Pulse Shaping for Ultra-Fast Adiabatic Quantum Gates

TL;DR

This paper introduces Delayed Leakage Reduction (DLR), a novel control technique for quantum gates that suppresses leakage without IQ modulation, enabling ultra-fast, high-fidelity operations in spin qubits within 9.4 ns.

Contribution

The paper presents DLR, a new leakage suppression method applicable to baseband signals, improving quantum gate speed and fidelity beyond existing techniques like DRAG.

Findings

Achieved >99.9% fidelity in 9.4 ns for spin qubits.

Demonstrated DLR's effectiveness without IQ modulation.

Assessed hardware sampling rate impact on fidelity.

Abstract

A fundamental challenge in quantum computing is to increase the number of operations within the qubit coherence time. While this can be achieved by decreasing the gate duration, the use of shorter signals increases their bandwidth and can cause leakage into energetically separated states. A common method to suppress leakage for short pulses is the Derivative Removal by Adiabatic Gate (DRAG) method, which however, relies on IQ modulation of radio-frequency (RF) signals, thus cannot be applied to the baseband signals, e.g., for semiconductor spin qubits. This paper proposes a novel technique, Delayed Leakage Reduction (DLR), that suppresses leakage at targeted frequencies even for baseband control by using time-delayed repetitions of the control signal to enable rapid, high-fidelity operations. We apply DLR on the adiabatic CZ gate between two spin qubits and achieve fidelities exceeding 99.9% within 9.4 ns for a resonance frequency difference of only 100 MHz. Towards the experimental realization of the proposed control method, we also assess the impact on the fidelity of the sampling rate of the electronic hardware generating the control pulse, thus setting the minimum hardware requirements for any experimental demonstration.