Takahashi Keisuke, Satoshi Maeda
Department of Chemistry, Hokkaido University North 10, West 8 Sapporo 060-8510 Japan
Institute for Chemical Reaction Design and Discovery (WPI-ICReDD), Hokkaido University Kita 21 Nishi 10, Kita-ku Sapporo Hokkaido 001-0021 Japan
RSC Adv. 2021 Jul 1;11(38):23235-23240. doi: 10.1039/d1ra03395f.
Data science is introduced to identify the reactant, product, and reaction path in the chemical reaction network. Cobalt catalyzed hydroformylation is investigated where the reaction network is built first principles calculations. The closeness centrality and high frequency node are found to be the reactant cobalt tetracarbonyl hydride. In addition, betweenness centrality uncovers three reaction paths which have the products of aldehyde, CHO, and CO, respectively. The energy profile determines that the reaction path leading to aldehyde is energetically favored; thus, the reaction path for cobalt catalyzed hydroformylation is identified without kinetics. Hence, the proposed approach can act as a first step towards understanding the complex chemical reaction network and towards further kinetic understanding of the chemical reaction.
引入数据科学以识别化学反应网络中的反应物、产物和反应路径。研究了钴催化的氢甲酰化反应，其中反应网络是基于第一性原理计算构建的。发现接近中心性和高频节点是反应物四羰基氢化钴。此外，中介中心性揭示了三条分别生成醛、CHO和CO产物的反应路径。能量分布确定生成醛的反应路径在能量上更有利；因此，在没有动力学信息的情况下识别出了钴催化氢甲酰化反应的反应路径。因此，所提出的方法可作为理解复杂化学反应网络以及进一步从动力学角度理解化学反应的第一步。
