🔬 materials science

Effect of Al-Zn alloy wafer grain boundary diffusion on the magnetism and microstructure of sintered NdFeB magnets

This study demonstrates that grain boundary diffusion using an Al-Zn alloy source at 900°C significantly enhances the coercivity of sintered NdFeB magnets by 21.7% through the formation of a continuous, high-anisotropy core-shell microstructure and improved grain boundary decoupling, offering an effective non-heavy-rare-earth strategy for high-temperature applications.

Original authors: Xi Liu, Wenxi Fang

Published 2026-07-27

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Xi Liu, Wenxi Fang

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

Imagine the invisible world of magnets as a bustling city made of tiny, magnetic skyscrapers. These skyscrapers are grains of a special material called Nd-Fe-B (neodymium-iron-boron), and they are the powerhouses behind everything from the motors in electric cars to the speakers in your headphones. But like any city, these magnets have a weak spot: when the weather gets hot, the "security guards" at the edges of the buildings get lazy, and the whole city starts to lose its magnetic power. This is a big problem because we need these magnets to work hard even when things get toasty.

To fix this, scientists usually try to reinforce the edges of these skyscrapers with heavy, expensive elements like dysprosium. Think of it like hiring a super-expensive, elite security team to guard the perimeter. It works, but it's costly and can make the buildings a bit weaker overall. A smarter idea is "grain boundary diffusion," which is like sneaking a small amount of that elite security team right into the cracks between the buildings so they can patrol the edges without needing to replace the whole building. However, getting these security guards to travel deep enough into the city is tricky; they often get stuck near the surface. This paper explores a clever new trick to help them travel further and do a better job, using a special alloy of aluminum and zinc as a delivery vehicle.

The Mission: A New Delivery Service for Magnets

In this study, researchers Xi Liu and Wenxi Fang set out to see if they could boost the heat resistance of sintered Nd-Fe-B magnets without breaking the bank or weakening the magnet's core strength. They used a method called grain boundary diffusion, but with a twist. Instead of grinding up aluminum and zinc and mixing them into the raw magnet powder (which is messy and hard to control), they took a neat sheet of Al80Zn20 alloy (80% aluminum, 20% zinc) and pressed it directly against the ends of the cylindrical magnets.

They then heated these magnets in a vacuum oven at two different temperatures: a "warm" 700°C and a "hot" 900°C, holding them there for 7 hours. Afterward, they cooled them down and checked how the magnets performed.

The Results: Heat Makes the Magic Happen

The results were clear: heat matters a lot.

The 700°C Treatment: When the magnets were heated to 700°C, the aluminum and zinc managed to sneak a little way into the cracks between the grains. The magnet's resistance to losing its magnetism (called coercivity) went up from 951.5 kA/m to 1039.6 kA/m. That's a nice improvement, but it was a bit like a security guard only patrolling the front door.
The 900°C Treatment: When they cranked the heat up to 900°C, the story changed dramatically. The coercivity jumped all the way to 1158.2 kA/m. That is a massive gain of 206.7 kA/m (a 21.7% increase). The researchers found that at this higher temperature, the aluminum and zinc traveled much deeper—about 150 to 200 micrometers inside the magnet—compared to only about 50 micrometers at the lower temperature.

There was a tiny trade-off. The "remanence" (how strong the magnet is when it's fully charged) dropped slightly. At 900°C, it fell from 1282 mT to 1256 mT. The authors explain this is because some aluminum atoms managed to slip inside the main grains and replace a few iron atoms, slightly diluting the magnetic power. However, the huge boost in heat resistance was worth this small price.

The "How": A Team of Two with Different Jobs

The most fascinating part of the paper is how the aluminum and zinc worked together, like a dynamic duo with very different personalities.

Zinc is the "Road Crew."
Zinc has a low melting point and loves to turn into gas (vaporize) at high temperatures. The researchers found that as the zinc heated up, it didn't just sit there; it bubbled and moved around in the liquid layer between the grains. The authors suggest this movement acted like a road crew clearing traffic jams. By creating little disturbances in the liquid, it kept the "highways" (the grain boundaries) open and flowing, allowing other atoms to move faster and deeper. Most of the zinc actually evaporated and left the magnet, which explains why the magnets lost a tiny bit of weight after the process.

Aluminum is the "Architect."
While zinc was busy clearing the roads, aluminum was doing the heavy lifting. Aluminum atoms are similar in size to iron atoms, so they could easily slip into the outer shell of the main magnetic grains. They didn't just stay in the cracks; they built a protective "shell" around the grains. This shell is tougher and better at stopping the magnetic "bad guys" (reverse magnetic domains) from taking over. The aluminum also helped the liquid between the grains spread out more evenly, creating a thin, continuous, and smooth layer that perfectly wrapped every single grain.

The Microscopic View: Building a Better City

When the researchers looked at the magnets under a powerful microscope (SEM), the difference between the two temperatures was like comparing a messy neighborhood to a well-planned city.

At 700°C: The gaps between the grains were still a bit patchy. Some grains were touching each other directly, which is bad because it lets the magnetic weakness spread. The protective layer was there, but it wasn't very uniform.
At 900°C: The view was stunning. The grains were perfectly separated by a thin, continuous, and smooth layer. It looked like every skyscraper was wrapped in a protective bubble. The researchers also saw a distinct "core-shell" structure: a dark center (the original grain) surrounded by a bright, thin shell (the aluminum-rich layer). This shell acts like a super-strong fence, making it incredibly hard for the magnet to lose its power.

The Verdict: A Smarter Way Forward

The paper concludes that using an Al-Zn alloy sheet at 900°C is a highly effective way to make sintered Nd-Fe-B magnets much stronger against heat. It suggests that this method works because the zinc keeps the diffusion channels open and fluid, while the aluminum builds a high-quality protective shell around the grains.

The authors are careful to note that while this is a great step forward, it's not a magic bullet that solves everything instantly. They suggest that this process could be used as a "pre-treatment" to prepare the magnet for even more advanced (and expensive) heavy rare-earth elements later on. By using this aluminum-zinc trick first, they might be able to push those expensive elements even deeper into the magnet in the future, getting even better performance without the massive cost.

In short, this study shows that by carefully choosing the right "delivery team" (Al and Zn) and the right "delivery temperature" (900°C), we can build magnets that are tougher, more efficient, and ready for the heat of tomorrow's technology.

Technical Summary: Effect of Al-Zn Alloy Wafer Grain Boundary Diffusion on the Magnetism and Microstructure of Sintered NdFeB Magnets

Problem Statement
Sintered Nd-Fe-B magnets are critical for high-performance applications but suffer from a sharp decline in coercivity at elevated temperatures. The conventional solution involves adding heavy rare earth elements (HREs) like Dysprosium (Dy) or Terbium (Tb) to enhance thermal stability. However, this approach incurs high costs due to HRE scarcity and significantly reduces remanence because HREs couple antiferromagnetically with iron in the main phase lattice. While Grain Boundary Diffusion (GBD) technology offers a more efficient alternative by forming a high-anisotropy shell around grains, it faces limitations in diffusion depth. Recent studies suggest that adding low-melting-point elements (such as Al and Zn) can improve the fluidity of the Nd-rich grain boundary phase, yet most existing methods involve mixing these elements into raw powders, which complicates processing and obscures their specific roles during the diffusion stage.

Methodology
This study investigates a novel approach where Al and Zn are introduced not as powders mixed into the bulk, but as a solid diffusion source placed directly on the magnet surface.

Diffusion Source: Self-prepared $Al_{80}Zn_{20}$ alloy sheets were used, leveraging the Al-Zn eutectic reaction (melting point ~382°C) to ensure early melting and wetting of the magnet surface.
Experimental Setup: Cylindrical sintered Nd-Fe-B magnets (Grade 38SH) were sandwiched between the alloy sheets.
Process Parameters: Diffusion treatments were conducted under vacuum ( $\le 5 \times 10^{-3}$ Pa) at two distinct temperatures: 700°C and 900°C, with a holding time of 7 hours, followed by tempering at 500°C.
Characterization: The study employed Scanning Electron Microscopy (SEM), Energy-Dispersive X-ray Spectroscopy (EDS), and X-ray Diffraction (XRD) to analyze microstructure and elemental distribution. Magnetic properties were measured using a Vibrating Sample Magnetometer (VSM).
Computational Modeling: The authors utilized a multiphysics model combining diffusion kinetics (Arrhenius-based), phase-field microstructure evolution, and micromagnetic simulations (Landau-Lifshitz-Gilbert equations) to validate experimental observations and decompose coercivity enhancement mechanisms.

Key Results

Magnetic Performance:
- Coercivity ( $H_{cj}$ ): The treatment significantly enhanced coercivity. The 900°C sample showed a substantial increase from 951.5 kA/m (untreated) to 1158.2 kA/m (+206.7 kA/m, 21.7%). The 700°C sample increased to 1039.6 kA/m (+88.1 kA/m, 9.3%).
- Remanence ( $B_r$ ): A modest decline in remanence was observed, dropping from 1282 mT to 1256 mT for the 900°C sample and to 1276 mT for the 700°C sample.
- Energy Product: Despite the trade-off between coercivity and remanence, the maximum energy product remained relatively stable across samples.
Microstructural Evolution:
- 900°C Treatment: Produced a thin, continuous, and uniform grain boundary phase. A distinct core-shell structure formed around the main-phase grains ( $Nd_2Fe_{14}B$ ), with a shell thickness of approximately 1–2 µm.
- 700°C Treatment: Resulted in a less continuous grain boundary phase with a shallower diffusion depth (~~50 µm) compared to the 900°C sample (~~150–200 µm).
Elemental Distribution (EDS Analysis):
- Aluminum (Al): Preferentially enriched in the shell region of the main-phase grains (1.24 wt% at the edge vs. 0.31 wt% at the center) and also present in the grain boundary phase. This indicates Al substitution for Fe in the main lattice.
- Zinc (Zn): Predominantly resided in the grain boundary phase (0.42 wt%) and was nearly undetectable in the main phase grains. Mass loss analysis confirmed significant Zn volatilization at 900°C.
Crystal Structure: XRD analysis confirmed that the main phase retained its tetragonal $Nd_2Fe_{14}B$ structure without secondary phase formation. A slight lattice expansion was noted, consistent with partial Al substitution for Fe.

Mechanisms of Action
The study attributes the coercivity enhancement to three synergistic factors verified by computational analysis:

Grain Boundary Decoupling: The formation of a continuous, non-magnetic grain boundary layer effectively isolates main-phase grains, weakening intergranular magnetic exchange coupling.
Domain Wall Pinning: The Al-enriched shell layer possesses higher magnetocrystalline anisotropy, strengthening domain wall pinning at the grain surfaces.
Nucleation Suppression: The smoothing of grain edges and the uniformity of the shell layer suppress the nucleation of reverse magnetic domains.

The synergistic roles of the elements are defined as follows:

Zn: Acts as an "activator." Its high vapor pressure and volatilization create local disturbances in the grain boundary liquid phase, preventing stagnation and keeping diffusion channels open.
Al: Acts as a "fluxing agent" and "positioning agent." It lowers the melting point and viscosity of the grain boundary phase, facilitating flow, and enters the main-phase lattice to prepare the structure for potential heavy rare earth diffusion.

Significance and Claims
The paper claims that using an $Al_{80}Zn_{20}$ alloy wafer as a diffusion source is an effective, non-heavy-rare-earth pathway for enhancing the coercivity of sintered Nd-Fe-B magnets, particularly for high-temperature applications. The study demonstrates that the 900°C treatment is considerably more effective than 700°C, primarily due to the Arrhenius-driven increase in diffusion coefficients and improved fluidity of the grain boundary phase.

The authors position this work as a step toward optimizing grain boundary diffusion processes without the immediate need for expensive heavy rare earths. They suggest that this Al-Zn pre-diffusion treatment creates a favorable microstructure (continuous boundaries and lattice-prepared shells) that could serve as a foundation for subsequent, deeper diffusion of heavy rare earths in a two-step process, thereby addressing the depth limitations of current GBD technologies. The research provides a clear mechanistic understanding of how low-melting-point elements interact during diffusion, offering a strategy to improve demagnetization resistance while minimizing remanence loss.

The Mission: A New Delivery Service for Magnets

The Results: Heat Makes the Magic Happen

The "How": A Team of Two with Different Jobs

The Microscopic View: Building a Better City

The Verdict: A Smarter Way Forward

More like this