Probabilistic Phase Labeling and Lattice Refinement for Autonomous Material Research

TL;DR

CrystalShift is a novel probabilistic algorithm for XRD phase labeling that enhances robustness and efficiency, enabling faster materials discovery through improved structural analysis without extensive training.

Contribution

It introduces a symmetry-constrained pseudo-refinement and Bayesian approach for phase estimation, eliminating the need for phase space information or training data.

Findings

Outperforms existing methods on synthetic datasets

Provides robust probability estimates for phase identification

Integrates seamlessly into high-throughput workflows

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.