With new, artificial intelligence approach, scientists discovered metallic glass 200 times faster than with an Edisonian approach. Credit: SLAC National Accelerator Laboratory

Researchers have created a new metallic glass that is stronger and lighter than some of the best steel, while standing up to corrosion and wear.

A team from the Department of Energy's SLAC National Accelerator Laboratory, the National Institute of Standards and Technology (NIST) and Northwestern University used a new system at SLAC's Stanford Synchrotron Radiation Lightsource (SSRL) that combines machine learning with experiments that quickly make and screen hundreds of sample materials at a time.

This allowed the researchers to discover three new blends of ingredients that form metallic glass about 200 times faster than before.

While there are millions of possible combinations of ingredients for metallic glass, only a few thousand combinations have been evaluated in the last 50 years, with only a select few being developed to the point where they can be commercially useful.

“It typically takes a decade or two to get a material from discovery to commercial use,” Chris Wolverton, the Jerome B. Cohen Professor of Materials Science and Engineering in Northwestern's McCormick School of Engineering, said in a statement. “This is a big step in trying to squeeze that time down. You could start out with nothing more than a list of properties you want in a material and, using AI, quickly narrow the huge field of potential materials to a few good candidates.”

The scientists ultimately want to be able to scan hundreds of sample materials and get immediate feedback from machine learning models, while having another set of samples available to test within an hour. This could enable researchers to make and screen about 20,000 samples in a given year.

“The unique thing we have done is to rapidly verify our predictions with experimental measurements and then repeatedly cycle the results back into the next round of machine learning and experiments,” co-author Apurva Mehta, a staff scientist at SSRL, said in a statement.

According to Mehta, the process could eventually be automated so scientists can focus on other aspects of their work that require human intuition and creativity. The new method could be used in searches of materials like metallic glass and catalysts, whose performance is strongly influenced by the way they are manufactured.

“One of the more exciting aspects of this is that we can make predictions so quickly and turn experiments around so rapidly that we can afford to investigate materials that don't follow our normal rules of thumb about whether a material will form a glass or not,” paper co-author Jason Hattrick-Simpers, a materials research engineer at NIST, said in a statement. “AI is going to shift the landscape of how materials science is done, and this is the first step.”

The study was published in Science Advances.