Metal Oxide Resistive Switching: Evolution of the Density of States Across the Metal-Insulator Transition

A. Mottaghizadeh, Q. Yu, P. L. Lang, A. Zimmers, and H. Aubin

Phys. Rev. Lett. 112, 066803 – Published 14 February 2014

Abstract

We report the study of gold- $SrTi O_{3}$ (STO)-gold memristors where the doping concentration in STO can be fine-tuned through electric field migration of oxygen vacancies. In this tunnel junction device, the evolution of the density of states (DOS) can be followed continuously across the metal-insulator transition (MIT). At very low dopant concentration, the junction displays characteristic signatures of discrete dopant levels. As the dopant concentration increases, the semiconductor band gap fills in but a soft Coulomb gap remains. At even higher doping, a transition to a metallic state occurs where the DOS at the Fermi level becomes finite and Altshuler-Aronov corrections to the DOS are observed. At the critical point of the MIT, the DOS scales linearly with energy $N (ϵ) \sim ϵ$ , the possible signature of multifractality.

Received 17 September 2013

DOI:https://doi.org/10.1103/PhysRevLett.112.066803

Authors & Affiliations

A. Mottaghizadeh, Q. Yu, P. L. Lang, A. Zimmers^*, and H. Aubin^†

Laboratoire de Physique et d’Etude des Matériaux, UMR 8213, ESPCI-ParisTech-CNRS-UPMC, 10 rue Vauquelin, 75231 Paris, France

^*Alexandre.Zimmers@espci.fr
^†Herve.Aubin@espci.fr

Article Text (Subscription Required)

Click to Expand

Supplemental Material (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 112, Iss. 6 — 14 February 2014

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
(a) False color SEM image of gold electrodes deposited on the STO substrate. Applying a voltage between the electrodes leads to the formation of a conducting filament that is identified by a switching event in the $I V$ curve shown in panel (c). (b) Perovskite structure of STO crystal. (d) Two-wire resistance of two different conducting filaments measured at low temperature.
Reuse & Permissions
Figure 2
(a)–(f) Seven hundred fifty differential conductance curves, indexed by $i = 0, \dots, n$ ( $n = 749$ ), measured at low temperature $T \sim 260 mK$ , are shown as a function of drain voltage. Before measurement of each curve, a voltage pulse of 1 ms is applied. For each curve, the amplitude of the pulse is indicated on the panel to the right of each differential curve panel. (a) The first differential curve measured ( $i = 0$ ) shows that the junction is highly conducting. A voltage pulse of 5 V turns the junction insulating, as shown by the curve ( $i = 1$ ). Then, applying many negative voltage pulses, from $- 2$ to $- 3 V$ , turns the junction conducting again as shown by the increasing value of the differential conductance between each curve. Between the panel (a) curves and panel (b) curves, one positive voltage pulse has been applied, which turns the junction insulating again. Applying negative voltage pulses leads again to a progressive increase in the differential conductance. This protocol has been repeated six times [(a)–(f)]. The transition from the insulating state to the conducting state can be reproducibly repeated as shown in panel (g), which shows that all the differential conductance curves, $i = 0, \dots, n$ ( $n = 749$ ), plotted separately in panels (a) to (f), can be superimposed on each other. Panel (h) shows that the zero bias conductance, measured at $V_{drain} = 0 V$ , increases by 5 orders of magnitude from the insulating to the conducting sample. Arrows labeled “fig3abc,” “fig3.d,” and “fig3.e,” indicate the zero bias conductance for the junctions whose DOS is shown in Fig. 3.
Reuse & Permissions
Figure 3
DOS at low bias voltage ( $- 0.1$ to 0.1 V) at various steps of the transition from the insulator to the metal. The curves have been normalized at $V_{drain} = 0.1 V$ and displaced for clarity. The horizontal dashed lines indicate the level of zero conductance. The curve labeled (a) is measured at very low doping concentration; it shows a hard gap and discrete electronic levels. At high carrier concentration (b), the DOS has a quadratic energy dependence, $N (ϵ) \sim ϵ^{2}$ (Efros and Shklovskii [25]), as shown by the dotted line. At very high carrier concentration [(d),(e)], the DOS becomes finite at zero energy with a cusp that can be fitted by the Altshuler-Aronov law $N (ϵ) = N (0) [1 + {| ϵ / E_{T} |}^{1 / 2}]$ , as shown by the dotted lines. $E_{T} \sim 0.1 eV$ for curve (e) and $E_{T} \sim 0.003 eV$ for curve (d). Finally, just at the critical point of the MIT, the DOS has a linear dependence with energy, $N (ϵ) \sim ϵ$ (dotted line).
Reuse & Permissions
Figure 4
(a) DOS at low bias voltage (18–100 mV) in the low doping regime showing a sharp discrete doping level. (b) Zoom (18–28 mV) on the discrete electronic level. The peak splits in two peaks due to Zeeman splitting. The magnitude of the Zeeman splitting suggests that these electronic levels arise from a $d$ ( $ℓ = 2$ ) orbital level, shown in panel (c).
Reuse & Permissions

Physical Review Letters