Numerical and Experimental Pore-scale Analyses of Inert and Reactive Multiple Colloidal Particles in Complex Flow Domains
Abstract
A fluctuating lattice-Boltzmann model was developed to simulate pore-scale flow and transport of multiple particles in geometrically complex porous and fractured domains. The model is based on the original work of Ladd [J. Fluid Mech., 271, 285, 1994] and the modeling approach based on the virtual intraparticle fluid nodes as proposed by Ding and Aidun [J. Stat. Phys., 112, 685, 2003]. The model has been improved by introducing two-body electrostatic and van der Waals potentials. Moreover, the commonly used bounce-back algorithm to simulate no-slip conditions has been replaced by an immersed boundary condition to simulate softer particle-wall interactions. The simulation results captured the wall and inertial effects on trajectories of a single particle in different Reynolds number flows in smooth-walled channels, consistent with earlier numerical simulation results. Multiple-particle simulations in porous and fractured domains captured trains of particles crossing multiple streamlines in fast-flow paths and lagged particles in slow-flow paths as has been observed in our experiments and reported in the literature. Experimental studies are focused on two-dimensional flow for three microflow cell geometries and use monodispersed particles in dense and dilute concentrations. The average particle sizes are 2, 10 and 30 microns, and the interaction between particle surfaces is controlled by the use of surfactants. Three two- dimensional flow cells with a 50 to 500 micron width have been manufactured to evaluate scale effects. Preliminary results are available for the flow of 2 micron poly(lactic-co-glycolic acid) [PLGA] microspheres dispersed in polyvinyl alcohol solution (PVA) in a 500-micron tube with inline flow obstruction with dilute and concentrated solutions. These results demonstrate particle streamlines and show particle-particle and particle-wall interactions. The experimental findings are compared with simulation results.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2008
- Bibcode:
- 2008AGUFM.H43E1062B
- Keywords:
-
- 1847 Modeling;
- 4809 Colloids