AlphaDiffract: automated crystallographic analysis of powder X-ray diffraction data
Materials identification and structural understanding from powder X-ray diffraction (PXRD) data are long-standing challenges in materials science, fundamental to discovering and characterizing novel materials. A prerequisite for full structure solution is the accurate determination of the crystal lattice, including lattice parameters and crystallographic symmetries. Traditional methods for this are iterative and typically require expert input, and while existing deep learning approaches have shown promise, a robust, single-shot method for comprehensive lattice determination from experimental data remains a key goal. Here, we introduce AlphaDiffract, a deep learning framework for jointly predicting the crystal system, space group, and lattice parameters directly from PXRD patterns. AlphaDiffract utilizes a 1D adaptation of the ConvNeXt architecture, a modern convolutional neural network that integrates key design principles from transformers, coupled with dedicated prediction heads for each crystallographic property. The model is trained and validated on the largest-to-date physics-based dataset of over 31 million simulated diffraction patterns, generated by augmenting 312,267 curated structures from the ICSD and Materials Project databases. Crucially, it demonstrates strong transfer from simulated training data to experimental measurements, achieving 81.7% crystal system accuracy and 66.2% space group accuracy on the RRUFF dataset while additionally predicting all six lattice parameters. By providing a unified model for rapid, simultaneous prediction of lattice symmetry and parameters from PXRD data, AlphaDiffract represents a significant step forward in leveraging deep learning for high-throughput materials discovery.
Authors
- Nina Andrejevic (ORCID: https://orcid.org/0000-0002-8648-5859)
- James P. Horwath (ORCID: https://orcid.org/0000-0002-7333-3768)
- Mathew J. Cherukara (ORCID: https://orcid.org/0000-0002-1475-6998)
- Michael H. Prince (ORCID: https://orcid.org/0000-0002-4196-3925)
- Brian H. Toby (ORCID: https://orcid.org/0000-0001-8793-8285)
- Aileen Luo
- Hemant Sharma (ORCID: https://orcid.org/0000-0002-0800-9275)
- Ming Du
- Xiangyu Yin
Institutions
- Argonne National Laboratory (US)
- Advanced Photon Source (US)
Publication Details
- Journal
- npj Computational Materials
- Published
- 2026-11-15
- DOI
- https://doi.org/10.1038/s41524-026-02388-0
- Citations
- 4
- Primary Topic
- Machine Learning in Materials Science
- Type
- article
- Field-Weighted Citation Impact
- 3.54
Funders
- U.S. Department of Energy
- University of Chicago
- Office of Science
- Basic Energy Sciences
- Argonne National Laboratory