Soot formation models assessment in turbulent diffusion jet flames: A RANS-based comparison

A RANS-based Comparison Due to the intricate interaction between turbulence, chemical kinetics, radiation, and soot particle dynamics, modelling soot formation processes in flames is a challenging task. To predict the level of soot formed, it is essential to accurately capture all stages of soot formation and oxidation. Using a RANS approach, this study focuses on the implementation, within the computational open-source tool OpenFOAM, and comparison of three detailed soot formation models, (i) the Interpolative Closure Method of Moments (MOMIC), (ii) the Hybrid Method of Moments (HMOM), and (iii) the Discrete Sectional Method (DSM), as well as a semi-empirical two-equation model. Both the combustion process and the formation of soot precursors in the gas phase are described using the Steady Laminar Flamelet model and a detailed chemical kinetic mechanism. Radiation effects are modelled using the optically thin method. The computational results obtained here are compared with the experimental data characterizing the Adelaide ENH1 jet flame and other past numerical results. The results reveal significant differences in soot formation source terms among the models and for each of the soot formation stages. DSM best matches the experimental peak soot position, which is attributed to its modelling of condensation and surface growth occurring downstream compared to MOMIC and HMOM.