Highlighting slowing-down effects in mixture permeation across microporous membranes

Binary mixture diffusion in microporous crystalline materials is characterized by a set of three M-S diffusivities: Ð 1 , Ð 2 , and Ð 12 . The Maxwell-Stefan (M-S) diffusivities Ð 1 , and Ð 2 characterize interactions of guest molecules with pore walls. The exchange coefficient Ð 12 quantifies correlation effects that result in slowing-down of the more mobile species due to correlated molecular jumps with tardier partners. Molecular Dynamics (MD) simulation data for a wide variety of binary mixtures in several zeolites, and metal-organic frameworks (MOFs) demonstrate that the M-S diffusivities are dependent on the pore occupancy, θ ; the quasi-chemical theory provides the appropriate theoretical framework to quantify the occupancy dependence. The M-S diffusivities Ð 1 , and Ð 2 in binary mixtures can be estimated with reasonably good accuracy from the corresponding unary diffusivities, evaluated at the same pore occupancy, θ . Correlation effects are of significant importance in structures with intersecting channels (MFI, ISV, BEA), cavities with large windows apertures (FAU, NaY, NaX, CuBTC, IRMOF-1, MOF-177), and 1D channels with channel dimension > 8 Å (AFI, MgMOF-74, MIL-47, MIL-53(Cr), BTP-COF). Correlation effects are considerably less significant in hosts such as LTA, CHA, DDR, and ZIF-8, that consist of cages separated by narrow windows in the 3.2 Å – 4.2 Å size range. Experimental data on unary and binary mixture permeation across SAPO-34, DDR, ZIF-8 and MFI membrane are analyzed to confirm the expectations based on MD simulations regarding the importance of slowing-down effects in various structures. For SAPO-34, DDR, and ZIF-8 the component permeances in mixtures are nearly the same as for the unaries; there are no slowing-down effects. For CO 2 /H 2 permeation across MFI membrane, the H 2 permeance in the mixture is about an order of magnitude lower than the corresponding unary value due to slowing-down effects caused by correlations.