An optimized model for ammonia/syngas combustion

Ammonia is a promising renewable carbon-free energy resource for alleviating environmental issues caused by greenhouse gas emissions. Many experimental data, including ignition delay time, laminar flame speed and species concentration, become available recently, providing a good foundation for improving and validating detailed reaction models. In addition, rate parameter adjustment of important reactions or competing reaction channels is frequently encountered in model development that can result in a discrepancy with theoretical and experimental rate constant data. Model optimization is a powerful tool for automated, data-driven adjustment in rate parameters and to gain valuable insight into the fundamental kinetics. A total of 41 influential reactions (123 Arrhenius parameters) were selected via sensitivity analysis and a two-stage optimization strategy was applied using 200 shock-tube and 172 rapid-compression-machine ignition delay time measurements, and 1411 laminar flame speed measurements. The temperature-dependent initial uncertainty was estimated and all Arrhenius parameters were optimized simultaneously. The optimized model showed an overall superior performance and is the only one among the investigated models without obvious prediction discrepancies in any of the tested experimental categories, reproducing well not only the ignition delay time and laminar flame speed data, but also the species profile data measured in ammonia pyrolysis, a jet-stirred reactor and a burner-stabilized flame. Most (39 out of 41) of the reaction rate adjustments were well within the corresponding uncertainty (<= 50%) and the reactions with a noticeable adjustment were examined and discussed with reference to the rate constant measurements and macroscopic combustion data. The kinetic interaction between ammonia and hydrogen under pyrolysis conditions, and between ammonia and carbon monoxide in propagating flames, was investigated using the optimized model.