Fast entangling gates enabled by micromotion in trapped-ion quantum processors

Abstract

A predominant challenge in realising near-future trapped-ion quantum computers (QCs) is the implementation of entangling gate operations that enable scalable quantum information processing, while maintaining both high fidelity and speed. Current approaches to trapped-ion entangling gates are based on adiabatic transitions, which scale poorly with the number of ions. A promising alternative is offered by so-called ‘fast gates’, which utilise carefully-designed sequences of ultrafast laser pulses to rapidly entangle qubits. However, fast gate design has historically neglected the RF-driven micromotion intrinsically experienced by trapped ions. While this is appropriate for few-ion systems, implementing fast gates in scalable QC architectures will require that micromotion is incorporated into fast gate design. This thesis accordingly investigates fast gates in the presence of micromotion as a step towards their realisation in scalable trapped-ion QCs. Importantly, this thesis builds off foundational results by Ratcliffe et al. [1], who demonstrated that large micromotion amplitudes can be leveraged to enhance fast gate performance. However, this study significantly constrained fast gate design to reduce the computational complexity of micromotion, and consequently only generated fast gates that require unrealistic experimental resources to implement. I remove these constraints in this thesis, and advance towards realising high-fidelity fast gates in the presence of micromotion that can be implemented with experimentally feasible resources. I demonstrate that micromotion-enhanced fast gates are not an artefact of the constraints used by Ratcliffe et al. [1]; removing these constraints reveals that micromotion dynamics can be exploited to decrease fast gate errors arising from unrestored ion motion, which has not previously been observed. I demonstrate that this enhancement allows fast gates in the presence of micromotion to be designed with fidelities above the fault-tolerant threshold required for error-correction (99.9%), with feasible laser repetition rates (100 MHz – 1 GHz), low pulse numbers (∼40) and realistic RF driving frequencies (20 – 80 MHz). As with previous fast gate studies [2, 3], I identify pulse imperfections as the limiting error that inhibits the experimental realisation of micromotion-enabled fast gates, and discuss pathways for overcoming this limitation using established laser control techniques. Show more