Metal-organic frameworks (MOFs) have been recognized as one of the ideal supporting sintering-resistant catalyst materials because of their high specific surface area and unique nano-porous structure capable of manipulating the sizes and shapes of encapsulated metal clusters. To explore the binding sites, the stability, and migration mechanisms of encapsulated metal clusters in MOF materials, a robust potential model that accurately describes the interaction between metal clusters and MOF materials is highly desired for large-scale atomic simulations. Herein, as a demonstration case, an artificial neural network potential for encapsulated platinum (Pt) clusters in MOF-808 was developed using the machine learning-based global neural network (G-NN) technique. The artificial G-NN potential was tested and validated against a series of density functional theory calculation data, including structure optimization, adsorption energies, and the migration energy barrier of Pt n (n = 1–13) clusters in MOF-808. The newly developed Pt-MOF G-NN potential was further used to predict the adsorption and migration behaviors of Pt n clusters in MOF-808. It is found that the most stable adsorption site varies with the Pt n cluster size. The migration possibility of the Pt n cluster is strongly correlated with the adsorption energies of the Pt n clusters. Finally, the CO adsorption on the single Pt atom would effectively promote the aggregation of Pt n clusters via the Ostwald ripening mechanism.

Electrochemical CO2 reduction to transportation fuels and valuable platform chemicals provides a sustainable avenue for renewable energy storage and realizes an artificially closed carbon loop. However, the rational design of highly active and selective CO2 reduction electrocatalysts remains a challenging task. Herein, a series of metal–organic framework (MOF)-supported flexible, self-adaptive dual-metal-site pairs (DMSPs) including 21 pairwise combinations of six transition metal single sites (MOF-808-EDTA-M1M2, M1/M2 = Fe, Cu, Ni, Pd, Pt, Au) for the CO2 reduction reaction (CO2RR) were theoretically screened using density functional theory calculations. Against the competitive hydrogen evolution reaction, MOF-808-EDTA-FeFe and MOF-808-EDTA-FePt were identified as the promising CO2RR electrocatalysts toward C1 and C2 products. The calculated limiting potential for CO2 electroreduction to C2H6 and C2H5OH over MOF-808-EDTA-FeFe is −0.87 V. Compared with an applied potential of −0.56 eV toward CH4 production over MOF-808-EDTA-FeFe, MOF-808-EDTA-FePt exhibits an even better activity for CO2 reduction to C1 products at a limiting potential of −0.35 V. The present work not only identifies promising candidates for highly selective CO2RR electrocatalysts leading to C1 and C2 products but also provides mechanistic insights into the dynamic nature of DMSPs for stabilizing various reaction intermediates in the CO2RR process.

The distribution of aluminum (Al) on the H-form zeolite framework strongly affects the Brønsted acidity, resulting in the corresponding acid-catalyzed reactivity of zeolite catalysts. In the present work, the effects of next-nearest-neighbor (NNN) Al locations and numbers on the acidity for the specific Brønsted acid sites (BAS) inside the channel and on the external surface of the HY zeolites with two Si/Al ratios were investigated using density functional theory calculations. The Gibbs free energy of ammonia adsorption at the BAS site with different local NNN Al environments was used to characterize the Brønsted acidity. It has been found that an NNN Al atom on the β cage slightly enhances the acidity, while NNN Al atoms at the hexagonal column and super cages significantly decrease the acidity. With the increasing number of NNN Al atoms, the acidity of the specific BAS becomes weaker. The effects of the NNN Al location and numbers on the Brønsted acidity are further confirmed using pyridine, indole, and quinoline as probe molecules.

The search for new catalytic agents for reducing excess CO2 in the atmosphere is a challenging but essential task. Due to the well-defined porous structures and unique physicochemical properties, metal–organic frameworks (MOFs) have been regarded as one of the promising materials in the catalytic conversion of CO2 into valuable platform chemicals. In particular, introducing the second metal (M) atom to form the MII–O–Zr4+ single-atom metal sites on the Zr nodes of MOF-808 would further greatly improve the catalytic performance. Herein, CO2 hydrogenation reaction mechanisms and kinetics over a series of MOF-808-encapsulated single-atom metal catalysts, i.e., MII–MOF-808 (MII = CuII, FeII, PtII, NiII, and PdII), were systematically studied using density functional theory calculations. First, it has been found that the stability for the encapsulation of a divalent metal ion follows the trend of PtII > NiII > PdII > CuII > FeII, while they all possess moderate anchoring stability on the MOF-808 with the Gibbs replacement energies ranging from −233.7 to −310.3 kcal/mol. Two plausible CO2 hydrogenation pathways on CuII–MOF-808 catalysts, i.e., formate and carboxyl routes, were studied. The formate route is more favorable, in which the H2COOH*-to-H2CO* step is kinetically the most relevant step over CuII–MOF-808. Using the energetic span model, the relative turnover frequencies of CO2 hydrogenation to various C1 products over MII–MOF-808 were calculated. The CuII–MOF-808 catalyst is found to be the most active catalyst among five MII–MOF-808 catalysts.