June 1978 / Vol. 2, No. 6 / OPTICS LETTERS
163
Remote SO2 measurements at 4 ,umwith a continuously
tunable source
Richard A. Baumgartner and Robert L. Byer
Applied Physics Department, Edward L. Ginzton Laboratory of Physics, Stanford University, Stanford, California 94305
Received February 27, 1978
Remote atmospheric measurements of S02 were performed using a differential absorption lidar with a continuously
A comparison of injected gas con-
tunable LiNbO 3 parametric oscillator and amplifier source in the 4.0-lim region.
centration in a remotely located sample chamber with the lidar measurements through the chamber showed measurement agreement within experimental error and a detection sensitivity for SO2 of 0.9 part in 106km.
lidar measurements of SO2 using a continuously tunable
parametric oscillator and amplifier source in the 4.0-,gm
cm Hz"/2 W'/2 , is located at the focal point. Atmospheric probing over a full hemisphere is possible with
the fully steerable altazimuth telescope mount. The
region. A sample chamber in the optical path of the
tunable source can be tuned on and off a molecular
This Letter presents results of remote atmospheric
lidar simulated a source of SO2, allowing a direct com-
parison of lidar measurements with in situ gas-concentration values.
In the present measurements we obtained a sensitivity of 0.9 part in 106 km with a transmitted pulse
energy of 0.6 mJ at 10 pulses per second for 500 averaged
pulses. Continuous tunability over the 1.4- to 4.0-gm
range offers flexibility in selecting the optimum infrared
spectral region and wavelength for pollutant measurements. This broad tuning range includes absorption
bands of many pollutants, such as CO, C0 2 , CH4 , and
other hydrocarbons and N20, in addition to SO2. The
details and background for differential-absorption and
Raman-lidar techniques for remote monitoring are
covered in review articles.'-3 The original measurement of SO2 was made by Inaba,' with several subsequent measurements in the ultraviolet.4 ' 5
Because of the importance of SO2 emission from
stationary sources, especially from combustion of coal
resonance for a differential absorption measurement.
Figure 1 shows a schematic
2 TO I
REDUCTION
TELESCOPE
infrared lidar is an important
The full daylight operation with total eye safety is an
advantage over the ultraviolet lidar system previously
described by Grant et al. 4 The capability of providing
more positive spectral identification of a gaseous constituent with the continuous tuning is an advantage over
discretely tuned infrared laser measurement.7 For
example, numerous hydrocarbons with CH stretch
resonances in the 3.3-,gmregion may be identified by
scanning appropriate parts of their spectra. The continuous tunability of the present system offers the
possibility of optimizing the measurement strategy for
a given pollutant molecule by adjusting the absorption
cross section and minimizing interferences.
The lidar is a monostatic system that uses topographic targets as retroreflectors. The beam is transmitted coaxially with a 40-cm-diameter Newtonian
r-
receiver telescope. A liquid-nitrogen-cooled InSb detector 1 mm in diameter, with a detectivity of 2 X 10"1
Fig. 1.
0146-9592/78/0600-0163$0.50/0
BEAM TO
ROOF
t MOUNTED
TELESCOPE
BEAM
SPLITTER
or fuel oil, a real-time remote measurement of SO2 with
measurement capability.6
diagram of the Nd:
YAG-pumped LiNbO3 parametric oscillator and amplifier that provides the pulsed infrared radiation over
the tuning range of 1.4-4.0 gm.8 The pump source is
a 6.35-mm-diameter Nd:YAG unstable-resonator oscillator. Half of the 250-mJ output energy pumps the
LiNbO3 angle-phase-matched parametric oscillator
through a Faraday rotator isolator. The remaining
125-mJ output energy is amplified in a 9.35-mm-diameter YAG amplifier and pumps a second 6-cm-long
LiNbO3 crystal parametric amplifier. The parametric
oscillator is a singly resonant' oscillator type with a
double-pass pump wave. The parametric signal wave
is resonated in a 13-cm-length cavity consisting of a
600-line/mm grating for wavelength tuning control and
-1-----
2 TO I
EXPANSION
TELESCOPEC
COMBINING
I BEAM SPLITTER
I
I
I
I
L
UNSTABLE
RESONATOR
YAG LASER
ROTATOR
ISOLATOR
PARAMETRIC
OSCILLATOR
75 % OUTPUT
COUPLING
I
I
1
/. A FARADAY
II I F Z
f: D:
3/8" YAG
AMPLIFIER
I
2 TO I BEAM
EXPANSION
TELESCOPE
I
I
PARAMETRIC I
AMPLIFIER
CRYSTAL
1.06 jm
BEAM DUMP
- …- -
Schematic diagram of the tunable source for the in-
frared lidar measurement.
©1978, Optical Society of America
164
OPTICS LETTERS / Vol. 2, No. 6 / June 1978
The sample chamber simulating the SO2 source was
located 120 m from the roof-mounted telescope. Oak
trees, located an additional 75 m beyond the sample
chamber, served as a topographic target. The chamber,
borrowed from SRI International, has a path length of
2.44 m along the lidar direction, with low-density
polyethylene windows mounted at a 15° angle to minimize direct backreflections into the receiver. Measurements were made by adding a known concentration
of SO2 to the chamber and measuring the concentration
by scanning the lidar several times over the frequency
range from 2945 to 2513 cm-', using the parametric
0
LU
I
0
V)
U)
SOURCE
U)
zn
EQUIVALENT S02
CONCENTRATION
6 PPM KM
4 CmI
n
2497
2501
2505
V IN Cm
2509
2513
1
idler wave with a 4-cm- 1 bandwidth.
I-)
z
100
0
LU
tiLl
U)
,-SOURCE ,
LI)
z
EQUIVALENT SO2
CONCENTRATION
13.7 PPM KM
4Cr"
I-
2495
2499
2503
V IN cm
2507
2511
I
Fig. 2. Examples of lidar frequency scans through a remotely
located sample chamber for two values of SO 2 concentrations
versus frequency in cm-'. Normalization scans are also
shown.
a 75%transmittance output coupler. The parametric
signal-wave linewidth of approximately
3.5 cm-' is
determined by the grating dispersion and crystal gain
linewidth. The parametric oscillatortuning and return
data plotting is under the control of a minicomputer,
permitting continuous scanning over a 10-cm-1 spectral
region in about 4 min. The pulsed parametric tunable
source is advantageous for remote monitoring with a
monostatic lidar since the high peak power is essential
to overcoming the typical attenuation encountered for
the returning lidar signals.
The return signal for this topographic target measurement was amplified and processed by a boxcar integrator. A second detector provided a normalization
signal to account for shot-to-shot source variation.
Recent advances in parametric oscillator performance
show that a substantial increase in output energy should
L
-
CONCENTRATIONMEASUREMENT
OF SO 2 USING LIDAR
-
z
Vs
CONCENTRATION OF S02
IN SAMPLE CHAMBER
a:
The
(n
:
signals were stored and the spectrum versus frequency
plotted under minicomputer control.
The concentration of SO2 in the chamber is determined from the expression,9
n (Poff) -2
For normalization
the chamber was purged of SO2, and scanned providing
the 100%reference values shown in Fig. 2. The error
bars illustrating standard-deviation values were averages over three or four frequency scans and represent
the overall response of the SO2 spectrum and the system
linewidth at each frequency point.
Figure 3 shows the excellent agreement between the
sample-chamber gas concentration values and the lidar
measurements over a large concentration range. The
concentration values were derived by averaging frequency points over a region equivalent to the source
linewidths (darkened regions in Fig. 2), on both a normalized on-resonance and off-resonance region. A
system sensitivity may be derived by extrapolating to
a unity signal-to-noise ratio. For the SO2 measurements the value was 0.9 part in 106 km, as indicated
earlier. The linewidth value of 3.5 cm-' for this measurement is quite appropriate since the S0 2-band
linewidth is pressure broadened by the other atmospheric constituents.
30
a.
LU
-j
2
aoff) dr,
10
2
where Poff and Pon represent the return power received
off and on the molecular resonance, respectively, L is
the optical path length in the chamber, Pso2 is the
0
-J
2
partial pressure of SO 2 in the chamber, and aon- aoff
is the on-resonance to off-resonance absorption coefficient. The absorption coefficient difference was obtained by a White-cell scan of SO2 at 1-Torr partial
a.
Un
ar
pressure over a length of 675 cm. The measured value
of aon - aoff = 0.363 (atm cm)-' compares quite favor-
ably with a value of 0.372 (atm cm)-' obtained in the
same wavelength region by averaging the data of Pine
and Moulton' 0 in a similar procedure used for the
sample chamber data.
3
LU
1
Fig. 3.
I, .I
, ,, ,I
3
10
PARTS PER MILLION KM
(GAS SYSTEM MEASUREMENT)
30
A comparison of lidar measurements with in situ gas
concentration for the remotely located sample chamber.
June 1978 / Vol. 2, No. 6 / OPTICS LETTERS
be possible. A transmitted energy of 15 mJ per pulse
at 4.0 gm would improve the sensitivity of the SO2
measurement to 30 parts in 109 km for 500 averaged
pulses, which is more than adequate sensitivity for
source monitoring.
The authors would like to acknowledge the suggestions and loan of the sample chamber by E. Murray and
E. K. Proctor of SRI International and the funding
provided by the Electric Power Research Institute and
by NSF/RANN through SRI International.
References
1. H. Inaba and T. Kobayashi, Opto-electronics 4,101-123
(1972).
2. V. E. Derr, ed., Remote Sensing of the Troposphere (U.
S. Government Printing Office, Washington, D. C., 1972),
NOAA Cat. # C55.602, T75.
3. R. L. Byer, "Review of remote air pollution measure-
ment," Opt. Quantum Electron. 7,147-177 (1975).
4. W. B. Grant and R. D. Hake, Jr., "Calibrated remote
165
measurements of SO2 and 03 using atmospheric backscatter," J. Appl. Phys. 46, 3019-3023 (1975).
5. J. M. Hoell, Jr., W. R. Wade, and R. T. Thompson, Jr.,
"Remote sensing of atmospheric SO2 using the differential
absorption lidar technique," Proceedings of the International Conference on Environmental Sensing and Assessment, Las Vegas, Nevada, September 1975.
6. R. L. Byer and E. R. Murray, "Remote monitoring techniques," in Handbook of Air Pollution Analysis, R. Perry
and R. J. Young, eds. (Chapman and Hall, London, 1977),
pp. 406-447.
7. E. R. Murray, J. E. van der Lann, and J. G. Hawley,
"Remote measurement of HCl, CH4 , and N2 0 using a
single-ended chemical-laser lidar system," Appl. Opt. 15,
3140-3148 (1976).
8. R. L. Byer and R. L. Herbst, "Parametric oscillation and
mixing," in Nonlinear Infrared Generation, Y. R. Shen,
ed. (Springer-Verlag, Berlin, 1977), pp. 81-133.
9. R. L. Byer and M. Garbuny, "Pollutant detection by
absorption using Mie scattering and topographic targets
for retroreflectors,"
Appl. Opt. 12, 1496-1505 (1973).
10. A. S. Pine and P. F. Moulton, "Doppler-limited and atmospheric spectra of the 4-gm v, + V3combination band
of SO 2 ," J. Mol. Spectrosc. 64, 15-30 (1977).