Machine learning assisted binary alloy catalyst design for the electroreduction of CO₂ to C₂ products

Abstract

The carbon dioxide reduction reaction (CO2RR) has become one of the most important catalytic reactions due to its potential impact on global emissions. Among the many products this reaction yields,...

The Synergistic Effect between Metal and Sulfur Vacancy to Boost CO2 Reduction Efficiency: A Study on Descriptor Transferability and Activity Prediction

“…Among them, some electronic structure descriptors need the aid of density functional theory (DFT) calculations. Instead, there are also some easily available descriptors, such as d-shell valence electron numbers, electron type and number, electronegativity, enthalpy of vaporization, coordination number, and coordination bond length of metal atom to the nearest neighbor atoms. − Moreover, some DFT-calculated parameters, such as charge transfer, catalyst mode, and work function, could be replaced by easily available descriptors. − As an illustration, the charge transfer between the catalyst and the adsorbates can be predicted by the formulated atomic ionization energy and electronegativity parameters . To the best of our knowledge, the ML model for describing the interplay between the vacancy and the metal sites has rarely reported.…”

Local descriptors-based machine learning model refined by cluster analysis for accurately predicting adsorption energies on bimetallic alloys

“…Descriptors based on averaged properties emphasize the signicance of incorporating details about each metal composing the alloy. [30][31][32][33] Furthermore, including a combination of semi-empirical parameters, such as geometric and tabulated atomic information, capable of differentiating between different surface alloys, enhances accuracy and robustness. [34][35][36] Similarly, predictive models incorporating electronic properties such as the d-band center, [37][38][39][40] Bader charges, 41 and surface energy 42,43 display better accuracy in capturing bonding interactions.…”

Analyzing the Active Site and Predicting the Overall Activity of Alloy Catalysts

“…In contrast, although the modeling strategy involving atomically disordered configurations is more complex, it more closely resembles real-world scenarios and facilitates in-depth analysis of the ALS. Statistical methods, such as analyzing the distribution of adsorption energies, − correlation matrices, and charge differences between sites, have been shown to reveal the relationship between the ALS and adsorption strength from complex results calculated based on disordered alloy surfaces. Finally, quantifying the overall activity of the alloy is a crucial step.…”