🔬 mesoscale physics

Nonlinear spin-Seebeck diode in $f$ -wave magnets, third-order spin-Nernst effects in $g$ -wave magnets and spin-Nernst effects in $i$ -wave altermagnets

This paper theoretically demonstrates that various unconventional magnetic orders ( $f$ -, $g$ -, and $i$ -wave) generate distinct nonlinear and transverse spin currents in response to temperature gradients without requiring spin-orbit interaction, while $p$ -wave magnets fail to produce such effects.

Original authors: Motohiko Ezawa

Published 2026-07-01

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Motohiko Ezawa

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 world where you can move tiny magnetic particles (spins) without moving any electric charge. This is the goal of a field called spintronics. Usually, to get these particles moving, scientists need to apply electricity or use a special "twist" in the material called spin-orbit interaction.

However, this paper by Motohiko Ezawa explores a different, simpler way to move these spins: just by heating one side of a material and cooling the other. Think of it like a river flowing because of a temperature difference rather than a pump.

The paper investigates different types of magnetic materials, which the author categorizes by their "shape" or symmetry (named after mathematical waves like d-wave, f-wave, g-wave, etc.). Here is what the paper discovers about these different shapes when you apply a temperature gradient (a heat difference):

1. The "Spin Diode" (f-wave magnets)

Imagine a one-way street for traffic. Usually, if you push a car forward, it moves forward; if you push it backward, it moves backward. But a diode is a device that only lets traffic flow in one direction, no matter how you push it.

The paper finds that in f-wave magnets, heat acts like a magical force that pushes spin current in a specific direction regardless of which way the heat is flowing.

The Analogy: Think of a ratchet wrench. You can wiggle the handle back and forth (fluctuating the temperature), but the bolt only turns in one direction.
The Result: The spin current is generated proportional to the square of the temperature difference. This means even if the heat fluctuates back and forth, the spins keep flowing in a single direction. The author calls this a "nonlinear spin-Seebeck diode." It's a perfect one-way valve for spin current.

2. The "Third-Order Twist" (g-wave magnets)

In g-wave magnets, the relationship between heat and spin flow is even more complex.

The Analogy: Imagine trying to push a heavy swing. A gentle push (linear) might not do much, but if you push with a specific rhythm that involves three distinct movements, the swing suddenly flies high.
The Result: Here, the spin current only appears when you look at the third power of the temperature difference. It's a very specific, high-order reaction where the heat gradient creates a spin current in a perpendicular direction.

3. The "Standard Cross-Flow" (i-wave magnets)

In i-wave magnets, the behavior is more straightforward but still special.

The Analogy: Think of a crosswind. If you blow hot air from the left, the spin current doesn't just go left; it gets pushed sideways, flowing from top to bottom.
The Result: This is the spin-Nernst effect. The heat gradient flows one way, but the spin current flows perpendicular to it (at a 90-degree angle). This happens even without the usual "twist" (spin-orbit interaction) that most materials need.

4. The "Dead End" (p-wave magnets)

Not all shapes work. In p-wave magnets, the paper finds that applying a temperature gradient does nothing.

The Analogy: It's like trying to push a car that is stuck in neutral with the parking brake on. No matter how much you heat it up, the spins refuse to move.

The Big Picture

The most exciting part of this paper is that all these effects happen without the need for "spin-orbit interaction." Usually, moving spins requires complex atomic structures that twist the electrons. These new materials (f, g, and i-wave) can do it using only their internal magnetic symmetry and temperature differences.

Summary of the "Wave" Family:

d-wave: The known star. Creates a cross-flow (spin-Nernst) with heat.
f-wave: The Diode. Creates a one-way spin current that works even if the heat fluctuates.
g-wave: The Complex Reactor. Needs a specific 3rd-order heat push to create a cross-flow.
i-wave: The Sidewinder. Creates a cross-flow with heat, but in a different order than d-wave.
p-wave: The Silent One. No spin current is generated by heat.

The author provides mathematical formulas to predict exactly how much spin current will flow in these materials, suggesting that these "one-way streets" for spin could be built into future devices using materials like specific crystals or twisted layers of magnetic atoms.

Technical Summary: Nonlinear Spin-Seebeck and Spin-Nernst Effects in Higher-Wave Magnets

Problem Statement
While linear spin current generation via the spin-Hall, spin-Seebeck, and spin-Nernst effects is well-established in systems with spin-orbit interactions (SOI), the generation of nonlinear spin currents induced by temperature gradients remains an open area of investigation. Specifically, there is a need to identify magnetic symmetries that support nonlinear spin transport—such as second-order spin-Seebeck or third-order spin-Nernst effects—without relying on SOI. This paper addresses the search for such phenomena in altermagnets and other higher-wave magnetic systems characterized by specific crystal symmetries (p-wave, d-wave, f-wave, g-wave, and i-wave).

Methodology
The author employs a semiclassical Boltzmann transport approach to derive analytic formulas for spin currents induced by a temperature gradient $\nabla T$ .

Formalism: Starting from the Boltzmann equation with a relaxation time approximation, the Fermi distribution function is expanded in powers of the temperature gradient ( $\partial_a T$ ). This allows for the calculation of current responses to any order $\ell$ .
Current Definitions: The $\ell$ -th order charge and spin currents are defined based on the expansion coefficients. The diagonal components (parallel to $\nabla T$ ) correspond to Seebeck effects, while off-diagonal components (perpendicular to $\nabla T$ ) correspond to Nernst effects.
Model Systems: The study focuses on two-dimensional systems with diagonal spin Hamiltonians ( $H = H_{\text{kin}}\sigma_0 + H_J\sigma_z$ $H = H_{kin} σ_{0} + H_{J} σ_{z}$ ), where the spin splitting is driven by exchange interactions rather than SOI. The analysis covers various wave symmetries defined by their nodal structures and mirror symmetries:
- Symmetry Analysis: The paper utilizes mirror symmetries ( $x \to -x$ and $y \to -y$ ) to determine selection rules. For instance, if a mirror symmetry exists in the energy dispersion, specific orders of nonlinear currents vanish due to integrand parity.
- High-Temperature Expansion: Analytic results are derived using a high-temperature approximation for the derivative of the Fermi distribution, which are then compared against numerical results without this approximation.

Key Contributions and Results
The paper derives explicit analytic formulas for spin currents in various wave-symmetry magnets and categorizes their nonlinear responses as summarized in Table I:

f-wave Magnets (Nonlinear Spin-Seebeck Diode):
- Finding: In f-wave magnets, which lack mirror symmetry $x \to -x$ , a second-order nonlinear spin-Seebeck current ( $j_{\text{spin}}^{(x^2;x)}$ ) is generated parallel to the square of the temperature gradient, $(\nabla T)^2$ .
- Significance: This current flows regardless of the direction of the temperature gradient (non-reciprocal), effectively functioning as a spin current diode. This is the first demonstration of a second-order spin-Seebeck effect in the absence of SOI.
- Contrast: No spin-Nernst effect is found in f-wave magnets due to remaining mirror symmetries.
g-wave Altermagnets (Third-Order Spin-Nernst):
- Finding: In g-wave altermagnets, where mirror symmetry $y \to -y$ is broken, a third-order nonlinear spin-Nernst current ( $j_{\text{spin}}^{(x^3;y)}$ ) is generated perpendicular to the temperature gradient.
- Result: The current scales with $(\nabla T)^3$ .
i-wave Altermagnets (Linear Spin-Nernst):
- Finding: In i-wave altermagnets, a linear spin-Nernst current ( $j_{\text{spin}}^{(x;y)}$ ) is generated perpendicular to the temperature gradient.
- Contrast: This contrasts with electric-field-induced responses in these systems, where linear spin currents are absent. The paper also notes the existence of higher-order nonlinear responses in these systems.
d-wave Altermagnets:
- Finding: The study revisits d-wave altermagnets, confirming the generation of a linear spin-Nernst current perpendicular to $\nabla T$ , consistent with previous literature.
p-wave Magnets:
- Finding: The analysis shows that no linear or nonlinear spin current is generated in p-wave magnets by a temperature gradient, due to the preservation of relevant mirror symmetries.

Significance and Claims
The paper claims that these phenomena are significant because they occur in the absence of spin-orbit interactions, relying solely on the crystal symmetry and exchange splitting of the altermagnetic band structure.

Diode Functionality: The identification of the second-order spin-Seebeck effect in f-wave magnets provides a mechanism for a "spin current diode," a crucial component for spintronic logic where current directionality is controlled by the magnitude rather than the polarity of the driving force (temperature gradient).
Universality of Symmetry: The results highlight that the order of the nonlinear response (second, third, etc.) is strictly dictated by the nodal structure and mirror symmetries of the specific wave-symmetry magnet.
Material Relevance: The author notes that while the theoretical framework assumes diagonal spin (neglecting SOI), the exchange interaction ( $J$ ) in proposed materials (e.g., Ba $_3$ MnNb $_2$ O $_9$ , FePO $_4$ , twisted bilayers) is typically orders of magnitude larger than the Rashba interaction. Therefore, the derived results are expected to remain applicable even in the presence of weak SOI.

The paper concludes by summarizing these findings in Table I, establishing a clear map between magnetic wave symmetry (p, d, f, g, i) and the allowed orders of spin-Seebeck and spin-Nernst effects.

1. The "Spin Diode" (f-wave magnets)

2. The "Third-Order Twist" (g-wave magnets)

3. The "Standard Cross-Flow" (i-wave magnets)

4. The "Dead End" (p-wave magnets)

The Big Picture

More like this