End-to-End Crystal Structure Prediction from Powder X-Ray Diffraction

TL;DR

This paper presents XtalNet, an innovative equivariant deep generative model that predicts detailed crystal structures directly from PXRD data, significantly advancing automated materials characterization and discovery.

Contribution

XtalNet is the first end-to-end model that uses PXRD as a condition to generate complex crystal structures, improving accuracy and automation over prior composition-only methods.

Findings

Achieves 90.2% top-10 match rate on hMOF-100 dataset.

Achieves 79% top-10 match rate on hMOF-400 dataset.

Enables direct, automated crystal structure prediction from experimental PXRD data.

Abstract

Powder X-ray diffraction (PXRD) is a prevalent technique in materials characterization. While the analysis of PXRD often requires extensive human manual intervention, and most automated method only achieved at coarse-grained level. The more difficult and important task of fine-grained crystal structure prediction from PXRD remains unaddressed. This study introduces XtalNet, the first equivariant deep generative model for end-to-end crystal structure prediction from PXRD. Unlike previous crystal structure prediction methods that rely solely on composition, XtalNet leverages PXRD as an additional condition, eliminating ambiguity and enabling the generation of complex organic structures with up to 400 atoms in the unit cell. XtalNet comprises two modules: a Contrastive PXRD-Crystal Pretraining (CPCP) module that aligns PXRD space with crystal structure space, and a Conditional Crystal Structure Generation (CCSG) module that generates candidate crystal structures conditioned on PXRD patterns. Evaluation on two MOF datasets (hMOF-100 and hMOF-400) demonstrates XtalNet's effectiveness. XtalNet achieves a top-10 Match Rate of 90.2% and 79% for hMOF-100 and hMOF-400 in conditional crystal structure prediction task, respectively. XtalNet enables the direct prediction of crystal structures from experimental measurements, eliminating the need for manual intervention and external databases. This opens up new possibilities for automated crystal structure determination and the accelerated discovery of novel materials.