Branching ratios of competing unimolecular reactions often exhibit a complicated temperature and pressure dependence that makes modeling of complex reaction systems in the gas phase difficult. In particular, the competition...

Methyl formate (MF) is the smallest carboxylic ester and currently considered a promising alternative fuel. It can also serve as a model compound to study the combustion chemistry of the ester group, which is a typical structural feature in many biodiesel components. In the present work, the pyrolysis of MF was investigated behind reflected shock waves at temperatures between 1430 and 2070 K at a nominal pressure of 1.1 bar. Both time-resolved hydrogen atom resonance absorption spectroscopy (H-ARAS) and time-resolved time-of-flight mass spectrometry (TOF-MS) were used for species detection. Additionally, the reaction of MF and perdeuterated MF-d 4 with H atoms was investigated at temperatures between 1000 and 1300 K at nominal pressures of 0.4 and 1.1 bar with H-ARAS. In the latter experiments, ethyl iodide served as precursor for H atoms. Rate coefficients of seven parallel unimolecular decomposition channels of MF and five parallel reaction channels of the MF + H reaction were calculated from statistical rate theory on the basis of molecular and transition state data from quantum chemical calculations. These calculated rate coefficients were implemented into an MF pyrolysis/oxidation mechanism from the literature, and the experimental concentration–time profiles of H (from ARAS) as well as MF, CH3OH, HCHO, and CO (from TOF-MS) were modeled. It turned out that the literature mechanism, which was originally validated against flow-reactor experiments, ignition delay times, and laminar burning velocities, was generally able to fit also the concentration–time profiles from the shock tube experiments reasonably well. The agreement could still be improved by substituting the original rate coefficients, which were estimated from structure–reactivity relationships, by the values calculated from statistical rate theory in the present work. Details of the channel branching are discussed, and the updated mechanism is given, also in machine-readable form.