Schematic of CBPFNet. (a) Graph attention network used to learn atomic energies and forces from molecular graphs. (b) Bond-stretching protocol that generates strained configurations via a prescribed displacement rule and predicts bond peak forces without (IMAGE)

Shanghai Jiao Tong University Journal Center

Schematic of CBPFNet. (a) Graph attention network used to learn atomic energies and forces from molecular graphs. (b) Bond-stretching protocol that generates strained configurations via a prescribed displacement rule and predicts bond peak forces without

Caption

  • CBPFNet converts 3D molecular structures into spatial graphs, processing interatomic distances and chemical environments through multi-head attention layers to predict atomic energy and forces directly from non-equilibrium states.
  • The automated protocol generates highly strained configuration ensembles by rigidly separating molecular fragments along a target bond axis.
  • Tracking restoring forces across tensile displacements identifies the covalent-bond peak force, serving as a direct descriptor to predict mechanochemical regioselectivity and bond scission pathways.

Credit

Feiyu Zhou, Siwei Zhao, Haowei Wang, Feng Wang, Sida Liu, Yilun Liu, Yan Chen, Shengying Yue.

Usage Restrictions

Credit must be given to the creator.

License

CC BY

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.