High-fidelity Quantum Error Correction Achieves Twofold Threshold Increase with Singlet-Triplet Spin Qubits

The challenge of maintaining quantum information during computation demands innovative approaches to error correction, particularly as quantum systems scale up. Adam Siegel and Simon Benjamin, researchers at the University of Oxford, have now demonstrated a significant advance in this field by exploring the potential of singlet-triplet spin qubits for fault-tolerant quantum computation. Their work establishes these dual-spin qubits as a natural fit for ‘erasure’ coding, a technique that protects against information loss during quantum operations. By introducing a novel leakage-detection protocol and combining it with established error correction techniques, the researchers achieve a substantial improvement in error correction thresholds and a significant reduction in logical error rates. This research paves the way for practical, high-fidelity quantum computation using shuttling-based architectures in semiconductor devices.

Erasure conversion for singlet-triplet spin qubits enables high-performance shuttling-based quantum error correction. Adam Siegel and Simon Benjamin investigated the potential of utilising singlet-triplet (dual-spin) qubits within semiconductor quantum dot devices to improve quantum error correction. Fast and high fidelity shuttling of spin qubits has already been demonstrated in these systems, and several architectures leveraging this capability have been proposed. The research focuses on a fault-tolerant framework specifically designed for shuttling-based quantum error correction, suggesting singlet-triplet qubits could be optimal for achieving the highest possible shuttling fidelities. This work presents a novel approach to maintaining quantum information during the physical movement of qubits, a crucial step towards scalable quantum computation.

Leakage-Aware Error Correction with Dual-Spin Qubits

Such dual-spin qubits establish them as a natural realisation of erasure qubits within semiconductor architectures. Researchers introduce a hardware-efficient leakage-detection protocol that automatically projects leaked qubits back onto the computational subspace, without the need for measurement feedback or increased classical control overheads. When combined with the XZZX surface code and leakage-aware decoding, the team demonstrates a twofold increase in the error correction threshold and achieves orders-of-magnitude reductions in logical error rates. This establishes the singlet, triplet encoding as a practical route toward high-fidelity shuttling and erasure-based, fault-tolerant quantum computation in semiconductor devices.

Quantum error correction (QEC) is one of the cornerstones of practical quantum computation, essential for mitigating the effects of noise and decoherence. Stabiliser codes, such as the surface code, are leading candidates for QEC due to their high threshold and relatively simple implementation requirements. However, these codes typically require a large overhead in physical qubits to protect a single logical qubit, necessitating the development of more efficient qubit encodings and error correction strategies. Recent proposals have explored the use of erasure codes, which offer the potential to reduce this overhead by encoding quantum information in a subspace and detecting when qubits leak out of this subspace.

Surface Code Stabiliser Measurements with Spin Qubits

Quantum error correction (QEC) is considered essential for building practical quantum computers, offering the potential to reduce logical error rates to levels suitable for complex algorithms. The principle relies on using logical qubits constructed from multiple physical qubits, and scaling error correcting codes like the surface code can achieve exponential error suppression if the physical component failure rate remains below a specific threshold. Implementing QEC involves repeatedly measuring parity operators, known as stabilisers, which can require billions of repetitions during a single computation. Therefore, a fast, high-fidelity and scalable implementation of these stabilisers is crucial for successful quantum computation.

Semiconductor spin qubits are emerging as a promising technology for this purpose, benefiting from long coherence times, scalability, and compatibility with existing manufacturing and cryogenic electronics. Compact silicon spin qubit processors have already been developed, and small-scale error correction codes have been successfully demonstrated. High-fidelity single- and two-qubit gates have exceeded 99%, with some reports indicating fidelities surpassing 99.9%, crucial for maintaining the integrity of quantum information. The CZ and CNOT gates are particularly well-suited for stabiliser measurements within silicon-based spin qubit systems, as they are established primitives that do not require complex transformations.

Furthermore, spin qubits exhibit exceptional shuttling capabilities, demonstrated by recent experiments achieving 99.99% fidelity per increment at high speed when moving electrons across a silicon structure. Theoretical studies suggest even higher fidelities are possible, and shuttling offers benefits such as mitigating dephasing noise and enabling motional averaging. This shuttling capability has significant implications for semiconductor quantum computers, enabling longer-range connectivity and reducing the overhead associated with compiling quantum algorithms. Novel architectures, including Crossbar, Spiderweb, and SpinBus paradigms, propose utilising shuttling to move ancilla and data qubits, effectively extending interaction ranges and creating space for control electronics, moving away from static qubit arrangements towards dynamic, mobile systems.

Singlet-Triplet Qubits Enhance Fault Tolerance Thresholds

This work establishes a fault-tolerant framework leveraging singlet-triplet qubits within semiconductor architectures, demonstrating their suitability as erasure qubits. Researchers developed a hardware-efficient leakage-detection protocol which projects leaked qubits back into the computational space without requiring measurement feedback or additional classical control. This protocol, when integrated with the XZZX surface code and leakage-aware decoding, achieves a twofold increase in the fault-tolerance threshold and significantly reduces logical error rates. The findings demonstrate that this encoding offers a viable pathway towards high-fidelity qubit shuttling and fault-tolerant computation in solid-state devices. The authors acknowledge limitations stemming from assumptions regarding measurement reliability, specifically the ability to perfectly differentiate between all qubit states, and note that modelling currently does not explore variations in the accuracy of state assignment. Future research could explore the implementation of these protocols with less precise measurement systems, potentially utilising simpler singlet-triplet and parity readout techniques, and investigate the impact of varying measurement accuracy on overall performance.