Computational methods

We formulated micromagnetic simulation models of nanocrystalline soft magnetic materials including effects of magnetostriction, and simulated motions of domain walls to clarify mechanism of energy dissipations in the nanocrystalline soft magnetic materials. The magnetostriction is nonuniformly distributed, and magnetic energy induced by external field is converted into elastic energy due to the magnetostriction. This energy consumption generates an excess loss in the nanocrystalline soft magnetic materials even there is no eddy current. The simulation results enable us to reduce the core loss in the nanocrystalline soft magnetic materials.

Weight-sharing is used to accelerate and to effectively pretrain neural network-based variational Monte Carlo methods when solving the electronic Schrödinger equation for multiple geometries.

Guided by high-throughput computational screening, we report the preparation of hydrogen-bonded cobaltoporphyrin frameworks and demonstrate the achievement of high activity and selectivity for electrochemical H₂O₂ production in acid.