Controlling Spin-Mixing Conductance in KTaO$_{3}$ 2DEGs by Varying Argon-Ion Irradiation Time

TL;DR

This work demonstrates that Ar$^+$ irradiation of (001)KTaO$_3$ creates a surface 2DEG with oxygen-vacancy–induced conductivity, confirmed by Ta valence changes and metallic transport down to low temperatures. By placing Py directly on the irradiated KTO surface, spin pumping induces additional damping, and the real part of the spin-mixing conductance $g_{}^r$ can be tuned by irradiation time, increasing from $\sim3.3$ to $\sim30$ nm$^{-2}$ as the 2DEG conductivity rises. The study links the enhanced spin transfer to a thicker, more conductive 2DEG layer (estimated $L\approx6.6$ nm) and confirms the linear, Gilbert-damping–dominated FMR behavior with negligible extrinsic damping. Overall, Ar$^+$-irradiated KTO 2DEGs offer a scalable, tunable platform for efficient oxide spintronics and spin-to-charge conversion optimization.

Abstract

The Rashba-split two-dimensional electron gas (2DEG) at the surface and interface of insulating oxides like KTaO$_{3}$ (KTO) shows great promise for all-oxide spintronics. However, efficient spin current injection into the adjacent 2DEG remains a key challenge. In this study, we report the spin-pumping experiments on a 2DEG formed on the (001)KTO surface via Ar$^+$ irradiation. We observed a significant increase in magnetic damping in the Ar$^+$-KTO/Py bilayer compared to a non-irradiated KTO/Py control sample, confirming spin pumping into the 2DEG. We demonstrate that the spin-mixing conductance ($g_{\uparrow\downarrow}^r$) can be substantially enhanced by controlling the Ar$^+$ irradiation time. The enhancement is attributed to increased 2DEG conductance, which results from a higher concentration of oxygen vacancies with longer irradiation times. This work provides crucial guidance for optimizing spin-to-charge conversion in KTO-based systems, highlighting the potential of Ar$^+$-irradiated KTO 2DEGs for future oxide spintronics.

Figures (7)

• Figure 1: (a) Schematic of the creation of 2DEG on a single-crystal KTO surface via $\mathrm{Ar}^+$ ion irradiation. The $\mathrm{Ar}^+$ ion irradiation generates oxygen vacancies, forming a metallic 2DEG layer just beneath the resulting amorphous KTO top surface. (b) Photographs of the KTO substrate before and after 20 min of $\mathrm{Ar}^+$ ion irradiation, showing a visual change. (c) Sheet resistance ($R_S$) as a function of temperature for samples irradiated for different durations. The metallic behavior across all temperatures, without a low-temperature resistance upturn, confirms successful 2DEG formation. (d) Dependence of room-temperature sheet resistance ($R_S$) on $\mathrm{Ar}^+$ ion irradiation time, illustrating a rapid decrease in resistance as irradiation time increases.
• Figure 2: (a) Atomic Force Microscopy (AFM) image of the single-crystal KTO substrate before $\mathrm{Ar}^+$ ion irradiation, showing the initial surface morphology. (b) AFM image of the KTO substrate after 20 minutes of $\mathrm{Ar}^+$ ion irradiation, illustrating the increase in surface roughness. (c) High-resolution X-ray Photoelectron Spectroscopy (XPS) spectra of the Ta 4f core levels and associated peak fits for the KTO substrate before irradiation. (d) XPS spectra and fits of the Ta 4f core levels for the KTO substrate after 20 minutes of $\mathrm{Ar}^{+}$ion irradiation. The appearance of a new shoulder near 24 eV in the Ta 4 f spectrum confirms the presence of reduced tantalum species ( $\mathrm{Ta}^{4+}, \mathrm{Ta}^{2+}$ ), which is direct evidence for the formation of oxygen vacancies induced by the $\mathrm{Ar}^{+}$ irradiation.
• Figure 3: (a) Experimental schematic depicting the configuration used for the ferromagnetic resonance (FMR) measurements. (b) Room temperature FMR derivative absorption spectra $(d I(H) / d H$, where $I$ is the microwave absorption intensity) recorded for microwave frequencies ranging from 4 to 10 GHz (in 0.5 GHz steps). The solid lines represent the Lorentzian best fits to the experimental data. (c) Frequency dependence of the FMR linewidth $(\Delta H)$ for the reference KTO/Py bilayer and the 20min $\mathrm{Ar}^{+}$-irradiated $\mathrm{Ar}^{+}-\mathrm{KTO} / \mathrm{Py}$ bilayer. Solid lines are fits to Equation \ref{['eq4']}. (d) FMR resonance frequency ( $f$ ) plotted against the resonance field ( $H_{r}$ ) for both the reference and $\mathrm{Ar}^{+}-\mathrm{KTO} / \mathrm{Py}(20 \mathrm{~min})$ samples. The solid line represents the fit using the Kittel equation (Eq. \ref{['eq3']}).
• Figure 4: (a) Effective Magnetization ( $M_{e f f}$ ) as a function of $\mathrm{Ar}^+$ ion irradiation duration. The $M_{e f f}$ values were derived by applying a fit of Equation \ref{['eq3']} to the FMR resonance frequency ( $f$ ) as a function of the resonance field $\left(H_{r}\right)$. (b) Effective Gilbert damping constant $(\alpha)$ versus $\mathrm{Ar}^{+}$ion irradiation time. The damping constant was calculated by fitting the frequency dependence of the FMR linewidth ( $\Delta H$ vs. $f$ ) using Equation \ref{['eq4']}. (c) Spin-mixing conductance $\left(g_{\uparrow \downarrow}^r\right)$ and room-temperature sheet resistance as a function of $Ar^{+}$ ion irradiation time for KTO samples. The left axis shows $\left(g_{\uparrow \downarrow}^r\right)$ values calculated via Eq. \ref{['eq5']}, while the right axis displays the corresponding sheet resistance.
• Figure S1: X-ray diffraction (XRD) of as-received single crystal KTaO$_3$ substrate