Realisation of high-fidelity nonadiabatic CZ gates with superconducting qubits

Abstract

Abstract Entangling gates with error rates reaching the threshold for quantum error correction have been reported for CZ gates using adiabatic longitudinal control based on the interaction between the |11〉 and |20〉 states. Here, we design and implement nonadiabatic CZ gates, which outperform adiabatic gates in terms of speed and fidelity, with gate times reaching $$1.25/(2\sqrt 2 g_{01,10})$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mn>1.25</mml:mn> <mml:mo>∕</mml:mo> <mml:mrow> <mml:mo>(</mml:mo> <mml:mrow> <mml:mn>2</mml:mn> <mml:msqrt> <mml:mn>2</mml:mn> </mml:msqrt> <mml:msub> <mml:mrow> <mml:mi>g</mml:mi> </mml:mrow> <mml:mrow> <mml:mn>01</mml:mn> <mml:mo>,</mml:mo> <mml:mn>10</mml:mn> </mml:mrow> </mml:msub> </mml:mrow> <mml:mo>)</mml:mo> </mml:mrow> </mml:mrow> </mml:math> , and fidelities reaching 99.54 ± 0.08%. Nonadiabatic gates are found to have proportionally less incoherent error than adiabatic gates thanks to their fast gate times, which leave more room for further improvements in the design of the control pules to eliminate coherent errors. We also show that state leakage can be reduced to below 0.2% with optimisation. Furthermore, the gate optimisation process is highly feasible: experimental optimisation can be expected to take less than four hours. Finally, the gate design process can be extended to CCZ gates, and our preliminary results suggest that this process would be feasible as well, if we can measure the CCZ fidelity separate from the initialisation and readout errors in experimental optimisation.