NMR Investigation of the Quasi-One-Dimensional Superconductor ${\mathrm{K}}_2{\mathrm{Cr}}_3{\mathrm{As}}_3$

We report ${}^{75}\mathrm{As}$ NMR measurements on the new quasi-one-dimensional superconductor ${\mathrm{K}}_2{\mathrm{Cr}}_3{\mathrm{As}}_3$ ($T_c\sim 6.1\mathrm{K}$) [J. K. Bao et al., Phys. Rev. X 5, 011013 (2015)]. We found evidence for strong enhancement of Cr spin fluctuations above $T_c$ in the $[{\mathrm{Cr}}_3{\mathrm{As}}_3{]}_{\infty }$ double-walled subnanotubes based on the nuclear spin-lattice relaxation rate $1/T_1$. The power-law temperature dependence, $1/T_{1}T\sim T^{-\gamma }$ ($\gamma \sim 0.25$), is consistent with the Tomonaga-Luttinger liquid. Moreover, absence of the Hebel-Slichter coherence peak of $1/T_1$ just below $T_c$ suggests an unconventional nature of superconductivity.

Figure 1

The crystal structure of quasi-1D superconductor ${\mathrm{K}}_2{\mathrm{Cr}}_3{\mathrm{As}}_3$ [5]. (a) Top view of four unit cells. A complete $[{\mathrm{Cr}}_3{\mathrm{As}}_3{]}_{\infty }$ DWST is in the middle. (b) Angled view of a unit cell, and (c),(d) DWST. The crystal structure was plotted with VESTA [9].

Figure 2

${}^{75}\mathrm{As}$ NQR spectra measured at (a) 200 K and (b) 6.5 K. We divided the raw NQR signal intensity by the square of the frequency to take into account variation of the sensitivity. We marked three distinctive peaks as $A$, $B_1$, and $B_2$. Inset: Temperature dependence of the ${}^{75}\mathrm{As}$ NQR peak frequency ${\nu }_Q$. All solid curves are guides for the eye.

Figure 3

${}^{75}\mathrm{As}$ nuclear spin-lattice relaxation rate$1/T_1$ measured by NQR techniques in $B=0$ for $A$, $B_1$, and $B_2$ sites. The solid line above $T_c$ is the best fit for the $A$ sites with a power law, $1/T_1\sim T^{1-\gamma }$ ($\gamma =0.25\pm 0.03$), while the dash-dotted line below $T_c$ represents $1/T_1\sim T^4$. The dotted line is the best linear fit of the normal state data for the $B_2$ peak, $1/T_1=0.27T{\mathrm{s}}^{-1}$.

Figure 4

$1/T_{1}T$ for $A$, $B_1$, and $B_2$ sites. The solid curve through the normal state data of the $A$ sites represents the power-law fit, $1/T_{1}T\sim T^{-\gamma }$, with the same $\gamma$ as in Fig. 3. The dotted straight line through the data points of the $B_2$ peak is the Korringa fit, $1/T_{1}T=0.27{\mathrm{s}}^{-1}{\mathrm{K}}^{-1}$ above $T_c$. Inset: Temperature dependence of $T_{1}T$ for the $A$ peak above $T_c$, with the best fit $T_{1}T\sim T^{+\gamma }$.