Tuning the electronic properties of ${\mathrm{LaAlO}}_3$/${\mathrm{SrTiO}}_3$ interfaces by irradiating the ${\mathrm{LaAlO}}_3$ surface with low-energy cluster ion beams

We have investigated the effects of low-energy ion beam irradiations using argon clusters on the chemical and electronic properties of ${\mathrm{LaAlO}}_3$/${\mathrm{SrTiO}}_3$ (LAO/STO) heterointerfaces by combining x-ray photoelectron spectroscopy (XPS) and electrical transport measurements. Due to its unique features, we demonstrate that a short-time cluster ion irradiation of the LAO surface induces significant modifications in the chemical properties of the buried STO substrate with (1) a lowering of Ti atoms oxidation states (from ${\mathrm{Ti}}^{4+}$ to ${\mathrm{Ti}}^{3+}$ and ${\mathrm{Ti}}^{2+}$) correlated to the formation of oxygen vacancies at the LAO surface and (2) the creation of new surface states for Sr atoms. Contrary to what is generally observed by using higher energy ion beam techniques, this leads to an increase of the electrical conductivity at the LAO/STO interface. Our XPS data clearly reveal the existence of dynamical processes on the titanium and strontium atoms, which compete with the effect of the cluster ion beam irradiation. These relaxation effects are in part attributed to the diffusion of the ion-induced oxygen vacancies in the entire heterostructure since an increase of the interfacial metallicity is also evidenced far from the irradiated area. This paper highlights the possibility of tuning the electrical properties of LAO/STO interfaces by surface engineering, confirming experimentally the intimate connection between LAO chemistry and electronic properties of LAO/STO interfaces.

Figure 1

Time evolution of the XPS atomic percentage of the different detected elements along the argon cluster ion depth profiling of the heterostructure H1. Inset: time evolution of the ratio $[2\times \%\mathrm{at}(\mathrm{O}\left)\right]/[\%\mathrm{at}$ (Sr) + %at (Ti) + %at (La) + %at (Al)]. The solid line is a guide for the eye.

Figure 2

Influence of the cluster ion beam exposure on titanium and strontium atoms during the complete depth profiling of the heterostructure H1. (a) and (b) The XPS spectra of the $\mathrm{Ti}2p_{3/2}$ and Sr $3d$ peaks, respectively, at different etching times. The sketches on the left schematically represent the depth of the ion etching at each step. The insets (at $t=0$) show high-resolution (pass energy 10 eV) measurements of $\mathrm{Ti}2p_{3/2}$ and Sr $3d$ spectra (with a counting time of $\sim 2$ h) performed prior to the beginning of the ion sputtering. The total counting time between each etching step was then fixed to 12 min. (c) and (d) Evolution of the $\alpha =[\mathrm{T}{\mathrm{i}}^{3+}+\mathrm{T}{\mathrm{i}}^{2+}]/[\mathrm{T}{\mathrm{i}}^{4+}+\mathrm{T}{\mathrm{i}}^{3+}+\mathrm{T}{\mathrm{i}}^{2+}]$ and $\beta =\mathrm{S}{\mathrm{r}}_B/\left[\mathrm{S}{\mathrm{r}}_A+\mathrm{S}{\mathrm{r}}_B\right]$ ratios, respectively, as a function of etching time (solid lines are guides for the eye).

Figure 3

The LEIS spectra (978-eV ${\mathrm{He}}^{+}$ ions, scattering angle of 130°) recorded on the STO surface (black curve) and a LAO/STO heterostructure at different times of the cluster ion beam exposure (2000 at./cluster, mean energy 4000 eV/cluster). Data have been normalized and corrected from the background.

Figure 4

(a) Schematic representation of the H2 and H3 samples. For each sample, zone (A) is defined as the ion-irradiated zone and (B) as the witness zone on the other side of the 2-mm bare STO area, denoted (C). (b) and (c) Evolution of the sheet resistance and carrier density, respectively, measured at room temperature on areas (A) (top panel) and (B) (bottom panel) of heterostructures H2 and H3 at different stages of the experiment ($t=0$, after ion beam exposure and after annealing). Arrows help to indicate the evolution.

Figure 5

The XPS results obtained on the irradiated area (A) of H2 and H3 during the cluster ion exposure and the following relaxation time. (a) and (b) Time evolution of the $\alpha =\mathrm{T}{\mathrm{i}}^{3+}/\left[\mathrm{T}{\mathrm{i}}^{4+}+\mathrm{T}{\mathrm{i}}^{3+}\right]$ and $\beta =\mathrm{S}{\mathrm{r}}_B/\left[\mathrm{S}{\mathrm{r}}_A+\mathrm{S}{\mathrm{r}}_B\right]$ ratios, respectively. (c) Evolution of the La/Al cationic ratio as a function of time. This ratio has been evaluated considering a stoichiometric single crystal of ${\mathrm{LaAlO}}_3$ as a reference.

Figure 6

High-resolution TEM images of post-annealed sample H2 showing the stacking of crystallized LAO layer upon STO substrate. The LAO layer remains mostly crystalline and exhibits visible variations in thickness from (a) 4.7 nm to (b) 2.7 nm. To ease LAO layer visualization, its surface is materialized with a dotted line. Arrows point out the existence of perturbed areas at the LAO/STO interface.