Intrinsic high electrical conductivity of stoichiometric $\mathrm{SrNb}{\mathrm{O}}_3$ epitaxial thin films

$\mathrm{SrV}{\mathrm{O}}_3$ and $\mathrm{SrNb}{\mathrm{O}}_3$ are perovskite-type transition-metal oxides with the same $d^1$ electronic configuration. Although $\mathrm{SrNb}{\mathrm{O}}_3$ $\left(4d^1\right)$ has a larger $d$ orbital than $\mathrm{SrV}{\mathrm{O}}_3$ $\left(3d^1\right)$, the reported electrical resistivity of $\mathrm{SrNb}{\mathrm{O}}_3$ is much higher than that of $\mathrm{SrV}{\mathrm{O}}_3$, probably owing to nonstoichiometry. In this paper, we grew epitaxial, high-conductivity stoichiometric $\mathrm{SrNb}{\mathrm{O}}_3$ using pulsed laser deposition. The growth temperature strongly affected the Sr/Nb ratio and the oxygen content of the films, and we obtained stoichiometric $\mathrm{SrNb}{\mathrm{O}}_3$ at a very narrow temperature window around 630 °C. The stoichiometric $\mathrm{SrNb}{\mathrm{O}}_3$ epitaxial thin films grew coherently on $\mathrm{KTa}{\mathrm{O}}_3$ (001) substrates with high crystallinity. The room-temperature resistivity of the stoichiometric film was $2.82\times {10}^{-5}\mathrm{\Omega }\mathrm{cm}$, one order of magnitude lower than the lowest reported value of $\mathrm{SrNb}{\mathrm{O}}_3$ and comparable with that of $\mathrm{SrV}{\mathrm{O}}_3$. We observed a $T$-square dependence of resistivity below $T^{*}=180\mathrm{K}$ and non-Drude behavior in near-infrared absorption spectroscopy, attributable to the Fermi-liquid nature caused by electron correlation. Analysis of the $T$-square coefficient $A$ of resistivity experimentally revealed that the $4d$ orbital of Nb that is larger than the $3d$ ones certainly contributes to the high electrical conduction of $\mathrm{SrNb}{\mathrm{O}}_3$.

Figure 1

(a) Two-dimensional detector images of the $\mathrm{SrNb}{\mathrm{O}}_3$ thin films grown on $\mathrm{KTa}{\mathrm{O}}_3$ single crystals at various temperatures. The tilt angle of the sample (χ) was 45°. The 001 and 101 peaks of the films and substrates overlap with each other. White circles indicate the streaky diffraction patterns from the films. The scale of the color bar representing the intensity is set to make the weak streak pattern clear. The 001 and 101 peaks are displayed with the saturated color, of which counts are over 3000 and 5000, respectively. (b) Out-of-plane lattice constants of the $\mathrm{SrNb}{\mathrm{O}}_3$ films plotted against growth temperature. (c) XPS intensity ratios between the Sr $4p$ and Nb $3d$ peaks of the $\mathrm{SrNb}{\mathrm{O}}_3$ thin films grown at various temperatures. These values are normalized by the photoionization cross section.

Figure 2

(a) XRD 2θ-θ pattern of the stoichiometric $\mathrm{SrNb}{\mathrm{O}}_3$ epitaxial thin film. The inset shows the pattern near the 002 peak taken with a high-resolution detector. (b) XRD reciprocal-space map image around the 103 diffractions of the $\mathrm{SrNb}{\mathrm{O}}_3$ thin film and $\mathrm{KTa}{\mathrm{O}}_3$ substrate. (c) Absorption coefficient spectrum of the stoichiometric $\mathrm{SrNb}{\mathrm{O}}_3$ thin film. The dashed line is a fit to the simple Drude model, using the carrier density evaluated by a Hall measurement. The inset shows a Tauc plot that assumes an indirect band gap.

Figure 3

Growth temperature dependence of (a) resistivity ρ, (b) carrier density $n$, and (c) Hall mobility μ of the $\mathrm{SrNb}{\mathrm{O}}_3$ thin films measured at room temperature. The dashed line in (b) indicates the nominal carrier concentration.

Figure 4

Temperature dependence of the (a) resistivity and (b) Hall mobility of the stoichiometric $\mathrm{SrNb}{\mathrm{O}}_3$ epitaxial thin film. The inset in (a) shows the resistivity plotted against $T^2$. The dashed line is a linear fit of the curve in the low-temperature region.

Figure 5

(a) Plot of the $T$-square coefficient $A$ of resistivity vs carrier density $n$ for single crystals or epitaxial films of perovskite-type transition-metal oxides $\mathrm{[}\mathrm{S}{\mathrm{r}}_{1-x}\mathrm{L}{\mathrm{a}}_x\mathrm{Ti}{\mathrm{O}}_3$ [22], $\mathrm{SrT}{\mathrm{i}}_{1-x}\mathrm{N}{\mathrm{b}}_x{\mathrm{O}}_3$ [23], ${\mathrm{K}}_{1-x}\mathrm{B}{\mathrm{a}}_x\mathrm{Ta}{\mathrm{O}}_3$ [24], $\mathrm{C}{\mathrm{a}}_{1-x}{\mathrm{Y}}_x\mathrm{Ti}{\mathrm{O}}_3$ [25], $\mathrm{CaV}{\mathrm{O}}_3$ [26], $\mathrm{SrV}{\mathrm{O}}_3$ [14], $\mathrm{C}{\mathrm{a}}_{1-x}\mathrm{S}{\mathrm{r}}_x\mathrm{V}{\mathrm{O}}_3$ [27], $\mathrm{SrMo}{\mathrm{O}}_3$ [28], and $\mathrm{SrNb}{\mathrm{O}}_3$ (this study)]. Dashed lines with a slope of −1 are drawn for several series of materials as guides for the eyes. The gray shaded area indicates the band with a slope of −1 from $m_{\mathrm{b}}\mathrm{\Phi }=5\times {10}^{-52}$ to $2\times {10}^{-50}\mathrm{kg}\mathrm{s}$. (b) Plots of the $m_{\mathrm{b}}\mathrm{\Phi }$ value vs the tolerance factor for $AM{\mathrm{O}}_3$ $\left(A=\mathrm{Sr},\mathrm{Ca},M=\mathrm{V},\mathrm{Nb},\mathrm{Mo}\right)$. Data are taken from the same references as (a). The tolerance factor was calculated from the ionic radius reported in Ref. [32].