Scientists in the US are using artificial intelligence and advanced chemistry to search for stronger permanent magnets that can support future energy and industrial technologies.
The effort is backed by a $2.7 million grant from the US Department of Energy’s Advanced Research Projects Agency-Energy (ARPA-E). The project aims to identify and develop new magnetic materials that outperform today’s widely used neodymium-iron magnets.
The research is led by Kirill Kovnir, a professor of chemistry at Iowa State University. The team includes experts from Iowa State University, Ames National Laboratory, the University of Texas at Austin, and Florida State University. Together, they will combine computer-based material discovery with laboratory testing to develop entirely new magnetic compounds.
The project is called MAGNUMS, short for Machine-learning Assisted Generation of Novel Ultra-strong Magnets via Synthesis. Its work supports ARPA-E’s MAGNITO program, which stands for Magnetic Acceleration Generating New Innovations and Tactical Outcomes. The program focuses on discovering new magnetic materials instead of making only small improvements to existing ones.
Machine Learning Guides
Machine learning is a type of artificial intelligence that enables computers to learn from large amounts of data and identify useful patterns. In this project, researchers will use machine learning to rapidly examine many possible material combinations before testing them in laboratories. This process helps scientists focus on the most promising candidates while avoiding less useful options.
James Chelikowsky from the University of Texas at Austin and Yongxin Yao from Ames National Laboratory will lead the computational work. Their systems will predict which crystal structures and chemical combinations may deliver stronger magnetic performance. These predictions will guide laboratory researchers toward the most promising starting points.
Yao said that advanced theoretical and AI-based tools make the search for new magnetic materials feel like a treasure hunt. Instead of relying only on traditional experiments, researchers can narrow down the possibilities much faster. This approach saves both time and research resources.
READ ALSO: Q-CTRL Unveils Airworthiness-Qualified Quantum GPS Backup at Farnborough Airshow
Magnets Get an Upgrade
Once promising materials are identified, chemists will produce and examine them in the laboratory. Julia Zaikina, Yaroslav Mudryk, Michael Shatruk, and Kovnir will lead the synthesis, testing, and detailed analysis of the new compounds. Their work will determine whether the predicted materials can actually be manufactured and perform as expected.
Creating new magnetic materials requires careful control over chemical ingredients, temperature, and manufacturing methods. Even small changes in these conditions can produce very different crystal structures and magnetic properties. The researchers aim to guide chemical elements into new arrangements that have never been produced before.
Mudryk explained that much of today’s magnet research focuses on improving known materials. The MAGNITO program instead encourages scientists to discover completely new compounds with better performance. That goal makes chemistry a central part of the research effort.
Zaikina said the chemistry team looks forward to working closely with the computational researchers. She explained that AI predictions will help identify where experiments should begin and which paths deserve further study. This partnership reduces the chances of spending valuable time on unsuccessful experiments.
Industry Gains Ahead
Permanent magnets play an important role in electric motors, wind turbines, generators, and many other modern technologies. Stronger magnets can increase efficiency while reducing the size and weight of important equipment. That makes them valuable for transportation, manufacturing, and clean energy systems.
The Iowa State-led project is one of several initiatives funded under ARPA-E’s broader $72 million investment. The program supports early-stage research designed to strengthen domestic magnet manufacturing and improve access to critical minerals needed for advanced technologies. These efforts also support more secure US supply chains for essential industrial materials.
The researchers hope their work will lead to magnets capable of generating and maintaining stronger magnetic fields than current materials. Such magnets may improve energy efficiency, lower electricity generation costs, and enable smaller, lighter motors for industrial applications.
As AI continues to accelerate materials discovery, the project represents an important step toward developing the next generation of high-performance permanent magnets.
