Exploring Challenging C2+ Products During CO2 Reduction via Machine Learning Acceleration

Abstract Although CO 2 reduction reaction (CO 2 RR) has achieved significant progress in past years, the C 2+ products are mainly limited to a few products, while many other products have rarely been reported in experiments with a limited understanding of the underlying mechanisms. Accordingly, in this work, machine learning (ML)‐based theoretical investigations is conducted to uncover the reaction mechanisms for the conversion to challenging C₂ + products (C 2 H 6 , CH 3 OCH 3 , CH 2 CO, and C 2 H 2 ) during CO 2 RR on graphdiyne‐supported atomic catalysts (GDY‐ACs) with well‐defined active sites. Using the first‐principles machine learning (FPML) predictions, key factors limiting the diversity of C 2+ products are identified. The conversions to C 2 H 6 are mainly hindered by large rate‐determining step (RDS) barriers (>4 eV). The formation of CH 2 CO meets the competitive reactions due to similar reaction pathways with C 2 H 6 , which also undergoes further hydrogenation easily to other C 2+ products. The CH 3 OCH 3 formation is hindered by large dehydration barriers caused by steric hindrance induced by the neighboring adsorption of C 1 intermediates. FPML predictions also reveal the significance of binding configuration parameters in realizing efficient and accurate predictions. This work offers not only important references to the low selectivity of specific C 2+ products but also critical theoretical insights into CO 2 RR mechanisms.