China's rare-earth magnet industry is weighing a shift toward samarium-iron-nitrogen (Sm-Fe-N) permanent magnets as a way to cut reliance on expensive heavy rare earths such as dysprosium and terbium, according to a deep-dive analysis published by Chinese outlet D1EV. The material, studied for over three decades, offers higher temperature tolerance than the neodymium-iron-boron (NdFeB) magnets now dominant in EV drive motors, but its path to mass production is blocked by a fundamental processing problem.
Why NdFeB is under pressure
NdFeB magnets, the third-generation rare-earth permanent magnet system, hold the largest share of the global market. But their relatively low Curie temperature of around 310°C means performance degrades in hot environments, forcing EV motor makers to add heavy rare earths like dysprosium and terbium to maintain coercivity. Those elements are scarce and costly: terbium oxide has traded between 5 million and 7 million yuan per tonne, and dysprosium oxide between 1 million and 2.5 million yuan per tonne, the report notes.
China controls more than 80% of global medium and heavy rare-earth reserves, according to 2025 data from the China Geological Survey cited by D1EV. In April 2025, Beijing tightened export controls on several medium and heavy rare-earth items, including samarium, gadolinium, terbium, dysprosium, lutetium and yttrium, underscoring their strategic importance.
Samarium, by contrast, is abundant in China and cheap. USGS 2025 average prices put 99.5% samarium oxide at about $2.82 per kilogram, versus roughly $69 per kilogram for 99% praseodymium-neodymium oxide. That price gap gives Sm-Fe-N a natural cost advantage, the report argues.
Strong on paper, hard to make
Sm-Fe-N's appeal lies in its intrinsic properties. Sm2Fe17N3 has a Curie temperature of about 470°C, well above sintered NdFeB's roughly 310°C, and an anisotropy field two to three times higher, giving it strong resistance to demagnetization. Its theoretical maximum energy product reaches 470 kJ/m3, comparable to high-end NdFeB.
The performance leap comes from an interstitial atom effect: nitrogen atoms slot into the Sm2Fe17 lattice, expanding it and strengthening 4f-3d exchange interactions. Peking University's Yang Yingchang team used neutron diffraction to map nitrogen occupancy and magnetic structure coupling, contributing core original research, D1EV reports.
But the material is thermodynamically metastable. Above 600°C, Sm2Fe17N3 decomposes irreversibly into non-magnetic phases, while conventional sintering of permanent magnets requires temperatures above 1,000°C. That contradiction blocks the production of fully dense blocks. Current commercial products are limited to bonded magnets, where polymer binders dilute the energy product to 5-22 MGOe and limit temperature resistance.
Researchers, including teams at Japan's Daido Steel and Hitachi Metals, are exploring spark plasma sintering and ultra-high-pressure powder metallurgy to densify the material below 500°C, or using low-melting-point metals like zinc as sintering aids. Laboratory samples show promise, but scaling to large industrial parts with consistent quality and controlled cost remains a major challenge, the report says.
D1EV concludes that Sm-Fe-N and NdFeB are more likely to coexist than to replace each other outright. In the near term, bonded Sm-Fe-N magnets could compete in consumer electronics, automation actuators, 5G communications and auxiliary micro-motors in vehicles. If industrial breakthroughs in low-temperature densification arrive, the material's high-temperature advantages could open up heavy-duty applications such as high-speed main drive motors.
