United States Office of Air Quality EPA-450/3-82-028
Environmental Protection Planning and Standards October 1984
Agency Research Triangle Park NC 27711
Air
&EPA Method 16A for
Determination of
Total Reduced
Sulfur Emissions -
Summary of
Comments and
Responses
-------
EPA-450/3-82-028
Method 16A for Determination of
Total Reduced Sulfur Emissions -
Summary of Comments and Responses
Emission Standards and Engineering Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
October 1984
-------
This report has been reviewed by the Emission Standards and Engineering Division of the Off ice of Air Quality Planning
and Standards, EPA, and approved for publication. Mention of trade names or commercial products is not intended to
constitute endorsement or recommendation for use. Copies of this report are available through the Library Services
Off ice (MD-35), U.S. Environmental Protection Agency, Research Triangle Park, N.C. 27711, or from National Technical
Information Services, 5285 Port Royal Road, Springfield, Virginia 22161.
-------
TABLE OF CONTENTS
Page
Chapter 1. INTRODUCTION 1
Chapter 2. SUMMARY OF CHANGES SINCE PROPOSAL 2
Chapter 3. SUMMARY OF PUBLIC COMMENTS AND RESPONSES 4
Table 1. LIST OF COMMENTERS 18
-------
Chapter 1
INTRODUCTION
On June 18, 1981, the U.S. Environmental Protection Agency published
Method 16A, "Determination of Total Reduced Sulfur Emissions from Stationary
Sources," in the Federal Register (46 FR 31904). This method can be used
by affected facilities to determine compliance with the regulations. This
method was proposed under the authority of Sections 111, 114, and 301(a) of
the Clean Air Act, as amended.
Public comments were solicited at the time of proposal. An invitation
to request a public hearing was issued to provide interested persons the
opportunity for oral presentation of data, views, or arguments concerning
the proposed test method, but no person desired to make an oral presentation.
The public comment period was from June 18, 1981, to August 17, 1981.
Letters concerning issues relative to the proposed test method were
received from 10 commenters. A detailed discussion of these comments and
responses are summarized in this document. The summary of comments and
responses serve as the basis for the revisions which have been made to the
test method between proposal and promulgation.
-------
Chapter 2
SUMMARY OF CHANGES SINCE PROPOSAL
1. Section 1. A new section discussing interferences, precision and
bias has been added.
2. Section 1.1. New minimum detection limits for the method when
sampling for 1 and 3 hours have been added.
3. Section 2. The probe specifications have been changed to minimize
the sampling of particulates. A heated filter has been added to the
sampling train.
4. Section 2.1. A 250-ml polyethylene bottle is needed for the
peroxide solution recovery.
5. Section 2.1.1. The two midget impingers used for the S02 scrubber
have been replaced by three larger Teflon impingers.
6. Section 2.1.3. The combustion tube furnace should be capable of
maintaining a temperature of 800 +_ 100 °C instead of 815 ;* 15 °C.
7. Section 2.1.4. The rate meters should be capable of measuring
flow rate to within 5 percent of the selected flow rate.
8. Section 2.3. The 10-ml buret should have 0.05 ml graduations.
9. Section 3.1.3. A pH adjustment of the citrate buffer solution is
requi red.
10. Section 3.1.4. The standard deviation for Method 11 analyses of
the calibration gas is specified. Gravimetrically calibrated permeation
devices are an allowable option for generating the calibration gas. The
gas concentration should be generated in the range of the stack concentration
or within 20 percent of the applicable standard level,,
-------
11. Section 3.1.5. For the combustion gas, an example system for
purifying ambient air is listed.
12. Section 4.1. An S02 scrubber conditioning period is specified.
13. Section 4.1.1. New directions for the S02 scrubber are given.
The oxidation furnace temperature is maintained at 800 + 100°C.
14. Section 4.1.3. The sample is collected over a 1- or 3-hour
period. The three-hour sample should not be intermittent and should have
a total volume of 360 liters. It is recommended that the probe be washed,
the filter replaced, and the S02 scrubber changed after each test run.
15. Section 4.3. The sample analysis has been modified to allow for
duplicate analysis.
16. Section 5.2. The performance check calibration system has been
changed by replacing the dry gas meters used to measure the calibration
and combustion gases with calibrated rotameters. The results of the recovery
check will be reported but will not be used to correct the emission data.
A recovery of 100 + 20 percent must be obtained for the data to be valid.
The performance check prior to testing is recommended but is not mandatory,
and no precision for this check is specified.
17. Section 6. This section has been expanded to include sample
volume, TRS concentration, and recovery check calculations.
18. Section 7. Two additional references have been added to the
bibliography.
-------
Chapter 3
SUMMARY OF PUBLIC COMMENTS AND RESPONSES
Method 16A
1. D-2
Comment: The definition of total reduced sulfur compounds (TRS) in 40
CFR 60.281 does not make reference to Method 16A. This should be done or
mention of Method 16 should be dropped.
Response: Mention of Method 16 is essential in defining the individual
components that make up TRS because the compounds are identified using gas
chromatography. Since Method 16A is an alternative method and makes no
distinction between sulfur compounds, there is no advantage in including it
in the definition.
2. D-2
Comment: When calculating compliance with § 60.283(a)(4), a total
mass emission rate is needed and relies on the known concentration of each
reduced sulfur compound. Since Method 16A oxidizes all sulfur compounds to
sulfur dioxide before analysis, the concentrations needed in § 60.285(d)(3)
are not available.
Response: Section 60.285(d)(3) is being revised to allow the calculation
of TRS concentrations on an equivalent H2S basis. This revision was proposed
in the Federal Register on January 19, 1984 (49 FR 2448) and will eliminate
the need to quantify individual TRS compounds.
3. 0-5, D-6, D-8, D-9
Comment: The following changes or additions are recommended for the
scrubber system.
-------
a. The specified 40 ml of citrate scrubber solution may not be
sufficient to collect high levels of S02 for extended periods of time.
Two large impingers containing 200 ml of scrubber each will effectively
remove the S02«
b. The midget impingers will not have the capacity to hold the volumes
of condensed moisture present in the streams of some sources.
c. The pH of the scrubber solution should be adjusted to 5.5 - 5.6
with citric acid or potassium citrate.
d. The scrubber should be conditioned with 5 to 10 ppm H£S in nitrogen
or source gas for 15 minutes prior to initiation of sampling.
Response: The scrubber system has been changed to use three large
Teflon impingers in series. The first two impingers will contain 100 ml of
scrubber solution each, while the third impinger will be initially dry. An
addition to Section 3.1.2 has been made requiring the pH adjustment of the
scrubber solution. A requirement to condition the scrubber before sampling
has also been added. Our studies show that a 10-minute purge with stack
gas provides sufficient conditioning.
4. D-5, D-8
Comment: In the background paper referenced in Section 7.1, it is
indicated that 83 ppm H2S is completely oxidized at 2 liters per minute but
not at 3 liters per minute with the furnace temperature at 850°C. An
earlier study suggests a temperature of 1000°C for the oxidation of TRS
compounds. The testing reported in the background paper also primarily
used H2$. It would appear that additional effort should be spent in
optimizing the oxidation temperature and flow rates for complete conversion
of mixtures of TRS compounds.
-------
Response: Studies by the EPA and other sources have shown that
temperatures significantly lower than 850°C will completely oxidize TRS
compounds at levels encountered e,t most sources. A temperature range of
800 +_ 100 °C will be specified for the oxidation furnace. The data listed
in the background paper included one sample set that showed complete
oxidation of a mixture of TRS compounds. Other laboratory studies have
verified the complete oxidation of all TRS compounds at the 2 liter per
minute flow rate.
5. D-B
Comment: Specifying a combustion furnace temperature of 815 + 15°C in
Section 2.1.3 is not critical. Our studies indicate that 460°C is sufficient
to oxidize all TRS compounds.
Response: We agree that the ^15°C range is not critical, but to
maintain uniformity in application, the furnace temperature range will be
specified as 800 + 100 °C.
6. D-5
Comment: At some facilities, the tube furnace may not be permitted
due to safety restrictions.
Response: At facilities where this alternative method is restricted,
Method 16 should be used.
7. D-5
Comment: In the background paper referenced in Section 7.1, the results
of a field study are presented wherein Method l&A is compared to Method 16
and the Barton Coulometric Method at a lime kiln, recovery furnace, and
smelt dissolving tank. From the data, it is very clear that Methods 16 and
16A do not give the same results on the kiln emissions. The report states
-------
that the difference was caused by C02 quenching of the photometric detector
used in Method 16, whereas Method 16A is not affected by C02. The emission
standard for pulp mills, however, is stated to be 5 ppm of TRS emissions
based on the use of reference Method 16. If Method 16A does not give
comparable results to the reference method, it cannot be used to regulate
emissions at this source without either changing the standard by specifying
different values for the method or by revising the reference method. The
values obtained on the H2S calibration standards during the field work
indicate that both methods give low values. In comparison, in the laboratory
portion of the study, high values were obtained by Method 16A. Further
study is needed in this area.
Response: Method 16 lists C02 as having a substantial desensitizing
effect on the flame photometric detector and states that acceptable systems
must demonstrate that they have eliminated this interference. The Method
16 system used at the lime kiln had not corrected this C02 problem; hence,
the values it gave were substantially lower than those obtained using Method
16A. The values obtained for the field calibration gases show that both
methods had recoveries around 90 percent. Recoveries above 80 percent are
acceptable. The recoveries for the Method 16A laboratory sample sets
averaged 102 percent, with all sets within 9 percent of the known sample
value.
Method 16A is proposed as an alternative method. This does not require
equivalency to Method 16, but that the method produce results adequate to
determine compliance.
-------
8. D-5
Comment: A considerably longer sampling time should be used in order
to provide a higher concentration sample that would reduce errors due to
the small amount of titrant needed, the faintness of the indicator endpoint,
and other errors common to sampling.
Response: Laboratory and field studies have shown that Method 16A can
accurately measure TRS in the range of the emission standards when 1- or 3-
hour samples are taken. If a 1-hour sample is collected from a source
emitting 5 ppm TRS, the amount of titrant needed for analysis would be
approximately 2 ml. This titration volume can be accurately measured to
within 0.025 ml using the buret specified in the method. The precision in
distinguishing the indicator endpoint for an experienced analyst is normally
within 1 drop (0.05 ml) of titrant. Therefore, the error associated with
this small titrant volume (less than 2 percent) and indicator endpoint
detection (2.5 percent) should be less than 5 percent.
9. D-8
Comment: The amount of thorin indicator added for titration is
critical to the end point. Consistency should be used for blanks and
samples.
Response: Four drops will be specified for samples and blanks.
10. D-5, D-6, D-7, D-8, D-9
Comment: More flexibility should be allowed in the calibration system.
The use of permeation devices should be permitted in place of the cylinder
gas for the following reasons:
-------
a. Method 16 uses permeation tubes.
b. Gas cylinders are not as safe as permeation systems.
c. It is difficult to accurately analyze H2S cylinder gas by
Method 11 when the concentration is less than 50 ppm. This would affect
the ability to meet the +15 percent accuracy criterion specified in the
method.
d. The dilution required to bring 30 to 50 ppm H2S into the applicable
range would introduce significant errors.
e. If desired, H2S from cylinders can be calibrated against permeation
tubes traceable to the National Bureau of Standards (NBS). This would
eliminate the need for dilution.
Response: Permeation devices will be allowed as alternative calibration
systems as long as they meet the performance check of Section 4.3. A recent
EPA study evaluating Method 11 placed the minimum detectable limit at 2.4
ppm and showed that accurate analyses could be performed in the 2-20 ppm range.
The Method 11 analysis is required due to the instability of cylinder gas
H2S over periods of time. The + 15 percent accuracy criterion has been
changed to +_ 20 percent.
11. 0-6, 0-8, 0-9
Comment: The field calibration system includes two gas cylinders and
two dry gas meters. It is fairly complex and requires adjusting several
flows and then taking simultaneous meter readings when starting and stopping.
The dilution factor cannot be accurately determined because it is calculated
by taking the difference between two large volumes. This setup should be
listed as an example system only. A setup that has been shown to work,
-------
uses a rotameter to monitor the flow from the H^S cylinder and a dry gas
meter to record the total flow. The flow rate through the rotameter will
remain constant during sample recovery checks.
Response: The proposed calibration system has been revised to make it
simpler and less prone to inaccuracies. The new example system utilizes
calibrated rotameters to set the HgS and combustion gas flow rates. This
will eliminate the need for dry gas meters to determine the individual
volumes of these gases. Alternative systems may be used provided the
acceptability criteria of Section 4.3 is met.
12. 0-6, D-9
Comment: The post-test procedure calls for introducing the recovery
gas through the probe. This is different from the more desirable Method 16
requirement of introducing the gas after the probe and filter. Why was
this change made? If the probe is heated, there should be no TRS loss.
This post-test loss check should oe performed after each third sample or
earlier, if desired.
Response: It has been found that components emitted by some facilities
may collect in the probe and attribute to significant TRS losses through
adsorption. This loss can be effectively determined by including the
uncleaned probe in the post-test procedure. This loss check through the
probe is not required for Method L6 because higher sampling rates are
employed and a very large amount of calibration gas would be needed.
Field tests have shown that one field recovery sample after each test
run is sufficient when using the modified probe design, the heated in-line
10
-------
filter, and when changing the S02 scrubber after each run. Recovery
checks performed on a more frequent basis are allowed.
13. D-6, D-8
Comment: Section 4.1.3 requires that a run consists of three 1-hour
samples or one 3-hour sample to be consistent with Method 16. One 3-hour
sample would not give any indication of precision. Also, Method 16 needs a
3-hour run to get a reasonable average of 16 individual injections. A
2-hour run has been found ideal since three can be performed in addition to
calibration in a single day, whereas the 3-hour runs would require an
additional day. A continuous sample for 2 hours gives a better average
than sixteen 10-second grab samples and does not invalidate the intent of
Method 16.
Response: The test run time was chosen primarily to be consistent
with reference Method 16 since this is an alternative method. A shorter
time would offset the similarity of the two methods.
Neither a 2- nor 3-hour sample would give any indication of precision.
The option to collect one continuous 3-hour sample as opposed to three
individual 1-hour samples may be preferred in certain situations.
14. D-6
Comment: More flexibility to use different apparatus should be allowed
provided the quality assurance tests are met.
Response: This flexibility is allowed and is stated in Section 2.1.
15. D-8
Comment: The preamble to the proposed method states that it has few
interferences. We have found that calcium and sulfate particulate from
11
-------
some lime kilns and recovery furnaces are severe interferences. These
compounds should be noted in the applicability section. Provisions should
be made to place a filter in the sample train to remove participate matter.
Response: The sampling train has been modified to include a filter.
Proper use of the filter to eliminate particulate interferences is now
mentioned in the applicability section.
16. D-8
Comment: The sample analysis procedure should be changed to follow
Method b exactly. The proposed procedure generates a large titration volume
with only enough sample for one analysis.
Response: The sample analysis procedure has been changed to reduce
the large titration volume and allow for duplicate analysis. The small
sample aliquot specified by Method 6 would create a dilute sample wherein
analytical accuracy would be lowered due to the decreased amount of titrant
required.
17. D-8
Comment: The preamble to the method states several advantages of
Method 16A over Method 16. The following comments are made.
a. It is stated that no gas dilution system is needed. A gas dilution
system is required for the system performance check.
b. We disagree that samples need not be analyzed in the field. Due
to the need to check system performance prior to, during, and after testing,
all samples must be analyzed on site, and all results must be calculated
at that time.
12
-------
c. The proposed method will not necessarily reduce the cost of
emission testing to demonstrate compliance with the standards. Our experience
with Method 16 and 16A indicates that Method 16A requires 80- to 100-1abor
hours to test a source whereas Method 16 would require 40 labor hours to
test the same source.
Response: The dilution system mentioned in the preamble refers to the
collection of the sample gas. A gas dilution system is used in the system
performance check but is not needed while collecting samples. The method
makes no requirement to analyze any samples in the field. It is wise to
analyze the system performance samples in the field to determine the validity
of the samples, but this is not mandatory. We agree that the amount of time
required to use a test method depends on the proficiency of the tester.
If it is determined by individuals that Method 16A is more costly and time
consuming than Method 16, then the individual has the option of not using
the alternative method.
18. D-l
Comment: The new method is a welcome change. We agree with the five
advantages noted in the preamble and as an additional advantage, the
availability of this technique in lieu of such traditional methods as the
continuous titrimetric techique which has been plagued with frequent
equipment malfunctions in field applications.
Response: No response needed.
19. D-8
Comment: We calculate the minimum detectable limit for the method to
be 1 ppm instead of 0.04 ppm.
13
-------
Response: Minimum detectable limits of 0.3 ppm for 1-hour samples
and 0.1 ppm for 3-hour samples based on the blank volume plus two standard
deviations have been calculated and will be listed.
20. D-8
Comment: The +2 percent flowmeters specified in Section 2.1.4 are
excessive. A flow rate of 2 liters +1Q percent is sufficient.
Response: The 2 percent specification refers to f'lowmeter accuracy,
not flow rate precision. This has been clarified by rephrasing the sentence.
21. D-a
Comment: The combustion gas in Section 3.1.5 should be air only since
dry gas meters are not calibrated for oxygen.
Response: Calibration of the dry gas meter is independent of gas
type. One liter of oxygen would occupy the same volume as 1 liter of air.
The ary gas meters are no longer used in the system performance check.
22. D-8
Comment: The system performance check should be performed before any
cleaning of the system is done.
Response: We agree, and this has been stated in Section 4.1.4.
23. D-8, D-9
Comment: Based on past studies with Method 16A, the 5 percent precision
for the system performance check is too stringent. This is also unnecessarily
restrictive for those sources which generally operate well below the
applicable standard. Precision within 10 percent and the option to run the
system performance check at the level of the applicable standard +20 percent
is more reasonable.
14
-------
Response: The precision requirement for the system performance check
has been dropped. An allowance to run this check at the level of the
applicable standard +_ 20 percent has been made.
24. D-8
Comment: The audit samples in Section 5.2 should be used to validate
the S02 titration but should not be needed as a substitute for a recovery
check.
Response: The original intent of this audit requirement was to
substitute an unknown HzS gas audit for one of the recovery checks during
a test. This requirement has been dropped in favor of an S02 audit of the
analysis procedure.
25. D-9
Comment: Method 16 measures TRS concentration on a dry basis. If
oxygen correction is needed (as required for most kraft sources), Method 3
should be specified to determine the oxygen content of the sample on a
dry basis.
Response: Use of Method 3 is already specified in paragraph 60.285 of
the regulations.
2fa. D-9
Comment: Since the present TRS standards require 0^ correction of the
measured TRS concentration for most sources, the calculation procedure for
02 correction should be added to the method.
Response: The equation for Og correction is listed in paragraph 6U.284
of the regulations.
15
-------
27. D-10
Comment: The proposed method should be revised to allow the use of 6
percent hydrogen peroxide and large impingers containing 200 ml of absorbing
solution.
Response: The use of large impingers in this method would be impractical
A large titration volume would result form taking an aliquot of sufficient
size to analyze accurately. A 20U-ml sample aliquot, plus 800 ml of
isopropanol, means titrating a 1-liter sample.
28. D-10
Comment: Section 4.2 should oe revised as follows;: "Pour contents of
the impingers of the Method 6 part of the train into separate leak-free
polyethylene bottles."
Response: The impinger contents of the Method 6 part of the train
should be poured into a container separate from the contents of the spent
scrubber impingers. The individual Method 6 impinger contents need not be
placed in separate bottles since the combined contents are used in the
analysis.
29. D-8
Comment: In Section 6.2, 12.020 y/meq should be 12.020 yl/meq.
Response: This correction has been made. This factor has been
converted to 12025 yl/meq to allow Vm(stCj) to be expressed in dry standard
liters instead of dry standard cubic meters.
30. D-8
Comment: The preamble to the method states that Method 16A reduces
the chances of measurement error Decause it involves fewer and less
complicated components. We think the chance of measurement error using
16
-------
Method 16A is greatly increased due to potential interferences (calcium
particulate) from the source, the components required for checking system
performance, and the method specified for calibration of the calibration
gas cylinder. The method should be modified and further field evaluations
should be performed to demonstrate that the method gives comparable results
to Method 16.
Response: The potential interferences from the source have been
eliminated by incorporating a filter into the sampling train. The components
required to check the system performance have been simplified by allowing
the use of permeation devices and rotameters to determine dilution ratios.
The permeation devices offer an alternative to calibration gas dilution and
verification of concentration by Method 11. In a recent evaluation of
Method 11 by the Agency, the accuracy of Method 11 was verified down to
a concentration of 2 ppm. Method 16A has been shown by the EPA and other
companies to give results comparable to Method 16.
31. D-l, 0-3, D-4, D-7
Comment: We approve of and support the proposed method for the
following reasons:
a. Method 16A is superior to Method 16 and provides more accurate
data at a lesser expense.
b. The unnecessary complexity of emission testing is reduced.
c. Plants will have the option of using their own staff and equipment
at a considerable savings over using expensive outside consultants having
the specialized equipment and instrumentation required by existing Method 16.
Response: No response needed.
17
-------
Table 1. LIST OF COMMENTERS
Docket Number A-80-38
Document number Commenter/affillation
IV-D-1 Peter E. DeRossi, Corporate Environmental
Engineer
Green Bay Packaging, Inc.
Post Office Box"1107
Green Bay, Wisconsin 54305
IV-D-2 Michael E. Fogle, Unit Coordinator
Georgia Department of Natural Resources
270 Washington Street, S.W.
Atlanta, Georgia 30334
IV-D-3 Jay Lipscomb
Mostardi-Platt Associates, Inc.
1077 Entry Drive
Bensenville, Illinois 60106
IV-D-4 T.C. Owen, Corporate Director
Union Camp Corporation
Post Office Box 1391
Savannah, Georgia 31402
IV-D-5 C.D. Malloch, Regulatory Management Director
Monsanto Company
800 N. Lindberg Blvd.
St. Louis, Missouri 63166
IV-D-6 Alan S. Rosenfeld, Sr. Research Engineer
Crown Zellerbach Environmental Services
904 N.W. Drake Street
Camas, Washington 98607
IV-D-7 Karl C. Ayers, Energy and Environmental
Manager
Mead
Eighth and Hickory Streets
Chillicothe, Ohio 45601
IV-D-8 Bruce B. Ferguson, Vice President,
Analytical Services
Harmon Engineering and Testing
Auburn Industrial Park
Auburn, A1abama 36830
18
-------
Table 1. LIST OF COMMENTERS
(Continued)
Docket Number A-80-38
Document number Commenter/affiliation
IV~D~9 Russell 0. Blosser, Technical Director
National Council of the Paper Industry
for Air and Stream Improvement, Inc.
260 Madison Avenue
New York, New York 10016
IV~D~10 Bill Stewart, Executive Director
Texas Air Control Board
6330 Hwy. 290 East
Austin, Texas 78723
19
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
EPA-450/3-82-028
3. RECIPIENT'S ACCESSION NO.
Method 16A for the Determination of Total Reduced
Sulfur Emissions
Summary of Comments and Responses
5. REPORT DATE
October 1984
6. PERFORMING ORGANIZATION CODE
Emission Standards and Engineering Division
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Emission Measurement Branch (MD-19)
Emission Standards and Engineering Division
U. S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
DAA for Air Quality Planning and Standards (MD-10)
Office of Air and Radiation
U. S. Environmental Protection Agency
Research Triangle Park. N. C. 27711
13. TYPE OF REPORT AND PERIOD COVERED
4. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This document addresses the public comments submitted after proposal
of Method 16A in the Federal Register. Changes made to the method as a
result of these comments are included. This document serves as the basis
for the revisions which have been made to the test method between
proposal and promulgation.
17.
a.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
18. DISTRIBUTION STATEMENT
Release Unlimited
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDIT.ON ,s OBSOLETE
b.lDENTIFIERS/OPEN ENDED TERMS
19. SECURITY CLASS (This Report)"
Unclassified
20. SE
CURITY CLASS.(Thispage/
Unclassified
c. COS AT I Field/Group
13B
21. NO. OF PAGES
19
22. PRICE
------- |