Xue Zhe, Tan Rui, Wang Hongxia, Tian Jinzhong, Wei Xiaolin, Hou Hua, Zhao Yuhong
School of Materials Science and Engineering, Collaborative Innovation Center of Ministry of Education and Shanxi Province for High-performance Al/Mg Alloy Materials, North University of China, Taiyuan 030051, China.
Physics and Electronic Engineering, Hengyang Normal University, Hengyang 421002, China.
J Colloid Interface Sci. 2023 Dec;651:149-158. doi: 10.1016/j.jcis.2023.07.128. Epub 2023 Jul 28.
Single-atom catalysts with particular electronic structures and precisely regulated coordination environments delivering excellent activity for oxygen-evolution reaction (OER) and oxygen-reduction reaction (ORR) are highly desirable for renewable energy applications. In this work, a novel tetragonal carbon nitride T-CN monolayer with remarkable stability was predicted by using the RG method. Inspired by the well-defined atomic structures and just right N aperture of T-CN substrate, the electrocatalytic performance of a series of transition metal single-atoms anchored on porous T-CN matrix (TM@CN) have been systematically investigated. In addition, machine learning (ML) method was employed with the gradient boosting regression GBR model to deeply explore the complex controlling factors and offer direct guidance for rational discovery of desirable catalysts. On this basis, the coordination environment of the central TM active sites has been tailored by incorporating heteroatoms. Impressively, the Co@CN/B-C, Rh@CN/SC and Rh@CN/SN exhibit significantly enhanced OER/ORR activity with notably low η/η of 0.39/0.32, 0.26/0.35 and 0.37/0.27 V, respectively. Our work provides insights into the rational design, data-driven, performance regulation, mechanism analysis and practical application of TMNC catalysts. Such a systematic theoretical framework can also be expanded to many other kinds of catalysts for energy storage and conversion.
具有特定电子结构和精确调控配位环境、对析氧反应(OER)和氧还原反应(ORR)具有优异活性的单原子催化剂在可再生能源应用中极具吸引力。在这项工作中,使用RG方法预测了一种具有显著稳定性的新型四方相氮化碳T-CN单层。受T-CN基底明确的原子结构和合适的N孔径启发,系统研究了一系列锚定在多孔T-CN基质上的过渡金属单原子(TM@CN)的电催化性能。此外,采用机器学习(ML)方法和梯度提升回归GBR模型深入探索复杂的控制因素,并为合理发现理想催化剂提供直接指导。在此基础上,通过引入杂原子对中心TM活性位点的配位环境进行了调整。令人印象深刻的是,Co@CN/B-C、Rh@CN/SC和Rh@CN/SN分别表现出显著增强的OER/ORR活性,其η/η分别低至0.39/0.32、0.26/0.35和0.37/0.27 V。我们的工作为TMNC催化剂的合理设计、数据驱动、性能调控、机理分析和实际应用提供了见解。这样一个系统的理论框架也可以扩展到许多其他用于能量存储和转换的催化剂。