Near-MHz Laser Absorption Diagnostics for Non-Equilibrium Thermometry, Barometry, and Velocimetry of Nitric Oxide in Non-Equilibrium Environments

Laser absorption spectroscopy (LAS) is a useful technique for acquiring high-precision measurements of quantitative gas properties such as internal temperatures (e.g., rotational, vibrational, and electronic temperatures), static pressure, species mole fractions, and velocity. LAS is particularly attractive for applications in harsh environments characterized by large vibrations, high-temperatures, -pressures, and -velocities. Additionally, LAS can be performed in a non-intrusive manner and can provide more accurate measurements of gas properties versus traditional methods (e.g. thermocouples which have long response times). The recent advent of quantum-cascade lasers (QCLs) and interband-cascade lasers (ICLs) has made LAS more practical in the mid-IR where many IR-active species have high linestrengths in their fundamental vibration bands. These lasers can also be rapidly wavelength tuned by sending them a modulated current which enables up to MHz-rate measurements of gas properties.

LAS has been applied many times in experiments characterizing combustion and thermal and chemical kinetics. Regarding kinetics, shock tubes are useful tools for characterizing thermal and chemical kinetics as the properties of the gas behind the shocks are near uniform in planes parallel to the shock. In shock-tube experiments, LAS measurements can provide researchers with time histories of temperature and species partial pressure. Additionally, the mole fraction and pressure can be separately measured if the pressure is accurately known (e.g., from shock-jump relations or transducer measurements) or if the collisional broadening coefficients of the bath gas are known to high-accuracy. However, performing LAS is challenging in high-temperature and high-pressure shock-tube experiments due to blending of spectral lines, beam steering, and the need for high-speed measurements to resolve thermal and chemical transients. Regarding high-speed measurement rates, the lasers used in the mid-IR have reduced scan depths at high scan rates (i.e., they scan over a smaller spectral region). The reduced scan depth and blending of spectral lines makes determining the absorbance-free laser intensity challenging. As a result, spectrally resolved LAS experiments have been limited to moderate pressure (≈1 atm).

This dissertation presents a LAS diagnostic for measurements of temperature, pressure, and nitric oxide mole fraction at 500 kHz in high-temperature (up to 5500 K) and -pressure (up to 12 atm) shock-heated air in shock-tube experiments. Also presented are diagnostics for non-equilibrium thermometry (rotational and vibrational temperatures), nitric oxide (NO) partial pressure, and velocimetry in hypersonic flows at rates up to 500 kHz. CFD simulations of the expanding flow in reflected-shock tunnels were performed to compare the accuracy of vibrational relaxation rates for NO and to quantify the impact of the non-uniform line-of-sight on path-integrated LAS measurements.

The conditions behind the incident and reflected shocks in shock tube facilities can be estimated from shock-jump relations. However, depending on the initial gas composition and shock speed there can be a large degree of uncertainty about the thermochemical state of the gas behind the incident and reflected shocks. This uncertainty mainly results from finite-rate chemistry and thermal relaxation. To address these challenges, a high-speed quantum-cascade-laser-absorption-spectroscopy (QCLAS) diagnostic was developed to measure temperature, pressure, and NO mole fractions behind both incident and reflected shocks at high temperatures and high pressures, while being insensitive to blending of spectral lines, emission, and beam steering. Measurements were acquired in the Sandia National Laboratories (SNL) Hypersonic Shock Tunnel (HST) running in shock-tube mode. A distributed feedback (DFB) QCL centered near 1976 cm^-1 was used to scan across two transitions of NO in its ground electronic state (X²Π_1/2). A measurement rate of 500 kHz was achieved using a single QCL by: (1) performing current modulation through a bias-tee, (2) targeting closely spaced transitions with a large difference in lower-state energy, and (3) the modified free-induction-decay (m-FID) data processing technique was used to improve the diagnostic SNR and reduce impacts from blended spectral lines and beam steering. The diagnostic was validated in a mixture of 95% argon and 5% NO which was shock-heated to ≈2000 to 3700 K. The diagnostic was then applied to characterize shock-heated air at high temperatures (up to ≈5500 K) and high pressures (up to 12 atm) behind either incident or reflected shocks. The LAS measurements are compared to theoretical predictions from shock-jump relations, pressure sensors mounted in the wall of the shock tube, and equilibrium values of NO mole fraction. Additionally, a comparison between a measured NO mole fraction time history and a time-stepped homogeneous reactor simulation performed using two different chemical kinetics mechanisms is presented. In a related work, a similar laser near 1979 cm^-1 was used to measure the temperature and NO partial pressure at 1 MHz in high-temperature shock-heated air in the Purdue High-Pressure Shock Tube (HPST). The high measurement rate enabled the diagnostic to clearly resolve the evolution of temperature and NO partial pressure behind strong shock waves at temperatures near 5500 K and a pressure near 1 atm. Comparisons to simulated time histories using two chemical kinetics mechanisms were also made.

The freestream conditions in shock-tunnel facilities are typically characterized by a large degree of thermal and chemical non-equilibrium. Modern tools (e.g., CFD) can attempt to predict the freestream conditions. However, they rely on many assumptions about the state of the gas in the reservoir and in the chemical and thermal kinetics rates. Additionally, shock tunnels have large transients during start up and wind down and accurately measuring the quasi-steady test time is difficult. Thus, measurements of the non-equilibrium freestream are needed to determine the quasi-steady test time and to quantify the non-equilibrium thermodynamic state of the freestream. A QCLAS diagnostic for measuring the partial pressure and internal temperatures (rotational and vibrational) of NO in hypersonic flows was developed. Two QCLs were used to measure four transitions of NO near 1887 cm^-1 and 1930 cm^-1 at 25 or 100 kHz using scanned-wavelength direct absorption (scanned-DA). Initial validation tests were performed in the Purdue HPST using an NO-Ar mixture to confirm the accuracy of the diagnostic and spectroscopic model. The diagnostic was then applied to characterize the HST which was configured as a reflected-shock tunnel. In the HST, two flow cutters fixed to the test section walls were used to direct the measurement line-of-sight through the quasi-uniform core flow exiting the nozzle, thereby avoiding line-of-sight non-uniformities associated with the thick boundary layers at the nozzle exit. In the HST, tests were performed at conditions targeting air velocities of 3, 4, and 5 km/s where the rotational and vibrational temperatures of NO varied from 150 K to 850 K and the partial pressure of NO was near 20 Pa. Additionally, dry bottled air and humid room air were used as test gases to quantify the impact of water contamination on the vibrational non-equilibrium of NO. Comparisons with two CFD predictions using unique rate constants for the vibrational relaxation of NO are also presented. The vibrational non-equilibrium of NO was more pronounced for 3 km/s tests, and water contamination at around 1% mole fraction had a negligible impact on the thermal non-equilibrium of NO. Lastly, the measured rotational temperature of NO agreed well with CFD predictions, the measured partial pressure of NO was consistently above CFD predictions, and the vibrational temperature had moderate agreement with CFD predictions for 4 and 5 km/s tests, and poor agreement for 3 km/s tests.

The QCLAS diagnostic previously developed for measuring the rotational and vibrational temperatures and partial pressure of NO was then upgraded to include measurements of velocity in hypersonic flows at rates up to 500 kHz. Compared to the first test campaign, measuring hypersonic non-equilibrium flows in the HST, the measurement rate was five times higher, two different probe designs were used, and some CFD rates for vibration-vibration relaxation between N₂, O₂, and NO were corrected. Additionally, some vibration-translation relaxation rates that were missing in the CFD discussed in the previous work were included into this work's CFD simulations. Two fiber-coupled QCLs and 3D-printed optical probes were used to measure the aforementioned properties via a single retro-reflected beam and a single detector. This novel approach was taken to minimize spatial averaging and sensor complexity while still providing a self-referenced calibration-free velocity measurement. The diagnostic was applied in the freestream of the HST for 3, 4, and 5 km/s test conditions. The quasi-steady operation of the facility was characterized, transients are discussed, and comparisons with CFD predictions are given. Lastly, measurements from the two 3D-printed probe designs are compared and design guidelines for minimally invasive probes in hypersonic test facilities are discussed.