Search

We describe the design of an inductor-capacitor (LC) network to increase the isolation of a dc SQUID phase qubit from its current bias leads and thereby increase the relaxation time T1 and coherence time T2. One junction in the SQUID acts as an ideal phase qubit while the second junction and the SQUID loop inductance act as a broadband inductive filter to isolate the first junction from the current bias leads. The LC isolation network provides an additional isolation factor at the junction plasma frequency and allows flexibility in the choice of SQUID parameters. Our thin-film on-chip LC isolation network has a 10 nH inductor and an 80 pF capacitor. The combination of the broadband filter and LC filter provides a maximum nominal isolation factor of about 108 at a qubit plasma frequency of about 6.7 GHz. To reduce dielectric loss and two level systems in the qubit junction, we use a relatively small area (4 mum2) Al/AlOx/Al qubit junction built on sapphire and add an external capacitor with 100 nm thick SiNx dielectric layers. Measurements revealed Rabi oscillations with an envelope decay time constant of about 42 ns, and an energy relaxation time of 32 ns, consistent with a loss tangent tan(delta)=7times10-4 in the SiNx.

We report measurements of spectroscopic linewidth and Rabi oscillations in three thin-film dc SQUID phase qubits. One device had a single-turn Al loop, the second had a 6-turn Nb loop, and the third was a first order gradiometer formed from 6-turn wound and counter-wound Nb coils to provide isolation from spatially uniform flux noise. In the 6 - 7.2 GHz range, the spectroscopic coherence times for the gradiometer varied from 4 ns to 8 ns, about the same as for the other devices (4 to 10 ns). The time constant for decay of Rabi oscillations was significantly longer in the single-turn Al device (20 to 30 ns) than either of the Nb devices (10 to 15 ns). These results imply that spatially uniform flux noise is not the main source of decoherence or inhomogenous broadening in these devices.

We have used 2 ns pulses of current to read out the states of a Nb/AlOx/Nb dc SQUID phase qubit at 25 mK. Plotting the number of escape events that occur versus the size of the current pulse reveals a series of steps corresponding to the occupancy of different energy levels. After calibrating the measurement pulses, we fit these steps to determine the population in each level. Rabi oscillations were viewed using this technique. The single-shot measurement fidelity was theoretically analysed and the optimal single pulse in our device was determined.

We present measurements of the spectroscopic coherence time, T2* of a Nb-AlOx-Nb dc SQUID phase qubit. T2* is calculated from the spectroscopic width of the |0rang to |1rang transition measured at 25 mK. We find that T2*increases at higher transition frequencies corresponding to measurements deeper in the potential well. The measured are consistent with root mean square current noise saturating at 4 nA.