&EFA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 78-CKO-13
July 1980
Air
Steel Industry
By-Product Coke Ovens
Emission Test Report
U.S. Steel Corporation
Clairton, Pennsylvania
-------
SOURCE TEST AT U.S. STEEL
CLAIRTON COKE OVENS
CLAIRTON, PENNSYLVANIA
M. W. HARTMAN
Contract No. 68-02-2812
Assignment No. 23
EPA Technical Manager: Gene Riley
Prepared for:
U. S. Environmental Protection Agency
Emission Standards and Engineering Division
Emission Measurements Branch
Research Triangle Park, NC 27711
TRW Environmental Engineering Division
3200 E. Chapel Hill Rd./Nelson Hwy.
P. 0. Box 13000
Research Triangle Park, NC 27709
-------
GLOSSARY
BaP - Benz-a-Pyrene
BSD - Benzene Soluble Organics
DSCM - Dry Standard Cubic Meters
GC - Gas Chromatograph
MeCl - Methylene Chloride
MS - Mass Spectrophometer
PAH - Polynuclear Aromatic Hydrocarbons
POM - Polycyclic Organic Matter
SASS - Source Assessment Sampling System
SAMPLE LABELING GLOSSARY
CD - Module Condensate
CICD - Impinger Contents
CIMC - Methylene Chloride Impinger Rinse
MR - Module Rinse Methylene Chloride
NMC - MeCl Nozzle Rinse
NPR - Nozzle Probe/Filter Rinse
P - Combined Probe
PF - Filter
PIB - Probe/Impinger Benzene Rinse
PICD - Probe/Impinger Condensate
PR - Probe/Filter Rinse
V - Vent Rinse
VB - Benzene Vent Rinse
VMC - Vent Rinse Methylene Chloride
XR - XAD-2 Resin
-------
TABLE OF CONTENTS
1.0 INTRODUCTION
2.0 SUMMARY
3.0 DESCRIPTION OF RESULTS. . .
4.0 PROCESS DESCRIPTION ....
5.0 LOCATION OF SAMPLING POINTS
6.0 TEST PROCEDURES
PAGE
1
2
5
8
12
14
APPENDIX A: COMPLETE RESULTS AND SAMPLE CALCULATIONS
APPENDIX B: LABORATORY RESULTS
APPENDIX C: FIELD DATA SHEET
APPENDIX D: TEST LOG
APPENDIX E: PROJECT PARTICIPANTS
-------
1.0 INTRODUCTION
During the week of July 30 through August 6, 1978, TRW Environ-
mental Engineering Division, under Contract No. 68-02-2812, with the
Environmental Protection Agency's Emission Measurement Branch, tested a
battery of coke ovens at U.S. Steel's Clairton, Pennsylvania plant. The
purpose of this sampling was two-fold: 1) to provide data associated
with emissions of polycyclic organic matter from topside leaks and 2) to
verify that a reduction in visible topside emissions would result in an
emissions reduction of polycyclic organic material.
Sampling was conducted in order to determine the emission rate
(mg/min) of pollutants from a simulated coke oven topside leak; two
different size leaks were tested. In addition, background ambient
samples were taken from battery topside. The leaks were simulated by
modifying an oven port lid to include a vent tube which utilized a ball
valve for controlling the leak rate. Samples were collected by placing
the nozzle of the sampling train probe directly above the vent tube.
Pollution emission rates were determined for Benzene Soluble Organics
(BSD), Polycyclic Organic Matter (POM), and Benz-a-Pyrene (BaP); determi-
nation of a Benzene emission rate was attempted but was not successful.
-------
2.0 SUMMARY
During the week of testing at the Clairton Coke Works, seven sam-
ples of BSO/POM were taken from the simulated oven leaks on Battery
Number 1. In addition, two ambient air samples were taken at the
battery topside to provide information on the background pollutant
level. During each run the simulated leak was adjusted to give either a
large leak (approximately a three to six-foot visible plume) or a small
leak (approximately a one-foot visible plume). Tests one through four
were taken during a large leak while tests five through seven were taken
during a small leak. Photographs of the sampling apparatus and the
simulated leaks were taken during the test program for EPA by U.S. Steel
personnel; these photographs are on file at the EPA. All sampling was
done on the number one coke oven battery at the Clairton Plant. In
order to minimize problems with taking the sampling equipment onto the
hot battery topside, the ovens closest to the pinion wall were selected
for testing. In all cases, the oven lid closest to the push side of the
battery was used. Sampling was begun between the first and second hours
after the oven was charged. The ambient samples were taken on battery
topside past the pinion wall (i.e., between oven A-l, and the end of the
battery); this position was approximately 40 feet from the simulated
leak. A more detailed discussion of the sampling locations is presented
in Section 5. A Summary of Results are shown in Table 2.1.
Separate sampling trains were used for the BSD and the POM/BaP
sampling. In order to obtain comparable BSO/POM data, a common nozzle
was used so that both trains could simultaneously sample the same leak
The sampling rates were adjusted so that approximately half of the leak
went to each train. Both trains were of a modified Source Assessment
Sampling System (SASS); SASS trains were used in order to obtain a high
sample rate.
-------
SUMMARY OF RESULTS
BSO
POM
BAP
z
• .
t/>
Ul
1
2
3
4
5
6
7
5A
5B
7A
Ul
z:
j_
Ul
i—
12:50
14:23
8:35
10:16
12:53
21:32
16:45
12:33
21:12
15:07
UJ
t—
g
6/2
8/2
8/3
8/3
8/4
8/4
8/5
8/4
8/4
8/5
Ul
CO
z
Ul
g
- A III
A I
A V
A VII
A II
A I
A I
Ambient
Ambient
Ambient
UJ
ISJ
•— *
^
«(
UJ
Lanje
Large
Large
Large
Small
Small
Small .
None
None
None
_^ .
c
^,
i
>a ^
Ul
_l
VI
24
6
15
8
16
15
15
56
55
70
*"-»
i*
*•— •
Q
^^j
a
_ )
8
to
i
784
1700
8819
1618
458
714
1312
9.3
9.0
1?
E -—
cn <_>
-E 2
^~*^ CJ
^ p
1— O
*%
o «t
Ul
tjj ^fc
.12.fi« 1.f>64
283.33 0.118
587.93 1.069
202.25 0.148
28.63 0.672
47.60 0.730
87.47 0.604
0.084 9.988
0.128 6.709
^^
C
5-
^
*J
1—
_J
S
30
9
15
10
15
15
15
55
55
69
c
•r~
? ^» 5
•*** 6 t/>
O *^-
c 1 s
_j ^^ 5T
8 2 1
sag
1.873
0.131
0.637
2189 218.9 0.404
0.823
0.880
511 34.0 0.803
9.608
7.6 0.11 6.347
^^
i*
***>
0
E
**•
g
5
35.83
17.91
0.12
c
*g
•Jj)
g
^"^
1
OC
o
to
a
3.58
1.19
0.0017
co
5A Collected During Run 5
5B Collected During Run 6
Table 2.1. SUMMARY OF RESULTS
-------
POLYCYCLIC ORGANIC MATERIAL ANALYSES
COMPOUND
NAPHTHALENE
BIPHENYL
DIMETHYLNAPHTHALENES
ACENAPHTHYLENE
DIBENZOFURAN
FLUORENE
DIMETHYLBIPHENYLS
DIBENZOTHIOPHENE
PHENANTHRENE
BENZO(h)QUINOLINE
CARBANZOLE
METHYLPHENANTHRENE
4H-CYCLOPENTA
(def) PHENANTHRENE
PHENYLNAPHTHALENE
PYRENE
FLUORANTHENE
BENZO(a)FLUORENE
METHYLPYRENES
BENZO(qhi)FLUORANTHENE
CHRYSENE
BENZO(a)ANTHRACENE
TERPHENYL
METHYLBENZOANTHRALENES
BENZOTHIOPHENE
BENZOPYRENES
BENZOPERYLENES
DIBENZOANTHRACENE
TOTAL POM
MOLECULAR WT.
128
154
156
152
168
166
182
184
178
179
167
192
190
204
202
202
216
216
226
228
228
230
242
240
242
276
278
BAP 4
MASS (mg)
320.2
13.6
10.1
128.8
71.8
81.0
18.3
21.2
374.0
4.4
44.9
33.8
32.0
9.2
153.8
146.2
26.6
46.2
9.8
8.5
150.0
5.4
29.4
18.7
296.1
109.8
24.7
2188.5
EMISSION
RATE (mg/m1n)
32.0
1.4
1.0
12.9
7.2
8.1
1.8
2.1
37.4
0.4
4.5
3.4
3.2
0.9
15.4
14.6
2.7
4.6
1.0
0.8
15.0
0.5
2.9
1.9
29.6
11.0
2.5
218.8
BAP 7
MASS (mg)
138.8
3.1
1.6
15.4
2.0
55.2
4.3
31.9
30.0
5.9
10.2
65.9
10.6
135.7
510.6
MISSION
ATE (mg/m1n)
9.2
0.2
0.1
1.0
0.1
3.7
0.3
2.1
2.0
0.4
0.7
4.4
0.7
9.0
33.9
BAP 7A
MASS (mg)
3.1
1.4
0.7
2.4
7.6
EMISSION
RATE (mg/mli)
0.045
0.020
0.010
0.035
0.110
TABLE 2.2. POM ANALYSIS
-------
The BSD fraction was passed through a dry impinger into a filter at
125 F. The filter and first impinger contents were extracted with
benzene; the extract was dried and the BSD determined gravimetrically.
The filtered gas flow proceeded through three more impingers containing
water and finally through silica gel before being measured at the meter
box. The impingers were extracted with benzene which was dried and
weighed.
The POM/BaP fraction was collected on a filter at 125°F and a solid
adsorbent resin (XAD-2) which were both extracted with methylene chloride
(MeCl). The gas then passed through a series of impingers containing
water and silica gel before being measured on the gas meter. The re-
covered water was extracted with MeCl. POM analysis was conducted on a
Gas Chromatograph/Mass Spectrometer (GC/MS) - by TRW's West Coast Ana-
lytical Department. The BaP analysis was conducted at the EPA's Research
Triangle Park location by thin layer chromatography/fluorescence. Only
runs 4, 7 and 7A were analyzed for POM/BaP. For the remaining runs the
samples were extracted and the extract retained for possible future
analysis.
An integrated bag was taken after the filter from the BSD stream.
It was analyzed on-site for benzene with a gas chromatograph. The
benzene results are missing from this report since it was determined
that the high concentrations of the volatile organic matter present in
the sample made the determination of benzene impossible. The gas chro-
ma tography column utilized in the field on the integrated bag sample did
not perform an adequate separation of benzene from the interfering
compounds.
-------
3.0 DISCUSSION OF RESULTS
Upon arrival at the plant the van was set up next to the coke
ovens. Generators were rented in order to provide power for the SASS
trains. The vent pipe was placed on an oven lid. The orifice plate
(Section 6) in the vent pipe not only caused a flow restriction but also
provided a flat surface on which tar formed a viscous layer resulting in
complete plugging of the vent in a short period of time. The orifice
plate was removed and a larger steadier flow resulted. All tests were
conducted without an orifice plate. Prior to the first run a problem
was encountered with plugging of the leak vent which seemed to be caused
by particles of coal fines. A new vent was placed on the lid and the
first test conducted. Although the leak vent did not totally plug
during the test, the flow appeared to be somewhat restricted. The
determination was made that this problem of flow restriction due to coal
fines would occur only during the first hour of testing, therefore
subsequent tests were begun one hour after oven charging. The partial
plugging of the vent is reflected in the results. The first test was of
longer duration (24 minutes) and collected less mass than the subsequent
tests at the large leak rate. The first test was considered an experiment
and three more complete tests were conducted. During the sampling the
probe nozzle was placed one half inch above the vent outlet to permit
visual monitoring of the leak. The observation of the leak assured that
plugging of the vent did not occur during the tests runs.
SASS trains were used to collect the sample. The POM/BaP train and
the BSO train used a common nozzle which was placed above the
simulated leak. After a photograph of the leak was taken, the
test was initiated. The pumping rate was adjusted so that each train
-------
sampled an equal amount and the entire leak was captured. The tests
were of a short duration due to the plugging of the filters on each
train. The runs were discontinued when the filters were overloaded.
The flow rates were adjusted by timing the volume and observing the
meter orifice pressure differential. This was not an accurate process;
therefore, the volumes collected were not exactly equal. The valve that
adjusted the leak was a coarse ball valve which only permitted approx-
imately the desired leak rate. The variability in the leak rate accounted
for the different sampling times.
The filter was placed on the wrong side of the supporting screen
during run 3 in the BSD train. This accounts for the large sampling
volume and mass collected. The sample volumes of run 4 disagree by a
factor of four. No apparent reason is known for this other than they
were both low volumes and this type of variance in flow rate could
easily have resulted from the flow approximating method.
Methylene chloride was to have been used as an impinger solvent,
however, during the first attempted leak check it was found to be
freezing the impingers at the tips and causing plugging. Water was
substituted and used in all the tests.
The ambient trains were operated approximately 40 feet away from
the oven lids. The ambient test runs were started before the BSO/POM
tests and were stopped after the sampling. Duration and volume sampled
were greater than the lid leak tests in an attempt to produce a measureable
sample.
Immediately following each test the samples were recovered. The
recovered samples were transported to TRW's west coast facilities where
the splitting, combining, extraction and analysis took place. MeCl was
used to rinse the sampling equipment as well as for the extraction of
the POM train filter and Resin. The complete results of the GC/MS
analysis are contained in Appendix B. Only samples 4, 7 and 7A were
analyzed for POM/BaP. The sample splits of these samples were sent to
EPA at Research Triangle Park, North Carolina. Where thin layer chromato-
graphy /spectroscopy for Benz-a-Pyrene analysis were performed. These
results are listed in Appendix B. For all other runs the samples were
extracted and the extract retained for possible future analysis.
-------
During testing, an integrated bag was collected from the BSD train
after the filter. This bag was analyzed on site for benzene utilizing
an OV-101 column on a dual flame ionization detector gas chromatograph.
The resulting chromatogram displayed numerous peaks in the retention
time frame of benzene. These peaks made it impossible to discern which
peak was benzene or if benzene was masked by the interfering peaks.
Although different column conditions were attempted, the high concentrations
of volatile organic matter present in the sample rendered separation
impossible. No benzene results are presented here.
Multiple extractions were performed on the BSD samples (filter) to
determine the efficiency of the first and subsequent extractions. The
average of the first extractions were 98.8% of the total with the remaining
1.2% present in the second extraction. No weight gains were recorded
for the third extraction. Complete results are listed in Appendix B.
Multiple extractions with Methylene chloride were conducted on filters
and resins for several POM samples in order to determine the efficiency
of extraction. The first extraction contained 99.9% of the measurable
ROMs and the second extract the remaining 0.1%. The third extractions
were below detection limits. Furthermore, cyclohexane extractions were
conducted on several samples after the MeCl extraction to determine the
efficiency of the MeCl extraction. The BaP cyclohexane extractions were
below the detection limits.
-------
4.0 PROCESS DESCRIPTION
Battery 1 is one of three batteries (1, 2, 3) of a coke oven oper-
ating facility located at the Clairton works. Table 1 lists design
parameters of Battery 1.
During the testing, Battery 1 was operated in a normal manner. The
average gross coking time was 18 hours. Table 2 lists the production
rates for each day of testing. Table 3 lists coal analyses (conducted
by U.S. Steel) for each day of testing.
-------
TABLE 1
COKE OVEN BATTERY DESCRIPTION'
l
Type
Number of ovens
Date of initial operation
Date of last pad up rebuild
Approximate oven width (average)
Approximate oven height (floor to roof)
Approximate oven length (between doors)
Free space above coal
Calculated capacity (cubic foot coal)
Charging holes
Charging hole diameter
Number of collecting mains
Type of oven doors
Type of charging
Wilputte
64
1918
1955
18.5 inches
11 feet 10 inches
37 feet 4 inches
13 inches
626
4 per oven
18 inches
2
self-sealing
gravity feed stage charging
i
Information from letter of September 6, 1974 from John G. Munson,
Assistant to Vice President of U.S. Steel, to Reid Iversen, EPA.
10
-------
TABLE 2
PRODUCT INFORMATION
Date Ovens Charged Per Day Gross Coking Time (hrs)
8/1/78 81 19.0
8/2/78 83 18.5
8/3/78 73 21.0
8/4/78 83 18.5
8/5/78 82 18.7
19.1
i
Information from letter of September 18, 1978 from R.J. Weiskircher
of U.S. Steel to C.E. Riley, U.S. EPA.
11
-------
ro
TABLE 3: COAL ANALYSIS
1
»
8/1/78
8/2
8/3
8/4
8/5
2°
6.5
6.3
6.2
5.9
6.8
%Sulphur
1.15
1.21
1.14
1.15
1.16
%Ash
7.45
7.75
7.65
7.80
8.25
%Volatiles
28.80
28.75
29.60
29.40
29.15
+ .500
inch
4.6
3.0
4.3
5.7
4.7
Screen
+ .250
inch
14.8
14.9
13.6
14.2
15.0
Size
+.125
inch
17.1
16.3
16.7
15.4
15.3
-.125
inch
63.5
65.8
65.4
64.7
65.0
Information from letter of September 18, 1978 from R.J. Weiskircher of U.S. Steel to C.E. Riley,
U.S. EPA.
-------
5.0 LOCATION OF SAMPLING POINTS
All sampling points were located on the topside of the number one
coke oven battery at the Clairton Works. The oven charging port closest
to the push side was chosen for sampling. One end of the oven battery
was chosen because it would least disturb the normal operation of the
charging trolley. All samples were taken from ovens A, through Ay,
depending upon the daily charging sequence. The ambient samples were
taken from topside at a location past the pinion wall. Figure 5.1 is a
schematic of the battery topside indicating the sampling locations.
13
-------
•SAMPLE POINT
o
o
o
o
ISHING ->^
SIDE
O
O
EXHAUST FUME
COLLECTOR
c20O Q O Q
OVERHEAD TROLLEY
•
A7® O O O
AeO O O O
AB® O O O
A4O O G O -
A3® G O O -
A2® G G G
AT® O G O
1234
ROOF HEIGHT-301 ABOVE GROUND LEVEL
PINION WALL
AMBIENT SAMPLE •
SAMPLE TRAIN
PREPARATION AREA
COKE
QUENCH
CAR
COKING SIDE
4 '
40'
i 1
FIGURE 5.1 : BATTERY 1 (TOPSIDE VIEW)
14
-------
6.0 TEST PROCEDURES
During the testing, the oven was charged with the overhead trolley
and the lids replaced and sealed by U.S. Steel personnel; a special lid
modified with a valve and vent pipe was replaced on one of the oven
ports (see Figure 6.1). After the lids had been sealed, a simulated
leak was attempted at the desired plume length. Condensation occurred
in the vent forming a viscous substance; consequently, the plume was
monitored continuously to assure clogging of the vent did not occur.
Once the plume was determined to be the correct size, the nozzle was
placed over the vent and the sampling trains were started. During
sampling, the vent was visually monitored continuously to assure that
the sampling rate was sufficient to assure total capture of the leak.
Sampling was continued until plugging in the vent occurred or until the
loading on the sample train filter caused an excessive decrease in the
sampling rate. The sampling rates of the BSO and POM sampling systems
were evenly maintained in order to evenly split the sample between the
two trains.
The sampling was conducted in accordance with the draft methods--
Determination of Benz-a-Pyrene Emissions from Stationary Sources and
Determination of Benzene Soluble Organics. Both of these methods were
supplied by EPA and both are draft procedures. Several exceptions to
the procedures of these Methods were made. SASS sampling trains were
substituted for the regular Method 5 trains and the two sampling trains
shared a common nozzle (see Figure 6.2). The ambient air sampling
system was a Method 5 train modified to accept the XAD-2 adsorbing
module; no nozzle was used for this train. Methylene chloride was not
used in the impingers because of a problem in freezing due to the heat
loss inherent in volatile organic compounds. This was determined during
the first run. Subsequent impinger fractions used water as a collection
15
-------
medium. Methylene chloride was used for cleanup and extraction of the
POM train in lieu of cyclohexane.
The basic sample train components are diagrammed in Figure 6.3.
All the components used were standard source assessment sampling system
(SASS) parts. A simulated leak was created by the pressure in the oven
flowing through the valve to the atmosphere. After visually measuring
the leak a nozzle common to both trains was positioned over the leak.
The nozzle was not connected but just positioned over the leak. The
train's pumps were adjusted to provide approximately half the leak to
each sample train. Some excess air was also sampled to insure all of
the vent gases were captured. The probe and heated oven were maintained
at 125°F for the BaP/POM train while only the probe was heated to 100°F
for the BSO train. The BaP/POM train filtered the sample to remove the
particulate matter. The non-particulate gases were then collected by
passing the gas sample through an XAD-2 absorbent resin maintained at
70-90°F. Cooled impingers and silica gel were used to collect the
remaining moisture prior to the pumps and the gas flow meters. The
BSO train collected the particulate in the first impinger and on the
filter. A dry impinger was placed prior to the filter in order to
minimize plugging. During the sampling, an integrated Tedlar bag was
taken after the impingers but before the silica gel on the BaP/POM
train. A sample identification log is contained in Appendix £. This
log lists each sample fraction and the solvent used in clean up and
recovery. They are generally shown in the following table. The GC/MS
analysis was performed according to the procedures outlined in Analysis
of Polynuclear Aromatic Hydrocarbons from Coke Oven Effluents by R. E.
Beimar, September, 1978 (EPA report 78-CKO-12). The Column utilized was
a Dexel 300 packed column for the separation of POMS!
16
-------
CfinfflSILJIOZZle
l/2"Stainless Steel Tube
(Insulated)
1" Nipple Threaded
Into Oven Port
t
1" Stainless Steel Ball Valve
Figure 6.1 Sampling Vent/Probe Configuration
-------
Probe
Filter
Modified
Batelle Trap
Adsorbent
Sampler
XAD-2
Impingers
00
Inclined Manometer
Hry Gas Moter
FIGURE 6.2
AMBIENT TRAIN
-------
To
Common
Nozzle
Probe
BSO TRAIN
Fi 1 ter
To Integrated Sampling Bag
Impinger
Inplngers
Pumps
and
Control
Module
To
Common <^ •
Nozzle ^
Probe
DAM
n
u
Filter
XAD-2
Adsorbent
Module
1 JR1
B
!«
f
H20
^r
&t
&
?;"«
i«
'«»
«'
/ — \
Bl
IWH
an
5r •
k S1
I
1:1
TT^
TT
_qe
n
•I
I
fa!
Pumps
and
Control
Module
Inpinners
Figure 6-3. SASS SAMPLING TRAINS
-------
Field Recovery of BSO Train
Component
Probe, Knock-out Impinger, Filter Housing
Filter
Impingers
1
Recovery Steps
Condensate collected
Acetone Rinse
Benzene Rinse
Collected
Condensate collected
Acetone Rinse
Benzene Rinse
Component
Vent Pipe
Common Nozzle
Field Recovery of BSO/BaP Trains
(Common Components)
Recovery Steps
Methylene Chloride Rinse
Benzene Rinse
Methylene Chloride Rinse
Benzene Rinse
Field Recovery of BaP Train
Component
Probe, Front of Filter Holder
Filter
XAD-2 Resin
Back of Filter Holder and Module
Impingers
Recovery Steps
Methylene Chloride Rinse
Collected
Collected
Condensate Collected
Methylene Chloride Rinse
Condensate Collected
Acetone Rinse
Methylene Chloride Rinse
Some Impingers contained activated charcoal which was not recovered
after each test. The same is true of the silica gel.
20
-------
Laboratory Sample Preparation
(BaP/BSO Common Fractions)
1. Nozzle - Methylene Chloride Rinse - Split
Benzene Rinse - Split
2. Vent - Methylene Chloride Rinse solit (A B C)
Benzene Rinse
Laboratory Sample Preparation
(BSD Fractions)
1. Nozzle Benzene Rinse , Probe, and
Knock-out Impinger Benzene Rinse combined and Split4 (D,E)
and Nozzle Methylene Chrloride Fraction
2. Probe Knock-out Impinger Acetone Rinse - Split (D,E)
3. Filter Benzene Extraction and Split (D,E)
4. Impinger Liquid - Split (D,E)
5. Impinger Acetone Rinse - Split (D,E)
6. Impinger Benzene Rinse - Split (D,E)
7. Knock-out Impinger Condensate - None in most cases
The rinses split in a ratio to correspond to the sampled gas volumes of
the BaP/BSO trains for each test.
2
Nozzle fraction (Methylene Chloride) after being brought to dryness
and re-extracted with Benzene.
3
Nozzle fraction derived from the BaP/BSO common fractions.
4
The samples were split in half. The "D" fractions were dried and
weighted gravimetrically by TRW. The "E" fractions were sent to EPA
and from EPA to U. S. Steel.
21
-------
Laboratory Sample Preparation (continued)
(BaP/POM Sample Fractions)
c
*1. Nozzle - Methylene Chloride Rinse g
Combined and Split (A,B,C)
Probe Front of Filter Holder Methylene Chloride Rinse
*2. Filter - First Methylene Chloride Extraction and Split (A,B,C)
- Second Methylene Chloride Extraction and Split (A,B,C)
- Third Cyclohexane Extraction and Split (A,B,C)
*3. XAD-2 Resin - First Methylene Chloride Extraction and Split (A,B,C)
- Second Methylene Chloride Extraction and Split (A,B,C)
- Third Cyclohexane Chloride Extraction and Split
(A.B.C)
*4. Module Condensate - Cyclohexane Extraction and Split (A,B,C)
*5. Module Back of Filter Holder - Methylene Chloride Rinse and Split
(A.B.C)
6. Impinger Liquid Cyclohexane Extraction and Split (A,B,C)
7. Impinger Rinse Acetone Rinse Combined an(J Spm (A>B)C)
Methylene Chloride Rinse
From BaP/BSO common fractions.
6"A" which was half of the volume was aliquoted and GC/MS'd by TRW
(see note *.)
"B" which was a quarter of the volume was sent to EPA.
"C" which was a quarter of the volume was retained.
*Equal aliquots from the "A" fractions of the split for each section of
the sampling train were combined, except the samples derived from the
impingers, to provide a solution for GC/MS analysis. GC/MS was used
for runs BaP4, BaP7, and BaP 7A, the results are reported. Samples
from the other runs were kept but not analyzed.
22
------- |