Stochastic modeling of kaolinite transport through a sand filter

Abstract

Accurately modeling the transport of clay particles through coarse-grained porous media is essential to engineering applications ranging from filtration and drainage, groundwater flow modeling, to contaminant transport. However, predicting the retention and clogging behavior of clay particles within a coarse-grained soil matrix is extremely challenging because clay particles can aggregate and form clusters with a variety of fabrics depending on the prevailing geochemistry of the pore fluid (i.e., pH and ionic strength). The work performed in this study developed a stochastic model to investigate the uncertainty of clay particle transport in porous media using random sampling at a given grain-size distribution to account for inherent uncertainty of the size of clay clusters being transported. Results demonstrated that the model proposed in this work can evaluate upper and lower boundaries of retention profiles of clay particles in a sand medium at given mean and standard deviation of grain-size distributions. In addition, the deterministic approach (using median sizes of sand and clay particles in the simulation) underestimated the mass of retained particles at small size ratios of clay particle size/sand particle size when compared with the stochastic prediction, which would result in nonconservative design.