A Lattice Boltzmann approach for predicting the capture efficiency of random fibrous media

We study the propagation of submicron airborne particles through random fibre networks such as paper. In our approach, we first construct a three-dimensional model of the network and then use a Lattice Boltzmann method to obtain the flow of air through that structure. We finally calculate the trajectories of airborne particles and determine the fraction of these particles that impinge on fibres in the network. The combined approach is used to obtain pressure drop and mechanical filtration efficiency curves for a variety of structures. Our results show that, at fixed pressure drop and flow rate, a filter with a high basis weight and porosity will perform better than one made from fewer fibres that are more densely packed, at least in the range of porosities considered. For filters with a bimodal fibre size distribution, we find that the minimum in the efficiency curve becomes sharper and moves to smaller particle sizes as the mean fibre diameter of the mixture decreases, as expected from single-fibre theory. The efficiency of capture by diffusion and interception exhibits a weaker dependence on surface area mean fibre diameter than that predicted by theory, in agreement with the observations of Brown and Thorpe. (C) 2010 Curtin University of Technology and John Wiley & Sons, Ltd.