A team led by Iowa State University chemist Kirill Kovnir won a $2.7 million grant from the U.S. Department of Energy's ARPA-E to find new magnetic materials that outperform today's strongest permanent magnets. The project, part of a $72 million push to strengthen domestic magnet manufacturing and secure critical mineral supply chains, will pair machine learning with experimental synthesis to discover compounds that can generate and maintain high magnetic fields.
The MAGNITO program and the hunt for new physics
The funding comes from ARPA-E's MAGNITO initiative - short for Magnetic Acceleration Generating New Innovations and Tactical Outcomes. The program wants to "find entirely new physics, chemistries, and structures for ultra-powerful magnets," according to the agency's summary. Current neodymium-iron magnets dominate electric motors and generators, but the program seeks materials that leap beyond incremental improvements.
"A lot of current research is about improving known compounds," said Yaroslav Mudryk, a scientist at Ames National Laboratory and Iowa State's materials science and engineering department. "The goal of the MAGNITO program is to discover new compounds. That's why chemists are involved."
Machine learning as a screening engine
Kovnir's team - dubbed MAGNUMS, for Machine-learning Assisted Generation of Novel Ultra-strong Magnets via Synthesis - will use machine learning to rapidly scan candidate materials and combinations for promising magnetic signatures. James Chelikowsky and Yongxin Yao will lead the computational work, building models to guide experimentalists toward the most likely winners.
Yongxin Yao, a scientist at Ames National Laboratory and adjunct associate professor at Iowa State, described the effort: "Armed with state-of-the-art theoretical and AI-driven tools, it is truly like embarking on a treasure hunt for new magnetic materials." That approach, which falls under AI for Science & Research, shifts the discovery process from trial-and-error toward hypothesis-driven synthesis.
Chemists guide elements into unprecedented structures
The experimental side of the project - led by Kovnir, Julia Zaikina, Mudryk, and Florida State University's Michael Shatruk - will synthesize, test, and characterize prototype magnets. Zaikina explained that chemists will "guide" elements into new arrangements by precisely controlling ratios, synthesis methods, and temperatures. "We look forward to working closely with the computational group that will provide guidance on where to start and where to go, while saving time and resources from exploring the 'dead ends,'" she said.
The aim is to create compounds "that have the superpower of generating and maintaining high magnetic fields," Zaikina added. Success could lead to magnets that improve energy productivity, lower electricity generation costs, and enable smaller, lighter motors for industry and transportation.
Why this matters for science and research professionals
For researchers working at the intersection of computation and materials, the MAGNUMS project illustrates a practical model: machine learning doesn't replace experimental expertise - it directs it. The grant's structure, with tight coupling between theorists and synthetic chemists, shows how AI tools can compress the discovery timeline for functional materials. The work also highlights a growing federal appetite for projects that combine fundamental physics with supply-chain resilience, creating opportunities for labs that can bridge data science and hard synthesis.
Your membership also unlocks:
