Ground state magnetic structure of Mn3Sn

Using spherical neutron polarimetry and density functional theory, this study identifies the ground state magnetic structure of Mn3Sn as an inverse triangular Type III configuration selected by subtle sixth-order anisotropy, while revealing that its magnetic domains are partially controllable by moderate fields at low temperatures but become decoupled and uncontrollable upon entering the incommensurate phase.

The Gist

Imagine a microscopic city built on a special kind of grid called a kagome lattice. In this city, the residents are Manganese atoms (Mn), and they are arranged in triangles. These atoms have a secret: they are magnets, but instead of all pointing in the same direction like a crowd of soldiers, they are "non-collinear." This means they arrange themselves in a triangle, each pointing 120 degrees away from its neighbor, like a peace sign made of three fingers.

The paper you shared is a detective story about figuring out exactly which way these magnetic fingers are pointing in a specific material called Mn3Sn.

Here is the story broken down into simple concepts:

1. The Mystery: Two Possible Poses

For years, scientists have been trying to figure out the "ground state" (the resting pose) of these magnetic atoms. They knew there were two main possibilities, which the paper calls Type III and Type IV.

• The Analogy: Imagine a group of three friends sitting at a round table. They all want to point their fingers at the center, but they also want to point their backs toward specific walls.
  • Type III is like them pointing their backs toward the walls that run North-South and East-West.
  • Type IV is like them pointing their backs toward the diagonal corners (North-East, South-West, etc.).

For a long time, scientists assumed Mn3Sn was doing the same thing as its cousin, Mn3Ge, which was known to be Type IV. But this paper says, "Wait a minute, we need to check for ourselves."

2. The Detective Tool: Magnetic Polarized Neutrons

To solve this, the researchers used a super-powerful microscope called Spherical Neutron Polarimetry (SNP).

• The Analogy: Imagine throwing a bunch of tiny, spinning tops (neutrons) at the crystal. These tops are "polarized," meaning they are all spinning in a specific direction (like a top spinning clockwise).
• When these tops hit the magnetic atoms in the crystal, they bounce off. By measuring exactly how the tops spin after they bounce, the scientists can deduce the exact orientation of the magnetic atoms inside. It's like figuring out the shape of a hidden object by watching how it deflects a stream of water.

3. The Big Reveal: It's Type III!

The experiment showed that Mn3Sn is actually Type III.

• The Twist: Even though Mn3Sn and Mn3Ge look almost identical and behave similarly, their magnetic "fingers" point in different directions.
• Why? The energy difference between the two poses is so tiny that standard computer models (called Density Functional Theory) couldn't tell them apart. It's like trying to decide if a coin is slightly heavier on the head or the tail when the difference is smaller than a single atom. The paper suggests that a very subtle force called "sixth-order anisotropy" (a fancy way of saying a tiny, complex preference for a specific direction) tips the scales in favor of Type III.

4. The Crowd Control Problem: Magnetic Domains

Here is where it gets tricky. The crystal isn't just one big block of magnets; it's divided into domains.

• The Analogy: Imagine a stadium full of people. In the "commensurate phase" (the high-temperature state), the people are organized into six different sections (domains).
• The researchers found that if they applied a magnetic field (like a loudspeaker shouting "Face North!"), they could get three of those six sections to align with the shout, while the other three stayed quiet.
• The Surprise: They expected only one section to align. Instead, three sections aligned equally. It's as if the crowd heard the shout and decided, "Okay, we'll all face North, but we'll split into three groups to do it."

5. The "Ghost" Phase: The Low-Temperature Trap

When the material gets very cold (below about 290 Kelvin), it enters a new phase called the Incommensurate (IC) phase.

• The Analogy: Imagine the organized stadium crowd suddenly turning into a swirling, chaotic dance. The neat rows of magnets start to rotate slightly as you go up each layer of the crystal, like a spiral staircase.
• The Problem: In this chaotic phase, the magnetic domains lose their connection to the outside world. Even if you shout "Face North!" with a massive magnet, the atoms inside don't listen. They are "decoupled."
• The Consequence: Because the researchers couldn't force the atoms to align in this cold phase, they couldn't get a clear picture of the structure using their neutron microscope. The data was too messy to solve the puzzle completely.

Why Does This Matter?

You might ask, "Who cares about which way tiny atoms point?"

• The Superpower: These materials have a "superpower" called the Anomalous Hall Effect. This allows them to generate electricity from magnetism without needing a permanent magnet. This is the holy grail for spintronics—the next generation of computers that are faster, smaller, and use less energy.
• The Goal: To build these devices, engineers need to control the magnetic domains (the "crowd"). If they can switch the direction of the magnetic fingers with a tiny electric current, they can create super-fast memory switches.
• The Takeaway: This paper tells us that Mn3Sn is a Type III magnet, not Type IV. Knowing this is the first step to learning how to control it. However, the fact that the cold phase is "uncontrollable" with magnets is a hurdle. The authors suggest we might need to use electricity or pulsed currents (like a quick zap) to control the atoms in that cold state, rather than just using magnets.

In summary: The paper solved a decades-old puzzle about the resting pose of Mn3Sn atoms, proving they are different from their cousins. It also highlighted a challenge: while we can control them when they are warm, they become "stubborn" and unresponsive when they get cold, requiring new tricks to harness their potential for future technology.

Technical Summary

1. Problem Statement

Mn$_3$Sn is a non-collinear antiferromagnet with a kagome lattice structure that exhibits a large anomalous Hall effect (AHE) at room temperature, making it a prime candidate for antiferromagnetic spintronics. The AHE arises from Berry curvature generated by Weyl points, which require broken time-reversal symmetry.

• The Core Issue: While the magnetic structure of the isostructural compound Mn$_3$Ge was previously determined to be "Type IV" using spherical neutron polarimetry (SNP), the ground state magnetic structure of Mn$_3$Sn remained ambiguous.
• The Ambiguity: Previous neutron diffraction studies narrowed the possibilities down to two similar inverse triangular structures: Type III (spins parallel to $\langle 100 \rangle$) and Type IV (spins parallel to $\langle 110 \rangle$). Both structures allow for six magnetic domains, and the population of these domains significantly influences the material's macroscopic properties (like the AHE). Without controlling the domain population, distinguishing between Type III and IV is difficult.
• The Gap: No study had definitively determined the zero-field ground state of Mn$_3$Sn, nor had the behavior of magnetic domains in the low-temperature incommensurate (IC) phase been fully characterized.

2. Methodology

The authors employed a multi-faceted approach combining experimental neutron scattering and theoretical calculations:

• Sample Synthesis: High-quality single crystals of Mn$_3$Sn were synthesized using a Bi-flux method.
• Spherical Neutron Polarimetry (SNP):
  • Performed at the Institut Laue-Langevin (ILL) using the D3 instrument and Cryopad.
  • Measured the full $3 \times 3$ polarization matrix ($P_{fi}$) for spin-up and spin-down neutrons in two crystal orientations: $(h, k, 0)$ and $(h, 0, l)$.
  • Crucial Step: To distinguish between models, the sample was subjected to a moderate magnetic field (1 T) to partially align the magnetic domains before measurement. This was essential because the domain population affects the diffraction signal.
• Unpolarized Neutron Diffraction:
  • Performed on the D23 instrument (ILL) in high magnetic fields (up to 15 T).
  • Used to investigate the stability of magnetic domains upon cooling from the commensurate phase to the incommensurate (IC) phase.
• Density Functional Theory (DFT):
  • Calculations performed using GPAW and the Atomic Simulation Environment (ASE).
  • Used Local Spin-Density Approximation (LSDA) to calculate the total energy of the four possible magnetic structure models (Type I, II, III, IV) for both Mn$_3$Sn and Mn$_3$Ge.

3. Key Contributions and Results

A. Determination of the Ground State (Type III)

• SNP Analysis: The experimental polarization matrices were fitted against the four theoretical models.
  • Result: Only Type III and Type IV provided reasonable fits.
  • Differentiation: In the $(h, 0, l)$ scattering plane, Type III provided a significantly better fit ($\chi^2_R = 37.14$) compared to Type IV ($\chi^2_R = 73.03$).
• Conclusion: The ground state magnetic structure of Mn$_3$Sn is Type III (Space group $Cmc'm'$), where spins are parallel to the $\langle 100 \rangle$ direction. This contrasts with Mn$_3$Ge, which adopts Type IV.

B. Domain Population and Control

• High-Temperature Phase (Commensurate, >290 K):
  • Applying a magnetic field along the $b$-axis did not select a single domain as expected. Instead, three of the six domains were populated with roughly equal probability (~33% each), while the other three were negligible.
  • This selection is driven by the Zeeman energy coupling the small net ferromagnetic moment of the domains to the external field.
• Low-Temperature Phase (Incommensurate, <290 K):
  • Upon cooling below $T_S \approx 290$ K, the system enters an incommensurate (IC) phase with propagation vectors $k_L \sim (0,0,0.08)$ and $k_T \sim (0,0,0.11)$.
  • Critical Finding: The magnetic domain structure is completely lost upon entering the IC phase. The domains become decoupled from the external magnetic field.
  • Experiments with fields up to 15 T showed no change in peak positions or intensities in the IC phase, indicating that no known method can currently control the domain distribution in this phase.

C. Theoretical Insights (DFT)

• Energy Degeneracy: DFT calculations revealed that the energy difference between Type III and Type IV is less than 1 $\mu$eV, effectively making them degenerate within the accuracy of standard DFT.
• Selection Mechanism: Since standard DFT cannot distinguish them, the selection of Type III in Mn$_3$Sn (vs. Type IV in Mn$_3$Ge) is governed by sixth-order magnetic anisotropy.
• Anisotropy Sign: The results imply that the sixth-order anisotropy constant ($K_6$) has the opposite sign in Mn$_3$Sn compared to Mn$_3$Ge. In Mn$_3$Ge, the easy axis is along the reciprocal lattice vectors ($a^*$), whereas in Mn$_3$Sn, it is along the direct lattice vectors ($a$).

4. Significance

• Resolution of Structural Ambiguity: The paper definitively resolves the long-standing debate regarding the magnetic ground state of Mn$_3$Sn, correcting the assumption that it shares the same structure as Mn$_3$Ge.
• Spintronic Implications: Understanding the specific magnetic structure (Type III) and the behavior of its domains is critical for engineering the Anomalous Hall Effect (AHE) in spintronic devices. The ability to switch the AHE sign via field or strain relies on manipulating these specific domains.
• Limitations in the IC Phase: The discovery that the IC phase is decoupled from magnetic fields presents a significant challenge for applications requiring domain control at low temperatures. It suggests that alternative control mechanisms (e.g., electric fields or pulsed currents) must be explored for low-temperature operation.
• Fundamental Physics: The study highlights the subtle role of high-order anisotropy terms in selecting magnetic ground states in kagome antiferromagnets, a mechanism too small to be captured by standard DFT but crucial for material properties.

5. Summary of Findings Table

Feature	Mn$_3$Sn (This Work)	Mn$_3$Ge (Previous Work)
Ground State Structure	Type III ($Cmc'm'$)	Type IV ($Cm'cm'$)
Spin Direction	Parallel to $\langle 100 \rangle$	Parallel to $\langle 110 \rangle$
Easy Axis	Direct lattice vector ($a$)	Reciprocal lattice vector ($a^*$)
DFT Energy Diff	$< 1 \mu$eV (Degenerate)	$< 1 \mu$eV (Degenerate)
Domain Control (High T)	3 domains selected by field	2 domains selected by field
Domain Control (Low T)	Lost (Decoupled from field)	Not explicitly detailed in this context

In conclusion, this work establishes that Mn$_3$Sn possesses a distinct magnetic ground state (Type III) driven by subtle sixth-order anisotropy, and highlights a critical limitation where magnetic domain control is lost in the low-temperature incommensurate phase.