Building Erasure Qubits Using Dual-rail Superconducting Cavities

Abstract

Erasure qubits are engineered so that their dominant errors can be detected without the use of additional ancilla qubits. In a quantum error correction (QEC) context, erasure qubits can achieve higher error thresholds and larger effective code distances than qubits limited by Pauli errors. As such, they have emerged as a promising path to reducing the overhead of scaling fault-tolerant quantum computation.

In this presentation, we discuss how we realize erasure qubits using a dual-rail encoding in a pair of superconducting cavities, where the logical states are defined in the single-photon manifold of the two cavity modes. The dominant error in this system is single-photon loss to the vacuum state, which can be readily detected as an erasure. Residual errors are primarily dephasing, at roughly an order of magnitude lower rate, and bit-flip errors are remarkably rare, occurring at the parts-per-million level.

We also present a full gate set for operating dual-rail cavity qubits and show that this error hierarchy is largely preserved across all operations. In particular, we focus on the implementation of our two-qubit gate and on the special properties that make it especially useful for QEC. We conclude with an outlook on how erasure detection and the error hierarchy can be used to realize efficient and scalable QEC.

Scroll to Top
Skip to content