Quantum error correction with silicon spin qubits

TLDR

Large‑scale quantum computers depend on quantum error correction, and silicon‑based spin qubits—compatible with mature nanofabrication—have recently achieved high‑quality one‑ and two‑qubit operations, yet demonstrating QEC with three or more coupled qubits remains an open challenge. The study demonstrates a three‑qubit phase‑correcting code in silicon that protects an encoded state against any single‑qubit phase‑flip error. The correction is achieved via a three‑qubit conditional rotation implemented as a single‑step resonantly driven iToffoli gate. The error‑correction protocol successfully mitigated one‑qubit phase‑flip errors and intrinsic dephasing, demonstrating functional QEC and highlighting silicon’s potential for large‑scale quantum computing.

Abstract

Abstract Future large-scale quantum computers will rely on quantum error correction (QEC) to protect the fragile quantum information during computation 1,2 . Among the possible candidate platforms for realizing quantum computing devices, the compatibility with mature nanofabrication technologies of silicon-based spin qubits offers promise to overcome the challenges in scaling up device sizes from the prototypes of today to large-scale computers 3–5 . Recent advances in silicon-based qubits have enabled the implementations of high-quality one-qubit and two-qubit systems 6–8 . However, the demonstration of QEC, which requires three or more coupled qubits 1 , and involves a three-qubit gate 9–11 or measurement-based feedback, remains an open challenge. Here we demonstrate a three-qubit phase-correcting code in silicon, in which an encoded three-qubit state is protected against any phase-flip error on one of the three qubits. The correction to this encoded state is performed by a three-qubit conditional rotation, which we implement by an efficient single-step resonantly driven iToffoli gate. As expected, the error correction mitigates the errors owing to one-qubit phase-flip, as well as the intrinsic dephasing mainly owing to quasi-static phase noise. These results show successful implementation of QEC and the potential of a silicon-based platform for large-scale quantum computing.