Scientists use artificial intelligence to find new magnetic materials that don't use critical elements. A research team at the U.S. Department of Energy's Ames National Laboratory has developed a new machine learning model for discovering permanent magnet materials that do not contain critical elements. The model predicts the Curie temperature of new material combinations. This is an important first step in using artificial intelligence to predict new permanent magnet materials. The model complements the team's recently developed ability to discover thermodynamically stable rare earth materials.
Scientists at Ames National Laboratory have designed a machine learning model that can predict new magnet materials without using scarce elements. This innovative approach focusing on the Curie temperature of materials offers a more sustainable path to future technology applications.
The Importance of High-Performance Magnets
High-performance magnets are critical for technologies such as wind energy, data storage, electric vehicles and magnetic refrigeration. These magnets contain key materials such as cobalt and rare earth elements such as neodymium and dysprosium. These materials are in high demand, but supply is limited. This situation has prompted researchers to look for ways to design new magnetic materials that reduce critical materials.
The role of machine learning
Machine learning (ML) is a form of artificial intelligence. It is driven by computer algorithms, using data and trial-and-error algorithms to continuously improve predictions. The research team used experimental data and theoretical modeling of Curie temperatures to train the ML algorithm. The Curie temperature is the highest temperature at which a material remains magnetic.
"Finding compounds with high Curie temperatures is an important first step in discovering materials that can remain magnetic at high temperatures," said Yaroslav Mudryk, an Ames Laboratory scientist and senior leader of the research team. "This aspect is crucial not only for the design of permanent magnets, but also for the design of other functional magnetic materials."
Mudrick believes that discovering new materials is a challenging activity because the search for new materials has traditionally been done through experiments, which is expensive and time-consuming. However, using ML methods can save time and resources.
Model testing and validation
To validate the model, the team used compounds based on cerium, zirconium and iron. The idea was proposed by Andriy Palasyuk, a scientist at Ames Laboratory and a member of the research team. He hopes to focus on unknown magnet materials based on Earth-abundant elements. Palaschuk said: "The next super magnet must not only have excellent performance, but also rely on abundant domestic components.
Palaschuk collaborated with research team member Tyler Del Rose, another Ames Laboratory scientist, to synthesize and characterize the alloy. They found that the ML model successfully predicted the Curie temperature of the candidate materials. This success is an important first step in a high-throughput approach to designing new permanent magnets for future technological applications.
"We are writing physics-informed machine learning for a sustainable future," Singer said.
