Researchers have shown that an informatics-based adaptive strategy, when used with experiments, can propel the discovery of new materials with targeted properties, states a paper recently published in the journal Nature Communications.
Turab Lookman, lead researcher and a physicist and materials scientist with the Physics of Condensed matter and Complex Systems group at Los Alamos National Laboratory, says that, “What we’ve done is show that, starting with a relatively small data set of well-controlled experiments, it is possible to iteratively guide subsequent experiments toward finding the material with the desired target.”
“Finding new materials has traditionally been guided by intuition and trial and error,” Lookman adds.”But with increasing chemical complexity, the combination possibilities become too large for trial-and-error approaches to be practical.”
To achieve this goal, Lookman and fellow scientists at Los Alamos and the State Key Laboratory for Mechanical Behavior of Materials in China made use of machine learning to make the process faster. The team designed and developed a framework that employs uncertainties to guide the next experiments towards looking for a shape-memory alloy with very low thermal dissipation. These alloys are important in the engineering industry
The goal is to cut in half the time and cost of bringing materials to market,
said Lookman. “What we have demonstrated is a data-driven framework built on the foundations of machine learning and design that can lead to discovering new materials with targeted properties much faster than before.” The study used the Los Alamos’ supercomputing machines.
The Materials Genome initiative issued by the White House in 2011 raised an interest in materials discovery, but this is the first study to concretely demonstrate how an informatics framework can help in discovering new materials. Lookman and his team used nickel-titanium-based shape-memory alloys, but their work can be applied to any other materials or target properties. It can also be adapted to other applications, such as in manufacturing.