EPA/600/A-93/251
Simultaneous Calibration of Open-path and Conventional
Point Monitors for Measuring Ambient Air Concentrations
of Sulfur Dioxide, Ozone, and Nitrogen Dioxide
Frank F. McElroy
Jimmie Hodgeson
Thomas A. Lumpkin
Kenneth A. Rehme
Robert K. Stevens
Atmospheric Research and Exposure Assessment Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Charles P. Conner
ManTech Environmental
Research Triangle Park, NC 27709
Hans Hallstadius
Opsis AB
Furulund, Sweden
ABSTRACT
A two-stage dilution system and an associated procedure to simultaneously calibrate both open-path
(long-path) and conventional point air monitors have been used successfully during a comparison test
study of open-path monitoring systems in Houston during August, 1993. Two open-path, differential
optical absorption spectrometers (DOAS) were calibrated using standard concentrations up to 50 ppm in
a 0.9 meter stainless steel optical cell connected to each DOAS analyzer via fiber optic cables. The
calibration apparatus consisted of various flow controllers, flow meters, and mixing chambers to
provide accurate dynamic flow dilutions; a high-concentration ozone generator; a suitable reaction
chamber for gas phase titration (GPT) of NO to generate N02 standard concentrations; and an output
manifold. High-concentration standards for S02 and NO were obtained by the primary dilution of
nominal 1000-ppm, NIST-traceable concentration standards in compressed gas cylinders. N02
concentrations were generated by GPT from NO concentrations. The ambient-level concentration
standards were provided by quantitative secondary dilution of the high concentration standards. Ozone
standard concentrations were generated by the high-concentration ozone generator and assayed, after
secondary dilution to ambient-level concentrations, by a commercially available UV photometric ozone
analyzer used as a transfer standard.
INTRODUCTION
Instruments that measure atmospheric pollutants over an extended, linear, open path through the
atmosphere offer several significant advantages relative to conventional air pollutant monitors (point
analyzers), which extract an air sample from the atmosphere at a single point for analysis inside the
analyzer. The primary advantages are much better spacial representation of a geographical area and
better probability of capturing measurements of local sources under variable wind directions. Other
advantages include greater siting flexibility, capability of monitoring areas or locations not readily
accessible with conventional monitors, measurement of pollutants in situ without risk of pollutant
alteration due to contact with surfaces and components in the inlet and analytical systems of

-------
conventional monitors, and capability for monitoring multiple pollutants simultaneously with the
same instrument.
Open-path, differential optical absorption spectrometer (DOAS) monitoring instruments have
been used for several years in Europe to monitor 03, S02, and N02 in the ambient air. Such
instruments are now commercially available in configurations suitable for potential application in the
United States by state and local air monitoring agencies. However, two things must happen before
open-path instruments can be used in EPA-required state and local air monitoring stations (SLAMS):
1) Open-path analyzers must be formally designated by EPA as equivalent methods,1and 2) the
EPA monitoring regulations^ must be amended^ to accommodate the use of open-path* pollutant
measurements and to govern the application, siting, operation, and quality assurance of open-path
analyzers. A critical part of the latter requirement is the quality assurance procedures associated
with the operation of open-path analyzers, particularly calibration and assessment of accuracy and
precision of open-path pollutant measurements.
In an effort to gather operational information and experience with open-path monitoring
analyzers in support of the development of regulatory amendments, EPA carried out a field study of
the technology in Baytown (Houston), Texas in August, 1993. This site, on property owned by the
Exxon Corporation, was selected because of the availability of an existing, well-equipped, conven-
i tional air monitoring station; substantial concentrations of S02, N02, and 03 typically recorded at
the site; suitable siting for the installation of open-path monitoring instruments; and convenient site
access and security. Since one objective of the study was to compare the performance and pollutant
. measurements of open-path and conventional analyzers, both types of analyzers were operated in the
study. In addition to the existing conventional monitoring station, a second monitoring station
I housing conventional analyzers for S02, N02, and 03 and two open-path DOAS analyzers were
installed for the study. The DOAS instruments were manufactured by Opsis AB (Furulund,
! Sweden); one was owned by EPA and the other was provided for the study by Opsis AB. Both
i DOAS instruments operated with parallel paths, each 400 meters long.
Another objective of the study was to test the efficacy of the calibration technique recommended
1 by the manufacturer and the audit and precision check procedures proposed by the manufacturer for
incorporation into the quality assessment requirements of the monitoring regulation/4* Because of
the comparative aspects of the study, it was deemed important to calibrate both the conventional
analyzers and the open-path instruments simultaneously, using the same calibration standards.
: Accordingly, a calibration system and associated calibration procedure for simultaneous calibration
of both types of analyzers for S02, N02, and 03 was assembled and used successfully during the
: field study.
CALIBRATION SYSTEM DESCRIPTION
The calibration technique prescribed by Opsis for its DOAS instruments utilizes an optical
calibration cell having quartz end windows and gas ports through which pollutant gas concentration
standards can be passed. The regular outdoor ultraviolet (UV) light transmitter and receiver used
for monitoring are temporarily replaced during calibration by a local UV light source and two
smaller receivers. These receivers are mounted to a calibration bench, which holds them and the
calibration cell in optical alignment, as shown in Figure 1. One receiver is connected, via a fiber
optic cable, to the DOAS analyzer in place of the fiber optic cable from the normal (outdoor)
*It is actually the "long-palh" nature of the pollutant measurement, «'. t. the averaging of pollutant concentrations in the
atmosphere over a path many meters long (path-integrated concentration), that is of concern with respect to the EPA monitoring
regulations. Unfortunately, the term "long path monitoring" has been (tentatively) defined by A&WMA Technical Committee EM-6
in its Remote Sensing Glossary to apply to optical instruments having a folded optical path, either with the path in a closed cell or
open to the atmosphere. In the later case (which should logicalfy be categorized as an open-path analyzer) the actual monitoring path
through the atmosphere is typically so short (less than 1 meter, for example) that EPA would consider the instrument to be a point
analyzer for regulatory purposes. Committee EM-6 defines "open-path monitoring" in the extended path sense; therefore, "open-
path" is used in this paper to be consistent with that definition (pending possible further refinement of the two definitions).

-------
monitoring receiver. The other receiver is connected to the local light source via another fiber optic
cable and thereby becomes, in effect, a light transmitter. The internal surfaces of the stainless steel
calibration cell are chemically polished to minimize surface reaction or adsorption of the pollutant
gases inside the cell, which could alter the standard concentrations.
The calibration apparatus used for the simultaneous calibration is shown schematically in Figure
2. The system was based primarily on two-stage, quantitative, dynamic dilution of NIST-traceable,
high-concentration standards of NO and S02 from compressed gas cylinders. Small flow rates (10
to 100 cm3/min) from either the NO or S02 standard gas cylinder were measured and controlled by
flow controller 1. Zero air could also be metered by this flow controller to maintain identical flow
conditions during periods when neither NO nor S02 was required. Zero air for the primary dilution
was metered and controlled by flow controllers 2 and 3. As shown, this dilution flow was divided
so that a portion of it passed through an ozone generator, which could be turned on or off electrical-
ly. These three flows then passed into reaction and mixing chambers (glass Kjeldahl-type mixing
bulbs) and into the DOAS optical calibration cell. The flow controllers were accurately calibrated
with a Gillian bubble flow standard, and the cell concentrations of NO or S02 were then calculated
from the dilution ratio, as Ccell = X Fj/tFj + F2 + F3). A bypass around the calibration cell
was also provided to allow a convenient way to check for possible 03 or NO losses in the cell by
comparing the concentrations at the inlet and outlet of the cell.
The effluent gas from the calibration cell was then secondarily diluted by passing a portion of
the flow (about 90 cm3/min) through flow meter 5 and diluting it with a flow (about 9500 cm3/min)
of zero air measured and controlled by flow controller 4. This diluted gas standard was passed
through another mixing device and into a vented manifold for connection of the conventional point
analyzers under calibration. The final diluted pollutant concentration standard at this output
manifold was calculated from the secondary dilution, as Cout = Ccell x F5/(F4 + F5). Excess
calibration gas effluent from the DOAS calibration cell was vented through a valve, which was
necessary to create backpressure to force the gas through flow meter 5 and the remainder of the
apparatus. This resulted in the pressurization of the calibration cell above atmospheric pressure and
necessitated a corresponding pressure correction in the DOAS measurement of the calibration cell
concentration. A manometer was connected at the outlet of the cell to measure the cell pressure for
the pressure correction. Originally, a mass flow controller was tried in place of flow meter 5 to
automatically regulate flow rate F5 for the secondary dilution. However, the considerably higher
backpressure required for proper flow controller operation greatly increased the potential for errors
due to leaks in the system and created a potential hazard from increased pressure in the glass
reaction and mixing chambers. Hence, flow rate F5 had to be regulated manually via the needle
valve on the cell vent.
Ozone standards were generated by a stable but uncalibrated, high-concentration ozone generator,
then assayed, after quantitative dilution to ambient-level concentrations, by a NIST-traceable ozone
transfer standard. The ozone generator utilized a mercury vapor lamp to generate ozone, similar to
ozone generators for ambient-level concentrations. To get the much higher concentrations needed,
the lamp was larger, and the air flow completely surrounded the lamp, which operated inside a
chamber several centimeters in diameter. A mechanical sleeve that could be slid over part of the
lamp was used to change the ozone concentration without changing the flow rate through the
generator. Since the lamp was turned off when ozone was not needed, the generator required about
20 minutes after being turned on to provide a stable ozone concentration. The ozone concentration
in the DOAS calibration cell was calculated as Ccell = Cout x (F4 + F5)/F5, where Cout is the
ozone concentration in the output manifold as assayed by the ozone transfer standard.
N02 standards were generated by gas phase titration (GPT).(5),(6) An NO standard concentra-
tion was established at a secondary dilution output of about 400 to 450 ppb, as measured by the
(previously calibrated) point NO analyzer. A small amount of the NO was converted to N02 by air
oxidation due to the high concentrations and fairly long residence time in the system. However, the
amount was usually small, of the order of a few ppb, as measured at the outlet manifold. The

-------
ozone generator was then turned on to generate sufficient 03 to titrate up to 80% - 85% of the NO
to N02. After the system restabilized, the remaining NO concentration was determined by the NO
point analyzer, and the N02 concentration was calculated as the difference between the original and
final concentrations of NO, with some adjustment to account for the NO2 present prior to addition
of the 03. As with 03, the concentration of N02 in the calibration cell was calculated as, Ccel] =
Cout x (F4 + ^5)^5. where Cout was the N02 concentration determined in the output manifold.
CALIBRATION OF THE DOAS INSTRUMENTS
The particular calibration cell used during our work had a measured internal length of 0.900
meters, while the installed DOAS monitoring path length was 407 meters. To be equivalent to
ambient concentrations, pollutant standard concentrations in the cell therefore would have to be
higher by a factor of 407/0.900 or 452. Ambient air pollutant monitors for S02, N02, and 03 are
normally calibrated over a concentration range of 0 to 0.5 ppm; thus, calibration standard concentra-
tions in the cell should be in the range of 0 to 226 ppm for an equivalent calibration range for the
open-path instruments. In our simultaneous calibration apparatus, the standard concentrations
obtainable were limited to about 50 ppm because of several constraints, including the ranges of the
flow meters that were available for the apparatus, the maximum flow rate capacity of the zero air
supply system, a minimum flow rate limit for the cell to maintain a reasonable flush time, and a
practical limit on the dilution ratio of the secondary dilution. A 50 ppm cell concentration was
equivalent to an ambient concentration of about 0.110 ppm, which covered most the ambient
concentration levels measured during the study. To provide consistency in the calibration results for
both point and open-path analyzers, all calibration cell concentrations were ultimately calculated as
equivalent ambient concentrations (effective concentrations), using Equation (1):
^cell j
m
where: Ceff = Effective cell concentration, equivalent to an average ambient concentration
over path length Lm, ppm
Ccei] = Actual cell concentration, ppm
Lc = Calibration cell length, m
Lm = Opsis monitoring path length, m
The Opsis DOAS instrument is programmed to correct its open-path pollutant measurements for
the ambient temperature and pressure. In our study, outdoor temperature was measured continuous-
ly with a temperature sensor, and the temperature corrections were based on this measured ambient
temperature. Ambient (barometric) pressure was not monitored with a sensor, so the average
barometric pressure for the site was manually entered into the Opsis software as a fixed value.
During calibration, the temperature in the calibration cell, which was located indoors, was likely to
be somewhat different than the outdoor ambient temperature, and the cell was slightly pressurized.
These temperature and pressure differences could be handled in several ways, either automatically
by the analyzer or manually, external to the analyzer. Choosing the former, we manually entered
the cell temperature and pressure values into the analyzer software and reprogrammed the analyzer
to use these special fixed values during the calibration rather than the average barometric pressure
and the temperature readings obtained from the outdoor sensor. The reason for this choice was that
the absorption cross section for S02 has a small temperature dependency, which the instrument takes
into account in its measurement calculations. For 03 and N02, which show no similar temperature
dependency, we could have just as well elected to make the corrections manually, external to the
analyzer. With the temperature and pressure corrections taken care of by the analyzer, the DOAS
instrument readings of the calibration cell concentrations needed only to be converted from jxg/m3
units to ppm, using Equation (2):

-------
where: C'
rdg
Urdg
273/298
F25
Measured concentration of the calibration cell contents, effective ppm
Cell concentration reading reported by the DOAS, ^g/m3 corrected to 0 °C
and 760 mm Hg and assuming monitoring path length L^,
Temperature adjustment from 0°C to 25°C
EPA units conversion factor at 25°C, ^g/m3/ppm;
F25(NO2) = 1880; F25(O3) = 1960; F25(SO^)=2620
The Opsis DOAS instrument reports all pollutant measurements in units of jtg/m3 corrected to 0°C
and 760 mm Hg. In converting these units to ppm, the 273/298 term is included so that the familiar
EPA unit conversion factors calculated at 25°C (Fy) can be used. (Alternatively, the unit
conversion factors could be recalculated at 0°C and the 273/298 term eliminated.) For convenience
during the actual calibration adjustment of the instrument, equation 2 can be rearranged and C'rdg
replaced with Ceff to calculate the correct reading to set the instrument, as in Equation (3):
= C


298
273
25
(3)
CALIBRATION OF THE CONVENTIONAL POINT ANALYZERS
Calibration of the conventional point analyzers was straight forward, using the calculated
standard concentrations at the output manifold. The 03 analyzer was calibrated directly with a
photometric primary ozone standard that had been recently intercompared with the NIST-traceable
Standard Reference Photometer maintained at EPA's Research Triangle Park laboratory. The
calibrated 03 analyzer was then used as a transfer standard to assay the ozone concentrations
generated and diluted by the calibration system.
CALIBRATION RESULTS
The results of the calibration of the various point and open-path analyzers used in the study
(after any calibration adjustment) are summarized in Table 1. The calibrations of the open-path
analyzers are given in terms of effective or equivalent ambient concentrations to make them readily
comparable with the point analyzers. All results show adequate calibration, with relatively minor
variations. Admittedly, the results of a one-time calibration of this nature would be expected to be
quite good, and no information was obtained to establish the stability and reproducibility of the
calibration system. Further testing of the apparatus will be required to determine those performance
characteristics.
As some confirmation of the good calibrations of both the open-path and point analyzers, Figure
3 presents preliminary ambient ozone measurement data from the three monitors for August 20,
1993. A high 03 peak occurred in the afternoon of that day, exercising the analyzers over a
considerable range of ambient concentrations. Ozone is usually well mixed in the ground-level
atmosphere, and the two types of 03 analyzers would be expected to agree well despite the
differences in the location between the two DOAS monitoring paths and the point analyzer inlet.
Figure 3 confirms good agreement among all three analyzers on this day over the substantial range
of concentrations, with minor discrepancies apparent only at the higher concentration levels.
Complete evaluation of the ambient pollutant measurements obtained during the study is currently in
progress, and the results of that evaluation will be reported subsequently.

-------
CONCLUSIONS
This work demonstrated the feasibility of simultaneous calibration of both open-path and
conventional point analyzers, using the same concentration working standards for both types of
instruments. Such simultaneous calibration is advantageous in situations where the two types of
monitoring instruments are under comparison. Although the two types of analyzers require
calibration standards in concentration ranges differing by up to 2.5 orders of magnitude, a calibra-
tion apparatus using a two stage (double) dilution system was successfully assembled and used in the
field study. An 03 generator of relatively simple design and using a conventional UV lamp was
shown to provide suitable 03 calibration concentrations for open-path instruments. Ozone concen-
trations up to 50 ppm were successfully generated, passed through an optical cell with minimal
losses, and quantitatively diluted for assay by a conventional 03 transfer standard at ambient-level
concentrations. These 03 concentrations also were successfully used to titrate NO to generate
accurately known standard concentrations of NO2, using the same GPT technique widely used to
generate N02 standard concentrations at ambient levels.
Calibration of the Opsis open-path DOAS air monitoring instrument presented only relatively
minor problems. Special calibration equipment is required, including an optical calibration cell,
calibration bench, two optical receivers, an auxiliary light source and power supply, additional fiber
optic cables, and a high-concentration ozone generator. Most of this equipment is available from
the instrument manufacturer. The high-concentration 03 generator may have to be assembled using
some custom-made components. Concentration standards up to several hundred ppm may be
required, depending on the ratio of the installed monitoring path length to the calibration cell length.
Conventional calibrators typically used for point analyzers may provide suitable dilution capability if
the pollutant concentration standards are in an appropriate range, although the 03 generators in such
calibrators probably will not provide high enough 03 concentrations for 03 calibrations or for N02
calibrations using GPT. Calibration of open-path analyzers for N02 using dilution of a high-
concentration N02 compressed gas standard may be a potentially advantageous alternative to GPT.
Techniques proposed for conducting accuracy audits and precision checks of open-path
analyzers for purposes of data quality assessment are very similar to the calibration techniques
discussed for calibration. In general, data quality assessment checks should test the entire monitor-
ing system in a configuration as close to the normal monitoring configuration as possible.
Therefore, one salient difference is the need to try to use the same outdoor light source (transmitter)
and optical path that are used for normal monitoring, rather than the alternate, local light source.
Two possibilities for using the normal monitoring light transmitter and receiver are potentially
available. In one, the same calibration bench, cell, and two small receivers are used, as shown in
Figure 1 for the calibration configuration. However, the small receiver operating as a transmitter
would be connected to the outdoor monitoring receiver rather to than the local light source. Thus,
this system would use the light from the normal light transmitter, receiver, and light path and would
thereby test the entire open-path monitoring system. The other possibility is to mount a special
optical cell directly onto the outdoor monitoring receiver, to remain in the optical path at all times.
Using this type of cell, precision or zero and span checks could be automated, greatly reducing the
need for visits to the open-path monitoring site. In both techniques, the test measurements would
have to be corrected because they would include both the prevailing ambient pollutant concentrations
as well as the test concentrations. There are other potential problems with these quality assessment
techniques, and further study and evaluation are needed before they are ready for adoption into the
EPA monitoring regulations or guidance.
ACKNOWLEDGMENTS
The authors gratefully acknowledge the contributions of Don Smith and Walt Buchholtz of the
Exxon Corporation Baytown (Texas) Plant for their invaluable assistance in making the field study
site available to us and in getting the field facilities set up; Opsis AB for furnishing an open-path

-------
analyzer and various calibration equipment needed for the study; Lee Ann Byrd, of EPA's Office of
Air Quality Planning and Standards for administrative planning and implementation of the field
study; and Avis Hines, AREAL, for intercomparison of our ozone photometric standard against the
NIST-traceable Standard Reference Photometer.
DISCLAIMER
The information in this document has been funded (wholly or in part) by the United States
Environmental Protection Agency under Contract No. 68-DO-0106 to ManTech Environmental. It
has been subjected to Agency review and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
REFERENCES
1.	U. S. Environmental Protection Agency, Title 40, Code of Federal Regulations, Part 53;
"Ambient Air Monitoring Reference and Equivalent Methods."
2.	U. S. Environmental Protection Agency, Title 40, Code of Federal Regulations, Part 58;
"Ambient Air Quality Surveillance."
3.	Proposed amendments to Title 40, Code of Federal Regulations, Part 58; publication pending.
Copies of the current draft are available from the first author.
4.	U. S. Environmental Protection Agency, Title 40, Code of Federal Regulations, Part 58,
Appendixes A, "Quality Assurance Requirements for State and Local Air Monitoring Stations
(SLAMS)" and Appendix B, "Quality Assurance Requirements for Prevention of Significant
Deterioration (PSD) Air Monitoring."
5.	U. S. Environmental Protection Agency, Title 40, Code of Federal Regulations, Part 50,
Appendix F; "Measurement Principle and Calibration Procedure for the Measurement of
Nitrogen Dioxide in the Atmosphere (Gas Phase Chemiluminescence)."
6.	Ellis, E. C.; Technical Assistance Document for the Chemiluminescence Measurement of
Nitrogen Dioxide, EPA-600/4-75-003; U. S. Environmental Protection Agency; Research
Triangle Park, NC, 1976.

-------
Table 1. Calibration data for both point and open-path analyzers.

Calibration
No.


Correlation
Analyzer
range*, ppm
points
Slope
Intercept
coefficient
Point analyzer, S02
0 - 0.500
6
0.980
0.002
0.99998
Open-path analyzer 1, S02
©
o
l
o
6
0.999
0.000
0.99997
Open-path analyzer 2, S02
0-0.110
6
0.997
0.000
0.99995
Point analyzer, 03
0 - 0.500
7
1.004
0.000
1.00000
Open-path analyzer 1, 03
0-0.110
7
0.997
0.001
0.99985
Open-path analyzer 2, 03
o
©
1
o
7
0.989
0.000
0.99991
Point analyzer, NO
0 - 0.500
6
0.986
0.000
0.99998
Point analyzer, N02
0 - 0.500
5
0.998
0.000
0.99960
Open-path analyzer 1, N02
0-0.110
5
1.013
0.000
0.99966
Open-path analyzer 2, N02
o
p
©
5
1.002
0.000
0.99981
•Effective (equivalent) ambient range for the open-path analyzers.

-------
Power
supply
Calibration gas ports
^ \
Auxiliary
light source

Calibration bench
Fiber-optic cable —
Receiver
Receiver
DOAS
analyzer
Figure 1. DOAS calibration system.
Opsiscal.drw

-------
ZERO AIR
SCRUBBER
PUMP
NO
1000 ppm
STANDARD
SO 2
1000 ppm
STANDARD
FLOW
CONTROLLER
No. 2
FLOW
CONTROLLER
No. 1
DOAS
CALI-
BRATION
CELL
VENT
MANOMETER
e-
3-WAY CELL BYPASS
REACTION MIXING VALVE
CHAMBER CHAMBER
FLOW
CONTROLLER
No. 3
OZONE
GENERATOR
FLOW
CONTROLLER
No. 4
FLOW
METER
No. 5
MIXING
CHAMBER
OUTPUT MANIFOLD
®= SHUTOFF VALVE
fc- NEEDLE VALVE
VENT"
n n iinnn
f
f
f


NOX

SO 2

O3
AMBIENT INSTRUMENTS
Figure 2. Simultaneous calibration apparatus.

-------
03- Hourly Data- DOAS and FRM
August 20,1993

180 j

160-

140-
n
120-
CL
Q.
100-
O
c
80-
o

O
60-
CO

O
40-

20-

o»

-20+

0
12
Time, Hour
e- FRM	-A- EPA DOAS	OPSIS DOAS
Figure 3. Comparison of ambient 03 measurements from point and open-path analyzers.

-------
TECHNICAL REPORT DATA
—
1. REPORT NO.
EPA/6G0/A-93/251
2.
:
4. TITLE AND SUBTITLE
Simultaneous Calibration of Open-path and Conventional Point
Monitors for Measuring Ambient Air Concentrations of Sulphur
Dioxide, Ozone, and Nitrogen Dioxide
5.REPORT DATE
October, 1993
6.PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Frank F. McElroy, Jimmie Hodgeson, Thomas A. Lumpkin, Kenneth
A. Rehme, Robert K. Stevens; Charles P. Conner (ManTech); Hans
Hallstadius (Opsis)
8.PERFORMING ORGANIZATION REPORT
NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
AREAL/ORD
Research Triangle Park, NC 27711
10.PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-DO-0106
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. EPA
AREAL/ORD
Research Triangle Park, NC 27711
13.TYPE OF REPORT AND PERIOD COVERED
Symposium paper
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
Prepared in cooperation with ManTech Environmental and Opsis AB, Furulund, Sweden.
16. ABSTRACT
A two-stage dilution system and an associated procedure to simultaneously calibrate both open-
path (long-path) and conventional point air monitors have been used successfully during a comparison test
study of open-path monitoring systems in Houston during August, 1993. Two open-path, differential
optical absorption spectrometers (DOAS) were calibrated using standard concentrations up to 50 ppm in a
0.9 meter stainless steel optical cell connected to each DOAS analyzer via fiber optic cables. The
calibration apparatus consisted of various flow controllers, flow meters, and mixing chambers to provide
accurate dynamic flow dilutions; a high-concentration ozone generator; a suitable reaction chamber for
gas phase titration (GPT) of NO to generate N02 standard concentrations; and an output manifold. High-
concentration standards for S02 and NO were obtained by the primary dilution of nominal 1000-ppm,
NIST-traceable concentration standards in compressed gas cylinders. N02 concentrations were generated
by GPT from NO concentrations. The ambient-level concentration standards were provided by
quantitative secondary dilution of the high concentration standards. Ozone standard concentrations were
generated by the high-concentration ozone generator and assayed, after secondary dilution to ambient-
level concentrations, by a commercially available UV photometric ozone analyzer used as a transfer
standard.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IDENTIFIERS/ OPEN ENDED TERMS
c.COSATI
Ambient monitoring, open-path monitoring
S02, 03, N02, calibration

18. DISTRIBUTION STATEMENT
Release unlimited
19. SECURITY CLASS (This Report)
Unclassified
21.NO. OF PAGES
11
20. SECURITY CLASS (This Page)
Unclassified
22. PRICE

-------