Spin-wave propagation at low temperatures in YIG thin films on YSGG substrates

TL;DR

The paper addresses the challenge of cryogenic spin-wave transport by replacing the conventional GGG substrate with YSGG for YIG thin films. It demonstrates robust spin-wave propagation down to 2 K under fields up to 150 mT, using PLD-grown YIG/YSGG films with narrow FMR linewidths. By comparing with YIG/GGG, it shows absence of Gd diffusion–related relaxation and reduced low-temperature inhomogeneous broadening, enabling a scalable cryogenic magnonics platform. The approach opens routes to on-chip magnonic interconnects and hybrid quantum devices operating at cryogenic temperatures.

Abstract

The use of spin waves in magnetic thin films at cryogenic temperatures has long been hindered by the lack of a suitable material platform. Yttrium iron garnet (YIG) is the leading candidate, yet it is typically grown on gadolinium gallium garnet (GGG) substrates, which develop a large paramagnetic moment at low temperatures. This substrate effect limits spin-wave propagation. In this work, we demonstrate that thin YIG films grown on yttrium scandium gallium garnet (YSGG) substrates support robust spin-wave propagation in the Damon-Eshbach geometry, measurable down to 2 K under applied magnetic fields up to 150 mT. Compared with YIG/GGG, YIG/YSGG films exhibit narrower ferromagnetic resonance (FMR) linewidths at low temperatures and are free from the atomic interdiffusion effects that degrade the performance of YIG/GGG systems. These results establish YIG/YSGG thin films as a promising low-temperature platform, overcoming the intrinsic limitations of YIG/GGG and opening new opportunities for scalable magnonic and hybrid quantum devices operating under cryogenic conditions.

Figures (7)

• Figure 1: (a) Temperature dependence of the magnetization of GGG (black) and YSGG (red) substrates measured under a 1 T applied field in the range $4-100$ K. The inset shows the YSGG data with an expanded y-axis scale. (b) XRD $\theta$-$2\theta$ scan of the YIG/YSGG film. (c) XRD $\theta$-$2\theta$ scan of the YIG/GGG film. The (444) reflections from the film and substrate are labeled.
• Figure 2: (a) FMR signal of the YIG/YSGG film measured at 6 GHz and 300 K. The black dots represent the experimental data, and the red line shows the fitted curve. (b) FMR signal of the YIG/YSGG film measured at 6 GHz and 2 K. (c) Temperature dependence of the extracted FMR linewidths for the YIG/YSGG film at different frequencies. (d) FMR signal of the YIG/GGG film measured at 6 GHz and 300 K. (e) FMR signal of the YIG/GGG film measured at 6 GHz and 2 K. (f) Temperature dependence of the extracted FMR linewidths for the YIG/GGG film. Note the different y-axis scales in (c) and (f). All FMR spectra were fitted using two differentiated Lorentzian lineshapes, and the linewidths in (c) and (f) correspond to the higher-amplitude Lorentzian component.
• Figure 3: (a) Heat map of the real part of S$_{21}$ as a function of frequency and applied magnetic field for the YIG/YSGG film at 300 K. (b) Same as (a) for the YIG/GGG sample at 300 K. (c) Same as (a) for the YIG/YSGG sample at 2 K. (d) Same as (a) for the YIG/GGG sample at 2 K. In all measurements, the excitation power was set to $-15$ dBm, using parallel antennas separated by 10 $\upmu$m. A reference spectrum acquired at higher magnetic fields was subtracted from the raw data to obtain the S$_{21}$ signal. Additional background subtraction was applied in (c) and (d) to remove non-magnetic artifacts.
• Figure 4: (a) Real part of S$_{21}$ as a function of frequency at 2 K and 5 mT, measured with parallel antennas separated by 10 $\upmu$m. (b) Same as (a) with 20 $\upmu$m antenna spacing. (c) Same as (a) with 50 $\upmu$m spacing. (d-f) Same as (a) with 10 $\upmu$m spacing at applied fields of 50 mT, 100 mT, and 150 mT, respectively. In all cases, a reference measurement at higher applied magnetic fields was subtracted from the raw data to obtain the S$_{21}$ signal, and additional background subtraction was applied to remove non-magnetic artifacts.
• Figure S1: Temperature-dependent parameters extracted from FMR measurements of the YIG/GGG and YIG/YSGG films: (a) saturation magnetization ($M_\mathrm{s}$), (b) out-of-plane magnetic anisotropy field ($H_\mathrm{k}$), (c) inhomogeneous linewidth broadening ($\Delta H_0$), and (d) phenomenological Gilbert damping parameter ($\alpha$).