AI-assisted X-ray absorption spectral analysis: Data reproducibility, database, and machine Learning

X-ray absorption spectroscopy (XAS) is a premier element-specific experimental technique widely used for materials characterization in battery research. XAS encodes rich local structural and chemical information around X-ray absorbing species, however such information is convoluted in an abstract form, making it very difficult and time consuming to analyze. In this talk, we will discuss the emerging opportunities in AI-assisted XAS analysis, which leverages first-principles theory, high-throughput computing and data analytics. Specifically, we will highlight key areas in this approach, including pipelines to ensure data quality and reproducibility, spectral database development, and machine learning applications. The AI-assisted analysis pipeline can enable real-time feedback in high-throughput studies and autonomous experimentation.