QSE Quantum Seminar: "Demonstrating high-fidelity operations using dual-rail cavity erasure qubits"

Please join us for the QSE Center Quantum Seminar with Nitish Mehta from the Quantum Circuits company who will give the talk "Demonstrating high-fidelity operations using dual-rail cavity erasure qubits" on Thursday February 5th from 12pm to 1:30pm.
Location: CE 1 105

Pizzas will be available at 12:00. All PhDs, postdocs, students, group leaders, and PIs are welcome to join us.

TITLE: "Demonstrating high-fidelity operations using dual-rail cavity erasure qubits"

ABSTRACT: 
Erasure qubits have re-emerged as a strategy to significantly reduce hardware requirements for quantum error correction through both higher thresholds and more favorable logical error suppression with increasing code distance. Such erasure qubits are designed to ‘self-report’ most of their errors without the need for additional ancilla qubits. In this talk, we will show how to realize erasure qubits using a simple superconducting cavity-based dual-rail encoding. In this dual-rail cavity qubit we encode a qubit in the single-photon manifold of two cavity modes. The dominant error channel is single photon loss to the vacuum state which can be efficiently detected, converting this photon loss error into an erasure-like error.  The residual dephasing and bit-flip Pauli errors in our system are much rarer, with error rates at least an order of magnitude smaller than our photon loss rates. This strong hierarchy of errors is necessary for our dual-rail cavity qubits to perform well in a quantum error correction setting such as in a surface code. We also describe a new controlled-Z gate, which uses an auxiliary transmon-based coupler as a source of dispersive coupling between two dual-rail qubits. We benchmark this gate and experimentally show that it has high-fidelity and low-erasure rates, while also exhibiting a rather novel error ‘asymmetry’, whereby the control qubit suffers from more decoherence than the target qubit, leaving the target qubit mostly error-free. We show how to leverage this error asymmetry for error correction. Finally, we combine all these operations into a multi-qubit system and show some early results of experiments on a system consisting of five dual-rail cavity qubits. We perform logical state preparation and measurement operations in a distance-2 surface code, showing how the code can correct for a single erasure and detect a single Pauli error.