Dynamical Sweet Spots Achieve 3-5x Coherence Enhancement in Superconducting Qubits

Scientists are tackling the persistent problem of decoherence , the loss of quantum information , in superconducting qubits, a major hurdle in building practical quantum computers. Zhen Yang, Shan Jin, and Yajie Hao, from the Institute of Fundamental and Frontier Sciences at the University of Electronic Science and Technology of China, alongside Xiu-Hao Deng et al., present a novel approach to extending qubit coherence times by meticulously engineering ‘dynamical sweet spots’ (DSSs). Their research introduces a fully parameterized framework optimising both energy relaxation and pure dephasing simultaneously, revealing fundamental limits to coherence extension and achieving a 3-5x improvement over existing DSS strategies, maintaining coherence in the hundred-microsecond range. This work not only identifies robust operating bands insensitive to both DC and AC flux noise, but also designs control protocols demonstrating high-fidelity gate operations, establishing a powerful and general method for Pareto-front engineering of DSSs and significantly advancing superconducting qubit performance.

Optimising Qubit coherence via Flux Modulation requires precise

Scientists have developed a fully parameterized framework to engineer dynamical sweet spots (DSSs) in Superconducting qubits, substantially improving their coherence and gate performance. The research, published on January 28, 2026, introduces a multi-objective periodic-flux modulation technique that simultaneously optimises energy relaxation (T1) and pure dephasing (Tφ), quantifying the inherent trade-off between these crucial coherence metrics. This innovative approach surpasses existing DSS strategies, enhancing Tφ by a factor of 3-5 while maintaining T1 in the hundred-microsecond range for Fluxonium qubits with realistic noise spectra. The team achieved this breakthrough by moving beyond conventional single- and two-tone drives, allowing for arbitrary periodic flux modulations to expand the search space for optimal waveform design.
The study establishes a Pareto front, a graphical representation of the trade-off between T1 and Tφ, revealing optimal operating points for superconducting qubits. Through rigorous numerical optimisation, researchers mapped this front, identifying solutions that significantly extend coherence times compared to previously established DSS methods. Importantly, the work proves that, despite the effectiveness of DSSs in suppressing first-order sensitivity to low-frequency noise, the relaxation rate cannot be arbitrarily reduced, thereby establishing a fundamental upper bound on achievable T1. This finding clarifies the inherent limitations of coherence enhancement strategies and guides future research directions.

Furthermore, the team identified “double-DSS” regions within the optimised operating points, demonstrating insensitivity to both DC and AC flux. These robust operating bands offer significant advantages for experimental implementation, providing stable and reliable conditions for quantum computations. To demonstrate the practical implications of this work, scientists designed single- and two-qubit control protocols at these operating points, numerically verifying high-fidelity gate operations. Simulations reveal a single-qubit X gate achieving 99.9993% fidelity over 10 nanoseconds, and a √iSWAP gate reaching 99.995% fidelity within 28 nanoseconds.

This research establishes a general and versatile framework for Pareto-front engineering of DSSs, offering a powerful tool for optimising coherence and gate performance in superconducting qubits. By systematically exploring the parameter space of periodic flux modulation, the team not only achieved substantial improvements in coherence times but also uncovered fundamental limits governing qubit performance. The identified double-DSS regions and demonstrated high-fidelity gate operations pave the way for more robust and reliable quantum computations, bringing us closer to realising the full potential of superconducting qubits for advanced quantum technologies.

Fluxonium Qubit Optimisation via Periodic Modulation enables enhanced

Scientists engineered a fully parameterized, multi-objective periodic-flux modulation framework to simultaneously optimise energy relaxation (T1) and pure dephasing (Tφ) in fluxonium qubits. This work addresses a key challenge in quantum computation: extending coherence times under dynamical sweet spot (DSS) strategies and identifying fundamental limitations. The research team developed a method that moves beyond conventional single- and two-tone drives, enabling arbitrary periodic flux modulations to expand the search space for waveform design and optimise qubit performance. Experiments employed a multi-objective optimisation process to map the T1-Tφ Pareto front, explicitly revealing the trade-off between relaxation and dephasing times and identifying optimal operating points.

This innovative approach involved numerically simulating fluxonium qubits with realistic noise spectra, allowing precise control over the periodic flux modulation parameters. The system delivers enhanced Tφ values by a factor of 3-5 compared to existing DSS strategies, crucially maintaining T1 in the hundred-microsecond range, a significant achievement for qubit coherence. Researchers further proved that, despite DSSs eliminating first-order sensitivity to low-frequency noise, the relaxation rate cannot be arbitrarily reduced, establishing an upper bound on achievable T1. This theoretical finding provides a fundamental limit to coherence enhancement and guides future optimisation efforts.

The study pioneered the identification of double-DSS regions, insensitive to both DC and AC flux, creating robust operating bands for experiments and improving qubit stability. As applications, the team designed single- and two-qubit control protocols at these optimised operating points and numerically demonstrated high-fidelity gate operations. This demonstrates the practical utility of the framework and its potential to improve quantum gate performance. These results establish a general and useful framework for Pareto-front engineering of DSSs, substantially improving coherence and gate performance in superconducting qubits and paving the way for more robust quantum computation.

Fluxonium Qubit Coherence via Pareto Front Mapping optimizes

Scientists have developed a fully parameterized, multi-objective periodic-flux modulation framework for fluxonium qubits, substantially improving coherence and gate performance. The research addresses a key question regarding the limits of coherence extension when employing dynamical sweet spots (DSSs) to suppress decoherence from low-frequency flux noise. Experiments utilising this new framework enhance coherence by a factor of 3-5 compared with existing DSS strategies, while maintaining energy relaxation times (T1) in the hundred-microsecond range. The team measured both energy relaxation (T1) and pure dephasing (Tφ) simultaneously, mapping out a T1-Tφ Pareto front to explicitly reveal the trade-off between these two critical figures of merit.

Results demonstrate that, despite DSSs eliminating first-order sensitivity to low-frequency noise, the relaxation rate (T1) cannot be reduced arbitrarily close to zero, establishing an upper bound on achievable coherence. Analysis of the Pareto front revealed this fundamental limitation, providing insight into the inherent constraints of coherence optimisation. At the optimised working points, researchers identified double-DSS regions insensitive to both DC and AC flux, creating robust operating bands suitable for quantum experiments. These double-DSS regions offer enhanced stability and reliability for qubit operation, paving the way for more precise measurements and control.

As applications, single- and two-qubit control protocols were designed at these operating points, and numerical simulations demonstrated high-fidelity gate operations. Tests prove a single-qubit X gate achieves 99.9993% fidelity over a 10 nanosecond evolution, while a √iSWAP gate reaches 99.995% fidelity within 28 nanoseconds. This breakthrough delivers a general and useful framework for Pareto-front engineering of DSSs, enabling systematic optimisation of coherence and gate performance in superconducting qubits. Measurements confirm the framework’s ability to extend Tφ by factors of 3-5, representing a significant advancement in qubit control and stability. The study establishes that, even with advanced modulation techniques, fundamental limits exist on achievable T1, guiding future research directions and optimisation strategies.

Optimised Flux Modulation Boosts Qubit Coherence times

Scientists have developed a novel flux-modulation framework employing periodic modulation and Pareto-front optimisation to address decoherence in fluxonium qubits caused by low-frequency flux noise and dielectric loss. This research quantifies the trade-off between energy relaxation time (T1) and pure dephasing time (Tφ), identifying optimal operating points that maximise both coherence parameters simultaneously. The resulting Pareto front directly reveals the achievable limits of T1 and Tφ through periodic flux modulation. At these optimised points, researchers designed control pulses demonstrating high-fidelity X and √iSWAP gates, achieving fidelities of 99.9993% and 99.995% respectively, under realistic experimental conditions.

These findings demonstrate the potential for realising high-coherence quantum control via dynamical modulation, enhancing gate performance in superconducting circuits. The authors acknowledge that practical considerations, such as device anharmonicity, may influence optimal designs, suggesting preference-based multi-objective optimisation to tailor operating points to specific experimental goals. Furthermore, the developed Pareto-front engineering techniques are broadly applicable and extendable to other superconducting platforms, including transmon qubits, promising advancements across the field of quantum computing.