In-situ synchrotron X‑ray instruments perform material characterization to determine properties such as phase nucleation and transformation under controlled heating. However, the complexity and amount of data from synchrotron X-ray diffraction (XRD) make the analysis challenging. This project develops a high-throughput computational workflow for automated extraction of key structural features from XRD data, including crystallinity, peak parameters, and phase-transition temperatures.

The SUDS Scholar will apply data science approaches such as distribution modeling and signal processing, as well as supervised/unsupervised machine learning methods, to evaluate physics-based candidate features and indicators. After identifying a workflow, students will work to automate the analysis for compatibility with high-throughput experimentation, identifying the phase evolution processes and corresponding structure information. Students will also practice software engineering skills necessary to document the workflow in an open-science framework. Final outcomes include open, reproducible analysis that accelerates materials discovery and demonstrates core data science competencies: algorithm design, scalable computing, and automated knowledge extraction.