Modeling of the Soot Oxidation in Gasoline Particulate Filters

• Published in: SAE International journal of engines Vol. 8; no. 3; pp. 1253 - 1260
• Main Authors: Nicolin, Per, Rose, Dominik, Kunath, Florian, Boger, Thorsten
• Format: Journal Article
• Language: English
• Published: Warrendale SAE International 14.04.2015; SAE International, a Pennsylvania Not-for Profit
• Subjects: Engines; Fluid filters; Fuel injection; Gasoline; Gasoline engines; Modeling; Oxidation; Oxygen; Particulate emissions; Particulate matter; Robust design; Soot; Vehicles
• ISSN: 1946-3936; 1946-3944; 1946-3944
• DOI: 10.4271/2015-01-1048

The share of gasoline engines based on direct injection (DI) technology is rapidly growing, to a large extend driven by their improved efficiency and potential to lower CO₂ emissions. One downside of these advanced engines are their significantly higher particulate emissions compared to engines based on port fuel injection technologies [1]. Gasoline particulate filters (GPF) are one potential technology path to address the EU6 particulate number regulation for vehicles powered by gasoline DI engines. For the robust design and operation of GPFs it is essential to understand the mechanisms of soot accumulation and oxidation under typical operating conditions. In this paper we will first discuss the use of detailed numerical simulation to describe the soot oxidation in particulate filters under typical gasoline engine operating conditions. Laboratory experiments are used to establish a robust set of soot oxidation kinetics. The filter model and oxidation kinetics are validated using a large set of experimental data from laboratory reactor as well as vehicle experiments. With the validated model and parameters the impact of key operating parameters will be discussed, with special focus on passive soot oxidation mechanisms under normal vehicle operating conditions. Results show that short oxygen pulses during fuel cuts are the dominant passive regeneration mechanism. Results also provide insight into potential “worst case” conditions. Based on the insight from the detailed model a reduced 0D model is developed and described, which could be used for real time soot estimation.