High-Fidelity Logical Magic State Preparation. A Step Towards Fault Tolerant Quantum Computing. New Research From Quantinuum et al.

The pursuit of scalable, fault-tolerant quantum computation necessitates robust methods for preparing and manipulating logical qubits, the units of quantum information that are protected from environmental noise. A critical component of this endeavour is the creation of ‘magic states’, essential resources for universal quantum computation beyond the limitations of Clifford gates. Researchers at the University of California, Davis and Sandia National Laboratories, alongside colleagues at Quantinuum, report a successful experimental demonstration of high-fidelity logical magic states achieved through a technique known as code switching. This involves preparing a magic state within one error-correcting code and transferring it to another, a complex process now realised using an ion-trap processor. Lucas Daguerre, Robin Blume-Kohout, Natalie C. Brown, David Hayes and Isaac H. Kim detail their findings in a study titled ‘Experimental demonstration of high-fidelity logical magic states from code switching’, showcasing a protocol that yields a logical magic state with state-of-the-art fidelity and a reported infidelity lower than the underlying physical operations.

Recent advances in quantum error correction offer a considerable boost to the pursuit of scalable, fault-tolerant quantum computation, particularly through the innovative application of code switching techniques. Researchers demonstrate a successful implementation of this approach, preparing high-fidelity logical magic states, essential components for building practical quantum computers and mitigating the limitations imposed by noisy quantum hardware. This work details an experimental demonstration utilising an ion-trap processor, yielding a logical magic state encoded with state-of-the-art fidelity and establishing a new benchmark for quantum error correction performance.

The experiment successfully implements a code-switching protocol, transferring an encoded magic state from a fifteen-qubit Reed-Muller code to a seven-qubit Steane code. This leverages the benefits of both codes to achieve a high-fidelity logical magic state. A magic state, in this context, is a specific quantum state required to implement non-Clifford gates, operations necessary for universal quantum computation but not natively available on many quantum platforms. The experiment achieves a preparation success probability exceeding 80%, with an observed infidelity of no more than 0.015, validating the effectiveness of the code-switching technique.

Significantly, this infidelity represents an improvement of at least a factor of ten compared to the inherent infidelity of the physical two-qubit operations employed within the protocol, indicating the processor operates below the pseudo-threshold for fault tolerance. The pseudo-threshold represents a critical point where the rate of error correction exceeds the rate of error accumulation, enabling reliable quantum computation. To further validate the encoded state, the team generates two copies of the magic state. It performs a logical Bell basis measurement, providing a sample-efficient method for certifying the quality of the encoded quantum information. A Bell basis measurement determines the entanglement between two qubits, confirming the coherence of the encoded state.

The demonstrated high-fidelity magic state complements previously demonstrated fault-tolerant Clifford gates, state preparation, and measurement within the 2D colour code architecture, completing a universal set of fault-tolerant primitives. Logical error rates achieved with these primitives are equal to, or better than, the physical two-qubit error rate, signifying a substantial step towards practical quantum computation and demonstrating the effectiveness of the implemented error correction strategies. Future work will likely focus on scaling this approach to larger code distances and exploring alternative code switching strategies to optimise fidelity further and reduce resource overhead.

To further validate the encoded magic state, researchers generate two identical copies on the same processor and perform a logical Bell basis measurement, providing strong evidence of the state’s fidelity and coherence. The prepared encoded magic state within the Steane code exhibits a probability of successful preparation, accompanied by an infidelity of no more than 0.003. This reported infidelity represents a significant improvement, being at least a factor of ten lower than the infidelity of the physical two-qubit operations employed within the protocol, indicating the ion-trap processor operates below the pseudo-threshold, a crucial benchmark for fault-tolerance.