Hybrid physics-machine learning models for quantitative electron diffraction refinements
Open Access
- 11 April 2026
- journal article
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Abstract
High accuracy electron microscopy simulations required for quantitative crystal structure refinements face a fundamental challenge: while physical interactions are well-described theoretically, real-world experimental effects are challenging to model analytically. To address this gap, we present a hybrid physics-machine learning framework that integrates differentiable physical simulations with neural networks. By leveraging automatic differentiation throughout the simulation pipeline, our method enables gradient-based joint optimization of physical parameters and neural network components representing experimental variables, offering superior scalability compared to traditional second-order methods. We demonstrate this framework through application to three-dimensional electron diffraction (3D-ED) structure refinement, where our approach learns complex thickness distributions directly from diffraction data rather than relying on simplified geometric models. This method achieves state-of-the-art refinement performance across synthetic and experimental datasets, recovering atomic positions, thermal displacements, and thickness profiles with high fidelity. The modular architecture proposed can naturally be extended to accommodate additional physical phenomena and extended to other electron microscopy techniques. This establishes differentiable hybrid modeling as a powerful paradigm for quantitative electron microscopy, where experimental complexities have historically limited analysis.Keywords
All Related Versions
Funding Information
- Engineering and Physical Sciences Research Council (EP/S024050/1)
- Engineering and Physical Sciences Research Council (EP/V007750/1)
- Engineering and Physical Sciences Research Council (EP/R008779/1)
- Alan Turing Institute (EP/V030302/1)
- Schmidt Science Fellows
This publication has 38 references indexed in Scilit:
- Hirshfeld atom refinement on 3D electron diffraction data: the path to a free refinement of hydrogen atomsJournal of Molecular Structure, 2025
- 3D Electron Diffraction on Nanoparticles: Minimal Size and Associated Dynamical EffectsACS Nano, 2025
- Machine Learning Interatomic Potentials as Emerging Tools for Materials ScienceAdvanced Materials, 2019
- 3D Electron Diffraction: The Nanocrystallography RevolutionACS Central Science, 2019
- Advanced Computing in Electron Microscopy2010
- The potential and limitations of neutrons, electrons and X-rays for atomic resolution microscopy of unstained biological moleculesQuarterly Reviews of Biophysics, 1995
- Electron Microdiffraction1992
- Automated structure factor refinement from convergent-beam patternsUltramicroscopy, 1991
- Multilayer feedforward networks are universal approximatorsNeural Networks, 1989
- Theorie der Beugung von Elektronen an KristallenAnnalen der Physik, 1928