Scaling of the spin Seebeck effect in bulk and thin film

K. Morrison,A.J Caruana,C. Cox

DOI: https://doi.org/10.48550/arXiv.1705.02491

2018-07-23

Abstract:Whilst there have been several reports of the spin Seebeck effect to date, comparison of the absolute voltage(s) measured, in particular for thin films, is limited. In this letter we demonstrate normalization of the spin Seebeck effect for Fe$_3$O$_4$:Pt thin film and YIG:Pt bulk samples with respect to the heat flux J$_q$, and temperature difference $\Delta$T. We demonstrate that the standard normalization procedures for these measurements do not account for an unexpected scaling of the measured voltage with area that is observed in both bulk and thin film. Finally, we present an alternative spin Seebeck coefficient for substrate and sample geometry independent characterization of the spin Seebeck effect.

Mesoscale and Nanoscale Physics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is related to the standardization issue in the measurement of the Spin Seebeck Effect (SSE), especially in thin - film and bulk materials. Specifically:

1. **Measurement standardization problem**: Currently, the absolute voltage measurement of the Spin Seebeck Effect, especially in thin - film materials, is difficult to be effectively compared. The author points out that the existing standardization methods do not fully consider the unexpected scaling phenomenon of the measured voltage with the area change.

2. **Relationship between heat flow and temperature difference**: The traditional definition of the Spin Seebeck coefficient depends on the temperature difference \( \Delta T \), but this method has limitations in thin - film samples because it is difficult to directly measure the temperature difference inside the active material. And the standardization method based on heat flow \( J_q \) is more reliable.

3. **Area - dependence problem**: It is found in experiments that the measured Spin Seebeck voltage \( V_{\text{ISHE}} \) is not only proportional to the contact spacing \( L_y \) and the temperature gradient \( \nabla T \), but also related to the sample area \( A_T \). This area - dependence is rarely discussed in the existing literature, and it is necessary to re - define a Spin Seebeck coefficient that is independent of the substrate and sample geometry.

### Main problems summary

- **How to accurately standardize the measurement results of the Spin Seebeck Effect to achieve effective comparison between different samples?**
- **Why do the existing standardization methods have limitations in thin - film samples?**
- **How to explain and handle the phenomenon that the measured voltage changes with the sample area?**

To solve these problems, the author proposes a new Spin Seebeck coefficient \( S_5 \), which eliminates the influence of geometry by introducing an area - dependent term, thereby achieving a more general standardization method. The specific formula is as follows:

\[ S_5=\frac{V_{\text{ISHE}}}{L_y\left(\frac{J_q}{\sqrt{A_T}}\right)} \]

This new coefficient can better describe the Spin Seebeck Effect and is applicable to samples with different geometries, thus providing a more reliable benchmark for future research.