Comparison of methane combustion mechanisms using concentration measurements

A large amount of concentration measurement data from combustion experiments carried out in flow reactors, jet-stirred reactors, and shock tubes for methane (+ H 2 /CO)-oxygen-diluent mixtures covering wide ranges of temperature, pressure, equivalence ratio, and diluent ratio were collected from 45 publications. Altogether, 35420 concentration data points in 965 datasets were encoded in 332 XML data files and made available on the ReSpecTh site ( https://respecth.hu ). The measured species are CH 4 , C 2 H 2 , C 2 H 4 , C 2 H 6 , aC 3 H 4 (allene), pC 3 H 4 (propyne), C 3 H 6 (propene), C 3 H 8 , C 4 H 2 (diacetylene), C 4 H 4 (vinylacetylene), C 6 H 6 (benzene), CH 2 O, CH 3 OH, C 2 H 5 OH, O 2 , CO, CO 2 , H 2 , and H 2 O. In some shock tube measurements, the concentrations of radicals CH, CH 3 , and OH were also measured. The diluents were N 2 , H 2 O, CO 2 , Ar and He. The performance of 13 methane combustion mechanisms was analysed based on their capability to reproduce concentration measurements across various experimental types and conditions. A novel consideration of temperature uncertainty was proposed and applied in the performance evaluation of models. The mechanisms Aramco-II-2016, Konnov-2017, NUIG1.1–2021, Glarborg-2018, Caltech-2015, and FFCM-I-2016 were the best-performing mechanisms in ascending order of error. Aramco-II-2016 reproduces the concentrations of flow reactors the best, especially for CH 4 , O 2 , CO, CO 2 , CH 3 OH, and C 2 hydrocarbons. Konnov-2017 is good at predicting concentrations in shock tubes, and FFCM-I-2016 has the overall lowest errors for helium-containing measurements. Local sensitivity analysis based on the Aramco-II-2016 mechanism identified 42 important elementary reactions. Directly measured and theoretically calculated rate coefficients (2316 data points in 176 datasets) were collected for these elementary reactions and compared with those used in the investigated methane combustion mechanisms.