Newswise — A team of scientists at Ames National Laboratory has developed a new machine learning A model for discovering critical element-free permanent magnet materials. The model predicts Curie temperatures in new material combinations. This is an important first step in application artificial intellect Prediction of new permanent magnet materials. This model is added to the team Recently developed Discovery of thermodynamically stable rare earth materials.
High-quality magnets are essential for technologies such as wind power, data storage, electric vehicles and magnetic cooling. These magnets contain critical materials such as cobalt and rare earth elements such as neodymium and dysprosium. These materials are in high demand, but availability is limited. This situation prompts researchers to find ways to design new magnetic materials with reduced critical materials.
machine learning (ML) is the form artificial intellect. It is guided by computer algorithms that use data and trial-and-error algorithms to continually improve its predictions. The team used experimental data on the Curie temperature and theoretical modeling to develop the ML algorithm. The Curie temperature is the maximum temperature at which a material retains its magnetism.
“Finding high-temperature Curie compounds is an important first step in discovering materials that can retain magnetic properties at elevated temperatures,” said Jaroslav Mudrik, an Ames Lab scientist and senior research team leader. “This aspect is crucial for the design not only of permanent magnets, but also of other functional magnetic materials.”
According to Mudryk, discovering new materials is a difficult activity because the search is traditionally based on experiments, which are expensive and time-consuming. However, using the ML method can save time and resources.
Prashant Singh, an Ames Lab scientist and member of the research team, explained that a key part of this effort was developing an ML model using fundamental science. The team trained their ML model using experimentally known magnetic materials. Information about these materials establishes a relationship between several electronic and atomic structural features and the Curie temperature. These patterns give the computer a basis for searching for potential candidate materials.
To test the model, the team used compounds based on cerium, zirconium, and iron. Andrii Palasiuk, an Ames Lab scientist and member of the research team, proposed the idea. He wanted to focus on unknown magnetic materials based on elements abundant on Earth. “The next super magnet will not only have to be excellent in performance, but also rely on a number of internal components,” Palasiuk said.
Palasiuk worked with Tyler Del Rose, another Ames laboratory scientist and member of the research team, on the synthesis and characterization of the alloys. They found that the ML model was successful in predicting the Curie temperature of the material candidates. This success is an important first step towards creating new high-throughput permanent magnet designs for future technology applications.
“We write physically informed machine learning for a sustainable future,” said Singh.
This study is further discussed in “Physics-informed machine learning prediction of the Curie temperature and its promise for guiding the discovery of functional magnetic materials”, wrote Prashant Singh, Tyler Del Rose, Andrei Palasiuk, and Jaroslav Mudryk and published in Chemistry of materials.
Ames National Laboratory is US Department of Energy Office of Science A national laboratory operated by Iowa State University. Ames Laboratory creates innovative materials, technologies and energy solutions. We use our experience, unique capabilities and interdisciplinary collaboration to solve global problems.
The Ames Laboratory is supported by the US Department of Energy's Office of Science. The Office of Science is the single largest supporter of basic research in the physical sciences in the United States and works to address some of the most pressing challenges of our time. For more information, please visit https://energy.gov/science.