Epitaxial strain modulated electronic properties of interface controlled nickelate superlattices

Perovskite nickelate heterostructures consisting of single unit cells of ${\mathrm{EuNiO}}_3$ and ${\mathrm{LaNiO}}_3$ have been grown on a set of single crystalline substrates by pulsed laser interval deposition to investigate the effect of epitaxial strain on electronic and magnetic properties at the extreme interface limit. Despite the variation of substrate in-plane lattice constants and lattice symmetry, the structural response to heterostructuring is primarily controlled by the presence of the ${\mathrm{EuNiO}}_3$ layer. In sharp contrast to bulk ${\mathrm{LaNiO}}_3$ or ${\mathrm{EuNiO}}_3$, the superlattices grown under tensile strains exhibit metal-to-insulator transitions (MIT) below room temperature. The onset of magnetic and electronic transitions associated with the MIT can be further separated by application of large tensile strain. Furthermore, these transitions can be entirely suppressed by very small compressive strain. X-ray resonant absorption spectroscopy measurements reveal that such strain-controlled MIT is directly linked to a strain-induced self-doping effect without any chemical doping.

Figure 1

(a) $1{\mathrm{EuNiO}}_3/1{\mathrm{LaNiO}}_3$ superlattice along pseudocubic [0 0 1]. (b) Schematic of sequence for layer-by-layer deposition used for this study. (c) Intensity variation of RHEED specular spot during the deposition on ${\mathrm{NdGaO}}_3$ substrate. (d) Final RHEED pattern obtained along the $\left[1-10\right]$ direction of NGO after cooling the film to room temperature. (e) XRD patterns of [1ENO/$1\mathrm{LNO]}\times 10$ superlattices on various substrates. The curves have been shifted along the $y$ axis for visual clarity. The vertical lines represent the expected ${\left(002\right)}_{pc}$ peak position for bulk ${\mathrm{EuNiO}}_3$ and ${\mathrm{LaNiO}}_3$.

Figure 2

(a) Experimental geometry of XLD measurement. TEY and TFY refers to total electron yield and total fluorescence yield respectively. (b) Ni $L_2$ XAS recorded in bulk sensitive TFY mode with horizontally (H) and vertically (V) polarized light and their differences are shown for these 1ENO/1LNO SLs. Due to strong overlap of the Ni $L_3$ edge with the La $M_4$ edge, only the Ni $L_2$ edge is shown. The data have been shifted vertically for clarity. (c) Splitting and ratio of holes $\left(X\right)$ between two $e_g$ orbitals area shown as a function of substrate's in-plane lattice constant.

Figure 3

(a) Temperature-dependent resistivity of [1ENO/1LNO] superlattice on various substrates. The data of the SL on NGO substrate have been adapted from Ref. [36]. (b) Resistivity analysis for 1ENO/1LNO SL on LAO substrate. Yellow triangles indicate the temperature range where the derivative can be considered constant within the noise. As is evident from this, $d\rho /d{T}^{4/3}$ is almost temperature independent (green curve) upto 230 K. This behavior changes to linear $T$ dependence after that (red curve). Similar analysis for other SLs are shown in the Supplemental Material [52]. (c) Determination of $T_N$ for the film on DSO and NGO substrates by plotting $d\left(ln\rho \right)/d(1/T)$ vs $T$. (d) Temperature dependence of $R_H$ for SLs grown on LAO and YAO substrates.

Figure 4

(a) Prepeak of O $K$-edge absorption around 528.5 eV for 1ENO/1LNO SLs measured at 300 K. The peaks above 530 eV are shown in the Supplemental Material [52]. (b) Energy shift and FWHM of the prepeak as a function of substrate's in-plane lattice constant. An additional strong peak, related to the transition to hybridized Sc $3d$–O $2p$ states is present around 533 eV for 1ENO/1lNO SL on DSO (see Supplemental Material [52]). (c) Schematic representation of the single-particle density of states in terms of charge removal and charge addition for a negative charge transfer material with $d^8{\underline{L}}^m$ as ground state, adapted from Ref. [77]. Left and right panels correspond to covalent insulator and $pd$ metal, respectively.