August 5, 2025

AIMNet2-NSE: A Transferable Reactive Neural Network Potential for Open-Shell Chemistry

View Full Paper

Ask AI

Bookmark

Authors

Discussion

AIMNet2-NSE demonstrates accurate predictions in open-shell systems, highlighting its utility in radical chemistry and reaction modeling.

AIMNet2-NSE enables accurate modeling of open-shell reactions, crucial for radical chemistry applications.
It utilizes neural network potentials to distinguish between spin states, enhancing predictive accuracy.
Trained on 20 million closed-shell and 13 million open-shell configurations, AIMNet2-NSE offers significant computational efficiency.
This model improves the understanding of radical reaction pathways that are typically challenging to compute using traditional methods.

Kalita et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark