Opportunities for Connecting Computational and Experimental Approaches to Design Zeolite Catalysts for Ethanol-to-Olefin Reactions

Increasing concern over global CO₂ emissions has led to recent attention toward routes to produce sustainable aviation fuel (SAF) as one strategy to decarbonize the aviation industry. In one established route for SAF production, bioethanol from agricultural waste and other biomass is first converted to light olefins via ethanol-to-olefins (ETO) reactions, then oligomerized and hydrogenated to C₈–C₁₆ paraffins that serve as jet-range blendstocks. The initial ETO step strongly governs carbon efficiency, selectivity, and deactivation behavior. Zeolites and metal-containing zeotypes have therefore received renewed interest as catalysts for ETO chemistries. In parallel, developments in computational catalysis and machine learning have expanded and accelerated opportunities for understanding the kinetics of zeolite-catalyzed reactions. These include machine learning interatomic potentials that can extend the length and time scales of molecular dynamics simulations by approximating ab initio energies and forces within the training domain. Obtaining a molecular-level understanding of ETO kinetics in metal-containing zeotypes and zeolites is complicated by the interactions of coproduced water on reactant adsorption and transition state stability, which in turn influence broader catalytic performance such as selectivity and deactivation. Herein, we discuss the landscape of microporous catalysts and reaction pathways, starting from bioethanol that can be used to produce olefins. We then provide our recommendations for how recent computational developments can be leveraged to obtain a fundamental understanding of ETO reactions in conjunction with experimental measurements. Finally, we provide our outlook on how the integration of these computational tools with experiments can lead to the discovery of next-generation microporous catalysts.