
1
y
1C
^MB
1C
FE
LI
SBR LATEX
SCALE 20" « 1 CM
* SAMPLING LOCATIONS
0
-7
-0
LATEX STORAGE 10 SMALL TANKS
5 2x5
0 /
OQ
S~\T\ STRIPPING
VJVJ AREA
°88 §*
rv^>
oa«
LATEX f\ ^^\ 1
STORAGE \^_J \_J
0 10 s~*\ S~*\
LARGE f ) ( )
TANKS \_J \*~S
/" OO
, OO
1
40 ' 200 80
NCE-
NE
\
STRIPPING
WEA VACUUM
ifENTS 4"
XIAMETER,
5 ABOVE
GROUND
FACING
DOWN
30
TANK
CAR
&
TRUCK
LOADING
35
BOILER
HOUSE
SOAP SOLUTION BUILDING
SBR LATEX
BUILDING
*
LATEX
BLDG.
CENTRAL VACUUM
VENT, 3" DIAMETER
40* FEET
ABOVE GROUND
I
10
OTHER 50
PROCESS, .
OFFIC
WAREHO
BUILDI
CENTRA
VACUUM
E & 10
Jit
NGS ^9
-| 5
" I IQ
FACING UP '
400
MONOMER STORAGE
CYLINDRICAL HORIZONTAL TANKS
1
50 65 25 |
BUTADIENE STORAGE
Figure 4-2. Plant Layout
4-4
-------
5. SAMPLE LOCATIONS
This section of the report describes the location of the sample
points in the styrene/butadiene rubber process tested at the General
Tire and Rubber facility. The sample points locations in the process
will be divided into five primary areas.
5.1 - Reactor blowdown tank system (BDT #4)
5.2 - Central vacuum system (CVS)
5.3 - Styrene stripper system (SSS)
5.4 - Product storage tanks (PST)
5.5 - Shaker screens
A brief discussion of the process occuring at these areas will be
presented with the relationship to the sample points. A complete
discussion of the areas in the process is presented in Section 4. The
labeling scheme for the sample location is presented in the glossary and
the plant layout with sample locations is presented in Figure 2-1. The
sample acquisition method and timing of process will be presented in
this section, and the complete procedure discussed in Section 6.
5.1 REACTOR BLOWDOWN TANK SYSTEM (BDT #4)
The reactor/blowdown tank system process is initiated at the
reactor. The highly volatile styrene/butadiene formulae is processed
from the reactors into a blowdown tank (blowdown tank #4 studied in his
test). The gaseous emissions from this processing step are vented to
the central vacuum system for emissions into the atmosphere. Sample
locations were maintained at the exiting gas vent to the central vacuum
system (sample location 1A) to obtain the gaseous emissions, and a latex
sample was obtained from a sampling valve of the reactor dumping into
the blowdown tank (sample position 4) to obtain the latex product
-------
constituents at this point in the process. The latex formulae is held
in the blowdown tank until three reactors have dumped in the tank and
had a degassing period of the blended formulae. The latex from the
blowdown tank is processed to the shaker screen system. A latex sample
location is maintained at the exit of the blowdown tank (sample point 5)
from a plant product sampling valve. The latex is analyzed for chemical
constituents that have not combined into the product.
Figure 5-1 is a side veiw indicating the arrangement of major
components of the reactor/blowdown system, including the reactor's
vessel, blowdown tank, condenser, and approximately locations of the
latex and gaseous sampling points. Figure 5-2 is the sample point
location at the gaseous emissions from the blowdown tank. This sample
point is labeled 1A because the location is a modification of sample
point 1 at the central vacuum system. This modification was to provide
a more accurate gaseous emission analysis from the blowdown tank
emissions before processing through a condenser in-line to the central
vacuum system. Figure 5-3 is a layout of the second floor of the
reactor building indicating the relationship between the reactors, the
blowdown tank (BDT #4) and sample position set-up. Figure 5-4 is a
schematic of the sample train utilized at sample position 1A.
Figure 5-5 specifies the dimensions of the blowdown tank and components.
The gas stream at position 1A was monitored by a continuous FID and
the modified EPA Method 110 was utilized in obtaining an integrated bag
sample for GC/FID analyses. The latex samples were taken and prepared
according to the sampling procedures from the integrated grab method
presented in Appendix G.
5.1.1 Sampling Dimensions Around the Reactor/Blowdown System
The temperature of the blowdown tank exhaust was measured with a
plant-installed dial thermometer at the point of the pipe tap used for
the blowdown tank exhaust vapor (Point 1). The tap was a one-inch (1")
diameter pipe threaded into a 12-inch (12") diameter pipe. The tap was
located ten (10) equivalent diameters (EQ) from downstream flow
disturbance and two (2) equivalent diameters (Er.) from the upstream flow
disturbance. The tap was installed in the top of the 12-inch main,
5-2
-------
.Temperature
Readout
en
i
co
Steam * N-
Injection'
Condenser
Figure 5-1. Rx/Blowdown System Schematic
-------
en
i
To Sample System from
Sample Point 1A
V Welded
Pipe Thread
Swage to
Teflon Tubing
D/P Gauge
Figure 5-2. Rx/Blowdown Tank Outlet -
Sample Point Location 1A
-------
tn
i
en
0
O
To Central
Vacuum Sys
s
Rx
E^Condenser]
om ^. J
Sampling System
with FID and
Integrated Bag
Set-ups
OOOJ
Rx
Control Valve
B1 owdown
Tank
#4
Figure 5-3. Rx/Blowdown Building Schematic (Plan View)
-------
which was 13 feet above the floor level. A 22-foot long V Teflon line
connected the 1-inch line to the sampling train. Due to safety
considerations, the sampling train equipment was positioned adjacent to
building ventiliation doors. The area required a hot work permit, which
plant safety personnel provided daily after inspection of the area.
The flowrate of blowdown tank exhaust was measured through the use
of an orifice. The orifice was installed 12 feet downstream from the
sampling tap and 10 feet upstream from the nearest flow disturbance.
The orifice was installed at the flange of the 12-inch line and had a
nominal diameter of 1 inch. Pipe taps for the orifice measurement were
located approximately 30 inches upstream and eight pipe diameters
downstream of the orifice. The positioning of the taps was specified by
2
the plant engineer according to standard practices. Each pipe tap was
equipped with a pressure gauge with a line from each pipe leading to a
differential pressure (D ) instrument. The instrument was a model 13A-1
supplied by Foxboro Instrument. The D cell (0-150" HLO) was
constructed of 316 stainless steel with a 15 psi working pressure and
monitored by a gauge (Moore Products Company), calibrated by plant
personnel for 0 to 100 percent of D scale.
Two changes were initiated in the blowdown tank flow measurement
system during the test program. The blowdown tank developed increased
temperature and pressure and the operating pressure of the D cell
exceeded 15 psi after 2 days of measurements. These conditions
indicated a restriction in the orifice and, therefore, it was replaced
with a second orifice, which had a 2-inch nominal diameter. Secondly,
the presure gauges installed in the orifice pipe tap were replaced with
vacuum gauges, which were required since the pressure of the system was
negative.
5.1.2 Sample Procedures at the Reactor/Blowdown Tank System
The sampling train was modified to minimize problems that arose
with the condensation of moisture, the maintenance of sample pressure,
and adequate dilution of sample of hydrocarbon analysis. The
modifications are shown in Figure 5-4.
13/26/80 and 3/27/80.
2
Derived from Industrial Instruments, pg. 206.
5-6
-------
Several pumps were used to provide adequate working pressure to the
system. The most effective pump was a Duo-Seal vacuum pump. Because of
heat buildup, the pump gradually lost the ability to maintain sample
flow against the vacuum of the blowdown tank exhaust. The pump oil was
periodically changed to minimize the problem. A quality control check
was implemented to compare the effect of the pump on sample integrity by
taking gaseous grab bag samples before and after the pump. The results
showed no addition or subreactions of hydrocarbons from the pump.
A series of condensers were added in an attempt to simulate the
condenser of the blowdown tank exhaust system. A condensate sample from
the BDT condenser system was not collected because the BDT condenser was
common to all four blowdown tanks at the plant. Two condensers were
placed prior to the pump to provide sufficient volume for the condensate
collected during the sampling period. The volume of these condensers
proved to be inadequate, particularly during the degassing stage of
blowdown tank operation. Therefore, the condenser bottles were changed
as necessary. A third condenser was placed after the pump to guard
against condensation of the vapor due to system pressure restriction.
Throughout the testing at this location difficulty was encountered
in supplying adequate sample pressure to all components of the sampling
train. The hydrocarbon component of the sampling stream required
extensive dilution with Np before analysis. Sample flow to the dilution
board was monitored with a rotometer to insure accurate dilution. The
dilution board was calibrated before and after field testing and is
described in Appendix H. The flows to the integrated bag and
hydrocarbon analyzer were monitored with flowmeters and were maintained
at 2.0 liters per minute and between .2 and .5 liters per minute,
respectively. Any excess sample pressure was removed at the point of
attachment of the grab bag (see Figure 5-4).
The latex samples were taken directly from sample values utilized
by the General Tire personnel to sample the product at the different
stages in the process. The preparation of the sample bottles was
performed beforehand in the field lab. The TBC solution was added to
the appropriate sample bottle and labeled. The sample position 4 sample
5-7
-------
en
i
TO
POINT 1A
SAMPLE
BLEED
VALVE
DILUTION
BOARD
EXHAUST
PRESSURE RELIEF VALVE
Figure 5-4. Sample Train Schematic
-------
vo
il ii
LATEX
LATEX
STEAM/N2
VAPOR
BUILDING WALL • *
1 !
1 I
t
i
»— •
.TO CYCLONE
RAW
LATEX
FROM
RtACTO!
^r
12'
TO
TRANSFER
LINE
RE-FLOC TANK
t
PolntT" *
U —
Figure 5-5. Slowdown Tank #4
-------
en
i
Sample
Train
System
With FID
And
Integrate*
Bag
Set-ups
Pump
Pump
ser
Gas Cylinders
Figure 5-6. Central Vacuum System (Floor Plan View)
-------
r
Steam
Injection.
To A
->
i
mospl
VSA
/
'-t
4" . .
s
VACUUM BUILDING
iere
' *^ 3 1 /D"
Relief * • ' .
RJvaive
.
S 7'°" ™ in1 1
\ ° /f\ 1D ^
Figure 5-7. Central Vacuum System - Exhaust Piping: Sample Point #1
-------
on
i—"
ro
To Stack
66"
Orifice
» \
10"
103"
From Knock-
out pot
\90°
i
4.
Thermal Couple
Differential Insulation
Pressure
Cell
Magnehelic
Figure 5-8. Vacuum System Flow Measurement (Side View)
-------
Dilution Board
en
i
POINT!
Sample
Bleed
Valve
Sample
Valve
Condensers
Before Pump
Bleed Valve
Integrated
Bag Sample
Valve
Bag Sample Can
Figure 5-9. CVS Sample Train Schematic (Point #7)
-------
was obtained from the reactor dumping into the blowdown tank. The
temperature of the sample was taken with a metal thermometer directly
from the sample bottle. The sample position 5 (Figure 5-5) latex was
sampled after the degassing period of the blowdown tank, and the latex
from the blowdown tank was being processed to the shaker screens.
5.2 CENTRAL VACUUM SYSTEM (CVS)
The central vacuum system is maintained as an outlet for the
process gas emissions. The system pulls the gas emissions from the
blowdown tank system through a condenser to the CVS. The CVS contains a
knockout tank as a control device before emitting the gas stream through
a stack to the atmosphere. A gas sample location was maintained after
the gas stream had been processed through the CVS and before the stream
is vented to the stack (sample point 1). The gas stream at point 1 was
sampled by a continuous monitor FID for total hydrocarbons, and the
modified EPA Method 110 was utilized in obtaining an integrated bag
sample for GC/FID analyses.
Figure 5-6 is a plan view schematic of the central vacuum system
(CVS). Figure 5-7 is a side view schematic of the central vacuum system
indicating the sample point. Figure 5-8 shows the arrangement of the
flow measurement equipment prior to the CVS sample acquisition point.
Figure 5-9 is a diagram of the sample train used to quantify the VOC
emission rate to the atmosphere from the central vacuum system.
5.2.1 Sampling Dimensions at the Central Vacuum System
The sample acquisition from the CVS was taken from the 4-inch line
exiting the central vacuum system. Positive pressure was encountered at
the points of sample acquisition and flow measurement. A 1 1/8" pipe
was tapped into the 4-inch exhaust line to provide an easier sampling
point. The tap was equipped with valves to regulate the amount of
sample desired. The exhaust line (4") from the CVS was partially
insulated and, therefore, in order to maintain sample integrity, the
1 1/8" line was also wrapped with insulation to prevent excess
condensation. The sample point was 30 equivalent diameters downstream
from the nearest flow disturbance. Steam was injected into the 4-inch
line approximately 8 feet after the sample point.
5-14
-------
The flow rate of the flue gas from the central vacuum system was
measured prior to the sample point by pressure differential across a
known orifice which was installed approximately 28 equivalent duct
diameters downstream from the nearest flow disturbance and approximately
25 equivalent duct diameters upstream from the nearest flow disturbance.
The orifice had a nominal diameter of 1 inch during the first week of
testing and a 2-inch nominal diameter during the second week of
testing. The pressure gages installed in the pipe upstream and
downstream from the orifice (P, and P2, respectively) were connected to
the D instrument, similar in design to the instrument utilized at the
blowdown tank flow measurement site. Because of the low pressure
differential across the D cell, a magnehelic (0" - 10" of water) was
installed in line with the D cell to monitor changes in the flow
profile at low flow conditions.
5.2.2 Sampling Procedures at the Central Vacuum System
The sampling activities at the central vacuum system operated
without any major disruptions. A few miscellaneous notations, however,
are in order. The sampling conducted on 3/27/80 was conducted by the
integrated bag - evacuated can method only. Total mass emissions to the
atmosphere monitored by the FID hydrocarbon analyzer were determined
only during the second week of testing.
The plant recorded both the central vacuum system temperature and
vacuum of the system at the vacuum pumps inside the CVS building. This
data was not identified as being available during the actual testing
period. However, examples of CVS temperature and vacuum are included in
the process description appendix. For testing purposes (flow
measurement), the temperature of the flue gas was measured with a type K
thermocouple wire against the outside of the 4-inch exhaust pipe. The
temperature was within ± 5°F from the plant's measurement system.
5.3 STYRENE STRIPPER SYSTEM (SSS)
The latex stream from the pre-blend tank is processed through the
styrene stripper system. The gas emissions from the stripping process
3/26 and 3/27.
5-15
-------
were vented to the atmosphere (sample point 2). The efficiency of the
stripper system was determined by sampling the latex at the inlet
(sample point 8) and at the outlet (sample point 9).
Figure 5-10 is a functional diagram of the styrene stripper system.
The SSS used steam to strip residual styrene out of the latex. The
vapor stripped from the latex is pulled under vacuum to a water-cooled
condenser (Figure 5-11). The cooled exhaust is pulled under vacuum from
the condenser to the styrene vacuum pump from which the exhaust gas is
vented to the atmosphere. Figure 5-12 is a schematic of the sample
train utilized to measure the atmospheric emissions from the styrene
stripper system exhaust (sample point 6).
The hydrocarbon emissions to the atmosphere were determined by
continuous monitoring with the Horiba hydrocarbon analyzer on 4/2/80.
Integrated bag samples (3) were taken by the modified EPA Method 110 to
quantify the individual components of the gas stream. Analysis of the
bag samples was by GC/FID. The inlet latex and the outlet latex were
sampled by the latex grab method (Appendix H), three (3) times during
the SSS exhaust monitoring. The condensate from the condenser could not
be sampled because of dilution with city water.
5.3.1 Sampling Procedures at the Styrene Stripper System
The inlet latex samples were taken from the product level tank
(Figure 5-10) by dipping with the sample bottle. Temperatures of the
system were checked at each of the vapor liquid separators, the
condenser vapor exhaust, and the condenser water inlet. The vacuum of
the system was monitored prior to and after the condenser. The outlet
latex sample was taken after the fourth stage pump from a sample valve.
The latex sample was sampled according to the standard procedure
(Appendix H). The latex at this point was pumped to the final product
storage tank. The system exhaust was measured for VOC concentration and
flow at sample point 2. This point was located outside the filter
building, approximately 8 feet above the ground. For sampling purposes,
a flexible exhaust hose was used to extend the vent to ground level. A
vane anemometer was installed in line to measure the volumetric flow. A
teflon line was also inserted into the flexible duct to draw out a vapor
sample for the sampling train (Figure 5-12).
5-16
-------
Under Vacuum
Floor
Control
Valve
Product
Level
Tank
(Sample
Point 6)
Steam
Figure 5-10. Styrene Stripper System Function Diagram
-------
en
i
CO
Syst
Temper
Gauge
em
at
— • —
Vacuum
ure O
'vt
3
•O
.c
X
LU
1.
O
O.
IO
^
''
1
, !
f- 5"
1
J
*t
Diameter
w
Hj
/
< .
MS'
\
7
i 1 1
v\
1"
,o
\
-o*-
-------
JP
Outside Vent
I
t—«
vo
Exhaust
1/4"
Flow Meter Tefl£n Une
Flexible Tubing
Figure 5-12. Sample Train Styrene Stripper Exhaust - Sample Point #6
-------
Two minor operational problems were encountered during the sampling
of the styrene stripper system. The latex inlet sample was taken by
dipping from the incoming product line which was plugged with coagulated
latex. However, the representativeness of the sample was then judged to
be adequate. A condensate sample from the vapor condenser was desired
for the purpose of formulating a more complete mass balance
determination. However, the system design and operation was such that a
condensate sample could not be collected prior to dilution with city
water. The amount of dilution was undeterminable and was estimated to
be large by the plant process engineer. Therefore, no condensate sample
was obtained from the styrene stripper system.
5.4 THE PRODUCT STORAGE TANK (PST)
A product storage tank was sampled to determine gas emissions to
the atmosphere (sample point 11) during the filling operation. Tank #39
was sampled during a final product filling operation from the
styrene stripper. The hydrocarbon vapors were sampled by the integrated
bag method, and the flow of exhaust was measured by a vane anemometer at
the tank vent. Figure 5-13 is an aerial view and Figure 5-14 is a side
view of the tank. The tank was heated with a steam jacket and was
agitated with a mechanical mixer. The gas sample point was maintained
3 feet inside the tank from the headspace of the tank. The point was
obtained by inserting the teflon sample line through the tank vent
exhaust. The gas integrated bag sample was obtained by the modified EPA
Method 110 for analyses for GC/FID.
5.4.1 Sampling Procedures at the Product Storage Tank
Tank #39 was approximately 25' in diameter. The top of the tank
was enclosed with a standard safety railing, where the 4-inch diameter
tank vent was located. A flexible hose extension was added to the tank
vent for the flow measurement. A large capacity dry gas meter was
intended to measure the tank flow during filling, but the flow from the
vent proved to be below the measuring capacity of the meter. Therefore,
a vane anemometer was used to monitor the flow during the second and
third integrated bag samples taken at this location on 4/2/80.
5-20
-------
Product Mixer
Tank Access-
Measurement Port
Tank Vent
Exhaust
Steam
Figure 5-13. Product Storage Tank - Aerial View
5-21
-------
66"
Probe from
Sample
Point 11
to Sample
System
Figure 5-14. Storage Tank: Sample Point #11 (Side View),
5-22
-------
The manhole on the top of the tank was fitted with a stainless
steel sampling probe and sealed during the bag sampling. Between
integrated bag samples, the latex volume of the tank was measured. The
manhole was reopened and the level in the tank was measured by a
500-inch tape measure. The tape was read as inches of tank outage, with
each inch representing 195 gallons of product. Outside of the flow
measurement difficulties during the first integrated bag, no sampling
problems were encountered.
5.5 SHAKER SCREENS
The latex stream from the pre-blend tank is processed through a
shaker screen system. The coagulated latex is extracted on the screens
and disposed of, while the filtered latex is pumped through the system
to temporary storage. The gaseous emission from the filtering process
(sample point 3) is emitted as a fugitive emission into the process
room.
Figure 5-15 is a schematic of a shaker screen and presents the
location of sample point 3. Figure 5-16 is a generalized schematic of
the shaker screen set in the process before the stripper system. Only
one integrated bag sample was taken by the evacuated can method
according to the modified EPA Method 110 for analyses by GC/FID. The
sample was obtained over the shaker screens during operation. The
ambient concentration of styrene was expected to be low.
5.5.1 Sampling Procedures at the Shaker Screens
A stainless steel probe was extended over one of the shaker
screens, approximately 20 inches above the first stage of the shaker
screen. A 30 minute bag sample was collected with the evaluated can
system. Temperature above the shaker screens was measured with a Type K
thermocouple attached to the probe assembly. No sampling problems were
encountered at this location. Only one bag sample was collected.
5-23
-------
Incoming Latex
I
ro
42"
Sample
Point
#3
Screen
67"
42"
9"
9"
\
Screen
\
Product
Storage
I
To Waste
Prefloc
Dumpster
Figure 5-15. Shaker Screens
-------
Incoming Latex
en
i
ro
en
1
To Waste
Prefloc
Dumpster
To Waste
Prefloc
Dumpster
i
To Waste
Prefloc
Dumpster
Figure 5-16. Shaker Screens - Plan View - 2nd Floor Level
-------
6. SAMPLING AND ANALYTICAL PROCEDURES
Section 6 discusses all the sampling and analytical procedures
applied to all the samples. Modifications of the draft methods are also
discussed in depth in this section. The section is divided into four
discussions:
Section 6.1 describes the sampling methods used at each sample
location.
Section 6.2 describes the analytical procedures used or specific
types of samples and types of analysis, for example hydrocarbon and
analysis of latex samples by GC/FID.
Section 6.3 discusses the hydrocarbon emission concentrations
(determined by continuous monitor FID) in relation to the process as
well as a comparison of the FID results and the integrated bag sample
results (determined by GC/FID).
Section 6.4 discusses the Quality Assurance and Quality Control
measures followed throughout the test as well as the results of a
degradation rate study of both the gaseous hydrocarbons and the latex
hydrocarbons.
6.1 SAMPLING PROCEDURES
Pertinent data compilation and sample collection provided
information for the analysis and characterization of VOC emissions and
the efficiency of the control devices in the SBR process. The process
data monitored were gas flows, liquid flows, and temperatures. The gas
flows were measured with a magnehelic, a vane annemometer, or plant
installed standard orifice with differential pressure readouts. The
liquid flows obtained from plant GPM meters or tank level readouts. The
temperature was monitored with K-type thermocouple wire with temperature
-------
readout or metal thermometers. Moisture determination was not monitored
because condensate from sample points were analyzed and found to be
liquid hydrocarbons. Gas sample were collected using the integrated bag
system with an evacuated can according to the modified EPA Method 110.
The gas stream was also continuously monitored with a Horiba OPE-405
hydrocarbon analyzer utilizing flame ionization detection (FID). A
dilution board adaption was required at the gas streams with high HC
concentration levels. The latex samples were collected using a latex
grab collection procedure. Appendix H, Field Sampling Procedures,
contains the procedure outlines utilized for the sampling at the
Mogadore facility. A following discussion presents the procedures used
at each sample location with the adaptions from the method procedures.
6.1.1 Slowdown Tank Gas Outlet, Sample Procedures
This sample location was labeled 1A to denote this location as a
modification of sample location 1. This modified sample location
allowed the gas stream to be sampled before processing through the
condenser in line to the central vacuum system. Sampling modification
with condenser bottles was required to eliminate the condensate from the
gas stream.
6.1.1.1 Flow Measurements. The flow measurements at the blowdown
tank gas outlet were determined with an orifice placed in-line between
the BDT #4 outlet and the condenser. The differential pressure gauge
was a Foxboro Instrument Model #13A-1 and was installed by the Mogadore
personnel. Problems developed during the second test day when liquid
formed around the 1-inch orifice originally installed. The restrictions
caused error in the differential pressure readout. This problem was
overcome by installing a 2-inch orifice.
6.1.1.2 Temperature Measurements. The temperature of the outlet
gas stream was monitored from a dial thermometer temperature gauge
installed by the plant inline between BDT #4 outlet and the condenser.
The temperature gauge was before the orifice as in Figure 5-1. The dial
thermometer was not calibrated for accuracy. But the precision of an
instrument of this type is approximately ±10°F.
6-2
-------
6.1.1.3 Gas Sampling Procedures. Integrated gas samples and
continuous HC monitoring were obtained at the blowdown tank outlet. The
gas samples collected utilized the modified EPA Method 110 presented in
Appendix H with the evacuated can. The sample system was modified with
the dilution board and condenser (Appendix G). The dilution board was
calibrated and operated according to the procedure in Appendix H. The
condenser bottles were placed before and after the sampling pump. The
amount of liquid captured in the condenser bottles was recorded and
analyzed. A significant problem occurred during the degassing of the
blowdown cycle. The injection of steam into the blowdown tank exceeded
the capacity of the condenser bottles. Therefore, the train was shut
down numerous times during this cycle to install new condenser bottles.
This condensate problem could have caused error in the accuracy of the
integrated bag concentration during this cycle because the train was
down approximately 12 percent of the steam degassing period.
6.1.2 Blowdown Tank Inlet and Outlet Latex Sample Procedures
The latex flow in and out of BDT #4 was sampled and analyzed for
residual styrene concentrations present in the latex at this point of
the process. The inlet latex sample (sample position 4) was taken from
the product sample valve of the reactor dumping into BDT #4
(Figure 5-1). The outlet latex (sample poisition 5) flow was obtained
from a product sample valve at the BDT #4 before transfer to the shaker
screens (Figure 5-5).
6.1.2.1 Flow Measurements. The latex volume flow measurements at
the inlet and outlet latex streams of BDT #4 were calculated and
recorded by the plant personnel assuming a typical percent completion of
the reaction. The total inlet volume flow of latex was based on the sum
of the theoretical volume of the three reactors dumping into BDT #4
during the test period. The outlet flow recorded and reported in
Appendix D was estimated from the change in the level of BDT #4 as it
was emptied and the corresponding percent of the total tank capacity
that this change represented of BDT #4 that was emptied. Outlet flow
measurement was not required in the calculations for emission rates.
The inlet flow measurements were used for the material balance
6-3
-------
calculations assuming that the same amount of latex exited the BDT as
was put into the tank.
6.1.2.2 Temperature Measurements. The temperature of the latex
sample at the inlet and outlet were taken from direct readouts with a
metal thermometer. The thermometer was placed in the sample jar while
the sample was being obtained.
6.1.2.3 Latex Sampling Procedures. The inlet latex sampling was
coordinated with the process engineer to obtain the timing of the
reactors' releasing into the blowdown tank. The reactors were monitored
for the valve to BDT #4 to open. The sample was taken 5 minutes after
the valve opened. The sample was taken from the product sample valve at
the bottom of the reactor. This valve was opened, and the latex was
allowed to flow at a constant rate into a bucket for approximately
30 seconds. The latex was then obtained and prepared according to the
latex sampling procedures in Appendix H.
The outlet latex sampling was coordinated with the plant process
engineer to obtain the timing of BDT #4 emptying its contents to the
shaker screens. The sample was taken approximately 5 minutes after the
valve had been opened to process the BDT #4 latex to the shaker screens.
The sample was obtained from a product sampling valve. The latex was
purged from the sample valve, obtained, and prepared the same as the
inlet latex sample.
6.1.4 Central Vacuum System Sampling Procedures
The central vacuum system exhaust sample location was labeled #1,
and sample position 1A was a portion of this gas stream. The gas stream
at #1 was the gas concentration emitted to the atmosphere from the blow
down process and other processes which were exhausted from the central
vacuum system.
6.1.3.1 Flow Measurements. The flow measurements at the central
vacuum system outlet were determined with an orifice placed in-line
between the central vacuum system and the stack outlet to the
atmosphere. The orifice was installed by plant personnel according to
manufacturer's specifications (see Section 5.2.1). A 0"-10" of water
magnehelic was installed in line with the D cell to allow for the
monitoring of the low flows.
6-4
-------
6.1.3.2 Temperature Measurements. The temperature of the outlet
gas stream from the central vacuum system was monitored during the
sampling period using a K-type thermocouple wire attached to a Digimite
temperature display.
6.1.3.3 Gas Sampling Procedure. Integrated gas samples and
continuous monitoring of the hydrocarbon level were obtained at the
central vacuum system outlet. The evacuated can system was required for
the integrated gas sample according to the modified EPA Method 110
described in Appendix H. The Horiba OPE-405 hydrocarbon analyzer was
utilized as the continuous monitoring of the outlet stream. The Horiba
OPE-405 utilizes the flame ionization detection method of analysis and
was operated according to the procedures in Appendix H. The timing of
the sampling was simultaneously with the blowdown tank outlet.
6.1.4 Shaker Screen Exhaust Sampling Procedure
This sample location was labeled position 6. This sample location
was the fugitive gaseous emissions escaping from the process of
filtering the latex through the shaker screens. A three tool metal
probe was added to the sample probe for maintaining the sample point in
the center of the shaker screen.
6.1.4.1 Flow Measurements. No flow measurements were obtained
from the shaker screen emission.
6.1.4.2 Temperature Measurements. The temperature of the gas
stream being sampled was monitored with a K-type thermocouple wire
®
connected into a Digimite temperature display. The point monitored was
at the sample point by securing the thermocouple to the sample probe.
6.1.4.3 Gas Sampling Procedure. An integrated gas sample was
obtained during the shaker screen operation. The evacuated can system
was required to obtain the sample according to the modified EPA
Method 110 presented in Appendix H.
6.1.5 Steam Stripper Vacuum System Outlet
This sample location was labeled position 2 and was the gas
emissions from the stripper system. This location was monitored during
a period when the steam stripper was operating.
6-5
-------
6.1.5.1 Flow Measurements. The flow measurement at the steam
stipper outlet were monitored with a vane anemometer. The anemometer
was installed in the gas stream by modifying the outlet vent with
flexible hosing (see Figure 5-12). The flow through the anemometer was
recorded during the test period.
6.1.5.2 Temperature Measurements. The temperature measurements
®
were obtained with a K-type thermocouple wire connected into a Digimite
temperature readout. The thermocouple was placed in the gas stream flow
through a port at the end of the modified outlet vent (flexible hose).
6.1.5.3 Gas Sampling Procedures. Integrated gas samples and
continuous HC monitoring were obtained at the styrene stripper outlet.
The integrated gas sample was obtained with an evacuated can sampling
system according to the modified EPA Method 110 as described in
Appendix H. The continuous monitoring was with a Horiba OPE-405
hydrocarbon analyzer. This instrument uses the flame ionization
detection method of HC analyses and was operated according to the
procedures in Appendix H.
6.1.6 Steam Stripper Inlet and Outlet Latex Sampling
The inlet latex flow to the stripper was labeled sample position 8,
and the outlet latex flow was labeled position 9. The samples were
obtained simultaneously and taken during the sampling of the outlet
gaseous emissions from the steam stripper.
6.1.6.1 Flow Measurements. The flow measurement for the latex
stream feeding through the steam stripper was moniotored on a GPM meter
by the plant personnel. This feed rate was recorded and utilized for
the inlet and outlet latex flow.
6.1.6.2 Temperature Measurements. The temperature of the latex
sample at the inlet and outlet were taken from direct readouts with a
metal thermometer. The thermometer was placed in the sample jar and
monitored while the was being taken.
6.1.6.3 Latex Sampling Procedures. The outlet latex sample was
taken directly from a sample production valve according to the latex
sampling procedure described in Appendix H. But a modification of this
latex sampling procedures was required at the inlet. The inlet latex
6-6
-------
sample was taken from an open tank feeding into the stripper. The latex
sample was obtained by dipping the sample jar into this tank.
Precautions were taken to obtain the sample from the center of the tank
and below the hardened layer of latex to ensure a representative sample.
6.1.7 Storage Tank Vent
The sample location at the storage tank vent was labelled
position 11. The sample timing was coordinated with the steam stripper
gas outlet sample and the operation of filling the storage tank.
6.1.7.1 Flow Measurements. The flow measurements were obtained
with a vane anemometer installed in the tank vent. The flow through the
anemometer was monitored during testing.
6.1.7.2 Temperature Measurement. The temperature measurements
were obtained with a K-type thermocouple wire connected into a Digimite
temperature readout. The temperature was monitored at the sample point
by securing the thermocouple to the sample probe.
6.1.7.3 Gas Sampling Procedures. An integrated gas sample was
obtained during the filling of the storage tank. The evacuated can
system was required in obtaining the sample according to the modified
EPA Method 110 as presented in Appendix H.
6.2 ANALYTICAL PROCEDURES
There were two types of samples analyzed by GC/FID: integrated gas
bag analyses and latex grab analyses. Since each type required special
handling and preparation a brief discussion of the methods and any
deviations from the methods are described below.
6.2.1 Analysis of Each Gas (Bag) Samples
Each gas (bag) sample was analyzed on-site, normally within thirty
minutes of collection. Prior to the field program, gas samples were to
be analyzed using two column types; 6' by 1/8 inch stainless steel
packed with AT-1200 Bentone 34 for the determination of benzene,
toluene, ethylbenzene, xylene, and styrene, and a Porpak Q column for
butadiene and C,-Cg hydrocarbons. Due to the high concentrations of
butadiene relative to C,-Cg hydrocarbons, the Porpak Q portion of the
gas chromatographic analysis was dropped. Subsequent gas analysis was
performed in the differential mode with two instruments (Shimadzu Mini I
6-7
-------
and Shimadzu Mini II) with dual AT-1200 Bentone 34 columns. Figure 6-1
shows the flow line and valve sequence applicable to both the Shimadzu
Mini 1 and Mini 2. Gas samples were flushed through a 1 ml sample loop
at 0.6 1pm for 30 seconds, allowed to equilibrate, and injected in the
column. Figure 6-2 shows the gas sampling system used for all heat
samples. Analysis time was approximately 15 minutes per gas sample
injection. Quantisation was by appropriate response factors with
®
integration by two Shimadzu-CRIA chromatopaks. A third Shimadzu gas
chromatograph equipped with a conductivity detector, and molecular sieve
5A and silica gel columns, was used to quantify stationary gases.
The analysis of the gas samples indicated a saturation effect of
the FID detectors when more than 5% total hydrocarbons were injected
into the gas chromatograph. An alteration in the standard analytical
procedure was required due to the high concentration of butadiene. This
required two separate analyses on each sample. The first analysis
required an injection of the neat sample to determine the concentrations
of the components excluding butadiene. The second analysis required
dilution of the butadiene sample to prevent saturation of the detector.
The identified and quantified compounds of the first injection were
utilized as internal standards to determine the dilution ratio for
butadiene.
6.2.2 Analysis of Latex Samples
Latex samples were analyzed by GC/FID using an AT-1200 Bentone 34
column and undecane as an internal standard. The residual sytrene in
latex was stabilized using TBC in methylene chloride solution. One
change in procedure was implemented on-site. The EPA Method for
Determination of Residual Styrene recommends that a 50 ml aliquot of
latex be dissolved in 150 ml of TBC/MeCl solution. The field samples
were insoluble. A series of extractions demonstrated that three
methylene chloride extractions would remove 95 percent of the residual
styrene. In addition, an initial saturation period of greater than
8 hours with the first extraction was found to be beneficial.
Therefore, 100 ml of TBC/MeCl solution was added to each latex sample
(50 ml), sealed, and allowed to stand up to 8 hours. The methylene
6-8
-------
K/F1D VALVE
INJECT
POSITION
tt/FID VALVE
LOAD
POSITION
.COLUMN
COLUMN
TO DETECTOR
CARRIER
IN
SAMPLE
OUT
CARRIER
IN
SAMPLE
LOOP
TO DETECTOR
SAMPLE
LOOP
SAMPLE
OUT
INJECTION
PORT
COLUMN
O
DETECTOR
PLOTS
DIFFERENTIAL
Figure 6-1. Flow Line
6-9
-------
en
i
FLOW HETER
GAS INLET SAMPLE LOOP
EXHAUST
CHROMATOPAC
ELECTRICAL CONNECTION
INTEGRATED BAG
Figure 6-2. GC/FID Gas Sampling System
-------
chloride (MeCl) was decanted and followed with two additional
extractions of 100 ml each to complete the extraction. Both the second
and third extractions required the addition of 100 ml of TBC/MeCl,
extraction of the latex for 10 minutes, and the combination of the
extraction volume with the previous one. The total extraction volume
was 300 milliliters.
6.3 CONTINUOUS MONITORING AT THE SLOWDOWN TANK OUTLET, CENTRAL VACUUM
SYSTEM OUTLET, AND STYRENE STRIPPER VACUUM OUTLET WITH A FLAME
IONIZATION DETECTOR
A Horiba OPE-405 hydrocarbon analyzer was utilized as the
continuous monitor at the blowdown tank outlet, central vacuum system
outlet, and styrene stripper vacuum outlet. The instrument was set up
at each position on separate tests day and operated according to the
Horiba Operational Manual. The FID was calibrated with standard gases
on a propane basis. Therefore, the FID results are given in total
hydrocarbons as propane. The following sections explain the FID results
from the test day at the particular location and compare the results to
the plant process and integrated bag results.
6.3.1 Continuous Monitor by FID at the Blowdown Tank Outlet
(March 27, 1980)
Figure 6-3 shows a comparison between the plant process, the FID,
which was set up at the blowdown outlet and gave a reading of THC as
propane, and the bag samples which were analyzed by compound as propane.
During each reactor dump into the blowdown tank, the FID showed high
4
peak readings between 18 and 24 x 10 ppm THC. During most of the
degassing period, no FID data was obtained due to failure of the sample
pump. It can be seen toward the end of the degassing that the THC
4 4
concentration decreased to between 2 x 10 and 10 x 10 ppm THC. The
bag samples represented averages taken throughout the day. The compound
of the highest concentration was butadiene which ranged from 2.7 to 11.2
4
x 10 ppm. Styrene was the next highest, ranging from 0.11 to 0.60 x
4
10 ppm. The concentration of each bag corresponded to be average FID
reading during the same time frame.
6.3.2 Continuous Monitor by FID at the Central Vacuum System Outlet
(March 31, 1980)
Figure 6-4 shows a comparison between the plant process and the FID
which was set up at the central vacuum system. No bag samples were
6-11
-------
VP
w
I
BUT
Transfer
V-
epT-1
v*~l
BDT-43
BffM
BDT 4-*Degm1ng
i
»—*
ro
i-ii
19
]4.S BuUdUiw 11— °'18 Sti
Hag A II 2.8 But^dUlM
Baa B
4.0 ButadUn*
BaoC
0.
0.31 Styrtr*
3 BuUdUiw
Bag 0
Bi
58 Sly.
11.2
tadl
Bag
f»|0.60St>j
E t.7 ButacleM
3/27/80
Bag F
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
19 No FID
-------
w
!
•or
Transfer
-»% Stripper 14
BDT-
-» I R12 —
T-3 | »BDT-4
807-4
I BDT-3 Degassing
IL8-*
*BDT-4
•J 15
|& 10
* 5
3/31/80
1100
1200
1300
1400
1500
1600
No teg sanplts MBit taken.
Figure 6-4. FID Results at the Central Vacuum
System Outlet on March 31, 1980
6-13
-------
VP
BOT
Transfer
M.
-4
•iDper-'-Peqasstng
R.8 -»
BOT-4
BDT-4
I BDT-3-Deaamnq
J
aDT-4-DeQassTng
i
i—•
.&
20
•fel «
s M
lh15
MX
.1- 10
4/1/80
7.8 BuUdlcM
Bag A
2.3 BuUdene Baa C
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
Ho FID data obtain*!.
Figure 6-5. FID Results at the Central Vacuum
System Outlet.on April 1, 1980
-------
•< to
Inlet latex — Stripper
20
^ gf 15
O *• X
2&">
*Et 10
y*§
p •».»•
-£ 5
20
»
E 2
"-
a
0.61Styren»
0.87 S^yrene
0.57Styrene
4/2/80
^0.03 Butadiem
Bag A
• 0.04 Butadiene
Bag B
•0.04 Butadiene
Bag C
1200
1300
1400
1SOO
1600
1700
9 Inlet latci feed stren remained constant durlMf test period.
Figure 6-6. FID Results at the Styrene
Stripper Outlet on April 2, 1980
6-15
-------
taken at this location on this day. (Prior to the reactor dumps the FID
4 • 4
showed readings between 1 x 10 ppm and 2 x 10 ppm THC as propane.)
During each reactor dump into the blowdown tanks, the FID showed high
4 4
peak readings between 13 x 10 and 18 x 10 ppm THC as propane. The
effects of the degassing process could be seen only inbetween.the
reactor dumps due to the low concentration of the emissions, which were
4
between 2 and 4 x 10 ppm THC as propane.
6.3.3 Continuous Monitor by FID at the Central Vacuum System Outlet
(April 1, 1980)
Figure 6-5 shows a comparison between the plant process, the FID
which was set up at the central vacuum system, and the bag samples.
During each reactor dump into the blowdown tank the FID showed high peak
4
readings between 7 and 14 x 10 ppm THC as propane. Butadiene was the
4
component with the highest concentration averaging 8.1 x 10 ppm as
propane during this reactor blowdown period. Styrene had a
4
concentration averaging 0.30 x 10 ppm as propane. The FID readings
4
decreased during the degassing period to lesss than 5 x 10 ppm. During
this time the butadiene concentration in the bag decreased to 2.3 x
4
10 ppm. During the last reactor dump the FID showed a peak of 18 x
4
10 ppm, then continued to decrease as the degassing process continued.
6.3.4 Continuous Monitor by FID at the Styrene Stipper Outlet
(April 2, 1980)
Figure 6-6 shows a comparison between the FID which was set up at
the stripper system and the bag samples that were collected. The FID
4 4
showed concentrations between 1.0 x 10 and 3 x 10 ppm THC as propane.
According to the analysis of the bag samples, the compound with the
highest concentration was styrene which ranged from 0.67 to 0.87 x
4 4
10 ppm as propane, butadiene ranged from 0.03 to 0.04 x 10 ppm as
propane.
6.4 QUALITY ASSURANCE AND QUALITY CONTROL
Attaining the necessary results required specialized analytical
techniques and considerations. Therefore, this section was divided into
three subsections to discuss each procedure in depth. The three
subsections are:
6-16
-------
6.4.1 CALIBRATION
6.4.2 DEGRADATION STUDIES
6.4.3 AUDIT SAMPLES
6.4.1 Calibration
The daily calibration procedure followed throughout the field
testing consisted of an instrument warm-up of thirty minutes followed by
area integration of known benzene and butadiene gas standards and area
integration of a liquid standard containing 1% styrene and 1% undecane.
Replicate standard injections were made with a deviation in the area
counts of less than 5%. Standards were injected during analysis to
assess the stability of the instrument response. Gaseous standards were
injected in the same manner as gaseous samples (i.e.,. the sample loop
was flushed with gas at a rate of .5 liter per min. for thirty seconds.)
From the area counts and the concentration of the standard, a response
factor was calculated daily and applied to the area counts of the
compounds of interest in the samples to determine concentration in ppm
(Appendix B). Prior to any sample analysis, the instruments were
calibrated with a 1300 ppm butadiene standard and a 98.1 ppm benzene
standard.
The liquid standard used in this test was 1% styrene and 1%
undecane in methylene chloride TBC solution. One pi injection was made
into the GC/FID and the area counts of undecane and styrene recorded
until they were reproducible to within 5%. A response factor for
styrene was calcualted from the area counts and concentration of styrene
in ppm. An average area count was determined for the undecane internal
standard.
Linearity of the instruments was determined for both benzene and
styrene. Benzene gas standards with ppm concentrations of 82, 106, and
495 were used in the analyses (see Figure 6-7) with calculations of mean
and standard deviation included. Liquid styrene standards were prepared
with percent concentrations of 0.1, 0.25, and 1.0. Figure 6-8 shows the
linearity of the instrument, with the mean and standard deviation
calculations included.
6.4.2 Degradation Studies
Degradation rates were not obtained from the original presurvey
samples due to the time delay of days between the sampling and the
6-17
-------
160,ooor
140,000
120,000
100,000
to
§ 80,000
o
o
OJ
60,000
40,000
20,000
Mean x
ppm Area Counts
8.2 2,646
106 36,639
495 156,957
I/Slope S.D.
0.0031 260.9
0.0029 3039
0.0031 4051
100 200 300 400
Benzene Concentration (ppm)
500
Figure 6-7. Instrument Linearity for Benzene
6-18
-------
3
O
LJ
ID
O)
2.000,000r
1,800,000
1,600,000
1,400,000
1,200,000
1,000,000
800,000
600,000
400,000
200,000
Mean Area
% Styrene Counts S.D. I/Slope
0.1 183,608 4602 5 x 10"!
0.25 433,245 4511 5 x 10~i
1.0 1,839,517 46,658 5 x 10"'
i t t
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
% Styrene
Figure 6-8. Instrument Linearity for Styrene
6-19
-------
initial analysis. Previous observations of the degradation behavior in
bags indicated the necessity of this type of study. TRW had confirmed
that degradation of gaseous hydrocarbon samples was severe and substance
specific. For example, a sample containing benzene, styrene, and other
hydrocarbons exhibit a 15% loss of benzene in 48 hours and a 65% loss of
styrene over the same time period. Hydrocarbons whose retention times
fall between benzene and styrene exhibit similar losses.
Another related consideration involves the temperature differential
affecting the sample between the 120°F stack and ambient conditions in
the field lab. The temperature effects were investigated at the field
site by initial analysis of a heated bag sample (~100°F) and repeated
analysis over a four-hour period. The time between collection and
analysis, not temperature, proved to be the major consideration.
Although the degradation, or.loss of butadiene and benzene (<10%) was
acceptable, the percent loss for molecules with a carbon number of C-,
and greater ranged from 15 to 59%. The degradation loss is shown in
Figures 6-9 to 6-12. Based upon the degradation study, all bag samples
during the test were analyzed within 1 hour of collection. In most
cases, analysis was within thirty minutes of collection.
Preliminary investigation of the latex samples indicated that
styrene in methylene chloride solutions containing 4-tertiary
butylpyrocathecol (TBC) would remain constant over a 3-day period.
Initial field analyses required the determination of stability of
styrene in methylene chloride containing TBC. This was demonstrated by
the repeated analysiis of two latex samples over a period of 9 days with
a degradation of less than 2 percent (Figure 5-6). Based upon these
observations, latex samples were analyzed within 9 days of sampling.
6.5 AUDIT SAMPLES
EPA supplied three gas samples of unknown compounds and
concentrations. The audit samples were analyzed after the daily
calibration procedures (Section 6.1) in the same manner as all the gas
samples. The compounds were identified by their retention times, and
the concentrations were calculated from the daily response factors
(Table A-10).
6-20
-------
25.
20- •
S
15- •
10- -
5- •
Butadiene
Styrene
— H
3
T(HOURS)
Figure 6-9. Degradation Study - Hydrocarbons
Bag Sample - Mogadore, Ohio
6-21
-------
£
o.
i
g
|
«_>
g
1000. .
Toluene
Ethyl Benzene
Xylenes
T(HOURS)
Figure 6-10. Degradation Study - Hydrocarbons
Bag Sample - Mogadore, Ohio
6-22
-------
7. .
6.
I 5-.
I «!
I 14.
• - Benzene
T(HOURS)
Figure 6-11. Degradation Study - Benzene
Bag Sample - Mogadore, Ohio
6-23
-------
4.0. •
3.5. •
»o
2 3.0..
E"
o.
8
§ 2.5..
8
£
.0
1.5-
1.0.
Latex Sample IT
•o
4-
4-
H 1 \—
3/25 3/26 3/27 3/28 3/29 3/30 3/31 4/1 4/2 4/3
* - Latex Sample II T(DAYS)
• - Latex Sample 12
Figure 6-12. Degradation Study - Latex (Styrene) -
Mogadore, Ohio - March 25, 1980-April 3, 1980
6-24
-------
EPA also supplied three liquid audit samples of unknown
concentrations. The audit samples were extracted and analyzed as
described in the latex analyses method. The concentrations were
calculated from the daily response factors (Table A-10). The summary or
results of the audit samples appear in Table 6-1. The most significant
error occurred in the analysis of the butadiene gas audit (24% from
audit valve). The reason for this deviation was probably the results of
analyzing a 25 ppm level standard with an instrument calibrated with a
1300 ppm level standard. The samples collected during the test had
extremely high levels of butadiene and even after dilution were closer
in range of 1300 ppm than 25 ppm.
6-25
-------
Table 6-1. AUDIT RESULTS
Sample #
288-BAL111
642-B117
1032-A10567
T
U
V
aValue supplied
k* Hifforenre -
Compound
Benzene
Benzene
Butadiene
Styrene
Styrene
Styrene
by project
measured -
Measured
(ppm)
344.7
100.8
25.9
1048
10261
4187
officer.
audit y inn
Audit
Values3
(ppm)
358
106
20.9
1000
10000
5000
% Difference
-3.7%
-4.9%
23.9
4.8%
2.6%
-16.3%
audit
6-26
------- |