Insulator-Metal Transition in Highly Compressed NiO

The insulator-metal transition was observed experimentally in nickel monoxide (NiO) at very high pressures of $\sim 240\mathrm{GPa}$. The sample resistance becomes measurable at about 130 GPa and decreases substantially with the pressure increase to $\sim 240\mathrm{GPa}$. A sharp drop in resistance by about 3 orders of magnitude has been observed at $\sim 240\mathrm{GPa}$ with a concomitant change of the resistance type from semiconducting to metallic. This is the first experimental observation of an insulator-metal transition in NiO, which was anticipated by Mott decades ago. From simple multielectron consideration, the metallic phase of NiO forms when the effective Hubbard energy $U_{\mathrm{eff}}$ is almost equal to the estimated full bandwidth $2W$.

Figure 1

The images of a NiO sample at 35 GPa (a) and 240 GPa (b). The four Pt leads are connected to the NiO sample in the central part of the cBN gasket. At the transition pressure (240 GPa), the NiO sample becomes brown in color (semiconducting phase) and the metallic phase forms black percolation paths in the region of the highest pressure between the two electrodes (b). The cBN gasket is nearly transparent at these conditions. (c) A strong nonlinear decrease of resistance in the NiO sample has been measured at room temperature under compression. We observe a sharp drop of resistance into a metallic state (see inset) at the transition pressure $P_T=240\pm 10\mathrm{GPa}$. The data from several runs are shown in the inset. The resistance is normalized to the resistance $R_T$ just before the transition.

Figure 2

(a) The dependence of ${\mathrm{Log}}_{10}(R)$ on $1/T$ at several pressures. (b) The transition from semiconducting to metallic-type resistance behavior in NiO at $\sim 240\mathrm{GPa}$. (c) The evolution of optical reflectivity spectra of NiO.The spectra are shifted along the vertical; the zero reflectivity level is shown by dashed lines. The reflectivity spectrum at ambient pressure (0 GPa) is from the work of Newman and Chrenko [36]. (d) The pressure behavior of the energy gap $E_g$ in comparison with the behavior of the thermal-activation energy estimated at room temperature. The model calculation [32] of $E_g$ [using Eq. (2)] is shown as solid lines. The pressure dependence of the thermal-activation energy $E_{\mathrm{ac}}$ is multiplied by a factor of 20 for better comparison. The line drawn through the experimental activation energy points is guide to the eye.