Probabilistic Phase Labeling and Lattice Refinement for Autonomous Material Research
This work addresses the bottleneck of matching XRD data analysis rates in high-throughput autonomous material discovery, offering a solution that accelerates materials identification.
The paper tackles the challenge of rapid, automated analysis of X-ray diffraction data for materials research by introducing CrystalShift, an algorithm for probabilistic phase labeling that outperforms existing methods on synthetic and experimental datasets.
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.