High-throughput Discovery of Magnetic Rare Earth Transition Metal Alloys
This paper presents an accelerated materials discovery framework combining diffusion-based crystal generation with hierarchical screening to identify over 300 low-energy rare-earth–transition-metal magnetic candidates, including stable phases with saturation magnetization up to ~1.8 T, while providing systematic insights into dopant selection and structural origins for future high-magnetization alloy design.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Modern life runs on invisible forces. The electric motors that drive cars, the generators that power wind turbines, and the hard drives that store our memories all rely on permanent magnets. For decades, the best magnets have been made from a specific family of metals: rare earth elements mixed with transition metals like iron or cobalt. These materials are powerful, but they are also expensive and often rely on supply chains that are difficult to secure. Scientists have long known that there are likely many other combinations of these atoms that could make even better magnets, but finding them is like searching for a needle in a haystack. The number of ways to arrange these atoms is so vast that trying them one by one in a lab would take centuries.
To solve this, researchers have turned to a new kind of search engine. Instead of mixing chemicals in a beaker, they use powerful computers to imagine millions of new crystal structures. These digital models act as a filter, testing which arrangements are stable and which might hold a strong magnetic field. The goal is to find materials that are not only powerful but also made from abundant elements, reducing the need for the most expensive rare earth metals. This approach allows scientists to explore the entire landscape of possibilities, rather than just the few paths they have already walked.
In a recent study, a team of researchers used this high-speed digital approach to hunt for new magnetic materials. They focused on mixing rare earth elements with iron, cobalt, and nickel, and then adding a third ingredient to see if it could stabilize the structure. They did not just look at a few guesses; they generated nearly 240,000 different crystal structures on the computer. To manage this massive number, they used a two-step screening process. First, they used a fast, machine-learning model to quickly discard the millions of arrangements that were likely unstable or weak. This left them with a much smaller group of promising candidates. Then, they used a more precise, but slower, method to double-check the physics of these survivors, ensuring the results were accurate.
The search was successful. The team identified more than 300 low-energy candidates that could exist in the real world. Among these, they found five specific combinations that are thermodynamically stable, meaning they are likely to form naturally without falling apart. The most exciting discovery was the strength of the magnetic field these materials could produce. In the best cases, the new alloys reached a saturation magnetization of about 1.8 tesla. This is a measure of how much magnetic force the material can hold, and it rivals or exceeds the performance of the best magnets currently in use. The strongest results came from materials rich in iron, specifically combinations involving samarium or yttrium mixed with iron and a small amount of a third metal like titanium or manganese.
The researchers also uncovered a pattern in how these new materials are built. They found that most of the successful ternary alloys—those with three types of metals—were not entirely new inventions. Instead, they were variations of known binary structures, created by splitting specific atomic positions to make room for a third element. This insight is valuable because it tells experimental chemists exactly where to look when they try to make these materials in a lab. They don't need to guess; they can start with a known structure and swap in the right atoms at the right spots.
The study also revealed how different third elements affect the magnetic strength. When the team added manganese, it aligned perfectly with the iron atoms, boosting the overall magnetic field with very little loss. However, when they added chromium, it acted in the opposite direction, fighting against the iron and weakening the magnet. This clear distinction provides a simple rule for future design: choose dopants that agree with the iron, not those that oppose it. While the current models were limited to structures with fewer than 20 atoms, the researchers suggest that pushing this search into larger, more complex structures could unlock even stronger magnets. Their work demonstrates that by combining artificial intelligence with deep physical understanding, it is possible to navigate the vast chemical landscape and find the materials that will power the next generation of technology.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.