Investigation on permeability of filter cake with different particle sizes: Experimental and simulation study

Cake permeability is a crucial macroscopic physical property in characterizing filtration behavior for coal slurry. However, there is still no appropriate model derived to establish the essential relationship between the cake permeability and complex particle property. In this study, the three-dimensional microstructures with different particle gradation and porosity were generated using particle discrete element software. The permeability of each constructed model was calculated utilizing Lattice Boltzmann Method (LBM). Simulation result shows that the Kozeny-Carman (KC) model is proper to predict the permeability of porous media with different gradations. When the gradation parameters are not considered, it shows that d30 can best represent the particle size characteristics of the particle systems. When the porosity and average particle size remain unchanged, the permeability increases with the increasing of the non-uniformity coefficient (Cu) and the curvature coefficient (Cc). Moreover, we proposed an analytical expression for the permeability estimation with different gradation characteristics, which contains the parameters that can be easily measured in engineering. Then a set of coal slurry samples with different particle size distributions were selected to conduct the filtration test, the permeability of each sample was measured to compare with the predicted results of the gradation permeability model. It is clear that the KC equation in this case can not calculate the cake permeability effectively. Although the gradation model is relatively appropriate to estimate the cake permeability, its accuracy still needs to be improved by considering the pore structure property to eliminate the effect of constant terms.