PNNL

The current process for discovering and designing materials to meet the needs of next-generation energy storage technologies is often limited by time and resource intensive trial-and-error approaches. The use of data-driven and physics-informed machine learning methods have the promise to significantly accelerate this process by enabling the virtual screening and design of materials with targeted properties. In this talk, Emily Saldanha will highlight work performed under Pacific Northwest National Laboratory’s Energy Storage Materials Initiative to leverage such machine learning techniques to support the development process for electrolyte materials. In particular, this presentation will discuss machine learning approaches for data extraction from literature, molecular property prediction, experimental design, and molecular inverse design.