Oleg Astafiev: Phase-slip qubit realization efforts

Oleg Astafiev (NEC Green Inovation Research Labs, Tsukuba) reported on very recent (last week!) experiments that promise to resolve a long-standing issue: observation of coherent dynamics of quantum phase slips (QPS) in narrow superconducting wires.

The device, a so-called phase-slip qubit, containes a superconducting loop with a narrow portion (“wire”). The wires, 50-130 nm wide and ~ 1 mm long, were fabricated from highly resistive 20-nm-thick InOx films prepared by Danny Shahar’s group at Weizmann University. The superconducting loops which contained these wires were coupled either directly to a coplanar waveguide (CPW) transmission line, or to a CPW resonator, and the phase of the transmitted signal was measured over a frequency range of 6-12 GHz as a function of the magnetic flux treading the loop. The coherent QPS dynamics leads to formation of two low-energy levels in the spectrum of the wire separated by a gap which is exponentially sensitive to the wire resistance. One of the experimental challenges is to fabricate a wire whose resistance at the superconducting coherence length scale would be comparable to the quantum resistance. InOx wires were up to this task. According to Lev Ioffe (Rutgers), who provided theoretical support to this work, InOx offers an additional important advantage: the superconducting gap in these films is not suppressed even at the sheet resistances as high as 6 kOhm/square, which translates into freezout of quasiparticle excitations at sufficiently low temperatures (<100 mK).

Both experiments, with loops coupled to a CPW resonator and directly to a CPW transmission line, provided an evidence of the resonance excitation of the two-level system formed by the coherent QPS dynamics. The corresponding resonance frequency varied periodically with the magnetic flux threading the loop. In particular, the avoided level crossing was observed when the resonance frequencies of the two-level system and the CPW resonator were brought together by the external magnetic field. The second-tone excitation of the TLS allowed tracing the resonance over a relatively wide range of magnetic fields. It was varified that the dependence of the resonance frequency on the magnetic flux in the loop is consistent with the kinetic inductance of a uniform InOx wire (this observation eliminates the possibility of formation of Josephson weak links in the wire). The resonance frequencies observed at a half-integer magnetic flux through the loop were 2.3 GHz and 5.8 GHz for two studied devices. Interestingly, the observed resonances were relatively broad: e.g. the 2.3 GHz resonance had a width 0f ~ 0.4 GHz. The sorce of the dissipation responsible for this relatively low Q-factor remains to be identified.

Blogged by Michael Gershenson