Thesis

SQuAre: scalable qubit arrays for quantum computation and optimisation

Downloadable Content

Download PDF
Creator
Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2024
Thesis identifier
  • T16820
Person Identifier (Local)
  • 202091625
Qualification Level
Qualification Name
Department, School or Faculty
Abstract
  • The work described in this thesis summarises some of the key milestones achieved in the development of the UK’s first quantum computer based on arrays of neutral atoms, delivered in partnership with M Squared Lasers Ltd. These milestones include the commissioning of the 1064 nm optical dipole trapping system for holographic array generation, assembly and testing of a low phase noise microwave source used for global single qubit rotations and the calibration of the Rydberg laser system locked to a narrow linewidth ultralow expansion reference cavity. Using 133Cs atoms, we demonstrate high-fidelity control over large scale arrays with over 225 trap sites (> 100 qubits) by performing randomised benchmarking (RB). The recorded average single qubit gate fidelity of 0.99993(2), achieved using a global microwave driving field with composite pulses, is the highest recorded fidelity on any platform of this size and comfortably meets the thresholds of many error correction codes. To further advance the field towards realising error correction, we also present the first practical application of a non-destructive readout (NDRO) procedure based on a cycling transition in alkali atoms. By increasing the trap depth and reducing the array size to 49, the largest array size achieved with this technique to date, we are able to demonstrate suppression of state preparation and measurement errors during RB by a factor of 1.7 without affecting the average gate fidelity. Finally, we present some early analogue simulation results of the quantum one-dimensional Ising spin model on arrays of up to 9 atoms. Using this simple, classically tractable problem, we characterise the performance of our platform as an analogue optimisation device and identify key areas of improvement as the experiment evolves into a state where more challenging two-dimensional geometries can be simulated with a view of demonstrating quantum advantage in optimisation problems with practical relevance.
Advisor / supervisor
  • Pritchard, Jonathan D.
Resource Type
Note
  • Previously held under moratorium from 1st February 2024 until 31st March 2025.
DOI

Relations

Items

Thumbnail Title Date Uploaded Visibility Actions
file details PDF of thesis T16820 2024-12-09 Public Download