&EPA
United States
Environmental Protection
Agency
Environmental Sciences Research
Laboratory
Research Triangle Park NC 27711
EPA-600 4-79-065
October 1979
Research and Development
Regional Air
Pollution Study
Carbon Dioxide
Effects on RAMS
Sulfur Monitors
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U S. Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are
1 Environmental Health Effects Research
2 Environmental Protection Technology
3 Ecological Research
4 Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7 Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL MONITORING series.
This series describes research conducted to develop new or improved methods
and instrumentation for the identification and quantification of environmental
pollutants at the lowest conceivably significant concentrations. It also includes
studies to determine the ambient concentrations of pollutants in the environment
and/or the variance of pollutants as a function of time or meteorological factors.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/4-79-065
October 1979
REGIONAL AIR POLLUTION STUDY
Carbon Dioxide Effects on RAMS Sulfur Monitors
D.H. Hern
Rockwell International
Environmetnal Monitoring & Services Center
Environmental & Energy Systems Division
11640 Administration Drive
Creve Coeur, MO 63141
Contract No. 68-02-2093
Task Order 121
Project Officer
Stanley Kopczynski
Atmospheric Chemistry and Physics Division
Environmental Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
ENVIRONMENAL SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
DISCLAIMER
This report has been reviewed by the Environmental Sciences Research
Laboratory, U.S. Environmental Protection Agency, and approved for publi-
cation. Approval does not signify that the contents necessarily reflect
the views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
-------
ABSTRACT
Data has been collected to verify and quantify the effect of sample
carbon dioxide content on the response of flame photometric sulfur gas
analyzers of two types, the Tracer model 270HA sulfur gas chromatograph and
the Meloy model SA 185 total sulfur analyzer. These analyzers were utilized
in the Regional Air Monitoring System (RAMS).
The effectiveness of the RAMS heatless air drier in removing carbon
dioxide from ambient air during the zero air generation process has been
investigated. The effects of variables such as atmospheric moisture,
ambient carbon dioxide concentration and dryer material have also been
determined.
A wet chemical analytical technique has been refined and adapted for
the determination of ambient carbon dioxide. The method is valid down to a
concentration of 10 ppm. The technique has been validated utilizing NBS
Standard Reference Material gases.
The diurnal variability of ambient carbon dioxide concentration at an
urban location has been determined with data taken for four individual
24-hour periods. Atmospheric stability for each sampling period was deter-
mined with upper air soundings.
This report was submitted in partial fulfillment of Contract No.
68-02-2093 by Rockwell International Corporation under the sponsorship
of the U.S. Environmental Protection Agency. This report covers a period
from April, 1977 to July, 1979 and work was completed as of July, 1979.
m
-------
CONTENTS
Abstract iii
Figures vi
Tables vi
1.0 Introduction 1
2.0 RAMS Zero Air System C02 Scrubbing Efficiency 3
2.1 Experimental Methods 3
2.2 Conclusions 5
3.0 Effect of Carbon Dioxide on Meloy and Tracer Sulfur Gas
Measurements 7
3.1 Experimental Methods 7
3.2 Conclusions 7
4.0 Simultaneous Operation of Meloy and Tracer Instruments 15
5.0 Diurnal C0? Sampling 17
6.0 Quality Assurance 19
Appendix
A. Procedures for the Determination of Carbon Dioxide in Air 20
-------
FIGURES
Number Page
1 Calibration Gas Configuration 8
TABLES
Number Page
1 C02 Content of Ambient and RAMS Zero Air 5
2 Tracor S02 Response - Site 106 9
3 Tracor Total Sulfur Response - Site 106 10
4 Meloy Total Sulfur Response - Site 107 11
5 Meloy Total Sulfur Response - Site 111 12
6 Meloy Total Sulfur Response - Site 111 (Duplicate) 13
7 Diurnal C02 Test Results - Site 106 18
vi
-------
1.0 INTRODUCTION
Data taken by the Rockwell International Air Monitoring Center,
Research Triangle Institute and others have shown that carbon dioxide can
be a suppress!ve interferent in the determination of gaseous sulfur com-
pounds by the flame photometric technique. Differences in response of up
to 20% have been reported for sample C02 concentration differences of zero
to 360 ppm.
Calibration techniques employed in the Regional Air Monitoring System
(RAMS) and other air monitoring programs utilize pollutant-free air (zero air)
as a diluent for permeation tube effluent and compressed gas standards. The
purification methods generally used to obtain this zero air frequently
deplete its C02 content. The use of this C02 deficient air to determine
instrument calibrations can result in the establishment of erroneous calibra-
tion coefficients which can lead to the understatement of ambient sulfur
gas concentrations. The primary objective of this experimental work was
to quantitatively determine the magnitude of this suppressive effect.
Effort was also expended to determine the possible effect of instrument
type and operational variables on this suppression.
Two types of sulfur gas analyzers were used in the RAMS network, the
Tracor model 270HA sulfur gas chromatograph and the Meloy model SA 185 total
sulfur analyzer. The effect of sample C02 content on the response of each
has been investigated.
The RAMS station gas analyzer calibration system utilizes zero air which
is generated by a heatless air dryer and purification device. Preliminary
analysis for the C02 content of the zero air effluent from this station zero
air generation system performed by Research Triangle Institute (RTI) has
indicated that the percentage of ambient level C02 (nominally 350 to 370 ppm)
which is removed by different systems within the RAMS network varies from
30 to 100 percent. The station zero air generation system influent and
1
-------
effluent have been analyzed daily at three RAMS stations for a twenty day
period to determine the system scrubbing efficiency and to attempt to
elucidate the factors, such as temperature and atmospheric moisture content,
which might effect this scrubbing efficiency.
Because of the possible effect of the ambient C02 concentration vari-
ability on the reported sulfur gas concentration and the zero air C02 content,
ambient air samples taken every four hours for four individual diurnal cycles
were analyzed for C02 content.
-------
2.0 RAMS ZERO AIR SYSTEM C02 SCRUBBING EFFICIENCY
2.1 EXPERIMENTAL METHODS
Determinations of CO^ in both influent and effluent of the heatless
dryer scrubbing system were performed to determine the efficiency of this
device for COp removal. These determinations were made daily at each of the
three stations involved in this study from 4 May 1977 to 24 May 1977, inclu-
sive. These analyses were performed by bubbling the sample air through gas
washing bottles (bubblers) which were prepared at the central facility,
sealed, and transported to the station for sampling. After sampling, the
bubblers were resealed and returned to the central facility for same-day
analysis.
The heatless dryer influent sample was obtained by pumping the inside
air from the shelter through a calcium sulfate drying cartridge (indicating
drierite), a mass flow meter, and then to the analytical absorption device
(bubbler). The drying cartridge was necessary because the moisture content
of the sample will effect its thermal conductivity and hence the ability to
be volumetrically measured with a mass flow meter. The heatless dryer
effluent was sampled from the station calibration system through an existing
test port on the calibration panel which allows pressurized zero air to be
sampled. This sample was volumetrically measured with the same mass flow
meter.
The relative humidity of the zero air system influent was measured by
taking a psychrometer reading inside the shelter at the time of COo sampling.
The enclosure containing the station zero air generation system is ventilated
with the air conditioning system output. The station inside air is, there-
fore, representative of the zero air system influent.
Table 1 contains the results of this study. Those days listed as "not
available", were due to problems at the RAMS site heatless dryer or computer.
Duplicate samples were collected on 18 May 1977 and 22 May 1977 from site
3
-------
TABLE 1. C02 CONTENT OF AMBIENT AND RAMS ZERO AIR
DATE SITE
5/4/77 111
5/4/77 107
5/4/77 106
5/5/77 111
5/5/77 107
5/5/77 106
5/6/77 111
5/6/77 107
5/6/77 106
5/7/77 111
5/7/77 107
5/7/77 106
5/8/77 111
5/8/77 107
5/8/77 106
5/9/77 111
5/9/77 107
5/9/77 106
5/10/77 111
5/10/77 107
5/10/77 106
5/11/77 111
5/11/77 107
5/11/77 106
5/12/77 111
5/12/77 107
5/12/77 106
5/13/77 111
5/13/77 107
5/13/77 106
AMBIENT
AIR C09
TIME L
(CST)
1145
0730
0915
1150
0655
0910
ZERO
AIR CO,
TIME i
(CST)
1225
0805
1005
1320
0740
1000
AMBIENT AMB. AIR ZERO AIR
RELATIVE
HUMIDITY
63%
86%
74%
62%
86%
64%
TEMP.
(°c)
27.5
17.8
22.7
21.1
21.1
18.8
C02 (ppm) (
(average of 2
359
440
327
360
441
332
;02 (ppm)
analyses)
54
16
105
49
14
95
Not Available
0755
1159
1016
0705
0840
1134
1102
0755
71%
60%
61%
64%
14.9
9.9
10.2
12.0
454
375
487
508
16
121
51
21
Not Available
1126
0704
0924
1041
0657
0843
1128
0706
0908
1151
0730
0845
1133
0712
0909
1136
0726
0940
1215
0750
1005
1134
0733
0920
1215
0750
0946
1258
0810
0924
1210
0750
0945
1241
0823
1020
50%
49%
53%
33%
32%
28%
29%
29%
31%
35%
43%
37%
35%
43%
43%
45%
46%
51%
18.6
21.9
17.2
17.5
17.8
20.6
23.3
20.0
21.1
22.0
17.8
21.4
22.5
16.9
20.0
25.5
19.7
23.0
455
510
347
398
422
379
400
457
398
403
445
424
398
517
360
372
454
340
16
21
no
51
21
118
45
40
95
46
22
133
46
18
118
50
40
145
(continued)
-------
TABLE 1 (continued)
DATE
5/14/77
5/14/77
5/14/77
5/15/77
5/15/77
5/15/77
5/16/77
5/16/77
5/16/77
5/17/77
5/17/77
5/17/77
5/18/77
5/18/77
5/18/77
5/18/77
5/19/77
5/19/77
5/19/77
5/20/77
5/20/77
5/20/77
5/21/77
5/21/77
5/21/77
5/22/77
5/22/77
5/22/77
5/22/77
5/23/77
5/23/77
5/23/77
5/24/77
5/24/77
5/24/77
SITE
111
107
106
111
107
106
111
107
106
111
107
106
111
107
106
107
111
107
106
111
107
106
111
107
106
111
107
106
106
111
107
106
111
107
106
AMBIENT
AIR C09
TIME
(CST)
0959
0747
0922
0717
1200
0915
1213
0738
1005
1158
0723
1006
0756
1100
0739
0924
1144
0726
0928
0853
1105
0707
1245
0748
0934
1009
1430
0947
1253
1143
0720
0930
ZERO
AIR CO
TIME
(GST).
"•• —
r\
2 RELATIVE
HUMIDITY
1108 46%
Not
0842
1015
Not
0812
1238
Not
0952
1307
0835
1049
1245
0825
1049
0853
1135
0815
1000
1226
0810
1012
0945
1143
0746
1330
0823
1049
1123
1516
1023
1331
1239
0807
1031
Available
47%
47%
Available
53%
47%
Available
50%
47%
43%
52%
47%
47%
59%
Duplicate
44%
51%
51%
46%
46%
49%
53%
40%
40%
42%
47%
45%
Duplicate
48%
56%
38%
58%
56%
59%
.
AMBIENT
TEMP.
(°O
26.5
25.0
27.5
24.5
27.5
27.5
27.5
26.5
27.0
27.5
25.5
25.0
Sample
27.5
25.5
25.5
27.0
26.5
26.5
27.5
28.5
23.5
26.0
25.5
24.0
Sampl e
28.0
24.5
31.5
26.5
24.5
24.5
AMB. AIR
C0? (ppm)
(average of
349
371
358
354
355
355
348
379
344
394
403
337
390
371
423
346
339
390
348
363
371
364
360
363
357
358
361
378
366
349
398
380
ZERO AIR
C0? (ppm)
2 analyses)
49
148
49
194
46
151
47
26
99
34
29
115
113
32
22
118
45
12
107
49
14
127
46
8
117
111
46
13
51
45
11
100
-------
107 and 106, respectively. These duplicates were taken and analyzed to
establish a confidence level on the accuracy of this sampling method.
The two RAMS sites that gave the most divergent COp concentrations in
the station zero air were sites 107 and 106. Sites 107 and 106 average 20
and 118 ppm C02, respectively. To determine the cause of this difference,
the two heatless dryer scrubber columns from both sites were removed and
disassembled. The scrubber from 106 was found to contain only activated
charcoal. That from site 107 contained a layer of 200 ml of Linde Molecular
Sieve 5A, with the remaining 800 ml of column consisting of activated
charcoal. This lends evidence that Mole Sieve 5A has a retentive effect on
co2.
Linde, in their materials catalogue, contend that Mole Sieve 5A will
effectively remove any molecule with an effective diameter of 5 angstroms or
less. C02 has an effective diameter of less than 4 angstroms.
Other parameters for the scrubbing efficiency of the heatless dryers
are the pressure at which scrubbing occurs, amount of air scrubbed, cycle
time for each half of the dryer, and the age of the column materials.
Because of time and material shortages, no attempt was made to determine if
any other column packings would have a decreased affinity for C02 (e.g.,
activated alumina), ^nd yet generate air free of pollutants and moisture.
2.2 CONCLUSIONS
From this study it can be concluded that the heatless dryers containing
activated charcoal do remove varying amounts of the C02 from the ambient air
influent. The amount of C02 scrubbed is enhanced by the added presence of
Linde Molecular Sieve 5A in the dryer. Other factors that are probably
involved, but were not tested, are the pressure at which the air is scrubbed
by the dryer and the total volume of air being withdrawn from the system at
the time of sampling. Relative humidity does not appear to have any affect
on the amount of C02 removed by the station zero air generation system.
-------
3.0 EFFECT OF CARBON DIOXIDE ON MELOY AND TRACOR SULFUR GAS MEASUREMENTS
3.1 EXPERIMENTAL METHODS
To determine the effect of the C02 sample concentration on Meloy and
Tracer response, span gas consisting of S02 concentration of 48 and 102 parts
per billion (ppb) were prepared using 4, 46, 92, 188 and 370 ppm COp.
The span gas samples were generated utilizing the portable dynamic
dilution system which was used for RAMS quality assurance field audits. The
source of S02 is a National Bureau of Standards (NBS) permeation tube
(elution rate 0.267 ul/min.) in a temperature controlled environment. The
various COo concentrations were generated by attachment of the required C02
spiked gas cylinder to the dilution gas input of the calibrator. Figure 1
gives a representation of the calibration system used to perform this test.
This test was done on the rack mounted Tracor at site 106 and the rack
mounted Meloy instruments at sites 107 and 111, with a duplicate test at site
111. All output data were generated from the RAMS data acquisition system and
recorded in Tables 2-6.
To determine the effect of flame support hydrogen flow on the C02 inter-
ference, the tests mentioned previously in this section were repeated using
varied hydrogen flow rates. These rates and their corresponding results are
also listed in Tables 2-6.
3.2 CONCLUSIONS
Results from Meloy sites 107 and 111 indicate that there is a total
sulfur gas response supression, ranging up to 20% for varied ambient C02
concentrations using constant sulfur gas concentrations. These results show
that the supression is not a function of sulfur gas present, but that some
slight variation does occur with varied hydrogen flows used to support the
detection flame. The factory recommended hydrogen flow rate of 52 ml/min.
-------
O
O
ie
0)
N
fc. C
o «t
(O
X J- E
i- jQ 0)
•O f- +J
E i— «/>
QJ «O >,
CO O CO
-------
TABLE 2. TRACOR S02 RESPONSE - SITE 106
(Run on 6/2/77)
C02 CONC.
(ppm)
4
46
92
188
370
4
46
92
188
370
4
46
92
188
370
INSTRUMENT RESPONSE
0 ppb S00 48 ppb SOQ
Run at 29
0.068 0.896
0.896
0.879
0.896
0.071 0.896
Run at 31
0.122 0.947
1.047
1.030
1.013
0.088 0.896
Run at 27
0.088 0.862
0.913
0.879
0.896
0.130 1.030
IN VOLTS
102 ppb S00
Ibs H2 (54.
1.770
1.738
1.704
1.789
1.790
Ibs H2 (64.
1.838
1.990
2.039
1.855
1.704
Ibs H2 (45.
1.670
1.719
1.636
1.704
1.704
INSTRUMENT
0 ppb S00 48
5 ml/min)
2.4
-
-
-
2.5
9 ml/min)
4.5
-
-
-
3.2
1 ml/min)
3.2
-
-
-
4-8
RESPONSE
ppb S00
33.9
33.9
33.2
33.9
33.9
35.8
39.6
39.0
38.3
33.9
32.6
34.5
33.2
33.9
39.0
IN ppb S02
102 ppb S00
67.1
65.9
64.6
67.8
67.8
69.7
75.4
77.3
70.3
64.6
63.3
65.1
62.0
64.6
64.6
Transfer equation: S02 (ppm) = 0.0380 (v) - 0.00018.
-------
TABLE 3. TRACOR TOTAL SULFUR RESPONSE - SITE 106
(Run on 6/2/77)
C02 CONC.
(ppm)
4
46
92
188
370
4
46
92
188
370
4
46
92
188
370
INSTRUMENT RESPONSE
0 ppb S00 48 ppb S00
"— ' f. •— ' £ •
Run at 29
0.042 0.900
0.920
0.901
0.884
0.173 0.864
Run at 31
0.061 0.977
1.069
1.013
0.994
0.134 0.884
Run at 27
0.098 0.920
0.957
0.901
0.938
0.190 0.920
IN VOLTS
102 ppb S00
L-l £
Ibs H2 (54
1.816
1.799
1.743
1.760
1.799
Ibs H2 (64
1.929
1.929
2.041
1.836
1.667
Ibs H2 (45
1.760
1.799
1.687
1.760
1.743
INSTRUMENT
0 ppb S00 48
^
.5 ml/min)
1.4
-
-
-
6.2
.9 ml/min)
2.1
-
-
-
4.8
.1 ml/min)
3.4
-
-
-
6.8
RESPONSE
ppb S00
' L.
33.0
33.8
33.1
32.5
31.7
35.9
39.3
37.2
36.5
32.5
33.8
35.0
33.1
34.5
33.8
IN ppb TS*
102 ppb S00
' c.
66.9
66.2
64.2
64.8
66.2
71.0
71.0
75.2
67.6
61.4
64.8
66.2
62.1
64.8
64.2
Transfer equation: TS* (ppm) = 0.03692 (v) - 0.000177.
*S02 Equivalent.
10
-------
TABLE 4. MELOY TOTAL SULFUR RESPONSE - SITE 107
(Run on 6/11/77)
C02 CONC.
(ppm)
4
46
92
188
370
4
46
92
188
370
4
46
92
188
370
INSTRUMENT RESPONSE
0 ppb S00 48 ppb S00
I—I {_ l-J £ •
Run
0.210 1.237
1.168
1.139
1.079
0.256 1.009
Run
0.257 1.195
1.181
1.099
1.029
0.255 0.973
Run
0.101 1.107
1.100
1.081
0.993
0.121 0.899
IN VOLTS
102 ppb S00 0
at 52 ml/min
2.315
2.219
2.191
2.081
1.900
at 48 ml/min
2.236
2.238
2.118
2.081
1.843
at 56 ml/min
2.094
2.085
2.089
1.842
1.790
INSTRUMENT RESPONSE
ppb S00 48 ppb S00
-4.4 45.3
41.9
40.5
37.6
-2.2 34.2
H2
-2.2 43.2
47.6
38.6
35.2
-2.3 32,5
H2
-9.7 39.0
38.6
37.7
33.5
-8.7 28.9
IN ppb TS*
102 ppb S00
97.4
92.8
91.4
86.1
77.3
93.6
93.7
87.9
86.1
74.6
86.7
86.3
86.5
74.5
72.0
Transfer equation: TS* (ppm) = 0.048393 (v) - 0.014615.
*S02 Equivalent.
11
-------
TABLE 5. MELOY TOTAL SULFUR RESPONSE - SITE 111
(Run on 6/12/77)
C02 CONC.
(ppm)
4
46
92
188
370
4
46
92
188
370
4
46
92
188
370
INSTRUMENT RESPONSE IN VOLTS
0 ppb S00 48 ppb S00 102 ppb S00 0
Run
0.061 1.167
1.127
1.116
1.045
0.083 0.977
Run
0.159 1.179
1.152
1.079
1.009
0.144 0.942
Run
-0.011 1.084
1.077
1.052
0.977
0.066 0.874
at 52 ml/min
2.266
2.165
2.150
2.046
1.865
at 48 ml/min
2.219
2.202
2.092
2.058
1.804
at 56 ml/min
2.078
2.061
2.058
1.926
1.760
INSTRUMENT RESPONSE
ppb S00 48 ppb S00
0.5 46.3
44.7
44.2
41.3
1.4 38.4
H2
4.6 46.8
45.7
42.7
39.8
4.0 37.0
H2
-2.5 42.9
42.6
41.6
38.4
0.7 34.2
IN ppb TS*
102 ppb S00
91.8
87.6
87.0
82.7
75.2
89.9
89.2
84.5
83.2
72.7
84.0
83.3
83.2
77.7
70.9
Transfer equation: TS* (ppm) = 0.041355 (v) - 0.002018.
*S02 Equivalent.
12
-------
TABLE 6. MELOY TOTAL SULFUR RESPONSE - SITE 111 (Duplicate)
(Run on 6/12/77)
C02 CONC.
(ppm)
4
46
92
188
370
4
46
92
188
370
4
46
92
188
370
INSTRUMENT RESPONSE
0 ppb S00 48 ppb S00 1
•— ' C. L.
Run
0.062 1.166
1.130
1.116
1.044
0.082 0.979
Run
0.158 1.177
1.145
1.077
1.088
0.139 0.945
Run
-0.011 1.079
1.075
1.050
0.977
0.061 0.877
IN VOLTS
02 ppb S00
L.
at 52 ml/min
2.265
2.166
2.171
2.042
1.861
at 48 ml/min
2.221
2.200
2.095
2.055
1.805
at 56 ml/min
2.077
2.060
2.059
1.925
1.761
INSTRUMENT RESPONSE
0 ppb S00 48 ppb S00
H2
0.5 46.2
44.7
44.1
41.2
1.4 38.5
H2
4.5 46.7
45.3
42.5
43.0
3.7 37.1
H2
-2.5 47.6
42.4
41.4
38.4
0.5 34.2
IN ppb TS*
102 ppb S00
C.
91.6
87.6
87.8
82.4
74.9
89.8
89.0
84.6
83.0
72.4
83.9
83.2
83.1
77.6
70.8
Transfer equation: TS* (ppm) = 0.041355 (v) - 0.002018.
*S0 Equivalent.
13
-------
was identical for each Meloy. By changing this flow rate to 56 and 48
ml/min., no appreciable improvement in response versus C02 concentration was
found.
Analysis of the data from the Tracor at site 106 indicated no change in
instrument response due to varied sample C02 concentration, at either 48 or
102 ppb S02. This held true for both the total sulfur and the S02 output
channels. There was a +_ 2% variation in response for both channels, but
this did not fit into any progression concerning CO,, concentration. An in-
crease in this variation did occur when the instrument was operated at both
a higher and lower hydrogen flow rate. As before, the variation was in no
orderly progression and can be attributed to instrument noise due to opera-
tion at the extreme hydrogen flow rates.
It was expected that the Meloy would be more sensitive to ambient (XL
concentrations. The Meloy instrument draws in between 50 and 60 ml/min. of
sample air for analysis and to support the hydrogen detector flame. This
air is subject to variation in C02 concentration depending on its source
(i.e. ambient or station zero air). The Tracor injects only a 5 ml sample
per injection. This amount is dwarfed in the hydrogen flame by 100 to 120
ml/min. of air from the RAMS air purification system. As the heat!ess
dryer generates air of a generally constant C02 concentration, the changes in
instrument sensitivity due to variations in ambient sample C02 concentration
are minimized.
14
-------
4.0 SIMULTANEOUS OPERATION OF MELOY AND TRACOR INSTRUMENTS
The comparability of ambient air sulfur gas content reported by the
Meloy and Tracor sulfur gas analyzers was investigated by operating both
instruments at RAMS station 106, 107, and 111. Stations 107 and 111 each
have normally installed Meloy instruments. These stations each were equipped
with a Tracor instrument mounted on the station workbench. Station 106 has a
normally installed Tracor instrument and an auxiliary Meloy was installed at
this site. Each of the auxiliary instruments derived its ambient air and
calibration gas sample identically to the station instument. A spare
run/calibrate valve was electrically connected to the sulfur analyzer run/
calibrate signal for the auxiliary instrument, allowing the selection of
either the ambient air sample manifold or the calibration manifold as the
sample source under either station computer or manual control. The signal
output from the auxiliary instruments was connected to the RAMS station data
acquisition system via unused signal input channels.
From 4 May 1977 to 24 May 1977, inclusive, both station and auxiliary
sulfur analyzers were zeroed/spanned manually for the 20 days of this test.
The total sulfur parameter for the Meloy instruments and the total sulfur and
sulfur dioxide parameters for the Tracor instruments were subjected to this
daily calibration only on the most sensitive operating ranges (0-0.2 ppm).
This calibration was in addition to the normal computer executed zero/span
each evening. The manually executed calibrations were performed each data at
approximately the same hour and immediately following the sampling activities
for the zero air COp concentration. Data obtained from these calibrations
were derived from the remote station data acquisition system and were
manually recorded. Slope and intercept values were computed manually and
these calibration values were input to the RAMS central computer for con-
version of level 1 to level 2 data exclusive for this study.
The central computer was programmed to extract both station and
15
-------
auxiliary instrument data from the level 1 tape file, and the data was
placed on a special tape file to be processed to level 2 (engineering unit)
using the manually input calibration constants. The level 2 data in this
file had only lower detection limit validation checking. All other valida-
tion was manually performed. An attempt was made to analyze the data;
however, the contract effort was terminated before any conclusions were
reached regarding the comparability of the Meloy and Tracer analyzers.
16
-------
5.0 DIURNAL C02 SAMPLING
To determine the fluctuation of ambient CCL concentration, diurnal
sampling was conducted for four 24 hour periods at site 106. These samples
were collected every 4 hours and analyzed within the hour. Samples were
gathered and analyzed per the procedures in Appendix A. These results are
presented in Table 7.
Results range from a high (XL reading of 409 ppm (XL at 0215 central
standard time on 21 May 1977 to a low of 322 ppm C02 at 1344 on 24 May
1977. A general tendency seen from these results is the increase in (XL
concentration during hours of darkness. Bacterial decay of plant materials
on the surface is most likely the predominant source of atmospheric (XL.
The increases in (XL concentration observed during hours of darkness are
attributed to this constant source coupled with the reduced vertical mixing
experienced during nighttime hours.
17
-------
TABLE 7. DIURNAL C02 TEST RESULTS - SITE 106
DATE TIME*
1st
2nd
3rd
4th
DIURNAL PERIOD
5/20/77
5/20/77
5/20/77
5/20/77
5/21/77
5/21/77
DIURNAL PERIOD
5/24/77
5/24/77
5/24/77
5/24/77
5/25/77
5/25/77
DIURNAL PERIOD
5/31/77
5/31/77
5/31/77
5/31/77
6/1/77
6/1/77
DIURNAL PERIOD
6/7/77
6/7/77
6/7/77
6/7/77
6/8/77
6/8/77
0928
1412
1815
2215
0215
0707
0930
1344
1800
2200
0200
0600
1000
1400
1800
2200
0200
0600
1030
1430
1830
2230
0230
0630
348
337
347
363
409
363
351
322
338
352
373
403
371
369
354
360
367
366
357
341
369
371
371
344
* All time is reported in central standard time,
18
-------
6.0 QUALITY ASSURANCE
A Bendix dynamic calibrator was used in performing the work described
in section 3.1. This device was calibrated prior to and checked after its
use by the positive soap film displacement technique. No discrepancy was
found between these two calibrations.
The permeation device used in this section to generate SCL* was a
National Bureau of Standards permeation tube, calibrated at 0.267 ul/min.
19
-------
APPENDIX A
PROCEDURES FOR THE DETERMINATION OF CARBON DIOXIDE IN AIR
A.I INTRODUCTION
In compliance with Task Order 121, an analytical method was developed
for the analysis of carbon dioxide (C02) in air samples. It was necessary
that this method be accurate, reproducible, relatively simple, have an
analytical range between 0 and 550 parts per million (ppm) C02» and not
require a large capital expenditure. A literature search revealed an
analytical method for COp using the reaction between C02 and barium hydrox-
ide (Ba(OH)2).(1'2)
The following sections detail the analysis, sampling, quality assurance,
and other related procedures used for these C02 analyses.
20
-------
A. 2 SUMMARY
The method which was used for the determination of CO^j involves the
dissolution and reaction of C02 in a known amount of a standard barium
hydroxide solution [Ba(OH)2]. An acid-base titration of the unreacted
Ba(OH)2 was then performed with a known concentration of oxalic acid
[(COOH)?]. These factors along with the known volume, at standard temper-
ature and pressure (STP), of sample passed through the absorbing solution,
determined the C02 concentration in parts per million.
The reaction between C02 and Ba(OH)2 absorbing solution is given by
the following equation:
C02 + Ba(OH)2 + BaC03 4- + H20
The titration of the remaining base with oxalic acid is given by the
following equation:
Ba(OH)2 + (COOH)2 -»• Ba(COO)2 + 2H20
Tests were run using certified standards to insure the accuracy of this
method. These tests proved this method accurate to +_ 3 ppm.
Because the concentrations of C02 found during these experiments were
substantially larger than those of any interfering gases (e.g. SO^, which
occur at maximum concentrations of less than 0.5 ppm), no special gas traps
or correction factors were needed.
21
-------
A.3 ANALYTICAL AND SAMPLING PROCEDURES FOR C02 DETERMINATION
The following paragraphs describe the sequence of preparations required
for carbon dioxide (COp) determination in ambient, station zero, and C0?
spiked standard air.
A.3.1 PREPARATION OF REAGENTS
A.3.1.1 Preparation of 0.1 molar (M) oxalic acid stock solution
Dilute 12.607 gm oxalic acid [(COOH)2 • Zh^O]1 to 1 litre using freshly
deionized water in a volumetric flask.2 Transfer this solution to a
stoppered glass bottle.3
A.3.1.2 Preparation of 0.005 M oxalic acid titration solution
Pipette 50 ml of 0.1 M oxalic acid stock solution into a volumetric
flask and dilute to 1 litre using freshly deionized water. Transfer this
solution to a stoppered glass bottle.3
A.3.1.3 Preparation of 0.1 M barium hydroxide [Ba(OH)2] stock solution
Dissolve 31.548 gm of Ba(OH)2 • 8H20 in a volumetric flask and dilute
to 1 litre using freshly deionized water. Due to carbonates and other
impurities normally found in crystaline Ba(OH)2, the resulting turbid
1. Oxalic acid is used as a standard for determining C02 because it is
available in a purity of > 99.99% and a stable hydrated form.
2. Freshly deionized water is free of any carbonates or carbon dioxide.
3. As oxalic acid is used as an analytical standard material, these
pipettings and weighings need to be exact. The oxalic acid is A.C.S.
grade in crystalline form.
22
-------
solution requires filtration prior to storage in a glass stoppered bottle."*
A.3.1.4 Preparation of 0.01 M Ba(OH)2 absorption solution
Pipette 100 ml of 0.1 M Ba(OH)2 stock solution into a volumetric flask
and dilute to 1 litre using freshly deionized water. Transfer this solution
to a stoppered glass bottle.1*'5
A.3.2 PREPARATION OF A C02 GAS SAMPLING BOTTLE
Pipette 100 ml of 0.01 M Ba(OH)2 absorbing solution into a 250 ml
fritted glass gas washing bottle (bubbler).6 Add 10 drops (0.5 ml) of
1-butanol to this solution.7 Plug both ends of the bubbler to preclude any
ambient C02 contamination.
A.3.3 SAMPLE GATHERING
Samples to be analyzed for C02 are bubbled through the C02 bubbler at
a rate of 1 litre per minute (1/min) at standard temperature and pressure
(STP) for 30 minutes.
The rate of 1 1/min was chosen as close to the maximum allowable rate
without an absorption fluid loss due to overflow. 30 minutes was chosen as
a sampling time in order that a mass of C02 in excess of what could be
4. Any pipetting or weighing need not be exact. Precise standardization
will be done with the 0.005 M oxalic acid solution.
5. Unlike the acidic oxalic acid solutions where C02 is insoluable, C02
is very soluable and reactive in the Ba(OH)2 working and absorbing
solution. Care is required to minimize ambient C02 contamination.
6. The exact amount of absorbing solution used to charge a bubbler is
critical. While any known amount could be used, 100 ml was used as it
filled the bubbler above the fritted glass and was not subject to
overflow during sampling.
7. 1-Butanol is used to relieve surface tension in the Ba(OH)2 solution
and generate a 10 cm bubble column to facilitate C02 absorption.
23
-------
absorbed would not be sampled.8 Ambient (XL sampling required the use of a
metal diaphram pump to force sample through the mass flow meter while
sampling station zero air required only a pressure regulator. This regulator
was used to reduce the station zero air pressure from 70 Ibs to the 10 Ibs
needed for use with the mass flow meter. Figure A.I illustrates the sampling
train used.
A.3.4 SAMPLE ANALYSIS
The exact molarity of unreacted Ba(OH)2 absorbing solution used to
charge the (XL bubbler(s), is determined by pipetting 25 ml of the stock
absorbing solution and titrating with 0.005 M oxalic acid. Two drops 1% thy-
mol phthalein are used as an indicator, with the solution titrated from blue
to colorless. The ml of oxalic acid used is recorded and later used for
calculating C02 concentrations.
This procedure is repeated when titrating a reacted bubbler absorbing
solution. As some of the barium carbonate (Ba(XL) precipitate formed by the
Ba(OH)2 reaction with (XL will enter into the titration, the titration
should be performed slowly to enable a complete Ba(OH)2 digestion from the
precipitate by the oxalic acid. A reference titration proves very beneficial
in determining the end point.9
It is preferable that no BaCXL precipitate enter the titration, but
without the use of centrifuge, this is almost impossible. As Ba(XL pre-
cipitate is soluable in acidic solutions, a loss of carbon dioxide is
possible through a "fizzing-out" process if the end point of the titration
has an acidic pH. Thymolphthalein is used as an indicator as its transition
range is 9.3-10.5. The final end point for this titration is basic and
allows no C02 loss through dissolution of the BaC03 precipitate.
All titrations were done using a 50 ml burette. By titrating 25 ml
8. The maximum C02 concentration that may be used for these reagent and
sampling parameters is 747 ppm C02.
9. A gas tube exhausting (XL free air in the titration flask proves
beneficial in retarding ambient (XL interference. Ascarite efficiently
absorbs all C02-
24
-------
QQ
LU
O
O
C£
C£
Z O
O D-
-------
of absorbing solution, at twice the normality of the titrating solution, no
need arose to use more than 50 ml 0.005 M oxalic acid solution and thusly
no need to refill the burette with its associated errors.
A.3.5 CALCULATING C02 CONCENTRATIONS
From the amount of 0.005 M oxalic used to neutralize 25 ml unreacted
Ba(OH)2 absorbing solution (section A.3.4) and the following equation, the
exact Ba(OH)2 molarity was determined, where:
M D tr\u\ - ml of oxalic acid used x oxalic acid molarity
M Ba^2 ml of Ba(OH)2 titrated
From the molarity of Ba(OH)2 and the amount of 0.005 M oxalic acid used
to neutralize the absorbing solution in a used C02 bubbler (section A.3.4),
the following equation applies:
A 22400 (100 xB-4xCxD)
A ~ E
where:
A = CCL cone, in the sample gas in ppm
B = Ba(OH)2 molarity
C = oxalic acid molarity
D = ml of oxalic acid used in neutralization of Ba(OH)2
E = amount of air sample in litres
This equation was derived from the reaction between C02 and Ba(OH)2
where:
C02 + Ba(OH)2 -»• BaC03 +•+ H20
that C02 (gas) occupies 22.4 litres at STP;
and that a mole per mole reaction occurs between oxalic acid and Ba(OH)2
during titration, where:
Ba(OH)2 + (COOH)2 -»• Ba(COO)2 + 2H20
26
-------
The constants in the equation used to determine CCL concentration
apply only if 100 ml of absorbing solution is used to charge a bubbler, and
if 25 ml of this solution is titrated.
Experiments showed that approximately 0.5 ml of absorbing solution was
lost for each 10 litres of sample collected. Allowing for the 0.5 ml added
from the 1-butanol required,, a correction (reduction) of 1% to the C02 concen-
tration is required for a 30 litre sample.
A.3.6 CALIBRATION OF MASS FLOW METERS
A Tylan mass flow meter was used to measure the flow rate of sample
through the CO? bubbler. This mass flow meter was calibrated for 1 1/min at
STP and rechecked approximately once every week during the period of sampling.
The original setting at the start of sampling was 0.164 volts and did not
change during the period of use. On 31 May 1977 a Brooks mass flow con-
troller was used with considerable success due to its incorporation of a
flow controller. The Brooks mass flow controller proved much more stable
at regulating a constant flow and in hindsight would have proved a better
initial choice. The Brooks mass flow controller and Tylan mass flow meter
were both calibrated against a 5 litre bubble tube. The setting of 0.352
volts for the Brooks at 1 1/min (STP) proved constant during its 10 day use.
A check was made to determine if any difference in flow rate occurred
for using the metal diaphram pump (section A,3.3) versus a pressurized air
source. This was investigated because of the pulsed output of all diaphram
pumps. No difference was seen for the two sources.
Because mass flow meters are sensitive to moisture, a CaSO, dryer was
incorporated into the ambient air sampling train (Figure A.I). The station
zero air is moisture free and hence required no dryer. The CaSO, in the
dryer was changed every two days to preclude any moisture breakthrough.
Many different materials could have been used to remove moisture from
the air sample, but many would also have an affinity for C02 (e.g. silica
gel). Calcium sulfate (CaSO^) and potassium carbonate (K2C03) are two of
the most common desiccants that have no affinity for C02. To prove this
contention a check was made of the amount of oxalic acid needed to neutralize
27
-------
30 litres of 300 ppm C02 (Linde) with and without the sample passing through
a CaSO^ dryer. The results are presented in Table A.I and indicate no C02
adsorption by the CaSO..
A.3.7 QUALITY ASSURANCE
Tests were performed to assess the accuracy and reproducibility of the
analyses done for Task Order 121. These tests included absorption efficiency
of C02, reproducibility of titrations and analyses, and agreement between
actual and standard C02 concentration results.
A.3.7.1 C02 Absorption Efficiency
In the bubbler absorption method of analysis, it is necessary that all
of the C02 be absorbed in the absorbing solution with none being allowed to
pass through. A test was devised to determine if any C02 was passing
through the (XL bubbler unreacted.
Using a Scott-Marrin analyzed tank of 370 +_ 18 ppm C02 in air10, two
identical C02 bubblers were connected in series and this C02 standard
sampled at 1 1/min (STP) for 46 min. While a higher C02 concentration
standard run at 30 min. would have been more desirable, this was the best
that could be accomplished using available standards. This standard is
comparable to a C02 concentration of 580 ppm C02 run for 30 minutes. The
results are presented in Table A.2 and indicate that no C02 passes through
the first bubbler. This test only proves that this analytical method is
accurate for concentrations up to 580 ppm C02 sampled at 1 1/min (STP) for
30 minutes. Any concentration in excess of 580 ppm, sampled at 1 1/min (STP)
for 30 minutes would require further checks af absorption efficiency. Note
that the maximum theoretical concentration that can be handled using these
sampling and analytical procedures11 is 747 ppm C02.
10. Analyzed at 378 ppm C02 by Rockwell personnel.
11. Sample bubbled at 1 1/min (STP) for 30 min. through 100 ml of 0.01 M
Ba(OH)2.
28
-------
A.3.7.2 Reproducibility of Titrations
Prior to any sampling, tests were run to determine if any variations
were present in the acid-base titrations used in analyses. The results are
presented in Table A.3 and show only a +_ 0.05 ml 0.005 M oxalic acid vari-
ation. Later duplicate titrations of sample analyses showed a maximum vari-
ation of 0.15 ml corresponding to an error of +_ 3 ppm CO,,.
A.3.7.3 Analysis of Scott-Marrin and NBS C02 Standards
Included in Table A.4 are the analyses of the 5 Scott-Marrin C02
standards used in this study. As may be seen, these results were within
analysis tolerances supplied by Scott. Each analysis was repeated using
the analytical procedures described in section A.3.4. The sampling procedure
was equivalent to that used for the station zero air (section A.3.3).
A certified C02 cylinder was purchased from the National Bureau of
Standards (NBS) for a final check of the accuracy of all analytical and
sampling procedures. A Bendix portable calibrator was used to generate
varying C02 concentrations. These C02 concentrations were sampled as in
section A.3.3 (ambient air samples) and analyzed as in section A.3.4. The
relatively low C02 maximum concentration (85 ppm) is a function of the
limitations of this particular calibrator. As may be seen from Table A.5,
the comparability between theoretical and analyzed (XL concentrations are
very good.
Tables A.4 and A.5 show that each analysis was performed twice. The
difference between the two of 2 ppm maximum is less than the 3 ppm maximum •
shown in section A.3.7.2. This decrease in error may be in part due to
improved laboratory techniques and in part to the incorporation of the
Brooks mass flow controller (section A.3.6).
A.3.7.4 Analytical Stability with Respect to Sample Volume
During the early part of this Task Order, tests were conducted on the
variability of analytical results as a function of sample volume. These
tests were done in accordance with all of the sampling (section A.3.3) and
analytical methods (section A.3.4), with sample flow controlled using the
Tylan mass flow meter. The results are presented in Table A.6. These datum
29
-------
show good comparative results regardless of sample flow used. The error
from this test is ±3 ppm COo. The gas used was a Linde CCL standard analyzed
at 300 ±18 ppm.12
12. Analyzed by Rockwell personnel at 293 ppm (XL.
30
-------
TABLE A.I. CaS04 ADSORPTION OF C02 TEST
No CaS04 Dryer CaS04 Dryer
1st titration 27.80 ml 1st titration 27.70 ml
2nd titration 27.75 ml 2nd titration 27.75 ml
Titrations using 0.005 M oxalic acid.
TABLE A.2. C02 ADSORPTION EFFICIENCY
Ba(OH)2 soln. (unreacted) Ba(OH)2 soln. 2nd bubbler
1st titration 49.0 ml 1st titration 49.0 ml
2nd titration 49.0 ml 2nd titration 49.0 ml
Titrations using 0.005 M oxalic acid
TABLE A.3. TITRATION REPRODUCIBILITY
Titration #1 36.40 ml oxalic acid
Titration #2 36.45 ml oxalic acid
Titration #3 36.40 ml oxalic acid
*Titration #4 36.40 ml oxalic acid
*14 hours later
Ba(OH)2 cone. * 0.01 M
Oxalic acid + 0.005 M
31
-------
TABLE A.4. ANALYSIS OF SCOTT-MARRIN C02 STANDARDS
Rockwell Analysis
Scott COp Standard Analysis #1 Analysis #2
4.2 + 0.4 ppm C02
46 +_ 2 ppm
92 +_ 5 ppm
188 +_ 9 ppm
370 +_ 18 ppm
4.4 ppm
47 ppm
94 ppm
190 ppm
379 ppm
4. 4. ppm
46 ppm
93 ppm
190 ppm
377 ppm
TABLE A.5. ANALYSIS OF NBS C02 STANDARDS
Rockwell Analysis
NBS Standard (listed - 1%) Analysis #1 Analysis #2
60 ppm 60 ppm 59 ppm
85 ppm 85 ppm 86 ppm
22.6 ppm 23 ppm 22 ppm
TABLE A.6. SAMPLE VOLUME ANALYSIS TEST
Amount of Sample (STP) Analysis
10 litres 296 ppm
20 litres 293 ppm
60 litres 290 ppm
32
-------
REFERENCES
1. Jacobs, M. B., The Chemical Analysis of Air Pollutants, p. 245, 1960,
Interscience Publications, Inc.
2. Lei the, W., The Analysis of Air Pollutants, p. 188-194, 1972, Ann Arbor
Science Publications, Inc.
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
I2T
EPA-600/4-79-065
3. RECIPIENT'S ACCESSION NO.
/I. TITLE AND SUBTITLE
REGIONAL AIR POLLUTION STUDY
Carbon Dioxide Effect on RAMS Sulfur Monitors
6. PERFORMING ORGANIZATION CODE
5. REPORT DATE
October 1979
7. AUTHOR(S)
D.H. Hern
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Rockwell International
Environmental Monitoring & Services Center
11640 Administration Drive
Creve Coeur, MO 63141
10. PROGRAM ELEMENT NO.
1AA603 AA-126 (FY-79)
11 CONTRACT/GRANT NO.
68-02-2093
Task Order 121
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory - RTP, NC
Office of Research and Development
J.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/09
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Effects of carbon dioxide (C02) content of the air on the response of flame photo-
metric sulfur gas analyzers of two types, the Tracer model 270 HA sulfur chromatograph
and the Meloy model SA 185 total sulfur analyzer, were studied. These analyzers were
used in the Regional Air Monitoring System (RAMS). For each instrument, measurements
were made to determine response to a matrix of five C02 levels and three sulfur dioxide
(S02) levels. Measurements were also made of C02 concentrations in the influent to and
effluent from heat!ess air dryers providing zero air for calibration at the RAMS sta-
tions.
Little effect on the Tracer response to increased C02 was detected on either the
SO, or total sulfur channel. The Meloy response was a suppressing effect of C02
which was linear over the values measured, averaging about a 20% suppression at the
highest CO, level used (370ppm). The percentage suppression was independent of S02
concentration and of detector flame hydrogen flow rate. The zero air contained
varying amounts of C02, apparently somewhat dependent on scrubber column packing and
operating parameters, slightly dependent on influent C02 content, and not dependent
on relative humidity. Because of the many RAMS component changes carried out during
the period of measurements, detailed, site-specific corrections to the Meloy readings
for the effects of C02 suppression would not be reliable and should not be made.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
*Air pollution
*Carbon dioxide
*Sulfur
*Monitors
Flame photometry
Regional Air Monitoring
Systems (RAMS)
St. Louis, MO
13B
07B
14B
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19 SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
40
20 SECURITY CLASS (This page)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (Rev. 4-771
34
------- |