(584au) Unraveling Fluoride's Structure Directing Role on Zeolite Self-Assembly Via Integrated DFT and Monte Carlo Simulation

Zeolites are nanoporous aluminosilicate materials with high specific surface area, regular pore sizes, strong acidity, and ion exchange capacity. Zeolites are widely applied in various fields, including catalysis, separations, and storage. There is increasing demand for zeolites, including tailor-made zeolite structures, to reach sustainability-related goals such as catalyzing biofuel production and capturing carbon dioxide. However, achieving precise control over the synthesis remains a significant challenge due to the complex interplay of factors within the sol-gel systems. As one of the key factors influencing zeolite formation, fluoride anion plays a significant structure-directing role in stabilizing key building blocks such as the double-4-membered ring (D4R). A thorough comprehension of the interactions between sol-gel silica and fluoride anions is dominant in unraveling the intricate zeolite self-assembly pathways.

We integrate density functional theory (DFT) calculations with Reactive Ensemble Monte Carlo (RxMC) simulations, which successfully capture the zeolite formation kinetics and the complex interaction between the Si framework and the organic structure-directing template. A multi-step strategy is taken: (1) Recognizing the structural diversity of fluoride-silica interactions and the need for energetic accuracy, gas-phase electronic energy calculations via DFT generate a comprehensive energy dataset detailing the evolution of silica clusters. (2) The constructed energy dataset refines the acceptance probabilities within the RxMC simulations, enabling an accurate representation of the energetic landscape governing fluoride-mediated zeolite assembly. (3) To validate the computational predictions, the local fluoride environments within the developing framework are benchmarked with experimental data obtained through various spectroscopic techniques, including Nuclear Magnetic Resonance and Raman spectroscopy. By capturing the essential energetic driving forces at a fundamental level and incorporating them into a statistical simulation, our approach aims to bridge the gap between atomistic complexity and mesoscale behavior, ultimately enabling the predictive modeling of nanoporous crystal formation through a computationally tractable methodology.