A study published this year in Applied Physics A raised the coercivity of a grain boundary diffused NdFeB magnet from 17.81 kOe to 26.47 kOe by using a terbium and copper mixed powder as the diffusion source, compared with 24.09 kOe for pure terbium hydride, under identical diffusion conditions and the same terbium dosage. Both magnets started from the same material and received the same amount of heavy rare earth, so the gain came from how the terbium was delivered and how it moved during heat treatment.
As a company developing integrated NdFeB magnet manufacturing, we read results like this through a production lens. They confirm something every magnet line learns early: a large share of a magnet’s final performance is decided after the alloy has already been made, inside stages that have to be controlled as carefully as the chemistry itself.
Why Coercivity Sets the Limits of an NdFeB Magnet
Coercivity is a magnet’s resistance to demagnetization, and it determines where a magnet can safely work. Traction motors, servo drives, robotic actuators, and wind generators all expose magnets to heat and opposing fields. Coercivity falls as temperature rises, so a grade that performs well on a room temperature test bench can lose output inside a hot, heavily loaded motor.
The standard way to raise coercivity is to add dysprosium or terbium. Reducing the use of these heavy rare earths has been a central research focus for the permanent magnet community over the past decade, because the earliest approach, alloying them directly into the bulk material, significantly reduced remanence and consumed large amounts of heavy rare earth. Remanence is the magnet’s strength, and bulk alloying spends it to buy thermal stability.
How Diffusion Changes Where Heavy Rare Earth Goes
A sintered NdFeB magnet is built from fine magnetic grains separated by thin boundary phases, and demagnetization tends to start at the grain edges. Grain boundary diffusion takes advantage of that structure. A heavy rare earth source is applied to the surface of a sintered magnet, and heat treatment carries it inward along the grain boundaries, where it forms a thin enriched shell around each grain. The magnet gains coercivity at the points where it is most vulnerable while keeping most of its strength.
Research continues to refine how precisely that material can be placed. One recent study explored selectively hardening only the regions of a magnet most susceptible to demagnetization, and found that coercivity gains level off as more diffusion material is added, likely because adjacent grains can be magnetically decoupled with only a small amount of it. Beyond a certain threshold, adding more terbium brings little benefit, and the quality of the process becomes the deciding factor.
The Problem the 2026 Study Set Out to Solve
Diffusion has well known limits. Heavy rare earth tends to penetrate only so far, and an overly thick enriched shell can build up near the surface, limiting both material efficiency and further coercivity gains. When heavy rare earth is used alone as the source, it frequently segregates at the surface and does not travel deep into the magnet.
The Applied Physics A team addressed that by adding a small amount of copper, a low melting point metal, to the terbium source, then systematically adjusting variables to identify the optimal diffusion parameters for the combined system. Low melting point additives help the grain boundary phase carry heavy rare earth inward at treatment temperatures, which is consistent with the higher coercivity the team measured. Results from a laboratory study still have to be validated at production scale for each grade and geometry, but they show where industry practice is moving: toward diffusion sources and heat treatment profiles engineered as carefully as the alloy.
The Variables That Decide the Outcome
Any manufacturer applying diffusion at commercial scale has to hold a demanding set of variables within tight limits across every batch. Coating uniformity decides whether each magnet receives the same amount of source material. Temperature and time determine how deep the heavy rare earth travels and how thick the surface shell becomes. Atmosphere control protects the magnet from oxidation at treatment temperatures, and the annealing that follows diffusion shapes the final grain boundary structure.
Geometry adds another layer. Because diffusion works inward from the surface, thin magnets respond differently from thick ones, and parts generally need to be machined close to final dimensions before diffusion so the enhanced layer is not ground away afterward. In a well run line, machining sequence becomes a performance decision as much as a dimensional one.
Why Heat Treatment Inherits Everything Upstream
Diffusion depends on the microstructure it receives. Strip casting establishes the starting structure of the alloy. Hydrogen decrepitation and jet milling set powder particle size and distribution. Magnetic alignment, pressing, and vacuum sintering determine grain size, density, and the boundary phases that heavy rare earth will later travel along. Variation at any of those stages reaches the furnace, and a heat treatment recipe tuned for one microstructure will behave differently on another.
This is the reason Neodyte is developing an integrated manufacturing platform that covers the complete production chain for sintered NdFeB permanent magnets, from rare earth oxide processing and metal production through sintering, heat treatment, precision machining, surface coating, magnetization, inspection, and traceability. When every stage sits inside one system, the conditions heat treatment depends on can be controlled at the source, and results can be traced back through each step that produced them.
What a Motor Manufacturer Actually Needs
A company qualifying a magnet grade for a motor platform tests samples at temperature, approves them, and then builds thousands of motors on the assumption that every future batch will behave like the ones it tested. Its real requirement is that the thousandth delivery matches the first. Advances in grain boundary diffusion make higher performance possible with less heavy rare earth, and turning that possibility into a dependable supply depends on process integration, validation, and traceability across the whole line. That is the capability we are building.
