University of New Hampshire researchers have built an AI system that reads scientific papers and extracts experimental data to create the Northeast Materials Database — a searchable catalog of 67,573 magnetic compounds. From this massive dataset, the AI identified 25 previously unknown materials that maintain magnetism at high temperatures without requiring rare earth elements, which are critical for EV motors.
The discovery addresses a major supply chain vulnerability: China processes 90% of the world's rare earth elements and controls 98% of dysprosium and 99% of yttrium processing. When China restricted rare earth exports in April 2025, neodymium-praseodymium oxide prices spiked 40% within a month. With 94.7% of light EVs using rare earth magnet motors, finding alternatives is essential for affordable, geopolitically secure electric vehicles.
Lead researcher Suman Itani said the work could "reduce dependence on rare earth elements and lower the cost of electric vehicles and renewable-energy systems." The study, funded by the U.S. Department of Energy, was published in Nature Communications.
