Uncovering the Chemical Processes during Atmospheric Entry of a Carbon/Phenolic Ablator: Laboratory Studies by In Situ Mass Spectrometric and Molecular Beam Techniques

Thermal protection systems (TPSs) are required to shield spacecraft from the high temperatures generated during atmospheric entry. At the start of this grant, analysis and design of new ablative heat shields relied on material response models based on 50-year-old methodologies. The Minton Group at Montana State University has used molecular beam and mass spectrometric techniques to collect fundamental data on the time- and temperature-dependent decomposition processes of a carbon/phenolic heat shield material and on the oxidation mechanisms of a carbon fiber network material at high temperatures. In addition, a finite rate oxygen-carbon ablation model has been developed in collaboration with modelers. The new data are being used by NASA collaborators to validate non-equilibrium models whose aim is to simulate the response of heat-shield materials during atmospheric entry of spacecraft. A key objective of the project was to develop the methodology to measure relative molar yields of pyrolysis products under conditions where (1) time and temperature are independent variables and (2) there is no interaction of products with each other, with the hot sample surface, or with any container walls after they leave the decomposing material. Our approach was to use a mass spectrometer to detect the pyrolysis products in situ as they desorbed from a decomposing sample that was heated in high vacuum and to determine relative molar yields of the products by a validated interpretation of the mass spectral data. This new methodology was used to improve our understanding of the decomposition mechanisms of the specific carbon/phenolic composite, PICA, as this material has been chosen for several NASA missions and is an important model material for ongoing theoretical efforts to describe the in-depth temperature response of ablative TPS materials. Another focus material was the flight heritage heat shield material, Avcoat, which is being used on NASA’s Orion crew capsule. The work began with a study of PICA pyrolysis on an existing molecular beam apparatus, and then another apparatus was repurposed and set up for dedicated studies of PICA pyrolysis. Finally, an entirely new apparatus was set up in a new lab for studies of the pyrolysis of a key resin component of Avcoat, epoxy-novolac (based on D.E.N. 438), and a composite material prepared with epoxy-novolac and resole phenolic microballoons. The thermal decomposition of PICA was investigated with the objective of measuring molar yields of pyrolysis products at heating rates that are relevant to Mars Science Lab flight conditions. Specifically, PICA was pyrolyzed at heating rates of 3.1, 6.1, 12.7 and 25 °C s-1, with sample temperatures that spanned the range 100-1200 °C. Samples were restively heated in a differentially pumped vacuum chamber and pyrolysis gases were measured with a carefully calibrated mass spectrometer. The relative molar yields of 14 pyrolysis gases were obtained in conjunction with mass loss measurements. These measurements allowed for the calculation of absolute molar and mass yields as a function of heating rate, as well as the simulation of TGA curves. Pyrolysis product yields changed as a function of heating rate, and this behavior was attributed to two mechanisms that compete during the initial stages of thermal decomposition. The first reaction involves the condensation of hydroxyl functional groups to form carbon-carbon bonds or ether functional groups. The second reaction involves the breakdown of the methylene bridge between phenol moieties to yield phenol and its derivatives. The second decomposition reaction competes more effectively at higher heating rates. Epoxy-novolac resin samples were prepared to mimic the epoxy-novolac component of Avcoat. D.E.N. 438 was mixed with methyl-5-norbornene-2,3-dicarboxylic anhydride (NMA) as a crosslinking agent and N,N-dimethylbenzylamine (BDMA) as the catalyst. The ratio of NMA to epoxy-novolac determines the extent of crosslinking in the cured resin system. The data suggest that the thermal degradation of the epoxy-novolac resin takes place in four temperature-dependent stages. The first stage involves the loss of methyl cyclopentadiene at low temperatures to produce an unsaturated polyester. The second stage involves the dehydration reactions and the breakdown of the unsaturated polyester to form H2O, CO, CO2, and oxygen containing heterocyclic rings. In the third stage, one of the heterocyclic rings decomposes to form combinations of CO2 and CH4. In the fourth stage, H2 is released as domains of polyaromatic hydrocarbons are formed. A second objective of the project was to develop an understanding of the interaction of the important boundary layer species, atomic oxygen, with the carbon fiber network (FiberForm), which is a component of PICA. To this end, a beam of atomic oxygen was directed at samples of FiberForm, and the reactive and non-reactive products that scattered from the sample were detected as a function of their velocity and scattering angle with the use of a rotatable mass spectrometer detector. These results were qualitatively similar to the results of earlier studies of the oxidation of vitreous carbon and highly oriented pyrolytic graphite, suggesting that the chemical mechanisms are similar but that surface morphology influences the relative importance of the mechanisms. In order to deepen our understanding of carbon oxidation, earlier data from the Minton Lab on the oxidation of vitreous carbon were analyzed in great depth and the new insight was used in a collaboration to develop an oxygen-carbon ablation model.