Reduced Kinetic Mechanism for Reactive Flow Simulation of Syngas/Methane Combustion at Gas Turbine Conditions

Abstract

The reduced kinetic mechanism for syngas/methane developed in the present work consists of a global reaction step for fuel decomposition in which the fuel molecule breaks down into CH2O and H2. A detailed CH2O/H2/O2 elementary reaction sub-set is included as the formation of intermediate combustion radicals such as OH, H, O, HO2, and H2O2 is essential for accurate predictions of non-equilibrium phenomena such as ignition and extinction. Since the chemical kinetics of H2 and CH2O are the fundamental building blocks of any hydrocarbon oxidation, the inclusion of detailed kinetic mechanisms for CH2O and H2 oxidation enables the reduced mechanism to predict over a wide range of operating conditions provided the reaction rate parameters of fuel-decomposition reaction is optimized over those conditions. Therefore, the rate coefficients for the fuel-decomposition step are estimated and optimized for the ignition delay time measurements of CH4, H2, CH4/H2, CH4/CO and CO/H2 mixtures available in the literature over a wide range of pressures, temperatures and equivalence ratios that are relevant to gas turbine operating conditions. The optimized reduced mechanism, consisting of 15 species and around 40 reactions, is able to predict the ignition delay time and laminar flame speed measurements of CH4, H2, CH4/H2, CH4/CO and CO/H2 mixtures fairly well over a wide range conditions. The model predictions are also compared with that of GRI3.0 mechanism. The reduced kinetic mechanism predicts the ignition delay time of CH4 and CH4/H2 mixtures far better than GRI mechanism at higher pressures. To demonstrate the predictive capability of the model in reactive flow systems, the reduced mechanism was implemented in Star-CD/KINetics commercial code using a RANS turbulence model to simulate CH4/air premixed combustion in a backward facing step. The CFD model predictions of the stable species in the exhaust gas agree well with the GRI mechanism predictions in a chemical reactor network modeling by approximating the backward facing step with a series of perfectly-stirred reactor and plug-flow reactor.