Scientists at the University of New Hampshire are using artificial intelligence to speed up the search for advanced magnetic materials. Their work has produced a searchable resource containing 67,573 magnetic compounds, including 25 materials that had not previously been recognized as magnets capable of staying magnetic at high temperatures.
“By accelerating the discovery of sustainable magnetic materials, we can reduce dependence on rare earth elements, lower the cost of electric vehicles and renewable-energy systems, and strengthen the U.S. manufacturing base,” said Suman Itani, lead author and a doctoral student in physics.
A Massive Magnetic Materials Database
The new resource, called the Northeast Materials Database, makes it easier for scientists to explore materials that are essential to modern technology. Magnets are critical components in smartphones, medical devices, power generators, electric vehicles, and many other everyday systems. However, today’s most powerful magnets depend on rare earth elements that are costly, largely imported, and increasingly difficult to secure. Despite the large number of known magnetic compounds, no entirely new permanent magnet has been identified from this pool.
The study, published in Nature Communications, describes how the team developed an AI system capable of reading scientific papers and pulling out important experimental data. That information was then used to train computer models to determine whether a material is magnetic and to calculate the temperature at which it loses its magnetism. The results were organized into one comprehensive and searchable database.
Reducing the Need for Rare Earth Elements
Researchers have long understood that many magnetic materials likely remain undiscovered. Yet testing every possible combination of elements, which could number in the millions, would take enormous amounts of time and money in a laboratory setting.
“We are tackling one of the most difficult challenges in materials science — discovering sustainable alternatives to permanent magnets — and we are optimistic that our experimental database and growing AI technologies will make this goal achievable,” said Jiadong Zang, physics professor and co-author.
Expanding the Role of AI in Science and Education
The research team also includes co-author Yibo Zhang, a postdoctoral researcher in both physics and chemistry. Looking ahead, the scientists believe the large language model used in this project could serve purposes beyond building this database, especially in higher education. For example, the technology could convert images into modern rich text formats, helping update and preserve library collections.
The project received support from the Office of Basic Energy Sciences, Division of Materials Sciences and Engineering, U.S. Department of Energy.
