The researchers used nanowires of indium arsenide since the material can form Schottky barrier-free contacts with metals. This gives an interface between the nanowire and the superconducting contacts that is highly transparent to electrons - something that’s necessary for the proximity effect to occur. The effect enables Cooper pairs of electrons to leak from the superconductor into the nanowires.

The team grew the nanowires by a vapour-liquid-solid process, producing monocrystalline wires with diameters of 40-130 nm and lengths of 3-10 µm. Next the researchers transferred the nanowires to a p+ silicon substrate coated with a 250-nm thick silica layer. They used this conductive substrate as a back gate to vary the electron density in the nanowires.

Finally the team created two electrodes by e-beam lithography and e-beam evaporation of 10 nm of titanium and 120 nm of aluminium.

The resulting nanowire devices had a normal state resistance of 0.4 - 4.5 kilohms. At temperatures less than 1.1 K - the superconducting transition temperature of the aluminium-based electrodes - the proximity effect took over and a dissipationless supercurrent flowed in the nanowire. The scientists say this can be thought of as the diffusion of Cooper pairs through the nanowire section between the superconducting electrodes.

The researchers were able to use the gate voltage to adjust the size of the critical current at which the device switched from a superconductive to a resistive regime. This enabled them to switch the supercurrent on or off. So the nanowires could operate as “tunable superconducting weak links”. The researchers say that using alternative gating geometries such as local top gates or gate-around configurations would give a much stronger coupling and might enable individual control of different nanowires on the same chip.

The researchers reported their work in Science.