🔬 mesoscale physics

Multiscale Vectorial Determination of Magnetic Order Parameters using Electron Magnetic Linear Dichroism

This paper demonstrates that electron magnetic linear dichroism, when combined with advanced dynamical diffraction simulations, enables the quantitative, nanometer-resolution reconstruction and real-space mapping of vectorial magnetic order parameters in both ferromagnetic and antiferromagnetic materials like FeRh.

Original authors: Jan Hajduček, Jáchym Štindl, Ján Rusz, Vojtěch Uhlíř

Published 2026-05-28

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Jan Hajduček, Jáchym Štindl, Ján Rusz, Vojtěch Uhlíř

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 figure out the direction a tiny, invisible arrow is pointing inside a piece of metal. This arrow represents the magnetic "spin" of atoms. In some materials, like the iron-rhodium alloy (FeRh) studied in this paper, these arrows are arranged in two different ways:

Ferromagnetic (FM): All arrows point in the same direction (like a crowd marching in unison).
Antiferromagnetic (AF): Neighboring arrows point in opposite directions (like a checkerboard of red and blue arrows). In this state, the arrows cancel each other out, leaving no net magnetic field. This makes them incredibly hard to "see" with standard tools, which usually only detect the overall crowd direction.

The researchers in this paper have developed a new, high-resolution way to map these arrows using a Transmission Electron Microscope (TEM). They call their method Electron Magnetic Linear Dichroism (EMLD).

Here is a simple breakdown of how it works, using everyday analogies:

1. The Problem: The "Invisible" Magnet

Think of the Antiferromagnetic state as a room full of people holding flashlights. Half are pointing North, and half are pointing South. If you stand outside the room and look in, the light cancels out, and it looks pitch black. Traditional tools can't tell you which way the individual people are pointing because the net result is zero.

2. The Tool: The "Flashlight" Electron Beam

Instead of a camera, the scientists use a beam of electrons (tiny particles) fired through the material. As these electrons pass through the crystal, they bump into the atoms and lose a tiny bit of energy. This is like throwing a ball through a forest; the way the ball bounces off the trees tells you about the trees' arrangement.

The key innovation here is that the electrons don't just bounce randomly. They interact with the magnetic "arrows" inside the atoms. The researchers realized that by carefully measuring how the electrons lose energy and where they scatter, they can detect the orientation of those hidden arrows.

3. The Trick: "Linear Dichroism" (The Polarized Sunglasses Effect)

The paper distinguishes between two types of signals:

Circular Dichroism (EMCD): This is like looking at a spinning top. It tells you if something is spinning clockwise or counter-clockwise. This works well for the "marching crowd" (Ferromagnetic) but is very picky about the angle you look at it.
Linear Dichroism (EMLD): This is the star of the show. Imagine wearing polarized sunglasses. If you rotate your head, the view changes depending on how the light is oriented. Similarly, EMLD measures how the electrons interact with the atoms based on the direction of the magnetic arrow relative to the electron beam.

The researchers found that even when the arrows cancel each other out (the Antiferromagnetic state), the shape of the interaction changes depending on the arrow's direction. It's like knowing which way a person is facing in a dark room by the specific shadow they cast on the wall, even if you can't see the person.

4. The Simulation: The "Digital Twin"

To make sense of the messy data coming from the microscope, the team built a powerful computer simulation. Think of this as a "Digital Twin" of the experiment.

They programmed the computer to know exactly how electrons should behave if the magnetic arrows pointed North, South, East, or West.
They included a specific "twist" in the math (called exchange splitting) that accounts for the tiny energy differences caused by magnetism.
By comparing the real experimental data with this digital twin, they can reverse-engineer the exact direction of the magnetic arrows in 3D space.

5. The Result: A 3D Map of the Invisible

The paper demonstrates that this method works on FeRh, a material that can switch between the "canceling out" state (Antiferromagnetic) and the "marching" state (Ferromagnetic) just by changing the temperature.

In the Ferromagnetic phase: They successfully mapped the direction of the magnetic arrows.
In the Antiferromagnetic phase: They successfully mapped the "Néel vector" (the direction of the opposing arrows), which was previously very difficult to do with this level of detail.

Why is this a big deal?

The authors claim this is a "multiscale" solution. It works whether you are looking at a large chunk of material or zooming in to the size of a single atom.

Robustness: Unlike previous methods that required perfect, needle-sharp conditions to work, this method is sturdy. It works even if the electron beam is slightly tilted or if the sample is a bit thick.
Separation: They figured out how to mathematically separate the "magnetic" signal from the "structural" signal (the shape of the atoms), ensuring they are actually seeing magnetism and not just the crystal shape.

In summary: The paper presents a new "magnetic compass" for electron microscopes. It allows scientists to see the direction of magnetic arrows inside materials that were previously invisible, even when those arrows are canceling each other out. This is done by firing electrons through the material, measuring the specific energy they lose, and using a sophisticated computer model to translate that data into a 3D map of magnetic order.

Technical Summary: Multiscale Vectorial Determination of Magnetic Order Parameters using Electron Magnetic Linear Dichroism

Problem Statement
Nanoscale reconstruction of magnetic order remains a significant challenge in spintronics and quantum materials, particularly for compensated magnetic phases such as antiferromagnets (AF) and the recently described altermagnets. In these systems, the antiparallel alignment of neighboring atomic magnetic moments results in a vanishing net magnetization, rendering them inaccessible to conventional magnetization-sensitive probes. While X-ray magnetic dichroism (specifically XMLD) has become a standard for element-specific magnetic imaging and Néel vector orientation, its application is limited by the spatial resolution of synchrotron-based techniques. Conversely, transmission electron microscopy (TEM) offers atomic-scale resolution but has historically struggled with the vectorial sensitivity required to fully reconstruct magnetic order parameters in compensated systems. Although Electron Magnetic Circular Dichroism (EMCD) has demonstrated sensitivity to AF order, its vectorial sensitivity remains partially explored, and full vector reconstruction has largely been conceptual. Furthermore, the magnetic extension of electron linear dichroism (EMLD), sensitive to anisotropic transitions, has been limited by detector sensitivity and, crucially, the lack of a predictive theoretical framework capable of handling arbitrary magnetic axis orientations.

Methodology
The authors present a unified vectorial simulation framework that integrates explicit vectorial core-level exchange splitting into mixed dynamic form factor (MDFF) simulations, accounting for dynamical diffraction. This approach enables the direct reconstruction of magnetic spin axes from momentum-resolved electron energy-loss spectra (EELS).

Key methodological components include:

Vectorial Simulation Framework: The method accepts first-principles transition matrix elements from Density Functional Theory (DFT) and employs Euler transformations to rotate spin and orbital reference frames. This allows for the construction of MDFFs combined with dynamical diffraction for arbitrary spin axis orientations.
Exchange Splitting Inclusion: The framework explicitly includes core-level exchange splitting ( $\Delta E_{j,j_z} = \lambda_j j_z$ ) to model the anisotropic transition probabilities that give rise to EMLD. This distinguishes EMLD from natural linear dichroism (ENLD), which arises from structural anisotropy.
Signal Separation: The study demonstrates that the dichroic signal can be decomposed into circular (EMCD), magnetic linear (EMLD), and natural linear (ENLD) components. Three separation strategies are proposed and validated:
1. Opposite-Magnetization (OM): Comparing spectra from reversed magnetization states to isolate EMLD (sum) and EMCD (difference).
2. Exchange-Splitting (ES): Comparing simulations with and without exchange splitting to isolate EMLD.
3. Spectral-Difference-Profile (SDP): Exploiting distinct line shapes of EMCD and EMLD contributions.
Reconstruction Scheme: A procedure is established where EMLD spectra acquired at selected aperture positions in the diffraction plane progressively constrain the possible Néel-vector directions. By utilizing both out-of-plane (OOP) and in-plane (IP) sensitive aperture geometries, the full vectorial orientation ( $\theta_{\vec{N}}, \phi_{\vec{N}}$ ) can be determined.

Key Contributions

Theoretical Framework: The paper establishes a general route to vectorial reconstruction of magnetic order parameters by unifying the treatment of EMCD and EMLD within a single simulation framework that handles arbitrary magnetic moment orientations.
EMLD as a Robust Probe: The authors demonstrate that EMLD is intrinsically separable from non-magnetic anisotropy (ENLD) and exhibits a well-defined dependence on the Néel vector or magnetization orientation.
Scalability: The approach is shown to be robust across a wide range of electron probe sizes, from micrometer-sized parallel beams down to atomically sized probes with large convergence angles, and remains effective across various sample thicknesses and crystal orientations.
FeRh Case Study: The methodology is applied to the prototypical metamagnetic alloy FeRh, which undergoes a first-order phase transition from a low-temperature AF phase to a high-temperature ferromagnetic (FM) phase. The study successfully simulates the reconstruction of the magnetic vector in both phases.

Results

Vectorial Sensitivity: Simulations confirm that EMLD signals persist in both AF and FM phases of FeRh, whereas EMCD is detectable only in the FM phase (or AF sublattices with atomically confined probes). The EMLD signal exhibits a $\sin^2$ -like harmonic dependence on the spin-axis orientation, analogous to XMLD, allowing for unambiguous determination of the spin axis.
Separation of Signals: The study validates that EMLD and ENLD can be experimentally separated by exploiting their distinct physical origins (e.g., by heating above the Néel temperature to suppress EMLD, or by rotating the magnetic axis while keeping geometry fixed).
Robustness: Unlike EMCD, which requires specific two- or three-beam diffraction conditions and is highly sensitive to beam orientation and sample thickness, the EMLD signal remains present across an extended angular phase space of incident directions and retains its sign and spectral shape over a wide range of sample thicknesses.
Convergent Beam Performance: The EMLD signal remains detectable even with highly convergent electron probes (up to 12 mrad), preserving sensitivity to AF order, whereas EMCD in AF systems becomes appreciable only with atomically confined probes.

Significance
The paper claims that these results open a pathway to nanoscale spectroscopy and imaging of antiferromagnets and altermagnets. By providing a direct access to different magnetic order parameters through a generalized approach to electron dichroism, the work enables the quantitative real-space mapping of magnetic vectors in compensated systems. The demonstrated vectorial sensitivity, combined with robustness against probe size and alignment, places the experimental realization within the reach of modern aberration-corrected microscopes. This offers a pathway to map local magnetic anisotropy, interfacial and two-dimensional magnetism, and symmetry breaking induced by strain, thereby providing access to emergent phenomena in antiferromagnets and altermagnets at the nanoscale. The work bridges the gap between X-ray dichroism methodologies and electron microscopy, allowing for the transfer of established XMLD protocols to the TEM for quantitative evaluation of magnetocrystalline anisotropy.