CrystalGRW: generative modeling of crystal structures with targeted crystallographic properties via geodesic random walks
9 affiliations
- 1Extreme Conditions Physics Research Laboratory and Center of Excellence in Physics of Energy Materials (CE:PEM), Department of Physics, Faculty of Science, Chulalongkorn University, Bangkok, 10330, Thailand
- 2Thailand Center of Excellence in Physics, 328 Si Ayutthaya Road, Bangkok, 10400, Thailand
- 3Department of Chemistry and Biochemistry, University of South Carolina, Columbia, SC, 29208, USA. [email protected]
- 4Chula Intelligent and Complex Systems Lab, Department of Physics, Faculty of Science, Chulalongkorn University, Bangkok, 10330, Thailand
- 5Theoretical Physics Division, Department of Physics, Chemistry and Biology (IFM), Linköping University, 581 83, Linköping, Sweden
- 6Department of Chemistry and Biochemistry, University of South Carolina, Columbia, SC, 29208, USA
- 7Extreme Conditions Physics Research Laboratory and Center of Excellence in Physics of Energy Materials (CE:PEM), Department of Physics, Faculty of Science, Chulalongkorn University, Bangkok, 10330, Thailand. [email protected]
- 8Thailand Center of Excellence in Physics, 328 Si Ayutthaya Road, Bangkok, 10400, Thailand. [email protected]
- 9Chula Intelligent and Complex Systems Lab, Department of Physics, Faculty of Science, Chulalongkorn University, Bangkok, 10330, Thailand. [email protected]
Abstract
Determining whether a candidate crystalline material is thermodynamically stable depends on identifying its true ground-state structure, a central challenge in computational materials science. We introduce CrystalGRW, a diffusion-based generative model on Riemannian manifolds that proposes candidate crystal configurations in stable phases, validated through density functional theory calculations. Our model is designed for de novo generation, which creates crystal structures together with their compositions. The crystal properties, such as fractional coordinates, atomic types, and lattice matrices, are represented on suitable Riemannian manifolds, ensuring that new predictions generated through the diffusion process preserve the periodicity of crystal structures. We also incorporate an equivariant graph neural network to account for rotational and translational symmetries within the model. CrystalGRW generates crystal structures that are stable and closely resemble their density-functional-theory ground states. The model also supports conditional control, such as enforcing a specified crystallographic point group, thereby accelerating materials discovery and inverse design by providing symmetry-consistent, energetically stable candidate crystals for experimental validation.
Related
- Synthesis and Photophysical Properties of Novel 1H-Indazole-3-carbohydrazides: Computational and Crystallographic Insights
- Synergistic fabrication and investigation of electrical and dielectric properties of Fe0.01Zn0.99O@C core@shell structures: Towards an in-situ energy storage application
- Crystal Structures and Luminescence Properties of the Oxonitridolithosilicates RE ₂ [LiSi ₃ N ₅ O ₂ ] ( RE = Ce, Pr, Nd, Sm, Gd, Tb)
- Hierarchical generative modeling for the design of multi-component systems
- Continuous SUN (stable, unique, and novel) metric for generative modeling of inorganic crystals
BibTeX
@article{tangsongcharoen2026crystalgrw,
title = {{CrystalGRW: generative modeling of crystal structures with targeted crystallographic properties via geodesic random walks}},
author = {Krit Tangsongcharoen and Teerachote Pakornchote and Chayanon Atthapak and Natthaphon Choomphon-anomakhun and Annop Ektarawong and Björn Alling and Christopher A. Sutton and Thiti Bovornratanaraks and Thiparat Chotibut},
journal = {Scientific Reports},
year = {2026},
month = aug,
doi = {10.1038/s41598-026-62470-x},
url = {https://doi.org/10.1038/s41598-026-62470-x},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - CrystalGRW: generative modeling of crystal structures with targeted crystallographic properties via geodesic random walks
AU - Krit Tangsongcharoen
AU - Teerachote Pakornchote
AU - Chayanon Atthapak
AU - Natthaphon Choomphon-anomakhun
AU - Annop Ektarawong
AU - Björn Alling
AU - Christopher A. Sutton
AU - Thiti Bovornratanaraks
AU - Thiparat Chotibut
PY - 2026
DA - 2026/08/31
JO - Scientific Reports
DO - 10.1038/s41598-026-62470-x
UR - https://doi.org/10.1038/s41598-026-62470-x
AB - Determining whether a candidate crystalline material is thermodynamically stable depends on identifying its true ground-state structure, a central challenge in computational materials science. We introduce CrystalGRW, a diffusion-based generative model on Riemannian manifolds that proposes candidate crystal configurations in stable phases, validated through density functional theory calculations. Our model is designed for de novo generation, which creates crystal structures together with their compositions. The crystal properties, such as fractional coordinates, atomic types, and lattice matrices, are represented on suitable Riemannian manifolds, ensuring that new predictions generated through the diffusion process preserve the periodicity of crystal structures. We also incorporate an equivariant graph neural network to account for rotational and translational symmetries within the model. CrystalGRW generates crystal structures that are stable and closely resemble their density-functional-theory ground states. The model also supports conditional control, such as enforcing a specified crystallographic point group, thereby accelerating materials discovery and inverse design by providing symmetry-consistent, energetically stable candidate crystals for experimental validation.
SN - 2045-2322
ER -