Persistent current readout for coherent quantum annealing
by
IFAE Seminar Room
In-Person
Supervisor: Pol Forn-Díaz
Quantum computing is one of the most intriguing challenges for physicists in this century. Among the different types of quantum computers, quantum annealers are perhaps the most promising in the middle term. Today, existing quantum annealers are limited by qubit coherence times, despite their complexity, and they are able to perform coherent quantum annealing processes only for a short time and cannot overcome classical computers. The main goal of this thesis is to advance towards the development of a coherent quantum annealer based on superconducting flux qubits by introducing a novel dispersive readout method. In this thesis’s defence, the first formulation and experimental results of a persistent current readout (PCR) circuit, composed of a dc-SQUID resonator, are presented.
Experiments on single, uncoupled flux qubits were realized with conventional dis persive readout methods to establish a benchmark for the PCR. After analyzing the conventional dispersive readout, other couplings are investigated. In particular, we investigate the galvanic coupling between a qubit and a resonator, and between two qubits. We show that, assuming a small shared inductance, the coupling Hamiltonian reduces to the classical energy stored in the coupling inductor itself. The first system has interesting applications for exploring different coupling regimes, such as the ultrastrong coupling regime, whereas the second system is relevant for studying a qubit coupled to a strongly nonlinear object. Similarly, the Hamiltonian of two qubits sharing part of the loops is derived. Applying the same methodologies, the coupling Hamiltonian is reduced to the classical energy stored in the shared inductance, in the limit of low coupling inductance, as in the previous case. This system is the simplest coupling between two qubits. Despite the limited practical value, the Hamiltonian describes the qubit behaviour when coupled to another strongly non-linear object, giving interesting insights for other inductive coupling. Building on these results, the PCR system is defined as dc-SQUID-based resonator coupled to a flux qubit via mutual geometric inductance. The dc-SQUID acts as a non-linear inductor, whose inductance depends on the qubit persistent current state. Fixing the dc-SQUID operational point, a change in the qubit persistent current causes a shift in the resonator frequency. Moreover, adjusting the flux in the dc-SQUID loop, it is possible to decouple the qubit from the readout circuitry, leading to an improvement of the qubit coherence times, while the readout is idle. Moreover, the Hamiltonian of the whole system is derived and particular emphasis is given to the two-photon contribution and the dispersive shift is calculated. To experimentally demonstrate the PCR, a single qubit device is designed and fabricated. On the chip, the flux qubit is coupled inductively to the dc-SQUID resonator and capacitively coupled to a coplanar waveguide (CPW) resonator. The presence of the CPW permits to perform the conventional dispersive readout measurements, being a benchmark for the PCR. Unfortunately, due to device imperfections, it was not possible to perform any qubit characterization. Nevertheless, the dc-SQUID shift due to the inversion of the qubit persistent current when crossing the sweetspot was observed through a flux period measurement. This result constitutes proof of the concept of the PCR we are presenting, pointing to a new way to design quantum annealers
Zoom link: https://us02web.zoom.us/j/89787514064?pwd=SkRaOElqanZRNFZXM2d2SE9PN1d0Zz09