Single wall diesel particulate filter (DPF) filtration efficiency studies using laboratory generated particles

Diesel engines offer higher fuel efficiency, but produce more exhaust particulate than conventional gasoline engines. Diesel particulate filters are presently the most efficient means to reduce these emissions. These filters typically trap particles in two basic modes: at the beginning of the exposure cycle the particles are captured in the filter holes, and at longer times the particles form a "cake" on which particles are trapped. Eventually the "cake" is removed by oxidation and the cycle is repeated. We have investigated the properties and behavior of two commonly used filters: silicon carbide (SiC) and cordierite (DuraTrap (R) RC) by exposing them to nearly-spherical ammonium sulfate particles. We show that the transition from deep bed filtration to "cake" filtration can easily be identified by recording the change in pressure across the filters as a function of exposure. We investigated the performance of these filters as a function of flow rate and particle size and found that the filters have the highest filtration efficiencies for particles smaller than similar to 80 nm and larger than similar to 200 nm. A comparison between the experimental data and a simulation using incompressible lattice-Boltzmann model shows good qualitative agreement, but the model over-predicts the filter's trapping efficiency. (C) 2009 Published by Elsevier Ltd.