Ultrafast X-ray sonography reveals the spatial heterogeneity of the laser-induced magneto-structural phase transition in FeRh

arXiv:2507.16638 · cond-mat.mtrl-sci · Submitted 2025-07-22 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: We're looking at the paper: "Ultrafast X-ray sonography reveals the spatial heterogeneity of the laser-induced magneto-structural phase transition in FeRh".

Mira: Here's the gist of the paper. Ultrafast X-ray sonography reveals the spatial heterogeneity of the laser-induced magneto-structural phase transition in FeRh by resolving phase-specific strain response to identify nucleation mechanisms and domain dimensions The Core Methodology The researchers introduce ultrafast X-ray sonography, a combination of time-resolved hard-X-ray diffraction with sound waves in form of high-frequency acoustic strain pulses of picosecond duration to capture spatio-temporal phase heterogeneity by resolving the phase-specific strain response.

Kai: This non-invasive pump-probe technique leverages transient structural markers, such as sound waves, to localize coexisting phases in both time and space by following their evolving X-ray signatures.

Mira: The concept hinges on the relation between phase heterogeneity and phasespecific strain response.

Kai: Identification of Nucleation Scenarios The study applies this technique to the antiferromagnetic-to-ferromagnetic (AFM–FM) magneto-structural phase transition in FeRh.

Title and authors: Kai: So we're looking at this paper, "Ultrafast X-ray sonography reveals the spatial heterogeneity of the laser-induced magneto-structural phase transition in FeRh," and it’s using ultrafast X-ray sonography to map out where and how this phase transition happens across the material.

Mira: It’s about using sound waves, specifically picosecond acoustic strain pulses, to capture that spatio-temporal heterogeneity by seeing how different structural phases react differently to that mechanical shock.

Lev: So what we're hearing is a non-invasive way to look at things in real time—like seeing the structure evolve under a laser pulse—by using these transient markers.

Kai: Right, and this technique lets them distinguish between five specific nucleation scenarios for the antiferromagnetic to ferromagnetic phase transition in FeRh, covering pure out-of-plane, pure in-plane, and mixed heterogeneities.

Mira: That’s interesting because they use analytical expressions derived from the relation between volume fraction and relative thickness to specify those scenarios.

Lev: From an engineering standpoint, that sounds like a way to model complex kinetics without needing incredibly slow measurement times.

Kai: And then they move on to quantifying the domain dimensions by looking at how much the FM Bragg peak broadens along the out-of-plane reciprocal coordinate qz.

Mira: They estimate the lateral diameter of these nucleated FM domains to be approximately thirty nanometers for FeRh, which they found by accounting for stochastic nucleation.

Lev: That thirty nanometer estimate sounds like a concrete number we can actually use when trying to design hardware that relies on these magnetic domains.

Kai: The validation showed that scenario three optimally describes the nucleation of the FM phase during this transition, and it fits their model for fluences above seven point seven mJ cm−two.

Mira: They also noted that these results align with previous experiments regarding the slower rise of macroscopic magnetization governed by domain wall motion.

Lev: That connection to domain wall motion is important because it grounds this ultrafast observation in something we already understand about how magnetism evolves over time.

Kai: The authors suggest a future direction involving double-pulse excitation schemes where a delayed second excitation launches a dedicated strain pulse to mark heterogeneity at a selected pump-pump delay.

Mira: That approach gives them flexibility because they can tune the X-ray photon energy to access buried layers and probe different degrees of freedom.

Lev: Tuning the energy is smart because it lets you select which structural elements you're looking at, which is crucial when you're trying to map things out in real hardware.

Kai: So, to wrap up this paper on ultrafast X-ray sonography revealing the spatial heterogeneity of the laser-induced magneto-structural phase transition in FeRh, it shows how this technique can spatially map these transitions by resolving phase-specific strain response.

Mira: It provides a method for identifying nucleation mechanisms and determining domain dimensions, giving us a better picture of how these magnetic structures form under ultrafast driving conditions.

Lev: The limitation they mention is that the uncertainty in determining domain size is estimated to be ten nanometers based on their analysis, which means we can't get down to atomic scale resolution with this specific measurement setup <ref:2507.16638#pg1>.

Kai: So, for someone who only listens to the show, it changes how you think about phase transitions in materials by showing you how spatial heterogeneity dictates the nucleation process.

Mira: It moves beyond just knowing if a transition happens to understanding exactly where and in what shape that transition begins across the sample volume.

Lev: It gives us a way to predict domain growth kinetics, which is something we need when we’re trying to design error correction codes for real hardware operating under these kinds of non-equilibrium conditions.

Kai: We'll keep an eye on how this concept of phase-specific strain response can be applied to other complex heterostructures in the future.

Mira: Exactly, and we'll see if that double-pulse excitation scheme proves useful for probing different structural elements in materials like FeRh.

Lev: We’re looking forward to seeing how this method scales up when you try to run these simulations on actual hardware platforms.

The paper's summary: Kai: So, we just looked at how ultrafast X-ray sonography maps out where this laser pulse actually triggers that phase transition in Iron Rhodium, FeRh.

Mira: It’s basically using sound waves that are picoseconds long to probe the material as it changes structure and seeing how different parts of the sample react differently to that shock.

Lev: I see they're looking at five different ways this transition can start—pure out-of-plane, pure in-plane, or a mix of both—and they’re using math based on volume fraction to figure out which one is most likely happening.

Kai: That five-scenario model helps them pinpoint the nucleation mechanism under these laser conditions.

Mira: The real payoff here is that they can quantify the size of those new magnetic domains by looking at how much the FM Bragg peak broadens along a specific direction, qz.

Lev: They estimate those domains to be about thirty nanometers wide for FeRh, which gives us a tangible number for domain sizing.

Kai: The paper confirms that scenario three is the best fit for the nucleation kinetics when you hit fluences above seven point seven mJ per centimeter squared.

Mira: It connects this fast structural change to slower macroscopic magnetization rise, which makes sense because that rise is governed by how those domains move their walls.

Lev: If we were trying to build a device that relies on these magnetic properties, knowing the domain size and the nucleation pathway is crucial for figuring out the error correction overhead you'd need.

Kai: The authors also proposed this double-pulse excitation scheme, where a delayed second pulse launches a strain wave to mark heterogeneity at a specific time delay.

Mira: That technique lets them tune the X-ray photon energy to access different layers of the sample, so they can get spatial information both in and out of the plane.

Lev: So it's not just about seeing one thing; it’s a way to fully parametrize the three-dimensional nature of this heterogeneity.

Kai: It gives us a tool that’s positionally sensitive, letting AI potentially predict where and how these ferromagnetic domains will nucleate in more complex heterostructures.

Mira: The limitation they mention is that the uncertainty in determining domain size is around ten nanometers based on their analysis, so we can't get down to atomic scale resolution with this specific setup.

Lev: That ten nanometer uncertainty means when you translate this to real hardware, you're looking at a certain level of noise or variation in your domain sizes that needs to be factored into the design parameters.

The paper's improvements: Tom: So, we're talking about how the authors suggest they can make this measurement tool better for future research on phase transitions in materials like FeRh.

Kai: They propose using double-pulse excitation schemes where a second, delayed pulse launches a dedicated strain wave to mark the heterogeneity at a specific pump-pump delay.

Mira: That lets them select which structural elements they are probing by tuning the X-ray photon energy, which is really flexible for sample design.

Lev: From an engineering standpoint, that flexibility is key because it allows selective probing of specific degrees of freedom within the material structure.

Kai: It also means they can access buried layers if they tune that X-ray photon energy correctly, so it’s not just surface stuff anymore.

Mira: So this moves beyond just volume fractions to fully parameterizing the three-dimensional nature of this phase heterogeneity, which is a big step.

Lev: I think that's where the real power is, getting full spatial maps instead of just averaged data points across the sample volume.

Kai: The goal is positioning this tool as a universal non-invasive probe for characterizing diverse structural changes in many different crystalline materials.

Mira: It’s about making sure you can map out the transition kinetics across the whole material, not just a tiny spot where it first starts.

Lev: That level of spatial control would be essential when you're building models for real hardware where local variations in material properties matter a lot.

Kai: So they're aiming for a tool that can give us quantitative access to phase heterogeneity across a broad range of materials, not just FeRh.

Mira: It’s about using this technique to see how different structural responses enable the direct localization along the layer’s depth.

Lev: That specificity is what makes it useful for understanding the underlying physics of how these transitions actually happen at a molecular level.

Conclusion: Tom: We've covered how ultrafast X-ray sonography maps out where and how this laser pulse triggers the phase transition in FeRh, using transient strain pulses to see different structural responses.

Kai: So, essentially, they are showing us that we can spatially map these transitions by resolving the phase-specific strain response in real time.

Mira: It gives us a much better picture than just knowing if a transition happens at all; it tells us exactly where and in what shape those new magnetic domains begin to form.

Lev: For someone building hardware, that spatial mapping is important because it helps you understand the local variations in material properties that you have to account for.

Kai: The paper confirms that scenario three is the best fit for the nucleation kinetics when you hit fluences above seven point seven mJ per centimeter squared.

Mira: It connects this fast structural change to slower macroscopic magnetization rise, which makes sense because that rise is governed by how those domains move their walls.

Lev: If we were trying to build a device that relies on these magnetic properties, knowing the domain size and the nucleation pathway is crucial for figuring out the error correction overhead you'd need.

Kai: The authors also proposed this double-pulse excitation scheme where a second, delayed pulse launches a strain wave to mark heterogeneity at a specific time delay.

Mira: That lets them tune the X-ray photon energy to access different layers of the sample, so they can get spatial information both in and out of the plane.

Lev: So it's not just about seeing one thing; it’s a way to fully parametrize the three-dimensional nature of this heterogeneity.

Kai: It gives us a tool that’s positionally sensitive, letting AI potentially predict where and how these ferromagnetic domains will nucleate in more complex heterostructures.

Mira: The limitation they mention is that the uncertainty in determining domain size is around ten nanometers based on their analysis, so we can't get down to atomic scale resolution with this specific setup.

Lev: That ten nanometer uncertainty means when you translate this to real hardware, you're looking at a certain level of noise or variation in your domain sizes that needs to be factored into the design parameters.

Kai: So, we wrap up on the paper "Ultrafast X-ray sonography reveals the spatial heterogeneity of the laser-induced magneto-structural phase transition in FeRh," showing how this technique can spatially map these transitions by resolving phase-specific strain response.

Mira: It provides a method for identifying nucleation mechanisms and determining domain dimensions, giving us a better picture of how these magnetic structures form under ultrafast driving conditions.

Lev: And the suggestion to use double-pulse excitation schemes shows where the future development of this probe is heading in terms of flexibility.

Kai: So, that's what we have on this one—a powerful way to see phase transitions spatially using sound waves and X-rays.

Mira: It moves beyond just knowing if a transition happens to understanding exactly where and in what shape that transition begins across the sample volume.

Lev: It gives us a way to predict domain growth kinetics, which is something we need when you’re trying to design error correction codes for real hardware operating under these kinds of non-equilibrium conditions.

Kai: We'll keep an eye on how this concept of phase-specific strain response can be applied to other complex heterostructures in the future.

Maximilian Mattern, Angel Rodriguez-Fernandez, Roman Shayduk, Jon Ander Arregi, Vojtš Uhlírš, Ulrike Boesenberg, ංorg Hallmann, Wonhyuk Jo, Aliaksandr Leonau, Rustam Rysov, James Wrigley, Alexey Zozulya, Stefan Eisebitt, ංnder Madsen

Max Born Institute for Nonlinear Optics and Short Pulse Spectroscopy · European X-ray Free-Electron Laser Facility · ංCEITEC BUT Brno University of Technology · Institute of Physical Engineering Brno University of Technology · Department Physik Universität Siegen Institute für Physik und Astronomie Technische Universitš Berlin

cond-mat.mtrl-sci

Submitted: 2025-07-22

Updated: 2025-07-22

Journal ref: Communications Materials 7 (1), 261 (2026)

DOI: 10.1038/s43246-026-01378-5

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 83/100

The gist: Ultrafast X-ray sonography reveals the spatial heterogeneity of the laser-induced magneto-structural phase transition in FeRh by resolving phase-specific strain response to identify nucleation…

Key concepts

Ultrafast X-ray Sonography
This technique combines hard X-ray diffraction with high-frequency acoustic strain pulses lasting picoseconds. It acts like a non-invasive pump-probe method, using sound waves to track and localize structural changes in both time and space by observing how different phases react to the laser excitation.
Magneto-structural Phase Transition
This is the change in material structure (from antiferromagnetic to ferromagnetic) triggered by a laser pulse. The study focuses on understanding how this transition happens spatially, meaning identifying exactly where and how these magnetic changes initiate within the material.
Phase-Specific Strain Response
Different structural phases in FeRh react differently to the laser's energy input, causing distinct strain patterns. By analyzing these unique strain signatures, researchers can distinguish between different nucleation scenarios—such as pure out-of-plane versus mixed phase heterogeneities—to pinpoint the exact physical conditions for phase nucleation.
Reciprocal Space Peak Broadening
This is a measurement technique used to quantify the size of crystalline domains. When a material has small, randomly distributed domains, its diffraction peaks in reciprocal space broaden. The study uses this broadening of the FM Bragg peak to estimate the lateral diameter of nucleated ferromagnetic domains, finding them to be around 30 nm.

Terminology

Summary

Ultrafast X-ray sonography reveals the spatial heterogeneity of the laser-induced magneto-structural phase transition in FeRh by resolving phase-specific strain response to identify nucleation mechanisms and domain dimensions

The Core Methodology

The researchers introduce ultrafast X-ray sonography, a combination of time-resolved hard-X-ray diffraction with sound waves in form of high-frequency acoustic strain pulses of picosecond duration to capture spatio-temporal phase heterogeneity by resolving the phase-specific strain response. This non-invasive pump-probe technique leverages transient structural markers, such as sound waves, to localize coexisting phases in both time and space by following their evolving X-ray signatures. p. 1

Where the paper says it

We leverage the strain pulses as a transient structural marker, enabling localisation of coexisting phases in both time and space by following their evolving X-ray signatures on ultrafast time scales. This non-invasive pump-probe technique can track the coherent growth and stochastic nucleation of structural domains with almost no requirements for sample preparation.

The concept hinges on the relation between phase heterogeneity and phasespecific strain response. p. 1
Where the paper says it

In the case of the coexistence of two distinct structural phases, the specific strain response of each phase enables its direct localisation along the layer’s depth and therefore provides access to spatial heterogeneities both in and out of the plane. This key concept of ultrafast X-ray sonography, namely the relation between phase heterogeneity and phasespecific strain response, is illustrated in Fig.

Identification of Nucleation Scenarios

The study applies this technique to the antiferromagnetic-to-ferromagnetic (AFM–FM) magneto-structural phase transition in FeRh. The analysis distinguishes between five realistic nucleation scenarios, covering pure out-of-plane (I), pure in-plane (II), and mixed phase heterogeneities (III–V). p. 4

Where the paper says it

Distinguishing between the other two key scenarios of pure out-of-plane and mixed heterogeneities is more subtle and requires a comparison of the experimental X-ray sonogram with simulated data, which can even quantify the transient relative thickness DFM(t) and in-plane coverage AFM(t) of the FM phase via Eq. Ultrafast domain nucleation in FeRh In the following, we discuss five realistic nucleation scenarios, which are sketched above each panel b–f in Fig.

The model uses analytical expressions derived from the relation between volume fraction and relative thickness to specify these scenarios. p. 4
Where the paper says it

(2), we can find simple analytical expressions to specify the individual scenarios, which are described in more detail in the Methods section. Irrespective of the nucleation scenario, the rise of the FM volume fraction is constrained via Eq.

Quantification of Domain Dimensions

The method allows for the quantification of domain dimensions by exploiting the broadening of the FM Bragg peak along the out-of-plane reciprocal coordinate qz. For FeRh, this broadening is used to estimate the lateral diameter of nucleated FM domains to approximately 30 nm. p. 7

Where the paper says it

However, the investigated FeRh film is composed of small mosaic crystallites with an average diameter of approximately 25 nm [32] significantly broadening the Bragg peaks along qx and qy. Instead, we employ the broadening of the FM Bragg peak along the out-of-plane reciprocal coordinate qz to estimate the lateral diameter of the nucleated FM domains to approximately 30 nm (further details are given in the Methods section and in Extended Data Figure E4).

This estimation accounts for the stochastic nature of domain nucleation by averaging situations as sketched in Extended Data Figure E4. p. 9
Where the paper says it

Estimating the in-plane domain dimension Extended Data Figure E1c displays an additional broadening of the FM Bragg peak that is not captured by our modelling. This additional broadening decays approximately on the nucleation timescale of the FM phase indicating an origin from the stochastic lateral domain nucleation.

Validation and Conclusion

Quantitative comparison between experimental and modelled sonograms shows that scenario III optimally describes the nucleation of the FM phase during the laser-induced magneto-structural phase transition for FeRh. The global quadratic deviation analysis identifies scenario III to describe the nucleation kinetics for fluences above 7. p. 6

Where the paper says it

We display χ 2 as a function of the final FM volume fraction V ∗ FM to exclude any local minima in Fig. Again, this global analysis identifies scenario III to optimally describe the nucleation of the FM phase during the laser-induced magneto-structural phase transition for the investigated FeRh layer, excited by an optical laser fluence of 7.

7 mJ cm−2. Furthermore, the results align with previous experiments regarding the slower rise of macroscopic magnetisation governed by domain wall motion. p. 6
Where the paper says it

7 mJ cm−2 . Importantly, our result is in line with a much slower rise of the macroscopic magnetisation, as shown in previous experiments [17, 25], governed by aligning the magnetisation of the nucleated FM domains that initially points equally along the different magnetic easy axes via domain wall motion.

Future Directions

The authors propose double-pulse excitation schemes where a delayed second excitation launches a dedicated strain pulse to mark phase heterogeneity at a selected pump-pump delay. This approach offers flexibility in sample design and allows access to buried layers by tuning the X-ray photon energy. p. 7

Where the paper says it

Utilising a photoacoustic transducer with a band gap enables two-colour double-pulse excitation schemes that selectively excite the transducer without modifying the transition kinetics of the actual sample, and vice versa. Although the choice of hard Xrays renders sonography very flexible in sample design and allows access also to buried layers [27], the X-ray photon energy can be tuned to absorption resonances down to the tender X-ray range and potentially even into the soft X-ray range to selectively probe different degrees of freedom and/or elements.

The tool is positioned as a universal non-invasive probe providing quantitative access to phase heterogeneity in a broad range of crystalline materials. p. 7
Where the paper says it

Although the choice of hard Xrays renders sonography very flexible in sample design and allows access also to buried layers [27], the X-ray photon energy can be tuned to absorption resonances down to the tender X-ray range and potentially even into the soft X-ray range to selectively probe different degrees of freedom and/or elements. Since the emergence of longrange ordered spins, charges, and orbitals is often coupled to the crystal lattice that parametrises the phase transition, ultrafast X-ray sonography is a universal non-invasive tool providing quantitative access to phase heterogeneity in a broad range of crystalline materials.

Sample Preparation and Modeling Details

The FeRh(001) film was grown by magnetron sputtering from an equiatomic FeRh target on a MgO(001) substrate. The modelling utilizes the modular Python library udkm1Dsim to calculate spatio-temporal lattice strain based on thermophysical parameters calibrated from previous experiments. p. 8

Where the paper says it

the sonogram, using the modular Python library udkm1Dsim [41]. In the first step, we calculated the spatio-temporal lattice strain within the FeRh layer.

The model explicitly considers the volume changing character of the phase transition by introducing additional stresses corresponding to an out-of-plane expansion of 0. p. 4
Where the paper says it

Here, we use essentially the parameters already calibrated in our previous experiment on the very same sample [18]. In addition, we explicitly consider the unit cell’s volumetric changes through the phase transition by introducing additional stresses that correspond to an out-of-plane expansion of 0.

6 % with the appearance of the FM phase. p. 2
Where the paper says it

In thermal equilibrium, the transition at ≈ 370 K is characterised by a change in the electronic band structure [29, 30], an arising FM magnetisation [31], and an isotropic expansion of the unit cell [32]. The latter enables tracking phase-specific strain responses, thereby facilitating the application of ultrafast X-ray sonography to probe the spatio-temporal appearance of local FM order.

Extended Data Insights

Extended Data Figure E4b displays the resulting average in-plane strain of the five nucleation sites including stochastic nature by averaging in a sketched dynamics for different tnuc weighted by probability. This spread translates to a transient spread of the out-of-plane lattice constant via the Poisson effect, which explains the additional broadening of the FM Bragg peak in the experiment. p. 7

Where the paper says it

The inplane expansion reduces the out-of-plane lattice constant via the Poisson effect. Thus, the total FM Bragg peaks becomes the superposition of an ensemble of Bragg peaks at different positions representing the individual FM domains nucleated at different delays and sites.

The uncertainty in determining domain size is estimated to be 10 nm based on this analysis.

Acknowledgements and Funding

The work acknowledges European XFEL for providing XFEL laser beamtime at the Materials Imaging and Dynamics instrument. The research was supported by the computational resources of the Maxwell at Deutsches Elektronen-Synchrotron (DESY), Hamburg, Germany. p. 0

Where the paper says it

Ultrafast X-ray sonography reveals the spatial heterogeneity of the laser-induced magneto-structural phase transition in FeRh Maximilian Mattern1*, Angel Rodriguez-Fernandez2 , Roman Shayduk2 , Jon Ander Arregi3 , Vojtˇech Uhl´ıˇr3,4, Ulrike Boesenberg2 , J¨org Hallmann2 , Wonhyuk Jo2 , Aliaksandr Leonau5,2, Rustam Rysov2 , James Wrigley2 , Alexey Zozulya2 , Stefan Eisebitt1,6, Anders Madsen2 , Daniel Schick1* , Jan-Etienne Pudell2* 1Max Born Institute for Nonlinear Optics and Short Pulse Spectroscopy, Berlin, 12489, Germany. 2European X-ray Free-Electron Laser Facility, Schenefeld, 22869, Germany.

Access to CEITEC Nano Research Infrastructure was supported by the Ministry of Education, Youth and Sports (MEYS) of the Czech Republic under the project CzechNanoLab. p. 0
Where the paper says it

2European X-ray Free-Electron Laser Facility, Schenefeld, 22869, Germany. 3CEITEC BUT, Brno University of Technology, Brno, 61200, Czech Republic.

The data is available from the authors upon request. p. 2
Where the paper says it

Here, t denotes the pump-probe delay, H(t) the Heaviside step function, and V ∗ FM the final FM volume fraction. The rapid modification of the electronic band structure of FeRh upon direct electron-photon interaction [36] perturbs the critical Rh-Fe hybridisation [34, 37, 38], resulting in a drastically reduced nucleation timescale of τ = 8 ps, as compared to 50 ps upon near-equilibrium heating [18, 26, 34].

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved material discovery by enabling precise spatial mapping of phase transition nucleation sites based on transient strain responses, as X-ray sonography can capture the spatio-temporal phase heterogeneity in great detail by resolving the phase-specific strain response. This allows AI to predict where and how ferromagnetic domains will nucleate in complex heterostructures.

  2. Development of predictive models for first-order phase transitions by incorporating stochastic nucleation kinetics, using Eq. (3) with the fixed nucleation timescale τ = 8 ps to simulate domain growth and identify optimal parameters like DFM(t) that describe the transition pathway.

  3. Enhanced analysis of experimental data by quantifying domain dimensions through reciprocal space peak broadening, allowing AI to estimate the lateral diameter of nucleated domains, such as to approximately 30 nm for FeRh.

  4. Creation of a versatile tool for characterizing diverse structural changes by leveraging the concept that the specific strain response of each phase enables its direct localisation along the layer’s depth and therefore provides access to spatial heterogeneities both in and out of the plane. This capability extends beyond just volume fractions to fully parametrize the three-dimensional nature of the phase heterogeneity.

  5. Improved experimental design through proposing novel probe schemes, such as double-pulse excitation schemes, where a delayed second excitation launches a dedicated strain pulse to mark phase heterogeneity at a selected pump-pump delay. This allows for selective probing of specific degrees of freedom and elements by tuning X-ray photon energy.

Abstract

Phase transitions are governed by both intrinsic and extrinsic heterogeneities, yet capturing their spatio-temporal dynamics remains a challenge. While ultrafast techniques track phase changes on femtosecond timescales, the spatial complexity and stochastic nature of the processes often remain hidden. Here, we present an experimental approach that combines well-established ultrafast hard-X-ray diffraction with a propagating strain pulse as a universal and non-invasive probe. This ultrafast X-ray sonography can capture the spatio-temporal phase heterogeneity in great detail by resolving the phase-specific strain response. We apply this approach to the antiferromagnetic-to-ferromagnetic magneto-structural phase transition in FeRh and identify the ferromagnetic phase to nucleate at the surface as narrow columnar domains of approximately 30, nm diameter. Besides reconciling the diverse experimental results in the literature on FeRh, X-ray sonography offers a versatile platform for investigating a wide range of phase transitions accompanied by structural changes.

Sources

Related papers