LES investigation of soot formation in a turbulent non-premixed jet flame with sectional method and FGM chemistry

Abstract

This study proposes a detailed soot modeling framework for large-eddy simulation (LES) to accurately predict soot formation and particle size distributions (PSD) in turbulent reacting flows. The framework incorporates Flamelet Generated Manifold (FGM) chemistry and a soot model based on the discrete sectional method (DSM) to predict both qualitative and quantitative sooting behavior while keeping the computational cost affordable. Two elementary modeling strategies are considered in the LES formalism for describing soot formation rates. These strategies rely on an a-priori tabulation of soot formation rates and their run-time computation. The LES formalism is applied to the simulations of a well-characterized, non-premixed, turbulent jet flame. A comparative analysis of strategies employed for filtered soot source term treatment is conducted to investigate their impact on the prediction of soot quantities and the evolution of soot PSDs. The LES results for the gas phase and soot phase are compared against the available experimental data. A good prediction of soot evolution is achieved with the two methodologies. The tabulation of soot formation rates leads to a significant reduction in computational cost compared to the model based on their explicit runtime computation. The LES results reveal that the modeling of filtered soot source terms has a significant impact on the quantitative prediction of soot formation. The possible reasons for the observed differences in the soot prediction are discussed. The run-time computation-based model provides a more consistent treatment of the non-linear interactions between the gas and soot phases in soot source terms compared to the tabulated soot chemistry approach. On the other hand, the tabulated soot chemistry model is an interesting and efficient modeling approach for predicting soot formation in turbulent conditions. Overall, both approaches have their strengths and limitations, and the choice of approach may depend on the specific needs of the application.