Date of Award

Fall 2014

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Mechanical Engineering

First Advisor

Robert Lucht

Committee Chair

Robert Lucht

Committee Member 1

Andrew Weiner

Committee Member 2

William Anderson

Committee Member 3

Timothy Zwier


The dissertation deals with the further development of chirped-probe-pulse femtosecond coherent anti-Stokes Raman spectroscopy (CPP fs-CARS) for applications of gas phase thermometry and extension to methane concentration measurements. The main effort has been to assess the usefulness and robustness of the technique in turbulent combustors of practical interest. A primary aim has been to evaluate the use of CPP fs-CARS for vibrational N2 thermometry in a highly turbulent environment. It has been suggested that due to the laser beam temporal overlap required for fs-CARS signal generation, the technique would be unsuccessful due to beam propagation retardation effects from density gradients in the flame. Chirping of the beams due to combustor windows has also been suggested to be a potentially significant complication. ^ First a series of experiments were performed in non-premixed free jet diffusion flames. Five kHz CPP fs-CARS thermometry of the hydrogen/air diffusion flame yielded temperatures between 300 and 2400 K, highlighting the Kelvin-Helmholtz instability caused by buoyant interactions of hot combustion products with the cold ambient air. These measurements demonstrated the excellent dynamic range of the CPP fs-CARS technique which can be further improved by the use of a dual detection channel scheme in which two spectrometers and two EMCCD cameras are used for split detection of the CARS signal. Measurements in a highly sooting methane/air diffusion flame were also successful. Radiation from the soot and/or absorbance of the laser beams did not impede detection of the CARS signal. ^ In collaboration with the German Aerospace Center (DLR) in Stuttgart, Germany, CPP fs-CARS measurements were performed for the first time in a highly turbulent combustor, the DLR Gas Turbine Model Combustor (GTMC). The combustor has significant levels of swirl and high levels of turbulence. Time resolved temperature measurements were performed at 73 locations in the GTMC for two different operating conditions. Every laser shot produced some resonant CARS signal; no significant loss of signal due to beam steering, pressure fluctuations, or shear layer density gradients was observed. Power spectral density analysis was performed on the CPP fs-CARS thermometry results yielding the characteristic thermo-acoustic pulsation and precessing vortex core frequency previously reported in the literature by DLR. The real limitation of CPP fs-CARS thermometry has proven to be that of detector dynamic range. Signal levels were observed to be approximately a factor of 1000 times higher in room air than at 2200 K. ^ The theoretical spectral fitting code for N2 was improved to include an instrument response function (IRF) to account for the spectral broadening due to the spectrometer and EMCCD detector. The inclusion of the IRF significantly improved the comparison of experimental and calculated CPP fs-CARS spectra. In addition, the laser parameters that are determined from calibration flames at different temperatures are now very consistent from flame to flame. This resolved a long-standing issue with our calibration procedure. Previously the laser parameters exhibited changes due to temperature. ^ In addition to improving CPP fs-CARS thermometry, the first simultaneous N2/CH4 CPP fs-CARS spectra were obtained. A theoretical code for CH4 concerned with accurately modeling the spectra of N2 and CH4 simultaneously from a set of room temperature pressure cell experiments was developed. This work expands the list of CPP fs-CARS molecules previously modeled, and points to interesting possibilities such as that of determining spectroscopic constants from theoretically fitting experimental spectra of more complicated hydrocarbon molecules and also different species.