Prediction of Metal–Organic Framework-Supported CuSn Double-Atom Catalysts with Decoupled Geometric/Electronic Descriptors for Targeting Electrosynthesis toward Ethanol

Two-dimensional metal–organic frameworks (2D MOFs) can serve as an ideal platform for targeting electrosynthesis toward ethanol via a carbon dioxide reduction reaction (CO 2 RR). In this work, the CO 2 RR performance on 18 distinct MOFs with the active phase “–CuSnN 2 X 1 X 2 –” (X 1 and X 2 = N, O, and S) was systematically investigated, aimed at achieving enhanced electrocatalytic performance and elucidating the structure–performance relationship. Combining density functional theory calculations and machine learning techniques, we identified two critical factors influencing catalytic performance: the microstructure of the active phase and the size of the ligand, which represent the inner and outer coordination spheres, respectively. Our findings revealed that the incorporation of O/S atoms boosts activity, while ligand elongation improves selectivity. These effects are linked to specific geometric descriptors: the perimeter of the CuX 1 X 2 triangle and the Cu–Sn distance. By decoupling the interplay between geometric and electronic properties, we further interpreted these descriptors through the carried charge of the active phase. Notably, the oxidation state of the entire CuX 1 X 2, rather than individual Cu or Sn, plays a decisive role in activity. It was found that the “–(CuSnN 2 S 2 )–” phase shows optimal performance, featuring a wide potential window, low barriers for C–C coupling and potential-determining steps, high stability under the working conditions, and strong synthesizability in a lab. Our work highlights the importance of optimizing the microenvironment of the active phase, offering valuable insights for the rational design of high-efficiency electrocatalysts toward C 2 H 5 OH production.