A self-contained robotic laboratory that discovers new materials

Researchers at MIT have developed a fully autonomous platform that combines an intelligent algorithm and a robotic system to rapidly discover new materials. This “self-contained laboratory” could significantly accelerate the development of key technologies such as batteries, solar cells, and drugs.

It's a so-called closed loop, where a genetic algorithm generates hundreds of promising "recipes" for new polymer blends. The recipes are sent to a robotic system that automatically mixes the chemicals, takes measurements, and sends the results back to the algorithm. The algorithm learns from them and makes even better suggestions for the next production run.

One of the biggest advantages of the new system is undoubtedly its speed. It can create and test up to 700 new materials every day. This represents a major leap forward in research, where developing new materials is usually a lengthy process. The platform enables the discovery of materials such as better electrolytes for batteries that are safer and more efficient, cheaper components for solar cells, and precise nanoparticles for drug delivery.

In initial tests, the advanced platform sought a polymer blend that kept enzymes stable at high temperatures. It found a blend that was 18 percent more effective than any of its individual components, demonstrating its potential to uncover unexpected solutions.