Discovery of an electrically-controllable superconducting memory effect

Abstract

If a computer could be assembled from superconducting components, the energy efficiency would far surpass that of conventional electronics. Historic research efforts towards this goal yielded pivotal breakthroughs in the development and discovery of scanning tunnelling microscopy and high temperature superconductivity. Although recent strides have been taken in advancing superconducting diode and switching technologies, harnessing read/writeable memory functionality in superconducting platforms has remained challenging. Here we show that bulk single crystal specimens of the triplet superconductor candidate uranium ditelluride (UTe$_2$) possess such properties. Upon applying a magnetic field to access an intermediate regime straddling two distinct superconducting phases, we find that direct current pulses can push the material in and out of a metastable state possessing an enhanced critical current $J_c$. This switching is controllable by the strength and duration of the stimuli, with the system 'remembering' whether it is in the high or low $J_c$ state for extended periods. We interpret this to be due to competition between two distinct vortex species, which can be perturbatively pushed into a non-equilibrium high-disorder configuration with stronger pinning forces and thus higher $J_c$. Rather than requiring proximate magnetic or semiconducting interfaces, this memory functionality appears to be an intrinsic property of UTe$_2$ rooted in the superconducting order itself. Our findings underscore the rich complexity of putatively $p$-wave vortex matter and demonstrate the viability of engineering a new class of superconducting memory elements with ultralow-power switching, which could be transformative for cryogenic computing and quantum hardware.

Figures (14)

• Figure 1: Electrical switching of an intrinsic superconducting memory effect. a, The low temperature phase diagram of UTe$_2$ for magnetic field $B$ applied along the hard crystallographic $\hat{b}$ direction tony2024enhanced. The region of phase space where memory effects are manifested is coloured purple, lying between two distinct superconducting phases (SC1 and SC2). b, The measured voltage $V$ across a UTe$_2$ single crystal as a function of dc current density $J$. Panels are arranged chronologically from left to right, with the insets (and colour coding) showing how $J$ was modulated as a function of time. When $J$ is ramped continuously, the profile of $V(J)$ is in its equilibrium state and no hysteresis is observed. By contrast, after a discontinuous change in $J$ from a large magnitude abruptly to zero (labelled 'pump'), hysteresis in $V(J)$ is observed, plotted in red and orange. Smoothly sweeping the current from a large negative value to zero ('erase') then resets the system back to the original $V(J)$ profile with lower $J_c$. All data were acquired at 50 mK with $B \parallel \hat{b}$. c, Data collected on sample S2 at 7 T and 50 mK. Here we plot just two traces, to highlight the hysteretic out-of-equilibrium form of $V(J)$ in the memory state. d, The lower panel shows $V$ measured as a function of time for modulations of $J$ depicted in the upper panel, cycling through a sequence of perturbation, measurement, and erasure (resetting) protocols, as described in e. By sitting at this point in the $J$--$V$ curve, successively switching in and out of the memory state yields a voltage versus time profile akin to supercurrent rectification by the superconducting diode effect SCDiode_Nature2020.
• Figure 2: Three tuning parameters to control the memory effect. Hysteretic response profiles for modulations of an applied excitation's a amplitude, b duration and c relaxation rate. In each case, the greater the perturbation to the system -- be it a larger amplitude $J_{\text{ex}}$, longer duration $\Delta t$ or more rapid relaxation rate $\alpha$ -- the greater the resulting hysteresis loop in $V(J)$, up to some saturation value (see Extended Data Fig. 1 for further analysis). For the heatmaps, $\Delta J$ is calculated as the change in $J$ at fixed values of $V$. The regions of highest $\Delta J$, coloured yellow, correspond to the strongest memory effect.
• Figure 3: Temperature dependence and modelling of the memory effect. a,$V(J)$ at incremental temperatures as indicated. The inset shows the sequence of current modulations. While large hysteresis loops are recorded at low temperature, these have closed by 650 mK. At 1.0 K, the curvature of $V(J)$ is markedly different compared to the equilibrium curve at 50 mK (see also Fig. S7). b, Heatmaps of $\Delta J$ at incremental temperatures for modulations of the excitation amplitude $J_{\text{ex}}$ up to 20 A cm$^{-2}$ measured on sample S2 at 14 T. Again, yellow colouring indicates the strongest memory effect, which sharply diminishes at elevated temperatures. c, Simulated $J-V$ curves for the two types of current pulses, which correspond to the equilibrium and non-equilibrium switching of current, see Methods. The memory state is when the system is out-of-equilibrium. The calculated curves are in good qualitative agreement with the measured low temperature data.
• Figure 4: Mapping the memory region between SC1 and SC2. a, Schematic phase landscape of UTe$_2$ for rotations of $B$ by an angle $\theta$ from $\hat{b}$ towards $\hat{c}$. The memory region (coloured purple) -- identified by nonzero $\Delta J$ at low $T$ -- is located at the intersection of the SC1 and SC2 domains. We refer to this as SC1.5. b, The equilibrium profile of $V(B)$ at 50 mK for successive $J$ values, as indicated by the colour scale. An anomalous non-monotonic kink in $V(B)$ at high $J$ is observed at the boundary of the memory region. c, Sequential modulations of $J$ at different fields for $B \parallel \hat{b}$ (0$\degree$) and d, modulations of $J$ at different magnetic field tilt angles $\theta$ at $B =$ 15 T. The inset defines the measurement protocol. e, Evolution of $\Delta J$ as a function of $B$ at 0$\degree$ and 15$\degree$ and f, as a function of $\theta$ at 15 T and 29 T. g, Equilibrium $V(\theta)$ for $J$ = 2.5 Acm$^{-2}$, $B=$ 15 T, $T=$ 50 mK measured on sample S2. The transition from the memory region into SC1 is marked by a sudden jump in $V$ at around $\theta =$ 20$\degree$, indicating a change in the vortex properties. This angle is where, in panel (f), $\Delta J$ reaches zero within the resolution of the measurement. h, Magnetic field sweeps of the effective resistivity $\rho$ for incremental $J$ values as indicated. The data in this panel were acquired by low frequency ac measurements, whereas all other data in this article are from dc measurements (see Methods). While zero resistance is observed over all $B$ in the low $J$ limit, at higher $J$ a complex profile of flux-flow behaviour is exhibited.
• Figure : Extended Data Fig. 1 | Saturation of the memory effect. a, Amplitude, b, duration and c, relaxation-rate tuning of the memory effect of UTe$_2$. For sufficiently large perturbative amplitudes $J_{\text{ex}}$ or durations $\Delta t$, the memory effect approaches a saturation value. $\Delta J$ is calculated here at a nominal fixed value of $V =$ 1 $\upmu$V as indicated by the dashed lines. All data were acquired on sample S2 at 14 T and 50 mK.