Researchers have trained an AI system to analyze decades of scientific literature, identifying potential candidates for next-generation magnets that could reduce dependence on scarce and often geopolitically sensitive resources.
The AI is capable of extracting experimental details to determine a material’s magnetic properties and the temperatures at which those properties are maintained. This information is then organized within a publicly accessible database called the Northeast Materials Database. This approach allows researchers to quickly identify promising materials without the need for extensive laboratory testing.
Currently, the most effective permanent magnets rely heavily on rare earth elements, the supply of which is often concentrated in specific regions, leading to price volatility and availability concerns. Although numerous magnetic compounds are known, finding one that can match the performance of rare-earth magnets in everyday applications has remained a significant challenge.
The database already contains over 67,000 magnetic materials, including 25 newly identified compounds that retain their magnetic properties even at high temperatures. According to Suman Itani, a physics doctoral student and lead author of the study, “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.”
Professor Jiadong Zang, a co-author of the perform, noted that the research addresses one of the primary challenges in materials science: identifying sustainable alternatives to permanent magnets. The team is confident in the potential of their database and AI technologies to make this goal achievable. The breakthrough underscores the growing role of artificial intelligence in accelerating scientific discovery and addressing critical supply chain vulnerabilities.
