In Silico Screening of Zeolites for the Highly Selective Adsorption of Central C-C Bonds toward More Effective Alkane Cracking

Developing Br?nsted acid zeolites that can enable more effective alkane cracking by promoting the breaking of its central bonds for minimal methane production plays a critical role in the petroleum industry. In this study, we computationally investigate more than 1 million distinct T-site environments to identify promising Br?nsted acid zeolites that favor the adsorption of alkane with its central bonds near the acid sites. To make such large-scale screening feasible, a new entropy estimation algorithm is developed. This approach requires orders of magnitude shorter computational time than the traditional configurational bias Monte Carlo (CBMC) technique and offers accurate predictions. Through analyzing top zeolites identified herein, ellipsoid-shaped confinements are found to be the key structural feature. The longer axis of the confinement needs to be sufficiently large to accommodate the adsorption of alkane with its central bonds near the T-site but sufficiently short to entropically penalize the alkane adsorption at their terminal bond. Such geometry can also be considered to be a sugar cane structure, with the smaller pore opening of zeolites representing its node. This work can guide the future development of Br?nsted acid zeolites, and the method developed can also be applied to study the adsorption of other molecules at a high temperature. ? 2021 American Chemical Society. All rights reserved.