Near-surface InAs two-dimensional electron gas on a GaAs substrate: Characterization and superconducting proximity effect

We have studied a near-surface two-dimensional electron gas based on an InAs quantum well on a GaAs substrate. In devices without a dielectric layer we estimated large electron mobilities on the order of ${10}^5{\mathrm{cm}}^2/\mathrm{Vs}$. We have observed quantized conductance in a quantum point contact, and determined the $g$ factor. Using samples with an epitaxial Al layer, we defined multiple Josephson junctions and found the critical current to be gate tunable. Based on multiple Andreev reflections the semiconductor-superconductor interface is transparent, with an induced gap of $125\mathrm{\mu }\mathrm{eV}$. Our results suggest that this InAs system is a viable platform for use in hybrid topological superconductor devices.

Figure 1

(a) Schematic of the layer structure. (b) Poisson-Schrödinger simulation of the conduction and valence band edges relative to the Fermi level ($E_c,E_v$, black) as a function of depth measured from the surface, and the simulated 3D electron density $n_{3\mathrm{D}}$ (blue). Dotted lines represent the layer boundaries from (a), without Al. For the calculation, a background density [17] of $4\times {10}^{16}{\mathrm{cm}}^{-3}$ was assumed in the ${\mathrm{In}}_{0.81}{\mathrm{Ga}}_{0.19}\mathrm{As}$ and ${\mathrm{In}}_{0.81}{\mathrm{Al}}_{0.19}\mathrm{As}$ layers. (c) Optical dark field and (d) AFM image of the wafer surface. The $x$ and $y$ axes are parallel with the $\left[\overline{1}10\right]$ and [110] crystallographic directions, respectively. The RMS surface roughness based on the AFM image is $3.1\mathrm{nm}$.

Figure 2

Magnetotransport on a nongated Hall bar (sample No. 04). (a) Longitudinal ($R_{xx}$, black) and Hall ($R_H$, red) resistance as a function of the out-of-plane magnetic field $B$ at $T=1.5$ K. The Hall density is $n=4.95\times {10}^{11}{\mathrm{cm}}^{-2}$, the mobility $\mu =86000{\mathrm{cm}}^2/\mathrm{Vs}$. Purple dashed lines highlight the expected plateau positions at integer Landau level filling factors $\nu$. Inset: optical photo of a similar Hall bar. (b) Longitudinal resistance of the same device at a series of temperatures.

Figure 3

Magnetotransport on top-gated Hall bar No. 11. (a) Resistance map as a function of $V_{\mathrm{TG}}$ and $B$ at 4 K. The dashed lines highlight crossings of the Landau levels of the first (green) and second (red) subbands. (b) The corresponding map of the Hall conductivity. Orange dots indicate the resistance maxima in (a), while white numbers are filling factors of the subbands on quantum Hall plateaus: ${\nu }_1+{\nu }_2$. (c) Landau level density of states (DOS) illustrations for the three stars on the lower green dashed line in (a) and (b), i.e., keeping ${\nu }_1=3.5$ constant and $E_F=E_{1+}^{\left(1\right)}$ (highlighted by a black dashed line in all subpanels).

Figure 4

Characterization of top-gated Hall bar No. 14. (a) Density $n$ (black), conductivity $\sigma$ (red), and mobility $\mu$ (blue) at 1.5 K. Single data points are values before ALD, while solid or dashed lines are measurements as a function of top-gate voltage $V_{\mathrm{TG}}$ after ALD and gate electrode fabrication. The abscissa of the single points was chosen as the conjunction of the density data. (b) Effective mass $m$ and Dingle temperature $T_D$ as a function of density $n$. Single data points are values before ALD.

Figure 5

Transport in a quantum point contact. (a) Four-point conductance $G_{4p}$ as a function of split-gate voltages $V_{\mathrm{sg}1,2}$ at 1.5 K and 0 T. The gray curve is the conductance along the diagonal $V_{\mathrm{sg}1}=V_{\mathrm{sg}2}$ (see the right vertical axis). Insets: optical microscope image of the sample (left), and scanning electron micrograph of a similar pair of split gates (right). (b) An equivalent map of $G_{4p}(V_{\mathrm{sg}1},V_{\mathrm{sg}2})$ at $B=8$ T. The gray curve is a diagonal cut. (c) Bias spectroscopy along the diagonal of the gate-gate map at 8 T, showing the numerical derivative $d{G}_{4p}/d{V}_{\mathrm{sg}1,2}$. $V_{\mathrm{dc}\mathrm{corr}}$ is the estimated dc voltage drop on the QPC. Inset: half-size $\mathrm{\Delta }V$ of the $1e^2/h$ plateau as a function of $B$.

Figure 6

Measurements on a top-gated Josephson junction. (a) The resistance $dV/dI$ on a logarithmic scale as a function of the applied (dc) current $I$ and $V_{\mathrm{TG}}$. Inset: an example $V\left(I\right)$ curve (solid line) at $V_{\mathrm{TG}}=-3.11\mathrm{V}$. The dotted line is the fit to the high-$\left|V\right|$ part. Red and purple arrows denote the critical current $I_c$ and excess current $I_{\mathrm{exc}}$, respectively. (b) A zoom of the low-density region as highlighted by the blue rectangle in (a). Purple curves plot the estimated $I_{\mathrm{exc}}$. (c) Smoothed $dV/dI$ data plotted versus $V$ at a series of gate voltages increasing from $V_{\mathrm{TG}}=-3.1\mathrm{V}$ (purple line) with 0.05 V step size to $V_{\mathrm{TG}}=-2.5\mathrm{V}$ (dark red). The red circle is equivalent to the point highlighted in (b) at $V_{\mathrm{TG}}=-2.7\mathrm{V}$. (d) Fraunhofer pattern as a function of out-of-plane magnetic field $B$ at $V_{\mathrm{TG}}=2\mathrm{V}$.