Materials genome engineering-based hydrogen storage materialsdatabase and its applications

Abstract

<p indent="0mm">The accelerated research and development of high-density hydrogen-storage materials are critical to the transition of the energy economy and the realization of the carbon peaking and carbon neutrality goals. The data-driven research paradigm-integrated high-throughput calculations, database, and machine learning is appealing to accelerate new material development. Due to their complex compositions, structures and morphologies, the data-driven research on hydrogen-storage materials is still limited, and a database on the comprehensive properties is necessary to be established. Herein, we show our results on establishing a Materials Genome Initiative database and the property prediction based on machine learning for hydrogen-storage materials. The datasets are constructed from published papers, Materials Genome Initiative databases, and high-throughput first-principle calculations. A hydrogen-storage materials database was established and hydrogen property predictions are carried out by using machine-learning methods. We believe that the accelerated development of new hydrogen-storage materials would be benefited from our database and platforms.