A computational fluid dynamics (CFD) study was performed to investigate the extent ofmethane recombination in an expansion tube to replicate Titan entry conditions. The
majority of spacecraft heating for Titan entry is a result of the formation of CN molecules in the non-equilibrium shock layer. This heating can be investigated in experimental tests via the use of transient flow facilities such as expansion tubes however for accurate modelling it is desirable for the conditions immediately before the bow shock of a test model to accurately reflect the actual Titan atmospheric composition.
A number of methods were used to investigate the expected conditions in the expansion tube considered for experimental testing, X2. An initial equilibrium analysis of the flow composition and conditions was conducted through the unsteady expansion by stepping through with constant entropy using the Chemical Equilibrium and Applications (CEA) package. This revealed that for the given experimental conditions the methane would indeed have completely reformed at the nozzle exit conditions if given sufficient time to move towards equilibrium. Further analysis using finite rate chemistry in an existing onedimensional
code, L1d, revealed that this was in fact not the case and the composition at the end of the expansion tube consisted largely of N2, HCN, and H2. The flow was found
to be largely chemically frozen down the length of the acceleration tube and there was very little recombination. This one-dimensional simulation had an excellent correlation with the experimental results and for the finite rate case the primary shock speed was found to match the experimentally measured speed and the secondary shock differed by just more than 5%. This result meant that the test gas did not have sufficient time to move towards equilibrium and any equilibrium analysis would provide erroneous results.
Further analysis of the flow was attempted using a two-dimensional code, MB_CNS, but incorrect input conditions meant that the results obtained did not exactly represent the desired situation. Information about the flow behaviour was still obtained from these results but it was not possible to use these as a comparison with the one- dimensional data.