🔬 materials science

Temperature dependence of spin-model parameters in ferrimagnets

Using Green's-function theory and atomistic spin-dynamics simulations, this study demonstrates that while exchange interactions in two-sublattice ferrimagnets decrease monotonically with temperature, their unequal thermal renormalization leads to a strongly nonmonotonic, polarization-dependent spin-wave stiffness, particularly near the angular momentum compensation region.

Original authors: Marta Yanguas, José M. Lendínez, Theodor Griepe, Rubén M. Otxoa, Levente Rózsa, Unai Atxitia

Published 2026-10-07

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Marta Yanguas, José M. Lendínez, Theodor Griepe, Rubén M. Otxoa, Levente Rózsa, Unai Atxitia

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

Magnetic materials are the silent engines behind much of modern technology, from the hard drives that store our memories to the sensors that guide our smartphones. For decades, engineers have relied on a specific class of magnets called ferromagnets, where tiny atomic magnets align in the same direction to create a strong, unified force. However, a different kind of magnet, known as a ferrimagnet, is gaining attention for its potential to power the next generation of ultra-fast electronics. In these materials, the atomic magnets are arranged in two opposing groups that do not cancel each other out completely, leaving a net magnetic force. What makes ferrimagnets particularly special is that their internal balance can be shifted simply by changing the temperature. At certain critical heat levels, the opposing groups can momentarily cancel each other's momentum, creating a unique state where the material behaves in ways that are impossible for standard magnets. Understanding how these materials react to heat is crucial for building devices that are faster and more energy-efficient than anything currently available.

A team of researchers has now mapped out exactly how the internal rules governing these ferrimagnets change as the temperature rises. Using a combination of advanced mathematical theory and powerful computer simulations, they investigated a model of a ferrimagnet made of two distinct layers of atoms. They focused on the invisible forces that hold these atomic magnets in place and how those forces weaken or shift as the material gets hotter. The study reveals that while the basic magnetic forces do indeed get weaker as the temperature increases, the story is far more complex than a simple, uniform fading. The two opposing groups of atoms do not react to heat in the same way; one group holds onto its magnetic order much longer than the other. This unequal response creates a surprising and counterintuitive effect: even though the underlying forces are steadily weakening, the way magnetic waves travel through the material can actually become stronger and faster for a specific type of wave as the temperature approaches a critical point.

To understand this, imagine the magnetic material as a vast grid of tiny spinning tops. In a ferrimagnet, these tops are arranged in two groups spinning in opposite directions. The researchers used a method called Green's function theory, which acts like a sophisticated calculator for predicting how these spins interact with their neighbors and with thermal energy. They compared their calculations against massive computer simulations that tracked the movement of billions of individual spins over time. The results showed a precise match between the two methods, giving the researchers high confidence in their findings. They discovered that the forces keeping the spins aligned, known as exchange interactions, and the forces that keep them pointing in a specific direction, called anisotropy, do not all fade at the same rate. The forces within one group of spins weaken much faster than those in the other group. This difference is driven by the fact that the two groups have different magnetic strengths and respond differently to the jostling of heat.

The most striking discovery concerns how these uneven changes affect the movement of magnetic waves, which are ripples of spin that carry information through the material. In most magnetic systems, heating the material simply makes these waves slower and less energetic, a process known as softening. However, in this ferrimagnet, the researchers found that one specific type of wave behaves differently. As the temperature rises toward a point where the total momentum of the two groups cancels out, the speed and stiffness of this particular wave actually increase. It becomes more rigid and travels faster, despite the fact that the forces holding the material together are getting weaker overall. This happens because the changing balance between the two groups of spins creates a temporary environment that favors the propagation of this specific wave. The other type of wave, by contrast, continues to slow down as expected.

This finding challenges the idea that we can predict how a magnetic material will behave at high temperatures just by looking at how its basic forces weaken. The researchers demonstrated that the collective behavior of the material emerges from a complex interplay between the weakening forces and the shifting balance of the two spin groups. Their work provides a clear roadmap for creating better models of these materials. Instead of assuming that magnetic properties fade uniformly, engineers can now use these new, temperature-dependent rules to design devices that take advantage of the unique, non-linear behavior of ferrimagnets. This is particularly important for devices that need to operate at high speeds or switch states rapidly, as the researchers showed that the material's ability to carry magnetic information can be enhanced rather than degraded in specific temperature ranges. By connecting the microscopic behavior of individual atoms to the large-scale movement of magnetic waves, this study offers a vital tool for the future of spintronic technology, allowing scientists to harness the unique thermal properties of ferrimagnets for practical applications.

Technical Summary: Temperature Dependence of Spin-Model Parameters in Ferrimagnets

Problem Statement
Ferrimagnets, characterized by antiferromagnetically coupled sublattices with unequal magnetic moments, offer unique dynamical properties for spintronic applications, such as ultrafast switching and high-speed domain-wall motion near compensation temperatures. While these materials are often described by multi-sublattice atomic spin Hamiltonians (e.g., Heisenberg models), realistic modeling of spintronic phenomena frequently requires coarse-grained approaches (e.g., micromagnetic models based on the Landau–Lifshitz–Bloch equation) to access computationally prohibitive length and time scales. A critical gap exists in understanding the temperature dependence of the effective parameters (exchange interactions and anisotropies) required for these coarse-grained models in ferrimagnets. Unlike ferromagnets and antiferromagnets, where thermal scaling is better established, ferrimagnets remain largely unexplored. Specifically, it is unclear how the inequivalent thermal evolution of the two sublattices and their distinct correlations affect the renormalization of effective spin-model parameters and the resulting long-wavelength spin-wave dynamics.

Methodology
The authors employ Green's-function (GF) theory to systematically investigate the thermal renormalization of effective spin-model parameters in two-sublattice ferrimagnets on a rock-salt-type lattice. The study utilizes a classical spin Hamiltonian including isotropic Heisenberg exchange interactions (intra- and inter-sublattice) and single-ion uniaxial anisotropy.

To validate the GF approach and provide a numerical benchmark, the authors perform large-scale atomistic spin-dynamics (ASD) simulations using the stochastic Landau–Lifshitz–Gilbert (sLLG) equation. The ASD simulations incorporate thermal fluctuations via Gaussian white noise and utilize spatial Fourier transforms of spin configurations to extract static transverse spin-spin correlation functions. These correlation functions are directly related to the spin-wave (SW) frequencies, allowing for the extraction of effective temperature-dependent parameters ( $J^{rs}(T)$ and $K_r(T)$ ) by fitting the ASD spectra to the finite-temperature dispersion relations derived from the GF theory.

The study focuses on a representative system characteristic of rare-earth transition-metal ferrimagnets (e.g., GdFe alloys), examining the temperature evolution from low temperatures up to the critical ordering temperature. Additionally, the authors analyze low-temperature scaling exponents to quantify the role of transverse spin correlations by varying the partitioning between intra- and inter-sublattice exchange strengths.

Key Contributions and Results

Quantitative Agreement between GF and ASD:
The GF theory demonstrates quantitative agreement with ASD simulations across a broad temperature range for sublattice magnetizations, angular momenta, and the full spin-wave spectrum. The critical temperatures estimated by both methods differ by approximately 3%, a deviation attributed to the decoupling approximation in GF theory and finite-size effects in ASD.
Distinct Thermal Renormalization of Parameters:
The effective intra- and inter-sublattice exchange interactions ( $J^{AA}, J^{AB}, J^{BB}$ ) and on-site anisotropies ( $K_A, K_B$ ) exhibit distinct thermal renormalizations. While all effective parameters decrease monotonically with increasing temperature, they do not follow simple uncorrelated scaling laws (e.g., $J \propto m^2$ ). Instead, transverse spin correlations cause the effective exchange interactions to decrease more slowly than predicted by mean-field approximations. The anisotropies show a stronger relative thermal reduction, particularly for the sublattice with weaker intra-sublattice coupling, and approximately follow single-ion Callen–Callen scaling ( $K \propto m^3$ ) at low temperatures.
Nonmonotonic Long-Wavelength Spin-Wave Dynamics:
A central finding is that the unequal thermal renormalization of parameters leads to strongly nonmonotonic behavior in the long-wavelength spin-wave response, despite the monotonic reduction of the underlying interactions.
- Zone-Center Gaps: The magnitude of the negative-frequency zone-center gap ( $|\omega^-_0|$ ) exhibits a pronounced nonmonotonic behavior, increasing in an intermediate temperature range before decreasing near the critical temperature.
- Spin-Wave Stiffness: The branch-resolved spin-wave curvature coefficients (related to exchange stiffness) show divergent behaviors. While the positive branch stiffness decreases monotonically, the negative branch stiffness ( $|A^-|$ ) increases substantially with temperature, reaching a maximum near the angular momentum compensation temperature ( $T_A$ ). This indicates an enhanced spin-wave stiffness for one circularly polarized mode in the vicinity of $T_A$ , driven by the interplay between exchange, anisotropy, and the temperature-dependent sublattice magnetizations.
Correlation Corrections and Scaling Exponents:
The study quantifies the influence of transverse spin correlations through low-temperature scaling exponents ( $\epsilon$ ). The results show that these exponents are not universal constants determined solely by lattice geometry or exchange type. In ferrimagnets, the scaling exponents depend on the distribution of exchange interactions among intra- and inter-sublattice channels and the resulting correlations between the inequivalent magnetic subsystems. Breaking the equivalence between sublattices splits the correlation corrections for intra-sublattice exchanges, a feature that vanishes only when the sublattices become dynamically decoupled.

Significance
The paper establishes a direct connection between temperature-dependent atomistic interactions and the effective parameters required for finite-temperature micromagnetic and multiscale modeling of ferrimagnetic materials. The authors argue that the effective long-wavelength dynamics cannot be inferred simply from the monotonic temperature dependence of microscopic exchange interactions. Instead, the collective dynamics emerge from the complex interplay of renormalized exchange, anisotropy, and sublattice magnetizations.

The GF framework presented offers a computationally efficient alternative to exhaustive ASD simulations for determining temperature-dependent micromagnetic parameters. This is particularly significant near compensation temperatures, where the strongly polarization- and temperature-dependent spin-wave stiffness can significantly influence the propagation and dynamics of magnetic textures. The work provides a consistent theoretical foundation for constructing quantitative finite-temperature descriptions of ferrimagnetic spin dynamics, essential for the development of next-generation spintronic devices.

More like this