Superconductivity and crystalline electric field effects in the filled skutterudite series Pr(Os$_{1-x}$Ru$_x$)$_4$Sb$_{12}$

TL;DR

This study investigates how superconductivity and crystalline electric field effects evolve in the filled skutterudite series Pr(Os$_{1-x}$Ru$_x$)$_4$Sb$_{12}$, revealing a competition between two superconducting types and consistent CEF splitting behavior.

Contribution

It provides detailed experimental analysis of superconducting transition temperatures and CEF effects across the entire series, highlighting the interplay between different superconducting states and electronic configurations.

Findings

Superconducting $T_c$ decreases linearly with Ru concentration.

A minimum $T_c$ of 0.75 K occurs at $x=0.6$, indicating competing superconducting states.

Crystalline electric field splitting increases monotonically with $x$, consistent with a $ ext{Gamma}_3$ doublet ground state.

Abstract

X-ray powder diffraction, magnetic susceptibility $\chi(T)$, and electrical resistivity $\rho(T)$ measurements were made on single crystals of the filled skutterudite series Pr(Os$_{1-x}$Ru$_x$)$_4$Sb$_{12}$. One end of the series ($x = 0$) is a heavy fermion superconductor with a superconducting critical temperature $T_{c} = 1.85$ K, while the other end ($x = 1$) is a conventional superconductor with $T_{c} \approx 1$ K. The lattice constant $a$ decreases approximately linearly with increasing Ru concentration $x$. As Ru (Os) is substituted for Os (Ru), $T_{c}$ decreases nearly linearly with substituent concentration and exhibits a minimum with a value of $T_{c} = 0.75$ K at $x = 0.6$, suggesting that the two types of superconductivity compete with one another. Crystalline electric field (CEF) effects in $\chi_\mathrm{dc}(T)$ and $\rho(T)$ due to the splitting of the Pr$^{3+}$ nine-fold degenerate Hund's rule $J = 4$ multiplet are observed throughout the series, with the splitting between the ground state and the first excited state increasing monotonically as $x$ increases. The fits to the $\chi_\mathrm{dc}(T)$ and $\rho(T)$ data are consistent with a $\Gamma_{3}$ doublet ground state for all values of x, although reasonable fits can be obtained for a $\Gamma_{1}$ ground state for $x$ values near the end member compounds ($x = 0$ or $x = 1$).