MIT's AI-powered 'autonomous lab' could lead to better batteries and solar panels
Researchers at MIT have developed a fully autonomous platform that uses an intelligent algorithm and a robotic system to rapidly discover new materials. This 'autonomous lab' could significantly speed up the development of next-generation technologies that impact daily life, from batteries to medicines.
The closed-loop system works by having a genetic algorithm intelligently design hundreds of promising new polymer blends. It then sends these recipes to a robotic platform that automatically mixes the chemicals and conducts tests on the new materials. The results are fed back to the algorithm, which learns from the outcomes and designs an even better set of materials for the next round of experiments.
This automated process is incredibly fast, allowing the system to create and test up to 700 new polymer blends every day. The researchers state this accelerated discovery workflow could be a game-changer for several critical fields.
The platform could facilitate the discovery of the following kinds of materials:
- Improved battery electrolytes for safer and more efficient energy storage.
- More cost-effective solar panels.
- Tailored nanoparticles for safer and more precise drug delivery.
In its initial tests, the system worked to find a polymer blend that could keep enzymes stable at high temperatures. The platform discovered a blend that performed 18% better than any of its individual components, showcasing its ability to find novel and non-obvious solutions. The work was published in the journal Matter.
