Concentration dependent transport of colloids in saturated porous media.

A series of column experiments was undertaken to explore the influence of colloid input concentration (2, 1, 0.5, and 0.25 times a reference concentration), colloid size (negatively charged 3.2 and 1.0 microm carboxyl latex), and sand grain size (360, 240, and 150 microm quartz sands) on transport and deposition. A similar mass of stable mono-dispersed colloids was added to each column. For a given input concentration, decreasing the sand size and increasing the colloid size resulted in increased mass retention in the sand near the column inlet and lower relative concentrations in the effluent. For a given sand and colloid, increasing the input concentration produced less deposition and higher mass recovery in the effluent, especially for coarser sands and smaller colloids. Results of a time dependent attachment (blocking) and detachment model were not consistent with this behavior because the simulations predicted much less retention near the column inlet and a decreasing number of favorable attachment sites (mass of deposited colloids) with increasing input concentration in a given system (colloid and sand). A time dependent straining model (filling of straining sites) provided a better description of the effluent and deposition data, but still could not account for the observed concentration dependent mass recovery. Alternatively, the straining model was refined to include a liberation term that assumed that straining was hindered at higher concentrations (collision frequencies) due to repulsive colloid (aqueous phase)-colloid (strained) interactions. Simulations that included straining, liberation, attachment, and detachment significantly improved the description of the experimental data.