Anomalous metallic phase and reduced critical current in superconducting nanowires due to inverse proximity effect

Abstract

Superconductor-to-metal transitions (SMTs) are key probes of mesoscopic superconductivity, but their interpretation can be complicated by device geometry and measurement conditions. Here, we study epitaxial InAs-Al nanowires and show that metallic contacts induce an inverse proximity effect (IPE), creating weak spots in the superconductor that strongly suppress the critical current and give rise to an anomalous metallic phase. Using transport measurements supported by Usadel theory, we demonstrate that this phase originates from the contact-induced weakening of superconductivity together with Joule heating, rather than intrinsic material properties. Our findings reveal an overlooked observer effect in mesoscopic superconductors and provide essential guidance for interpreting SMTs and for designing devices based on these systems.

Figures (4)

• Figure 1: Device geometry. (a) Sketch of the four-terminal geometry with inner (outer) electrical contacts separated by a distance $L$ ($L_{full}$). The width of the electrical contacts is shown as $w$. The voltage drop, $V$, and the differential resistance, $dV/dI$ are measured as a response to a DC bias current, $I$, mixed with a small AC excitation. The inset schematically depicts a superconducting weak-spot induced in the superconducting wire by inverse proximity to a normal lead. (b) Schematic of the cross-section of a full-shell InAs-Al nanowire, with inner diameter, $d_c$, and Al shell thickness, $t_s$. (c) Differential resistance as a function of bias current and magnetic field applied parallel to nanowire axis for devices contacted by normal (device A) and superconducting leads (device B). The switching current, $I_s$, and retrapping current, $I_r$, oscillate due to the Little-Parks effect. The vertical dashed lines in the lower panel mark the approximate critical field of the fabricated superconducting contacts, and the inset highlights a region of anomalous resistance. (d)$V(I)$ traces taken at $B_\parallel = 0$ mT and $B_\parallel=35$ mT, showcasing the drastic reduction of $I_s$ and emergence of anomalous resistance in device B once the electrical contacts turn normal.
• Figure 2: Regimes of inverse proximity effect in a superconducting wire. (a) Differential resistance as a function of bias current and magnetic field applied perpendicular to nanowire axis. The yellow and cyan dashed lines indicate fits to eq. (\ref{['eq:ICpower']}) and eq. (\ref{['eq:ITpower']}) respectively, with $\gamma_C = 5/2$, $\gamma_T=1$. (b) Schematics of the the distinct nanowire phases, $\mathcal{N}$, $\mathcal{N}_S$, and $\mathcal{S}$, in the $I-B_{\perp}$ parameter space. A weak-to-strong IPE crossover occurs because $\gamma_C > \gamma_T$ and $I_C(0)>I_T(0)$. (c) Illustration of the mechanism setting $I_C$, where current is limited by the reduced gap under the metallic contact. (d) Depiction of the thermal balance between Joule heating and Wiedemann-Franz cooling governing $I_T$.
• Figure 3: Effect of the magnitude of the inverse proximity effect on $I_C$ and $T_C$. (a, b) Differential resistance measurements of Devices A and C as a function of bias current and bath temperature, $T$. Cyan dashed lines displays $I_T(T) = I_T(0)\sqrt{T^2_C-T^2}$ used to extract $T_C^A$ and $T_C^C$. (c, d) Theoretically obtained estimates of $I_C/I_{C0}$ and $T_C/T_{C0}$ as a function of normal metal-superconductor interface conductance, $G_I$. Parameters are chosen to match Device A in (c) and Device C in (d).
• Figure 4: Emergence of anomalous metallic phase with $B_{\perp}$.(a)$dV/dI(I, B_{\parallel})$ measurements of device A taken with $B_{\perp}$ = 0 (top panel), 46 (middle panel) and 60 mT (bottom panel). The yellow and cyan dashed lines correspond to $I_C(B)$ and $I_T(B)$ calculated using parameters obtained from the fit in Fig. \ref{['fig:2']}(a) and $\gamma_C = 5/2$, $\gamma_T = 1$. (b)$dV/dI(I, B_{\parallel})$ near $\Phi = 0.5 \Phi_0$ with $B_{\perp}$ = 0 (left panel) and 30 mT (right panel). (c) Zero-bias $dV/dI$ taken at $\Phi = 0.5 \Phi_0$ ($B_{\parallel}$ = 90 mT, highlighted in panel (b) as a light green point) as a function of the bath temperature for various $B_{\perp}$ from 0 to 45 mT.