XQueryer—Intelligent Phase Identification for Powder X-Ray Diffraction

Artificial intelligence (AI) is revolutionizing material science, offering unprecedented capabilities for analyzing and predicting material properties. Powder X-ray diffraction (PXRD), a standard technique for crystal identification, now faces a crucial transformation. With the rapid advancement of AI technologies, there is a growing demand to enhance traditional PXRD-based analyses, transforming them into intelligent structure analyzers to fully harness the opportunities in AI-driven research. In this paper, we present XQueryer, a novel machine learning model designed for intelligent structure identification from PXRD data. XQueryer outperforms current models and traditional search-matching methods by seamlessly integrating with existing hardware systems and offering superior accuracy. This model is capable of identifying crystal structures across the entire Materials Project database, ensuring wide applicability and precision. To support this new interdisciplinary approach, we have developed a comprehensive PXRD database and a machine learning pipeline. We provide the first extensive benchmark for PXRD-based crystal identification, comparing 11 different baseline models. Furthermore, we offer a free online platform powered by XQueryer, allowing users to perform crystal structure analysis using experimental PXRD data with ease.