(452a) Computational Generation of Low-Energy Configurations of Zeolite NaY

With growing interest in the potential of zeolites to cost-effectively separate gas mixtures of industrial significance [1-2], computational methods are used to model adsorption and diffusion in these microporous materials. Many computational studies focus on pure-silica zeolites due to their readily-available and perfectly-periodic crystal structures [3-4]. Most zeolites, however, have never been synthesized in pure-silica form, and nearly all zeolites used in practice are defined by a Si/Al ratio and the presence of extra-framework cations. Averaged crystal structures of zeolitic materials are available from the International Zeolite Association [5], for example, with atoms described by fractional occupancies. However, to consider such materials in molecular simulations, it is necessary to generate realizations of typical configurations of the aluminum and cations.

We have developed a novel computational framework for modeling cation-modified zeolite structures. The approach consists of dividing a zeolite structure into small clusters centered around each T-atom site, which can take on either Si or Al. The possible states of the neighborhood around each T-atom site are sampled, and for each configuration the energy difference between a Si or Al center is estimated using electronic structure calculations [6]. The cluster energies are combined to construct a larger section of modified zeolite crystal with low energy using mixed-integer linear optimization.

The approach is demonstrated with zeolite NaY, selected for its importance to the petrochemical industry and the simplicity of its faujasite framework in which all T-atoms are symmetrically equivalent. For Si/Al ratios between 2.25 and 2.85, we have generated low-energy configurations of a periodic unit cell of NaY suitable for use in molecular simulations of adsorption and diffusion. This opens the possibility for modified zeolites to be considered in computational screenings of zeolites for applications in separations and catalysis.

References:

1. Hasan, M. M. F., First, E. L., and Floudas, C. A. Cost-effective CO₂ capture based on in silico screening of zeolites and process optimization. Physical Chemistry Chemical Physics, 15(40):17601–17618, 2013.

2. First, E. L., Hasan, M. M. F., and Floudas, C. A. Discovery of novel zeolites for natural gas purification through combined material screening and process optimization. AIChE Journal, 60(5):1767–1785, 2014.

3. First, E. L., Gounaris, C. E., Wei, J., and Floudas, C. A. Computational characterization of zeolite porous networks: an automated approach. Physical Chemistry Chemical Physics, 13(38):17339–17358, 2011.

4. First, E. L., Gounaris, C. E., and Floudas, C. A. Predictive framework for shape-selective separations in three-dimensional zeolites and metal–organic frameworks. Langmuir, 29(18):5599–5608, 2013.

5. Baerlocher, C. and McCusker, L. B. Database of Zeolite Structures. http://www.iza-structure.org/databases/, 2014.

6. Frisch, M. J. et al. Gaussian 09, Revision A.02. Gaussian, Inc., Wallingford CT, 2009.