Ames Laboratory Leverages AI to Design Rare-Earth-Free Permanent Magnets

Scientists at the U.S. Department of Energy’s Ames National Laboratory have created a systematic workflow that merges physics-based modeling, high-throughput simulations, and reasoning-based artificial-intelligence (AI) tools to identify permanent-magnet candidates without rare-earth elements. The AI framework evaluates material candidates before any laboratory synthesis, allowing researchers to focus experimental effort on the most promising compositions.

Interesting Engineering

Permanent magnets are integral to data storage, electric-vehicle drivetrains, high-resolution medical imaging, defense systems, and energy-generation equipment. The United States currently relies on foreign refining capacity for rare-earth elements, a situation that raises both cost and national-security concerns. Reducing this dependence has become a strategic priority for U.S. research programs.

Interesting Engineering

The effort is led by Prashant Singh, a senior scientist at Ames National Laboratory. Singh’s team draws on the laboratory’s long-standing magnet-materials database, which the lab describes as “deep expertise and a long history of data in the magnet space that no other institution has.”

Interesting Engineering

The AI model incorporates atomic-scale structure and electronic behavior to predict key magnet performance metrics: saturation magnetization, coercivity (resistance to demagnetization), energy-product (storage capacity), and high-temperature stability. By evaluating these parameters computationally, the workflow can rank thousands of hypothetical alloys, narrowing the field to a manageable set for experimental validation.

Interesting Engineering

The approach directly addresses supply-chain fragility by targeting materials that are both abundant and inexpensive. If successful, the resulting magnets could be produced domestically, supporting U.S. goals to lessen reliance on external sources for critical components in transportation, healthcare, and defense technologies.

Interesting Engineering

In a press release, Ames Laboratory highlighted its combined theoretical and analytical tools that predict how element combinations affect magnetic performance before any experiment is run. The release emphasized that AI accelerates this prediction process and expands the searchable material space. Additionally, the AI system evaluates material availability and cost, ensuring that discovered candidates are scalable and industrially viable.

Interesting Engineering

• “Understanding the physics of materials is important to include in AI frameworks when you are trying to design new materials. If you just use the data to train your models, you are going to get only the predictions within the range of information you have,” — Prashant Singh, Scientist, Ames National Laboratory

Interesting Engineering

• “But once you understand the physics of what controls specific properties, then you and your agentic tools or AI frameworks can search arbitrary material space,” — Prashant Singh

Interesting Engineering

• “Ames Lab’s strength comes from its deep expertise and a long history of data in the magnet space that no other institution has,” — Prashant Singh

Interesting Engineering

• “In any material design problem, you need to know how combining two elements will change their performance before you ever run an experiment. We have been building both theoretical and analytical tools to answer that question, and we are now bringing AI into that process to make it faster and broader in scope.” — Prashant Singh

Interesting Engineering

The research findings were published in *Materials Science and Engineering*. The team plans to advance the most promising candidates toward pilot-scale production, completing the pipeline from AI-guided discovery to industrial availability.

Interesting Engineering