Exploring the energy landscape of resistive switching in antiferromagnetic $\mathrm{S}{\mathrm{r}}_3\mathrm{I}{\mathrm{r}}_2{\mathrm{O}}_7$

We study the resistive switching triggered by an applied electrical bias in the antiferromagnetic Mott insulator $\mathrm{S}{\mathrm{r}}_3\mathrm{I}{\mathrm{r}}_2{\mathrm{O}}_7$. The switching was previously associated with an electric-field-driven structural transition. Here we use time-resolved measurements to probe the thermal activation behavior of the switching process and acquire information about the energy barrier associated with the transition. We quantify the changes in the energy-barrier height with respect to the applied bias and find a linear decrease of the barrier with increasing bias. Our observations support the potential of antiferromagnetic transition-metal oxides for spintronic applications.

Figure 1

(a) Resistive switching in $R\left(I\right)$ characteristic of $\mathrm{S}{\mathrm{r}}_3\mathrm{I}{\mathrm{r}}_2{\mathrm{O}}_7$ ($T=78.6\mathrm{K}$). Black (gray) trace represents the $I$ sweep from $-15\mathrm{mA}$ to $+15\mathrm{mA}$ (from $+15\mathrm{mA}$ to $-15\mathrm{mA}$). Arrows indicate switching events. Two switching events to a higher resistance for increasing bias (at $I_c=\pm 8.5$ and $\pm 12.7\mathrm{mA}$) and one to a lower resistance for decreasing bias are clearly visible. The inset shows the temperature dependence of the critical current associated with the first switching event at $I_c=8.5\mathrm{mA}$.

Figure 2

A time trace of the process of switching from low bias to the “first high-resistance” state. The vertical axis plots the voltage across the sample captured with an oscilloscope where the first abrupt increase is caused by a sudden rise in applied current. After a time τ, the switching occurs, resulting in a second abrupt rise in measured voltage due to an increase in sample resistance, indicating that switching has occurred. This measurement was repeated 600 times. The inset shows a histogram of the measured delay-time values and a corresponding Poisson fit.

Figure 3

Average delay time between application of current step and resistive switching as a function of the final bias measured at 77 K. All switching events were executed from a beginning bias of 8.0 mA. The solid line shows exponential fit. The inset depicts the change in the barrier height, $\mathrm{\Delta }$, as a function of the final bias with a linear fit corresponding to a modulation of 172 meV/mA.

Figure 4

(a) Temperature effects of the average delay measurement shown in Fig. 3. Temperatures range from 77 K at the rightmost curve to 80 K at the leftmost curve. Black and white symbols serve to separate data associated with neighboring temperatures. (b) $\mathrm{\Delta }$ calculated as a function of temperature $({\tau }_0=1\mathrm{ns})$. (c) The change in the current, $I_{\mathrm{eff}}$, associated with $\mathrm{\Delta }$ equal to 60 meV, as a function of temperature with a linear fit corresponding to a variation of $110\pm 1\mu \mathrm{A}/\mathrm{K}$.