Modeling transport of colloidal particles through polydisperse fibrous membrane filters under unfavorable chemical and physical conditions

Numerical simulations of filters with polydisperse fibers under chemically and physically unfavorable conditions were conducted to predict particle retentions on polypropylene fibrous filters. The fibrous filters were analyzed by a scanning electron microscopy to obtain fiber size distributions, and on the basis of a randomized algorithm, 2-D calculation domains were generated, and flow field calculations were conducted. For tracking particles, the standard Lagrangian tracking method in ANSYS Fluent software was modified by user-defined functions to incorporate particle deposition via interception and interaction energy calculations. Particle release from collector surfaces was considered by performing torque analysis for particles attached to the surfaces. Numerical retention efficiencies of fibrous filters showed very good agreement with experimental data without any free parameters or empirical correction factors. The parametric studies of polydisperse fibrous filters were conducted, and the retention efficiency was varied from 0 to 100% depending on chemical and physical conditions. [Display omitted] •Modeling of membranes with polydisperse fiber size distributions.•Incorporating interaction energy between a particle and filter in particle tracking.•Accurate prediction of deposition behaviors of colloidal particles on membranes.•Insignificant effect of membrane thickness on collision efficiency.•Lower collision efficiency with smaller fiber size and lower solid volume fraction.