脱氢
催化作用
丙烷
化学
Atom(片上系统)
氧化物
选择性
化学物理
过渡金属
多相催化
物理化学
协调数
结晶学
光化学
有机化学
离子
嵌入式系统
计算机科学
作者
Ping Hu,Qingyu Chang,Wei Zhang,Minglei Yang,Ming Lei,De Chen,Xinggui Zhou,Zhi‐Jun Sui,Yi‐An Zhu
标识
DOI:10.1016/j.jcat.2024.115299
摘要
Tailoring atomically dispersed catalysts by various methods has been the subject of intense current research in heterogeneous catalysis. In this work, the structural stability and propane dehydrogenation performance of Pt-TiO2 catalysts with single, dual, and triple Pt atoms doped on the reduced TiO2 surface have been examined. Experimental results indicate the atomically dispersed Pt is thermally very stable and shows exceptionally high activity (TOFpropane = 4.93 s−1) and selectivity (∼98.5 %) for the dehydrogenation reaction. The recently developed machine-learning-based SSW-NN method is then employed to explore the global potential energy surface of the atomically dispersed Pt atom-doped TiO2, which suggests that the single and dual Pt atoms can be tightly held by the vacancies created on the reduced transition-metal oxide, as confirmed by DFT + U calculated energetics. After that, advanced characterization techniques are used in conjunction with microkinetic analysis to establish the quantitative structure–activity relationships. It is found that the dual-Pt-atom structure with two Pt atoms embedded in adjacent Ti and O vacancies is the only active ensemble that dramatically promotes the CH bond activation and hydrogen recombination, as compared to single Pt atoms accommodated by either Ti or O vacancies. These results provide new and direct evidence in support of the idea that specific function can be imparted to specific active sites by tuning their coordination environments, which makes it possible to tailor the catalytic properties of metal oxides to a particular reaction such as propane dehydrogenation.
科研通智能强力驱动
Strongly Powered by AbleSci AI