(688h) Modeling Single Component Transport in Mesoporous Membranes Under Non-Equilibrium Conditions

Authors: 
Ford, D. M., University of Arkansas
Rathi, A., University of Massachusetts Amherst
Kikkinides, E., Professor
Monson, P. A., University of Massachusetts Amherst
Mesoporous membranes have potential for energy efficient removal of pollutants like CO2 and volatile organic compounds from their mixture with air in the form of industrial effluents. Logical design of mesoporous materials requires in-depth knowledge of transport of both condensable vapors and gases in pure and mixture form through the pores. Characterization of these materials through measurement of pore size distribution as well as application to gas separation, require partial pressure gradients and are thus inherently nonequilibrium. Modeling of these systems with classical modeling techniques such as computational fluid dynamics require predetermined models for various phenomenon such as Knudsen diffusion, capillary condensation and surface adsorption while molecular simulations require computational effort too large to be used on regular basis. Dynamic mean field theory (DMFT), a coarse-grained lattice based theory, computationally efficient technique has been used to model mesoporous membranes. We used this approach to study permporometry, a characterization technique where permeation of a light gas in presence of condensable vapor is used to determine pore size distribution and deviation from equilibrium is small. We have further investigated separation process associated with use of significant pressure gradient to separate condensable vapor from its mixture with light gas. This represented a large deviation from equilibrium. An interesting phenomenon was revealed as a result where capillary condensation is confined to high pressure side of the system and pore filling is incomplete at steady state. This phenomenon has been previously hypothesized in the literature but concrete understanding is still lacking. We used a simple slit pore with pure fluid to understand the transport in partial capillary condensation state using both DMFT and dual control volume grand canonical molecular dynamics.