Multiband Superconductivity in the Chevrel Phases ${\mathrm{SnMo}}_6{\mathrm{S}}_8$ and ${\mathrm{PbMo}}_6{\mathrm{S}}_8$

Sub-Kelvin scanning tunneling spectroscopy in the Chevrel phases ${\mathrm{SnMo}}_6{\mathrm{S}}_8$ and ${\mathrm{PbMo}}_6{\mathrm{S}}_8$ reveals two distinct superconducting gaps with ${\Delta }_1=3\mathrm{meV}$, ${\Delta }_2\sim 1.0\mathrm{meV}$ and ${\Delta }_1=3.1\mathrm{meV}$, ${\Delta }_2\sim 1.4\mathrm{meV}$, respectively. The gap distribution is strongly anisotropic, with ${\Delta }_2$ predominantly seen when scanning across unit-cell steps on the (001) sample surface. The spectra are well fitted by an anisotropic two-band BCS $s$-wave gap function. Our spectroscopic data are confirmed by electronic heat capacity measurements, which also provide evidence for a twin-gap scenario.

Figure 1

(a) Zero-field 35 nm trace on ${\mathrm{SnMo}}_6{\mathrm{S}}_8$ taken at $T=0.4\mathrm{K}$, junction resistance $R_T=0.03\mathrm{G}\Omega$. (i) Topography showing steps two unit cells high; (ii) spectroscopic trace; (iii),(iv) raw spectra taken on a flat terrace (1) and above a topographic step (2). (b) Zero-field 40 nm trace on ${\mathrm{PbMo}}_6{\mathrm{S}}_8$ taken at $T=0.5\mathrm{K}$, $R_T=0.015\mathrm{G}\Omega$. (i) Spectroscopic trace; (ii) average spectrum from entire trace; (iii) topographic variation.

Figure 2

(a)–(c) ${\mathrm{SnMo}}_6{\mathrm{S}}_8$ spectra and fits: $T=0.4\mathrm{K}$, $R_T=0.03\mathrm{G}\Omega$. (d)–(e) ${\mathrm{PbMo}}_6{\mathrm{S}}_8$ spectra and fits: $T=0.5\mathrm{K}$, $R_T=0.015\mathrm{G}\Omega$. (f) ${\mathrm{PbMo}}_6{\mathrm{S}}_8$ spectrum from 19: $T=1.9\mathrm{K}$, $R_T=0.025\mathrm{G}\Omega$. See the text for details and Table  for fit parameters. $2{\Delta }_1/k_B{T}_c$ in ${\mathrm{PbMo}}_6{\mathrm{S}}_8$ is similar to data from 26.

Figure 3

Temperature variation of the large gap ${\Delta }_1(T)$ in (a) ${\mathrm{SnMo}}_6{\mathrm{S}}_8$ and (b) ${\mathrm{PbMo}}_6{\mathrm{S}}_8$, measured by STS. The gap value was determined by fitting spectra acquired on a flat terrace (i.e., with a negligible ${\Delta }_2$ component) using a BCS single-band anisotropic $s$-wave model.

Figure 4

(a),(b) $C_{\mathrm{elec}}/\gamma T$ with two-band $\alpha$-model fits [25]. Insets: $C/T$ at 0 and 28 T. (c),(d) Field-dependent contributions to $C_{\mathrm{elec}}$ at $T=0.35\mathrm{K}$ with linear fits above and below the crossover field $H_x$ (see the text). The crossover in ${\mathrm{PbMo}}_6{\mathrm{S}}_8$ is rather broad compared to ${\mathrm{SnMo}}_6{\mathrm{S}}_8$; this is due to the lower homogeneity in the ${\mathrm{PbMo}}_6{\mathrm{S}}_8$ crystal as seen by increased transition widths in HC and ac susceptibility data.