Machine-Learning-Augmented Chemisorption Model for CO2 Electroreduction Catalyst Screening

详细信息    查看全文

• 作者：Xianfeng Ma ; Zheng Li ; Luke E. K. Achenie ; Hongliang Xin
• 刊名：The Journal of Physical Chemistry Letters
• 出版年：2015
• 出版时间：September 17, 2015
• 年：2015
• 卷：6
• 期：18
• 页码：3528-3533
• 全文大小：382K
• ISSN：1948-7185

文摘

We present a machine-learning-augmented chemisorption model that enables fast and accurate prediction of the surface reactivity of metal alloys within a broad chemical space. Specifically, we show that artificial neural networks, a family of biologically inspired learning algorithms, trained with a set of ab initio adsorption energies and electronic fingerprints of idealized bimetallic surfaces, can capture complex, nonlinear interactions of adsorbates (e.g., *CO) on multimetallics with 鈭?.1 eV error, outperforming the two-level interaction model in prediction. By leveraging scaling relations between adsorption energies of similar adsorbates, we illustrate that this integrated approach greatly facilitates high-throughput catalyst screening and, as a specific case, suggests promising {100}-terminated multimetallic alloys with improved efficiency and selectivity for CO₂ electrochemical reduction to C₂ species. Statistical analysis of the network response to perturbations of input features underpins our fundamental understanding of chemical bonding on metal surfaces.