You are here

Automated trace gas molecular analyzer using rotational spectroscopy


OBJECTIVE: Construct an automated molecular spectrometer operating in the mm/submm wavelength spectral region that can analyze a low-pressure gas mixture and identify the composition and concentration of its constituents. DESCRIPTION: There is a long-standing and growing need to measure the composition and concentration of mixtures of unknown gases. Among the many examples that may be cited, of principal interest to AMRDEC is the need to monitor the health and degradation of rapidly aged missile propellant formulations by measuring the composition and concentration of out-gassed decomposition by-products. Other examples of interest to AMRDEC include the monitoring of vapor from toxic industrial chemicals (TICs) [1] and the analysis of the molecular composition and concentration of rocket exhaust. Low-pressure gas phase molecular rotational spectroscopy in the millimeter/submillimeter (mm/submm) wavelength, terahertz frequency region has been well established for more than 60 years.[2],[3] Known to produce spectra of polar molecules with incredible recognition specificity (essentially no false positives) and sensitivity (at the parts per trillion level), this technique has been limited to laboratory scale application because of the immaturity of the source and detector technology. As these technologies mature, the opportunity exists to develop a spectrometer as easy to use as Fourier transform infrared (FTIR) spectrometers but with much greater recognition specificity and sensitivity.[4],[5],[6] For this instrument, it may be assumed that the gas, which may have been pre-concentrated, is provided in a sample container whose volume is<1 liter with pressures of<1 Torr after obscuring atmospheric gases (N2, O2, Ar) have been removed through some combination of pristine collection, cryopumping, and/or release from an appropriate sorbent. A portion of this gas sample will be measured by the spectrometer at sufficiently low pressure that line center frequencies and strengths may be measured precisely and quantitatively (i.e. lines are Doppler-broadened with linewidths of ~1 MHz). The remaining challenges facing the construction of such a diagnostic instrument are the need to (1) measure and (2) analyze the spectra reliably in order to identify the constituent gas composition and quantify their respective concentrations. It must be assumed that the constituent gases are unknown prior to measurement. Therefore, it is not sufficient that the spectrometer only measure and identify lines in its library, it must measure all lines over its scan region, identify lines not in its library, and ascertain whether these indicate the presence of additional unknown species. By no means must the spectrometer scan the entire mm/submm wavelength spectral region, but must scan far enough with enough resolution that molecules of military interest (including the TICs with permanent dipole moments) or variants thereof may be detected, analyzed, recognized, and quantified with confidence. To build such an instrument, two advances are needed. First, a high-resolution spectrometer operating in the mm/submm wavelength region must be constructed that can perform sufficiently broad spectroscopic sweeps with sufficiently high signal-to-noise ratios to achieve detection thresholds well below 1 ppb. Second, a sophisticated analysis algorithm must be developed that uses a library of previously measured spectral features (including spectra from naturally abundant isotopomers and thermally populated vibrational levels) to identify the composition and concentration of constituent molecular gases. The algorithm must also provide a spectroscopic listing of unassigned measured lines that will require subsequent analysis to identify. This analytical tool must be expandable as new data becomes available, preferably by measurements performed with the spectrometer and subsequently"learned"through a calibrated library-building feature that may also be used to populate the initial library. The user output will be a simple table that displays the measured gases and their concentrations, plus an auxiliary table that lists the unassigned lines. PHASE I: Design a spectrometer capable of ascertaining the composition and concentration of a low pressure gas mixture using measured and/or calculated molecular rotational spectra in the mm/submm wavelength region. It may be assumed that the user has previously"purified"the gas by removing ambient atmospheric gases N2, O2, and Ar and that the spectrometer will be able to measure the pressure of the fraction of the sample introduced into the diagnostic cell. The deliverable at the end of Phase I is the completed design of the spectrometer and the demonstration of a working gas analysis algorithm to convert measured spectra into quantitative estimates of constituent composition and concentration with false recognition probability<10^-6 and partial pressure sensitivity<1 mTorr, respectively. The spectral region covered, the anticipated molecular library, the anticipated acquisition time, the anticipated analysis time, and the detection threshold must all be estimated. PHASE II: Based on the Phase I design, construct and deliver to AMRDEC a spectrometer capable of ascertaining the composition and concentration of a low pressure gas mixture using measured molecular rotational spectra in the mm/submm wavelength region. The deliverable at the end of Phase II is a working spectrometer and gas analysis algorithm that provides quantitative estimates of constituent composition and concentration with false recognition probability<10^-6 and sensitivity<1 ppb, respectively. As part of the gas analysis algorithm, a large and expandable library of spectra must be provided as well as a means for adding new spectra through measurements in the spectrometer, predictions from on-line databases, and/or calculations based on published rotational constants. PHASE III: An improved version of this spectrometer with an automated gas sampling/purifying stage and low unit cost would find tremendous military and civilian markets in areas that include non-destructive test, treaty verification, early warning sentinel for toxic gas release, chemical health monitoring, chemical quantitative analysis, medical screening, detection of illicit drug production, location of hazardous waste sites, atmospheric monitoring of plant emissions, and border/port patrol, among many others. Therefore, the goal of a Phase III effort is to reduce size, weight, power, and cost while increasing the instrument"s autonomy and the size of its library to cover an ever-increasing realm of applications. REFERENCES: [1] TIC list: [2] C.H. Townes and A.L. Schawlow,"Microwave Spectroscopy,"Dover (New York, 1955). [3] W. Gordy and R.L. Cook,"Microwave Molecular Spectroscopy, 3rd edition"Wiley (New York, 1984). [4] S. Albert et al., Analytical Chemistry 70, 719A (1998). [5] I.R. Medvedev et al., Optics Letters 35, 1533 (2010). [6] DARPA Mission Adaptable Chemical Sensor solicitation and final report: & mode=form & id=511848ae4527165e81738d8197bdec7e & tab=core & _cview=0,
US Flag An Official Website of the United States Government