A research group led by Assistant Professor Ryuhei Sato of the Graduate School of Engineering at the University of Tokyo, together with Professor Shin-ichi Orimo and Professor Hao Li of Tohoku University's Advanced Institute for Materials Research(WPI-AIMR), and Professor Chris Pickard of the University of Cambridge, has successfully reproduced the formation reaction of superhydrides under high pressure using a machine learning model. They revealed a universal reaction mechanism that explains how and why pressurization accelerates the hydrogenation process at the atomic scale.
In this study, a machine learning potential capable of handling unknown reaction pathways was developed based on first-principles calculations. Using molecular dynamics simulations, the researchers reproduced the transformation reaction of calcium hydride into calcium polyhydride under high temperature and high pressure. They uncovered a reaction pathway in which the surface melts, absorbs hydrogen, and eventually solidifies, proposing a general reaction mechanism in which melting is promoted by pressure and molecular interactions.
This mechanistic insight into hydrogenation is expected to be a breakthrough in enabling the controlled synthesis of superhydrides, which contain significantly higher hydrogen content than conventional hydrides, paving the way for their application as hydrogen storage materials. Furthermore, the study serves as a pioneering example of applying machine learning to predict unknown chemical reaction pathways.
Figure: Synthesis reaction of Calcium superhydride under high pressure reproduced by machine learning potential molecular dynamics simulation.
Papers
Journal: Proceedings of the National Academy of Sciences (PNAS)
Title: Surface Melting-Driven Hydrogen Absorption for High-Pressure Polyhydride Synthesis
Authors: Ryuhei Sato*, Lewis J. Conway, Di Zhang, Chris J. Pickard, Kazuto Akagi, Kartik Sau, Hao Li, Shin-ichi Orimo*
Links
Photocatalytic solar hydrogen production from water on a 100m2-scale
Development of Self-Growing Protocell Model: Exploiting the Stabilization of Phase Separation by DNA
