Highlighting a variety of unusual characteristics of adsorption and diffusion in microporous materials induced by clustering of guest molecules

Abstract

In this work, we highlight several unusual characteristics of adsorption and diffusion of a variety of guest molecules, such as linear and branched alkanes with a number of C atoms in the 1-6 range, CO₂, and Ar in microporous structures such as zeolites (FAU, NaY) and metal organic frameworks (IRMOF-1, CuBTC, MIL-47, MIL-53 (Cr)-lp, PCN-6') that have channel or cavity sizes larger than about 0.75 nm. Clustering of guest molecules is found to manifest at temperatures below the critical temperature, T_c, of the guest species. The degree of clustering is increased as the temperature, T, is reduced increasingly below T_c. For linear alkanes, T_c increases with chain length and, consequently, at a given T the degree of clustering increases with increasing chain length. For C4, C5, and C6 alkane isomers, the linear isomer shows a higher degree of clustering than the corresponding branched isomers. Mixture adsorption characteristics are significantly influenced by clustering; specifically, the separation selectivity is found to increased significantly with lowering T. We also discuss the interesting possibility of separating alkane isomer mixtures by exploiting the differences in the degrees of clustering, induced by differences in T_c of constituent species. An important characteristic of clustering is that the inverse thermodynamic factor 1/Γ _i ≡ (d ln c_i)/(d ln f_i) exceeds unity for a range of molar concentrations ci within the micropores. For the concentration ranges for which 1/Γ _i >1, the Fick diffusivity, D_i, for unary diffusion is often lower than both the Maxwell-Stefan, Ð_i, and the self-diffusivity, D_i,self. Correlation effects in diffusion are significantly lowered as a consequence of clustering; this reduction in correlation effects is found to have a significant influence on the mixture diffusion characteristics. The diffusion selectivity is significantly affected with increased clustering.