MaterialEyes – Looking into Renewable Energy Materials with Theory- and AI-Guided Characterization

The understanding and design of materials for sustainability -- such as photovoltaics, optoelectronics, and energy storage materials -- rely on measurements not just of the functional properties, but also of fundamental structural and electronic properties. Data driven approaches such as machine learning (ML) for interpreting these measurements via microscopy and spectroscopy rely on labeled, reliable, balanced training data. We will discuss the use of AI including computer vision and large language models (LLMs) to obtain such training data from literature, and also the use of theory-guided AI approaches to interpret new microscopy and spectroscopy data. We will also discuss the use of generative AI to understand interfaces and structural evolution in renewable energy materials.