Impact of the out-of-plane conductivity on spin transport evaluation in a van der Waals material

This study demonstrates that accounting for out-of-plane conductivity in layered materials like PtTe$_2$ is crucial for accurate spin transport evaluation, as conventional isotropic assumptions significantly overestimate key parameters such as the out-of-plane spin diffusion length and spin Hall conductivity.

The Gist

Imagine you are trying to move a crowd of people (electrons) through a building. In most buildings, the hallways are wide and open in every direction, so people can run north, south, east, west, up, or down with equal ease. This is how scientists usually think about electricity in metals.

But in the world of Van der Waals materials (like the one studied in this paper, PtTe₂), the building is very different. It's like a stack of pancakes. People can sprint easily along the flat surface of the pancake (the "in-plane" direction), but trying to jump from one pancake to the next (the "out-of-plane" direction) is like trying to climb a steep, slippery cliff. It's incredibly hard.

This paper is about fixing a major misunderstanding in how we measure spintronics—a futuristic technology that uses the "spin" of electrons (imagine them as tiny spinning tops) to store and process data, rather than just their electric charge.

Here is the breakdown of the story:

1. The Problem: The "Flat Earth" Mistake

For years, scientists studying these "pancake-stack" materials made a big assumption: they treated the material as if it were a solid block of concrete where movement is the same in all directions. They used a "one-size-fits-all" map to calculate how far the spinning electrons could travel before they stopped spinning (a distance called the spin diffusion length) and how efficiently they could be turned into electricity (the Spin Hall Effect).

Because they ignored the "pancake" structure, they were essentially using a map of a flat field to navigate a mountain range. Their calculations were wrong. They thought the electrons could travel much further vertically than they actually could, and they thought the material was more efficient at converting spin to electricity than it really was.

2. The Solution: A 3D GPS with a Stretchy Map

The researchers at Osaka University decided to stop guessing and start simulating. They built a 3D digital model of the material, but they didn't just simulate it normally. They used a clever mathematical trick called renormalization.

Think of it like this:
Imagine you have a rubber sheet representing the material. Because it's so hard to move up and down, the sheet is "stiff" vertically. To make the math easier, the researchers stretched the vertical axis of their map until the "stiffness" looked the same as the "easy" horizontal movement.

• Before: The map was distorted; vertical movement looked easy, but it wasn't.
• After: They stretched the map so the physics looked "isotropic" (the same everywhere), did the math, and then shrank the map back to reality to get the true numbers.

3. The Discovery: We Were Overestimating Everything

When they applied this new, accurate method to PtTe₂ (a material that looks like a shiny, layered crystal), they found some surprising results:

• The "Vertical" Travel is Short: The electrons can run a long way across the pancake, but they barely make it to the next layer. The "out-of-plane" spin diffusion length is actually very short (about 1–2 nanometers).
• The Efficiency is Lower: Because the electrons get stuck vertically, the material isn't quite as efficient at converting spin to electricity as previous studies claimed. The old "flat Earth" models had overestimated the performance.
• Two Different Modes: They also discovered that the material behaves differently depending on how "dirty" or "pure" the crystal is.
  • In pure crystals (high conductivity), the spin conversion is driven by the material's internal structure (like a perfectly designed highway).
  • In slightly dirty crystals (lower conductivity), the conversion is driven by electrons bumping into impurities (like cars hitting potholes).

4. Why This Matters for the Future

This isn't just about fixing a math error; it's about building better computers.

If you are designing a new type of computer memory (like MRAM) using these materials, you need to know exactly how far the "spin signal" can travel. If you use the old, wrong numbers, you might design a chip that is too thick or too wide, causing the signal to die before it reaches its destination.

The Takeaway:
This paper is a warning label for the future of electronics. It tells engineers: "Stop treating layered materials like solid blocks. They are stacks of pancakes. If you want to build fast, efficient spintronic devices, you must account for the fact that electrons hate jumping between layers."

By correcting the map, the researchers have given us a clearer path to designing the next generation of super-fast, low-energy computers.

Technical Summary

Here is a detailed technical summary of the paper "Impact of the out-of-plane conductivity on spin transport evaluation in a van der Waals material" by Nakamura et al.

1. Problem Statement

Layered van der Waals (vdW) materials, such as transition metal dichalcogenides (TMDs), are promising for spintronic applications due to their unique electronic structures and strong spin-orbit interactions (SOI). However, these materials exhibit strong anisotropic conductivity, where the out-of-plane conductivity ($\sigma_\perp$) is significantly lower than the in-plane conductivity ($\sigma_\parallel$).

• The Core Issue: Conventional evaluation methods for spin transport parameters (specifically spin diffusion length $\lambda_s$ and spin Hall conductivity $\sigma_{SH}$) typically assume isotropic resistivity.
• The Consequence: In layered materials where $\sigma_\perp \ll \sigma_\parallel$, this isotropic assumption leads to inaccurate quantitative evaluations. Specifically, neglecting anisotropy can cause significant overestimation of the out-of-plane spin diffusion length and the spin Hall conductivity, hindering the accurate design of spintronic devices.

2. Methodology

The authors investigated the prototypical TMD PtTe$_2$ (a Type-II Dirac semimetal) using a combination of experimental measurements and advanced theoretical modeling.

• Experimental Setup:

  • Materials: Mechanically exfoliated PtTe$_2$ flakes (some forming nanowires) and bulk crystals.
  • Device Structures: Non-local Spin Valve (NLSV) devices with Py/Cu/PtTe$_2$ structures to measure spin accumulation and the Inverse Spin Hall Effect (ISHE).
  • Measurements:
    • Four-terminal resistivity measurements to determine in-plane resistivity ($\rho_\parallel$) as a function of film thickness and width.
    • Out-of-plane resistivity ($\rho_\perp$) measurements on bulk crystals using a specialized electrode configuration to ensure uniform current injection.
    • ISHE and NLSV signal measurements at various temperatures (50 K – 300 K).

• Theoretical Modeling:

  • 3D Finite Element Method (FEM): The authors developed a 3D numerical model to simulate spin transport in anisotropic systems.
  • Renormalization Approach: To treat anisotropic spin diffusion similarly to conventional isotropic models, they introduced a coordinate renormalization: $z' = \sqrt{\sigma_\parallel/\sigma_\perp} z$. This transforms the anisotropic system into an effectively isotropic one with conductivity $\sigma_\parallel$, allowing the extraction of distinct in-plane ($\lambda_\parallel^s$) and out-of-plane ($\lambda_\perp^s$) spin diffusion lengths.
  • Comparison: Results were compared against a conventional isotropic model (assuming $\sigma = \sigma_\parallel$) and a 1D spin diffusion model.

3. Key Contributions

1. Development of an Anisotropic Theoretical Framework: The paper presents a robust theoretical model that treats anisotropic spin diffusion by renormalizing the spatial coordinates, enabling the separate extraction of $\lambda_\parallel^s$ and $\lambda_\perp^s$.
2. Quantitative Correction of Spin Parameters: The study demonstrates that the conventional isotropic assumption systematically overestimates key spin transport parameters in layered materials.
3. Mechanism Identification in PtTe$_2$: The authors identified a crossover in the Spin Hall Effect (SHE) mechanism in PtTe$_2$ from intrinsic (low conductivity) to extrinsic (high conductivity, dominated by skew scattering).
4. Interface vs. Bulk Contribution: The results suggest that the observed high spin-charge conversion efficiency in PtTe$_2$ is likely a combination of bulk SHE and interfacial Rashba-Edelstein effects (REE), supported by the extremely short out-of-plane spin diffusion length.

4. Key Results

• Conductivity Anisotropy:

  • PtTe$_2$ exhibits extreme anisotropy. For a 30 nm film, the anisotropy ratio $\rho_\perp / \rho_\parallel$ exceeds 100 at low temperatures and is approximately 30 at room temperature.
  • $\rho_\parallel$ is highly sensitive to film thickness and width (size effects), whereas $\rho_\perp$ remains relatively constant and close to bulk values.

• Spin Diffusion Length ($\lambda_s$):

  • In-plane ($\lambda_\parallel^s$): The anisotropic analysis yields values nearly identical to the conventional isotropic model ($\lambda_\parallel^s \approx \lambda_{iso}^s$).
  • Out-of-plane ($\lambda_\perp^s$): The anisotropic model reveals that $\lambda_\perp^s$ is significantly shorter than the isotropic estimate. Specifically, $\lambda_\perp^s = \sqrt{\sigma_\perp/\sigma_\parallel} \lambda_\parallel^s$.
  • Finding: The conventional isotropic model overestimates $\lambda_\perp^s$ because it fails to account for the suppression of spin current diffusion in the out-of-plane direction due to low conductivity.

• Spin Hall Conductivity ($\sigma_{SH}$):

  • The study found that $\sigma_{SH} < \sigma_{SH}^{iso}$. When anisotropy is accounted for, the calculated spin Hall conductivity is lower than values derived from isotropic assumptions.
  • This reduction is attributed to the suppression of the shunting effect by the Cu electrode in the presence of conductivity anisotropy.

• SHE Mechanism Crossover:

  • Low Conductivity Regime: $\sigma_{SH}$ is nearly constant, indicating a dominant intrinsic mechanism (Berry curvature).
  • High Conductivity Regime: $\sigma_{SH}$ scales linearly with conductivity ($\sigma_{SH} \propto \sigma$), indicating a crossover to an extrinsic mechanism dominated by impurity skew scattering.
  • The fitted intrinsic spin Hall conductivity is $\sigma_{SH}^{int} \approx 4.8 \times 10^2 (\hbar/2e)\Omega^{-1}\text{cm}^{-1}$.

• Discrepancy with Theory:

  • First-principles calculations for bulk PtTe$_2$ yielded a $\sigma_{SH}$ value roughly 3 times smaller than the experimental effective value.
  • Explanation: The authors attribute this discrepancy to interfacial spin-charge conversion (Rashba-Edelstein effect) at the Cu/PtTe$_2$ interface, which enhances the total conversion efficiency. The extremely short measured $\lambda_\perp^s$ (1–2 nm) supports the conclusion that spin-charge conversion is spatially confined to the interface.

5. Significance

• Paradigm Shift in Evaluation: This work establishes that assuming isotropic resistivity in layered vdW materials leads to erroneous conclusions regarding spin transport efficiency. Future studies on TMDs and other layered materials must incorporate anisotropic models to obtain accurate $\sigma_{SH}$ and $\lambda_s$ values.
• Device Design Implications: For spintronic devices utilizing vdW materials, the out-of-plane spin diffusion length is much shorter than previously thought. This implies that spin injection and detection must be optimized for the specific anisotropic geometry, and interface engineering is critical for maximizing spin-charge conversion.
• Physical Insight: The study clarifies the interplay between intrinsic band-structure effects and extrinsic scattering mechanisms in Type-II Dirac semimetals, while highlighting the significant role of interfacial effects (REE) in enhancing spin-charge conversion in thin-film heterostructures.