🔬 materials science

Nanoscale Imaging of Strain-Controlled Altermagnetic Domains in {\alpha}-MnTe

This study utilizes scanning nitrogen-vacancy magnetometry to reveal that mechanical strain controls altermagnetic domains in bulk $\alpha$ -MnTe through a hysteretic process of domain coalescence during compression and metastable fragmentation upon unloading, establishing a pathway for strain-programmable spintronic devices.

Original authors: Alex L. Melendez, Sijie Xu, Liangbo Liang, An-Ping Li, Pengcheng Dai, Hu Miao, Zhaoyu Liu, Huan Zhao

Published 2026-07-30

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Alex L. Melendez, Sijie Xu, Liangbo Liang, An-Ping Li, Pengcheng Dai, Hu Miao, Zhaoyu Liu, Huan Zhao

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 you are trying to organize a chaotic crowd of people at a music festival. Some people are wearing red shirts, some blue, and they are all jumbled together. Now, imagine you have a giant, invisible hand that can gently squeeze the crowd from the sides. In the world of advanced electronics, scientists are trying to do something similar, but instead of people, they are organizing tiny magnetic "spins" inside a material. These materials, called altermagnets, are a special new type of magnet. They are like the best of both worlds: they act like regular magnets in how they move electricity, but they don't have the messy, stray magnetic fields that usually mess up nearby electronics. This makes them perfect candidates for building super-fast, super-dense computer chips that don't overheat or interfere with each other.

The big question scientists have been asking is: How exactly does squeezing these materials change their magnetic order? Do the tiny magnetic spins just slowly turn like a compass needle? Or do they suddenly snap into a new pattern, like a crowd suddenly clearing a path? Understanding this is crucial because if we can control these patterns with a simple squeeze (strain), we could build computers that remember information just by how they were bent, leading to a new generation of "programmable" electronics.

In this study, researchers decided to take a direct look at what happens inside a specific material called $\alpha$ -MnTe (a type of manganese telluride) when they squeeze it. They used a high-tech "microscope" called a nitrogen-vacancy (NV) magnetometer, which acts like a tiny, sensitive nose that can smell magnetic fields from just a hair's breadth away. They placed a chunk of this material in a special machine that could squeeze it along a specific direction while they watched the magnetic patterns change in real-time.

Here is what they found: When they started squeezing the material, the tiny magnetic regions didn't just rotate smoothly. Instead, they acted like a crowd of people merging into larger groups. Small, scattered patches of magnetic "red" and "blue" spots began to stick together, forming massive, connected islands. This process, called coalescence, made the biggest magnetic domains grow huge while the walls between them disappeared. It was as if the squeeze forced the crowd to organize into one giant, unified block.

However, the most surprising part happened when they let go. When the researchers released the pressure, the material didn't just snap back to how it looked before. The giant, unified magnetic islands didn't just shrink back to their original small size. Instead, they fragmented into a completely new, messy pattern that was different from the starting point. It was like squeezing a sponge: when you let go, it doesn't just return to its exact original shape; it gets stuck in a slightly different, squished configuration.

This means the material has a memory. The way the magnetic domains are connected and organized holds onto the history of how much it was squeezed. The researchers found that while the total amount of "boundary" between magnetic regions changed only a little, the size of the big magnetic islands and the strength of the magnetic fields showed a huge difference between squeezing and releasing. This proves that the material's magnetic state isn't just a simple, reversible turn; it's a complex dance of merging and breaking apart that leaves a lasting mark.

In short, this paper shows that squeezing $\alpha$ -MnTe doesn't just turn a knob; it reorganizes the entire magnetic landscape. The material remembers the squeeze by getting stuck in a new, metastable state. This discovery is a big step toward understanding how we might one day "write" information into materials just by bending them, paving the way for a future where our electronics can be reconfigured on the fly.

Technical Summary: Nanoscale Imaging of Strain-Controlled Altermagnetic Domains in α-MnTe

Problem Statement
Altermagnets represent a distinct magnetic order combining compensated real-space magnetization (like antiferromagnets) with momentum-dependent spin splitting (like ferromagnets), offering potential for spintronic applications free from stray fields. A central challenge in altermagnetic spintronics is the deterministic manipulation of the Néel order. While mechanical strain has been shown to modify the anomalous Hall response and lift the degeneracy of magnetic variants in the prototypical altermagnet $\alpha$ -MnTe, the microscopic pathway remains unresolved. Previous studies have offered conflicting interpretations: one suggesting strain-induced "detwinning" (redistribution of magnetic variant populations toward a single-domain state) and another suggesting continuous local rotation of the Néel vector. Crucially, these interpretations have largely relied on spatially averaged measurements, leaving a gap in understanding how strain-driven Néel-vector reorientation is realized through the motion, merging, and reorganization of individual nanoscale domains.

Methodology
To address this, the authors integrated a piezo-driven uniaxial strain cell with scanning nitrogen-vacancy (NV) magnetometry. This setup enabled the imaging of the same microscopic region of bulk $\alpha$ -MnTe while compressive strain was varied in situ at room temperature.

Sample Preparation: Bulk $\alpha$ -MnTe crystals were polished, and a specific region of interest (ROI) was characterized via atomic force microscopy (AFM) to ensure minimal surface roughness ( $h_{rms} = 3.7$ nm).
Strain Application: Strain was applied along the $[110]$ crystallographic direction using a piezo-driven cell. The strain magnitude was calibrated using large-area AFM scans tracking surface particle displacement.
Imaging: A diamond-tipped probe containing a single NV center scanned the sample at a height of $\sim$ 40 nm. The NV center sensed the projection of the stray field ( $B_{\parallel}$ ), which was reconstructed into out-of-plane (OOP) magnetic field ( $B_z$ ) maps using Fourier-space reconstruction enforcing $\nabla \cdot B = 0$ .
Analysis: The study tracked the evolution of magnetic textures through a complete compression-release cycle. Key metrics included the maximum positive-domain area, total area of large domains ( $>5 \mu m^2$ ), domain-wall density, and the full-width at half-maximum (FWHM) of the $B_z$ distribution.

Key Results
The study reveals that strain controls the magnetic texture in $\alpha$ -MnTe through a process of domain coalescence and hysteretic fragmentation:

Loading (Compression): As compressive strain increases along $[110]$ $[110]$ , the magnetic texture undergoes substantial reorganization. The strain promotes the coalescence of neighboring regions with compatible projected OOP moments ( $M_z$ $M_{z}$ ) into larger connected domains. This is evidenced by:
- An increase in the maximum connected positive-domain area and the total area of large domains.
- A reduction in the visible domain-wall density.
- A broadening of the $B_z$ distribution (increased FWHM), indicating reduced cancellation between intermixed positive and negative regions.
- The texture remains multidomain but shifts toward a strain-selected state, with the largest connected domains growing significantly.
Unloading (Release): Upon releasing the strain, the domain network does not retrace the loading pathway. Instead, the large connected regions fragment into a new metastable configuration. This results in:
- A rapid decrease in the largest connected domain size, which becomes smaller than the initial unstrained state.
- Pronounced hysteresis in the maximum domain size and the $B_z$ distribution width.
- A narrowing of the $B_z$ distribution as the network fragments, increasing local field cancellation.
Domain-Wall Density Hysteresis: Unlike the domain size metrics, the total domain-wall density exhibits only weak hysteresis. This suggests that the total visible boundary length is governed primarily by the instantaneous applied strain, whereas the connectivity and topology of the domain network carry the magnetic memory.

Significance and Claims
The paper claims to resolve the debate between "detwinning" and "continuous rotation" models by demonstrating that these are not mutually exclusive but represent different stages of the same strain-driven evolution.

Microscopic Pathway: The work establishes that local Néel-vector reorientation is realized through domain-boundary motion, specifically coalescence during loading and fragmentation during unloading.
Memory Mechanism: The authors identify domain connectivity and topology, rather than just the total amount of domain wall, as the key carriers of strain-induced magnetic memory. The system retains a "memory" of the strain trajectory in the form of a metastable domain configuration.
Implications for Transport: The observed real-space hysteresis provides a microscopic basis for the persistent strain dependence of the anomalous Hall effect in $\alpha$ -MnTe. The incomplete recovery of the domain network during unloading suggests that transport responses may encode the history of the strain trajectory, not just the instantaneous strain state.
Platform for Engineering: The study establishes scanning NV magnetometry under in situ strain as a platform for resolving and engineering altermagnetic domain dynamics, offering a route toward strain-programmable altermagnetic textures and reconfigurable spintronic devices.

More like this