Experimental and Numerical Investigation of Ablation Kinetics

The objective of this research was to perform a detailed investigation of pyrolysis gas and plasma interactions for a commonly studied ablative thermal protection material known commonly as PICA. Experiments were conducted at University of Vermont (UVM) to provide benchmark data on key interaction species at well-documented plasma test conditions. Numerical studies were carried out at University of Michigan (UMi) to aid in the design of the experiments and to compare the experimental results with predictions based on current finite-rate chemical kinetic models. Four different experimental activities were conducted during this investigation, starting with tests of PICA and FiberForm (the PICA substrate) in different plasmas to characterize the spectral emission and identify the key species resulting from pyrolysis-plasma interaction as a function of time from heating onset. Concurrently with these initial tests, two new gas-injection probes were developed to allow steady-state pyrolysis gas flow simulation. The third activity entailed using the better injection probe to identify the most suitable gases and flow rates to inject into the plasma stream to reproduce the spectral signature of PICA. Tests were conducted with the optimal injection conditions to examine the spatial distribution of the key emitting species as they interacted with different plasma test gases under steady injection and heating conditions. Finally, Laser-Induced Fluorescence (LIF) measurements of key species were made to assess ground state populations of CO and H. Numerical simulations performed by UMi were compared with the experimental results for two injection cases, and these comparisons provided a critical assessment of the finite-rate kinetics that pertain to these cases. Specific accomplishments from the experimental investigation included: 1) demonstrating that buffering the a reactive plasma gas with argon slowed pyrolysis and substrate burning significantly, while apparently having minimal impact on the chemical kinetics; 2) after an initial period of intense emission from carbon bearing species (C2-Swan, CN-Violet, CN-Red), the spectral signature for air plasma tests included emission from OH, NH, CN and H from the pyrolysis gas and plasma interaction, along with Na and Cs as impurities from substrate material; 3) toward the end of the PICA tests the spectral signature asymptotically approached that of the substrate, FiberForm; 4) emission from the cyanogen radical provides a reliable visual indication Page 2 – Revised 02/2018 of the pyrolysis gas penetration depth; 5) steady-state injection of a H2-CO2 gas mixture through a porous graphite plug reproduces the spectrally resolved emission from PICA pyrolysis in the interaction zone near the surface; 6) local thermodynamic equilibrium analysis of the Ar (and H) atomic transitions provided reliable temperature values for characterizing plasma stream conditions; and 7) the numerical simulations used CFD (LeMANS) combined with a radiative transport model (NEQAIR v14.0) to show good agreement with the experimental emission spectra for the atomic species (Ar and H), but over-predicted the CN and OH molecular band emission. The difference was attributed to the chemical kinetic rates used in the simulation. All of the experimental data have been archived for NASA researchers and others interested in testing computational models of pyrolysis and plasma interaction. Through multiple conference proceedings, the UVM design and the development of the gas injection probe to replicate steadystate pyrolysis has been made available to the aero-thermal testing community. Researchers at the University of Texas have recently used the UVM approach for a similar study [1 (p. 99)].