RESEARCH TRIANGLE 1NSTITUT
RTI/1808/05-1F
Final Report
Analysis of Commercial Cylinder Gases
of Nitric Oxide, Sulfur Dioxide, and Carbon Monoxide
at Source Concentrations
Results of Audit 3
Prepared for
Quality Assurance Division
Environmental Monitoring Systems Laboratory
Environmental Research Center
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
EPA Contract No. 68-02-3222
September 1979
RESEARCH TRIANGLE PARK, NORTH CAROLINA'27709
-------
RTI/1808/05-1F
Final Report
Analysis of Commercial Cylinder Gases
of Nitric Oxide, Sulfur Dioxide, and Carbon Monoxide
at Source Concentrations
Results of Audit 3
by
C. E. Decker
R. E. Encke
Prepared for
Quality Assurance Division
Environmental Monitoring Systems Laboratory
Environmental Research Center
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
EPA Contract No. 6&02-3222
September 1979
-------
TABLE OF CONTENTS
Section Page
1.0 SUMMARY 1
2.0 INTRODUCTION 4
3.0 CYLINDER GAS ANALYSIS AND TRACEABILITY PROCEDURE 8
3.1 Analysis of NO Cylinders 8
3.2 Analysis of S02 Cylinders 9
3.3 Analysis of CO Cylinders 10
3.4 EPA Quality Assurance Program 11
4.0 RESULTS OF CYLINDER GAS ANALYSES 13
5.0 SUMMARY OF RESULTS OF AUDITS 1, 2, and 3 19
-------
SECTION 1
SUMMARY
Manufacturers of commercial cylinder gas, if requested, will
supply gas standards with a certified analysis and a statement of
accuracy. Generally, the level of accuracy is specified as _+! to 3
percent of the component value. In order to ascertain the accuracy of
commercially available cylinder gas, EPA has initiated a national
performance audit program of commercial gas manufacturers. Audit 1 was
performed from May to July 1978 and Audit 2 was performed in January
1979. Audits 1 and 2 included cylinders of sulfur dioxide and nitric
oxide at source concentrations. Results of these audits were reported
in the publication entitled "Analysis of Commercial Cylinder Gases of
Nitric Oxide and Sulfur Dioxide at Source Concentrations - Results of
Audits 1 and 2."* A summary of these results is also included in
Tables 10 and 11 of this report.
The purpose of Audit 3 was two-fold. The first objective was to
analyze the concentration of cylinder gases purchased from a repre-
sentative sample of manufacturers and determine the accuracy of the
manufacturers' certified concentrations. The second objective was to
reanalyze the cylinder gases analyzed in Audit 2 to determine stability
of these gases since the initial analysis in January 1979.
Audit 3 of commercial cylinder gas included cylinders of sulfur
dioxide (302) at 90 and 50° PPm> nitric oxide (NO) at 50 and 300 ppm,
and carbon monoxide at 50 and 500 ppm obtained from twelve different
manufacturers. Audit 3 was performed in August and September 1979.
The procedure used to analyze Audit 3 cylinder gases was EPA's "Trace-
*Report available from the Quality Assurance Division, Environmental
Monitoring Systems Laboratory, Environmental Protection Agency,
Research Triangle Park, North Carolina 27711.
-------
ability Protocol for Establishing True Concentrations of Gases Used for
Calibration and Audits of Continuous Source Emission Monitors (Protocol
No. 1, June 15, 1978). A summary of Audit 3 results is presented in
Table 1.
-------
TABLE 1. SUMMARY OF RESULTS FOR AUDIT 3 OF COMMERCIAL CYLINDERS FOR
SULFUR DIOXIDE, NITRIC OXIDE, AND CARBON MONOXIDE
Sulfur Dioxide
Manufacturer Nominal Percent
Cone, (ppm) Accuracy
Airco Industrial
Gases
Air Products 4
Chemical, Inc.
Big Three Indus-
tries
Ideal Gas Products
Liquid Carbonic
Corp.
Matheson Gas
Products
MG Scientific Gas
Products
North East
Cryogen I cs
Scientific Gas
Products, Inc.
Scott Environ-
mental Tech., Inc.
Scott-Marrln, Inc.
Llnde Div. of
Union Carbide
500
500
500
500
90
500
500
500
90
500
90
500
90
500
90
500
500
90
- 0.8
1.0
- 5.5
- 8.8
- 7.0
- 1.0
0.4
- 0.8
- 3.3
- 3.8
-21.6
-12.2
- 3.0
1.7
3.2
0.2
8.9
3.8
Nitric Oxide
Nominal Percent
Cone, (ppm) Accuracy
300
300
300
300
50
300
300
300
50
300
50
300
50
300
50
300
300
500**
3.5
5.6
- 6.3
3.1
4.7
- 3.4
-12.4
5.7
22.2
4.0
>100*
0
0
-.1.0
- 2.8
- 0.6
8.3
- 1.6
Carbon Monoxide
Nominal Percent
Cone, (ppm) Accuracy
500
500
500
500
50
500
500
500
50
500
50
500
50
500
50
500
500
50
3.3
1.9
- 3.5
9.2
- 1.2
- 1.1
- 1.1
- 7.5
7.9
- 1.5
- 3.9
2.6
0.2
1.2
- 3.3
2.8
1.6
17.8
* Actual percent accuracy not calculated but was greater than lOOJf.
** Nominal 50 ppm ordered but 500 ppm shipped.
Percent Accuracy = 100(Manufacturer - RTD/RTI
-------
SECTION 2
INTRODUCTION
There are basically three types of gas standards available from
commercial manufacturers. These are primary standards, analyzed or
certified standards, and unanalyzed standards. The primary standards
are prepared gravimetrically on a high load, high sensitivity analyt-
ical balance with a tolerance of _+! percent of the component. The
analyzed or certified standards are prepared by a variety of
gravimetric and pressure-volume-temperature techniques. The mixture is
then analyzed by instrumental and/or wet chemical methods with a
component tolerance of +3 percent. The unanalyzed standards are
prepared in the same manner as the analyzed standards, but a chemical
analysis is not performed. The unanalyzed standards may have a
tolerance of _+15 percent or greater from the nominal concentration
ordered.
Commercial cylinder gas manufacturers will supply gas standards
and provide a certified analysis of gas concentration when requested.
The purpose of this project was to analyze cylinders obtained from a
representative sample of gas manufacturers and compare the measured
concentrations with the manufacturers' quoted analyses. The gases of
interest were NO, CO, and SOg, each in a balance of nitrogen.
A survey was conducted of manufacturers who routinely provide
cylinder gases that are used for calibration and auditing. A list of
these companies is given in Table 2. Each company was contacted as to
the types of gases produced and/or supplied. From the thirty-one
companies shown in Table 2, twelve were selected to supply gas
cylinders for analysis. These twelve manufacturers are listed in Table
3. The companies not included were eliminated because: (1) they do
not supply gases used in environmental studies (i.e., they supply hos-
pital, welding, or nuclear tagged gases), or (2) they do not blend and
analyze their own gases but merely label another manufacturer's
product.
-------
TABLE 2. SURVEY OF GAS CYLINDER MANUFACTURERS
Air Products and Chemicals, Incorporated
Airco Industrial Gases
Automotive Environmental Systems, Incorporated
Big Three Industries
Chemetron Industrial Gases
Edmond Scientific Company
Essex Chemical Corporation
Ideal Gas Products
J. T. Baker Chemical Company
J. W. Goodliffe Air Products Company
Liquid Carbonic Corporation
Matheson Gas Products
MG Scientific Gases
National Welders Supply Company, Incorporated
New England Nuclear
Northeast Cryogenics
Nuclear Sources and Services, Incorporated
Ohio Chemical and Manufacturing Company
Pipe Welding Supply Company, Incorporated
PPG Industries, Incorporated
Puritan Compressed Gas Corporation
Scientific Gas Products, Incorporated
Scientific Products, Division of American Hospital Supplies
Scott Environmental Technology, Incorporated
Scott-Marrin, Incorporated
Stauffer Chemical Company
Sulfrian Cryogenics, Incorporated
Supelco, Incorporated
Texas Gulf, Incorporated
Union Carbide Corporation, Linde Division
Virginia Chemical, Incorporated
-------
TABLE 3. CYLINDER GAS MANUFACTURERS INCLUDED IN
AUDIT PROGRAM
Manufacturer's Name
Manufacturer's Address
Airco Industrial Gases
Air Products and Chemicals, Inc.
Big Three Industries
Ideal Gas Products
Liquid Carbonic Corporation
Matheson Gas Products
MG Scientific Gases
North East Cryogenics
Scientific Gas Products, Inc.
Scott Environmental Technology,
Inc.
Scott-Marrin, Inc.
Union Carbide Corporation,
Linde Division
Research Triangle Park,
North Carolina
Tamaqua, Pennsylvania
Houston, Texas
Newark, New Jersey
Baltimore, Maryland
Morrow, Georgia
Somerville, New Jersey
Newtonville, Massachusetts
South Plainfield, New
Jersey
Plumsteadville,
Pennsylvania
Riverside, California
Raleigh, North Carolina
-------
The following gas mixtures were obtained through a third party to
maintain anonymity, since the results of the first two audits had been
published. The nominal concentrations ordered were as follows:
(1) 300 ppm NO; balance Nitrogen.
(2) 500 ppm $02; balance Nitrogen.
(3) 500 ppm CO; balance Nitrogen.
In addition, from selected manufacturers, a second cylinder of
each gas at a lower concentration was obtained. The nominal concentra-
tions of these cylinders were as listed below:
(1) 50 ppm NO; balance Nitrogen.
(2) 90 ppm S02J balance Nitrogen.
(3) 50 ppm CO; balance Nitrogen.
Specifications as to the type of analysis or the type of cylinder
(steel or aluminum) were not 'included on the purchase request. The
third party purchased the cylinders under the pretense of requiring the
cylinders for field applications, thereby insuring representative
samples. The only specifications given to the manufacturers were
cylinder size, component, nominal concentration levels, and a request
for a certified analysis.
-------
SECTION 3
CYLINDER GAS ANALYSIS AND TRACEABILITY PROCEDURE
Each cylinder was received and then cataloged by manufacturer,
pollutant species, and concentration. The NO, CO, and S02 analyses
were performed separately using source emission analyzers. Each
commercial cylinder was analyzed three times. Between each analysis
an NBS-SRM (National Bureau of Standards-Standard Reference Material)
or a GMPS (Gas Manufacturers Precision Standard) which had met the EPA
criteria of less than an average of 1 percent per month degradation was
introduced as a control check on the analyzer. All procedures met or
exceeded the guidelines set forth in EPA's "Protocol for Establishing
the Traceability of Calibration Gases Used with Continuous Source
Emission Monitors" (Protocol No. I, June 15, 1978). Immediately prior
to analysis of the commercial cylinders, RTI was required to partici-
pate in an EPA audit for NO, CO, and S02-
3.1 ANALYSIS OF NITRIC OXIDE CYLINDERS
A Thermo-Electron Corporation (TECO) chemiluminescent NO-N02-
NOX analyzer-Series 10 (S/N 10AR-6795-88) was used to analyze the NO
cylinder concentrations. The multipoint calibration procedure speci-
fied in EPA's "Protocol for Establishing the Traceability of Calibra-
tion Gases Used with Continuous Source Emission Monitors" was used to
calibrate the TECO analyzer. Two NBS-SRM NO in N2 cylinders and zero
gas were used to generate calibration concentrations in the range of 0
to 500 ppm. A multipoint calibration (zero level and five upscale con-
centrations) was performed prior to each set of cylinder gas analyses.
The multipoint calibration was accomplished by dilution of the highest
NBS-SRM (approximately 500 ppm) with zero gas using a calibration flow
system. At the conclusion of the multipoint calibration, a check of
the calibration curve was performed using the lower NBS-SRM without
dilution. The response of the instrument based on the original cali-
-------
bration curve (dilution of higher NBS-SRM) was compared to the response
to the true concentration of the lower NBS-SRM. If the difference
between the apparent concentration (based on dilution of higher
NBS-SRM) was less than 3 percent of the concentration of the lower NBS-
SRM, then the analysis of the commercial cylinders was performed. If
not, the multipoint calibration was repeated. Audit cylinders provided
by EPA were then analyzed. Agreement to within +5 percent between EPA
audit and measured concentration v/as required prior to analysis of the
commercial cylinders. Zero and span checks were performed during the
analysis period, according to EPA Protocol Mo. 1, to insure stable
instrument performance.
The analyses of Audit 3 cylinder gases were performed during the
period September 14-20, 1979. Cylinders of NO in M2 from Audit 2
were also reanalyzed during this period to determine the stability of
these gases since Audit 2, which was performed in January of 1979.
Because of the large number of cylinders (35) to be analyzed and
the relatively high flow rate (2 cfh) required by the analyzer, it was
decided that a GMPS (Gas Manufacturers Precision Standard) would be
used for span checks rather than an NBS-SRM. A cylinder of MO in N2
prepared by Scott-Marrin for Audit 2 was compared with the SRM and
found to have had degradation of less than the average 1 percent per
month specified in EPA Protocol No. 1. This cylinder was used as the
GMPS for Audit 3.
3.2 ANALYSIS OF SULFUR DIOXIDE CYLINDERS
A TECO pulsed fluorescent S02 analyzer-Series 40 (S/N SDM-6997-
90) was used to analyze the S02 cylinder concentrations. The analy-
zer v/as calibrated as described above in Section 3.1, except that only
one NBS-SRM (500 ppm) was used to provide calibration concentrations by
dilution. SRMs for S02 below 500 ppm are not yet available from NBS.
A lower concentration cylinder of S02 in nitrogen referenced to the
NBS-SRM was used without dilution to check the calibration curve.
After the multipoint calibration, audit cylinders provided by EPA were
analyzed. Agreement to within j^5 percent between the EPA audit and
-------
the measured concentrations was required prior to analysis of the com-
mercial cylinders. Zero and span checks were performed during the
analysis, according to EPA Protocol No. 1, to insure stable instrument
performance.
The analyses of Audit 3 cylinder gases were performed during the
period August 28-September 13, 1979. Cylinders of S02 in N2 from
Audit 2 were reanalyzed during this period to determine the stability
of these gases since Audit 2, which was performed in January 1979.
Because of the large number of cylinders (35) to be analyzed and
the relatively high flow rate (3 cfh) required by the analyzer, it was
decided that a GMPS would be used for span checks rather than an
NBS-SRM. A cylinder of S02 in N2 prepared by Scott Environmental
Technology for Audit 2 was compared with the SRM and found to have had
degradation of less than the average 1 percent per month specified in
EPA Protocol No. 1. This cylinder v/as used as the GMPS for Audit 3.
3.3 ANALYSIS OF CARBON MONOXIDE CYLINDERS
A Bendix nondispersive infrared analyzer Model 8501-5C(S/N 54351)
was used to analyze CO cylinder concentrations. The analyzer was cali-
brated as described above in Section 3.1, except that only one NBS-SRM
(470 ppm) was used to provide calibration concentrations by dilution.
A lower concentration cylinder of CO in nitrogen referenced to the
NBS-SRM was used without dilution to check the calibration curve.
After the multipoint calibration, an audit cylinder provided by EPA v/as
analyzed. Agreement to within +5 percent between the EPA audit and the
measured concentrations was required prior to analysis of the commer-
cial cylinders. Zero and span checks were performed during the
analysis, according to EPA Protocol No. 1, to insure stable instrument
performance.
The analyses of Audit 3 cylinder gases were performed during the
period August 16-21, 1979.
10
-------
3.4 EPA QUALITY ASSURANCE PROGRAM '
In order to assess the quality of data obtained during the analy-
sis of commercial cylinders, EPA conducted an external quality assur-
ance program on RTI consisting of a performance audit of the source
analyzers used to determine cylinder concentrations. EPA provided
audit cylinders of NO, CO, and S0£ in nitrogen to RTI for analysis
immediately prior to analyses of Audit 3 cylinder gases. RTI analyzed
the EPA audit cylinders, determined the concentration based on the
calibration curve of the source analyzers, and reported the data to
EPA. If the agreement between the audit cylinder and measured concen-
tration was within _+5 percent, RTI .proceeded to analyze each commer-
cial gas cylinder. If the measured level exceeded the +5 percent
limit, then a recalibration of the source analyzer and reanalysis of
additional audit cylinders were required. The data obtained from the
EPA performance audit of RTI are shown in Table 4. The agreement
between the RTI measured concentrations and the EPA audit concentra-
tions was well within +5 percent for all cylinders.
11
-------
TABLE 4. EPA PERFORMANCE AUDIT OF RTI SOURCE ANALYZERS
Cylinder RTI RTI EPA Difference Percent
no. Pollutant analysis measured cone. audit cone. RTI-EPA accuracy
date (ppm) (ppm) (ppm)
SD10020 CO Sal. 1 August 16, 1979 518
520
- 2
-0.4
FF6739
FF1167
S02 Bal.
S02Bal.
August 30, 1979 481
August 30, 1979 158
481
158
0
0
FF6395 NO Bal. N2 September 14, 1979 236
FF1577 NO Bal. N- September 18, 1979 48.5
243
48.5
-2.9
0
Percent Accuracy = 100(RTI - EPAJ/EPA
12
-------
SECTION 4
RESULTS OF CYLINDER GAS ANALYSES
The certified accuracy of analysis quoted by most manufacturers
typically ranges from +2 to 3 percent of the component concentration.
Based upon this specification, the manufacturer's analysis should be
within _+5 percent of the mean concentration measured by RTI. The
results for Audit 3 are given in Tables 5 through 7. Each table
includes manufacturer, cylinder type, manufacturer's analysis, mean of
RTI analysis, standard deviation, ppm difference, and percent accuracy.
Table 5 gives the results for sulfur dioxide. Table 6 gives the
results for carbon monoxide, and Table 7 gives the results for nitric
oxide. Table 8 gives the reanalysis results for sulfur dioxide
cylinder gases from Audit 2. Table 9 gives the reanalysis results for
nitric oxide cylinder gases from Audit 2. The results in Tables 8 and
9 show the stability of the Audit 2 cylinder gases from January 1979 to
September 1979.
13
-------
TABLE 5. RESULTS OF SULFUR DIOXIDE CYLINDER ANALYSES (AUDIT 3)
Manufacturer
Alrco Indus-
trial Gases
Air Products 4
Chemical, Inc.
Big Three
Industries
Ideal Gas
Products
Liquid Carbon-
ics Corp.
Matheson Gas
Products
MG Scientific
Gas Products
North East
Cryogen i cs
Scientific Gas
Products, Inc.
Scott Environ-
mental Tech.,
Inc.
Scott-Marr i n
Inc.
Linde Div. of
Union Carbide
Cy 1 i nder
Construction
A 1 urn i num
Steel
A 1 urn i num
Steel
Steel
A 1 urn i num
A 1 urn i num
Aluminum
Aluminum
Steel
Steel
Steel
Steel
A 1 urn i num
A 1 urn i num
A 1 urn i num
Steel
Steel
Manufacturer's
Analysis*
ppm
490
490
484
480
80
488
506
493
88
450
80
542
98
536
93.9
483
500
100
Mean of RTI
Analysis**
ppm
494
485
512
526
86
493
504
497
91
468
102
617
101
527
91
482
459
104
STD Dev. of
RTI Analyses
ppm
1.5
1.5
3.1
1.0
1.0
1.0
1.0
2.6
0
1.7
1.0
0.6
1.0
0.6
1.2
2.6
1.0
0.6
Difference
MFG - RTI
ppm
- 4
5
-28
-46
- 6
- 5
2
- 4
- 3
-18
-22
-75
- 3
9
2.9
1
41
- 4
Percent
Accuracy
- 0.8
1.0
- 5.5
- 8.8
- 7.0
- 1.0
0.4
- 0.8
- 3.3
- 3.8
-21.6
-12.2
- 3.0
1.7
3.2
0.2
8.9
- 3.8
* Manufacturer's Certified Analysis
** Mean of three analyses
Percent Accuracy = lOOCManufacturer - RTI)/RTI
14
-------
TABLE 6. RESULTS OF CARBON MONOXIDE CYLINDER ANALYSES (AUDIT 3)
Manufacturer
Alrco Indus-
trial Gases
Air Products &
Chemical, Inc.
Big Three
Industries
Ideal Gas
Products
Liquid Carbon-
Ics Corp.
Matheson Gas
Products
MG Scientific
Gas Products
North East
Cryogen I cs
Scientific Gas
Products, Inc.
Scott Environ-
mental Tech.,
Inc.
Scott-Marrln,
Inc.
Linde Dlv. of
Union Carbide
Cylinder
Construction
Aluminum
Steel
Aluminum
Steel
Steel
Aluminum
\
Aluminum
Steel
Steel
Steel
Steel
Steel
Steel
Aluminum
Aluminum
Steel
Steel
Manufacturer's
Analysis*
ppm
510
490
483
520
56.5
455
516
455
52.2
523
56.5
520
51
500
50.2
515
504
50.4
Mean of RTI
Analysis**
ppm
494
481
500
476
57.2
460
522
492
48.4
531
58.8
507
50.9
494
51.9
501
496
42.8
STD Dev. of
RTI Analyses
ppm
0
0
1.0
0.6
0.0
0.6
1.2
0.6
0
0.6
0
0
0
0
0
0.6
1.2
0
Difference
MFG - RTI
ppm
16
9
-17 '
44
- 0.7
- 5
- 6
-37
3.8
- 8
- 2.3
13
0.1
6
- 1.7
14
8
7.6
Percent
Accuracy
3.3
1.9
- 3.5
9.2
- 1.2
- 1.1
- 1.1
- 7.5
7.9
- 1.5
- 3.9
2.6
0.2
1.2
- 3.3
2.8
1.6
17.8
* Manufacturer's Certified Analysis
** Mean of Three Analyses
Percent Accuracy = 100(Manufacturer - RTI)/RTI
15
-------
TABLE 7. RESULTS OF NITRIC OXIDE CYLINDER ANALYSES (AUDIT 3)
Manufacturer
Airco Indus-
trial Gases
Air Products 4
Chemical, Inc.
Big Three
Industries
1 dea 1 Gas
Products
Liquid Carbon-
ics Corp.
Matheson Gas
Products
MG Scientific
Gas Products
North East
Cryogen i cs
Scientific Gas
Products, Inc.
Scott Environ-
menta 1 Tech . ,
Inc.
Scott-Marrin,
Inc.
Linde Div. of
Union Carbide
Cy 1 i nder
Construction
A 1 urn i num
Steel
Aluminum
Steel
Steel
Aluminum
Aluminum
Steel
Steel
Steel
Steel
Steel
Steel
Aluminum
Aluminum
A 1 urn i num
Steel
Steel
Manufacturer's
Analysis*
ppm
322
318
285
298
45
280
276
299
45.4
288
45
314
50
307
52.5
306
503
287
Mean of RTI
Analysis**
ppm
311
301
304
289
43
290
315
283
37
277
11
314
50
310
54
308
511
265
STD Dev. of
RTI Analyses
ppm
0
0
1.2
0
0
0.6
0.6
0.6
0
0
0
0.6
0
1.0
0.6
0
0.6
2.9
Difference
MFG - RTI
ppm
11
17
-19
9
2
-10
39
16
8.4
11
34
0
0
- 3
- 1.5
- 2
- 8
22
Percent
Accuracy
3.5
5.6
- 6.3
3.1
4.7
- 3.4
-12.4
5.7
22.2
4.0
>100*»*
0
0
- 1.0
- 2.8
- 0.6
- 1.6
8.3
* Manufacturer's Certified Analysis
** Mean of three analyses
*** Actual percent accuracy not calculated but was greater than 100?
Percent Accuracy = 100(Manufacturer - RTD/RTI
16
-------
TABLE 8. STABILITY OF SULFUR DJOXIDE CYLINDER GASES FROM AUDIT 2
Manufacturer Cylinder .
Construction
Airco Indus- Aluminum
trial Gases
Air Products Steel
& Chemicals,
Inc.
Big Three Steel •
Industries
Ideal Gas Steel
Products Steel
Liquid Car- Aluminum
bonics Corp.
Matheson Gas Steel
Products
MG Scientific Aluminum
Gas Products
North East Steel
Cryogenics
Scientific Aluminum
Gas Products,
Inc.
Scott Envi- Aluminum
ronmenta 1 Tech . ,
Inc.
Scott-Mar r In, A 1 urn 1 n urn
Inc.
Linde Div. of Steel
Union Carbide Steel
Manufacturer's
Analysis
ppm
496
478
498
460
81
513
526
577
450
532
520
499
500
90
RTI Analysis RTI Analysis
January 1979 Sept. 1979
ppm** ppm**
499 496
466 460
491 483
447 430
126 1 18
517 511
530 524
569 561
514 510
544 542
524 520
502 500
524 520
92.3 84
Difference Percent
ppm Difference
(1) (2)
- 3 - 0.6
- 6 - 1.3
- 8 - 1.6
-17 - 3.8
- 8 - 6.3
- 6 - 1.2
- 6 - 1.1
- 8 - 1.4
- 4 - 0.8
- 2 - 0.4
- 4 - 0.8
- 2 ' - 0.4
- 4 - 0.8
- 8.3 - 8.9
* Manufacturer's Certified Analysis
** Mean of three analyses
(1) Difference = September 1979 analysis - January 1979 analysis
(2) % Diff. = 100(Sept. 79 analysis - Jan. 79 analysis)/Jan 79 analysis
17
-------
TABLE 9. STABILITY OF NITRIC OXIDE CYLINDER GASES FROM AUDIT 2
Manufacturer
Alrco Indus-
trial Gases
Air Products
& Chemicals,
Inc.
Big Three
Industries
Ideal Gas
Products
Liquid Car-
bon ics Corp.
Matheson Gas
Products
MG Scientific
Gas Products
North East
Cryogenics
Scientific
Gas Products,
Inc.
Scott Envi-
ronmental
Tech., Inc.
Scott-Marrin,
Inc.
Linde Div. of
Union Carbide
Cyl inder
Construction
Aluminum
Steel
Steel
Steel
Steel
Al urn in urn
Steel
Aluminum
Steel
Steel
Aluminum
Aluminum
Al urn in urn
Aluminum
Steel
Steel
Manufacturer" s
Analysis
ppm
310
295
288
285
50
295
330
326
325
50
290
312
51.4
302
302
45
RTI Analysis
January 1979
ppm**
304
294
294
255
57.4
301
323
344
302
32.5
298
311
50.0
299
292
44.8
RTI Analysis
Sept. 1979
ppm**
294
285
289
232
59
297
314
337
302
13
294
307
51
300
287
46
Difference
ppm
(1)
-10
- 9
- 5
-23
- 1.6
- 4
- 9
- 7
0
-19.5
- 4
- 4
1
1
- 5
1.2
Percent
Difference
(2)
- 3.3
- 3.1
- 1.7
- 9.0
2.8
- 1.3
- 2.8
- 2.0
0
-60
- 1.3
- 1.2
2.0
0.3
- 1.7
2.7
* Manufacturer's Certified Analysis
** Mean of three analyses
(1) Difference = September 1979 analysis - January 1979 analysis
(2) % DIff. = 100(Sept. 79 analysis - Jan. 79 analysis)/Jan 79 analysis
18
-------
SECTION 5
SUMMARY OF RESULTS OF AUDITS 1, 2, AND 3
The results for Audits 1, 2, and 3 of commercial cylinder gases
are summarized in Tables 9 and 10. Table 9 gives the results for
sulfur dioxide and Table 10 gives the results for nitric oxide. Audit
3 was the first audit performed for carbon monoxide; therefore, a
summary table is not presented.
19
-------
TABLE 10. SUMMARY OF RESULTS FDR AUDITS 1, 2, and 3 OF COMMERCIAL CYLINDERS FOR
SULFUR DIOXIDE
Manufacturer
Alrco Industrial
Gases
Air Products &
Chemical, Inc.
Big Three Industries
Ideal Gas Products
Liquid Carbon ics
Corp.
Matheson Gas
Products
MG Scientific Gas
Products
North East Cryo-
genics
Scientific Gas
Products, Inc.
Scott Environ-
mental Tech., Inc.
Scott-Marrin, Inc.
Linde Div. of
Union Carbide
Cy 1 1 nder
Construction
A 1 urn 1 num
Steel
Steel
Aluminum
Steel
Steel
Steel
Aluminum
Steel
Aluminum
A 1 urn i num
Aluminum
Steel
Steel
Steel
Steel
Steel
Steel
A 1 urn I num
Steel
Aluminum
Aluminum
A 1 urn i num
Steel
Steel
Nominal
Cone, (ppm)
500
500
500
500
500
90
500
500
500
500
500
90
500
90
500
500
90
500
500
90
500
90
500
500
90
Percent Accuracy
Audit 1
(*)
9.0
7.5
4.0
-8.0
2.1
(*)
-7.2
-9.4
14.7
-10.1
- 4.8(1)
-- 1.0(1)
-19.7(1)
- 2.4
- 1.6
-11.3
0.2
5.8
2.2
Audit 2
- 0.6
2.6
1.4
2.9
-35.7
- 0.8
- 0.8
1.4
-12.5
-15.6
- 2.2
- 0.8
- 3.1
- 0.6
- 4.6
- 2.5
Audit 3
- 0.8
1.0
- 5.5
- 8.8
- 7.0
- 1.0
0.4
- 0.8
- 3.3
- 3.8
-21.6
-12.2
- 3.0
1.7
3.2
0.2
8.9
- 3.8
(*) Company not included in audit survey.
(1) These cylinders were not part of the first audit but were procured and analyzed after
reviewing the results of the first audit.
Percent Accuracy = 100(Manufacturer - RTD/RTI
20
-------
TABLE 11. SUMMARY OF RESULTS FOR AUDITS 1, 2, AND 3 OF COMMERCIAL CYLINDERS FOR
NITRIC OXIDE
Manufacturer
Airco Industrial
Gases
Air Products &
Chemical, Inc.
Big Three Industries
Ideal Gas Products
Liquid Carbon ics
Corp..
Matheson Gas
Products
MG Scientific Gas
Products
North East Cryo-
genics
Sci-entific Gas
Products, Inc.
Scott Environ-
mental Tech., Inc.
Scott-Marrin, Inc.
Llnde Dlv. of
Union Carbide
. Cylinder
Construction
A 1 urn i num
Steel
Steel
Aluminum
Steel
Steel
Steel
Aluminum
Steel
Aluminum
Steel
Steel
Steel
Steel
Steel
Steel
Steel
Steel
Aluminum
Aluminum
A 1 urn i num
Steel
Steel
Nominal
Cone, (ppm)
300
300
300
300
300
50
300
300
300
300
300
50
300
50
300
50
300
50
300
50
300
300
50
500
Percent Accuracy
Audit 1
(»)
2.2
11.1
>100(**)
>100(**)
- 7.0
(*)
- 4.8
>100(*»)
>100(**)
- 0.7(1)
4.0(1)
- 2.0
0.0
0.2
2.3
2.3
0.8
Audit 2
2.0
0.2
- 2.0
11.8
-12.9
- 2.0
2.2
- 5.2
7.6
53.8 '
- 2.8
0.3
2.8
1.0
3.4
0.4
Audit 3
3.5
5.6
- 6.3
3.1
4.7
- 3.4
-12.4
5.7
22.2
4.0
>100«*
0
0
- 1.0
- 2.8
- 0.6
8.3
- 1.6
* Company not included in audit survey.
** Actual percent accuracy not calculated but was greater then 100$.
(1) These cylinders were not part of the first audit but were procured and analyzed
after reviewing the results of the first audit.
Percent Accuracy = 100(Manufacturer - RTI)/RTI
21
------- |