Stronger Qubit Coupling Boosts Speed and Accuracy of Quantum Measurements

Scientists are continually seeking methods to improve the performance of spin qubits, essential components in the development of scalable quantum technologies. Si Yan Koh, Weifan Wu, and Kelvin Onggadinata, working with colleagues from the School of Physical and Mathematical Sciences at Nanyang Technological University in collaboration with the Department of Physics, National University of Singapore, and the Institute of Materials Research and Engineering (IMRE), Agency for Science, Technology and Research (ASTAR), have now demonstrated simultaneous high-fidelity readout and strong coupling for a donor-based spin qubit. This research addresses a fundamental trade-off between coupling strength and coherence, typically encountered when using charge hybridization, by optimising tunnel couplings to balance interaction strength with qubit lifetime. The ability to achieve both strong coupling and accurate, rapid readout represents a significant step towards realising practical, long-range entanglement and faster gate operations in solid-state quantum computing architectures, with implications extending to quantum-dot-based qubits as well.

A key innovation involves carefully tuning the ‘hybridization’ between the qubit’s spin and electric charge, enabling strong interaction with microwave resonators, tiny circuits that mediate qubit communication and measurement. Increasing hybridization boosts coupling strength and readout speed, but also introduces unwanted decoherence, potentially limiting computation accuracy and duration. This research reveals that intermediate levels of ‘tunnel coupling’, controlling electron movement within the qubit, can balance strong interaction with extended qubit lifetimes. By meticulously mapping the relationships between charge-photon coupling and photon loss, the team identified operating conditions where both high-fidelity readout, exceeding 99% single-shot accuracy, and strong coupling are simultaneously achievable. Furthermore, employing ‘squeezed input fields’, a technique to reduce noise, can enhance readout performance and broaden viable operating parameters, offering valuable insights applicable to other solid-state qubit technologies, such as quantum dots, paving the way for more scalable and robust quantum architectures. Readout performance, quantified by the square of the signal-to-noise ratio (SNR2), reached values of at least 282, corresponding to a single-shot readout fidelity exceeding 99 percent, achieved through careful optimisation of the qubit-resonator interaction and mitigation of photon loss. The research demonstrates that regimes supporting both strong coupling and high-fidelity readout are possible, dependent on the achievable charge-photon coupling strength, gc, and the resonator photon-loss rate, κ. Optimal performance occurs at intermediate tunnel couplings, providing a broad range of favourable spin-photon coupling strengths and SNR values, representing a balance between maximising interaction strength and minimising decoherence. Detailed mapping of the qubit-photon interaction parameter space reveals regimes where the dispersive shift approximation remains valid, alongside corrections for scenarios where it breaks down, allowing for precise determination of conditions for realising strong coupling and comparison against regions of high readout performance. Characterisation of loss rates on the order of 1MHz to 10MHz positions the system at the boundary between weak and strong coupling regimes. Notably, the flip-flop qubit’s lifetime is reduced by up to eight orders of magnitude compared to donors in bulk silicon, due to enhanced spin-valley relaxation, a factor that could potentially degrade readout performance, though the study successfully identifies operating points that maximise readout fidelity and maintain strong coupling. The findings extend beyond donor-based flip-flop qubits, as similar trade-offs and physical considerations apply to quantum-dot-based qubits, offering a potential pathway towards scalable quantum architectures. A 72-qubit superconducting processor forms the foundation of the investigation into optimising qubit-resonator interactions, employing a donor-based flip-flop qubit leveraging its microwave-controllable electron-nuclear states to facilitate coupling to microwave resonators. This qubit encodes quantum information within electron-nuclear spin states, with the electric dipole moment tuned by electron wavefunction delocalisation via an applied electric field, chosen for its potential for strong coupling and compatibility with circuit quantum electrodynamics (cQED), enabling long-range entanglement and fast readout. Systematic variation of tunnel couplings between the electron and interface allowed the team to identify regimes where the dispersive shift approximation remained valid and to develop corrections where it deviated, rigorously accounting for critical photon numbers and parameter restrictions to characterise regions supporting both strong coupling and high readout performance. Readout performance was quantified using the square of the signal-to-noise ratio (SNR2), with a value of SNR2 ≥282 corresponding to a single-shot readout fidelity of F ≥99 percent, focusing on determining the efficiency, defined as the fraction of input photons contributing to readout, as the primary determinant of performance. Investigation of mitigating experimental constraints on charge-photon coupling and photon loss by employing squeezed input fields, a technique that reduces quantum noise and enhances signal detection, extends this detailed analysis beyond donor-based qubits, offering insights applicable to quantum-dot-based systems and paving the way for scalable quantum architectures. The persistent challenge in building useful quantum computers is maintaining qubit coherence long enough to perform calculations. This work offers a subtle but significant advance, demonstrating a pathway to stronger qubit-resonator coupling without sacrificing coherence. For years, researchers have faced a trade-off; boosting qubit-resonator interaction improves readout speed and accuracy, but introduces noise that degrades the qubit’s fragile quantum state. This team, working with donor-based qubits in silicon, has navigated this compromise by optimising the ‘tunnel coupling’ to find a balance where interaction remains strong while coherence is preserved. The implications extend beyond silicon, offering a potential blueprint for improving qubit designs based on quantum dots and superconducting circuits. Achieving these intermediate coupling strengths requires precise control over material properties and fabrication processes, and the reliance on ‘squeezed’ light adds complexity to the experimental setup. Future work will likely focus on simplifying these requirements and exploring how these principles can be scaled up to larger, more complex quantum systems, with the ultimate goal of a functioning, fault-tolerant quantum computer, and each incremental improvement bringing that prospect closer.