Phenomenological soot modeling with solution mapping optimization of biodiesel-diesel blends in diesel engines

It is widely accepted that the particulate matter emissions of a diesel engine may be substantially reduced when a blend of diesel and biodiesel fuels is used. Here we propose a phenomenological model aimed at evaluating the extent of soot formation and oxidation in a diesel engine fueled with diesel–biodiesel blends and providing a better understanding of these processes. These blends are represented by three hydrocarbon functional groups: aromatic, aliphatic, and ester fuels, in different proportions. In a previous study, the phenomenological soot model of Foster and Reitz and colleagues was modified to capture the influence of the diesel–biodiesel molecular composition on soot formation and emissions. In the present work, this model was implemented in computational fluid dynamics (CFD) diesel engine combustion simulations. These CFD simulations were performed using the finite volume method commercial code ANSYS Fluent 16.1. The solution mapping optimization method developed by Frenklach and colleagues was used to calibrate this model for the engine. Validation was performed by comparing the following simulated variables with experimental results: pressure, heat release rate, averaged soot density inside the piston bowl, maximal temperature inside the piston bowl, and exhaust soot emissions. Reasonable agreement was obtained between our computational results and the published experimental results of Mancaruso and colleagues. The soot evolution in the piston bowl in these simulations was to a certain extent qualitatively and quantitatively similar to that in the experiments. The maximal experimental in-cylinder two-color soot density was 44 g/m³ and its computational equivalent was 60 g/m³. The maximal space-averaged two-color soot densities in the experiments and the computations were 9.3 and 7.73, respectively. The experimental soot reduction potentials of biodiesel blending for B20, B50, and biodiesel were 12.0%, 27.3%, and 47.2%, respectively. The respective computational soot reduction potentials were 34.8%, 36.2%, and 43.4%.