A comparison between head-on quenching of stoichiometric methane-air and hydrogen-air premixed flames using Direct Numerical Simulations

Abstract

Head-on quenching of statistically planar stoichiometric methane-air and hydrogen-air flames has been compared based on a Direct Numerical Simulation (DNS) database. Due the absence of OH at the wall CO cannot be oxidised anymore which leads to the accumulation of carbon monoxide in the near-wall region during flame quenching of stoichiometric methane-air flames. Furthermore, for both fuels, low-temperature reactions at the wall have been found to give rise to accumulation of HO2 and H2O2 during flame quenching. As a result of this, a non-zero heat release rate can be observed at the wall during flame-wall interaction and this effect is particularly strong for the head-on quenching of the stoichiometric hydrogen-air premixed flame. The minimum Peclet number (i.e. normalised flame quenching distance) and the normalised wall heat flux magnitude are found to be smaller in the stoichiometric hydrogen-air flame than in the stoichiometric methane-air premixed flame. Moreover, it has been found that the flame quenching distance tends to decrease under turbulent conditions for the head-on quenching of the stoichiometric hydrogen-air premixed flame but the quenching distances for laminar and turbulent conditions remain comparable for the stoichiometric methane-air premixed flame. The mean reaction rate of reaction progress variable is not properly predicted in the near-wall region by well-known closures for the Flame Surface Density (FSD) or scalar dissipation rate (SDR). For FSD based mean reaction rate closure, it has been observed that recently proposed, simple chemistry DNS based, near-wall corrections perform satisfactorily for both fuels for the detailed chemistry case without adjustment of the model parameters. However, the previously proposed near-wall modification to an algebraic SDR closure based on simple chemistry data performs satisfactorily for the head-on quenching of stoichiometric methane-air premixed flame but it is found to be less effective in the case of the head-on quenching of stoichiometric hydrogen-air premixed flame. Moreover, the prediction of the mean heat release rate for head-on quenching of the stoichiometric hydrogen-air premixed flame cannot be achieved by the mean reaction rate closure of the reaction progress variable based on the mass fraction of a major species because of the important role played by intermediate species in the heat release rate at the wall.