资源环境科技发展态势分析平台

A 3-D refractive-index matching Lagrangian particle tracking (3D-RIM-LPT) system was developed to study the filtration and the clogging process inside a homogeneous porous medium. A small subset of particles flowing through the porous medium was dyed and tracked. As this subset was randomly chosen, its dynamics is representative of all the rest. The statistics of particle locations, number, and velocity were obtained as functions of different volumetric concentrations. It is found that in our system the clogging time decays with the particle concentration following a power law relationship. As the concentration increases, there is a transition from depth filtration to cake filtration. At high concentration, more clogged pores lead to frequent flow redirections and more transverse migrations of particles. In addition, the velocity distribution in the transverse direction is symmetrical around zero, and it is slightly more intermittent than the random Gaussian curve due to particle-particle and particle-grain interactions. In contrast, as clogging develops, the longitudinal velocity of particles along the mean flow direction peaks near zero because of many trapped particles. But at the same time, the remaining open pores will experience larger pressure and, as a result, particles through those pores tend to have larger longitudinal velocities.