Multi-wavelength Differential Absorption Measurements of Chemical Species

Open Access
Author:
Brown, David Michael
Graduate Program:
Electrical Engineering
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
June 30, 2008
Committee Members:
  • C Russell Philbrick, Committee Chair
  • Zhiwen Liu, Committee Member
  • Kultegin Aydin, Committee Member
  • Hampton Nelson Shirer, Committee Member
Keywords:
  • remote sensing
  • supercontinuum
  • spectroscopy
  • differential absorption
  • lidar
Abstract:
The probability of accurate detection and quantification of airborne species is enhanced when several optical wavelengths are used to measure the differential absorption of molecular spectral features. Characterization of minor atmospheric constituents, biological hazards, and chemical plumes containing multiple species is difficult when using current approaches because of weak signatures and the use of a limited number of wavelengths used for identification. Current broadband systems such as Differential Optical Absorption Spectroscopy (DOAS) have either limitations for long-range propagation, or require transmitter power levels that are unsafe for operation in urban environments. Passive hyperspectral imaging systems that utilize absorption of solar scatter at visible and infrared wavelengths, or use absorption of background thermal emission, have been employed routinely for detection of airborne chemical species. Passive approaches have operational limitations at various ranges, or under adverse atmospheric conditions because the source intensity and spectrum is often an unknown variable. The work presented here describes a measurement approach that uses a known source of a low transmitted power level for an active system, while retaining the benefits of broadband and extremely long-path absorption operations. An optimized passive imaging system also is described that operates in the 3 to 4 μm window of the mid-infrared. Such active and passive instruments can be configured to optimize the detection of several hydrocarbon gases, as well as many other species of interest. Measurements have provided the incentive to develop algorithms for the calculations of atmospheric species concentrations using multiple wavelengths. These algorithms are used to prepare simulations and make comparisons with experimental results from absorption data of a supercontinuum laser source. The MODTRANTM model is used in preparing the simulations, and also in developing additional algorithms to select filters for use with a MWIR (midwave infrared) imager for detection of plumes of methane, propane, gasoline vapor, and diesel vapor. These simulations were prepared for system designs operating on a down-looking airborne platform. A data analysis algorithm for use with a hydrocarbon imaging system extracts regions of interest from the field-of-view for further analysis.