Error-detected entangling gate operations between dual-rail cavity cavity qubit – Part 2

Encoding dual-rail qubits in superconducting cavities is a promising platform which exhibits a strong hierarchy of errors. In these qubits the dominant error, single photon loss, may be detected and either post-selected or converted to an erasure. The remaining undetectable errors due to cavity dephasing are expected to be much smaller, creating a favorable error hierarchy. It is crucial that this error hierarchy is preserved throughout all qubit operations, especially the two-qubit gate.

In part 2, we focus on the experimental implementation of the gate. We detail our gate calibration protocol and benchmark gate performance, extracting both the fidelity after error detection and the effective erasure rate. Our results show our gate performs well for both metrics – with post-selected fidelities on par with the best two-qubit gates in any qubit platform. Finally, detailed study of the two-qubit error structure confirms these gates are well-suited for error-correction tasks.