Effect of Irradiation-Induced Disorder on the Conductivity and Critical Temperature of the Organic Superconductor $\kappa \mathrm{\text{−}}(\mathrm{BEDT}\mathrm{\text{−}}\mathrm{TTF}{)}_2\mathrm{Cu}(\mathrm{SCN}{)}_2$

We have introduced defects into clean samples of the organic superconductor $\kappa \mathrm{\text{−}}(\mathrm{BEDT}\mathrm{\text{−}}\mathrm{TTF}{)}_2\mathrm{Cu}(\mathrm{SCN}{)}_2$ in order to determine their effect on the temperature dependence of the interlayer conductivity $\sigma$ and the critical temperature $T_{\mathrm{c}}$. We find a violation of Matthiessen’s rule that can be explained by a model of $\sigma$ involving a defect-assisted interlayer channel which acts in parallel with the bandlike conductivity. We observe an unusual dependence of $T_{\mathrm{c}}$ on residual resistivity, inconsistent with the generalized Abrikosov-Gor’kov theory for an order parameter with a single component, providing an important constraint on models of the superconductivity in this material.

Figure 1

Interlayer resistivity of an x-ray damaged sample as a function of temperature for a range of irradiation doses, showing the violation of Matthiessen’s rule. Data from the proton-irradiated samples are qualitatively the same. Insets show the ET molecule and a magnified view of the region near $T_{\mathrm{cross}}$ (defined in the text).

Figure 2

The low-temperature region of Fig. 1. The upper inset shows that $\Delta {\rho }_0$ increases with x-ray dose in an approximately linear fashion. Data from proton-irradiated samples show a similar pattern as a function of disorder. The lower inset shows three methods of defining the critical temperature: a lower $T_{\mathrm{c}1}$, an upper $T_{\mathrm{c}2}$, and $T_{\mathrm{c}}$, the maximum of the first derivative, which is the definition used in this Letter.

Figure 3

The dependence of the change in interlayer conductivity at the peak, $\Delta {\sigma }_{\mathrm{p}}$, on the change in residual resistivity, $\Delta {\rho }_0$, which is approximately linear, except for at large $\Delta {\rho }_0$. Because of the different nature of the defects, the slope of proton (two different samples shown with open circles and squares) and x-ray (one sample indicated by solid circles) data differ, though the effect remains qualitatively the same. Inset: the position $T_{\mathrm{p}}$ of the resistivity peak shifts slightly to higher temperature with increasing ${\rho }_0(x)$.

Figure 4

The suppression of $T_{\mathrm{c}}$ compared to the interlayer residual resistivity ${\rho }_0(x)$. The suppression is approximately linear for low damage ${\rho }_0(x)<{\rho }_0^{*}=2\Omega \mathrm{cm}$, but there is a change in gradient for ${\rho }_0>{\rho }_0^{*}$. The inset shows the dependence of the width $\Delta T_{\mathrm{c}}$ of the superconducting transition on ${\rho }_0(x)$. Symbols correspond to those in Fig. 3 and data for an additional undamaged sample are shown by a solid diamond.