HIGH QUALITY COMPUTATIONAL SCREENING OF METAL-ORGANIC FRAMEWORKS FOR CONTAMINANT REMOVAL
MetadataShow full item record
High-throughput computational screening of thousands of metal-organic frameworks (MOFs) have been performed for separation applications using selective adsorption. First, a MOF-specific benchmarking study of DFT functionals for predicting MOF structural parameters, elastic properties, and atomic point charges was performed. To achieve this task, a test set of diverse MOFs with high accuracy experimentally derived crystallographic structures was compiled. Results indicate that the discrepancies in the properties predicted by the various functionals is small compared the accuracy necessary for most practical applications. Motivated by these observations, the PBE functional was used to assign atomic point charges derived from periodic DFT electronic structure calculations for thousands of MOFs. As an example of using these charges, each MOF was examined for adsorptive removal of tert-butyl mercaptan (TBM) from natural gas. Monte Carlo (MC) simulations revealed many candidate MOF structures with high selectivity for TBM. Based on results from the benchmarking study, DFT was used to predict the energy minimized structure of over 800 MOFs. These energy minimized structures are used to analyze the relationship between nanopore structure and gas adsorption properties. Results indicate that structure precision is crucial for MC prediction of CO2 adsorption in MOFs. Given the findings, preliminary studies of impact of MOF flexibility on the MC prediction of adsorption properties of CO2 and xylenes were performed.