Researchers have developed a bold new strategy to optimize the development of new materials.
A team of scientists from Los Alamos National Laboratory has released a study that claims a new strategy could be extremely useful in developing new materials. According to a report from UPI, scientists have employed machine learning to cut down the trial-and-error period of testing out new materials that can be used for a wide range of applications.
Testing out new materials can be extremely time-consuming. As the chemical makeup of materials intended for different purposes becomes increasingly complex, scientists have begun to feel the weight of a nearly impossible challenge before them.
Enter supercomputers. The team at Los Alamos National Laboratory employed an “informatics-based adaptive design strategy” to speed up the process by crunching massive sets of data to determine which compounds could be applicable in the world.
According to Turab Lookman, a physicist and materials scientists from Los Alamos, “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.”
Lookman and his team developed a machine-learning algorithm that significantly cut down on the trial-and-error period of the materials development process. Uncertainties in the algorithm could be addressed by human analysis after the process was complete.
The study, published in the journal Nature Communications, could have profound impacts on the materials science field. Up until now, scientists had to meticulously test and tweak materials before they would be ready for most applications. Now, this process takes only a fraction of the time that it used to.
“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.”
A press release from Los Alamos National Laboratory describing the details of the study can be found here.