Thermal conductivity and tunable thermal anisotropy of magnetic CrSBr monolayer

First-principles calculations reveal that monolayer CrSBr exhibits significant thermal anisotropy (with a $\kappa_{xx}/\kappa_{yy}$ ratio of approximately 1.8) driven by phonon velocities and lifetimes, which can be further tuned by controlling flake size to suppress long mean free path phonons.

The Gist

Imagine a microscopic world made of a single, ultra-thin sheet of atoms. This sheet is a material called CrSBr (Chromium Sulfide Bromide). It's special because it's magnetic (like a tiny fridge magnet) and it's only one atom thick in some places, making it a "2D material."

Scientists wanted to understand how heat moves through this tiny sheet. Think of heat not as a warm feeling, but as a crowd of invisible runners (called phonons) trying to sprint across a track. The faster they run and the longer they can keep running without tripping, the better the material conducts heat.

Here is the story of what the researchers found, explained simply:

1. The "One-Way Street" Effect (Anisotropy)

The most surprising discovery is that this material is a one-way street for heat.

• If you try to send heat from Left to Right (the x-axis), the runners sprint very fast and don't trip much.
• If you try to send heat from Top to Bottom (the y-axis), the runners are slower and trip over each other more often.

The result? Heat flows 1.8 times faster in one direction than the other. It's like having a super-highway in one direction and a bumpy dirt path in the other. This happens because the atoms are arranged in a way that naturally guides the "runners" better in one direction.

2. The "Traffic Jam" Problem (Why previous studies disagreed)

Before this study, other scientists had tried to calculate this heat flow, but they got very different answers. Some said the heat flow was very low; others said it was huge.

The researchers in this paper realized the previous studies were using a bad map. They were using a simplified rule that assumed every runner acts alone. But in this material, the runners actually help each other. Sometimes, when two runners bump into each other, they don't stop; they just swap places and keep moving forward together. This is called a "Normal" scattering event.

The old maps ignored this teamwork. The new study used a super-complex simulation (like a high-tech traffic control system) that accounts for how the runners interact. This gave them the correct, much higher numbers for how well the material conducts heat.

3. The "Size Matters" Trick (Tuning the Anisotropy)

Here is the coolest part: You can change the material's behavior just by cutting it to a different size.

Imagine the runners are trying to cross a park.

• If the park is huge (Infinite size): The runners have plenty of room. The "Left-to-Right" runners are so much better at avoiding obstacles than the "Top-to-Bottom" runners that the difference is huge. The material is very "directional."
• If the park is tiny (Nanometer size): Now, the runners hit the walls of the park very quickly. It doesn't matter if they are good at avoiding each other; they all get stuck at the wall. The "Left-to-Right" advantage disappears because the walls block everyone equally.

The scientists found that by making the flake smaller, they could dial down the difference between the two directions. If you make the flake very small, the heat flows almost the same in both directions. If you make it big, the strong directional difference returns.

4. The "Magnet" That Wouldn't Change

The researchers also tried to squeeze and stretch the material (like stretching a rubber band) to see if they could flip its magnetic switch from "North-South" to "South-North." They wanted to see if they could force the material to change its magnetic personality.

Result: No luck. The material was too stubborn. No matter how much they squeezed or stretched it, it stayed magnetic in its original "North-South" way. This tells us the material is very stable and robust, which is good news for building future gadgets.

Why Should We Care?

This isn't just about math. Understanding how heat moves in these tiny, magnetic sheets is crucial for building the electronics of the future.

• Better Computers: As computers get smaller, they get hotter. We need materials that can move heat away efficiently.
• Smart Control: Because we can change how heat flows just by changing the size of the material, engineers could design tiny chips that manage heat differently depending on where they are on the chip.

In a nutshell: This paper tells us that a single layer of magnetic atoms acts like a heat highway that is much faster in one direction than the other. We can fix the "traffic jams" in our calculations to get the right numbers, and we can even tune how directional this heat flow is simply by cutting the material into smaller or larger pieces.

Technical Summary

Here is a detailed technical summary of the paper "Thermal conductivity and tunable thermal anisotropy of magnetic CrSBr monolayer."

1. Problem Statement

The paper addresses the significant discrepancies in reported values for the lattice thermal conductivity ($\kappa$) of monolayer (ML) CrSBr, a van der Waals magnetic 2D material. Previous theoretical studies yielded vastly different results:

• Xuan et al. reported ultralow values ($\kappa_{xx} \approx 0.8$ W m$^{-1}$K$^{-1}$).
• Liu et al. reported moderate values ($\kappa_{xx} \approx 25.5$ W m$^{-1}$K$^{-1}$).
• Han et al. reported high values ($\kappa_{xx} \approx 80.8$ W m$^{-1}$K$^{-1}$).

These studies used similar Density Functional Theory (DFT) codes but differed in the solution of the Boltzmann Transport Equation (BTE) and the inclusion of higher-order anharmonic scattering (e.g., 4-phonon processes). Furthermore, the physical origin of the strong in-plane thermal anisotropy ($\kappa_{xx} \neq \kappa_{yy}$) and the potential to tune this anisotropy via external stimuli (strain or size) remained unclear.

2. Methodology

The authors employed first-principles calculations to resolve these inconsistencies and analyze transport properties:

• Electronic Structure: Calculations were performed using VASP with the PBE-GGA functional, DFT-D3 van der Waals corrections, and DFT+U ($U_{eff} = 4.0$ eV) to treat localized Cr $d$-electrons.
• Force Constants: Second- and third-order interatomic force constants (IFC2 and IFC3) were computed using finite differences (Phonopy and Thirdorder.py) on a $6 \times 5 \times 1$ supercell.
  • Crucial Step: Rotational sum rules were enforced using hiPhive to ensure correct quadratic dispersion for the ZA (flexural) acoustic modes, which are prone to numerical instability in 2D materials.
  • Non-analytical corrections were omitted to correctly reproduce the suppression of LO-TO splitting in 2D systems.
• Thermal Transport: The almaBTE suite was used to solve the linearized BTE iteratively, going beyond the Relaxation Time Approximation (RTA). This allows for the inclusion of Normal (N) scattering processes, which are critical in 2D materials.
• Comparative Analysis: The study compared full BTE solutions against RTA approximations and investigated the effects of:
  • Temperature (100–150 K, below the Curie temperature).
  • Strain (uniaxial and biaxial up to $\pm 5\%$).
  • Flake size ($L = \infty$, $10$$\mu$m, and$100$ nm) to simulate boundary scattering.

3. Key Contributions

A. Resolution of Thermal Conductivity Discrepancies

The authors demonstrate that the large variability in previous literature stems largely from the use of the RTA approximation.

• RTA Failure: The RTA underestimates thermal conductivity by $\approx 25\%$ for $\kappa_{xx}$ and significantly misrepresents the anisotropy ratio. It incorrectly treats Normal scattering events as resistive.
• Full BTE Results: By solving the full BTE, the authors obtained $\kappa_{xx} \approx 86.3$ W m$^{-1}$K$^{-1}$ and $\kappa_{yy} \approx 43.1$ W m$^{-1}$K$^{-1}$ at 150 K. These values align closely with Han et al. (who included 4-phonon scattering), suggesting that 4-phonon processes are not the primary source of scattering in CrSBr, but rather the accurate treatment of 3-phonon Normal processes is.

B. Origin of Thermal Anisotropy

The study identifies that the strong anisotropy ($\kappa_{xx}/\kappa_{yy} \approx 1.8$) arises from a combined effect of:

1. Phonon Velocities: Group velocities along the $x$-axis ($v_x$) are higher than along the $y$-axis ($v_y$) at low frequencies ($\omega \lesssim 4$ THz).
2. Phonon Lifetimes: Phonons propagating along the $x$-axis have significantly longer lifetimes ($\tau_x \gg \tau_y$) at low frequencies due to reduced scattering rates.

C. Tunability via Size Modulation

A major finding is that thermal anisotropy is not an intrinsic constant but can be tuned by controlling the flake size.

• Mechanism: Boundary scattering suppresses long Mean Free Path (MFP) phonons. Since $\kappa_{xx}$ is dominated by phonons with longer MFPs than $\kappa_{yy}$, reducing the flake size suppresses $\kappa_{xx}$ more aggressively.
• Result: As the flake size decreases from infinite to 100 nm, the anisotropy ratio drops from $\sim 1.8$ to $\sim 1.6$. At very small scales ($L \sim 10^{-8}$ m), the transport becomes less anisotropic, approaching the RTA limit where Normal processes are suppressed.

D. Magnetic Stability under Strain

The authors investigated whether mechanical strain could induce a phase transition from the Ferromagnetic (FM) ground state to an Antiferromagnetic (AFM) state.

• Finding: The FM state remains the ground state across all uniaxial and biaxial strains ($\pm 5\%$). The energy difference ($\Delta E_{FM-AFM}$) remains negative, indicating that mechanical deformation alone cannot switch the magnetic order in monolayer CrSBr within this range.

4. Key Results

Parameter	Value / Observation
Thermal Conductivity (150 K)	$\kappa_{xx} \approx 86.3$ W m$^{-1}$K$^{-1}$; $\kappa_{yy} \approx 43.1$ W m$^{-1}$K$^{-1}$
Anisotropy Ratio ($\kappa_{xx}/\kappa_{yy}$)	1.8 (Full BTE) vs. 1.5 (RTA)
RTA Error	Underestimates $\kappa_{xx}$ by $\sim 25\%$; fails to capture correct anisotropy.
Magnetic Ground State	Ferromagnetic (FM) is stable; AFM is higher in energy by 49 meV/f.u.
Strain Effect	No FM $\to$ AFM transition observed up to $\pm 5\%$ strain.
Size Effect	Reducing $L$ from $\infty$ to 100 nm reduces $\kappa$ by $\sim 7\times$ and lowers anisotropy ratio.

5. Significance

• Methodological Benchmark: The paper highlights the critical necessity of solving the full BTE (including Normal scattering) rather than relying on RTA for accurate thermal transport predictions in 2D magnetic materials.
• Material Design: It establishes CrSBr as a highly anisotropic thermal conductor, useful for thermal management applications where directional heat flow is required.
• Control Knob: The discovery that geometric confinement (flake size) can tune thermal anisotropy offers a new degree of freedom for engineering thermal properties in 2D devices without chemical doping or complex external fields.
• Magnetic Robustness: The confirmation that the FM state is robust against strain simplifies the design of strain-based sensors, as the magnetic order will not be inadvertently switched by mechanical stress.