The discovery of functional inorganic materials remains slow, costly, and heavily dependent on expert intuition. This constrains progress in critical areas such as carbon dioxide capture and conversion, water electrolysis, optoelectronics, and energy storage, where improved materials are urgently needed. A key bottleneck is synthesis: inorganic materials often form through complex chemical processes that are difficult to understand, control, or reproduce. Minor variations in conditions—such as temperature, precursor ratios, or reaction environment—can lead to substantially different outcomes, making reliable and rational synthesis design challenging.

Our self-driving lab for inorganic chemistry is a closed-loop, autonomous platform that integrates multiple synthesis approaches (including solution-based, solid-state, and electrochemical methods) with real-time data collection, adaptive process control, and AI-guided planning. Rather than relying on fixed protocols or trial-and-error experimentation, the platform continuously captures detailed, time-resolved data during synthesis, including transient intermediate states. These data enable reconstruction of material formation pathways and iterative improvement of synthesis strategies.

The lab transitions inorganic materials development from an empirical, trial-and-error process to a property-driven approach informed by mechanistic understanding. This is enabled by modular, well-instrumented reactors that are tightly integrated with materials characterization and application-relevant performance testing, such as electrocatalysis and battery evaluation. By enabling faster, more reproducible discovery of complex inorganic materials and synthesis strategies, we will accelerate progress in energy and climate technologies while contributing shared modules, protocols, and data standards that strengthen the broader SDL ecosystem.