Robots, AI Could Help the Hunt for New Materials

Watch this video to see how — with the help of AI-guided robots — A-Lab could help identify and fast-track materials for several research areas, including batteries, energy storage, solar cells, fuel cells, and more.

“Our vision is using AI to discover the materials of the future,” said Yan Zeng  , a staff scientist leading the A-Lab at the Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab).

Transcript

00:00:01 Scientists have predicted hundreds of thousands of new materials. But finding the useful ones is a slow process. Researchers at Berkeley Lab are speeding it up with artificial intelligence and automation. At A-Lab, robots guided by AI will make and test materials quickly processing more than 100 samples a day. The robots weigh out and mix the starting ingredients then transfer them to crucibles and into furnaces that can reach different temperatures and incorporate different gases.

00:00:52 Robots then break up the new material to fine powders. Analyzing what has been created and sending data back to the system. The system can process 50 to 100 times as many samples in a day as a human can. Combining automation and AI allows the system to rapidly iterate and use data to make predicted materials faster. To start, A-Lab will look for new materials for batteries that could store energy more effectivley or less expensively. Researchers plan to integrate A-Lab with other labs and big data projects, such as The Materials Project. This will enable A-Lab to request new computations

00:02:00 and return real world data, guiding future scientists in making the materials of the future.