🔬 condensed matter

Dislocation-Driven Nucleation Type Switching Across Repeated Ultrafast Magnetostructural Phase Transition

Using in situ transmission electron microscopy, researchers demonstrate that repeated ultrafast laser irradiation induces dislocation networks in FeRh thin films, which switch the antiferromagnetic-to-ferromagnetic phase transition from homogeneous to heterogeneous nucleation, lowering the transition temperature and stabilizing sub-micron magnetic vortices.

Original authors: Jan Hajduček, Antoine Andrieux, Jon Ander Arregi, Martin Tichý, Paolo Cattaneo, Beatrice Ferrari, Fabrizio Carbone, Vojtěch Uhlíř, Thomas LaGrange

Published 2026-01-15

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Jan Hajduček, Antoine Andrieux, Jon Ander Arregi, Martin Tichý, Paolo Cattaneo, Beatrice Ferrari, Fabrizio Carbone, Vojtěch Uhlíř, Thomas LaGrange

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 a thin sheet of metal, just 15 nanometers thick (about 5,000 times thinner than a human hair), made of an alloy called FeRh. Under normal conditions, this metal is a bit of a mood swing. When it's cool, it's "antiferromagnetic," meaning its tiny internal magnets are pointing in opposite directions, canceling each other out. When you heat it up, it suddenly snaps into a "ferromagnetic" state, where all the magnets line up in the same direction, turning the sheet into a magnet.

This switch isn't just a gentle change; it's a violent, first-order phase transition, like water suddenly turning to ice. Usually, when this happens, the new magnetic state starts forming in a few random spots and then spreads out evenly across the sheet, like a drop of ink slowly diffusing in water.

The Experiment: Zapping the Metal
The researchers in this paper wanted to see what happens if they zap this metal sheet with a laser, over and over again, while watching it through a super-powerful microscope (a Transmission Electron Microscope). They didn't just heat it once; they gave it a cumulative "workout" of laser pulses.

Think of the laser pulses like a drummer hitting a drum. At first, the drum skin (the metal) just vibrates. But if you hit it hard enough and fast enough, the skin itself starts to change shape.

The Big Discovery: From Smooth to Spotty
Here is the surprising part:

The First Time: When they first zapped the clean metal, the magnetic change happened smoothly and evenly (homogeneous nucleation). It was like a calm, uniform wave rolling across the surface.
After Many Zaps: After they repeated this process thousands of times, something changed. The metal had developed tiny scars and wrinkles inside its crystal structure, called dislocations. These are like microscopic cracks or tangles in the metal's atomic grid.

Once these "scars" formed, the magnetic switch changed its behavior completely. Instead of a smooth wave, the new magnetic state started popping up in specific, chaotic spots right where the scars were. It switched from a smooth wave to a "staccato" pattern of many tiny, isolated islands of magnetism.

The Vortex Effect
Even more interesting, these new magnetic islands didn't just look like solid blobs. They formed vortices. Imagine a whirlpool in a bathtub. The magnetic spins in these tiny islands were swirling around a center point, creating a stable, topological shape.

The paper shows that these whirlpools were "pinned" or stuck in place by the dislocation networks (the scars). The metal's internal damage actually acted as a trap, forcing the magnetic swirls to form in specific patterns.

Why It Matters (According to the Paper)

Lower Energy Needed: Because the metal was "pre-damaged" by the laser, it took less energy (about 50% less laser power) to trigger the magnetic switch the second time around. The scars made it easier for the change to happen.
Lower Temperature: The metal would switch to its magnetic state at a lower temperature (about 20 degrees Celsius lower) after the laser treatment.
The "Memory" of Damage: The paper emphasizes that the laser didn't just heat the metal; it physically rearranged the atomic defects. These defects then dictated how the metal would behave in the future.

The Takeaway
The study reveals that if you keep hitting a material with ultrafast lasers, you aren't just heating it; you are rewriting its internal map. You are creating a landscape of defects that forces the material to change its magnetic state in a completely different, more chaotic, and vortex-filled way than it would on its own.

The researchers conclude that this is a direct link between defects (the scars) and nucleation (how the new phase starts). They showed that by controlling these defects with light, you can fundamentally change the rules of how the material switches states, turning a smooth transition into a textured, vortex-filled one.

Technical Summary: Dislocation-driven nucleation type switching across repeated ultrafast magnetostructural phase transition

Problem Statement
While disorder and dislocations are known to modulate phase transitions in functional materials, particularly first-order phase transitions (FOPTs), the microscopic role of dislocations in controlling the earliest stages of phase nucleation remains elusive. This gap is critical in nonequilibrium, laser-driven systems where repeated stimuli can dynamically create or stabilize defects, potentially altering the phase transition pathway. Existing studies on FeRh have focused on the collective influence of disorder or the temporal dynamics of laser-driven transitions, but the local mechanisms by which dislocations and their strain fields mediate phase nucleation during cyclic excitation have not been directly observed. Furthermore, time-resolved stroboscopic studies often lack the spatial resolution to detect these defect-mediated irreversibilities.

Methodology
The authors employed an in situ methodology combining high-spatial-resolution imaging with controlled, repeatable laser excitation within a transmission electron microscope (TEM).

Sample: 15-nm-thick epitaxial FeRh films were grown on MgO(001) substrates and subsequently detached to create freestanding flakes on Cu TEM grids. This geometry facilitates strain relaxation and allows for the examination of the interplay between lattice defects, strain, and magnetic phase formation.
Imaging: Lorentz TEM (LTEM) was used to visualize magnetic domain walls and AF/FM phase boundaries. Transfer-of-Intensity Equation (TIE) reconstruction was applied to generate vector-resolved maps of in-plane magnetic induction.
Excitation: The samples were subjected to cumulative femtosecond laser pulses (1030 nm wavelength) with systematically varied parameters: pulse energy ( $E_p$ ), repetition rate ( $f_{rep}$ ), and pulse duration ( $t_{pulse}$ ).
Simulation: Numerical thermal simulations (COMSOL Multiphysics) were performed to model temperature excursions and cooling rates, while Eshelby's inclusion model was used to estimate hydrostatic stresses.
Structural Analysis: High-resolution scanning TEM (HR-STEM) and two-beam diffraction conditions were utilized to characterize the structural environment of nucleation sites, specifically identifying dislocation types (edge and screw).

Key Results

Nucleation Regime Switching: The study demonstrates that cumulative laser irradiation fundamentally alters the nucleation pathway of the antiferromagnetic (AF) to ferromagnetic (FM) transition.
- Homogeneous Nucleation: In pristine films or under low-energy/low-repetition conditions, the FM phase emerges via spontaneous formation of a limited number of nucleation centers, leading to uniform domain growth.
- Heterogeneous Nucleation: Upon repeated exposure to laser pulses exceeding a specific threshold (pulse energy > 500 nJ, repetition rate < 10 kHz), the mechanism switches to heterogeneous nucleation. This occurs at specific, structurally favorable sites, resulting in a significantly higher density of nucleation centers.
Thermodynamic Shifts: The switch to heterogeneous nucleation lowers the transition temperature by approximately 20 K (from ~350 K to ~330 K) and reduces the required laser fluence threshold by up to 50%. The hysteresis loop width also increases by more than 5 K.
Emergence of Magnetic Vortices: In the heterogeneous regime, the dominant magnetic motif shifts from continuous domains to submicron magnetic vortices. These vortices are pinned in the film by underlying dislocation networks and remain stable across repeated laser cycles.
Defect Origin: High-resolution structural analysis reveals that the heterogeneous nucleation sites correspond to dense networks of dislocations (both edge and screw types) formed and rearranged during repeated phase transitions. These dislocation networks create localized strain fields that lower the energy barrier for FM nucleation.
Stress Mechanism: The authors estimate that the combination of the 1% lattice volume expansion during the AF-FM transition and rapid thermal gradients generates hydrostatic stresses on the order of gigapascals. In freestanding films, these stresses mobilize preexisting misfit dislocations, which reorganize into stable Frank networks that act as preferential nucleation centers.

Significance and Claims
The paper establishes a direct link between defect formation and the microscopic morphology of the nucleated ferromagnetic phase. The authors claim that their results:

Provide the first direct observation of how cumulative laser exposure reshapes the local microstructure (specifically dislocation networks) to alter phase nucleation kinetics.
Demonstrate that defect engineering can be used to tune nucleation pathways, shifting them from homogeneous to heterogeneous regimes and stabilizing topological magnetic textures (vortices).
Highlight a critical consideration for ultrafast stroboscopic experiments: long-term laser exposure may inadvertently alter the energy landscape of the material by inducing defect-mediated changes, which could lead to misinterpretations of time-resolved data if not accounted for.
Offer a pathway for controlling phase transitions in functional materials via photo-induced structural modifications, with potential relevance for reconfigurable magnetic memory and ultrafast spintronic devices, though the paper frames these as potential implications rather than demonstrated applications.

The work underscores that the interplay between laser-induced strain, defect stabilization, and transformation dynamics is a crucial factor in the behavior of magnetic materials under cyclic ultrafast excitation.

More like this