Machine learning magnetic interactions from neutron powder diffraction data
Abstract
Neutron diffraction is a versatile experimental technique capable of probing a material's magnetic properties. While diffraction is typically used to determine the magnetic structure of a material, magnetic diffuse scattering data from a diffraction experiment are also sensitive to the magnetic interactions in its Hamiltonian. However, accurately determining magnetic interaction parameters from neutron-scattering data involves an inverse scattering problem that is challenging to solve in general. Here, we investigate the effectiveness of a machine learning approach to predict the interaction parameters given magnetic diffuse-scattering data measured on powder samples, for a comprehensive survey of isotropic interactions on eight high-symmetry lattices. Across all lattices we considered, the machine-learning approach estimates the interaction parameters with high (~2%) accuracy, while avoiding the issue of false minima that is encountered with non-linear least squares refinement. Our results highlight that powder diffuse-scattering data can provide a compact "fingerprint" of the magnetic interactions for many materials.
Related
- Water, vacancies, and competing exchange interactions in Prussian blue analogues: a neutron diffraction study of field and dehydration-driven magnetic transitions
- Polarized neutron diffraction with ex-situ 3He neutron spin filter and two-dimensional detector for noncollinear incommensurate magnetic order
- Field evolution of the multi-k magnetic states in NdB4 revealed by neutron diffraction
- Magnetic-field-induced detwinning of orientational domains in altermagnet α-MnTe: A single-crystal neutron diffraction study
- Scalable machine learning framework for multiphase identification from powder X-ray diffraction
BibTeX
@misc{desai2026machine,
title = {{Machine learning magnetic interactions from neutron powder diffraction data}},
author = {Adit S. Desai and Yongqiang Cheng and Joseph A. M. Paddison},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.21970},
eprint = {2609.21970},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.21970},
}
RIS
TY - UNPB
TI - Machine learning magnetic interactions from neutron powder diffraction data
AU - Adit S. Desai
AU - Yongqiang Cheng
AU - Joseph A. M. Paddison
PY - 2026
DA - 2026/09/18
PB - arXiv
DO - 10.48550/arxiv.2609.21970
UR - https://arxiv.org/abs/2609.21970
AB - Neutron diffraction is a versatile experimental technique capable of probing a material's magnetic properties. While diffraction is typically used to determine the magnetic structure of a material, magnetic diffuse scattering data from a diffraction experiment are also sensitive to the magnetic interactions in its Hamiltonian. However, accurately determining magnetic interaction parameters from neutron-scattering data involves an inverse scattering problem that is challenging to solve in general. Here, we investigate the effectiveness of a machine learning approach to predict the interaction parameters given magnetic diffuse-scattering data measured on powder samples, for a comprehensive survey of isotropic interactions on eight high-symmetry lattices. Across all lattices we considered, the machine-learning approach estimates the interaction parameters with high (~2%) accuracy, while avoiding the issue of false minima that is encountered with non-linear least squares refinement. Our results highlight that powder diffuse-scattering data can provide a compact "fingerprint" of the magnetic interactions for many materials.
ER -