LIDAR FOR ATMOSPHERIC TRACE GAS DETECTION

Jonathan Rall
NASA/Goddard Space Flight Center
Greenbelt, Maryland

I. Introduction

Trace gases, unlike the constituent gases oxygen and nitrogen, which are permanent members of the Earth's atmosphere, are highly variable in the atmosphere both temporally and spatially. The constituent gases are well mixed and therefore their abundance can be calculated accurately from measurement of atmospheric pressure alone. Trace gas concentration, however, must be measured directly using either in-situ methods or remote sensing methods to record their distribution and fluctuations. Certain trace gases, primarily water vapor, CO2, CH4 and to a lesser extent O3 and N2O, have strong absorption features in the near to mid infrared (ir). Variations in the atmospheric concentration of these trace gases may lead to significant global climatological change by altering the absorption and re-radiation of long wave terrestrial radiation (Greenhouse Effect). Long term global monitoring of these trace gases is needed to further improve modeling of global change processes. In addition some trace gases are considered pollutants (O3, SO2, HNO3) which are harmful to mankind and therefore need to be monitored over populated areas. Table 1 shows some of the common trace gases and the center wavelengths of their prominent near and mid ir absorption features. These absorption lines can be probed with narrow spectral laser radiation using the differential absorption lidar technique. Open-path laser sensing using light detection and ranging (lidar) instruments enable accurate remote measurements of atmospheric trace gas concentration over ranges of several meters to tens of kilometers.

Table 1
Trace GasBand Center (mm)Laser
Water Vapor0.72, 0.82, 0.94, 1.1, 1.4, 1.9 AlGaAs, InGaAs, OPO
CO2 1.3, 2.0, 4.3 InGaAs, Pb-salt, OPO
CH42.4 InGaAs, Pb-salt, OPO
O30.29Excimer, Solid-state
N2O2-4, 8-9 InGaAs, Pb-salt, OPO
SO20.3, 5, 8-9Excimer, Solid-state, Pb-salt

II. Status of Technology

Lidar measurements of atmospheric trace gases have historically employed two basic techniques, elastic scattering differential absorption lidar and inelastic scattering Raman lidar. Fluorescence lidar while capable of making atmospheric trace gas measurements has primarily been used in ocean surface and vegetation scattering measurements and more recently chemical agent identification for military battle field applications. This article will further explore the dial and Raman lidar techniques only.

III. Lidar Techniques

eqn
Several techniques have been used to demonstrate the DIAL technique including wavelength modulation and fm spectroscopy. Wavelength modulation usually employs a low frequency dither (5 kHz) and derivative spectroscopy techniques to lock the laser to the molecular absorption line for the on-line measurement and then an additional measurement is made off-line. The ratio of on-line to off-line LIDAR measurements is used to infer the absorber concentration using the dial equation above. Fm spectroscopy utilizes a high frequency small signal modulation applied to the laser. The modulation frequency must be greater than the spectral width of the absorption feature in order to probe the molecular absorption line with one of the FM sidebands. As one sideband is tuned through an absorption line, a balanced receiver generates a beat frequency as the upper and lower sidebands become unbalanced. The quadrature component of the detected signal is proportional to the differential absorption. FM spectroscopy techniques using tunable diode lasers provides the most sensitive means to measure trace gas concentration using differential absorption LIDAR1.

The Raman LIDAR technique involves detecting transmitted laser radiation which has been shifted in wavelength due to interaction with the scattering molecule. This wavelength shift, or Stokes shift, is equal in energy to a vibrational-rotational or rotational transition in the scattering molecule. The backscattered power of the wavelength shifted signal is proportional to the concentration of scattering molecules and inversely proportional to l-4. Thus, the primary advantage of Raman LIDAR of DIAL is that it offers a direct measure of species concentration or mixing ratio by comparing the Raman signal of the scatterer to the Raman signal of N2 or O2. However, Raman scattering is a very weak process and the signal can be two to four orders of magnitude weaker than the elastic backscattered signal. Also, the weak scattering cross section typically limits Raman lidar to nighttime measurements at ranges of less than 10 km. To increase the Raman signal and make daytime measure ments, high power lidar systems have been developed to operate at wavelengths from 248.5 to 268.5 nm. Unfortunately at these wavelengths absorption by molecular oxygen and ozone can attenuate the transmitted beam and solar irradiance can obscure the backscattered Raman signal.

IV. Laser Transmitters

Semiconductor lasers are the smallest, most efficient lasers available. Their light output can be directly modulated using bias current and they can be tuned in wavelength using both temperature and bias current. Two fundamentally different types of laser diodes are currently available 1) high power, single mode lasers available from 635 - 1650 nm using III-V semiconductor compounds: AlGaInP, AlGaAs, InGaAs, InGaAsP and 2) lead-salt lasers operating between 3 - 30 mm. Semiconductor laser hold great promise in the realization of small, affordable dial instruments for trace gas monitoring.

Excimer lasers (KrF, XeF) have been used extensively in Raman LIDAR instruments for the measurement of atmospheric water vapor, Mie/Rayleigh backscatter ratio, N2, and O2. However, excimer lasers are expensive and involve hazardous gases which will likely preclude their widespread use in trace gas monitoring instruments.

Solid state lasers have been used directly as transmitters for both DIAL and Raman LIDARS and as optical pump sources for optical parametric oscillators (OPO), Raman cells, dye lasers, and second, third, and fourth harmonic generation using non-linear crystals. Several different solid state lasers including, Nd:YAG, Ti:sapphire, Alexandrite have been used in fielded LIDAR systems. The Lidar Technology Experiment (LITE) which recently flew on the space shuttle employed a flash-lamp pumped Nd:YAG lasers which yielded the first lidar measurements from orbit. However, solid state lasers tend to be complex, inefficient, and expensive which will likely prevent their widespread use as ground based, airborne, and spaceborne atmospheric trace gas monitors. OPO's provide tunable infrared monochromatic light which is useful for near to mid IR trace gas sensing but require a solid state laser to optically pump them. Efficient harmonic generation is possible and has been used with Nd:YAG lasers to produce UV radiation for Raman LIDAR applications.

V. Receiver Narrowing / Optical Bandpass Filters

All LIDAR systems require spectrally selective components in their receivers, whether bandpass, blocking, or notch filters, to suppress background noise. Improvements in dichroic filters have led to 1 Angstrom FWHM bandpass filters with >50% peak transmission available in the visible through the near IR. Temperature controlled etalons can be used in series with bandpass filters to further narrow the optical pass band but at the expense of reduced transmission and additional electronics. Volume holographic narrow band optical filters2 are available in the visible spectrum with 0.125 Angstroms FWHM and >10% transmission but so far have not been demonstrated in the near IR. Finally, Faraday anomalous dispersion optical filters have been demonstrated at 780 nm with a 1 GHz bandwidth and 63% transmission3. However, Faraday filters require an atomic resonance or transition to work which will limit their efficacy for trace gas measurements.

VI. Detectors

For detection of weak signals (i.e. photon-counting) from 200 nm through 1mm, photomultipliers have been preferred for their high-gain and low noise characteristics. Recent advances in night vision goggles for the military have improved pmt technology for the near IR including higher quantum efficiency (10-20%) GaAs photocathodes and more rugged vacuum APDanodes. Development of Geiger-mode silicon APD's have provided a choice of photon-counting detectors for ground based and airborne LIDAR systems working between 400 and 900 nm. These photon counting modules exhibit the high quantum efficiency (>50% @ 700 nm) of silicon, the high gain of APD's, and low dark counts (<50 counts/sec typical) of PMT's. However, their ability to count fast is limited to ~ 10 MHz due to the need to quench or recharge the diode junction and their performance degrades if exposed to radiation levels typical of space. Development of photon-counting InGaAs apd modules is underway but performance is not well documented and no commercial devices are available. However successful development of photon-counting InGaAs APD's would enable sensitive trace gas detection using the well established fiber-optic communication lasers around 1.3 and 1.5 mm.

VII. Recommendations for Further Activity

1. Development of photon counting detectors beyond 1 mm wavelength using InGaAs or Germanium APD's.

2. Development of high Q.E. GaAs photocathode, vacuum apd photomultipliers.

3. Continued development of high power AlGaAs and InGaAs lasers (1-10 Watt cw) using master oscillator / power amplifier designs (MOPA)

4. Continued development of OPA and non-linear crystals for efficient generation of tunable IR and second, third, and fourth harmonic generation.

5. Continued development of ultra-narrow band, high throughput, optical filters.

REFERENCES

1. L. G. Wang, H. Risis, C. Carlisle, and T. Gallagher, "Comparison of approaches to modulation spectroscopy with GaAlAs semiconductor lasers: application to water vapor," Appl. Opt., 27 (10) 2071-2077, (1988).

2. G. A. Rakuljic and V. Leyva, "Volume holographic narrow-band optical filter," Opt. Lett., 18 (6) 459-461, (1993).

3. D. J. Dick and T. M. Shay, "Ultrahigh-noise rejection optical filter," Opt. Lett., 16 (11), 867-869, (1991).

[ TOP OF PAGE ] [ BACK TO TECH BRIEFS ] [ BACK TO IFT ]