Numerical simulation of soot formation in a turbulent diffusion flame: comparison among three soot formation models

Combustion and soot formation in a turbulent diffusion flame are simulated. Chemistry of combustion is treated with a detailed reaction mechanism that employs 49 species and 277 reactions. Turbulence is taken into account via the corrected k-epsilon model. Radiation heat transfer from flame is modelled by the P-1 model. An empirical model proposed by Khan and Greeves and two semi-empirical models proposed by Tesner and Lindstedt are used to simulate the soot formation in the flame. Khan and Greeves model showed to underpredict the maximum soot volume fraction. Nevertheless, the main shortcoming of Khan and Greeves model which undermines the applicability of this model to prediction of soot formation in turbulent diffusion flames is the inability to locate the highly sooting regions of the flame properly. Tesner model underpredicts the soot formation significantly, although the predicted shapes of the soot profiles are in accordance with the experimental measurements. Lindstedt model performs well in predicting both the maximum soot formation and the soot profile shapes in the chamber. Therefore, Lindstedt model can be considered as the most suitable model for the prediction of soot formation in turbulent diffusion flames.