Table of Contents
Fetching ...

Imaging of Gate-Controlled Suppression of Superconductivity via the Meissner Effect

P. J. Scheidegger, K. J. Knapp, U. Ognjanovic, L. Ruf, S. Diesch, E. Scheer, A. Di Bernardo, C. L. Degen

Abstract

It was recently discovered that supercurrents flowing through thin superconducting nanowires can be quenched by a gate voltage. This gate control of supercurrents, known as the GCS effect, could enable superconducting transistor logic. Here, we report that the GCS also manifests in a suppression of Meissner screening, establishing the phenomenon as a genuine feature of superconductivity that is not restricted to transport. Using a scanning nitrogen-vacancy magnetometer at sub-Kelvin temperatures, we image the nanoscale spatial region of GCS suppression in micron-size niobium islands. Our observations are compatible with a microscopic hot-spot model of quasiparticle generation and diffusion, and in conflict with other candidate mechanisms such as Joule heating or an electric field effect. Our work introduces an alternative means for studying quasiparticle dynamics in superconducting nanostructures, and showcases the power of local imaging techniques for understanding emergent condensed matter phenomena.

Imaging of Gate-Controlled Suppression of Superconductivity via the Meissner Effect

Abstract

It was recently discovered that supercurrents flowing through thin superconducting nanowires can be quenched by a gate voltage. This gate control of supercurrents, known as the GCS effect, could enable superconducting transistor logic. Here, we report that the GCS also manifests in a suppression of Meissner screening, establishing the phenomenon as a genuine feature of superconductivity that is not restricted to transport. Using a scanning nitrogen-vacancy magnetometer at sub-Kelvin temperatures, we image the nanoscale spatial region of GCS suppression in micron-size niobium islands. Our observations are compatible with a microscopic hot-spot model of quasiparticle generation and diffusion, and in conflict with other candidate mechanisms such as Joule heating or an electric field effect. Our work introduces an alternative means for studying quasiparticle dynamics in superconducting nanostructures, and showcases the power of local imaging techniques for understanding emergent condensed matter phenomena.

Paper Structure

This paper contains 6 sections, 3 equations, 4 figures.

Figures (4)

  • Figure 1: Experimental setup and superconducting devices.a, Sketch of the scanning NV magnetometer and superconducting sample. We image the stray field above the superconducting Nb device (dark blue) by scanning with the diamond NV probe (red) at constant height ($z_0\sim 100\,\mathrm{nm}$). b, False-colored scanning electron micrograph of a device identical to D2a and D2b. The superconducting Nb structure (blue) sits on top of a SiO$_2$ substrate (gray). A voltage $V_\mathrm{G}$ is applied between island and gate in constant-current mode (leakage current $I_\mathrm{L}$). Scale bar, $1\,\mathrm{\,\mu{\rm m}}$. c, $I_\mathrm{L}/V_\mathrm{G}$ curves measured on device D2b. Multiple curves (colors) are acquired over time, showing significant variation in the onset voltages of $I_\mathrm{L}$. The triangles mark the onset voltages $V_\mathrm{G}^\mathrm{ON}$ defined by the condition $P = I_\mathrm{L}V_\mathrm{G}^\mathrm{ON} = 50\,\mathrm{nW}$.
  • Figure 2: Observation of gate-controlled suppression of the Meissner effect.a, Principle of Meissner screening in an out-of-plane magnetic field $B_0$. The stray field lines (purple) are partially expelled from the thin Nb film (blue). The green dashed line indicates the scan height $z_0$. b, Model calculation of stray field at height $z_0$. Triangles mark the field maxima that approximately coincide with the edges of the superconducting film. The dashed line is the applied background field $B_0$. c, Experimentally measured stray field maps plotted for increasing leakage power $P$. Blue regions of low field reflect the diamagnetic Meissner screening by the sample. Note the presence of superconducting vortices in the first panel (red $B_z$ maxima) that are absent in the later panels taken at higher $P$. $B_0 = 1\,\mathrm{mT}$. d, Corresponding simulated magnetic field maps, assuming a circular suppression of superconductivity with radius $r_0$ centered at the gate apex and no vortices. Simulations are performed using the SuperScreen software package (Ref. bishopvanhorn2022 and Methods). Scale bars in c,d, $1\,\mathrm{\,\mu{\rm m}}$.
  • Figure 3: Power-dependence of GCS effect and hot-spot model of suppressed superconductivity.a, Suppression radius $r_0$ plotted as a function of gate power $P$, for positive and negative gate polarities (open and closed circles, respectively) and two bath temperatures (colors). The dashed line is a fit to Eq. (\ref{['eq:ro']}). Data are from device D3, gate G1. b, Data of a plotted against gate voltage $|V_\mathrm{G}|$ showing only weak correlation thereby ruling out a direct field effect. c, Hot-spot model. Non-equilibrium particles (black dots), generated near the gate apex where the current density is highest, diffuse through the SiO$_2$ substrate. In the Nb film, they break up Cooper pairs creating quasiparticles (blue dots), turning the superconductor (SC) into a normal conductor (NC) within the critical radius $r_0$.
  • Figure 4: Suppression of superconductivity for various gate geometries.a, Scanning electron micrograph of device D3. Scale bar, $5\,\mathrm{\,\mu{\rm m}}$. b-e, Electrode configurations. The top row shows a FEM simulation of the electric field strength at $V_\mathrm{G}=10\,\mathrm{V}$. The middle row shows the magnetometry map ($B_x$) for zero gate voltage ($P=0$). The bottom row shows the corresponding magnetometry map for non-zero gate power ($5\,\mathrm{\,\mu{\rm W}}$ in b, $2\,\mathrm{\,\mu{\rm W}}$ in c-e). Dotted contours indicate approximate boundaries of the Nb island and gates (horizontally shifted to align with the stray field edge for clarity). Arrows indicate suppression with applied power. Note that the stray field is incompletely screened by the superconductor, presumably due to the presence of vortices and non-superconducting grains. This leads to regions of negative $B_x$ (blue in b,c) or positive $B_x$ (red in d,e) above the Nb island, respectively. See Fig. S5 for an overview scan of the entire island. Scale bars, $1\,\mathrm{\,\mu{\rm m}}$.