Evolution of shape and volume fraction of superconducting domains with temperature and anion disorder in (TMTSF)$_2$ClO$_4$

TL;DR

This study models how superconducting domains in (TMTSF)$_2$ClO$_4$ evolve in shape and volume with temperature and disorder, revealing oblate domains and anisotropic percolation affecting superconductivity.

Contribution

It introduces a Maxwell-Garnett approximation-based method to analyze the shape and volume fraction of superconducting domains in an anisotropic organic superconductor, considering disorder effects.

Findings

Superconducting domains are oblate, shortest along the interlayer z-axis.

Cooling rate and annealing influence domain shape differently.

Anisotropic resistivity and $T_c$ can be explained by domain shape and sample geometry.

Abstract

In highly anisotropic organic superconductor (TMTSF)$_2$ClO$_4$, superconducting (SC) phase coexists with metallic and spin density wave phases in the form of domains. Using the Maxwell-Garnett approximation (MGA), we calculate the volume ratio and estimate the shape of these embedded SC domains from resistivity data at various temperature and anion disorder, controlled by the cooling rate or annealing time of (TMTSF)$_{2}$ClO$_{4}$ samples. We found that the variation of cooling rate and of annealing time affect differently the shape of SC domains. In all cases the SC domains have oblate shape, being the shortest along the interlayer $z$-axis. This contradicts the widely assumed filamentary superconductivity along $z$-axis, used to explain the anisotropic superconductivity onset. We show that anisotropic resistivity drop at the SC transition can be described by the analytical MGA theory with anisotropic background resistance, while the anisotropic $T_c$ can be explained by considering a finite size and flat shape of the samples. Due to a flat/needle sample shape, the probability of percolation via SC domains is the highest along the shortest sample dimension ($z$-axis), and the lowest along the sample length ($x$-axis). Our theory can be applied to other heterogeneous superconductors, where the size $d$ of SC domains is much larger than the SC coherence length $\xi$, e.g. cuprates, iron based or organic superconductors. It is also applicable when the spin/charge-density wave domains are embedded inside a metallic background, or vice versa.