Highlights

• University of Birmingham team demonstrates viable 'magnet-to-magnet' short-loop recycling of NdFeB magnets from hard drives.
• Achieves coercivity up to 1466 kA/m—better than original scrap—by separating oxygen-rich grain boundaries after hydrogen processing.
• New spiral jet mill process splits HPMS powder into coarse (low-oxygen, high-performance) and fine (oxygen-rich waste) fractions.
• Enables recycled magnets with ~98–99% density and magnetic properties matching or exceeding virgin material.
• Breakthrough offers strategic alternative to China-dominated rare earth processing.
• Provides Europe and allies a credible pathway to circular NdFeB supply for EVs and wind turbines without traditional chemical refining routes.

---

Viktoria Kozak (opens in a new tab) and colleagues at the University of Birmingham, working with industrial partners linked to the EU’s SUSMAGPRO and REEsilience projects, have unveiled a promising way to turn waste hard-disk-drive magnets into high-performance neodymium-iron-boron (NdFeB) magnets again—without sending material back through traditional, China-centric chemical refining routes.

In a new open-access paper in Sustainable Materials and Technologies (December 2025), the team shows that by carefully separating oxygen-rich grain-boundary material after hydrogen processing, they can make “short-loop” recycled magnets with magnetic performance close to, and in the case of some scrap streams, better than, the original parts. For investors watching China’s grip on rare earth processing, this is a serious signal that magnet-to-magnet recycling is moving from lab concept to engineering reality.

Turning Old Hard Drives into New Magnets

The study focuses on Hydrogen Processing of Magnet Scrap (opens in a new tab) (HPMS), a Birmingham-pioneered method that uses low-pressure hydrogen to crack magnets embedded in products like hard disk drives, motors, and speakers. Instead of shredding everything and losing magnets in mixed scrap, HPMS produces a demagnetized NdFeB alloy powder that can be re-used.

Kozak’s team went a step further. They took HPMS powder from two real scrap streams—hard disk drives (HDD) and production scrap (PS)—and ran it through a spiral jet mill with an integrated cyclone classifier. This dry system, operated under nitrogen, splits the powder into:

• a coarse fraction with lower oxygen and slightly lower total rare earth content, and
• a fine fraction enriched in rare-earth-rich grain boundary phase (GBP), but also in oxygen and other light elements.

Both fractions were then blended with 5 wt% neodymium hydride (NdH₂.₇), pressed, and re-sintered into new magnets for detailed testing.

What the Scientists Actually Achieved

The numbers matter:

• Oxygen levels in the coarse fraction were about 0.29–0.34 wt%, while the fine fraction climbed to 1.39–1.60 wt%—a clear concentration of oxidized GBP in the fines.
• The fine fraction also carried higher total rare earth content (TREE), confirming that most of the valuable but oxygen-rich GBP ended up there.

When only the coarse powder was used (plus NdH₂.₇), the recycled magnets showed:

• Coercivity (resistance to demagnetization) of up to 1466 kA/m (PS coarse), higher than the starting magnets in that scrap stream.
• Remanence (magnetic strength) around 1.28 T, comparable to or slightly above the original magnets.
• Densities close to 98–99% of theoretical, indicating good microstructural quality.

By contrast, magnets made from a blend of coarse + fine powder saw a drop in coercivity and density, consistent with more oxygen, more porosity, and less ideal grain boundaries. In simple terms, removing much of the oxygen-laden grain-boundary phase upgrades the recycled magnet.

Why This Matters in a China-Dominated Market

Today, less than 5% of rare earth elements are recycled, and most NdFeB magnet material ultimately flows into, or back through, China’s processing ecosystem. Short-loop “magnet-to-magnet” routes like HPMS plus grain-boundary separation offer a way to:

• Keep more value in-region (UK/EU or any adopter) by bypassing full chemical reprocessing of oxides.
• Cut environmental and permitting risk tied to new mines and solvent-extraction refineries.
• Build a secondary, circular supply of NdFeB that can feed EVs, wind turbines, and electronics without first being shipped as oxide or metal to Chinese refiners.

For policymakers and OEMs, the key takeaway is strategic: if HPMS-based recycling scales, it becomes a credible buffer against export controls, licensing games, and price shocks originating in Beijing.

Not a Silver Bullet: Limits, Yield Losses, and Next Steps

The authors are explicit about limitations:

• The current cyclone “cut point” throws away too much good material; yield losses of the matrix phase are significant and must be reduced by tuning gas flow and classifier speed.
• Oxygen can’t be fully eliminated—only managed. Further gains will require degassing, post-sinter annealing, and possibly heavy rare earth additions or grain-boundary diffusion to push coercivity higher.
• Economic viability at industrial scale is not answered here; the work is technical proof-of-concept, not a full cost curve.

There’s also a strategic caveat: improved recycling doesn’t magically erase China’s dominance in primary mining and separation. But it does give Europe, the UK, Japan, and others a tangible lever—particularly when tied to automated magnet extraction from scrap and coordinated policy like the EU’s ReSourceEU and Critical Raw Materials Act.

A Small Process Step with Big System Implications

Kozak and colleagues show that by treating HPMS powder not as waste, but as a tunable feedstock, engineers can selectively remove oxygen-rich phases and “upgrade” scrap into higher-grade magnets. For the rare earth ecosystem, this is more than an incremental tweak. It hints at a future where a meaningful share of NdFeB demand is met by short-loop recycling plants next to data centers, auto dismantlers, and e-waste hubs—chipping away at China’s processing near-monopoly one batch at a time.

Citation: Kozak V, Pickering L, Awais M, et al. Improving short-loop recycling of sintered NdFeB magnets by grain boundary phase separation following hydrogen processing. Sustainable Materials and Technologies. 2025;46:e01744. doi:10.1016/j.susmat.2025.e01744 (opens in a new tab).

© 2025 Rare Earth Exchanges™ – Accelerating Transparency, Accuracy, and Insight Across the Rare Earth & Critical Minerals Supply Chain.