Scaling of the spin Seebeck effect in bulk and thin film

TL;DR

This paper investigates how the spin Seebeck effect scales with sample size and heat flux in bulk and thin film materials, revealing unexpected area dependence and proposing a new normalized coefficient for consistent comparison.

Contribution

It introduces a normalization method accounting for heat flux and temperature difference, and proposes an area-independent spin Seebeck coefficient for better comparison across samples.

Findings

Standard normalization does not account for area scaling.

Observed voltage scales unexpectedly with area.

Proposed a new spin Seebeck coefficient for consistent measurement.

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.