A Method to Screen Nanoporous Catalysts for Transition-State Selectivity

3-dimensional nanoporous materials such as zeolites and metal-organic frameworks are used as catalysts for some of the most important industrial chemical reactions. The unique ability of these materials to selectively promote certain transformations originates from the subtle (mis)match between structures of the catalyst frameworks and those of the reacting species, and is usually classified into reactant, product, and transition-state selectivity. While the first two types have been successfully studied or even predicted at a large scale using Monte Carlo and molecular dynamics methods, analyzing transition-state selectivity requires much more computationally expensive quantum chemical techniques, and is limited to examination of a few sites on a case-by-case basis. Drawing on our findings from investigating zeolite-catalyzed reactions involved in catalytic cracking, we introduce a method based on the configurational-bias Monte Carlo algorithms that allows us to sample transition-state structures over the entire pore space of a material, and show that it is able to yield energies in good correlation with those obtained by density-functional theory calculations. The low-energy configurations identified in the sampling procedure can optionally be refined automatically for better accuracy via higher-level calculations. This method is easily applicable to long-chain molecules with large numbers of conformational degrees of freedom, for which current quantum chemical techniques can be very challenging to use. In addition, it also allows us to screen known zeolite structures for the wide array of different reactions involved in energy conversion processes.