Spin transport from order to disorder

TL;DR

This paper uses Schwinger boson mean-field theory to analyze spin transport in magnetic systems across ordered and disordered phases, revealing temperature-dependent behaviors and the potential of the spin Seebeck effect as a probe of magnetic correlations.

Contribution

It introduces a unified theoretical framework to evaluate spin transport in magnetic systems at all temperatures, including disordered phases, and explains experimental observations in gadolinium gallium garnet.

Findings

Spin Seebeck coefficient changes sign near the transition temperature.

At high temperatures, the behavior aligns with Curie-Weiss physics.

Deviations from susceptibility indicate short-range correlations in disordered phases.

Abstract

Schwinger boson mean-field theory (SBMFT) is a non-perturbative approach which treats ordered and disordered phases of magnetic systems on equal footing. We leverage its versatility to evaluate the spin correlators which determine thermally-induced spin transport (the spin Seebeck effect) in Heisenberg ferromagnets (FMs) and antiferromagnets (AFs), at arbitrary temperatures. In SBMFT, the spin current, $J_s$, is made up of particle-hole-like excitations which carry integral spin angular momentum. Well below the ordering temperature, $J_s$ is dominated by a magnonic contribution, reproducing the behavior of a dilute-magnon gas. Near the transition temperature, an additional, paramagnetic-like contribution becomes significant. In the AF, the two contributions come with opposite signs, resulting in a signature, rapid inversion of the spin Seebeck coefficient as a function of temperature. Ultimately, at high temperatures, the low-field behavior of the paramagnetic SSE reduces to Curie-Weiss physics. Analysis based on our theory confirms that in recent experiments on gadolinium gallium garnet, the low-field spin Seebeck coefficient $\mathcal{S}(T) \propto \chi(T)$, the spin susceptibility, down to the Curie-Weiss temperature. At lower temperatures in the disordered phase, our theory shows a deviation of $\mathcal{S}(T)$ relative to $\chi(T)$ in both FMs and AFs, which increases with decreasing temperature and arises due to a paramagnetic liquid phase in our theory. These results demonstrate that the SSE can be a probe of the short-ranged magnetic correlations in disordered correlated spin systems and spin liquids.