Industrial applications for zeolites include catalysis, ion exchange, and separations. The scope of applicability would increase with the discovery of new zeolites (currently fewer than 200 zeolites are known), which can be stimulated with computational predictions of stable, zeolite-like structures. Recently, R.S. Pophale and M.W. Deem of Rice University, in collaboration with P.A. Cheeseman of Purdue University, refined their previously published computational approach by accounting for the Pauli exclusion principle, and constructed a database of predicted, zeolite-like materials.
As reported recently in the online edition of Physical Chemistry Chemical Physics (DOI: 10.1039/c0cp02255a), Pophale, Cheeseman, and Deem developed a Monte Carlo technique to randomly sample the structural space of low-energy, zeolite-like structures. Geometry optimizations were performed with two interatomic potentials—the Sanders-Leslie-Catlow potential, which is accurate for zeolites, and the van Beest-Kramer-van Santen potential, which agrees well with experimentally determined enthalpies of formation. Enforcing the Pauli exclusion principle makes the structures resulting from the two potentials much more realistic, stable, and similar to each other.
Over 2.6 M zeolite-like structures were found and about 10% have energies in the range of known zeolites. Calculated powder x-ray diffraction patterns for the database structures are similar to those of known zeolites.
The researchers said that the low-density structures in their database are of particular interest because they expect them to have large rings and large pores, and expressed the hope that their work will spur attempts at their synthesis. The researchers also said that they are collaborating with other research groups in order to explore the adsorption and diffusion of small molecules in a subset of their zeolite database as well as how their predicted zeolites can be used for carbon sequestration.
The researchers said that their approach “may serve as a guide for construction of analogous databases for other materials such as metal-organic frameworks or crystal hydrates. Diversity in structures and possible functionality among these classes of materials could reveal themselves through such efforts.”
