Ion-gel-gating-induced oxygen vacancy formation in epitaxial $\mathrm{L}{\mathrm{a}}_{0.5}\mathrm{S}{\mathrm{r}}_{0.5}\mathrm{Co}{\mathrm{O}}_{3-\delta }$ films from in operando x-ray and neutron scattering

Ionic-liquid/gel-based transistors have emerged as a potentially ideal means to accumulate high charge-carrier densities at the surfaces of materials such as oxides, enabling control over electronic phase transitions. Substantial gaps remain in the understanding of gating mechanisms, however, particularly with respect to charge carrier vs oxygen defect creation, one contributing factor being the dearth of experimental probes beyond electronic transport. Here we demonstrate the use of synchrotron hard x-ray diffraction and polarized neutron reflectometry as in operando probes of ion-gel transistors based on ferromagnetic $\mathrm{L}{\mathrm{a}}_{0.5}\mathrm{S}{\mathrm{r}}_{0.5}\mathrm{Co}{\mathrm{O}}_{3-\delta }$. An asymmetric gate-bias response is confirmed to derive from electrostatic hole accumulation at negative gate bias vs oxygen vacancy formation at positive bias. The latter is detected via a large gate-induced lattice expansion (up to 1%), complementary bulk measurements and density functional calculations enabling quantification of the bias-dependent oxygen vacancy density. Remarkably, the gate-induced oxygen vacancies proliferate through the entire thickness of 30–40-unit-cell-thick films, quantitatively accounting for changes in the magnetization depth profile. These results directly elucidate the issue of electrostatic vs redox-based response in electrolyte-gated oxides, also demonstrating powerful approaches to their in operando investigation.

Figure 1

(a) Device and experimental setup schematic for synchrotron x-ray diffraction on epitaxial $\mathrm{L}{\mathrm{a}}_{0.5}\mathrm{S}{\mathrm{r}}_{0.5}\mathrm{Co}{\mathrm{O}}_{3-\delta }$ films. (b) Gate-bias-($V_{\mathrm{g}}$)-dependent specular diffraction (00L) scans, where L is in reciprocal lattice units (r.l.u.) of the LAO substrate. (c) Change in $c$-axis lattice parameter ($\mathrm{\Delta }{c}_{\mathrm{op}}$, left axis) and cell volume (ΔV, right axis) with $V_{\mathrm{g}}$. (d) Change in Scherrer thickness ($\mathrm{\Delta }{t}_{\mathrm{s}}$, left axis), and $t_{\mathrm{s}}$ itself (right axis), vs $V_{\mathrm{g}}$. (e) Change in film thickness from Laue fringes ($\mathrm{\Delta }{t}_{\mathrm{f}}$), and $t_{\mathrm{f}}$ itself (right axis), vs $V_{\mathrm{g}}$. Blue dotted lines are guides to the eye.

Figure 2

(a) Powder x-ray diffraction patterns of bulk $\mathrm{L}{\mathrm{a}}_{0.5}\mathrm{S}{\mathrm{r}}_{0.5}\mathrm{Co}{\mathrm{O}}_{3-\delta }$ before and after the thermogravimetric analysis (TGA) scans shown in (c), which induce progressive reduction. The δ values from TGA are labeled on the right side of (a), and the curves are color-coordinated with (c). * denotes samples no longer in the perovskite structure, which are omitted from further analysis. (b) Reference powder diffraction pattern for cubic $\mathrm{L}{\mathrm{a}}_{0.5}\mathrm{S}{\mathrm{r}}_{0.5}\mathrm{Co}{\mathrm{O}}_3$. (c) Bulk TGA scans, i.e., sample mass change (left axis) and conversion to δ (right axis) vs temperature, for reduction in flowing ${\mathrm{N}}_2$. (d) Change in bulk cell volume (ΔV) and bulk lattice parameter ($a$) as a function of δ. The maroon dotted line is a straight line fit and the solid black lines are the theoretical ΔV due to formation of oxygen vacancies ($U_{\mathrm{Co}}=5$ and 6 eV). (e) Change in δ (Δδ) with $V_{\mathrm{g}}$ in epitaxial films, obtained by combining Figs. 1 and 2. The blue dashed line is a guide to the eye.

Figure 3

(a) Device and experimental setup schematic for polarized neutron reflectometry on epitaxial $\mathrm{L}{\mathrm{a}}_{0.5}\mathrm{S}{\mathrm{r}}_{0.5}\mathrm{Co}{\mathrm{O}}_{3-\delta }$ films. (b) Reflectivity (R) vs scattering vector magnitude, Q, from a 165 Å film at zero gate bias ($V_{\mathrm{g}}$), 30 K, and 3 T. Black and red denote the non-spin-flip “$R^{++}$” and “$R^{--}$” channels, respectively, for both data (points) and the fits to the model discussed in the text (lines). (c)–(e) Spin asymmetry at 30 K in 3 T for $V_{\mathrm{g}}=0$, 2, and 3 V, respectively. Solid lines are fits to the model discussed in the text. (f) Extracted magnetization (M) depth profile at 30 K in 3 T for $V_{\mathrm{g}}=0$, 2, and 3 V; $z=0$ is the substrate/film interface.