AIMNet2‐NSE: A Transferable Reactive Neural Network Potential for Open‐Shell Chemistry
Bhupalee Kalita, Roman Zubatyuk, Dylan M. Anstine, Maike Bergeler, Volker Settels, Conrad Stork, Sebastian Spicher, Olexandr Isayev
Highlight
Abstract Open‐shell systems such as radical intermediates are central to radical polymerization (RP), combustion, catalysis, and many other chemical and industrial processes, yet their accurate modeling presents significant computational challenges.
Abstract
Abstract Open‐shell systems such as radical intermediates are central to radical polymerization (RP), combustion, catalysis, and many other chemical and industrial processes, yet their accurate modeling presents significant computational challenges. Most of the current machine learning interatomic potentials do not distinguish between different spin states, making them unsuitable for open‐shell reactive chemistry. Here we present AIMNet2‐NSE (neural spin‐charge equilibration), a neural network potential that incorporates spin‐charge equilibration for accurate treatment of molecules and reactions with arbitrary charge and spin multiplicities. Built upon the AIMNet2 framework, AIMNet2‐NSE is trained on an extensive dataset comprising 20 million closed‐shell neutral and charged molecules, 13 million open‐shell radical configurations, and 200K radical reaction profiles. With explicit handling of spin charges, AIMNet2‐NSE enables prediction of spin‐resolved properties with near‐DFT accuracy while maintaining a favorable linear scaling compared to the polynomial scaling of electronic structure methods. The predictive capabilities and generalizability of our model are confirmed by evaluations on large‐scale radical test sets, the industrially relevant BASChem19 benchmark, and RP reactions. Overall, AIMNet2‐NSE represents a significant advancement in machine learning interatomic potentials, allowing efficient exploration of complex open‐shell systems, and significantly advancing our ability to model radical reaction pathways and reactive intermediates in chemical processes where traditional quantum mechanical methods are computationally prohibitive.
Keywords
Cite This Paper
@article{Kalita2025b,
author = {Kalita, Bhupalee and Zubatyuk, Roman and Anstine, Dylan M. and Bergeler, Maike and Settels, Volker and Stork, Conrad and Spicher, Sebastian and Isayev, Olexandr},
title = {AIMNet2‐NSE: A Transferable Reactive Neural Network Potential for Open‐Shell Chemistry},
year = {2025},
journal = {Angewandte Chemie},
doi = {10.1002/ange.202516763},
url = {http://dx.doi.org/10.1002/ange.202516763},
publisher = {Wiley},
keywords = {neural network, machine learning},
researchAreas = {ml-potentials, quantum-chemistry, reactions-reactivity},
highlight = {Abstract Open‐shell systems such as radical intermediates are central to radical polymerization (RP), combustion, catalysis, and many other chemical and industrial processes, yet their accurate modeling presents significant computational challenges.}
} Copied to clipboard!
Related Research Areas
Related Publications
Teaching a neural network to attach and detach electrons from molecules
Nature Communications 12
Abstract Interatomic potentials derived with Machine Learning algorithms such as Deep-Neural Networks (DNNs), achieve the accuracy of high-fidelity quantum mechanical (QM) methods in areas traditionally dominated by empirical force fields and allow performing massive simulations.
AIMNet2‐NSE: A Transferable Reactive Neural Network Potential for Open‐Shell Chemistry
Angewandte Chemie International Edition
Abstract Open‐shell systems such as radical intermediates are central to radical polymerization (RP), combustion, catalysis, and many other chemical and industrial processes, yet their accurate modeling presents significant computational challenges.
Machine Learning-Accelerated Screening of Hydroquinone Analogs for Proton-Coupled Electron Transfer
Proton-coupled electron transfer (PCET) mediated by hydroquinone and related molecules is key to natural and artificial energy conversion.
Approaching coupled cluster accuracy with a general-purpose neural network potential through transfer learning
Nature Communications 10
Abstract Computational modeling of chemical and biological systems at atomic resolution is a crucial tool in the chemist’s toolset.
ANI-1, A data set of 20 million calculated off-equilibrium conformations for organic molecules
Scientific Data 4
AbstractOne of the grand challenges in modern theoretical chemistry is designing and implementing approximations that expedite ab initio methods without loss of accuracy.