Equivariant Many-body Message Passing Interatomic Potentials for Magnetic Materials
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918436818386944 |
|---|---|
| author | Ho, Cheuk Hin van der Oord, Cas Darby, James P. Keane, Theo Benson, Raz L. Espinoza, Cristian Rebolledo Kulkarni, Rutvij Spinu, Elina Papanikolaou, Michail Tomsett, Richard Forrest, Robert M. Bean, Jonathan J. Csányi, Gábor Ortner, Christoph |
| author_facet | Ho, Cheuk Hin van der Oord, Cas Darby, James P. Keane, Theo Benson, Raz L. Espinoza, Cristian Rebolledo Kulkarni, Rutvij Spinu, Elina Papanikolaou, Michail Tomsett, Richard Forrest, Robert M. Bean, Jonathan J. Csányi, Gábor Ortner, Christoph |
| contents | Magnetism governs key properties of materials used in energy, data storage, and spintronic technologies, yet its complex coupling to lattice and electronic degrees of freedom challenges conventional first-principles approaches. We introduce an equivariant message-passing graph neural network that embeds atomic magnetic moments as explicit degrees of freedom, enabling the learning of magnetic interactions beyond collinear approximations. The model learns physically consistent and transferable representations of magnetic behaviour and can incorporate spin-orbit coupling, achieving near density-functional-theory accuracy with strong data efficiency across diverse magnetic systems by fine-tuning from a pre-trained model. Applications to structural transformations, finite-temperature magnetic phenomena, and materials screening for strongly spin-orbit coupled materials demonstrate transferable magnetic behaviour, establishing a practical foundation for data-driven, high-throughput discovery of complex magnetic materials. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_08143 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Equivariant Many-body Message Passing Interatomic Potentials for Magnetic Materials Ho, Cheuk Hin van der Oord, Cas Darby, James P. Keane, Theo Benson, Raz L. Espinoza, Cristian Rebolledo Kulkarni, Rutvij Spinu, Elina Papanikolaou, Michail Tomsett, Richard Forrest, Robert M. Bean, Jonathan J. Csányi, Gábor Ortner, Christoph Materials Science Magnetism governs key properties of materials used in energy, data storage, and spintronic technologies, yet its complex coupling to lattice and electronic degrees of freedom challenges conventional first-principles approaches. We introduce an equivariant message-passing graph neural network that embeds atomic magnetic moments as explicit degrees of freedom, enabling the learning of magnetic interactions beyond collinear approximations. The model learns physically consistent and transferable representations of magnetic behaviour and can incorporate spin-orbit coupling, achieving near density-functional-theory accuracy with strong data efficiency across diverse magnetic systems by fine-tuning from a pre-trained model. Applications to structural transformations, finite-temperature magnetic phenomena, and materials screening for strongly spin-orbit coupled materials demonstrate transferable magnetic behaviour, establishing a practical foundation for data-driven, high-throughput discovery of complex magnetic materials. |
| title | Equivariant Many-body Message Passing Interatomic Potentials for Magnetic Materials |
| topic | Materials Science |
| url | https://arxiv.org/abs/2604.08143 |