Bridging the gap between diffusivities from experiment and molecular dynamics: n-hexane and 2,2-dimethyl butane in zeolite BEA

When do molecular dynamics (MD) simulations yield better agreement with experimental macroscopic measurements of diffusivities of guest molecules in zeolites? We report studies to show that simulations of n-hexane and 2,2-dimethyl butane in 'powder' samples of zeolite BEA (as compared to single-crystal sample) lead to a better agreement with ZLC diffusivity measurements previously reported by Barcia et al. The results suggest that main reason why MD simulations until now could not reproduce the measurement from macroscopic techniques such as uptake and gas chromatography is because all simulations employed 'single-crystal' sample. Implications of these results on 'sample preparation' which is considered very important in experiments but never planned in simulations are discussed. The results further demonstrate that the changes in diffusivities, as well as activation energies that are observed between single crystal and 'powder' sample are not only dependent on the nature of the zeolite sample, but also on the type of guest molecule, i.e., the changes on going from 'single crystal' to 'powder' sample for n-hexane and 2,2-dimethyl butane are not the same.