An apparatus based on line-of-sight resonant absorption of ultraviolet light yields measurement data from which one can calculate the concentrations of nitric oxide (NO) and of hydroxyl radicals (OH) in a laboratory flat flame at a pressure up to 30 atm ≈3 MPa). The basic measurement principle is distinct from the principles of laser-induced fluorescence and other laser diagnostic techniques; hence, the data generated by this apparatus could provide independent verification of data from laser-based instruments.
Line-of-sight resonant absorption of ultraviolet light has been in use at least since 1976; however, until now, it had not been experimentally verified to be useful for determining NO and OH concentrations at pressures above 2 atm (≈0.2 MPa). The high-pressure-flame regime of the present development was chosen because it is representative of conditions at the exit of advanced combustors. The design of the apparatus and the measurement principle are simple enough that it should be possible to develop the apparatus into a portable optoelectronic instrument that could be set up in combustor or engine test cells.
In the apparatus (see figure) a water-cooled hollow-cathode lamp generates ultraviolet light, which is collimated and directed through a test cell that contains the flame to be probed. The portion of the collimated beam that remains after passing through the test cell then enters a fiber-optic cable, through which it travels to the entrance slit of a computer-controlled grating spectrometer equipped with a linear array of 1,024 photodiodes at its output plane. The spectrometer measures the spectrum of light that has passed through the test cell, at wavelengths from 208 to 280 nm (for NO) or 300 to 330 nm (for OH) with a spectral resolution of 0.3 nm.
For measuring the concentration of NO, a glow discharge in flowing air at a pressure of 5 to 10 torr (≈0.7 to 1.3 kPa) is created in the hollow-cathode lamp. The light from this discharge includes discrete emission spectral lines generated by the recombination of O and N with N2 and O2. Thus, the emitted light includes components that resonate with the absorbing species of interest. Although one could use a continuum light source (at least in principle), resonant absorption offers the advantage of a greater signal-to-noise ratio. For measuring the concentration of OH, the lamp is operated in a similar manner except that the glow discharge is created in an atmosphere of argon saturated with water.
The spectrum of transmitted light is well approximated by a mathematical model of transmissivity as a function of wavelength, the temperature of the flame, the length of the optical path through the flame, and the concentration and optical-absorption characteristics of the gas species (NO or OH) of interest. The model was developed to nearly its present form in 1980 and was refined, for use in the present application, by incorporating terms to account for shifting and pressure broadening of spectral lines of both NO and OH.
In use, an assumed value of the concentration of the species of interest is inserted in the model of transmissivity along with the known values of the other quantities and the model is convolved with a spectrometer-slit function to obtain a predicted spectrometer output. This computation is repeated, if necessary, using different values of assumed concentration. The concentration of the species of interest is then deemed to equal whichever value of assumed concentration results in the best match between the predicted and actual spectrometer outputs.
In tests, the apparatus was used to measure spectra during operation of the cell with flames at two different fuel/air mixture ratios (fuel concentrations of 0.98× and 1.3× stoichiometric) and at several pressures from 1 to 30 atm. (≈0.1 to 3 MPa). Concentrations of NO and OH were measured independently by a conventional gas-sampling technique. The absorption spectra measured by the apparatus agreed, within 25 percent, with absorption spectra predicted by the mathematical model. Continuum absorption in hot oxygen was found not to be strong enough to interfere significantly in interpretation of the data on absorption in NO.
This work was done by D. S. Liscinsky, B. A. Knight, and J. A. Shirley of United Technologies Research Center for Glenn Research Center. For further information, access the Technical Support Package (TSP) free on-line at www.nasatech.com/tsp under the Test and Measurement category.
Inquiries concerning rights for the commercial use of this invention should be addressed to
NASA Glenn Research Center
Commercial Technology Office
Attn: Steve Fedor
Mail Stop 4 —8
21000 Brookpark Road
Refer to LEW-16914.