Accelerating Experiments with AI and Automation: Powder Materials and their Compositional Characterization

by Andrea Giunto (University of California, Berkeley)

Thursday 15 Jan 2026, 10:00 → 11:00 Europe/Zurich

OHSA/E13

AI and computational science are fueling the hunt for high‑performance materials, yielding thousands of new inorganic candidate compounds. Yet, their experimental realization lags behind, constituting the crucial bottleneck in the material discovery pipeline. In this talk, I will highlight community efforts to overcome this gap through automated and self‑driving laboratories, including our own A‑Lab for solid-state synthesis of materials [1]. I will focus on the challenges of high‑throughput materials characterization and where AI can make the difference.

 

I will then present our work on automating the compositional characterization of powders using energy‑dispersive X‑ray spectroscopy with a scanning electron microscope (SEM‑EDS) [2]. We refined the EDS quantification schemes to obtain accurate compositional data directly from as‑deposited powders, eliminating the need for cumbersome resin embedding and polishing. When combined with automation, this approach enables rapid, large‑scale data acquisition and supports machine‑learning‑assisted identification of individual phase compositions in complex multiphase samples. Our work lays the foundation for integrating high-throughput phase‑level compositional analysis of powders into fully self‑driving laboratories, accelerating the process of material discovery and optimization.

 

Reference:

[1] Szymanski, N.J., et al. An autonomous laboratory for the accelerated synthesis of novel materials. Nature 624, 86–91 (2023). https://doi.org/10.1038/s41586-023-06734-w

[2] Giunto, A., et al. Harnessing Automated SEM-EDS and Machine Learning to Unlock High-Throughput Compositional Characterization of Powder Materials, 14 October 2025, PREPRINT [https://doi.org/10.21203/rs.3.rs-7837297/v1]