In view of the considerable interest in the use of hydrocarbon-based jet fuels for scramjet propulsion, and the extremely short residence time within the combustor, on the order of milliseconds, chemical kinetics assumes a critically significant role in the development of scramjet technology. Consequently it is essential that the oxidation of hydrocarbon fuels be modeled well in CFD codes aiming to simulate the combustor behavior. The challenge here is that hydrocarbon reaction mechanisms are extremely complex, consisting of multitudes of intermediate species and reactions. Furthermore, the associated temperature-sensitive reaction rates can span over many orders of magnitude, thereby imparting severe stiffness to the code. The proposed program aims to perform rigorous and systematic reduction of detailed reaction mechanisms to levels that are computationally adaptable to complex CFD codes with moderated stiffness, while preserving chemical fidelity and comprehensiveness. The reduction will be based on theories of directed relation graph and computational singular perturbation. The effort will involve: reducing a validated, detailed ethylene oxidation mechanism for situations relevant for scramjet operations, using existing reduction algorithms; further improvements of these algorithms in terms of stiffness moderation; and assessment of effects of fuel preheat on the chemical mechanism and the combustion response
