Mott Transition and Transport Crossovers in the Organic Compound $\kappa \mathrm{\text{−}}(\mathrm{B}\mathrm{E}\mathrm{D}\mathrm{T}\mathrm{\text{−}}\mathrm{T}\mathrm{T}\mathrm{F}{)}_2\mathrm{C}\mathrm{u}[\mathrm{N}(\mathrm{C}\mathrm{N}{)}_2]\mathrm{C}\mathrm{l}$

We have performed in-plane transport measurements on the two-dimensional organic salt $\kappa \mathrm{\text{−}}(\mathrm{B}\mathrm{E}\mathrm{D}\mathrm{T}\mathrm{\text{−}}\mathrm{T}\mathrm{T}\mathrm{F}{)}_2\mathrm{C}\mathrm{u}[\mathrm{N}(\mathrm{C}\mathrm{N}{)}_2]\mathrm{C}\mathrm{l}$. A variable (gas) pressure technique allows for a detailed study of the changes in conductivity through the insulator-to-metal transition. We identify four different transport regimes as a function of pressure and temperature (corresponding to insulating, semiconducting, “bad metal,” and strongly correlated Fermi-liquid behaviors). Marked hysteresis is found in the transition region, which displays complex physics that we attribute to strong spatial inhomogeneities. Away from the critical region, good agreement is found with a dynamical mean-field calculation of transport properties using the numerical renormalization group technique.

Figure 1

Pressure-temperature phase diagram of the $\kappa$-Cl salt. The crossover lines identified from our transport measurements delimit four regions, as described in the text. The spinodal lines defining the region of coexistence of insulating and metallic phases (hatched) are indicated, as well as the line where $d\sigma /dP$ is maximum. The latter yields an estimate of the first-order transition line, ending in a critical end point. The transition line into an antiferromagnetic insulating phase has been taken from Ref. 5, while the superconducting phase [3, 5] below 13 K has been omitted.

Figure 2

Pressure dependence of the in-plane conductivity $\sigma$ at different temperatures. The inset shows the derivative $d\sigma /dP$ which reveals a sharp peak at low temperature.

Figure 3

Conductivity measured for increasing (${\sigma }_{\mathrm{i}\mathrm{n}\mathrm{c}}$) and decreasing (${\sigma }_{\mathrm{d}\mathrm{e}\mathrm{c}}$) pressure sweeps, at different temperatures up to 39 K. Hysteresis at 35 and 39 K is more apparent in the difference ${\sigma }_{\mathrm{d}\mathrm{e}\mathrm{c}}-{\sigma }_{\mathrm{i}\mathrm{n}\mathrm{c}}$ (inset).

Figure 4

Pressure dependence of the paramagnetic gap. The inset displays $\Delta =2d\ln \rho /d(1/T)$ vs $T$ for a few pressures, from which the crossover from a Mott insulator to a semiconductor is identified.

Figure 5

Pressure dependence of the $T^2$ coefficient $A$ and residual resistivity ${\rho }_0$. Inset: Plot of $\rho$ vs $T^2$, also showing the increase of $T^{*}$ with pressure (straight lines are guides to the eyes).

Figure 6

Temperature dependence of the resistivity at different pressures. The data (circles) are compared to a DMFT-NRG calculation (diamonds), with a pressure dependence of the bandwidth as indicated. The measured residual resistivity ${\rho }_0$ has been added to the theoretical curves.