REMOTE SENSING OF OZONE USING AN INFRARED
DIFFERENTIAL ABSORPTION SYSTEM
by
J. L. Guagliardo, R. T. Thompson, Jr.,
D. H. Bundy
Environmental Monitoring Systems Laboratory
and
M. H. Wells
Nevada Power Company
Las Vegas, Nevada
Project Officer
John A. Eckert
Advanced Monitoring Division
Environmental Monitoring Systems Laboratory
Las Vegas, Nevada 89114
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89114
-------
REMOTE SENSING OF OZONE USING AN INFRARED
DIFFERENTIAL ABSORPTION SYSTEM
by
J. L. Guagliardo, R. T. Thompson, Jr.,
D. H. Bundy
Environmental Monitoring Systems Laboratory
and
M. H. Wells
Nevada Power Company
Las Vegas, Nevada
Project Officer
John A. Eckert
Advanced Monitoring Division
Environmental Monitoring Systems Laboratory
Las Vegas, Nevada 89114
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89114
-------
DISCLAIMER
This report has been reviewed by the Environmental Monitoring Systems
Laboratory--Las Vegas, U.S. Environmental Protection Agency, and approved for
publication. 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.
NOTICE
The authors were affiliated with the Environmental Monitoring Systems
Laboratory when they worked on the infrared differential absorption system for
remote sensing of ozone; As of the date of publication, J. L. Guagliardo is
with Computer Genetics Corporation of Wakefield, Massachusetts; R. T.
Thompson, Jr., is on assignment to the U.S. Environmental Protection Agency
from Old Dominion University in Norfolk, Virginia; D. H. Bundy is with the
Environmental Monitoring Systems Laboratory in Las Vegas, Nevada; and M. H.
Wells is with the Nevada Power Company in Las Vegas, Nevada.
~
ii
-------
CONTENTS
Introduction. . . . . . .
. . . . . . . . . . .
. . . . . . . . . . .
Coincidence Measurements.
. . . . . . . . . . . . . . . . . . . . . .
Ground Testing.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Theory. .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Apparatus
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Results
. . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . .
Conclusions
. . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
References.
. . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
LIST OF FIGURES
1.
A comparison of experimental and theoretical values of ozone
absorption coefficient at several COz laser lines identified
by J-value for the 9.4 ~m band P-branch . . . . . . . . . .
......
2.
Schematic diagram of the system used for ground testing of
differential absorption using a topographic reflector. . . .
.....
3. Lasers, ozone test cells and receiver telescope used for the
,. . ground tests are shown in this photograph. '. . . . . . . . .
.....
4.
Description of a data set from a pair of laser firings where
NI, Nz is "the noise generated by the initial laser discharge,
POI' Poz is the power monitor response, and PI' Pz is the
measured strength of the signal backscattered from the target
for laser 1 and laser 2 respectively. . . . . . . . . .
.....
s.
A comparison of ozone concentration in test cells as measured
by differential IR absorption to that measured by UV absorption
. . . .
Hi
.. 1
1
3
3
4
6
8
8
2
5
6
7
7
-------
INTRODUCTION
It was concluded by Kildal and Byer1 that the resonance absorption
technique is a far more sensitive method for remote measurement of atmospheric
pollutants than either Raman or resonance backscatter methods. Byer and
Ga~buny2 proposed the use of topographic targets as retroreflectors and
stated that single-ended absorption methods with ranges of up to 10 km and
sensitivities of less than 0.01 ppm were possible. This fossibility has been
further analyzed by Seals and Bair.3 Menzies and Shumate and Murray, et
al.S, have proposed methods in which differential absorption and topographic'
reflection are utilized for the measurement of air pollutants.
The method we describe ultimately is one in which the output from the C02
TEA lasers, tuned to two different wavelengths, is pulsed downward from an
aircraft. The differential attenuation of the two pulses at different.
wavelengths reflected from a topographic target is then evaluated to infer the
integrated concentration of ozone in a column between the aircraft and the
local topographic feature. .
We will use the differential absorption ozone device to study oxidant/
precursor transport problems. Oxidant or oxidant precursor transport over
distances on the order of 100 miles or more has been observed in studies such
as those of Zeller et al.6 and Spicer et al.7 In order to develop
appropriate control strategies, the actual magnitude of the impact of the
transported oxidant and/or precursors must be quantified, and if possible, a
model developed to relate emissions to their distant impact. Data from the
earth-reflected differential absorption device, when coupled with windspeed~
can be used to determine ozone flux from one area to another, a measurement
which can only be approximated by other methods.
COINCIDENCE MEASUREMENTS
The prototype device was designed to utilize the high energy pulses
available from TEA lasers. Most studies of the wavelength coincidence b~tween
C02 laser lines and ozone absorption lines employ lower pressure (2 kPa to
4 kPa) lasers,8,9,10 whereas the TEA lasers operate at atmospheric pressure
(=100 kPa). Since the gain lines in the latter case are broader than those in
the former, coincidence measurements were undertaken to investigate the
suitability of the TEA lasers for use in a differential absorption system.
Although the results presented here cannot be applied to all C02 TEA lasers,
they are probably representative of what is to be expected.
The lasers employed in the laboratory and in ground field testing were
grating tuned flowing gas C02 lasers, with outputs of 250' millijoules (at
10.5 ~m) and a pulse duration of 250 ns, manufactured by GEN-TEC (model
1
-------
DDL-2SH). The ozone cells employed were made from 30-cm PVC tubing with
IRTRAN II windows 25 mm in diameter. The laser pulses propagated through the
cells were detected by GEN-TEC ED-lOa power monitors. An ozone/oxygen mixture
from a glass corona discharge cell was passed through the cells and the ozone
concentration was monitored by ultraviolet absorption.
Results of the wavelength coincidence measurements are presented in
Figure 1. The lasers were tuned over the P branch of the 9.4-pm C02 band.
The measured attenuation coefficients for ozone absorption obtained with the
TEA lasers are plotted for comparison with the calculated and measured values
reported by Patty et al.8 and Shewchun et al.9
16
"",,,~ Present data (experimental)
-"-0 Ref. 8 (experimental)
-----0 Ref. 9 (experimental)
-. Ref. 9 (theory)
CD
-
-
E 12
...
«J
-
E 10
CD
(.)
:= 8
CD
o
u
5 6
...
c.
~
"~ 4
.c
«
CD
c: 2
o
N
o
.
.
,
,
,
'~
o
8
10
12
14
16
18
20
22
24 26
28
30
32
34
36
Figure 1. A comparison of experimental and theoretical values of ozone
absorption coefficient at several C02 laser lines identified by
J-value for the 9.4 pm band P-branch.
2
-------
The measured attenuation coefficients were of th~ same order as those
reported for lower pressure lasers. However, the variation of 20 percent to
50 percent at P(14) and P(28) by different observers could be attributed to
variations in coincidence of a cavity mode controlled laser line position with
the broadened ozone absorption lines. It is anticipated that detailed
measurements would support the following argument.
There are several ozone absorption lines (=7 within to.1 cm-I) in the
vicinity of the C02 P(14) and P(28) laser lines.9,11 In general,- TEA lasers
will lase at several wavelengths under the gain profile as determined by the
laser cavity parameters. When the pressure-broadened absorption coefficient
of ozone due to all absorption lines in the vicinity of a given laser line is
compared with the pressure-broadened gain profile of the laser, one expects
the observed absorption coefficient to be strongly dependent upon the active
modes o~ the laser within that gain profile. It is this type of limited
wavelength coincidence which could lead to variations of the magnitude
observed in Figure 1.
It is possible to stabilize laser cavity modes by introducing a saturable
absorber12, injection locking13, or some other technique. An alternative
to mode locking the laser is to incorporate an ozone absorption cell14 in
the system which would use a part of each transmitted laser pulse «1 percent)
obtained from a beam splitter to constantly monitor the effective ozone
absorption coefficient.
In view of the apparently strong C02 J value dependence of the ozone
absorption, the anticipated stability enhancement available through mode
locking and the constant calibration updating available through the use of an
ancillary absorption cell, it would appear that TEA lasers are suitable for
use in an earth-reflected differential absorption system for monitoring ozone.
GROUND TESTING
THEORY
The backscattered radiant power from a Lambertian reflector is given
by:2
Pr(R)
=
A
Kp~2 0 exp[ -ZofRE;a (r)dr],
1TR
(1)
where K is the optical efficiency of the receiving telescope (including
overlap of laser field on telescope field), p is the reflectivity of the
diffuse target, A is the telescope area, R is the distance to the target, Po
is the transmitted power, and E;a(r) is the volume attenuation coefficient,
which is the sum of terms due to atmospheric scattering and absorption and to
pollutant absorption.
3
-------
If we consider measurements at two adjacent wavelengths and assume that
atmospheric attenuation, target reflectivity, telescope optica~ efficiency,
etc. are equal at both wavelen3ths, then the average concentration over
distance R of pollutant C is:l
[C)
=
P2 POI
(lnpl P02)/2RAac,
(2)
where Pn (n = 1,2) is the backscattered peak power at wavelength An and
POn is the transmitted peak power on wavelength An, and Aac is the
difference in the absorption coefficients of pollutant C measured at each of
the two wavelengths.
A real-time signal processing system based on Equation (2) was designed
and built16 with values for backscattered peak power and transmitted peak
power levels obtained from the system detectors. The R distance is obtained
by timing the ground return, and the absorption coefficients are obtained
through independent laboratory measurements. The processing system proved
unsuitable for use in the EMI (electromagnetic interference) noise environment
produced by the lasers, so detector response voltages were collected and
processed subsequent to the experiment.
APPARATUS
A block diagram of the apparatus is shown in Figure 2. The lasers were
identical to those used in the coincidence studies except that they were
modified to utilize spark gaps and trigger generators purchased from a
different manufacturer. The lasers and telescope assembly were isolated from
the signal processing equipment by fiber optic LED's, photodiodes and cable.
A pulse generator was used in a double pulse mode to trigger the waveform
digitizer and to trigger the lasers 20 ~s apart. The 20 ~s is slow enough for
double path round trip time for nonmultiplexed recording but fast compared to
atmospheric scintillation and aircraft groundspeed. The zero-order reflection
from the grating in each laser was detected by a fast (:4 MHz) pyroelectric
detector and used to measure the instantaneous power output of the laser for
each pulse. The return pulse was collected by a 32 cm f/4 Newtonian telescope
and focused onto a HgCdTe-LN2 cooled photoconductive detector. The four
pulses were digitized by a Biomation 8100 waveform digitizer and displayed on
an oscillographic strip chart recorder.
A 3O-cm cell was placed in front of each laser. The beams
through the cells were allowed to propagate over a distance of
meters to be diffusely reflected from the side of a building.
the apparatus mounted on a test cart for this purpose.
transmitted
about 300
Figure 3 shows
The same corona discharge ozone generator as used in the coincidence
measurements was used to alternately flow pure oxygen and oxygen/ozone
mixtures through the cells. The ratio of ozone to oxygen was controlled by
4
-------
PULSE
GENERATOR
TRIGGER
. WAVEFORM
DIGITIZER
RECORDER
I FIBER OPTICS
Ia_--------.
I
- - - - I
r- - - - - --,
I I I I
I . .#:
I~' f I
. FIBER OPTICS I
I I
. . I I
I DRIVER I
, I
I I
PHOTO DIODES I I
TRIGGER TRIGGER I I
GENERATOR GENERATOR I I
I I
I 4 MHz . I
LASER POWER LASER POWER I PYROELECTRIC HgCdTe . I
SUPPLY SUPPLY OmCTOR
I OmCTOR I
I I
.------------...
EMI-SHIELDED ENCLOSURE
LASER 1
.LASER 2
Figure 2. Schematic diagram of the system used for ground testing of
differential absorption using a topographic reflector.
5
-------
Figure 3. Lasers, ozone test cells and receiver
telescope used for the ground tests
are shown in this photograph.
varying the oxygen flow rate. The concentration of ozone Lp--Erach cell was
determined by measuring the change in ultraviolet absorption through a 10-cm
quartz cell placed in series between the infrared absorption cells. The
254-nm line of a mercury pen-ray lamp isolated by a Shoeffel monochromator was
used to measure the ozone concentration in the lo-cm quartz cell. Ambient
ozone concentration was measured with a Dasibi ozone monitor. -
RESULTS
A representative ozone measurement data set is shown in Figure 4. In
particular, Figure 4 displays the noise, -Nn, picked up from each laser
discharge, n, followed in approximately 8 microseconds by a signal, Pon,
proportional in amplitude to the laser output energy and, in approximately 3
more microseconds, by a signal, Pn, proportional in amplitude to light
backscattered from a topographic feature. The two lasers were tuned to
wavelengths of Al - 9.584 ~m (P24) and A2 - 9.501 ~m (P14).
i
A comparison of data obtained with the differential absorption system to
those obtained by the ultraviolet absorption method is shown in Figure 5.
Each data point in Figure 5 is computed from an average of 10 four-peak data
sets similAr to that shown in Figure 4. The cells were purged with oxygen
before and after each variation in oxygen flow rate (ozone concentration is,
within limits, inversely proportional to oxygen flow rate through the ozone
generator).
6
-------
P2
N 1 P1' ~
. ~ ' 'I
~ Po 1 ~ P02 :~
;-.-:: A ;'L'i /~. f\
. II I , \ ;'
,t, I \j . :. : \J ~
o : ~., T-' "~ t -...-.........
~ ;
I .. If
o 10 20 30
Time (~ sec.)
.... . . ~ --.-- ... . .. . . . .
Figure 4. Description of a data set from a pair of laser firings where
NI, N2 is the noise generated by the initial laser discharge, POI' P02
is the power monitor response, and PI, P2 is the measured strength
of the signal back-scattered from the target for laser 1 and
laser 2 respectively.
Q)
en
c:-
O~
Co .-
enC
Q) ~
a:Q)
>
~.-
0"'"
.....ca
(J-
Q) Q)
.....~
Q)-
C.
1
N2
.
-"
40
II
- -.
- 1 1
E
-
CO
o::t 9 8- --8 UV
,
o 0-0 If
..-
-
('I)
0 7
....
0
c:
0
.-
-
CO 5 ~
...
-
c: ~
Q) '-
(J
c:
0
U
2
3
4
5
Oxygen Flow Rate (ft 3/hr)
Figure 5. A comparison of ozone concentration in test cells as measured
by differential IR absorption to that measured by UV absorption.
7
-------
Dasibi readings indicated that ambient ozone levels averaged about 21 ppb
during the course of the experiment. Ambient ozone accounted for about 1
percent additional infrared absorption over the pathlength between"source/
receiver and topographic target.
The variation between each set of four peaks, i.e., the results of each
In[(PZPOl/PIPOZ)]' was only 5 percent without the cells in place. However,
the variation was 20 percent at best with the cells in place, even when the
cells were purged with nitrogen and the alignment maximized. This variation
is thought to be due to a combination of interference effects and the fact
that the telescope field of view was less than. the divergence of the lasers
(0.8 and 5 milliradians, respectively).
In addition, the lasers were known to operate in multiple high order
transverse modes, as was indicated by several output burn patterns obtained on
carbon paper. Modifications planned for the prototype should eliminate these
problems by establishing mode control on the lasers, combining both beams in
o
order to make the system coaxial, using a beam expander to control divergence
and by using Brewster angled windows in the cell to eliminate interference
effects.
CONCLUSIONS
Ground tests of the initial system design have disclosed certain
limitations, which have been eliminated for the next modification. In spite
of serious limitations imposed by a strong EMI field, an existence of many
active laser modes, a lack of divergence control of the laser and a cumbersome
test environment for conducting controlled tests, the system was able to
demonstrate a correlation between infrared measurements of ozone and reliable
ultraviolet measurements as indicated in Figure 5. The measurements reported
in Figure 5 show the system capable of measuring ozone concentrations from 300
to 800 ppm in a 3D-cm cell when operating 300 meters from a topographic
target. This would correspond to average concentrations of 0.15 to 0.4 ppm
over the 300 meters. Additional tests are planned for the redesigned system
to determine the range dependence and concentration dependence of sensitivity
for an earth-reflected differential absorption system.
REFERENCES
1.
H. Kildal and R. L. Byer, Proc.
IEEE 59, 1644 (1971).
2.
R. L. Byer and M. Garbuny, Appl. Opt. 12, 1496 (1973).
3.
R. K. Seals, Jr. and C. H. Bair, I.S.A. Paper 71-1083 from: The Second
Joint Conference on Sensing of Environmental Pollutants, Washington,
D. C., December 1973.
8
-------
4.
R. T. Menzies and M. S. Shumate, Appl. Opt. 15, 2025 (1976).
"
5.
E. R. Murray, J. E. van der Laan, and J. G. Hawley, Appl. Opt. 15, 3140
(1976).
6.
K. F. Zeller, R. B. Evans, C. K. Fitzsimmons, and G. W. Siple, J.
Geophys. Res. 82, 5879 (1977).
7.
C. W. Spicer, D. W. Joseph, and G. F. Ward, EPA 600/3-76-109, November
1976.
8.
R. R. Patty, G. M.. Russwurm, W. A. McClenny, and D. R. Morgan, Appl. Opt.
13, 2850 (1974).
9.
J. Shewchun, B. K. Garside, E. A. Ballik, C. C. Y. Kwan, M. M.
Elsherbiny, G.. Hogenkamp, and A. Kazandjian, Appl. Opt. 15, 340 (1976).
10.
W. Schnell and G. Fischer, Appl. Opt. 14, 2058 (1975).
11.
A. Barbe, C. Secroun, P. Jouve, N. Monnanteuil, J. C. Depannemaecker, B.
Duterage, J. Bellet, and P. Pinson, J. Molec. Spect. 64, 343 (1977).
12.
A. Nurmikko, T. A. DeTemple, and S. E. Schwarz, Appl. Phys. Lett. 18, 130
(1971).
13.
R. B. Gibson, A. Javan, and K. Boyer, Appl. Phys. Lett. 32, 726 (1978).
14.
J. M. Hoell, Jr., W. R. Wade, and R. T. Thompson Jr., Proceedings of the
International Conference on Environmental Sensing and Assessment, Las
Vegas, Nevada, Paper 10-6. September 14-15, 1975.
15.
J. L. Guagliardo and D. H. Bundy, Proc. of Intn. Tele. Conf. 10, 414
(1974). .
16.
J. L. Guagliardo and D. H. Bundy, Proc. of 7th Intn. Laser Radar Conf.,
Palo Alto, California 1975.
9
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 12. 3. RECIPIENT'S ACCESSION NO.
4. TITL.E ANO SUBTITL.E 5. REPORT DATE i
REMOTE SENSING OF OZONE USING AN INFRARED
DIFFERENTIAL ABSORPTION SYSTEM 5. PERFORMING ORGANIZATION CODE
7. AUTHOR(S) 8. PERFORMING ORGANIZATION REpORT NO.
J. L. Guagliardo, R. T. Thompson, Jr., D. H. Bundy
and M. H. Wells.
9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGRAM EL.EMENT NO.
Environmental Monitoring Systems Laboratory lAD883 ..
Office of Research and Development 11. CONTRACT/GRANT NO.
U.S. Environmental Protection Agency
Las Vegas, NV 89114
12. SPONSORING AGENCY NAME AND ADDRESS 13. TYPE OF REPORT AND PERIOD COVERED
U.S. Environmental Protection Agency--Las Vegas, NV Tech Report - FY-78
Office of Research and Development 14. SPONSORING AGENCY CODE
Environmental Monitoring Syste~s Laboratory
Las Vegas, NV 89114 EPA/600/07
15. SUPPL.EMENTARY NOTES
15. ABSTRACT
A prototype airborne downlooking infrared differential absorption system using C02
TEA (transverse excited atmospheric) lasers is described. The system uses two
wavelengths and topographic reflection to measure the integrated column concentration
of ozone between the laser source/receiver and a noncooperative target. A comparison'
is made between ozone absorption coefficients measured with TEA lasers and values
reported from other sources. Ground tests utilized two 30-cm long ozone-filled test.
cells, one in each laser path. A correlation was observed between measurements of TEA
laser pulses backscattered from a building and ultraviolet determination of ozone
concentration in the cells.
-
17. KEY WORDS AND DOCUMENT ANAL.YSIS
a. DESCRIPTORS b.IDENTIFIERS/OPEN ENDED TERMS C. COSA TI Field/Group
C02 lasers 20E
Ozone 07B
Air pollution monitors 14G
Electromagnetic absorption infrared 20C
18..DISTRIBUTION STATEMENT 19. SECURITY CL.ASS (This Report) 21. NO. OF PAGES
UNCLASSIFIED
RELEASE TO PUBLIC - 20. SECURITY CL.ASS (This page) 22. PRICE
UNCLASSIFIED
EPA Form 2220-1 (Rev. 4-77)
PREVIOUS EOITION IS OBSOLETE
. . --"-
------- |