Quantum corrections to the conductivity in a perovskite oxide: A low-temperature study of LaNi1-xCoxO3 (0

Abstract

In this paper we propose to study the evolution of the quantum corrections to the conductivity in an oxide system as we approach the metal-insulator (M-I) transition from the metallic side. We report here the measurement of the low-temperature (0.1 KT100 K) electrical conductivity of the perovskite-structure oxide system ${\mathrm{LaNi}}_{1\mathrm{\ensuremath{-}}\mathit{x}}$ ${\mathrm{Co}}_{\mathit{x}}$${\mathrm{O}}_{3}$ (0\ensuremath{\le}x\ensuremath{\le}0.75). ${\mathrm{LaNiO}}_{3}$ is a metal and ${\mathrm{LaCoO}}_{3}$ is an insulator. The system is metallic for x\ensuremath{\le}0.65. For all x, at low temperatures, the conductivity (\ensuremath{\sigma}) rises with temperature (T). Below 2 K, \ensuremath{\sigma} follows a power-law behavior, \ensuremath{\sigma}(T)=\ensuremath{\sigma}(0)+\ensuremath{\alpha}${\mathit{T}}^{\mathit{m}}$. For samples in the metallic regime, away from the metal-insulator transition (x\ensuremath{\le}0.4), m\ensuremath{\approxeq}0.3--0.4. As the transition is approached [i.e., \ensuremath{\sigma}(0)\ensuremath{\rightarrow}0], m increases rapidly; and at the transition [\ensuremath{\sigma}(0)=0, ${\mathit{x}}_{\mathit{c}}$\ensuremath{\approxeq}0.65], m\ensuremath{\approxeq}1. On the insulating side (x>0.65), m takes on large values and \ensuremath{\sigma}(0)=0. We explain the temperature dependence of \ensuremath{\sigma}(T), for T2 K, on the metallic side (x\ensuremath{\le}0.4), as arising predominantly from electron-electron interactions, taking into account the diffusion-channel contribution (which gives m=0.5) as well as the Cooper-channel contribution. In this regime, the correction to conductivity, \ensuremath{\delta}\ensuremath{\sigma}(T), is a small fraction of \ensuremath{\sigma}(T). However, as the M-I transition is approached (x\ensuremath{\rightarrow}${\mathit{x}}_{\mathit{c}}$), \ensuremath{\delta}\ensuremath{\sigma}(T) starts to dominate \ensuremath{\sigma}(T) and the above theories fail to explain the observed \ensuremath{\sigma}(T).