United States
Environmental Protection
Agency
Environmental Sciences Research r
Laboratory "V,
Research Triangle Park NC 27711 '/li
Research and Development
EPA-600/S3-81-026 June 1981
Project Summary
Atmospheric Measurements of
Trace Pollutants: Long Path
Fourier Transform
Infrared Spectroscopy
Ernesto C. Tuazon, Arthur M. Winer, Richard A. Graham, and James N
Pitts, Jr.
A four-year study to measure at-
mospheric concentrations of trace
pollutants by kilometer pathlength
Fourier transform infrared (FT-1R)
absorption spectroscopy was con-
ducted at two sites in the California
South Coast Air Basin (CSCAB) from
1976 to 1979. During 1976 and 1977
the FT-IR facility was operated in
Riverside, California, and provided
valuable benchmark data. These in-
cluded the first reported direct spec-
troscopic detection of trace levels of
nitric acid and formaldehyde in the
polluted troposphere, and confirmation
of the suspected prevalence of high
(NH3) concentrations (>100 ppb in
some instances) in the Riverside area
which originate primarily from upwind
agricultural sources.
During the last two years of the
study (1978 and 1979), the FT-IR
facility was operated in Claremont,
California, a mid-basin site chosen to
characterize episodes closer to the
Pasadena-Azusa area which presently
experiences the highest smog levels in
the CSCAB. The 1978 study focused
on extended monitoring periods and
succeeded in recording a "classic"
stagnant air episode with pollutant
carryover and progressively increasing
oxidant levels during the week of
October 9-13, 1978. The FT-IR mea-
surements included a 38-hour contin-
uous monitoring period for the more
sever episodes of October 12 and 13,
1978. The most intense episode of
that week (October 13) was charac-
terized by peak pollutant levels of 454
ppb O3, 37 ppb PAN. 49 ppb HNO3,
19 ppb HCOOH, and 71 ppb HCHO.
Research in 1979 consisted of col-
laborative studies to validate newly
developed analytical and sampling
techniques for measuring HNO3, NHs.
and HCHO. In these studies the km
pathlength FT-IR spectroscopic tech-
nique served as the reference method.
Among the studies conducted was a
major EPA-sponsored field program
to compare current analytical methods
for gaseous HNOa and particulate
nitrate.
Ambient air data which we have
obtained for trace pollutant concen-
trations over a four-year period, to-
gether with the results of the collabo-
rative studies to validate new methods
of analysis of "non-criteria" pollu-
tants, will provide a critically needed
data base for stringently testing the
chemical kinetic submodels of the
current generation of widely used
urban airshed models.
This Project Summary was developed
by EPA's Environmental Sciences
Research Laboratory, Research Tri-
angle Park, NC, to announce key
-------
findings of the research ^project that is
fully documented in a separate report
of the same title (see Project Report
ordering information at back).
Introduction
During the past decade, the major
efforts of federal, state and local agen-
cies in the acquisition of an air monitor-
ing data base have largely focused on
the regulated or "criteria" pollutants,
i.e., ozone (Os), nitrogen dioxide (NOz),
total hydrocarbons, sulfur dioxide (SOz)
and total suspended participates (TSP).
At the same time, laboratory and smog
chamber studies, as well as more com-
prehensive computer kinetic models of
photochemical air pollution have in-
creasingly stressed the importance of a
number of trace atmospheric species
and their roles in photochemical air
pollution. These important species
include formaldehyde (HCHO), formic
acid (HCOOH), nitric acid (HN03), nitrous
acid (MONO), and hydrogen peroxide
(H202>, as well as ammonia (NH3). Until
recently, comparatively little progress
had been made in characterizing the
concentration ranges and the temporal
and geographical distributions of these
significant, but presently unregulated,
air pollutants. This lack of data con-
cerning the ambient concentrations of
the so-called "non-criteria" pollutants
was striking in view of their potential
importance to atmospheric chemists
and modelers, to control officials, and to
others who are concerned with impacts
on agriculture and human health.
In response to this need, in 1976 we
initiated measurements of such species
under an Environmental Protection
Agency (EPA) sponsored program em-
ploying a kilometer pathlength FT-IR
spectrometer originally designed and
assembled by Dr. P.L Hanstof the EPA-
Research Triangle Park (RTP) Labora-
tories. This system provided an in-situ
part per billion (ppb) detection capability
for many oxygenated and nitrogenous
pollutants for which reliable, alternative
analytical methods were unavailable.
Instrumentation and Methods
The eight-mirror multiple reflection
optics employed in this study consisted
of four rectangular, in-focus (nesting)
mirrors, separated by 22.5 m from four
out-of-focus (collecting) mirrors. All
mirrors were optically polished and
gold-coated for maximum reflectivity in
the infrared.
The cell consisted of a sectional
rectangular aluminum frame (total
dimensions: 0.81 x 0.84 x 23 m) with a
lining of 50-micron thick FEP Teflon
film. Figure 1 depicts the multiple-
reflection cell coupled to the Fourier
transform spectrometer system, which
was housed in a 3.7 x 3.7 m air-condi-
tioned building. Interfaced with the long
path optics was a Digilab Model 296
Michelson IR interferometer (resolution
>0.5 cm"1) and an associated data
system. A magnified image of the source
element was focused at the entrance
aperture of the cell and the beam exiting
from the cell was sent to either of two
Ozone
Analyzer
NO-N02 Analyzer
Michelson
Interferometer
Figure 1. Kilometer pathlength Fourier transform infrared spectrometer.
2
\
-------
liquid-Na cooled detectors: a photovoltaic
InSb detector for the 2000 to 4000 cm"1
range or a photoconductive HgCdTe
detector for the 600 to 2000 cm"1
region.
The sampling procedure consisted of
drawing air into the cell at a rate of 330
liters sec"1 for a minimum of four
minutes before the start of an interfer-
ometer scan. This corresponded to a
displacement of the previous air sample
by a minimum of five volumes of fresh
sample. Total pathlengths of 900 and
1080 meters and a resolution of 0.5 crrf1
(16384 digitized points per interfer-
ogram) were routinely employed. Thirty
or 40 interferograms were usually co-
added to enhance the signal-to-noise
ratio in the interferogram, and thus in
the computer spectrum.
Results
Data from monitoring activities during
1976 and 1977 at Riverside, a site
approximately 60 miles east and 8-12
hours downwind of the primary emission
sources in the CSCAB, have been pre-
sented in literature and are only briefly
discussed in the report. Results included
the first direct spectroscopic detection
of nitric acid and formaldehyde in the
atmosphere, and confirmation of the
previously suspected prevalence of
relatively high levels of NH3 in Riverside.
Detailed data from 5-8 hour monitoring
periods for a total of 3 episode days in
October 1976 and 10 episode days in
July-October 1977 are reported, along
with supplemental air quality data for
NO, NO2, CO and nonmethane hydro-
carbons.
In the summer of 1978, the FT-IR
facility was moved from the UC-Riverside
campus to Claremont, California, and
was installed on the roof of the Jacobs
Science Center of Harvey Mudd College.
Although moderate pollution episodes
were recorded for various days during
the summer months, this report focuses
on the week of October 9-13, 1978
during which extended periods of air
monitoring were carried out, including
data collection for a continuous 38-hour
period for the more severe smog epi-
sodes on October 12 and 13. This week-
long stagnant air episode was charac-
terized by a monotonic rise in the daily
peak 03 readings from 0.16 ppm on
Monday, October 9, to 0.45 ppm on
Friday, October 13 and the measured
concentrations of Os. PAN, HN03,
HCOOH and HCHO are reported as a
function of time for each day of this
episode. The maximum concentrations
observed each day are shown in Table 1.
Morning HCHO concentrations in
Claremont were observed to be in the
range of 20-40 ppb. A slight decrease in
these concentrations occurred around
the noon hours and prior to the peak Os
readings. Well-defined HCHO peaks
corresponding to Os maxima were ob-
served during the more severe episodes
of Thursday and Friday, October 12 and
13, with HCHO reaching a peak concen-
tration of 71 ppb at about 1600 hr on the
13th. The observations in Claremont
suggest that the high levels of HCHO
reported in earlier studies in the 1950's
and 1960's may indeed be approached
under conditions of intense photochem-
ical air pollution. Formaldehyde was
also found to persist fat concentrations
of ~20 ppb) during the night, an impor-
tant observation in modeling the role of
formaldehyde as an early morning
catalyst for smog formation.
Formic acid levels were generally low,
typically less than 10 ppb even during
moderate episodes. This agrees with
the majority of our measurements in
1977 at Riverside. The highest concen-
tration of HCOOH observed at Claremont
was 19 ppb.
Although we had accumulated sub-
stantial nitric acid data during the 1977
study in Riverside, these profiles con-
sisted only of upper limits during many
periods (i.e., concentrations were below
the detection limit) and thus a detailed
characterization of HN03 concentra-
tions as a function of photochemical
activity was not possible. In the Clare-
mont study the rise in HN03 concentra-
tion was strongly coincident with the
increase in oxidant levels (e.g., 03 and
PAN). The highest HNOs concentration
measured was 49 ppb which coincided
with the Os concentration of 454 ppb
recorded on October 13.
The average NH3 concentration mea-
sured in Claremont during the week of
October 9-13 was~8 ppb. An examina-
tion of our data obtained during the
months of July and August, 1978, also
indicated that the NH3 level in Claremont
is generally less than 10 ppb. Thus, the
NH3 concentrations measured by FT-IR
spectroscopy in Claremont were ap-
proximately five times lower than those
found in Riverside during our 1976 and
1977 studies.
The recognition of artifactual errors in
the widely accepted method for mea-
suring nitrates in the atmosphere by
sample collection on fiberglass filters
provided impetus for researchers to
develop new, more-sensitive and selec-
tive techniques for the measurement of
both gaseous and particulate nitrates. In
1978 the EPA developed a plan to
conduct a field experiment that would
bring together appropriate research
groups to compare these measurement
methods for nitrate and HN03. It was
recognized at that time that the kilometer
pathlength.FT-IR spectroscopy could
serve as a standard for nitric acid
determination since this method involves
an in-situ, nondestructive technique,
and identification is made unambigu-
ously by recognition of characteristic
spectral features.
During the period August 27-Septem-
ber 3,1979, a field study was conducted
at the site of our long-path FT-IR facility
in Claremont, California. While detailed
results of the FT-IR monitoring are
provided in this report, a full interpreta-
tion of the nitric acid intercomparison
data, along with its implications for the
data simultaneously gathered for partic-
ulate nitrate is the subject of a separate
EPA-Battelle report.
Attempts made to detect other atmos-
pheric species resulted in the following
upper limits to their ambient concentra-
tions for the periods and locations
involved in this study: nitrous acid
(HONO), 10 ppb; hydrogen peroxide
(H202), 40 ppb; peroxynitric acid
(HO2NOz), 8 ppb; peroxyalkyl nitrates
(R02N02), 6 ppb; hydrochloric acid
(HCL), 8 ppb; acrolein (CH2=CHCHO), 15
ppb; ketene (CH2=C=0), 6 ppb.
Table 1 . Daily Maximum Concentrations (ppb) Observed in Claremont. CAby Kilo-
meter Pathlength FT-IR Spectroscopy
Date 03 PAN NH3 HNOa HCOOH HCHO
October 9
10
11
12
13
163
227
280
360
454
6
14
13
22
37
11
23
13
13
25
18
28
30
29
49
5
5
7
17
19
23
30
31
52
71
H U.S GOVERNMENT PRINTING OFFICE, 1961-757-012/7140
-------
Conclusions
This program has demonstrated the
utility of long pathlength FT-IR spectro-
scopy for quantitatively measuring
parts-per-billion levels of a number of
nitrogenous and oxygenated compounds
in ambient air.
Our results verified that HNOs, and
not PAN, is the major nitrogen-contain-
ing product of photochemical smog and
showed that HMOs levels correlate
negatively with prevailing NHa concen-
trations. The data obtained in Riverside
included numerous measurements of
simultaneous NH3 and HNOa concentra-
tions; and these results support the
hypothesis that paniculate NhUNOs is in
equilibrium with its gas-phase precur-
sors NH3 and HMOs. In areas such as
Claremont where average NH3 levels
are low (<10 ppb), the HMOs concentra-
tion during the peak of a smog episode
may be expected to amount to approxi-
mately 10% of the Oa concentration or
more.
Our data established that ambient
concentrations of HCOOH are low and
will rarely be above 20 ppb even during
severe smog episodes. This is approxi-
mately 4-10 times lower than earlier
values reported from wet chemical and
long pathlength FT-IR measurements.
On the other hand, our results are
consistent with early measurements by
wet chemical techniques showing that
high concentrations (—100 ppb) of
HCHO may prevail during an intense
siege of photochemical air pollution
associated with high vehicular activity.
The value of long pathlength FT-IR
spectroscopy as an analytical tool for air
pollution studies should not be limited
to the characterization of species found
in photochemical smog. Given the
increasing awareness of the potential
hazards posed by atmospheric releases
of toxic materials and their degradation
products, the kilometer pathlength FT-
IR technique should continue to serve
as a highly developed method for atmos-
pheric monitoring.
Ernesto C. Tuazon, Arthur M. Winer. Richard A. Graham, and James N. Pitts, Jr.,
are with the Statewide Air Pollution Research Center, University of California,
Riverside, CA 92521.
B. W. Gay, Jr., is the EPA Project Officer (see below).
The complete report, entitled "Atmospheric Measurements of Trace Pollutants:
Long Path Fourier Transform Infrared Spectroscopy, "(Order No. PB 81-179 848;
Cost: $11.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Sciences Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
PS 0000329
U S ENVIR PROTECTION AGENCi
REGION 5 LIBRARY
230 S DEARBORN STREET
CHICAGO IL 60604
1
------- |