Spin-gapped Mott insulator with the dimeric arrangement of twisted molecules Zn(tmdt)${}_2$

${}^{13}\mathrm{C}$ nuclear magnetic resonance measurements were performed for a single-component molecular material Zn(tmdt)${}_2$, in which tmdt's form an arrangement similar to the so-called $\kappa$-type molecular packing in quasi-two-dimensional Mott insulators and superconductors. A detailed analysis of the powder spectra uncovered local spin susceptibility in the tmdt $\pi$ orbitals. The obtained shift and relaxation rate revealed singlet-triplet excitations of the $\pi$ spins, indicating that Zn(tmdt)${}_2$ is a spin-gapped Mott insulator with exceptionally large electron correlations compared to conventional molecular Mott systems.

Figure 1

(a) Molecular structure of Zn(tmdt)${}_2$. (b) Molecular arrangement in the Zn(tmdt)${}_2$ crystal. (c) End-on projection of the “$\kappa$-type” molecular arrangement in the tmdt layer. (d) Temperature dependence of two-probe resistivity for Zn(tmdt)${}_2$ measured on a single crystal of approximately 70 $\mu \mathrm{m}$. The inset shows the Arrhenius plot and an activation energy $E_{\mathrm{a}}=91$ meV was obtained from the data for 200–280 K (indicated by a red curve).

Figure 2

(a) Temperature dependence of ${}^{13}\mathrm{C}$ NMR spectra for the polycrystalline Zn(tmdt)${}_2$. Solid (red) curves are fits described in the text. (b) Temperature dependence of the three principal values of the shift tensor ${\delta }_{ii}$ ($i=x$, $y$, and $z$). The symbols correspond to those indicated in (a).

Figure 3

(a) Temperature dependence of the isotropic part of NMR shift determined by the first moment of the spectra. The red line indicates the isotropic part of chemical shift ${\sigma }_{\mathrm{iso}}$. (b) Temperature dependence of the anisotropic parts of NMR shifts; for example, the level of the blue line is equal to the anisotropic part of chemical shift ${\sigma }_1$ and the deviation from that corresponds to $-K_{\mathrm{aniso}}$. (c) Temperature dependence of spin susceptibility (with a Curie contribution subtracted) of the polycrystalline Zn(tmdt)${}_2$,which was fitted on the basis of the singlet-triplet excitation model (red curve). (d) $K_{\mathrm{iso}}–\chi$ and (e) ${}^i{K}_{\mathrm{aniso}}–\chi$ plots. Solid lines are the fits of the data. The curves in (a) and (b) are the $\chi$ curve [the red curve in (c)] multiplied by respective hyperfine coupling constants determined by the slopes in (d) and (e).

Figure 4

(a) Nuclear spin-lattice relaxation rate $T_1^{-1}$ for Zn(tmdt)${}_2$. The symbols of squares and triangles represent the relaxation rates of the sample and a minor impurity phase (see text). (b) Temperature dependence of the stretched exponent $\beta$ in the fitting of the nuclear relaxation (see text). (c) The activation plot of $T_1^{-1}$ for 200–300 K, where the relaxation rate of the sample is well separated from that of the minor impurity phase. The inset is the activation plot of ${\left(T_{1}T\right)}^{-1}$. (d) Spin model for the $p\pi$ electrons in the “$\kappa$-type” configuration of tmdt's in the two-dimensional layer.

Figure 5

Scheme 1.

Figure 6

Scheme 2.