Accelerated experimental materials discovery through integration with theory and artificial intelligence

Abstract

Solar photochem. is a promising energy technol. limited by the performance and compatibility of its component materials. A variety of approaches for discovering solar fuels materials have been and continue to be deployed, with photoactive components proving to be particularly difficult to identify due to their need to simultaneously exhibit a broad range of chem. and phys. properties. The development of high throughput screening techniques in the Joint Center for Artificial Photosynthesis has opened new avenues for materials discovery with unique challenges and opportunities. Most notably, rapid synthesis and screening enables evaluation of candidate materials proposed by high throughput computational screening, and some expts. produce datasets that are too large and complex to be interpreted by std. techniques or even expert materials scientists, motivating the development of artificial intelligence algorithms. Integration of the resp. computational techniques with expts. accelerates both materials discovery and understanding of known materials, creating new paradigms for conducting materials research and developing multi-functional materials.