Accelerated development of energy materials through automation and artificial intelligence

Public summary

* Self-driving laboratories integrate AI, automation, and computation to accelerate energy materials discovery.

* Human–AI collaboration and physical insights can transform energy materials discovery.

* Closing experimental and instrumental bottlenecks is now the key barrier to fully autonomous materials discovery.

* Inverse design with better and reproducible data can bridge computation and experiments to accelerate discovery.

* Open and affordable AI and automation tools can broaden access to materials innovation.

Abstract

Batteries, fuel cells, and electrolyzers mostly rely on advances in solid-state inorganic materials, yet discovering and optimizing these materials remains a complex and time-intensive challenge. Self-driving laboratories (SDLs), through advances in computation, automation, and artificial intelligence (AI), are revolutionizing the pace of solid-state material discovery. SDLs can rapidly navigate vast chemical spaces, optimize multi-dimensional synthesis and processing conditions, and accelerate the identification of high-performance materials relevant for commercialization. The approach not only enhances efficiency and experimental precision but also mitigates the trial-and-error limitations of traditional methods. In this perspective, we examine how computation, high-throughput automation, and advanced AI techniques can accelerate the discovery of energy materials, highlighting key capabilities and current applications and providing a perspective on future directions in the field. We also discuss how, despite challenges such as high implementation costs and the complexity of automating solid-state synthesis, the continued success of SDLs relies on more democratized, accessible, and collaborative frameworks.