Quantum information processing with tunable and low-loss superconducting circuits
Abstract: The perhaps most promising platform for quantum information processing is the circuit-QED architecture based on superconducting circuits representing quantum bits. These circuits must be made with low losses so that the quantum information is retained for as long as possible. We developed fabrication processes achieving state-of-the-art coherence times of over 100 µs. We identified the primary source of loss to be parasitic two-level systems by studying fluctuations of qubit relaxation times. Using our high-coherence circuits, we implemented a quantum processor built on fixed-frequency qubits and frequency-tunable couplers. The tunable couplers were lumped-element LC resonators, where the inductance came from a superconducting quantum interference device (SQUID). We achieved a controlled-phase gate with a fidelity of 99% by parametric modulation of the coupler frequency. Using this device, and another similar to it, we demonstrated two different quantum algorithms, the quantum approximate optimization algorithm, and density matrix exponentiation. We achieved high algorithmic fidelities, aided by our carefully calibrated gates. Additionally, we researched parametric oscillations using frequency-tunable resonators. Previously, degenerate parametric oscillations have been demonstrated by modulation of the resonant frequency at twice that frequency. We use this phenomenon to implement a readout method for a superconducting qubit with a fidelity of 98.7%. We demonstrated correlated radiation in nondegenerate parametric oscillations by modulating at the sum of two resonant frequencies of a multimode resonator. We showed an excellent quantitative agreement between the classical properties of the oscillations with a theoretical model. Moreover, we studied higher-order modulation at up to five times their resonant frequencies. These types of parametric oscillation states might be used as a quantum resource for continuous-variable quantum computing.
CLICK HERE TO DOWNLOAD THE WHOLE DISSERTATION. (in PDF format)