Computational Investigation of Window Size Effects on Gas Transport In High Silica 8-Ring Zeolites
Preeti Kamakoti1, Sang-Eun Jee2, Ronald R. Chance1, Sebastian C. Reyes1, and David S. Sholl2. (1) ExxonMobil Research and Engineering, Annandale, NJ 08801, (2) Georgia Institute of Technology, School of Chemical & Biomolecular Engineering, 311 Ferst Drive, N.W., Atlanta, GA 30332-0100
The diffusion of gases and hydrocarbons in microporous materials is of great practical importance in the chemical and petrochemical industry. Small differences in window size and shape can have dramatic effects on molecular transport, an example being methane in 8-ring zeolites, where window size and molecular size are comparable. Pulsed field gradient NMR experiments  show that subtle differences in window dimensions of around 0.05-0.35 Å between siliceous LTA, CHA and DDR can cause room temperature methane self diffusivities to span two orders of magnitude, ranging from 1.42x10-6 cm2/s for LTA to 1.6x10-8 cm2/s for DDR. Currently, there is little understanding of the factors affecting window architecture, and no means identified for prediction of the significant structural variations that have been observed experimentally. To provide a better understanding of window architectures, we have performed Density Functional Theory (DFT) calculations on the same three siliceous zeolites LTA, CHA and DDR, wherein a complete relaxation of the cell volume, shape and atomic positions was performed. Results show that DFT is an excellent tool for quantitatively predicting structural information in siliceous zeolites. DFT is also surprisingly successful in capturing the subtle differences between window sizes and puckering in all three structures, those differences being critical to understanding molecular transport. The optimized structures obtained with DFT are used in conjunction with an atomistic model to simulate the adsorption and diffusion of CO2 and CH4. We have used molecular simulations to examine the diffusion of CO2 and CH4 in DDR, CHA, and LTA, both as single components and as mixtures and performed a comparison with experimental rseults. These simulations give insight into the strong dependence of the observed diffusivity on zeolite structure and suggest opportunities for using the unusual properties of diffusing mixtures in some materials for practical applications.