Probabilistic Phase Labeling and Lattice Refinement for Autonomous Material Research
arXiv:2308.07897 · doi:10.1038/s41524-025-01627-0
Abstract
X-ray diffraction (XRD) is an essential technique to determine a material's crystal structure in high-throughput experimentation, and has recently been incorporated in artificially intelligent agents in autonomous scientific discovery processes. However, rapid, automated and reliable analysis method of XRD data matching the incoming data rate remains a major challenge. To address these issues, we present CrystalShift, an efficient algorithm for probabilistic XRD phase labeling that employs symmetry-constrained pseudo-refinement optimization, best-first tree search, and Bayesian model comparison to estimate probabilities for phase combinations without requiring phase space information or training. We demonstrate that CrystalShift provides robust probability estimates, outperforming existing methods on synthetic and experimental datasets, and can be readily integrated into high-throughput experimental workflows. In addition to efficient phase-mapping, CrystalShift offers quantitative insights into materials' structural parameters, which facilitate both expert evaluation and AI-based modeling of the phase space, ultimately accelerating materials identification and discovery.
arXiv comment13 pages, 6 figures
References in corpus (7)
- On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning
- Complex solid solution electrocatalyst discovery by prediction and high-throughput experimentation
- A probabilistic deep learning approach to automate the interpretation of multi-phase diffraction spectra
- Crystallography companion agent for high-throughput materials discovery
- Autonomous synthesis of metastable materials
- On Identifiability of Nonnegative Matrix Factorization
- Autonomous decision making for solid-state synthesis of inorganic materials