On the adequacy of OH* as heat release marker for hydrogen–air flames

Abstract

The correlation between the distributions of OH* and heat release rate (HRR) is numerically investigated for laminar and turbulent hydrogen–air flames. First, laminar premixed one-dimensional flames are considered, observing a peak shift between OH* and HRR regardless of the equivalence ratio or OH* sub-mechanism. Nevertheless, OH* and HRR well correlate for methane-hydrogen fuel mixtures, suggesting that the reasons for such peak shift are to be searched in the hydrogen flame intrinsic properties. In particular, the H-radical is pivotal, given its different role in the main OH* formation and HRR reaction pathways. The chain-branching nature of hydrogen oxidation enhances the formation of H-radical pool, leading to higher OH* production in the post-flame region, while HRR peaks upstream, being linked to the consumption of HO2 generated by the recombination reactions of back-diffused H-radicals. Methane oxidation, instead, is chain-terminating, hence H-consuming, releasing heat and preventing the radical pool formation in the post-flame zone. Similar analyses are then performed for strained counterflow diffusion hydrogen–air flames, where the OH* distribution shows to be, at least for strain levels not close to extinction, an adequate HRR marker. Indeed, differently from premixed flames, HRR is here found to be dominated by H direct consumption on the fuel side. The observations made for laminar one-dimensional flames are confirmed by Large Eddy Simulations (LES) of three-dimensional turbulent hydrogen–air diffusion and partially premixed flames, stabilized in the HYLON injector at IMFT laboratory. When compared with the experimental OH* field, LES-computed HRR correctly retrieves OH* position in the diffusion flame, while a mismatch in the axial direction is observed between the two distributions for the lifted partially premixed flame. An overall good match, instead, is observed between measured and LES-computed OH* fields, emphasizing the importance of including OH* kinetics to accurately compare simulations and experiments of multi-regime hydrogen–air flames.