An InAsSb surface quantum well with in-situ deposited Nb as a platform for semiconductor-superconductor hybrid devices

Abstract

We present a novel semiconductor-superconductor hybrid material based on a molecular beam epitaxially grown InAsSb surface quantum well with an in-situ deposited Nb top layer. Relative to conventional Al-InAs based systems, the InAsSb surface quantum well offers a lower effective mass and stronger spin-orbit interaction, while the Nb layer has a higher critical temperature and a larger critical magnetic field. The in-situ deposition of the Nb results in a high-quality interface that enables strong coupling to the InAsSb quantum well. Transport measurements on Josephson junctions reveal an induced superconducting gap of 1.3 meV. Furthermore, a planar asymmetric SQUID is realized, exhibiting gate-tunable superimposed oscillations originating from both the individual Josephson junction and the full SQUID loop. The large induced superconducting gap combined with strong spin-orbit interaction position this material as an attractive platform for experiments exploring gate-tunable superconductivity and topological superconducting devices.

Figures (9)

• Figure 1: (a) STEM overview of the top part of the heterostructure, showing the InAsSb/InAlSb surface QW and the Nb superconducting layers. Schematic sketches of the layers and electron confinement in the surface QW are shown on the left and right of the figure, respectively. (b) Zoom in of the framed area in (a) with higher resolution. (c) XRD measurements along the (004) direction to determine the components of each layer. The x-axis has been converted from angle to lattice constant.
• Figure 2: (a) False-color SEM image of the device indicating the source-drain electrodes, top-gate as well as the measured AC voltage and DC voltage. (b) The measured resistance $R$ as a function of bias current and gate voltage. (c) Temperature dependent measurement of $R-I_{\rm{SD}}$ diagrams. Each trace is offset by 50 $\Omega$ for clarity. The peaks associated to the first three MAR resonances are indicated in the figure. (d) $R$ as a function of measured $V_{\rm{SD}}$ at 15 mK. Indicated are the MAR peaks. The inset shows the first six MAR peaks plotted as a function of 1/$N_{\rm MAR}$. The dashed line represents the fit associated with an induced gap of 1.34 meV. (e) The temperature dependence of $V_{\rm{SD}}$ where MAR peaks appear. $V_{\rm{SD}}$ for MAR with $N_{\rm MAR}$=1 (red) ,2 (blue), and 3 (green) are plotted (dots) and fitted with equation \ref{['BCSstar']} (lines). (f) The switching current of the device plotted as a function of temperature. The solid line is the fit of the combined model for the escape mechanism of the JJ. The dashed line indicates the crossover temperature between two regimes of different dominating mechanisms.
• Figure 3: (a) A false-colored SEM image of the a representative SQUID device on the same chip before gate deposition. The measurement circuit configuration is illustrated. (b) Measurement on the large JJ while the small JJ on the SQUID is pinched off. $R$ is plotted as a function of applied magnetic field $B$ and source-drain bias current $I_{\rm{{SD}}}$. (c) Same as in b but for the small JJ. (d) The interference pattern observed with both junctions fully open. (e) A high resolution measurement of a small area of $B$, indicating the abrupt changes in switching current. (f)-(h) The interference patterns of the SQUID for a narrow region of magnetic field with the small JJ always open, but the large JJ completely pinched off (f), partially closed (g), and completely open (h).
• Figure 4: XRD measurement results for QWs with a graded buffer (a) and an InAlSb superlattice buffer (b), with the data converted to out-of-plane lattice constants.
• Figure 5: Low-temperature DC characteristics of the superconducting film. (a) $T_{\rm{c}}$ measurement. (b )$B_{\rm{c2}}$ as a function of temperature (b). The value of $B_{\rm{c2}}$ at $T=0$ K is extrapolated from the fit (dashed line).