AI Discovers Greener Pastures

Pacific Northwest National Laboratory has partnered with Microsoft to employ artificial intelligence (“AI”) to successfully identify a novel battery material that uses less lithium. The discovery was achieved using AI computing to search through 32.6 million possible battery materials (many not found in nature) in 80 hours – a task the team estimates would have typically taken 20 years.

“This is a new way to do science” said Nathan Baker, an author of the scientific paper outlining the experimental method and senior director of partnerships for chemistry and materials at Microsoft. Members of the joint project were effectively able to “pick through the periodic table” and “drop new atoms” into various locations through a process of substitution. The AI-software then weeded out materials that were poorly suited for battery manufacturing (e.g., because of poor conductivity, high cost or insufficient stability), ultimately providing material experts with the answer to build a prototype battery in the laboratory.

These efforts not only drive a new generation of batteries that are less dependent on environmentally damaging lithium, but fundamentally redefine the conventional scope of research and development initiatives. What’s more, AI’s role in unearthing the new material demonstrates its potential to revolutionize materials science (a field commonly believed to exceed the capabilities of machine learning) and actively participate in building a more sustainable future.