Manipulating the polar mismatch at the $\mathrm{LaNi}{\mathrm{O}}_3/\mathrm{SrTi}{\mathrm{O}}_3$ (111) interface

Heteroepitaxial growth of transition-metal oxide films on the open (111) surface of $\mathrm{SrTi}{\mathrm{O}}_3$ results in significant restructuring due to the polar mismatch. Monitoring the structure and composition on an atomic scale of $\mathrm{LaNi}{\mathrm{O}}_3/\mathrm{SrTi}{\mathrm{O}}_3$ (111) interface as a function of processing conditions has enabled the avoidance of the expected polar catastrophe. Using atomically resolved transmission electron microscopy and spectroscopy as well as low-energy electron diffraction, the structure of the thin film, from interface to the surface, has been studied. In this paper, we show that the proper processing can lead to a structure that is ordered, coherent with the substrate without intermediate structural phase. Angle-resolved x-ray photoemission spectroscopy shows that the oxygen content of thin films increases with the film thickness, indicating that the polar mismatch is avoided by the presence of oxygen vacancies.

Figure 1

Stacking sequence of $\mathrm{LaNi}{\mathrm{O}}_3/\mathrm{SrTi}{\mathrm{O}}_3$ in (a) [111] and in (b) [001] directions. The structure of $\mathrm{LaNi}{\mathrm{O}}_3/\mathrm{SrTi}{\mathrm{O}}_3$ in the [111] and [001] directions is shown, respectively. It is easily seen that the packing factor of the [111] direction is considerably larger than the [001] direction.

Figure 2

(a) RHEED oscillations for $\mathrm{LaNi}{\mathrm{O}}_3/\mathrm{SrTi}{\mathrm{O}}_3$ (111) are presented for 15 uc. The inset shows the RHEED pattern before and after growth. The streaklike pattern after growth is an indication of 2D growth mode. (b) HAADF-STEM image of 16-uc $\mathrm{LaNi}{\mathrm{O}}_3/\mathrm{SrTi}{\mathrm{O}}_3$ (111) along the $\left[1\overline{1}0\right]$ direction. The interface is marked by the yellow line and the ball model mapped on the image shows the schematic of $\mathrm{LaNi}{\mathrm{O}}_3/\mathrm{SrTi}{\mathrm{O}}_3$ (111). (c) The EELS elemental mapping and line profiles for Ti, Sr, and La. The change in the formal valence of Ti is shown across the interface.

Figure 3

Oxidation state of Ti ions for the $\mathrm{LaNi}{\mathrm{O}}_3/\mathrm{SrTi}{\mathrm{O}}_3$ film. (a) Background-subtracted EELS spectra of Ti $L$ edges across the $\mathrm{LaNi}{\mathrm{O}}_3/\mathrm{SrTi}{\mathrm{O}}_3$ interface layer. The terminated Ti layer was set as $N=0$. (b) The energy position of $\mathrm{Ti}{L}_3$ (red) and $L_2$ (blue) $e_g$ peaks.

Figure 4

(a), (b) LEED pattern of 3- and 5-uc $\mathrm{LaNi}{\mathrm{O}}_3$ (111) thin films. Red circles show the position of fractional spots which would be associated with a $\mathrm{L}{\mathrm{a}}_2\mathrm{N}{\mathrm{i}}_2{\mathrm{O}}_5$ surface. (c) Surface of $\mathrm{L}{\mathrm{a}}_2\mathrm{N}{\mathrm{i}}_2{\mathrm{O}}_5$ (111). The rows of oxygen vacancies are shown with yellow arrows. The simulated LEED pattern for this surface is shown. (d) Surface of $\mathrm{LaNi}{\mathrm{O}}_3$ (111). The simulated LEED patterns are shown next to each structure. For simulated LEED patterns, green spots are fractional. Red and blue spots are integers, each color accounting for a domain.

Figure 5

(a) ABF-STEM image of 16-uc $\mathrm{LaNi}{\mathrm{O}}_3/\mathrm{SrTi}{\mathrm{O}}_3$ (111) along $the\left[1\overline{1}0\right]$ direction. The interface is marked by the yellow line. (b) FFT of the ABF-STEM image. Red circles indicate the position of fractional spots for $\mathrm{L}{\mathrm{a}}_2\mathrm{N}{\mathrm{i}}_2{\mathrm{O}}_5$ phase. Absence of fractional spots in the FFT image indicates no ordered oxygen vacancy. (c), (d) Schematic of $\mathrm{LaNi}{\mathrm{O}}_3$ and $\mathrm{L}{\mathrm{a}}_2\mathrm{N}{\mathrm{i}}_2{\mathrm{O}}_5$ projected along $\left[1\overline{1}0\right]$. The simulated electron diffraction patterns are shown in the inset. The Fourier transform of $\mathrm{L}{\mathrm{a}}_2\mathrm{N}{\mathrm{i}}_2{\mathrm{O}}_5$ shows fractional spots which are absent in $\mathrm{LaNi}{\mathrm{O}}_3$.

Figure 6

(a)–(d) XPS spectra of Ni 3p for different thicknesses at normal emission. (e) Left: Change in binding energy of Ni $3p$ core level as a function of thickness. Right: Change in area ratio of $\mathrm{N}{\mathrm{i}}^{3+}/\mathrm{N}{\mathrm{i}}^{2+}$ for Ni $3p$ core level. (f) The difference in binding energy of La $4d$, O $1s$, and Ni $3p$ in normal emission and $\theta ={75}^{\circ }$. The inset shows the schematic angle-dependent XPS.