In this talk, I will review the basic types of Josephson qubits, discuss the IBM Josephson flux qubit extensively, and review the some of the systems issues in making a quantum computer using Josephson technology.

Our qubit consists of three Josephson junctions and three loops coupled to a fixed-length superconducting transmission line. The bare qubit has two control parameters, the flux and the control flux. This allows the qubit to have a tunable difference frequency between the ground and first excited states and at the same time to be biased at a degenerate point with respect to the flux parameter.  This condition can be met for a wide range of junction critical currents. This flexibility of our structure is a very desirable property for a scalable qubit. To stabilize the operation of our qubit and increase its coherence time, we couple the bare qubit to the lowest mode of a superconducting transmission line, which we model as a harmonic oscillator. Using harmonic oscillator stabilization and pulsed dc operation, we have observed Larmor oscillations with a single shot visibility of 90% and a coherence time of 100 ns. In another qubit the visibility was 60% and there was no measurable visibility reduction after 35 ns.

Our system has several unique features that offer good prospects for scalability, compared with other Josephson qubits. The transmission line frequency, depending only on one geometric parameter, the length, can be fixed with very high precision. Its insensitivity to the magnetic flux environment when biased at the oscillator stabilized operating point will greatly diminish the degree of unintended couplings between qubits. The gradiometric design of the qubit can be exploited to further reduce crosstalk during gate operations, when the qubit is moved off the oscillator stabilized operating point. These issues of scalability are the most important ones to be explored in the next round of experiments on this qubit system.

Last updated on Monday, 20 February 2006 by Victor Yakovenko