Partenaires

CNRS
Universite Pierre et Marie Curie - Paris Univertsitas


Rechercher

Sur ce site


Accueil >

Superconducting qubits

1) Topologically protected qubits. With Lev Ioffe (Rutgers University and now at LPTHE since July 2012), we have conceived Josephson junction arrays which simulate the famous Kitaev toric code model in its low-energy limit. We have shown that such arrays have a doubly-degenerate ground-state which is very well protected from the influence of local noises. These theoretical ideas have been partly confirmed by experiments done by M. Gershenson at Rutgers University. We have optimized the design of these circuits in order to be able to correct the effects of both charge and fux noises, and we have proposed various protocoles to measure and manipulate such unsual qubits. I have also participated in the theoretical modelling of an experience done in Grenoble which has evidenced quantum interference effects in the dynamics of vortices in a chain of Josephson junctions.

2) Origin of noise in superconducting circuits. The general goal of Lara Faoro’s research work is to understand the microscopic origin of the different sources of noise (and their temporal and spatial correlations) in superconducting circuits at low temperatures, in order to identify strategy to control them and to improve the performance of superconducting qubits.
Briefly, since 2007 the most significant achievements are :
- We have identi-ed a novel microscopic origin of the charge and critical current noise : Kondo-like traps (KT) formed by localized spins at the Superconductor Insulator (SI) interfaces that competes with more conventional charged TLS to dominate the observed noise in superconducting structures.
While the TLS noise exists also in the normal state, the KT noise is speci-c to the superconductors. We showed that the KT mechanism might be responsible for a very slow relaxation of quasiparticles that is the root of the "quasiparticle poisoning" because it leads to the formation of the strongly-localized states of quasiparticles at the SI interface.
- We made signi-cant progress towards the understanding of excess low frequency magnetic noise in dc-SQUIDs. Speci-cally, we proposed a theoretical picture consistent with data byWellstood et al. in which the noise it is due to the spins at the SI interface coupled via RKKY interaction. In contrast to the alternative models this mechanism explains all puzzling features of the fux noise : its apparent temperature independence down to 20mK, its persistence to at least 20 MHz and the rough SQUID loop area independence. This mechanism generates roughly 1/f noise in a broad frequency range but the details of its spectrum at very high and very low frequencies need to be computed for a better comparison with the data. To this aim, very recently we showed that a system of weakly interacting disordered spin systems can generate high frequency noise. We developed a technique to compute igh-frequency asymptotics of spin correlators and found that the high frequency spin correlators decreases exponentially at high frequencies, SS exp(-w t*) and we computed the characteristic time, t*, of this dependence.
- Prompted by a number of recent experiments performed on microwave high quality superconducting coplanar waveguide resonators, we argued that the monochromatic radiation applied to the resonator does not saturate the TLS located at the interface oxide surfaces of the resonator (as it is predicted by the conventional theory of microwave absorption of TLS in glasses) and we suggested that this could be an indication of the importance of TLS-TLS interactions. We then estimated the microwave loss due to interacting TLS and we showed that the interactions between TLS lead to a drift of their energies that result in a much slower, logarithmic dependence of their absorption on the radiation power in perfect agreement with the data.