Atomic structure and chemical bonding analysis using artificial intelligence

1.7 Quantum computing
3.2 Mathematical and physical sciences
3.3 Engineering
3.4 Chemistry

Teruyasu Mizoguchi

Institute of Industrial Science

Professor

Atomic structures and the chemical bonding of materials play crucial roles for their functions. The analysis of those atomic and electronic structures has been performed using time-consuming simulations based on quantum chemistry theory. In this study, we are developing artificial intelligence-aided methodology to accelerate the atomic and electronic structure analysis.

Schematic illustration of chemical bonding prediction using artificial intelligence
Teruyasu Mizoguchi

Prediction of bonding strength, bond length, Fermi energy, and covalency
Teruyasu Mizoguchi

Research collaborators

National Institute of Advanced Industrial Science and Technology (AIST)

Related publications

"Accurate prediction of bonding properties by a machine learning–based model using isolated states before bonding",E. Suzuki, K. Shibata, and T. Mizoguchi,Appl. Phys. Exp. 14 (2021) 085503-1-6 doi: 10.35848/1882-0786/ac083b
"Dataset on structure and physical properties of stable diatomic systems based on van der Waals density functional method",K. Shibata, E. Suzuki, and T. Mizoguchi,Data in Brief, 36 (2021) 106968.
"Quantum Deep Field: Data-Driven Wave Function, Electron Density Generation, and Atomization Energy Prediction and Extrapolation with Machine Learning",M. Tsubaki and T. Mizoguchi,Phys. Rev. Lett., 125 (2020) 206401-1-6.
"Fast and Accurate Molecular Property Prediction: Learning Atomic Interactions and Potentials with Neural Networks",M. Tsubaki and T. Mizoguchi,J.Phys. Chem. Lett., 9 (2018), 5733-5741.